Scholars continue to argue passionately about how exchange rate devaluation affects trade balance in developed and developing nations, despite the fact that numerous theoretical and empirical research have looked at this link. This research explores the effect of exchange rate devaluation on Ethiopia's trade balance. Utilizing annual time series data collected from the National Bank of Ethiopia; from 1984 to 2018 the study employs descriptive and econometric methods for analysis. In the econometric analysis, the Autoregressive Distributed Lag model (ARDL) is utilized, incorporating variables such as trade openness, real GDP, money supply, inflation rate, and real effective exchange rate index.
The results indicate a stable cointegration relationship between trade balance and its determinants. Trade openness emerges as the most influential factor, affecting both short and long-term trade balance. Real effective exchange rate devaluation is found to have a positive significant effect on trade balance in the short term, but showing a deteriorating effect in long term. The study challenges Marshall-Learner condition, revealing that devaluation strategy employed by Ethiopia does not align with condition for improving trade balance. While devaluation has positive short-term effects, it is insufficient for long-term correction of trade deficits. The research suggests that policies focusing on internal supply-side factors, encouraging production of exportable and substitute goods, are crucial for sustained improvement in trade balance. In conclusion, the research sheds light on the complexities of trade balance dynamics in Ethiopia, highlighting the multifaceted nature of factors influencing the effectiveness of exchange rate devaluation as a policy tool.
Table of Contents
2.1. Theoretical Literature Review.. 3
2.1.1. Definition of Balance of payment and Its Component 3
2.1.2. Definition of Devaluation. 3
2.1.3. Economic Impact of devaluation. 3
2.2. Empirical Literature Review.. 4
3.1. Types and Source of Data. 6
3.2. Method of Data Analysis. 6
3.3.1. The Autoregressive Distributed Lag (ARDL) Model 6
4.1. Descriptive Data Analysis. 8
4.1.1. Devaluation in Ethiopia. 8
4.1.2. Devaluation and Ethiopian Export and Import 9
4.1.3. Devaluation and Ethiopia’s Trade Balance. 10
4.2.1. Structural Break Test with an Unknown Break Point 13
4.2.2. Model Specification. 13
4.2.3. ARDL Bound Test for Cointegration. 14
4.2.4. Hetroscedasticity Test 15
4.2.5. Auto Correlation Test 15
4.3. Regression Result and Interpretation. 16
4.3.3. The Marshal Learner Condition and the J Curve Phenomena. 18
4.3.4. Stability Test for the Model 19
5. Conclusion and Policy Implication. 21
6. Availability of data and materials. 22
1. Introduction
In developing countries, various economic problems have deteriorated their economic situation. Internal and external problems are common among the economic problems. External balance problems, particularly the trade deficit problem, are severe in Africa.To overcome these problems, countries have suggested implementing a number of policy measures. The most commonly used policy measures to correct the internal and external balance of countries are expenditure changing and expenditure switching policies. In particular, to solve the trade deficit of a country,expenditure switching policy devaluation is the most adopted one(Salvatore, 2004).
The structural adjustment program (SAP) was introduced in the 1980s in over 30 sub-Saharan African (SSA) countries with poor growth performance and severe macroeconomic disequilibrium. At the center of this program is the exchange rate adjustment aimed at improving economic growth and stabilization by enhancing the incentive to produce tradable goods, particularly agricultural tradable goods(Befekadu and Berhanu 2000).
In October 1992, Ethiopian Birr (ETB) saw a major free fall when it was devalued by 142% from its pegged rate of 2.07 per US dollar to 5 per US dollar, signaling the first major onslaught on the value of ETB, which since then has been virtually on a slipperyslope. However, the trade deficit continued to widen from 1992/93, from about 625 million US dollars when the exchange rate reforms were initiated, to more than 6.3 billion US dollars in 2008/09 (an increase of more than 904%!)(Alekaw, 2012).
The Ethiopian People Revolutionary Democratic Front (EPRDF) started implementing extensive macroeconomic and structural changes in order to address this economic issue. Among these reforms is depreciation. The International Monetary Fund (IMF) believed that devaluation was the final resort for developing nations looking to adjust their trade imbalance and balance of payments (BOP).
After the devaluation in 1992, the exchange rate was changed from fixed to flexible in order to control overvaluation through a gradual depreciation of the domestic currency every year, and the gap between the unofficial and official rates also decreased compared to the period when the exchange rate was fixed. After 1992 the government took a devaluation effect in September 2010 and had the rate jumping from 13.6 ETB/USD to 16.3 ETB/USD amounting to 17 percent. The latest one is the official devaluation of birr against US dollar by 15% in November 2017,which moved up the exchange rate from 23.3 ETB per US dollar to 27 ETB per US dollar This huge devaluation was expected to decrease overvaluation and increase competitiveness(NBE, 2018).
Following the 16.7% devaluation of the ETB in 2010, the Real Effective Exchange Rate (REER) started appreciating again in 2011. Since 2014, not with standing the depreciation of the official exchange rate against the USD by 6% in 2015/16 and 5.8% in 2016/17, the pace of appreciation has continued unabated. The REER overvaluation of ETB was estimated to be 30% in 2015 and 20% in 2017. Recent empirical analyses conducted by the World Bank, IMF, and others have concluded that the Ethiopian economy has been suffering from an overvalued exchange rate. It was estimated that a 10% depreciation of the REER would reduce the current account deficit by approximately 2% of GDP (through a 5% increase in exports and a 6% decrease in imports) and increase real GDP growth by more than 2% per year. However, in the short term, devaluation will have an immediate adverse effect of raising external debt and debt service payments in the local currency(Ababa et al., 2017).
In general, empirical evidence of the impact of devaluation on trade balance is mixed. A study by(Alekaw, 2012),(Temesgen, 2016), Taye, H (1999), and (Borena, 2013) on the Ethiopian economy, and(Muhammad, 2011) on Pakistan economy found that devaluation is effective in improving the trade balance of Ethiopia. However, studies by(Haile, 2008) and(Yilkal, 2012)on the Ethiopian economy and ,(Zeeshan, 2016) on Pakistan economy found that devaluation deteriorates the trade balance of Ethiopia. Loto, M. A. (2011) tested the Marshal Lerner Condition (MLC) effectively on the Nigerian economy and proved that the MLC did not hold. The implication of this is that devaluing the Nigerian Naira will not improve the trade balance of Nigeria.
Over the course of 27 years, the trade deficit has continued to rise despite the government's best efforts to enhance the export performance sector and lower import expenditure through its currency devaluation strategy. Many studies have been carried out as a result of the widespread interest in the effects of devaluation on Ethiopia's trade balance. Aside from their vacuous theories, the majority of studies solely take into account the 1992 devaluation, although others also take into account the 2010 devaluation. Nevertheless, the claims and empirical data are conflicting and call for additional empirical proof. Consequently, this study investigates the effects of devaluation on trade balance by taking into account the three-devaluation era, verifying the J curve phenomenon, and testing the Marshal Lerner Condition for Ethiopian trade balance by taking into account structural break tests. The findings of the research are used by policymakers to influence trade and exchange rates.
2. Literature Review
2.1. Theoretical Literature Review
2.1.1. Definition of Balance of payment and Its Component
One of the most important sources of information about a country’s international economic position is its balance of payments. This is a summary statement of all transactions between the residents of a country and the rest of the world. The balance of payment account consists of the following two accounts.The first is that the current account is a part of the balance of payments in both visible and invisible trade. Visible trade, imports, and exports consist of physical merchandise, whereas invisible imports and exports are services, transfer payments, interest, profit, and dividends. The second component of the balance of payments is the capital account, which records long-term and short-term capital movements between countries(Devkota, 2004). However, this study considers only the trade account of the current balance, which is visible trade imports and exports.
2.1.2. Definition of Devaluation
Devaluation has received several definitions based on the experience of different economic conditions at different points in time. According to the World Bank’s (2001) definition, devaluation is a measure that a government may take to reduce the value of its currency in terms of foreign currencies.Devaluation is expected to be a useful measure to correct trade imbalances. It has several effects on major macroeconomic variables. It reduces expenditure and stimulates the level of output through the multiplier effect. At the other extreme, it has an inflationary effect and increases import costs; if the nation involved is import-dependent, the cost of production increases. This also increases the cost of servicing foreign debt ( Loto, M. A. 2011)
2.1.3. Economic Impact of devaluation
In theory, devaluation promotes exports and aggregates economic activity through the “multiplier effect.” However, currency devaluation may not produce desired outcomes for several reasons. First, the MLC may not hold in the short run. According to the MLC, ceteris paribus, a country will improve its current account deficit by devaluing its currency provided that the sum (in absolute value) of the elasticity of demand for exports and imports is greater than one; otherwise, it does not improve the trade balance.
Second, if we allow for changes in some variables, such as changes in national income, devaluation will improve the trade balance only if the improvement in trade balance generated by currency depreciation more than offsets the improvement in imports brought about by a rise in national income,Third there may be a J-curve effect of devaluation for two major reasons: a) export receipts may not increase in the short run due to supply side constraints associated with time logs, which is largely the case for agricultural commodities that need several months to harvest; and b) most imports are less responsive, if not, non-responsive at all despite the increase in their prices after devaluation. This applies to most capital goods and raw materials (such as oil) that have an inelastic demand in capital – deficient and oil-importing countries. As a result, the fall in foreign spending on the county’s exports and the increase in domestic spending on imports will cause the trade deficit to worsen before it improves(Pavle Petrović & Gligorić 2010).
2.2. Empirical Literature Review
Devaluation is a common adjustment tool used by developing nations, especially to improve trade and payment balances. However, evaluating the effects of devaluation can be challenging. Since certain least developed countries (LDCs) have frequently used devaluation as a key tool in their policies, it should be possible to get a good picture of how this strategy has affected these nations' experiences.
Loto, M. A. (2011) used the Ordinary Least Square (OLS) method to analyze the effectiveness of devaluation in improving Nigeria’s trade balance (test of the ML condition). The import demand function is positive and significant. The elasticity is 0.15926. The price variable also has an expected sign that is both negative and significant. The result of the export function also shows the positive and significant impact of the world income growth rate on export performance. The elasticity is 0.0619. The relative price variable is negative and significant. The study was able to effectively test the Marshall – Lerner condition. It was proved that for the Nigerian economy, between 1986 and 2008, the ML condition did not hold. The implication of this is that devaluing the Nigerian Naira will not improve the trade balance of Nigeria. Zeeshan ,(2016) Investigated the impact of devaluation on balance of trade by using ARDL over the period 1980-2014, this study verifies the long-run relationship between the balance of trade, currency devaluation, and external debt. The negative coefficient of the real effective exchange rate indicates theabsence of a J-curve in Pakistan. These results clearly indicate that devaluation will disfavor trade balance in the case of Pakistan.Muhammad Asif, (2011)analyzed the impact of devaluation on trade balance in Pakistan in both the long run and short run by using a bound testing approach to co integration and error correction mechanism (ECM). He found that devaluation is effective in improving trade balance, and there is a cointegrated relationship between the real effective exchange rate and trade balance using the ARDL model.
In the Ethiopian case,(Temesgen, 2016) and(Borena, 2013) examined the short- and long-run effects of the real effective exchange rate on Ethiopia’s trade balance. The ARDL approach was used by Temesgen for analysis of time series annual data from 1979/80 to 2013/14. Both the Temesgen and Borena econometric results reveal that in the short run and long run, the real<<<<<<<<<<< exchange rate has a significant and positive effect on trade balance. Accordingly, the real effective exchange rate has a short-as well as a long-run effect on Ethiopia’s trade balance.(Alekaw, 2012) examines the short- and long-run relationships between trade balance and real exchange rates in Ethiopia. The bound testing approach of the cointegration and error correction model, developed within the ARDL model framework, is applied to annual data for the period from 1970/71 to 2010/11. Variance decomposition (VDC) and impulse response functions (IRF’s) were used for further inferences. Using this approach, the estimated results show that exchange rate appreciation (depreciation) is negatively (positively) related to the trade balance in the long and shortrun, consistent with economic theories. (Haile, 2008 ) also examined the impact of devaluation on the economic performance of a subsistence economy using a medium-sized macro econometric model. The simulation results indicate that devaluation would help improve the current account balance but would be stagflationary.(Yilkal, 2012) analyzed the short- and long-run effects of currency devaluation on output growth in Ethiopia. The study was conducted using quarterly time-series data over the period ranging from 1997/98 to 2009/10 and employing a Vector Auto regression model. By controlling for monetary and fiscal policies, the study found that currency devaluations are contractionary in the long run and neutral in the short run.Fassil, (2017) examined the effect of Birr devaluation on Ethiopia’s trade balance for the period 1970-2014 using the Vector Error Correction Model. The key results of the present study revealed that Birr devaluation deteriorates Ethiopia’s trade balance in the short run and improves it in the long run, as well as the impulse response function, which revealed that a J-curve phenomenon exists for Ethiopia’s tradebalance. This means that the findings of this study show that the Marshall-Lerner Condition holds only in the long run.
Aleto A. (2018) studied the economy-wide impact of thedevaluation of Ethiopian currency (birr) using the Dynamic Computable General Equilibrium(DCGE) model.The findings reveal thatdevaluation results in a decline in overall GDP growth and domestic absorption. However, the trade balance improved under all simulation scenarios.
3. Research Methodology
3.1. Types and Source of Data
This study used annual time-series data for the period ranging from1984-2018. Data were collected from the National Bank of Ethiopia (NBE).
3.2. Method of Data Analysis
Descriptive and econometric styles were used to examine and dissect the different issues. In the econometric analysis section, the researcher used the ARDL method and impulse responses to examine the effects of trade openness, real GDP, money supply, inflation rate, and the real effective exchange rate index on Ethiopia's trade balance. The model is handed in a natural logarithmic form to grease the interpretation and analysis. Tables, percentages, and basic statistical tools were among the instruments used in the descriptive statistical analysis.
3.3. Model Specification
A country's balance of trade is defined by its net exports (exports minus imports), and is affected byall factors that affect international trade.
TB=X-M
Where TB is the trade Balance, X is the export earnings, and M is the import payment
The ordinary least square method can be applied using a multiple regression model. Therefore, the functional model for this study is as follows:
TB = f (RGDP,REERI, OPN, MS, INFR,) -------- (1a)
Where RGDP is Real Gross Domestic Product, REERI is Real Effective Exchange Rate index, OPN is trade openness which is the sum of exports and imports per real gross domestic product, MS is the money supply and INFR is Inflation Rate
From equation 1(a), the trade balance depends on the real effective exchange rate index, real gross domestic product, money supply, trade openness, and inflation rate. Therefore, the trade balance function for this investigation is specified in a log-linear form, as follows:
LnTB = β0 + β1LnRGDP + β2LnREERI + β3LnOPN + β4LnMS + β5LnINFR + Ui------- (1b)
3.3.1. The Autoregressive Distributed Lag (ARDL) Model
(Pesaran et al, 2001)proposed an ARDL/Bounds Testing approach to investigate the existence ofcointegration relationshipsbetween variables. The ARDL model version of Eequation (1b) is expressed as follows:
Where t = 1.2.3…..T is the number of years and k is the lag of the independent variable. The first part of the equation with β i (i = 1, 2...5) represents the short-run dynamics of the model, where as the parameter αi(i = 1, 2...5) represents the long-run relationship. The researcher applied bounds testing to determine the long-run relationships between variables. The F-test was used for the null hypothesis of no cointegration against the alternative; it is not true and compare the computed values of the F-statistic with the critical values given by the (Pesaran, 1997)and(Pesaran et al, 2001). The null hypothesis of the model is as follows.
H0: α1= α2= α3= α4= α5= 0
H1: α1≠ α2≠ α3≠ α4≠ α5≠ 0
If the bounds testing procedure confirms the long-run relationship, then in the next step, we estimate the following long-run ARDL model:
The next step is to obtain the short-run dynamic parameters by estimating an ECM associated with the long-run estimates to determine the speed of adjustment back to long-run equilibrium after a short-run disturbance. The equation for the standard ECM is as follows:
Here, β i (i = 1, 2...5) are the short-run dynamic coefficients of the model’s convergence to equilibrium, and λ is the speed of adjustment if its positive indicates divergence, while a negative indicates convergence. The coefficient of the lagged error correction term (λ) is expected to be negative and statistically significant, to further confirm the existence of a cointegrating relationship.
4. Results and Discussion
4.1. Descriptive Data Analysis
4.1.1. Devaluation in Ethiopia
By converting the former central and command economy into a market-based economy, the Ethiopian transitional government aimed to end poverty and promote peace, security, and economic progress. In November 1992, the transitional administration implemented a new economic policy with the aim of fostering development. In the area of foreign exchange, the government depreciated the ETB by 142% in October 1992, from its nominal value of 2.07 Birr per US dollar to 5.00 Birr per US dollar. A weekly action system for foreign exchange was also created in May 1993, and it operated reasonably effectively. State governors tightly controlled and manipulated the country's economic structure during the Derg era. Aid from the socialist state of the Soviet Union was utilized. It promotes social benefits over private ones, as economic activity is always focused on government investment made possible through public funding. The economy known as the command economy (autocracy) did not have any competitive international trade. That period's exchange rate was estimated to be less than one US dollar, about 2.07 birr. The strong purchasing power of this currency was caused by government regulation rather than the competitiveness of foreign trade. International traders were never able to reach Ethiopia since the government made society produce and consume domestically and imposed financial customs duties on importable items, despite the fact that there was a need for foreign commodities. In addition to the economic sanctions imposed, government officials discouraged foreign investment. Even original investors were required to demonstrate their financial capacity to the regime of investment coil (finance), and those with a total wealth exceeding 10,000 birr were subject to higher rates of taxation. As a result, exports decreased, even after paying export duties, which allowed Ethiopian exports to quickly transform the open economy's economic structure and facilitate international trade with Ethiopia. Due to the government's devaluation and the liberalization of trade with other nations, Ethiopia's import and export volumes increased. The Transitional Government of Ethiopia (TGE) devalued the Birr several times over the years, increasing the rate of depreciation from 1993 to 1997 (2.765, 5.001, 5.7704, 6.06, 6.69 per US dollar, and an official exchange rate of 22 Birr per US dollar in 2016).
Needless to say, the Ethiopian economy showed strong economic performance, especially from 2003 to2015, due to improvements in the service, construction, and agricultural sectors. Since then, the country has beenon a sustained, strong path of economic performance, with a mean real GDP growth of 11 percent for 2003–2015. On the other hand, Ethiopia has experienced a long period of unfavorable trade balance deficits, which means that the external sector is characterized by structural trade deficits. This may be partly due to thecomposition of imports and exports, and the import-intensive nature of the Ethiopian economy. (Fassil, 2017)
Ethiopia's exports are primarily composed of agricultural items, which have erratic demand in the global market. These products include coffee, hides and skins, raw materials, chat, pulses, oilseeds, and others. The primary imports into Ethiopia are capital goods, energy, raw materials, and consumer durables, all of which the nation cannot afford to reduce even in light of their increased cost. The Ethiopian government has worked very hard to reduce the external imbalance and boost production growth through the devaluation of the birr or its expenditure switching policy. The value of the Birr was devalued by 15% on October 10, 2017, and by 16% on August 31, 2010, when it was worth 16.35 US dollars instead of 13.63. These actions were reportedly made to improve export performance and implement fundamental changes in the economy. This was a brave and significant step since it suggests that Ethiopia's industrial development and external performance have been hampered by its recognition by the Ethiopian government, which is a policy setting.
4.1.2. Devaluation and Ethiopian Export and Import
Figure1.illustrates the trend of Ethiopia's export and import expenditures from 1984 to 2018.The graph shows that, prior to the devaluation, import and export trends remained steady after 1992 devaluation both the value of exports and import expenditures increased at an increasing rate. Specifically, the growth rate of imports during the 2010–2017 periods was 19.42%, higher than the growth rate of exports, which was 15.6% during the same period. This indicates that 2010 devaluation was insufficient to close the country's trade deficit, as evidenced by the fact that import expenditure growth outpaced export value growth even after the 2017 devaluation. Devaluation by itself cannot improve Ethiopia's trade balance unless it is accompanied by other policy measures, as Figure1.generally explains that the country's deficit trade balance frequently increases even when devaluation measures are unable to prevent this trade balance problem.
Figure1. Ethiopian export and import trend
Editor’s note:
Images are not included in the reading sample
4.1.3. Devaluation and Ethiopia’s Trade Balance
The net amount of a nation's imports and exports of products and services is known as its trade balance. If a nation's exports surpass its imports, its trade balance is positive, or it registers a surplus. Conversely, a nation's trade balance is negative or shows a deficit if the value of imports exceeds that of exports. Net exports are officially referred to as the trade balance in the current account.Since the beginning of its international trade, Ethiopia has had a negative trade balance. In order to address the problem of a trade deficit, the EPRDF has conducted measurements. Devaluation, or the revision of the exchange rate, was one measurement. This action is predicated on the improvement of the trade balance, a decrease in imports, and an increase in exports to the global market. However, the trade deficit issue remains unresolved despite this measurement.
Editor’s note:
Tables are not included in the reading sample
Source: NBE and own computation
The average trade deficit during the pre-devaluation period was 1353.89 million birr, with a growth rate of 7.3443%, from 1984 to 1991. To balance its trade, the Ethiopian currency was devalued in September 1992 by the country's transitional administration. But the trade balance deficit increased from -24,102.9 million Birr in 2004 to -82,884.69 million Birr in 2009; these trade deficits increased even after the country devalued in 2017, and the average trade balance deficit reached -337106.4 million Birr, indicating that the trade balance deteriorates.
The nature of Ethiopia's imports and exports accounts for a large portion of the country's trade imbalance, which has increased since the devaluation of the national currency. Devaluation makes a nation's exports less expensive than its imports. Since fuel, semi-finished goods, consumer goods, capital goods, and raw materials make up the majority of the nation's imports, these imports are crucial to the nation's development. The nation is unable to decrease imports of certain components, even if their prices rise.In terms of exports, the price decline resulting from devaluation and export demand abroad is not proportionate since Ethiopian exports are agricultural items that are income and price inelastic.Ethiopia's value of exports decreases as a result of the global drop in the price of coffee, the nation's main export good.
Editor’s note:
Tables are not included in the reading sample
Source: NBE and own computation
Generally,table2.shows that after devaluation, the trade deficit increases,which means that devaluation has a negative impact on trade balance, but we do not know whether it isstatistically significant; therefore, this study uses an ANOVA test to check whether devaluation has a real impact on Ethiopia’s trade balance.
Editor’s note:
Tables are not included in the reading sample
Source: own computation
The ANOVA result in table3.shows that it is significant Anova F-testis less than 5% (0.0002) that means it is statistically significant that devaluation have impact on trade balance in other word 1992, 2010 and 2017 devaluation can able to create a mean difference on trade balance.
Due to the agricultural foundation of Ethiopian economy, better economic growth demands developmental goods. Periodically, this results in a rise in import expenses. As we've shown in figure2.devaluing the trade balance doesn't help with the trade deficit; rather, it makes matters worse. The export trend shows that, despite increasing at a faster rate in the post-devaluation era, the export value is still unable to surpass the cost of imports. This suggests that improving trade imbalances through devaluation alone is not possible.
Figure2. Ethiopian Trade Balance
Editor’s note:
Images are not included in the reading sample
Ethiopia's trade deficit was consistent prior to the devaluation phase. On the other hand, the trade deficit issue got worse over time following the devaluation phase. Because capital goods and semi-finished commodities make up the majority of Ethiopia's imports and its exports are mostly agricultural products, devaluation has little positive impact on the country's trade balance.
4.2. Econometric Analysis
4.2.1. Structural Break Test with an Unknown Break Point
Before performing any regression, it is preferable to check for the presence of structural breaks. Hence, in this study, a preliminary structural break test was conducted using multiple breakpoint tests withan unknown break point. The multiple breakpoint tests tested whether there were a structural changein all equation parameters. Hence, it allows us to test whether there is astructural change in a subset of parameters. The year 1998/99 was selected as the break time.
In order to estimate the long- run and short-run correlations as well as the dynamic interactions among the variables of interest, this study used the ARDL/bounds testing cointegration approach. Finding the best lag option for the model is the first step in executing ARDL. To choose the best lags for the model, the Akaike information Criteria (AIC) approach was applied.
TB= f (REERI, RGDP, MS, INFR, OPN)
The AIC model with a lag order of 1,1,1,1,1,1,1,1,1,1,1,1,1 was selected.
4.2.2. Unit Root Test
Variables' order of integration was determined by testing prior to starting the ARDL bounds test. The unit root test could assist us in determining whether or not the ARDL model should be employed, even though the ARDL framework does not require pre-testing of variables. It could also help us make sure that the variables were not I(2) stationary or of different order than I(1).Because the limits test is predicated on the idea that the variables are I(0) or I(1). Outtara (2004) claims that the computed F-statistics given by Pesaranet al. (2001) are invalid in the presence of I(2) variables. The times series must thus be examined in order to identify the data-generation procedure and prevent erroneous outcomes.
Editor’s note:
Tables are not included in the reading sample
Source: Own computation
4.2.3. ARDL Bound Test for Cointegration
Table 4. Bound test for Cointegration Analysis
Tables are not included in the reading sample
In the ARDL regressions, the computed F-statistics are shown in the table4.when trade balance is taken into account as a dependent variable (normalized).The fact that the F-statistic (48.36) is greater than the upper-bound critical value (5.59) at the 1% level of significance indicates that there is a long-term link between the variables and that the null hypothesis that there is no cointegration among the variables is rejected.
4.2.4. Hetroscedasticity Test
Table 5. Hetroscedasticity Test: Breusch-Pagan-Godfrey
Tables are not included in the reading sample
4.2.5. Auto Correlation Test
When the value of Ut (the random term at time t) is correlated with (depends upon) its prior value, it is known as auto correlation. The stochastic or disturbance variables in any regression model may or may not be independent; if this condition is not met, then an issue can occur. OLS estimators remain unbiased yet inefficient as a result of the autocorrelation problem in time series analysis. Therefore, identifying this issue is important.
Table6. auto correlation test: Breusch-Godfrey Serial Correlation LM Test
Editor’s note:
Tables are not included in the reading sample
Source: own computation
As the result indicates in table6.the p-value is higher than 5% of significance (0.5472); it is statically insignificant, which means rejecting the alternative hypothesis of serial correlation and accepting the null hypothesis that there is no serial correlation.
Based on the above tests, we can say that the specified model is powerful enough to show a relationship between the dependent and independent variables.
4.3. Regression Result and Interpretation
4.3.1. Long Run Dynamics
Editor’s note:
Tables are not included in the reading sample
The ARDL approach was utilized to obtain the long-run coefficients. Results from table7.showed that all factors, with the exception of real gross domestic product and inflation rate, had a substantial impact on trade balance. This suggests a positive correlation between the real effective exchange rate index and trade balance.The actual effective exchange rate index's positive coefficient indicates that a 1% devaluation of our currency will eventually result in a 0.32% improvement in the trade balance. This demonstrates how a decline in the real effective exchange rate promotes exports and discourages imports, increasing a country's ability to compete globally and improve its trade balance.
These results are consistent with the findings of(Temesgen, 2016),(Alekaw, 2012)and(Borena, 2013)as their empirical findings also suggest that, in the long run, the real exchange rate has a significant and positive effect on trade balance. Further, all variables except money supply and inflation rate have a positive effect on the trade balance in the long run.
As shown in the table7, for a 1% increase in money supply and trade openness, the trade balances of the country increases by 0.606% and 0.607%, respectively. As we see from this, according to the result, the long-run trade balance of the country can be mostly affected by the money supply and trade openness of the country, as a country creates a liberalized trade system that becomes more exposed to international markets, which encourages specialization in sectors that have economies of scale that contribute to improving efficiency and productivity by creating more markets for domestic goods and services in the rest of the world. As a result, the trade balances of the country improved.
4.3.2. Short Run Dynamics
The results of the error correction model for trade balance are presented in table8.
Table 8. Error Correction Model for Trade Balance: ARDL(1,1,1,1,1,1,1,1,1,1,1,1,1)
Editor’s note:
Tables are not included in the reading sample
Source: own computation
The result has impressive diagnostic statistics, suggesting that short-term changes in the trade balance may be sufficiently explained by the factors selected in themodel. The goodness of fit of the model is high; the R-squared value of 0.99indicates that up to approximately 99% of the systematic short-run variations in trade balance at anygiven time is explained by the explanatory variables and the ECM term. Similarly, the F-statistic passes the significance test at the 1% level; thus, we cannot reject the hypothesis of a significant short-term relationship between trade balance and all the independent variables combined.
According to table8.real effective exchange rate, trade openness, and money supply havea significant effect on the trade balance of the country in the ECM, while the real gross domestic product and inflation rate are insignificant, implying that it takes sometime for the inflation rate to affect the trade balance. More specifically, in the short run, a 1% increase in the trade openness trade balance will increase by 0.82%. From this, wecan see that trade openness has a dominant effect on the trade balance in both the short and long run. The country can be mostly affected by the trade openness of the country as itcreatesa liberalized trade system that becomes more exposed to international markets, creating more markets for domestic goods and services in the rest of the world as a result of the trade balance ofthe country.
The equilibrium error correction coefficient (CointEq(-1)) estimate of -1.61 is highly significant, with the correct sign. This implies a high-speed adjustment to equilibrium after a shock.The coefficient -1.61 indicates a high rate of convergence to equilibrium, which implies that deviation from the long-run equilibrium is corrected by 161 percent over eachyear, or approximately 161 percent of disequilibria from the previous year’s shock converge back to the long-run equilibrium in the curren tyear.
4.3.3. The Marshal Learner Condition and the J Curve Phenomena
This paper's primary goals are to investigate the long-term relationship between Ethiopia's trade balance and exchange rate devaluation, as well as to validate the Marshal-Learner condition using a conventional econometric technique. As per the requirement, if the REERI coefficient's sign is significant and more than one, or if the total of the export and import demand elasticity is greater than one, devaluation will improve the trade balance. Regression estimates show that the REERI coefficient has a positive sign, is significant, and is smaller than one. Since the Marshal Learner requirement is not met by this outcome, currency depreciation does not enhance Ethiopia's trade balance.
4.3.3.1.Impulse Response Function
In order to trace the dependent variable's response to a unit and a one-time shock to other variables in the system, impulse response functions, or IRFs, are employed to determine the dynamic relationship between the dependent variable (TB) and the other independent variables.The trade balance's 10-year response to an initial, one-time shock to the real effective exchange rate index is depicted in the impulse response functions plotted below.
Figure3. The response of DLTB to Cholesky one standard error DLREERI innovation
Editor’s note:
Images are not included in the reading sample
The actual effective exchange rate index's reaction to a shock of one standard deviation to Ethiopia's trade balance is depicted in Figure3. This suggests that soon after the devaluation of a birr, devaluation increases trade balance. The impulse response function is therefore utilized to examine whether the J-curve phenomena holds in Ethiopia using time series data for the years 1984–2018. However, the trade balance of Ethiopia begins to deteriorate just two years later. Figure3.show that Ethiopia's response to the J-curve phenomena in terms of trade balances does not follow this pattern.
4.3.4. Stability Test for the Model
The stability of the ARDL model and the result of post-estimation diagnostics could affect the validity and robustness of the results, which should be tested prior to further analysis. It is tested using the cumulative sum (CUSUM) and cumulative sum square (CUSUMSQ). The CUSUM of recursive residuals and CUSUMSQ tests are applied to assess parameter stability (Pesaran, 1997).
Whereas the cumulative sum of squares test finds unexpected deviations from the constancy of the regression coefficients, the cumulative sum test finds systematic changes in the regression coefficients.
Figure4. Cumulative sum (CUSUM) graph at 5% significance level
Editor’s note:
Images are not included in the reading sample
Figure5.Cumulative Sum Square (CUSUMSQ) graph at 5% significance level
Editor’s note:
Images are not included in the reading sample
Source: own computation
The CUSUM and CUSUMSQ test results are presented in figures4.and figure5. Because the plots of the CUSUM and CUSUMSQ statistics lie inside the crucial bands of the 5 percent confidence intervals of parameter stability, the results show that there is no instability in the coefficients. As a result, throughout the Ethiopian sample period, there is stability in the coefficients.
5. Conclusion and Policy Implication
Currency devaluation has been used as a tool to correct global trade balance deficits. It was proposed by the IMF as a strategy for balance of payment adjustment, especially in developing countries. Ethiopia is a developing country that has adopted this strategy. The country devaluated and experienced its official devaluation of currency since 1992, after a change in government.
This research uses econometric and descriptive analysis to look at the effects of this policy and if it meets the Marshall Learner requirement between 1984 and 2018. Exports and imports are higher in the post-devaluation period than they were in the pre-devaluation period, according to the descriptive analysis, which evaluates the patterns of export profits, import spending, and trade balance in the pre- and post-devaluation periods. In contrast, the import value is higher than the export value when we compare the two, suggesting that the trade gap widens following the devaluation phase.
The ARDL model and econometric analysis were utilized to examine the effects of the real effective exchange rate index on Ethiopia's trade balance.
The main discovery of this research is that the trade balance and its determinants—real effective exchange rate index, real gross domestic product, money supply, trade openness, and inflation rate—have a stable cointegration connection.The empirical findings offer compelling proof that all explanatory factors—aside from real GDP and inflation rate—have a greater influence on predicting the long- and short-term trends of the trade balance.
Moreover, the most important aspect is the country's degree of trade openness, which both positively and significantly affects the trade balance of the nation over the long and short terms. Although the purpose of this research is to examine the short- and long-term effects of changes in the actual effective exchange rate of the Birr on Ethiopia's trade balance, it is clear that these changes have a major influence both ways. Based on the descriptive and econometric results, even if devaluation has a positive impact on the trade balance of Ethiopia, exchange rate reform is not sufficient to diversify the economy and change the structure of imports; that is, only exchange rate devaluation cannot correct the trade deficit.
Long-term trade balance improvement requires policies other than devaluation, including concentrating on the internal supply side, which creates an environment that is favorable to exportable commodities and goods that may be substituted for imports. Nonetheless, relying exclusively on exchange rate policy may not produce the anticipated improvement in Ethiopia's trade balance. Exchange rate policy remains the most significant tool for improving trade balance.
Even though trade liberalization increases an economy's exposure to global trade, which in turn encourages more resource efficiency, boosts competitiveness, and aids in the nation's economic growth, Ethiopia must develop trade liberalization in line with a solid regulatory structure in order to be safeguarded against financial crises in the nation that establishes trade partnerships with. The analysis's findings show that the money supply is a major factor that negatively affects trade balance. In line with economic theory, a country's money supply expansion results in lower interest rates, lower rates of return on deposits made in local currency, higher prices for domestic goods, and a desire among economic agents to use more imports as replacements, all of which lead to trade deficits. For this reason, the nation needs a solid strategy to stabilize its money supply.
6. Availability of data and materials
The data supporting the findings of this study are available from the corresponding author upon reasonable request.
7. LIST OF ABBREVATION
ADF: Augmented Dickey Fuller
AIC Akaike Information Criterion
ARDL: Autoregressive Distributed Lag
BOP: Balance of Payment
CUSUM: Cumulative Sum
CUSUMSQ: Cumulative Sum Square
ECM: Error correction Model
EPRDF: Ethiopian People Revolutionary Democratic Front
ETB: Ethiopian Birr
IMF: International Monetary Fund
INFR: Inflation Rate
LDCs: least developed countries
MLC: Marshal Learner condition
MS: Money Supply
NBE: National Bank of Ethiopia
OLS: Ordinary Least Square
OPN: Trade Openness
REERI: Real Effective Exchange Rate Index
RGDP: Real Gross Domestic Product
SAP: Structural Adjustment Program
SSA: Sub Saharan African country
TB: Trade Balance
TGE: Transitional government of Ethiopia
8. References
Ababa, A., Washington, D. C., & Monetary, I. (2017). Ethiopia : Impacts of the Birr Devaluation on Inflation1 . 5(25), 2015–2019.
Aleto, A. A. (2018). The Economy Wide Impact of the Devaluation of Ethiopian Currency: A Recursive Dynamic Computable General Equilibrium Approach (Doctoral dissertation, Addis Ababa University Addis Ababa, Ethiopia).
Alekaw. (2012). Determinants of Trade Balance in Ethiopia : an ARDL Cointegration Analysis . By Alekaw Kebede Yeshineh Ethiopian Development Research Institute ( EDRI ) . April 2017. https://doi.org/10.2139/ssrn.2854178
Befekadu D. and Berhanu N. (2000). Annual Report of the Ethiopian economy, EEA, Addis Ababa.
BorenaLencho, D. (2013). The effect of exchange rate movement on trade balance in Ethiopia. Tokyo University..
Devkota, S. C. (2004). Impact of exchange rate change on foreign trade balance in Nepal (No. 0410003).University Library of Munich, Germany.
Fassil, A. (2017). Birr devaluation and its effect on trade balance of Ethiopia: An empirical analysis. Journal of Economics and International Finance , 9(11), 103–119. https://doi.org/10.5897/jeif2017.0864
Petrović, P., &Gligorić, M. (2010). Exchange rate and trade balance: J-curve effect. Panoeconomicus, 57(1), 23-41.
Taye, H. K. (1999). The impact of devaluation on macroeconomic performance: The case of Ethiopia. Journal of Policy Modeling , 21(4), 481-496.
Haile, A. (2008). The Impact of Devaluation on Trade Balance: The Case of Ethiopia (Doctoral dissertation, MA Thesis, University of Oslo, Department of Economics)..
Loto, M. A. (2011). Does devaluation improve the trade balance of Nigeria?(A test of the Marshall-Lerner condition). Journal of Economics and International Finance , 3(11), 624.
Muhammad. (2011). Impact of Devaluation on Trade Balance in Pakistan. Oeconomics of Knowledge, 3(3), 16–25.
NBE. (2018). National Bank of Ethiopia 2017/18 Annual Report . 34(1), 126.
Pesaran et al. (2001). BOUNDS TESTING APPROACHES TO THE ANALYSIS. 326(February 1999), 289–326. https://doi.org/10.1002/jae.616
Salvatore, D. (2004). (Schaum’s) Dominick Salvatore, Eugene Diulio - Schaum’s outline of theory and problems of principles of economics-McGraw-Hill (1995) .
Temesgen. (2016). SCHOOL OF GRADUATE STUDIES THE IMPACT OF REAL EFFECTIVE EXCHANGE RATE ON TRADE Project submitted to the School of Graduate Studies of Addis Ababa . June.
Yilkal. (2012). THE EFFECT OF CURRENCY DEVALUATION ON OUTPUT : THE CASE OF ETHIOPIAN ECONOMY
Zeeshan. (2016). Asian Journal of Economic Modelling IMPACT OF DEVALUATION ON BALANCE OF TRADE : A CASE STUDY OF Contribution / Originality . 4(2), 90–94. https://doi.org/10.18488/journal.8/2016.4.2/8.2.90.94
Frequently asked questions about the Language Preview
What is the main focus of this document?
This document is a language preview focusing on analyzing the impact of devaluation on the trade balance of Ethiopia. It includes an introduction, literature review (theoretical and empirical), research methodology, results and discussion, and conclusion with policy implications.
What is the Structural Adjustment Program (SAP) mentioned in the Introduction?
The Structural Adjustment Program (SAP) was introduced in the 1980s in over 30 sub-Saharan African (SSA) countries. A key component of this program was exchange rate adjustment, aiming to improve economic growth and stabilization by incentivizing the production of tradable goods, especially agricultural ones.
What happened to the Ethiopian Birr (ETB) in 1992?
In October 1992, the Ethiopian Birr (ETB) experienced a significant devaluation of 142%, changing from 2.07 Birr per US dollar to 5 Birr per US dollar.
What were some other devaluations of the ETB after 1992?
After the 1992 devaluation, the government implemented a devaluation effect in September 2010, increasing the rate from 13.6 ETB/USD to 16.3 ETB/USD (a 17% increase). Another official devaluation of the birr against the US dollar by 15% occurred in November 2017, moving the exchange rate from 23.3 ETB per US dollar to 27 ETB per US dollar.
What is the Marshal Lerner Condition (MLC)?
The Marshal Lerner Condition (MLC) states that, all things being equal, a country will improve its current account deficit by devaluing its currency if the sum (in absolute value) of the elasticity of demand for exports and imports is greater than one. Otherwise, the trade balance will not improve.
What is the J-curve effect of devaluation?
The J-curve effect describes a scenario where devaluation initially worsens the trade balance before it improves. This can be due to supply-side constraints and time lags (especially for agricultural commodities) and the inelastic demand for certain imports (like capital goods and raw materials).
What are the main components of the balance of payments?
The balance of payments consists primarily of two accounts: the current account and the capital account. The current account includes visible trade (imports and exports of physical goods) and invisible trade (services, transfer payments, interest, profit, and dividends). The capital account records long-term and short-term capital movements between countries.
What types of data and sources are used in the research methodology?
The study uses annual time-series data from 1984-2018. Data were collected from the National Bank of Ethiopia (NBE).
What methods are used for data analysis?
The study employs both descriptive and econometric styles of analysis. The econometric analysis utilizes the Autoregressive Distributed Lag (ARDL) method and impulse responses to examine the effects of trade openness, real GDP, money supply, inflation rate, and the real effective exchange rate index on Ethiopia's trade balance.
What model specification is used in the research?
The functional model is: TB = f (RGDP, REERI, OPN, MS, INFR), where TB is the trade Balance, RGDP is Real Gross Domestic Product, REERI is Real Effective Exchange Rate index, OPN is trade openness, MS is the money supply, and INFR is the Inflation Rate. The model is specified in a log-linear form: LnTB = β0 + β1LnRGDP + β2LnREERI + β3LnOPN + β4LnMS + β5LnINFR + Ui.
What are the key findings regarding devaluation and Ethiopia's trade balance?
The research findings show that Ethiopia's trade deficit increases after devaluation, suggesting that devaluation alone does not improve the trade balance. The study also shows that devaluation has an impact on trade balance, but is not statistically significant. The paper shows that the money supply and trade openness are important factors affecting the balance of trade. Trade openness has a dominant effect on trade balance both in the short and long run.
Does the research validate the J-curve phenomenon in Ethiopia?
No, the impulse response functions indicate that Ethiopia's response to the J-curve phenomena does not follow the traditional pattern.
Are there any policy implications discussed in the conclusion?
Yes, the conclusion recommends policies beyond devaluation, including focusing on the internal supply side to create a favorable environment for exportable commodities and import substitution. The document concludes that relying exclusively on exchange rate policy may not be sufficient for improving Ethiopia's trade balance. It also discusses the impacts of trade liberalization and money supply on the Ethiopian trade balance.
- Arbeit zitieren
- Teshome Mihret Abate (Autor:in), 2023, Unraveling Trade Dynamics. The Short and Long-Term Impact of Exchange Rate Devaluation on Ethiopia's Trade Balance, München, GRIN Verlag, https://www.grin.com/document/1509869
