Declaration

This is to certify that student Wasyehun getie has conducted senior essay research paper, titled
on the determinant of economic growth in Ethiopia. This work is completed with satisfactory
evaluation of the advisor and examiners as paper the requirements of the university.

Advisor________________________ Date________ Signature________________

Examiner______________________ Date_______ Signature___________________

I
Acknowledgement
It is not exaggeration to say that without insisting the Almighty God, Jesus Christ, would not
have been in a place to complete successfully this thesis. Thus, glory to him.

My gratitude and appreciation goes to my advisor Mubarak juhar (MSC) for his constructive
comments, technical support, welcoming approach in every step of my work and helped me in
shaped this study.

I would like to extend my gratitude to my families. Without whose moral support, my

achievement was not possible. And also I want to say thank you all to my relatives and
colleagues who have helped me while I was writing this thesis.

II
ABSTRACT
The main objective of this study is to investigate the determinants of economic growth in
Ethiopia during the period 1976-206. The Autoregressive Distributed Lag (ARDL) Approach to
Co-integration and Error Correction Model are applied in order to investigate the long-run and
short run relationship between the dependent variable (real GDP) and its determinants. The
finding of the Bounds test shows that there is a stable long run relationship between real GDP,
capital stock, labor, export, aid, real effective exchange rate, Investment and inflation.

The empirical results reveal that capital stock, export, aid, and Investment have positive impact
on economic growth while real effective exchange rate and inflation are negatively and statically
insignificant. However, the study found out that export of goods and service and capital stock
have statistically significant impact on economic growth in the long run. This study has also an
important policy implication. The findings of this study imply that economic growth can be
improved significantly when capital stock and export increases. Hence policy makers and /or the
government should strive to increase capital stock and export which is believed as a back bone
of growth and has allocate adequate finance for export, which will help to work on quality and
performance of export of good and service.

III
 Tables of Contents
Contents…………………………………………………………………………….. ……..Pages

Declaration

Acknowledgement

ABSTRACT

Tables of Contents

Lists of Tables

Lists of Figures

Acronyms

CHAPTER ONE

INTRODUCTION

1.1Background of the study

1.2 Statement of Problem

1.3 Research question

1.4 Objective of the study

1.4.1 Specific objective of the study

1.5 Significant of the study

1.6 Scope of the study

1.7. Limitation of the study

1.8. Organization of the study

CHAPTER TWO

2. LITERATURE REVIEW

2.1 Definition of economic growth

2.2. Theoretical Literature

2.2.1. Theories of economic growth

IV
 2.2.1.1. The Keynesian theory of growth

2.2.1.2. The neo-classical theory of Growth

2.2.1.3. The Endogenous theory of Growth

2.3 Empirical Literature

CHAPTER THREE

3.METHODOLOGY AND DATA SOURCE

3.1. Types and sources of data

3.2. Method of data analysis

3.3 model specification

3.4 Expected sign and Description of variables

3.5. Long run co-integration test

3.6.Unit root test ………………………………………………………………………………………………………………………………..19

3.7. Error correction model (ECM)

CHAPTER FOUR

4. DATA ANALYSISE AND DISCRIBITION

4.1 descriptive analyses

4.1.1 Economic performance the country

4, 1.1.Export and Ethiopian Economy

4.1.2Aid and Ethiopian economy

4.1.3 Trends of Inflation in Ethiopia

4.2 Economic analyses
4.2.1. The Unit Root Test Analysis
3.2.2 Lag Length Determination
4.2.3 Bodes test
4.2.4. Long Run ARDL Model Estimation
4.2.5. Short Run Error Correction estimation

V
 4.2.6 Diagnostic Test

4.2.6.1 Normality test

4.2.6.2Auto-coloration tests

4.2.6.3Heteroskedasticity tests

CHAPTER FIVE

5. CONCLUSION AND POLICY RECOMMENDATION

5.1 CONCLUSION

5.2 POLICY RECOMMENDATION

Reference

Annex

VI
Lists of Figures
Table 4.1 Augmented Ducky Fuller test

Table 4.2 lag length criteria determination

Table 4.3 bounds regression results

Table 4.4 Estimated Long Run Coefficients using the ARDL

Table 4.5 Error Correction Representation for the Selected ARDL

Table 4.6 Auto-coloration tests

Table 4.6Heteroskedasticity test

VII
Lists of Tables
Figure 4.1Trend of GDP growth

Figure 4.2 Trend of export

Figure 4.3 Trend aid

Figure 4.4 trend of inflation

Figure 4.5 test of normality

VIII
Acronyms
ADF Augmented Dickey Fuller

AFDB African Development Bank

AIC Akaike Information Criterion

ARDL Autoregressive Distributed Lag

CSA Central Statistical Agency

ECM Error Correction Method

ECT Error Correction Term

EEA Ethiopian Economic Association

GDP Gross Domestic Product

GTP Growth and Transformation Plan

IMF International Monetary Fund

KPSS Kwiatkowski Phillips Schmidt Shin

MoFED Ministry of Finance and Economic Development

NBE National Bank of Ethiopia

OLS Ordinary Least Square

PP Phillips Perron

R&D Research and Development

RGDP Real Gross Domestic Product

SBC Schwarz Bayesian Criterion

SDPRP Sustainable Development and Poverty Reduction Program

VAR Vector Autoregressive

WBI World Bank Index

WD Durban Watson

IX
 CHAPTER ONE

INTRODUCTION

1.1Background of the study

Economic growth is one of the most important issues in international political economy and one
of the major goals of all countries. It is the increase of gross domestic product. Therefore, it
shows only the quantity of goods and services produced in a country in given period of time.
According to Nafziger (2005), economic development refers to economic growth gone with an
improvement in the material well-being of the poor; a decline in agriculture’s share of national
output; increase in the output share of industry and services; an increase in the education and
skills of the laborforce and technical advances originating within the country. The economic
achievements lead to the improvement of the standard of life, adequate conditions of medical
care, improvement of the educational system and a better redistribution of incomes (Haller,
2012). The world economy grows 2.6 percent in 2014 to reach $77 trillion in current prices, and
growth is projected to accelerate to 3 percent in 2015. Developing economies grew an estimated
4.4 percent in 2014 and projected to grow 4.8 percent in 2015. Growth in high-income
economies has updated from earlier forecasts to 1.8 percent in 2014 and 2.2 percent in 2015
(World Bank, 2015).
The economic growth of Ethiopia has shown various changes in different regime. These changes
in government structure create a problem of inconsistency in implementing the policy of the
previous regime as well as natural disasters like famine, drought, political instability and war
(Alemayehu and Befkadu, 2005) had depressing effect in the history of Ethiopian economic
growth. In the modern Ethiopian political economic growth theory there are three regimes that
followed unique macroeconomic policy with its impact on macroeconomic performance of the
country.
The economy of Ethiopia has experienced change from market -oriented to centrally planned and
back again to market -oriented economy. Under the monarchy (up to 1874), the economy was
primarily agricultural. The economy was based on feudal system under which land ownership
was highly inequitable with major portion of farmland being in the hand of wealthy property

1
owner of very few industries, most were owned by foreigner. During the military regime (1974-
1991), the economy shifted to command economy where socialist principles and ideology
rule .EPRDF (1991) followed a market oriented economy. It has supported process of economic
reform based on privatization of state enterprise ,promotion of agricultural export and move
towards to free market economy and GDP growth from 2% to 11.6% between 1974/75-2005/06
(MoFED&CSA,2005/06) placing among the top performing economy in sub-Sahara Africa
(NBE 2013/16). In 2015/16 GDP grew by 8% it slow down from 10.9%regested in
2015/15(AEO, 2017).the service and industrial sector lead growth during these period. Growth in
the agricultural sector was negatively affected by the El Niño induced drought (Edward,
BattSennog, Admit, Wondifraw Hailed Kebret, 2016). for 2016/17 and 2017/18 investment in
energy and transport infrastructure ongoing reform spur industrialization ,such as the
development of industrial park and continued progression in service are expected to lead
growth (AEO, 2017).

Ethiopia is steadily recovering from the 2015/16 and 2017 droughts, with continued expansion of
services and industry and a rebound in agriculture. At 39.3%, services accounted for the largest
share of GDP in 2016/17, 39.3%, driven by trade, transport, and communications, although this
share decreased from 47.3% in 2015/16. Industry’s share of GDP increased from 16.7% in
2015/16 to 25.6% in 2016/17, driven by construction, electricity, and manufacturing.
Implementation of the export-led industrialization strategy supported growth in industry.
Although agriculture’s share of GDP stagnated at 36%, the sector’s growth rate increased from
2.3% in 2015/16 to 6.7% in 2016/17 due to rising commodity prices, notably for coffee. Growth
continues to be led by investment in line with stable public infrastructure spending and higher
foreign direct investment (FDI). Real GDP growth during 2017/18–2018/19 will be led by
greater agricultural productivity and strong industrial growth (NBE, 2016).

1.2 Statement of Problem

Economic growth is the total amount of good and service that produce in country which is
determined by different factors. In most macroeconomic models, aggregate demand and
aggregate supply interact to determine the short-run performance of the economy, but when it
comes to the long-run analysis of economic growth, aggregate demand usually makes its exit and

2
aggregate supply rules the roost. Endogenous growth theory incarnation, share this neglect of
aggregate demand. These theories imply that the rate of growth of per-capita income in long-run
equilibrium depends on supply-side factors. They do not introduce aggregate demand into the
analysis at all; assuming that the economy is always at full employment and that all saving is
invested. Thus, for mainstream macroeconomists, aggregate demand is relevant only for the
short run and in the study of cycles, but irrelevant for the study of growth. The apparent reason
for this is that the market mechanism, in the form of flexible wages working through assets
markets, or government policies, solves the problems of unemployment and the deviation of
aggregate demand from aggregate supply in the longer run. The neglect of aggregate demand
from current mainstream growth theory is ironic, because in Harold’s (1939) growth model argue
by the key pioneering contribution to modern growth theory—aggregate demand plays a central
role. Harrod distinguished clearly between saving and investment behavior, and his warranted
rate of growth, at which saving and investment behavior were mutually consistent, could be
different from the natural rate of growth, or the rate of growth of what we can call aggregate
supply. Other growth theories in which aggregate demand played a major role, such as those of
Robinson (1962) and Kahn (1959) were also earlier considered a part of growth theory.
Growth theories in which aggregate demand plays a role have not disappeared entirely, however.
Several heterodox economists, who can be called post-Keynesian or structuralism, give center
stage to aggregate demand. These economists have developed models of aggregate-demand
determined growth imply that the rate of growth of the economy in the long run can be increased
by increasing aggregate demand, for instance, government spending. However, these models,
while reinstating aggregate demand, appear to jettison aggregate supply, somewhat implausibly
implying that the aggregate supply factors, so dear to mainstream growth theorists, are irrelevant
for long-run growth. In short, aggregate demand has disappeared from mainstream growth
theory, which focuses entirely on the supply side. Growth theories that focus on aggregate
demand, however, ignore aggregate supply considerations.
Previously there are money studies that are conducted on the determinant of economic growth in
Ethiopia in different time period .such as Demesse and Balecha (2005) and EstubdinkSeleshi
(2014).those research tries to identify the determinant of economic growth in Ethiopia. I.e.
inflation, foreign aid, rainfall, population, investment, internal debit and export are included in
their study. There is other macroeconomic variable that may determine economic growth in

3
Ethiopia which is not addressed in their studies. This study tries to fill the gap by considering
productivity of labor and capital and their contribution to economic growth in Ethiopia and time
gap by taking 40 years data from 1974-2016.

1.3 Research question

The study is addressing the following research question:-

- Is there a long run relationship between capital and labor productivity with real GDP in
Ethiopia?

-how to test the causality between labor, capital and exchange rate with economic growth?

1.4 Objective of the study

The main objective of the study is to address empirical evidence on the determinant of economic
growth in Ethiopia.

1.4.1 Specific objective of the study

- To indicate the long run relationship of labor and capital productivity with Ethiopian economic
growth.

-To test the causality between productivity of labor, capital and exchange rate with economic
growth in Ethiopia.

1.5 Significant of the study

 It use by other researchers who want to have further investigations in to the determinant of
economic growth in Ethiopia. By invoking the other researcher who wants to see the other
factors that I missed.

 It used for as an input for policy maker or other concerned bodies to design effective policy
to improve the economic growth of Ethiopia.

1.6 Scope of the study

The study is time series analysis to identify major factors which affect GDP of Ethiopia. There
are variable which determine the economic growth of Ethiopia .these study focus only in

4
identifying the effect of macroeconomic variable such as capital, Labor and Exchange rate on the
GDP growth of Ethiopia from 1974-2016

1.7. Limitation of the study

This study is limited to the period between 1976-2016, which covered forty years of time-series
data on macroeconomic variable that can affect economic growth. The limitation of this study
was the one associated with data availability. There are shortages of data, particularly foreign aid
data. The most challenge while doing this study came from inconsistency of data from different
organizations. So as to avoid such inconsistency attempt is made to stick to the same source of
data. The main aim of this study is to analyze the macroeconomic determinants of economic
growth. However, there are also non-economic factors that affecting growth like political
stability, rules of economic regulation (monitoring and fiscal policy), and rules of law (property
right) are not addressed here and might be consider other limitations of this study.

1.8. Organization of the study

The study is organized in to five chapters. on the first chapter introduction about the general
aspect of the study such as background of the study ,statement of the problem ,research
question ,objectives of the study ,significance of the study and the scope of the study are
presented .the second chapter contains review literature and empirical literature of the study ,the
third chapter contains the research design (methodology),model specification ,type source of data
analysis .chapter four contain data analysis and discussion of the result another last chapter
contains conclusion and possible policy recommendation about the determinant of economic
growth in Ethiopia.

5
 CHAPTER TWO

2. LITERATURE REVIEW

2.1 Definition of economic growth

Economic growth is the increase in price of goods and services produced by an economy,
typically a nation, over a long period of time. It is measured as percentage increase in real gross
domestic product (GDP) which is gross domestic product (GDP) adjusted for inflation. GDP is
the market value of all final goods and services produced in an economy or nation. Generally
speaking, gross domestic product is an economic model that reflects the value of a country's
output. In other words, a country's GDP is the total monetary value of the goods and services
produced by that country over a specific period of time.

2.2. Theoretical Literature

2.2.1. Theories of economic growth
The process of economic growth and the sources of differences in economic performance across
nations are some of the most interesting, important and challenging areas in modern social
science. The analysis of the process of economic growth was a central feature of the work of the
classical economists, as represented chiefly by Adam Smith, Thomas Malthus, David Ricardo,
and Karl Marx were all concerned with the growth of the economy (I.e., the increase in the
production of goods and services over time).

The interest of these economists in problems of economic growth was rooted in the concrete
conditions of their time. Specifically, they were confronted with the fact of economic growth and
social changes taking place in contemporary English society as well as in previous historical
periods. According to A. Smith (1776), the importance of ‘invisible hand’ (the force of
supply/demand in a competitive market), specialization/division of labor, accumulation of
physical capital (investment) and technological progress were the most determinants of
economic growth in the long term and hence the prosperity of nations. A wide range of studies
have investigated the factors underlying economic growth. Using different conceptual and
methodological viewpoints, these studies have placed emphasis on a different set of explanatory
parameters and offered various insights to the sources of economic growth.
6
The broad consensus highlighted in these studies is that a country’s growth over a long period is
basically determined by three factors, namely: (1) the efficient utilization of the existing stock of
resources, (2) the accumulation of productive resources such as human capital, and (3)
technological progress (Dewan and Hussein, 2001, Ndambiri et al., 2012). Moreover, research
and development, economic policy and macroeconomic condition, openness to trade and
institutional framework are among the most important determinants of economic growth. These
broad categories can be further broken down into various determinants of economic growth. The
influences considered here include human capital, physical capital, exports, Aid, government
policies, inflation, external debt, government expenditure, financial systems and technological
progress.

A variety of studies have addressed the issue of economic growth, mostly using either cross-
country or panel data approach (Barro, 1997, 2003). While most of these studies utilize the
standard neo-classical growth Model, More recent studies focus on endogenous growth models.
There have been two periods of powerful work on growth theory, the first was in the 1950s and
1960s, and the second (30 years later) in 1980s and 1990s. In the first period, the neoclassical
theory of growth was best known contribution by Robert Solow (1956).

2.2.1.1. The Keynesian theory of growth

Unlike Smith’s conception of saving as an important prerequisite of growth, Keynesians
approach gives emphasis to the demand aspect. According to the Keynesian theory of growth
demand from consumer and state were the prerequisite for economic growth. This assumption
means that changes in income, especially disposable income, is the prime influence on
consumption expenditures. If the household sector has more income because of the economy is
expanding, then they increase consumption expenditures. If the household sector has less income
because of the economy is contracting and a large group of workers is unemployed, then they
decrease consumption expenditures.

Following the publication of Keynes’s General Theory in 1936, some economists sought to
dynamism Keynes’s static short-run theory in order to investigate the long-run dynamics of
capitalist market economies. Roy Harrod (1939, 1948) and Evsey Domar (1946, 1947) were
developing the growth model independently that relate an economy’s rate of growth to its capital
stock. However, the assumptions and results are, basically the same. While Keynes emphasized

7
the impact of investment on aggregate demand, Harrod and Domar emphasized how investment
spending also increased an economy’s productive capacity (a supply-side effect). The Harrod-
Domar (H-D) model considers a closed economy in which one homogenous good Y is, where Y
is gross output. This good may be either used as an investment good, I, or as a produced
consumption good, C. The model suggests that the economic rate of growth depends on the level
of savings, and the productivity of investment (i.e. in order to grow, economies must save and
invest a certain portion of their GDP). The labor force is assumed to grow at a constant
exogenous rate n and thus, = n. Thus, an aggregate production function with fixed
technological coefficient was given as:

Yt=min¿] ----------------------------------------------- [1]

Kt = physical capital stock at a time t v = utilized capital-output ratio (constant, i.e. k)

Et = effective labor force at a time t

u = employed effective labor-output ratio (constant, i.e. L)

Assuming a two-sector economy (households and firms), we can write the simple national
income equation as:

Yt= Ct+St ------------------------------------------------------- [2],

That is It=St---------------------------------------------------- [3]

Subtitling [2.3] into [2.2] yields [2.4]

Yt= Ct+It--------------------------------------------------------- [4]

The evolution of the capital stock over time is given by:

Kt+1= (1-δ) Kt+It ------------------------------------------- [5]

Where δ is the rate of depreciation of capital stock. By assuming that total saving (S t) is some
Proportion of GDP (Yt ),

8
 K
St=SYt ------------------------------------------------------------ [6] we know that v= , from this
Y
K=vY and It = St =sYt, it follows that we can rewrite equation [2.5] as:

vYt+1 = (1-δ) vYt+sYt --------------------------------------- [7]

Dividing both sides by v, and subtracting Yt from both sides of equation [2.7] yields equation

(2.8):

Yt+1¿ ¿- δ) Yt----------------------------------------------------------8

Dividing both sides [2.8] by Yt yields that:

∆y

∆Y
Y
s
()
=+ −δ , →gy =¿) --------------------------------- [9]
v

This simply states that the growth rate (gy) of GDP is jointly determined by the savings ratio (s)
divided by the capital–output ratio (v). The higher the savings ratio and the lower the capital–
output ratio and depreciation rate, the faster will an economy grow (Brian Slowdown and
Howard R. Vane, 2005).

2.2.1.2. The neo-classical theory of Growth

School of thoughts and theories on economic growth can be traced back to the classical
economists of the eighteenth and nineteenth century, whose works are briefly reviewed alongside
the transition to neo classical growth theory. The basic framework of neoclassical growth models
was first developed by Robert Solow (1956) and Trevor Swan (1956). This neo-classical model
states that, at any point in time, the total output of the economy depends on the quality and
quantity of physical capital employed, the quantity of labor employed and the average level of
skills of the labor force. However, once the economy reaches the full equilibrium level,
additional growth in the stock of capital per worker will only take place if productivity increases,
either through enhanced capital stock or through improvements in the quality of the labor force.
The basic assumptions of the Solow model include constant returns to scale, diminishing
marginal productivity of capital, exogenously determined technical progress and substitutability

9
between capital and labor. And his basic question was “what are the main determinants of
economic growth in the long term”.

This model (Solow) was started by criticizing the Roy Harrod and EvseyDomar (Harrod, 1939
and Domar 1946) models for its weakness. The production takes place under conditions of fixed
proportions and there is no possibility of substituting labor for capital in production. “Theory of
Economic Growth (Solow, 1956)”

Based on his growth model, high investment rate (saving rate), high level of technology, skilled
human capital, low level of population growth rate and low rate of capital depreciation are the
most determinants of economic growth in long run. According to this simple mathematical
model, economic growth can be measured as follows.

∂ Y ∆ Kt ∂ Y ∆ At
∆Yt= +∂ Y ∆< ¿ + ¿ ---------------------------------------------- [10]
∂K ∂L ∂A

When we divide both sides of [2.10] by Yt, it becomes that:-

ΔYt ∂ Y ∆ Kt ∂Y ∆ At
= + ∂Y ∆ < ¿ + ¿------------------------------------------------- [11]
Yt ∂ KYt ∂ LYt ∂ AYt

The above equation decomposes GDP growth into portions that can be attributed to growth in the
capital stock, the labor force, and the technology level. Then using same methodology for labor
and technology, reduced form of Equation [2.11] in growth form is as follows.

ΔK
=βk
ΔKt

=βk𝗀k

Using the same methodology for labor and technology, reduced from Equation [11] in growth
form is as follows.

𝗀y=βk𝗀K+βL𝗀L+βA𝗀A---------------------------------------------- [12]

Or βA𝗀A=𝗀Y–(βk𝗀k+βL𝗀L) ------------------------------------------ [13]

10
Since the Solow’s growth model assumption was constant return to scale and perfect competitive
market, the summation of the share of capital and labor is a unity. So if share of capital is βk,
then the share of labor is 1-βk =βL and the above equation can be rewrite as

βA𝗀A =𝗀Y – (βk𝗀k+ (1–βk) β

Where gY= Growth rate of Real GDP

𝗀K=Growth rate of physical capital

𝗀L= Growth rate of human capital

𝗀A= Growth rate of technology and βK, βL, βA are the marginal elasticity of capital, labor force
and technology respectively. So if we have observations on the growth rate of output, the labor
force, and the capital stock, we can have an estimate on the growth rate of total factor
productivity. Equation [2.13] defines as the “Solow residual” in its long run growth model.

According the neo classical theory of growth, the model makes three important forecasts. First,
increasing capital relative to labor creates economic growth, since people can be more productive
given more capital. Second, poor countries with less capital per person grow faster because each
investment in capital produces a higher return than rich countries with sufficient capital. Third,
because of diminishing returns to capital, economies eventually reach a point where any increase
in capital no longer creates economic growth and which is called a steady state.

2.2.1.3. The Endogenous theory of Growth

In the mid-1980s it the standard neoclassical growth model was theoretically unsatisfactory as a
tool to explore the determinants of long-run growth .The main limitation of the Solow model is
that technological progress is assumed exogenous and common across countries (Elhanan
Helpman, 2004). The model without technological change predicts that the economy will
eventually converge to a steady state with zero per capita growth. The basic improvement of the
new growth theory or endogenous growth theory over that model is that it explicitly tries to
model technology rather assuming it to be exogenous. In essence, it looks for determinants of
technology frontier upward continuously. In this theory, the central motive of profit
maximization of business firms are considered to determine technological progress as these
firms involve in research and development (R&D) seeking new and better idea.

11
During the mid-1980s several economists, most notably Romer (1986) and Robert Lucas (1988),
sought to construct alternative models of growth where the long-run growth of income per capita
depends on investment decisions rather than unexplained technological progress. Romer (1986)
and Lucas (1988)) was developed an economic growth model that includes mathematical
explanation of technological advancement, which incorporates a new concept of human capital
(skill and knowledge) that make workers productive. Unlike physical capital, human capital has
an increasing rate of return. Therefore, overall there are constant rate of return to capital and
economies never reach a steady state. According to the Endogenous theory of growth, economic
growth does not slow as capital accumulation, but the rate of growth depends on the type of
capital a country invest in. As research indicates that increasing human capital (education) and
technological change (innovation) fast economic growth in long run (Elhanan Helpman,
Hulyauilka, (2004).

The first version of endogenous growth theory was AK theory, which did not make an explicit
distinction between capital accumulation and technological progress. In effect it lumped together
the physical and human capital whose accumulation is studied by neoclassical theory with the
intellectual capital that is accumulated when innovations occur. An early version of AK theory
was produced by Frankel (1962), who argued that the aggregate production function can exhibit
a constant or even increasing marginal product of capital. In the special case where the marginal
product of capital is exactly constant, aggregate output Y is proportional to the aggregate stock
of capital K.

Thus: Y=AK ------------------------------------------------- [14], where A is a positive constant.

According to AK theory, an economy’s long-run growth rate depends on its saving rate. For
example, if a fixed fraction s of output is saved and there is a fixed rate of depreciation, the rate
of aggregate net investment is:

K=SY-δK--------------------------------------------------- [15]

Dividing both side of equation [2.15] by K and simplifying it yields:

12
 =s -δ → gk=s--------------------------------------------(16)where 𝗀k is capital growth rate and A=
K y
k k
Y
K

, Hence an increase in the saving rate s will lead to a permanently higher growth rate.

Romer (1986) produced a similar analysis with a more general production structure, under the
assumption that saving is generated by inter-temporal utility maximization instead of the fixed
saving rate of Frankel. Lucas (1988) also produced a similar analysis focusing on human capital
rather than physical capital.

The Romer model tries to explain why and how advanced countries of the world exhibit
sustained growth. According to his model, technological progress is driven by R&D sector in
advanced world and indigenizes the technological progress by introducing an R&D sector, i.e.
search of new ideas by researcher interested in profiting from their invention. The aggregate
production function in the Romer model is:-

Y=k α ¿ALy) 1- α ------------------------------------------- [17], where A denotes for technology and

Ly is part of labor force, which are numbers of people engaged in production of output. One of
the key equations in endogenous growth model (Romer) is the one describing the research and
development (R&D) sector. Accordingly, technology (A) is the number of ideas, or stock of
knowledge accumulated up until time t. The number of new idea is equal to the number of
people devoting their time in discovery new idea (La), multiplying by the rate at which they
discover new ideas (φ ). Thus:

dL a(t)
=φLa (t) ----------------------------------------------------- [18]
dt

Labor is used either to produce goods, Ly, or to produce new ideas La. So the economy faces the
following resource constraint:

L = Ly + La

The rate, at which new ideas are discovered, might be constant, or an increasing function of A.
Therefore the rate at which new ideas discovered may be rewritten as:
13
φ=θAy----------------------------------------------------------- [19], whereθ andγ are constants.

Notice that with γ > 0 the productivity of research increases with the stock of ideas that have
already been discovered. On the contrary with γ < 0, discovering new ideas becomes harder over
time. With γ = 0 the discovery rate is independent from the stock of knowledge.

It is possible that new ideas are more likely when there are more persons engaged in research.
Thus, the effect of La is not proportional. Hence, it can be assumed that it is Lμa that enter in the
production function of new ideas, with 0 < μ< 1. Therefore general production function of new
ideas is:

A=θ Lμa Ay----------------------------------------------------------- [20]

Dividing both side of [2.20] by A yields [2.21]

μ
Ȧ La
=θ ------------------------------------------------------------------------------- [21]
A A 1− y

 Ĺ
Along the balanced growth path (BGP), =gA, =n. Thus, the numerator and the denominator
A L
should grow at the same rate, which means that: - Rossi (2012), Simplifying this equation, finally
it yields that:

𝗀A=
nμ
----------------------------------------------------------- [22], both µ and Ѳ are parameters,
1−θ

Such as subsidies to research and development. The implication is that the growth rate of
Researchers and parameters of production function for ideas determine the long-run growth.
Intuitively, it is highly likely to get people with brilliant mind in large population than small one
(Jones, 1998, as cited in Seid, 2000)

2.3 Empirical Literature

There are many empirical studies conducted on determinants of economic growth in Ethiopia
(Tadesse; (2011) WoubetKefle; (2006) Soressa; (2013) Hailegiorgis (2012); Teklu; (2014),
Hailemariam (2011) Asemamawe (2012). Those studies determine the major source of economic
growth. According to empirical studies of the above researchers’, Country’s economic growth is

14
influenced by several macroeconomic variables like inflation, export level, Aid, investment,
money supply and government expenditure. Similarly, a study done by Soressa (2013) Appling
the Autoregressive distributed lag model (ARDL) and Grange causality for the period 1960 to
2011 found that, a one percent increase in export will lead to increase the economic growth by
0.57 percent in long run. Moreover, the research done by Hailegiorgis (2012) during 1974 to
2009 by using granger causality found that, there is an evidence of unidirectional causality
between export and economic growth in Ethiopia (I.e. export growth causes economic growth).In
addition a study done in Ethiopia for the period 1970-2009, indicates that there is a positive and
statistically significant relationship between foreign aid and economic growth in the long run but
insignificant in short run (Tasew, 2011). According to his result the long run elasticity of growth
with respect to aid is 0.45. This implies that every unit increase in the foreign aid was likely to
grow GDP by 0.45 percent in long run. But the research done by Asmamaw (2012), there is
negative relation.

This section presents a simple growth model that attempts to capture some of the major
macroeconomic factors affecting economic growth in Ethiopia. Macroeconomic theory has
identified various factors that influence the growth of a country from the classical, neo classical
and the new growth theories. These factors include natural resources, investment, human capital,
innovation, technology, economic policies, foreign aid, trade openness, institutional framework,
foreign direct investment, political factors, socio-cultural factors, geography, demography and
many others. Understanding characteristics and determinants of economic growth requires an
empirical framework that can be applied to a relatively long time frame. In order to examine the
empirical evidence of the macroeconomic determinants of economic growth in Ethiopia, the
study considers most of these factors. As we discussed in the theoretical literature review, origin
of the econometric model is extended neoclassical growth model

Thus: y = f (k, L) -------------------------------------------------------- [23], where k and L represents

capital and labor, respectively

Following broadly the approach of extended neoclassical growth model,the specification of the
economic growth function for Ethiopia as follows:- Real GDP is a function of capital stock, real
effective exchange rate, labor, exports of goods and service, foreign aid, investmentand inflation.

15
 CHAPTER THREE

3. METHODOLOGY AND DATA SOURCE

3.1. Types and sources of data

The study use time series secondary data. The data has been collected from different institution
like National Bank of Ethiopia (NBE), aid from World Bank (WB), International Monetary Fund
(IMF) and Minister of Finance and Economic development (MOFED).

3.2. Method of data analysis

In this study both simple descriptive and econometrical methods of data analysis are employed.
To analyze the macroeconomic performance of Ethiopia during the study period, we used tools
of descriptive statistical such as graphs. On the other hand standard econometrical technique
would apply to analyze the major determinants of economic growth under the study period.
Finally, Eview 9.0 versions have been used as statistical software package for the entire analyze
running this study.

3.3 model specification

Economic growth in Ethiopia can be affected by different factors. Different models have been
suggested by various researchers to investigate the determinant variable that affect economic
growth in Ethiopia. This study will have adopted model by econometric techniques. The factors
are real effective exchange rate(REER), capital (k), export(X), foreign aid (AID), investment
(I), labor(y), and inflation (INF). This paper tested by using ARDL model for short and long run
relation with its determinant variable.

Therefore the mathematically relationship between real GDP and its major macroeconomic
determinant are expressed as follows:

16
GDP=f (ER, K, X, FA, I, L, IN)

LnYt=β0+β1LnREERt+β2LnKt+β3LnXt+β4LnFAt+β5LnIt+β6LnLt+β7LnINFt+ε t …… (24)

Yt represents real GDP at a time t; Kt represent capital ; FAt represent foreign aid , REERt
represent real effective exchange rate , Lt represent labor ,It represent total investment, and
INFtis for general inflation;.(ε) The error term ( assumed to be normally and independently
distributed with zero mean and constant variance, which captures all other explanatory variables
which influences real gross domestic product in a country which are not captured in the model.
βiare the Where partial elasticity’s of real GDP with respect to macroeconomic variables listed
above.

p p p p p
ECTt=ΔLnGDPt-1-[αo+∑ β oΔLnGDPt-1+∑ β 1ΔREERt-i+∑ β 2-iΔKt-i+∑ β 3ΔXt-i+∑ β 4ΔFAt-i+
i=1 i=o i=o i=o i=o

p p p

∑ β 5ΔIt-i+∑ β 6ΔLt-t+ ∑ β 7ΔINFt-i]

i=o i=o i=1

Here Δ is the first difference operator; β's are the coefficients relating to the short-run dynamics
of the model's convergence to equilibrium, and measures the speed of adjustment

3.4 Expected sign and Description of variables

Real GDP(Y) is the market value of the goods and services produced by an economy over time.
It is conventionally measured as the percent rate of increase in real Gross Domestic Product.
Since most economists argue that economic growth can be measured as growth in real GDP, it
includes in the model as main dependent variable in order to measure economic growth. In order
to avoid the inconsistency associated with different base year price while computing real GDP,
this study was used the real GDP (constant value), which is deflated by Ministry of Finance and
Economic Development (MoFED) based on the constant price of 2010/11.

REER is the average of a country's currency relative to an index or basket of other major
currencies, adjusted for the effects of inflation. The weights are determined by comparing the
relative trade balance of a country's currency against each country within the index and its
expected sign is expect to be negative.

17
K; Capital refers to financial asset or the financial value of assets, such as funds held in deposit
accounts, as well as the tangible machinery and production equipment used in environments such
as factories and other manufacturing facilities. Additionally, capital includes facilities, such as
the buildings used for the production and storage of the manufactured goods. Materials used and
consumed as part of the manufacturing process do not qualify that its expected value is positive.

Exports of goods and service (EXT) are defined as the total exports of goods and service to the
rest of the world. It is believed that export of a country’s is one of the macroeconomic
determinants of economic growth. For this reason and due to researcher’s interest this variable is
entered as explanatory in order to analyze it effect on Ethiopian economic growth. The expected
sign of this variable is expected to be positive

Foreign Aid (AID) is defined as aid inflows from external assistances. As we know Ethiopia is
one of the poor countries in the world. As result Ethiopia is getting from external assistance in
the form of aid. To see its effect on the economic growth this variable is chose as one
explanatory variable and expected to have positive sign.

L;labour means any valuable service rendered by a human agent in the production of wealth,
other than accumulating and providing capital or assuming the risks that are a normal part of
business undertakings. It includes the services of manual laborers, but it covers many other kinds
of services as well. It is not synonymous with toil or exertion, and it has only a remote relation to
“work done” in the physical or physiological senses. The application of the physical energies of
people to the work of production is an element in labor and it is one of the explanatory variable
with its expected sign is positive.

General Inflation (INF) = Inflation is defined as an increase in the overall price level in a
country and measured in percent. In Ethiopian history inflation was not a problem of economic
growth. However, starting 2008 it is a serious problem. Therefore to analyze its effect on
economic growth, it is the other interest of the researcher’s, which is included in this study as
independent variable. The coefficient of this variable would be expected a negative sign. As we
discussed earlier, this study was used annual secondary data from 1974 to 2016 sourced from
Ministry of Finance and Economic Development (MoFED); National Bank of Ethiopia (NBE);

18
Central Statistical Agency (CSA) and World Bank Indicator (WBI). Except inflation, all the
variables are measured in millions of birr

I; investment can be defined allocating money to assets with the hope that in the future it would
provide some benefit such as generation of income. Property and gold are some common
examples of the traditional type of investment. The value of property, gold, mutual funds and
shares bought today may see a significant increase in the future with its expected sign is positive.

3.5. Long run co-integration test

In analysis a time series data regression of non-stationary data on another non stationary data
may give as spurious regression. In this case, multiple time series variable is nonintegrated and
shares some common stochastic trend. One can interpret the existence of co integration in the
long run between two or more stationary variables then the deviation from this long run path are
stationary. A time series can, however be co integrated if it is integrated of some order(Eagle and
Granger, 1987)and Johnson(1990)show some times in a time series economics even if a time
series data on two or more variable are non-stationary their liner combination may be stationary.
If there is an evidence of long-run relationship (co integration) of the variables, the following
long-run ARDL model will be estimated:

p p p
LnGDP=αo+∑ β oLnGDPt-1+∑ β 1REERt-i+∑ β 2-iKt-i+
i=1 i=o i=o

p p p p

∑ β 3Xt-i+∑ β 4FAt-i+∑ β 5It-i+∑ β 6Lt-t+

i=o i=o i=o i=o

∑ β 7INFt-i+εt............................................................................................................................................................ (25)
i=1

3.6. Unit root test

It is fundamental to test for the statistical properties of variables when dealing with time series
data. Time series data are rarely stationary in level forms. Regression involving non-stationary
(I.e., variables that have no clear tendency to return to a constant value or linear trend) time
series often lead to the problem of spurious regression. This occurs when the regression results

19
reveal a high and significant relationship among variables when in fact, no relationship exist.
Moreover, Stock and Watson (1988) have also shown that the usual test statistics (t, F, DW, and
R2) will not possess standard distributions if some of the variables in the model have unit roots.
The other necessary condition for testing unit root test when we applying ARDL model is to
check whether the variables enter in the regression are not order two (I.e. I(2)), which is
precondition in ARDL model. Therefore, it is necessary to test for time series variables before
running any sort of regression analysis.

Non-stationary can be tested using Augmented Dickey-Fuller (ADF) test and Phillips Perron
(PP) test. However, to ensure reliable result of test for stationary, the study employs Augmented
Dickey-Fuller (ADF) tests. The testing procedure for the ADF unit root test is specified as
follows:

p
∆y=α+δ t+γY t-i+∑ λ ∆ Yt-i+ϵ t------------------------------------------ [26] Where Yt is a time series
i=1

variables under consideration in this model at time t, t is a time trend variable; Δ denotes the first
difference operator;ϵ tis the error term; p is the optimal lag length of each variable chosen such
that first-differenced terms make a white noise. Thus, the ADF test the null hypothesis of no unit
root (stationary).

That is: H0:= 0; H1:= 0

If the t value or t-statistic is more negative than the critical values, the null hypothesis (I.e. H0) is
rejected and the conclusion is that the series is stationary. Conversely, if the t-statistic is less
negative than the critical values, the null hypothesis is accepted and the conclusion is that the
series is non-stationary.

3.7. Error correction model (ECM)

In addition to learning potential long run relationships between two series, the concept of co
integration also enriches the kinds of dynamic models. An ECM is a short run model which
reflects the current error in achieving the long run equilibrium relationship among variable. ECM
is a short run model, the coefficients of the independent variables show short run relationships of

20
them with the dependent variables. ECM is used to estimate the short run relation; we will use to
study the short run relationship among the variables under consideration

CHAPTER FOUR

4. DATA ANALYSISE AND DISCRIBITION

4.1 descriptive analyses

4.1.1 Economic performance the country
Ethiopia is steadily recovering from the 2015/16 and 2017 droughts, with continued expansion of
services and industry and a rebound in agriculture. In 2016/17, registering a 10.9 percent
expansion compared to 8 percent in 2015/16 At 36.3%, services accounted for the largest share
of GDP in 2016/17, 36.3%, driven by trade, transport, and communications, although this share
decreased from 50% in 2015/16. Industry’s share of GDP increased from 38.8% in 2015/16 to
40.4% in 2016/17, driven by construction, electricity, and manufacturing. Implementation of the
export-led industrialization strategy supported growth in industry. Although agriculture’s share
of GDP stagnated at 36%, the sector’s growth rate increased from 2.3% in 2015/16 to 6.7% in
2016/17 due to rising commodity prices, notably for coffee. This gradual but steady shift in the
structure of the economy reflects the government’s policy of developing manufacturing sector
and promoting export-led growth while continuing to give due attention to modernizing the
agriculture sector which has dominated for long the country’s economic base. Rapid and
sustainable economic growth over the last 15 years has led to improvements in income inequality
and poverty reduction. Accordingly the level of per capita income has reached USD 863 in
2016/17 compared to USD 784 last year. Poverty has declined to 22 percent from 38.7 percent in
2004/05. Investment to GDP ratio has increased to 39 percent and domestic savings t GDP ratio

21
to 24.1 percent Real GDP growth during 2017/18–2018/19 will be led by greater agricultural
productivity and strong industrial growth.

The government pursued a contractionary fiscal policy stance in 2016/17, prioritizing spending
in pro-poor and growth-enhancing sectors, including education, health, agriculture, and roads.
Capital expenditure accounted for a large share of the budget, though it decreased from 51% in
2015/16 to 46% in 2016/17. The 2016/17 budget deficit was 1 percentage point lower than
programmed; the ratio of tax revenue to GDP remained low, at 12.9%. Revenue enhancing
measures are expected to increase tax collection. The monetary policy stance has been consistent
with the Central Bank’s objective of maintaining low and stable inflation, which was below the
8% target in 2016/17. The Central Bank is implementing a contractionary monetary policy to
address inflationary pressures due to rising food prices. In October 2017, the birr was devalued
15% to boost exports. Merchandise exports increased 1.4%, while imports decreased 5.5%,
reducing the current account deficit. Remittances remained stable at $4.4 billion (6% of GDP) in
2016/17; FDI increased 27.6%, to $4.2 billion. The GDP of country The Ethiopian economy
which had exhibited 9 percent average annual growth during 2012/13-
2016/17, registered 10.9 percent growth in 2016/17, depicting recovery from challenging
macroeconomic and weather conditions of the previous year. The registered growth rate in real
GDP was 0.2 percentage point lower than base case scenario GTPII target set for the fiscal year
although growth estimated for Sub-Saharan Africa (World Economic Outlook Update, October
2017).The real values of the economic sectors in 2016/17 were based on 2015/16 base year; and
others are on 2010/11 fiscal year. As a result, there will be some adjustments when the rebasing
of national Accounts Statistics is finalized. The figure below show that the trend of GDP growth
from 1976 to 2016.as we see the graph below the economic growth of Ethiopia is increase from
time to time as discussed in the theoretical part above.

22
Figure 4.1Trend of GDP growth

LNRGDP
12.4

12.0

11.6

11.2

10.8

10.4
1980 1985 1990 1995 2000 2005 2010 2015

SOURCE; NBE

4, 1.1.Export and Ethiopian Economy

The export sector has played an important role to bring about rapid economic growth in
developing countries. However, most of them largely depend for their source of currency earning
on a single product or a very narrow range of low value products, mostly agricultural
commodities and minerals. Likewise, the Ethiopian commodity export sector is basically
characterized by the dominant share of agricultural raw commodities in generating the greater
proportion of the export earning of the country. These export commodities together have
accounted more than 86 percent (NBE, 2012/13) of the total merchandise export earnings. The
major export items, in order of their significance in the total commodity export value include
coffee, gold, oil seeds, hides and skins, pulses, chat, flower, fruits and vegetables. The total
amount of export value in 1974 was 8,192 Million Birr. It has steadily increased to 10,136
Million Birr in 1983 at constant value. In the year 1984, however, it was fluctuated and starts to
decline and reached 4,727 Million Birr in 1991. After the reform period i.e. 1992, export
earnings increased with little fluctuations and recorded 122,632 Million Birr.

23
Figure 4.2 Trend of export

LNEX
20

18

16

14

12

10

4
1980 1985 1990 1995 2000 2005 2010 2015

Source; NBE

Despite the increment of export as share of real GDP from 3.6 percent in 1991 to 18 percent in
2013, the trade balance as share of real GDP (resource balance) continued to rise to 17.8 percent
during 2013/14 from 14.2 percent in 2002/03 (NBE, 2013/14). This huge gap resource balance
indicates that exports of goods and service have insignificant impact on the Ethiopian economic
growth. According the GTP of the Ethiopian government, Export of goods and non-factor
services were expected to play an important role in accelerating the growth of the economy.
However, the export sector played an insignificant role particularly in 2011/12 and 2012/13
(MoFED, 2012/13). The fluctuated growth of exports of goods and service (see figure4.2) were
associated with the agricultural commodity export, since agriculture, in general, is under the
vagaries of nature, particularly in the Ethiopian case, the high concentration on non-traditional
export goods resulted in an unstable export performance. Ethiopia’s export sector is
characterized by over dependence on few agricultural products, with very limited exports of
manufactured and semi-manufactured goods. This structure of concentrating on few agricultural
commodities has not significantly changed over time. Besides, Coffee has still remained to be the
dominant export commodity, though its share in the value of total exports fluctuates from time to
time. It accounted for, an average, 40 percent of export earnings in 2016/17 is Gold, Oil seeds,
chat, flower, pulses, and live animal have share of 7.2, 12.1, 9.4, 7.1, 3.1 and 2.3 percent,
24
respectively during the period. The combined share of coffee along with the above six items
were 60 percent. This indicates that the dominant shares of Ethiopian export sector are
agricultural commodities and it confirms that the diversification of the export sector is limited to
these agricultural raw commodities.

4.1.2Aid and Ethiopian economy

As we know that, Ethiopia is one of the poorest countries in the world, which depend its export
sector on agricultural primary commodities (68 percent) that suffers from low levels of domestic
saving and insufficient amount of foreign exchange required to purchase capital goods. Not only
this but also the country suffers from budget deficit, which required external assistance to
enhance economic growth. Foreign aid, could fix this problem by supplementing domestic
saving or foreign exchange reserves. Foreign aid donated to Ethiopia from different donors
increased during the study period starting from 131.95 Million of Birr in 1976 to 3.455 billion of
Birr in 2010. The total Financial and technical aid during the military power (1976-1991) was
45,285 Million Birr while 583,497 Million was registered post 1992 (1992-2013) (from MoFED,
time series data),even though the Aid ratio to real GDP continuously grew from 1.2 percent in
1976 to 11.5 percent in 2003, the economic growth fluctuated, which registered the highest
growth rate (13.1 percent) in 1986 and lowest growth rate (8.8 percent below zero point) in 1984,
which was due to best rain seasons and catastrophic drought and famine, respectively. As a result
the aid was served for consumption of the society rather than financing economic developments.
However, Aid ratio to real GDP increases to 14.5 percent in 2010 (i.e. the highest value under
the study period) from 11.5 percent in 2003, the economic growth also went in the same
direction and registered an average growth rate of 11 percent for the period of 2003-2013. Aid
increase with 4.074 billion in 1016/17 than in 2015/16 that registered by 3.234 billion birr (as
seen figure 4.3 below).

25
Figure 4.3 Trend aid

LNAID
22.5

22.0

21.5

21.0

20.5

20.0

19.5

19.0

18.5
1980 1985 1990 1995 2000 2005 2010 2015

SOURSE; WB

4.1.3 Trends of Inflation in Ethiopia

Trends of inflation show the change in the inflation over the years. Looking at the trends of the
Inflation would enable the reader to understand the change of inflation during the study periods.
Ethiopia has recorded strong economic growth over the past 10 years mainly due to the
Government-led development policies emphasizing public investment, commercialization of
agriculture and non-farm private sector development. Inflation remained at a reasonable low
level rate before 2000/03. However, post 2003/04 sow sharp increase despite rapid economic
growth during the same period (Alemayehu and Kibrom, 2008). According to Alemayehu and
Kibrom (2008), the sharp increasing of general inflation was caused primarily by food inflation,
which is the effect of food demand triggered and international food price hike. The official
headline inflation during 2008 stood at about 33 percent with food inflation being about 49
percent. This was huge macroeconomic shock in the history of Ethiopia for the last five decades

26
and until 2003, was below 5 percent per annum (Ibid). This high rate of inflation continued until
2011/12. The 12 month moving average general inflation rate, which shows a longer inflation
situation, was 18 percent for June 2010/11 and 33.7 percent in June 2011/12.
68 According to the MoFED report (2012/13) the high inflation rate, particularly in the year
2011/12 adversely affected the wellbeing of people and the effect to promote private investment.
As a result, the Ethiopian government had taken policy measurements (prudent fiscal and
monetary policy) and price stabilization intervention. Consequently, as of June 2012/13, the
general inflation declined to 13.5 percent. As the National Bank of Ethiopia report indicates
(2013/14), the general inflation declined from 13.5 percent in 2012/13 to 8.1 percent in 2013/14.
Generally, we can say that, in the Ethiopian history inflation was at reasonable low level (i.e.,
does not harm the economy significantly) except for the period 2008-2012.In 2016/17, the
annual average headline inflation slowed down to 7.2 percent from 9.7 percent a year ago. This
was largely owing to 3.9 percent decline in food & non-alcoholic beverages inflation from
11.2percent to 7.3 percent and 1.1 percent drop in nonfood inflation from 8 to 7 percent. Figure
4.4 show the trend of inflation from 1976-2016.

27
Figure 4.4 trend of inflation

LNINF
5.2

4.8

4.4

4.0

3.6

3.2

2.8

2.4

2.0

1.6
1980 1985 1990 1995 2000 2005 2010 2015

Source; NBE

4.2 Economic analyses

4.2.1. The Unit Root Test Analysis
In order to determine the degree of integration, a unit root test is carried out using the standard
Augmented Dickey-Fuller (ADF) test. Moreover in applying ARDL model all the variables
entered in the regression should not be integrated of order two. To check these conditions, unit
root test is conducted before any sort of action taken. Even though the ARDL framework does
not require per-testing variables to be done, the unit root test could convenience us whether or
not the ARDL model should be used. The ADF test solves this problem by considering a higher
order and augmenting the random walk equation with some more lags. It is suggested to allow
both an intercept and time trend in the regression model used to test the presence of unit root.

In both tests the null hypothesis is that the variable is non-stationary against the alternative
stationary. The null hypothesis is rejected only when there is strong evidence against it at the
conventional levels of significant. It is an augmented version of the Dickey and Fuller (1979) test
for a larger and more complicated set of time series models.
28
The augmented Dickey–Fuller (ADF) statistic, used in the test, is a negative number. I.e. the
more negative it is, the stronger the rejections of the hypothesis that there is a unit root at some
level of confidence. The null hypothesis of a unit root is rejected if the t-statistic associated with
the estimated coefficients exceeds the critical values of the test.

The practical rule for establishing the number of lags is that it should be relatively small in order
to save degrees of freedom, but sufficient to remove serial correlation in the residuals. The
weakness in this test is that the power of the test may be adversely affected by miss-specifying
the lag length (Rao, 1994). Variables (RGDP, INF, L, EX, AID and INV) are stationary at first
differencing according to augmented ducky fuller tests but REER and K are stationary at level
and at second difference respectively.

If the value of computed ADF statistics is greater than the absolute value of critical statistics the
model is stationary. In this model absolute value of computed ADF statistics model exceeds than
the absolute statistics of the critical value at first differencing stationary tests in 1%, 5%, 10%
level of significance for variables (RGDP,INF,L,EX,AID and INV) except REER and K are
stationary at level and at second difference respectively (see table4.1).

29
Table 4.1 Augmented Ducky Fuller test
Variable ADF test at ADF test at ADF test at second
level first deference deference
Test of critical value

LNRGD -1.058004 -7.066497 -6.099455

P 1% level -3.610453

5% level -2.938987
LNREE -8.131380 -5.967474 -1.612932
R 10% level - 2.607932

0.669046 -5.549555 -8.499170

LNINF

-0.771155 -5.802854 -6.969747

LNL
-0.032913 3.453086
-8.178573
LNK
-0.212920 -6.245042 -7.002718

LNEX
LNAID -1.321151 -7.937932 -7.63167

LNINV 0.427693 -8.703500 -7.013789

Source: Eview 9.0 results

30
3.2.2 Lag Length Determination
Since the dependent variable is time series type, the previous year observation of either the
dependent or independent variable may be considered as an explanatory cofactor. This implies
that the previous year value of the dependent variable may be also taken into account as an
independent variable in the regression. This process could force the research to determine the lag
length that will be considered in the operation. In some occasions three of them may show
different level of lag. In this research four of them (LR, FBE, HQIC and HQ) methods show that
the maximum lag length that should be considered in the operation is three. But SC shows that it
is possible to consider up to the third lag length. It is not uncommon to find consistent result
when alternative strategies for model choice are used, it may lead to different outcomes that
make some subjective judgment necessary (Hill et al., 2011). Given this, the research considers
only third lag of the dependent as well as the independent variables as an additional explanatory.

Table 4.2 lag length criteria determination

Lag LogL LR FPE AIC SC HQ

0 -160.0788 NA 9.60e-07 8.846250 9.191005 8.968911

1 147.2435 469.0708 2.83e-12 -3.960184 - -2.856233
0.857389*
2 216.9903 77.08861 3.35e-12 -4.262649s 1.598186 -2.177409
3 346.2846 88.46451 5.11e-13* - 0.919684 -
* 7.699190* 4.632661*

* indicates lag order selected by the criterion

31
4.2.3 Bodes test
According to Pesaran et al.(2001) supply bounds on critical value for asymptotic distribution of
F- statistics, based on his assumption there is upper and lower bonds on critical value, in this
case the lower bond is based on the assumption of variables in order zero(I(0)),and the upper
bond is all the variables order one (I(1)).If the calculated F-static fall below the lower bound, we
conclude that variables are order zero (I (0)) and If the F- statistic greater than the upper bond,
we conclude we have co-integration. IN our regression the calculated F-statistics exceeds the
order one (upper bond) in the critical value of 10% and 5% level of significance. The bond test
for existence of long run relationships between dependent and independent variables showed that
there is relationship GDP and all independent variables. This is because the calculated F-statists
(4.4) are greater than the upper bond of the critical F-statistics at 10% and 5% level of
significance (see table 4.3). This implies that the null hypothesis of no long run relationship is
rejected. Rather than accepted the alternative hypothesis (there is long run relationship) based on
Pesaran et al.
(2001) and Narayan
Table 4.3 bounds regression results
(2004).

Test Statistic Value No

variables

F-statistic 3.469530 7

Critical Value Bounds

Significance I0 Bound I1 Bound

10% 2.03 3.13

5% 2.32 3.5
32
2.5% 2.6 3.84
1% 2.96 4.26
4.2.4. Long Run ARDL Model Estimation
After confirming the existence of long-run co-integration relationship among the variables, the
next step is running the appropriate ARDL model to find out the long run coefficients, which is
reported in table 4.4 below.

Table 4.4 Estimated Long Run Coefficients using the ARDL

Variable Coeffici Std. Error t-Statistic Prob.

ent

LOG(REER) - 0.134364 -0.485348 0.6340

0.06521
3
LOG(INF) - 0.182738 -0.841310 0.4126
0.15373
9
LOG(L) - 0.749176 -1.476477 0.1592
1.10614
1
LOG(K) 1.41393 0.680722 2.077116 0.0543
9
LOG(EX) 0.04528 0.017938 2.524479 0.022
4
LOG(AID) 0.09253 0.096097 0.962971 0.3499

33
 9
LOG(INV) 0.07802 0.050425 1.547293 0.1413
2
C 10.3117 6.161870 1.673483 0.1137
85

Source Eview 9.0

This result is in-line with the theory of economic growth which states that capital stock is the
major determinates of economic growth (Keynesian theory of growth, Solow’s theory of
growth).
Since we have specified the growth model in a log-linear form, the coefficients can be
interpreted as elasticity with respect to real GDP. Capital stock (k) is statically significance at 5%
level of significance and has positive relation with RGDP and 1 percent increase in capital stock
results increase the GDP of Ethiopia by 1.4 percent. This indicates that, capital stock has a long
run impact and yield higher return for the Ethiopian economy growth. The findings of this
research concerning the long run positive impact of capital stock on Ethiopian economic growth
are consistent with the endogenous growth theories in the long run which argue that capital stock
leads to productivity improvement that enhances output.
In line with the research conducted in Ethiopia by Gezehegn (2012), he found that, total exports
of goods and service was insignificant on Ethiopian economic growth even though there is
positive impact, while there is an inverse relationship between economic growth and export
volatilely, but the result of this study revealed that the impact of total exports of goods and
service on Ethiopian economic growth during the study period has a positive relationship and it
is statistically significant at 5% level of significance. As a result 1percent increase in export
increases GDP by 0.045 percent.
As indicated in table 4.4 above foreign aid has positive impact on Ethiopian economic growth.
But it is statistically not significant during the study period, which is also in line with the study
of Dasha Chatrna (1980-2007) conducted in 85 developing countries including Africa. An
empirical result provide in the literature part also report a mixed results. In my opinion, there
might be two possible reasons behind the insignificant result. In the first case, the inflow of
foreign aid might finance for consumption than for investment, which is used to accelerating
economic growth. If so, it does not have any impact on economic growth due to value add to

34
macroeconomic growth. The second reason behind the insignificant result may be associated
with the data inconsistency, which I took from the organization (World bank, 1976- 2016) to
solve the problem of no long time series (40 years) data availability for this variable only.

4.2.5. Short Run Error Correction estimation

After the acceptance of long-run coefficients of the growth equation, the short-run ECM model is
estimated. The error correction term (ECM), indicates the speed of adjustment to restore
equilibrium in the dynamic model. It is a one lagged period residual obtained from the estimated
dynamic long run model. The coefficient of the error correction term indicates how quickly
variables converge to equilibrium. Moreover, it should have a negative sign and statistically
significant at a standard significant level (i.e. p-value should be less than 0.05).

35
Table 4.5 Error Correction Representation for the Selected ARDL
Variable Coefficient Std. Error t-Statistic Prob.

DLOG(RGDP(-1)) 0.274073 0.266575 1.028124 0.3192

DLOG(REER) -0.041322 0.078909 -0.523663 0.6077
D(INF) -0.097415 0.098533 -0.988651 0.3376
DLOG(L) -1.133719 4.803081 -0.236040 0.8164
DLOG(L(-1)) 13.932056 7.547861 1.845828 0.0835
DLOG(L(-2)) -15.688199 5.390731 -2.910217 0.0102
DLOG(K) -0.051717 0.101982 -0.507118 0.6190
DLOG(K(-1)) 0.271086 0.135507 2.000534 0.0627
DLOG(K(-2)) -1.242658 0.694870 -1.788333 0.0927
DLOG(EX) 0.024319 0.010472 2.322302 0.0337
DLOG(EX(-1)) 0.005245 0.009588 0.547014 0.5919
DLOG(EX(-2)) -0.016636 0.008908 -1.867686 0.0802
DLOG(AID) -0.012776 0.044573 -0.286635 0.7781
DLOG(INV) 0.016871 0.017553 0.961180 0.3508
DLOG(INV(-1)) -0.027523 0.016905 -1.628161 0.1230
ECM-1 -0.633637 0.261200 -2.425873 0.0275

Source Eview 9.0

The error correction coefficient, estimated at -0.633637 is highly significant, has the correct
negative sign, and imply a very high speed of adjustment to equilibrium. According to Bannerjee

36
et al. (2003) as cited in Kidanemarim (2014), the highly significant error correction term further
confirms the existence of a stable long-run relationship. Moreover, the coefficient of the error
term (ECM-1) implies that the deviation from long run equilibrium level of real GDP in the
current period is corrected by 63.4 % in the next period to bring back equilibrium when there is a
shock to a steady state relationship in this regression results error correction model is negative
and statically significant. The variable (L) is positive and statistically significant at 10%level of
significance in first difference and negative and statically significance at 5% level of significance
in second differences. Capital is positive and statistically significant at 10% level of significance
in the first but negative and statically significance at 10% level of significance in the second
difference and also export is positive and statistically significance at 5% level of significance at
level. This indicates the positive relationship between export and Ethiopian economic growth.
On the other hand, with two period lagged, export is significant at 10 percent significance level
with negative coefficient.

4.2.6 Diagnostic Test

To check the verifiability of the estimated long run model, some diagnostic test is undertaken.
Priority in doing any analysis, we required to check the standard property of the model. In this
study we carried a diagnostic checking, which includes Serial correlation test (Brush & Godfray
LM test), Normality (Jaque-Bera test), and Hetroscedasticity test. In order to reject or accept the
null hypothesis, we can decide by looking the p-values associated with the test statistics. That is
the null hypothesis is rejected when the p-value are smaller than the standard significance level
(i.e. 5%).

4.2.6.1 Normality test

The testes for normality revealed that there is no problem of normality. That means, the residual
data are normally distributed. Because in our regression model the probability is exceeds that of
5% (0.95>0.05).

37
 Figure 4.5 test of normality

4.2.6.2Auto-coloration tests
The null hypothesis of no serial correlation (Brush Cod fray LM test) is failed to reject for the
reason that that the p-values associated with test statistic is greater than the standard significant
level (I.e.0.65 >0.05). Her LM test for testing serial correlation is applied because unlike the
traditional Durbin Watson test statistic which is totally inapplicable when the lagged dependent
variable appear as a regressors, LM test avoid such limitation of DW test

Table 4.6 Auto-coloration tests

F-statistic 3.022237 Prob. F(3,13) 0.0681

Obs*R-squared 15.61334 Prob. Chi-Square(3) 0.6514

Source Eview 9.0

4.2.6.3Heteroskedasticity tests
The testes result for Heteroskedasticity implies there is different variance exists in the model.
This model is not Heteroskedasticityi.e. chi-squre is >standard significance level (1.0000>0.05).
This means the null hypothesis of homoscedastic is accepted (see table 4.6).

38
Table 4.6Heteroskedasticity test

F-statistic 0.67964 Prob. F(21,16) 0.7989

2
Obs*R-squared 17.9157 Prob. Chi- 0.6543
6 Square(21)
Scaled explained SS 3.26153 Prob. Chi- 1.0000
9 Square(21)

Source Eview 9.0

CHAPTER FIVE

5. CONCLUSION AND POLICY RECOMMENDATION

5.1 CONCLUSION
The main objective of this study is to analyze the determinants of economic growth in Ethiopia
during the specified period. To determine the long run and short run relationship among the
variables, Autoregressive Distributed Lag (ARDL) model was applied. Before applying the
ARDL model, all the variables are tested for their stationarity properties using the ADF. As a
result, real effective exchange rate is stationary at level, while real GDP, inflation, labor, export
and aid are stationary at first difference and also capital is stationary at second difference.

Next to testing for time series property, the model stability was done by testing the diagonal
testing techniques. The result revealed that no evidence of serial correlation, the residual is
normally distributed and no evidence of hetroscedasticity problem. As we discussed above, this
study applied the methodological approach called ARDL model also known as bound test
approach. As the result indicted the bound test (F-statistic) value is larger than the upper bound
critical value at5% and 10%level of significance. This indicates there is a long run relationship
between real GDP and its determinants in long run during the study period.

39
The empirical result showed that both capital and export are found to have positive impact on
Ethiopian economicgrowth during the study period and statistically significant at 5% level
significance in the long run and one percent increases in capital results 1.4 percent increase in
real GDP in long run. This indicate that capital have a great return for the growth of GDP of the
country. The study found out that export of goods and service statistically significant impact on
economic growth with positive sign in the long run. From this one can understand that exports of
goods and service.
The other finding of this study is the negative and insignificant impact of inflation, and real
effective exchange rate on Ethiopian economic growth during the study period. Despite their sign
are inconsistent in long run and short run, they have insignificant impact on Ethiopian economic
growth. From this one can understand that significantly those variables are not harming the
Ethiopian economic growth during the study period.
There is also one outcomes i.e productivity of labour have negative impact and spastically
insignificant in the long run onGDP of Ethiopia which is unexpected. I couldn’t find any
justification for it. It is an open issue which needs further study.

5.2 Policy Recommendation

Based on the finding of this study, the following policy recommendations are forwarded.
 In order to enhance the contribution of the capital stock government of Ethiopia has to set
policies to increase domestic saving which is believed as a back bone of growth. This
includes increase saving mobilization like selling of government Bonds, expanding
financial institutions and by strengthening existing saving tools.
 Exports of goods and service have significant and positive impact on Ethiopian economic
growth. Therefore, the Federal Government of Ethiopia should strengthen the existing
strategies in export development and promoting investment particularly in the
manufacturing sector for export and import substitution as well as government work on
the quality of export products to compute in international market.

40
41
 References

Alemayehu G. and Befekadu D. (2005), “Explaining Africa Growth Performance”: The case of
Ethiopia, Final Report (FR), August, 2005.

Asmamaw Mulusew (2012),” Relationship between Inflation and Economic Growth in

Ethiopia”, Msc thesis in Economic, University of Gonder, June 2012

Banerjee, A., Dolado, J., Galbraith, J., and D. Hendry (2003).Co-integration, Error Correction,
And the Econometric Analysis of Non-stationary Data: Advanced Text in Econometrics New
York: Oxford university press

Barro Robert J. (1991),”Economic Growth in a Cross Section of Countries”, the Quarterly

Journal of Economics, Vol. 106, No. 2 (May, 1991), pp. 407-443

Barro Robert J. (1997),”Determinants of Economic Growth: A Cross-Country Empirical Study”,

NBER Working Paper No. 5698

Barro Robert J. (2003),” Determinants of Economic Growth in a Panel of Countries”, Annals Of

Economics and Finance 4, 231–274 (2003)

Befekadu D. and Birhanu N. (1999/2000), “Annul report on Ethiopian economy”, Addis Ababa
Ethiopia,

Domar Evsey (1946), Capital expansion, rate of growth and employment. Vol.14, pp. 137-147.
Domar Evsey (1947). Expansion and employment American Economic Review, Vol. 37, pp.34-
55.

E. M. Ekanayake, and DashaChatrna, (2008),” The effect of foreign aid on economic growth in
Developing countries”, Journal of International Business and Cultural Studies

Elhanan Helpman (2004), the mystery of Economic Growth, Harvard University Press,
Cambridge (2004)
“Macroeconomic Determinants of economic growth in Ghana: Cointegration Approach”,
European Scientific Journal, July 2013 edition vol.9, No.19

X
Eshetu and Makonnen (1992), “The Macroeconomic Performance of the Ethiopian Economy
1974-90”, the Ethiopian Economy: Structure and Policy Issues

Gezahegn G. (2012), “Long-run effect of Export volatility on GDP: Case of Ethiopia”, Södertörn
University Department of Social Sciences| Economics, Master Program, Thesis, 2012

Hailemariam A. (2011), “The Impact of External Debt on Economic Growth and Private
Investment in Ethiopia: A Vector Auto Regressive Approach”, Ethiopian Economic Association.

Harrod, Roy F. (1939), an essay in dynamic theory. Economic Journal Vol. 49, pp. 14-33.

Harrod, Roy F. (1948), towards a dynamic economics, London: Macmillan.

IMF (2013), “Country Report to Ethiopia No. 13/308, International Monetary Fund Publication
Services, Washington, D.C

IMF working paper WP/02/69 (2002), “Eternal debt and Growth”, by Cotherine pattillo, Helene
Poisons and Luca Ricci, Apr 2002.

Keynes, J.M (1936), The General theory of Employment, Interest and money

Kidanemariam G. (2014), “The Impact of Human Capital Development on Economic Growth in

Ethiopia: Evidence from ARDL Approach to Co-Integration”, A Peer-ReviewedIndexed
International Journal of Humanities & Social Science, Vol-II, Issue-IV, April- 2014.

MoFED (2013), “Ethiopia Macroeconomic Performance (2001-2012) final report (Published),

Addis Ababa.

MoFED (2012/13) “Growth and Transformation Plan: Annual Progress Report”, for F.Y.
2012/13

NBE (2012/13), National Bank of Ethiopia Annual Report (bulletin), 2011/12, Addis Ababa

NBE (2012/13), National Bank of Ethiopia Annual Report (bulletin), 2012/13, Addis Ababa

NBE (2012/13), National Bank of Ethiopia Annual Report (bulletin), 2013/14, Addis Ababa

XI
S. Seid M. (2000), “Determinants of economic growth in Ethiopia”, unpublished Master Thesis,
Addis Ababa University

Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations, first
Published in 1776, The Glasgow Edition of the Works and Correspondence ofAdam Smith, two
voles, Oxford: Oxford University Press.

Solow Robert M. (1956), “A Contribution to the theory of Economic Growth”, The Quarterly
Journal of Economics, Vol. 70, No. 1 (Feb., 1956), pp. 65-94, The MIT Press

Tadesse Demissie. (2011), “Sources of Economic Growth in Ethiopia: A Time Series Empirical
Analysis”, Master thesis, University of Oslo

Tasew Tadesse. (2011), “Foreign Aid and Economic Growth in Ethiopia: A Co integration
Analysis”, the Economic Research Guardian – Vol. 1((2), Semi-annual Online Journal,

Teklu K., Pr. D.K Mishra and Melesse A. (2014),” Public External Debt, Capital formation and
Economic growth in Ethiopia”, Journal of Economics and Sustainable Development Vol.5,
No.15,

WB (2015), World Bank, World Bank Data base last updated: 01/30/2015

Master thesis, Addis Ababa University

Woubet K. (2006), “Human Capital and Economic Growth in Ethiopia”, unpublished Master
Thesis, Addis Ababa University

XII
Appendixes

Appendixes 1lag length criteria determination

Lag LogL LR FPE AIC SC HQ

0 -160.0788 NA 9.60e-07 8.846250 9.191005 8.968911

1 147.2435 469.0708 2.83e-12 -3.960184 - -2.856233
0.857389*
2 216.9903 77.08861 3.35e-12 -4.262649s 1.598186 -2.177409
3 346.2846 88.46451 5.11e-13* - 0.919684 -
* 7.699190* 4.632661*

Appendixes 2 test of normality

XIII
 Appendixes 3 Auto-coloration tests

3.02 0.0
F-statistic 2237 Prob. F(3,13) 681
15.6 Prob. Chi- 0.6
Obs*R-squared 1334 Square(3) 514

Appendixes 4Heteroskedasticity test

0.6 0.
796 79
F-statistic 42 Prob. F(21,16) 89
17. 0.
915 Prob. Chi- 65
Obs*R-squared 76 Square(21) 43
3.2 1.
615 Prob. Chi- 00
Scaled explained SS 39 Square(21) 00

ARBAMINCH UNIVERSITY
XIV
 COLLEGE OF BUSINESS AND ECONOMICS

DEPARTMENT OF ECONOMICS

DETERMINANTS OF ECONOMIC GROWTH IN ETHIOPIA

(TIME SIERES ANALYSIS)

By:-Wasyehun Getie

Advisor:-Mubarek J. (Msc)

A Research submitted to Arba Minch University college of Business and

Economics Department of Economics in partial fulfillment of the
requirement for Bachelor of Art (BA) Degree in economics.

JUNE, 2018

Arba Minch Ethiopia

XV
XVI