ADDIS ABABA UNIVERSITY

COLLEGE OF BUSINESS AND ECONOMICS

DEPARTMENT OF ECONOMICS

THE FACTORS AFFECTING GROWTH OF INVESTMENT IN THE

MANUFACTURING SECTOR IN ETHIOPIA

A SENIOR ESSAY PREPARED FOR PARTIAL REQUIREMENT

OF BACHELOR OF ART DEGREE IN ECONOMICS

ACKNOWLEDGMENT
First and foremost, I praise to God, the almighty, for providing me with the strength to
accomplish this study and for all the blessing in my life; without him nothing is achieved.
Secondly, I would like to extend my deepest gratitude to my advisor Dr. Jonse Bane for his
advice, encouragement and insightful comments that initiated me to work hard and exert my
effort, which also further initiated me to like the economics discipline more than before in
addition to doing the paper well. Lastly, my deepest gratitude also goes to my family for their
moral and financial support, and to all others who directly or indirectly contributed to the
successful completion of the study

Abbreviations

ADF-Augmented Dickey Fuller

AEG-Augmented Engle Granger

ARCH-Autoregressive Conditional Heteroscedasticity

CSA-Central Statistical Agency

DF-Dickey Fuller

DW-Durbin Watson

ECM-Error Correction Model

FR- Foreign Reserve

GDP-Gross demotic product

GTP-Growth and transformation plan

IGDP-Industrial Gross Domestic Product

IMS -Investment in manufacturing sector

INFR-Inflation Rate

LDCs-Least developed countries

MPK-Marginal product of capital

NBE-National bank of Ethiopia

OLS-Ordinary least squire

PASDEP-Plan action for sustainable development and Redaction poverty

REER-Real Effective Exchange Rate

OLS-Ordinary Least Square

RGDP- Real Gross Domestic Product

VIF-Variance Inflection Factor

WB- World Bank

SAP-Structural adjustment program

SDPRP-Sustainable development and poverty reduction

LIST OF FGURES

FIGURE 1: MANUFACTURING VALUE ADDED FOR THE TOTAL GDP.............................................20

FIGURE 2: TRENDS OF INDUSTRIAL VALUE ADDITION (% OF GDP) AND INVESTMENT IN

MANUFACTURING SECTORS FOR ETHIOPIA............................................................................21

FIGURE 3: TRENDS OF TOTAL FOREIGN RESERVES IN MONTH IMPORT AND INVESTMENT IN

MANUFACTURING SECTORS FOR ETHIOPIA BY THE GRAPH....................................................23

FIGURE 4: TRENDS OF CAPITAL EXPENDITURE AND INVESTMENT IN MANUFACTURING SECTORS
FOR ETHIOPIA BY THE GRAPH.................................................................................................24
FIGURE 5: TRENDS OF GENERAL INFLATION RATE AND INVESTMENT IN MANUFACTURING
SECTORS FOR ETHIOPIA BY THE GRAPH..................................................................................25
FIGURE 6: TREND OF REEL EFFECTIVE EXCHANGE RATE AND INVESTMENT IN MANUFACTURING
SECTORS OF ETHIOPIA............................................................................................................27
FIGURE 7: TRENDS OF DEPOSIT INTEREST RATE AND INVESTMENT IN MANUFACTURING SECTORS

OF ETHIOPIA............................................................................................................................27

LIST OF TABLES

TABLE 4.1 SUMMARY OF DESCRIPTIVE ANALYSIS..........................................................................28

TABLE 4.2 UNIT ROOT TESTS.........................................................................................................30
TABLE 4.3 OPTIMAL LAG LENGTH SELECTED BY AIC...................................................................31
TABLE 4.4 RESULTS OF AUGMENTED ENGEL GRANGER (AEG) TEST...........................................31
TABLE 4.5 ESTIMATION RESULTS OF THE LONG RUN IMS MODEL................................................32
TABLE 4.6 COMPARISON OF EXPECTED SIGN AND MODEL RESULT................................................32
TABLE 4.7 ESTIMATION RESULTS OF THE SHORT-RUN BASED ON THE ERROR CORRECTION MODEL.
................................................................................................................................................34
TABLE 4.8 TEST FOR MULTICOLLINEARITY PROBLEM...................................................................36

ABSTRACT
This study investigated the factors affecting the growth of Investment in the Manufacturing
Sector in Ethiopia over the period 1990–2021 using data from NBE, WB, CSA and investment
commission. The study applied the ARDL and short run error correction model (ECM)estimation
technique to capture the factors affecting the growth of investment in the manufacturing sector in
Ethiopia. Descriptively, graphs and tables are used to show the trend and performance of
Ethiopian manufacturing sectors. Specifically, the ARDL and ECM model shows that the real
Industrial Domestic Product and foreign reserve have positive significant impact on the
manufacturing investment growth both in the short run and long run. Furthermore, the co-
integration test indicates that growth of investment in manufacturing sectors and its determinants
have long run relationship. Moreover, in the short run only industrial GDP, total foreign reserves,
and deposit interest rate are significant factors affecting the growth of Investment in the
Manufacturing Sector. The overall result indicates that industrial GDP, total foreign reserves,
deposit interest rate are the key significant factors affecting the growth of Investment in the
Manufacturing Sector in Ethiopia both in short run and long run. And Real effective exchange
rate, inflation and Capital expenditure have negative impact on the manufacturing investment
growth, but their impact is significant only in long run, in short run their impact is found out to
be insignificant. Policy makers and any stakeholders, therefore, have to enhance these key and
significant variables in order to improve the country’s investment in manufacturing sectors.
Contents
CHAPTER ONE..............................................................................................................................................1
1. INTRODUCTION.......................................................................................................................................1
1.1. Background of the Study......................................................................................................................1
1.7. Organizations of the Study...............................................................................................................5
CHAPTER TWO.............................................................................................................................................6
2.LITERATURE REVIEW.................................................................................................................................6
2.1. Basic Concepts and Definitions of Investment......................................................................................6
2.2 Theoretical Review............................................................................................................................7
2.2.1 Neo-Classical theory of investment............................................................................................7
2.2.3. Accelerator theory of investment........................................................................................8
2.2.7. Theories Related to Some Factor affecting Investment.............................................................8
2.3 Empirical Review of Related Literature............................................................................................10
2.3.1. Evidence from developing countries..................................................................................11
2.3.2. Evidence from Ethiopia......................................................................................................12
2.4. Conceptual Framework...................................................................................................................14
CHAPTER THREE........................................................................................................................................17
3. Methodology of the Study.....................................................................................................................17
3.1 Model Specification.............................................................................................................................17
3.2 Data type and source.......................................................................................................................19
3.3 Estimation Method..........................................................................................................................19
3.3.2 Co-integration Test...................................................................................................................20
3.3.3 Error Correction Model (ECM)..................................................................................................20
3.4 Post-Estimation Diagnostic Tests.....................................................................................................20
3.4.1 Heteroscedasticity Test.............................................................................................................20
3.4.2 Auto-correlation test................................................................................................................21
3.4.3 Multi-collinearity test...............................................................................................................21
3.4.4 Normality Test..........................................................................................................................21
4. Result, Discussion, and Analysis.............................................................................................................23
4.1 An Overview of Patterns, Trend and Performance of Ethiopian Investment.......................................23
4.1.1. Structure and Performance of Manufacturing Investment in Ethiopia(IMS)...................................23
4.1.2. The Structure and Performance of Ethiopian industrialization........................................................25
4.1.3 Trends of Explanatory variables and Investment in Manufacturing Sectors.....................................25
4.1.4 Summary of descriptive statistical analysis.......................................................................................34
4.2 Pre-estimation Test 4.2.1 Tests for Stationarity (Unit Root Test)........................................................35
4.2.2 Optimal Lag Length...........................................................................................................................37
4.3 Estimation............................................................................................................................................37
4.3.2 Long Run Estimation Results.............................................................................................................38
4.3.3 Results of the Short Run Model Based on ARDL-ECM......................................................................41
4.4 Post Estimation/ Diagnostic Tests........................................................................................................43
4.5 Analysis of Review Literature and Results of the Study.......................................................................45
CHAPTER FIVE............................................................................................................................................47
5. Concluding Remarks, Policy Implications, and Limitations of the Study................................................47
5.1 Conclusion...........................................................................................................................................47
5.2 Policy Implication.................................................................................................................................48
5.3 Limitations of the Study.......................................................................................................................49
References.................................................................................................................................................50
CHAPTER ONE
1. INTRODUCTION
1.1. Background of the Study
Several economists define investment in various ways and at various times. Jack Harvey (2000)
define Investment as a real capital asset that is added to the stock of existing real capital assets
because it includes the value of new structures, new producer's durable equipment, and changes
in inventories. Investment is the flow of money spent on projects that produce commodities and
services that are primarily meant for consumption in the future rather than immediate use today.
These projects produce inventories as well as physical and human capital. Positive net present
value is produced by the capital flow. Yet, in economics analysis, investment is the stream of
spending that boosts real capital for the future.

A favorable environment that encourages both domestic and foreign investors to put their money
into industrial growth is necessary for the mobilization of both domestic and foreign resources
for industrialization. Specialization in light manufactured goods, followed by intermediate, and
eventually heavy capital goods, was the traditional pattern of economic expansion (Ghatak,
2018).

Investment is multifarious in nature. It may undertake by both private and public sector. Private
sector may invest in fixed capital like machinery, equipment and building and in working raw
materials. It may also undertake by public authority in social capita like school building, health
services, housing and other infrastructures. The investments that are solely undertaken by the
government are mostly without any profit motives rather it intended to give free service for the
public, and that are undertaken by private enterprises are induced by profit motive under the
influence of change in income or consumption (Sintayehu, 2013).

Manufacturing sector is one part of investment in industrialization that produces goods within
the economy. It creates a greater contribution for the growth of agricultural productivity by
creating demand for agricultural output and producing input for the sector. It also helps to reduce
unemployment rate and stimulate other economic sectors like trade and service related
economies. Investment in manufacturing sector is capable of increasing job opportunity,
increasing productivity, increasing country GDP, country product of export and substituted

1
import of the country. The role of industrial sector is to raise productivity per head (physical
output per head) and to develop the economies that become nationally integrated, flexible and
capable of self-generated and self-sustained (Bezawork, 2007).

In order to increase national technological capacity, industrial capability, create a wide range of
job opportunities, and boost income, the manufacturing sector must grow. Also, the growth of
the manufacturing sector contributes to an increase in the general economy's factor productivity
and competitiveness, as well as its cascading effects up and down the supply chain. The
government of Ethiopia provided numerous incentives for the growth of the manufacturing
sector due to the existence of a variety of comparative advantages within the nation and to
increase its competitiveness. Although the manufacturing sector might be a way to promote
sustainable economic growth, its expansion is not without difficulties. The major ones include
unskilled labor forces with limited experience; limited infrastructure; external pressure from
global market, shallow industrial research and development activities, underdeveloped
market system, problems associated with trade logistics and limited promotion made on the
resources and other opportunities (Ibid!).

Therefore, this paper tried to identify major determinants which are either motive or constraint
factors for investment in the manufacturing sector and answers the question of what are factors
responsible for the low growth and less diversifying of the sector, by using reference results that
are obtained from other related study that are already done associated with private and public
investment in the manufacturing sector. Generally, this paper focuses on important trends,
patterns, performance and major factors for investment in the manufacturing sector in the case of
Ethiopia.

1.2. Statement of the Problem

Investment influences the rate at which physical capital is accumulated and plays a significant
role in the economic growth of developing nations (Jongwanich & Kohpaiboon, 2006).
According to Mohammed (2016) Investment is an engine for economic development and is one
of the most important weapons to reduce poverty. It improves the productive capacity of the
nation and also creates job opportunities for many people. Investment is central to economic
development because economic development is a result of rapid capital accumulation which is
induced by new investment. Thus to make investment in least developed countries the transfer of

2
surplus labor from the subsistence sector to the modern manufacturing sector is essential since
the capitalist sector uses reproducible investment while the subsistence sector does not.

Since the 2000s, Ethiopia has emerged as one of the fastest-growing economies in Africa.
Nevertheless, Ethiopia’s manufacturing sector is still far from being an engine of growth and
structural change. The manufacturing sector plays a marginal role in employment generation,
exports, output, and inter-sectoral linkages. In some ways, the structure and performance of the
Ethiopian manufacturing sector mirrors the wider sub-Saharan African experience (Lawrence
2005). The Ethiopian manufacturing sector has had two distinct features: first, “… a low level of
industrialization in terms of the sector’s share in GDP, export earnings, industrial intensity, and
competitiveness. The Second, the industrial structure is dominated by small firms and resource
based industries (in particular the food industry) and concentrated around the capital city”
(Arkebe Oqubay, 2015).

Although Ethiopia has emerged as one of Africa’s fastest-growing economies, its manufacturing
sector is still far from being an engine of growth and economic transformation. It currently plays
a marginal role in employment creation, exports, and output, and falls short on stimulating
domestic linkages. The sector is dominated by small firms and resource-based industries, low-
value and low technology products, and weak inter-sectorial and intra-sectorial linkages. The
manufacturing sector’s export orientation has been low and stagnant (Arkebe Oqubay, 2019).

According to the (NBE, 2018/19) report the manufacturing sector in Ethiopia operates at a very
low technological level well behind world technological standards even in the activities in which
it specializes. Imports of licensed technology are negligible with no sign of increase. This may
not be unusual in less industrialized countries. However, increasing the export of manufactures
cannot be achieved without the rapid introduction of modern technology at least for those
industries of comparative advantage aiming to produce for export.

The growth of real GDP has positive and significant impact on investment in the manufacturing
sector. Increasing the productivity of agricultural sector alone will not improve the living
standard of the people but 85%of the people are engaged in agricultural sector. So it demands a
greater improvement in the manufacturing sector since its growth leads to increase the demand
for agricultural product and provide modern input adoption to the agricultural
sector(Remedan,2012).

3
Various studies were done on the possible factors affecting growth of investment in Ethiopia’s
manufacturing sectors. For instance, money supply, capital expenditure and gross saving has
positive impact for the development of manufacturing investment while war, interest rate,
inflation has negative impact (Bezawork, 2007). Real GDP, capital expenditure and inflation rate
have positive impacts on investment in manufacturing sector while real lending interest rate and
real exchange rate affect manufacturing sector negatively both in short run and long run
(Remedan, 2012). Moreover, the test employed in many researches fail to test important
variables like the sector GDP of the industrial itself and international reserves, while others have
contradictory conclusions about the same determinants of investment in the manufacturing
sector. Lastly, the explanatory variables used in these studies are not the only determinants of
investment in Ethiopia manufacturing sectors. As explained above, the results of these studies
are different; for example, the effects GDP of the industrial sector itself and international
reserves. The difference in each of these empirical studies are based on the difference in the time
period of the study, methods employed, factors considered and indicators used in the study.

In this study, however, additional relevant variables are controlled; for example, amount of GDP
of the industrial sector itself, international reserves, Corruption, Conflicts and Property right
include, which is believed to importantly determine the country’s growth of investment in
manufacturing sector was included. This study will also try to fill the time gap using long time
period (30 years) from 1991/2 to 2020/1 than which is used by the previous studies discussed
above by using recent data.

1.3 Objective of the Study

1.3.1 The General objective of the study

 The major objective of this study focuses on the examination of the main factors that affect
investment in the manufacturing sector in Ethiopia.

1.3.2 The Specific objectives of the Study are:

 To examine the major factors affecting an investment in the manufacturing sector.

 To examine the performance of an investment in the manufacturing sector of Ethiopia in
terms of trend and pattern.
 To examine the long run relationship between investment and its major determinants.

4
1.4 Research Question

The major research questions of this study are the following

 What are the factors affecting investment in the country’s manufacturing sector?
 What is the exact relationship between investment and its determinant variables?

1.5. Scope of the study

This study limit to the different macro- economic variable that are affecting Investment in
manufacturing sector of Ethiopia. It covers the time span from 1992/3 to 2020/1.

1.6 Significance of the Study

Specifically identifying what the factors that affect investment in manufacturing sector of the
country is highly difficult task. Given this fact, the study will attempts establish the major factor
affecting growth of investment in manufacturing sector of Ethiopian. It is in this fact that the
significance of this study will lies. In addition, the paper will be expected to serve as a beginning
input for any future study by economics students and scholars who may have an interest and
which will be conducted in this area and related areas. Research is done before on this study area
by many researchers; for example, (Remedan, 2012), (Gizachew, 2017). I also include GDP of
the industrial sector itself and international reserves as a new relevant variable. This is very
helpful in clearly investigating the factors that affect investment in manufacturing sector of the
country.

1.7. Organizations of the Study

There will be five key chapters in the paper. The first chapter discusses a method for solving the
issue. Review of related literature is included in the second chapter. The study's methodology is
presented in the third chapter, while the study's econometric analysis is covered in the fourth.
Finally, the chapter concludes with policy implications.

5
CHAPTER TWO
2.LITERATURE REVIEW
2.1. Basic Concepts and Definitions of Investment
Investments are often defined as the flow of costs that raise the actual stock of capital. According
to Dornbusch and Fischer (1994), investment spending is significant because it accounts for a
significant portion of the movement in the business cycle. Broadly speaking, investment is a
significant macroeconomic factor that affects economic growth (Collier & Gunning, 1999).

According to Parker (2013), economists often reserve the term "investment" for actions that
boost the economy's overall real wealth. This primarily refers to the acquisition (or creation) of
brand-new, genuine, durable assets like factories and machinery. Investment, as defined by the
International Centre for Settlement of Investment Disputes Agreement, includes any justifiable
activity or asset that contributes to the current capital formation of a country and so has a
positive effect on the gross output of a country.

Investment is generally classified into four major components: private domestic investment,
public domestic investment, FDI and portfolio investment. Private domestic investment refers to
gross fixed capital formation plus net changes in the level of inventories whereas public
investment includes investments made by the government and public enterprises on social and
economic infrastructures, real estate and tangible assets. The combination of private investment
and public investment is normally referred to as gross fixed capital formation and this is
distinctive from their counterpart – foreign investment. When foreign investment is on a tangible
asset, it is referred to as a direct foreign investment; when it is in shares, bonds, securities, etc.; it
is called portfolio investment (Bakare, 2011).

As cited on Desta kidanu, 2010; as defined by International Standard Industrial Classification

(ISIC) in 2008, manufacturing is physical and chemical transformation of materials, substance or
components into new products. These material and components are outcome of the
manufacturing activity itself. Using the number of employees, the central statistical agency of
Ethiopia classified industries as follows in 2005/06 survey of its study.

I. Large and medium scale manufacturing establishments engaging ten or more workers and uses
power driven machinery.

6
II. Small scale establishments engaging less than ten workers and uses power driven machinery.

III. Cottage/handicraft manufacturing establishments performing their activity by hand (non -

power driven machinery).

Investment in the manufacturing sector, in particular, can boost employment opportunities, factor
productivity, national GDP, and balance of payments stability by producing surpluses for export
and substituting domestic goods for imported ones. In order to investigate the significant
conflicts that now exist and to emphasize the central idea, some hypotheses connected to
investments and the manufacturing sector are discussed in this section in a concise manner.

2.2 Theoretical Review

As a matter of understanding the essence of trade, and hence the determinants of trade balance,
the following reviewing theories are of very important.

2.2.1 Neo-Classical theory of investment

The neoclassical model analyzes the advantages and disadvantages of capital asset ownership for
businesses. The model demonstrates the relationship between the degree of investment—the
increase in the stock of capital—and the marginal product of capital, the interest rate, and the tax
laws that apply to businesses. This model explains that the net investment is the difference
between actual capital stock and the desired level of capital stock. i.e., I t=Kt-Kt-1=λ (k*-kt-1)
Where; It= net investment, Kt=the existing capital stock, Kt-1= last year capital, K*=desired level
of capital, and λ=measures the fraction of the gap between the actual and the desired level of
capital. This shows that the desired capital stock (K*) has a negative relationship with the capital
rental cost and a direct relationship with the expected level of output (N. Gregory Mankiw,
2009). So, cost, return, and expectations are the three factors that determine a business
investment, according to the neoclassical approach.

2.2.2. Keynesian theory of investment

The Keynesian theory of investment was developed by John Maynard Keynes in 1930s. Keynes
(1936) has given consideration for the existence of independent investment behavior in an
economy. He considers that the marginal efficiency of capital in relation to some interest rate
that reflects the cost of invested capital, and investment is the function of this prospective
marginal efficiency of capital. He also argued that individual investment decision is related to
7
uncertainty. According to him uncertainty reflects option difference between individuals about
the future interest rate. Therefore, individual decision to invest or not invest largely depends on
their expectation about the future possible outcome of investment and on the cost of capital, that
is, the interest rate. Generally according to Keynes investment decisions are taken by comparing
the marginal efficiency of capital (MEC) or the yield with the real rate of interest (r).

2.2.3. Accelerator theory of investment

Contrary of Keynes, there is a concept of accelerator which is the numerical value of the
relationship between the increase in investment resulted from an increase in income. The net
induced investment will be positive if national income increases and induced investment may fall
to zero if the national income or output remain constant. The accelerator principle states that an
increase in the rate of output of a firm will require a proportionate increase in its capital stock.
The capital stock refers to the desired or optimum capital stock, K.

Assuming that capital-output ratio is some fixed constant, v, the optimum capital stock is a
constant proportion of output so that in any period.

Kt = VYT

Where Kt is the optimal capital stock in period t, v (the accelerator) is a positive constant, and Y
is output in period t. Any change in output will lead to a change in the capital stock. Thus

Kt – Kt-1 = v (Yt – Yt-1)

And Int = V (Yt – Yt-1) [Int=Kt– Kt-1= V∆Yt, where ∆Yt = Yt – Yt-1, and Int is net
investment.

This equation represents the naive accelerator. In this equation, the level of net investment is
proportional to change in output. If the level of output remains constant (∆Y = 0), net investment
would be zero. For net investment to be a positive constant, output must increase.

2.2.7. Theories Related to Some Factor affecting Investment

A. Interest rate

If interest rates are high, investors will find it expensive to borrow money, which would lead to a
decline in investment as many investments are financed by borrowing. Saving money is also

8
more appealing in the current economic climate due to high interest rates. Retained profits are
frequently used to finance investments. High interest rates make investing less appealing than
bank savings in general.

Mankiw (2007), the quantity of investment is determinant by the level of interest rate if the
interest rate is rise, fewer investment projects are profitable and the quantity of investment goods
demanded falls. When the overall changes of price level is essential to make distinction between
the nominal and the real interest rate nominal interest rate is the rate at which the inventory pay
for borrowing money, while the real interest rate is the rate at which is after adjusted the rate of
inflation. Therefore, Mankiw summarized that inverse relationship between investment and the
real interest rate.

B. Real Gross Domestic Product

Real gross domestic product (GDP) is an inflation-adjusted measure that reflects the value of all
goods and services produced by an economy in a given year (expressed in base-year prices) and
is often referred to as constant-price GDP, inflation-corrected GDP, or constant dollar GDP.

Economic growth represents the increase in the amount of the goods and services produced by an
economy over time. It is conventionally measured as percentage rate of increase in real gross
domestic product. Domestic investment represents gross fixed capital formation or gross
domestic fixed investment. The rate of economic growth also affects the level of investment.
Business investment tends to be quite volatile. If businesses see an improvement in economic
forecasts, they will increase investment to meet future demand. Therefore, an improvement in the
rate of economic growth can cause a substantial rise in investment. But, if there is an economic
downturn and a fall in the rate of economic growth, business will cut back on investment (World
Bank, 2008)

C. Inflation rate

Inflation is a rise in the average cost of goods and services over time. Rising inflation erodes the
purchasing power of a bond's future (fixed) coupon income, reducing the present value of its
future fixed cash flows. Accelerating inflation is even more detrimental to longer-term bonds,
given the cumulative impact of lower purchasing power for cash flows received far in the future.
The goal with an investment is to increase returns and long-term purchasing power, but inflation

9
puts this goal at risk. As inflation increases, the value of the investment diminishes, and the
consumer ends up paying more for less because of the decreased value of the dollar (U.S. Bank
2020).

D. Capital Expenditure

Capital expenditure is the part of the government spending that goes into the creation of assets
like schools, colleges, hospitals, roads, bridges, dams, railway lines, airports and seaports.
Capital expenditure also covers the acquisition of equipment and machinery by the government,
including those for defense purposes. Capital expenditure also includes investment by the
government that yields profits or dividend in future. This reveals that government capital
expenditure has positive impact on manufacturing sector output and others.

E. Real exchange rate

Investment is a determinant factor of output level, productivity and growth. Changes in exchange
rate may cause changes in the profitability of production and investment incentives. Exchange
rate increase may increase the demand of domestic products and the cost of imported capital and
other imported inputs. It will cause investment increase only if the impact on demand is more
than the cost effect. There are some factors affecting the optimal response of an industry
investment policy to exchange rate changes. It includes the reliance on imported inputs and the
share of foreign sales in total sales. If a firm is more dependent on imported inputs, there will be
more variable costs and less marginal value of capital. So a depreciation of exchange rate causes
a reduction in the level of industrial investment. By contrast, there will be increase in price
competitiveness a firm following an exchange rate appreciation. This is likely leads to an
increase in the expected value of capital and its level of investment. There is a negative and
statistically significant impact of real exchange rate movements on manufacturing investment.
(Reza L, 2013).

2.3 Empirical Review of Related Literature

Any expert research is a response to the persistent problem in the field. The majority of empirical
research is conducted in relation to the investment in the manufacturing sector in many nations,
according to approaches to this idea. The study conducted in Ethiopia and other developing
nations will be briefly discussed in this part.

10
Empirical studies conducted in Africa, Asia and Latin America have shown a critical relationship
between investment and growth rate (Ghura DA, 1996), (Collier, 1999).

In the 1990s, the ratio between total gross domestic investment and gross domestic product in
Asia, which had a high average growth rate compared to the rest of the world, was around 27%,
while in Latin America and Sub-Africa-Sahara the corresponding rates were 20% or 17%
(Herandez-Cato, 2000)

2.3.1. Evidence from developing countries

Alan et.al (2008) aimed to investigate the determinants of increased investment in Sub-Saharan
African manufacturing sectors. The outcome demonstrates that Sub-Saharan Africa's
manufacturing sector investment growth is flat. The research listed factors like political unrest,
corruption, and a lack of infrastructure as the causes of this increase. This study also looks at the
lack of a connection between the manufacturing sector's investment and independent variables
including FDI, interest rates, inflation, and government incentives in Sub Saharan African
countries in the year 2008-2010.

In Rwanda large manufacturing firms are productive than small firms since most of them are
dominated by foreigner which have skilled man power. Enterprises which are found around the
capital city (Kigali) are more productive than other because they have access to infrastructure,
skilled employer and information. Beginner firms have higher probability to invest than the old
one and credit did not play considerable role in the investment decision of Rwanda (Kamarudeen
and Soderbon, 2013).

A paper which was conducted by Elhiraika (2008) examine the main determinant of investment
in the manufacturing sector for 36 countries by analyzing total output in relation to the growth of
growth domestic product and growth volatility. The data used in the model are both time series
and cross sectional data between the years 1980-2007 to estimate the result. Investment in
African countries was stagnant it shows that lower performance of the sector with smaller
difference among the countries of the continent. But very few countries had high investment rate
compared to other country. For example, the ratio of domestic investment to GDP in Botswana
and Mauritius was 29%, in Gabon it was 30% and in Lesotho it was 59% between the years

11
1980-1995.However the general level of investment to GDP ratio was higher in Algeria that was
recorded about31.7% over the period (Bezawork, 2007).

According to Mbugua (2000) which analysis private investment in Kenya by using OLS method
of estimation between the year 1964-1998.And these shows that private investment in the
manufacturing sector is positively related to determinant factors like real interest rate, public
expenditure on infrastructure, lagged credit to private sector, real GDP lagged and public
expenditure on education. But public debt has negative impact on private investment in the
manufacturing sector of Kenya.

The result shows that the level of income at the initial level of investment had positively and
significant effect on aggregate level of manufacturing output. It also tells that industrialization
and population growth go in the same way. That means as the number of population increase the
growth of industrialization will be improved since it results to increase the demand manufactured
output. But the effect of population on the environment is not mentioned in this paper

Greene and Villanueva (1991) made a study on 23 countries to capture the determinants of
investment in developing countries. They did not actually use and explicitly specified model for
their estimation. This is because of complications involved in applying standard investment
models to private investment behaviors in developing countries. Greene and Villanueva found
real per capita income growth rate, per capita GDP level, real interest rate, public investment,
domestic inflation rate and external debt burden to be significant. They found private investment
to be positively affected by the level of per capita GDP, higher growth rate of real per income
and public investment. On the other hand, they found private investment to be negatively
affected by real deposit interest rate, high debt burden and high inflation.

2.3.2. Evidence from Ethiopia

A study by Workie (1996) suggested on constraints to entry, operation and expansion of private
investment in Ethiopia. Using investor level information showed that bureaucratic procedures, a
lack of infrastructure, power supply problems and access to finance were the leading constraints
for operations. The other areas of the business environment (such as political/policy uncertainty
and labor regulations) were relatively less important. The survey ultimately confirmed that the
availability of finance rather than the interest rate is a crucial determinant of private investment

12
in Ethiopia. Macroeconomic instability and political/policy uncertainty were not found to be
significant determinants of private investment.

Getachew (1997) studied the determinants of private industrial investment in Ethiopia using
descriptive statistics to analyze micro-level determinants. He found that the real interest rate did
not have a significant impact on private investment in Ethiopia. The study revealed that private
investment was positively affected by credit disbursement to the private sector in Ethiopia. It
also found that severe constraining factors to private manufacturing investment were market,
finance, infrastructure, policy, technology, and input related ones. Dawit (2010) showed that the
following are the success factors for private investment; the maintenance of good accounting
records by firms, good managerial skill, experience, government support and training. The major
problems are a lack of proper planning and feasibility studies, lack of skilled staff, delays in
securing bank loans, a lack of market for output, infrastructure problems and inflation.

Ambaye, Berhanu and Abera (2014) 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.

2.3.3 Determinants of Investment in the Manufacturing Sector in Ethiopia

The paper lacks description of all variables included in the model like Sintayehu Nigussie (2013)
and Bezawork Hailu, (2007). For example, budget deficit and capital expenditure were included
in the model but in the regression analysis there is no any result showing the relationship
between these independent variables and the dependent variable. The other limitation is that the
study did not show tests of regression such as stationary test, co-integration test and the paper did
not indicate normality of the variables. The research also did not describe anything about the
skewness and kurtosis of the variables.

Soderbom (2011) uses the data between the years 1998 -2008 to analyze the structural change
manufacturing sector Ethiopia. This paper focuses on to examine firm size and their economic
performance measured by value added per worker. The main issue of the paper is that whether
the product category is reasonably homogenous or not. The researcher investigates that product

13
selection difference exists across small and large firms. For 14 products under consideration, the
size coefficient is statistically significant at 10% level of significance. As the result, large firms
are likely to produce beer, soft drink, cotton, cotton fabrics, cotton yarn, leather garment, hides,
cement and wires. On the other hand, small firms are more likely to produce edible oil, cakes,
bread, gravel, plastic, foot wear and cement blocks.

The economic climate in Ethiopia in 2002 was examined by Kaufmann (2003). They came to
several conclusions, including that the business environment in Ethiopia was worse than the
average in sub-Saharan Africa in terms of "Voice and Accountability," "Political Stability,"
"Government Effectiveness," and "Regulatory Quality," but that the country's situation regarding
"Rule of Law" and "Control of Corruption" was better than the sub-Saharan average.

According to the study large firms produce more physical output than small firms since large
firms have more skill than the smaller one. But the researcher does not test this possibility
because of lack of data on human capital of employees. However, the result does not show the
long run trend of the Sectoral change since it uses cross sectional data. To indicate the long run
trend of the sector, we have to use long run determinants of investment such as; GDP, inflation,
interest rate etc. Second; the paper mainly focuses on firm size while there are many variables
which affect industrial performance. Some the main factors are FDI to the manufacturing sector,
government incentive to the industry and inflation.

The majority of reviews considered the factors relating to socio-economic to identify the
determinants of investment in the manufacturing sector and also applied to other African
countries and Ethiopia. Further, the certain factors were not considered in this reviews like
corruption, foreign reserves and Industrial gross domestic product. Hence, the researcher fills the
gap to ascertain the factors affecting growth of investment in manufacturing sectors the case of
Ethiopia.

2.4. Conceptual Framework

The empirical studies reviewed so far described different factors that causing low growth
investment in the manufacturing sector. Also, theories reviewed above portray these factors are
in turn influenced by the amount of investment. Yet, those determining factors have an impact on
the nation's investment, either favorably or unfavorably. However, what is empirically observed,

14
in developing countries like Ethiopia, is very far from what investment theory and different
literatures suggests for. The conceptual diagram below illustrates this fact.

Diagram1

Investment in MS

Decline incentive to Invest rises incentive to invest

In MS (low investment ` in MS (high investment

•High Corruption low Conflicts

•Low property right high property right

•High deposit interest rate high real IGDP

•High level of inflation low growth of low deposit interest rate

•High capital expenditure investment in MS low level inflation

•Low Foreign reserve low capital expenditure

•High real exchange rate low real exchange rate

•High Conflicts low corruption

Source: ow.

15
Clearly, an increase in real GDP of the country increases per capita income of customer. This
would increase the demand for saving by households. Owing to this, Availability of credit with
low interest rate increase, resulting in increased investment.

High capital expenditure is another factor causing decreased incentive to invest. High capital
expenditure is essentially increased budget deficit because high government borrowing and less
fund available for private investors. Increased in budget deficit causes incentive volume to invest
to decrease which ultimately results in decreased investment.

High deposit interest rate can also decrease investment. High deposit interest rate implies return
from save greater than from investing. And also, it implies an Investors save rather than
investing. This would, together, cause an increase in the amount of save volume which indeed
brings decreasing investment.

Still important, a high level of inflation will bring a declining investment. The reason is that a
high inflation disincentive domestic inventors to purchase foreign goods and capital due to the
rise real coast of imported input, and also due to the fact the price of imported goods has now

decreased which is the enabling factor for increased consumption of imported goods and high
opportunity cost of investing. This together may result in a declining level of Ethiopia’s
investment.

Lastly, but not the least, the high real exchange rate of the country causes the decline in real
value of domestic currency. Ethiopia, largely, exports primary commodities which are mainly
agricultural products in kind. These products are low quality products in that there is no value
addition that makes increased demand for it, that means, it is of inferior quality for importers.
This is partly a cause for cost push inflation and low country’s investment.

16
CHAPTER THREE
3. Methodology of the Study
3.1 Model Specification
Several works of literature conducted on the determinants of growth of investment in
manufacturing sectors in different countries have detected that it is affected by several factors.
Given the data constraints in Ethiopia, the study uses the following model.

Following Alan et.al (2008) we specify the investment in manufacturing sectors function as:

IMS =f (IGDP, FRS, CEX, REER, INFR, DIR) (IGDP, FRS, CEX, REER, INFR, DIR,)

LogIMS = α + β1LogIGDP+ β2LogFRS + β3LogCEX + β4LogREER+ β5INFR+β6DIR+

β7CORP+ ε

Expected sign (+) (+) (-) (-) (-) (-)

Where, IMS = investment in manufacturing sector

IMSX=IMS and its determinants

IGDP =Industrial Gross domestic product

FRS=Foreign reserves

CEX = Capital expenditure

REER = Real effective exchange rate

INFR = Inflation rate

DIR= Deposit interest rate

CORP=Corruption

βi= is coefficient of variable

α = is constant term

17
ε = is error term

The Explanatory variables

1.Industrial GDP: It is the value of all final goods and services produced in the industrial sector
of the country within a given period of time. It was measured by Industry Value Added (% of
GDP): It comprises value added in mining, manufacturing (also reported as a separate subgroup),
Textile, leather, consumption goods, construction, electricity, water, gas, etc. The data was
collected from the World Bank development indicators for Ethiopia.

2.Foreign Reserves: It is international reserves or foreign currency that the government or

central bank holds. It is measured by Total reserves in months of imports which show reserves
expressed in terms of the number of months of imports of goods and services they could pay for.
The data was collected from the World Bank development indicators for Ethiopia.

3.Capital expenditure; Capital expenditure is the money spent by the government on the
development of machinery, equipment, building, health facilities, education, etc. It also includes
the expenditure incurred on acquiring fixed assets like land and investment by the government
that gives profits or dividend in future. The data was collected from the National Bank of
Ethiopia (NBE).

4.Real effective exchange rate: exchange rate data was collected from the National Bank of
Ethiopia (NBE) and the coefficient of the variable representing exchange rate is expected to be
negative. This is because devaluation of domestic currency will result to increase the cost of
import in term of domestic currency. Most developing countries import machinery which leads
to increase the cost investment activities.

5.Inflation: it is simply defined as sustained growth of the general price level over time. High
inflation rate adversely affect long term investment by increasing the cost of raw material and
other supply of input. High inflation rate is an indicator of macroeconomic instability. The data
was collected from the National Bank of Ethiopia (NBE).

6.Deposit interest rate: It is the amount of money paid to the depositor as a return when they
deposit money in the bank. The data was collected from the National Bank of Ethiopia (NBE)

18
and the coefficient of the term representing interest rate is expected to have negative with the
dependent variable. This is because as the levels of deposit interest rate increases investors are
not investing their money and capital in the manufacturing sector rather they save it.

3.2 Data type and source

The data that is going to be used in this paper are secondary, which will be accessed from
different sources such as Central Statistical Agency (CSA), and National Bank of Ethiopia
(NBE), Investment commission of Ethiopia (ICE), World Bank (WB), international monetary
fund (IMF) and other concerned organization. For general descriptive analysis, the study will use
processed data from international journals, magazines, and periodicals which are related to the
study and different publications.

3.3 Estimation Method

In this study time series econometric analysis are used and the estimation technique will be
employed, as a matter of doing this paper, is Ordinary Least Square (OLS) method. The
estimation procedure will have three stages. The first is the test of stationary in order to eliminate
the possibility of spurious regression results; the second stage is test for co-integration. The
essence of co-integration is to ascertain whether the residual of the regression estimated using the
non-stationary variables is stationary. In the last stage, the short run models will be estimated
using Error correction model (ECM).

3.3.1 The Unit Root test for stationary

The first step in time series regression analysis is to test stationarity of each variable. The need to
test stationary of the variables arises because estimating regression using non-stationary base on
OLS leads to spurious and inconsistent result (Gujarat, 1995). In addition, if variables are non-
stationary it is difficult to conduct hypothesis testing as the classical assumption on the property
of the error term namely that it has zero mean, constant variance, and is non-auto correlated is
violated (Rao, 1994). Therefore, stationarity test is important.

There are different ways of testing stationarity. In this paper, the two widely applicable (and
most available in statistical software) test of unit root, namely the Dickey-fuller (DF) and
Augmented Dickey-Fuller (ADF) will be used.

19
It is known that most time series variables are non-stationary at level. Differencing the respective
variables and running regression on the same can handle the non-stationarity problem. However,
this method suffers from the problem that information about the long run relationship between
the variable is lost, since in the long run first difference of these variable are zero (Yuan and
Kochhar, 1994).

3.3.2 Co-integration Test

The concept of co-integration means that despite being individually non-stationary, a linear
combination of two or more-time series variables can be stationary (Rao, 1994). Co-integration
of two (or more) time series variables suggests that there is a long-run equilibrium relationship
between the variables. The two widely employed approaches for testing co-integration
relationship are the Engle-Granger (1997) two step procedure and Johansen (1988) maximum
likelihood approach. In this paper, the Engle and Granger two-step residual based procedure is
used. In the Engle-Granger approach the first step is to estimate the co-integrating regressions
and then to test whether the residual obtained from the co-integrating regression is stationary or
not, if the residual is stationary then the independent and dependent variables have long-run
relationship (Rao, 1994).

3.3.3 Error Correction Model (ECM)

The ECM indicates the short run dynamics of the OLS estimation results and its adjustment
towards the long-run equilibrium. Error correction Model (ECM) removes non-stationary from
the individual series in order to make the conventional classical regression techniques applicable
and to correct the disequilibrium error created in the short run (Gujarati, 1995). Regressing the
first difference of the dependent variables with the first difference of the independent variables
(s) using OLS shows the short-run dynamics of the model, but we can use the one period lagged
error term to tie the short run behavior of the dependent variable to its long run value (Gujarat,
1995).

3.4 Post-Estimation Diagnostic Tests

As a matter of detection, before interpreting the findings of the regression it is necessary to test

for Heteroscedasticity, autocorrelation, multicollinearity, normality and model specification.

20
3.4.1 Heteroscedasticity Test

It occurs when the variance of the error term changes with the changes in the value of regressors.
The existence of Heteroscedasticity is a major concern in the application of regression analysis
including the analysis of variance as it can invalidate statistical tests of significance that assume
that the modelling errors are uncorrelated and uniform, and hence that their variances do not vary
with the effects being modeled. But if there is Heteroscedasticity in the specified model the
variance varies and there is also correlation in the model. The paper will use Breech-Pagan /
Cook-Weinberg test for testing the presence or absence of Heteroscedasticity in the model. This
method requires regressing the square of the residual on the regresses (including the constant
term). Then multiplying the number of observations by the R-squared from the auxiliary
regression. The decision rule is to accept the null, which says there is Heteroscedasticity in the
model, if p-value is greater than 5 % critical value.

3.4.2 Auto-correlation test

In the regression estimation, the most important assumption is that the consecutive error terms
are not correlated or there is no auto-correlation. It arises when the error terms are serially
correlated. Running regression estimation by disregarding auto-correlation will result in
inefficiency on the estimated result and its standard errors are estimated in the wrong way. In this
study, will use Dubbin-Watson d test because it most important method in testing for the
presence or absence of serial correlation is the. To be the residuals serially uncorrelated the DW
statistics should be two(d=2). If d is close to 4 there is negative correlation and if it approaches to
0 then it is hilly positively correlated. The tolerable serial correlation is if d is between 1.5 and
2.5 in the DurbinWatson d test.

3.4.3 Multi-collinearity test

Multi col-linearity refers to a case in which two or more explanatory variables in the regression
model are highly correlated, and making it difficult to isolate their individual effects on the
dependent variable. The existence of the problem of multi-collinearity is tested using correlation
coefficient test and variance inflation factor (VIF). Correlation above 0.5 between independent
variables indicates the existence of problem of multi-collinearity. Furthermore, VIF above 10

21
shows the existence of multi-collinearity (Gujarati, 2007). The decision rule is that if the mean
value of VIF (βk) > 10, there is multi-collinearity problem and may deserve farther investigation.

3.4.4 Normality Test

There are various methods to test for normality such as the Graphical method, Anderson-Darling
test, Normal probability plot, Shapiro-Will W test, Jarque-Bera test etc. However, in this study
the Shapiro-Will W test for normality will be employed. The Shapiro-will test is chosen in this
study because it works better with a sample size less than 50. This test is a way to tell if a
random sample is from normal distribution, meaning that, the test tests the null hypothesis that a
sample x1… xn come from a normally distributed population. In order to determine whether the
long run model is normally distributed we use the W test.

3.4.5 Model Specification Test

The model specification error can occur when one or more variables that are relevant have
omitted and/or some variables, which are irrelevant, are included in the model. There are several
methods of detecting specification error in different literature's. This study however, will
employs Ramsey, RESET (regression specification error test) for omitted variables using the
power of fitted values. The decision rule is that if the P-value exceeds the chosen level of
significance, it indicates that the model does not have miss specification problem.

22
CHAPTER FOUR

4. Result, Discussion, and Analysis

4.1 An Overview of Patterns, Trend and Performance of Ethiopian Investment
Industry is one of the major economic sectors which should play significant role in the Ethiopian
economy. Ethiopia is one of the few African countries which its manufacturing sector highly
depends on imported capital goods and raw materials. The effect of such a pattern of industrial
development is well known. In this section, the study concentrates on the analytical part of the
relationship between manufacturing investment and the explanatory variable in Ethiopia. But
before going to the more rigorous econometric part of the study, it is important to see its trend in
Ethiopia. In addition, the review of the policy environment and the performance of
manufacturing investment in Ethiopia would also be presented. Investment in Ethiopia especially
the manufacturing sector development has been witnessing different experiences in the Derg and
EPRDF regimes of the country.

4.1.1. Structure and Performance of Manufacturing Investment in Ethiopia(IMS).

Ethiopia’s manufacturing sector is among the key productive sectors of the economy identified
under GTP I (2010-2015) which can spur economic growth and development because of its
immense potential for wealth creation, employment generation and poverty alleviation. The
manufacturing sector makes an important contribution to the Ethiopian economy and employs
most thousand people in the year 2017/2018. Structural changes within the manufacturing sector
have consequences for the patterns of investment, employment and policy formulation. The
structural change in the sector can be roughly represented by changes in the shares of the
industrial branches in total manufacturing output.

23
After the overthrow of the military government, the transitional government of Ethiopia
recognized the need for increased participation of the private sector in economic development.
Proclamations that encouraged private investment were issued. Economic and social
development of the country was intensified by increasing the supply of goods and services and
promoting domestic investment, particularly in the production sector in order to ensure linkages
and suitable development. Economic reform was introduced, which include short term economic
stabilization and structural adjustment through macro-economic stability measure. During this
time the economy showed a little growth in general and the manufacturing sector in particular.
The industrial policy was implemented through various sub-sector strategies and through the
successive five years ’development plans such as Sustainable Development and Poverty
Reduction Program (SDPRP) 2003-2005, the Plan of Action for Sustainable Development and
Eradication of Poverty (PASDEP) 2006-2010, the Growth and Transformation Plan (GTP1)
2011-2015, and the second growth and transformation plan (GTPII). The SDPRP gave great
emphasis to smallholder agriculture while in the PASDEP, GTPI and GTPII ones the policy
scope was broadened to encompass urban and the industrial sector development. The main
emphasis of the industrial strategy is to actively support the export-oriented and labor-intensive
sectors.

Figure 1: Manufacturing Value added for the Total GDP

Manfucturing value added (%GDP)

8

0
90 992 994 996 998 000 002 004 006 008 010 012 014 016 018 020
19 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2

Source: own computation using WB Data.

24
The figure above shows that up to the period-1992 there was little growth of value addition in the
manufacturing sector, and after the year 1992 up to 1997 there was slightly good growth of
investment on the sector this may be because of early policy reforms by newly formed
transitional government of Ethiopia that favored to the sector. But, in year 1997 there was a
sudden decline of the value addition for this year, and after this the investment in the sector was
decreasing at decreasing rate up to year 2012. After year 2012, it starts to spark up with greater
rate and reach maximum at year 2017. However, from year 2018 onward there was a sharp
decline in investment in the sector, and this poor performance of the sector may also have
resulted from the internal political instability that the country faced in this period.

4.1.2. The Structure and Performance of Ethiopian industrialization

Industrialization is regarded as the key to economic development and cultural change. It has
therefore become a declared aim of all countries. The preoccupation of development economists
has been how to maintain economic growth in advanced countries and the industrialization of
developing countries.
National strategy for industrialization remained more or less the same both before and after the
institutional changes of 1974. As was reflected in the Ten-Year Perspective Plan, the government
chose to emphasize light manufacturing, home market orientation and employment generation.
The choice of an industrial strategy will have an added significance for the national economy in
accordance with the growing importance of the manufacturing sector. It is known that Ethiopia’s
Industry Value added is dominated by value added in mining, manufacturing (also reported as a
separate subgroup), Textile, leather, consumption goods, construction, electricity, water, gas, etc.
To bring industrialization Ethiopian government undertake series of Industrial policies and
strategies. Since 1990s, Ethiopian industrialization has undertaken extensive structural reforms to
bring the country toward market economy. This first step was started by adopting the structural
adjustment program (SAP) guided by international financial institutions mainly World Bank and
IMF. In mid 1990s the government also formulated Agricultural Development Led
Industrialization (ADLI) as its development vision. To strengthen the role of ADLI in increasing
export volume to alleviate the trade balance problems the Export Promotion Strategy (EPS) was
adopted in 1998 which later on developed in to Industrial Development Strategy (IDS) in 2003.
26 The other industrial policies were implemented through various sub-sector strategies and
through the successive five years ’development plans such as Sustainable Development and

25
Poverty Reduction Program (SDPRP) 2003-2005, the Plan of Action for Sustainable
Development and Eradication of Poverty (PASDEP) 2006-2010, the Growth and Transformation
Plan (GTP1) 2011-2015, and the second growth and transformation plan (GTPII). The SDPRP
gave great emphasis to smallholder agriculture while in the PASDEP, GTPI and GTPII ones the
policy scope was broadened to encompass urban and the industrial sector development. Through
this process the Ethiopian government has been tried to bring industrialization for past 30 years.

4.1.3 Trends of Explanatory variables and Investment in Manufacturing Sectors.

4.1.3.1 Trend of industrial GDP and Investment in Manufacturing Sectors.
Industrial GDP: It is the value of all final goods and services produced in the industrial sector
of the country within a given period of time. It was measured by Industry Value Added (% of
GDP): It comprises value added in mining, manufacturing (also reported as a separate subgroup),
Textile, leather, consumption goods, construction, electricity, water, gas, etc. Therefore, based on
the briefing from different reports the trends of industrial GDP and its value adding and
Investment in Manufacturing Sectors from year 1990 to 2020 was provided a follows:

Figure 2: Trends of Industrial Value Addition (% of GDP) and Investment in

Manufacturing Sectors for Ethiopia.
30

25

20

15

10

0
90 992 994 996 998 000 002 004 006 008 010 012 014 016 018 020
19 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2

IMS IGDP

Source: WB (2020/2021) and own sketch

The above figure illustrates that up to 1992 there was sharp decline in industrial value adding.
This phenomenon may be occurred because of the past poor performance of the Ethiopian

26
economy under dergue regime. Also, the result of internal political instability- prolonged civil
war and the socialist economic policy that hinder the freedom for private investors to invest in
the sector and undergoing regime change in the country may also be the factor for poor
performance of the sector.

During the period from1992/93 to 1997/98, there was sharp increases in the value addition of the
industrial sector. Since 1998 there was stable growth in value adding of the industrial sector.
This is because of early economic and political reform by new government and the shift of
economy of from command to market oriented- this reform increased participation of the private
sector in economic development, and it allows private sector to invest in industrial sector. The
other positive reform was early Industrialization policies that have undertaken extensive
structural reforms to increasing the productivity of industrial sector under private ownership and
non-economic factor may also attribute to the relative political stability under new government
that create enable environment for foreign investors to invest in the country. Moreover, from
2012-2018 there was a sharp increase in value adding in the sector, and this may also because of
the acceleration of Ethiopia’s real gross domestic product (GDP) growth accelerated during that
period. Industry, mainly construction, textile factories by foreign investors, and services
accounted for most of the growth.

Lastly, from 2018, there was decline in the industrial sector value addition. This may also have
attributed to the weakness of EPRDF in its last regime, and the resulting protest by youth that
embarked on internal political instability. This internal political problem creates bad image of the
country, and reduces the inflow of the FDI during this period. During this period even those
industries that were under operation before were faced difficulties in smooth daily production,
and they forced to stop producing by closing their factory. In summary, the highest value adding
was recorded in year 2018 which is 27.3% and the lowest value adding was recorded during
1992- which is 6.1% from the given time period. Generally, the graph depicts that Industrial
Value Addition (% of GDP) and Investment in Manufacturing Sectors are moving together in the
same direction.

4.1.3.2 Trend of Foreign Reserves and Investment in Manufacturing Sectors.

Foreign Reserves: It is international reserves or foreign currency that the government or central
bank holds. Foreign reserves may include treasury bills, bonds, bank deposits, banknotes, and

27
other government securities. Some people include IMF funds or gold reserves. Since the dollar is
the most traded currency globally, the majority of these reserves are stored in this currency.
These assets have multiple uses, but they are primarily kept as a safety net in case a central
government agency's home currency depreciates sharply or goes bankrupt altogether. The ideal
way to retain foreign exchange reserves, according to economists, is in a currency that is not
connected to the nation's own currency.

Despite the fact that the country's foreign reserves have a lot of economic importance, for this
study we concentrated on the analysis of the role on financing imports and its role in reducing
volatility of the value of domestic currency in the global market. Permission is required for all
international payments, and the NBE approved authorized dealers must handle all foreign
exchange transactions. The National Bank of Ethiopia accounts must be used to fund imports. To
accomplish this, the National Bank of Ethiopia should maintain sufficient foreign reserves to pay
importers on demand from their foreign currency account. Since most of the inputs of the
manufacturing sector are imported, the availability of foreign reserve in the safe of national bank
of Ethiopia is necessary to ensure the smooth growth of the investments in the manufacturing
sector. The cross-country evidence shows that an increase in foreign exchange reserves raises
external debt outstanding and shortens debt maturity. The results also imply that increased
foreign exchange reserves may lead to a decline in consumption, but can also enhance
investment and economic growth. The increased availability of the foreign reserve has a positive
impact on manufacturing investment and its output.

The trends of Ethiopia's national bank's foreign reserves, as measured by total reserves in months
of imports, were provided based on the framework mentioned above as follows:

Figure 3: Trends of Total Foreign Reserves in Month Import and Investment in

Manufacturing Sectors for Ethiopia by the graph.

28
 8

0
90 992 994 996 998 000 002 004 006 008 010 012 014 016 018 020
19 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2

IMS TFRS

Source: own computation using WB and NBE data.

Figure 3 above indicates that even though total foreign reserve and investment in manufacturing
sectors move together in the general trend, when one is increasing the other is decreasing over
the trend. From 1990-1996 there was sharp growth of international reserves or total holding of
foreign reserve; from 1997-2001it was declined; from 2001-2007 it was fluctuated. In year 2008
it was declined to almost equal with the payment needed for imports of the month (that means
the amount of foreign reserve that available in the safe of national bank of Ethiopia and the
amount of foreign currency that needed to finance the import of the month is equal) and this
obligates monetary authority to postpone some imports for next time because of this shortage of
foreign currency and the investors faced the delay in inputs for their investment. This
phenomenon is associated with the world financial crisis of the year 2008 that attacked global
economy, especially US and China. Moreover, from 2009-2016 it showed rise of reserves with
some periodic fluctuation. Lastly, from 2016-2020 it was increased. Generally, the graph depicts
that trade balance and real effective exchange rate are moving together in the same direction

4.1.3.3. Trends of Capital expenditure and Investment in Manufacturing Sectors of

Ethiopia.
Capital expenditure is an amount spent to acquire or improve a long-term asset such as
equipment or buildings. It is one of the two kinds of government expenditure of which the other
is recurrent expenditure, which refers to an amount spent on the day-to day running of the state

29
and on the payment of salaries within a period of 12 months or a financial year. Furthermore,
capital expenditure refers to an outlay of cash for a project that is expected to produce a cash
inflow over a period of time exceeding one year. It is used to provide infrastructure. It also
increases labor participation, takes stock of the economy and raises its capacity to produce more
in future.

Figure 4: Trends of Capital Expenditure and Investment in Manufacturing Sectors for

Ethiopia by the graph.
300000000000

250000000000

200000000000

150000000000

100000000000

50000000000

0
90 92 94 96 98 00 02 04 06 08 10 12 14 16 18 20
19 19 19 19 19 20 20 20 20 20 20 20 20 20 20 20

IMS CXP

Source: own computation using WB and NBE data

The above graph depicts the trend of capital expenditure and investment in manufacturing
Sector. From 1990-2006 government final capital expenditure was recorded as low, on average,
of 4,654.4 million birr per year. From 2006-2008 it was increased; this is because of the
government commitment toward the improvement wellbeing of society at whole and its
ideological change to development state. From this, we conclude that, Ethiopian government
have to started to launch huge investment by itself and give subsidies for the private sector to
invest in huge mega project that can push the economy toward development. This would increase
investment in manufacturing sectors of the country. Both variables are moving together over the
general trend, moreover.

4.1.3.4. Trends of General Inflation rate and Investment in Manufacturing sectors of

Ethiopia

30
Inflation is a phenomenon characterized by the general rise in prices for both producers and
consumers for some time. When the general price level rises, each dollar buys fewer goods and
services, hence inflation indicates a reduction in purchasing power per dollar. Today,
manufacturers are facing a combination of unprecedented events that strongly impact their
operating margins. They are dealing with higher input costs across their business, not only from
raw materials but also from energy, components, packaging, and transportation. This is leading
to an overall increase in the prices consumers pay for their products and services as
manufacturers need to protect their margins. Based on the above framework from past studies
and the recognized negative effect of the inflation on investment in manufacturing sector, the
trends of inflation in Ethiopia was provided as follows:

Figure 5: Trends of General Inflation Rate and Investment in Manufacturing Sectors for
Ethiopia by the graph.
60

50

40

30

20

10

0
0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0
99 199 199 199 199 200 200 200 200 200 201 201 201 201 201 202
-101

-20

IMS INFR

Source: own computation WB and NBE Data.

Inflation is explained as the main determinant of economic growth. Instability in inflation creates
uncertainty in investment environment. The above figure illustrates that general inflation has
rapid up and down movement; the highest deflation was recorded during the period 1996 and
2001 on the other hand the highest inflation period was seen during year 1991 and 2008. This is
because of high growth of money supply; growth of money supply is supposed to be less than the
nominal money supply if it is greater leads to inflation. The other reason is agricultural

31
production is low during 2007/08 especially crop production and also the high import price of
oil. From 2013 to 2017 there has been a tolerable price level which was almost less than 10%. In
year 2021 the general inflation was shoot up to almost 21.5%. This is because of the disruption
of international supply chains as a result of the COVID-19 pandemic, the increase in the price of
fuel and raises in devaluation is the reason for the current inflation as stated by Economist,
Wossenseged Assefa. Generally, the above graph depicts that inflation affecting investment in
manufacturing sectors negatively.

4.1.3.5. Trend of Reel Effective Exchange Rate and Investment in Manufacturing Sectors of
Ethiopia

The real effective exchange rate (REER) is defined as real worth of foreign exchange in terms
of a given domestic currency. The appreciation or depreciation of domestic currency against
foreign currencies (change in exchange rate) is important variable in the demand for money
functions of developing countries. In Ethiopia, the nominal exchange rate was pegged to US
dollar for over three decades (including the last years of the imperial regime) and has been
administratively fixed at 2.07per US dollar. The fixed exchange rate policy coupled with other
restrictive trade policies were experienced in Ethiopia for several years results the development
of parallel foreign exchange market which further causes balance of payment deficit.

Figure 6: Trend of Reel Effective Exchange Rate and Investment in Manufacturing Sectors
of Ethiopia.

32
Source: own computation WB and NBE Data.

The above figure illustrates from 1990-2000 the real effective exchange rate is decreased, on
average, by 1.17%. This phenomenon is associated with the currency devaluation of the year
1992. Moreover, from 2000-2010 it showed an average growth by a small figure of 0.247%.
Lastly, from 2010-2020, it increased averagely by 0.87%. Generally, the graph depicts that
Investment in Manufacturing Sectors and real effective exchange rate are moving together in the
same direction.

4.1.3.6. Trends of Deposit interest rate and investment in manufacturing sectors of

Ethiopia.

Deposit interest rate: It is the amount of money paid to the depositor as a return when they
depose money in the bank. Deposit interest rate is controlled by central bank and kept constant
for a long period of time except some revisions made by NBE in certain times. For
manufacturing investment, the real interest rate is important for determining the viability of
investment. Lower interest rates make it cheaper to borrow. Therefore, lower interest rates tend
to encourage business investment. High real interest rates discourage investment.

Figure 7: Trends of Deposit interest rate and investment in manufacturing sectors of

Ethiopia.

Source: own computation WB and NBE Data.

33
The above graph depicts that from 1990-1992 the deposit rate was fixed at 4%. After that
nominal interest rate on saving deposit was raised to 10 percent in 1992/93 after being 6 percent
for about 4 years. In year 1997 it was set again at 6 percent and further declined to 6 percent by
1998 and it has remained the same for the following four years. Moreover, from 2001-2007 it
further declined to 3%. Lastly, from 2007-2020 the national bank of Ethiopia starts to rises
nominal interest rate because of high rate of inflation witnessed in Ethiopia. Recently it was
fixed at 7%. Both variables are moving together.

4.1.4 Summary of descriptive statistical analysis

This section presents the summary statistics of the variables considered in our model. The
original unit of measurement for each variable is in millions of birr except for inflation, deposit
interest rate and real effective exchange rate which measure in interest rate and unit of domestic
currency.

Table 4.1 summary of descriptive analysis

Variable Observation Mean Std. Dev. Min Max

LogMI 31 .6746606 .0911008 .4932489 .8633883

LogIGDP 31 4.556746 .4631962 3.99697 5.60669

INF 31 11.07097 14.09537 -10.8 55.2

LogREERI 31 2.137573 .1189449 1.972203 2.537189

LogCEX 31 4.153788 .7481301 2.961563 5.310005

LogTFRS 31 3.948882 .8145203 1.621176 4.894299

DIR 31 5.451613 2.188435 3 10

Source: Own computation using Stata software

From the table given above, the result shows that the average and standard deviation of the log
value of dependent variable i.e. IMS variables during the 1990-2020 is 6746606 million birr and
0.0911008 respectively. The minimum average value of IMS is recorded during 1992 and the

34
maximum average value is recorded during 1997 i.e. 0.5 million birr and 0.8634 million birr
respectively.

For the independent variable I included in my model (IGDP, TFRS, REERI, CEX, DIR, and
INF) is analyzed below. The variable IGDP reveal that the value of all final goods and services
produced in the industrial sector. The average and standard deviations of the log value of IGDP
is 4.556746 million birr and is .4631962 units respectively. The variable TFRS show that the
value of international reserves or foreign currency that the government or central bank holds.
The average and standard deviation of the log value of TFRS is 3.95 million birr and 0.8145
million birr respectively. The variable CEX reveal that value of spent to acquire or improve a
long-term asset such as equipment or buildings. The average and the standard deviation of log
value of CEX is 4.15 million birr and 0.75 million respectively. The lowest capital expenditure
that government put into economy is 2.96 million birr- which was recorded in year 1992 and the
highest was 5.31 million birr- which was recorded in last year 2020. The variable REER display
that the value of the weighted average of a country's currency in relation to an index or basket of
other major currency. The average value of Ethiopia birr relative to others is 2.138 units and it
vary by 0.119 units. The lowest value of LogREERI is about 1.98 units it recorded during 2002
and the highest value was 2.538 it recorded during 1992.

The variable INF show that the value of sustained growth of the general price level over time.
The average and standard deviation of INF is 1.07 percent and 14.09 percent respectively. The
lowest Inflation rate that was recorded during period of 2001is -10.8 that is deflation, and the
highest Inflation rate that was recorded is 55.2 percent which is recorded in year 2008. The
variable DIR reveal that value of the amount of money paid to the depositor as a return when
they deposit money in the bank. The average interest rate the depositors get as return from the
banks under this period is 5.45 percent and its standard deviation is 2.189 percent. The highest
interest rate that the depositors get as interest under this period was 10 percent which was
recorded from year 1993 to 1996 and the lowest was 3 percent which was recorded from year
2002 to 2007.

4.2 Pre-estimation Test

4.2.1 Tests for Stationarity (Unit Root Test)
The reason for knowing whether a variable has a unit root, that is, whether the variable is

35
nonstationary is that under the alternative hypothesis variables exhibits stationarity means that
they
undergo reversion characteristics, finite variance, covariance between two values of the series
depends only on the difference apart in time and autocorrelation die out as number of lag length
increases. However, under non-stationarity they don’t. Despite being non-stationary at level,
however, stationarity of economic variables obtained after differencing them once, twice, and so
on. This process tests the nature of the time series and examines their order of integration.

I used the ADF unit root test to evaluate the order of integration of variables to verify the
stationarity of variables. According to the ADF test results, all variables, except annual Foreign
reserve and inflation rate, are individually non-stationary i.e. first difference stationary I (1),but
there is the possibility that this regression result is spurious. However, there is still a possibility
of obtaining stationarity i.e. I (0), by subjecting the residuals obtained from the regression to unit
root test (Gujarati, 2004). Since there is mixed stationary, it is possible to apply the ARDL
estimation technique to capture the investment in manufacturing sectors in Ethiopia, because of
ARDL approach has the advantage that it does not require all variables to be I (1) as the
Johansen framework does and it is still applicable if we have I (0) and I (1) variables in our set.
That is, the variables included in the ARDL model are combinations of stationary and non-
stationary time-series for ARDL bounds testing approach proposed by Pesaran et al. (2001). That
is, based on these results we can apply the ARDL technique.

Table 4.2 Unit root tests

Variables DF Critical t 1% critical 5% critical 10% critical

statistics value value value
value

logIMS I(1) -5.967 -3.723 -2.989 -2.625

logIGDP I(1) -3.944 -3.723 -2.989 -2.625

logCEX I(1) -5.574 -3.723 -2.989 -2.625

logREERI I(1) -7.811 -3.723 -2.989 -2.625

DIR I(1) -4.971 -3.723 -2.989 --2.625

36
logTFRS I(0) -4.295 -3.716 -2.986 -2.624

INF I(0) -4.996 -3.716 -2.986 -2.624

Source: own computation

4.2.2 Optimal Lag Length

In estimating the ARDL model, determining the optimal lag length of each variable in the model
(both dependent and independent variables) is crucial. The best model fitted by STATA-17 has
different lag lengths for each variable in the model.

Table 4.3 Optimal lag length selected by AIC

Variables Optimal lag length

logIGDP 1
INF 0
logREER 1
logCEX 1
logTFRS 1
DIR 1

According to Lutkepohl (2006) the dynamic link among the series can be captured if proper lags
are identified and used. In our study, the optimal lag length for each variable is determined using
the Akaike information criterion (AIC) so that the selected model is ARDL (1 1 0 1 1 1 1). The
summary of the optimal lag length for each variable is given in Table 4.

4.3 Estimation
4.3.1 Test of Co-integration
Co-integration means that despite individually being non-stationary, a linear combination of two
or more time series can be stationary. In essence, it shows the existence of long-run relationship
among non-stationary variables. Testing for it is significant because differencing the variables to

37
 obtain stationarity may produce a model that does not show long-run behaviors of variables.
Hence, testing for co-integration is the same testing for long-run relationship. To be specific, if
variables which are integrated of order I‟ produce a liner combination that are integrated of order
less than I‟, say v‟, then the variables are co-integrated and hence have long-run relationship
(Gujarati, 2004). For this study, the researcher has implemented Engle Granger (EG) test to
whether there is long run relationship between variables, because we can’t use Johansson co-
integration test to for this data as it was not stationary in order one I (1)-which is necessary
condition to use Johansson integration test, rather this data displays mixed stationarity.

Table 4.4 Results of Augmented Engel Granger (AEG) test

Variable Test statistic 1% critical value 5% critical 10% critical value

value
Residual -6.375 -3.716 -2.986 -2.624

Source: own computation

Using AEG test methods, the result of the test indicates the existence of a long-run relationship
as the test statistics is the greatest at all critical levels for both I(0) and I(1) which is 6.375 and
3.716 respectively at the 1% level of significance (Table 4.3).This justifies a long-run
relationship between logMI, logIGDP logTFRS, logCEX DIR, logREERI and INF.

4.3.2 Long Run Estimation Results

The long run model is estimated by autoregressive distributed lag (ARDL) model is an ordinary
least square (OLS) based model which is deals with a single integration and was introduced by
Pesaran and Shin (1999) and further extended by Pesaran et al. (2001). The long run result model
that estimated by autoregressive distributed lag model is provided below.
Table 4.5 Estimation results of the long run IMS model

Variables Coefficient Standard error T P>|t|

logIGDP .4683971 .092665 5.05 0.000
INF -.0004412 .0011212 -0.39 0.699
logREER -.083222 .2947749 -0.28 0.0181
logCEX -.4104391 .1194841 -3.44 0.003

38
logTFRS .1490848 .1018557 1.46 0.0162
DIR .0007894 .008998 0.09 0.931
_cons -.1055082 .3740819 -0.28 0.781
Source: own computation

Number of obs = 30 Log likelihood = 63.319056

F (12, 17) = 12.26 Root MSE = 0.0389

Prob > F = 0.0000

R-squared = 0.7697

Adj R-squared = 0.6071

The estimation result of long run model is stated as

logMI=-.1055+ .4683971logIGDP -.083222 logREER- 4104391 logCEX+ .1490848

logTFRS- .0004412INF+ .0007894DIR

Table 4.6 Comparison of expected sign and model result

Variables Expected sign Model result sign

logIGDP +ve +ve
INF -ve -ve
logREER -ve -ve
logCEX -ve -ve
logTFRS +ve +ve
DIR -ve +ve

Where, +ve is the positive sign and -ve is the negative sign.

The above result displays that 76.79 percent of the variation in investment in manufacturing
sectors is explained by the given independent variables mutually. Putting differently, the model
explains 76.79 percent of the change in the explained variable. The F-test shows that all the
independent variables are jointly significant. And indicates that, except deposit interest rate and

39
inflation rate, all the independent variables are statistically significant and their sign are as
expected. The constant term, with t-value of -.1055, is insignificant.
The estimation results of the long run model reveal that investment in manufacturing sectors in
the country is positively affected by industrial GDP; total foreign reserves; deposit interest rate
and negatively affected by capital expenditure; real effective exchange rate and inflation rate.

The coefficient of industrial GDP is statistically significant, and its sign is what is expected.
Specifically, a 1% increase in industrial GDP results in a 46.83971 % increase in investment in
manufacturing sectors, ceteris paribus, which is statistically significant at the 1% level of
significance. This implies that, industrial GDP of the current year forces country to increase its
per capita income through increase household saving and various police, which thus rise
availability of credit with low interest rate in order to improve investment.

The coefficient of deposit interest rate is statistically insignificant with unexpected sign. This
result enables as to the conclusion that deposit interest rate is not the significant determinant of
investment in manufacturing sector of Ethiopia, and hence it doesn’t improve investment.

The econometrics result indicated that capital expenditure has expected sign, and there is
statistically significant relationship between capital expenditure and investment in manufacturing
sector. Specifically, a 1% increases in capital expenditure would lead, on average, to 41.04391%
decrease in investment in manufacturing sectors, ceteris paribus. This implies that increase in
government capital expenditure leads to budget deficit which is financed by government
borrowing and result in less fund available for private investors then it follows by decreasing in
investment because of crowding out effect of massive government borrowing that discourage
private investors by increase interest rate.

The coefficient of REER is statistically significant with an expected sign. This indicates that
devaluation of currency would discourage investment through increasing import price more than
export by making import expensive and export cheaper. Increased import price decreases peoples

40
purchasing power and increase cost of imported inputs then leads to cost push inflation. As such,
devaluation discourage investment since it depends on imported intermediate goods and the user
cost of capital. This is in line with economic theory. The coefficient of it is -.083222; it implies
that a 1% increase in REER of the country will decrease investment in manufacturing sectors by
8.322%, ceteris paribus.

The coefficient of total foreign reserve during the month of import is statistically significant, and
its sign is positive. This indicates that foreign reserve during the month of import is estimated to
increase investment in manufacturing sector by 14.90848%, ceteris paribus. This is because most
of the inputs of the manufacturing sector are imported, the more availability of foreign reserve in
the safe of national bank of Ethiopia increases the smooth growth of the investments in the
manufacturing sector by enable countries import machinery, equipment and other inputs then
surplus of input decrease operation cost to the firms as the result improve investment.

The coefficient of inflation rate is statistically insignificant with expected sign. This result
enables as to the conclusion that inflation rate is not the significant determinant of investment in
manufacturing sector of Ethiopia, and hence it discourages investment.

4.3.3 Results of the Short Run Model Based on ARDL-ECM

The ECM (error correction model) is significant for obtaining the short-run relationships among
the variables though they are in a state of long-run equilibrium. The result of ARDL-ECM
estimation is provided below.

Table 4.7 Estimation results of the Short-run based on the error correction model.

Variables Coefficient Standard error T P>|t|

logIGDP .1160701 .0978111 -1.19 0.0252

logREER -.2766254 .2092451 -1.32 0.204

logCEX -.1833688 .108517 1.69 0.109

logTFRS .1429405 .0623224 -2.29 0.035

41
 DIR -.0202311 .009002 -2.25 0.038

INF -.0002986 .0007324 -0.41 0.689

_cons -.1055082 .3740819 -0.28 0.781

Number of obs = 30

Coefficient of ECM = -.6768288

F (12, 17) = 12.26

Prob > F = 0.0000

R-squared = 0.8964

Adj R-squared = 0.8233

Root MSE = 0.0389

Table 4.5, reveals that the values of R2 and adjusted R2 are 89.64 percent and 82.33 percent
respectively in the short run model, that means, 89.64% and after adjustment 82.33% of the
variations in investment in manufacturing sectors is explained by all explanatory variables in the
short run model. The coefficient of deposit interest rate implies that a 1% increase in deposit
interest rate decrease investment in manufacturing sectors by 2.02311 %, holding others factor
constant. Also, the coefficient of foreign reserve during the month of import is positive and
statistically significant determinant of investment in manufacturing sectors in the short run. It
implies that, ceteris paribus, a 1% increase in foreign reserve during the month of import leads to
14.29405% increase in investment in manufacturing sectors. This was resulted from the more
availability of the foreign reserve in the banks during short run, the more investor will import
inputs for their project, and the more they invest in the sector.

The coefficient of industrial GDP show that Specifically, a 1% increase in industrial GDP result
in a 11.6% increase in investment in manufacturing sectors, which is statistically significant at
the 1% level of significance. This implies that If the output of industrial sector increases, and the

42
investors in sector become profitable, then those investors who are investing in other sector
would be attracted by high profit in the sector and they turn their investment capital to the
manufacturing sector as a result the investment in the sector increases.

Furthermore, the short run model shows that only industrial GDP, foreign reserve during the
month of import and deposit interest rate are statistically significant in explaining investment in
manufacturing sectors in the short run while others are insignificant variables. The sign of capital
expenditure is positive and statistically insignificant. Also, the sign of real effective exchange
rate is negative and statistically insignificant. These results, however, show that explanatory
variables did not adequately describe the model as the long run model did so. The reason is due
to the insensitiveness to change of investment in manufacturing sectors, in the short run, in most
cases (Dr. JONSE).

The error correction term is interpreted as the speed of adjustment towards long run equilibrium
or the measure of removing the disequilibrium of the investment in manufacturing sectors due to
various shocks to its long run equilibrium. Thus, the ECM coefficient measures how
quickly/slowly the relationship returns to its equilibrium path and it must have a statistically
significant coefficient with a negative sign.

In our short run model, the coefficient of the error correction term of the ARDL investment in
manufacturing model is negative and less than one. It is also statistically significant at less than a
1% level of significance. This result ensures that investment in manufacturing sectors converges
to its long run equilibrium. Its magnitude indicates that about 68% of the disequilibrium in the
investment in manufacturing sectors is corrected per year implying that the disequilibrium due to
various shocks is nearly corrected within one year. This shows that it takes about one year to
return to the long-run equilibrium level after the shocks which is a reasonably short period and
implies that there is long run relationship between the investment in manufacturing sectors and
its determinants, and a possibility of convergence to equilibrium level in the long-run. Hence, it
will take approximately one year and three months to recover from a single shock and restore to
long run equilibrium. (Table 4.5).

43
 4.4 Post Estimation/ Diagnostic Tests
Six diagnostic tests have been employed in the long run model of the study to check the problem
of multicollinearity, heteroscedasticity, serial correlation, non-normality of error term and
misspecification of the model. The result of all diagnostic tests are summarized on the following
two tables below.

Table 4.8 Test for multicollinearity problem

Variable logCEX logIGDP logTFRS logREER DIR INF

VIF 24.89 12.50 11.40 2.05 1.31 1.27

1/VIF 0.040171 0.07997 0.087709 0.486838 0.762002 0.785949

5
Mean VIF 8.91

Source: own computation using STATA software

From the table given above, the result shows that all variables have a VIF values less than ten.
The mean VIF of these variables is 8.91, that there is no problem of multicollinearity among
explanatory variables.

Table 4.7 Diagnostic/post estimation tests results for the long run model

Tests Test statistics P-value

ARCHLM Test 0.905 0.3414

Breusch-Godfrey LM Test 0.866 0.3521

Shapiro-Wilk W Test 0.96818 0.47040

Ramsey RESET(Ovtest) Test 3.57 0.3140

Link Test -0.39 0.699

44
Source own competition using STATA software

Table 4.7 gives the diagnostic test of the long run model. According the test statistics, the model
has no specification problem, has no heteroscedasticity problem, the residuals are normally
distributed with constant variance and there is no serial correlation. Thus, the classical
assumptions are satisfied indicating that the estimates of the efficiency and policy implications of
the model are reliable. More detail explained below.
ARCHLM test statistics has values of 0.905 with P-value of 0.3414, which is higher than the 5%
level of significance implying that there is no heteroscedasticity problem in the disturbance term
of the model. Therefore, the null hypothesis of homoscedasticity has fail to be rejected; meaning
that the variance of the disturbance term is constant.

The test for serial correlation of the residuals is undertaken using the Breusch-Gog Frey LM test.
The LM test for residual serial correlation results from table 4.7 confirm that we fail to reject the
null hypothesis of no serial correlation among the residuals of the long run model. The F-statistic
has the values of 0.866 with the P-values of 0.3521 which is greater than the chosen level of
significance i.e. 5% implying that there is no a problem of serial correlation.

In order to check whether the error terms are normally distributed the Shapiro-Wilk W test is
undertaken. As table 4.7 displays for an alpha level of 0.05, the data set has W-statistic value of
0.96818 with a P-value of 0.47040. Therefore, we do not reject the null hypothesis that the data
are from normally distributed population. This is due to the P-value which is greater than the
chosen level of significance (5%) implying that the error terms are normally distributed.

To identify whether the model is specified correctly or not, Ramsey RESET test for omission of
relevant variables and the link test, which performs the link test for model specification error has
employed. Accordingly, the result of RESET test confirms that we do not reject the null
hypothesis that there is no omission of relevant variables having the test statistics values of 3.57
with P-value of 0.3140. The result of link test also confirms that the t-test for hat-square is
insignificant having test statistics value of -0.39 with P-value 0.699 indicating that there is no
problem of link between the dependent and explanatory variables. Hence, the result of these two
tests supports that the long run model has no specification error.

45
In conclusion, the diagnostic (post estimation) tests shows that the long run model does not
suffer from multicollinearity, nonmorality, serial correlation, heteroscedasticity and
misspecification problems. Thus, the classical assumptions are satisfied indicating that the
estimates of the efficiency and policy implications of the model are reliable.

4.5 Analysis of Review Literature and Results of the Study.

Many literatures argue that industrial GDP affect investment growth in manufacturing sectors
positively (Dagnechaw,2019). In contrast, others posit it to affect investment growth in
manufacturing sectors negatively. Nevertheless, according to this study, industrial GDP is found
to affect investment growth in manufacturing sectors in a positively. This is could be brought
about by capturing an increases the output of industrial sector and the profitable investors in
sector. We also observed that the coefficient of industrial GDP is positive both in the short run
and long run, and hence it is the significant determinant of the investment growth in
manufacturing sectors The result is, thus, in line with many economic literatures and theories.

Very importantly, capital expenditure is found to affect investment growth in manufacturing

sectors negatively. Although savings-investment gap and the IS-LM Mundell-Fleming model
argue for government expenditure to have a positive effect on investment growth, the result of
our finding, however, indicates that government capital expenditure discourage investment
growth. This contradiction might be due to the fact that the country where this model is
developed and Ethiopia is different in level of development, and the time span used in
developing that model may also be different. These and the like reasons may make the results of
the study to be different.

No matter how our result is different, our finding would make economic intuitive for developing
countries like Ethiopia. This is because of massive government spending on mega projects in
short
run leads to a deficit that is financed by increased borrowing, then the government borrowing can
increase interest rates, leading to a reduction in private investment because of crowding out
effect in long run. The results of our study, therefore, has disproved the existing theories and
some literatures.

46
Interestingly, the coefficient of foreign reserve is positive both in the short run and long run, and
hence it is the significant determinant of the growth investment in manufacturing sectors. This
due to of many input of manufacturing sectors is imported; the more reserve of foreign currency
the more import then more investment in manufacturing.

Lastly, but not the least, inflation is found to affect investment in manufacturing sectors
negatively, and hence it is the insignificant determent of the growth investment in manufacturing
sectors. Since goal with an investment is to increase returns and long-term purchasing power, but
inflation puts this goal at risk. As inflation increases, the value of the investment diminishes, and
the consumer ends up paying more for less because of the decreased value of the dollar (U.S.
Bank 2020). This implies that our finding is in line with existing theory and literature.
CHAPTER FIVE
5. Concluding Remarks, Policy Implications, and Limitations of the Study
5.1 Conclusion
This study investigated the factors affecting the growth of Investment in the Manufacturing
Sector in Ethiopia over the period 1990–2021 using data from NBE, WB, CSA, and investment
commission. The study applied the ARDL and short run error correction model (ECM)
estimation technique to capture the factors affecting the growth of investment in the
manufacturing sector in Ethiopia. Specifically, industrial GDP, foreign reserves, inflation rate,
real effective exchange rate, deposit interest rate, and capital expenditure are the main
determinants chosen in this research.

The estimation results of the ARDL model show that all variables included in the model had
expected signs except the deposit interest rate. It indicates that in the long run investment in
manufacturing sectors of the country was positively and statistically significantly affected by
industrial GDP and total foreign reserves. Very importantly, their sign is as expected and they
are all statistically significant. The sign of the deposit interest rate and the inflation rate is not
what is expected, and it is statistically insignificant in the long run. When the industrial GDP
increased by 1%, the resulting investment in manufacturing sectors increased by 0.116%
implying that in Ethiopia investment in manufacturing pressure is a fiscal and monetary
phenomenon with nearly one-on-one effects. However, the current level of inflation had negative

47
and statistically insignificant effects on the investment in the manufacturing sectors of the
country.

Coming to the short-run model, only industrial GDP, total foreign reserves, and deposit interest
rate are statistically significant and, therefore, these are the only short-run determinants in
improving the growth of investment in manufacturing sectors of the country. The rest of the
variables have neither increase nor decrease the growth of investment in manufacturing sectors
in the short run because of their statistical insignificance. The coefficient of ECM is 0.6768
implying that 67.7 percent of the shock is corrected each year, and therefore it moves to long-run
equilibrium by this speed of adjustment. This shows that it takes about one year to return to the
long-run equilibrium level after the shocks which is a reasonably short period.

5.2 Policy Implication

Depending on the analysis of this study, as the policy directive and/or the way out to increase
Ethiopia’s competitiveness in the industrial sector and to achieve high investment growth in the
manufacturing sector, the following policy measures may help in achieving high investment
growth in the manufacturing sector of the country.

 Since the interest rate transmission channel is weak in developing countries like Ethiopia,
So the main emphasis for the policymaker should be to increase the aggregate demand in
the economy so that industrial output can be raised, which will help raise the general
investment level in the manufacturing sectors of the country.
 The government should have to give special emphasis on variables which have a significant
impact on the level of investment in the manufacturing sector since all the factors included
under study should be considered in conducting monetary and fiscal policy.
 The government should have to give incentives for the provision of loans, imported inputs,
acquired technologies, and research and development subsidies for investors so as to
increase investment in the sector and encourage the participation of the private sector
which plays a leading role in the expansion and development of the industrial sector.

48
  Keeping the value of the domestic currency stable as much as possible to avoid the
volatility of the exchange rate and create a good image of the country to increase the inflow
of foreign direct investment was also one policy implication of this study.
 The Government should also find different sources that brought foreign currency to its safe
so as to increase external reserve in the bank since it has a positive and significant effect on
investment in the manufacturing sector of the country both in the short run and long run.
 The government should also decrease its huge capital expenditure which leads to a deficit
that will be financed by increased borrowing, and then the government borrowing can
increase interest rates, leading to a reduction in private investment because of the crowding
out effect in the long run.
 Financial institutions should decrease the deposit interest rate so as to increase investment
in the manufacturing sector since it has a negative and significant impact on the level of
investment in the sector.

5.3 Limitations of the Study

Despite various attempts made to effectively examine the factors affecting growth of investment
in the manufacturing sectors of Ethiopia, it is not without limitations. To list a few, simple time
series econometrics is followed in this study. However, if panel econometrics is used the
estimation result becomes more effective since comparison of countries across time dimension
and country dimension is possible, and also it is possible to see which country’s investment in
manufacturing sectors shock is adjusted at a fast rate and which is not. Moreover, since I am a
practitioner rather than a research expert, the policy recommended is to be accepted to the limit
of my level. In addition, had the study used advanced models like vector error correction model,
the model would have better examined the factors affecting growth of investment in the
manufacturing sectors of Ethiopia well. Lastly, the variables used in this study are not the only
the factors affecting growth of investment in the manufacturing sectors of Ethiopia.

Hence, further studies have to include more explanatory variables that can explain the growth of
investment in the manufacturing sectors of Ethiopia well than simply depending on some
macroeconomic variables to fill the gap created because of the above problems. Panel
econometrics and advanced econometrics like vector error correction model have to be used in
further studies, and also this study has to be further done by research experts.

49
References
AACCSA (2014): A Survey of Manufacturing sector in Ethiopia, Addis Ababa

Dagnachew Assefa, (2018): Determinants of investment in manufacturing sector in ethiopia,

Addis Ababa.

Adam. B. Elhiraika, (2008): Promoting manufacturing to Accelerate Economic Growth and

Reduce Volatility in Africa. African Economic Conference: Globalization, Institutions and
Economic Development of Africa, 12-14 November 2008, Tunis, Tunisia.

Alan Harper, Raufu.O. Sokul and JP Morgan Chose (2008): Determinant of manufacturing
sector growth in Sub-Saharan African countries. Research in Business and Economics Journal,
Vol.12, pp 9-10.

Bezawork Hailu, (2007): Determinants of investment in manufacturing sector in ethiopia, Addis

Ababa University for the fulfillment of B.A degree in Economics Department.

Collier P. & Gunning J.W. 1999. ‘Explaining African economic performance,’ Journal of
Economic Literature, 37(1), 64-111.

Bakare, A. 2011. ‘The determinants of private domestic investment in Nigeria,’ Far East Journal
of Psychology and Business, 4(2), 27-37.

50
Bayai, I. & Nyangara, D. 2013. ‘An analysis of determinants of private investment in Zimbabwe
for the period 2009-2011,’ International Journal of Economics and Management Sciences, 2(6),
11-42.

Ambaye, G. 2014. ‘Modeling the determinants of domestic private investment in Ethiopia.’

Agris on-line papers in Economics and Informatics.

Arther lewis, 1954, Economic Development with unlimited supplier of labor, Manchister School.

EIC, (2020) Ethiopian annual report on investment capital and project operation, Addis Ababa.

GalbisV, 1979; money, investment and growth in Latin America, 1961-1973. Economic
Development and Cultural Change.

Getachew, M. 1997. ‘Determinants of private industrial investment in Ethiopia.’ Unpublished

MSc thesis, Addis Ababa, University of Addis Ababa.

Greene, J. & Villanueva, D. 1991. ‘Private investment in developing countries: An empirical

analysis.’ International Monetary Fund Staff Papers.

Gregory Mankiw, 2007; Macroeconomics, Harvard University, U.S.A 6th Edition.

Gujarati, D. N. (2009). Basic Econometrics. Southern Clifornia: The McGraw Hill Campanies.

Kamarudeen S, Soderbom M. 2013. Constraints and Opportunities in Rwanda's Industrial Sector.

International Growth Centre. London.

Mankiw, N. G. (2007). Macroeconomics. Harvard University, U.S.A 9th Edition.

Mbugua, T. 2000. ‘The micro and macroeconomic determinants of private investment in the
manufacturing sector in Kenya.’ Unpublished MSc thesis Addis Ababa, Addis Ababa University.

Michael P. Todaro and Stephen C. Smith (2012). Economic development; New York University
and Gorge Washington University, USA 11th edition. P. 594.

Mohammad J, (2006). An analysis of macroeconomic determinants of investment in serial Leon,

Serial Leone.

51
Mulu Gebreeyesus (2007). Firm turnover and Productivity Differentials in Ethiopian
manufacturing; Gothenburg University, Sweden.

Opaluwa David and others (2010). The effect of exchange rate fluctuation on the Nigerian
manufacturing sector. Benue State University, Makurdi, Nigeria.

Ramadan M, 2012 macroeconomic determinants of investment in manufacturing sector in

Ethiopia, Addis Ababa.

Sintayeu Nigussie, (2013) macroeconomic determinant of investment in manufacturing sector in

Ethiopia Addis Ababa.

Teshome Adugna (2014). The impact of manufacturing sector on economic growth in Ethiopia:
A Kaldorian Approach; Ethiopian Civil Service University, Ethiopia.

Vergara, R. 2004. Taxation and private investment: Evidence for Chile. Santiago: Universidad
Catolica de Chile.

Workie M. 1996. ‘Determinants and constraints of private investment in Ethiopia,’ Ethiopian

Journal of Economics.

52
Annexes
Annex-1 data

logTFR
Periods logMI INF logREER logCEX S DIR logIGDP
1990 0.66513 5 2.27132 3.1384 1.62118 4 4.08919
1991 0.507991 45 2.27433 3.07335 2.24969 4 4.03608
1992 0.493249 2.1 2.53719 2.96156 2.53995 4 3.99697
1993 0.596179 4.7 2.17319 3.23353 3.12795 10 4.10121
1994 0.630794 6.3 2.17696 3.41716 3.50425 10 4.12197
1995 0.688962 14.8 2.13988 3.4743 3.59058 10 4.15495
1996 0.717038 -9 2.08458 3.52401 3.75726 10 4.17843
1997 0.863388 -2.7 2.05652 3.62729 3.59863 7 4.19709
1998 0.72073 0.1 2.04999 3.54694 3.49214 6 4.21287
1999 0.758443 10.4 2.04883 3.61494 3.55495 6 4.2422
2000 0.748591 1.9 2.03782 3.53509 3.46564 6 4.24842
2001 0.761642 -10.8 1.99034 3.6913 3.46355 6 4.27031
2002 0.760912 -1.2 1.9722 3.79799 3.75919 3 4.30446
2003 0.761176 17.8 2.00389 3.80567 3.90245 3 4.32991
2004 0.728411 2.4 1.98408 3.82736 4.05336 3 4.37433
2005 0.685114 10.7 2.00303 4.05474 4.13688 3 4.41295

53
2006 0.666303 10.8 2.04999 4.14742 4.0031 3 4.45379
2007 0.660837 15.1 2.07078 4.1409 4.07836 3 4.48636
2008 0.613641 55.2 2.17754 4.38239 3.93995 4 4.5304
2009 0.588865 2.7 2.14829 4.4857 4.2359 4 4.57113
2010 0.599217 7.3 2.0835 4.59464 4.43599 4 4.62313
2011 0.566155 38 2.0892 4.7267 4.71224 5 4.69706
2012 0.534039 20.8 2.14426 4.86315 4.60316 5 4.77493
2013 0.56846 7.4 2.14675 4.95991 4.64483 5 4.86871
2014 0.601114 8.5 2.1486 5.03094 4.70436 5 4.93705
2015 0.643906 10.4 2.19756 5.06873 4.82489 5 5.01568
2016 0.755219 7.5 2.20222 5.1493 4.87015 5 5.5323
2017 0.791452 8.4 2.23528 5.18351 4.86849 5 5.60669
2018 0.76549 16.8 2.21484 5.15756 4.88996 7 5.26823
2019 0.747754 15.3 2.29798 5.24291 4.8943 7 5.29823
2020 0.724276 21.5 2.25382 5.31 4.89201 7 5.3241
Source: NBE and world bank (WB)

Where, IMS is investment in manufacturing sector; IGDP is Industrial Gross domestic product;
FRS is Foreign reserves; CEX is Capital expenditure; REER is Real effective exchange rate;
INFR is Inflation rate and DIR is Deposit interest rate. All the variables are in their natural
logarithmic form.

54
Annex-2 STATA out-put as a picture

1.STATA Result for Summarization

. sum

Variable Obs Mean Std. dev. Min Max

Periods 31 2005 9.092121 1990 2020

logMI 31 .6746606 .0911007 .493249 .863388
INF 31 11.07097 14.09537 -10.8 55.2
logREER 31 2.137573 .1189454 1.9722 2.53719
logCEX 31 4.153787 .7481294 2.96156 5.31

logTFRS 31 3.948882 .8145196 1.62118 4.8943

DIR 31 5.451613 2.188435 3 10
logIGDP 31 4.556746 .4631962 3.99697 5.60669

1. STATA Result for Unit Root Test

55
. dfuller logMI

Dickey–Fuller test for unit root Number of obs = 30

Variable: logMI Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -1.847 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.3572.

. dfuller d.logMI

Dickey–Fuller test for unit root Number of obs = 29

Variable: D.logMI Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -5.967 -3.723 -2.989 -2.625

MacKinnon approximate p-value for Z(t) = 0.0000.

. dfuller logIGDP

Dickey–Fuller test for unit root Number of obs = 30

Variable: logIGDP Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -0.034 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.9556.

56
. dfuller d.logIGDP

Dickey–Fuller test for unit root Number of obs = 29

Variable: D.logIGDP Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -4.812 -3.723 -2.989 -2.625

MacKinnon approximate p-value for Z(t) = 0.0001.

. dfuller INF

Dickey–Fuller test for unit root Number of obs = 30

Variable: INF Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -4.996 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.0000.

. dfuller d.INF

Dickey–Fuller test for unit root Number of obs = 29

Variable: D.INF Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -9.077 -3.723 -2.989 -2.625

MacKinnon approximate p-value for Z(t) = 0.0000.

. dfuller logREER

Dickey–Fuller test for unit root Number of obs = 30

Variable: logREER Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -2.271 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.1815.

57
. dfuller d.logREER

Dickey–Fuller test for unit root Number of obs = 29

Variable: D.logREER Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -7.811 -3.723 -2.989 -2.625

MacKinnon approximate p-value for Z(t) = 0.0000.

. dfuller logCEX

Dickey–Fuller test for unit root Number of obs = 30

Variable: logCEX Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) 0.088 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.9652.

. dfuller d.logCEX

Dickey–Fuller test for unit root Number of obs = 29

Variable: D.logCEX Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -5.574 -3.723 -2.989 -2.625

MacKinnon approximate p-value for Z(t) = 0.0000.

. dfuller logTFRS

Dickey–Fuller test for unit root Number of obs = 30

Variable: logTFRS Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -4.295 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.0005.

. dfuller d.logTFRS

Dickey–Fuller test for unit root Number of obs = 29

Variable: D.logTFRS Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -4.182 -3.723 -2.989 -2.625

MacKinnon approximate p-value for Z(t) = 0.0007.

58
3. STATA Result of Co-Integration Test
. predict ehat,resid

. dfuller ehat

Dickey–Fuller test for unit root Number of obs = 30

Variable: ehat Number of lags = 0

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -4.450 -3.716 -2.986 -2.624

MacKinnon approximate p-value for Z(t) = 0.0002.

4.Autoregressive distributed lag (ARDL) model Estimation Result

. ardl logMI logIGDP INF logREER logCEX logTFRS DIR,lags(1 1 0 1 1 1 1) ec

ARDL(1,1,0,1,1,1,1) regression

Sample: 1991 thru 2020 Number of obs = 30

R-squared = 0.7697
Adj R-squared = 0.6071
Log likelihood = 63.319056 Root MSE = 0.0389

D.logMI Coefficient Std. err. t P>|t| [95% conf. interval]

ADJ
logMI
L1. -.6768288 .1799902 -3.76 0.002 -1.056575 -.2970827

LR
logIGDP .4683971 .092665 5.05 0.000 .2728911 .6639031
INF -.0004412 .0011212 -0.39 0.699 -.0028068 .0019243
logREER -.083222 .2947749 -0.28 0.781 -.7051427 .5386987
logCEX -.4104391 .1194841 -3.44 0.003 -.6625285 -.1583496
logTFRS .1490848 .1018557 1.46 0.162 -.065812 .3639816
DIR .0007894 .008998 0.09 0.931 -.0181947 .0197736

SR
logIGDP
D1. -.1160701 .0978111 -1.19 0.252 -.3224336 .0902934

logREER
D1. -.2766254 .2092451 -1.32 0.204 -.718094 .1648433

logCEX
D1. .1833688 .108517 1.69 0.109 -.0455821 .4123196

logTFRS
D1. -.1429405 .0623224 -2.29 0.035 -.2744293 -.0114516

DIR
D1. -.0202311 .009002 -2.25 0.038 -.0392236 -.0012386

_cons -.1055082 .3740819 -0.28 0.781 -.8947519 .6837356

59
5. STATA Result of Post Estimation (Diagnostic) Test
. vif

Variable VIF 1/VIF

logCEX 24.89 0.040171

logIGDP 12.50 0.079975
logTFRS 11.40 0.087709
logREER 2.05 0.486838
DIR 1.31 0.762002
INF 1.27 0.785949

Mean VIF 8.91

. estat archlm
LM test for autoregressive conditional heteroskedasticity (ARCH)

lags(p) chi2 df Prob > chi2

1 0.905 1 0.3414

H0: no ARCH effects vs. H1: ARCH(p) disturbance

. estat bgodfrey

Breusch–Godfrey LM test for autocorrelation

lags(p) chi2 df Prob > chi2

1 0.866 1 0.3521

H0: no serial correlation

. swilk ehat

Shapiro–Wilk W test for normal data

Variable Obs W V z Prob>z

ehat 31 0.96818 1.036 0.074 0.47040

. linktest

Source SS df MS Number of obs = 31

F(2, 28) = 27.89
Model .165766346 2 .082883173 Prob > F = 0.0000
Residual .083213972 28 .002971928 R-squared = 0.6658
Adj R-squared = 0.6419
Total .248980317 30 .008299344 Root MSE = .05452

logMI Coefficient Std. err. t P>|t| [95% conf. interval]

_hat 1.611438 1.569712 1.03 0.313 -1.603971 4.826847

_hatsq -.4626362 1.183361 -0.39 0.699 -2.886641 1.961368
_cons -.1994693 .5182648 -0.38 0.703 -1.261087 .8621481

60
THANK YOU!!

61