DETERMINANTS OF PRIVATE INVESTMENT IN ETHIOPIA (FROM

1990 TO 2021)

MSc THESIS

By

SORA KULA MOLU

BULE HORA UNIVERSITY

JUNE 2024
BULE HORA, ETHIOPIA
DETERMINANTS OF PRIVATE INVESTMENT IN ETHIOPIA (FROM
1990 TO 2021)

A THESIS SUBMITTED TO DEPARTMENT OF ECONOMICS,

COLLEGE OF BUSINESS AND ECONOMICS, SCOOL OF
GRADUATES STUDIES, BULE HORA UNIVERSITY

FOR THE DEGREE OF MASTER OF SCIENCE IN DEVELOPMENT

ECONOMICS.

SORA KULA MOLU

MAJOR ADVISOR: GIRUM DAGNE (PH.D)

CO-ADVISOR: DABA D (MSC)

JUNE 2024
BULEHORA, ETHIOPIA
 Declaration
I hereby declare that “The Determinants of Private Investment in Ethiopia” is my
original work and has not been presented for a degree in any other university, and all
sources of material used for this MSc thesis have been duly acknowledged by
means of complete reference.

Submitted by:
Full Name: --------------------------- Signature-------------------------------Date----------------

Approved by:

This thesis has been submitted for examination with my approval as advisor.

Name of Advisor----------------------------Signature--------------------------Date---------------

i
 Approval Sheet

BULEHORA UNIVERSITY

SCHOOL OF GRADUATE STUDIES

DEPARTMENT OF ECONOMICS

EXAMINERS’ APPROVAL SHEET

We, the undersigned, members of Board of Examiners of the final open defense by
Sora Kula Molu, have read and evaluated his thesis entitled ‘Determinants of Private
Investment in Ethiopia’, and examined the candidate. This is, therefore, to certify that
the thesis has been accepted in partial fulfillment of the requirements for the degree
of MSc in Development Economics.

1. Chair person: -------------------------------- Signature: ------------------Date: -----------------

2. Major Advisor: ----------------------------- Signature: ------------------ Date: ------------------

3. Co-Advisor: -------------------------------- Signature: -------------------Date: -------------------

4. Internal Examiners: -------------------------Signature: ----------------- Date: ------------------

5. External Examiners: ------------------------ Signature: ----------------- Date: ------------------

6. Dean of SGS: -------------------------------- Signature: ----------------- Date: ------------------

Final approval and acceptance of the thesis is contingent up on the submission of

the final copy of the thesis to the School of Graduate Studies (SGS) through the
Department/School of Graduate Studies Committee (DGC/SGC) of the candidate’s
department.

Stamp of SGS Date: --------------------------

ii
 Table of Contents

Declaration i
Approval Sheet ii
Table of Contents iii
Acknowledgments v
Dedication vi
Acronyms and Abbreviations vii
List of Tables viii
List of Figures ix
Abstract x
CHAPTER ONE 1
INTRODUCTION 1
1.1 Background of the Study 1
1.2. Statement of the Problem 4
1.3. Research Questions 6
1.4. Objective of the Study 6
1.4.1. General objective 6
1.4.2. Specific objective 6
1.5. Scope and Limitations of the Study 6
1.6. Significance of the Study 7
1.7. Organization of the Study 7
CHAPTER TWO 8
LITERATURE REVIEW 8
2.1. Concepts and Definitions 8
2.2. Theoretical Literature Review 9
2.2.1. The accelerator theory 9
2.2.2. Flexible accelerator theory 9
2.2.3. Neoclassical investment theory 9
2.2.4. The Tobin Q theory of investment 10
2.3. Empirical Literature 10
2.3.1. Empirical literature in the rest of the world. 10
2.3.2. Related studies in Ethiopia 15

iii
 Table of Contents (Continued)
2.4. Research Gap 16
2.5. Conceptual Frameworks 17
CHAPTER THREE 18
RESEARCH METHODOLOGY 18
3.1. Research Design and Approach 18
3.2. Data Type and Sources 18
3.3. Methods of Data Analysis 18
3.4. Model Specification for ARDL Model 18
CHAPTER FOUR 22
RESULTS AND DISCUSSION 22
4.1. Descriptive Data Analysis 22
4.1.1. Trends of Private Investment in Ethiopia 23
4.1.2. Trends of RGDP of Ethiopia in percent, 1990-2021 24
4.1.3. Trends of FDI inflows to Ethiopia, 1990-2021 25
4.2. Econometric Results of the Model 26
4.2.1 Unit root test 26
4.2.2 Lag Length Determination 27
4.2.3. ARDL-Bounds Test Approach to Co-integration 28
4.2.4. Long Run Econometric Results of the Model 29
4.2.5. Short Run ARDL Model Estimation Results 33
4.3. Diagnostics Checks for Short Run 36
4.3.1 Autocorrelation test 36
4.3.2 Heteroscedasticity test 37
4.3.3 Normality test 37
4.3.4 Stability test 38
4.3.5 Multi-collinearity test 39
4.3.6. The goodness of fit model 39
4.3.7. Granger causality test 40
CHAPTER FIVE 41
CONCLUSIONS AND RECOMMENDATIONS 41
5.1. Conclusions 41
5.2. Policy recommendations 42
6. REFERENCES 44
7. APPENDICES 47

iv
 Acknowledgments
First, may all honor and glory be extended to the exalted Heavenly God for his help,
guidance, and encouragement throughout my life.
I am very grateful to my principal advisor, Girum Dagne (Ph.D), for his continuous
guidance, suggestions, encouragement, sharing experiences, and reactions in the
construction of this thesis. In addition, I sincerely thank my co-advisor Daba Deresa
(MSc) for his extensive intervention by providing constructive comments and
suggestions. I would also like to extend my appreciation to those whose names are
not mentioned here, but have contributed a lot to my success.

Last but not least, thanks to all my families, especially to my beloved wife, Nagele
Guyo, my son, Borora Sora and my brother Dadacha Kula, for their love and affection
throughout my life.

v
 Dedication

This dedication is to my beloved wife, Nagele Guyo and my lovely and chunky son,
Borora Sora and my elder brother, Dadacha Kula.

vi
 Acronyms and Abbreviations
ADF Augmented Dickey-Fuller

AR Auto Regressive

ARDL Autoregressive-Distributed Lag

EDS External Debt Servicing

EIC Ethiopia Investment Commission

FDI Foreign Direct Investment

FDRE Federal Democratic Republic of Ethiopia

GDP Gross Domestic Product

IMF International Monetary Fund

IR Real Interest Rate

MOFED Ministry of Finance and Economic Development

NBE National Bank of Ethiopia

NR National Reserve

OLS Ordinary Least Square

PI Private Investment

PuI Public Investment

RGDP Real Gross Domestic Product

SSA Sub-Saharan Africa Countries

WB World Bank

vii
 List of Tables

Table 1. Measurement and description of the study variables 21

Table 2. Summary statistics of variables in the study 22
Table 3. Unit root test results 27
Table 4. Optimal lag length selection criteria 28
Table 5. Bounds-F test for co-integration result 28
Table 6. ARDL long run estimation results 29
Table 7. Long run diagnostic test result 32
Table 8. ARDL short run results 34
Table 9. Autocorrelation test results 36
Table 10. Heteroskedasticity results 37
Table 11. Multicollinearity test 39
Table 12. Granger causality test result 40

viii
 List of Figures
Figure 1. Conceptual framework of the Study 17
Figure 2: Share of private investment in GDP of Ethiopia, 1990-2021 24
Figure 3: Trend of RGDP of Ethiopia in percent, 1990-2021 24
Figure 4: Trend of FDI inflows in current USD, 1990-2021 26
Figure 5. Long run Stability Test 33
Figure 6. Normality test result 38
Figure 7. Short run stability test result 38

ix
 Abstract
Private investment is viewed as a powerful tool for maintaining and expanding the
capital formation and production capacity of an economy. Hence, the study sought
to analyze the determinants of private investment in Ethiopia using the time series
dataset from 1990-2021. To meet its goal, the study employed an Autoregressive
Distributed Lag (ARDL) bounds testing approach through Stata version 15. The error
correction model estimation showed the existence of a shorter period of adjustment
between the short run and the long run dynamics of private investment in Ethiopia in
the long run. The findings of this study reveal that the real interest rate, inflation,
bank credit to private and national reserves negatively and significantly predicts the
private investment in Ethiopia in the long run. However, FDI, external debt and Real
GDP have a positive and significance impact on private investment performance in
the long run. An appropriate interest rate policy to boost saving on one hand and
measures to widen and modernize the scope of credit disbursement to the private
sector on the other hand is recommended. Moreover, enhancing the real per-capita
income of people by creating various employment opportunities and income thereby
to increase FDI inflows into the country are some of the presumed measures to be
taken so as to boost private investment in the country.

Keywords: Private Investment, ARDL, ECM, Ethiopia

x
xi
 CHAPTER ONE
INTRODUCTION

1.1 Background of the Study

Investment is an act of current spending for expected future return. It expands the
productive capacity of a nation and plays a crucial role in the long term economic
growth and development process (World Bank Group, 2013). Investment has been
regarded as one of the primary engines of growth (Wade, 1989) and the proposed
solutions to SSA's economic and geographic disadvantages (Sachs, 2005).

In the process of economic growth of countries, investment plays a crucial role to

raise productivity through encouraging technological progress and promotes new
techniques of production. It also plays an enormous role in the long run capital
accumulation since investment increases productive capacity and creates new
capital goods. Hence, as investment rates increases, the rate of accumulation of
capital stock increases rapidly (Majeed and Khan 2008).

Evidence on economic development indicates that private sector in effective with the
public sector has the way to overcome poverty in Africa. Among the main way that
private sector participate on economic development is through private investment,
since investment may be defining as expenditure in capital goods (World book,
1992).

Furthermore, Private investment is playing a great role in country’s development

especially in developing countries whose capital is scarce and their government
lacks enough capacity to cover all constraints and bring economic change for
development. It thrives and delivers sustained growth when number of factors
combines to produce conductive environment for the private sectors to develop.
Private investment is led by market forces and thus more efficient than public
investment. Public investment is usually through national planning in which it is
difficult to address a single issue in the economy. Nevertheless, in the case of
private investment as they led market forces and their mainly goal is profit
maximization, every specific issues are likely to be addressed and thus higher
efficiency in resources mobilization and allocation can be achieved. Private
investment is although very important economic ideology in bringing economic

1
development, its working is determined by different socio-economic, environmental
and political factors. These different factors have either positive or negative effects
on private investment.

That is, because of some negative effects in some factors, private investment may
become less efficient, or on the other hand, because of positive changes in some of
the determinants, private investment may be encouraged and results in more
efficient (Apple Yard, 2006). When we see in case of Ethiopia, private investment in
Ethiopia has been changing with the change of government due to change in
ideological outlook they follow, the so-called economic system of each regime.
During the period prior to the 1974 revolution known as feudo-capitalistic economic
system, there was an encouragement of private sector participation in economic
development (Taye, 2005). The issuance of the commercial code in 1960 was a clear
indication of the government to encourage domestic and foreign entrepreneurs. Tax
incentives and other encouragement provided for foreign investors by giving less
importance to domestic private investment. Following the 1974 revolution or during
the Derg regime, the centrally planned command economy came up with
nationalization of most industries and government control and ownership of all
economic activities. Consequently, foreign investors who owned and operated most
of the industries are left out and domestic private investment virtually closed. This
brought the doomsday for private investment. After elapsed of more than one and
half decades, the government declared a mixed economy in which encourages
private sector once again. With the fall of communism and emergence of global
economy, beginning from 1991, change in economic and political condition
happened all over the World. At the same period similar event happened in Ethiopia
following the downfall of military government and the new government presently
called Ethiopia People’s Revolutionary Democratic Front (EPRDF) adopted market
oriented economy in 1992. The new government has introduced a number of
reforms encompassing currency devaluation, trade liberalization, deregulation of
markets, removal of restriction on private sector participation and modest
privatization and state owned enterprises to boost the role of private sector as
owner of productive resources.

Rate of investment is one of the key indicators of countries’ economic development.

This indicator indicated that high-growth countries have high investment, whereas
2
low growth countries have low investment. Hence, the low investment indicated that
the productive capacity of the economy goes to down and resulted lower rates of
growth, job creation opportunity, and difficult for the poor to improve their livelihoods.
This leads to developing countries such as Ethiopia, their economic growth mainly
depends by increment of investment (White, 2005).

Likewise, the experiences of most developing countries have enlightened that their
rapid economic growths have come from increased investment. Through investment,
many countries raise their productive capacity of an economy, create new capital
goods and increase capital accumulation. Hence, the more increment in the rate of
investment resulted that capital stock rapidly grow. Therefore, Investment is best
way for inducing economic growth and reducing poverty (Abate, Y. 2016).

The Ethiopian government, for instance, has made an effort to attract private
investment through the recently revised investment policy by giving recognition on
the role of private sector investment increment in all sectors of the economy
resulted in increasing productivity and accelerating the economic development of
the country, and hence ensure its sustainability, strengthen domestic production
capacity and thereby improve the living standards of its people (Investment
Proclamation no. 1180/2020).

Even though the Ethiopia government has been made various policy reforms to
stimulate private investment, the actual performance is very low (Hailu A. 2013).
Thus, unregulated and high inflation rate, exchange rate fluctuation, interest rate
increment could be a signal of macroeconomic instability. Besides, the prolonged
poor governance, rampant corruption, frequent changes of investment policies and
requirements, inefficient and poor infrastructural investment environment would
deteriorate investors’ confidence and ability to invest (Abate Y. 2016).

Thus, understanding the status and determinants of private investment is essential

for successful and effective implementation of sustainable development goals
(SDGs). According to UN World Investment Report (UN, 2014) SDGs require huge
levels of both public and private investment in all countries. Even if public finances
are considered as central to investment in SDGs, they cannot meet all SDG-implied
resource demands. There has been mounting evidence that private investment
depends on number of variables which significantly contributes to its growth.

3
Monitoring and maintaining these variables well is a precondition for growth and
development and it is believed to be the most certain way of enhancing private
investment.

The thesis uses time-series data to empirically examine the major macro-economic
determinants of private investment in Ethiopia based on the data gathered from
1990 to 2021. Further, the study period is selected based on the major macro
economic reforms encompassing currency devaluation, trade liberalization,
deregulation of markets, removal of restriction on private sector participation and
modest privatization and state owned enterprises. Therefore, in order to study the
effect of private investment on the performance of an economy, we need to identify
first the factors that are affecting it. By doing so, this study able to understand why
and how changes in private investment occurred and set the policy formulation
incentive provision to the sector.

1.2. Statement of the Problem

Being one of the least developing countries, Ethiopia has been characterized by low
and fluctuating private investment performance which partly contributed for the very
low advancement in the economy. The private investment has shown different
trends of growth and performance in the case of Ethiopia due to shifts in the
regimes and political processes.

During the imperial era (1960/61- 1973/74), private investment as a percent of GDP
at market price was about 10.5 percent and growing by about 6 percent per annum
on average. The reasons for a relatively good performance of private investment in
this period include the existence of import substitution strategy, market-oriented
financial sector policies, a developing share market, the free market where prices
were determined purely by supply and demand (Alemayehu and Befekadu, 2002).

However, this promising trend was reversed due to the socialist ideology persuaded
by the military government that replaced the monarchy in 1974. Thus, the period
1974/75-1990/91 witnessed centralized economic system, where the state was
given a significant role in all sphere of the economy. As a result, the private
investment was given little rooms due to the ideology. During this period (1974-
1991), the ratio of private investment as a percent of GDP declined to 4.5 percent.
4
This ratio is far below the SSA countries where the average rate of private
investment to GDP was 10.6 percent (IMF, 1995).

After the down fall of the Derg regime in 1991, the country adopted the market-
oriented economic policy, placing due emphasis on the role of the private sector in
its development efforts. In line with this policy, new investment policy and
declaration was publicize and successively amended with the aim of expanding
private investment. As a result, the share of private investments in the GDP has
registered an encouraging improvement.

The economy has undergone series of reforms i.e. from a liberalized economy (till
1974) to command type (from 1974-1989/90), and again liberalized after 1991
(following the fall of military regime). Since 1991, a comprehensive structural
adjustment programs (SAP) such as macroeconomic stabilization, privatization and
public sector development, decentralized governance and other reform packages
have been implemented in line with the free market ideals.

The private sector is, thus, identified as an essential stakeholder in fighting poverty,
creating employment opportunities, ensuring long-term economic growth and hence
a pathway towards inclusive economic growth and transformation of the economy
(Khan, 2008). Along with opening of more sectors for private investors, packages of
investment incentives are also being given. In spite of such macro-economic and
political reforms and ranges of investment incentives given, private investors have
remained very shy to make significant strides thus far.

Now a day in Ethiopia, private investment has got important place as engine of
economic growth and development. However, enormous development potential is
far under fetched till now. This suggests that, in the face of present-day international
competitions, promoting private investment should be approached with locally fit
and globally responsive investment policies and strategies. But, the shortage of
contemporary and contextual research on the area has remained to be a serious
bottleneck for such policy making and action taking efforts (Mustefa, 2014).

Most of the previous studies investigating the major determinants of private

investment and their impacts on the growth of private investment are cross country
5
in nature, with few focusing on individual developing countries’ economies (Erden
and Holcombe, 2005; Cavallo and Daude, 2011; Mustefa, 2014; Candelon et al., 2013).
Hence, in-depth investigation of incorporating new explanatory variables that are not
yet studied is important since the effect of such variables varies depending on the
time period covered and methodology followed.

In Ethiopia, only a few explanatory variables were addressed under the previous
studies that focus on determinants of private investment using ARDL approach and
there is no the same sign agreement on the significance of explanatory variables on
dependent variable. Thus the existence of contradicting conclusions, adding
irrelevant variables, and omission of relevant variables in the previous studies
initiates the researcher to another new investigation. Accordingly, the primary aim of
this study is to investigate the major determinants that determine the performance
of private investment in Ethiopia and set the policy formulation incentive provision to
the sector.

1.3. Research Questions

The study is aimed to answer the following core research questions:

 How looks the trend of private investment in Ethiopia over the period under study?

 What are the effects of major macro-economic variables on private investment?

 What is the causal direction of private investment with major explanatory

variables?

1.4. Objective of the Study

1.4.1. General objective

The main objective of this study is to examine the major determinants of private
investment in Ethiopia over the period between 1990 and 2021.

1.4.2. Specific objective

More specifically, the study addresses the following specific objectives:

 To investigate trend of private investment in Ethiopia.

6
  To examines the effects of major macro-economic variables on private
investment.
 To determine directional causality of private investment with major
explanatory variables.

1.5. Scope and Limitations of the Study

This study was delimited to assess the determinants of private investment in

Ethiopia over the period between 1990 and 2021. The study period would also be
selected based on the availability of data about the private investment and placing
due emphasis on the determinants of private investment on development efforts.
On top of this, this study focused on the major variables that affect the private
investment, and not every variable that influence decisions to invest was included in
the model. As such, some of the missing variables that are qualitative in nature
such as political uncertainty, corruption and natural disaster were excluded to
narrow down the scope of the study to manageable dimensions.

1.6. Significance of the Study

The study provided scientific findings on the subject of private investment; in

particular, it would be of significant importance to policy makers in their quest of
providing the appropriate incentives to promote private investment in Ethiopia. It also
helps researchers and policy makers in designing appropriate long run and short run
effect of selected explanatory variables that determines the level of private
investment in the country. Moreover, researchers who want to undertake further
investigation in the related topic will also benefited from the findings.

1.7. Organization of the Study

The paper was organized into five chapters. Chapter one was dealt with the
introduction and it covered background of the study, problem statement, objectives
of the study, research questions, scope and limitations of the study, significance and
organization of the study. Chapter two presented review of relevant theoretical and
empirical literature. It would also present overview and improvement of private

7
investment climate and its determinants in Ethiopia. Chapter three would be
discussed about the methodology employed in the study. Chapter four
encompassed the empirical results and discussions. Chapter five would present the
conclusions and policy recommendations of the study.

CHAPTER TWO
LITERATURE REVIEW

2.1. Concepts and Definitions

Private investment is defined as all additions to the stock of assets or purchases

and own account gross capital formation, less any sales of secondhand and
scrapped assets (Asante, 2000). Capital asset is simply property that is not easily
sold and is generally purchased to help an investor to generate a profit (land, building,
machinery and equipment).

Scott (1976) defines investment for proper growth process understanding as ‘all
expenditures which would not be made in a stationary state’. He further explained
that such expenditures consist of new production capital such as new machinery
and vehicles, building and construction, research and development, expenditure on
marketing, planning et cetera.

In economics, the concept of investment is use to mean the purchase of capital

goods that actually end up improving the welfare of a population i.e. goods which are
used in the production of other goods e.g. railroads, a factory, clearing land, or
putting oneself through education. In other words, they increase output.

Private investment behavior is primarily influenced by profit motive and always

characterized by risk and uncertainty. Risk is explained as a measurable possibility
of losing money or not gaining interest on one’s investment. Though investment is
considered as a risky venture, individuals invest with the hope of earning a capital

8
gain at the time of sale Weirich (1983). Again, people invest because they want a
return to compensate them for the time, the expected rate of inflation (a general
increase in the price of goods and services over time) and the uncertainty of the
return (Pollack and Heighberger, 1998).

2.2. Theoretical Literature Review

Investment theories can be broadly categorized into the simple accelerator theory
associated with Clark (1917), flexible accelerator model associated with Keynes
(1936), the neoclassical model associated with Hall and Jorgenson (1971) and the
Tobin’s Q model also associated with Tobin (1969).

2.2.1. The accelerator theory

After Keynes, the accelerator principle was the dominant theory of investment
behavior especially during the 1950s and early 1960s (Ghura and Goodwin, 2000).
The accelerator theory of investment postulated the linear relationship between
investment and output. The theory stipulates that capital investment outlay is a
function of output. According to this theory, given the incremental capital/output
ratio, it is easy to compute the investment requirements associated with a target for
output growth which means that an increase in the rate of output of a firm will
require a proportionate increase in its capital stock.

2.2.2. Flexible accelerator theory

Flexible accelerator model theory is a modified version of accelerator theory/model

based on the optimal accumulation of capital with an assumption that investment is
the function of the level of output and the user cost of capital. The basic idea of
flexible accelerator model is to take account of lagged effects (Khan and Khumar,
1997). This model assumes that investment is not only determined by the current

9
change in output but also its earlier changes. However, this theory has limitations
with regard to its assumption of perfect competition and exogenously determined
output.

2.2.3. Neoclassical investment theory

The restrictive assumptions behind the accelerator theory led Jorgenson et al (1967),
Jorgenson, to formulate the neoclassical approach. In this approach, which is a
version of flexible accelerator model, the desired (or optimal) capital stock depends
on the level of output and the user cost of capital (which in turn depends on the price
of capital goods, the real interest rate, and depreciation rate). Lags in decision
making and delivery creates a gap between current and desired capital stocks, giving
rise to an investment equation, that is , an equation for the change in the capital
stock.

The foundation of this approach has been criticized on the ground that the
assumptions of perfect competition and exogenously given output are inconsistent,
the assumption of static expectations about future prices, output, and the interest
rate is inappropriate since investment is essentially forward-looking, and the lags in
delivery are introduced ( Birhanu, 2004).

2.2.4. The Tobin Q theory of investment

Another theory of investment is the Tobin's q model advanced by Tobin (1969).

Tobin argues that firms’ investment level depends on the ratio of the present value of
installed capital to the replacement cost of capital. In the Tobin Q theory of
investment, the ratio of the market value of the existing capital stock to its
replacement cost (the Q ratio) is the main force driving investment (Chirinko, 1993,
Ghura and Goodwin, 2000). That is to say, enterprises will want to invest if the
increase in the market value of an additional unit exceeds the replacement cost.

2.3. Empirical Literature

2.3.1. Empirical literature in the rest of the world.

The purpose of this section is to review related studies in Ethiopia and elsewhere to
have a deeper understanding of the factors contributing for private sector
investment growth. Many researchers have studied private investment from different
10
perspectives in both developed and developing countries focusing on different
variables that determine private investment.

Oshikoya (1992) investigated how interest rate deregulations have affected private
investment in Kenya over the period 1970-1989. The dependent variable was the
private investment ratio, while explanatory variables were the real economic growth
rate, real deposit rate of interest, changes in terms of trade, public investment ratio,
inflation rate, and the lagged debt service ratio. The results showed that the real rate
of interest is significant and positively related to the private investment rate. The
other variables like inflation rate, terms of trade, and external debt service payments
have negative and significant coefficients. The coefficient on the public investment
ratio was positive and significant, thus implying that public investment is
complementary to private investment.

The paper of Greene and Villanueva (1991) looks policy effects and macroeconomic
environments on the response of private investment in LDCs during 1975-87 and
they concluded that real and per capita GDP level and public investment positively
affects private sector investments while real interest rate, external debit and debit
service and the incidence of high inflation has negatively associated with private
investment rates.

As Pfeffermann and Madarassy (1993) summarizes what determines private

investments in developing countries and it is stimulated by the growth of demand in
the economy. Likewise, credit obtained from banks, which helps to finance
investment projects are directed to public and private firms and when the public
sector needs highest proportion of it leaves negative impact on the amount required
by the private sector investments. Moreover, fiscal deficit adversely affects the
availability of finance to the private sector firms. Similarly, exchange rate movements
and the existence of inflation affect private investments negatively as of distorting
effects on the relative prices of items. High inflation in the economy associated with
the devaluation of currency leads to increase the price of imported goods and items
and affects the private sector investment negatively. Ghura and Goodwin (2000)
confirmed that private investment in LDCs has positively associated with real GDP
growth, investment activity by government and improvement of financial
intermediation.

11
Ronge and Kimuyu (1997) examined the determinants of private sector investment
for Kenya using data over the period 1964-1996. A double-logarithmic form of the
investment equation was estimated using ordinary least squares (OLS). The results
indicated that both the availability of credit and foreign exchange exerts significant
positive effects on private investment confirming the results in most empirical
studies.

Mbanga (2002) investigated the impact of external debt on private investment in

Cameroon from 1970-1999 using time series data and found out that investment
accelerator effect was in existence since a significant positive real GDP-private
investment relationship was found. The ‘‘debt overhang” hypothesis was also
confirmed in the case of Cameroon as well as the ‘‘crowding-out” effect of debt
service ratio. Public investment however crowded-in private investment while the
investment climate, captured by the lagged value of private investment, stimulates
current levels of investment. There was also a confirmed positive and significant
relationship between credit expansion and private investment whereas deteriorating
terms of trade and depreciating real exchange rate had negative effects on private
investment.

Erden (2005) uses a panel data set of 19 developing countries from 1980 to 1997
using a reduced-form neoclassical model of private investment that allows the
estimation of both the short-run and long-run determinants of private investment.
The results show that in both the long and short run public investment had a positive
effect on private investment. Real interest rate was negatively significant impact on
the level of investment, but credit availability to the private sector was positively
significant. Macroeconomic uncertainty also has a negative impact in the long run,
although its short-run impact is insignificant. Public investment also serves as a
stimulus to private investment in developing economies.

Lesotho (2006) investigated the determinants of private investment in Botswana

using a time series data over the period 1976-2003, in both the short and long run. It
employed the techniques of co-integration and error correction modeling. The results
show positive and significant coefficients for public investment, bank credit and real
interest rates for the short run model. GDP growth and real exchange rates are
significant in the long run. Inflation was insignificant in both cases. This means that

12
in the long run, the variations in private investment level is underpinned by GDP
growth, public investment, and the real exchange rate, while real interest rates, public
investment and the availability of credit affect investment behavior in the short term.

Osmond (2014) studied the determinants of private investment in Nigeria for the
period from 1970- 2012 by estimating the investment rate function derived from life
cycle hypothesis while taking into account the structural distinctiveness of
developing country. The results of the study confirmed that investment rate is
positively influenced by the growth rate of disposable income and the real interest
rate on bank deposits. Investment rate in Nigeria is found to be influenced negatively
due low public infrastructure, high lending rate, low saving rate and political
insatiability.

Badawi (2004) investigated the impact of macroeconomic policies on private

investment in Sudan employing annual data over the period 1969-1998. The results
suggested significant crowding-out effect of public investment on private investment
in Sudan, devaluation policies also contributed to discouraging private sector capital
expansion.

Acosta et al. (2005) investigated the short run and long run determinants of private
investment in Argentina for the period 1970 to 2000. The results from the ARDL
model revealed that exchange rate, inflation, trade liberalization and shocks in the
aggregate demand were the main determinants of private investment in the short run.
Further, public investment was found to have a crowding effect on private
investment in Argentina. In the case of the long run, external debt and domestic
credit markets were found to determine private investment.

Ribeiro (2001) employs the Johansen (1988) multivariate co-integration technique

and Engle Granger Two Step approach to model private-sector investment in Brazil
during the period 1956- 1996. The findings reveal a positive impact of the output,
public investment and financial variables and a negative effect of the exchange rate.
A test for weak exogeneity and super exogeneity were carried out and the result
showed credit to the private sector and public investment to be an important
economic policy instruments.

13
Erden and Randall (2005) have examined the impact of public investment on private
investment. They applied several pooled specifications of a standard investment
model to a panel of developing economies from the period 1980 to 1997. Their study
find out that public investment crowds in private investment i.e. an average, a 10%
increase in public investment is associated with 2% increase in private investment.
Moreover, the results also indicate that in developing economies availability of bank
credit is the major constraint for private investment.

Frimpong et al (2010) carried out a study seeking to present an empirical

assessment of factors that have either stimulated or dampened private sector
investment in Ghana using ARDL framework covering the period 1970 to 2002. From
the results it emerges that private investment is determined in the short-run by public
investment, inflation, real interest rate, openness, real exchange rate and a regime of
constitutional rule, while real GDP, inflation, external debt, real interest rate, real
exchange rate and openness significantly influenced private investment response in
the long-run. On the policy front, the study indicate that improving the productivity of
sectors such as agriculture and manufacturing by providing more efficient advanced
technologies as input subsidies could go a long way to increase private investment
levels and growth in output.

Kazeem et al (2012) in the study which covered the period 1970 to 2010 used
advanced econometric technique of ARDL bounds testing approach in modeling long
run determinants of domestic private investment. Findings from the study showed
clearly that difference exist between long and short run determinants. Interest rate,
real GDP, exchange rate, terms of trade, external debts, public investments, credit to
the private investment and reforms dummy are the key long run determinants of
domestic private investment while real GDP, public investment and terms of trade
are statistically significant in the short run. Kazeem et al (2012) recommended that
necessary infrastructures to complement domestic private investment should be put
in place and that external debts reduced to the barest minimum and negative effects
of external shocks endangered by foreign direct investment uncertainty and deficit
terms of trade should be prevented all together.

Quattara (2004) employed the ARDL modeling for the period 1970 to 2000 to analyze
the determinants of private investment in Senegal. Further, the paper employed

14
Johansen co integration and ARDL bounds approach to estimate the long run
relationship between private investment, public sector investment, real GDP, credit to
the private sector, foreign aid and terms of trade. The paper found that private
investment was positively influenced by public investment, real income and foreign
aid flows, but negatively influenced by credit to private sector and terms of trade.

Ajide and Bello (2013) investigated the determinants of private investment placing
greater emphasis on the role of governance, using an annual data over 1970 to 2010
periods in Nigeria. The Auto-Regressive Distributed Lag (ARDL) bounds testing
approach was use to ascertain the long and short run association of various
variables. Emanated from the estimated models showed that saving, real GDP,
degree of openness, real interest rates, inflation rates and governance measures are
strong determining variables on private investment but political stability and
accountability indicators appear both to be negatively and significantly affect the
private investment in Nigeria.

Akpalu (2002) used annual time series data from 1970 – 1994 to study the
determinants of private investment. He employed the Engle-Granger Two Step
procedure and the Johansen multivariate test. The study reveals that in relative
terms private investment in the short-run responds more too real per capita income
growth, credit availability and public investment. Public investment was found to
crowd-out private investment. There was also a significant negative relationship
between cost of capital and private investment in both the short and long run.
Further, a significant positive relationship between real GDP and private investment
was found in both the short and long run models but was not significant in the short-
run. This result indicates a confirmation of the accelerator theory of investment in
Ghana. The Consumer Price Index however was found not to be significant in both
situations.

Mallick (2002) investigated the determinants of long-term economic growth in India.

Private investment was found to be one of the key variables that accelerated
economic growth in India during the review period after employing the VAR model for
the Indian data from 1950 to 1995.

In Nigeria, Ariyo and Raheem (1991) investigated the determinants of private

investment and found that public investment, rate of GDP growth, domestic credit to

15
private sector and interest rate impacted positively on private investment.

Chete and Akpokodje (1998) findings show that private investment in Nigeria is
influenced by public investment, inflation rate, real exchange rate, and domestic
credit to the private sector in addition to the private foreign capital inflow.

Frimpong and Marbuah (2010) also look into factors that have either stimulated or
damped private investment in Ghana. With the use of unit root tests, co-integration
and error correction techniques within an Autoregressive Distributed Lag (ARDL)
framework the results of the study indicate that private investment in the short-run is
positively related to public investment, inflation, real interest rate, openness, real
exchange rate and a regime of constitutional rule. Private investment in the long run
is positively related to real output, inflation, real interest rate, openness and real
exchange rate; while negatively affected by external debt.

In a recent study, Tchouassi (2014) examined how private investment contributed to

economic growth in selected African countries, namely Cameroon, Cote d ’lvoire,
Tunisia, South Africa and Zambia. Using the recently developed Autoregressive
Distributed Lag (ARDL)-bounds test approach and the Error Correction Model (ECM),
private investment was found to be positively related to economic growth in all the
selected study countries, both in the short run and long run. The findings add support
to the call for African countries to adopt market-friendly policies that promote private
sector growth.

2.3.2. Related studies in Ethiopia

Ambachew (2010) study on the determinants of domestic private investment in

Ethiopia identified that domestic credit given to the private sector reduces domestic
private investment because the credit may be diverted to non-productive activities.
The study further identifies that the appreciation of the real exchange rate
discourages domestic private investment and vice versa. In short, the high value of
local currency constrains domestic investment.

Adugna (2013) undertook a study covering the period 1981-2010 using Ordinary
Lease Square (OLS) regression to model the determinants of private investment in
Ethiopia. Findings from the study showed that public investments in basic
infrastructures and social overheads are essential for private investment. In addition,

16
the rising real per-capital income of the people has a crucial positive effect on
private investment by way of increasing market demand for goods and services.
These in turn trigger private investment. Likewise, external debt has a favorable
effect on private investment in countries like Ethiopia where there is a serious
shortage of finance.

Siraj (2014) tried to evaluate the inter-relationship between private investment and
economic growth both in the long and short run. He argued that there is evidence of
uni-directional causality between economic growth and private investment. The
findings showed that both private and public sector investment have a positive
significant impact on real output/economic growth while in the short run public
investment has a negative impact on growth and private investment has a positive
impact on growth.

Sisay (2010) carried on the study of the determinants of private investment in

Ethiopia over the period ranging from 1950-2003 motivated by modified flexible
accelerator model by applying multivariate single equation ECM estimation
methodology. According to his study, private investment in Ethiopia is influenced
positively by the domestic market, infrastructural facilities and FDI and negatively by
macroeconomic uncertainty.

2.4. Research Gap

Most empirical studies focus on the role of private investment on economic growth
ignoring the major macro-economic variables that determine the performance of
private investment in country’s economy. Furthermore, previous related studies
applied OLS estimation technique to examine the determinants of private sectors
investment, and there is no the same sign agreement on the significance of
independent variables on the dependent variable (private investment). However, this
study adopts ARDL econometric method of analysis after checking the possible
assumptions of the model. In Ethiopia, despite this study is similar to other studies,
the presence of little empirical analysis, incorporating irrelevant variables and
existence of contracting conclusions make this study vital to add the existing
literature on the key determinants of the private investment in the economy and to
help the policy formulation incentive provision to the sector.

17
2.5. Conceptual Frameworks

The conceptual framework of the study in figure 1 below illustrates the private
investment as a dependent variable, and the explanatory variables such as Real GDP,
Inflation Rate, Interest Rate, External Debt Service, Public Investment, Access to
Bank Credit, National Reserves and Foreign Direct Investment respectively.

Independent Dependent
Variables Variabe

Figure 1. Conceptual framework of the Study

Source: Own Construction, 2023/2024

18
 CHAPTER THREE
RESEARCH METHODOLOGY

3.1. Research Design and Approach

The study employed quantitative research approach to investigate the trends and
determinants of private investment in Ethiopia. The study used secondary data
which is readily available for convenience, in terms of time available. The 32 years
data (i.e. from 1990 to 2021) were gathered to some important variables. The data
set was restricted to this period due to availability of consistence of information
especially about the private investment. The collected secondary data were
summarized using tables and graphs. Then, data were analyzed using ARDL model.

3.2. Data Type and Sources

This study used secondary data for better accomplishment of the study. Accordingly,
the annual time series data were collected from 1990 to 2021 from World Bank
Database (World Development Indicators), National Bank of Ethiopia (NBE),
Ethiopian Investment Commission (EIC) and Ministry of Finance and Economic
Development (MoFED).

3.3. Methods of Data Analysis

The study used both descriptive and econometric methods of data analysis.
Descriptive data analysis was employed to describe summary of descriptive data
statistics and overall private investment trends using simple tools like frequency,
tables, percentages and graphs. Econometric data analysis was further adopts ARDL
model with other appropriate econometric techniques to examine major
determinants of macroeconomics variables on private investment in Ethiopia.

19
3.4. Model Specification for ARDL Model

The theoretical models such as Keynesian, Neoclassical and Neo-liberal alone

cannot identify the determinants of private investment. Hence, the study used the
Autoregressive Distributed Lag (ARDL) bounds testing approach developed by
Pesaran et al (2001) to determine the determinants of private investment in Ethiopia.
The ARDL proves to be a suitable method to estimate the influence of time lag
effects of explanatory variables in multivariate time series analysis. Unlike the
competing VAR models, ARDL model treats the parsimonious lags of the right hand
variables as an independent variable (Pesaran et al, 1999). The approach has
advantageous over the other co-integration techniques such as the Engle and
Granger (1987) approach and the Johansen and Juselius (1990) approach. First, the
variables included in the model do not have to be integrated of the same order, it can
be a mixture of I (0) and 1(1) (pesaran et al, 2001). Second, it reduces serial
correlation and endogeneity problems. Third, it renders robust results for small
samples. Fourth, ARDL bounds test approach estimates the long and short run
parameters of the model simultaneously. Fifthly, ARDL approach to co-integration
helps in identifying co-integrating vectors and if one co-integrating vector is the
ARDL model of co-integrating vector is re-parameterized into ECM. The re-
parameterized result gives short run dynamics and long-run relationship of variables
of a single model. Hence, the empirical model used to investigate the determinants
of private investment follows that of Karagoz (Ngome and Berke, 2020).

Based on the above discussion, this study applies the ARDL- bounds testing
methodology to examine the dynamic impact of selected variables on private
investment in Ethiopia. Accordingly, the functional relationship is specified as:

PIt= f (PUIt, RGDPt, FDIt, IRt, INFt, BCRt, EDSt, and NRt)……….. (3.1)

Where PIt stands for Private Investment at time t, RGDPt stands for Real Growth
Domestic Product at time t, PUIt stands for Public Investment at time t, IRt stands
for Interest Rate at time t, FDIt stands for Foreign Direct Investment at time t, INFt
stands for Inflation at time t, BCRt stands for Bank Credit to private sectors at time t,
EDSt stands for External Debt Service at time t, NRt stands for National Reserve at
time t. Transforming equation (3.1) into mathematical equation by applying partial

20
logarithmic relations, therefore, the short-run ARDL econometric model is specified
as follows.

ΔLnPIt
n n n n n

= β0 + ∑β LnPIt - i + ∑β LnPUIt - i + ∑β LnRGDPt - i + ∑β LnFDt - i + ∑β LnIRt - i

i=0
1
i=0
2
i=0
3
i=0
4
i=0
5

n n n n

+ ∑β LnINFt - i + ∑β LnBCRt - i + ∑β LnEDSt - i + ∑β LnNRt - i + ϵ ……(3.2)

i=0
6
i=0
7
i=0
8
i=0
9 i

For t = time period measured by years (1990, 1991,1992…, 2021), β0 is the value of
private investment when there are no predicting factors, β1-β9 measures the partial
slope coefficients for the respective variables, Δ is the first difference operator, n is
the lag length and ε = error term.

The next step involves the use of the F-statistic based on the joint significance of all
the lagged levels of the variables. There are two sets of critical values that are used,
as reported by Pesaran et al. (2001), in the test of co-integration - the lower and the
upper critical values. The null hypothesis of no long-run relationship is rejected if the
calculated F-statistic is above the upper critical value. But, the study fails to reject
the null hypothesis of no co-integration if the F-statistic measure is below the lower
critical value. However, the co-integration test becomes inconclusive if the
calculated F-statistic falls within the lower and upper critical values.

After establishing the existence of the long-run relationship of the variables, the next
step is to obtain the long-run and error-correction estimates of the ARDL model.
Accordingly, the long run and error correction model of the ARDL model is expressed
as follows:

ΔLnPIt
n n n n n

= β0 + ∑β LnPIt - i + ∑β LnPUIt - i + ∑β LnRGDPt - i + ∑β LnFDt - i + ∑β LnIRt - i

i=0
1
i=0
2
i=0
3
i=0
4
i=0
5

n n n n

+ ∑β LnINFt - i + ∑β LnBCRt - i + ∑β LnEDSt - i + ∑β LnNRt - i + θECMt-

i=0
6
i=0
7
i=0
8
i=0
9 i

+ϵi……(3.3)

21
Where ECMt-i the error correction term is lagged once and θ is the coefficient.
The coefficient of the error-correction term measures the speed of adjustment to the
long-run position. To re-affirm the established co-integration relationship, the error-
correction term is expected to be statistically significant and negative.

Lastly, the study has ensured the fitness of the model by using pre-estimation test
including unit root test, maximum lag length and co-integration test, and post
estimation test which include stability test, normality test, autocorrelation,
heteroscedasticity test and multicorrelation test to ensure that the data fits the basic
assumptions of ARDL model.
3.5. Description of Variables
Table 1. Measurement and description of the study variables

Variables Measurement and description of the variables Expected

Sign of the
Variables
Dependent Variable
Private Private investment is an act of current spending for
Investment expected future return by private sector. It is
positively/negatively affected by different explanatory
variables.
Independent Variables
Public Sector As crowd-in: When public investment increases +
Investment aggregate demand increases, then demand for
money increases that causes to raise interest rates.
Thus, interest sensitive private investment
decreases. As crowd-out: Further public investment
is funded by deficit, which leads to increase interest
rate, credit rationing and impose current and future
tax burdens
Bank Credit to The availability of credit to private investors with fair +
the Private interest rates.
Sector
External Debt It is measured by debt service ratio to export receipts, _
the ratio of external debt to export receipts and the
ratio of external debt to GDP, and measured in
current US dollars.

22
Inflation It represents changes in the general price level or _
inflationary conditions in the economy.
Foreign Direct It is a foreigner or an enterprise wholly owned by +
Investment foreign nationals, having invested foreign capitals in
Ethiopia or a foreigner or an Ethiopian incorporated
enterprise owned by foreign nationals jointly
investing with a domestic investor and is measured in
current US dollars.
Real Gross The sum of gross value added by all resident +
Domestic producers in the economy plus any product taxes and
Product minus any subsidies not included in the values of the
products that adjusted for inflation, and measured in
US dollars.
Real Interest It represents the country’s nominal interest rate -
Rate adjusted for inflation and used as proxy to measure
macroeconomic stability.
National It refers to net gold and foreign currency reserve in +
Reserves current price, and measured in current US dollars.

CHAPTER FOUR
RESULTS AND DISCUSSION

4.1. Descriptive Data Analysis

Descriptive data deals with the summary statistics of data analysis and trends of
private investment using tables, graphs and percentages.

Table 2. Summary statistics of variables in the study

Variable Obs Mean Std. Dev. Min Max

LnPI 32 11.43 1.27 9.037 14.23
LnPuI 32 10.51 2.32 6.1 16.13
LnRGDP 32 12.75 0.78 11.14 14.75
LnBCR 32 8.77 2.91 4.062 19.55
INF 32 12.13 13.50 -8.99 55.24
LnIR 32 2.47 0.43 0.97 4.02
LnFDI 32 3.64 2.61 -1.73 8.298
LnEDS 32 7.35 1.06 6.02 9.77
LnNR 32 12.31 14.28 5.24 90.02
Source: Own computation using Stata 15, 2023/24
23
In Table 2, private investment over the past 32 years reached a maximum of 14.23
units having a mean value of 11.43. Public investment had a mean of 10.51 and a
standard deviation of 2.32 with a minimum value of 6.1 and a maximum value of
16.13 for the period under study. Real GDP is one of the most primary economic
indicators of the country. If GDP grows, the likelihood of the selling manufacturing
products also grows and private investments are likely to benefit from that apprise
of higher profits. The RGDP had a mean of 12.75 and a standard deviation of 0.78
with a minimum of 11.14 and a maximum of 14.75. Bank credit to private investors
had a mean of 8.77 and a standard deviation of 2.91 with a minimum and maximum
value of 4.06 and 19.55 respectively. Inflation had a mean of 12.13 and a standard
deviation of 13.5 with a minimum and maximum value of (8.99) and 55.24
respectively. The real interest rate had a mean of 2.47 and a standard deviation of
0.43 with a minimum value of 0.97 and maximum value of 4.02. The mean of foreign
direct investment was 3.64 and a standard deviation of 2.6 with a minimum and
maximum value of (1.72) and 8.29 respectively while external debt service had a
mean value of 7.35 and a standard deviation of 1.06 with its minimum and maximum
value of 6.02 and 9.77 respectively. Finally, the mean of national reserves is 12.31
and a standard deviation of 14.28 with the minimum and maximum value of 5.28 and
90.01 respectively for the period under study.

4.1.1. Trends of Private Investment in Ethiopia

Investment is regarded as one of the engines of growth and prosperity of nations.

Since it mobilizes idle resource, be it material or human, investment has special
importance for developing countries. Ethiopia as a developing country needs a huge
surge of investment from both domestic and external sources. More or less various
investment policies have been designed and implemented since long time ago.
However, the private investment performance trend of Ethiopia has been very low for
a long time since 1990. Figure 2 shows that the private investment in Ethiopia
increases throughout the year due to suitable investment climate for the investor
and appropriate investment policy. Nevertheless, in 1997 (6.5%) the trends shows
the downfall of private investment due to high inflationary rate and then starts to rise
in 2000 (11.847%). The private investment as a percentage share of GDP shows
different trends in Ethiopia from 8.43% GDP in 1992 to the end of period specified in

24
this study 12.543% of GDP in 2021. In the study period, on average, the percentage
share of private investment to GDP is 17.34%, the Ethio-Eritrean war period (1998-
1999) registered a smallest while the Ethiopian Millennium year (2008) largest share
of private investment. The percentage share of private investment in Ethiopia
declined between 2018 (21.43%) to 2021 (12.54%) for the reasons of COVID-19
pandemic disease, political tensions, and a devastating conflict in the Tigray region
(figure 2). The economic impact of COVID-19 includes the increased price of basic
foods, rising unemployment, a slowdown in growth, and an increase in poverty. An
underdeveloped private sector investment would limit the country’s trade
competitiveness and resilience to shocks (World Bank, 2021).

According to Primary Manufacturing Survey (2020) 84.64% of the interviewed private

sector investors reported that COVID-19 adversely affected their investment (EIC,
2021). In general last three years record (2019-2021) shows a minimum
development of private investment in Ethiopia relative to the 2018 (21.43%).
16
14
12
10
8

1990 2000 2010 2020

LnP I LnR G D P

Figure 2: Share of private investment in GDP of Ethiopia, 1990-2021

Source: Authors’ computation, 2023/24; various years report of EIC and MoFEC

25
4.1.2. Trends of RGDP of Ethiopia in percent, 1990-2021

The growth trends of private investment in Ethiopia from 1990 to 2021 can also be
revealed by economic growth during the period. Figure 3 shows the growth of
economic output adjusted for inflation in Ethiopia from 1990 to 2021.
1 5 .0 0
1 4 .0 0
LnR G D P
1 3 .0 0
1 2 .0 0
11 . 0 0

1990 2000 2010 2020

Ye a r

Figure 3: Trend of RGDP of Ethiopia in percent, 1990-2021

Source: World Bank, 2023
As can be seen in figure 3, real GDP growth slowed down from 2019-2021 periods
due to multiple shocks including COVID-19, with growth in industry and services
easing to single digits. However, agriculture, where over 70% of the population is
employed, was not significantly affected by the COVID-19 pandemic, and its
contribution to growth slightly improved compared to previous years. The
consistently high economic growth over the last decade resulted in positive trends in
poverty reduction in both urban and rural areas. However, gains are modest when
compared to other countries that saw fast growth, and inequality has increased in
recent years. Furthermore, conflicts in various parts of Ethiopia risk undermining the

26
economic and social development progress the country has achieved.
The government has launched a 10-Year Development Plan, based on its Home
Grown Economic Reform Agenda, which runs from 2020/21 to 2029/30. The plan
aims to sustain the high growth achieved under the Growth and Transformation
Plans of the previous decade while facilitating the shift towards a more private
sector driven economy.

4.1.3. Trends of FDI inflows to Ethiopia, 1990-2021

As seen in figure 4, the pre‐1996 low private investment performance pattern is also
reflected by fluctuations of the percentage share of the stock of FDI to GDP. The
figure shows the low percentage of FDI inflows to GDP of Ethiopia which had been
below average of the world and developing countries almost up to 1996 and its
sharp rise during the post 1996 period, Surpassing all in the 2019-2021 period. It
seems due to this fact that UNCTAD (2019) has classified Ethiopia in a group of
countries with ‘low potential but high performance to FDI’.
8
6
4
LnF D I
2
0
-2

1990 2000 2010 2020

Ye a r

27
Figure 4: Trend of FDI inflows in current USD, 1990-2021
Source: World Bank, UNCTD, FDI online data base, 2023

According to UNCTAD’s World Investment Report 2022, a FDI inflow to Ethiopia is

4.26 USD billion, a 77.79% increase from the previous year. However, the stock of FDI
inflows to Ethiopia decreased to 2.55 (2019) and 2.4 (2020) USD billion down from
3.36 USD billion in 2018 period. This might be due to the Covid-19 pandemic, ethnic
conflict of Tigray region, protests from Oromo ethnic groups, and vulnerability to
climate conditions and changes in world commodity prices.

4.2. Econometric Results of the Model

4.2.1 Unit root test

A unit root test is a common practice and a first step that are to be undertaken in
macro-level data analysis to address the non-stationary problem of variables. In this
study, testing for presence of unit root is done using the Augmented Dickey Fuller
(ADF) test where the null hypothesis test is that the series has unit root problems.
Therefore, working with non-stationary variables lead to spurious regression results
from which further inference is meaningless. Though the ARDL bound testing
approach does not require all variables to be integrated of the same order, it is not
applied to variables whose order of integration is greater than I (2). Therefore, the
test for the order of integration to make sure that there are no variables that are
integrated of order two and above is done. If a series is found to be non-stationary in
levels it’s differenced in order to make it stationary.
Table 3. Unit root test results

Variables ADF test statistics

st
Level 1 difference Decision rule
Test statistics 5% critical Test 5% critical
value statistics value
st
LnPI -3.339 -3.756 -6.264 3.580 Stationary at 1
difference
st
LnPuI - 2.971 -3.756 -6.491 -3.580 Stationary at1
difference
LnRGDP -5.196 -3.576 Stationary at
level

28
LnBCR -4.646 -3.576 Stationary at
level
INF -5.530 -3.576 Stationary at
level
IR -3.262 -3.576 -3.975 -3.580 Stationary at 1st
difference
LnFDI -2.827 -3.576 -5.884 -3.580 Stationary at 1st
difference
LnEDS -1.729 -3.576 -4.370 -3.580 Stationary at 1st
difference
LnNR -5.575 -3.756 Stationary at
level

Accordingly, the results from ADF test illustrates that the Real GDP, bank credit to
private sector, inflation and national money reserves are stationary in level form
while the private investment, public investment, interest rate, foreign direct
investment and external debt burdens are stationary after their first difference. All
the variables for the study country are either integrated of order 0 [I (0)] or 1 [I (1)]
based on ADF test results. Hence, the proposed model to estimate the linkage
among the variables is the Auto-Regressive Disturbed Lag (ARDL) model, and the
study concludes that the time series data does not suffer from stationary problems.

4.2.2 Lag Length Determination

The researcher uses optimal lag length selection of the variables in order to estimate
the ARDL model. However, choosing the optimal number of lags for a multivariate
time series model can be challenging, as it depends on various factor such as the
data characteristics, the model assumptions and specifications, and the model
performance criteria. To do this, the study performed different criteria to select the
optimum lag such as the Final Prediction Error (FPE), Akaike Information Criterion
(AIC), Hannan-Quinn Criterion (HQ) and Schwartz Bayesian Criterion (SBC). For
optimal lag selection the criteria with the lowest value and the lag length with the
most number of asterisks (*) is chosen. Hence, the Akaike Information Criterion (AIC)
was used to select the optimal lag length for the ARDL model. The researcher
selected maximum lags as two due to capturing relevant information, avoiding
multicolinearity and overfitting of data that leads to unbiased and inefficient results.
Accordingly, the AIC based ARDL (2 0 1 2 0 2 2 2 2) was found to be the appropriate
specification for small sample (below 60 observations) as all the variables in the

29
data are stationary either on I (0) or I (1) or both.
Table 4. Optimal lag length selection criteria

Lag LL LR FPE AIC HQIC SBIC

0 -472.385 6932.6 34.3846 34.5155 34.8128
1 -344.289 256.19 322.491 31.0206 32.3297 35.3027
2 -182.044 324.49 10.3664* 25.2174 27.7047 33.3534
3 -7376.19 15116 -508.871 -505.206 -496.881
4 7533.81 315.23* -520.129* -516.464* -508.139*
Note: * lag order selected by the criterion, Stata 15 is used.

4.2.3. ARDL-Bounds Test Approach to Co-integration

Peseran et al. (2001) bounds test to co-integration indicates rejection of null

hypothesis of no co-integration at 5 % level of significance. Accordingly, the study
estimates the result using ARDL-bounds test approach discussing both the short run
and long run relationship among the variables (Table 5). Hence, the F-bounds test
result is 37.24 which is above all the upper critical value at 1 percent level of
significance, implying that the variable share the long run relationship with its
selected explanatory variables.

Table 5. Bounds-F test for co-integration result

Ho: no levels relationship F-statistics 37.24

Critical values (0.1-0.01), F statistics
Peseran et al. 10% 5% 1%
(2001) critical
values I(0) I(1) I(0) I(1) I(0) I(1)
k_8 1.95 3.6 2.22 3.39 2.79 4.10
Own computation 2023/24

The next step in the analysis is to estimate the long-run and short-run coefficients of
variables using the ARDL approach.

4.2.4. Long Run Econometric Results of the Model

Table 6. ARDL long run estimation results

R-squared = 0.9958 No of observation = 30

Sample = 1990-2021 Adj R-squared = 0.9847 ARDL (2 0 1 2 0 2 2 2 2 )
Coef. Std. error P>t
Long run coefficients

30
Public investment (LnPuI) -0.0280789 0.0106074 -2.65 0.029
Real GDP (LnRGDP) 1.263764 0.0823313 15.35 0.000
Bank credit to private (LnBCR) -0.047808 0.0132157 -3.62 0.007
Inflation -0.0068603 0.0021348 -3.21 0.012
Real Interest rate (LnIR) - 0.1050562 -5.09 0.001
0.05352377
Foreign Direct Investment 0.0907381 0.0192781 4.71 0.002
(LnFDI)
External Debt Services 0.2977586 0.214151 13.90 0.000
(LnEDS)
National reserves (LnNR) -0.0243484 0.0027828 -8.75 0.000
Constant -5.42667 1.211356 -4.48 0.002
Own computation 2023/24

The long-run results in Table 6 for Ethiopia show that the coefficient of real GDP is
positive as expected and statistically significant at 5%. The results further reveal that
the coefficient of public investment (PuI) is negative and statistically significant. The
other long-run results indicate that the coefficients of credit to the private sector
(BCR), inflation rate (INF), interest rate (IR) and national reserves (NR) are negatively
and significantly determine the performance of private sector in the long run in
Ethiopia. However, foreign direct investment and external debt service are positively
related to private investment in the long run.

In particularly, in long run, one percent changes in public investment (PuI) decreases
the private investment by approximately 2.8 percent, ceteris paribus. The magnitude
and size of public investment expenditure coefficient implies that public investment
has long-run crowding out effect on private investment in Ethiopia. The crowding out
effect of public investment, in this case, underscores the market reforms and policy
tools that promote the growth of private investment of nation. On the first hand,
public investment inhibits private investment growth when it produces commodities
in direct competition with the private sector.

This can be the case when the state participates in commercial activities in which
the private sector has higher and growing marginal productivity than its counterpart
(Devarajan et al., 1996). The resulting private sector displacement in the market will
lead to a slowdown in private investment growth. On the other hand, public
investment can also crowd out private investment when spent on inefficient and
31
state-subsidized industries. The resulting financial losses of such state enterprises
require that the state continues to extend subsidies to keep them afloat. Again, this
arrangement implies that private sector projects, which are more efficient and
productive, can be displaced in resource allocation and this would eventually retard
economic growth (Nazmi & Ramirez, 1997). Although unexpected to this study,
empirical studies in different countries have shown that public investment spending
crowds out the amount of private investment (Voss (2002), Khan (1984), and Gupta
(1984)).

On the other hand, the results reveal that the real GDP has an expected sign. The
long run empirical results show that a one percent growth of real GDP increases the
performance of private investment by 1.26 percent, ceteris paribus. This implies that
Real GDP growth can serve as source of increase in aggregate and effective demand
thus motivate firms to invest more due to higher sales volume and profitability.
Similarly, increase in RGDP is believed to have raised the effective demand in the
economy through increased disposable income. Such increased in effective
demands for goods and services have stimulated more private investment in the
economy over the period under study. This result is consistent with the findings of
Hailu A. (2013), Kazeem et al (2012), and Woldemariam F. (2018), among others.
Hence, this result is in line with what was explained in the previous empirical studies.

The results further reveal that the bank credit to the private investors (BCR) has a
negative significant impact on private investment at 5% significant level in the long
run. The figure shows that a one percent increases in rate of credit to private
investors declines the motives to investment about around 4.7 percent in the long
run in the country. This is might be due to lack of transparency among private credit
funds, shortage of capital and NBE restrictions of loans to private investors inhibits
the growth of private investment in the long run. Further, the negative sign of BCR
suggests that when the government financed its expenditure through domestic
borrowings, it limits the private investors’ capacity to access to loans rather than
raising funds to finance investment activities. As a result, the private sector investors
hardly get credit facilities from domestic financial institutions to finance their
investment projects. Although contrary to expectations of this study, this finding is
consistent with other empirical studies on the subject see (Ambachew, 2010;

32
Quattara, 2004).

Based on the empirical results reported in table 6, foreign direct investment

significantly determines private investment, and contributes more to the
performance of private investment in the long run. The long run ARDL results from
the table illustrates that a one percent inflows in FDI resulted in the improvement on
the growth private investment approximately by 0.091 units in Ethiopia, ceteris
paribus. This finding suggests that FDI inflow on its own improves the performance
of private sectors growth of nations by generating spillovers through diffusions of
new technology and creates a positive working environment to private investors.
This is also in line with the findings of Pineli et al. (2019) and Mulhen E. (2020) in
developing countries.

Similarly, the results in the table 6 reveal that in the long run, real interest rate and
inflation indicators negatively and significantly predict the private investment of
Ethiopia at 1% level of significant. In the long run, holding all other explanatory
variables constant, a percentage increase in rate of inflation causes the level of
private investment to decrease by approximately 0.06%, which is statistically
significance at 1% level of significance. The negative sign of inflation causes
production costs to increases, creates macroeconomic instability on economic
environment; thereby reducing production and productivity of private investors. This
is consistent with the findings of Acosta et al. (2005), Fimpong et al. (2010) and
Esubalew (2014). As per the prior expectation, real interest rate has the negative
impact on private investment but statistically significant at 1% significance level in
the long run. This means, a percentage increase in real interest rate causes the
private investment to decrease by approximately 53.5%, ceteris paribus. This
suggests that an increase in the cost of debt leads to higher users cost of capital as
well as the lower rate of private investment, thus made the investment disincentive.
The findings compares favorably with many other studies on the subject including
Kaputo (2011), Oshikoya (1992), Lesotho (2006), Marbuah and Firinpong (2010),
Osmod (2014) and Kasem et al. (2012).

The results further reveal that in the long run, the coefficient of external debt has
statistically positive significant impact on private investment at 1% level. That is, a
33
one percent increase in external debt service increases the private sector investment
by around 29.7% in the nation. This suggests that the long term external debt could
provide access to financial capital to fund investment, increases financial
globalization and promotes better macroeconomic policy and governance in the
borrowing country thereby improving private investment performance. Although
contrary to the expectation of this study, this finding is consistent with some
previous studies on the subject (see, among others, Frinmong et al. 2010; Adugna,
2013; Sakr, 1993; Manba, 2002).

Based on the results estimated in the table 6, the long run coefficient of national
reserve negatively influences the growth of private investment; but statistically
significant at 1% level of significance. The result further reveals that level of
significance. The result further reveals that a percentage increase in country’s
reserve causes the private investment to decline by approximately 2.43 percent in
the country. The negative relationship might be due to the fact that piling large
national currency reserves can lead to currency appreciation, distort domestic
monetary policy and inflationary pressures, which may harm the private investors
competitiveness in the economy.

Diagnostics Checks: To check the validity of models, the researcher conducted

different diagnostic checks, including the Breusch-Godfrey test for serial correlation
of the error term, Breusch-Pagan-Godfrey test for heteroskedasticity, the VIF for
multicollinearity and the CUSUM graph test for checking parameter stability. The
result in the table 7 indicates that null hypothesis is accepted stating no serial
correlation, heteroskedasticity, multicollinearity or normality problem at the 5 percent
level of significance.

Table 7. Long run diagnostic test result

I. Breusch Godfrey LM test for autocorrelation

Ho: no autocorrelation at lag order

2 2
Lags(p) chi Df Pro>chi
2 0.85 2 0.355
II. Breusch-Pagan-Godfrey test for heteroskedasticity
Ho: Homoskedasticity
2 2
chi (1) Pro>chi

34
 0.30 0.584
IV. VIF test for multicollinearity
Ho: No multicollinearity problem
Mean VIF 2.08
Source: own computation, 2023/24

Furthermore, the researcher applied CUSUM graph test to check the model stability
in the study country. An inspection of the plots indicates that the model is stable in
study country as the graph lies within the lower and upper bound limits (Figure 5).

20

10
CUSUM

-10

-20

1990 2000 2010 2020

Ye a r

Figure 5. Long run Stability Test

4.2.5. Short Run ARDL Model Estimation Results

According to ARDL model estimation result in the short run foreign direct investment

35
(FDI), past period public investment and lagged period private investment are
significantly and positively predicting the change in private investment in Ethiopia.
Other variables including bank credit to private investors and national reserves are
negatively and significantly predicted the variation in private investment performance
in the Ethiopian economy. While other variables including GDP, external debt and
interest rate were having a mixture of sign in predicting the variation on the
magnitude of private investment (Table 8).

Table 8. ARDL short run results

ARDL(2 0 1 2 0 2 2 2 2 ) Sample: 1990-2021 Number of obs. = 32

Variables F(23, 6) = 285.31 Prob > F = 0.0000
Coef Std. Error T P>t
Private Investment (PI)
Lag-1 0.1372556 0.1337962 1.03 0.345
Lag-2 -0.2455181 0.756331 -3.25 0.018
Public Investment (PuI)
---- -0.371811 0.162987 -2.28 0.063
Lag 1 0.0154595 0.185073 0.84 0.436
Real GDP
--- 1.696142 0.1148643 14.77 0.000
Lag 1 -0.4309317 0.2844516 -1.51 0.181
Lag 2 0.1627079 0.1710406 0.95 0.378
Bank credit for private (BCR)
---- 0.0038475 0.0097129 0.40 0.706
Lag 1 -0.0370615 0.0094563 -3.92 0.008
Lag 2 -0.146761 0.0147657 -0.99 0.359
Inflation (INF)
------ -0.0083188 0.0020162 -4.13 0.006
Real interest rate (IR)
----- -0.9825741 0.1313988 -7.48 0.000
Lag 1 -0.2516304 0.1467262 -1.71 0.137
Lag 2 0.579353 0.102574 5.65 0.001
Foreign Direct Investment (FDI)

----- -0.0176993 0.0161369 -1.10 0.315

Lag 1 0.0561127 0.0215313 2.61 0.040
Lag 2 0.054447 0.023251 2.68 0.037
External Debt Service (EDS)
----- 0.069392 0.420256 1.65 0.150
Lag 1 -0.0229398 0.0610375 -0.38 0.720
36
Lag 2 0.265913 0.0726336 3.66 0.011
National reserve (NR)
--- 0.0006043 0.0018256 0.33 0.752
Lag 1 -0.0103584 0.0014849 -6.94 0.000
Lag 2 -0.017594 0.0021966 -7.81 0.000
ECM(-1) -1.144512 0.1224352 -9.35 0.000
Constant -5.42473 1.43794 -3.77 0.009
Source: Own computation 2023/24

In specifically, a percentage change in the first lag of public investment is associated

within 1.54 percent increase in private investment of Ethiopian economy, ceteris
paribus. This result indicates that public investment has crowds in effect in the short
run. This implies that a provision of public infrastructures to complement private
investment reduces private cost of productions and hence to raise profitability which
in turn stimulates private investment. This finding is consistent with previous studies
of Oshikoya (1994) and Gijini et al. (2012).

Moreover, in the short run, a percentage increase in the past two years’ GDP is
capable of increasing the current amount of private investment by 16.2 percent,
ceteris paribus. This shows that increase in the second lagged value of real GDP
tends to improve private investment thereby fostering private sector output in the
country. In contrast, the immediate current period of GDP negatively predicts the
current amount of private investment.

The short run results in table 8 reveal that the past period bank credit to private
investors (BCR) is negative; but statistically significance impact on growth of private
sector investment. The result further implies that loan disbursement to private
investors has immediate negative effect on private investment of Ethiopian economy
when the investors used it for unproductive business activities; the public produces
goods and services in direct competition with private sectors by making long-term
projects unattractive via higher interest rates, and impositions of credit cap on the
banking industry by government. Although contrary to expectations of this study, this
finding is consistent with other empirical studies on the subject see (Ambachew,

37
2010; Quattara, 2004).

The results further indicated that the past year coefficients of foreign direct
investment is positive and significantly predicting the current amount of private
investment in Ethiopia. This means, one percent increase in lagged foreign direct
investment causes the current growth of private investment to increase
approximately by 5.6 percent. Similarly, a percentage increase in past two years FDI
improves the current level of private investment of nations around by 5.44 percent in
the short run.

The results of other variables show that in the short run, ceteris paribus, a
percentage increase in one period lagged private investment has positive and
insignificant impact on current amount of private investment by 13.72 percent. This
result is in line with the rational expectation theory, where investors are rational and
do have information about the industry. Hence, the more vibrant industry in the last
period stimulates investment in the current period. However, a percentage increases
in the past two year’s period of private investment is capable of reducing growth of
current private investment in Ethiopia approximately by 24.5 percent in the short run.
The econometric results in the table 8 illustrates that the coefficients of national
reserves (NR), interest rates (IR) and inflation have negative and significance impact
on private investment as expected, which is statistically significance at 1%. The
results further reveal that the first lag external debt is negatively associated with
private investment in Ethiopia, while the second lag external debt has positive and
significant impact on private investment of Ethiopia in the short run. The coefficient
of error correction term (ECM (-1)) is -1.144 and statistically significance at 1%
implying oscillatory convergence. This indicates that an adjustment to long run
equilibrium at annual rate of 114.4% when there disequilibrium shocks to private
investment performance in the previous periods.

4.3. Diagnostics Checks for Short Run

4.3.1 Autocorrelation test

A common problem in using time series regressions is that the estimated residuals

38
tend to be correlated across time. Testing for autocorrelation helps to identify any
relationships that may exist between the current values of the regression residuals
and any of its lagged values. The presence of serial correlation in ARDL leads to
estimates that have small standard errors, inefficient, biased and inconsistent
especially when lagged dependent variables are included on the right hand side of
the test equation (Hamilton, 1994). This study tests for the presence of
autocorrelation using the Breusch-Godfrey test. The null hypothesis of the LM
Breusch-Godfrey test for autocorrelation is there is no serial correlation on residuals.
Hence, the P-value for Breusch-Godfrey test is greater than 5 percent level of
significance concluding no serial correlation in the dataset.

Table 9. Autocorrelation test results

I. Breusch-Godfrey LM test for autocorrelation

H0: no autocorrelation at lag order

Lags (P) Chi2 Df Prob >chi2
4 5.597 4 0.2313

4.3.2 Heteroscedasticity test

The classical linear regression assumes that the variance of the error term is
constant over time which is homoskedastic. The study applies Breusch-Pagan-
Godfrey test for heteroscedasticity problem. The null hypothesis of this test is
homoscedasticity or constant variance with the 5% level of significant; the p-value of
chi-square should be greater than 5% to conclude that there is no heteroscedasticity
problem. Hence, the chi-square p-value is 0.6562 which is greater than 5 percent
significant level, the researcher fails to reject the null hypothesis and concludes that
the dataset doesn’t have heteroscedasticity problem.

Table 10. Heteroskedasticity results

Breusch-Pagan-Godfrey test for Heteroscedasticity test

H0: Homoscedasticity
2 2
Chi 0.20 Prob > Chi 0.6562

Own computation using stata, 2023

39
4.3.3 Normality test

The researcher plotted the histogram that is bell-shaped, with only one peak, and is
symmetric around the mean. A normal distribution histogram shows a true
symmetric distribution of observed values. When histogram is constructed on values
that are normally distributed, the shape of columns form a symmetrical bell shape.
Accordingly, the histogram is approximately bell-shaped and symmetric about the
mean, the residuals are normally distributed (Figure 6).

Figure 6. Normality test result

4.3.4 Stability test

Furthermore, regarding diagnostics test, the researcher applies CUSUM graph test to
check the model stability in the study country. An inspection of the plots indicates
that the model is stable in study country as the graph lies within the lower and upper
bound limits (Figure 7).

40
 20

10

CUSUM
0

-10

-20

1990 2000 2010 2020

Ye a r

Figure 7. Short run stability test result

4.3.5 Multi-collinearity test

Multicollinearity reveals the linear association among the independent variables.

Multicollinearity exists when there is a correlation between multiple independent
variables in a multiple regression model. The study employed VIF technique to test
for multicollinearity among variables. The VIF measures how much the variance of
estimated regression coefficient is increased due to multicollinearity. The study
suggests that the high VIF value usually above 10 has the multicollinearity problems.
The estimated variance inflation factor is less than 10 and the researcher concludes
that there is no multicollinearity among variables.

Table 11. Multicollinearity test

Variable VIF 1/VIF

LnRGDP 4.28 0.233764
LnFDI 2.92 0.342155
LnBCR 2.00 0.500978
LnIR 2.00 0.501226
LnEDS 1.45 0.688586
LnNR 1.43 0.697486
LnPuI 1.29 0.776765

41
INF 1.26 0.791835
Mean VIF 2.08
Source: Own Computation, 2023/24

4.3.6. The goodness of fit model

The Goodness of fit statistics illustrate how ‘close’ the regression line fitted to all of
the data points taken together. Beside the reliability and validity concept, the
goodness of fit statistics also measured commonly through R2 and adjusted R2.
Coefficient of determination or R2 measures what percentage of a change in the
dependent variable can be measured or explained by the change in the independent
variables. It is also explains the level of the explanatory power. Its description is the
value of R2 and adjusted R2 always lie between 0 and 1 and if this correlation is high,
the model fits the data well, while if the correlation is low (close to zero), the model
is not a good fit to the data. The estimated regression result shows that the value of
R2 and adjusted R2 is 0.995 and 0.984 respectively. The value of R2 (=0.995) signifies
that 99.5 percent of variations in private investment is explained by independent
variables. Accordingly, the high value of R2 indicated that the explanatory variables
are well explained the trends of private investment.

4.3.7. Granger causality test

The study employees the granger causality tests to verify the directions of causal
relationship between private investment and its major determinants. Accordingly,
table 12 presents the directional causality between private investment and
explanatory variables to predict the usefulness of explanatory variables on future
values of private investment. The alternative hypothesis is that there is granger
causality from the independent variable to the dependent variable at 5% significant
level. Therefore, the study concludes that all explanatory variables granger causes
private investment in Ethiopia at 5 percent level of significant and not the vice versa.

Table 12. Granger causality test result

Granger Causality Test

Sample: 1990-2021
Lags: 2
Null Hypothesis Observation F-Statistics Prob.

42
PI does not granger cause PUI 32 1.4 0.497
PUI does not granger cause PI 1.8661 0.393
RGDP does not granger cause PI 32 4.0986 0.129
PI does not granger cause RGDP 4.843 0.089

FDI does not granger cause PI 32 3.0683 0.216

PI does not granger cause FDI 9.6851 0.008

BCR does not granger cause PI 32 3.2253 0.199

PI does not granger cause BCR .83324 0.659

INF does not granger cause PI 32 4.0505 0.132

PI does not granger cause INF 1.0807 0.583
IR does not granger cause PI 32 .01249 0.994
PI does not granger cause IR .19593 0.907
EDS does not granger cause PI 32 .35857 0.836
PI does not granger cause EDS 6.0059 0.050
NR does not granger cause PI 32 .30383 0.859
PI does not granger cause NR 1.394 0.498
Source: Own computation 2023/24

CHAPTER FIVE

CONCLUSIONS AND RECOMMENDATIONS

5.1. Conclusions

The empirical results of ARDL model revealed that in the short run, public investment
crowds in private investment in Ethiopia. To this end, the provision of public
infrastructures to complement private investment reduces private cost of
productions and hence to raise profitability for private investment. This finding is
consistent with previous studies of Oshikoya (1994) and Gijini et al. (2012).

In contrast, public investment has long-run crowding out effect on private investment
in Ethiopia. The crowding out effect of public investment, in this case, underscores
the market reforms and policy tools that promote the growth of private investment
of nation. First, public investment inhibits private investment growth when it
produces commodities in direct competition with the private sector. Second, public
investment can also crowd out private investment when spent on inefficient and
state-subsidized industries. Then, the resulting financial losses of such state
enterprises require that the state continues to extend subsidies to keep them afloat.

43
The short run results show that the foreign direct investment and past two year’s real
GDP significantly associated with the present private investment performance of
nations. In the long run, the external debt predicts private investment positively and
significantly.

In Ethiopia, the real GDP and foreign direct investment have a positive and significant
impact on private investment in the long run. Whereas, bank credit to private
investors, real interest rate, inflation and national reserves are negatively and
statistically significantly predicted the variation in private investment in Ethiopia in
the long run.

The private investment is converging towards their long-run equilibrium for any
external shocks that might occur in the Ethiopian economy.

5.2. Policy recommendations

Based on the empirical findings from this study, the following policy
recommendations can be reached:

1) The government should undertake public investment in a way that could

remove bottlenecks that undermine private investment as the long run finding
reveal that public expenditure is a competent of private sector and hence,
reduces the amount of money available for private sector investment.

2) Policy makers should ensure macro-economic stability by containing the

inflationary trend persisted over a longer period. Inflation should be kept at a
manageable level because with the uncertainties that it brings, it hurts private
investment in Ethiopia.

3) Policy makers should enhance the real per-capita income of people by

creating various employment opportunities and income generating means,
efficiently utilize external sources of finance (loan) for productive investment
activities, since real gross domestic product is an important variable that

44
 determine private investment in the Ethiopia economy.

4) The Ethiopian government should encourage the policy measures that aims
to improve real interest rates should maintain the tradeoff between saving
and investment. To this end, the government should adjust the real interest
rate in a way that encourages saving and boost private sector investment.

5) Generally, the policy makers should create the fertile working environment
that stimulates the growth of private investment in the Ethiopian economy.
Likewise, the policy seekers should encourage FDI-inflows, promotes and
expand banks and other financial institutions for providing credit facilities to
private investors of the nations.

6) On the other hand, the government and central bank of Ethiopia should
consider increasing the stock of national money reserves and level of
financial development, thereby optimizing the amount of external debt
services in the country. In addition, the institutional quality of Ethiopia
demands improvement, and the government should consider other
alternatives for enhancing deficit financing, like domestic borrowings. In
addition, the government of Ethiopia should identify the domestic sources of
reserves to mitigate the problem and thereby optimize external debt services.

7) The researcher recommends that others conduct research on related topics

by increasing the time-frame and including other macroeconomic factors that
affect the growth of private investment in Ethiopian economy.

45
 6. REFERENCES

Alemayehu, G. and Befekadu, D. (2005). ‘Explaining African economic growth’. The

case of Ethiopia. AERC growth working paper, AERC, and Nairobi, Kenya.

Ariyo, A. and Raheem. M. I. (1991). ‘Effect of fiscal deficit on some macroeconomic

aggregates in Nigeria’. Final Report Presented at AERC Workshop, Nairobi, Kenya.

Asante, Y. (2000). ‘Determinants of private investment behavior’. Vols. 100. AERC

Research Paper No.

Bank, W. (2005). A better investment climate for everyone. World Bank. Washington
DC: World Development Report.

Bank, W. (2012.). World Development Indicators [CD]. Washington DC: Author.

Chete, L. and Akpokodje, G. (1998). ‘Macroeconomic determinants of domestic

46
private investment in Nigeria': An Empirical Exploration, Vol. 35(1). CBN Economic
and Financing Review.

Erden, M. and Randall, G. (2005). ‘The effects of public investment on private

investment in developing economies’. In Public Finance Review, 33 (5): 575-602.

Frimpog, J.M. & Marbuah G. (2010). The determinants of private sector investment.
european journal of social science , 250-261.

Ghura, D. and B. Goodwin (2000). ‘Determinants of private investment: A Cross

Regional Empirical Investigation. Journal of Applied Economics.

Greene, J. and Villanueva, D. (1990). Determinants of private investment in LDCs. In

Finance and Development. West publishing Co UK.

Greene, J., Villanueva, D. and Smith, K. (1991). ‘Private investment in developing

countries. An empirical analysis. IMF Staff Papers.

Gujarati, D. (2003). In M. Hill., Basic econometrics (4th edition.). New York.

Gupta, K. (1984). Finance and Economic Growth in Developing Countries. London:

CroomHelm.

H, A. (2000). Factors Determining Private Investment in Ethiopia.

Johansen, S. and Juselius, K. (1990). ‘Maximum likelihood estimation and inference

on Cointegration-with application to the demand for money. Oxford Bulletin of
Economics and Statistics.

Kazeem, A., Molapo, S. & Olukemi, L. (2012). ‘Modeling the long run determinants of
domestic private investment in Nigeria’. Canadian Center of Science and Education.

Keynes, M. (1947). The General Theory of Employment, Interest and Money.

NewYork: Macmillan.

Khan, S. & Reinhart, M. (1990). Private investment and economic growth in

developing countries. World Development Indicators.

King’ori, Z. I. (2007). Factors influencing private investment in Kenya. Unpublished

47
manuscript .

Lesotho, P. (2006). An investigation of the determinants of private investment: the

case of Botswana. University of the Western Cape, Department of Economics,
Botswana.

Mbanga, G. N. (2002). External Debt and Private Investment in Cameroon. African

Journal of Economic Policy, 9(1), 109-125.

Mustefa, S. (2014). ‘Private investment and economic growth evidence from

Ethiopia’. Mekelle University, School of Graduate Studies.

Oshikoya, T. Macroeconomic Determinants of Domestic Private Investment in Africa:

In E. D. Change, An Empirical Analysis. Vols. 42(3), pp. 573-595. Economic
Development and Cultural Change.

Pesaran, H. Shin, Y. & Smith, R. J. (2001). Bounds testing approaches to the analysis
of level relationships. Journal of Applied Econometrics, 16(2), 289-326.

Pfeffermann, G. & Madarassy A. (1991). ‘Trends in private investment in developing

countries. Discussion Paper 11: International Finance Corporation: Washington, DC.

Ronge, E. & Kimuyu, P. K. (1997). Private investment in Kenya: Trends, composition

and Determinants. Institute of Policy Analysis and Research, Nairobi, Kenya.

S, A. (2002). Private investment and public Policy in sub Saharan Africa an emperical
analysis.

Sakr, K. (1993). Determinants of private investment in Pakistan. International

Monetary Fund, IMF Working Paper No. 93/30.

Seruvatu, E. & Jayaraman, T. (2001). Determinants of Private Investment in Fiji.

Working Paper No. 2001/02.

Tobin, J. (1969). ‘A general equilibrium approach to monetary theory’. Journal of

Money, Credit and Banking, vol. 1, no.1, 15‐29.

Worke, M. (1997). ‘Determinants and constraints of private investment in Ethiopia’.

Addis Ababa University, Addis Ababa Ethiopia.

48
 7. APPENDICES

Appendix I: Dickey-Fuller unit root test for Private Investment, Public Investment,
RGDP and Bank Credit to Private Investor

49
Appendix II: Dickey-Fuller unit root test for Inflation, FDI, Real Interest Rate, EDS and
National Reserve

50
Appendix III: Bounds Co-integration Test Result

51
Appendix IV: ARDL Long-run Test Result
52
53
Appendix V: ARDL Short-run Test Result

54
Appendix VI: Breusch-Godfrey LM test for autocorrelation

Appendix VI: Breusch-Pagan test for heteroskedasticicty

Appendix VI: White test for homoskedasticity

55