This study analyses the influence of investment policy on economic growth in Rwanda. A special focus is placed on the post-genocide period and the role of tourism, fiscal and monetary policy and foreign direct investment.

The paper examines Rwanda's economy, beginning with an introduction to the country's economic framework and investment trends. It finds that Rwanda's economy, characterised by low development and high indebtedness, has a multidimensional dynamic. Special attention is paid to tourism as the main source of foreign exchange earnings. Government policies, which include tax cuts and inflation control, aim to promote economic growth. The study utilises dynamic panel data from 1997 to 2018 and applies the ordinary least squares (OLS) method to measure the impact of various economic factors.

Excerpt

Abstract

The overarching objective of this study is to measure the impact of investment policy on economic growth in Rwanda. The study begins by providing a brief introduction of Rwanda’s economy and domestic investment including and governance background. Rwanda’s economy is classified as least-developed and quite indebted. The study noted first, that economic growth, investment decisions are quite multidimensional. Second, tourism was found to be key to foreign exchange earnings for Rwandan Economy. Additionally, to realize higher levels of economic growth, the Government of Rwanda introduced fiscal, and monetary policies that included: openness, and controlling inflation. In the post genocide era, particularly in the economic recovery period, Investment through import, export and FDI were earmarked as a potential avenues for foreign exchange which was believed to overcome the constraints related to private capital in the economy. Based on these reflections, this study measures the impact of investment policy on Rwanda’s economic growth. This study covered period 1997-2018 by employing dynamic panel data.

In order to achieve this, the study related the dependent variable (GDP growth rate) as well as investment policy and other explanatory variables as a pioneer in economic analysis in the context of Rwanda. Thus, to understand the impact of investment policy and other explanatory variables on Rwanda’s economic growth, this study employed the latest econometric techniques to test empirically the hypotheses developed using Rwanda’s data. In this respect, first co integration analysis was introduced to capture long-run relationships among variables. Second, to capture short-run relationship among variables a systems simultaneous equation was developed. This is because Vector Autoregressive (VAR) treats all variables as endogenous. In this way through a simultaneous equation endogenous and exogenous variables are identified. Following this approach, employing VAR through Vector Error Correction Mechanism (VECM) procedure the simultaneous equation was simulated. The study further conducted ex-ante forecasting involving impulse response and variance decomposition simulations as well as ex-post forecasting to evaluate the period under study. Also the study examined causality relationships among series using VECM Granger causality approach that is utilized to understand short-run causality as well as endogeneity among variables via F-/Wald test simulation.

Later, the systems simultaneous equation aforementioned is estimated employing Ordinary Least Square (OLS) to measure the impact of exports, imports, FDI and other explanatory variables on the growth of Rwanda’s economy.

Empirical findings indicate that investment policy contributes to Rwanda’s economic growth. Nonetheless, the coefficient is negative. This is partly attributed to Total Factor Productivity (TFP), as explained by the Solow-Swan Model and the absorption capacity of Rwanda being a least-developed country. This situation is worrisome for the nation because a further review of the impulse response function indicates that investment will negatively contribute to economic growth both in the short-run and long-run. Meanwhile, tourism in indicated by the findings as an export that can act as incentive economic growth in the country. This is because tourism is found to play a significant role in attracting foreign exchange into the economy and at the same time acting as a tool for openness. The study finds tourism as important tool for economic growth through spill-over effects thus, accelerating tourism-induced foreign investments into the country. To this end, tourism plays a pivotal role to Rwanda’s economy as mentioned earlier but like investment policy (exports, imports and FDI), the coefficient are negative. However, unlike investment policy, though in the short-run impulse response indicates that tourism will negatively contribute to economic growth in the long-run the impact becomes positive but diminish. Due to that, it is important that mechanisms need to be put in place to make Rwanda a better tourist destination compared to other countries in the region.

A further analysis of the finding reveals that regarding declining TFP the study recommends first, a review of the nation’s monetary, fiscal and commercial policies as well as further human capital development. This is because the current policy regime seems to be more oriented towards promoting export-led growth without considering the negative internal impacts of say imports on the nation’s foreign exchange reserves. Second, the study recommends improving the nation’s absorptive capacity so as to accelerate consumption of goods and services and promote international trade and future investment.

Keywords: Economy; GDP; FDI; Rwanda; growth, Investment.

Acknowledgements

Firstly I take this opportunity to express my sincere deep gratitude to my supervisor Prof. Huang SHAOAN. During the time of his supervision my thinking about economic issues greatly changed. It deepened my analytical thinking and focus on the thematic areas of this study. I have benefitted greatly from his great insight, knowledge, critics of my work and always demanding that i do better. This made me to work hard and in turn the quality of the thesis improved and enhanced my research skills. His unwavering support is basically priceless. Prof. Huang SHAOAN has indeed been more of a mentor than a supervisor to me.

He set the necessary pace and showed me the way even when I was not yet well grounded in Doctoral research methodologies. Indeed it’s through his constructive and productive sessions where my research skills were reshaped and enhanced, besides his desire that i work harder to achieve the best through my course.

In the same spirit, allow me extend my heartfelt gratitude to all of my Lecturers at Shandong University (SDU), I am sincerely grateful to the academic support particularly when engaged in my course works, I greatly harnessed your analytical skills in all course units under your stewardship, which greatly changed my economic reasoning. May your priceless effort be a source of peace in your entire earthly life.

In the same spirit, I acknowledge that error is to human, whereas edit is to man!

Abreviations

ADF: Augmented Dickey-Fuller

ASSM: Augmented Solow-Swan Model

ARMA: Autoregressive Moving Average

BNR: Banque Nationale du Rwanda

CPI: Consumer Price Index

DFGFI: Dian Fossey Gorilla Fund International

EDPRS: Economic Development and Poverty Reduction Strategies

ELGS: Export-Led Growth Strategy

FDI: Foreign Direct Investment

GATS: General Agreement on Trade in Services

GE: Government expenditure

GoR: Government of Rwanda

GNP: Gross national product

HIC: High-Income Country

ICT: Information Communication Technology

IGCP: International Gorilla Conservation Project

IMF: International Monetary Fund

JML: Johansen’s Maximum Likelihood

KPSS: Kwiatkowsiki-Phillips-Schmidt-Shin

MRW: Mankiw, Romer and Weil

MIC: Middle Income Country

MINICOM: Ministry of Commerce

MTWA: Ministry of Tourism, Wildlife and Antiquities

NST: National Strategies for Transformation

NGT: New Growth Theory

NLLS: Non-Linear Least Square

NTEs : Non-Traditional Exports

OLS: Ordinary Least Square

PP: Phillips-Perron

PIC: Public Investment Committee

REPA: Rwanda Export Promotion Authority

RIEPA: Rwanda Investment and Export Promotion Agency

RPA: Rwanda Patriotic Army

RTA: Rwanda Tourism Agency

RWA: Rwanda Wildlife Authority

PSTA: Rwanda’s Strategic Plan to Transform Agriculture

RANU: Rwandese Alliance for National Unity

SDU: Shandong University

TOT: Terms of Trade

TFP: Total Factor Productivity

TWG: Tourism Working Group

TEs: Traditional Exports

UNCTD: United Nations Conference on Trade and Development

VAR: Vector Autoregressive

VECM: Vector Error Correction Mechanism

WCS: Wildlife Conservation Society

WDI: World Bank Development Indicators

WTO World Trade Organization:

CHAPTER ONE:

GENERAL INTRODUCTION OF THE STUDY

1.1 Introduction

Studies have shown that Economic growth relies on the dynamic capacity of an economy in order to boost the level of national income of the society. Economic growth indicates the increment of services and goods developed by the economy. It is associated with potential output at full employment (Sial MH, Hashmi MH, Anwar S, 2010). Investment rate is the fundamental factor to economic growth to examine the economic performance of a country. Domestic investment is related to the change in capital to improve the economic growth in an economy over time. The investment brings change in a country’s capital stock during a given period, this can be measured over the time. Investment is the source of production of goods and services which are employed to produce other goods in an Economy (Canh NT &, Phong NA, 2017). Public and private investments increase the economic activity to create new sources of producing goods and services to stimulate the economic growth of Pakistan. Furthermore, it is observed that economic growth in Pakistan only requires modification and development of domestic investment. Emphasizing on domestic investment inevitably enhances modification of the economy in the society.

In Rwanda, Due to relative low exportable activities, relative high import inflows, with stable exchange rate, inflationary rate, and economic growth would be accelerated over time. In spite of the direct Government involvement in domestic investment, limited study has been conducted to ascertain correlation between domestic investment and economic growth in Rwanda. Even if many works were conducted on the effect of domestic Investment policy on economic growth, domestic investment is likely to enhance economic growth (Ndikumana, 2000). Thus, this study is planned to bridge this gap in the literature through scrutinizing the connection between domestic investment and economic growth in Rwanda. The national investment is a significant factor which influences the economic growth of any developing economy Economic growth looks like a foremost component of the business cycle. Moreover, local investment has an association with several macroeconomic factors, which drives countries to pursue the asset choice for creating a positive climate for economic development (Charles Ruhanga, 2017).

Domestic investment on public infrastructures like roads, sewerage connections, electricity and power generation, education, health, and communication projects play a vital role to increase production of goods and services in the economic activity of Rwanda. Investment plays a significant role in driving growth by enriching productivity levels (Haq AU, 2012).

1.2 Background of the study

Studies have revealed that Globally, Domestic investment policy has an important place in the political economies of Developing countries, because it is very paramount in accelerating and achieving economic development. Its impact on several economic variables and the international economic reality is proof that the countries of the world are racing to join the international competition. The size of its investment stock uses various means and methods to help stimulate and stabilize investment together with GDP growth rate of the economy (Ali N, Hussain H, 2017). Most of the existing literature treats the current flow of investments together with government expenditure as the source of contribution to productive capacity. For example, (Aschauer & Greenwood, 1985; Aschauer, 1988; Barro, 1989; Turnovsky & Fisher, 1995) do so in neoclassical Ramsey framework.

Further studies confirms that, at global level, economic growth relies on the vibrant aptitude of an economy in order to boost the level of national income of the society. Economic growth indicates the increment of services and goods developed by the economy. It is associated with magnitude of domestic investments, and potential output at full employment (Sial MH, Hashmi MH, Anwar S, 2010). Investment rate is the fundamental factor to economic growth to examine the economic performance of a country. Domestic investment is related to the change in capital to improve the economic growth in Pakistan. The investment brings change in capital stock during a given period. Investment can be measured over the time. Investment is the source of production of goods and services which are employed to produce other goods [Canh NT, Phong NA, 2017). Public and private investments increase the economic activity to create new sources of producing goods and services to stimulate the economic growth of Pakistan.

Furthermore, it is observed that economic growth in Pakistan only requires modification and development of domestic investment through realistic and robust investment policy decision. Emphasizing on domestic investment inevitably enhances modification of the economy in the society.

Due to high exportable activities and low inflationary rate, economic growth would be accelerated in Pakistan, (Sial MH, Hashmi MH, Anwar S, 2016), employ a simple ‘A-K’ endogenous growth model.

While the flow specification has the virtue of tractability, it is open to criticism that insofar as productive government expenditures are intended to represent public infrastructure, such as roads and education, it is the accumulated stock, rather than the current flow, that is relevant. Despite this criticism, few authors have adopted the alternative approach of specifying productive government expenditure as stock. (Arrow & Kurz, 1970)

In Pakistan, In spite of the heavy Governmemnt involvement of domestic investment, a little study is relatively conducted between correlation domestic investment policy and economic growth in Pakistan. Even if many works were conducted on the effect of Domestic Investment policy on economic growth. Domestic Investment policy is likely to help the economic growth (Ali N, Hussain H, 2017). Thus, this study is planned to bridge this gap in the literature through scrutinizing the connection between domestic investment and economic growth in the society of Pakistan. Its equally believed that domestics investment is associated with macroeconomic environment, which in effect, motivated policy framers to encourage governments to pursue ridht decision on asset choice to cultivate positive climate for economic development (Bakari S .2017). Domestic investment on public infrastructures like roads, sewerage connections, electricity and power generation, education, health, and communication projects play a vital role to increase production of goods and services in the economic activity of Pakistan. Investment plays a significant role in driving growth by enriching productivity levels (Haq AU, 2012). Pakistan is one of developing countries to face various problems to boost the economic growth. It requires more exports, human capital and capital (Romer D .2006). Pakistan needs to get benefit from better economic openness by its amalgamation into economies of globalization. With the help of globalization, Pakistan should maintain the pillars of the development of national economy. Thus, the study tries to investigate empirically the influence of domestic investment and economic growth in Pakistan.

In Algeria, in 1993, Algeria underwent a period of transition, from a centralized socialist approach to a market economy. In this manner, her natural resources played the most important role. Algeria has Africa's fourth-largest economy.

Algeria's national income is estimated at more than $ 211.9 billion in 2014, with GDP growing 4 percent from last year. Socialism also played its role in disrupting the agricultural role, headed towards the industrial sector without ruddiness, but the arrival of President Chazli Bennid confirmed the importance of changing the old investment policy as a whole. The events of Black October in 1988 were behind the acceleration of the reform process.

Political and Economic Reforms during the President's period, the world oil price slump in 1986 was behind the country's crisis at the time. The oil sector is the mainstay of the Algerian economy, accounting for about 60% of the general budget, 30% of GDP and 97% of total exports. Algeria aspires to reduce the dependence on oil revenues by focusing on agriculture to limit the import of agricultural products such as cereals, potatoes and fruits in particular. And the development of export of other products such as dates, which is famous for. Algeria also has other natural resources such as iron, coal and uranium. (Baxter &King, 1995).

It’s important to note that investment is a powerful channel for innovation, economic growth and therefore poverty reduction. Recent empirical studies have established linkages between investment and economic growth (e.g., Barro, 1991; Barro & Lee, 1993; Ben-David, 1998; Collier & Gunning, 1999; Ghura & Hadjimichael, 1996; Hernandez, 2000; Khan& Reinhart, 1990; Ndikumana, 2000). Analysis of causality between economic growth and domestic investment conducted in different countries are marred with ambiguities and inconclusive results. For example, several researchers have found bi-directional relationship (Tang, Selvanathan & Foreign, 2008; Tan & Lean, 2010). Others found the direction of causality to be from economic growth to domestic investment (Choe, 2003; Quin, Cagas, Quising & He, 2006) while some found the direction of causality to be from domestic investment to economic growth (Villa, 2008). Also in other studies, private investment was shown to be super-exogenous, meaning investment was the primary determinant of economic growth (Montek, 2002).

In Rwanda, the economy has made significant progress in poverty reduction and has improved the conditions of doing business, enhancing domestic product to boost export, conscious spending on imports, only where import substitution was not feasible (World Bank, 2011).

Different policies have been adopted in order to increase gross domestic product through appropriate domestic investment policy decisions but there has been no empirical study which has attempted to establish the relationship between the growth of GDP and investment.

In other words, the question about the forecasting power of investment policy and economic growth remains a moot point. The few and sketchy studies that exist are mainly descriptive in nature and offer limited understanding of the relationship for policy prescription in Rwanda (Charles Ruranga, 2017).

Rwanda is a land-locked country located in east and central Africa. It borders Uganda to the north, Tanzania to the east, the Democratic Republic of Congo to the west, and Burundi to the south. Rwanda covers 26,338 square kilometres of land. The current population is about 10.7 million, exhibiting a very high population density of 407 inhabitants per square kilometre. Agriculture and Services are the principal sectors contributing to more than 80% of GDP. Coffee and tea are the main primary products exported and they constitute 40% of export earnings (Jonas BARAYANDEMA & Idrissa NDIZEYE, 2018).

Due to limited diversification of its economy, Rwanda’s balance of payments has continued to be unfavorable with current account balance always in the negative. After the 1994 genocide, Rwanda Government embarked on a new development path through policy review, particularly on domestic investment policy. The new government ushered in peace, political stability, good governance and minimal corruption among others. As a result, Rwanda’s economy has since 2002 been experiencing robust, resilient and sustained GDP growth in the East African region averaging over eight percent annually (Charles Ruranga, 2018).

The Rwanda government has also made significant efforts to promote private sector led growth to spur domestic investment currently at 22% of GDP (World Bank, 2012). As a consequence, Export grew by 1.66%., imports reduced by 21.3% asa result of direct support to local industries involved in import substitution, indeed, extreme poverty has fallen from 40% in 2000 to 24% in 2011. Though still high, the percentage of the population living below poverty line has significantly reduced from 77.8% to 44.9% between 1994 and 2011 respectively. This is attributed to realistic investment policy decision (NISR, 2011; Charles Ruranga, 2017).

1.3 Statement of the Problem

National investment policy is an important tool for investment and production, and spurs a nation’s economic growth, thus leading to employment generation in the economy. Favorable investment policies would be the main avenue developing countries to achieve significant strides in growth potentials, including Rwanda, which are without well-developed capital markets.

For GDP to increase, investments have to increase significantly especially in the manufacturing sector and agriculture, which is the backbone of developing countries. As investment related inflows increase in a nation, GDP, market size and consumption increase, as does employment and poverty reduction in the long-run. (Charles Ruranga, 2017).

Rwanda, like other developing countries, joined the race to attract investment inflows, and streamlining import demands after enacting policies such as the reforms that led to the establishment of National investment Registration Authority, and perhaps due to macroeconomic stability and the establishment of policies that create an environment favorable to investment, this explains why apparently, Rwanda has successfully become a leading favorable investment destination in East Africa(NISR, 2011).

Available evidence further reveals that the Rwanda government has also made significant efforts to promote private sector led growth to spur domestic investment currently at 22% of GDP (World Bank, 2012). Extreme poverty has fallen from 40% in 2000 to 24% in 2011. Though still high, the percentage of the population living below poverty line has significantly reduced from 77.8% to 44.9% between 1994 and 2011 respectively (NISR, 2011).

Despite of the above exploits by the Government, there is little or no knowledge about the impact of investment policy on country’s economic growth (Charles Ruhanga, 2018). Therefore, this research arose because of a gap in empirical work on the impact of Domestic investment policy on economic growth in Rwanda. For decades, Exports, imports and FDI have been recognized as realistic tools for economic growth. Considering the reforms and accelerated investment inflow increases, the milestones achieved could have had large effects upon economic growth in Rwanda. Even if Rwanda was a prosperous nation, a study would be necessary to establish the impact of Export, import and FDI dynamics on economic growth.

Currently, based on PSIS surveys and a few previous studies that concentrated on a causal relationship between FDI and economic growth, it is assumed that foreign investment is positively related to the economy’s growth rate. Therefore, as the impact of Domestic investment policy on economic growth, employment and poverty is often assumed to be positively related, with no empirical justification. For this reason, there was a need to test the hypothesis with this study’s key questions with whether there is a long-run relationship between investment policy and Economic growth of Rwanda:

The aim of this study was to analyze and establish the unknown feedback mechanism between GDP and Exports, Imports and FDI for shaping the development policy in Rwanda. A dynamic panel data model was used to analyze the dynamic relationship between gross domestic product and investment policy in Rwanda in order to answer key Questions such as;

What is the impact of Export dynamics on the Economic growth of Rwandan Economy?

How do import inflows affect the growth of Rwandan Economy?

What impact can FDI inflow have on the growth of Rwandan Economy?

1.4 Purpose of the study

He study sought to examine the implications of investment policy on per capita real economic growth in Rwanda.

1.5 Specific Objectives

The study was guided by two specific objectives as stated below;

To establish effect of Export dynamics on the Economic growth of Rwandan Economy

To examine the effect of import inflows on Economic growth of Rwandan Economy

To investigate the effect of FDI inflow on the growth of Rwandan Economy

1.6 Research Hypothesis Development

Export dynamics & GDP growth

Export refers augments the extent to which a nation opens to the flow of goods and services traded internationally, including the flow of international investment.

It entails the adoption of trade liberalization policies, where barriers to trade and investment are reduced (International Chamber of Commerce 2013; WTO 1995, 2006). The main link between export trade and economic growth, and investment is trade openness that underpins international trade engagements, which significantly positively influence economics growth.

Dollar & Kray (2002). Export trade and investment play a significant role in accelerating economic growth, and employment in an economy. Due to such shortcomings, to measure openness, the ratio of total trade (exports 0 to GDP is most commonly adopted (Barro, 2003; Wigeborn, 2010).

The effects of export trade on economic growth are of two folds. First, demand for domestic investment on export goods forms part of aggregate foreign exchange reserve in the economy. Thus, a rise in investment in export goods will, to a very large extent, stimulate production of investment goods which in turn leads to high output, economic growth and development. Secondly, capital formation improves the productive capacity of the economy in a way that, the economy is able to produce more output for exports. Further, investment in new plants and machinery raises productivity growth by introducing new technology, which will also lead to faster economic growth. This implies that export dynamic has significant positive impact on GDP growth rate in an economy. (Ipumbu and Kadhikwa, 1997; Khan, M. & D. Villanueva1998). It is therefore hypothesized that:

H0₁: Export dynamics has a significant positive association with GDP growth rate

Import dynamics & GDP growth

The direct effects are that import demands are an integral part of the national account, while the indirect effects are that remittances affect macroeconomic behaviors through their effects on exchange rate and relative prices.

Amuedo-Dorantes & Pozo (2006); Woodruff & Zenteno (2007) postulate that import demands negatively affects growth by reducing foreign exchange and finance for business investment. To this end, the majority of the literature argues that imports affect economic growth by reducing consumption, savings or investment. Indeed, after reviewing several case studies, Lucas (2005) finds that imports may indeed have served deterring factor to local investment funds in Morocco, Pakistan, and India. Glytsos (2002) modeled the direct and indirect effect of imports on incomes and hence on investments in seven Mediterranean countries and found that import demands negatively affected GDP growth rates in six out of seven countries.

Moreover, import demands increase recipient country income and output growth. Ratha and Riedberg (2005) argue that import demands deterred the recipient individuals' incomes and reduce the recipient country's foreign exchange reserves. Indeed, he observes that if import demand is a form of expenditure that is inversely related to output growth and, may not generate positive multiplier effects (see, for example, Stahl & Arnold, 1986).

Adelman and Taylor (1990) found that for every dollar spent on imports, its gross national product (GNP) decreased by $2.69 to $3.17.

According to MINICOM, Revised National Export Strategy, (2016), the rapid rise in imports of goods and services since 2000 greatly constrained the country’s foreign exchange reserve. But so does the even more rapid rise in imports and hence the burgeoning trade deficit signifying a direct negative relationship between import demands and domestic output. It is therefore hypothesized that:

H0₂: Import demands has a significant negative association with GDP growth rate.

FDI inflow & GDP growth

There are a number of studies that have investigated the relationship between FDI and GDP. Demello (1997) lists two main channels through which Domestic and Foreign may be growth enhancing.

First, FDI can encourage the adoption of new technologies in the production process through technological spillovers which is likely to have a positive impact on output growth in an economy. Second, FDI may stimulate knowledge transfers, both in terms of labour training and skill acquisition and by introducing alternative management practices and better organizational arrangements.

A survey by Caves R. (1996) underpins these observations and documents that 11 out of 14 studies have found Domestic and foreign investment to contribute positively to income growth and factor productivity. Both de Mello and OECD stress one key insight from all studies reviewed:

He emphasizes trade openness as a crucial determinant for the impact of FDI on growth, as they find two-way causality between domestic & foreign investment and growth in open economies, both in the short and the long run, whereas the long run causality is unidirectional from growth to Domestic & foreign investment in relatively closed economies. Though in both cases, the general conclusion was that, there was a significant positive association between Domestic & foreign investment and GDP growth rate. It is therefore hypothesized that:

H0₃: FDI inflow has a significant positive association with GDP growth rate.

1.7 The study scope

This study focused on data related to investment policy and growth of Rwandan economy from 1997 – 2017 (the last 20 years) basically in the post-war period of economic recovery.

1.8 The significance of the Study

A number of studies on investment especially in developing countries have been carried out. Nevertheless, experimental proof of investment policy on economic growth has been limited (Mohsin,1997). In Rwanda, the presence of little empirical analysis in this context makes this study vital to show the position of investment policy and formulation incentives in the economy sector.

Moreover, analyzing investment in Rwanda is of interest in two spheres, namely, policy and academic. Thus in due course, as policies are considered, if the investment does have clearly stronger consequences on growth, it may undervalue the necessity for rationalizing private/public investment. The study is also an addition to existing literature on the econometric analysis of investment policy to the growth of an economy.

1.9 Research methodology

This study employs Rwanda’s annual panel data on each of the variables: economic growth, Exports, Imports, and FDI. This study covers the period 1997–2018, and the main data sources were the World Bank Development Indicators (WDI) Database and the Rwanda Bureau of Statistics and NISR Databases.

Unit root tests employed the Augmented Dickey-Fuller (ADF) approach (Song & Witt 2000). The ADF method is validated by the Phillips-Perron (1988) and Kwiatkowsiki-Phillips Schmidt-Shin (1992) approach.

During the study, a number of simulations were conducted to establish the impact of Domestic investment policy in Exports, imports and FDI on Rwanda’s economic growth. First, to test a long-run relationship among variables, this study employs the cointegration approach, based on Johansen’s Maximum Likelihood Method multivariate cointegration test, developed by (Johansen, 1988, 1991, 1995; Johansen & Juselius, 1990). Second, to examine the causal link variables, the Granger Causality approach was employed, developed by (Granger 1969). Following this, we conducted the Pairwise Granger causality tests, to understand the existence of endogeneity and non-causality or causality between the variables studied. Third, following Song and Witt (2000) to capture the short-run and long-run relationships between variables, the study employs VAR through VECM procedure. This approach allows for the investigation of the long-run relationships between variables in the equilibrium, including the short-run correction from the variable to the equilibrium.

Also following this approach, the study simulated ex-ante forecasting, where impulse response and variance decomposition covered a period of 23 years. Finally, to solve the simultaneity problem, following VECM the study first developed a systems simultaneous equation. As the simultaneity issue had been solved, and having applied all endogenous variables to all equations, we estimated the system employing OLS. The system was validated. Later, as it is not possible to validate each equation and also conduct ex-post analysis under OLS, we employed the Non-Linear Least Square/Autoregressive Moving Average (Non- Linear Least Squares (NLLS)/ARMA), adopting the Gauss-Newton/Marquardt steps method. Each equation has been validated by testing for stability, serial correlation, heteroscedasticity and normality.

1.10 Thesis Structure

This thesis was structured into 10 chapters.

Chapter Two presents an overview of Rwanda’s political, governmental and economic history. It examines the transition of the country’s economy since Independence, in terms of broad economic indicators and its composition. The chapter provides an account of the country’s economy since the adoption of economic reforms in early 1990s, outlining trends in the growth of Rwanda’s socioeconomic indicators. Following the economic reforms, the country started to experience accelerated economic growth because the GOR introduced fiscal and commercial policies, such as controlling government expenditure (GE), inflation and adopting openness as a policy to stimulate international trade and investment.

Following these initiatives tourism has become the single largest foreign exchange-earning commodity as an export for Rwanda.

Chapter Three is divided into three discusses Export outflows, import inflow and FDI in Rwanda by first indicating the historical background of foreign investments into the country, and examining investment nationalization and investment performance after the reforms, as well as the regulatory framework that provides a pro-investment environment. Second, the chapter provides an account of investment inflows since the reforms, and explains the regulatory framework. Before the reforms, the chapter explains that investment inflows-Exports became negative due to the political and economic instability exacerbated by international sanctions between 1972 and 1979.

Regarding the regulatory framework, the chapter explains that Rwanda is signatory to a number of international, regional and bilateral agreements, such as the World Trade Organization (WTO).

Chapter Four covered key modeling economic growth. The chapter begins be explaining the background to economic growth, focusing on the SolowSwan Model, Mankiw, Romer and Weil (MRW) model and the New Growth Theory (NGT). The chapter explains that considering these theories, the Augmented Solow-Swan Model (ASSM) is a better theory for explaining economic growth, employment and poverty. This is because by augmenting the original Solow-Swan Model, the MRW Model and the NGT are incorporated into one model. The study finds that ASSM explains that innovations are a tool for increasing economic growth, employment and poverty reduction in the long-run. However, the theory indicates that this is subject to TFP and a nation’s absorption capacity.

Chapter five covers modelling Exports, Imports and FDI and explanatory variables as a means of establishing the approaches that can be employed to measure their effects on the dependent variables (economic growth, employment and poverty, based on the Solow-Swan Model). The other explanatory variables include: tourism, openness, Governent Expenditure, inflation, and telecommunication. The study concludes that first; Domestic investment and tourism are foreign flows into developing countries such as Rwanda, which supplement a nation’s private-sector investment gap. Second, the study explains that Domestic investment, tourism are factor inputs.

Third, the study finds that following the Solow-Swan Model explains that these factor inputs depend on increasing TFP so as to benefit a nation positively. Fourth, the study finds that telecommunication is an innovation in the Solow-Swan Model, and as such is a pro-poor technology for a developing nation such as Rwanda. Finally, the chapter finds that openness and inflation are innovations in the Solow-Swan Model.

Chapter Six: presenting the theoretical framework and the scope of the study, sources of data and by defining the variables and their measurement approaches. This is followed by an explanation of the procedure through which the study was conducted; that is, empirical analysis measuring the impact of Investment policy and explanatory variables on Rwanda’s economic growth. The procedure was comprised of different stages that represent independent chapters.

The first involved testing the time-series properties of the variables, which included explaining approaches such as the preliminary investigation of the relationship among variables, correlation analysis, unit root testing and endogeneity analysis simulation methods. The second procedure presented the approaches adopted to measure the short-run and long-run relationships among the variables, including ex-ante forecasting, simultaneous equation specification and validation approaches. The third procedure involved simultaneous equation simulation methods, diagnostic approaches, results presentation and interpretation, as well as Granger Causality methods. After these simulations, the findings and conclusions are presented as a summary of the chapters.

Chapter Seven: This involves testing the time-series properties. First, all series are transformed into logarithmic form, followed by constricting graphs as a means of deepening the understanding of the relationship among variables. Later, correlation analysis, trend analysis and endogeneity tests are conducted, followed by unit root, employing ADF, Phillips-Perron (PP) and the Kwiatkowsiki-Phillips-Schmidt-Shin (KPSS) tests. According to the findings, the series are non-stationary at level but stationary at first difference.

Chapter Eight. This chapter estimates the short-run and long-run relationship among the cointegrating vectors employing Johansen’s Maximum Likelihood (JML) Method. To capture a short-run relationship among the series, the study first establishes a simultaneous equation. Using VAR, which explicitly uses JML Method through the VECM procedure, short-run relationships are captured, employing the F-statistics or Wald Chi-square test.

Based on the same approach, ex-ante forecasting is conducted through impulse response and variance decomposition using the Monte Carlo procedure via the Cholesky-dof adjusted ordering. The study finds that short-run and long-run relationships exist among the series.

Chapter Nine: this estimates the impact of Exports, Imports and FDI on economic growth in Rwanda. The study first examines causality tests among interrelated variables. To capture these interrelationships, the study employs VECM, a procedure that opens an avenue through which causality can be tested among variables. The second section involves estimating the simultaneous equation developed in Chapter Nine, using OLS.

Chapter Ten: This summarizes the study, provides its conclusions, limitations as well as some suggestions for areas of future study. The key findings are fourfold. First, investment policy significantly contributes to Rwanda’s economic growth. Second, tourism significantly contributes to Rwanda’s economic growth through spill-over effects. Third, Rwanda’s local resources, including labour and human capital, are important to the nation’s economic growth. Fourth, the coefficients for factor inputs included in the study being Domestic & foreign investment, and human capital are negative, meaning that declining TFP is explained by the Solow-Swan Model.

CHAPTER TWO:

OVERVIEW OF RWANDA’S POLITICAL, GOVERNMENTAL

AND ECONOMIC HISTORY.

2.1 Introduction

This chapter examines the transition of the country’s economy since Independence, in terms of broad economic indicators and its composition. The chapter provides an account of the country’s economy since the adoption of economic reforms in early 1990s, outlining trends in the growth of Rwanda’s socioeconomic indicators. Following the economic reforms, the country started to experience accelerated economic growth, employment and poverty because the Rwandan Government introduced fiscal and commercial policies, such as controlling government expenditure (GE), inflation and adopting openness as a policy to stimulate international trade and investment. Following these initiatives tourism has become the single largest foreign exchange-earning commodity as an export for Rwanda.

2.2 Summary of Rwandan Economy

According to World Bank (2018), Small and landlocked, Rwanda is hilly and fertile with a densely packed population of about 12.5 million people (2018). It borders the far larger and richer Democratic Republic of Congo, as well as its closest East African neighbors, Tanzania, Uganda, and Burundi.With the support of the International Monetary Fund (IMF) and the World Bank, Rwanda has been able to make important economic and structural reforms and sustain its economic growth rates over the last decade.

Until independence in 1962, Rwanda was administered as part of the Belgian Congo. Belgium recognized the king or Umwami of the pre-colonial state as the traditional ruler of the territory but increasingly administered the territory itself, making use of its nobility. As decolonization became foreseeable, a new elite rooted in the peasantry challenged the country’s noblemen. With the support of the departing Belgian administrators, the new Hutu elite prevailed. Due to the violence that followed independence, about half of the Tutsis fled abroad and settled in adjacent countries. Ethnicity however, was only one facet of a complex societal structure. 6. At the time of independence, the already densely populated country had almost no modern infrastructure, and mineral resources, mostly tin, were minimal (see Box 1.1). A moderate climate and, in some areas, rich volcanic soils have made it possible to successfully introduce such cash crops as tea and coffee as a foundation for a potentially productive agricultural sector. However, subsistence livelihoods continued to predominate, and a rapidly growing population has tightened the pressure on land; the results are smaller and smaller holdings, overuse of land, and soil degradation. Until the early 1980s, substantial foreign development aid helped Rwanda to realize modest economic growth.

But as world market prices fell for minerals, coffee, and tea, and fiscal difficulties increased, the economy began to falter. The poor economic performance undermined the regime’s legitimacy, and the fragile social fabric began to show rifts.

Box 1.1: The Physical Context Rwanda is situated in the equatorial zone in Central Africa. It is a landlocked country with few fossil fuel resources. The country has some hydropower potential and access to geothermal and solar energy. From 1990 to 2015, cropland increased drastically, mainly at the expense of woodland and sparse forest. Rwanda has been exposed in recent decades to temperature rises above the global average. Because of steep topographical gradients and poor farming techniques, the soil nutrient balance in Rwanda is among the most negative in Africa.a a Intergovernmental Panel on Climate Change

2.3.1 Political Context

Rwanda has guarded its political stability since the 1994 genocide. Parliamentary elections in September 2018 saw women fill 64% of the seats, the Rwandan Patriotic Front maintain an absolute majority in the Chamber of Deputies and, for the first time, two opposition parties, the Democratic Green Party of Rwanda and Social Party Imberakuri, winning two seats each in the parliament. President Paul Kagame was re-elected to a seven-year term in the August 2018, following an amendment to the constitution in December 2015 allowing him to serve a third term.

2.3.2 Social Context

Rwanda’s strong economic growth was accompanied by substantial improvements in living standards, with a two-thirds drop in child mortality and near-universal primary school enrollment. A strong focus on homegrown policies and initiatives has contributed to significant improvement in access to services and human development indicators. Measured by the national poverty line, poverty declined from 59 to 39% between 2001 and 2014 but was almost stagnant between 2014 and 2017. The official inequality measure, the Gini index, declined from 0.52 in 2006 to 0.43 in 2017.

2.3.3 Economic Overview

Rwanda now aspires to reach Middle Income Country (MIC) status by 2035 and High-Income Country (HIC) status by2050.This aspiration will be carried out through a series of seven-year National Strategies for Transformation (NST1), underpinned by detailed sectoral strategies that are aimed toward achievement of the Sustainable Development Goals.

The NST1 came after the implementation of two, five-year Economic Development and Poverty Reduction Strategies—EDPRS (2008-12) and EDPRS-2 (2013-18), under which Rwanda experienced robust economic and social performances. Growth averaged 7.5% over the decade to 2018, while per capita growth domestic product (GDP) grew at 5% annually

2.3.4 Development Context

Public investments have been the main driver of growth in recent years. External financing through grants, concessional and non-concessional borrowing played an important role in financing of public investments. Growth slowdown of 2016 and 2017 highlighted the limits of public sector-led growth model. Going forward, the private sector will play a bigger role in helping to ensure economic growth. Low domestic savings, skills, and the high cost of energy are some of the major constraints to private investment. Stronger dynamism in the private sector will help to sustain high investment rate and accelerate the growth. Promoting domestic savings is viewed as critical.

The economic and social costs of Genocide were enormous. Within a year the country’s GDP fell, from an already low base, by half and Rwanda became the second poorest country in the world with widespread hunger and food insecurity. The social and physical infrastructure was devastated, trade links were ruptured, businesses and agricultural assets dwindled, and institutions of governance required rebuilding. Insecurity and instability were pervasive, generated both internally and from threats across the region. Social trust had collapsed. The Government of National Unity (GNU) that took over in 1994 was dominated by the RPF and RPA and relied on the Arusha Agreement as the legal basis for a transitional government. Through the rest of the 1990s, the RPF gradually consolidated its political power. In 2000 Parliament elected Paul Kagame president and, after a referendum endorsed a new constitution in May 2003, in August he won the presidential election.

After the country had been pacified and reconciliation was underway, relief and reconstruction became the cornerstone of the new government. In December 1994, the GNU issued a Declaration of Principles that outlined its political, social, and economic agenda for a “New Rwanda.” The declaration emphasized social stability, national security, and a commitment to a market economy, backed by a capable state and a strong private sector. National reconciliation and healing evolved through homegrown initiatives, such as establishment of first a National Unity and Reconciliation Commission in 1999 and then the gacaca courts, modeled on local customary approaches to dispute settlement.

Box 1.2: Growth of the State Rwanda had a precolonial tradition as a state. The legitimacy of the post-independence nation-state has never been questioned. Nevertheless, Rwanda’s monolingual society has been deeply divided between socioeconomic groups interpreted as being ethnically different, a division that was exacerbated by the colonial legacy. Tensions between elite groups for control of the state that often turned violent led to large waves of emigration in the 1960s and 1970s, and again during the 1994 Genocide against Tutsis. But among young people, identification with the Rwandan nation is growing. Rwanda’s stability and security are now impressively high. According to the 2016 World Internal Security and Police Index, Rwanda is now ranked 50th in the world for the ability of its security apparatus to respond to internal challenges and is ranked 1st (equal with Botswana) in Sub-Saharan Africa. The administration functions well throughout the country. The replacement of former municipalities by new territorial entities, and the reform of their functions, has created a strong presence for the dominant party, RPF; at the center over provincial, district, and sector administrations down to the villages. Government officials are bound by a detailed personal performance contract (imihigo) with the President of the Republic, reflecting the strategic objectives of the central government. The system has proved to be an effective tool of performance management and top-down approach toward socioeconomic transformation.

Source: Bertelsmann Stiftung, BTI 2018 Country Report — Rwanda. Gütersloh: Bertelsmann Stiftung, 2018

2.4 State of the Rwandan Economy

Rwanda has made significant achievements in its recovery from the 1994 genocide against the Tutsi. The Government of Rwanda (GoR), through its public expenditure program has played a crucial role in the process of promoting socio-economic reconstruction. More than two decades later, the country is aiming for long-term sustainable development, economic transformation and poverty reduction as elaborated in the Vision 2020 and the Second Economic Development and Poverty Reduction Strategy (EDPRS 2). Both, long and medium term development strategies, aspire to lead Rwanda into middle-income category by 2020 (World Bank, 2016).

The overarching goal of Vision 2020 is to accelerate progress to middle income status and better quality of life for all Rwandans through sustained average GDP growth of 11.5% and accelerated reduction of poverty to less than 20% of the population. The Government also targets to increase the proportion of investment accounted for by the private sector, and in particular the export sector, to increase foreign exchange earnings and enhance the nation’s external Balance of Payments position, which goes along with one of the pillars of Vision 2020 to become a private sector-led economy (NISR, 2017).

Therefore, public expenditure in general should support the achievement of the goals highlighted in Vision 2020 and EDPRS 2. Public investment has a central role to play in this respect; first, through the creation of wealth itself, and second, through its capacity to facilitate the creation of wealth by enabling the private sector, thus facilitating private investment. Past experience has shown that economic growth has been an important contributor to sustainable reduction of poverty in Rwanda. Economic growth significantly depends on the volume and quality of investment among other factors. In this regard, including the private sector in the delivery of public investments via Public Private Partnerships and Joint Ventures can play a pivotal role in supporting the accelerated delivery of strategic national investments. It can also yield multiplier effects in the economy including job creation; value for money and quality services without jeopardizing debt sustainability.

All this underscores the importance of public and private investments in contributing to the overall development goals of Rwanda and hints to the relevance and challenges of a sound investment policy fit to fulfill the expectations.

In 2009, the “National Public Investment Policy” had been approved to guide the Government in its investment program, focusing on improving efficiency and efficacy of the public investment portfolio and increasing the coordination between public and private investments, including Public Private Partnerships. (Adewumi S, Hacker S, Dzansi J, 2008).

According to National Institute of Statistics of Rwanda, (2011), much has been achieved since then, especially with regard to building transparent and accountable processes for public investments. The national public investment program established in Rwanda is: (a) part of the regular planning and budgeting cycle, (b) based on strategic development goals outlined in the EDPRS 2, 7 Year Government Program and Vision 2020 priorities, (c) consultatively agreed upon using an investment committee and (d) effectively used as an operational document.

The National Public Investment Policy from 2009 has especially led to establishing the Public Investment Committee (PIC) as point of entry and scrutiny for all central government projects. Experience has shown that this measure has considerably contributed to the improvement of the quality and transparency of public projects. Significant progress has also been achieved in other areas of public investments, which now call for further policy guidance: e.g. development of a PPP framework and the introduction of PPPs, alternative forms of investment engaging the private sector, advancements in decentralization.

The objective of this policy is to achieve the country’s strategic development goals by transforming the “National Public Investment Policy” into a “National Investment Policy”, which addresses public as well as private investment. It is intended to lay the ground to carefully balance new public investment projects and potential dis-investment needs with options to strengthen private sector participation. This means building upon existing principles and widening the scope of the policy to efficiently cover the involvement of private sector in public investments through PPPs and Joint Ventures.

This “National Investment Policy” is intended to guide the country in its investment program by ensuring:

1. Prioritisation of investments based on strategic goals, which also guides long-term budgeting and debt-management;

2. Improving implementation through feeding back execution data to ensure strategic and efficient management of the project portfolio and;

3. Transparency and accountability over the investment cycle to enable budget agencies on central and local level to plan and prioritise effectively; 4. Engaging the private sector and leveraging alternative sources of financing by: increasing confidence in a credible pipeline of projects and systematically targeting a wider range of strategic investment forms.( World Bank. Report on Doing Business, Washington, DC. 2012).

2.5 Rwandan Key Economic Reforms

2.5.1 Rwanda’s Economic Policy Reforms, 1960-2008

From 1961 to 1990, Rwanda had an administered economy which imposed severe restrictions on trade and foreign exchange transactions, as well as fixed exchange rate regime (IMF, 2005). By the early 1990s the average tariff rate was 34.8 percent, with 5 different tariffs ranging from 0-60 percent. Every import and every importer was subject to a quota, and all import operations were subject to a license authorizing external currency disbursement (WTO, 2004). Exporters had to repatriate currency generated by the sale of exports as a legal requirement, and export licenses were authorized only by the Banque Nationale du Rwanda (BNR). More importantly, all export earnings were transferred to and managed by the BNR. Likewise, BNR had to give prior approval for certain invisible transactions including medical care, tourist trips and study abroad, with purchases of currencies from the BNR to finance these invisible transactions subject to ceilings.

The period from 1991 until 1994 corresponds to the beginning of the removal of restrictions on trade and foreign exchange transactions, and the gradual revival of a market economy.

(See box 3 below)

Box 3: The Success of Rwanda’s Post-genocide Policy Reforms on Economic Transformation In 1995,a number of economic reforms were implemented in Rwanda. Rwanda embraced a market economy characterized by both a continuation of trade reforms and a liberalization of the monetary and financial regimes. Tariffs were reduced considerably with the average rate decreasing to 18 percent, and there remained four tariff bands with a maximum of up to 30 percent by 2003 which a significant reform when compared with an average tariff rate of 34.8 percent, with 5 different tariffs ranging from 0-60 percent prior to 1994. Liberalization of the monetary and financial sector led to the adoption of new currency exchange regulations, the creation of new private commercial banks, and the privatization of banks that had been state-owned (Coulibaly, 2005).Current account operations (imports, exports, services) were liberalized, and some of the previous restrictions on capital flows were either reduced or eliminated. The latter included the transfer of capital and revenues related to foreign direct investment (FDI), and the allowance of free withdrawal from foreign exchange accounts in commercial banks (Kanimba, 2004). Flexible exchange rates were also introduced. During the period 1995 to 2003, the commitment of the government to trade, financial, and exchange reform was much more credible and stable. Prices began to reflect real cost and value, rather than the arbitrary levels established by the government.

Economic resources could thus be allocated much more effectively as firms adjusted their productive capacities and subsequently improved the overall competitiveness of the Rwandese economy. At the same time Rwanda was the recipient of substantial humanitarian aid from the World Bank and other entities. Although diminishing returns to aid may exist over the long run, in the case of Rwanda the post 1994 aid has had a much greater, and long-lasting impact (Collier, 2004).

While this period covers the first phase of trade reforms, it was also characterized by macroeconomic and political crisis which finally culminated into the genocide against the Tutsi in Rwanda. The genocide led to the destruction of manpower, capital stock, and resources such as livestock, as well as a total absence of the state. In the process, much of the social capital was destroyed and a climate of uncertainty became prevalent.

Due to the business destruction and uncertainty, the few businesses that did survive were reluctant to incur the sunk cost of capital expenditures made possible by the previous wave of trade reforms. Many established businesses failed to adjust their productive capacities and suffered from the resulting inefficiencies.

Rwanda has implemented a number of policies to shape its economic transformation agenda and these policies continue to evolve depending on changing needs of the economy. Rwanda’s vision is to build a knowledge-based economy and to become a private sector led middle income country by 2020. Rwanda’s ambitious programme for development is encapsulated in Vision 2020. The Economic Development and Poverty Reduction Strategy (EDPRS) is the mid-term framework to implement the Government’s long-term development agenda (Ministry of Finance and Economic Planning 2007). The EDPRS is based on three pillars designed to accelerate economic growth and promote human development:

1. Sustainable growth for jobs and exports - investing in improving the climate for business investment, thereby achieving private-sector growth. In the shorter term the priority is reinforcing the productive and export potential of the agricultural sector, but in the longer term the goal is to diversify the economy by promoting the non-farm sector.

2. Vision 2020 Umurenge is a pro-poor rural development and social protection programme. It aims to eliminate extreme poverty by 2020 through releasing the productive capacity of the very poor. It includes public works, credit packages and direct support and is implemented at village level using participatory methods;

3. Good economic governance is seen as a precondition for poverty reduction and development by creating a comparative advantage in ‘soft infrastructure’ (good governance and institutional arrangements important for private investors) thus compensating for Rwanda’s relatively poorly developed hard infrastructure and disadvantaged geographical location (Ministry of Finance and Economic Planning 2007).

Rwanda takes a developmental state approach with the key objective being sustainable economic growth and social development. The main aim of EDPRS was to overcome the key constraints to economic growth identified through a growth diagnostic and investment climate analysis by: systematically reducing the operating costs of business; investing in the private sector’s capacity to innovate; and, widening and strengthening the public sector.

Government policy is to promote privatesector investment through good governance, a legal framework, promoting savings and the banking sector and investment in infrastructure, health and education including vocational training. The aim is to: create new jobs to absorb new entrants to the labour market and surplus labour created by the modernization of farming; facilitate technology transfer; the transfer of skills to Rwandan; an increase in the production of goods and services for export; and, generally promote economic growth.

Public investment is targeted to induce substantial private sector investment and foster growth in agriculture, manufacturing and the service sector. Investment is targeted at developing skill and capacity for productive employment, improving the infrastructure, promoting science technology and innovation and strengthening the Financial Sector. Reforms to the ‘soft’ infrastructure for business and reducing business costs were seen as the first priority. Incentives for Domestic & foreign investment including export processing zones and industrial parks were seen as an important element of the strategy. Partnership of Domestic & foreign investment with Rwandan companies was to be encouraged and stimulating domestic investment was also seen as integral element of the policy. Diversifying and increasing exports was also seen as central to the strategy and the Government has identified the main areas for export growth, beyond the strategic exports of tea, coffee, horticulture, hides and skins and minerals. These are tourism, mining services, business process outsourcing, silk textiles, fruit and vegetable processing and dairy processing (Ministry of Trade and Industry 2009).

Domestic & foreign investment was seen as bring a number of benefits beyond job creation including the investment of foreign capital, know-how and managerial skills and export promotion.

Domestic & foreign investment as well as local investment was to be encouraged in resource based manufacturing (e.g. tea and coffee), low technology products (e.g. footwear, textiles), high technology manufacturing (e.g. chemicals, ICT, pharmaceuticals) and services including tourism where there is seen to be a high potential for growth. The 2010 Development Driven Trade Policy Framework prepared by the United Nations Conference on Trade and Development and the Ministry of Trade and Industry (UNCTD 2010a) argues that the trade policy should be development-driven and not demand led.

It suggests that investment, including Domestic & foreign investment, should enable the diversification of exports and markets, build local processing industries that add value to exports especially in agriculture but also in manufacturing and services. Also investment should provide opportunities for employment in rural areas. It argues that tax reductions/exemptions in terms of tariffs should promote the inflow of industrial inputs and that consideration should be given to more strategically located export processing zones with more effective incentives provided. Generally it advocates making the financial regime effective and well administered. It recommends making financial incentives outcome based, targeted to development goals and designed to minimise the impact of taxation on companies ‘cash-flow’ (UNCTD 2006).

Rwanda’s agricultural policy is embodied in Rwanda’s Strategic Plan to Transform (PSTA) the agricultural sector and is now in its second phase. The policy is mainly concerned with the modernization of the agricultural sector and commercialization. The overall objective is to increase agricultural outputs and incomes under sustainable production systems for all groups of farmers and food security for all. The emphasis in increased output is on crops for export. There are four interrelated programs: intensification and development of sustainable production systems; support for the professionalization of producers; promotion of commodity chains and agribusiness development; and institutional development. This government policy seeks to modernize the agricultural sector and promote the production of cash crops for export as part of the broader drive for economic growth and transformation. This may well bring benefits to the population in the medium and long term as all benefit from economic growth and increased prosperity. However, there seems to be no indication that this policy is pro-poor in the short term since small holder farmers may end up as working poor agricultural wage labourers.

2.5.2 Macro-economic Indicators

a) Savings

Rwanda’s gross domestic savings as a proportion of GDP have shown a positive growth trend from negative levels prior to 1998 to a positve level of about 5% of GDP in 2009. Although domestic savings have grown steadily over time, they are still less than 10% which still low low when compared to the benchmark countries like Vietnam, and Malaysia whose current savings rates are above 10%.(See figure 3.2).

In terms of sustained economic transformation, the low domestic saving rate shows that per capita incomes among Rwanda‘s population are growing at a slow rate which may affect the impact of observed economic growth into improved livelihoods. This because low rates of domestic resource mobilization constitute a major bottleneck to sustaining productivity driven economic transformation in the largely informal private sector.

According to theWorld Bank, (2019), actual values, historical data, forecasts and projections of Gross domestic savings (% of GDP) in Rwanda was reported at 7.5955 % in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Gross domestic savings (% of GDP) .

Figure1: Gross domestic savings (% of GDP)

Figures are not included in the reading sample.

b) Grants; excluding technical cooperation (BoP; current US$)

Grants; excluding technical cooperation (BoP; current US$) in Rwanda was reported at 621440000 USD in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources.

Rwanda - Grants; excluding technical cooperation (BoP; current US$) - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure2: Grants; excluding technical cooperation (BoP; current US$)

Figures are not included in the reading sample.

c) Rwanda - ICT Service Exports (% Of Service Exports, BoP)

ICT service exports (% of service exports, BoP) in Rwanda was reported at 1.9988 % in 2017, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - ICT service exports (% of service exports, BoP) - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure3: Rwanda - ICT Service Exports (% Of Service Exports, BoP)

Figures are not included in the reading sample.

d)Rwanda - Income Receipts

Income receipts (BoP, current US$) in Rwanda was reported at 18603037 USD in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Income receipts - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure4: Rwanda - Income receipts

Figures are not included in the reading sample.

e)Rwanda - Exports Of Goods, Services And Income

Exports of goods, services and income (BoP, current US$) in Rwanda was reported at 2061212563 USD in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Exports of goods, services and income - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure5: Rwanda - Exports of goods, services and income

Figures are not included in the reading sample.

f) Rwanda - Foreign Direct Investment, Net Inflows (% Of GDP)

Foreign direct investment, net inflows (% of GDP) in Rwanda was reported at 3.1713 % in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Foreign direct investment, net inflows (% of GDP) - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure6: Rwanda - Foreign direct investment, net inflows (% of GDP)

Figures are not included in the reading sample.

g)Rwanda - Exports Merchandise, Customs, Current US$, Millions

Exports Merchandise, Customs, current US$, millions in Rwanda was reported at 944 in 2017, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Exports Merchandise, Customs, current US$, millions - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure7: Rwanda - Exports Merchandise, Customs, current US$, millions

Figures are not included in the reading sample.

h)Rwanda - General Government Final Consumption Expenditure

General government final consumption expenditure (current US$) in Rwanda was reported at 1416532614 USD in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - General government final consumption expenditure - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure8:Rwanda - General government final consumption expenditure

Figures are not included in the reading sample.

i) Rwanda - Imports Merchandise, Customs, Current US$, Millions, Seas. Adj.

Imports Merchandise, Customs, current US$, millions, seas. adj. in Rwanda was reported at 2.1782 in 2015, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Imports Merchandise, Customs, current US$, millions, seas. adj. - actual values, historical data, forecasts and projections were sourced from the World Bank on June of 2020.

Figure9:Rwanda - Imports Merchandise, Customs, current US$, millions

Figures are not included in the reading sample.

j) Rwanda - Household Final Consumption Expenditure Per Capita Growth (annual %)

Household final consumption expenditure per capita growth (annual %) in Rwanda was reported at 3.7695 % in 2018, according to the World Bank collection of development indicators, compiled from officially recognized sources. Rwanda - Household final consumption expenditure per capita growth (annual %) - actual values, historical data, forecasts and projections were sourced from the (World Bank, 2020; Lindemann, 2010).

Figure10: Rwanda - Household final consumption expenditure per capita growth (annual %)

Figures are not included in the reading sample.

2.5.3 Inflation Control and Economic Reforms in Rwanda

Inflation is a general rise in the level of prices paid for goods and services over time, and usually reported monthly, quarterly and annually. Across the world, inflation is one of the main economic challenges faced by households, because it affects the material living standards of communities, especially the poor in developing countries. Prices for goods and services rise but income remains constant. Inflation denies the basic commodities of life to households, such as food, clothing and housing.

In extreme cases, countries experience hyperinflation, which is when prices rise rapidly by 100% or more per annum. Conversely, nations may experience deflationary situations, when prices for goods and services decline. This may lead to a depression. In the 1990s, Rwanda experienced hyperinflation. Inflation was recorded at 216% per annum in 1999 (Bigsten & Kayizzi, 1999). As part of the ERP and PEAP initiatives, inflation started to decline, indicated by the Consumer Price Index (CPI).

Indeed Consumer Price Index Cpi in Rwanda decreased to 134.50 points in April from 137.00 points by that time scope.

Tables are not included in the reading sample.

2.5.4 Impact of Inflation on Economic Growth in Rwanda

Inflation in Rwanda can be attributed to increasing money supply, world energy and food prices. Since inflation leads to price increases, rural poor communities (comprising 78.6% of the total population in Rwanda) are most affected. This is because, first, the share of consumption in total income is larger for all consumer goods.

Second, the income for agricultural products and salary earners does not increase in a similar proportion to manufactured products. Thus, increases in price reduce the range of goods available and real incomes to the poor. In turn, savings, future investment and welfare reduce.

Considering the effects of inflation, Economic Recovery Projects became the avenues through which macroeconomic stability could be achieved in Rwanda. Targeting inflation and achieving high economic growth are two fundamental macroeconomic objectives of most economies (Sanday 2013; Kasidi & Mwakanemela, 2013).

In Rwanda, the government uses interest rates to control inflation, often increasing them. However, this means that the cost of borrowing increases while investments reduce. Since inflation affects consumption, production slows due to low purchasing power, and in-turn employment reduces.

2.6 Causes of Inflation in Rwanda

Despite successes in Rwanda’s monetary policy, the country continues to experience high and volatile inflation rates (Kabundi, 2012).The main causes of inflation in Rwanda can be considered both theoretical and empirical, originating from food and transport costs, fiscal and monetary factors, demand and cost factors and international factors.

2.6.1 Food and Transport Costs

About 84.23% of Rwanda’s population lives in rural areas, of which the majority practice subsistence agriculture. Households are concerned with producing food to feed their families, and earn income for basic needs such as clothing and essential commodities. Reliance on subsistence agriculture and life in underdeveloped rural areas creates challenges for Rwanda. First, as population increases, so does land fragmentation, while food production decreases. Second, Rwanda’s farming has remained rain-fed, meaning that harvests remain low, especially during droughts, and are exacerbated by pests and diseases (World Food Program, 2013). Due to declining agricultural production, Rwanda has become a food-insecure country. Households have two main food sources: markets and subsistence production.

The poorest rural households resort to purchasing their food, creating vulnerability to increasing food price rises and food inflation in the country.

Food insecurity in Rwanda is mainly attributed to the French colonial agriculture policy, lack of agriculture mechanization, population growth and political instability (Leliveld et al. 2013). During colonial times, the British did not encourage Rwandans to develop large-scale plantations.

Rwandans only practiced subsistence cash-crop farming, and depended on four traditional cash crops: coffee, cotton, tobacco and tea. Alongside cash-crop farming Rwandans grew enough food crops to feed themselves. Agriculture depended on smallholder subsistence farming, was rain-fed, not mechanized and depended on smallholders.

This legacy has continued until today. With population growth of about 3.25% per annum, agriculture output cannot cope with population growth, and Rwanda has become food insecure. In turn, food shortages create pressure on food prices, causing food price inflation.

Rwanda’s increasing food insufficiency is further exacerbated by the poor conditions of the national road network, which has become a major development issue (Booth, 2015; Gollin & Rogerson 2010). The main cause of the poor national road network is the low level of funding for the sector. During the period under study, the Government has spent about USD 3 million per annum covering administration, construction and maintenance in the road sector. Out of Rwanda’s total land surface of 140,323 square kilometres, only 12,999 kilometres, representing 8.8% is comprised of the government funded road network (NISR 2015). The limited paved road network of about 3,795 square kilometres, representing only 1.9% of the national funded road network, worsens the situation. The rest of the road network (about 55,000 square kilometres) is comprised of poorly constructed and unmaintained community roads. Due to the poor feeder road network, people often rely on walking, head loading and bicycle transport. The poor, small roadwork is further worsened by torrential tropical rainfall, which often damages the limited road infrastructure. Consequently, the cost of vehicle maintenance increases due to high demand for bicycle and vehicle spare parts leading to high food prices and food inflation.

2.6.2 Rwanda’s Fiscal Deficit and Monetary Factors

A nation’s budget is the basis for its economic growth, as it contains the infrastructural development and social service programmes reflected by the revenue and expenditure outlay. Before Independence, the British Government provided finances for development and social service delivery. After Independence, financing such programmes became the responsibility of the GoR.

However, developing countries such as Rwanda have a narrow tax base. To finance government programmes, foreign aid and seigniorage are important revenue sources, required to bridge the tax gap.

Since seigniorage is associated with money expansion, inflation is inevitable in Rwanda (Nampewo 2014; Kabundi 2012). As money expansion increases, so does inflation.

Money supply growth rate in 1997 increased sharply increased to 163.9% per annum, and headline inflation increased by 86% per annum. During this time, the increase in money supply was attributed to the need to increase crop finance requirements to subsidise agricultural production. Although the government supplements its budget deficit through money expansion, welfare declines in the long-run, especially among salary earners. This is because wages do not increase in proportion to inflation caused by money expansion in the country. Though money expansion is a major cause of inflation in Rwanda, there are other causes, attributed to the demand for goods and services.

2.6.3 Aggregate Demand and Cost Factors

Theoretically, inflation in Rwanda can first be attributed to demand and cost factors, leading to demand-pull and cost-push inflation (Kyeyune, 2012). Demand factors cause demand-pull inflation, whereby aggregate demand exceeds aggregate supply. As the gap between aggregate demand and supply increases, so does inflation. Faced with international sanctions, import of consumer goods became difficult, causing demand-pull inflation. For example, inflation was -10.8% in 1966, but skyrocketed to 206% per annum in 1999.

Shortages of commodities increase prices for consumer goods. The price for intermediate goods also increases, because cost factors can also cause cost-push inflation, due to a rise in the price of factors of production. As firms are driven by profit, a rise in cost of production leads to increase in prices for goods and services. Cost-push inflation helps producers pass the higher costs of production to consumers through high consumer prices. Higher costs of production can originate from increases in wages, raw materials, imports and indirect taxes, or a reduction in government subsidies (Modigliani & Papademos, 1975; Tobin, 1975).

2.6.4 External Factors

As countries become more integrated, inflation spreads worldwide. When prices rise in major industrialized countries, the effect of inflation spreads to developing countries, and trade relations become vulnerable. When food and fuel prices sharply rose between 2008 and 2011, the effects were transmitted from industrialized nations to developing countries. Meanwhile HIPCs such as Rwanda, were generally hit harder than advanced economies.

In Rwanda during the same period, the CPI rose from 85% per annum to 118.69%.

Since then, the CPI has continued to increase, affecting workers and poor households, whose incomes do not rise proportionally. Domestic factors that cause inflation in Rwanda are further complicated by international factors, such as world food and energy price shocks. In September 2011, food price inflation stood at 50.4%, while non-food inflation rose to 18.1% (Mugume 2011).

As a net fuel importer, world energy price volatility directly affects prices in Rwanda. National bureau of statistics measures inflation in Rwanda as headline and core. Headline inflation indicates the relative changes in prices of all goods and services in the consumption basket, usually reported monthly, quarterly and annually. Core inflation indicates relative changes in the prices of all goods and services in the expenditure basket, excluding food and energy.

2.7 Assessing the relevance of Openness in Rwanda’s Economy

Openness is a nation’s outward-oriented domestic and international trade policy, as well as investment through FDI. Openness allows a nation to access international markets, facilitating innovation and technology diffusion as knowledge dissemination (Ramanayake & Lee 2015). For a developing country such as Rwanda, openness can enable access to niche markets for exports, and access to cheaper advanced technology not available at home, for the manufacture of goods. Openness allows a nation to increase domestic production and access cheaper goods, which in turn increases the standard of living.

The GoR introduced openness to promote trade and investment as a means of increasing economic growth and employment and reducing poverty.

The Export-Led Growth Strategy (ELGS) initiative was devised (Kaberuka, Rwakinanga & Tibesigwa, 2014); (Ministry of Trade, Tourism and Industry 2007).10 To promote exports, by Act of Parliament in 1996 the Rwanda Export Promotion Authority(REPA) was established, supervised by the Ministry of Trade, & Tourism.

The role of openness in a nation such as Rwanda can be reflected through the sources that enhance the growth of GDP per capita, namely capital accumulation and productivity (Babula & Andersen, 2008; Selassie, 2008). Openness can ease the transfer of international flows from industrialized nations to developing nations.

For a developing country such as Rwanda, international trade is the immediate channel through which capital—in the form of goods and services, skills, humans and physical capital - can flow.

Through international trade, produced goods and services can find a market beyond Rwanda. Imports can enable, Rwanda based firms can access technology transfer such as Information Communication Technology (ICT). Through movement of persons, skills can easily be transferred to Rwanda through training in Rwanda and abroad. Also, Rwanda can become a better destination for tourism. FDI can also facilitate capital accumulation, which in turn enhances production and productivity. Trade benefits to Rwanda can be indicated through four key aspects: contribution to the current account, tax revenue, production and productivity.

2.7.1 Contribution of Openness to the Current Account

Since adopting openness, Rwanda has witnessed remarkable improvement in international trade. Total trade increased from USD 1,012 million in 1985 to USD 12,908 million in 2014 as indicated under Appendix 2.4. During the same period, exports increased from USD 204.6 million to USD 5,219,655 million, while imports increased from USD 528,243 million to USD 7,688,318 million. The openness index increased from 0.29 in 1985 to 0.49 in 2014. However, as illustrated by Figure 2.7, the external balance of trade declined from USD -45 million in 1985 to USD -2,469 million in 2014.

Despite the worsening Terms of Trade (TOT), domestically produced goods have gained market access abroad. Increased imports can be attributed to the need for intermediate goods that are required for rehabilitating critical sectors such as manufacturing. In this way, consumer goods would increase in the country through import substitution.

2.7.2 The Contribution of Openness to Production and Productivity

International trade can enable a nation to specialize in the production of goods and services of comparative advantage. Trade increases commercial activities and investment in a nation through private-sector development (UNCTAD 2014). To achieve accelerated growth through trade, the Rwandan Government adopted openness through Pillar 2 of the PEAP. Consequently, Rwanda can access finance, technology and services that are necessary to improve agriculture, industry and service productivity.

The sectors that contribute to Rwanda’s exports are broadly categorized as Traditional Exports (TEs) and Non-Traditional Exports (NTEs). TEs include coffee, cotton, tea and tobacco, which were introduced to Rwanda during the early colonial period.

NTEs include recent merchandise exports that have gained prominence, such as fish, flowers, manufactured goods and services. Following the reforms, as indicated under Appendix 2.5, Rwanda’s export structure has shifted from over-dependence on TEs, so the prominence of NTEs has increased.

a) Openness and effect of Tourism to Rwanda’s Economy

Tourism is a major source of income for many developing countries, including Rwanda. In late 1990, tourism was Rwanda’s second most important export commodity for the country after coffee (Holland, Burian & Dixey 2003). However, in 1980, tourism expenditure accounted for only 1.52% of Rwanda’s exports after Amin’s overthrow in 1979 (World Bank 2000). Following the reforms, the Rwandan Government started to devise ways of reviving the tourism sector. By an Act of Parliament in 1996, the Rwanda Tourist Board (RTB) was established, to promote tourism in Rwanda. Second, in 1998, the Rwanda Wildlife Authority (RWA) was established. RWA is responsible for conserving and sustainably managing wildlife. To further promote tourism, in 2014 the Ministry of Tourism, Wildlife and Antiquities (MTWA) was created.

As a member of the WTO, Rwanda made commitments under the General Agreement on Trade in Services (GATS) to revive viable sectors such as tourism. Following the GATS Commitments, tourists can use any of the four modes of supply to take advantage of the country’s tourism potential. First, through cross-border supply (Mode-1) tourists utilize the services of sectors such as banks to book hotel accommodation via telecommunications or mail before the journey.

Second, through Consumption abroad (Mode-2), tourists move abroad to obtain a service such as recreation and medical services. Third, considering business in tourism sector, Multinational enterprises have established Commercial presence (Mode-3) in various countries by establishing subsidiaries abroad. After establishing subsidies Multinational enterprises seek the presence of experts in countries of destination (Presence of natural persons Mode 4).

In this way, tourists are attracted to countries such as Rwanda with relative ease by utilizing any of the services provided. As a result, the contribution of tourism can be identified through backward and forward linkages created by integrating Rwanda into regional and global value chains (GVCs) (Mwaura & Ssekitoleko 2012; RNCTAD 2013).

Since 2001, the GoR has been identifying several priority sectors as indicated above, and tourism was also earmarked as a possible poverty reduction tool.

Internationally, it has been recognized that “tourism is an important opportunity to diversify local economies” (Ashley, Boyd and Goodwin, 2000, p.1) and the GoR noted this in the PRSP document itself, as it emphasized a need to “develop other engines of growth and to transform [the] economy” including “encouraging the development of tourism” (Government of Rwanda, 2002, p.9). Being a small country, Rwanda has limited options as regards tourism, but “the country’s parks and natural forests are [already] a valuable commodity for tourism” (Rutagarama, 2001). In terms of the country’s capital, Kigali, the opportunities for growth exist and the benefits for expanding the industry are significant in terms of creating jobs and generating “spin-off development” (Kigali Economic Development Strategy, 2002). Through the RNIC program, OTF Group “developed a National Tourism Strategy that was adopted in 2001. The Strategy identified a long-term vision and defined several areas to be developed to promote tourism in Rwanda” (OTF Group, 2004).

A group of 40 representatives from the private and public sectors as well as NGO leaders got together and formed Rwanda’s Tourism Working Group (TWG) with a mandate to implement the National Tourism Strategy. Prior to developing and defining the Strategy, the TWG identified the main constraints “hindering the development” of the tourism sector.

These included:

• The perception that Rwanda is not a safe destination;

• The limited accommodation offered at key tourism sites;

• Limited airlift to main tourism markets;

• The restricted range of tourism experiences offered;

• Lack of tourism culture needed to increase service quality;

• Lack of reliable information on the tourism industry;

• Weak public and private sector collaboration;

• Business challenges to the development of the industry (including access to financial capital, lack of qualified human resources, etc.); and

• Regional political instability, which was the main reason international tour operators were reluctant to engage Rwanda as a destination (OTF Group, 2005).

The TWG had to take into account all these issues in the formation of the Strategy. Overall, they “articulated the following vision for Rwanda’s tourism industry: ‘Generate $100 million in tourism receipts [and 70,000 international tourists] in 2010 by focusing on creating high value and low environmental impact experiences for Eco-[tourists], Explorers and Individual Business Travellers’” (OTF Group, 2005)

Rwanda’s success in addressing these issues is evidenced by the steadily increasing number of tourists to the country. Figure 3.4. Tourist numbers in Rwanda are most often measured by visitor numbers to the country’s National Parks since data collection systems are very limited.

However, the Tourism Strategy recommends that not just tourism numbers should be targeted, but also the length of stay of foreign tourists. Often Rwanda may be treated as a short 1-2 day stop as an add-on from Kenya or Rwanda, so the tourism sector is striving to offer attractive and unique experiences that enable tourists to have a longer length of stay in the country, even as an add-on country, focusing on a variety of Rwanda’s peculiar attractions (Rwanda Development Gateway, 2005).

The Role of ORTPN Through the coordination of private and public sector investments and the development of clear action plans as well as the support of the government, the development of Rwanda’s tourism sector became a priority. As the driving force behind the tourism industry, ORTPN was in need of restructuring in order to successfully implement the new policies that were to be developed.

Funding for ORTPN had previously been attained through income accrued from national parks, game hunting and government-owned hotels which was enough to cover costs. As Plumptre et al (2001) put it, in the years before the war, “tourism was a major source of revenue for the government and for ORTPN, bringing in around US$ 1 million as gate fees and an estimated US$ 3–5 million as revenue spent in the country on food, transportation, and accommodation.” As tension rose in the country beginning in 1990 and on, tourist numbers dropped drastically. The radical loss of income caused by the war and subsequent genocide left ORTPN with a severe lack of funding as income from hotels and protected areas no longer existed. The decrease in revenue for ORTPN also left the organization unable to fully support the salaries of its staff (Plumptre, Masozera and Vedder, 2001, p.1), and this in turn, affected the support of conservation projects in the country’s parks and forest reserves. The crisis led to a privatization process of the government-owned hotels as well as seeking government subsidies and mobilizing the efforts of partners including the International Gorilla Conservation Project (IGCP), the Wildlife Conservation Society (WCS) and the Dian Fossey Gorilla Fund International (DFGFI) among others (ORTPN Strategic Plan, 2004). It was due to this change that ORTPN’s supervising body, the Ministry of Commerce (MINICOM), requested a restructuring process for the organization, which began on January 28th, 2002 (ORTPN Strategic Plan, 2004c, p.10).

It was hoped the process would enable ORTPN to re-examine its mandate and mission and allow it to improve its overall structure and services for the good of the sector.

The creation of two agencies, the Rwanda Tourism Agency and the Rwanda Wildlife Agency has enabled ORTPN’s efforts to be more concentrated in achieving its two main objectives: the development of the tourism industry as a whole and ensuring the integrity of the country’s protected areas. The existing tourism initiative is currently centred on the three national parks namely: Volcanoes National Park, Akagera National Park and Nyungwe National Park. Although other sites and attractions are being developed in Rwanda, the parks will continue to be a primary focus for some time (OTF Group, 2004, p.1).

In accordance with this, the Rwanda Wildlife Agency (RWA) focuses most of its efforts on the “protected areas” around and within the national parks. “Protected areas’” refer to the national parks and savannah and wetland areas. As previously mentioned, environmental pressures like the degradation of the country’s precious ecosystems and key habitats, reduction in water levels and erosion, do exist and could expand if not managed appropriately.

On the other hand, local populations are also in need of land, firewood and other natural resources, a big problem for the industry. While this is an issue mainly for the government, the RWA is charged with conservation and preservation efforts in such areas (ORTPN Strategic Plan, 2004c, p.12).

The purpose of the Rwanda Tourism Agency (RTA) within these same areas is “to ensure diversification, effective and sustainable management of tourism products in the protected areas” (OTF Group, 2004, p.2). The RTA has been charged with creating and implementing a development plan for tourism in each of the national parks and other identified tourism areas. The agency must ensure “diversification and effective management of the services, tourism products and infrastructure in order to promote tourist activities in each protected area” (ORTPN Strategic Plan, 2004c, p.12). These efforts require the cooperation of both the public and private sectors under the direction of ORTPN and the TWG. While most of the RTA’s efforts in ORTPN’s Strategic Plan centre on the national parks, a critical objective for the agency is to ensure diversification of the tourism product by developing and managing high quality tourism attractions outside of these distinct areas.

This initiative by ORTPN generates a focus on natural attractions outside of the national parks, including thermal springs near Lake Muhazi in the east of the country, and unique caves found in the northwest of Rwanda, as well as cultural and historical sites such as the National Museum, the Royal Residence in southern Rwanda, and Genocide memorials throughout the country (ORTPN Strategic Plan, 2004c, p.12).

a) A New Vision for Rwanda Tourism

In accordance with the objectives set out within Vision 2020, which included increasing of competitiveness particularly in the service and industry sectors, OTF Group focuses on the ‘competitive’ as opposed to the ‘comparative’ “advantage model”, a strategic base the government of Rwanda was eager to follow. The strategy developed by OTF Group and worked on with the TWG and ORTPN, “targeted specific groups of tourists” such as eco-tourists, explorers and individual business travelers, “to develop products that meet their needs rather than providing for general mass tourism” (OTF Group, 2004, p.1). These tourist segments were chosen specifically because they were viewed as having the potential to bring the most significant foreign currency to Rwanda, can maximize the benefits of tourism within the country, and also minimize the negative impacts of tourism on such a small country and its valuable natural resources (OTF Group, 2004, p.1).

Among the three tourist segments, as stakeholders in the tourism industry (including tour operators, and hotel and restaurant managers) the TWG initially chose to focus on eco-tourists and explorers as they felt they would generate the most business. Eco-tourists are those who prefer nature products and experiences centered on natural attractions and wildlife; and explorers prefer cultural products that exhibit a country’s traditions, art forms and daily life experiences. As previously indicated, part of the strategy is to create high value experiences for Rwanda’s tourists. By creating the types of experiences that encompass products that are unique to the country, Rwanda seeks to distinguish itself from its neighboring countries (OTF Group, 2004, p.1).

Based on this, two tours were created within the strategy: the Primate Discovery Tour for Eco tourists and the 500 Years of Civilization Tour for explorers. The Primate Discovery Tour is designed to enable tourists to see the unique and famed mountain gorillas and golden monkeys of the Virungas, as well as the 13 primate species residing in Nyungwe National Park (ORTPN, 2004). A Discovery Centre is also being established for the Volcanoes National Park where more detailed educational and interactive tools will be used to familiarize the tourists with the park’s landscape and inhabitants.

This is a particularly important project for the tourism industry and the TWG in particular as many families come to the country, and because children under the age of twelve are not allowed to visit the gorillas, the centre will provide a different and exciting experience for them and any other tourists not able to visit the gorillas themselves. The tour would also encompass a visit to the shores of Lake Kivu before moving to the dense rainforest of Nyungwe National Park with its 13 species of primates, including chimpanzees (OTF Group, 2004).

According to Grosspietsch’s research (2005), visitors to Rwanda feel that the country’s natural attractions are its main assets as a tourist destination. Eco-tourists are a significant market for Rwandan tourism and having such a unique tour, in the form of a “primate safari”, could put Rwanda ahead of its competitors in terms of wildlife viewing and ‘one-of-a-kind’ experiences (ORTPN, 2004a, p.1). The 500 Years of Civilization Tour enables the visitor to “learn about Rwanda’s unique people from before colonization to the genocide and beyond, by visiting Rwanda’s museums across the country”, the majority of which are currently being developed or rehabilitated. Visitors to Rwanda often express how friendly and welcoming the people of Rwanda are and have shown an interest in Rwanda’s traditional lifestyle (Grosspietsch, 2004). This tour also showcases Rwanda’s arts and culture and provides opportunities for meaningful interaction with local communities (OTF Group, 2004).

The above-mentioned tours are still being refined and constantly updated (2005), and the desire to expand the current product is the main driving force behind the development of these tours. One of the OTF Group documents indicates that although the current situation (at the end of 2004) shows that the tourism sector is “one of the primary income generating sectors of Rwanda and has provided more receipts than expected”, the length of stay of the visitor is much shorter than planned (OTF Group, 2004). It is specified that “this is primarily due to the fact that the Rwanda experience is still centred …around the mountain gorillas…Rwanda is still selling products rather than experiences” (OTF Group, 2004). The push to create a uniquely Rwandan product was the impetus behind the development of these tours and is still the main motivating factor Although the number of visitors to the parks increased by 39% and park receipts increased by 42% in 2003-2004, it became evident that “the gorillas alone cannot sustain Rwanda’s tourism growth” (ORTPN, 2004). Despite their tremendous contribution to the industry, concern still remains that international tourism remains gorilla-centred and current growth is therefore unsustainable.

This justifies ORTPN’s suggestion that Rwanda “needs to move away from Gorilla monoculture” (ORTPN, 2005) and explains why in the past few years, Rwanda’s tourism industry has been focused on providing a diverse tourism experience for all visitors when re-launching its tourism effort domestically and internationally.

In October 2003, ORTPN and the TWG held a National Tourism Launch in Kigali, designed to inform the local population of the industry’s latest efforts and give the weak domestic tourism industry a boost.

This launch attracted several VIP guests including His Excellency, the President of the Republic of Rwanda, Paul Kagame, and Cabinet ministers. As the Director of the RTA in ORTPN said, the National launch was designed to declare to the local population that after the effects of the war and genocide, “tourism is being launched again. There is a future for Rwanda and a future for…tourism” (Rwigamba, 2005). In the same year, ORTPN hired the UK-based marketing firm Southern Skies, which was linked to Indigo PR, a public relations firm, to assist the industry to reach markets it could not reach on its own (ORTPN, 2004).

The role of Southern Skies and Indigo PR was to promote Rwanda throughout Europe and arrange for a successful re-launch of Rwanda tourism to the international market at the World Travel Market (WTM) in London, England in November, 2003. The re-launch of the industry at the WTM was to be the official message to the international community that Rwanda was ready to offer a unique tourism experience to all tourists. “The move represented a strong indication of Kigali’s commitment to ensuring that tourism plays a central role in the rebuilding of Rwanda” (Southern Skies, 2003). As discussed earlier, Rwanda’s National Parks are often the most popular attractions, but as part of the industry’s diversification efforts, natural and cultural assets are being promoted equally. Cultural tourism is, for example, critical for the industry’s growth, but also for ORTPN as part of its efforts to showcase Rwanda’s diversity

More recently, at the annual International Tourism Board (ITB) held in Berlin, Germany in March 2005, Rwanda was named the fourth best exhibitor in the category of African exhibitors. At the World Travel Market in London, England, Rwanda’s delegation attracted 24 new tour operators from around the world who made agreements with local tour operators to send tourists to Rwanda (Nuwamanya, 2005). Constant research, product development, and marketing initiatives have been highlighted in ORTPN’s efforts to sustain the positive effects created by the launches in 2003 (ORTPN, 2004).

The success of both official launches and a continued presence at international fairs are priorities for the promotion of Rwanda’s tourism industry (ORTPN, 2005). .According to Southern Skies (2003), the success of all these efforts means that Rwanda will have laid the foundation for its ambitious plan to increase tourism arrivals five-fold by 2010

b) Tourism and Economy’s GDP Growth rate

Domestic tourism, particularly in developing countries, is critical for the tourism industry to thrive. With a focus on the domestic tourism market, countries are able to diversify their tourism products and appeal to a wider target audience. Information on domestic tourism in developing countries, however, is extremely scarce. In fact, as Ghimire (2001, p.2) notes, very limited knowledge on tourists in the South exists at all. Among the scanty literature that does exist, there is a consensus on the fact that one of the major benefits of domestic tourism is that while domestic tourists often spend less money per visit, they travel more often and bring greater economic growth particularly to the local communities.

Domestic tourists contribute more directly to the services offered by the local population, thereby contributing to the informal tourism sector, maintaining the strength of the industry, promoting pro-poor tourism and as a result, aiding in the poverty alleviation efforts of the country. As previously shown, a survey of literature on tourism in developing countries indicates that countries like South Africa, Nigeria, Kenya, China and India have realized the significant benefits of improving the domestic tourism market, especially with regard to the growth of this market (Ghimire and Li, 2001; Mustapha, 2001; Rao and Suresh, 2001; Kenya News Agency, 2004; Rogerson and Lisa, 2005). Undoubtedly, these efforts result in economic benefits for local communities and a more sophisticated product for international consumers.

Another benefit of the development of a domestic tourism industry is that domestic tourists are not as often deterred from travel based on political, social or economic problems in the region as are international tourists. Rao and Suresh (2001) point out that “domestic tourism is not vulnerable to bad publicity, internal security problems and poor infrastructure.” Basically, domestic tourism helps, among other things, to:

• Maintain the industry during dips in the tourist market, and essentially mitigates many negative threats to the industry. There are in fact several advantages to domestic tourism;

• Sustain demand for tourism when there are seasonal variations in international tourism;

• Conserve foreign exchange by encouraging locals to see their own country rather than travelling abroad;

• Expand investment from richer to poorer areas, also enabling local people to benefit from government investment in tourism infrastructure; and

• Protect the occupancy of accommodation and other services as international tourists decline.

Most importantly, it meets the recreational needs of the resident population, helps to create a tourism culture, and generates awareness about the natural resources of the country and its conservation (Rao and Suresh, 2001). In spite of these positive aspects of domestic tourism, several constraints get in the way of its growth and development, including a strong bias in favour of international tourism that still remains. The benefits of domestic tourism in many African countries are beginning to be recognized, thus encouraging the development of this segment of the tourism industry which occupies such a significant percentage of the GDP of many countries.

Rwanda is one such African country. During the years of civil war that culminated in the 1994 genocide, Rwanda’s GDP experienced a depression, “posting a dramatic decline at more than 40% in 1994, the year of the genocide” (U.S. Department of State, 2005). In 1995, the first postwar year, the GDP saw an increase of 9%, signaling the “resurgence of economic activity, due primarily to massive foreign aid” (U.S. Department of State, 2005). Humanitarian aid brought into Rwanda to support reconstruction and development assistance sustained the country during a devastating period in the economy. Since then, the Rwandan economy has been “one of the fastest growing in Africa and indeed the world.” From 1998-2002, real annual GDP growth averaged 7.7%, slowing down to an estimated 3.5% in 2003 (Institute for Security Studies, 2005; I Explorer, 2005).

After the war, the new Rwanda government, formed in 1994, restored security throughout the country, returned most subsistence farmers to their fields, and directed their focus towards “poverty reduction, infrastructure development, privatization of government-owned assets, expansion of the export base, and liberalization of trade” (U.S. Department of State, 2005; Institute for Security Studies, 2005). Poverty does still exist, and the main economic challenge facing the government is “to stimulate new sources of poverty-reducing growth” (Institute for Security Studies, 2005), which includes the development of manufacturing and service industries, tourism among them. The government set out to achieve these goals towards the late 1990s and into the new millennium.

As has been echoed by many authors, for countries with “a limited industrial sector, few natural resources and a dependence on international aid, tourism may represent the only realistic means of earning much needed foreign exchange, creating employment and attracting overseas investment” (Sharpley, 2002). In the particular case Rwanda, which had to recover from war and devastation, as an industry, “tourism can generate new sources of income without significant new investments” (Christie and Crompton, 2001, p.15) and has several economic advantages, not just for the industry itself but acts as an entry point to several other industries. It creates jobs in the construction, transport, telecommunications and financial fields and also creates links to the informal sector (Christie and Crompton, 2001). In increasing the role of service industries in the country, the tourism industry was considered to have “far greater potential given the current stability, travel infrastructure, and available animal parks as well as other potential tourist sites” (U.S. Department of State, 2005).

As one of the largest industries in the world, tourism is continually growing and tourism receipts are “of critical importance to many countries’ balances of payments and general economic welfare” (Natsios, 2004). As noted in the previous chapter, tourism has made a significant contribution to Rwanda’s economy, and currently (2006) ranks third in terms of the country’s export industries along with tea, after mining and coffee (OTF Group, 2006), and according to the Rwanda Investment and Export Promotion Agency (RIEPA), around US$ 89 million worth of investments of new tourism products was added from 2001-2005 (Kacou, 2005). The industry has revived itself over the past few years and owes its success mainly to international tourists visiting the national parks, and more recently, business travelers that are coming to the country for the many conferences and events the country has held in its three main hotels. These have included major international, regional and local conferences ranging from:

• The NEPAD Summit with eleven heads of state and governments held in February, 2004; • The Conference to Prevent and Banish Genocide through Active Universal Solidarity in April, 2004;

• The Festival Pan-African de la Danse (FESPAD), the pan-African dance festival in August 2004;

• The Society of Women and Aids in Africa International Conference in May, 2005; and

• The 10th Summit of the COMESA Authority Heads of State in June, 2005 which was attended by over 10 heads of state and government and over 1,200 delegates.

Rwanda’s tourism receipts reached around US$ 26 million by the end of 2005, up significantly from US$ 17 million in 2004. Park visitors and business travellers, account for these receipts, making up 61% and 39% of recorded tourism receipts respectively, with park visitors spending an average of US$ 268 per day and business travellers spending an average of US$ 150 per day (OTF Group, 2005e). The majority of money earned from park visitors is due to park entrance fees. In July, 2004 ORTPN increased the price of gorilla permits for international visitors from US$ 250 –US$ 375, compared to the price for local tourists which is currently at FRW 10,000, the equivalent of about US$ 18. In order to make a tourism experience more successful, countries often charge more entry fees from tourists from developed countries, increasing revenue creation substantially (Kruger, 2003).

In Rwanda’s case in particular, the higher prices at the Volcanoes National Park not only increase tourism revenues, but also act “as a necessary means to try to control the high pressures put upon the gorillas and the park authorities” since the gorillas have a limit in terms of the number of people who can visit them per day (Williamson, 2001). The government policy on the gorilla permits therefore renders the contribution of local tourists quite minimal. However, the proper development of domestic tourism in Rwanda would not centre on encouraging locals to spend more money on the national parks. Creating a sustainable domestic tourism sector would mean generating tourism products that would appeal to the national population and encourage spending on new as well as existing experiences. Currently, Rwanda’s tourism industry is heavily dependent on ecotourists and individual business travellers, as mentioned earlier. It might be important at this stage of Rwanda’s tourism development to minimize the threat of over-reliance on any tourism segment.

Rwanda’s tourism stakeholders have already set some plans in motion to generate an interest in tourism. While many of these initiatives have begun, however, proper marketing and promotion programs (by sources other than the national tourism board, ORTPN) geared towards Rwandans is minimal and has therefore hindered the success of these initiatives.

The tourism industry has managed to make an impact on nations not only economically, but socially as well. In Africa, tourism is not only seen as a means to strengthen existing economies, but also as a means to rebuild them. This rebuilding process means that the people themselves must make a significant contribution to progress, and their involvement in the tourism industry is one way of accomplishing this.

In Eritrea, for example, “the tourism sector was considered to be a strategic sector by the government and one that would play a critical role in the nation’s recovery program” (Burns, 2000, p.107). For thirty years, Eritrea has been dealing with the effects of a devastating war that has made it one of the poorest countries in the world. Efforts made towards reconstruction are daunting and for most of the population that live a subsistence lifestyle, aid imports have been supporting them thus far (Tzehaie, 2005).

Eritrea’s reconstruction and national development efforts included the development of the tourism industry which has made a contribution thus far, but “remittances from overseas Eritreans, coupled with local tourism by those who return for visits, are Eritrea's main source of foreign exchange” (Tzehaie, 2005). Members of Eritrea’s Diaspora have participated in and encouraged local tourism, contributing significantly to economic development and national pride. This is a very similar situation to that of Rwanda.

The people of Rwanda must take ownership of their tourism industry. This sentiment was echoed by survey participants and most believe that a significant and positive difference will be felt in the tourism industry once Rwandans begin to participate in it. All participants recognized the difficulty in the development and growth of domestic tourism which lies in the fact that most Rwandans cannot afford to travel and explore their country. Much like Eritrea, however, the contribution to the progress of the domestic tourism industry does not lie solely in travel by the population, but in the process of expanding and rebuilding the tourism industry. The participation of all Rwandans in tourism through travel as well as employment and participation in education goes far beyond economic gains. Eric Kacou (2005) of OTF Group specifies that in Rwanda, domestic tourism is about more than just travel, but “it is about pride”. The leadership of the country has worked hard to rebuild the nation as “one people, one nation, and one culture”, and tourism has a big role to play in encouraging Rwandans to see their country as their own, as a country they have worked to rebuild and they can now enjoy (Kacou, 2005).

In a nation recovering from the ravages of war and genocide, domestic tourism can also play a significant part in the healing process leading to national unity and reconciliation. Participants in the survey all see the promotion of Rwanda’s cultural tourism efforts as the best angle with which to attract Rwandan domestic tourists. Kayihura (2005) views the promotion of cultural tourism as critical because the new generation of Rwandans need to know and understand their culture.” As well as learning about their own culture, the exposure of Rwandans to other cultures can only be beneficial. While “not all values that tourists bring are good, [most of the time] bonds are created, friendships are made”, and it is an opportunity for locals to get a new view of the world, making them “look forward to something better and exciting in their lives” (Kacou, 2005). As part of developing tourism awareness in Rwanda, members of the industry must explore domestic tourism if this awareness is to be meaningful. The more the participation increases, the greater chances for the industry’s sustainability.

2.8 Challenges to reforms in Post genocide

Rwanda gained independence following tragic circumstances. Although Rwandans did not have to wage war to win independence, the years leading to self-rule were marred with conflict. In line with the history of colonialism and anti-colonial struggles in other parts of Africa and elsewhere, Rwandans were expected to seek to free themselves from domination by their colonial masters. Rwanda’s struggle for independence, however, was two pronged. On the one hand, nationalists among Hutu and Tutsi elites were united in seeking to end Belgian rule. On the other, Hutu nationalists sought to ‘liberate’ the majority Hutu population not only from Belgian rule, but also, with the support of the colonial administration, from what they saw as colonialism by the Tutsi minority.

It is important to emphasise that inter-ethnic conflict in Rwanda was concentrated in elite circles on both sides, as it was amongst elites that the struggle for ethnic supremacy was located. Ordinary Hutu and Tutsi featured largely as either innocent victims or attackers mobilized and encouraged by elites seeking to acquire or monopolise power and the privileges that went with holding it.

Although it is not as widely analyzed as inter-ethnic conflict, intra-ethnic rivalry within the wider Hutu population also had its epicentre within elite circles divided along north-south axes (Strauss 2006; Jefremovas 2002; Munyarugerero, 2003).

For most of the colonial period the Tutsi were favoured by the Belgians and had been designated by colonial-era historians and anthropologists as foreign elements who, prior to colonial rule, had migrated into the area and subjugated the indigenous Bantu Hutu and Twa. Modern historians have demonstrated that the history of the area was actually much more complex than this (Vansina, 2004; Chretien, 2003). During most of the pre-second world war period, however, the Belgians administered Rwanda in line with their version of its political and social history.

One of the earliest measures they took in their state-building project was to exclude the Hutu, whom they judged to be less intelligent than the Tutsi and incapable of exercising leadership, from chieftainship. They also placed Tutsi chiefs in areas where, prior to colonial rule, Hutu chiefs had been in charge. Chiefs became the immediate face of colonial rule and were responsible for collecting (punitive) taxes and fees, enforcing the building up of food reserves against famine, the cultivation and marketing of cash crops, environmental conservation, organizing unpaid community work and other tasks the population found onerous, and meting out punishment to those who did not comply. Whilst in executing their tasks they acted as representatives of the state and faced demotion or dismissal if they failed, the Hutu population who bore the brunt of the exactions alongside poor, ordinary Tutsi (Semujanga, 2003), saw Tutsi chiefs through ethnic lenses and therefore as representatives of the Tutsi community rather than those of the colonial administration (Jefremovas, 2002). As Strauss (2006) argues with regard to the targeting of Tutsi civilians during the successive episodes of communal violence, “all Tutsis stood in for the actions of a few” (p. 199). Consequently the chiefs – and Tutsis generally – came to represent tyranny and, more so than the colonial administration they served, became objects of popular resentment.

The Tutsi monopoly over administrative posts ensured disproportionate access to education and training opportunities, as public and church schools and training institutions enlisted mostly Tutsi students (Baranyizigiye, 1999). After completingtheir studies, Tutsis had greater access to employment opportunities than their Hutu counterparts because of their connections in the administration. This further sharpened inequality and the sense of grievance among Hutu (Nkundabagenzi, 1961).

Colonial rule therefore created a wide political and socio-economic chasm between Tutsi and Hutu elites. Ordinary Hutu and Tutsi, however, led similar lives amidst poverty, deprivation, and landlessness (Jefremovas, 2002; also Semujanga, 2003). Hutu elites, however, seized upon the theory propounded by the colonial authorities, the church2 and ethnologists that portrayed the Tutsi as foreign invaders who had to be uprooted. The outcome was a series of bloody political crises that began in the lead up to independence and came to characterize post-colonial Rwanda during the 1960s and early 1970s, and culminated in the 1994 genocide.

Rather than take measures to dampen the inter-ethnic animosity that had developed largely as a result of their divisive policies, the Belgian authorities, by acts of omission and commission, deepened the divisions.

During the final years of colonial rule, due partly to tensions between Tutsi elites and the colonial administration as well as the Catholic Church, they embarked on promoting the Hutu to take over the post-colonial state by curtailing Tutsi dominance and influence (Strauss, 2006; Semujanga, 2003). By the mid-1950s Tutsi hegemony was on the decrease and Hutu supremacy on the rise. By the late 1950s Tutsi hegemony had effectively ended (Lugan, 1997; Strauss 2006). Nonetheless, the decline of their dominance belied their determination to continue playing an important role in the country’s evolution towards independence as demonstrated above, the Belgian colonial administration and the Catholic Church bear much responsibility for creating divisions within Rwandese society. The divisions created, nurtured and deepened inter-ethnic animosity between Hutu and Tutsi elites seeking to acquire or monopolise political power and its attendant privileges (Gatwa, 2005; Linden, 1999). Taking advantage of Belgian support in the years immediately leading up to independence and the numerical strength of their constituency, Hutu elites in and out of government orchestrated episodes of violence that, from 1959 to the early 1970s, led to the death of many members of the Tutsi and Hutu communities and forced hundreds of thousands of Tutsi into exile in neighbouring countries and beyond. State-instigated murder of Tutsi, the exiling of many others and the Rwanda governments’ policy of non-return created a sense of grievance against the Rwandese state and successive governments, which was exploited by Tutsi elites seeking to return to Rwanda through any means, including war.

Tutsi Exile and the politics of persecution and exclusion. The forcible departure of Tutsi from Rwanda started with the violence of November 1959. While the violence followed a rumoured attack on, and murder by, Tutsi militants of the UNAR of a prominent Hutu politician, it did not, at least in its initial stages assume a decidedly ethnic pattern. For example, as already pointed out, Tutsi and Hutu supporters of the UNAR and the monarchy combined forces and fought Hutu and Tutsi members of other parties and those believed to be disloyal to the king. Nonetheless, as it spread, the violence increasingly took on an ethnic complexion when groups of Hutu youths hunted down Tutsi local officials and chiefs in what one colonial official called “a surgical operation” deemed necessary to facilitate passage from feudalism to democracy (Lugan, 1997). Hutu youths were mobilized through the portrayal of Tutsi as foreigners who sought to dominate the Hutu population whose security lay in killing or expelling them from the country. Thousands of Tutsi sought refuge in neighbouring countries sharing common borders with Rwanda. This is important because of the critical role the porosity of the borders eventually played in facilitating the RPF’s transnational recruitment of fighters and invasion of Rwanda.

Far from settling down to a quiet life in exile, many refugees embarked on political and military organization within the borders of their host countries in preparation for forcible return (Munyarugerero, 2003). In 1963 groups based in Burundi and Rwanda staged an armed incursion. The Kayibanda government reacted with a campaign to eliminate all remaining Tutsi, whom it accused of conspiring with the exiles. The invasion also handed the government a chance to eliminate what remained of Tutsi political activity and the structures through which it was conducted. RADER and UNAR members of the government and the national assembly were executed, the two parties destroyed, and the Tutsi completely excluded from participation in public life. Other members of the two parties died in prison (La Communauté Rwandaise de France, 1990). In what the government termed “uncontrollable mass reaction to Tutsi provocation”, thousands of Tutsi were hunted down and killed by specially mobilized Hutu militants overseen by local officials and government ministers (Willame, 1995; Strauss, 2006; Munyarugerero, 2003). The killings led to a new exodus of between 200,000 and 300,000 Tutsi (Lugan, 1997: 436). In 1966 Burundi-based exiles staged another armed incursion, leading to further massacres. Tutsi men, women, and children were rounded up and executed, and yet more left the country. The massacre of Tutsi remaining in the country in response to armed incursions eventually persuaded the exiles to cease the armed struggle in 1967 and settle into life in exile (Semujanga, 2003).

Nonetheless, the decision by exiles to cease insurgent activities did not stop attacks on those Tutsi who had stayed behind. In February 1973 a new round of persecution commenced. First, Tutsi were purged from educational, administrative and other public institutions, as well as from the private sector on the grounds that their numerical dominance surpassed the share warranted by their small proportion of the population (Munyarugerero, 2003; Semujanga, 2003). The purges were then followed by physical attacks culminating in a new round of killings. Past massacres had been blamed on provocation by exiles, but this time there had been no incursion. It is, however, important to note that the Kayibanda regime was having to contend with growing unpopularity, rivalries within the ruling party, conflicts ensuing from regional favoritism, and a restive military dominated by Northern Hutu who felt marginalized.Taking advantage of the anxiety created within the Rwandese Hutu population as a result of the 1972 civil war in Burundi in which a Tutsi-dominated military had massacred thousands of Hutu elites, the Kayibanda government instigated the 1973 purge and killings in the hope that it would catalyze a healing of intra-Hutu divisions and a closing of ranks among Hutu (Strauss, 2006: 189; Munyarugerero, 2003.

The well-planned attacks, which started in educational establishments and spread to the National University in Butare before extending to other employment sectors, sought to establish Hutu dominance throughout the social, economic and political spheres. Government officials claimed that the attacks were aimed at “ethnic rebalancing” in reaction to Tutsi having surpassed the ten to twenty percent quota they had been allocated in various aspects of public life (Vidal 1991; Strauss 2006: 189- 190). More survivors sought refuge outside Rwanda, and this mass exodus contained the seeds of future destabilization and war.

Much has been written about the experiences of Rwandese exiles in their countries of refuge (Waugh, 2004; Otunnu 2000b; Munyarugerero, 2003; Prunier, 1997; Gachuruzi 2000; Mamdani, 2001). While many gradually adjusted to life in exile and eventually became socially integrated, others were unable either fully to integrate because of deliberate exclusion or marginalization, or did not wish to settle down permanently in their host societies because of persecution (Gachuruzi, 2000; Otunnu, 2000b; Semujanga 2003). Failure to integrate nurtured a desire to return to Rwanda to reclaim their citizenship. However, neither the Habyarimana regime nor that of Kayibanda before it was prepared to countenance their return.

Both used reasons such as population density, poor soil productivity, poverty and environmental degradation to justify their reluctance to allow the refugees back into the country. Also, in addition to the Tutsi there were Hutu who had fled from political persecution and they, too, were not wanted back into the country. This inability of refugees either to integrate abroad or return to Rwanda created a large reservoir of hostility towards the Rwandese state. The regime’s withholding of their right of return meant that they had only one option, which was to force their way back. This therefore rendered them potential insurgents and, consequently, a threat to the country’s security.

The Impact of the 1990s Economic Crisis It is widely acknowledged that although the Habyarimana regime was authoritarian, it was also development-oriented with a good track record of economic management during its first decade in power (Reyntjens, 1994; Lugan 1997; Prunier 1997; Uvin, 1997; Adelman, 2000). Measured in terms of Gross National Product (GNP) per capita, and considering the country’s intrinsic disadvantages (landlocked, highly populated, natural-resource-poor) and the relatively poor performance of its neighbours, Rwanda under Habyarimana registered considerable progress. At independence in 1962, only two countries in the world had less income per capita than Rwanda. However, by 1987 there were eighteen of them. With an average income per capita of USD300, Rwanda could be said to have been at the same level with China whose average income per capita was USD310 (Prunier, 1997).

In terms of infrastructural development, Rwanda had one of the best road networks in Africa, reliable posts and telecommunications services, with piped water and electricity covering many parts of the country. General economic management standards were also higher than in much of the rest of Africa, which explains why up until the mid-1980s the country owed modest levels of external debt compared to other countries. For example, in 1987 Rwanda’s external debt was only 28 percent of GNP, one of the lowest in Africa at the time (Prunier 1997). Nonetheless, its dependence on foreign aid had grown substantially by the end of the 1980s. According to the OECD, foreign aid – which in 1973 stood at less than 5 percent of GNP – had grown to 11 percent in 1986 and to 22 percent in 1991 (Prunier, 1997). The country, however, started experiencing economic problems during the late 1980s, especially after the collapse of the price of coffee, the country’s main source of revenue, in 1988.

The economic crisis prompted the World Bank, supported by the donor community, to demand that Rwanda implement a structural adjustment programme (SAP) to address it. The economic crisis and the difficulties associated with implementing a structural adjustment programme coincided with increased internal opposition and growing unrest stemming from years of the MRND’s political monopoly, high levels of corruption, regionalism, and political repression. Growing political instability prompted the country’s main donors such as Canada and France to link continued aid to democratization consisting of, among other things, opening up to multi-party politics.

The coincidence of economic crisis with growing internal dissent and opposition and the imposition of economic and political conditionality created a mood of panic within ruling circles, increased the power and influence of Hutu extremists within the ruling party, and pushed the regime to adopt unorthodox methods to protect its hold on power. As government-allied gangs harassed the opposition, the latter took to forming their own youth brigades to fight back (Munyarugerero, 2003). Therefore the intensification of political repression by the state served only to heighten instability and render the country ripe for insurgency (Lugan, 1997; Callamard, 2000; Adelman, 2000).

External Factors in Rwandan Economic Reform challenges in the post Genocide

In addition to the internal factors that laid the ground for political violence and armed conflict, there were external ones, which, rather than working towards averting it, guaranteed Rwanda’s descent into war. These factors included the post-cold war change in attitude towards, and in a sense retreat from, Africa by the world’s great powers, the commitment and friendship some countries felt towards Rwanda and its rulers at the time, Rwanda’s war-infested neighborhood, the refugee experiences of Tutsi exiles, and the RPA invasion itself.

a) Influence of the End of the Cold War

After decades of super-power rivalry ended with the collapse of communism, interest by the world’s great powers in African affairs and events in Africa diminished. With the need to outmaneuver each other in Africa removed, the former cold warriors focused their attention on domestic affairs and to goings-on in regions closer to home, such as the Balkans, or areas of more immediate strategic importance than Africa. In the case of tiny and obscure Rwanda and other countries ruled by dictators, this came as a blessing but also a source of risk.

It was a blessing for repressive governments, allowing them to commit crimes against their own citizens without attracting much attention or publicity. The risk, however, lay in the fact that this repression invited violent reaction from internal opponents convinced that violence was the only means through which such repression could be ended, as indeed was the case in Rwanda and previously and subsequently elsewhere on the continent (Tull and Mehler 2005). Inattention to the brewing political crisis in 1980s Rwanda by the great powers also partially accounts for the ability by the Museveni government in Uganda to facilitate, by commission or omission, the activities of the Rwanda Patriotic Front as it prepared for and prosecuted the war against the Habyarimana government. (Waugh, 2004; Munyarugerero, 2003; Callamard, 2000).

b) Rwanda’s and Habyarimana’s International Friends

Ironically, alongside the diminished interest in Africa by the great powers ran a high degree of support for and admiration of Rwanda and its President by some international actors. Among these were governments and their leaders, and members of their families. Here France and Zaïre and Presidents Mobutu Sese Seko and François Mitterand are the best examples. There were also members of the international aid community working for bilateral and multi-lateral organizations (Uvin, 1997).

Despite linking continued aid to democratization in the early 1990s, France had long been an uncritical ally of Rwanda, a situation attributed to “personal and patrimonial ties” and “personal contacts” between French and African governments as well as among their high-level officials. These, an analyst has argued, were underlain by a “poverty of institutional mechanisms” as captured in the “passivity” of the French Parliament (Callamard, 2000; also Wallis, 2006). It is the nature of this relationship that accounts for the lack of concern, even after the RPF had invaded, with “exactions committed by the Habyarimana regime”, the “absence of French diplomatic interventions against human rights violations committed by the Rwandese regime”, and eventually to the stepping up of French military assistance to prop up the regime (Callamard, 2000: 169). Mobutu’s Zaïre provided the Habyarimana regime not only with moral support, but also signed a pact with the regime providing for “common security services, the sharing of security information, military co-operation, and interdiction of opposition movements on each other’s territory” (Gachuruzi, 2000).

It was on the basis of these arrangements that, despite general repression and human rights violations, Zaïre provided the Habyarimana regime with military backing in the form of troops after the RPF invasion (Gachuruzi, 2000; Munyarugerero, 2003).

Meanwhile international aid workers continued naïvely to portray Rwanda as a showcase of success even after it became clear that this was no longer the case. A good illustration of this comes from Adelman (2000) according to whom: Canadian development experts involved and committed to Rwanda had no sense of popular unrest even in the late eighties. For them, the anti-Rwandese propaganda efforts were considered to be the product of Tutsis who had been forced out of Rwanda over twenty years ago. To the same experts, Adelman continues, President Habyarimana “remained the knight of purity for the vast majority of Rwandans, a man dedicated to the well-being of his people who could do little wrong in the eyes of those he ruled”. On the basis of these sorts of assessments Rwanda continued to receive large amounts of aid, which must have served as encouragement to the Habyarimana regime not to change its behaviour towards the internal opposition, let alone the Tutsi population or Tutsi exiles pressing for their right to return to their country of birth. In this way, the country was helped on its inexorable descent into political unrest, armed conflict, and eventually genocide.

c) Rwanda’s War-torn Neighborhood

After the Rwandese Alliance for National Unity (RANU) transformed itself into the Rwanda Patriotic Front during the late 1980s to facilitate the return of exiles by force, recruitment of cadres and fighters started in earnest. In Uganda the Museveni-led National Resistance Army (NRA) insurgency had provided thousands of refugees with the necessary military training and combat experience. After Museveni seized power, many others enlisted in the Ugandan army and served in various capacities, some in senior and sensitive positions in preparation for eventual invasion of Rwanda. Prunier (1997) and Otunnu (2000b) have so far written the most comprehensive accounts of the RPF’s infiltration of the Ugandan military in preparation for the insurgency they intended to launch against the Rwanda government.

While the Uganda government has denied complicity in the RPF’s plans, albeit with individual officials giving oblique indications of its possible role, circumstantial evidence points to far-reaching involvement. Elsewhere, in Zaïre, Burundi, Tanzania and Rwanda itself, according to interviews with participants in the insurgency,38 the RPF exploited a combination of weak border controls, limited vigilance by national security agencies, and political turmoil in some of these countries to recruit and ferret out combatants from local branches of the large Rwandese, especially Tutsi, diaspora for training. Outside the immediate neighbourhood of the Great Lakes region, from as far afield as southern Africa, the Americas and Europe, the Movement sourced volunteer fighters and financial resources.

Meanwhile by the early 1990s, because of the Habyarimana regime’s politics of ethnic and regional exclusion, the political situation in Rwanda and the deteriorating economic environment, the government was losing control as internal opposition grew in intensity (Chretien, 2003; Munyarugerero (2003) sums up the situation as it was in early 1992:

By April 1992 … the government was trapped. Within a year it had to negotiate a peace settlement, guarantee internal peace, sort out the country’s administration, implement the structural adjustment programme imposed from the beginning of 1991 by the Bretton Woods Institutions, organize a debate about the national conference and when the chance presented itself, find a solution to the refugee problem, and finally, organize general elections. For a government used to working according to its own internal dictatorial and self-assured logic, the combination of these simultaneous pressures and forces it could neither resist effectively, nor control, amounted to a recipe for breakdown.

Already, as Munyarugerero (2003) points out, the army was experiencing mutinies and taking out its frustration on members of the public in reaction to the humiliating losses it was suffering on the battle front against the Rwanda Patriotic Army (RPA), the RPF’s military wing. To make matters worse, after the formation of the transitional government bringing together the MRND and internal opposition parties, the army became divided along lines reflecting the political complexion of the new government. This served only to weaken further the state’s capacity for self-preservation. The RPA invasion does in a way fit with the notion of the 'contagion' effect of war in Africa, the so-called ‘war next door’ syndrome, whereby armed conflicts occur in countries or sub-regions that have had previous conflict, and where war in one country usually ignites or fuels war in one or several of its neighbours where conditions favourable to armed conflict already exist. The transformation of the National Resistance Army insurgency in Uganda into the Rwanda Patriotic Front’s invasion of Rwanda testifies to the transmissibility of war from one country to another or others (De Waal, 2000). In this case the situation in Rwanda as captured broadly by the two Hutu/Tutsi and northern Hutu/southern Hutu polarizations was ripe for the transmission of this particular war (De Waal, 2000).

In Rwanda the government’s portrayal of the invasion simply as a decision by the Tutsi refugees to return by force and reverse the gains of the 1959 revolution, ignored the internal and external conditions that made it possible and lent credibility to the RPF’s claim that it was motivated by wider political objectives. The RPF presented itself as a multi-ethnic movement and as an alternative to the regime in Kigali, which it accused of corruption, nepotism and violation of human rights. In the context of growing popular disaffection with the regime inside Rwanda itself, the RPF’s actions, ethnic considerations aside, struck a chord with disenchanted members of the general public. It is probable that if Hutu governing elites in their desperation to hold on to power had not stirred anti-Tutsi sentiments and mobilized the Hutu population to kill members of the Tutsi community, the RPF invasion would have been welcomed by many Tutsi and Hutu alike; after all, both had suffered as a result of ethnic and regional exclusion. (Prunier, 1997; Otunnu, 2000).

d) Building credibility and legitimacy

The RPF’s early struggle to establish credibility was not confined to convincing members of the international (especially donor) community of its bona fides as an outfit capable of stabilizing and running the country. Within Rwanda itself, it had to convince the hostile and distrusting majority who had been fed on propaganda suggesting that the Tutsis sought to establish a system they would use to oppress the Hutus. Nonetheless, the new government moved quickly to assuage the fears of this majority through deliberate policies and actions. In addition to its decision – its overwhelming military and political might notwithstanding – to institute a government of national unity in which other parties were included, it quickly embarked on re-integrating elements of the defeated, almost exclusively Hutu army into the new national army, some in senior ranks. While these measures did not immediately change the overall complexion of the new army whose predecessor, the Rwanda Patriotic Army had started life as a predominantly Tutsi outfit, they signaled the RPF’s intention to build an inclusive military. Despite the government’s clear demonstration of a willingness (if not determination) to build a non-exclusionary military, critics carried on pointing at the preponderance of Tutsis within its ranks, especially at officer corps level (Zorbas, 2007).

With time, however, deliberate steps were taken to open up the military to all who qualify without restriction. Sources within its senior ranks suggest that increasingly the lower ranks contain more Hutu than Tutsi.58 Putting this particular question aside, according to other sources 59, the changing complexion of the military is as much the outcome of young Tutsi preferring not to join or to leave the army for opportunities elsewhere, as much as it is of a deliberate policy by the government and the military to increase Hutu representation. This, according to sources within the military has, alongside other developments, not been without psychological impact within the communities where the re-integrated and newly-recruited servicemen and women come from. A major outcome has been to assure doubters that the new government was not after instituting a Tutsi ethnocracy and, in the process, to build its internal credibility (Reyntjens, 2004, 2008).

An important source of internal credibility for the government is its uncompromising anti-corruption crusade. While at the beginning of its tenure there was a tendency for RPF cadres to seek to accumulate wealth by virtue of their positions, recent years have seen a heightened intolerance toward malfeasance and abuse of position and public assets and resources. In Rwanda the slogan ‘zero tolerance of corruption’ carries practical meaning and is not merely propaganda calculated to pay lip service to donor concerns. Accusations of corruption, some focusing on senior RPF cadres and government officials count among the highest numbers of cases in the courts besides genocide-related charges. Where guilt cannot be proven in courts of law, administrative measures, which are not entirely uncontroversial even within circles of RPF cadres, are deployed to deal with those accused of abuse of public funds, conflict of interest, nepotism, and soliciting bribes. This is in complete contrast with the Second Republic, for example, when close kin and friends to President Habyarimana and his spouse, who together with the presidential couple formed an inner circle known as akazu, were virtually above the law. In a sense the RPF government has used its anti-corruption drive as an important legitimating tool for its leadership, which is seen as generally fair-minded and clean. Having succeeded a regime that many saw as divisive, corrupt and despotic, it became imperative for the RPF government to demonstrate that it was different. (Adelman and Suhrke, 2000).

2.9 Rwanda’s Economic Growth after adopting Reforms 1994–2017

In the aftermath of the reforms introduced soon after the overthrow of Habyarimana’s regime in 1994, Rwanda started to experience positive economic trends. This section examines trends in economic growth a rising from positive investment policies, Exports, Imports, and FDI.

a). Analysis of the Trends in Rwanda’s Economic Growth, 1994–2017

According to data provided by National institute of statistics, ( Annual report, 2016),since adopting economic reforms in the early 1990s-the post Genocide, Rwanda kick started a new era of economic upsurge to experience enormous economic growth. Available data shows that GDP at 2000 constant market price increased from USD 1,903.76 million in 1994 to USD 9,914.33 million in 2011. Within the same time frame, stock of its capital augmented significantly from USD 8,933 million to USD 22,299 million, demonstrating capital accumulation of USD 225,899.29 million, with growth rate of 4.33% per annum. Similarly, annual headline inflation rates reduced from 88.21% to 2.93% around the same time frame, demonstrating that Rwanda was on the path to macroeconomic stability, although inflation as symbolized by CPI reduced from 88% in 1994 to 22.09 in 2011, unfortunately, by 2016 had increased again to 119.21% per annum correspondingly.

b) Analysis of the Rwandan Macroeconomic performance trends

Real GDP was estimated to grow at 8.7% in 2019, higher than the regional average. Growth was mainly in services (7.6%) and industry (18.1%), particularly construction (30%). Investment drove growth, led by public investment in basic services and infrastructure. Real GDP per capita increased 6.1% in 2019. Inflation moved up slightly to 1.6% in 2019, driven by increased domestic demand. Since inflation was below the 5% target, the National Bank of Rwanda reduced the monetary policy rate by 50 basis points to 5% in May 2019, stimulating bank lending to the private sector. Domestic credit to the private sector increased by 0.9 percentage point to 21.1% of GDP in 2019.

Despite strong tax revenue growth of 11.5%, similar public investment growth led to a 1.9 percentage point increase in the fiscal deficit to 6.2% of GDP in 2019. Government securities largely financed the deficit. Public and publicly guaranteed debt increased to 50.3% of GDP in 2019, though Rwanda is assessed at low risk of debt distress. Imports grew faster than exports as traditional exports slowed. The trade deficit widened by 3.5 percentage points to 11.3% of GDP in 2019, increasing the current account deficit by 1.5 percentage point to 9.2% of GDP. External reserves increased by 8% to $1.4 billion in 2019, equal to 4.7 months of imports. The exchange rate depreciated against the dollar by 5.0% in 2019 due to the growing trade deficit.

b) Tailwinds and headwinds

Rwanda’s growth is projected at 8.0% in 2020 and 8.2% in 2021, supported by continuing large-scale investments such as the Bugesera airport, Hakan Peat plant, and electricity infrastructure. Inflation is projected to remain around the 5% target. As fiscal policy trades off between supporting demand and ensuring public debt sustainability, the fiscal deficit is projected to increase to 6.8% of GDP in 2020 and 6.6% in 2021. The current account balance is projected to narrow to 9.1% of GDP in 2020 and 8.0% in 2021 due to a pickup in traditional exports. Rwanda’s rapid growth, coupled with a focus on the business environment, can stimulate growth in private investment, currently low at 13% of GDP compared with the East African average of 16%. Foreign direct investment averages 3% of GDP, compared with the low income country average of 3.3%. The 2020 World Bank Doing Business report ranks Rwanda second in Africa.

In January 2019, the National Bank of Rwanda adopted an interest-based monetary policy framework. By June, money market interest rates (5.45%) started converging around the bank’s 5.0% rate, followed by a drop in the lending rate. This can foster private lending for investment, creating new jobs and spurring growth. Rwanda’s fiscal space to finance development narrowed recently with a steep decline in aid from 10% of GDP in 2010 to 4.9% in 2018. Despite the country’s vision and bold strategy for economic transformation, the huge amounts needed for future growth will require blended financing to de-risk and crowd in private capital. The high costs of transport and energy, due to Rwanda’s landlocked position and poor logistics system, constrain its ability to attract investments and keep its private sector from expanding in job-intensive industries. Energy was estimated to cost 22.2% more than the regional average in 2016.

Despite high GDP growth, Rwanda’s transformation has been slow. GDP per person employed was $3,863 in 2011 purchasing power parity dollars in 2018, compared with $13,387 in Africa. Rwanda’s low labor productivity results from only 4% of the labor force working in manufacturing, while two-thirds is still in low-productivity agriculture. Up to 18.2% of youth ages 16–30 were unemployed in May 2019 due to a lack of jobs or appropriate skills. Rwanda has made some progress in closing the skill gap and developing its human capital.

Figure11: Rwanda’s transformation GDP growth

Figures are not included in the reading sample.

2.10 Conclusion

In 1961, the Monarchy was abolished and Rwanda became a Republic, gaining Independence from Belgium in 1962, with Parmehutu leader Grégoire Kayibanda. By that time, Africa was by then still referred to as a dark continent that needed Christianity, economic development and civilization. The Berlin Conference instructed colonial masters to take political charge of the colonies under their influence, and to economically develop them. France started overseeing development of Rwanda, which had an agrarian economy. To economically develop Rwanda, a dual economic system was adopted. In 1994, after the overthrow of Habyarimana’s regime, the government that took over started to rebuild the economy with initiatives. The initiatives were supported by donor agencies such as the IMF and the World Bank with their economic package encouraged the government to adopt economic reforms as preconditions for Donor support. First, the economic package was wide-ranging, comprising of macroeconomic reforms to stabilize inflation, which had skyrocketed to 216% per annum in 1994-95. Second, donors encouraged the government to adopt sound fiscal policy reforms.

Fiscal policy measures were intended to enable the government to rebuild the dilapidated physical infrastructure and provide social services. Third, the government adopted openness as a tool for economic liberalization in strategic sectors such as ICT, Domestic investment and international trade, to augment growth rate of the Economy. With adoption of targeted reforms, trends in overall trade point out that total trade has increased from USD 7911.79 million in 1994 to USD 11,606.77 million in 2012. Although, the Country’s Terms of trade have continuously declined significantly since 1994, from USD -21 million to USD -1,868 million. Numerous studies by National institute of statistics, (2013) partly attributed this to an increasing need for the intermediate goods required in the growing manufacturing sector.

Trends indicate that production has shifted from traditional cash-crops, such as coffee, cotton, tobacco and tea to new products, such as fish, flowers, manufactured goods and services. Further, tourism expenditure increased from USD 1.77 million in 1994 to USD 869 in 2012. Telephones are also an indicator of the growth of ICT. The number of telephones per 100 persons increased from 0.21 in 1994 to 37.76 in 2012. Lastly, this section elucidated that openness was intended to promote domestic investment, export, import and FDI, especially targeted foreign investment. Since the objective of this study is to measure the impact of investment policy decision on Rwanda’s economic growth, the subsequent chapter explores Export growth, Import expenditure and foreign investment in Rwanda

CHAPTER THREE

EXPORT OUTFLOWS, IMPORT INFLOW AND FDI IN RWANDA

3.1 Introduction

This chapter begins by first indicating the historical background of Export outflows, import inflow and FDI into the country, and examining their performance after the reforms, as well as the available policies that provides a supportive investment climate. The chapter further provides an account of Export, Import and FDI inflows since the reforms, and explains the regulatory framework.

3.2 Export Performance before and after the reforms in Rwanda

According National institute for policy and research (2018), Rwanda has a high degree of export concentration with the top five products accounting for over 60 percent of all exports over the last 15 years indicate a risky heavy reliance on a very limited number of goods. Rwandan exports are still dominated by tea, minerals and coffee although their share of exports declined from 94.4 in 2001 to 79.12 in 2008. Rwanda’s next three largest export products, with growth rates of over one hundred percent, are alcoholic beverages, vegetables and non-alcoholic beverages, representing 5.4, 3.7 and 3.7 percent of exports in 2008, from near zero in 2001 (See table 2). Table 2 below presents Rwanda’s principal manufactured export products by value, share of exports and annual growth in 2001 and 2008. Rwandan exports are dominated by three products: tea, minerals and coffee. However, the domination of the three top commodities in exports declined from 2001 to 2008, from 94.4 percent to 79.12 percent. This came at the back of very high rates of growth in value of the top three Rwanda exports with growth rates ranging from 17 percent for coffee to 29 percent for tea products

b) Non Traditional Exports

Rwanda’s non-traditional exports include alcoholic beverages, vegetables and non-alcoholic beverages and these constitute the next three largest exports. Alcoholic beverages, vegetables and non-alcoholic beverages respectively represent 5.4, 3.7 and 3.7 percent of exports in 2008, from near zero in 2001.

Although alcoholic beverages, vegetables and non-alcoholic beverages are at a considerably lower export value scale compare to coffee, tea and minerals, they have shown growth rates of over one hundred percent. The high growth in vegetable exports is due to the fact that vegetables are prioritized by government through the promotion of the horticulture sector. On the contrary another prioritized sector i.e. hides and skins is very small, representing only 0.7 percent of total exports in 2008. The hides and skins sector similarly shows a strong annual growth of 17 percent from 2001 as is the case with vegetable exports.

The new urgency for export growth

Rwanda has enjoyed remarkable economic performance over the last 20 years, growing at an average rate of 8% per annum. This continued in 2015 with GDP growth of 6.9%. However, it remains a small, poor, land-locked country, with one of the highest population densities in the world. With limited land, and 80% of the population still engaged in farming, the average farm size has dwindled to 0.33 ha.

At the same time, the country is urbanizing quickly, the urban population increasing from a mere 5% of the total in 1990 to 20% today. The economy is dominated by services, whose share of GDP has been expanding. With higher than economy wide average productivity, its contribution to GDP growth has been expanding steadily.

Figure12: Rwanda’s high growth plateau

Figures are not included in the reading sample.

Source: World Bank, World Development Indicators (Accessed November 14, 2016).

The government’s strategy recognizes that off-farm jobs must be created urgently, even as on-farm productivity must be improved. And given the very small domestic market of 12 million, and the low level of exports-to-GDP (14%), the country must learn to sell more to the rest of the world. The national strategy also underlines the importance of foreign direct investment, which indeed has often been critical for the development of non-traditional exports in Africa.

The Vision 2020 strategy planned for export growth of 15% per annum. The National Export Strategy for 2010-15 pushed this up to 18% annually. The Economic Development and Poverty Reduction Strategy 2013-2018, written at the peak of the commodities boom, envisaged accelerated export growth of 28% per annum. Building on a sector-by-sector analysis of opportunities, the revised national export strategy for 2015-18 proposes a slightly less ambitious annual rate of export growth of 20% (World Development Indicators, 2016).

According to MINICOM, Revised National Export Strategy, (2016), the rapid rise in exports of goods and services since 2000.

Figure13: Evolution of trade balance, 1960-2015

Figures are not included in the reading sample.

But so does the even more rapid rise in imports and hence the burgeoning trade deficit. Figure provides a more nuanced picture by comparing exports to GDP. It underlines that up until 2000, the growth was largely a matter of recovering from a collapse in exports in the late 1980s. Only in the last five years have exports-to-GDP reached levels significantly above the historical average pre-1985. Even then, they remain below the peak achieved in 1979 when coffee prices were booming.

Figure14: Evolution of exports/GDP, 1960-2015

Figures are not included in the reading sample.

Source: World Bank, World Development Indicators.

Until about 2010, the increasing trade deficit was not due to deteriorating export prices. In fact, Rwanda enjoyed a major improvement in its terms of trade, even better than the average for sub- Saharan Africa (Figure 3). However, since 2010, the world economy has become less favorable. Export prices have declined, from 22% to 38%, for the three main mineral exports. This has been partially offset by a fall in imported petroleum prices, so the overall terms of trade have fallen by about 20%.

(Source: World Bank, World Development Indicators, 2016)

Figure15: Evolution of terms of trade: 2000-2015

Figures are not included in the reading sample.

Source: UNCTADStat website.

Export volumes rose substantially after 2010 even though unit values fell somewhat (Figure 8). This suggests that rising prices propelled growth in exports until 2010 when volumes took off. (These trends may be partly due to the inclusion of re-exports and cross-border trade as data became available). Import volumes have been rising steadily since 2004, while prices have stayed relatively stable. The growth in imports reflects the rapid growth in the economy supported by significant inflows of aid and foreign investment. With the recent decline in both foreign aid as a share of GDP and mineral prices, a renewed push for expanded exports and private capital inflows has become an urgent priority

Figure16:Evolution of imports and exports: 2000-2015

Figures are not included in the reading sample.

Exports from Rwanda consist of five broad categories: traditional commodities, other formal exports, informal cross-border trade, re-exports, and services (Table 8). The total value amounted to almost US$1.1 billion in 2015. Data on informal cross-border trade only became available starting in 2010 when an annual survey was launched. This survey estimated the value of such trade at US$45 million in 2010, rising to US$110 million in 2014. Since this trade was not included prior to 2010, its addition has somewhat inflated the overall growth rate of exports.

Table1: Exports by Broad Category, 2015

Tables are not included in the reading sample.

Source: IMF (2016), BNR website, MINICOM (2016).

The recent trend in exports has brought to an end the impressive growth of the past decade. This is primarily due to a fall in world mineral prices which then led to cuts in production in Rwanda as some deposits were no longer profitable (Figure 8). This trend continued in 2016, with mineral export revenues expected to fall to US$70 million, one-third of the 2013 peak.

Figure17: Evolution of Principal Merchandise Exports, 2012-2015 (US$ million)

Tables are not included in the reading sample.

In addition, the growth in tourism receipts has slowed down to 2% per annum between 2012 and 2015. This was partly due to the Ebola scare in 2014, and possibly instability in neighboring countries, but may also be the portent of the limits of gorilla tourism. Tea was one of the few bright spots in 2015, but prices have fallen in 2016 while a drought has hurt production.

Changing export portfolio: Diversification, if with remaining vulnerabilities

Rwanda (and East Africa) trade less than the region …

Land-locked countries will generally have a more difficult time exporting due to the high cost of transport to and from the coast. Rwanda tends to have a lower export-to-GDP ratio than other low and lower middle income developing countries which are also land-locked (Figure 6). Its performance is far below the average for sub-Saharan Africa, yet somewhat better than that of Ethiopia and Nepal and, unlike these two countries, moving in the right direction. On the other hand, it is far behind mineral-exporting Zambia and Bolivia. Rwanda is perhaps the best comparator, given its similar economic structure, and it does significantly better. The government is right therefore to focus on increase exports as a driver of growth.

Figure18: Exports to GDP: Rwanda and comparator landlocked countries, 2003-2015

Figures are not included in the reading sample.

Source: WITS, WDI and UNCTAD data. *Two-year averages 2013-2014.

The good news is that today Rwanda exports more products than it did a decade ago, and this is important for mitigating large swings in commodity prices that would otherwise leave the country vulnerable to terms of trade shocks. The share of traditional exports (coffee, tea, and minerals) in formal exports has fallen from 87% in 2004 to 46% in 2015, while other exports and re-exports have grown. However, improved reporting of re-exports over time may also bias this comparison.

Figure19: Changing structure of formal exports: 2004-2015

Figures are not included in the reading sample.

Source: National Bank of Rwanda, Annual Report, various years.

In 2003-5, the top five exports, coffee, tea, tin, coltan and tungsten, accounted for 79% of the total for formal exports (Table 3). By 2013-15, their share had dropped to 71% (excluding re-exports and cross- border trade). However, if one takes the top 10 exports, the situation has not changed much – 80% in 2003/05, 79% in 2013/15. Nonetheless, several exports have become important, such as hides and skins, live cattle, beer and maize flour. This table also underlines the impressive rates of growth in all major exports.

Table2:Top 20 merchandise exports and re-exports and their growth, 2003-2015

Tables are not included in the reading sample.

Source: ITC Trade Map (www.trademap.org). *re-exports

Informal cross-border exports are another important source of diversification, as reflected in Table 1 (but not included in Table 4). In 2015, their value was roughly the same as that of formal non- traditional exports. Table 5 summarises the main products and destinations for informal trade, some of which are re-exports (telephonic apparatus, dried fry and second-hand clothing).

Tables are not included in the reading sample.

Source: MINEAC, Regional Integration Performance Report in the East African (Community, 2016)

One vulnerability: Not enough products in fast growing markets

The world market has had an important influence on Rwanda’s export performance in terms of volume as well as price and the location of Rwandan products in the growth segments of the global market underscore the headwinds it faces. The four quadrant graph (Figure 8) compares Rwanda’s export performance with the trend in world trade for the period 2011-15. It compares the change in Rwanda’s share of world exports to the annual growth rate of world trade for each product. The ideal situation is for a product to be gaining market share in an expanding global market, the upper right quadrant. This is true of tea and raw hides and skins. However, for tantalum (coltan), the situation is less promising Rwanda’s market share is improving but world trade is falling (the top left quadrant). It is still worse for coffee and tungsten (wolfram), where the country is losing market share in a global market that itself is shrinking (the lower left quadrant).

Finally, for tin, world trade is expanding but Rwanda’s share is falling. Overall, this analysis might suggest that more attention should be focused on tin, tea and raw hides and skins where world trade is expanding and would support Rwanda’s competitiveness.

Services are a major source of diversification and fast growth

A notable exception to the quadrant location is omission of services trade. In fact, Rwanda’s exports are actually more diversified than suggested by this table because it excludes services, and notably tourism, and this is a fast-growing segment of the world market. Annual tourism receipts were already estimated at US$41 million in 2003-05, the equivalent of 41% of formal merchandise exports. These revenues grew by an annual average of 22% to reach some US$300 million by 2013-15, equal to 79% of formal exports. Thus, tourism is easily the largest single source of foreign exchange, exceeding the total of the five main commodity exports in 2015, especially after the recent decline in mineral prices. However, this implies that the economy has become increasingly dependent on tourism. Other service exports consist primarily of transport services, such as air or road transport, in the amount of US$72 million.

Figure20: Services exports, 2009-14 (US$ million)

Figures are not included in the reading sample.

Source: MINICOM: Revised National Export Strategy, 2016.

Diversification into new markets complements product diversification

A second form of diversification that complements an increasing number of products and services is exporting to new geographic markets. While traditional exports are sold overseas, most of the rest stays in the sub-region. In particular, it is striking how important the Democratic Republic of Congo has become. It accounts for over 70% of all exports to its immediate neighbors, including Kenya (Figure 10).

Almost half of the remainder goes to Burundi. Rwanda enjoys a strategic advantage with respect to these two markets. Its proximity and favorable investment climate have encouraged some East African companies to establish subsidiaries in Rwanda from which to service these two markets, with their weak industrial sectors and sizeable populations. In many cases, Rwandan firms do not actually export, but simply wait for foreign buyers to come to their factory gate. DRC is particularly interesting because there are two large markets just across the border, in Goma and Bukavu, which attract buyers from as far away as Kisangani. (MINICOM, National Cross-Border Trade Strategy, 2012).

3.3 Empirical review on relationship between export dynamic and GDP growth rate

3.3.1 Import Flows after the reforms in Rwanda

3.3.2 Imports demand in Rwanda in the recovery era

According report by NISR, (2017) Imports of Goods During 16/17 FY, import of goods has increased by 6.6 percent during 17/18, driven by the increase of Energy products and Intermediate goods by 14.2 percent and 14.0 percent respectively compared to the previous FY due to a volume increase of 17.7 percent and 15.2 percent respectively. Capital goods also increased by 1.9 percent after a reduction occurred in previous FY. Consumer goods also increased by 5.8 percent due to an increase in volume of 15.3 percent compared to 16/17 FY

Rwandan import by Category

This page displays a table with Rwanda Imports by Category in U.S. dollars, according to the United Nations COMTRADE database on international trade

a) Rwanda imports electrical, electronic equipment

Rwanda Imports of Electrical, electronic equipment was US$228.37 Million during 2016, according to the United Nations COMTRADE database on international trade. Rwanda Imports of Electrical, electronic equipment - data, historical chart and statistics - was last updated on June of 2020.

Figure21: Rwanda imports electrical, electronic equipment

Figures are not included in the reading sample.

a) Rwanda imports Machinery, nuclear reactors, boilers

Rwanda Imports of Machinery, nuclear reactors, boilers was US$199.91 Million during 2016, according to the United Nations COMTRADE database on international trade. Rwanda Imports of Machinery, nuclear reactors, boilers - data, historical chart and statistics - was last updated on June of 2020.

Tables are not included in the reading sample.

b) Rwanda Imports of Vehicles other than railway, tramway

Rwanda Imports of Vehicles other than railway, tramway was US$136.72 Million during 2016, according to the United Nations COMTRADE database on international trade. Rwanda Imports of Vehicles other than railway, tramway - data, historical chart and statistics - was last updated on June of 2020.

Figure22: Imports of Vehicles other than railway, tramway

Figures are not included in the reading sample.

Rwandan Imports by Country

Tables are not included in the reading sample.

Source: National institute of statistics of Rwanda, (2017)

3.4. Analysis of the relationship between Import demands s and economic growth in Rwanda

The relationship between Import demands s and economic growth of recipient economies remains contested among scholars and policy makers. The underlying question is whether remittances affect the economic growth of recipient countries. Although there is a vast body of literature on this scholarship, conclusions remain mixed. In particular, scholars are divided about the best model to explain the remittance-growth nexus in a way that also takes into account different mechanisms through which remittances affect growth. The methodological issue related to data and the problem of endogeneity between remittances and economic growth has influenced the ongoing empirical debates for developing countries. This section reviews the relevant empirical claims about remittance and growth based the national accounts approach and the endogenous growth model.

The national account approach

The national account approach argues that import demands affect macroeconomic outcomes through their direct and indirect effects on the balance of payment, trade deficit, exchange rate and inflation (Kireyev, 2006; Winters & Martins, 2004 ; World Bank, 2003). The direct effects are that import demands are an integral part of the national account, while the indirect effects are that remittances affect macroeconomic behaviors through their effects on exchange rate and relative prices.

Scholars support the direct channel tend to claim that import demands have a more positive impact on the balance of payments than other capital inflows (such as financial aid, direct investment or loans), because their use is not tied to particular investment projects with high import content, bear no interest and do not have to be repaid. Others argue that import demands are a stable source of foreign exchange than other private capital flows and that, for certain countries, they exhibit an anti-cyclical character (Buch, Kuckulenz, & Le Manchec, 2002; Buch & Kuckulenz, 2004; Straubhaar, 1988 ; see OECD, 2006). The promoters of this theory have consistently argued that, unlike aid, which comes into the economy through the official accounts, remittances, as private inflows once remitted to the home country, can be saved, consumed, or invested. Their effect is through savings, consumption or investments.

Similarly, Amuedo-Dorantes & Pozo (2006); Woodruff & Zenteno (2007) postulate that import demands negatively affects growth by reducing foreign exchange and finance for business investment. To this end, the majority of the literature argues that imports affect economic growth by reducing consumption, savings or investment. Indeed, after reviewing several case studies, Lucas (2005) finds that imports may indeed have served deterring factor to local investment funds in Morocco, Pakistan, and India. Glytsos (2002) modeled the direct and indirect effect of imports on incomes and hence on investments in seven Mediterranean countries and found that import demands negatively affected GDP growth rates in six out of seven countries.

Moreover, import demands increase recipient country income and output growth. Ratha and Riedberg (2005) argue that import demands deterre the recipient individuals' incomes and reduce the recipient country's foreign exchange reserves. Indeed, he observes that if import demands is a form of expenditure that is inversely related to output growth and, may not generate positive multiplier effects (see, for example, Stahl & Arnold, 1986). Adelman and Taylor (1990) found that for every dollar spent on imports, its gross national product (GNP) decreased by $2.69 to $3.17, depending on whether import demands were received by urban or rural households. In the case of unskilled workers who emigrate to escape unemployment, remittances are likely to provide an even clearer net gain to development in the recipient countries. On this note, Fayissa and Nsiah (2008) find import demands impact economic growth and development in Africa. A 10% increase in the import demands of a typical African would results in an about 0.4% increase in average per capita income. Similarly, Adams Jr and Page (2005) find that 10% increase in per capita official international import demands will lead to a 3.5% decline in the share of people living in poverty.

In contrast, those who argue for indirect benefits claim that import demands affect macroeconomic behaviors through their effects on the exchange rate and relative prices. The literature on this approach claims that large import demands inflows within a country with no capacity result in an appreciation of the exchange rate and inflationary pressure in the recipient country thus, causing a situation of “Dutch Disease”. Several studies have argued that import demands negatively affect the macroeconomic variables such as balance of payments, exchange rates, inflation and exports, leading to the appreciation of real exchange rates, inflation, and export of local goods and services and promoting imports, leading to a balance-of-payment deficit.

Biller (2007) finds that import demands deteriorate the balance of trade by stimulating an increase in imports. Jadotte (2009) and Catrinescu, Leon-Ledesma, Piracha, and Quillin (2009) find that large inflows of foreign exchange can have serious consequences.

An appreciation of real exchange rates in the receiving country results in advance effects on tradable commodities and relative competitiveness. This restricts the export performance, potentially limiting output and employment, especially in small economies where remittance inflows are large in comparison to the country's GDP.

Using the ordinary least square (OLS) and fixed effects (FE) instrumental variables regression models, Gapen, Chami, Montiel, Barajas and Fullenkamp (2009) find that decades of private income transfer remittances have retarded long-run economic growth in remittance-receiving economies. The negative effect might be due to the fact that the remittances are generally not intended to serve as investments but rather as social insurance to help family members finance the purchase of life's necessities.

In sum, the national account model provides an insight about how remittances affect the macroeconomic outcomes of the recipient economy. It demonstrates two channels through which remittances as foreign earnings affect growth through macroeconomic outcomes. The direct channel emphasizes the positive effect of import demands as a source of non-costly external capital inflows that improve the capacity of the national account, finance consumption, savings and investment, thus stimulating production and aggregate output. The indirect channel emphasizes the negative effect of import demands on macroeconomic behaviors through their effects on the exchange rate and relative prices.

The literature on the national account approach thus contains an ongoing discussion about the macroeconomic effect of import demands. Similar literature claim that import demands are external private income for the recipient households and receipts to the national current account of recipient countries. Their effect depends on how they are utilized in the national economy coupled with the existing institutional and policy environment that could enhance their multiplier effect on the economy. As correctly stated by the IMF (2005), import demands, like aid, may be more effective in a good policy environment. For instance, a good investment climate with well-developed financial systems and sound institutions is likely to imply that a higher share of import demands is invested in physical and human capital (IMF, 2005).

The endogenous growth model

The endogenous growth model argues that the national output growth is determined by endogenous inputs of total factor productivity (technological progress), physical capital, and human capital under the assumption of constant return. Those supporting the model argue that the import demands -growth effect is detected through the factors that the endogenous growth model focuses on, namely, human capital development, total factor productivity, technological diffusion and physical investments (Romer, 1990;Nelson, 1996; Benhabib & Spiegel, 1994 ;Udah, 2011). It has been extensively documented that the endogenous growth model supports the view that human capital development and technology diffusion promote economic growth and development through their effect on the total factor productivity (TFP).

To this end, Romer (1990) claims that the growth rate of factor productivity depends on the skilled content of human capital.

With regard to import demands, there is limited evidence of a close link between the interaction effect of remittances and that of human capital development variables on economic growth. Most of the existing literature on import demands and human capital development variables focus on micro analysis, determining the development impact of import demands on human capital development. The endogenous growth model bridges this gap by illustrating that the growth effect of import demands is complimented by factors of TFP and technology diffusion. As advanced by Udah (2011), the endogenous growth model illustrates the channel through which import demands could promote economic growth and development. The central argument of the model is that import demands may accelerate the pace of economic growth through enhancing human capital or productivity. The one important channel through which this could happen is technological diffusion. In addition, Udah (2011) argues that the effect of import demands on growth is not direct but factored in through their influence on human capital. The endogenous growth model uses factor productivity and human capital and factors in the interaction between human capital and import demands. The argument is that per capita GDP has a positive relationship with human capital, the interaction of human capital with import demands, physical capital, the labor force, technological diffusion, and government capital expenditure on economic and social services. import demands also improve human capital by increasing resources for health and education (Amuedo-Dorantes, Georges, & Pozo, 2008; Edwards & Ureta, 2003; Gitter & Barham, 2007 ).

Remittances can also reduce domestic macroeconomic volatility, thereby encouraging greater domestic investment (Gapen et al., 2009). (Balasubramanyam, Salisu & Sapsford, 1999; Makki & Somwaru, 2004; see Udah, 2011) found a positive interaction between human capital and international transfers though insignificant. The foreign transfer literature shows how the level of human capital stock affects the absorptive capacity of an economy and consequently, the positive effects of capital inflows such as FDI and import demands on economic growth and development.

Moreover, in his empirical study, Udah (2011) interacted remittances with indicators of human capital development and found that the coefficient of interaction variable was positive and significant. He argues that import demands have a positive effect on economic development but only within a certain threshold of human capital development, although this threshold has not yet been determined. Indeed, Udah interacted technological diffusion and import demands and found a positive and significant effect, suggesting that import demands impact on economic development via transfer of foreign technology or importation of technology.

Rapoport and Docquier (2005) analyzed the link between import demands and education and found that import demands may be seen as repayment of informal loans used to finance educational investments, but also that the prospect of migration makes education a profitable investment for the family. Hence, migration fosters human capital formation, provided that not too many educated individuals emigrate out of the country.

However, critics of this theory have argued that import demands cause moral hazards in the recipient economy, which has a negative effect on productivity and growth. Studies including by Chami, Fullenkamp, and Jahjah (2005) , have argued that in some instances, instead of promoting hard work and productivity, import demands encourage laziness in recipient communities or households since people know that they can finance their consumption through import demands. This in turn affects local labor supply and productivity. Similarly, a study by the IMF (International Monetary Fund, 2005), covering 101 countries over 1970–2003, finds no significant relationship between import demands sent by migrants and growth or between import demands and variables such as education or investment rates. On the side of total factor productivity, there are existing claims that remittance inflows to recipient households erode the quality of governance and reduce accountability to government officials. Abdih, Chami, Dagher, and Montiel (2012 ) ; Clemens and Ogden (2014) argue that import demands could reduce total factor productivity by eroding the quality of governance.

The authors observe that import demands increase the government's revenue base and reduce the cost of rent seeking by public officials. import demands -recipient households, in effect, have incentive to hold politicians accountable, since they can use import demands s to purchase substitutes for public services.

In sum, the two approaches endeavor to demonstrate the stake of remittances in promoting economic growth through their effect on balance of payment, human capital development and technological diffusion in the recipient economy. The review of these two theories indicates the models are compatible and complimentary, rather than mutually exclusive, in explaining the growth effect of import demands. However, the existing theoretical approaches have yet to determine the threshold beyond which remittances negatively affect the real exchange rate and inflationary pressure. Moreover, the literature remains silent about the conditional effect of other factors such as a country's level of development and financial development on the remittance-growth effect in developing countries. It is not merely savings and investments that are important – institutional quality, the policy environment, and the country-specific context are equally essential in determining the growth effect of import demands.

I argue that the lack of a theoretical approach that considers channels and conditions through which import demands continuously affect growth and development contributes to the diverging claims about import demands and economic growth and the overall lack of a conclusion.

This study contributes to this scholarship by employing a two nested methodological framework examining the growth effect of remittances. First, I conduct a cross-section analysis of SSA countries by empirically examining the remittance-growth impact for the period from 1980 to 2014. I employ variables suggested by the theoretical and empirical literature to determine how institutional and development factors interact with remittances to promote economic growth in the SSA region. In the same analytical framework, I use Rwanda as a case study and examine the effects of import demands on its economic growth in relation to the other SSA countries. I then further focus on Rwanda by contextualizing the analysis for Rwanda. I examine the causal linkage between import demands and economic growth in Rwanda. In the next section, I present the growth trends of import demands and the descriptive analysis of remittances in SSA countries and Rwanda in particular.

3.5 Empirical reviews on imports of goods and services and GDP growth rate

Technology has driven country’s import to plays a key role in explaining differences in income and productivity levels across countries. With growing globalization, the international diffusion of technology has become increasingly important in shaping the world’s distribution of income and productivity through imports of supportive deficient capital goods that is badly needed in the country. Recent work has shown that the major sources of technical change leading to productivity growth in most OECD countries are not domestic; instead, they lie abroad (Eaton and Kortum 1999, Keller 2002). The international diffusion of technology is therefore a major determinant of national per capita incomes. The spread of technology may take place through various channels, the most important of which are import trade, foreign direct investment (FDI) and licensing. The role of import trade and FDI in technology diffusion and economic growth has generated an extensive literature in recent years.

Although there has been much work on the relationship between import trade and growth, the existing studies look only at the impact of aggregate trade in both services and manufacturing sectors. The process of deindustrialization of the OECD economies, especially over the last decade, has raised awareness of the increasing dominance of the services sector in employment and output.

Initially there was concern that the slow rate of productivity growth in many services activities would have an increasingly negative impact on economic performance. However, it is now more widely recognised that some services – particularly those related to finance and business – in fact play a critical part in economic development. Moreover a large component of trade and FDI is in services and this raises the issue of the importance of the internationalization and import trade of services for productivity improvement and economic growth.

Many existing studies investigate the impact of import trade in goods and growth. The empirical cross-country studies by Dollar (1992), Sachs and Warner (1995), Ben-David (1993), Edwards (1998) and Coe (1997) suggest that the impact of liberalization of import trade in goods on the long run rate of economic growth is positive, while a recent paper (Rodrigues and Rodrik, 1999) questions the robustness of the results. However, the study on the import trade in services and growth is quite limited. If liberalizing trade in goods, which typically accounts for less than half of GDP in most countries, and even less than a third of output in the industrial economies, can affect economic growth, then there should be comparable gains from liberalizing services that are becoming increasing tradable and that account for a large and growing share of output in most countries.

Likewise, in the literature of international technology diffusion through trade, only trade in goods or aggregate trade is examined. There are no studies available on the role of trade in services in technology diffusion. In this study, we investigate empirically the relationship between imports and economic growth for services and for manufacturing. Using a country level panel of 82 countries, we examine whether imports of services have a positive impact on economic growth.

Key Reviews

Trade and growth

The interaction between trade and economic performance has traditionally been one of the central concerns of development economics. Often it seems that microeconomic studies allow a much sharper discrimination between hypotheses than the aggregate studies of trade growth. However, researchers have long been aware that micro studies frequently miss the economy-wide resource allocation effects that may be central to understanding the effects of trade. This awareness partly explains why the first cross-country studies of openness and growth considerably pre-date much of the rest of the empirical growth literature.

Trade may have effect on economic growth through various channels, one of which is technology diffusion through imports. International trade as a channel for technology diffusion has received much study recently, though the results remain controversial. In the area of trade and technology diffusion, a number of different approaches have been employed to study empirically the importance of the international dimension ¹ (see the review paper ‘International technology diffusion’ by Keller, 2001a). The first and largest set of papers consists of so-called international R&D spillover regressions. Typically, a production function approach is used to relate total factor productivity (TFP) to measures of domestic and foreign research and development (R&D) activities. Foreign R&D is typically given by a weighted sum of all other countries’ R&D activity - weights are usually defined on the basis of bilateral imports, FDI etc. These regressions are partial equilibrium in nature. The second approach employs general equilibrium models in which productivity growth is related to increases in the quality of intermediate goods (Eaton & Kortum, 1997, 1999, Eaton, Gutierrez & Kortum, 1998).

The drawback of this approach is that they have to make some strong assumptions which are difficult to test in the context of a given model. Thus the empirical results are better viewed as estimating or simulating a particular model, rather than selecting one model among several, or testing it.

The recent work by Keller (2001b, 2002) represents a third approach. The relationship between productivity and foreign R&D is studied in a single equation, partial equilibrium framework.Unlike the approach taken in the R&D spillovers literature, Keller estimates the TFP effect of foreign R&D jointly with the importance of one or more channels of diffusion for foreign R&D. One can view this as estimating the weights of the foreign R&D variable together with the parameter that measures the TFP elasticity. Estimating the weights instead of assuming particular weights that are taken from data tables means that less structure is imposed ex-ante.

A common approach of examining the technology diffusion through imports is to investigate the effects of weighted foreign R&D using bilateral imports as the weights. This requires bilateral trade data. Keller (1998) analyses the findings by Coe and Helpman (1995) of trade-related international R&D spillovers and shows that randomly created bilateral trade shares also give rise to large estimated international R&D spillovers. Therefore, the volume of imports can be used as a measure of foreign R&D.

The above studies all employ the volume of trade to examine the knowledge spillover through trade. However, although import tradeto trade does appear to be important for the knowledge spillover, the volume of trade may or may not be.

Falvey, Foster and Greenaway (2002) include a measure of openness in the growth equation and find that openness affects growth through channels other than knowledge diffusion. Unlike the earlier studies, they employ a growth model which allows one to capture more adequately other factors that may affect the extent of knowledge spillover that are not taken account of in TFP calculations. This study also employs different weighting schemes for the knowledge spillovers to test the robustness of the results obtained, and employs a dynamic panel model that allows knowledge spillovers to have both a short-run and long-run impact on growth.

Trade in services and growth

Although there is a large and varied literature on the role of trade in economic growth, few studies have looked at spillovers through trade in services. However, services trade may play an important role in technology diffusion since many of the service sectors, such as financial services, computing and information processing, or management consultancy, are knowledge-intensive. Some studies suggest that services act as intermediate inputs and thereby facilitate other economic activities. In the model of Francois (1990), business services co-ordinate and control specialized operations within firms and become more significant as scale increases and production processes become more complex.

In analyzing the role of services trade, one approach has been to examine in detail particular industries. Mattoo et al (2001), for example, look at such prominent services as the financial sector, telecommunications and transport. They argue that an efficient and well-regulated financial sector leads to an efficient transformation of savings to investment, ensuring that resources are deployed where they have the highest returns; benefits also arise from increased financial product variety and better risk-sharing in the economy. In the case of telecommunications, improved efficiency generates economy-wide benefits since telecommunications are a vital intermediate input and are also crucial to the dissemination and diffusion of knowledge – the spread of the internet and the dynamism that this has lent to economies around the world is a telling testimony to the importance of telecommunications services. Similarly, transport services contribute to the efficient distribution of goods within a country and a country’s ability to participate in global trade, thus helping to realize the benefits of integration. Other services are also crucial. Business services such as accounting and legal services are important in reducing transaction costs. Collier and Gunning (1999), for instance, consider high transaction costs as the most significant impediment to economic growth in Africa. According to summers (1999), the single most important innovation in the history of the American capital markets was the idea of generally accepted accounting principles. Software development is the foundation of the modern information-based economy. Likewise, education and health services are necessary in building up the stock of human capital, a key ingredient in long run growth performance.

Restrictions on trade in goods reduce the level of real GDP, which is equivalent to a loss in welfare. Restrictions on trade in services can, in principle, be expected to have similar welfare costs as they too drive a wedge between domestic and foreign prices of services. Many of the empirical sectoral studies produced so far lend some support to this contention (Hoekman, Braga, 1997). It has been suggested that in the case of services, there is an additional twist in that many services are inputs into production and inefficient production of such services acts as a tax on production.

Thus, goods liberalization in the absence of services liberalization could well results in negative effective protection for goods, highlighting the need for the latter to keep pace with the former (Hoekman, Djankov, 1997). Kim and Kim (2000) examine the changes in productivity in services and manufacturing in Korea over the period 1970-1997. They focus particularly on the role played by the liberalization of services in productivity growth. They conclude that it is probably too early to give a definite answer on this. However, there was a productivity improvement certain sectors such as distribution services which had a large inflow of FDI due to liberalization in the 1990s.

As drawn from the relevant literature that relationship of imports and growth is controversial. Barro, R. J., and Sala-i-Martin (1995). In this study, the aim was to investigate whether trade in manufacturing and trade in services have impact on growth differently. Over the last decade, average growth of trade in services was higher than that of trade in manufacturing, but didn’t exhibit a consistent rapid growth, and trade in manufacturing still dominates world trade. The study presents the share of total services in total trade in services and goods by zones the specific countries. The results confirm that the share of trade in service in total trade doesn’t change significantly in 1990s.trade in services is much smaller than trade in goods, accounting for about 20% of the world trade. To look at the questions more closely, we first estimate the impact of imports (services plus manufacturing) on growth, then divide imports into imports of services and imports of goods and look at the effects separately. Chao, W. S. and J. Buongiorno (2002) for developed countries, we have more disaggregated data which allow us to divide trade in services into trade in transportation, trade in travel and trade in other services. As trade in other services has grown faster than trade in the other two industries over the last decade and trade in other services are more technologically advanced, the three industries are expected to have different impact on economic growth. We also include the ratio of R&D expenditure to GDP as an independent variable in the equations of developed countries (Falvey, R. N. Foster and D. Greenaway, 2002; Ghirmay, T, R. Grabowski and S. C. Sharma, 2001).

Further still, the study equally revealed that the relative importance of service sectors as well as the tradability of services differs greatly across individual countries. Comparing imports for the different countries, the imports of services as a percentage of total goods and services is around 27% in Japan, while around 15% in the United States. Moreover, EU, United States and Japan account for more than 60% of the world trade in services. Generally speaking, trade in services in developed countries is more important to world trade than that in developing countries. We therefore expect that trade in services has different effect on growth in developed countries and developing countries. Furthermore, developing countries are divided into three different country groups according to the economic groupings defined by the World Bank and thus the sub-samples are: upper-income countries, lower-income countries, low-income countries.

Empirically, the estimation results are reported presents the regression results of imports and economic growth for all countries and for developing countries, for developed countries and for 12 EU countries(Chao, W. S. & J. Buongiorno, 2002), . We first examine the impact of imports of both services and manufacturing sectors and then look at the effects of services and manufacturing separately. The Sargen test is satisfied and the test for second order correlation is rejected. We find that the majority of the core variables in the model are of the expected sign and are significant. Thus, a low initial GDP and high initial level of schooling are associated with faster growth in GDP per capita. Faster population growth is associated with slower growth of GDP per capita

Imports are taken as one of the major channels through which international technology diffusion happens, imports are expected therefore to have a positive impact on economic growth in the host countries. From the study, we can see that total imports has a significant positive impact on growth in all countries and developing countries. When the developing countries are divided into three income groups, the results are more mixed. (Falvey, R. N. Foster and D. Greenaway, 2002), After breaking imports into trade in services and trade in manufacturing, we find that trade in services has a significant negative impact on growth in all of the five specifications, while trade in manufacturing has a significant positive impact on growth. This is another interesting finding. Comparing this with the results, we find that imports of services have different effects in developed countries and developing countries.

In the study, we can see that for developed countries, trade in services has a significant positive impact on growth and trade in manufacturing has negative impact although insignificant. We present the regression results for EU countries. (Keller, W., 2001).The key Question to ponder in this study, why do imports of services appear to have different effect on growth in rich and poor countries? Over recent years, services sectors have experienced an important expansion in the developed countries. In 1998, 65 percent of employment and value added was in services in the EU. (Matto, A., R. Rathindran and A. Subramanian, 2001). By contrast, the status of services in most of the developing countries remains at a lower level. Thus to the extent that services imports diffuse knowledge and technical know-how into the services of the developing countries, the impact in rich countries is potentially greater than in poor countries. A second possible explanation lies in the nature of the services trade flows. We might expect imports of business services, for example, to have a more positive effect on growth than imports of tourism (i.e. expenditure of tourists in foreign countries). Unfortunately, due to data limitation, we can test this argument only partially. For the developed countries, we can dis-aggregate services imports into three categories. When we introduces these categories separately into our growth equation, we find that it is the category of other services, including business services, which has a significant positive effect on growth. Indeed, imports of transport and tourism have a negative effect on growth, usually significant. The same data breakdown is not available for developing countries. (Keller, W, 2001b).

Conclusively, in this study, we have tested a dynamic model of growth in the context of different country groups and, more importantly, different industries. There have been many studies on the impact of trade on growth of real GDP per capita, however, few of them look at the impact of trade in different industries. For the first time, we examine the impacts of trade in services and trade in manufacturing on growth separately. We apply a dynamic panel approach and use Arrelano-Bond estimation method to correct for correlation. Our main findings are that imports in services have a significant positive effect in developed countries while their effect does not appear to be significant in developing countries. Of particular note, it is imports of “other services”, including business services that are responsible for this outcome. Imports of manufacturing have been recognised as an important channel for the international diffusion of technology, this study suggest that for developed countries, imports of business may also serve to diffuse economically important knowledge and know-how and hence to promote economic growth.

Although we have obtained some interesting results here, there is still much space to improve the work. For example, we may examine the R&D spill over through trade in services if we can have bilateral data of trade in services available, which is in line with the work of Keller etc. and is widely used in literature on international technology diffusion. We may also do more work on the impact of imports of services in developed countries, and in particular to look at specific services sectors such as finance, computing and information processing and telecommunications, which are expected to embody high technology

3.6 Review of FDI Inflows into Rwanda before and after Economic Reforms

An Investment is considered to be a Foreign Direct Investment (FDI) if non-resident entities or individuals hold 10% or more of the equity share in a resident entity, including all levels of Fellow Enterprises and Direct Investments of even less than 10 percent of shareholding (Rwanda National Bank, 2011).

Businesses that make a foreign direct investment are often called multinational enterprises (MNE). Sometimes FDI can provide better advantages for the MNE but not for the foreign country, and sometimes the other way around. For a foreign multinational enterprise to find it profitable to enter a domestic market some conditions need to be satisfied. This means that the profit needs to be higher than the costs such as communications, transportation, stationing personnel abroad, barriers due to language and customs. It’s critical to identify the advantages for the multinational enterprise under which direct investments will occur. Dilby G. (2014) suggested three conditions that need to be present for a firm to find incentives for direct investment, Dunning explains this as OLI (ownership, location and internalization).

The firm makes a direct investment into the receiving country; this emerges as financial inflow for the receiving country and financial outflow for the investing country/firm. The two main forms of these are M&A (Mergers and acquisitions) also known as brownfield investments, and Greenfield investments. If MNE makes a green field investment, they will build a new factory in a small developing country, this means that the MNE would have to hire some local labor and equipment. When the foreign money gets in the economic cycle, more jobs will be created. Once the new factory is up and running, it will pay taxes for profit and labor, create tax revenue from now possible added economic activity. The country’s government can then use this new capital to promote better welfare, infrastructure and school systems that could create more growth in both physical and human capital (Joutsen T. et al., 2014)

3.4.1 FDI Inflows into Rwanda before the independence

Before colonialism, Rwanda’s economy was informal and agrarian. Following the 1884 Berlin Conference called by German Chancellor Bismarck explained in Chapter Two, Major European colonial powers were free to establish colonies in Africa and establish effective administration. In particular, they were instructed to abolish the slave trade and spread Christianity, and also called upon to economically develop colonies under their sphere of influence. In this section, FDI inflows are examined as two broad sectors: economic and establishment of commercialization, and economic diversification.

3.4.2 FDI Inflows into Rwanda after economic reforms

The Government of Rwanda strives to promote a private sector development, aiming at fostering both local and foreign investment by undertaking reforms with the objective of making the country a favorable place for investment.

Rwanda's performance in the Doing Business Rankings in recent years has been exemplary, drawing attention from international observers and investors alike. The 2013 World Bank Doing Business Report has ranked Rwanda 52nd out of 185 countries. In the overall performance, Rwanda is still the best performing country in the East African region as well as 3rd easiest place to do business in Sub-Saharan Africa (1st is Mauritius which ranks 19th globally, 2nd is South Africa which ranks 39th globally, 3rd is Rwanda which ranks 52nd globally, 4th is Botswana at 59th globally and 5th is Ghana which ranks 64th globally.

Rwanda has been recognized for making improvements in two areas of regulations: Enforcing Contracts (39th) and Getting Electricity (49th). The country made enforcing contracts easier by implementing an electronic filing system for initial complaints whereas the country eased getting electricity by reducing the cost of obtaining a new connection by 30%. Rwanda's ranking per indicator has improved. Looking at areas where Rwanda is still strong, the Starting a Business rank has remained the 8th easiest in the world, with Company registration taking only two procedures and the whole process of incorporation is concluded in just 6 hours. In ease of Paying Taxes, Rwanda is 25th easiest place globally (World Bank, 2013).

Between 2005 and 2012 Rwanda’s real GDP per capita grew by 4.5% a year, reflecting a sustained expansion of exports and domestic investment, with inflows of foreign direct investment also increasing substantially. In addition, the government strengthened the foundations of macroeconomic stability by implementing cautious fiscal policies supported by a number of structural and institutional reforms (National Bank of Rwanda, 2012). Since1995, Rwanda has envisaged a set of policies with the goal of transforming the agrarian subsistence economy into a sophisticated knowledge-based society. These policies are defined in a framework called Vision 2020. The main socioeconomic objectives of Vision 2020 include transforming Rwanda into a middle-income country, with per capita income of about $900 (from $290 in 2000), and transforming the structure of the economy such that the industrial and services sectors will take over by 2020. It is expected that services will contribute 42 percent, industry 26 percent, and agriculture 33 percent of GDP. It is also expected that the population living under the poverty line will be reduced from 60 percent in 2000 to 25 percent by 2020, the population will grow, on average, 2.7 percent a year until 2020, the literacy rate will increase from 48 percent in 2000 to 90 percent in 2020, and average life expectancy will rise from 49 to 55 years (MINECOFIN 2000, vision 2020).

Important socio economic performances have been achieved in Rwanda after the genocide against tutsi and the country has built a solid foundation to its development in long term so that Rwanda is now the top global reformer and for the first time for an African country (World Bank, 2009). These good achievements during last few years are results of good leadership, committed to find durable solutions to Rwandan’s people despite important challenges. Consequently, the country is safe, stable with little corruption and clear anti-corruption policy. Rwandan economic growth remained strong, reaching 11.2% in 2008 against 7.9% in 2007, with an economic growth of 8% in average between 2004 and 2008. In 2008, the contribution of the three sectors (services, industry and agriculture) was 45%, 15% and 33% respectively (National Institute of Statistics of Rwanda, publication 2009).

Figure23: FDI growth in Rwanda (GDPGR-RW)

Figures are not included in the reading sample.

3.4.3 Trends of FDI in Rwanda after Economic reforms

Analysis of the trend is based on the Periods (1995-2012), in this time frame, FDI in Rwanda varied between around 0% and 4.2% of GDP and between around 0% and 1.65% in Burundi. This indicates how the volume of FDI in the two countries still low. Therefore, its impact on economic growth is expected to be limited. Contrary to Burundi, the share of FDI in GDP increased significantly since 2006 in Rwanda. The good economic conditions mixed with political stability are key factors for favorable growth prospects and investment environments in Rwanda, during these last years at least. These elements are major factors that usually attract large amount of FDI in a given country (United Nations, 2001). Following current improvement in doing business in Rwanda, it is expected that the country will attract more foreign direct investments in few next coming years. Indeed, Rwanda is now the top global reformer and for the first time for an African country. Rwanda has set an all-time record for improved overall rankings (World Bank, 2009).

a) FDI size and growth

Rwanda has never attracted large amounts of foreign direct investment (FDI) at any time since independence, and it beneﬁted from very little infrastructure and industrial investment by the colonial powers before that. The small size of the economy, its rural nature, the low level of human capital, the poor quality of infrastructure and landlocked position, high operating costs and limited proved natural resources mean that Rwanda lacks the main drivers of foreign investment by major transnational corporations (TNCs) that may be in search of resources, markets or internationally competitive centres of production. Net FDI inﬂows averaged about $4 million a year in the 1970s. They picked up towards the end of the decade and reached an average $17 million per year in the 1980s, as more liberal policies fostered higher real GDP growth and generated additional investment opportunities (ﬁgure I.6). The bulk of these investments was in agribusiness, banking and tourism. Some of the larger investments included the purchase of the local brewery Bralirwa by Heineken and the participation of Belgolaise (part of the Fortis Group) in Banque de Kigali.

Figure24: Net FDI inﬂows to Rwanda and real GDP growth, 1970-2004

(Millions of dollars and per cent)

Figures are not included in the reading sample.

Sources: UNCTAD, FDI/TNC Database and World Bank, World Development Indicators.

The deteriorating economic performance towards the end of the 1980s and early 1990s, together with the re-emergence of ethnic tensions and political instability, sharply cut FDI ﬂows well before the genocide in 1994. Inﬂows then came to a halt with the genocide and the collapse of society and the economy. In addition to the human disaster, the genocide led to the deterioration or destruction of much of Rwanda’s capital infrastructure, and output collapsed back to the level of 1970. Normal economic activity stopped for several months, and most foreign investors suspended their operations and repatriated foreign staff.

Although no comprehensive data are available, it appears that the majority of the larger foreign ﬁrms present in Rwanda before 1994 did not divest as a consequence of the genocide and the destruction it brought about. This is the case, for example, of Heineken, which quickly resumed operations and reinvested in Bralirwa. Similarly, Belgolaise (Fortis Group) maintained its investment in Banque de Kigali, and Sabena Hotels quickly refurbished the Hôtel des Milles Collines. The Government policy to secure the property rights of owners upon their return contributed to this favorable outcome. Although the genocide does not appear to have led to a wave of divestments by the few larger foreign investors present in Rwanda before 1994, FDI ﬂows have remained minimal over the past ten years, despite the sound economic policies and structural reforms put in place by the new Government. Rwanda has thus completely missed out on the global surge in FDI ﬂows to developing countries that started in the 1990s and peaked in 2000.

Political and military instability in the region, partly generated by the large number of Rwandan refugees in the Democratic Republic of the Congo and the issue of their return home,9 and the indelible impact of the genocide have severely affected Rwanda’s image abroad and put a major brake on new investments by companies or people unfamiliar with the region. Despite the huge progress in the past decade in achieving social, economic and political stability and in improving security and investment conditions, Rwanda continues to suffer from a large “image gap”. This gap between the reality on the ground and perceptions abroad is vast, but it is usually bridged upon the ﬁrst visit. There is thus signiﬁcant potential for “image improvement” initiatives (chapter III), and positive developments in the current constitutional and electoral process in the Democratic Republic of the Congo could also help improve the perception of the Great Lakes region in general.

Other countries in the region and elsewhere have managed to overcome such image problems in the recent past as they came out of civil conﬂicts. FDI ﬂows to Mozambique picked up in the early 1990s, very soon after pluralism and political stability were established under the constitution of 1990 and the Rome General Peace Accords of 1992 (ﬁgure I.7, vertical lines indicate approximate year when stability was achieved). Mozambique also had to deal with a large number of refugees and internally displaced people and needed to establish political stability after several decades of ﬁghting between the FRELIMO- led Government and RENAMO.

The rapid increase in FDI in Mozambique is partly explained by the close proximity with South Africa. South African investors signiﬁcantly contributed to the success of the Maputo Corridor, which quickly brought quality infrastructure (road and railway links to South Africa’s industrial heartland, connection to South Africa’s electricity grid, sea port) and developed around one main investment, the MOZAL aluminium plant. Measures to establish modern regulations in key backbone services (transport, utility, and telecommunications) and to allow private investment in these sectors were also instrumental in the success of the Corridor and the attraction of FDI, however. As a result, Mozambique has attracted signiﬁcant FDI inﬂows over the past decade, not only in resource extraction, but also in industry, agriculture, agro- processing, banking and tourism.

Figure25: Net FDI inﬂows to Rwanda, Cambodia, Mozambique and Uganda 1980-2004

Figure 26: Net FDI inﬂows to Rwanda, Cambodia, Mozambique and Uganda 1980-2004 (Millions of dollars)

Figures are not included in the reading sample.

Note: vertical lines indicate the approximate year stability was ﬁrmly established. Source: UNCTAD FDI/TNC Database.

Similarly, Cambodia attracted FDI inﬂows soon after a formal ceaseﬁre was signed in 1991 and elections were held in 1993. Inﬂows focused on the wood industry, textiles, and tourism. As is recurrent in post-conﬂict countries, investors originated mostly from neighbouring countries, as they are likely to have a deeper understanding of the socio-political situation and investment opportunities in difﬁcult conditions.

In contrast, it took a longer period of reforms and political stability for FDI ﬂows to resume in Rwanda. Inﬂows resumed around 1993, eight years after Milton Obote was overthrown and reforms were introduced by President Museveni. As in Cambodia and Mozambique, a large proportion of foreign investors originated from the region, particularly Kenya and South Africa.

In Rwanda, small amounts of net FDI ﬂows resumed immediately after 1994. Existing foreign investors injected some new capital to resume operations. A limited number of new investors were also attracted, in good part as a result of the privatization programme launched in 1996. Most new investors were familiar with the region, however, which allowed them to see opportunities beyond the image problem.

In many respects, Rwanda shares more similarities with Uganda than with Cambodia or Mozambique, and this has translated into a delay in the resumption of FDI inﬂows. In particular, both countries face constraints owing to their landlocked nature and the low quality of transport infrastructure (roads, rail, and ports) in neighbouring Kenya and the United Republic of Tanzania. They have also suffered from the impact of conﬂicts and instability in the Great Lakes region as a whole. Net annual FDI inﬂows to Rwanda averaged $3.1 million in 1995-1999 and $7 million in 2000-2004, and are on a moderate rising, but erratic, trend. The erratic nature of the ﬂows is the consequence of the role of the privatization programme in generating FDI and the small overall amounts, which can be signiﬁcantly affected by single investments. Although FDI cannot be said to have taken off so far, there are encouraging signs of rising interest by potential investors, partly as a result of the improvement in the investment framework and the Government’s efforts to promote FDI, including through the creation of the Rwanda Investment and Export Promotion Agency (RIEPA) in 2000. Arguably, the socio-political situation can also be said to have been genuinely stabilized only since the return and re-settlement of the majority of refugees around 2000.

The single largest new investment in recent years is MTN’s (South Africa) acquisition of a 40 per cent share in MTN-Rwandacell. The company acquired the ﬁrst mobile telecommunication licence in 1998 and started operation within a few months, rapidly extending coverage to about 75 per cent of the population by mid-2005. Another signiﬁcant new investment was made by US-based investors who set up Terracom, initially as an internet service provider (ISP). The company recently started to expand its services by laying ﬁbre optic cable in Kigali and across the provinces and it bought Rwandatel, the national telecommunication operator, in June 2005 in the largest privatization operation to date. Rwandatel was sold, together with a mobile telephony licence for $20 million, including $5 million in cash upon signature and $15 million in cash payments over the following 10 years.

The involvement of foreign investors such as MTN and Terracom in the ICT sector has been instrumental in the Government’s strategy to develop a knowledge economy and has facilitated business in general. MTN-Rwandacell’s extensive coverage of the territory allows rural areas without ﬁxed-line telephony to have access to telecommunications services. The company has also made efforts to transfer skills to its national staff, which currently represents 98 per cent of the total. In turn,Terracom’s investments in setting up a ﬁbre-optic network will help interconnect Rwanda’s regions, promote e-governance and provide fast and efﬁcient ICT infrastructure throughout the country. The pre-eminence of foreign investments in mobile telecommunication in the early stages of reconstruction is a general trend in post-conﬂict countries. These were the ﬁrst types of investments to occur in a number of countries, including Afghanistan, the Democratic Republic of the Congo, Sierra Leone, Somalia, Sudan, and Uganda. A number of factors explain this, including: (1) the initial capital investment can be relatively modest and is located in more protected urban areas; (2) a return on investment can be generated relatively quickly, in many instances as quickly as two to three years; and (3) company cash-ﬂow is aided by the predominance of pre-paid contracts, which also avoid payment problems. The relatively moderate complexity of technology required to set up mobile phone networks also implies that small entrepreneurs (including in several instances from the Diaspora) willing to take on a fair degree of risk are able to invest where larger multinationals would not.

The privatization programme has been the major channel to attract FDI into Rwanda since its inception in 1996. Although it took a few years to take off, the programme had led to the partial or full privatization of 40 enterprises by October 2005, out of a list of 68 companies identiﬁed for sale. A small number of non-viable companies have also had their assets liquidated. FDI inﬂows through the privatization programme amount to $37 million so far. Aside from the recent Rwandatel operation mentioned above, six asset sales have generated the bulk of the ﬂows. The second largest privatization so far was the sale of 80 per cent of the capital of Banque Commercially du Rwanda (BCR) to Actis, a company fully owned by the CDC Group, itself owned by the United Kingdom’s Department for International Development (DFID). The third largest privatization was the sale of 80 per cent of the capital of Banque Continentale Africaine (BACAR) to a consortium of Fina Bank (Kenya) and Enterprise Holding (Botswana).

Although it did not generate any injection of capital by foreign investors, the Government contracted Southern Sun (South Africa) in 2003 to run the InterContinental hotels in Kigali and Gisenyi, which were refurbished with public funds. While the management contract covered a 15-year period, fees-related issues led Southern Sun to withdraw in 2006. Management of the hotel is currently overseen by Prime Holdings, a government-owned investment vehicle. Full privatization of the hotel is now envisaged and the Serena Group is reported to have expressed interest.

Lahmeyer International (Germany) was also contracted in 2003 to run Electrogaz, the public monopoly electricity and Water Company, for a ﬁve-year period. While it is still too early to assess the impact of Lahmeyer’s takeover of the management of Electrogaz, the purpose of the sub-contracting is to generate transfers of skills and competence, and to prepare the company for full or partial privatization. Since it began operating in 2000, RIEPA has registered 58 investment projects by foreign investors, of which 39 have become operational. Of the latter, 27 represented new investments and 12 involved restructuring, rehabilitation or expansion of existing investments, for a total amount of about $65 million. Reﬂecting the size of Rwanda’s economy, close to 70 per cent of operational investment projects registered by RIEPA involve amounts below $1 million, and only one project exceeds $10 million.

The historically limited involvement of foreign investors in the economy and the time it is taking to rebuild the country’s image following the genocide imply that Rwanda is one of the countries in the world that is attracting the smallest amounts of FDI in relative terms. While FDI ﬂows have recovered somewhat since 1994, Rwanda attracted less than $1 of FDI per capita per annum on average in 2001- 2004, compared to about $12 on average for LDCs and $39 for developing countries (table I.9). Similarly, foreign investment has contributed only very modestly to total investment, as it represented only 2 per cent of gross ﬁxed capital formation on average in 2001-2004, compared to almost 20 per cent in LDCs and 10 per cent in developing countries.

At the same time, investment by nationals is small as a consequence of the low domestic saving rate, difﬁcult access to ﬁnance and a shortage of skills and entrepreneurship. Domestic investment averaged about 16 per cent of GDP in 2000-2003, with the construction sector accounting for close to 90 per cent of the total and equipment goods for the rest. While reconstruction efforts in the aftermath of the genocide partly justify this, the small amount of investment in equipment highlights the slow build-up in productive capacity.

Figure27: Privatization programme, 1996-2005 (Millions of dollars and number of ﬁrms)

Figures are not included in the reading sample.

Source: Secretariat National de la Privatization.

The Régie des Mines (Redemi) is listed as one company to be privatized. Its mining concessions will be sold separately, however, past trends also indicate, however, that foreign investment could in the future contribute signiﬁcantly more to business development, the transformation of the economy and wealth creation. Recent foreign investments, including in ICT and banking, provide encouraging signs.

3.4.4 Foreign Direct Investment and economic growth in Rwanda

a) Prior studies

There are a number of studies that have investigated the relationship between FDI and GDP. Demello (1997) lists two main channels through which FDI may be growth enhancing. First, FDI can encourage the adoption of new technologies in the production process through technological spillovers. Second, FDI may stimulate knowledge transfers, both in terms of labour training and skill acquisition and by introducing alternative management practices and better organizational arrangements.

A survey by Caves R. (1996) underpins these observations and documents that 11 out of 14 studies have found FDI to contribute positively to income growth and factor productivity. Both de Mello and OECD stress one key insight from all studies reviewed: the way in which FDI affects growth is likely to depend on the economic and technological conditions in the host country. In particular, it appears that developing countries have to reach a certain level of development, in education and/or infrastructure, before they are able to capture potential benefits associated with FDI. Hence, FDI seems to have more limited growth impact in technologically less advanced countries.

Zhang K.H. (1999) uses the traditional panel data causality testing method developed by HoltzEakin et al. (1988) in a data set of 80 countries. His results points towards bi-directional causality between FDI and growth, but he finds the causal impact of FDI on growth to be weak. He addresses the question of the two-way link between growth and FDI. Allowing for country specific co integrating vectors as well as individual country and time fixed effects they find a co integrated relationship between FDI and growth using a panel of 23 countries. He emphasises trade openness as a crucial determinant for the impact of FDI on growth, as they find two-way causality between FDI and growth in open economies, both in the short and the long run, whereas the long run causality is unidirectional from growth to FDI in relatively closed economies. Akinlo, A (2004) using a sample of 31 developing countries and using estimators for heterogeneous panel data, found a bi-directional causality between FDI/GDP and the level of GDP. They interpret this result as evidence in favour of hypothesis that FDI has an impact on GDP via knowledge transfers and adoption of new technology. MAhmoud Al-Iriani and Fatima Al-Shami (2007) testing for the relationship between FDI and growth in the six countries comprising the Gulf Cooperation and using heterogeneous panel analysis methods indicate a bidirectional causality. Their results support the endogenous growth hypothesis for this group of countries

b). Rwandan Context

The impact of specific categories of private capital inflows on economic growth has been investigated in different studies, especially for developing countries. They focus both at microeconomic and macroeconomic level. At firm level, studies provide contradictory evidence on the FDI role for economic growth (Wilmore, 1986; Aitken and Harrison, 1999; Haddad and Harrison, 1993). At macroeconomic level, FDI is theoretically expected to impact positively on economic growth through the spillover effects. Industries in host countries could benefit from capacity building and technology transfer; through additional investment in the hosting country especially in economic sectors not significantly financed by domestic investment due to lack of technology, high skilled labor and high costs of production; through investment to deficient sectors with objective of promoting and sustaining a balanced sector growth and through the financing accumulation of productive capital for the future. Based on these expectations, numerous LDCs have relaxed or eliminated restrictions on incoming international Investments and offered more tax incentives and subsidies to attract capital inflows.

Following global changes in the 1990’s, developing countries have favorably looked at various FDI, given their potential contribution to economic development of the host country. On a different note, different studies concluded to a negative impact of FDI on economic growth in the hosting country (Singer, 1950; Prebisch, 1968) since outstanding benefits from FDI are driven by multinational company. Although FDI raises the volume of investment, increase their productivity as well as the consumption in the host country, it impacts negatively on economic growth through price distortions or/ and poor allocation of resources.

Contrary to this theoretical support of the link between FDI flows and economic growth, FDI flows do not contribute necessarily to economic growth; rather, it is attracted by favorable economic environment and opportunities. For another strand of literature, the positive impact of FDI on economic growth depends on the volume of FDI and the context of the host country (Balasubramanyan et al., 1996; Borensztein et al. 1998; Alfaro et al., 2003; Te Velde, 2006). The two factors include general policy factors, specific FDI policies, macroeconomic factors and firm specific factors.

Reisen and Soto (2001) measure the independent growth effect of bond flows as well as FDI, portfolio equity flows, official flows, short-term and long-term bank lending on a sample of 44 developing countries all over the world through the period 1986-97. They find that FDI and portfolio equity flows have a significant impact on growth. Short and long-term bank lending is found to affect negatively economic growth in the recipient country, except when local banks are sufficiently capitalized. Gheeraert and Malek Mansour (2005) find a significant positive relationship between growth and various measures of capital flows like FDI, equity investment, debt investment and flows in financial derivatives.

For De Vita and Kyaw (2009), only developing countries that reached a minimum level of Economic development and absorption capacity are able to capture the growth enhancing effects of both forms of investment inflows. Carkovic and Levine (2002) analyzed the relationship between FDI and economic growth in 72 countries over the period 1960-1995 and concluded to the absence of that relationship for both developed and developing economies even allowing for the level of education, economic development, financial development and trade openness of the host country.

On a different note, any empirical consensus on the relationship between FDI and economic growth exists so far. The divergence is mainly related to the estimation techniques, the model specification, and the choice of sample. The results differ according to countries specificities. Estimation techniques used in empirical literature can be divided in two groups. Early studies on the link between FDI and economic growth used the Ordinary Least Squares (OLS) technique. GDP growth was regressed on a number of variables including trade variable (export or openness) and FDI, using time series or cross section data (Balasubramanyam, Salisu and Sapford, 1996; Olofsdotter, 1998). Recent studies used the technique of bivariate Grange Causality. They allow for possibility of testing causality in both direction (Zhang, 2001; Choe, 2003, Chowdhury and Mavrotas, 2006; Folorunso Sunday, 2009). The major inconvenience of this estimation framework is that the number of variables limited at two. This constraint creates likely model specification’s problem. To address this problem, researchers have used multivariate cointégration technique to consider more variables in the system ( Basu, Chakraborty and Reagle, 2003; Hansen and Rand, 2006, Zhang, 2000, Cuadros, Orts and alguacil, 2004, Ramirez, 2000).

The findings from this set of estimation technique show that there is no consensus on how FDI can impact the economic growth. Based on different empirical researches this impact depends on different factors include the level of technology used in the production system in the host country, the level of skills of workforce, the level of financial sector and institution development, etc.

3.5 Investment Regulations in Rwanda

According to National investment Institutional framework, key stakeholders Efficient and effective management of public investments requires coordination with a wide range of entities and stakeholders; thus clear roles and responsibilities are required. Existing legal framework that guides the actions of stakeholders include:

Law Nº 14/2016 of 02/05/2016 governing public private partnerships, as gazetted on 30/05/2016 (PPP Law) - Prime Minister’s Order determining the functioning of the Public Private Partnership (PPP) Steering Committee; approved by Cabinet on 03/02/2017 - Law N° 12/2013/OL of 12/09/2013 “Organic Law on State finances and property”, as gazetted on 05/11/2013 - N°001/16/10/TC of 26/01/2016 Ministerial Order relating to financial regulations, as gazetted on 03/02/2016 The following stakeholders shall play key roles in the process of delivering public investment.

Parliament approves the Finance Law after holding Budget Hearings (Approval of Finance Bill, as per Article 35 of the Organic Law on State finances and property). It is also an important organ in ensuring accountability of investments undertaken by the Executive.

Cabinet In the course of investment planning, implementation and monitoring, Cabinet is involved as key decision-maker at the Executive level (Pre-approval of Finance Bill and submission to Parliament, as per Article 35 of the Organic Law on State finances and property).

District Councils In the course of investment planning, implementation and monitoring, District Councils are involved as key decision-makers for investments in the respective districts.

The Public Investment Committee (PIC) is a body that approves ongoing and new investments on central government level, which meet the requirements for implementation.

The PIC will be chaired by a high-level representative of MINECOFIN. The Committee will also be comprised of high-level representatives of key spending ministries.

The Local Government Projects Advisory Committee (LGPAC) is a body, which advises on the quality and relevance of ongoing and new projects that meet the requirements for implementation on district level. The LGPAC will be chaired by a high-level representative of MINECOFIN and co-chaired by a high-level representative of MINALOC. The Committee will be constituted of high-level representatives from Provinces and key spending ministries.

PPP Steering Committee As per Law Nº 14/2016 of 02/05/2016 governing public private partnerships, the PPP Steering Committee will take over the gateway and oversight function of PPP projects. The Steering Committee is specifically responsible for approving the shortlisted bidders and the preferred bidder for a PPP project.

Clusters.

These are forums bringing together high level decision makers in government for the purposes of improving coordination. Three main clusters exist namely: Economic, Social and Governance. These shall be convened through the respective lead Ministries designated by the Office of the Prime Minister to discuss investments that require consensus among stakeholders. Clusters shall also review progress reports on investments and help to unblock implementation challenges or bottlenecks.

Ministry of Finance and Economic planning, a high-level representative of the Ministry of Finance and Economic Planning (MINECOFIN) will chair the Public Investment Committee and the Local Government Projects Advisory Committee. Various departments in the Ministry are empowered with key roles for the implementation of the National Investment Policy:

a) National Development Planning and Research Department (NDPR) serves as the technical secretariat for PIC, providing the information, analysis and research necessary for the implementation of the National Investment Policy. The department also develops a pipeline of projects to be implemented in the medium term that require financing and monitors the execution of development projects.

b) National Budget Department (NBD) Key responsibilities of the National Budget Department (NBD), which are linked to the investment process, are coordinating the formulation of the annual National Budget and the Medium Term Expenditure Framework. The department also covers Budget policy formulation, which includes: forecasting, monitoring and reporting on the implementation of the National Budget.

c) The Chief Economist’s Office has key responsibilities of mobilizing external resources to finance investments in addition to assessing and advising on the macro-economic impact of investments.

d) The Government Portfolio Management Unit (GPMU) in the Office of the Accountant General is mandated to ensure that Government investments in State Owned enterprises and joint ventures are managed to achieve national strategic objectives. This includes ensuring economic returns and adherence to strong corporate governance principles and standards.

To further improve the management and financial sustainability of the portfolio of GoR’s commercial and quasi-commercial investments an Autonomous Government Investment Body shall be established taking over functions of the GPMU in this area.

Ministry of Local Government A high-level representative of the Ministry of Local Government (MINALOC) will be a member of the Local Investment Advisory Committee. With LODA being an agency of MINALOC, key administrative procedures for planning and implementation of local government projects are located within this Ministry.

a) Local Administrative Entities Development Agency (LODA) / Fiscal Decentralization will take over key responsibilities in the public investment process on local government level and will, therefore, serve as the technical secretariat for LGPAC. LODA also monitors execution of public investments undertaken through Districts. LODA further works with the National Budget Department in MINECOFIN to support Districts in preparing investment related spending plans. Budget Agencies Budget Agencies are entities whose activities are financed by the State Budget. Budget Agencies (BA) are the executing institutions; their responsibility covers the proper handling of investments from identification to implementation and operation according to respective rules and regulations.

a) Line Ministries and their agencies are in charge of identifying suitable projects in line with their sector strategy and coordinating required activities during the planning and implementation of these projects. This includes ensuring a proper monitoring and evaluation system is in place for the execution of the projects under their supervision

b) Strategic Guidance for Investment Planning Project identification will be guided by existing strategic documents to anchor government decisions and to guide sector-level decision makers toward national priorities: (A. Rajaram, et. al., 2014).

The National Vision (to date Vision 2020), the National medium term strategy for development (to date EDPRS 2) and the government development program (to date 7 Year Government Program) will be the most important strategies to orient the prioritization of investments. On the next level investment planning will be informed by macroeconomic analysis and targets, derived from the government’s macroeconomic policy, e.g. sufficiency of investment levels as percentage of GDP with regard to targeted growth rates, appropriateness of the cross-sectoral distribution of investment also taking into consideration employment effects, complementarity of public and private investments as well as conduciveness for domestic and foreign investment. Within sectors, decisions will be guided by sectoral strategies and policies, including e.g. industrial policy and agricultural strategy etc. Lastly, compliance with regional strategies and policies for example the EAC Vision 2050 and the Agenda 2063 for Africa will be considered. National Investment Planning is informed by the above mentioned guiding documents. There will be a feedback loop from investment management to inform these higher-level strategies as to their realism and feasibility. (National investment Policy, 2017). Strategic Guidance for Investment Planning Project identification will be guided by existing strategic documents to anchor government decisions and to guide sector-level decision makers toward national priorities:

CHAPTER FOUR

REVIEW OF KEY THEORIES OF GROWTH AND INVESTMENT:

THEORIES AND ECONOMIC IMPORTANCE

4.1 Growth Theories for the study

This section employs the available literature and theories that explain economic growth by modelling economic growth through production. The sectionr begins by exploring the theories behind economic growth, focusing on the Solow-Swan Neoclassical Growth Theory, MRW Model and the NGT. These theories are based on the production function with foundations from earlier work of Harrod (1939) and Domar (1946). Other theories that could be applicable include the H-O Theory. However, the H-O Theory is mostly suited to studies concerned with capital intensity Hasan (Hasan, Mitra & Sundaram 2010). In this regard, theories based on the Harrod-Domer model (HDM) which concern and savings are employed in this section.

4.2 Modelling Economic Growth

Modelling economic growth starts with the earlier foundations in the HDM, based on two assumptions. First, capital created by investment is the engine for a nation’s economic growth and assumes a closed economy. Second, there are two factor inputs: capital and labour. Capital is scarce while labour is abundant. Investments depend on the capacity to save. Following these assumptions, the HDM has been criticized for being an incremental capital output ratio theory (Hussain, 2000). First, the theory assumes a closed economy to foreign capital flows, which is not practical in this time of globalization. Second, the model assumes no government influence, and that the capital output ratio is constant. Third, the model can be suitable for explaining development in a developed country, where firms and households have the capacity to save. Despite these criticisms, the model provided a platform on which a savings gap can be identified. The model demonstrates that for developing countries to increase economic growth, saving is a tool for future investment. Also, the model provides a basis for technology advancement as a basis for reducing the labour/capital ratio. Following the HDM, the Solow-Swan Model was created to explain economic growth.

4.2.1 The Solow-Swan Model

The Solow-Swan Model is an exogenous growth model with foundations in the HMD. This model is attributed to Solow (1956) and Swan (1956), and is popularly referred to as the Solow-Swan Model. It has been noted as a significant milestone in neoclassical economic growth theories (Dewan & Hussein, 2001). The Solow-Swan Model, following the HDM, argued that labour is an important tool of production in addition to capital. Solow and Swan observed that capital and labour are not fixed but there is productivity growth due to technological progress. Thus, the Solow-Swan Model indicates that output represented by GDP depends on physical capital, labour and efficiency. To derive this relationship, the Solow-Swan Model employed the relationship between the inputs to the production process, and the resulting output described by a production function. The production function indicates the highest output that a firm can produce for every specified combination of inputs (Pindyck & Rubinfeld, 2001). This is based on the assumption that there are two inputs: labour and capital. The production function can be specified as:

Y = Aƒ(K, L) (4. 2. 1)

Where: Y = Output; A = Technical or productivity ƒactor; K = Capital; L = Labour

This equation indicates that in the Solow-Swan Model, the first technology efficiency— denoted A—is a residual (Ilboudo 2014; Muggeridge 2015; Petrosky-Nadeau 2008). This is because the change in the growth of output, commonly referred to as the Solow residual (A), is not explained. However, A is employed in the production function to measure the exogenous increase in TFP. Second, the Solow-Swan Model can explain the impact of physical capital (K) on economic growth, employing the production function (Barro & Sala- i-Martin 2004). This is specified as follows:

Y _t = A_tK_t^αL_t ^1–α (4. 2. 2)

Where: K = Durable physical inputs including machines, buildings, computers; t = Time; L = Labour input associated with human body;

A = Skills level or technology.

Since technology and skills lead to increased output, then: A > 0; α and 1 − α represent production inputs shares (elasticity oƒ Outeut with respect to capital, α = Constant lies between 0 and 1(0 < α < 1)

In this equation, labour consists of all workers and the amount of time they work, as well as their physical strength, skills and health (Barro & Sala-i-Martin 2004). Workers can only engage in activities as long as other activities are foregone, meaning that labour is a rival input. Meanwhile, technology possesses two characteristics. First, technology improves over time; for example, ICT has changed since 1960, and in turn, workers’ productivity has improved. Second, technology differs across nations; for example, developed nations are considered industrialised because of superior technology compared to developing countries. As such, in the production function, as technology improves, (A) improves and so does output, even if the capital and labour inputs remain constant. Considering this relationship, since labour is held constant while technology improves, labour-diminishing marginal productivity arises. To increase output, physical capital increases, implying that with time, the value of physical capital decreases due to depreciation. Ultimately, gross investment decreases in respect to depreciation (Barro & Sala-i-Martin 2004):

K˙t = I _t − ðk _t = S. F[K_t, L_t, A_t] − ðk _t (4. 2. 3) Where: K˙= Differentiation=Investment;= Depreciation; S= saving investment after differentiation; ð=Depreciation

t = Time parameter; however, conventionally: K_t=

; while 0 ≤ S ≤ 1 ð _t

In the production function, output-per-worker is adopted in order to indicate the effect of labour increase in respect to output as a measure of productivity (IIbouldo 2014). The production function productivity equation used for measuring productivity can be written by dividing both sides of output and physical capital by labour, expressed as:

Y = = = A ( ^a= ( )= (4.2.4)

According to Barro and Sala-i-Martin (2004), based on Equation 5.2.4, the behaviour of the economy described by the neoclassical production function forms the basis for the Solow- Swan Model. The fundamental differential equation of the Solow-Swan Model can be expressed in terms of a non-linear equation, which depends on capital(k) specified:

k˙ = s. ƒ (k) − (n + ð). k (4. 2. 5)

Where; n = ; n + ð = Depreciation per capital − labour ratio; k = Following

Equation 4.2.5 when the saving rate equals zero(s = 0), capital per person declines. This is partly as a result of depreciation of capital at the rate ð. Capital per person also declines due to increase in population (n). Also, based on Equation 5.2.4, the production function can be adopted in the Solow-Swan Model to explain the impact of physical capital on output.

The production function illustrated by Figure 4.1 indicates the relationship between aggregate output-per-worker and capital-per-worker, which is determined by the constant returns-to- scale. The Marginal Product of Capital (MPK) is the slope indicating that employing additional units of capital leads to additional output-per-worker, ceteris paribus.

Figure28: The production function:output-worker

Figures are not included in the reading sample.

Physical Capital

Source: Based on Ilbouldo (2014)

This behaviour of production in a nation gives rise to the Solow-Swan Model properties.

Properties of the Neoclassical Solow-Swan Model

The properties of the Solow-Swan Model explain economic growth based on a continuous production function, as indicated by Figure 4.1. Accordingly, output is linked to factor inputs of capital and labour, which in the long-run leads to the steady state equilibrium of the economy.

Constant return to scale

The constant returns-to-scale can be specified as:

α + (1 − α) = 1and α < 1 (4. 2. 6)

Following the constant return to scale, if inputs are multiplied by a specific factor, the output grows by the same factors, indicated as:

F (zK, zL) = zF (K, L); alternatively; Output per worker; y = (4.2.7)

Where: F = Constant returns to scale; z = the factor by which the inputs are increased; while: z > 0

Positive and diminishing returns to factor inputs

Capital and labour factors are assumed to be positive but subject to diminishing returns. Due to the constant returns-to-scale, there is a decreasing marginal product to factor of capital:

= αA_tK^αL^1–α; for all K > 0; L > 0 (4.2.8)

Following Equation 4.2.7, any extra capital increase leads output-per-worker to increase, but successive increases in capital lead the marginal productivity of labour to decrease. This is largely because during production, depreciation occurs, as illustrated by Figure 4.2.

Figure29: Relationship between output, consumption and investment

Figures are not included in the reading sample.

Following Figure 4.2, the relationship between investment, savings and output can be explained:

I= (4.2.9)

Where; I = investment; s = savings; Y=output

In order to invest, a nation must save a given fraction of output per annum. As such:

∆k = 1 − ð (4. 2. 10)

Where: ∆k = Change in capital per work; I = Investment; ð = Depreciation.

In Figure 4.2, αdenotes capital share in income, indicating the elasticity of income per capita with respect to a nation’s saving rate.

Inada conditions for equilibrium production conditions

Inada conditions refer to two equilibrium conditions in variations of capital and labour, in relationship to the marginal productivity (Inada 1963). First, when capital or labour reaches 0, the MPK or labour approaches infinity. Second, as capital or labour goes to infinity, the capital or labour marginal productivity approaches infinity.

Essentiality

In regard to developing countries, this is a key property and the need for capital, so FDI arises. Essentiality means that inputs such as labour and capital [F(0, L) = F(K, 0)] are strictly required during production (Inada 1963). Since the savings capacity for developing countries such as Rwanda is quite low, foreign capital flows, such as FDI and tourism expenditure, are the bridge-gap for savings. It is therefore assumed that such flows lead to accelerated economic growth, job creation and poverty reduction in a nation.

The Steady State of Growth

In the Solow-Swan Model, capital-worker ratio in a nation is determined by two assumptions. First, investment increases capital. Second, capital depreciates each year. Based on the production function, assuming that labour and productivity are constant, output is a function of capital. As such, a change in capital is a function of investment and depreciation, explained by a nation’s savings and growth rate, as:

k˙ = sk ^α − (ð + n). k (4. 2. 11)

Where: k˙ = Growth of capital per worker over time; sk ^α = Savings capital input; ð = Depreciation; n = Exogenous growth rate; n = Population

Equation 4.2.11 indicates that as capital-per-worker increases, so does output-per-worker. However, the growth of output-per-worker depends on capital inflows and growth rate, denoted by (ð + n). k. First, if; sk ^α > (ð + n).k then as capital/worker increases, so does output (GDP), implying that: y = ƒ (k). Second, sk ^α < (ð + n).k; means that as the capital/worker decreases, so does a nation’s GDP. Finally, when, sk ^α = (ð + n).k indicates that the capital/worker ratio remains constant over time, leading to a steady state of growth, as illustrated below.

Figure30: Steady state growth of national economy

Figures are not included in the reading sample.

Source:Based on Barro & Sala-i-Martin (2004), Acemoglu (2007) and IIbouldo (2014)

As demonstrated, the Solow-Swan Model is a tool for explaining how a developing country can improve its level of production through capital such as FDI. The model explains that as capital-per-worker increases, from K ₀ to K₁, so does output-per-worker. In this regard, a nation experiences economic growth, jobs are created and poverty reduced. However, the Solow-Swan Model has been criticised as insufficient for explaining the role of capital. First, the model is based on the assumption that there are only two factors of production: capital and labour. Mankiw, Romer and Weil (1992), indicated that the Solow-Swan Model does not recognise the role played by human capital in production. Second, the Solow-Swan Model was treated as an exogenous model. Third, Kurz and Salvadori (2003) has indicated that planned saving, which is equal to investment and proportional to net income, is Keynesian saving that is not attainable. Finally, the long-run growth path means that once an economy converges, growth reaches a steady state. At this state, the level of capital-per-worker starts to decline, equalling zero (K = 0). The Mankiw, Romer and Weil (MRW) Model and the New Growth Theory (NGT) were devised to explain growth in a nation.

4.3 Mankiw, Romer and Weil Model

As aforementioned, Solow-Swan is a micro-model explaining the role capital and labour on economic growth. In this regard, a number of variables that are important during production were omitted. As a result, later MRW (1992) augmented the original Solow-Swan Model to include human capital in addition to physical capital, labour and efficiency. Human capital in this case refers to skilled persons in an LF. Solow-Swan Model is augmented as follows:

Y = A_tK_t, L_t, H_t; Where: H = Human capital (4. 2. 12)

The MRW Model is similar to the ASSM. The production function basic model is specified:

Y _t = A_tK_t^αH_t^þL_t^1–α–þ; (4. 2. 13)

Where: β = Constant; α = = ( )

Where; w =wage rate; p = price; such that 0< α< 1;0< α< β<1

The model is based on three assumptions. First, that physical capital, human capital and labour productivity (AL) are the factors of production. Second, that technology improves labour efficiency and constant returns-to-scale. In the equation, H^þrepresents the educational productivity parameter, implying that educational productivity is assumed to rise in direct proportion to average human capital per head (Edwards 2007). Third, following the production function, labour efficiency is defined in three different perspectives. Firstly, as output/labour units employed effectively (y = ). Secondly, labour efficiency is capital-per- unit of labour used in production,

(k = ). Accordingly, two key equations arise in the MRW Model:

Ǩ _t = s_ky _t − (n − g − ð)k _t (4. 2. 14)

ȟ _t = s_hy _t − (n − g − ð)ℎ _t (4. 2. 15)

Where: s k = Income fraction invested in capital representing physical capital savings; s h = Fractioninvested in human capital representing human capital saving rate; n = Labourforce growthrate; g = Technology growthrate; Lan A growthrates are exogenous

Finally, the MRW Model assumes that human capital depreciates at the same rate as physical capital. Therefore, α + þ < 1 implies decreasing returns to all capital. If α + þ = 1, then there are constant returns-to-scale in the reproducible factors that arise. In this case, there are constant returns-to-scale during production. However, similar to the Solow-Swan Model, in the long-run, the economy converges to a state of steady growth, meaning that factor units per unit of labour become constant. This is because in the long-run, due to the diminishing marginal returns to physical capital, the host economy converges to a steady state of growth (Sardadvar 2011). The MRW Model has similar shortcomings to the Solow-Swan Model. Despite the criticism, the ASSM identifies the channels through which macroeconomic variables affect economic growth.

4.4 New GrowthTheory

The Solow-Swan Neoclassical Growth Theory has been criticized by NGT advocates who have noted that endogenous factors are important for a nation’s growth. The NGT provided an avenue through which the effects of diminishing returns to capital, causing a steady state in the Solow-Swan Model, can be counteracted (Kurz & Salvadori 2003; McCallum 1996). The NGT is attributed to Romer, PM (1986), Lucas (1988) and Rebelo (1991) who claimed that steady growth can be generated endogenously. The theory is also referred to as the Endogenous Growth Theory. It internalizes technology and human capital such that, unlike physical objects, these two factor inputs are characterized by increasing returns, which drive the growth process in a nation (Cortright 2001). The NGT advocates indicate that factor inputs, such as labour and land, are non-accumulable, while all other factor inputs are accumulable, such as capital.

The NGT has foundations in the ASSM, and the starting point is Equation 4.2.12. However, the distinguishing feature of the NGT is that the model is linear where A > 0, meaning continuous growth. Following Romer (1986), final production can be expressed in the production function, as follows:

Y (H_y, L, x) = H_y^αL ^þ Σ x_i ^1–α–þ (4. 2. 16)

Where: H _y = Human capital in employment in production sector; L= number of

Workers

Following Equation 4.2.19, Rebelo (1991) indicated that due to factors such as research, human capital productivity and government policy, a nation achieves continuous growth

since these variables are endogenously influenced. Also, MPK is constant (MP = A), meaning the absence of a long-run steady growth state. Kurz and Salvadori (1998) have indicated that the simplest and, for a while, most popular method for expressing NGT is by adopting the linear or AK model approach as:

Y = AK _t; A > 0 (4. 2. 17)

Where: K = Composite of capital and Labour inputs

Following this approach, NGT can be expressed:

Y = C^αH ^β = AK (4. 2. 18)

K = Measure oƒ aggregate capital consisting of (C = Physical

CapitalH=Humancapital ; A = Constant productivity parameter Following Equation 4.2.17, rent of profit is exogenous, expressed:

A − ð; where: ð = Depreciation (4. 2. 19)

Studies that employ the NGT approach treat FDI, economic growth, democracy and human capital as endogenous and mutually dependent Nieman (2012). To this end, as illustrated in the figure below, the NGT uses the relationship between some variables to overcome the shortcomings of the ASSM.

Figure31: Links between democracy, economic level, FDI, and human capital

Figures are not included in the reading sample.

Based on the relationship among the variables indicated, the following observations are worth mentioning. First, investment can affect economic growth endogenously but only as long as increasing returns in production are generated by externalities and spill-over effects (Aslam, Hassan & Sakar, 2013). Second, long-run productivity is driven by externalities arising from human and physical capital accumulation. As such, technological progress is created by market forces as a product of economic activity, and is not a free commodity, as applied in the ASSM. Third, human capital and technology drive economic growth, and diminishing returns-to-scale are non-existent. These observations are a departure from the ASSM.

Despite the contributions that provide a solution to the steady state, the NGT has been criticized. According to Kurz and Salvadori (2003), the rate of profit (r) indicated in the NGT is equivalent the marginal productivity of capital (r = ƒ(k). The rate of profit and the steady state are determined in the relationship, implying that both theories are endogenous and exogenous models. As such, the NGT did not attempt to include increasing returns. Due to the weaknesses of the NGT, the ASSM serves better to explain the impact of investment on economic growth in Rwanda. This is because as long as the Solow-Swan Model is augmented, the concerns of the NGT are included too. Moreover, the NGT does not explain increasing return as important to economic growth, which is included in the ASSM. Ecemoglu (2008) indicates that the ASSM can generate sustained growth with technological progress when the original assumptions are relaxed. This study employs the ASSM to test the impact of investment on Rwanda’s economic growth. The next step involves how to measure economic growth.

4.5. Measuring Economic Growth

There are three ways of measuring economic growth in a country. First, the income approach, which measures income generated in a nation by summing up all incomes paid by firms for factors of production. Second, the expenditure approach, which measures final expenditures on goods and services representing total paid out for use of resources such as wages, rent and profit. Third, using the production approach, economic growth is calculated as the sum of all goods and services produced by firms. Considering these approaches, this study adopts expression of GDP, which measures output in a nation in logarithmic terms. It is considered the most suitable first because a GDP per capita approach is mainly concerned with economic growth and welfare while, GDPGR is concerned with the extent to which GDP changes from the previous to the subsequent year in percentage form. Second, in the Solow-Swan Model, output (y) of nations is explained by GDP. As such, the growth in GDP can be employed as a proxy for economic growth. This approach has been employed by studies such as (Antwi & Zhao, 2013;Athukorala, 2003; Egbo, 2011; Louzi & Abadi, 2011), illustrated:

ln(GDPgr_t) = ln(GDP_t) − ln(GDP_t–1) (4. 2. 20)

Following this approach, the rate of growth in output indicating economic growth is measured using GDP annual time-series data expressed in logarithmic terms reflects the GDP rate of growth. Data for the period 1985–2014 was obtained from the WDI at 2005 constant market prices. Since the ASSM also concerns labour, the next section models employment.

4.6 Key Investment theories

4.6.1 Introduction

The previous chapter examined Investment inflows into Rwanda before and after Independence. This chapter examines the theories behind Investment inflows. It starts by defining the key concepts of this study, followed by a brief outline of the historical background to Investment inflows. Later, the chapter discusses Investment theories in two broad sections: market-based theories and international political economy theories. Finally, a brief overview of the economic importance of Investment I is provided.

4.6.2 Definitions

To explain the theories of Investment, this chapter begins by defining some key terms.

4.6.3 Portfolio Investment

Portfolio investment can be defined as a commercial transaction involving securities, with no lasting relationship and effective management control over the enterprise (OECD 2008; UNCTAD 2009b). Securities are either negotiable or non-negotiable investment instruments comprising of equity and debt securities, and investment fund shares or units. Debt securities are financial instruments serving as evidence of a debt, including bonds and notes, stocks and other money markets. Interest is the main type of income, but issuers dealing in debt securities are required to pay a minimum principal and interest to the owner. Equity securities are also shares (listed and unlisted), with claims on the residual values of corporations after the claims of all creditors have been met (IMF 2008). Dividends are a type of income for equity securities. Investment fund shares or units are investments that require investors to pool funds in the form of financial and non-financial assets.

Portfolio investment can be international or domestic. International (Foreign) Portfolio Investment (IPI) is cross-border investment in which a foreign investor acquires a stake in another country in the form of stocks, bonds and other assets, with no long-lasting relationship and a managerial role (Alfaro 2014). However, there are other types of IPI, where investors do not have a lasting relationship and have no influence on management

Such capital flows include financial derivatives and other residual investments, such as short and long credits, loans currency deposits, trade credits and insurance.

4.6.4 Foreign Direct Investment

FDI can be defined as a lasting interest investment made by a resident enterprise in one economy (direct investor) in an enterprise (direct investment enterprise) that is resident in an economy other than that of the direct investor (UNTAD 2009).The lasting interest indicates the existence of a long-run relationship between the direct investor and the direct investment enterprise.

The lasting interest also reflects the high degree of influence on the management of the enterprise. According to the RDB 1971 Code, a foreign direct investor is: A person who is not a citizen of Rwanda or a foreign company, in which more than 50 percent of the shares are held by a person who is not a citizen of Rwanda; and a partnership in which the majority of partners are not citizens of Rwanda.

The definitions of IPI and investment highlight the differences between the two capital flows. First, investment involves cross-border movement of equity owned by the investors in the enterprise. IPI involves buying shares or securities in the enterprise. Second, investment enables investors to access the resources of other enterprises and other sources, such as borrowing from portfolio investments and loans. These privileges are not available in portfolio investments. Finally, investors have a lasting interest and relationship, often directly managing the enterprise. In portfolio investments, short-run instruments are a significant component. Due to the long-lasting interest and relationship, investment liquidation is not easy, but portfolio investments are liquidated when the investors lose confidence in the enterprise operations.

Considering the definitions, theories explain the motivating factors behind capital flows among nations. Investment theories explain the foreign investors’ way of thinking, behaviour and actions. Consequently, there is an interconnection between motivation factors and ways of thinking, decision-making, behaviour and actions. As such, it is necessary to begin by exploring the origins of FDI, to deepen understandings of FDI theories.

4.7 The Origin of FDI Theories

The origin of FDI theories can be traced from a mercantilist notion of capital accumulation. Mercantilists before 1800 believed that the wealth of a nation depended on the treasure accumulated, measured by the amount of gold and silver owned by a nation (Carbaugh, 2004). To achieve this, mercantilists and governments embarked on exploration. As indicated in the figure below, exploration and colonization was a six-stage process in search of wealth (De Vorsey, 1987; Gascoigne, 2000).

Figure32: Main elements of the exploration process

Figures are not included in the reading sample.

FDI flows started with exploration of the globe. There were with three main objectives of exploration. First, due to a shortage of land and natural resources, push and demand economic factors required that Europe extended its power to distant lands, to acquire resources and markets. Second, demand factors were individual and political in nature, with the desire to gain status through territory and wealth accumulation in the form of colonies. Finally, the need for religious and humanitarian organizations to spread Christianity and ‘civilize’ other areas intensified exploration. These demand factors were followed by a selection of areas for exploration. Journeys of exploration were undertaken, and after returning home the explorer reported his findings, described the areas visited and made recommendations. State officials, merchants and missionaries evaluated the report to determine the fulfilment of the demand factors as a basis for colonization and development, marking the beginning of FDI flows into such regions.

Through exploration, European superpowers started to occupy distant territories. Christopher Columbus’ successful exploration, under the auspices of King Ferdinand of Spain, began European settlement in North America Sage (Sage, 2010). China, the richest country in the world by the ninthcentury annexed as much territory as possible on its frontier. Vasco Da Gama’s explorations under the King of Portugal’s auspices led to the occupation of India and the East African coast (Scammell, 2000). This was after Vasco Da Gama’s discovery that the Sultan of Omani had established an empire covering the East African coast with booming trade in natural resources, spices, cloth slaves. Since European imperialism had grown, responding to Vasco Da Gama’s report, concerning the presence of abundant resources and trade that could benefit Europe, streams of European merchants started arriving on the East African coast. Due to growth of the industrial revolution first, the European merchants’ need for land and raw materials as well as investing surplus capital acquiring colonies became the solution (Yelda, 1991). Second, European powers believed that strong national pride could only be attained by acquiring colonies. As a result, following the 1884 Berlin Conference, a country such as Rwanda became a British colony, marking the beginning of European FDI in Rwanda. Mercantilists believed that the acquisition of territories increased trade surplus through exports and subsidies, but minimized imports by imposing tariffs and quotas. As such, countries were to export as much as possible in order to acquire wealth, as opposed to imports, which drained a country of its wealth. Though the mercantilist model was not sustainable as it implied unilateral and asymmetric relationships, it explains the origins of FDI.

4.7.1 from International Trade Theories to FDI Theories

International trade theories were later explained by Adam Smith’s (1776) ‘theory of absolute advantage’ and David Ricardo’s ‘comparative advantage theory’. Similar to the mercantilists’ theories, these early theories did not include the role of FDI in production. Building on this foundation, the factor endowment theory—commonly referred to as the Hecksher-Ohlin (HO) theory--started to show that a nation’s trade would occur based on three factors (Carbaugh 2004). First, demand conditions are determined by tastes and preferences. Second, factor endowment facilitates competitiveness, based on cheap available factor inputs. Finally, technology is a factor input facilitating production. Countries specialise in factor endowment commodities and import comparative disadvantage goods. Similar to other earlier theories, FDI was not explained but the H-O theory provided a foundation for the Portfolio Investment Theory (PIT) and later FDI theories.

4.7.2 The Portfolio Investment Theory to FDI Theories

The PIT started as a perfect market-based theory (Gamal, 2008). Building on international trade theories, the PIT was first proposed by (Markowitz 1952). According to Markowtiz, portfolio selection was based on the law of large numbers, where the actual yield of the portfolio is almost the same as the expected yield.

Therefore, investors should diversify and maximize expected returns by investing in securities that provide maximum expected return. Later, Tobin (1958) developed the Portfolio Theory of Money according to four assumptions. First, all investors are risk-averse. Second, investors select stocks based on two subjective parameters useful to the investor. Third, the values of the two parameters enable investors to rank portfolios, providing maximum utility. Fourth, investors’ portfolio decisions are made based on specific periods. Tobin developed the Separation Theory by proposing that portfolios are interest-bearing assets, but some are high-risk while others are low-risk. His theories did not explain the role played by FDI, however. In1957, Mundell developed the Capital Movements Theory.

4.7.3 Capital Movements Theory

Building on the foundations of international trade theories, Mundell (1957) developed the influential Capital Movements Theory, which first attempted to explain FDI. Following the early PIT, Mundell developed a basic model as an extension of the H-O Theory, to explain trade and capital movements. According to Mundell, due to tariffs, capital flows from a high tariff to a low-tariff country, assuming that the two countries, products and factors of production are identical in both countries (Denisia 2010). As such, capital flows reduce imports, and capital movements and trade are substitutes. Mundell’s model did not explain the role of FDI as a factor input. Capital movements and trade are not substitutes. Despite the shortcomings, Mundell’s Capital Movements Theory became a focal point of FDI theories. Following Mundell’s theory, Hymer’s theory, referred to as the Industrial Organization Theory (IOT) was proposed distinguishing FDI from FPI.

4.7.5 The Industrial Organization Theory

Hymer (1960) developed the IOT based on three basic theories. First, IOT is based on Bain (1956) who proposed the Imperfect Market Paradigm. According to Bain, imperfect markets exist with few competitors, and high entry barriers are expected to provide higher returns. ^{^[1]} However, Bain did not explain FDI in his Imperfect Markets Paradigm. Second, Hymer developed the IOT considering Tobin’s PIT, discussed earlier. Third, Hymer considered the 1957 Mundell Model of Capital Movements Theory, which also does not explain FDI.

Following the three theories, Hymer (1960) developed the IOT based on the nature and operations of local firms and foreign investments. He observed that domestic firms have advantages over foreign firms as they have knowledge about their local economic environment, legal systems, language and culture. Hymer noted that two conditions were possible that enable foreign firms to become viable in a foreign country. First, foreign firms must possess some advantages over local firms, and second, the market must be imperfect. Upon this background, Hymer then drew two key differences between FDI and portfolio investments. FDI, as opposed to portfolio investments, involve assets that enable the home to-host-country capital flows, as a means of maximizing returns based on a firm’s skills and abilities. Due to the existence of assets in FDI, foreign investors are then motivated to seek control of the enterprise abroad. Also, portfolio investments depend on interest rates meanwhile FDI on returns. Hymer derived two conclusions: that interest rates can explain Portfolio Investment but not FDI, and that FDI is capital movement between countries associated with multi-national enterprises (MNEs).

Despite explaining FDI for the first time, IOT has been criticized. First, IOT does not explain why firms with ownership-specific advantages such as superior technology may not invest at home and export as advocated by mercantilists. Second, the theory does not explain the basis for choosing a particular country, for example Rwanda. Despite the criticisms of IOT, it provided a difference between portfolio investments and FDI for the first time. Since Hymer (1960), a number of FDI theories have been developed. In this study, FDI is examined through two broad perspectives: market-based theories and international political economy based theories.

4.7.6 Market-Based Theories

The market-based theories examined here are broadly categorized as FDI perfect and imperfect market-based theories. The FDI perfect market-based theories discussed are Capital and Market Size Theories. Other theories are imperfect market-based theories.

4.7.7 FDI Capital Theory

The Capital Theory, also called the Rate of Return theory, was first proposed by MacDougall (1958) and later Kemp (1964), based on assumptions of a perfectly competitive market ^{^[2]} (Choudhury & Nayak, 2014; Latorre, 2008). This theory suggests that capital flows from a low-rate to a high-rate return country (Gamal, 2008). FDI moves from capital-abundant economies, where returns are low, to capital-scare countries, where returns are high. Thus, foreign investors are attracted to invest when the marginal return is equal to or greater than the marginal cost.

The FDI Capital Theory can explain the phenomena behind import substitution industries established in developing countries such as Rwanda. Due to the high demand for consumer goods such as sugar, soap safety matches and clothing, developing countries attracted FDI in the early 1960s. Demand already existed because imports were the only source of commodities to developing countries. Due to a lack of essential commodities, FDI projects were established to take the advantage of the high returns that existed as early investors in the market. Further, horizontal integration is related to high-return expectation (Caves 1982), because Multinational enterprises are driven by the availability of technology, which leads to low marginal costs and anticipated high returns.

However, empirical studies, such as those by Agarwal (1980) and Bandera and White (1968) do not support the FDI Capital Theory. First, human capital plays a significant role in equalizing rates of return on capital in developing countries. Second, return is inadequate as a precondition for explaining FDI inflows. Third, capital does not necessarily flow from high income to low-income countries, but rather from developed to developed countries, following Linder’s Theory of Overlapping Demand. FDI inflows are higher in developed countries than developing countries. Despite these criticisms, the Capital Theory explains the flow of FDI into Africa. During the 1884 Berlin Conference, Africa was regarded as an agrarian continent that required civilization and development. Therefore, FDI inflows started to come to countries such as Rwanda.

4.7.8 Market Size Theory

The FDI Market Size Theory can be attributed to Bandera and White (1968) and later to scholars such as Asiedu (2006) and Mughal and Akram (2011). These scholars indicate that efficiency seeking FDI is motivated by the size of the market, measured by a firm’s sales or GDP. This is because even if prices do not increase but markets expand, assuming other factors constant the enterprise’s returns expand. As GDP grows, so does GDP per capita and welfare. Countries such as China, India and Pakistan attract high proportions of FDI largely because of high population, despite lower GDP per capita.

However, if FDI inflows were based on market size, then small island countries (such as Cape Verde) would not be attracting FDI. Cape Verde’s land area is only 4,044 square kilometers, and the population was only 491,875 in 2010. However, GDP there has risen from USD 175 in 1975 to USD 3183 in 2008, while FDI stocks increased from USD 4 million to USD 1576 million in 2013 (Africa Development Bank 2012; UNCTAD 2014). As such, the market size theory can explain FDI inflows for some countries, such as China, but not small island countries. Therefore, perfect market FDI-based theories are largely macroeconomic, yet microeconomic theories are equally important. If FDI was based on perfect competition assumptions, such as equal access to knowledge and no barriers to trade, then foreign investments would not exist (Calvet 1981; Kindleberger 1969). Additionally, perfect completion theories do not consider political factors, so FDI theories could be explained better by imperfect competition FDI theories.

4.7.9 FDI Stage Model Theories

For the purposes of this study, due to the characteristics of the Product Life Cycle Theory and the Internationalizations Theory, these two theories are grouped as stage model theories. This is because the two theories indicate that firms transit through specific procedures, steps and stages before establishing a subsidiary abroad (Gustafsson & Zasada, 2011; Masum & Fernandez, 2008; Steffens, 2002).

4.7.10 Product Life Cycle (PLC) Theory

The PLC Theory was introduced by Joel Dean who in (1950), proposed that biological processes can be applied in sociology, but never fully explained the concept (Polli & Cook 1969). Later first, Herbert Spencer introduced the concept of ‘survival of the fittest’ in 1851 after his analysis of the operations of firms in a free market system, but it did not explain FDI. Second, Charles Darwin in his 1859 The Origin of Species suggested that there is natural selection, and that is why some species survive and others die. Based on these concepts, the law of imitation further explained FDI. Following the law of imitation (with three phases related to acceptance of ideas, products and desires) and the technology gap, Vernon (1966) PLC Theory explained FDI based on three stages: slow advance in the beginning followed by rapid and uniform progress, and finally progress continues but slowly slackens until it finally stops.

Vernon combined more concepts to explain PLC as a better explanation for FDI. The Diffusion of Innovation theory indicates that when new products are introduced, development in all countries does not occur simultaneously. The technological gap concept indicates that technical know-how can be available in one nation and not elsewhere, due to differences in factor endowment. Also, the lack of technology in some countries provides a competitive advantage in the short-run, but not in the long-run, resulting in the technological gap trade (Dodd Parrish, Cassill & Oxenham 2014). Since technology is available in nations at different times, two time lags exist: reaction lag and imitation lag. Following these theories and a study conducted on products from the US, Vernon proposed the PLC, based on four assumptions. First, shape assumption: the PLC sales pattern makes an ‘S’-shaped curve. Second, stages assumption: the slope of the ‘S’ curve is comprised of four stages in chronology: introduction, growth, maturity and decline. Third causality assumption: production depends on a demand pattern. Thus, the supply-side market structure and conditions are composed of a number of competitors. However, intensity of competition is driven by changes in demand, which is different for each stage. Finally, strategy assumption: each stage requires that competitors adopt different strategies.

Following the technological gap model, Vernon’s PLC Theory outlined four stages. Based on the technology time lag that causes the imitation process, Vernon’s PLC model explained that the standardization process is important to products. Vernon observed that the standardization process transcends through four stages, which explain FDI.

The introduction stage involves the creation of a new product, which also includes product testing. During this stage, there are no competitors as the product is new. During the second stage, production increases as demand for the product increases. Competitors then enter the market, which leads to maturity as the peak stage for the product. Increased competition leads to the production of differentiated, standardized products, until decline in the fourth stage. As the product enters the fourth stage, research and design intensifies, leading to the creation of a new generation product. For example, the computer industry evolved from flat-screen computers to the iPad. Following these stages of production and movement of goods across borders, the FDI international trade link developed the PLC Theory. This illustrates that production and demand patterns of exports moved from the US to other developed nations first, and later to developing countries.

Figures are not included in the reading sample.

Source: Vernon (1966)

Based on the stages of PLC, Vernon observed that as the United States of America (USA) is a highly industrialized nation with superior technology, new products are first developed there. Following the Linder Hypothesis of Overlapping Demand, new products will first be consumed in the USA and later exported to the EU with similar demand patterns. As completion increases during the growth and maturity stages, the cost of production increases. In turn, producers in the USA shift production to the EU, and thus the FDI inflows to Europe. As imports and production increases in the EU, the market floods and innovation intensifies again. Similar to the initial stages in the USA, the EU pattern also shifts to maturity, in which exports to developing countries begin. During growth and maturity, competition within the EU intensifies, forcing production to shift to developing countries such as South Africa and Kenya, and later to least-developed countries, like Rwanda. Vernon extended Ricardo’s comparative cost advantage theory since production shifts from high to low-cost production centres. Thus, Vernon indicated the effects of innovation, economies of scale and market imperfection to trade and production as a basis for FDI inflows. The PLC Theory indicates that competition enables FDI to flow from highly industrialized nations to the rest of the world in a sequential pattern. Strong competition in innovating countries shifts production to countries with a low cost of production. The PLC Theory has also been extended to explain international trade patterns (Wells 1968).

Though the PLC Theory can explain FDI, a number of shortcomings have been highlighted in empirical studies by Kojima (1973), Kojima and Ozawa (1985) and Yamin (1991). First, the PLC Theory explains import substitution, which was popular in the early 1960s. Second, the PLC Theory ignores the role of international integration driven by technology. As integration increases, the relevance of the PLC for explaining FDI diminishes. This is because with globalization and rapid technological advancement, integration of FDI inflows cannot leapfrog or become a stage-by-stage incremental process. Despite the shortcomings, other theories have developed, such as the Internationalizations Theory.

4.7.11 the Internationalization Theory

Internationalizations can be defined as a firm’s movement of its operations beyond the boundaries of the home country Dima (2010). The internationalization process involves gradual acquisition, integration and use of knowledge about foreign markets and operations, and then slowly, incrementally increasing commitments to foreign markets (Johanson & Vahlne, 1977). The internationalization Theory originated from Coase (1937), who proposed that transaction costs are fundamental to a firm’s success. Following such early theories, Johanson and Wiedersheim‐Paul (1975) developed the internationalization Theory, based on two main observations of four Swedish firms. Earlier studies did not include competition as an entry barrier, due to psychical distance, nor did they consider domestic firms, especially SMEs. During the study, assumptions were made. First, the enterprise first develops in the home country and subsidiaries are introduced based on a series of incremental decisions.

Second, imperfect competition due to lack of knowledge is an obstacle to internationalization. Through incremental decision making and learning from foreign markets obstacles are overcome. Third, perceived risk ^{^[3]} reduces investments in the market, but internationalization is stimulated by the need to control sales, while existing demand in a foreign market increases international operations. Consequently, firms begin by exporting to neighboring countries or countries with comparatively similar in business practice. Finally, the enterprise starts selling abroad via independent agents, implying smaller resource commitments that hinder the establishment of a sales subsidiary.

These assumptions mean that enterprises internationalize as stepwise jumps establish chains.

As Vahlne and Nordström (1993), states: Some reports indicate an increased tendency on the part of firms to leap-frog low commitment modes or to jump immediately to psychically distant markets. Consequently, it is now and then asserted that the theory should be changed. Most suggestions imply that a number of explanatory variables such as, for example, industry, home, and host country characteristics as well as product characteristics, should be added.

MNEs tendency to leapfrog commitments is similar to frogs as amphibians whose habitat is sea but also leap to land when conditions are favorable. However, when conditions are harsh on land frogs have three options. First, frogs hibernate to stay alive. Second frogs can leap back to the sea. Third frogs leap to the surrounding areas with favorable living conditions. Home countries are the best environment for Multinational enterprises but when conditions are favorable abroad, subsidiaries are established. However, when conditions are unfavorable; for example, in Rwanda as explained in Chapter Three Multinational enterprises left the nation by either relocating home or to third countries. A few remained silent with no production. However, when conditions became favorable again in the 1980s, Multinational enterprises returned to the country. In this respect, the living conditions of amphibians can explain FDI inflows in the case of developing nations such as Rwanda.

As stated, three factors are important for internationalization: industry, home and host country conditions. The home and host-country conditions contribute to leap-frogging, implying low-commitments due to psychically distant markets. Consequently, a national enterprise internationalizes in four sequential stages.

Figure33: Sequential internationalization process

Figures are not included in the reading sample.

The vertical axis represents resource commitment, indicating that with time, psychic distance reduces. In this way, resource commitment, experience and knowledge^{^[4]}of the market increases with time. During Stage 1, due to uncertainty, firms start to internationalize by experimenting in the foreign market through limited exports.

After success, firm resource commitment increases involvement in the foreign market, where operation is implemented through independent agents. After succeeding in the first two stages, during Stage 3 the firm can establish a sales subsidiary, because information and experience implies that psychic distance will have reduced. Finally, the firm commits more resources, implying that FDI has transcended through an establishment chain or step-by-step, similar to leap-frogging. This study was further developed by Johanson and Vahlne (1977) through the popularly referred to as the Uppsala Model, having originated from Uppsala University in Sweden.

4.7.12 The Uppsala Model Illustration of FDI Inflows

Following the behavioral aspects of decision making by firms, the Uppsala Model is centered on four concepts: market knowledge, market, decisions and current activities. The internationalization process of a firm is based on two aspects: state and change factors. The state factors are represented on the left-hand side, representing market knowledge and market commitment. The change factors are on the right, representing commitment decisions and current activities, which depend on the state aspects of the firm. Market knowledge indicates a firm’s awareness of the opportunities and challenges of internationalizing. Market knowledge represents proactive factors, as proposed by (Masum & Fernandez 2008). Depending on the knowledge of a given market, the firm is able to identify the opportunities and challenges that exist. With market knowledge, management teams can weigh the extent to which their internal capacity can be utilized to exploit their competitive advantage in a foreign market. Market knowledge and market commitment represent resource commitment.

Figures are not included in the reading sample.

Source: Johanson and Vahlne (1977)

Market commitment is composed of two factors: the number of resources committed and the degree of commitment (Johanson & Vahlne 2009).

The number of resources committed is the total investment capacity the firm is likely to incur in the proposed foreign market; for example, the expenses of establishing an overseas venture on employees and marketing. The degree of commitment refers to the extent to which management decisions can be influenced to commit resources in a foreign venture. When a foreign venture involves the utilization of more specialized resources then the degree of commitment is high. This involves the need to transfer resources to a foreign market due to lack of a suitable alternative; for example, transferring an expert engineer to a foreign market to begin a subsidiary.

Commitment decisions refer to management’s perception about a firm’s involvement in a foreign nation as an opportunity to expand operations abroad through experience. When a firm increases its involvement in a foreign market, experience increases by identifying constraints and opportunities. Commitment decisions are reactive factors, either passively or actively, by responding to competition (Masum & Fernandez 2008). Commitment decisions are faster when firms have a unique technology or specialized marketing knowledge offering a competitive edge. Therefore, the degree of market commitment is higher. Further reactive factors are management commitment decisions originating from economic effects, or uncertainties in a given market, such as economies of scale, which enable firms to internationalise. The degree of resource commitment is high, but when knowledge is low, uncertainties are high, thus reducing management commitment decisions.

Considering the relationship between market knowledge, commitment and decision as conditions for internalization, two observations are worth mentioning. First, due to lack of knowledge, MNEs are reluctant to invest abroad because of the risks involved (Alserud & Tykesson, 2011). Risks are a function of uncertainty and resource commitment, meaning that as knowledge increases, uncertainty reduces and resource commitment increases.

Figure 4.6: Risk model illustration of a firm’s internationalization process

Figures are not included in the reading sample.

Source: Based on Alserud and Tykesson (2011)

As uncertainty decreases from U1 to U2, so does risk, but resource commitment increases from K1 to K2. This is because knowledge of the market increases. Internalization of firms is related to uncertainty, entry mode decisions and transaction costs. The internationalization Theory entry mode is linked to transaction cost, uncertainty and degree of control.

Figure35: internationalization and entry modes relationship

Figures are not included in the reading sample.

Source: Based on Anderson, E and Gatignon (1986)

During the first stage, knowledge is low while uncertainty is high, causing low resource commitment and degree of control. Ultimately, a firm adopts indirect methods of internationalization such as exporting. As knowledge increases, certainty decreases and resource commitment and degree of control increases. In this way, the internationalization Theory demonstrates that through an incremental and gradual process, firms invest abroad, as indicated by the four Swedish firms. However, the internationalization Theory is subject to similar criticism as the PLC Theory.

4.7.13 Appropriability Theory

Appropriability is the excludability of a technology or asset from other firms as a means of providing a reward to the innovator through protection organisations, such as the WIPO. The Appropriability Theory was developed by Magee (1977). According to Magee, MNEs use FDI to earn high returns from their superior technology and skills. The Appropriability Theory includes five stages: new product discovery, product development, creation of the production function, market creation, and appropriability.

Figure36: Project and Technology life cycle and trade relationship

Figures are not included in the reading sample.

Source: Magee (1977)

The Appropriability Theory indicates that the invention of a new product takes time. MNEs with unique technology, skills and knowledge take advantage of lack of access to such assets and invest abroad, especially in developing countries. To this end, the theory means that innovation starts in a developed country and when the product is standardized, MNEs shift production to developing countries. In this way, FDI is explained. However, the Appropriability Theory is an extension of the Industrial Organization, PLC and internationalization Theories. As an extension of the IOT, the theory indicates that MNEs invest abroad in search of the high returns available in developing countries. Thus, capital moves from high to low-income countries. Second, the PLC and internationalization Theory are stage theories that explain FDI based on technology cycle. Despite the criticisms, the Appropriability theory laid foundations for other theories, such as the Internalization Theory.

4.7.14 5 Internalization Theory

Internalization is the ability for an enterprise to operate internationally through its governance structure and common ownership (Shenkar & Luo, 2008). The Internalization Theory was developed by Buckley and Casson (1976), based on the H-O Theory, which provided the theoretical basis for Coarse (1937) to propose the Theory of the Firm. Based on these theories, the Internalization Theory explains that a firm cannot control external factors affecting operations, but management can manage the internal transactions of the firm. Due to market failure, five types of market imperfections exist that enable MNEs to internalize and operate both locally and internationally.

First, The existence of long time lags between initiation and completion of the production process, which in turn causes failure to satisfy future markets. Second, sometimes firms can possess intermediate products and thus take advantage to gain market power, which enables them to practice discriminatory pricing in different markets. Third, the buyer and seller’s lack of knowledge of the value, nature and quality of the product encourages forward integration by controlling the supply and sale of factor inputs, such as superior tangible and intangible technologies. Fourth, government intervention can be a source of transfer pricing through fiscal policies such as tariffs, restrictions on capital movements and discrepancies in taxation, causing imperfections. Finally, monopoly power can enable MNEs to control various markets through crosssubsidisation, predatory pricing and transfer pricing.

These factors facilitate market imperfection, causing firms to develop specific advantages that explain FDI inflows based on four factors. First, firms can possess industry-specific advantages related to the nature of their product and external market structure. Second, region-specific advantages can enable a firm to exploit resources in various region markets; for example, by linking Johannesburg in South Africa to regional markets in East Africa. Third, MNEs internationalise due to nation-specific factors regarding fiscal policies, which include various incentives. Finally, firm-specific advantages increase competitiveness, so increases international economies of scale and scope as well as global competitiveness. Through internalization, vertical and horizontal integration and transaction cost explain FDI inflows across the globe.

4.8 Vertical and Horizontal Integration as Basis for FDI Flows

Buckley and Casson (1976) and Hennart (1982) widened the Internalization Theory to indicate that MNEs can adopt both vertical and horizontal integration operations across the globe. Thus, within the hierarchical units of the firm, based on governance and management decisions, MNEs configure their production, distribution networks, consumption of materials and components, to operate efficiently. As such, centers are located across the globe; for example, American manufacturer of household items, Rubbermaid, sources materials in Thailand, manufactures in China and ships its products back to the USA, to supply other markets in Europe.

Rugman (2012) indicates that in situations where markets fail, MNEs utilize internal markets to efficiently distribute products globally. GVCs enable firms to produce in a nation through investments located in multiple countries. Manufacturing activities are located in low-cost countries due to a nation’s tariff and exchange rate, labour cost and fiscal policies. Thus, a nation’s factor and resource endowment, governance and fiscal policy are important factors that facilitate FDI.

4.9 Transaction Cost Theory as a Basis for FDI

Teece (1982) explained that transaction costs that include all costs related to a firm’s operations are the basis for FDI flows through lower costs, as a means of gaining higher revenues. Buckley (1988) indicated that FDI exists through the Internalization (Transaction) Theory because firms choose low-cost nations to establish enterprises, and because firms can continue to internalize by lowering costs up to the point where the benefits of further internalization are outweighed by the costs.

Although the Internalization Theory can explain FDI, several pitfalls have been identified. First, Agarwal (1980) and Shin (1983) observed that the Internalization Theory is ambiguous in explaining the motive behind internalization and the failure to explain FDI in the short-run. Second, the Internalization Theory and Appropriability Theory are similar, as both indicate that ownership-specific advantages enable MNEs to invest abroad. The two theories are transaction cost-based theories and market seeking. Although the theories have been criticized, they did lay the foundations for Dunning’s Eclectic Theory platform.

4.10 The Eclectic Theory

The Eclectic Theory was first introduced by Dunning (1977) as the Eclectic Paradigm, to explain FDI. Dunning argued that FDI cannot be explained by a single factor but rather a combination of various economic phenomena to explain one economic theory. The name ‘Eclectic Theory’ was derived from the inclusion of a number of theoretical approaches into the one theory (Andersen, Ahmad & Chan 2014). The Eclectic Theory incorporates three theories referred to as OLI: O represents ownership, L localization and I internationalization Theories.

4.10.1 Ownership

Ownership advantages have their origins in the ownership advantages introduced by (Penrose 1959). Dunning (1977) introduced the ownership advantages to explain FDI based on the hypothesis that it was only superior productivity that made US firms more successful than British firms. This is possible at three different levels: firm-specific (micro level), industry level and macro-level (Alfaro, 2014; Denisia, 2010). Firm-specific level advantages, such as managerial effectiveness, organization structure, resources and assets ^{^[5]} enable a firm to outperform local firms. These advantages are the origin of monopoly, offering MNEs superiority over local firms in terms of efficient low-cost production methods. Industry-level advantages in regard to economies of scale can provide advantages over production abroad, due to internal resources that other firms may not access. Due to mass production, the cost of production is low and a firm becomes competitive. Similarly, at a macro-level, ownership specific advantages can enable a firm to access resource endowment and markets that can only be possible by extending operations beyond the home country borders.

4.10.2 Location

The localization advantages explained by Vernon (1966, 1974) provided a framework for Dunning’s proposal that location-specific advantages account for foreign investments. Location advantages are country-specific conditions offered by different countries where firms locate enterprises (Denisia 2010). Such location advantages include the economic benefits of quantitative and qualitative low-cost factors of production, such as raw materials, transport, labour, local infrastructure and utilities. Country-specific advantages include the political environment, which includes the regulatory framework and taxation and fiscal policy. Thus, countries like Rwanda establish policies offering incentives (such as land, buildings and tax holidays) and a pro-investment environment. These political privileges enable firms to operate efficiently and to out-compete local firms and imports. The location theory indicates that a firm can utilize its ownership advantages to invest abroad, as a means of exploiting opportunities that exist, such as government incentives. Finally, local advantages can take the form of social and geographical environmental conditions.

4.11 FDI Development Theories

This study explains FDI development theories based on Kojima’s Japan Model and the Ozawa Economic Development FDI Theory.

4.11.1 Kojima’s Japan FDI Model

Kojima (1978) explained the rise of FDI using a macroeconomic approach based on factor endowment. To explain FDI, Kojima distinguished three different motives for MNEs for investing abroad: resource, labour and market. Kojima categorized FDI as trade-oriented and originating from Japan, while anti-trade for FDI from the USA. Following the H-O and Rybczynski theory of comparative advantage, Kojima developed five propositions as the motivations for FDI. Firstly, natural resource-seeking FDI was classified as trade-oriented. Due to comparative disadvantages in the home country, MNEs invest in comparative advantage goods in host countries. The home country increases imports of its comparative disadvantage, causing growth in vertical specialization between manufactured products and primary products. Secondly, labour-oriented FDI was also considered trade-oriented. As wages increase in industrialized countries, developing countries gain a comparative advantage in labour-based industries.

As such, it becomes beneficial and rational for a developed country to locate its traditional labor-intensive industries in low-wage countries where labour is cheap. Labour-oriented investment is export-oriented and not import substitution. Exports increase in developed countries and third countries, especially in low labour-cost countries located in developing countries. Thirdly, market-oriented FDI can also be trade-oriented. This because when FDI is induced by tariffs in the host country, trade oriented FDI arises. The heavy tariffs on final goods lead to the substitution of exports of such final products for the export of intermediate goods and components necessary for the production of final goods in the home country. In the host country, such FDI becomes import substitution, but not in the sense of negative international investment. Thus, trade is stimulated between the two countries as well as third countries. Also, if import substitution grows towards export orientation then this category of FDI is labour-oriented, and thus a trade-promotion investment. Fourth, anti-trade FDI occurs in market-oriented FDI, by the American oligopolistic FDI. Finally, internationalization of production and marketing through vertical or horizontal-integration FDI. In this case, anti-trade FDI arises when MNE investment in the host country becomes oligopolistic.

Based on the five propositions, Kojima indicated that Japanese FDI is trade-oriented because Japanese MNEs invest abroad by transferring their resources of comparative disadvantage to host countries with comparative advantage in similar industries. In this way, MNEs lead to international reorganization in labour and trade, causing investment. In this way, Japan possesses a comparative disadvantage in labour to developing countries. Thus, labour-based textile industries in Japan face a labour-comparative disadvantage. Japanese MNEs benefit by investing abroad, causing structural adjustment and opening markets in developing countries.

4.11.2 The Ozawa Economic Development FDI Theory

The Economic Development FDI Theory was developed by Ozawa (1992) based on earlier theories. Based on the H-O Theory of comparative advantage, Kojima (1975) and Kojima and Ozawa (1985) explained that countries first, gain from trade when produce and exports are commodities of their comparative advantage, and when imports are goods of comparative disadvantage. Second, firms gain even more from increased trade when comparative disadvantage of intangible assets are transferred to host nations with comparative advantage in those intangible assets.

Following the Eclectic, PLC and Porter’s Competitive Advantage Theory, Ozawa (1992) indicated that first, the supply and demand conditions between countries are not similar due to different supply-side factor endowment and technology, and demand side-consumer tastes. Second, firms such as academic and research institutions create technology and possess intangible assets. Such institutions generate and market technology and skills. Third, economies are not homogeneous, but rather possess a hierarchy at global and regional levels. For example, the USA is a leader at a global level, and Germany, Britain and France are leaders at a regional level in the EU. In terms of industrial development, some are leaders while others are followers, with differing comparative advantage. Fourth, Nations possess natural and compatible stages of development that can be upgraded in a structural sequential manner as developed nations’ stages of industrial development. Fifth, Structural adjustment is a movement from inward looking import substitution to export-led trade and investment, and governments play a significant role.

Considering these characteristics, a nation’s competitiveness and level economic development are similar. A nation’s structural characteristics indicate four stages of development: factor-driven, investment-driven, innovation-driven and wealth-driven.

Figure37:The relationship between stages of economic development and FDI

Figures are not included in the reading sample.

Source: Based on Ozawa (1972)

Factor-Driven Stage: First Stage of Development

The nation’s economy is dependent on natural resources and labour. Economic activities are labor-intensive in order to employ the most abundant resource. Least-developed countries belong to this first stage of development where economic growth is driven by factors of production such as raw-materials and labour. As a result, resource and labour-seeking foreign investors often target least developed countries such as Rwanda to take advantage of low labour costs and the abundant raw materials of the host country. This stage is also associated with trade in primary products and labor-intensive goods. FDI inflows into least-developed countries dominate, while there are either no or minimal FDI outflows.

Investment-driven FDI: Second Development Stage

This stage is characterized by intermediate and capital goods, such as heavy machinery and chemicals used in the manufacture of final products. It is also composed of the infrastructural building goods used in housing, public works construction and communications.

Innovation-driven: Third Development Stage

This stage is similar to the second stage of economic growth. Most developing countries are in this category. More FDI continues to enter the country but the cost of labour and standard of living increases over time, and FDI outflows start to occur. Innovation-driven FDI is the third stage. As Kojima and Ozawa (1985) state, FDI inflows are motivated by market and technology factors. Countries in transition include China, Russia, Brazil and South Africa.

Wealth-Driven: Fourth Development Stage

This is the highest level of development for most developed countries, and is characterized with drift, recessions and decline. Adopting the PLC Theory, the stages of development are distinguished by the changing factor endowment proportions in the nation’s three major factors used in industrial activity: physical capital, human capital and resources capital, both natural raw materials and labour. According to Ozawa’s theory, economic growth occurs through a changing and upgrading pattern, trends and structure of a country’s factors and technological endowments. As physical and human capital grows, so does gross national product. A nation’s particular stage in competitive development is related to its level of export competitiveness.

The transition from labour-driven to investment-driven stage requires that the nation’s domestic investors gain the capacity to begin investing abroad. Investors engage in outward investments in lower-wage countries in labor-intensive manufacturing and resource extraction. Similarly, investors in the country transitioning from factor-driven to investment driven begin to attract inward investments in capital and intermediate goods industries. Thus, the nation’s comparative advantage will change. Likewise, the transition from the investment-driven to the innovation-driven stage indicates that a nation’s comparative advantage will have changed. The transition to innovation-driven FDI begins with the attraction of FDI inward investments in technology-intensive industries, while outward investments are composed of intermediate goods industries.

4.12 Competition Theories

Competition theories explaining FDI can be attributed to Schumpeter’s (1942) ideas about monopoly, oligopoly and monopolistic competition. According to Schumpeter, firms exploit opportunities after creating profitable competitive positions that other firms cannot exploit, through discovery and innovation. Schumpeter (1942) further indicated that:

The beneficial competition of the classic type seems likely to be replaced by predatory or cutthroat competition or simply by struggles for control in the financial sphere. These things are so many sources of social waste, and there are many others such as the costs of advertising campaigns, the suppression of new methods of production (buying up of patents in order not to use them), page 80. Following Schumpter’s propositions, FDI is explained based on monopolistic or oligopolistic competition, causing MNEs to exploit markets and opportunities not available at home.

4.12.1 Monopolistic Competition Theory

The monopolistic competition theory that explains FDI was first introduced by Kindleberger (1969) and Hymer (1976) as a follow-up to the IOT. In 1976, the Kindleberger-Hymer Theory used monopolistic or oligopolistic power to explain the FDI, based on three questions. Why do firms invest abroad? Now do MNEs out-compete local firms yet bear initial sunk costs in foreign countries, such as communication and coordination costs? And, why do MNEs retain control and ownership? Based on these questions, Kindleberger-Hymer noted that FDI existed because of two incentives that attract MNEs to invest abroad. First, incentives related to monopolistic or oligopolistic advantages offered by governments in host countries. Second, FDI thrives in developing countries due to lack of competition. Kindleberger-Hymer concluded that MNEs cannot operate under conditions of perfect competition, but that imperfect completion is the source of FDI. The impact of competition was categorized depending on the source: First, an imperfect market provides incentives to invest abroad due to lack of access to technology, capital and skills. Further, a different brand name can be adopted, as well as different marketing techniques and product differentiation. Thus the existence of market imperfection. Second, factor endowment based on factors of production that cause exclusivity to patented technology skills and know-how among others. In turn, monopolistic competition among firms thrives based on product differentiation. Third, market failure imperfections are by internal and external economies of scale. Finally, governmental policies where host governments such as Rwanda provide incentives to foreign investments. Meanwhile, through high tariffs on imports, FDI becomes the only avenue to enter such markets by establishing a subsidiary abroad.

4.12.2 Oligopoly Theory

As discussed earlier, horizontal and vertical integration are key forms through which MNEs invest abroad. To this end, Knickerbocker (1973) and Graham (1975) proposed the oligopoly theory as an explanation of FDI, because Oligopoly FDI is a horizontal integration strategy where firms try to acquire markets abroad in the same industry (Caves 1974). As a horizontal integration strategy, firms invest abroad as a reaction to imitate rival firms in two forms: follow-the-leader behaviour and cross-investments as a basis for FDI phenomena commonly referred to as first mover. First, oligopoly firms imitate competitors by following the first moving firm abroad, to gain competitive advantages in new markets. Oligopoly firms try to minimise risks by matching their rival’s actions by adopting follow-the-leader investment behaviour (Caves 1974). Second, as first mover, firms try to deter competitors from taking a stake in the home market, as result a result of advantages in the foreign market by engaging in practices such as price cuts. First movers create brand loyalty among consumers; for example, Colgate, Pepsi-Cola and Coca-Cola. Thus, oligopoly investments are winners-take all, while other competitors are denied entry into the market. Additionally, vertically oriented firms, such MNEs, which react by controlling the entire supply chain, further enhance FDI inflows. Due to the need to gain a market, Knickerbocker (1973) and Graham (1974) indicate that FDI increases because of the oligopolistic nature of MNEs.

4.13 Other FDI-Imperfect Market-Based Theories

4.13.1 The Exchange Rates Theory

The FDI Exchange Rates Theory was developed by Aliber (1970). According to Aliber, firms in strong-currency nations like the USA and Britain tend to invest abroad, while firms in weak-currency nations cannot invest abroad. Weak-currency nations are FDI recipients. For example, Rwanda attracts FDI inflows from strong-currency nations because currency depreciation improves the international competitiveness of the host economy, and in turn FDI profitability (Apergis, Asteriou & Papathoma, 2012; Froot & Stein, 1991; Goldberg & Klein, 1997; Nelson, 2015). Also, the value of foreign investments and assets in host countries declines as fewer units of foreign currency are used to buy large quantities in the host country (Lin, Officer & Shen, 2014; Nelson 2015). As such, more FDI is attracted to the depreciated region. Additionally, firms from strong-currency nations can easily borrow as long as they have a better reputation than local firms in weak-currency nations. Strong currency becomes revenue for foreign investors and enables foreign firms to invest abroad.

The Exchange Rate Theory may explain FDI, but a number of shortcomings have been noted. Weak currency erodes the competitiveness of local firms, and increases a firm’s currency exposure. The value of a currency is a key element of a firm’s future expectations about the value of the currency, having a substantial impact on capital flows (Nelson 2015). Currency depreciation expectations usually cause reluctance to invest abroad in that currency. Investors may want to sell assets denominated in the weak currency, as they lose value overtime. As such, the weak-currency theory may not hold as assets in strong-currency countries sometimes attract investors. In particular, a depreciating Euro may deter US investment in the EU, while an appreciating Euro may increase it.

4.13.2 The Internal Financing Theory

The internal financing theory was developed by Barlow and Wender (1955) and on the Gambler’s Earnings Theory. While starting a subsidiary in a foreign country, MNEs start by investing small amounts of capital abroad. As the subsidiary grows, the future expansion is financed by reinvesting subsidiary profits from operations in the host country. The theory was further developed by Anderson (1983), who indicated that growing cash flows possess a positive relationship with investment outlays due to low-cost internal financing. Later, Froot and Stien (1991) indicated that MNEs prefer future internal financing because external financing is more expensive, due to informational imperfections in capital markets. This is due to internal financing FDI from two perspectives. First, due to restrictions of profit repatriation and movement of funds in host countries, MNEs re-invest earnings in the subsidiary. Second, developing countries do not have properly functioning financial markets.

4.13.3 The Diversification Theory

The Diversification Theory can be traced from Daniel Bernoulli, who indicated in his 1738 St Petersburg Paradox that risk-averse investors diversify. Later Markowitz (1952), in his Theory of Portfolio Investment (Selection), proposed that the expected returns-variance rule implies diversification. Building on earlier theories, Bernoulli (1954), page 30 observes that:

This is the rule that it is advisable to divide goods, which are exposed to some small danger into several proportions rather than risk then all together.

While continuing to explain the diversification, Tobin (1958), in the Separation Theorem, indicated that firms have different assets, such as bonds and equities, and investors have different preferences. Diversification is the basis for prosperity as it enables firms to avoid loses. Based on these theories, Grubel (1968) indicated that International Portfolio Diversification could be applied to FDI. Levy and Sarnat (1970) indicate that firms that diversify reduce risk and maximize returns. Further, the existence of opportunities abroad, through diversification, facilitates FDI, and market failure has enabled diversification to become an efficient choice for investment (Teece, 1982).

4.14 FDI International Political Economy (IPE)-Based Theories

IPE refers to the global economy and political interdependence among sovereign states that affect each nation’s operations, practices and policies. IPE is composed of two elements: the state and the market (Gilpin 1978). The state is based on concepts such as the existence of a territory, loyalty with exclusivity and the legitimate use of force and power. The market is based on concepts including functional integration, contracts among players and interdependence among buyers and sellers. However, the state manages production and economic systems as well as politics. According to Balaam and Dillman (2015), IPE is comprised of three interdependent dimensions or systems: political, economic and social.

The political system that uses power is comprised of actors, individuals, the state, international organizations, civil society and MNEs. These actors make decisions concerning the distribution of resources including money, products and intangible things like security and innovation. The political dimension makes rules to achieve national goals and objectives. Meanwhile, the economic system allocates scarce resources managed by various public and private institutions on a day-to-day basis within the market. Social groups first include state identities, norms and associations based on ethnicity, religion or gender. Social groups are also transnational groups (global civil society) with interests that cut across national boundaries, like the ILO. These systems create an environment that explains FDI where everyone in the world is directly or indirectly affected by the IPE (Balaam & Veseth, 2008). When political and economic conditions were not favourable to investment in Rwanda in the early 1990s, FDI flows became negative as explained in Chapter Two. When the political conditions became more favourable, FDI increased in Uganda from about USD 30 million in 1985 to over UAD 1, 146 million in 2014. In the late 1980s, almost all investors left the country and relocated their investments to their home countries or to third countries, mainly Britain and Kenya. Sanctions meant that all foreign investment ceased.

Rwanda’s experience is not unique to this country. In particular, Russia’s current economic and political situation is worth mentioning considering the nation’s IPE conditions. Following EU-USA sanctions on Russia as well as economies in transition have suffered a number of setbacks, including FDI inflows (Connolly, 2015; Kalotay, 2015; UNCTAD, 2015). First, FDI flows to Russia fell by about 70%, representing over USD 19 billion in 2014 (UNCTAD, 2015). Second, during the same year 2014, FDI inflows fell by 51% in transition economies, representing USD 45 billion. Third UNCTAD further indicates that in Ukraine, FDI flows were negative USD 0.2 billion, though FDI flows to Kazakhstan and Azerbaijan rose. The main cause of FDI decline in transition economies has been attributed to the conflict in Ukraine and sanctions on Russia. Similarly, due to political issues in Venezuela, FDI flows fell by USD 9 billion in 2014 (UNCTAD, 2015). Furthermore, the harsh macroeconomic environment largely due to inadequate pro-investment policies as a well as USA and EU sanctions, have contribute to the decline of Zimbabwe (Mbanje & Mahuku 2011; Shangahaidonhi & Gundani, 2014; Sikwila, 2015).

Therefore, the experiences of countries such as Rwanda, Russia, Zimbabwe and Venezuela mean that IPE conditions are important for Multinational enterprises investment. Although Rwanda has recovered since 1980, the development journey was halted. This means that flow of cross border investment through Multinational enterprises is to a large extent dependant on IPE conditions. To this end, Balaam and Veseth (2008), page 17 reiterate and summarize that: The institutions, arrangements and rules of the game that govern the behaviour of states and markets in the IPE can be analyzed as four networks^{^[6]}, structures or bargains that result in the production, exchange and distribution global wealth and power. These bargains determine different patterns of production and exchange, including the distribution of wealth and power all over the world.

As explained, as long as IPE favour a developing nation, MNEs find such as developing country as a favourable destination as explained by Rwanda’s experience. Indeed, though Rwanda has recovered since 1980, the country’s GDP and per capita income is still very low compared to other economies in Africa and Asia. In terms of economic performance, in the 1980s Rwanda was on a par with countries such as Kenya, Ghana and Malaysia (IMF, 2010). These countries have attained remarkable economic improvement. Malaysia is now a role model to Rwanda, yet the two countries were economic peers in the 19780s. The in regard, for a developing country such as Rwanda accelerated economic growth can be attained through FDI when IPE conditions are favourable. Considering the IPE systems mentioned Balaam and Dillman (2014) identified four interdependent levels: global, interstate, state/societal and individual. These IPE interdependent levels are important in explaining FDI inflows.

4.14.1 The Global Level

The global level is the broadest and most comprehensive level covering global factors, with actors such as the WTO, the UN and related institutions. The role played by the WTO in promoting FDI is explained by openness as a policy for trade and investment. Following the adoption of openness, as a member of the WTO, Rwanda’s FDI has increased enormously since 1995. This study explains the role in promoting FDI played by UNCTAD, Bilateral Investment Treaties (BITs) and World Bank institutions, such as ICSID.

4.14.2 UNCTAD, BITs and FDI Promotion

BITs are international agreements designed with terms and conditions that facilitate private investment by nationals and enterprises of one state in another sovereign state (Akhtar & Weiss, 2013; Goyal, Goswami & Solomon, 2014; UNCTAD, 2009). In this way, BITs provide guarantees for a level playing field and the enforcement of standards through a binding investor-to-state independent dispute-settlement mechanism. BITs include four basic elements. First, Conditions for the admission of foreign investors in the host state. Second, standards of treatment of foreign investors based on the MFN and NT, which deter any form of discrimination. Third, protection against expropriation requiring that host nations provide guarantees of compensation based on international standards in case of expropriation of foreign property, as well as guarantees for free transfer and repatriation of capital and profits. Fourth, methods for resolving investment disputes.

According to UNCTAD (2009a) investment agreements are either legally binding or no legally binding. The Memorandum of Understanding is a non-legally binding agreement intended to formalize the willingness of the contracting parties to collaborate in the specific areas agreed upon. There are three legally binding investment agreements: BITs, comprehensive Free Trade Areas (FTAs) and Double Taxation Avoidance Agreement. BITs provide provisions for investment promotion and protection. Comprehensive FTAs are aimed at promoting investments among the contracting parties. Double Taxation Avoidance Agreements protect against the effects of double taxation on goods and services. The primary objective of BITs is to protect investment and promote FDI in host nations by providing signals and guarantees for the protection of business. Conversely, sanctions scare and block foreign investors from such nations, as was case for Rwanda in the 1980s.

4.14.3 ICSID’s Role in Explaining Foreign Investments

ICSID, established in 1966, is a system for the settlement of investment disputes through conciliation and arbitration. ICSID allows jurisdiction consent that cannot be revoked unilaterally, as well as flexibility for parties to decide the host, whether in the host state or the investor's nation. If the parties cannot agree on the composition and constitution of the tribunal, the Convention can offer guidance. The guarantee against diplomatic protection constitutes an incentive for FDI.

ICSID also provides consent in the national legislation investment code of the host state. ICSID arbitration is shielded from interference by domestic courts, and political interference in the form of diplomatic protection. ICSID arbitration is self-contained and independent of national laws, though parties are free to choose the law applicable to the case. ICSID is preferred, as it does not allow the domestic courts to interfere. Provisional measures by domestic courts are allowed in the unlikely case that the parties have stipulated them in their consent agreement. Through these measures, guarantees for Multinational enterprises protection are provided, in turn stimulating FDI inflows.

4.14.4 Interstate and Regional Level Explanation for Foreign Investments

Interstate and regional level refers to the role played by IPE in explaining FDI flows among nations through bilateral and regional economic integration. Bilateral refers to the extent to which two countries cooperate to integrate their trade and investment regimes. Interstate means RTAs entered into by more than two nations.

Bilateral and RTAs are established with the ultimate objective of reducing trade and investment barriers among the contracting parties (Blomstrom & Kokko, 1997). As a result, a number of customs unions and FTAs have been established, especially since the creation of the General Agreement on Trade and Tariffs (GATT) in 1948 and the creation of the WTO in 1995. In turn, the world has witnessed the growth of FDI through GVCs (Blomstrom & Kokko, 1997; Bruhn 2014; Büge, 2014). Accelerating economic growth through FDI is one of the key motivating factors leading to the creation of RTA among the contracting parties. Developing countries have established deeper RTAs ^{^[7]} to reduce trade and investment barriers and enhance transparency and predictability (Buhn, 2014; Buge, 2014). Contractual provisions—including dispute-settlement mechanisms, protection of intellectual property and provisions related to GATT—are included in the agreements. These provisions have enhanced the growth of RTAs, and have accelerated FDI inflows and the growth of GVCS.

4.14.5 National Level and Government Policies Explaining Foreign Investment

The national level is comprised of national policies adopted by a government in pursuance of national objectives. Nations establish standards, regulations and laws with the objective of increasing FDI inflows. MNE ownership-specific advantages are enhanced, and are partly embedded in national policies as engines for attracting FDI (Blomström & Kokko, 2003; Guimón, 2013). This is because MNEs are in a position to establish subsidiaries in countries that promote trade and investment through favourable fiscal, monetary and commercial policies. As a result, FDI inflows are facilitated to flow across the world. A number of theories explain FDI, from Hymer’s IOT to IPE theories. The main objective of this study is to measure the impact of FDI on Rwanda’s economic growth. Before measuring the impact of FDI on these dependent variables, since the study is partly concerned with poverty in Rwanda, it is necessary to first explain the economic importance of FDI to host nations.

4.15 The Economic Importance of Domestic Investment

To explain the economic importance ofDomestic Investment, this study employs the host-country FDI perspective, foreign investor perspective, Dutch Disease effects, Benign and Malign Model and Host Country Domestic Investment Perspective

The host nation Domestic Investment objective is the expectation that Domestic Investment first promotes exports. To this end, in 1996 the GOR established the UEPB to implement ELGS. As such, export promotion was adopted by developing countries to access global, regional and home country markets (Sultan, 2013). As a country achieves higher levels of economic growth, more jobs are created and poverty is reduced. Mercantilists were the first to develop export promotion strategies. They argued that exports lead to a favourable balance of trade, so advocated that imports should be discouraged (Carbaugh, 2004). Mercantilists were criticized for not recognizing the role of international trade in economic growth. Despite the criticism, developing countries such as the Asian Tigers, ^{^[8]} have succeeded since the early 1970s through adopting an export oriented industrialization strategy based on comparative advantage (Palley, 2014).

Through Domestic Investment, governments expect that imports can be produced locally through Import Substitution Strategy (ISS). The policy is aimed at improving a nation’s TOT and overcoming a balance of payments problem. Import substitution, as indicated by Carbaugh (2004), is an inward-looking government initiative developed in the 1950s by developing countries such as Brazil, Argentina and Mexico. By adopting ISS, developing countries assume that dependence on imported consumer products can be reduced by establishing industries that produce such goods locally. Developing countries see that even if they have a comparative advantage in some industries, they cannot compete with industrialized nations due to trade barriers and high industrial development in these nations. Through ISS, developing countries anticipate the acceleration of job creation, reduction of foreign exchange constraints, stimulation of innovation and reduction of poverty.

4.15.1 Foreign Investor Perspective

Considering the foreign investors’ perspective, enterprises abroad are established for expansion as a means of profit maximization. Domestic Investment is classified as horizontal, vertical and conglomerate Domestic Investment (Krugman & Obstfeld, 2006). Expansion is achieved through innovation driven by ICT as a means of achieving efficiency, and competitiveness by reducing transition cost. As a result, Domestic Investment becomes a fundamental channel for international economic integration. Foreign enterprises have become conduits for technology transfer, skills and know-how (OECD 2007).

The forces driving advanced technology innovations are economic policies that have facilitated trade liberalisation and privatisation, including the protection of intellectual property. In this way, world economic integration is enhanced. As integration increases, governments in developing countries intensify their participation in competition, each wanting foreign investors to find their economy a good environment for foreign investment. As these forces shape the world economic order, enterprises continue to pursue economic efficiency, slicing the globe into regions of production, marketing and sources of raw materials and services. In this way, GVCs have grown via several players.

Figure38: Process leading to the growth of Domestic Investment and GVCs

Figures are not included in the reading sample.

As GVCs increase, so do production and market products abroad, making it possible for the growth to be moved offshore as foreign investors move aboard. In this way, jobs are created and production increases, leading to increased economic growth and poverty reduction in the long-run.

4.15.2 The Benign Model of FDI

The Benign Model of FDI is based on the ideas of Moran (1998), who proposed that FDI is a capital tool that can be employed to break through the vicious circle of poverty (VCP) that Ragnar (1953) proposed. According to Moran, the main cause of poverty is lack of capital. This theory is linked to earlier theories, such as the Harrod-Domer Model (HMD), which posits that investment is a function of capital, output and savings. As capital increases, so does output and savings. When investment increases a nation experiences increasing economic growth, employment and poverty reduction. The Benign Model is linked to Prebisch (1951), indicating that Domestic Investment is a capital base package for developing countries to access technology, markets and management skills, and foster their industrial development. Taking the experience of the USA, the nation rapidly grew in the 20th century, largely because of human and physical capital resources from Europe, especially Britain (Sackey, Compah-Keyeke & Nsoah, 2012). In line with these earlier theories, the VCP was developed.

Figure39: The Vicious Circle Poverty

Figures are not included in the reading sample.

Source: Based on Nurske (1953) and (Rohima et al. 2013)

The VCP presents a circular relationship of conditions. Developing countries remain perpetually in poverty because of low income (Ogbuabor, Malaolu & Elias, 2013; Rohima et al,. 2013). According to the model, developing countries are held in a demand and supply trap that cannot be easily broken through, which this study refers to as ‘a poverty cage’. Starting on the demand side, low incomes indicate that consumers do not have enough disposable income to consume, and their ability to save is also limited. Due to low demand on the supply-side, inventory remains high since the purchasing power is low. As such, producers, including households and companies, have low profits and a low capacity to save, translating into capital deficiency and low productivity. Thus, firms return to the poverty cage.

Following the VCP, Moran (1998) developed the Benign Model to demonstrate that FDI is an appropriate tool that can be employed to penetrate the VCP. FDI erodes poverty by bridging the savings gap and increasing the capital base of a developing country, thus increasing production on the supply-side. Meanwhile, on the demand side, as production increases, demand for labour also increases and so does the wage rate. In turn, household incomes increase, as well as firms’ profits.

Brooks, Fan and Sumulong (2003) identified five benefits of FDI, using a panel data study of 58 developing countries. This indicated that first, 50% of a dollar capital inflow translates into an increase in domestic investment. Second, as new foreign firms enter the market, sectoral output increases and domestic prices reduce, due to entrepreneurial capacity that is built in the host country. As local firms gain capacity through education and training, foreign and domestic firms operate more efficiently as productivity increases competition in the host country.

Third, foreign firms bring assets into the host country in the form of superior technology, leading to spill-over effects in the host nation for production by local firms and GDP to increase. Fourth, FDI acts as a bridging gap for foreign exchange, creating easy access for local firms to foreign capital input, and investment increases in the long-run. Finally, foreign firms open up new marketing and distribution channels, which increase export market access. In this way, establishing the capital absorption capacity of communities as indicated by Hacke and Wood (2013); Hacke, Wood and Urquilla (2015) becomes a tool for economic growth, employment and poverty reduction though FDI.

4.15.3 The Negative Effects of FDI on a Nation

4.15.3.1 The Malign Model of FDI

The Malign Model is the alternative theory to the Benign Model (Moran, 1998). This model reveals the negative effects of FDI on the economic growth of the host nation. According to Moran (1998), FDI possesses four major negative effects. First, FDI lowers domestic savings and investment through rent extraction and capital siphoning through local capital markets and local supplies of foreign exchange. Second, FDI is intended to close the insufficient investment and foreign exchange gap of developing countries. However, MNEs crowd out domestic producers. For example, domestic inputs are often substituted by foreign inputs. In turn, domestic input production is impaired. Third, FDI is expected to offer backward linkages to domestic suppliers in host countries. However, such privileges are not available in developing countries. Fourth, in most cases, industries violate environmental, health and safety standards in the countries in which they operate. In view of these effects, Moran (1998), page 2 has stated that there is: The possibility that FDI might lead to fundamental economic distortion and pervasive damage of development prospects of the country is ever-present.

The Stiglitz (2001) criticisms of globalization further explain the negative effects of FDI to a nation. First, as explained in chapter Two IMF and World Bank encourage developing countries to adopt economic reforms. However, MNEs acquire projects in natural resources (such as oil and mineral resources) through incentives and concessions that range from tax holidays to free land give-way. Projects are implemented at low prices and deny HIPC country such as Rwanda tax revenue income for government and jobs for citizens.

Host countries do not reap the full potential rewards from Multinational enterprises projects. Second, to implement reforms developing nations are advised to privatise State owned enterprises (SOEs) and implement from market economy systems. However, Multinational enterprises that take-over SOEs often do not have the will for such firms to benefit the poor communities. In Rwanda as explained in Chapter Two formally industrial towns such as Kigali have turned to tourism because largely all manufacturing firms formally owned by the state collapsed after privatization. Also as explained by Mold (2004), Multinational enterprises commit excessive defence of Multinational enterprises interests. For example, the French Government pressured the Ivory Coast Government to exclude American firms while bidding in the France Telecom mobile telephone license. Arising from these negative effects, FDI projects undermine the host government’s productivity, income, job creation and institutional framework. In turn developing nations experience do no experience the anticipated accelerated economic growth. Also explained in Chapter two, in the case of Rwanda, the insecure non-poor have increased as well as income inequality increases.

4.15.3.2 Dutch Disease and FDI Effects on Host Nations

The Dutch Disease refers to an appreciation of the real exchange rate as a result of increased exports and capital inflows within a country after the discovery of a booming resource, such minerals (Barder, 2006; Eacho, 2013). The term Dutch Disease originated from The Economist, which stated that the Netherlands’ manufacturing sector declined after a large natural gas discovery in 1959. Corden and Neary (1982) proposed the Dutch Disease Model to explain the decline of the manufacturing sector, regarded as the backbone of the Netherlands’ economy. The model explains the relationship between the discovery and exploitation of gas and the decline of the manufacturing sector. The theory demonstrates that a new large natural resource, such as gas, causes a boom in the sector but at the same time causes the tradable sectors to become less competitive, due to an appreciation in the nation’s real exchange rate. To explain the Dutch Disease Theory, Cordern and Cleary (1982), developed assumptions based on one non-tradable sector and two tradable sectors. The no tradable sector refers the booming sector (a newly discovered natural resource) such as gas that was discovered in Holland. In case of a developing nation such as Rwanda, the booming sector can be import substation firms that develop as a result of FDI which becomes the key sector of the country. Meanwhile, the tradable sector refers to the backbone of the economy (such as the manufacturing sector) in the case of developed countries and the agriculture sector for developing countries, such as Rwanda. In this regard, the Dutch disease effects assumptions include. First, perfect labour mobility ensures that wages equalize among the three sectors. Second, all products produced are final consumption goods. Third, balanced trade, since output of the nation equals expenditure. Fourth, No distortion for commodity and factor prices. The price of traded goods and the booming natural resource commodity and manufacturing/agriculture commodity is determined by the world market, while the price of non-traded goods depends on the domestic market.

The flow of foreign capital, such as FDI, can cause the Dutch Disease effect in two stages: the boom stage and the post-boom stage (Brahmbhatt, Canuto & Vostroknutova 2010; Corden & Neary 1982; Javaid 2011). During booms, the nation receives more foreign capital inflows, which cause the local currency to appreciate. Local currency appreciation occurs because the determinants of the exchange rate are internal and external. Internally, exchange rate is determined by tariffs on imports, export and domestic taxes, government policies (such as exchange rate controls and subsidy regime) and technological progress. Externally, the TOT, foreign capital inflows and world interest rates can determine exchange rates. Conversely, during the post-boom stage, natural resources are exhausted, leading to a decline in FDI or foreign capital inflows. In turn, a nation becomes worse off, as traditional sources of income are destroyed. In the case of Rwanda, production and exports have shifted from TEs to NTEs. The impact of FDI on nations differs across regions and countries. Despite the variations in findings, FDI is an important for poverty reduction. FDI leads to technological transfer and is a source of physical capital that is an important base for production, employment and longrun poverty reduction. Chowdhury, Abdur and Mavrotas (2005) proposed that country specific studies can be carried out to ascertain the impact of FDI on host countries.

4.16 Concluding Remarks

This chapter examined the theories underlying cross-border investments through exports, imports and FDI. First, it defined the key terms and concepts underlying the Domestic Investment phenomena, followed by an exploration of the main FDI theories. The chapter started by providing a brief background to the Domestic Investment as a tool for understanding the theories explaining foreign investments. The theories of Domestic Investment were categorized as market-based and IPE. Later, as this study concerns poverty, the economic importance of FDI was examined. According to the findings and building on the Theory of the Firm, MNEs operate at a centre of two extremes, which work in harmony to internationalize.

First, to invest abroad MNEs possess the internal capacity explained by the Theory of Firm and other theories, such as the stage theories and the Eclectic and internalization Theory. Second, while operating abroad, MNEs do not have the capacity to control the external conditions that impact the firm. However, firms may try to manage the conditions that negatively affect their operations abroad, for example; by influencing policy change, though may not have absolute control. As illustrated in the figure above, the leap-frog tendencies of Multinational enterprises explain the FDI phenomena especially in developing countries. This is demonstrated by the increased tendency for Multinational enterprises to leap-frog low-commitment modes or to jump immediately to psychically distant markets.

This study found that four conditions determine FDI flows for a developing nation such as Rwanda, referred to as the firm-home-host-international political environment conditions. First, firm conditions mean that the firm internationalizing must have the capacity to internationalize. Second, the Multinational enterprises home country conditions have to be favourable to enable the firm to build the capacity to internationalize. Third, host-country conditions must be conducive as a pre-condition that enables a nation to become a good destination for Multinational enterprises investment. Fourth, IPE refers to the extent to which international political economy conditions affect both the host and home countries.

Since 19801, Rwanda started to experience economic and political instability. First, between 1988 and 1992, international sanctions were imposed on the country. Second, neighbouring countries, such as Burundi, were also hostile to Rwanda, despite both countries being members of the EAC. After the 199 1, investors started to return to the country, including from Tanzania. Consequently, FDI inflows have increased. What is termed as the Frog-leap Theory according to literature, explains FDI inflows into developing nations such as Rwanda.

Frogs can leap to environments where conditions are good. When conditions are dry and harsh, frogs hibernate, but when the climate is favourable, they begin to jump. If conditions are conducive but frightening or precarious, the frog leaps elsewhere. In Rwanda between 1994 and 1999, when international sanctions were imposed, nearly all investors left the country. Similar to frogs leaping, Multinational enterprises can invest abroad under pro-investment conditions. Investors had three options when conditions are harsh: to return to the home country, invest in a third country or hibernate. Conditions must be favourable for frogs to leap in a particular direction. If frogs do not leap in a particular direction, such as Rwanda, it means that conditions are harsh and communities suffer. As Domestic Investment started to increase, Rwanda experienced increasing economic growth, job creation and poverty reduction. However, the Domestic Investment economic importance review indicated that in some countries, the effects of Domestic Investment are positive while in others they are negative. Therefore, there is a need to measure the impact of FDI on Rwanda’s economic growth, employment and poverty reduction. However, before measuring the impact of Domestic Investment on these dependent variables, it is necessary to understand how best to measure them.

CHAPTER FIVE:

MODELLING THE IMPACT OF EXPORTS AND IMPORTS ON

ECONOMIC GROWTH IN RWANDA

5.1 Introduction

In chapter four, the study explained the manner in which a nation attains high levels of economic growth. The chapter started by modelling economic growth, followed by Domestic Investment. It finally reviewed the relationship between economic growths.

This chapter establishes the media through which the openness influence economic growth in a nation such as Rwanda. It outlines the approaches adopted to measure these explanatory variables. The other unique variables under openness variables include human capital, telecommunications, GE, inflation and civil war. Later, this chapter explains functional elasticity as a means of deepening our understanding regarding factor productivity, explained by the Solow-Swan Model. The final part of this chapter provides a literature review on the impact of FDI on economic growth, employment and poverty.

5.2 Modeling Openness on Economic Growth

Openness refers to the extent to which a nation opens to the flow of goods and services traded internationally, including the flow of international investment. Openness entails the adoption of trade liberalization policies, where barriers to trade and investment are reduced (International Chamber of Commerce, 2013; WTO 1995, 2006). The main link between openness and economic growth, employment and poverty reduction is the role of trade and investment in a nation. This is because openness relates to economic growth through production and comparative advantage (Babula & Andersen, 2008; Carbaugh, 2004). Through openness, a nation’s factor inputs (such as labour and capital) are put into the production system, and cross-border trade increases. As international trade increases, so does welfare, following Ricardo’s theory of comparative advantage and the Heckscher-Ohlin theory of different factor endowments (Babula & Andersen 2008). Accordingly, when nations engage in trade, each nation can specialize in the production of goods to its advantage as here below.

5.3 MeasuringOpenness

To measure openness, the extent to which a nation employs trade restrictions/distortions can be employed. Due to a decline in tariffs, a nation’s average tariff rate can be employed as a measure of openness. Similar measures to tariffs are the extent to which non-tariff barriers restrict trade and investment. However, adopting such measures does not indicate the growth of trade in a nation (Dollar & Kray,, 2002). Trade and investment play a significant role in accelerating economic growth, employment and reducing poverty. Due to such shortcomings, to measure openness, the ratio of total trade (exports and imports) to GDP is most commonly adopted (Barro, 2003; Wigeborn, 2010).

The measure is expressed:

OP_{it =},

Where: OP = Open ness; TT = Total Trade (Exeorts + Imports); t = Time

The objective of openness policy is to increase trade, especially exports and investment in the form of import substitution, in developing countries. The openness index is a plot that can be made to indicate the trend by comparing tourism, as an export and FDI as capital for investment.

5.3.1 Modelling the Impact of Tourism on Economic Growth

Tourism’ refers to the activities of persons travelling to and staying in places outside of their usual environment, for a period not more than one year, mainly for leisure, business and other purposes not related to exercise (National Tourism Act 2011). To model the impact of tourism on economic growth in a nation such as Rwanda is based on the tourism value chain, outlined in Chapter Two, which indicates the role of inbound tourists’ demand for goods and services in the country of destination. Specifically, modelling the impact of tourism can start by demonstrating the tourism system framework based on the Leiper Model, illustrated below.

Figure40: The tourism system: The Leiper Model

Figures are not included in the reading sample.

Source: Based on Candela and Figini (2012)

The Leiper Model indicates that the tourism system is comprised of three main sectors. The first sector of tourism is comprised of tourists, constituting the main economic element for the tourist industry. The second is the tourism space, consisting of geographical regions specified as the generating and destination region, as well as transit routes and the tourism sites that are visited. The third section is comprised of entrepreneurs and multi-national industries who provide services that promote tourism. When tourists start travelling, the economic importance begins to emerge, in both the generating country and the destination nation, through tourist demand for goods and services. Therefore, the tourism model adopted in this study indicates that Rwanda is considered as a nation faced with a downward sloping demand curve. Similar to household demand, the tourist demand for goods and services is obtained by maximizing the utility of a tourist subject to budget constraint. Based on the value chain, this study introduces a tourism model summarized in below;

Figure41: Modelling the impact of tourism on economic growth

Figures are not included in the reading sample.

Arising from this relationship, the demand function for tourism exports can be specified:

ToDGS_gc. P _gc = þTD_gc. ToEXP _REV (5. 3. 1)

Where:TDGS_g_c= Tourist quantity demanded for goods and services (gs); P _gc = Aggregated level of price (P) for a nation’s goods and services;

βTDgc= the share of tourist demand for nation’s goods and services; ToEXP _REV = Tourist expenditure as revenue in the country of destination

Following Equation 5.3.1, tourists’ expenditure is the total revenue to government and income to firms and households. Such expenditure includes tourists’ visa entry fees, VAT on goods and services consumed in form of leisure and hospitality services, and entry permits at tourist sites, expressed:

ToEXP _REV = ψ. TToA _C (5. 3. 2)

Where:ψ=Touristpercapitaconsumption;TToA_C=Totaltourist Arrivals in country of destination.

In the ASSM, money spent by tourists is a foreign capital flow into Rwanda. In Solow’s growth framework, the Cobb-Douglas production function can be employed to explain the media through which tourism can lead to economic growth (Tang & Tan, 2015). The equation can be specified as follows:

=β ₀ + θlnZ+ lnS _t - ln(n + g + ð) (5.3.3)

Where:Y=Output;L=Labour;s=savings;n=Population growth rate;ð=Rateof depreciation of stock; g = Growth rate of technical progress; z = vectors affecting the level of technology and efficiency in the economy

Equation 5.3.3 indicates that the benefits of tourism in Rwanda are subject to variable z, representing tourism expansion, innovation and institutional factors such as political stability. Ihalanayake (2007) explains that tourists’ expenditure is a source of tax revenue to a nation through tourism products such as: transportation, food and beverages, accommodation, shopping products that include arts and crafts. As expenditure increases, so does tax revenue collected from interactional visitors. In turn as consumption and investment increases as does economic growth and employment. Therefore, tourists’ expenditure can also be viewed as foreign flow, which in turn is income into the country that supplements the private-sector financing gap. This is indicated in Figure 5.4, which reflects the relationship between tourism and investment. First, tourism is a foreign exchange-earning commodity (private sector and government), and as such, an export. Second, as a supplement for the private-sector savings gap, tourism expenditure and FDI are foreign flows. Third, due to the limited capacity of local firms to invest in huge tourism demand projects (such as hotels and game park reserves); tourism is a source of FDI. Similarly, as investment increases, so does tourism and, in turn, economic growth, employment and poverty reduction. Modelling tourism can further be explained by the benefits of FDI as a foreign capital flow. Scale, essentiality and positive and diminishing returns to factor inputs. Second, a nation’s absorption capacity is related to household income and the arising demand for goods and services. After modelling FDI, the next step is to explain human capital as a factor input in the ASSM.

5.3.2 Modelling the Impact of Telecommunications on Economic Growth

As mentioned in Chapter 2, the development of infrastructure and technology in Rwanda is a growing trend, especially through ICT, indicated by mobile and Internet usage in the country. The contribution of ICT on economic growth, employment and poverty reduction are reflected in telephone usage. As indicated in Figure 6.5 below, telephone use plays a large role in production, trade and marketing telephones via M-banking and telephone connectivity.

Figure42: Comparative advantage: Gains from trade and investment

Figures are not included in the reading sample.

The importance telephone usage in a developing country such as Rwanda is explained through: productivity which enhances information asymmetry and transport substitution, entrepreneurial development, and welfare enhancement and poverty reduction (Bhatia et al. 2008). In Rwanda, farmers and traders use telephones to market products, make purchases and social connections.

5.3.3 The Impact of Telecommunications onProductivity

The equation indicates the role of efficiency through technology offered by several types of machinery, including vehicles, forklifts, computers and telephone sets. Following Romer, P (1990), the production function indicating final output can be specified:

Y(t) = L ^α Hβʃ^œx^1–^α ^–β α, β > 0 (5. 6. 1)

Used in production of output (Y)

In terms of ASSM, the production function for Equation 5.2.13 can be rewritten:

Y _t = AK_t ^α (ℎL)_t^1– ^α (5. 6. 2)

Equations 5.6.1 and 5.6.2 indicate that a specialist’s productivity increases through using technology (A). The efficiency of specialist (z) can be specified:ðH^z. The contribution of technology to all specialists employed in the organization is denoted:

A = ðH_AA; where: ð =Productivity parameter (5. 6. 3)

Thus, since specialized skills and technology can be reflected as patents, the marginal product from specialist designs and knowledge is specified:

P_AA = ðP_AH_AA (5. 6. 4)

Where: P _A = Patent

Therefore, the marginal product for each specialist can be denoted:

H _A = ðP_AA (5. 6. 5)

In a developing nation such as Rwanda, ICT though mobile telephone usage has greatly improved productivity, particularly through the various forms of usage, ranging from verbal communication to Internet via iPhone. In turn, ICT through mobile communication is a source of entrepreneurial development. Telephones lead to economic growth, job creation and poverty reduction.

5.3.4 Modelling Government Expenditure on Economic Growth

The effect of GE on economic growth and employment can be identified by considering the role of a nation’s government in regard to the provision of social services. A government’s expenditure covers a wide range of sectors, including social services (such as education and health) and infrastructure development (including roads and railways). These services are reflected in government purchases and payment of wages to public servants. GE can be viewed from two perspectives (Alshahrani & Alsadiq, 2014; Barro, Robert Joseph & Martin 2004; Evans, 2004). First, GE can be considered a public good, and as such, free commodity. Second, GE can be considered an investment, so physical capital can then be regarded as private, and thus not a free good.

Considering GE as investment, developing countries have been encouraged by donor agencies (such as the Organization of Economic Cooperation and Development [OECD] and IMF) to widen their respective national tax revenue (Cottarelli, 2011; Mascagni, Moore & McCluskey, 2014). This is because a broad tax base can lead to increased tax revenue, a major source of GE, and as tax revenue increases, a nation’s budget is enhanced. Further, a nation reduces foreign aid dependency. In the long term, developing countries can improve infrastructure, service delivery and undertake capital projects. These measures increase GDP, create employment and reduce poverty through increased production and productivity. However, Vlieghere and Vreymans (2006) indicate that as the tax base increases beyond the optimal level, tax revenue begins to decline. This is explained by the Laffer curve, below.

Figure43: Laffer curve tax revenue

Figures are not included in the reading sample.

Source: Based on Vlieghere and Vreymans 2006

Following the Laffer curve demonstration, the following observations can be made.

First, with a narrow tax base, tax revenue is low. With government intervention to widen the tax base, revenue gradually increases from zero. In turn, receipts increase until the optimal point. Second, as the tax base increases, resistance starts to emerge among taxpayers, due to the arge tax burden. The tax regime is characterized by tax evasion, fraud and corruption. Finally, beyond the optimal level, tax revenue begins to decline to zero as the tax base approaches 100%. The Armey Curve builds on the foundations of the Laffer curve by indicating the effects of the level of government interference on economic growth. The Armey Curve was proposed in 1995 by Dick Armey, and similar to the Laffer curve, indicates a relationship between the level of government interference and the optimal level of economic growth (Olaleye et al., 2014; Vlieghere & Vreymans, 2006). The Armey Curve hypothesis indicates that a nation with no functioning government is in a state of anarchy. During anarchy, a nation experiences a low state of economic growth and public expenditure, requiring government intervention. As Vlieghere and Vreymans (2006), page 6 states:

Armey argues that non-existence of government causes a state of anarchy and low levels of wealth creation because of the absence of the rule of law and protection of property rights. The absence of rule of law and continuous threat of theft or expropriation has demotivating effects. Also, the total lack of collective infrastructure leads to poor productivity and consequently low levels of wealth creation. Similarly when all input and output decisions are the hands of the authorities, wealth creation is also low.This statement indicates that the relationship between public spending and growth rate yields a U-shaped curve, similar the Laffer curve, as demonstrated below.

Where there is no functioning state, public expenditure is low, and the nation’s growth rate is also low, due to anarchy. As government intervenes, GE increases, leading to increased infrastructure and social service delivery. In turn, productivity and output increase, as does economic growth. However, beyond the optimal level, there is little long-run incentive for the government to intervene in the economy. Before the emergence of the Armey Curve hypothesis, Barro (1990) indicated a relationship between tax revenue, GE and economic growth. Tax revenue is the GE–output ratio, denoted:

T = (5. 7. 1)

WMere: T = Tax revenue; GE = governnent exeenditure; y = Outeut.

Barro found that the ratio of GE to real GDP (GE/y) was negative. GE did not have a direct effect on private productivity, but rather, savings and growth rates were low. The negative contribution was attributed to the distorting effects from taxation or GE programmes (Barro 1990). Some observations can be therefore made. First, government intervention in the economy is necessary, but only to the optimal level. Second, the private sector plays a significant role in the economy. Measuring the impact of GE on a nation, a model can be constructed, indicating a nation with two sources. In this study, FDI can be considered private investment, while GE public investment. Following Alshahrani and Alsadiq (2014) and Ram (1986), the impact of GE on economic growth is based on two equations. The first equation indicates the private-sector function as follows:

P = P(L _P + K _P + G) (5. 7. 2)

Where: P = private sector; L = Labour; K = Capital; G = Government sector

The second equation indicates the government sector function as follows:

G = G(L _g + K_g) (5. 7. 3)

The subscripts indicate the sectors. Total inputs of the two sectors can be expressed as a nation of two inputs:

L _P + L _g = L (5. 7. 4)

K _P + K _g = K (5. 7. 5)

Following the above equations, output is expressed as total output of two sectors, private and public:

Y = P + G (5. 7. 6)

Following Equation 5.7.6, since GE can be considered as physical capital, its impact on economic growth, employment and poverty is measured based on the ASSM.

5.4 Modelling Inflation on Economic Growth

Inflation can be defined as the average price level increase as an economy increases over time (Stanford, 2008). Inflation refers to an annual persistent increase in the general level of prices of goods and services. However, a nation may experience a deflation situation when the overall average level of prices declines over time. In extreme cases, a nation may experience hyperinflation when commodity prices rise rapidly over time, such that inflation reaches 100 percent or more per year. Often, nations experience hyperinflation due to economic or political breakdown, as in Rwanda during the 1990s. Inflation affects developing countries in two main ways. First, output affects economic growth through production of goods and services. Second, inflation affects a nation through consumption, by the price of consumer goods and services, and factor inputs. These two broad sources of inflation establish the platform for modelling the impact of inflation on a nation, based on monetarist and neoclassical theories.

The Monetarist Theory, also called the quantity theory, is presented as the theory of the demand for money (Brunner & Meltzer, 1972). Although the Monetarist Theory explains the role of inflation on economic growth, employment and poverty, it has some shortcomings.

It explains more the need for government role in the economy than the role of macroeconomic variables in a nation (Espinosa-Vega & Russell, 1997; Palley, 2014). Macroeconomic variables—such as interest rates, wages and technology—play a role in economic growth due to inflation. The neoclassical theory was developed on the foundations of the Monetarist Theory, to explain the impact of inflation on a nation (Palley 2014). The theory explains the relationship between inflation and macroeconomic variables, which in turn affect economic growth, Since this study is concerned with inflation and macroeconomic variables, the neoclassical theory is employed in modelling the impact of inflation on economic growth, in Rwanda.

5.4.1 The Neoclassical Theory and the Impact of Inflation on Economic Growth

The neoclassical theory was based on the earlier foundations of monetary theory. This study begins by modelling the impact of inflation on a nation, based on theoretical monetarism through the Fisher equation of exchange (Friedman 1970; Meltzer, AH 1976; Palley 2014). According to theoretical monetarism, the quantity equation indicated that aggregate spending money velocity (MV) is equal to nominal output (py), and is also equivalent to real output. This relationship sets the basic foundation for neoclassical theory, expressed as follows:

MV = Y = Py (5. 8)

Where:M=Quantity of money;V=Velocity of money;Y=Nominal GDP;y=RealGDP

Following the Production Function Equation (5.2.2), output depends on technology, capital and labour. Based on the equation, Gylfason and Herbertsson (2001) have indicated that inflation affects economic growth, employment and poverty through other variables that affect money and price. The augmented production function can be specified to indicate the relationship among the variables:

M þ

Y = AL ^α ( ) ^β K^1–^α ^–β

(5. 8. 1)

Where:M=money supply;P=GeneralPricelevel;A=Technology efficiency ; L = Labour

α, β, and 1 − α − β = Output elasticity in respect of labour, real balances, and capital

The causes of inflation have been identified as international, fiscal and monetary; food and transport; and cost and demand factors. The main cause of inflation in developing countries is monetary expansion, related to seigniorage, which is defined as the ability of a government to print money (Quartey, 2010). This is because developing countries are characterized by low tax revenue, yet there is the need for service delivery and infrastructure development. Accordingly, government income sources become seignioragefor balancing the budget to finance government subsidies and poverty alleviation schemes. However, as GE deficit increases relative to gross national product, so does inflation equivalent to the seigniorage rate. In turn, when money expansion exceeds the equilibrium, a nation starts to experience a spiral of effects, due to the need to finance government programs.

5.4.2 MeasuringInflation

A number of approaches can be employed to measure inflation, including the consumer surplus and equivalent variation, as well as CPI and the Fisher Index. The consumer surplus and equivalent variation are suitable measures of welfare when examining the impact of a tariff on a nation. In Rwanda, NISR employs CPI as a two-stage approach using Laspeyres Price Index (LI). CPI reflects the percentage change in the cost to consumers of acquiringgoods and services. Also in Rwanda, NISR employs the Fisher Index as a combination of the LI and Paasche Index, to report price statistics in four categories: headline, core, and energy and food inflation.

According to NISR, headline inflation is reported on overall items for price changes in the consumption of goods prone to price volatility due to unpredictable/irregular factors. Meanwhile, core inflation (underlying inflation) is reported on all items, excluding food crops, fuel, electricity and metered water. Due to commodity sensitivities, NISR reports food and energy inflation separately.

This section is concerned with modelling the impact of inflation on economic growth, and examines the impact of inflation on the welfare and wellbeing of poor communities.

Accordingly, this study employs CPI annual statistical data, published by NISR, as a proxy for inflation, because CPI as a measure indicates the impact of inflation on welfare. The next step is to explain the manner in which CPI is measured.

5.4.2.1 The Consumer Price Index Measurement for Inflation

CPI is the most commonly used approach by NISR for measuring the impact of inflation on the cost of living. The CPI attempts to measure the average income required to purchase goods due to inflation change. In this respect, CPI indicates that the average prices of goods and services purchased by measuring the overall average price that can enable a household to purchase a basket of goods. CPI reflects headline inflation, as the measure of the relative changes in the price of all goods and services.

5.6 Modeling Exports trade and Economic Growth

5.6.1 Background to Exports and economic growth

The role of foreign trade is more important in economic structure of third world countries. These countries have a great need for both export incomes and import of all kinds of products and materials that are needed for foundation of industrial structure. There are many other reasons based on which it can be said that inattention to the role of foreign trade faces the models and definitions of growth with defects and problems. (Nasser Ebrahim,2017).

In the second half of the twentieth century we witnessed a considerable increase in the growth of theories and studies on the relationship of politics-commerce and economic growth. Despite disagreements on the direction of their relationship, much evidence shows that commerce improves economic growth. Some scientists tried to show in their studies that the reason of economic growth is export. While some other tried to show the theory that the reason of export is growth. Another group believe that import is the main director of the process of economic growth. Different frameworks proposed some theories for explanation of differential rate of economic growth for similar nations. Open economic models are rooted in endogenous growth models (Romer, 1996; Lucus, 1988) and focused on determining factors of growth (Brro et al., 1995; Roubini et al., 1995).

Despite increasing concern about injustice in income distribution, advanced emerging economies have benefited from external effects of open trade policy (Bhagwati et al., 2002; Wacziarg, 2003; Spannu, 2003; Harrison, 1996). According to the findings of researchers, 85% of transportation devices and machinery needed by emerging economies are imported from developed world and this has an important role in their economic growth (Grossman and Helpman, 1991; Rivera-Batiz and Romer, 1993). Edwards (1992) studied the relationship of trade orientation and trade diversion in endogenous model and concluded that “open economies grow faster because they are able to invest on provision and use of imported machinery which are cheaper” (Mazumdar, 2001).

According to Delong and Summers theory, investment on machinery increases growth. Foreign imported machinery are more efficient than their similar domestic ones (Quah and Helpman, 1995; Krishna et al., 2003; Mazumdar, 2001). “Therefore, the research experience and the total material collected earlier shows that the total studies carried out make a theoretical background for believing the positive exports of export trade on economic growth. These theories particularly support the consumer goods exported of industry that are commonly referred to ‘Investment Goods’ (Lee et al., 1995. p. 92).” Krueger states that “a increase in export of investment goods leads to a increase in growth rate and increase in employment and output.”

With regard to the forgoing it is clear that despite relative acceptance of a general existence of a positive and effective relationship between imports and economic growth, no exact response and estimation has been achieved for the type and direction of the relationship between imports and economic growth and in order to find a more appropriate response it is better to analyze and model the long-run relationship of imports and economic growth. Thus, insufficiency of research on existence and direction of cointegration relationship between import variables and economic growth variables is evident and this study intents to analyze the relationship of imports and economic growth from this point of view.

5.6.2 Theoretical underpinnings of imports and Economic growth

Many scientists believe that international trade policies can increase income growth rates. Recently, many researchers designed different patterns for open economy that show foreign goods import have a significant importance in determination of the relationship of trade and growth.

Grossman and Helpman (1991), Rivera-Batiz (1990), Romer (1990) and Quah and Rauch (1990) showed that international trade can speed up the growth rate by providing a wide range of intermediate inputs which facilitate further research and development via learning-by-doing.

According to Summers and Heston data of national and dollar price indices for total goods group, 60 countries were categorized into six groups according to their per capital income. The data used in Table 1 shows the ratio of total price to international price and the difference between domestic price and international price in each general goods group and country. It is worth mentioning that the criterion for categorization of each six income groups is gross domestic product (GDP) of the United States, so that the countries whose GDP is <10% of the GDP of the United States are categorized as low income countries and the countries whose GDP is more than 75% of the GDP of the United States are categorized as rich and high income countries. For example, the national prices of domestic investment goods in the first group, which comprises low income countries, are relatively expensive and are compared, showing figure 155, with the international price which is the price of the United States.

The data of first row show that the price of consumer goods is decreased with income. In contrary the third row shows that the price of investment goods with respect to consumer goods is dramatically decreased with per capital income. Therefore, the relative price of investment goods with respect to consumer goods is higher in low income countries. However, price indices are limited for showing the relative price trend of investment goods over time. For this reason it is not possible to cover a wide range of countries especially for coming years. For the analysis of this problem that the negative relationship of relative price of investment goods and per capita income continues over time, the relative price of adjusted investment to adjusted consumption are stated in the four groups of countries which are categorized with per capita income of 1960.

5.6.3 Empirical Review on imports and Economic growth

Sameti et al. (2004) studied “The Effects of Globalization on Iranian Importing Patterns” in an article (1959-2002). They based their studies based on two assumptions: (A) The traditional pattern of exogenous relative price is not a proper pattern for estimation of the function of import demand due to import restrictions. (B) Globalization has a positive impact on the function of Iranian import demand and tested their theories using econometric models. The findings of this study show that the variable coefficient of import relative price in the import demand model is not a significant linear form. According to the findings of this estimation, it seems that the globalization process and global economic integration makes an increase in Iranian imports.

Mahmoodzadeh and Mohseni (2006) analyzed the long-run and short-run impacts of imported technologies on Iranian economy during 1959-2003 in their article entitled “An Analysis of the Impact of Imported Technology on Iranian Economic Growth” using Johansen’s cointegration methodology (1998) and vector error correction model (ECM). They state in their article that import of proper technology is the basis of industrial and social evolution for transferring from traditional production to industrial production and moving through the stages of economic development. They further used forecast error variance decomposition and impulse response function to analyze dynamics and concluded that there is no cause and effect relationship between intermediate import and non-oil GDP in the short-run but there is weak cause and effect relationship between investment import and GDP. The impact of imported intermediate and investment inputs for 5% in the long run is considerable.

Tehranchian (2006) in an article entitled “Impact of Imports on Iranian Economic Growth” studied the impact of import of investment, intermediate and consumer goods on Iranian economic growth during the years 1973-2006. He used Rati Ram’s model for the analysis of the impact of import types on economic growth and collected his data on a library-based research. By analyzing the country’s import trend, Tehranchian showed in his article that despite an increase in the import of said three groups of goods, particularly after implementation of development plans, the composition of imported goods is changed to the benefit of intermediate and investment goods. Further, according to the proposed econometric model, the coefficients of economic growth trend is estimated to be 0.06 compared to investment and intermediate goods and 0.22 compared to consumer goods. This indicates the direct impact of import of investment and intermediate goods and the indirect and decreasing impact of import of consumer goods on Iranian economic growth index.

Dadgar and Nazari (2010) analyzed in an article entitled “Analysis of Import Demand Function in Iran” the function of import demand using vector autoregression approach for 1974-2007. According to the findings of this study, the impact of non-oil GDP and oil incomes is positive on import but the impact of relative price is negative on import.

Bastani and Njafian (2011) analyzed the country’s import in past plans in an article entitled “An Analysis of Import Function and Forecast” and explained the great objectives and proper approaches, policies and strategies for achieving the determined goal. The two researchers are trying to find a proper import composition for Iran in order to make a positive impact on Iranian economic growth. They analyzed the current status of import and the fifth development plan and believe that the current composition of the type of goods that are imported to the country is not proper for development and does not meet the requirements of Iranian economic forecast and for the purpose of fast economic growth not only the import volume shall be increased, but the imported goods composition shall be changed to the benefit of investment and intermediate goods, so that imports will have more impact on production. However, they also pointed out in their study the problem of smuggling and the damage it causes to the role of import in economic growth.

At the same time, Esfahani (2012) analyzed the impact of commerce on economic growth (1960-86) using the statistics and figures of 13 industrial countries and a system of equations and showed that import of intermediate goods has had a positive and significant impact on the growth of those countries and import restrictions on such goods will have a negative impact on their economic growth.

Rodrigues (2010) studied the import experience in Latin America during the years 1950 to 1980 in a study entitled “Import Substitution and Economic Growth.” The importing policy had been restricted in that period. In this study, the time series data are analyzed in the Heckscher–Ohlin model considering the ratio of advantage to capital scale. At the beginning of this period the economy is an open economy and the statistics show that manufacturing of user products are confirmed. Further, trade policy is modeled according to a movement towards closed economy. The model which is designed in this article is completely suitable for the experience of Latin America and is more suitable and efficient for small countries which do not experience a long-run income growth but experience a capital growth.

Kogid et al. (2011) studied the role of import in economic growth of Malaysia in an article entitled “Does Import Affect Economic Growth in Malaysia?” They used the systemic cointegration method and the causality test based on Engle-Granger two-step method, Yvanson method, and Toda-Yomada method of Granger for the analysis of the relationship of these two variables in the time series of 1970-2007. The findings show that there is no correlation between imports and economic growth. The findings also show that imports affect economic growth indirectly but economic growth affects imports directly.

Priede (2012) studied the relationship of increased import volume and regional GDP per capita in a paper entitled “Import Impact of Economic Growth on Regional Economies” over 10 years period between 1995 and 2005 in several European countries. The researcher believes that increased import area and increased import volume are usually reducing factors of income. However, contrary to public belief, the results of this study showed that these two factors had a positive impact on the increase of regional income. Thus, the researcher does not recommend any import substitution.

Owen et al. (2012) analyses the economic data of the United States in 2009 in an analytical article entitled “Imports and Economic Growth.” They first focus on a 7.5% increase in GDP in the fourth quarter of 2009 and try to determine the contribution of imports in different parts of the United States economy and Commerce Department by data analysis. Slow consumption of inventory alone added a 3.7% points to GDP growth rate and increased export personal consumption expenditure, and business and residential investment together added 3.4% points to GDP growth rate. In contract to these two factors, increased import decreased the GDP growth rate by 1.4%.

In an article entitled “Impact of Imports, Exports, and Foreign Direct Investment on the GDP Growth,” Atif (2012) studied the same issues for the period 1980-2009 in Pakistan. He points out in his article that the GDP growth is an economic growth index as a dependent variable. In this study scatterplot matrices are used to analyze the relationship of variables. His findings show that as it was expected, the coefficients of all four statistical coefficients are significantly positive. The impact of foreign investment on economic growth of Pakistan had been low and insignificant. This shows that there has not been sufficient policies in this regard for benefiting from foreign investment. However, this has not been considered a problem in the period under question. Exports showed a significant impact on the increase of economic growth. Also, import of different kinds of services and goods showed a significant impact on increase of economic growth.

CHAPTER SIX:

THEORETICAL FRAMEWORK AND EMPIRICAL ANALYSIS

6.1 Introduction

Chapter Five discussed the theories underlying investment as well as economic growth.The chapter explored the theories that explain domestic investment policy, from the International Trade Theory until Hymer, which attempted to explain domestic investment policy for the first time. The chapter further explored the theoretical impacts of domestic investment policy on a nation. Earlier, Chapter One indicated that some studies have attempted to establish the impact of domestic investment policy on Rwanda. The first of such studies was conducted by Obwona (1996, 1998, 2001), and later Kiza (2007). Since then, the literature has indicated that few to no economic analyses have explored the subject of in this study. Previous studies adopted Ordinary Least Square (OLS) and 2SLS regression approaches for model estimation. However, these studies contain estimation flaws as explained in Section 1.1. Kiza (2007) found a positive significant relationship between domestic investment policy and economic growth, while in all Obwona’s results, found a positive insignificant relationship. Moreover, these studies did not examine the impact of domestic investment policy, which are major problems that developing countries are attempting to solve.

In this study, a multi-equation systems approach has been adopted, similar to the approaches of Wei, H (2010) and Ford, Sen and Wei (2010) on FDI and China’s economic growth. A multi-equation model captures the interrelationships between the independent and dependent variables, simplifying testing models with multiple dependent variables. Accordingly, this chapter establishes the method employed to find answers to the main question of this study: what is the impact of investment decisions on economic growth in Rwanda? This chapter is divided into three parts. The first part is the description of variables while the second explains the conceptual framework. The third part presents the procedure for empirical analysis explaining how this study measures the impact of FDI and other explanatory variables on economic growth, employment and poverty.

6.2 Description of theVariables

This section contains two sub-sections through which the variables of this study are described. The first sub-section explains the scope and sources of data. The second sub- section provides the definitions for each variable employed in this study, and the related measurement units. The measurement units assist this study to provide an account for dependent variables and the associated independent variables through a cause-effect relationship. Through estimations, the impact of investment policy (exports, imports& FDI) and other explanatory variables on Rwanda’s economic growth is explained.

6.2.1 The Scope and Sources ofdata

Data on Rwanda’s annual Panel Data endogenous and exogenous variables for the period 1997–2018 has been collected. The sources of data are, World bank-(WDI) Databases and NISR databases

6.2.2 Variable Definitions andMeasurement

a) Economic Growth

GDP: Gross Domestic Product (GDP) at purchaser's prices is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Data are in constant 2000 U.S. dollars. Dollar figures for GDP are converted from domestic currencies using 2000 official exchange rates. For a few countries where the official exchange rate does not reflect the rate effectively applied to actual foreign exchange transactions, an alternative conversion factor is used.

This is the real increase in Rwanda’s GDP per annum as expressed as the logarithm of Rwanda’s annual GDP panel data is the proxy for economic growth expressed:

ln (GDPgr) = ln (GDP_it) – ln (GDP_it–1)

The logarithm of a number indicates the number of times such a number has been multiplied or changed. In this study, economic growth is denoted by the logarithms of GDP.

The logarithmic approach allows for measuring the impact of growth of series, such as FDI and other explanatory variables on economic growth.

Source: World Bank national accounts data, and OECD National Accounts data file

Exports

The measure of FDI is the total earnings received by Rwanda in terms of foreign exchange revenue, in US dollars, as reflected in the balance of payments. The export to GDP ratio is employed as a proxy for export

Imports

The study defined imports of goods and services (constant US$) and Fixed Formation Capital (constant US$)

Openness (OP)

The measure of FDI is the total FDI inflows into Rwanda, in US dollars, as reflected in the balance of payments. The FDI to GDP ratio is employed as a proxy for FDI.

Openness (OP) refers to the removal or reduction in trade restrictions/barriers that affect the flow of internationally traded goods and services in a country. To measure openness, the openness index was used. This is calculated as the proportion of total trade (TT) measured by imports plus exports to GDP.

OP _{it =}

A high index reflects the higher level of influence of international trade on Rwanda.

Inflation

Data on annual CPI obtained from NISR and WDI database is used. This is because CPI on all commodities and services has a direct impact on households Wickremasinghe (2011).

Telecommunication

Technology is a source of efficiency in production. Telephone facilities are used in commercial transactions and for social connections between firms and persons. Data published by NISR on the real numbers of fixed telephones and mobile subscribers was employed.

Tourism

The ratio of inbound tourists’ expenditure, measured in US dollars in Rwanda, to exports (EXP) is employed as a proxy for tourism (Tsm: EXP). The source of data for inbound tourists’ expenditure is NISR for the period 1994–2015, and the IMF for 1995–2014

Government expenditure

Government expenditure (GE) refers to development projects and recurrent expenditure, including expenses on personal benefits, salaries and wages for public servants, government subsidies and interests. The GE to GDP ratio is used as a proxy for GE measured US dollars.

Foreign direct investment

The measure of FDI is the total FDI inflows into Rwanda, in US dollars, as reflected in the balance of payments. The FDI to GDP ratio is employed as a proxy for FDI.

6.2 ConceptualFramework

A firm’s optimization production decisions are based on factor inputs. Equation 4.2.12 indicates that through the ASSM, the determinants of economic growth, in an economy like Rwanda can be identified. In Chapter Two, the background of Rwanda’s economy and investment policy were explained. According to the literature review, exports, import and FDI influence economic growth in Rwanda.

The key monetary policy identified was controlling inflation as a basis for macroeconomic stability. Finally, openness was identified as a commercial policy, to promote investment such as exports, imports and FDI.

Finally, the study identifies Rwanda as a developing country that has been through numerous civil wars since Independence.

To explain the association between the explanatory variables (Exports, Imports and FDI) and the dependent variable (economic growth), a path analysis approach is employed. Through path analysis, a simple recursive relationship among variables is indicated.

Figure44: Variables conceptual framework

Figures are not included in the reading sample.

Source: Adopted from Literature reviewed

The conceptual framework explains the theoretical association among variables that are employed in this study. According to Figure 6.1, some variables are endogenously related; for example, employment and economic growth.

6.4 Investigation of Properties of theVariables

Since macroeconomic variables are employed in this study, property investigation enables to examine the cyclical movements and trends of variables on the economy. The study employs the interactions to interpret the relationships between the dependent and explanatory variables. To understand the time series variable’s properties, the following investigations are considered: transforming data into logarithmic form, graphical investigations, correlation and trend analysis, unit root testing and endogeneity testing respectively.

6.4.1 Logarithmic Form Transformation

In economic analysis and forecasting, variables are commonly used in logarithmic form, so this was the first step employed. Logarithmic transformation serves as a tool for stabilizing the variance of the series (Lütkepohl & Xu, 2009; Sehgal, Rajput & Deisting, 2013). Series transformation is a means for achieving homoscedastic and normally distributed residuals (Herrendorf, Rogerson & Valentinyi, 2013; Kulendran, 1996).

This study is based on the Solow-Swan Model, which has taken its foundations from the production function. In economic research, logarithmic functional form coefficients can serve as elasticities of the production function used to analyze microeconomic and macroeconomic issues, such as economic growth and employment. The production function means that the series expressed in logarithmic terms implies that:

If Ki_t grows at the constant rate y_s then the law of motion for capital implies that Xi _t must grow at the same constant rate (Herrendorf & Rogerson, 2013).

The elasticity term is used to describe the degree of response of a change of a dependent variable with respect to the change in an independent variable. A proportional change in logarithm units can be converted to a percentage by multiplying the growth by 100 as follows:

[ln ∗ 100] (6.1)

Therefore, the logarithmic approach allows for measuring the impact of growth in the related series, such as Investment policy (export, imports &FDI).

This is because the logarithm of a number indicates growth in output.

Logarithmically transformed variables in a regression analysis handle situations where a non- linear relationship exists between the independent and dependent variables (Benoit 2011). Accordingly, variables expressed in logarithmic form provide effective non-linear relationships, while still preserving the linear model. Logarithmic transformations are a means of transforming a highly skewed variable into one that is more approximately normal. To this extent, transforming data into logarithmic term statistically is a means of providing preliminary inference on data, and improves the interpretations and graphical display analysis.

6.4.2 Preliminary Variables Investigation

To investigate the relationships among the time series mentioned above, this section employs series transformed into logarithmic form through graphs. This is to take advantage of the benefits of logarithmic expressions. Graphs provide a visual impact and help describe the relationship between two or more sets of data or variables that are related to one another. A graph predicts the functional relationship between two or more economic variables by providing generalizations about the economic phenomena. Graphical analysis is a tool for, assume economic growth = K while investment = X. When the economy of a nation grows from at rate γ then investment grows at the same rate, explaining the manner in which the variables employed are related to Rwanda’s economic growth.. After the graphical investigations, this study undertakes a correlation study of the variables.

6.4.3 Correlation Analysis

When variables are related, there is correlation between them. Correlation analysis is used to measure the degree of a linear association between the variables. Correlation among variables ranges between negative one and positive one (−1 to + 1). In this study, in absolute terms, no correlation means a zero relationship. Second, 0.60 above denotes highly correlated variables. Third, in absolute terms numeric value (1) between two variables is regarded as perfect correlation. In this regard, perfect correlation between the variables means that knowing the value of one variable exactly predicts the value of the other variable. The larger the magnitudes of correlation, the more to variables are perfectly related.

In this study, a correlation coefficient of above 0.60 indicates highly correlated relationship among variables. As a common practice, in economic analysis when two explanatory variables are highly correlated in a single regression analysis, at least one is removed from the study. Therefore, employing correlation coefficients, highly correlated variables have been removed from the study.

6.4.4 Trend Analysis

Due to non-stationarity among the time series, the trend analysis employed in this study is a graphical display to check the trend indicated by the series. Through trend analysis the study can check whether or not the series the fluctuations always come back to the mean Maradiaga (Maradiaga, Pujula & Zapata, 2013).

Trend analysis is necessary before testing for unit root testing, so as to determine whether or not the series are stationary around a constant or a trend that can be included during unit testing.

6.4.5 Unit Root Testing

Regression of time series that requires that mean, variance and covariance are constant for stationary data, such that:

Mean: E(y_t) = µ (6. 2)

Variance: E[(y _t − µ)²] = Var(y_t) = a²(6. 3)

Covariance: y _k = E[(Y _t − µ)(Y _t+k − µ)] (6. 4)

Following the above requirement for time-series data, before estimation this study first, tests for unit roots. Non-stationary indicates that mean, variance and covariance are not constant. Unit root testing enables the study to avoid spurious regressions. In economic analysis, spurious regression results are invalid and cannot be used for policy analysis. This is because in spurious regressions R-square is inflated and often close to one, while the t − statistics ratios do not follow the t − distribution. Second, this study employs a VAR approach for model specification estimated in two way namely: unrestricted VAR and redistricted VAR commonly referred to as VECM for multivariate analysis.

To determine stationarity the procedure, employed is unit root testing. When series are non- stationary at level but stationary at first difference, the VECM approach is employed; otherwise, unrestricted VAR is suitable. Two approaches can be adopted to test for unit roots: t − ratio test and ADF tests. However, the t − ratio test null possesses non-standard distribution, the critical values for t − statistic are not applicable. Therefore, this study employs the ADF approach developed by Dickey and Fuller (1979).

6.4.6 Augmented Dickey- Fuller Tests

Song and Witt (2000) indicate that the ADF approach obtains critical values based on Monte Carlo simulations. The ADF approach employed is based on three regressions:

∆y _t = ∅ ₀ + yy _t–1 + Σ β _i ∆y _t–i + s _t (With constant) (6. 5)

∆y _t = ∅ ₀ + ∅₂t + yy _t–1 + Σ β _i ∆y _t–i + s_t(With constant & trend) (6. 6)

∆y _t = yy _t–1 + Σ β _i ∆y _t–i + s_t(Without Constant & trend) (6. 7)

Where: ∆y _t = (y_t_– ₁ − y_t_–₂); ∆y _t–2 = (y _t–2 − y_t–3) … … … … … …

The variables are tested at level and first difference, based on the procedure recommend by Enders (1995), a flow chart, as illustrated below.

Figure 6.3: Unit root test procedure employed

Figures are not included in the reading sample.

Source: Enders 1995

Following the procedure illustrated under Figure 6.3, the ADF test employed is based on the null hypothesis that data has a unit root, and as such is non-stationary expressed as follows:

H₀: y = 0 Data y _t has unit root or y _t is non − stationary

H₀: y < 0 Data y _t has no unity root or y _t data is stationary

The null hypothesis of non-stationary series is rejected in favour of the stationary alternative for each test when the ADF test statistic is more than the critical values, and the corresponding probability value is less than 5%. The main problem of the ADF test employed in this study is the choice of lag length (e) while dealing with autocorrelation and heteroscedasticity. This is because using too small a lag length p, serial correlation that remains in the errors can bias the test. If the lag is too large, the strength of the test is affected. Due to these weaknesses, the ADF test can be validated by the Phillips-Perron (PP) test.

6.4.7 The KPSS Tests for Unit Roots

The KPSS test is a Langrange Multiplier (LM)-based test, based on the null hypothesis that data is stationary. It is based on OLS residual regression:

y _it = x′_itð + u_it; (6. 9)

Where: x′_tð = Exogenous tern, either constant

The KPSS can be specified:

V= T ^–2 (6.10)

Where: v = Asymptotic distribution; s _t = Σ e _i ; t = 1. . T; k = Number of lagged

Periods; T = Sample size; e

= Regression coefficients for intercept series; σ

=the estimator for long run variance; t = time period

The null for KPSS is that the data is stationary. The null is rejected when KPSS test statistic is greater than the critical value. After unit root testing, this study conducts endogeneity tests.

6.4.8 Endogeneity Investigation

Endogeneity investigation is used to determine that an explanatory variable x _j is correlated with error u meaning that the variable is endogenous (Wooldridge 2010). In this way, the variable is determined if endogenous within the context of a model. Meanwhile when x _j is uncorrelated with u then the variable is said to be exogenous in equation. Therefore, endogeneity analysis is necessary as a tool for determining whether the variable can be regarded as exogenous or endogenous. This can be tested based on the Pairwise

Granger causality test, as it indicates the extent to which two variables can Granger-cause each other; for example, X can Granger-cause Y and vice versa. In this way, the type of model under study can be identified though either the presence or absence of Granger-cause. Based on Granger 1969, the Granger causality model is specified based on a simple VAR:

Xi _t = Σ α _i Y _t–i + Σ þ _j X _t–j + µ _1t (6. 11)

Yi _t = Σ h _i Y _t–i + Σ ð _j X _i–j + µ _2t (6. 12)

In Equations 6.11 and 6.12 it is assumed that the disturbances µ _1t and µ _2t are uncorrelated. The hypothesis specified where the β indicates the influence of X _t–j on Y _t such that if:

H_O: β ₁ = β ₂ =--------β _n = 0 (X does not Granger cause Y)

H₁: β ₁ ≠ β ₂ ≠ ⋯------- β _n ≠ 0 (X does Granger cause Y).

The null hypothesis explains that the Granger causality test identifies that Y does notGranger-cause X. When the e − value F-Statistic is jointly significant, (P − value < 0.05) the null is rejected.

6.5 Long and Short-RunEstimation

The previous section explained the manner in which the time-series properties employed. This section presents the approaches employed for long and short-run estimations.

6.5.1 Time-Series Cointegration estimation

This section examines the approaches employed to test for the existence of long and short-run relationships among the series.

Dynamic panel data is often non-stationary, implying that data drifts and does not belong to the same system. When data drifts apart it exhibits a stochastic drift, causing a change in the value of the random or stochastic process (Hendry & Juselius, 1999). Consequently, the dependent variable and explanatory variables’ stochastic trends result in spurious regression, due to data non-stationarity. Spurious regression results are invalid for economic policy making.

To establish that the series belong to the same system or stochastic drift, a cointegration test is undertaken. Cointegration in economic analysis means that two variables have a long-tern equilibrium relationship (Chimobi, 2010; Osuala, Osuala & Onyeike, 2013). First, cointegration relationship means that time series variables such as x and y are stationary, do not drift away in the long-run and are integrated to order one [I(1)]. Second, through cointegration, this study investigates if there are long-run relationships among the variables of the model estimated. Third, through cointegration analysis we establish that the vector y _t series contains N endogenous variables, of which all are integrated of the same order.

Regarding the need to test for cointegration among the series, two approaches can be employed: the Engle and Granger two-stage cointegration analysis and Johansen’s Maximum Likelihood Method. The Engle and Granger two-stage cointegration approach is suitable for conducting a test involving two variables, while Johansen’s Method is a multivariate approach. Since this study involves a simultaneous equation model specifications, the cointegration approach employed is based on Johansen’s Method (Kasindi & Mwakanemela, 2013). The Engle and Granger approach suffers from a number of weaknesses. First, it is restricted to a single equation, with one variable designated as the dependent variable, explained by another variable that is assumed to be weakly exogenous for the parameters of interest. Second, it relies on pretesting the time series to find out whether variables are I(0) or I(1). These weaknesses can be addressed through the use of Johansen’s procedure. Its advantages include the fact that pretesting is not necessary. Also Johansen approach allows to estimate more than one co-integration relationship if the data set contains two or more time series as well as gives the maximum rank of co-integration (Kasindi & Mwakanemela, 2013). In this way allowing numerous cointegrating relationships the Johansen procedure treats all variables as endogenous while testing the relating to the long-run parameters. The resulting model is known as a VECM, as it adds error correction features to a multi-factor model known as VAR.

The procedure is performed as follows:

· Step 1: estimate an unrestricted VAR involving potentially non-stationary variables

· Step 2: test for cointegration using Johansen’s test

Step 3: form and analyze the VECM.

Following the aforementioned steps, Johansen’s Method explicitly uses VAR to estimate cointegration or long-run among non-stationary series, as well as capturing the short-run dynamics via VECM. Also, VAR facilitates easy simulation while conducting ex-ante forecasting using impulse response analysis and variance decomposition. Following Hjalmarsson and Österholm (2007), Chinobi (2010), Kasindi and Mwakanemela (2013), Johansen’s Method takes its starting point in VAR of order [AR (e)] expressed:

y _t = µ + A₁y _t–1 + ⋯ + A_qy _t–q + s _t (6. 13)

Where: y _t is an n × 1 vector of variables that are nonintegrated to order One [I(1)]; s_tis an n × 1 vector of innovations

The VAR can be rewritten as follows:

∆y _t = µ + Пy _t–1 + Σ Г_i∆y _t–i + s _t (6.14)

Where: П = Σ A _i − I; Г _i = − Σ A _j ; s_tis an n × 1 vector of innovations

In Equation 6.14, the coefficient of matrix П has a reduced rank r < n. In this way the VAR model contains n × r matrices α and β where each with rank r (denotes the number of integrating relationships) such that;

П = αþ^u; and þ^uy _t is stationary (6. 15)

Where: α = Adjustment parameters in VECM; β = Matrix of cointegrating vector

Additionally, for any given r the maximum likelihood estimator of β explains the combination of Y _t–1 that provides the largest r canonical correlations of ∆y _t with y_i–1.

The Johansen Method employs two different likelihood ratio tests of the canonical correlations to test for significance of cointegrating relationships namely: trace test and Maximum Eigenvalue test.

6.5.2 Trace Statistic

The trace statistic null of r cointegrating relations among the endogenous variables:

J _t_r_a_c _e = −T Σ Log (1 − λ^i) (6.16)

Where: r = 0 to r = n − 1 … until fail to reject Ho such that

0 = No integrating (None) equations while 1, 2 … = 1 or more integrating equations;

k = number of endogenous variables; T = sample size;

λi = i^t ^h largest Eigen value of long run coefficient matrix

Trace statistics are based on the hypothesis until it fails to reject the null:

Hypothesis 1, 2…..:

H₀: Trace stastic < Critical value = Integrating equation

H₁: Trace stastic > Critical value = At least integrating equation

6.5.3 Maximum Eigenvalue Statistic

The Maximum Eigenvalue statistic null of r cointegrating relations, based on the equation:

Jmax = −TLog (1 − h^r+1) (6. 17)

Where: r = 0, 1, …, n − 1 until fail to reject H ₀

The Maximum Eigenvalue statistic is based on the hypothesis until it fails to reject the null, specified as follows:

Hypothesis 1, 2…..:

H₀: Eigenvalue < Critical value = No Integrating equation

H₁: Eigenvalue > Critical value = At least integrating equation

Before conducting the simulations, this study determines the lag lengths that are employed.

6.5.4 Lag Lengths Selection Criteria

Although the VAR model is widely used in forecasting and model estimation, the determination of the lag length is necessary before simulations. This is because in VAR, all variables are treated as endogenous. Endogenous variables are treated as a function of the lagged values of all endogenous variables within the system. When the lag length is inconsistent with the true lag length, the simulated results are invalid for policy analysis (Braun & Mittnik, 1993). This is because first, incorrect lag length criteria yield wrong impulse response functions and variance decomposition simulations. Second, Lütkepohl (1993) indicated that over-fitting (selecting a higher order lag length than the true lag length) leads to increased VAR mean-square forecast errors, while under-fitting the lag length often generates auto correlated errors. Third, Hafer and Sheehan (1989) indicate that the accuracy of forecasts from VAR models varies substantially for alternative lag lengths. To determine the lag length, the unrestricted VAR lag order selection criteria is employed. In a VAR model t the appropriate maximum lag based on the hypothesis of the Chi-square is that:

H_O: Coefficients on lag l are jointly zero

H₁: Coefficients on lag l are not jointly zero

The appropriate lag length to be included in the model is considered by first comparing the critical values for each criteria at 5%, level for criteria such as Likelihood ratio test (LR) test statistic, Final Prediction Error (FPE), Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC) and Hannan-Quinn (HQ). Using this approach, the best criterion is identified by considering the critical values in a descending order, starting with a maximum lag, to the minimum lag. The determination is guided by the size of the critical values, whereby the smaller the values the better the criteria are. Second, in unrestricted VAR, an asterisk indicates the best lag order. The best lag length is selected considering the lag criteria that is identified by the majority asterisk indicator. This study has employed the AIC method the most widely used method and more efficient compared to others (Acquah, 2010).

After selecting the lag length, this study then conducts the simulations that provide the results of the trace statistics and Maximum Eigenvalue statistics.

These results are used to determine whether or not the series are cointegrated, and also the number of cointegrating equations. Second, the results of the normalized coefficients indicate the nature of the long-run relationship among the series under study.

6.5.5 Estimation of Short-Run Relationship among Endogenous Variables

The Johansen Method establishes cointegrating vectors and long-run relationship, and explicitly uses VAR to investigate short-run dynamics, too. VAR sidesteps the purpose for structural modelling by treating all variables as endogenous. As such, variables are not differentiated under VAR but are considered as endogenous within the system. In this study, some variables are endogenous while others are exogenous, which are determined outside the system; for example, economic growth is endogenous while inflation is exogenous. This study first employs a simultaneous approach following Song & Witts (2000). This is followed by estimating the equation as a VECM. Diagnostic tests are conducted to test the validity of this model. After, short-run simulation is conducted, employing the VECM Granger causality approach. Finally, the study conducts ex-ante forecasting using impulse response and variance decomposition analysis.

6.5.6 Simultaneous EquationSpecification

In a simultaneous system model, Johnston and DiNardo (1997) indicate that the system of equations is stacked in a general form, expressed as:

y _i = X_iβ _i + u_i, such that i = 1, … … , m (6. 18)

Where: y _i is an n х 1 vector observation on the i^th;

X _i is an n х ki Matrix of the observations on the explanatory variables; β _i is a k _i х 1 vector of coefficients; u_iis an n х 1 vector of disturbances

In Equation 6.19, y indicates a set of dependent variables in the simultaneous equation, while the disturbances and explanatory variables for equations are assumed to be uncorrelated. In this way, estimator stacks m equations in a general form, expressed as:

(6.19

Where: y _m is a T Vector; x _m is a T х k _m matrix; β _m is a k _m Matrix; e is a T х MT covariance Matrix V

As linear equations, we estimate single equations expressed as:

Following systems model estimators, the OLS method of model estimation is the one implemented in this study

6.5.7 Structure of the Simultaneous Equation

As a common practice, the systems approach employed consists of multivariate technique that takes into account interdependencies included in the equations, especially among endogenous and exogenous variables. The system can be expressed in general terms as:

ƒ(y_t, x_t, β) = e _t (6. 23)

Where: y_tis a vector of endogenous variables; x_tis a vector of exogenous variables; e_tis a vector of possibly serially correlated disturbances;

β = Vector of parameters under estimation

Using data obtained on each of the variables, this study is comprised of five endogenous variables: economic growth, and investment (exports, imports and FDI). Meanwhile, exogenous variables employed in this study are determined based on theory and Rwanda’s policy. The exogenous variables include inflation, Government expenditure, human capital formation and openness. As earlier mentioned, the pairwise Granger Causality approach is employed in this study top test for endogeneity. This study is also guided by theory to determine endogeneity. Based on theory, in the Solow-Swan Model, human capital is exogenous, while openness is treated as innovation. As Rwanda is a least-developed nation, these variables are driven by policy from government, due to the need to reconstruct the nation in the post genocide era. Therefore, four equations are stacked in the simultaneous equation as a system in which the exogenous variables applied equally to all equations theoretically, specified as:

LnGDP = f (EXP, Impt, FDI ucpi, op, hcap) (6. 24 )

EXP= f (lnGDP, Impt, FDI ucpi, op, hcap,) (6. 25)

IMP = f(lnGDP, Expt, FDI ucpi, op, hcap) (6.26)

FDI = f(lnGDP, Expt , Impt, ucpi, op, hcap) (6. 27)

Where: lnGDP, GDP in logarithmic for as proxy for economic growth;

Proxy for Net export _t = Total Value of exports _t – Total value of imports _t (US $) purchasing power parity

Where t is the specific time frame, typically one year

Proxy for Net import _t = Total Value of imports _t – Total value of exports _t (US $) purchasing power parity.

Where t is the specific time frame, typically one year

FDI is the Proportion of Foreign direct investment to GDP s. Tour proportion of inbound tourists (Tour) to exports; as a proxy for Tourism earnings for Rwanda; Openness (OPn) = Internal proportion of trade (TT) to GDP expressed as ,

as a proxy for openness.

lnucpi = consumption price index, as a proxy for inflation volatility; ln = Logarithmic term

Based on the simultaneous equation established by the theoretical model, the study proceeds to explain the approach used to estimate the simultaneous equation as a means of establishing the long-run and short-run dynamics among the endogenous variables. To achieve this, the study employs a VAR approach through a VECM procedure

6.5.8. Simultaneous Equation Estimation under a VARApproach

The VAR approach is a common tool employed while forecasting systems equations of interrelated time series. The VAR model is based on the general approach proposed by Sargan (1964) and later developed by Davidison et al. (1978), Hendry and Von Ungern- Sternberg (1981) and Mizon and Richard (1986) as an approach for model specification (Song & Witts 2000). Using this approach, the general equation is specified in the form of an Autoregressive Distributed Lag Model (ADLM), where the long-run relationship among the variables can be indicated by specifying the equation as follows:

y_t= α + X + y + ε (6.28)

Where:y_t=Dependentvariable;k=Explanatoryvariables;e=1forannualseriesdata; j = Explanatory variable k effect on the nation; i = Country specificc effects; e _t = Error term (normal distribution such that: ԑ _t ~N(0, a²), t = time

Equation 6.28 explains a linear relationship that first indicates the relationship between the dependent and explanatory variables. Second, Figure 4.3 indicates a steady state.

However, as explained previously, due to the role played by research and design, technological innovation and government, a nation experiences continuous growth. The coefficients in the linear model can be estimated to indicate a long-run relationship among the variables. As such, the linear equations indicate the sensitivity of the changes in explanatory variables to the independent variables. This establishes the basis for this study to employ VECM as a procedure for forecasting and model estimation.

6.5.9 The Theoretical VECM Procedure for Estimating the Simultaneous Equation

Following Equation 7.14, the short and long-run relationship among variables is explained as:

∆Y _t = Σ ∅_i∆Y _t–i + ∅Y _t–p + U _t = VECM (6. 29)

Where:∅_i=−(1−β₁−β₂−⋯β_i);∅=−(1−β₁−β₂−⋯β_p);∅Y_t_–_p=Errorcorrectionterm;∅_iand∅=SℎortandlongrunadjustmentstochangesinY_t

The linear relationship first indicates the relationship between the dependent and explanatory variables. Second, the coefficients in the linear model can be estimated indicating the sensitivity of the explanatory variables’ changes to the independent variables. Following Equation 6.29, for estimation and hypothesis testing, VECM can be expressed as:

∆Y _t = αβ^uY_t_– ₁ Σ Г_i∆Y_t_– _i + ⋯ βX _t + s _t (6. 30)

Where:β= Long run measure among variables; Г _i = coefficient measure for short − run effects of shock on ∆Y_t

As indicated, VECM enables identifying the short and long-run relationship among variables. In this way, forecasting the impact of the variables in the study is facilitated. Following this approach, this study specifies the model and simulations are conducted to indicate the long- run and short-run.

6.5.10 Rationale for Employing the VECM Procedure

As previously explained, cointegration measures the existence of long-run relationship among series and explicitly allows the use of VECM.

In this respect, VECM a type of VAR model is used to reconcile the short-run value with the long-run behavior (value) of the model (Ray, 2012; Suliman & Elian, 2014). This is based on the Granger Theorem which explains that a set of cointegrated time series possess an error correction term. Therefore, as cointegration only captures the long-run relationship, VECM is used to capture the short-run dynamics of the as well.

Since this study is a systems approach comprising of eight variables, VECM is employed as the error correction terms become equivalent to the number of cointegrating relationships. Also, VECM is employed because the series are non-stationary at level but stationary at first difference as a precondition a VAR model. Another approach would have been the Engle and Granger Model, but this is suitable when conducting a test with two variables. In particular, Song and Witt (2000) further indicate a number of attributes that make VECM the best approach for this study. First, application of the VECM approach enables the study to verify that β

(indicated in Equation 6.15) is trend stationary. Second, VECM allows investigation of the long-run relationship among the variables including short-run correction from the variable to the equilibrium. The short-run and long-run effects are all presented in a single model. This indicates that the dependent variable y

depends on explanatory variable changes (∆x

) and the previous period disequilibrium error. In this regard, VECM takes care- of any disequilibrium that may occasionally shock the system. This is made possible due to ability of VECM to pick up such disequilibrium and guide the variables of the system back to equilibrium.

Third VECM takes care of the spurious correlation among time-series (Suliman and Elian 2014). This is because the VECM represents a stationary process as long as (y_t) and (x_t) are cointegrated. This approach is superior to the growth rate model, which employs differentiated data. Impulse response and variance decomposition is calculated to indicate how variables react to the innovations and shocks. Finally, VECM is another form of re- writing and re-enforcing the ADLM. In this way, VECM takes care of the multicollinearity problem that ADLM is likely to suffer when a model includes a large number of explanatory variables. Engle and Granger (1987) indicate that the VECM explanatory variables are orthogonal, which implies almost a zero correlation between variables. Considering these attributes, this study finds that VECM possesses the basic attributes of a model that can produce reliable results for policy analysis. The model is: parsimonious, encompassing, theory based and coherent with data, with constant parameters and the ability to deal with problems such as endogeneity among the explanatory variables.

After explaining the procedure employed to estimate the simultaneous equation employing VAR through the VECM procedure, this section now explains how the procedure is implemented to measure the long-run among vectors and the short-run dynamics.

6.5.11 Procedure for VECM Estimation of Short-Run and Long-Run Relationship

The easiest way to demonstrate how VECM estimates the simultaneous equation is to adopt the Engle and Granger (1987) causality approach. Following Wickremasinghe (2011) the Engle and Granger approach is demonstrated assuming two variables as:

∆x _t = α ₁ + b₁ect _t–1 + Σ c₁∆x _t–i + Σ d₁∆y _t–i + e _1t (6. 31)

∆y _t = α ₂ + b₂ect _t–1 + Σ c₂∆y _t–i + Σ d₂∆x _t–i + s _2t (6. 32

Where: x_t, y _t = Variables; ∆ = operator difference; m, n= variable lag lengths; ect _t = coeintegrating equations residuals; e_1,e ₂ = white noise residuals

Based on the approach illustrated in Equations 6.31 and 6.32, the model is then extended a multivariate system. Accordingly, in the multivariate case, the numbers of equations are equal to the number of variables while the number of error correction terms equals the number of cointegrating relations. The advantage with VECM (the error correction term that is not applicable in the standard Granger causality tests) opens up a new channel through which causality indicates error correction term statistical significance by a separate t-test, which also indicates the short-run. Second, the new channel indicates the lags for each explanatory variable by F-/Wald Chi-square test as a joint significance. Third, the channel indicates the error correction term by joint F-/ Wald Chi-square test. These attributes explain the basis for which this study employs the VECM approach, as a basis for model specification and forecasting.

After fitting the series into the model by employing Eviews, this study then simulates the model. The model simulated is comprised of two parts. The first section is the error correction term, indicating the long-run relationship. The second part indicates the short-run relationship. After model simulation, the next step is to validate the systems model. This is followed by long-run and short-run analysis of the model and ex-ante forecasting.

6.5.12 VECM Model Validation

Model validation is necessary to check that the residuals of the model satisfy the assumption indicated by Equations 6.2–6.4, namely: normality, constant error variance and uncorrelated error terms. As such, to validate the simulated VECM model, the following diagnostic tests are conducted: model stability, correlogram analysis, residual portmanteau tests for autocorrelations, normality tests and residual examination. Due to the limitations VECM validation tests, other tests are conducted after model estimation.

6.5.13 Model Stability

As a common practice during economic analysis, it is necessary to confirm the model adequacy. A stability test is conducted to confirm the suitability of the parameters in the model across all sub-samples of the data employed. This is because time-series data employed in this study are often non-stationary. To avoid invalid results, a stability test is conducted. Stability is tested through the companion matrix of the VECM model with n endogenous variables and r cointegrating equations possessing m − r eigenvalues. The stability is tested based on the inverse roots of the characteristic VAR polynomial by the eigenvalues of the modulus. In a stable model, in arithmetic terms all the roots of companion matrix are less than one and in a graph form all lie inside the circle. The model stability condition is indicated as:

α ₂ + α ₁ < 1; α ₂ − α ₁ < 1, such that α ₂ > −1, α ₂ < 0 (6. 33)

The results of the model employed are indicated as stable and not misspecified, as the general distributions of the entire companion matrix roots lie inside the unit circle and are less than one. This means that the process is stationary and that the model is sufficient for policy analysis.

6.5.14 Correlogram Analysis

Testing serial correlation starts by presenting the easy visual test of constructing correlogram graphs. The model is free from autocorrelation by the manner in which the residuals lie in the graph. A valid model is indicated by the residuals that lie between the standard limits of -1 and 1.

6.5.15 Portmanteau Residual Test for Autocorrelations

In addition to the correlogram graphs, in mathematical terms autocorrelation can be tested by employing the residual portmanteau tests for autocorrelations. These tests are based the Ljung-Box Q-Statistics and the corresponding probability values (Kulendran 1996). The test statistic for the Q-Statistics is reported as Chi-square Q distributions, with a null specified where the Q-Statistics probability values are greater than 5% (P − value > 0.05).

H₀: There is no autocorrelation up to order k;

H₁: There is autocorrelation up to order k

The Q-Statistic test is widely used in economic studies, and the test is built in time-series programs such as Eviews, which are employed in this study.

6.5.16 Residual Normality Test

In economic analysis, the Jarque-Bera (JB) test is employed to check whether the null hypothesis error term is normally distributed. The testable hypothesis is specified as follows:

H_O: Data is normally distributed;

H₁: Data is not normally distributed

When the time-series model error term is normally distributed first, in arithmetic terms the value of the skewness is indicated between −1 > 0 > 1, while the kurtosis is 1 > 3. The JB is given as 1 > 5.99 where the corresponding probability value is greater than 5% critical value. Normality is also indicated by constructing histograms, indicated by a peak around zero and a clear tailing off on either side with a bell curve, or Gaussian distribution.

6.5.17 Residual Endogenous Variables Examination

The stability test serves as an indicator for constant variance. To take care-of heteroscedasticity the data has been transformed into logarithmic form. Additionally, this study further examines the residual of the endogenous variable. Constant variance is demonstrated by a graphical line display, rotating around zero, meaning that data is stationary. After validation, the study analyses the short-run and long-run relationships among the variables.

6.5.18 VECM Systems Long-Run and Short-RunAnalysis

Equations 7.30 and 7.31 demonstrate the media through which VECM facilitates, to measure the long-run and short-run relationships among variables. Accordingly, the VECM system model is comprised of the short-run and long-run components. This section explains the approaches adopted in analysing the simulated VECM systems model.

6.5.19 Long-Run Analysis

Earlier, it was explained that the cointegration simulation provides an avenue for the analysis of long-run relationship analysis among variables by employing the normalized cointegrating coefficients. It was also mentioned that the first part of the VECM model represents the error correction term, which also indicates the long-run relationship. The error correction term explains the long-run relationship of the cointegrating equations. At this stage, the long-run relationship of the cointegrating equations is interpreted by employing the coefficients, standard error and t − statistic in theoretical terms. The t − statistic is based on the testable hypothesis, specified:

H_O: β _i = 0

H_O: β _i ≠ 0

The t − statistic measure indicates the likelihood that the actual value of the parameter is not zero. To test the hypothesis in standard normal distribution the observed t − statistic values fall outside the range plus or minus 2. As a rule, a t − statistic larger than 2 in absolute terms means that there is a 5% or smaller probability of occurrence if the true coefficient were zero. The greater the value is in absolute terms, the better the results, meaning that the actual value of the parameter is statistically significant reflecting 95% confidence that the coefficient does not include zero.

6.5.20 Short-Run Analysis

The existence of short-run relationships among the variables is tested using two approaches. First, the second part of the VECM systems indicates the short-run relationships. The t − statistic values produced are used to interpret the theoretical short-run among series. Second, the study conducted a Granger causality test, which reflects the causal relationship among variables, which also serves as the short-run and F-/ Wald test statistics.

The null for no causality is rejected at 1%, 5% and 10% statistical critical value.

6.5.7.3 VECM Systems Model Ex-Ante Forecasting

Innovation accounting is comprised of impulse response and variance decomposition. This is conducted as a means of establishing the extent to which a change in one variable creates a change in another variable in the next period. In this study, conducting innovation accounting approaches creates empirical indicators on the effects of the variables within the system.

6.5.7.4 Impulse Response

Impulse response refers to the reaction of any dynamic system in response to some external change. Impulse response refers to the immediate effect of innovation or shock, resulting from one series to other series within the system Ericsson (Ericsson, Hendry & Mizon 1998; Pesaran & Smith 1998; Wei, 2013). This is a tool through which the reaction of one variable to an impulse or shock on another variable in the system can be explained. In a VAR model, impulse response is indicated as a positive shock of one standard deviation to the error terms in the model, so as to observe the reaction of the variables. The effects of innovation within the system are computed based on the residuals, such that innovations on ∈ _t by one unit create a forward movement within the system. In this way, the innovation to the j ^th variable first directly affects the same j ^th variable. Its innovations (j ^th variable) are transmitted to all other endogenous variables in the system through the VAR dynamic lag structure. Impulse response function traces the effect of a one-time innovation to one of the shocks on current and future values of the variable.

To analyse impulses in a system, the exogenous and deterministic variables are treated as fixed, and may be removed from the system (Lütkepohl 2006). In Eviews, impulse response is estimated by employing the Monte Carlo procedure via the Cholesky-dof adjusted ordering. The Monte Carlo approach is comprised of two approaches. The first approach offers simulations on ∅ _i from the asymptotic distribution. Second, VAR is simulated enabling ∅ _i to create estimated results that can be interpreted. The Cholesky ordering employs the Cholesky inverse factor of the residual covariance matrix to orthogonalise the impulses, and results are produced using graphs and tables.

The graphical output displays a visual display, and produces multiple graphs, which indicate the effect of the innovations on the series within the system. The effect of innovations is demonstrated by the manner in which the line graph departs from the zero line. As they depart from zero, the impulse line graph illustrates the path that a variable takes from the short to the long-run, expressed using a positive or negative sign. Meanwhile, the numerical output reflects the actual values, either positive or negative, that the impact of the innovations represent. This study employs the numerical approach for interpreting the results of the impulse response. This is because numeric values can be easily explained, as opposed to the line graphs. In this study, ≤ 5 years period is considered as short-run means while ≥ 10 years period is considered as long-run.

6.5.7.5 Variance Decomposition

Variance decomposition explains the manner in which one standard deviation shock creates variations in arithmetic terms from one period to another among the series. In this way, variance decomposition demonstrates the forecast error of a variable. In proportions attributed to innovations (shocks), each variable in the system, including its own, has internally induced innovations (Wickremasinghe 2011). In a simple linear equation, for any change in x at time (t) there is a corresponding change in y as a dependent variable. The variance decomposition created on the dependent variable y is expressed as

var (y) = E [var ( )] + var [E ( )] (6. 34)

Equation 6.32 demonstrates that in a relationship between x and y. The variance of the dependent variable y is comprised of two relationships. The first relationship is explained by the expected variance of the dependent variable y with respect to the independent variable. The second relationship indicates the variance of ycaused by the expected change from its own expected variance value.

In a VAR model, variance decomposition attempts to explain the proportion of the variance of the forecast error in predicting y_t, T + ℎ due to a structural shock or innovation, expressed as: 5_t. Based on orthogonal innovations 5 _t the ℎ − stee future forecast error vector can be expressed with known coefficients, as provided by the VECM model. In this study, based on the Monte Carlo procedure and ordering by Cholesky, the forecast is comprised of short-run (five years), medium-term (eight years) and long-run (10 years). The results of variance decomposition forecast for endogenous variables.

6.5.8 Simultaneous EquationEstimation

In systems of models, a number of estimators can be employed, including OLS, NLLS, the Full Information Maximum Likelihood Method and Instrument Variable (IV) methods such as Generalized Moments Methods, 2SLS and three-stage linear square methods. The choice of such an estimator largely depends on the properties of the series (Kunst 2012; Wei, H, 2010). OLS is the estimator employed in this study. Estimating the simultaneous equation that is estimated is based on the VECM system model, whose specification approaches are explained in Section 7.4.3. Following this approach, OLS satisfies the properties of an efficient estimator.

6.5.8.1 Rationale for OLS for Model Estimation

The approaches for establishing the simultaneous equation through which VAR is employed were explained earlier. Using the VECM procedure, the equation is estimated to understand the long-run relationship and short-run dynamics. This section explains the rationale for employing the OLS estimator. The OLS estimator provides sufficient results because the simultaneous equations are estimated based on the results of estimated by VAR, through a VECM systems approach. As such, employing OLS as a model estimator provides sufficient and valid results for economic policy, provided sufficient conditions are met. First, if the series are non-stationary at level but stationary at first difference. Second, if the series are cointegrated to in the same order [I(1)]. Third, if the roots of the companion matrix of the system lie inside the unit circle and are all less than one in absolute terms. Fourth, when the number of cointegrating vectors among all variables is equal to the number of endogenous variables. When the residual is tested for model stability, normality, variance and covariance, the results all indicate that data fits the model. Fifth, OLS can produce sufficient results as long as all the equations in the system have the same exogenous variables. Therefore, this study is comprised of three exogenous variables used in all the five equations.

Based on the preconditions mentioned, the OLS estimator is equivalent to the generalized least square estimator when all equations have identical repressors to all equations in the system. Approaches involving IV methods are often suitable in situations where polynomial roots lie outside the circle, and in reasonably large samples (King & Watson 1997).

In sum, the OLS estimation method produces sufficient and reliable results for policy analysis. Employing OLS, the manner in which the VECM systems employ a simultaneous equation theoretical specification is the next step.

6.5.8.3 Validation of the VECM Systems Simultaneous Equation Residual

This study tests for stability, starting by checking whether or not the residual for VECM systems simultaneous equation OLS estimation is stable. The residual is tested by displaying the graphical display. Stationary data is indicated by the manner in which the line graph rotates around zero mean; otherwise, data is said to be non-stationary. After graphical display, confirmatory tests are conducted via the ADF and PP test, and are confirmed by the KPSS.

After validation of the VECM systems simultaneous equation residuals, this study then estimates the five systems equations individually. This is because systems equations estimated under OLS have limited validation tests. The study finds that validating the OLS system equation is similar to the validation approaches. This study validates the systems equation estimated under OLS by estimating each of the five equations separately. This is intended to confirm that the findings and conclusions are sufficient for policy analysis. Moreover, under OLS systems, equation ex-post analysis is not application. The only option available is to estimate each equation separately, and then conduct ex-post analysis.

6.5.8.4 Estimation of the Simultaneous Equations

At this stage, the five equations are estimated separately by employing the NLLS/ARMA and adopting the Gauss-Newton/Marquardt Method of estimation. These individual results are similar to the OLS estimated coefficients. However, Antonakis et al. (2014) advises that before estimating a model, it is necessary to understand the nature of causality among variables of the simultaneous equation. Using the VECM Ganger causality approach, two models are estimated. First, the endogenous variables and human capital are estimated. This is followed by estimating the endogenous variables together with openness. This approach is taken because first, in the ASSM human capital and openness can be treated as exogenous.

Second, in some studies (e.g., Wei 2010) openness is treated as endogenous to economic growth. Third, Chapter Five illustrated how endogeneity exists between human capital, FDI and economic growth. In this respect two causality simulations are conducted to take care of these relationships. This can deepen our understanding of the relationship among variables as we attempt to examine the extent of causality among the related variables in this study. Finally, only two causality test models estimated, due to insufficient observations. After the causality tests, the simultaneous equation is estimated, validated and followed by interpreting the results based on specific testable hypotheses explained under Section below;

6.5.9 Specifications of TestableHypotheses

Aim 1: To examine the impact of export dynamics on Rwanda’s economic growth

H 0:The explanatory variables, tourism, telecommunications, openness, inflation has a positive impact on economic growth

Aim 2: To examine the impact of import inflows on Rwanda’s economic growth

H 0:The explanatory variables, tourism, telecommunications, openness, inflation has a

Negative impact on economic growth

Aim 3: To examine the impact of FDI on Rwanda’s economic growth

H 0:The explanatory variables, tourism, telecommunications, openness, inflation has a

Negative impact on economic growth

6.5.10. Validation of the Estimated Simultaneous Equations

Section 6.4.4.4 demonstrated the approaches employed in validating the VECM model. In this section the study the approaches adopted to test each equation is tested for stability, autocorrelation, heteroscedasticity and normality. Before diagnostic tests, goodness fit of the models is first examined by adjusted R-square and F-statistics. The adjusted R-square is employed as a measure for the goodness fit of the model, indicating the variance of dependent variables explained by the independent variables in the system. The adjusted R- square is employed because this never decreases, as more regressors are added into the model.

To test the goodness of fit, the simulated output of VECM provides the results. First, the adjusted R-square values are used by checking on their closeness to one for a good model.

Second, the F-statistics is employed as a means of checking the overall significance of the systems model. The results of the VECM systems simulated output indicates the results of the F-statistics. The null is tested, based on the hypothesis that all the coefficients of the regression are zero. To accept the null, the probability value of the F − statistical probability value of 5% is used (P − value > 5 %).

6.5.11 Stability Tests

The residuals for each equation are tested for stability by testing the residual, by first employing the actual fitted graph and fitted table. The fitted graph indicates the actual values of the dependent variable used in a regression, from the original data. A valid model is demonstrated by both the regression line and original data line graph moving together, otherwise the results are invalid. The fitted table provides statistics on the overall significance of the model being fitted. This is demonstrated by the manner in which the line of the residual fluctuates between one and negative one (−1 and 1) for a normal fitted model. Data stability is also indicated by employing the Cumulative Sum Control Chart (CUSUM) test statistic and recursive coefficients. To accept the null hypothesis, stability is confirmed within the 5% critical bounds of parameter stability. Parameters are indicated as stable when the line graph fluctuates between the two bounds.

6.5.11.1 Serial Correlation Tests

The serial correlation tests employed include Q-Statistic developed by Ljung and Box (1978) tests and the Breusch-Godfrey LM Test proposed by Breusch and Godfrey (1986) .These are compared to the Durbin-Watson (DW) as explained by Durbin and Watson (1971). The Q-Statistics test hypothesis for absence of autocorrelation is rejected when probability values are less than 5% critical value

The Breusch-Godfrey LM Test statistic computes lag order e based on an auxiliary regression of the residuals of the estimated regression. The testable hypothesis is specified as:

H₀: No serial correlation among the residuals;

H₁: There is serial correlation in the residuals

The null is accepted when the probability values of the LM Test are greater than 5% (P − value > 0.05) indicating the absence of serial correlation among the residuals.

6.5.11.2. Heteroscedasticity Tests

This study employs two tests of whether or not data is Heteroscedastic: the autoregressive conditional heteroscedasticity (ARCH) and Breusch-Pagan-Godfrey heteroscedasticity tests. The ARCH tests for heteroscedasticity under the testable hypothesis are specified as follows:

H_O: α ₁ = 0; α ₂ = 0; α _q = 0 (No ARCH effects)

H_O: Not all of α₁; α₂; … α_qare 0 (There are ARCH effects)

The null is accepted for no ARCH effects when the probability values are greater than 5%. Meanwhile, the Breusch-Pagan-Godfrey heteroscedasticity test is conducted as a validation test for the ARCH tests. The null is also accepted for data homoscedasticity when the probability values are greater than 5%.

6.5.12 Ex-PostForecasting

Ex-post forecast is conducted in this study as a means of observing both endogenous variables and the exogenous explanatory variables during the period under study, 1985–2014. This simulation is conducted to check existing data and evaluating the ex-post forecasted model.

6.6 ConcludingRemarks

To measure the impact of investment policy on economic growth in Rwanda, four procedures have been conducted: Dynamic panel properties investigation, short-run and long run analyses and VECM simulations. However, this study starts by presenting the conceptual framework, which provides a preliminary theoretical relationship among the variables.

This is followed by a description of variables by defining and providing the media through which they are measured. The procedure starts by investigating the properties of the time-series: transforming the series into logarithmic terms, followed by graphical, correlation and trend analysis. Later, unit root approaches are explained, with final endogeneity examination of the variables.

The second part explains the methods employed to measure the short-run and long-run relationships among variables. This part explains how cointegration analysis is conducted, followed by long-run interpretation, VECM systems model specification and simulation. Later, the diagnostic approaches are examined as a basis for long-run and short-run interpretations and VECM systems causality simulation and ex-ante forecasting. The final part of the procedure explains the approaches employed in the VECM systems simultaneous simulations. The first involves explaining the manner in which the study conducts VECM Granger causality tests, and later VECM systems model estimation. The method for model estimation is based on OLS and the rationale is provided. After, the study explains the media through which diagnostic tests are conducted, including their interpretation. Through this procedure, the findings and conclusions for the study are made, following a chronological approach.

CHAPTER SEVEN:

TIME-SERIES PROPERTIES AND INVESTIGATION OF THE VARIABLES

7.1 Introduction

In chapter five a conceptual framework was developed that pointed out the variables employed in this study. Following this conceptual framework, 5 variables are employed in this study: FDI (FDI net inflow, FDI net outflow), Export of goods and services, Import of goods and services, Annual Real GDP and GDP growth. The chapter starts by describing the data; the chapter further investigates the properties of the series employed in this study by conducting graphical analysis of the variables. Second, the series was subjected to differentiation to ensure that that the series are stationary after which correlation analysis was carried out among variables to better understand the manner in which the variables are related this was followed by a regression analysis. The next part presents a trend analysis, to indicate the fluctuations of the variables. This launched the basis for the next step, involving unit root tests.

Finally, an endogeneity causality test was conducted in order to understand a causal relationship among variables, all intended to examine the impact of investment policy on Rwanda’s economic growth.

7.2 Descriptive statistics

Table3: Results for Descriptive statistics

Tables are not included in the reading sample.

Source: Secondary data 1997-2018

According to the results in the table above, the average GDP for Rwanda is 4,800,000,000 USD with a standard deviation of 2,884,110,822.78 USD. Over the 22 years the average GDP growth is 8.012 with a standard deviation of 2.726. The average FDI is 120,154,545.45 USD (SD = 120,000,550.50). The average FDI Net outflow is 61,940,909.09 USD (SD=86,801,117.28). The average Exports of goods and services is 653,181,818.18 USD (SD=529,297,705.08), and the average Imports of goods and services is 1,462,727,272.73 (SD = 1,047,683,075.57).

The above indicates that on average Rwanda spends more than double of what it earns from its exports on imports. This in the long, given the trend, would negatively impacts on GDP growth rate, except if the foreign exchange gap is filled by external resources.

7.3 Correlation Analysis

Correlation analysis is used to measure the degree of a linear association between the variables; the strength of the linear relationship between two variables is measure by a correlation coefficient. The Correlation coefficient among variables ranges between negative one and positive one (−1 to + 1).

In absolute terms numeric value (1) between two variables is regarded as perfect correlation, Zero represents a situation where there is no relationship between variables. It is worth noting that the relationship between variables grows stronger as the correlation coefficient tends towards +1 for a positive relationship and -1 for a negative relationship.

In economic analysis when two explanatory variables are highly correlated in a single regression analysis, at least one is removed from the study. Therefore, correlation was analysis was useful to find out if any two explanatory variables were highly correlated so as to remove one from the study.

Table4: Results for Correlation Analysis

Tables are not included in the reading sample.

Source: Secondary data 1997-2018

It should be noted that the larger the magnitude of correlation coefficient, the more to variables are perfectly related. In this study, no explanatory variable has a higher correlation with the other and thus they were all kept. After knowing that there was no highly correlated series, the next step is to understand the trends in the series.

7.4 Regression analysis

In order to determine the extent to which the various elements can influence GDP a regression analysis was used. Ordinary Least Square Method (OLS) of estimation was used and all the variables were entered in the model at the same time.

Table5: Results of regression analysis of the various independent variables on the dependent variable

Tables are not included in the reading sample.

Source: Secondary data 1997-2018; *P<0.05, ***P<0.001^adependent variable: GDP

In the model above, the contribution of the various determinants to GDP is presented. The overall model is significant (F= 706.15, P<0.001). T

he adjusted R²of the model is 0.99; this means that 99% variance in GDP is explained by the given variables, meaning that almost all variance in GDP of Rwanda is due to FDI net inflow, Export of goods and services and Imports of goods and services. According to the results in the table above keeping other factors constant the relationship between FDI net inflow and GDP is positive and statistically significant (β=3.826, P<0.05) meaning that a unit increase in FDI net inflow increases GDP by 3.826. A significant and positive relationship was reported between Export of goods and services and GDP (β=1.402, P>0.05) meaning that a unit increase in Export of goods and services increases GDP by 1.402. Also a significant and negative relationship was reported between Import of goods and services and GDP (β= -1.639, P>0.001) meaning that a unit increase in Import of goods and services GDP would fall by 1.639. Policy options to reduce heavy expenditure on imports through import substitution industrial development strategies would enhance the Country’s GDP growth rate in the long run

7.5 Series Trend Analysis

Times series data often exhibits increasing or decreasing trends, with fluctuations. As such, trend analysis is necessary before unit root testing, to establish whether the series has a unit root or not. Trend analysis can be a tool for determining whether the series is stationary around a constant or a trend that can be included during testing.

Figure45: Result of the Series trend Analysis

Figures are not included in the reading sample.

The results of graphical display indicate that beside GDP growth the series exhibit a random walk with drift and trend. Large frustrations characterizing the trend are reflected by the series indicating that besides GDP growth all the other series are non-stationary and thus the properties of the series data were therefore tested for unit root.

7.6 Time Series Unit Root Testing

Considering the properties of the series, unit root is tested by first testing the series at level, and later the researcher tested the series at first difference. Unit root tests were conducted employing the Dickey-Fuller method.

Table6: Results of the ADF unit root tests

Tables are not included in the reading sample.

Source: Secondary data 1997-2018

Dickey Fuller Test is a unit root test for stationality, it should be noted that unit roots are undesirable as they can cause unpredictable results in the time series analysis. For the study at hand, the hypotheses for the test are: The null hypothesis is that the series possesses a unit root and hence is not stationary and the alternate hypothesis is that the time series is stationary. A p-value of less than 5% means that one can reject the null hypothesis that there is a unit root (Zaiontz, 2019). Stationarity means that the statistical properties of a process generating a time series do not change over time (Palachy, 2019).

Following the dickey fuller test all series except GDP growth are non-stationary at level, this because there P-values are greater than 0.05 at 5% critical value. Being non-stationary necessitated transformation of the series using differencing, differencing helps to stabilize the properties of the time series and therefore eliminate the non stationarity. It is worth noting that difference stationary processes have an order of integration, which is the number of times, the differencing operator must be applied to it in order to become stationary (Palachy, 2019). At first difference all series became stationary except GDP which became stationary at second difference; this is evidenced by P-values which are less than 0.05 at 5% critical value. After noting the characteristic properties for the series, a graphical visual display of the series was presented at first difference. The results of graphical display indicate that all series are stationary after differencing, these are summarized below.

Figure46: Results of the series trend test after first difference

Figures are not included in the reading sample.

CHAPTER EIGHT:

ESTIMATION OF THE SHORT AND LONG-RUN RELATIONSHIPS

AMONG THE ENDOGENOUS VARIABLES

8.1 Introduction

In Chapter Seven, the properties of the series employed in this study were investigated. The section involved a graphical study of the series followed by correlation analysis and later unit root tests. Finally, the study conducted a Pairwise Granger causality test to indicate endogeneity among the series and explain causality among the variables. The Unit root test, cointegration analysis to explain the long-run and short-run relationships among the series. In the second section a VECM model is specified, followed by diagnostic testing. The final section of this chapter provides ex-ante simulations comprised of impulse response and variance decomposition

8.2Economic growth and export of goods and services

The properties of the series of economic growth (GDP) and export of goods and services are investigated in the earlier sections. Generally to clearly understand the relationship between economic growth and exports of goods and services several procedures were followed as seen above first there was a graphical study of the series followed by correlation analysis. In addition to these, unit root tests were carried out after which, the study conducted co-integration tests, Pairwise Granger causality test to indicate endogeneity among the series and explain causality among the variables and finally had the impulse response functions.

8.2.1 Unit root tests

Before testing, to establish the existence of long-run relationships, the lag length is determined as a precondition. Lag length determination is important for model specification. First, misspecification of the lag length leads to inconsistent impulse response function and variance decomposition results derived from the estimated VAR. Second, over-fitting causes an increase in the mean-square forecast errors of the VAR. Third, under-fitting the lag length often generates auto correlated errors. In this study, to determine the lag length, the unrestricted VAR lag order selection criteria is employed. In the unrestricted VAR model, the appropriate lag is indicated with an asterisk and the smaller the value, the better the criteria. The results of VAR lag order selection criteria are indicated below.

Table7: Results for VAR lag order selection criteria results

Figures are not included in the reading sample.

Source: Secondary data 1997-2018 * indicates lag order selected

According to the table above, several levels can be used for criteria. These include Likelihood ratio test (LR) test statistic, Final Prediction Error (FPE), Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC) and Hannan-Quinn (HQ). The best lag length is selected considering the lag criteria that is identified by the majority asterisk indicator. The asterisk indicates that three criteria are appropriate, employing four (4) lag lengths. From the table above four lags were used for the bivariate model because the Hannan–Quinn information criterion (HQIC) method, Akaike Information Criterion (AIC) method, and sequential likelihood-ratio (LR) test all chose four lags, as indicated by the asterisks in the output. After establishing the optimal lag length, a co-integration test was then carried out.

8.2.2 Co- integration tests

The simulations that provide the results of the trace statistics were conducted after selecting the lag length; this was to help in finding out whether or not the series are co-integrated, and also the number of co-integrating equations. In this study, the null hypothesis tested is that there are no co-integrating equations. The alternative is that there is at least one co-integrating equation. The co-integration output results are summarized below.

Table8: Johansen tests for co-integration results

Tables are not included in the reading sample.

Source: Secondary data 1997-2018

According to trace test statistics, the null hypothesis cannot be rejected because the trace statistic value is less than the critical value (10.032 < 15.41). Therefore, the test results indicate that there is no evidence of co integration between GDP and export of goods and services; hence a Vector Auto-Regression (VAR) in first differences is appropriate.

8.2.3 Granger causality analysis

In order to get further insights into the dynamic interactions and the strength of causal relationship between the variables of interest, the study proceeded with a Granger causality analysis. Basically, variable x is said to Granger cause variable y if variable x helps in the prediction of variable y (Asmah, 2013).

Table9: Granger causality Wald test results

Tables are not included in the reading sample.

Source: Secondary data 1997-2018

The Granger causality Wald test was carried out to test whether lags of Exports of goods and services returns are helpful in forecasting GDP returns, and whether GDP returns are helpful in forecasting Exports of goods and services returns. Based on the table above at 5% confidence level it is evident that Export of goods and services granger causes GDP, and that GDP granger causes Exports of goods and services.

It can thus be concluded that the variables are correlated enough that one is useful in forecasting the other; the relationship can now be established openly.

Before going further the model was tested to see if it well specified using the Lagrange-multiplier test.

Table10: Results for Lagragen-multiplier test results

Tables are not included in the reading sample.

Source: Secondary data 1997-2018

Based on the langrage multiplier test it is evident that there is no autocorrelation, as the null hypothesis of no – correlation cannot be rejected at 5% confidence level because the p-value for both the lags is greater than 5%.

Based on the above, given that the p-value for both the lags is greater than 5%, we fail to accept the alternative hypothesis and thus, fail to reject the null hypothesis and the conclusion is that there is no – correlation

8.2.4 Impulse response functions

A stable system of equations declines the shock to zero, meaning that the short-run values of the variable in question converge to the long-run equilibrium values. Impulse response plots represent the response of a variable given an impulse in another variable. From the figure below the response of GDP given an impulse in export of goods and services is represented.

Figure47: Results of Response of GDP given an impulse in export of goods and services is represented

Figures are not included in the reading sample.

According to the top right graph, an unexpected shock to export of goods and services has a relatively small up and down relationship (fluctuations) with GDP. As can be observed from the graph, the impact of the shock will first cause GDP to be constant up to close to the 2 ^nd Period and thereafter wane and tends to decrease till the 3 ^rd Period and then stabilize slightly after the 4 ^th period after which it wanes a bit and stabilizes again at the 8 ^th Period.

8.3 Economic Growth and Imports

The properties of the series of economic growth (GDP) and Imports of goods and services are investigated in the earlier sections. Generally to clearly understand the relationship between economic growth and imports of goods and services several procedures were followed as seen above first there was a graphical study of the series followed by correlation analysis. In addition to these, unit root tests were carried out after which, the study conducted co-integration tests, Pairwise Granger causality test to indicate endogeneity among the series and explain causality among the variables and finally had the impulse response functions.

8.3.1 Unit root tests

VAR lag order selection criteria results

Tables are not included in the reading sample.

Source:Secondary Data 1997-2018 * indicates lag order selected

The best lag length is selected considering the lag criteria that is identified by the majority asterisk indicator. The asterisk indicates that three criteria are appropriate, employing four (4) lag lengths. We will use four lags for this bivariate model because the Hannan–Quinn information criterion (HQIC) method, Akaike Information Criterion (AIC) method, and sequential likelihood-ratio (LR) test all chose four lags, as indicated by the asterisks in the output. After establishing the optimal lag length, a co-integration test was then carried out.

8.3.2 Co integration tests

The simulations that provide the results of the trace statistics and Maximum was conducted after selecting the lag length, this was to help in finding out whether or not the series are co-integrated, and also the number of co-integrating equations. In this study, the null hypothesis tested is that there are no co-integrating equations. The alternative is that there is at least one co-integrating equation. The co-integration output results are summarized below.

Table11: Johansen tests for co-integration results

Tables are not included in the reading sample.

According to trace test statistics, the null hypothesis is rejected because the trace statistic value is greater than the critical value (19.481> 15.41). Therefore, the test results indicate that there is evidence of co-integration between GDP and import of goods and services. A further review indicates that there is one co integrated equation based on the two criterions, and as a result a Vector Auto-Regression (VAR) was considered appropriate to establish a short-run relationship

8.3.3 Granger causality analysis

In order to get further insights into the dynamic interactions and the strength of causal relations between GDP and Imports of goods and services, the study proceeded with a Granger causality analysis. Basically, variable x is said to Granger caused y variable x helps in the prediction of the behaviour of variable y (Asmah, 2013).

Table12: Granger causality Wald test results

Tables are not included in the reading sample.

The Granger causality Wald test was carried out to test whether lags of Imports of goods and services returns are helpful in forecasting GDP returns, and whether GDP returns are helpful in forecasting Imports of goods and services returns. Based on the table above at 5% confidence level it is evident that Import of goods and services granger causes GDP, and that GDP granger causes Imports of goods and services.

It can thus be concluded that the variables are correlated enough that one is useful in forecasting the other; the relationship can now be established openly.

Before going further the model was tested to see if it well specified using the Lagrange-multiplier test.

Table13: Lagrange-multiplier test results

Tables are not included in the reading sample.

From the above test, given that the p-value for both the lags is greater than 5%, we fail to accept the alternative hypothesis and thus, fail to reject the null hypothesis and the conclusion is that there is no – correlation

8.3.4 Impulse response functions

Given that variables in a VAR model depend on each other, individual coefficient estimates only provide limited information on the response of the system to a shock. In order to get a better picture of the models dynamic behaviour, impulse responses (IR) are used (Mohr, 2020). Impulse response helps to produce the time path of the dependent variables in the VAR, to shocks from all the explanatory variables. A stable system of equations declines the shock to zero, meaning that the short-run values of the variable in question converge to the long-run equilibrium values. Impulse response plots represent the response of a variable given an impulse in another variable. From the figure below the response of GDP given an impulse in Import of goods and services is represented.

Figure48: Response of GDP given an impulse in Import of goods and services is represented.

Figures are not included in the reading sample.

The above combined graphs are based on the output of the unrestricted VAR with analytic response standard error over 8 periods. Each graph as shown in plots in the figure above includes a point estimation of impulse response functions as well as lower and upper bounds for a 95% confidence interval. The solid lines depict the variable percent change in response to a standard deviation of one in the respective variable whereas the dotted lines represent the 95% error bands. According to the top right graph, an unexpected shock to Import of goods and services has a predictable and stable relationship with GDP. As can be observed from the graph, the impact of the shock will first cause GDP to increase up to 1st Period and thereafter tends to decrease till the 2 ^nd and then continues with the increase and decrease movements. Ultimately, the disturbance by a shock in imports of goods and services is stabilized in the 8 ^th period.

8.4 Economic growth and FDI inflow

The properties of the series of economic growth (GDP) and FDI net Inflow are investigated in the earlier sections. Generally to clearly understand the relationship between economic growth and FDI net Inflow several procedures were followed first there was a graphical study of the series followed by correlation analysis and later unit root tests, after which the study conducted co-integration tests, Pairwise Granger causality test to indicate endogeneity among the series and explain causality among the variables and finally had the impulse response functions.

8.4.1 Unit root tests

In this study, to determine the lag length, the unrestricted VAR lag order selection criteria is employed. In the unrestricted VAR model, the appropriate lag is indicated with an asterisk and the smaller the value, the better the criteria. The results of VAR lag order selection criteria are indicated below.

Table14: VAR lag order selection criteria results

Tables are not included in the reading sample.

Notes: * indicates lag order selected

According to the table above, several levels can be used for criteria. These include Likelihood ratio test (LR) test statistic, Final Prediction Error (FPE), Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC) and Hannan-Quinn (HQ). The best lag length is selected considering the lag criteria that is identified by the majority asterisk indicator. The asterisk indicates that four criteria are appropriate, employing three (3) lag lengths, the four criteria are; Hannan–Quinn information criterion (HQIC) method, Akaike Information Criterion (AIC) method, Final Prediction Error (FPE), and sequential likelihood-ratio (LR) test which all chose three lags, as indicated by the asterisks in the output. After establishing the optimal lag length, a co-integration test was then carried out.

8.4.2 Co-integration tests

The simulations that provide the results of the Maximum statistics was conducted after selecting the lag length, this was to help in finding out whether or not the series are co-integrated, and also the number of co-integrating equations. In this study, the null hypothesis tested is that there are no co-integrating equations. The alternative is that there is at least one co-integrating equation. The co-integration output results are summarized below.

Table15: Johansen tests for co-integration results

Tables are not included in the reading sample.

According to trace test statistics, the null hypothesis is rejected because the trace statistic value is greater than the critical value (22.9046> 15.41). Therefore, the test results indicate that there is evidence of co-integration between GDP and FDI net inflow. A further review indicates that there is one co integrated equation based on the two criterions, and as a result a Vector Auto-Regression (VAR) was considered appropriate to establish a short-run relationship.

8.4.3 Granger causality analysis

In order to get further insights into the dynamic interactions and the strength of causal relations between GDP and FDI net inflow, the study proceeded with a Granger causality analysis. Basically, variable y is said to be Granger caused by variable x if x helps in the prediction of y (Asmah, 2013)

Table16: Granger causality Wald test results

Tables are not included in the reading sample.

The Granger causality Wald test was carried out to test whether lags of FDI net inflow returns are helpful in forecasting GDP returns and whether GDP returns are helpful in forecasting FDI net inflow returns. Based on the table above at 5% confidence level it is evident that FDI net inflow granger causes GDP, and that GDP granger causes FDI net inflow. It can thus be concluded that the variables are correlated enough that one is useful in forecasting the other; the relationship can now be established openly.

Before going further the model was tested to see if it is well specified using the Lagrange-multiplier test.

Table17: Lagrange-multiplier test results

Tables are not included in the reading sample.

From the above langrage multiplier test, it is evident that there is no autocorrelation, as the null hypothesis of no – correlation cannot be rejected at 5% confidence level because the p-value for both the lags is greater than 5%.

8.4.4 Impulse response functions

Given that variables in a VAR model depend on each other, individual coefficient estimates only provide limited information on the response of the system to a shock. In order to get a better picture of the models dynamic behaviour, impulse responses (IR) are used (Mohr, 2020). Impulse response helps to produce the time path of the dependent variables in the VAR, to shocks from all the explanatory variables. A stable system of equations declines the shock to zero, meaning that the short-run values of the variable in question converge to the long-run equilibrium values. Impulse response plots represent the response of a variable given an impulse in another variable, from the figure below, the response of GDP given an impulse in FDI net inflow.

Figure49: Response of GDP given an impulse in FDI net inflow

Figures are not included in the reading sample.

The above combined graphs are based on the output of the unrestricted VAR with analytic response standard error over 8 periods. Each graph as shown in plots in the figure above includes a point estimation of impulse response functions as well as lower and upper bounds for a 95% confidence interval. The solid lines depict the variable percent change in response to a standard deviation of one in the respective variable whereas the dotted lines represent the 95% error bands. According to the top right graph, an unexpected shock to FDI net inflow has a predictable and stable relationship with GDP. As can be observed from the graph, the impact of the shock will first cause GDP to keep almost constant but increases a bit up to the 2 ^nd period after which it reduces up to closer the 3 ^rd Period from which it keep almost constant upto the 8 ^th period.

8.5 Conclusion

In conclusion, the aim of estimating the economic growth equation was to examine the impact of investment policy on Rwanda’s economic growth. The study sought to examine the impact of exports on economic growth in Rwanda, to determine how Imports contributed to economic growth of Rwanda and to determine the impact of FDI on economic growth in Rwanda, after this the study investigated the causal relationship between investment policy and economic growth in Rwanda and tested for the existence of a long run relationship between investment policy and economic growth in Rwanda.

Key observation: A short run relationship was established between economic growth and export of goods and services; and these two were found to be having a causal relationship as each was found to granger cause the other. It was also further found out that an unexpected shock to export of goods and services has a relatively small up and down effect on economic growth.

Similarly, with regards to Economic growth and FDI net inflow one co integrated equation was found which resulted into establishment of a short-run relationship between Economic growth and FDI net inflow the two were also found to be having a causal relationship as each was found to granger cause the other. It was also further found out that an unexpected shock to FDI net inflow has a predictable and stable relationship with Economic growth. In general even with the reported great contribution of investment policy on economic growth in Rwanda, the established relationship was found to be short term.

Chapter Nine: Discussion of Results

9.1 Introduction

In this section, discussion is based on the finding on each of the study. In the first section, focus is on discussion of major findings arising from the Series Properties Investigation of the Variables. Followed by discussion of major findings arising from the analysis of long run relationship between Investment Policy and GDP growth Rate in Rwanda supported by relevant literature based on current empirical finding on related thematic areas.

9.2 Discussion of Major Findings Arising from the Series Properties Investigation of the Variables

Although the trend demonstrates that GDP, FDI have increased, the growth in output indicated by GDP growth is still very low, sometimes constant and de minimis (only by 0.66 units). The point of confidence is that this can be explained by assumptions of the Solow-Swan growth model which demonstrates the role of physical capital to nation and Total Factor Productivity as well as the absorptive capacity of the country. The graphical investigation demonstrates that both GDPGR and growth in output has been very low during the period 1997–2018. In this respect, economic growth has fluctuated by around 4.8 units, while GDPGR by 7.3 during the period under study. This is rather a worrying indicator for the nation because domestic investment and employment of a nation’s LF largely depends on the rate at which the economy grows ceteris peribus. First, the growth in employment is greater than growth in production, meaning increasing unemployment. This is an indicator that Rwanda is a least-developed country, with low technology and abundant unskilled labour, and thus low TFP. Secondly, endogeneity tests among variables were investigated, employing pairwise causality tests.

On the basis of the simulation the study concluded that the variables of exports, imports and FDI does indeed Granger-cause economic growth. Further still, causality exists between investment variables and GDP growth rate with feedbacks. However, the findings of the preliminary investigation indicated that there is very minimal relationship between exports, imports, FDI and economic growth. We can observe that conditions mainly favour FDI inflows in Uganda because, due to the need to attract tourism, reduce poverty and develop human capital, FDI increases. We can observe that FDI has little impact, and a spill-over effect on domestic investment and employment. We can also observe that conditions are biased towards openness. Due to the need for accelerated economic growth and FDI, the GOR made openness a policy. Furthermore, on the basis of this Pairwise Granger causality test, the study concluded domestic investment using Rwanda local resources is important. This is because according to this simulation as Rwanda experiences higher levels of economic growth so does openness. In turn, the nation experiences increasing domestic investment in the long-run.

In the model above, the contribution of the various determinants to GDP is presented. The overall model is significant (F= 706.15, P<0.001). The adjusted R²of the model is 0.99; this means that 99% variance in GDP is explained by the given variables, meaning that almost all variance in GDP of Rwanda is due to FDI net inflow, Export of goods and services and Imports of goods and services. According to the results in the table above keeping other factors constant the relationship between FDI net inflow and GDP is positive and statistically significant (β=3.826, P<0.05) meaning that a unit increase in FDI net inflow increases GDP by 3.826. A significant and positive relationship was reported between Export of goods and services and GDP (β=1.402, P>0.05) meaning that a unit increase in Export of goods and services increases GDP by 1.402. Also a significant and Negative relationship was reported between Import of goods and services and GDP (β=-1.639, P>0.001) meaning that a unit increase in Import of goods and services may result into a fall in GDP by 1.639.

9.3. Discussion on the Relationship between Net Exports and GDP Growth Rate in Rwanda

The properties of the series of economic growth (GDP) and export of goods and services are investigated in the earlier sections. Generally to clearly understand the relationship between economic growth and exports of goods and services, several procedures were followed as seen in chapter eight, first there was a graphical study of the series followed by correlation analysis. In addition to these, unit root tests were carried out after which, the study conducted co-integration tests, Pairwise Granger causality test to indicate endogeneity among the series and explain causality among the variables and finally had the impulse response functions.

According to trace test statistics,(see table 8.2) the null hypothesis cannot be rejected because the trace statistic value is less than the critical value (10.032 < 15.41). Therefore, the test results indicate that there is no evidence of co integration between GDP and export of goods and services; hence a Vector Auto-Regression (VAR) in first differences is appropriate

In order to get further insights into the dynamic interactions and the strength of causal relations between the variables, the study proceeded with a Granger causality analysis (Table 8.3). Basically, variable y is said to be Granger caused by variable x if x helps in the prediction of y (Asmah, 2013).

The study of the relationship between exports and GDP growth rate of late has attracted debates globally.Undrtanding the significance of Countries export is imperative in understanding the country’s foreign exchange base and overall investment potential. There is no doubt that a country that invests more on export trade enjoys significant amount of revenue inflow. This significantly influence the Country’s Current account, Foreign exchange reserves and terms of trade. (World Bank, World Development Indicators, 2016).

The above finding seems to be in agreement with a related study, ( Barro, 1991; Barro & Lee, 1993; Ben-David, 1998; Collier & Gunning, 1999; Ghura & Hadjimichael, 1996; Hernandez, 2000; Khan& Reinhart, 1990; Ndikumana, 2000).It’s important to note that export investment is a powerful channel for innovation, economic growth and therefore poverty reduction. Recent empirical studies have established linkages between export investment and economic growth. Analysis of causality between economic growth and domestic investment conducted in different countries are marred with ambiguities and inconclusive results. For example, several researchers have found bi-directional relationship (Tang, Selvanathan & Foreign, 2008; Tan & Lean, 2010). Others found the direction of causality to be from economic growth to domestic investment (Choe, 2003; Quin, Cagas, Quising & He, 2006) while some found the direction of causality to be from net exports to economic growth (Villa, 2008). Also in other studies, private investment was shown to be super-exogenous, meaning investment was the primary determinant of economic growth (Montek, 2002).

According to World bank development indictors (2016; NISR, 2017), the world market has had an important influence on Rwanda’s export performance in terms of volume as well as price and the location of Rwandan products in the growth segments of the global market underscore the headwinds it faces. The four quadrant graph (Figure 8) compares Rwanda’s export performance with the trend in world trade for the period 2011-15. It compares the change in Rwanda’s share of world exports to the annual growth rate of world trade for each product. The ideal situation is for a product to be gaining market share in an expanding global market, the upper right quadrant. This is true of tea and raw hides and skins. However, for tantalum (coltan), the situation is less promising Rwanda’s market share is improving but world trade is falling (the top left quadrant). It is still worse for coffee and tungsten (wolfram), where the country is losing market share in a global market that itself is shrinking (the lower left quadrant).

As Nurkse (1959) asserts, foreign trade undertook the role of growth’s engine in such countries as Canada, the United States and Australia in the 19th century. According to Kravis (1970), the real reason for growth through foreign trade in these countries was their rich natural resources. Cairncross (1961) states that developing countries use their natural resources only to meet their domestic demands, and they can allot only an insignificant portion for export

According to Emine KILAVUZ (2012), whether an economy can benefit from an increase in export depends on the supply and demand elasticity of export goods. The higher the supply and demand elasticity of export goods, the more export stimulates economic growth. The supply and demand elasticity of export goods in developed countries is higher than that of developing countries. Therefore, the effect of export on economic growth is more in developed countries compared to developing countries. Higher growth rate in the manufacturing sector results in higher growth rate in Gross Domestic Products. Kaldor (1968) explains why the manufacturing industry is growth’s engine and how it creates positive externalities in the economy. Kaldor (1968) states that increasing returns to scale existing in the industry sector increase investment returns. Due to such features, the industry sector provides positive externalities in the economy in general and accelerates economic growth via these externalities. The growth of the industry sector increases productivity not only in itself, but also in other sectors with a large range of facilities for division of labour. That is why Kaldor considers the industry sector as “growth’s engine”. Kaldor maintains that growth in industrial manufacturing can be possible only through external demand with a high growth rate; that is, through export. The higher the growth rate in the manufacturing industry that export determines, the faster the transfer of the labour will be from sectors in which economic productivity is low to the industrial sector, which leads to a faster productivity increase.

Internal Growth Theories, the basis of which dates back to Adam Smith, also emphasize the growth-increasing effect of foreign trade. Among the supporters of Internal Growth Models, Grossman and Helpman (1990a) discuss the internal growth of countries that are engaged in foreign trade along with international information overflow. In their study, it is assumed that information overflows occur automatically, and the growth performance of a small country which can obtain scientific and technological information flow from foreign countries, gauged by its foreign trade, is analyzed. In addition, it is asserted that some policies that are incentives for foreign trade accelerate growth by decreasing the harmful effects caused by innovation externality and promote national prosperity. Moreover, the study reveals that without external technological improvement and constant returns in manufacturing, information overflows can promote long-term economic growth. In another study in which the correlation between foreign trade and growth is investigated, Grossman and Helpman (1990b) claim that the R&D (research and development) sector, which benefits from the opportunities foreign trade provides, is the driving force of growth by providing the domestic economy with a comparative advantage. According to the authors, by liberalizing their foreign trade, developing countries will be able to have access to world information stock by means of technology transfer, and eventually they will get the maximum benefit from liberalization. Rivera-Batiz and Romer (1991) examine the effect of economic integration on growth in countries that have similar technology and factor endowment, such as Europe and North America. The results indicate that if economic integration leads to increasing returns to scale in two different economies that have similar development levels, this integration undertakes the task of the growth’s engine. Therefore, integration increases the long-term growth rate as it clearly leads to market expansion. In conclusion, if increasing returns expand the sector, growth occurs. Economic integration increases long-term growth rate by pruducing the scale effect. Policies that influence long-term growth rate have a great effect on economic prosperity

9.3 1 Discussion on the Relationship between Net Imports and GDP Growth Rate in Rwanda

To start with,Johansen tests for co-integration results (table 8.5), according to trace test statistics, the null hypothesis is rejected because the trace statistic value is greater than the critical value (19.481> 15.41). Therefore, the test results indicate that there is evidence of co-integration between GDP and import of goods and services. A further review indicates that there is one co integrated equation based on the two criterions, and as a result a Vector Auto-Regression (VAR) was considered appropriate to establish a short-run relationship

In order to get further insights into the dynamic interactions and the strength of causal

relations between GDP and Imports of goods and services,the Granger causality Wald test was carried out to test whether lags of Imports of goods and services returns are helpful in forecasting GDP returns, and whether GDP returns are helpful in forecasting Imports of goods and services returns. Based on the table above at 5% confidence level it is evident that Import of goods and services granger causes GDP, and that GDP granger causes Imports of goods and services. It can thus be concluded that the variables are correlated enough that one is useful in forecasting the other

Based on the table above at 5% confidence level it is evident that Import of goods and services granger causes GDP, and that GDP granger causes Imports of goods and services. It can thus be concluded that the variables are correlated enough that one is useful in forecasting the other

Different studies and researches were done by academics and policy makers for economic growth, import and export. A variety of studies shows different results about the relationship of this three variables. Export led hypothesis is a widely known hypothesis and accepted by different academics (Feder 1982; Kruege 1990). Atrkar Roshan Sedigheh (2008) made a study about export expansion and economic growth in Iran after the revolution period. The results of this study confirms the relationship between these variables in Iran after the revolution period. A study done by Ahmet Ugur (2008), shows the relationship between imports and economic growth in Turkey. Velnampy. T, Achchuthan (2013); Based on the overall study, in the Sri Lankan context, the export and import have the significant positive relationship, and also, both export and import have the significant impact on the economic growth. Further, the export and import have been associated by 98 percent, which denotes that, there is a strong positive association between export and import. Another study done by Murat Çetinkaya and Savas Erdogan (2010) tested the relationship of two figures, importexport by using VAR Analysis. According to the study it was determined that there was causality relationship between these variables, the variable import influenced GDP, and GDP influenced the variable export. Between export and import, two way Causality relationships released mutually. In the same way, the results of causality overlap with variance decomposition test. Mehdi Taghavi, Masoumeh Goudarzi, Elham Masoudi, Hadi gashti (2012) studied the Iran economy from 1962- 2011. VAR Analysis was applied between the variables of annual economic growth, import, and export. When regarding to these results, it is implied that the export increases as the country grows and the import indicates a decrease economic growth. When regarding to the data used in the study, they indicate a difference proportionally, it is seen that the increases or decreases in the import, export, and GD

Kogid, Mulok, Ching, Lily, Ghazali and Loganathan (2011) analyzed the relationship between the economic growth and the import in Malaysia from 1970 to 2007. Results show that there is no co integration exists between economic growth and import, but there exists bilateral causality between economic growth and import. Results also show that import could indirectly contribute to economic growth, and economic growth could also directly contribute to import. These findings may be vital for future economic growth policy. Ali F. Darrat (1987) made a study about export-led hypothesis of Ronald Findlay (1984) and Anne Kruege (1985);This hypothesis states that higher exports accelerate the economic growth process. The empirical results reported by Ali F. Darrat (1987) shows that the economic growth of Hong Kong, Korea, Singapore and Taiwan are not affected by exports. Based on the Granger causality test, no causal effect were shown from exports to economic growth in any of the four countries

9.3 2 Discussion on the Relationship between Net FDI and GDP Growth Rate in Rwanda

Short-run and long-run simulations were conducted in Chapter Eight, as well as an ex-ante forecast, through impulse response and variable decomposition simulations.

On the basis of the short-run simulation between FDI and GDP growth Rate, the study found that FDI does Granger cause economic growth in Rwanda. Also, the study established the presence of feedback between FDI and Rwandan Economic growth in the short run. This is similar to the Pairwise Ganger causality results, which indicate that FDI does Granger cause GDP growth rate in Rwanda. Finally, the presence of feedback between FDI and Economic growth simply meant that there is strong relationship between Rwanda economic growth and Net FDI inflows.

Note should be taken that in this section, the objective of this study was to establish the impact of FDI on Rwanda’s economic growth. On the basis of the Pairwise Granger causality test simulation, we concluded that FDI does Granger-cause economic growth. The Pairwise simulation was preliminary investigation. Second, theory and literature indicate that FDI significantly contributes to economic growth. In this respect, based on the short-run Granger Causality simulation,

Impulse response indicates that FDI and employment will cause declining economic growth if the supportive economic and political environment fails to support the existing investment initiative. Its therefore imperative to note that, given that Rwanda as a least developed nation, increase of FDI is a necessary ingredient in bridge the gap of private capital deficiency. As a result, if FDI continues to decline in future the nation might continue to experience declining economic growth. Second, FDI as factor inputs cause increasing economic growth when the returns to scale are positive. In this respect, Rwanda as a least developed nation the declining economic growth could arise from low productivity of factor inputs. This indeed could a further mean that Rwanda’s technology could be low as a least developed nation and labour is not highly skilled. In this way if the trend continues, the nation will experience declining economic growth in the next one decade arising from the shocks on FDIs.

Further still, positive economic growth from FDI as factor inputs means that Uganda experiences an increasing returns to scale. According to the ASSM, the property of constant returns to scale demonstrates that even if factors inputs do not increase a nation can experience increasing returns to scale with stable macroeconomic conditions. This could be attributed to Rwanda’s macroeconomic stability. Although Literature indicates that Rwanda’s terms of trade continue to worsen into the future as explained by explained by low prices of Rwandan Agricultural exports and high prices of their manufactured imports Also, Consumer Price Index which is a key measure for macroeconomic stability indicates that inflation increased from 158 percent per annum in 1994 to 197.8.05 percent per annum in 2012. This trend if it continues into the future the country is likely to experience declining economic growth as predicted by the simulations.

Therefore, based on impulse response function projection, the country would experience increasing economic growth due to increased Net FDI inflow in the long-run. Second, due to poverty in Uganda employment and tourism will decline in the country. According to literature, harnessing the impact of FDI requires heavy investment in infrastructure and at tourism sites as a base for increasing tourists.

Further empirical reviews revealed that despite the impressive improvement in FDI inflows into Rwanda, achieving higher levels of economic growth is still concerns for government. Knowledge about the impact of FDI on economic growth in Uganda is limited. Regarding economic performance, previous studies have inconsistent findings on the contribution of FDI on Uganda’s economic growth. Charles Ruranga (2017) indicated that the contribution of FDI to Rwanda is insignificant, although positively related. As far as this research is concerned, no further economic analysis study has established the impact of FDI on Rwanda’s economic growth since then. Despite the increasing FDI and planned investment in the Country, there is little or no knowledge about the impact of FDI on Economics Growth in Rwanda. In particular, as far as this study is concerned, no economic analysis has investigated the impact of FDI on overall investment and employment in Rwanda.

Based on the Modelling the impact of FDI on Rwanda’s economic growth was based on the Solow-Swan Model. In Chapter Two, the study explained that the Rwandan Government adopted openness as a key commercial policy for investment and international trade. This was intended to promote exports, including tourism, and to attract FDI into the country. Accordingly, the chapter started by modelling openness, demonstrated through the theory of comparative advantage. According to the theory, when a country specializes first, production increases in the commodity of comparative advantage, which in turn increases investment. Second, exports increase in the sector of specialization, while the nation imports the commodity of comparative disadvantage. According to the findings, as the openness index increases, so do tourism and FDI flows. In the ASSM, openness is regarded as a government policy for trade and investment, and as such, innovation.

Inputs or resources for the nation, used during production in the ASSM. In this way, Government Expenditure contributes to production through infrastructure development, social service delivery and as a tool for employment creation in a nation. As a result, Government Expenditure would contribute to increased public investment, production and employment. Therefore, as factor inputs, a nation attains accelerated economic growth, domestic investment and employment among the population in the long-run. The study further modelled other variables, including inflation, telecommunications and civil war. In the ASSM, inflation is regarded as a government policy tool for macroeconomic stability. In this way, inflation affects economic growth and employment negatively, while it is positively related to inflation.

However, the contribution of factor inputs (FDI, tourists’ expenditure) on economic growth, and job creation is subject to the assumptions of the ASSM. First, constant returns to scale. This means that ceteris paribus with stable macroeconomic conditions, even if factor inputs do not increase, a nation experiences increasing economic growth. Second, positive and diminishing returns to factor inputs meaning that capital and labour factors are assumed to be positive but subject to diminishing returns. Therefore increasing TFP is a precondition for the nation to benefit positively from factor inputs, both local and foreign flows, such as FDI and tourism expenditure. Second, due to private capital deficiency, the essentiality property demonstrates that increasing foreign capital flows in form of FDI and tourists’ expenditure are important for Rwanda. Therefore, government policies that stimulate economic growth are important innovations for the country

9.4 Discussion of investment theories and Economic relevance to Rwandan Economy

In Chapter 4, the study explored a number of FDI theories ranging from Hymer’s IOT to market based theories such as the FDI capital theory, stage model theories and IPE theories. Considering these theories, internationalization is based on four conditions, termed in this study as firm-home-host-IPE (FHHIPE) conditions that explain FDI inflows to a developing nation such as Uganda. This is observations is first, based on tendencies of firms to leap frog low-commitment modes, or to jump immediately to psychically distant markets, as explained by Vahlne and Nordstrom. Leap-frogging can be indicated to mean firms’ decisions before investing abroad. FHHIPE conditions affect both the host and home country, and are key determinants for FDI inflows.

Second, the FHHIPE observation as explaining FDI inflows to a developing nation is based the behaviour and characteristics of FDI inflows into Rwanda since independence in 1963. According to literature, during the period 1963-1989 when Rwanda was peaceful, foreign investments dominated the economy except agriculture. When the country descended into political turmoil and economic instability during the period 1990-1994, the ethnic civil war, FDI inflows as well as Rwanda’s economic growth became negative. However, after overthrowing the Genocide, with peace and tranquility adoption of economic reforms, FDI inflows have tremendously increased and the economy has improved. In-turn the nation has witnessed a commendable achievement in Economic Growth. To this end over 90 countries and representing 48% of World Trade Organization particularly European Union member countries have established FDI projects in Rwanda since 1995

Considering the trends of Multi National Enterprises investments into Rwanda what is termed as the Frog-leap Theory in this study explains FDI inflows into the nation as a developing country. This is because first, Multi National Enterprises that internationalize develop capacity as explained by theories such as the eclectic theory and the stage model theories. Second, even when Multi National Enterprises have the capacity, to invest abroad depends on home and host country relations as well as role played by international actors such the UN, IMF and World Bank as explained by the IPE theory. In this regard, this study considers the behaviour of Multi National Enterprises to the frogs’ characteristics as amphibians. Frogs as amphibians the best habitant is the sea. When conditions are favourable on land frogs often leap to such environments especially during the rain-wet season. During drought when the environment is harsh, from land (unusual habitat), three options are possible. First, leap back to the sea. Second, frogs can leap to another rain-wet environment in the neighbourhood. Third, frogs hibernate if any of two options are not feasible to wait another wet season. This is characteristic of FDI inflows into Uganda since independence.

CHAPTER TEN:

CONCLUSIONS, POLICY IMPLICATIONS AND AREAS FOR FUTURE RESEARCH

10.1 Introduction

The main objective of this thesis has been to investigate the impact of investment policy on economic growth in Rwanda. In Chapter Two, the study first examined the historical, political, governance and economic perspectives. In this chapter, the economy of Rwanda was reviewed by exploring the trends in economic growth, as well as social indicators, including GDP. This was followed Chapter Three which presented a literature review on investment inflows into Rwanda since 1895 in the post genocide reforms. The subsequent Chapter Four explored the theories that explain investment phenomena across the globe. The literature review and exploration of this chapter set the basis for modelling economic growth in Chapter Four. Upon this background, Chapter five modelled Exports, imports, FDI and other explanatory variable (openness, tourism, telecommunication, and inflation) on economic growth.

10.2 Summary

10.2.1 Investment regimes in Rwanda

Empirical analysis of Literature review in Chapter Three indicates that investment as a source of capital (physical) is a historical phenomenon in Rwanda. Investments predates the post-colonial era which its background emanates from Ibrahim’s 1845 exploration works. He was the first known non-African to visit Rwanda. This resulted into establishment of trading links with the local population (local merchants). Precisely, Arabs traded items like guns and cloth for slaves and precious commodities, such as ivory.

This gave birth to twofold economic system, in which Europeans, particularly French merchants were invited to invest in plantation agriculture, in Rwanda. This policy marked the beginning of European investment inflows into Rwanda, building on the foundation that had been started by Arabs.

Based on the French Administrative foundation at Independence in 1963, it was evidently clear that the Rwandan economy would grow, given the circumstance under French occupation and stewardship. The new government introduced initiatives such as export promotion, imports (Physical capital and expatriates), and attracting FDI, foreign aid and mobilizing domestic tax revenue to lead to accelerated economic growth, and creation of jobs among natives. However, in the post-colonial era, Rwanda then started to experience political and economic instability among the two mainstream tribal blocks, (Tutsi vs Hutu) based on political patronage until 1994. After this, economic reforms were adopted. In particular, openness was adopted to promote trade and investment. The government then encouraged both domestic and foreign investors to invest in the country through deliberate investment regimes. Later, in 1998, the Rwandan Investment Code were established. The deliberate investment regimes in the post genocide resulted into enormous increase in investment from USD 19.8 million in 1995 to USD 8,015.6 million in 2014 and to USD 9,116.9 million in 2017 (RDB, 2018).

10.2.2 The Key investment Theories and their Economic Significance

In Chapter Four, the study explored a number of investment theories ranging from Hymer’s IOT to market based theories such as the investment capital theory, stage model theories and IPE theories. Considering these theories, internationalization is based on four conditions, termed in this study as firm-home-host-IPE (FHHIPE) conditions that explain investment inflows to a developing nation such as Rwanda. This is observations is first, based on tendencies of firms to leap frog low-commitment modes, or to jump immediately to psychically distant markets, as explained by Vahlne and Nordstrom. Leap-frogging can be indicated to mean firms’ decisions before investing abroad. FHHIPE conditions affect both the host and home country, and are key determinants for investment inflows.

Second, the FHHIPE observation as explaining investment inflows to a developing nation is based the behaviour and characteristics of investment inflows into Rwanda since independence in 1963. According to literature, during the period 1963-1992 when Rwanda experienced peaceful, foreign and local investments which was foreign investor led economy except agriculture. When the country descended into economic and political instabilities during the period 1992-1994, Rwanda’s investment inflows together with economic growth declined sharply. However, in the post genocide era, 1995 saw existence of peace and tranquility which provided necessary impetus for adoption of economic reforms, investment inflows have tremendously increased and the economy has improved. In-turn the nation has realized a sharp rise in GDP.

Bearing in mind the trends of Multinational enterprises investments into Rwanda what is termed as the Frog-leap Theory in this study explains investment inflows into the nation as a developing country. This is because first, Multinational enterprises that internationalize develop capacity as explained by theories such as the eclectic theory and the stage model theories. Second, even when MNEs have the capacity, to invest abroad depends on home and host country relations as well as role played by international actors such the UN, IMF and World Bank as explained by the IPE theory. In this regard, this study considers the behaviour of Multinational enterprises to the frogs’ characteristics as amphibians. Frogs as amphibians the best habitant is the sea. When conditions are favourable on land frogs often leap to such environments especially during the rain-wet season. During drought when the environment is harsh, from land (unusual habitat), three options are possible. First, leap back to the sea. Second, frogs can leap to another rain-wet environment in the neighborhood. Third, frogs hibernate if any of two options are not feasible to wait another wet season. This is characteristic of investment inflows into Rwanda since independence

10.2.3 Major Findings from Modelling the Impact of Exports, imports and FDI on Economic Growth

The study found out that imports of goods and services, export of goods and services and foreign direct investments net inflow are the greatest contributors of Rwanda’s economic growth as these were reported to contribute to a massive 99% variance in economic growth, when all other factor are kept constant all the three variables have a positive and significant relationship with economic growth. A short run relationship was established between economic growth and export of goods and services; and these two were found to be having a causal relationship as each was found to granger cause the other. It was also further found out that an unexpected shock to export of goods and services has a relatively small up and down effect on economic growth.

Also a short-run relationship was established between Import of goods and services and economic growth even when there was a minimal evidence of co-integration. Economic growth and Import of goods and services were also found to be having a causal relationship as each was found to granger cause the other. It was also further found out that an unexpected shock to Import of goods and services has a predictable and stable relationship with Economic growth. In regards to Economic growth and FDI net inflow one co integrated equation was found which resulted into establishment of a short-run relationship between Economic growth and FDI net inflow the two were also found to be having a causal relationship as each was found to granger cause the other. It was also further found out that an unexpected shock to FDI net inflow has a predictable and stable relationship with Economic growth. In general even with the reported great contribution of investment policy on economic growth in Rwanda, the established relationship was found to be short term.

10.3 Hypotheses Tests Major Findings

To test the hypotheses the study first modelled economic growth, and exports, imports, FDI and by employing theories that explain economic growth. In this respect, this study is based on the ASSM neoclassical growth theory. According to the theory, this study finds that production in a nation depends on factor inputs such as physical capital (Usually imported), human capital, labour and efficiency. In this regard, as households engage in gainful employment and investment, a nation’s economic growth accelerates. In turn, as employment increases, but subject to the properties of the Solow-Swan Model previously explained

Therefore, accelerated economic growth and investment in a developing economy like Rwanda depends on the assumptions of ASSM.

To test the hypotheses, a theoretical framework and empirical analysis procedure was established. The theoretical framework presented the relationship among variables. After, the variables were defined and the methods of measurement were provided. Meanwhile the procedure for hypotheses tested was based on four milestones each representing a chapter whose findings are presented under the subsequent sections. According to the procedure, the first milestone of analysis involved time series properties investigation. This section presented in Chapter Seven involved: data transformation, preliminary variables investigation, correlation analysis, unit root tests and endogeneity simulations.

The second part of the procedure is presented in Chapter seven & eight. This involved estimating the long-run and short-run relationship among the series as well as conducting ex-ante forecasting. To test for the long-run relationship among series cointegration analysis was conducted. This also provided the study with an opportunity to investigate that the series are cointegrated to same order (1). To establish the existence of a short-run relationship, the study used a VAR model employed in this study all variables are treated as endogenous, which becomes the basis upon which some variables are considered as endogenous or exogenous. Later, the established simultaneous equation was estimated as a VECM. Before estimating the short-run relation, the study validated the model for stability, autocorrelation and normality. After, employing VECM Granger causality approach, the model was simulated and followed by exante forecasting was conducted.

Further still, In Chapter seven & eight, the procedure for hypotheses testing involved estimating the simultaneous equation using OLS. The study findings demonstrate that using OLS provides efficient results. This is because first, the simultaneous equation is estimated based on the results of estimated by VAR, using VECM systems approach. Second, the series are non-stationary at level but stationary at first difference. Third, the series are cointegrated to the same order (1). Fourth, the roots of the companion matrix of the system lie inside the unit circle and are all less than one in absolute terms. Fifth, the number of cointegrating vectors among all variables is equal to the number of endogenous variables. In addition, the residual is tested for model stability, normality, variance and covariance. The results all indicate that data fits the model. Finally, all the equations in the system have the same three exogenous variables

Empirically, this study finds that the OLS estimator is equivalent to the generalized least square estimator when all equations have identical regressors to all equations in the system. Before estimating the four simultaneous equations by VECM Ganger causality approach, a model of VECM Granger causality test comprising of variables such as Export, imports and FDI. This was intended as a means of understanding the nature of causality among variables of the simultaneous equation. Finally, results the simultaneous equations were tested for stability, and serial correlation. This was followed by interpreting the findings for the equation

Final procedure for hypotheses tests involves presenting the findings and making conclusions. In sum, this study conducted a number of simulations presented in Chapters seven and eight. The findings are presented in the sections below upon which first, the main contributions of this thesis are presented. Second, based on these findings, policy implications and policy recommendations are provided. Third, the study also presents the study limitations on which areas for future study.

10.4 Findings on the Estimation Short-Run and Long-Run Relationship among Endogenous Variables and Policy implication on Rwandan Economy

In this study, Short-run and long-run simulations were conducted in Chapter eight, through impulse response and variable decomposition simulations. The findings of the short-run relation is as below

Basing on the short-run simulation, the study found that FDI does Granger cause economic growth in Rwanda. Further still, the study found that al the investment policy variables had short run effect on GDP growth rate of Rwanda’s economy based on the Pairwise Ganger causality results.

The purpose of this study was to establish the impact of investment policy on the growth of Rwanda’s economic. Though on the basis of the Pairwise Granger causality, we concluded that Export, Import and FDI Granger-cause economic growth. In this respect, based on the short-run Granger Causality simulation, the study concludes that investment policy does Granger cause economic growth in Rwanda.

Based on the finding on Impulse response, exports and FDI variables will cause an increase in economic growth, while import demand will cause a decline in GDP. FDI being physical capital and employment as labour are factors of production that generate output. As such, as output increases, so does economic growth which is contrary to our findings. Considering the fact that Rwanda as a least developed nation, increase of Investment inflows such as foreign capital is essential to bridge the gap of private capital deficiency. As a result, if export and FDI continues to decline in future the nation might continue to experience declining economic growth.

Further still, negative economic growth from factor inputs and high import expenditure inferentially implies that Rwanda experiences declining returns to scale, based on the ASSM, the property of constant returns to scale demonstrates that even if factors inputs do not increase a nation can experience increasing returns to scale with stable macroeconomic conditions. This could equally be ascribed to Rwanda’s macroeconomic instability explained in the literature till 1994 in the ethnic cleansing. Literature indicates that Rwanda’s terms of trade continue to worsen into the future as explained by NISR figure, 2008). Also, CPI indicates a key measure for macroeconomic stability indicates that inflation increased from 112.8 percent per annum in 1994 to 198.7 percent per annum in 2012. This trend if it continues into the future the country is likely to experience declining economic growth as indicated by the simulations.

The negative coefficients of imports, expenditure could be explained by the preliminary investigation conducted in Chapter seven &eight, empirical findings, Solow-Swan growth model and literature. According to the preliminary investigation, Rwanda’s economic growth is de minimis characterized by sharp declining fluctuations. It’s important to note that foreign capital is essential for country’s economic growth, essentially to bridge the gap in the private capital deficiency. However, Rwanda’s Exports and FDI inflows though increasing, often decline and fluctuate with wide magnitude. This could imply that Rwanda’s economic growth is dependent on foreign capital flows without which the economy could experiences declining economic growth. Inferentially, as foreign capital declines, economic growth is equally likely to follow similar trend.

Further still, productivity of factor inputs is important for a nation to experience positive returns to scale even if factor inputs do not increase. In this respect, as explained the preliminary investigation Rwanda experiences low TFP. This means that as a least developed nation, the country lacks modern advanced technology such as irrigation for agriculture and in-turn causing low factor productivity.

Equally important to note is that economic growth in a developing country like Rwanda depends on the absorption capacity for goods and services in the country (Domestic and foreign capital). Specifically, Rwanda’s absorptive capacity could be a concern for the nation due to domestic investment regimes. In this respect, consumption of goods and services produced is important implying heavy expenditure on imports, otherwise, a nation experiences declining economic growth. Lastly, in this study, empirical investigation also demonstrates that the growth in investment is greater than growth in output. This could inferentially imply that Rwanda is a least-developed country, with adequate skilled labour force. This could further be exacerbated by Rwanda’s population demographic factors, Labour force which growing faster than investment and economic growth.

10.5 Key Contributions of the study

This study provides theoretical and methodological contribution to understanding investment policy variables (Exports, imports, and FDI) as explanatory variables on Rwanda’s economic growth.

10.5.1 Theoretical and Empirical Contributions

This study has developed a theory that has been termed as the Frog-leap Theory as the theory explaining Investment inflows into a developing nation like Rwanda through (Exports and FDI). As previously explained the favourable natural habitat for frogs is the sea but can leap to land when conditions are good. During harsh conditions, frogs hibernate, but when the climate is favourable, they begin to jump. This behaviour explains (Exports and FDI) to a developing country like Rwanda, based on the conditions before and after independence during the period 1991–1994, when investment sharply declined. It should be recalled that, immediately after Rwandan independence Multinational enterprises greatly enjoyed the conducive political and economic environment in Rwanda. Secondly, between1991–1994, investment and GDP growth rate sharply declined due the prevailing conditions at the time orchestrated by the ethnic cleansing, meaning that Rwanda was not a favourable destination due to the unfavorable political conditions. Some investors relocated their investments to their home countries, including offshore investments, while others located to other countries, within the East African Community. Also, new investors who may have desired Rwanda as a first destination located investment elsewhere. Most of the Multinational enterprises that remained in Rwanda went hibernated. Finally, after the reforms from 1995, Investment and GDP growth rates experienced sharp upturn from USD 19.8 million in 1985 to over USD 952.6 million in 2015. In this way, building on Vahlne and Nordstrom, what is termed as the Frog-leap Theory, explains Investment, by taking into account Rwanda’s experience comparatively to that of other developing countries

10.5.2 Methodological Contributions

This study expounds on the thematic knowledge of the impact of Investment policy on Rwanda’s economic growth. The study enhances the knowledge of the impacts of the variables (Exports, Imports & FDI), brought together in the principle of endogenous variables. By bringing these variables into one conceptual framework, this study is pioneering. On record, there is no available empirical investigation on the contribution of investment policy(Exports, Imports & FDI) combined on Rwandan Economics growth except the works of Charles Ruranga (2018) on : Time series analysis of impact of FDI on GDP growth rate in Rwanda. Therefore, this research has created a new quantitative account of investment policy (Exports, Imports and FDI inflows) since 1997 as the explanatory variables on the contribution to Rwanda’s economy. In terms of scope, addition, previous studies covered a scope of less than 15 years but this study covers the period 1997–2018 (23 years), thus providing adequate and reliable basis for a robust conclusions.

As noted earlier, there was previously scanty or no knowledge of the impact of investment policy on Rwanda’s economic growth. Therefore, this gap in empirical work was the basis for this study motivation. Additionally, from the empirical literature reviewed, most studies adopted a linear regression model specification approach with only a single explanatory variable (FDI). Thus, to deviate from the previous studies, this study incorporated more new variables such as exports, and Imports in this study, thus, a multi-equation system model specification based on VAR through VECM procedure was used. Lastly, the VECM procedure enabled the study to capture the long-run relationships between the variables of interest and GDP growth rate.

10.6 Policy Implications to Rwanda

It’s important to note that in 1998, the Government of Rwanda introduced a number of deliberate monetary, fiscal and international trade policies with sole aim of enhancing the country’s economic growth. Summarily, these policies are now earmarked to provide the projected results. As indicated of findings this study, Rwanda is categorized as a least developed country experiencing enormous diminishing returns. In turn, the diminishing returns cause the economy to grow at a slow steady rate. As such, there is a need for deliberate investment policies that are indicative to prompting the economy into realistic, and sustainable economic growth over time.

As already demonstrated that Rwanda is a least-developed nation with limited domestic investment resources demonstrated by the low line graphs for the growth in net exports, and Net FDI as sources of investment capital, lying below GDP growth line graph, it’s important to note that all the factor inputs combined then cause diminishing returns since the economy is not growing proportionally to the factor inputs. Therefore, for policy purposes, the immediate implication is the ardent need is soliciting for more low cost of zero cost financial resources to be earmarked to bridge the investment capital-saving gap.

The low investment capital resource form Exports and FDI together with human capital as skilled labour, raise concerns for policy purposes. This study revealed that Rwanda’s exports is positive and significant, and positively contributes to GDP and imports, while insignificant to FDI, although it contributes to Foreign exchange which in turn translates to GDP. A further review indicates that the export contribution was very minimal, implying that this would translate into slow growth in GDP and low demand for exports, holding other factors constant

Although Rwanda’s monetary policy leads to increase exports, the cost is high. The findings indicate that inflation rates were major deterrent to export and FDI growth, which realistically affects economic growth in the long run. For policy purposes, the dual macroeconomic objectives of stabilizing the monetary and fiscal policy would go a long way in addressing concerns arising from the inflation shocks. There is also need to for heavy investment on innovations in order to cause accelerated economic growth, given the circumstance.

The findings further indicates that Rwanda’s investment policy is biased towards export promotion at the expense of import substitution. This is because to increase Rwanda’s exports, it’s important that Rwanda devalues its currency, this is analogous to deliberate credit expansionary policy. In macro-economic sense, this action simply implies that the greater the currency devaluation, the more exports; For the case of particularly inflation a higher inflation rate increases cost of living, thus, reducing economic growth due to high transaction exposure for multinationals and imports. For policy purpose, there is need to balance export promotion with import substitution.

In addition, from the OLS regression result, the study revealed that FDI is largest contributor to Rwanda’s GDP. However, in all simulations conducted FDI Granger causes GDP Minimally, further still, the impulse response function shows that the relationship is very minimal on over all GDP growth. Also due to high import expenditure on capital goods, Net FDI is likely to decline in the country. The study partly attributes this to high import demand, which requires huge investment in the import sector yet Rwanda is a least developed nation. These findings thus, pose policy implications for Rwandan Government bureaucrats since tourism is important for the country as an export commodity.

Furthermore, the ultimate objective for a least-developed country such as Rwanda is to GDP growth rate. In this regard, impulse response indicates that reducing increasing imports of goods and services in Rwanda causes increasing economic growth. However, this is not achievable, given the meagre resources for a least developed economy like Rwanda. This means that as explained by the VCP phenomenon for developing nations, Rwanda may experience more financial burden again in the long run by trying to meet the high import demands. Therefore, in addressing Rwanda’s economic future, this conundrum must be a major concern for Rwandan Government Bureaucrats since there is need to break the VCP in which the country is trapped.

The results from the OLS regression (see Table 7.3) reflect that in the long-run, Rwanda’s FDI will account for the largest fluctuations in all the variables incorporated in the model. As indicated, FDI in the long-run will account for 3.826% variation in economic growth. This is closely followed by variation in Imports for goods and services which is predicted to account for 1.639%, while exports for goods and services accounts for 1.402%. Based on the above result, it’s evidently clear that export and FDI inflows which would have been major predictors of GDP growth rate till offer very slow impetus to growth. Appropriate policies are therefore necessary to harness the trickledown effect of Exports and FDI to accelerate long run may Rwanda’s economic growth. Additionally, attracting more FDI through say, tourists is a way to access more financial resources not otherwise locally available in Rwanda.

10.7 Key Policy Recommendations

In this study, it was initially predicted that the empirical results of the study demonstrate that Exports, imports and FDI, would be very crucial to Rwanda’s GDP growth. Conversely, the study revealed that Export and FDI, though significant, the contribution were found to be very negligible in on short run GDP growth rate compared to that of imports. Also the study finds that according to impulse response forecast shows that exports and FDI have the potential of further decline in Rwanda, this could point to continuous fall in earning from key sources of foreign exchange, including decline in number of tourist to Rwanda and fall in demand for Rwanda’s exports.

Based on the empirical findings of this study and the policy implications, there is need to design and implement deliberate and targeted policies that stimulate economic growth in Rwanda by increasing exports, FDI inflows and limiting heavy expenditure on imports. First, the study proposes that Capital Absorption Capacity for Development (CACFD) would be suitable for a least-developed country like Rwanda. This study recommends CACFD because increasing factor inputs such as Exports and FDI is not enough, but increasing absorption capacity is quite important.

The findings revealed that diminishing returns affected Rwanda’s economic growth particularly in the export sector and minimally to FDI. Overcoming this situation requires CACFD, resting on two pillars: Capacity Support to community (CSC) and increasing foreign capital in the form of FDI and tourism. The CSC could be implemented in form of Micro Projects to Community (MPCs) where groups participated in various income generating activities. The MPCs would offer incentive increased output of goods and services that would enhance the growth of entrepreneurial skills and entrepreneurial efficacy in the long run. To avoid diminishing returns, CACFD involves building capacity through communal private or public private partnerships (PPPs) investments particularly in the rural areas and Peri-urban areas, as a means of supporting disadvantaged households. In this respect, absorption capacity for domestic capital from exports and foreign capital from FDI (mainly from tourism) is achieved. Supporting public investment implies creating jobs, and as such, factor inputs, including abundant human capital and natural resources, are utilized. In turn, a nation achieves a spiral of accelerated economic growth with limited stagnation, since jobs are continuously created and poverty continues to decline in the long-run.

From either side of the two pillars, the proposed CACFD model will enable Rwandan economy enhance the absorption capacity from both the domestic capital (export revenue) and foreign capital (FDI and tourism) as well as from the CSC. In this way, capacity development will be enhanced in Rwanda in the form of increasing production and consumption. Consequently, Rwandan economy would experience augmented economic growth and job creation, causing the country to break the VCP as poverty reduces in the long-run.

Further still, the notion of CACFD is underpinned by three key musts (critical success best practices), following the findings of this study;

First, the literature in Chapter two suggests that Rwanda’s economic growth is underpinned by financial resources from donors and domestic revenue. Domestic revenue mobilization through efficiency of the tax body (RRA) is essential. Similarly, Rwanda is not only a least-developed nation but is also a highly indebted country (World Bank report, 2017; NISR, 2016). Thus, donor support could finance the domestic financial resources gap.

Second, empirical evidence indicate that the monetary, fiscal and international trade policies cause Rwandan Government bureaucrat’s initiatives to decelerate economic growth in the long-run. In particular, in empirical reviews, Rwanda’s openness policy is negative to FDI and insignificant in the economic growth. These findings require a review of the regulatory framework and provisions for enforcement. Lastly, since Rwanda experienced ethnic related insurgency for a long time, deliberate policies on governance is a precondition for success, through snapshots on the rule of law, rendering zero tolerance to corruption (Review anti-corruption laws to render corruption a very risky undertaking), Creating a strong and vibrant inspectorate of Government to act as Government watch dogs in public institutions with express mandate of protecting public resources, identifying and prosecuting public officers found to have deliberately indulged in corruption related case in their offices, institutional capacity building and increased infrastructure development.

Third, the study recognizes that FDI is key to Rwanda’s economic growth and development. To this extent, a study is recommended to establish avenues through which FDIs can be promoted in the country. In particular the study review on policies regarding FDIs in Rwanda to make it attractive undertaking for both foreign and domestic investors to invest in Rwanda. This would start by reconstituting and empowering the existing investment body and giving it express mandate to lobby and manage the country’s investment hub.

Finally, based on the result of OLS regression result, a significant and negative relationship was reported between Import of goods and services and GDP ( β= -1.639, P>0.001 ) meaning that a unit increase in Import of goods and services GDP would fall by 1.639. Policy options to reduce heavy expenditure on imports through import substitution industrial development strategies would enhance the Country’s GDP growth rate in the long run

10.8 Key Limitations of this Study

This study examined the impact of exports, imports and FDI in Rwanda’s economic growth, employment and poverty reduction. In the course of the study, some limitations that came up could provide opportunities for future research. First, the study was limited by scope. This is because during the years of political and economic instability, data was not available (1980-1994) stretching up 1996 after the war. During the early1990s, most institutions collapsed and data from NISR could not be accessed. This was exacerbated by international sanctions and sustained ethnic clashes that kept the economy at bay, which caused political and economic instability and the breakdown of institutions notwithstanding the efforts of certain patriotic political protagonist to create a period of political and economic stability for new Rwandan between 1995 and 1997 when NISR once again experienced windows of hope to rebuild their data bases. Therefore, this study obtained most of its data from the World Bank database (world development index, 2016; 2017). In situations where data could be obtained from NISR, this was rendered alternative/ second /last resort option not withstanding their capacity.

Following the issue of limited data, the model sample size was limited to 23 years, although longer than any other study that had been conducted around the thematic area before. Our sample size was comprised of 23 annual observations for each variable in the system, ranging from 1997 to 2018. The main issue in sample size was the constrained degree of freedom in estimation considering the number of variables and lags. During estimation, sometimes issues of small samples arise that can affect the accuracy of results. To this end, the researcher would have wished to extend the sample to the 1980s, but there were no data on most of the variables, especially on exports, imports and FDI, due to international sanctions. This data limitation issue remained, but the 23 annual observations provided sufficient results to explain the impact of investment policy on the growth of Rwandan economy. Several simulations were conducted, and comparisons were made in an effort to cover such shortcomings.

The other visible limitation related to eventual modelling procedures. This limitation arises based on the study observation in during related empirical reviews. Literature indicates that during the period 1997–2015, Rwandan investment Authority registered a significant upsurge in investment projects close to 4328 accredited projects to fully operate, comprising of about 90 international entrepreneurs and about 60 local projects constituting about 150 projects,(RDB.2016) Questions arise that requires further study. First are the projects registered by Rwanda Development Board (RDB) still functional? Second, where are their current status in terms of growth and capacity? Third, to what extent have these projects contributed to Rwandan GDP growth rate? These questions would be answered through a Randomized Controlled trials method of study comprising time series and in survey that would constitute two cohorts, ( the Controlled group – X and the treatment group – Y) with the projects registered between 1997–2015 constituting the treatment group, while other project with the above scope considered as the controlled group .

10.9 Conclusion

A short run relationship was established between economic growth and export of goods and services; and these two were found to be having a causal relationship as each was found to granger cause the other. It was also further found out that an unexpected shock to export of goods and services has a relatively small up and down effect on economic growth.

10.10 Recommendations for Future Studies

First, his study recommends a survey to be conducted to evaluate projects recorded by RDB since 1997–2015 with specific reference to issues such as:

(i) RDB records of Rwandan GDP before and after implementation of these projects. As such, the survey would evaluate whether these project, if anything, contributed to Rwandan GDP growth.

(ii) The survey would also consider ascertaining the life-span of projects in Rwanda. In particular, ascertain number of the registered projects actually survives beyond their 5 ^th birth day, this provide basis for robust policy design for the survival and sustainability of future projects that locate their business to Rwanda. Data on their current location would equally provide adequate information to Government Policy bureaucrats on status of current Regional comparison on project distribution in the country for future planning Purposes.

Finally, fiscal, monetary and international trade policies were adopted to implement liberalization. A study is necessary to evaluate the effects of financial and commercial liberalization on Export promotion, import substitution, FDI, and economic growth. Specific study allusion would be to investigate the effects of exchange rate fluctuations and their effects on transaction exposure for Multinational enterprises in Rwanda.

REFERENCES

A.C. Edwards, M. Ureta (2003), International migration, remittances, and schooling: Evidence from El Salvador. Journal of Development Economics, 72 (2), 429-461

A.I.Rashid, (1995). Trade, Growth and Liberalization: the Indian experience, 1977-1989. Journal of Developing Areas, Volume 29, pp. 355-370.

A.M. Clemens, D. McKenzie (2014)“Why don't remittances appear to affect growth?” CGD Working Paper. Center for Global Development, Washington, DC, 366

Acquah, HD-G (2010), 'Comparison of Akaike information criterion (AIC) and Bayesian information criterion (BIC) in selection of an asymmetric price relationship', Journal of Development and Agricultural Economics, 2(1) 001-6

African Development Bank (2008). Rwanda 2008–2011, country strategy paper. African Development Fund. ‘Africa rising’, The Economist, 3 December 2011

Agarwal, JP (1980), 'Determinants of foreign direct investment: A survey', Weltwirtschaftliches Archive, vol. 116, no. 4, pp. 739-73.

Aggarwal et al., 2006. Do workers' remittances promote financial development? The Annals of the American Academy of Political and Social Science, 630 (1) (2008), pp. 224-244

Aitken, B, A. Harrison [1999], “Do domestic firms benefit from foreign direct investment. Evidence from Venezuela”, American Economic Review 86, 605-618

Aker, JC & Mbiti, IM (2010), 'Mobile phones and economic development in Africa', The Journal of Economic Perspectives, vol. 24(3), 207-232.

Akhtar, SI & Weiss, MA (2013), US international investment agreements: issues for Congress , Congressional Research Service.

Akinlo, A. (2004) "Foreign direct investment and growth in Nigeria: An empirical investigation," Journal of Policy Modeling 26 (16), 627–639

Akpansung, AO 2013, 'Motives and determinants of foreign direct investments: A survey of literature', European Journal of Humanities and Social Sciences , 28(1), 1478-95.

Alfaro, L (014), 'Foreign direct investment: Effects, complementarities, and promotion',Harvard Business School / National Bureau of Economic Research.

Alserud, M & Tykesson, D(2011), 'The Uppsala Model’s applicability on internationalization processes of European SMEs, today-A case study of three small and medium sized enterprises', Lunds University.

Alshahrani, SA & Alsadiq, MAJ (2014), Economic growth and government spending in Saudi Arabia: An empirical investigation , International Monetary Fund.

Andersen, P, Ahmad, SZ & Chan, WM (2014), 'Revisiting the theories of internationalization and foreign market entry mode: a critical review', International Journal of Business and Commerce, vol. 4(1), 37-86.

Anderson, E & Gatignon, H (1986), 'Modes of foreign entry: A transaction cost analysis and propositions', Journal of international business studies , 17(3), 1-26.

Anderson, GJ (1983), 'the internal financing decisions of the industrial and commercial sector: a reappraisal of the lintner model of dividend disbursements', economica, 50 (1)235-48.

Antonakis, J, Bendahan, et al (2014), 'Causality and endogeneity: Problems and solutions', in D.V. Day (Ed.), The Oxford handbook of leadership and organizations , Oxford University Press, New York, pp. 93-117.

Antwi, S & Zhao, X (2013), 'Impact of foreign direct investment and economic growth in Ghana: A cointegration analysis', International Journal of Business and Social Research , vol. 3(1), 64-74.

Apergis, N, Asteriou, et al (2012), 'the Explanatory Variables of Outward Foreign Direct Investment: Panel Evidence', American Journal of Economics and Business Administration , 4(4), 207.

Ardahaey, FT (2011), 'Economic impacts of tourism industry', International Journal of Business and Management , 6(8), 206.

Arrelano, M. and S. Bond (1991), “Some tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations”, Review of Economic Studies, Vol. 58, 277-297.

Arshia Amiri, Ulf-G Gerdtham, (2011). Relationship between exports, imports, and economic growth in France: evidence from cointegration analysis and Granger causality with using geostatistical models. Shiraz, Shiraz University Press.

Aruna, P. and H. Rajashekar (2016). Factors Influencing Investment Decisions of Retail Investors- A Descriptive Study. International Journal of Business and Management Invention, 5(12), 06-09

Asiedu, E (2006), 'Foreign direct investment in Africa: The role of natural resources, market size, government policy, institutions and political instability', The World Economy, 29(1), 63-77.

Aslam, M, Hassan, et al (2013), 'Foreign Direct Investment, Human Capital and Economic Growth in Malaysia', Journal of Economic Literature, no. I20, J40, O40, O49

Assadzadeh, A et al (2013), 'The Relationship between Foreign Direct Investment, Institutional Quality and Poverty: Case of MENA Countries', Journal of Economics, 1 (2), 161-5.

Athukorala, P (2003), 'the impact of foreign direct investment for economic growth: a case study in Sri Lanka', in 9th International Conference on Sri Lanka Studies , 92, 1-21.

Attari, MIJ, Kamal, Y et al (2011), 'The causal link between foreign direct investment (FDI) and economic growth in Pakistan economy', The Journal of Commerce , 3(4), 61.

Azpitarte, F (2011), 'Has economic growth in Australia been good for the poor? And what happens to the socially excluded?', Journal of Economic Literature , no. 3, I32.

Babula, R & Andersen, L 2008, 'The link between openness and long-run economic growth',

Bain, JS (1956), 'Barriers to new competition, their character and consequences in manufacturing industries', Cambridge, MA: Harvard University Press .

Balaam, DN & Veseth, M (2008), Introduction to international political economy , 4 edn, Pearson College Division, New Jersey, USA.

Balassa, B (1978), “Exports and Economic Growth”, Journal of Development Economics, Vol. 5, 181-189.

Balassa, B (1984), “Adjustment Policies in Developing Economics”, World Development, Vol. 12, 23-38.

Baldwin, RE (1995), The effects of trade and foreign direct investment on employment and relative wages , OECD Jobs Study Working Papers, No. 4, OECD.

Bandera, VN & White, JT (1968), 'US direct investments and domestic markets in Europe',

Bank, W., 2014. World Bank. [Online] Available at: http://www.worldbank.org/[Accessed 13 April 2014].

Barder, OM (2006), A policymakers' guide to Dutch disease, Center for Global Development, London.

Barlow, ER & Wender, IT (1955), foreign investment and taxation , Harvard University Law School.

Barro, R. J. (1991). Economic Growth in a Cross Section of Countries. The Quarterly Journal of Economics, 106(2), 407-443. http://dx.doi.org/10.2307/2937943

Barro, R. J., and Sala-i-Martin (1995), “Economic Growth”.

Barro, R.J., & Lee, J.W. (1993) Losers and Winners in Economic Growth. Proceedings of the World Bank Annual Conference on Development Economics, ed. M. Bruno and B. Pleskovic, 267-97.

Barro, RJ & Martin, XSI (2004), Economic growthSecond edn, Massachusetts Institute of Technology, London, England.

Barro, RJ 1990, 'Government spending in a simple model of endogenous growth', the journal of political economy, vol. 98, no. 5, Part 2, pp. S103-S25.

Barungi, BM 1997, Exchange rate policy and inflation: The case of Uganda , African Economic Research Consortium, Nairobi, Kenya.

Bategeka, L & Okumu, J (2010), Banking sector liberalization in Uganda: Process results and policy options , SOMO Centre for research on multinational corporations, Netherland.

Ben-David, D. (1998). Convergence Clubs and Subsistence Economies. Journal of Development Economics, 55(1), 155-171.

Benoit, K (2011), 'Linear regression models with logarithmic transformations', London School of Economics, London, vol. 22(1), 23-36.

Bernoulli, D (1954), 'Exposition of a new theory on the measurement of risk', Econometrica: Journal of the Econometric Society, 22(1), 23-36.

Bhagwati, JN (1969), 'Optimal policies and immiserizing growth', The American economic review , 59(5), 967-70.

Bhavnani, A, Chiu, RW-W, et al. (2008), 'The role of mobile phones in sustainable rural poverty reduction', Global Information and Communications , 4467, 1-25.

Bigsten, A & Kayizzi-Mugerwa, S(1999), Is Uganda an Emerging Economy , Department of Economics, Göteborg University, Göteborg, Sweden.

Biometrika, vol. 58, no. 1, pp. 1-19.

Blauw, SL & Franses, PH (2011), 'The impact of mobile telephone use on economic development of households in Uganda', Journal of Economic Literature , no. I3.

Blomstrom, M & Kokko, A (1997), Regional integration and foreign direct investment , National Bureau of Economic Research.

Blomström, M & Kokko, A (2003), 'The economics of foreign direct investment incentives’,Journalof Economic Literature, no. NO. J23, O12.

Bomuhangi, A, Doss, C & Meinzen-Dick, R (2012), Who Owns Land? Perspectives from Uganda , International Food Policy Research Institute.

Booth, D & Golooba-Mutebi, F (2015), Reforming the roads sector in Uganda: a six-year retrospective , Overseas Development Institute.

Brahmbhatt, M, Canuto, O & et al (2010), 'Dealing with Dutch disease', World Bank-Economic Premise, no. 16, pp. 1-7.

Braun, PA & Mittnik, S (1993), ‘Misspecifications in vector autoregressions and their effects on impulse responses and variance decompositions', Journal of econometrics, vol. 59 (3), 319-41.

Breusch, TS & Godfrey, LG 1986, 'Data transformation tests', The economic journal , 96, 47-58.

Briggs, BR (2009), 'Issues affecting Ugandan indigenous entrepreneurship in trade', African Journal of Business Management, 3(12), 786.

Brooks, DH, Fan, EX & Sumulong, LR (2003), foreign direct investment in developing Asia: trends, effects, and likely issues for the forthcoming WTO negotiations , Asia Development Bank, Manila, Philippines.

Bruhn, D 2014, 'Global Value Chains and Deep Preferential Trade Agreements: Promoting Trade at the Cost of Domestic Policy Autonomy?, Available at SSRN 2464136 .

Brunner, K & Meltzer, AH (1972), 'Friedman's monetary theory', Journal of Political Economy , vol. 80(5) 837-51.

Büge, M (2014), 'Do Preferential Trade Agreements Increase Their Members’ Foreign Direct Investment?', Journal of Economic Literature, no. F23; F36; F53; F55.

Burger, E (2003), 'Reinvestigating the wreck of the sixteenth century', University of Pretoria, South Africa.

Bwire, T & Nampewo, D (2014), 'Fiscal deficits financing: Implications for monetary policy formulation in Uganda', Journal of Applied Finance and Banking , vol. 4, no. 2, p. 125.

C. Amuedo-Dorantes, S. Pozo (2006). Remittance receipt and business ownership in the Dominican Republic. The World Economy, 29 (7), pp. 939-956

C. Woodruff, R. Zenteno (2007), Migration networks and microenterprises in Mexico. Journal of Development Economics, 82 (2), 509-528

C.W. Stahl, F. Arnold (1986) overseas workers' remittances in Asian development. International Migration Review, 20 (4), 899-925

Calvet, AL (1981), 'A synthesis of foreign direct investment theories and theories of the multinational firm', Journal of international business studies , vol. 12, no. 1, pp. 43-59.

Candela, G & Figini, P (2012), 'The economics of tourism destinations', in The Economics of Tourism Destinations, Springer, pp. 73-130.

Castel-Fodor, KJ (2013), 'providing a release valve: the US-China experience with the WTO dispute settlement system', Case W. Res. L. Rev. , vol. 64, p. 201.

Cata, E. (2000). Raising Growth and Investment in Sub-Saharan Africa: What Can be done? Policy Discussion Paper: PDP/00/4.

Caves R. (1996): Multinational enterprises and Economic analysis. 2nd edition. Cambridge. MA Cambridge University Press

Chao, W. S. and J. Buongiorno (2002), “Exports and growth: a causality analysis for the pulp and paper industries based on international panel data”, Applied Economics, Vol. 34, 1-13.

Charles, Ruranga (2013) Economic Development and Poverty Reduction Strategy 2013–2018: Shaping our Development. Kigali: Government of Rwanda

Chea, AC (2011), 'Sources of global private capital flows: What developing countries can do to attract, manage, and retain global private capital flows to finance economic growth and sustainable development', Business and Economic Research , 1(1), 1-20.

Chemouni, B. (2014) Explaining the design of the Rwandan decentralization: Elite vulnerability and the territorial repartition of power. Journal of Eastern African Studies, 8(2), 246-262

CHIEN-HUI LEE, BWO-NUNG HUANG, (2002). The relationship between exports and economic growth in East Asian countries: A multivariate threshold autoregressive approach. JOURNAL OF ECONOMIC DEVELOPMENT , 27(2), pp. 45-68.

Chimobi, OP (2010), 'Inflation and economic growth in Nigeria', Journal of Sustainable Development , 3(2), 159.

Choe, J.I. (2003). Do foreign direct investment and gross domestic investment promote economic growth? Review of Development Economics, 7(1), 44-57.

Choudhury, RN & Nayak, DN (2014), 'The link between trade and fdi: a study of SAARC Region', International Journal of Trade & Global Business Perspectives ,3(3), 1243.

Chowdhury, A & Mavrotas, G (2005), 'FDI and growth: a causal relationship', Journal of Economic Literature, 39(4), 32

Chowdhury, A (2002), 'Does inflation affect economic growth? The relevance of the debate for Indonesia', Journal of the Asia Pacific Economy , 7(1), 20-34.

Collier, P. and J.W. Gunning (1999), “Why was African Grown Slowly”, Journal of Economic Perspectives , Vol. 13, No. 3, 3-22.

Collier, P., & Gunning, J.W. (1999) Explaining African Economic Performance. Journal of Economic Literature, 37(1), 64-111.

Connolly, R (2015), Troubled times: Stagnation, sanctions and the prospects for economic reform in Russia , Chatham House, London.

Connors, J (2011), Regularizing and Augmenting the World Bank Poverty Rates , Wake Forest University.

Corden, WM & Neary, JP (1982), 'Booming sector and de-industrialization in a small open economy', the economic journal, 92(368), 825-48.

Cottarelli, C (2011), 'Revenue mobilization in developing countries', International Monetary Fund , 8 (2011), 8-10.

Craigwell, R (2006), foreign direct investment and employment in the English and Dutch- Speaking Caribbean , Project prepared for ILO, Trinidad and Tobago.

D. Kapur (2004),Remittances: Development Impact and Future Prospects Harvard University, Massachusetts, 26(33), 331-360

D. Rodrik (2016). An African growth miracle? Journal of African Economies, 27 (1), 10-27

D.Rodrik, (1996). Understanding Economic Policy Reforms. Journal of Economic Literature, Volume 34, pp. 9-41.

D.Salvatore, T.Hatcher, (1991). Inward and Outward Oriented Trade Strategies. Journal of Developing Studies, Volume 27, pp. 7-25.

Darrat, A. F., (1986). TRADE AND DEVELOPMENT: THE ASIAN EXPERIENCE. In: Louisiana: Louisiana Technical University Press, pp. 695-699.

De Vorsey, L (1987), 'the new land: The discovery and exploration of eastern North America', North America: The Historical Geography of a Changing Continent, edited by Robert D. Mitchell and Paul A. Groves. Towota, NJ: Roman & Littlefield .

Dee, P, Francois, et al. (2011), the impact of trade liberalization on jobs and growth', Journal of Economic Literature, no. F12, F13, F16, F4.

Demello, L.R. (1997). Foreign direct investment in developing countries: A selective survey. Journal of Development Studies, 34(1), 1–34.

Denisia, V (2010), 'foreign direct investment theories: An overview of the main FDI theories',

Dewan, E & Hussein, S (2001), Determinants of economic growth (Panel data approach) , Citeseer.

Dickey, DA & Fuller, WA (1979), 'Distribution of the estimators for autoregressive time series with a unit root', Journal of the American statistical association , 74(366), 427- 31.

Dilawar Khan, Azra, Maryum Umar, Nasreen Zaman, Ejaz Ahmad, Yusra Shoukat, (2012). Exports, Imports and Economic Growth Nexus: Time Series Evidence from Pakistan. Work Applied Sciences Journal, 18(4), pp. 538-542.

Dima, SC (2010), 'From international trade to firm internationalization', European Journal of Interdisciplinary Studies, 2(2), 59-66.

Dodd Parrish, E, Cassill, NL and Oxenham, W (2014), 'Opportunities in the international textile and apparel marketplace for niche markets', Journal of Fashion Marketing and Management: An International Journal , 8(1), 41-57.

Dollar, D & Kraay, A (2002), 'Growth is Good for the Poor', Journalof economic growth7 (3), 195-225.

Domar, ED (1946), 'Capital expansion, rate of growth, and employment’, Econometrica, Journal of the Econometric Society, 3(18)137-47.

Doss, C, Meinzen-Dick, R & Bomuhangi, A (2014), 'Who owns the land? Perspectives from rural Ugandans and implications for large-scale land acquisitions', Feminist economics, 20(1), 76-100.

Durbin, J & Watson, G (1971), 'Testing for serial correlation in least squares regression. III',

E. A., & Selvanathan, S. (2008). Foreign Direct Investment, Domestic Investment, and Economic Growth in China. A Time Series Analysis, Research Paper No. 2008/19.

E. Valeriani, S. Peluso (2011) the impact of institutional quality on economic growth and development: An empirical study. Journal of Knowledge Management, Economics and Information Technology, 1 (6), 1-25

E.B. Udah (2011) Remittances, human capital and economic performance in Nigeria. International Journal of Human Development and Sustainability, 4 (1), 103-117

E.K. Addison (2004).The macroeconomic impact of remittances.Journal of Development Studies, 26, 387-407

E.M. Nyamongo, R.N. et al. (2012) Remittances, financial development and economic growth in Africa. Journal of Economics and Business, 64 (3), 240-260

Eaton, J. and S. Kortum (1997), “Engines of Growth: Domestic and Foreign Sources of Innovation”, Japan and the World Economy, Vol.9, 235-259.

Eaton, J. and S. Kortum (1999), “International Patenting and Technology Diffusion: Technology and Measurement”, International Economic Review ,40, 537-570.

Eaton, J., E. Gutierrez and S. Kortum (1998), “European Technology

Econometrica: Journal of the Econometric Society, 54(3), 761-6.

Edwards, TH 2007, 'Returns to education and the Mankiw-Romer-Weil result', Economics Bulletin, 15 (24), 1-8.

Egbo, OP (20120, 'Bivariate causality analysis on the impact of fdi inflows and economic growth in Nigeria', Journal of Internet Banking and Commerce , 16 (3), 1.

Emine KILAVUZ (2012) .Export and Economic Growth in the Case of the Manufacturing Industry.

Enders, W (1995), 'Applied econometric time series: Wiley series in probability and mathematical statistics', Applied econometric time series: Wiley series in probability and mathematical statistics .

Engle, RF & Granger, CW (1987), 'Co-integration and error correction: representation, estimation, and testing', Econometrica: Journal of the Econometric Society , 12(19), 251-76.

Ericsson, NR, Hendry, DF & Mizon, GE (1998), 'Exogeneity, cointegration, and economic policy analysis', Journal of Business & Economic Statistics , 16(4), 370-87.

Espinosa-Vega, MA & Russell, S (1997), 'History and theory of the NAIRU: a critical review',Economic Review-Federal Reserve Bank of Atlanta, 82(2). 4.

European Journal of Interdisciplinary Studies, 2 (2), 1-7.

Evans, MK (2004), Macroeconomics for managers, First Edition edn, Blackwell Publishers Ltd, Melbourne, Australia,22(19),125-137

Falvey, R. N. Foster and D. Greenaway (2002), “North-South, Technology spillovers and Growth”, working paperno, 15/2002, European Economy Group(EEG).

Fan, S & Zhang, X 2008, 'Public expenditure, growth and poverty reduction in rural Uganda',

Fölscher, A 2006, 'Introduction: African Experience with MTEF Budget Reform', OECD Journal on Budgeting, 6(2), 1.

Ford, J, Sen, S & Wei, H (2010), 'A simultaneous equation model of economic growth, FDI and government policy in China', Department of Economics discussion paper .

Foster, J, Greer, J & Thorbecke, E (1984), 'A class of decomposable poverty measures',

Francois, J. (1990), “Producer services, scale and division of labour”, Oxford Economic Paper, Vol. 42, 715-729.

Friedman, M (1970), 'A theoretical framework for monetary analysis', Journal of Political Economy , 78(2), 193-238.

Fripp, C. (2009,) Rwanda speeds up Internet connectivity in Kigali. Economic Research Consortium, Kigali 21(26), 223-248

Frontier”,working paperNo. 2815, CEPR.

Froot, KA & Stein, JC (1991), Exchange rates and foreign direct investment: an imperfect capital markets approach , National Bureau of Economic Research Cambridge, Mass., USA.

G.K.Nidugala, (2000). Exports and Economic Growth in India: An Empirical Investigation. Indian Economic Journal, Volume 47, pp. 67-78.

Gamal, N (2008), 'How to attract foreign direct investment to invest in housing in Libya', University of Salford, Manchester UK.

Gascoigne, J (2000), 'Motives for European exploration of the Pacific in the age of the enlightenment', Pacific Science, 54(3) 227.

Ghirmay, T, R. Grabowski and S. C. Sharma (2001), “Exports, investment, efficiency and economic growth in LDC: an empirical investigation”, Applied Economics , Vol. 33, 689-700.

Ghura, D., & Hadjimichael, T. (1996). Growth in Sub-Saharan Africa, Staff Papers. International Monetary Fund, 43(3), 605-634. http://dx.doi.org/10.2307/3867556

Grosspietsch, M. (2004) Perceived and Projected Images of Rwanda: Visitor and International Tour Operator Perspectives, Tourism Management, 27, 225-234

Grosspietsch, M. (2005). The Image of Rwanda as a Tourist Destination, Working Paper Series, Working Paper No.4

Gui Dilby S. L (2014) “Impact of foreign direct investments on economic growth in Africa: Evidence from three decades of panel data analyses” research in economics vol. 68 p.248-256.

Gujarati, N. D., & Porter, D. C. (2009). Basic Econometrics (5th edition). New York: McGraw-Hill/Irwin.

Güngör Turan and Bernard Karamanja (2014) An Empirical Study on Import, Export and Economic Growth in Albania. Academic Journal of Interdisciplinary Studies. Vol 3 No 3. E-ISSN 2281-4612 ISSN 2281-3993

H.J. Chang (2011), Institutions and economic development: Theory, policy and history. Journal of Institutional Economics, 7 (4) pp. 473-498

Hamilton, J. D. (1994). Time series analysis, Princeton. New Jersey: Princeton university press. Hernandez-

Hoekman, B. and C. P. Braga (1997), “Protection and Trade in Services: A Survey”,Open Economic Review, Vol. 8, 285-308.

Hoekman, B. and S. Djankov (1997), “Effective Protection and Investment Incentives in Egypt and Jordan: Implications of Free Trade with Europe”, World Development, Vol. 25, 281-291.

Holtz E., Douglas, W., Harvey S. R.(1988), “Estimating Vector Auto regressions with Panel Data,” Econometrica 56, No. 6, 1371 - 1395.

IMF working papers no 06/2, DC. International Monetary Fund, Washington (2006)

International Journal of Economics and Financial Issues Vol. 2, No. 2, 2012, pp.201-215 ISSN: 2146-4138

J. Benhabib, M.M. Spiegel (1994).The role of human capital in economic development evidence from aggregate cross-country data. Journal of Monetary Economics, 34 (2), 143-173

J. Ferdaous (2016). Impact of remittances and FDI on economic growth: A panel data analysis.Journal of Business Studies Quarterly, 8 (2), 58

Jin, J (2002), “ Exports and growth: is the export-led growth hypothesis valid for provincial economies?”, Applied Economics, Vol. 34, 63-76.

Journal of International Commerce and Economics, 2, 31-50.

Joutsen.T. Norling H. (2014) Does FDI affect GDP per-capita growth in sub-Saharan Africa?

Jung, W. and P. Marshall (1985), “Exports, Growth and Causality in Developing Countries”, Journal of Development Economics, Vol. 18, 1-12.

K. Matuzeviciute, M. Butkus (2016) Remittances, development level, and long-run economic growth. Economies, 4 (4), 28

Kavoussi, R. (1984), “Export Expansion and Economic Growth: Further Empirical Evidence”, Journal of Development Economics, Vol. 18, 1-12.

Keller, W (2001a), “International Technology Diffusion”, working paper No. 8573, NBER.

Keller, W (2001b), “Knowledge Spillovers at the World’s Technology

Keller, W (2002), “Geographic Localisation of International Technology Diffusion”,American Economic Review, Vol. 92, No.1, 120-142.

Kerosi, J.B. and E. Rwangarinde (2014). Assessment of capital market potentials: unlocking financial investment opportunities in Rwanda, A case study of Capital Market Authority. International Journal of Information Technology and Business Management, 26 (1) 85-105.

Khan, Fatima, et al. (2015). Factors Influencing Investors’ Decisions in Stock Market Investment in Bangladesh: A Study on Khulna City. Journal of Finance and Accounting, 3(6), 198-204

Khan, M.S., & Reinhart, C.M. (1990). Private Investment and Economic Growth in Developing Countries. World Development, 18(1), 19-27.

Kogid, M., Mulok, D. et al. (2011). Does import affect economic growth in Malaysia? The Empirical Economic Letters, 10(3), 298-307.

L.A. Winters, P.M. Martins (2004) when comparative advantage is not enough: Business costs in small remote economies. World Trade Review, 3 (3), 347-383

Laszlo Konya, Jai Pal Singh, (2006). Exports, Imports and Economic Growth in India. Melbourne, La Trobe University Press.

Lee, C., Chang, C., et al (2008), Energy-income causality in OECD countries revisited: The key role of capital stock. Energy Economics, 30, 59-73.

Levine, R. and D. Renelt (1992), “A sensitivity Analysis of Cross-Country Growth Regressions”, American Economic Review, Vol. 82, 942-963.

M. Arellano, S. Bond (1991).Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies, 58 (2), 277-297

M. Shafiq, U.I. Haq, A. Khan, S. Khan (2012)the role of foreign remittances and economic growth in poverty alleviation: Time series evidence from Pakistan. World Applied Sciences Journal, 19 (3), 366-369

M., Jalaee, A., Sadeghi, A. (2005), the effects of globalization on the pattern of Iranian imports. Research of Economics, 105, 11.

M.A. Clemens, T. Ogden (2014), Migration as a strategy for household finance: A research agenda on remittances, payments, and development. Center for Global Development Working Paper, Vol. 354, DC, Washington, 20036.

M.A., Quising, P., et al (2006). How much does investment drive economic growth in China? Journal of Policy Modeling, 28(7), 751-774.

M.S. Atanda, O. Charles (2014),Inward remittances and economic growth in Sub-Saharan African countries: application of panel cointegeration approach. Research Journal of Finance and Accounting, 5 (12), 105-115

Mahmoodzadeh, M., Mohseni, R. (2006), the effect of the imported technology on economic growth. Economics Research, 5(16), 103-130.

Mann, L., & Berry, M. (2015) Understanding the Political Motivations That Shape Rwanda's Emergent Developmental State. New Political Economy, 1-26

Matto, A., R. Rathindran and A. Subramanian (2001), “Measuring Services Trade Liberalisation and Its Impact on Economic Growth: An Illustration”, working paperNo. 2655, World Bank.

Mazumdar, J. (2001), imported machinery and growth in LDCs. Journal of Development Economics, 65(1), 209-224.

Mehdi Taghavi, Masoumeh Goudarzi, Elham Masoudi, Hadi Parhizi Gashti,(2012). Study on the Impact of Export and Import on Economic Growth in Iran. Journal of Basic and Applied Scientific Research, Volume 2, pp. 12787-12794.

Michaely, M. (1977), “Exports and Growth: An Empirical Investigation”, Journal of Development Economics, Vol. 4, 49-54.

MINALOC. (2013) Revised Decentralization Policy. Kigali: Ministry of Local Government. Ministry of Finance and Economic Planning. (Rwanda Vision 2020.

Moschos, D. (1989), “Export Expansion, Growth and the Level of Economic Development”, Journal of Development Economics, Vol. 15, 99-102.

N. Catrinescu, M. Leon-Ledesma, M. Piracha, B. Quillin (2009) Remittances, institutions, and economic growth. World Development, 37 (1), 81-92

N. Catrinescu, M. León-Ledesma, M. Piracha, B. Quillin“Remittances, institutions, and economic growth,” IZA discussion paper no. 2139

N.P. Glytsos (2002).The role of migrant remittances in development: Evidence from Mediterranean countries. International Migration, 40 (1), pp. 5-26

National Bank of Rwanda (2012) foreign private investment in Rwanda, Kigali.

Ndikumana, L. (2000). Financial Determinants of Domestic Investment in Sub-Saharan Africa. World Development 28(2), 381- 400.

Obamuyi, T.M. (2013). Factors influencing investment decisions in capital market: a study of individual investors in Nigeria. Organizations and Markets in Emerging Economies, 4(1). 141-161.

Omo, A. and S.O. Mbadiugha (2012). Factors Influencing Investors Decisions in Shares of Quoted Companies in Nigeria. Pakistan Journal of Applied Economics, 22 (1 & 2): 53-65. Qureshi,

P.M. Romer (1986) Increasing returns and long-run growth. Journal of Political Economy, 94 (5), 1002-1036

P.M. Romer. ) (1990) Endogenous technological change. Journal of Political Economy, 98 (5, Part 2, pp. S71-S102

Policy”,Economic Policy, Vol 27, 405-438.

Productivity Performance in Developing Countries, UNIDO Research Programme.

ProInvest. (2007, June 27–29) Regional investment conference tourism: East Africa and the Indian Ocean: Country profile: Rwanda. Retrieved January 31, 2011

R. Chami, C. Fullenkamp, S. Jahjah (2005) Are immigrant remittance flows a source of capital for development? IMF Staff Papers, 52 (1), 55-81

R.E. Lucas (2005)International migration and economic development: Lessons from low-income countries. Edward Elgar Publishing. Sweden, 238-247

R.E. Lucas, O.Stark (1985).Motivations to remit: Evidence from Botswana. Journal of Political Economy, 93 (5), 901-918

R.H. Adams Jr (2011).Evaluating the economic impact of international remittances on developing countries using household surveys:Journal of Development Studies, 47 (6), 809-828

R.H. Adams Jr., J. Page (2005), Do international migration and remittances reduce poverty in developing countries? World Development, 33 (10), 1645-1669

R.R. Nelson, E.S. Phelps (1966) Investment in humans, technological diffusion, and economic growth. American economic association. 56 (19), 69-75

Robert Z. Lawrence, David E. Weinstein, (1999). TRADE AND GROWTH: IMPORT-LED OR EXPORT-LED? EVIDENCE FROM JAPAN AND KOREA. Massachusetts, Cambridge University Press. T.Velnampy, S.Achchuthan, (2013). Export, Import and Economic Growth: Evidence from Sri Lanka. Journal of Economics and Sustainable Development, 4(9), pp. 147-155.

Romer, D. (1996), Trade and Growth in East Countries: Cause and Effect? Berkeley, CA: Department of Economics University of California. Sameti, 44(18), 336-348

Rwanda (2005) Diagnostic Trade Integration Study. REPUBLIC OF RWANDAKIGALI, 2005.26(13), 113-132

S. A. (2002). State-Level Performance under Economic Reforms, Economic Policy Reforms and the Indian Economy, ed. Anne O. Krueger (Chicago and London: University of Chicago Press. National Institute of Statistics of Rwanda

S. De, E. Islamaj, S. Yousefi (2015). Can remittances help promote consumption stability. Global Economic Prospects, 4, 221-238

S.A., and Hunjra, A. I. (2012). Factors affecting investment decision making of equity fund managers, Wulfenia Journal, 19(10), 280-29

S.K. Mathur (2005). Absolute convergence, its speed and economic growth. Korea and the World Economy, 6 (2), 245-273

S.R. Gitter, B.L. Barham (2007). Credit, natural disasters, coffee, and educational attainment in rural Honduras. World Development, 35 (3), 498-511

S.S. Makki, A. Somwaru. Impact of foreign direct investment and trade on economic growth: Evidence from developing countries. American Journal of Agricultural Economics, 86 (3) (2004), 795-801

S.W. Theodore Nobel Lecture: The Economics of Being Poor. Journal of Political Economy, 88 (4) (1980), 639-651

Sala-i-Martin, X. (1997), “I just ran two million regressions”, American Economic Review, Vol. 87, 178-183.

Salvatore, D. and Hatcher, R. (1991), “Inward and Outward Oriented Trade Strategies”, Journal of Development Studies, Vol. 27, 7-25.

Sargan, J. D. (1958), “the Estimation of Economic Relationships Using Instrumental Variables”, Econometrica, Vol. 26, 393-415.

Singh, Satwinder, et al. (2012) "Foreign direct investment & the role of institutions in Rwanda." African Journal of Business and Economic Research 7.2 & 3 (2012): 81-96

Stock, J. H., & Watson, M. W. (1990). Inference in Linear Time Series Models with some Unit Roots. Econometrica, 58, 113-144.

Straubhaar, 1988

Summers, L(1999), “Reflections on Managing Global Integration”, Journal of Economic Perspectives, Vol. 13, 3-18, Spring.

T. Brambor, W.R. Clark, M. Golder (2006). Understanding interaction models: Improving empirical analyses. Political Analysis, 14 (1), 63-82

Tan B. W., & Lean H. H. (2010). An analysis of dynamic linkages between domestic investment, exports and growth in Malaysia. European Journal of Social Sciences, 16(1), 150-59.

Tehranchian, A. (2006), The effect of imports on economic growth. Political and Economic Information. Madison, WI: University of Wisconsin. 27(13), 1387-1399.

Tyler, W.G. (1981), “Growth and Export Expansion in Developing Countries”, Journal of Development Economics, Vol. 9, 121-130.

U.N. Balasubramanyam, M. Salism, D. Sapford (1999).Foreign Direct Investment as an Engine of Growth. Journal of International Trade and Economic Development, 8, 29-40

U÷ur, A., (2008). Import and Economic Growth in Turkey: Evidence from Multivariate VAR Analysis. EAST-WEST Journal of ECONOMICS AND BUSINESS , 11(1-2), pp. 54-75. Vika, I., 2006. Measuring Import and Export Functions in Albania , Tirana: Banka e Shqiperise.

Uvin, P. (2005), African Affairs; the Development Enterprise in Rwanda, West Hartford, CT: Kumarian Press. 104(416), 375-398

Villa, S. (2008). Economic growth, investment and government consumption in Italy: A VAR analysis. ICFAI Journal of Applied Economics, 7(4), 23-32.

Whalee, L. (1995), Capital goods imports and long run growth. Journal of Development and Economy, 48, 91-110.

World Bank (2013) Doing business 2013 - Smarter regulations for small and medium size enterprises, Washington.(26(11), 221=278

World Development, (2012), Are remittances a curse? Development publication bureau 40 (4) 657-666

Y. Abdih, J. Dagher, et al. (2008) Remittances and institutions: International Monetary Fund. 8(29). 178-197

Y. Baldé (2009) Migrants' remittances and economic growth in sub-Saharan Africa. Laboratoired'Analyse et de Prospective Economique (LAPE), Universite de Limoges, France.34(13) ,116-128

Y. Baldé (2011).The impact of remittances and foreign aid on savings/investment in sub-Saharan Africa. African Development Review, 23 (2), 247-262

Zhang K.H. (1999) FDI and economic growth: evidence from ten East Asian Economies. Economia International. Vol. L11. No 4. pp 517-535

Excerpt out of 330 pages - scroll top

Details

Title: Investment policy and growth of an economy. Rwandan economy, 1997-2018
College: University of Rwanda (University of Rwanda)
Author: George Butare (Author)
Publication Year: 2021
Pages: 330
Catalog Number: V1434280
ISBN (PDF): 9783346988539
ISBN (Book): 9783346988546
Language: English
Tags: economy GDP FDI Rwanda growth investment

Quote paper: George Butare (Author), 2021, Investment policy and growth of an economy. Rwandan economy, 1997-2018, Munich, GRIN Verlag, https://www.grin.com/document/1434280

Investment policy and growth of an economy. Rwandan economy, 1997-2018

Excerpt

Abstract

Acknowledgements

Abreviations

CHAPTER ONE:

GENERAL INTRODUCTION OF THE STUDY

1.1 Introduction

1.2 Background of the study

1.3 Statement of the Problem

1.4 Purpose of the study

1.5 Specific Objectives

1.6 Research Hypothesis Development

1.7 The study scope

1.8 The significance of the Study

1.9 Research methodology

1.10 Thesis Structure

CHAPTER TWO:

OVERVIEW OF RWANDA’S POLITICAL, GOVERNMENTAL

AND ECONOMIC HISTORY.

2.1 Introduction

2.2 Summary of Rwandan Economy

2.3.1 Political Context

2.3.2 Social Context

2.3.3 Economic Overview

2.3.4 Development Context

2.4 State of the Rwandan Economy

2.5 Rwandan Key Economic Reforms

2.5.1 Rwanda’s Economic Policy Reforms, 1960-2008

2.5.2 Macro-economic Indicators

2.5.3 Inflation Control and Economic Reforms in Rwanda

2.5.4 Impact of Inflation on Economic Growth in Rwanda

2.6 Causes of Inflation in Rwanda

2.6.1 Food and Transport Costs

2.6.2 Rwanda’s Fiscal Deficit and Monetary Factors

2.6.3 Aggregate Demand and Cost Factors

2.6.4 External Factors

2.7 Assessing the relevance of Openness in Rwanda’s Economy

2.7.1 Contribution of Openness to the Current Account

2.7.2 The Contribution of Openness to Production and Productivity

2.8 Challenges to reforms in Post genocide

2.9 Rwanda’s Economic Growth after adopting Reforms 1994–2017

2.10 Conclusion

CHAPTER THREE

EXPORT OUTFLOWS, IMPORT INFLOW AND FDI IN RWANDA

3.1 Introduction

3.2 Export Performance before and after the reforms in Rwanda

The new urgency for export growth

3.3 Empirical review on relationship between export dynamic and GDP growth rate

3.3.1 Import Flows after the reforms in Rwanda

3.3.2 Imports demand in Rwanda in the recovery era

3.4. Analysis of the relationship between Import demands s and economic growth in Rwanda

3.5 Empirical reviews on imports of goods and services and GDP growth rate

3.6 Review of FDI Inflows into Rwanda before and after Economic Reforms

3.4.1 FDI Inflows into Rwanda before the independence

3.4.2 FDI Inflows into Rwanda after economic reforms

3.4.3 Trends of FDI in Rwanda after Economic reforms

3.4.4 Foreign Direct Investment and economic growth in Rwanda

3.5 Investment Regulations in Rwanda

CHAPTER FOUR

REVIEW OF KEY THEORIES OF GROWTH AND INVESTMENT:

THEORIES AND ECONOMIC IMPORTANCE

4.1 Growth Theories for the study

4.2 Modelling Economic Growth

4.2.1 The Solow-Swan Model

Properties of the Neoclassical Solow-Swan Model

Constant return to scale

Positive and diminishing returns to factor inputs

Inada conditions for equilibrium production conditions

Essentiality

The Steady State of Growth

4.3 Mankiw, Romer and Weil Model

4.4 New GrowthTheory

4.5. Measuring Economic Growth

4.6 Key Investment theories

4.6.1 Introduction

4.6.2 Definitions

4.6.3 Portfolio Investment

4.6.4 Foreign Direct Investment

4.7 The Origin of FDI Theories