RESEARCH PROPOSAL ON DETERMINANTS OF INFLATION IN ETHIOPIA

OROMIA STATE UNIVERSITY

COLLEGE OF BUSSINESS AND ECONOMICS

DEPARTMENT OF ECONOMICS WEEK END PROGRAM 2012 E.C ENTRY

Group Name: ID. No

1. Mezgebu Asrat …………………….. W/12/06043

2. Bezawit Mideksa …………………... W/12/O6017
3. Matiwos Wakwoya ………………… W/12/06067
4. Lelisa Regasa ……………………….. W/12/06038
5. Fetene Bacha …………………………W/12/06025

Submitted to; Instructor Kedir

Lege Tafo Ethiopia
May: 2023
Table of Contents

Acknowledgement …………...............................

Abstract …………………………………………

1. Introduction.......................................................

2. Literature review…...........................................

2.1 The theoretical framework ...........................

2.2 The empirical frame work …………………..

3. Research methodology.......................................

3.1 Source of Data...................................................

3.2 Methods of Data Analysis.................................

3.3 Model specification and Variable description...............

3.4 Augmented Dickey Fuller Test........................................

3.5 Lag order selection for VAR............................................

3.6 Co integration Test ………................................................

3.7 Vector Error Correction Model Specification................

4. Empirical results and discussion..........................................

4.1 Trend and analysis..............................................................

4.2 Unit root test........................................................................

4.3 Choosing optimal lag length...............................................

4.4 Co integration Analysis ............................................................

4.5 Two Long run Equations............................................................

4.6 Vector Error Correction Model.................................................

4.7 Model Checking............................................................................

4.8 Impulse Response Function.........................................................

5. Conclusion......................................................................................................

6. References..........................................................................................................
Acknowledgement

We praise God for giving us time, strength, support and for the guidance to surpass all the trials that we
have encountered to finish this proposal. Then we would like to express our deepest gratitude and
appreciation to our instructor Mr. Kedir, for his advice and guidance throughout this proposal. Finally,
we want to thank our families for their love and support.

Abstracts

This proposal examines the determinants of inflation in Ethiopia, using Vector Error Correction Model
(VECM) by using annual time series data from 2012 to 2022. Augmented Dickey-Fuller unit root test
indicated that the variables are integrated of order one. However, the variables transformed to
stationary by taking the first difference. The Johansen co-integration test revealed that the existence of
long-run relationship between variables. Furthermore, the coefficient of vector error correlation model
indicates that there is a positive and significant relationship between inflation, budget deficit and
national debt. However, the effect of money supply on inflation is only on the short run. Finally, the
model is stable if a shock happens in the future.

1. Introduction

1.1 Back ground of the study

Inflation measures a rise in the overall price level of goods and services in a given economy. It is a
decline of purchasing power of a given currency (Ethiopian Birr in this case). A quantitative estimate of
the rate at which the decline in purchasing power occurs can be revealed in the increment of an average
price level of a basket of selected goods and services in the country during the study period. Individuals
with tangible assets such as property and stocked commodities may like to see some inflation as that
raise the value of their asset. But those holding cash may not like inflation, as it erodes the value of their
cash.

The majority of the developed countries have low rate of inflation and stable economy over years.
Sweden, UK and USA have registered 2.04%, 2.48%, 2.44% as inflation rate in 2018, respectively. On the
other hand, developing countries have unstable economy and high inflation rate. Ethiopia, Sudan and
Angola have registered 13.83%, 63.2% and 19.63% as inflation rate in 2018, respectively.

Ethiopia was ruled by the military junta between 1974 to 1991. During this regime, the government was
following a command market system and prices were controlled by the government. The government
was also rationing goods at a fixed price to the public which in turn had contributed to attain lower
inflation rate. The annual average inflation was 5.2 percent between 1980-2002 (Menji, 2008).

After the overthrow of the military Junta in early 1990s, the Ethiopian economy switched to a market
system. The first ten years of the new government was characterized by a low inflation rate and a low
economic growth. However, post 2002 Ethiopian economy is one of the fastest growing economies in
Africa and had been praised by World Bank (2010), International Monetary Fund, Economic commission
for Africa (Atilola, 2007). At the same time, inflation began to emerge as a major issue following some
policy changes such as less conservative monetary and fiscal policy implemented by the current
government (Geda & Tafere, 2008).

The data from National Bank of Ethiopia indicates that the rate of inflation in Ethiopia was 10.69%,
13.83%, 15.84%, 20.36%, 26.84%, 33.8% in 2017, 2018, 2019, 2020, 2021, 2022 respectively which is
increasing in the past few years. Double digit inflation which is caused by several factors has become
worrying for the policy makers and for the citizens, especially for those who lives in the capital city.

The determinants of inflation are different from one country to another. From the economic
perspective, these determinants have been classified as supply side and demand side factors. Supply
side factors are those economic factors which cause inflation by increasing cost of production. Some
important supply side factors are output growth, capital formation, import prices, exchange rate, tax
and wage. On the other hand, demand side factors lead to inflation by decreasing the purchasing power
of money. Some relevant demand side factors are increment of money supply, private consumption and
government expenditure. Several studies have attempted to identify the determinants of inflation in
Ethiopia. Based on their result, authors give recommendation to policy makers to control high rate of
inflation. Similarly, the purpose of this study is to identify the determinants of inflation in Ethiopia from
2012 to 2022 by using annual time series data.

1.2 Statement of the problem

Inflation is usually considered as one of macroeconomic problems of an economy since maintaining a

stable price level is one of the key indicators of macroeconomic stability and economic prosperity Early
studies (Tafere 2008,Tezera 2013) on the determinants of inflation in Ethiopia were focused on
identifying the determinants of national inflation. Tafere(2008) by developing a synthesis model of
monetarist and cost-push inflation theories and estimated using a vector autoregressive (VAR) and
single equation error correction models to identify the short run and long run determinants of inflation
at national level.

1.3 Objectives of the study

1.3.1 General objective

The general objective of the research is to investigate the determinants of inflation in Ethiopia evidence
from Addis Ababa city nature and causes of the inflation.

1.3.2 Specific objective

To identify the determinants of inflation in Ethiopia.

To describe relationship between inflation and other macro-economic variables in Ethiopia.

To find out major determinants of inflation in Ethiopia.

To suggest possible course of actions to solve the problem.

1.4 Significance of the study

The study will results confirm that both cost push and demand pull factors contribute to inflation in
Ethiopia. The finding from this study will reveal that inflation can be controlled by reducing money
supply, government expenditure and oil price and by increasing external debt and real output.

1.5 Scope of the study

This proposal will cover the macroeconomic determinants of recent inflation in Ethiopia. The study
covers the time period from 2012 up to 2022. The period was chosen because it can be explain the
inflationary trend experienced in Ethiopia.

1.6 Organization of the thesis

The proposal is organized into five chapters where chapter one provides introduction of the topic.
Theoretical as well as empirical reviews of literatures on inflation are organized in chapter two. Chapter
three presents the research methodology and design, data sources and variables, as well as model
specification. Chapter four deals with descriptive analysis on the overview of inflation in Ethiopian
economy. Finally, chapter five presents the conclusions of the proposal.

2. Literature review

2.1 The theoretical framework

2.1.1 Structuralist Theory of Inflation

The concept of structural theory of inflation is discussed by Myrdal (1968) and Streeten (1972) for the
first time (Canavese, 1982). The theory explains the inflation in the least developing countries (LCDs) in
terms of the structural features of the countries. Both Streeten and Myrdal (Canavese, 1982) have
argued against the direct application of the orthodox aggregative analysis to the LDCs. The orthodox
aggregative analysis assumes the existence of balanced and integrated structures in the economy where
production, consumption, backward and forward linkages in response to market signals are reasonably
smooth and fast, such that it is rational to talk in terms of aggregate demand and aggregate supply.
However, the majority LDCs are characterized by unstable economy, backward agriculture, weak
institutions, underutilization of natural resource, and frequent war etc. Because of this, it is difficult to
apply aggregative analysis to the LDCs.

Structuralists believe that inflation in LDCs is bound with developmental effort and structural response
to this effort is expressed through gaps of various kinds in these countries. The gaps that have
mentioned in literatures are Resource gap, food bottleneck, foreign exchange bottleneck and
infrastructural bottleneck. They suggest that, to understand the true nature of inflation in LDCs one
must identify the determinants that force to generate bottlenecks of various kinds in the normal process
of development, study how the bottlenecks lead to price increases and how these increases spread to
the whole economy. Since Ethiopia is one of the least developing countries, the structuralism theory and
suggestion hold in Ethiopia too.
2.1.2 Cost push inflation

Cost-push inflation occurs when overall prices increase due to increases in cost of wages enforced by
trade unions and cost of production. This type of inflation was identified during the medieval period, but
it was reviewed in the 1950s and 1970s as the main cause of inflation. There are many causes of cost
push inflation. High increment of wages more rapidly than the productivity of labor is one of them.
Trade unions push employers to increase wage considerably, thereby raising the cost of production of
commodities. Consequently, employers increase the price of their commodities. Even if the wage is
increased, the workers would buy only as much as before because of the price adjustment on products
by the producer.

Once again, the trade unions demand higher wages, and the producer will set higher price. This vicious
circle process leads to cost-push or wage inflation. Another cause of cost-push inflation is profit push
inflation. To maximize their profits, Oligopolies’ and monopolist firms charge high prices for their
products to offset production and labor costs. Because of the nature of oligopolies’ and monopolist
market, firms are in charge of setting price of their products: that is why profit- push inflation is also
called price-push inflation.

2.1.3 Demand-pull Inflation

Demand-pull inflation is the upward pressure on prices as a result of increase in aggregate demand. John
Maynard Keynes and his followers emphasized the increase in aggregate demand as the source of
demand-pull inflation. The aggregate demand consists of investment, consumption and government
expenditure. Inflation arises when the value of aggregate demand exceeds the value of aggregate supply
at the full employment level. Keynesians did not deny this fact that even before reaching full
employment production factors and various constraints can cause increase public price. According to
Keynesians, policy that causes decrease in each component of total demand is effective in reduction of
pressure demand and inflation.

2.1.4 Quantity Theory of Money

Quantity theory of money (QTM) states that money supply and price level in the economy are directly
proportional to one another. Irving Fisher showed the relationship between them as follows:

M*V= P*T Where,

M is Money supply

V is Velocity of money

P is price level

T is Volume of the transactions.

In the above relationship the velocity of money is assumed to be constant. When there is increment in
money supply, the price will adjust by the same or lower proportion rate immediately.
2.2 Empirical Literature Review

Empirical studies examine the determinants of inflation by using different econometric tools such as co
integration, vector autoregressive and vector Error Correction Model etc. A large number of studies
have investigated the determinants of inflation across the world (Ochieng et al., 2016; Olatunji et al.,
2010).

When we think about the causes of inflation, two dimensions might be useful: domestic and external
factors. The study result of African development bank (AFDB, 2011) indicates that the causes of inflation
in Ethiopia, Uganda, Tanzania and Kenya are world food and oil prices, domestic production and
monetary, fiscal and exchange rate policies. In the short run, external factors are outside the control of
these countries, because of production capacity constraints. The frequent drought in the region
worsened the food situation, causing a sharp increase in food prices. In addition to this, the rising world
oil prices have been transmitted to domestic inflation, aggravated rapid depreciation in exchange rates
across all four countries (AFDB, 2011).

In Nigeria the total export, total import, agricultural output, interest rate, government expenditure,
exchange rate and crude oil was included in the model as a determinant of inflation (Olatunji et al.,
2010). The study result reveals that the previous year total imports, government expenditure, and
exchange rate have negative influence on inflation rate (Olatunji et al., 2010). However, export,
agricultural output, interest rate and crude oil exports have negative impact on inflation (Olatunji et al.,
2010).

The empirical studies stated above have presented different results. The difference in results depends
on the economy of the countries, period of the study, the method used, and the variables included in
the model.

3. RESEARCH METHODOLOGY

3.1 Source of Data

This study discusses Macroeconomic determinants of inflation in Ethiopia from 2012 – 2022 by using
annual time series data. All the secondary data was collected from National Bank of Ethiopia.

3.2 Methods of Data Analysis

After the required secondary data was collected, the researcher used both descriptive and econometric
analysis to identify determinants of inflation in Ethiopia from 2012-2022. The descriptive analysis was
used to analyze the trend of inflation in Ethiopia while econometric analysis was applied to examine the
relationship between inflation and the explanatory variables by applying different tests.

3.3 Model specification and Variable description

The model used in this study includes Macroeconomic variables as a determinant of inflation in Ethiopia.
The variables computed in this model are the result of economic factors and governmental policies
which are assumed to affect inflation. The dependent variable is Consumer Price index as a proxy to
inflation and it is the variable whose behavior in a relation to explanatory variables has been
investigated. According to Ethiopian central statistics agency report in December 2016, about 54 % of
the household expenditure spent on food, beverages other goods and services. On the other hand,
Producer Price Index (PPI) is not a good measure in Ethiopian economy since the proportion of income
spent on the purchase of raw materials is low in the country. Therefore, it is appropriate to use CPI as a
measure of price change in Ethiopian economy.

The empirical model used in this study can be specified as follows:

log CPIT = 𝛽0 + 𝛽1logGM2t + 𝛽2logBDt + 𝛽3logGNDt + ut

Where, the parameters β 1, β 2 and 𝛽3 is the long run elasticity’s of the independent variables and 𝛽0 is
the value of the dependent variable when all independent variables are zero.

Consumer price index (CPI): measures the average change in prices over time that consumers pay for a
basket of goods and services. Such changes in the prices of goods and services have an effect on the real
purchasing power of consumer's income and their welfare.

Money Supply: traditionally, money supply is defined from its narrow and broader sense. Narrow
money supply is a measure of money stock proposed primarily for the use of transactions. The National
Bank of Ethiopia takes the broader definition of money as a money supply. Similarly, this study used the
growth rate of broad money in local currency unit as a money supply. As discussed earlier in the
theoretical part, we expect positive relationship between inflation and money supply.

Budget Deficit: it occurs when government expenditure exceed revenue and indicate the financial
health of the economy. This gap between revenue and expenditure is subsequently filled by government
borrowing. Inflationary effect of budget deficits depends upon the means by which the deficit is
financed. The effect of budget deficit on inflation has been controversial in the field of Economics.

National Debt: is the total outstanding borrowing of the central government comprising of internal and
external debt incurred to finance its expenditure especially from International Monetary Fund (IMF) and
World Bank in case of Ethiopia. Increment of national debt will have a positive effect on inflation. Since,
the National debt of Ethiopia has been increasing in the last two decades we expect positive effect of
national debt on inflation.

3.4 Augmented Dickey-Fuller Test

In time series data most of the variables are non-stationary, which means that they usually exhibit unit
root. When the variables exhibit a unit root, it indicates that their mean and variance changes over time
or non-constant. Because of this, a regression based on non-stationary variables leads to misleading
result with high 𝑅2 while there is no meaningful relationship between variables.

In order to avoid the misleading regression problem, with its related non-stationary pattern of the
variables, differencing has become the most common method of converting non-stationary time series
variable to stationary. When a variable is stationary at level, then it is integrated of order zero. However,
a variable is said to be integrated of order one or I (1), if it is stationary after differencing once. Most of
the time series variables become stationary after the first difference (Stock H. & Watson W., 2019). This
study used the Augmented Dickey-Fuller (ADF) test, which follows the same features as the Dickey-Fuller
statistic by adding the lagged value of the dependent variable (Gujarati & Porter, 2009). ADF test aims at
checking for the presence of unit root in a time series under the null hypothesis that a unit root is
rejected in favor of the alternative to be stationary.

3.7 Vector Error Correction Model Specification

Anoruo and Ahmad recommend using an ordinary VAR in the first difference if the variables in a data set
are not co integrated (Anoruo & Ahmad, 2001). However, if they are co integrated, a VECM which
combines levels and differences can be estimated instead of VAR in levels (Maitra, 2019). Additionally,
VECM allows analyzing the short-run dynamics and long-run equilibrium relationships in the data set
(Razaghi Khamsi, 2016).

VECM includes the lagged error correction term to measure the duration of the deviation of the
variables from the long run- equilibrium. In Practice most empirical applications analyze multivariate
systems. Since, there are four variables in this proposal, we can say that it is multivariate analysis. Let us
consider a VAR with P lags:

Yt =  + Σ𝛾𝑖𝑝𝑖=1Yt-1+ ut

Where, Yt is a K × 1 vector of variables,  is a K × 1 vector of parameters, 𝛾𝑖 is K × K matrices of

parameters, and Ut is a K × 1 vector of error term. Since all variables have a unit root and there is co
integration the best model is VECM. We can transform the VAR model to Vector error correction model
by taking the first difference of the variables.

Vector error correction model can be written as follows by taking the first difference of any VAR(P):

ΔYt =  + Σ𝛾𝑖Δ𝑝−1𝑖=1Yt-1+ 𝜆Et-1+ Ut t= 1……T

Where, ΔYt =Yt -Yt-1

If there is co integration among variables in the long run, the error correction term ECT will adjust
gradually the deviation from the long-run equilibrium through a series of partial short-run adjustments.
ECT is represented by Et-1 and the coefficient 𝝀 is the speed of adjustment. In VECM the dependent
variable is a function of its own lag, a function of the lagged values of explanatory variables in the
model, error correction term and disturbance term (Stock H. & Watson W., 2019). After VECM is applied,
the dynamic relationships among variables can be best understood by examining the impulseresponse
function (Dibooglu & Enders, 1995). Impulse response function measures the effect of a shock caused by
an endogenous variable on itself or another endogenous variable (Kilian & Lütkepohl, 2017).
4. EMPIRICAL RESULTS AND DISCUSSION

4.1 Trend and analysis

Looking at the trends of variables would enable the readers to understand how the variables changed
over time. Trend analysis is a graphical illustration of the variables in the model for a given period of
time. It depicts the ups and downs (fluctuations) of the interest variables due to several factors such as
policy changes, increment in world price, drought in the region and political instability in the region for
example starting from 2015 to still know there was change of government leaders, war between
Ethiopian army and TPLF, in most regions of the country there is no peace and security.

4.2 Co integration Analysis

We have discussed that the variables in the model are integrated of order one and the optimal lag
length is two. when all variables are integrated of order one and multivariate, Johansen co integration
test is one of the tests that could be applied to check if there is a long run relationship between the
variables based on predetermined number of lags. The decision criteria to determine the number of co
integration equation in the model is based on comparing the trace statistic value with the 5% critical
value. There is co integration equation in the model if the trace statistic value is greater than the critical
value and vice versa.

4.3 Two Long run Equations

The long run coefficients are exactly estimated using VECM approach to show the long run response of
the dependent variable to changes in the independent variable. Co integration equations in the model is

Ln CPI = - 2.74 + 2.38lnGND +0 .609lnBD………………………. (1)

Since all variables are written in logarithm form and co integrating vector is estimated, the coefficients
can be interpreted as a long-run elasticity. Thus, in the first long run equation, a 1% increment in
national debt and budget deficit will increase consumer price index (inflation) by 2.38 % and 0.61%
respectively. Both variables are statistically significant at 1%. The result implies that the increment of
budget deficit and growth of national debt in the past few years caused increment of price in Ethiopia.

lnGM2 = - .044 +.243lnGND + .0155lnBD……………………….......... (2)

The second long run equation can be interpreted as follows: a 1% increment in national debt and budget
deficit will increase consumer price index (inflation) by 0.24% and 0.015% respectively. Both variables
are statistically significant at 1%. To sum up everything that has been said so far this test results agree
with economics theory that increment of national debt and budget deficit has a positive relationship
with money supply.
4.4 Vector Error Model Correct

We have seen that the variables are co integrated in the long run. This section reviews the short-run
behavior of the variables in the model. The error correction term in VECM indicates the speed of
adjustment to reach the equilibrium when there is a shock in the short run. It is good when the
coefficient of the error correction term is negative and statistically significant.

4.4.1 Model Stability condition Test

We should also evaluate the stability of the estimated VECM. Model stability is used to check if the
model stays stable or explode if a shock happens in the future. If a VECM has n endogenous variables
and m co integrating vectors, there will be n − m unit model in the companion matrix. For stability, the
model of the remaining m eigenvalues should be strictly less than unity.

4.5 Impulse Response Function

Impulse response function (IRF) can be used to show the change in the dependent variable as a result of
a shock in the independent variable. IRF also shows that the sign and how long the shock stays. The
presence of integrated order one variables in VECM implies that the shocks can be permanent or
transitory. Consequently, in this study the shocks are positive and permanent.

5. Conclusion

In the previous chapter we have discussed that budget deficit and growth of national debt has a positive
and significant effect on inflation in the long run. However, money supply affects inflation only in the
short run. Depending on the result of the model, we can conclude that growth of budget deficit (Excess
government expenditure than revenue) and national debt are the causes of price increment in Ethiopia.

6. References

• Ahuja (2001). Modern macroeconomics second edition

• Alina carare, Andera Schaechter, Mark, stone and Zelmer (2002) establish in initial

•Anderson Yahyak (1989). Structural disequilibrium and inflation in Nigeria and New.

• Asheber (2005). Couses and effects of inflation on economic growth in Ethiopia.

• Central statistical (2013/14). Report on the Ethiopian economy.

• Connell MC. Campbell rand Brue, L. Stanely (1986). Contemporary labor economics.

•Dammar N. Gujarati (1995). Basic econometrics: fourth edition

•Ethiopian economic association (Ethiopian economic policy research institute (2013). Report on the
Ethiopian economy.

• Kibrom Teferi (2008) food price inflation in Ethiopia. Reports on the Ethiopian economy.