Devaluation: The Wrong Bitter Pill – By Tesfaye Kidane
In his June 8, 2010 budgetary speech to the parliament, Minister Sufian Ahmed indicated that the 2010/11 budget is prepared under the assumption that Birr will depreciate by 5% during the 2010/11 budget year. However, on September 1, 2010, in less than two months since the onset of the budget year, the national bank devalued birr against dollar by 20%. Though the National Bank of Ethiopia (NBE) officials gave some justification for the need to devalue birr, why the magnitude of the devaluation exceeded Ministry of Finance and Economic Development’s (MoFED) projected depreciation by 300% (5% vs. 20%) remains unanswered.
This opens up a wide space for conjecture.
The likely explanations include, first, it may be because something has gone terribly wrong since Minister Sufian’s June 8 speech, requiring devaluation of birr by four times higher than what is projected to be at the end of the year but done at the beginning of the year. Second, MoFED did a bad job in its projection because of lack of capacity or a good insight of the economy. Third, the policy makers do not have a coherent framework for policy making to the extent that they are ignorant of their projected targets and plans based on such targets. Four, the future trajectory of the economy has been changed dramatically (but not officially announced) as a result of the new five year plan that may require significant devaluation.
According to Addis Fortune’s news report (Vol. 11, No. 542, September 19, 2010), PM Meles Zenawi and the state minister Arkebe Oqubay gave some explanations for the need to depreciate birr significantly during EPRDF’s eighth convention. PM Meles contends that birr was overvalued due to high inflation over the last couple of years; and maintaining the right value of birr (equilibrium real exchange rate) is necessary to encourage exports and maintain stability in the coming five years. On a similar line of argument, Arkebe listed three rationales for the devaluation: to encourage exports, discourage imports, and attract more remittances. Arkebe also added that devaluation is one of the instruments to achieve the 10 billion USD in export earnings targeted in the growth and transformation plan. The NBE officials also alluded to the same.
The policy makers’ explanations above seem to suggest that devaluation is necessary to achieve the targets in the growth and transformation plan. If this is really the case, it may imply that the plan is drafted after the 2010/11 budget is ratified and hence it is not integrated with the budget. This casts doubt on the government’s capacity in long term planning. Letting this as it may, the important question is whether the justifications for the devaluation have any merit or not.
Theoretically the justifications for devaluation are sound. In practice, the justifications may lose their merits as the efficacy of devaluation in achieving those objectives depends on the structure of the economy. So, let’s first evaluate PM Meles’ argument in Ethiopian context. Meles argued that birr is overvalued and hence devaluation is required to correct the overvaluation. The PM is right on the overvaluation part of the story, and the empirical evidence supports this claim (see, for instance, IMF Country Report No. 09/296). However, devaluation does not necessarily solve the overvaluation of birr as the PM asserts. I will elaborate why this is so.
In the economics jargon, there are two types of exchange rates: nominal (NER) and real exchange rates (RER). Nominal exchange rate is the price of one currency in terms of another. It’s usually expressed as the domestic price of the foreign currency i.e., 1 USD=16.35 Birr. But the nominal exchange rate isn’t the whole story. The person or firm who buys another currency is interested in what can be bought with it. That’s where the RER comes in. RER is the purchasing power of two currencies relative to one another. While two currencies may have a certain exchange rate on the foreign exchange market, this does not mean that goods and services purchased with one currency cost the equivalent amounts in another currency. This is due to different inflation rates with different currencies. Real exchange rates are thus calculated as a nominal exchange rate adjusted for the different rates of inflation between the two currencies. So, what really matters for policy is the value of RER. The important question is whether depreciation of nominal exchange rate leads to depreciation of RER or not.
As stated above, RER is calculated as a nominal exchange rate adjusted for the different rates of inflation between the two currencies. Let’s note out different scenarios of the RER movement.
1.For a constant nominal exchange rate if domestic prices increase more than foreign prices, our real exchange rate would appreciate (overvalued). On the other hand, if the rate of increase domestic prices is lower than that of foreign prices, our real exchange rate would depreciate.
2.For constant price levels, depreciation of nominal exchange rate would lead to depreciation of real exchange rate, and vice versa.
3.For variable nominal exchange rate and price levels, depreciation of nominal exchange rate would lead to depreciation of real exchange rate if the rise in domestic price is lower than the sum of the rise in foreign price plus the rate of depreciation of the nominal exchange rate. For instance, suppose domestic and foreign prices rise by 30% and 5%, respectively, and the nominal exchange rate depreciate by 20%. In this case, our RER would appreciate because the combined effect of the nominal exchange rate depreciation and the rise in foreign price is not as high as that of the rise in domestic price. As a result, our goods and services will be more expensive.
Turning back to the Ethiopian case, the condition in (3) should be fulfilled to bring about depreciation of the RER. However, the existing empirical evidence suggests that the conditions in (3) may not hold true in Ethiopia due to the high response of the domestic inflation rate for nominal exchange rate depreciation. There are two likely routes through which nominal exchange rate depreciation induces inflation. First, through creating supply shortages as depreciation discourages domestic production by raising cost of production, which would eventually lead to soaring prices. Second, the effect of the exchange rate depreciation may be directly transferred into prices by producers/traders.
A recent study by Kibrom (2010)1 found that, other things remaining constant, one percent depreciation in birr would result in a 1.94% increase in the domestic prices. That is, the current depreciation of 20% will push the domestic prices up by around 39%. Given a foreign inflation rate of 5 – 9%, our RER could rather appreciate by 10- 14%. In this case, depreciation of nominal exchange rate would exacerbate the problem it is supposed to solve: overvaluation of real exchange rate.
The PM argument, that devaluation is “a bitter pill that the economy has to swallow and move into the next five years with stability” (see Addis Fortune, Vol. 11, No. 542, September 19, 2010), loses its merit once the empirical evidences on the inflationary impact of nominal depreciation are factored in. The PM seems to shred the inflationary effect of the devaluation as a temporary phenomenon and he argued farther that inflation will be controlled through tight monetary policy. In addition, the PM emphasized the low inflation outlook due to the expected surplus agricultural output in the coming harvest season, which is expected to dampen the inflationary pressure. Sadly, the PM is completely wrong on these points.
First, tight monetary policy is effective to control demand push inflation. However, nominal exchange rate depreciation leads to cost push inflation that cannot be managed through tight monetary policy. Second, devaluation erodes the real value of domestic producers’ working capital as imported raw materials become expensive. Producers demand additional financing in the forms of bank loan. However, with tight monetary policy, producers’ demand for loan may not be satisfied, leading to downsizing of production and hence a fall in output. This would in turn fuel inflation and further appreciation of the RER. Third, Kibrom’s study shows that the inflationary effect of nominal exchange rate depreciation dominates that of surplus output. That is, to absorb the inflationary effect of the depreciation, a 1293% (or around 13 fold) increase in surplus output is required. This is further aggravated by the rise in the budget deficit that may follow the rise in the cost of imports for the government sector itself which invariably is financed by monetization.
The more disturbing issue is that even if the nominal devaluation leads to depreciation of the RER against all odds, it is not necessarily true that our trade balance (Imports – Exports) will improve following the depreciation. For the trade balance to improve, both exports and imports should be highly responsive to the change in exchange rate, which is not the case in Ethiopia. Our imports are mainly basic necessities such as fuel, raw materials, spare parts, and medicine that cannot be cut significantly even if their prices rise in birr terms. On the other hand, our exports are mainly primary commodities and their demand is less responsive for change in prices due to exchange rate depreciation. In addition, their supply is also less responsive because of time lag in expanding their production. In short our exports are supply constrained while our imports are not price sensitive. This is in addition to the problem of Ethiopia’s trade imbalance in the context of which what economists call the ‘Marshall-Learner’ condition is not effective even if both import and exports are responsive. One need not expect attracting remittances (or competing with the parallel market) using this policy either as the government claims because the demand served by the parallel market cannot be served by the formal banking sector in Ethiopia – and hence the rational persistence of the parallel market premium.
So, what is the alternative for devaluation?
A viable alternative is to go beyond exchange rate and examine the structural factors that constrain the export sector. One study, based on a survey of exporting firms in Ethiopia, showed that the most important impediments for the operation of exporting firms are access to land, customs and tax regulations, tax rates and administration, and corruption. In addition, the exporting firms exhibit considerable level of inefficiency with an average capacity utilization rate of around 55%. Lack of working capital, shortage of raw material, technological constraints, and insufficient demand stand as the most important reasons for the under capacity operation. Such structural (invariably supply) problems may be much more important than the overvalued exchange rate. The policy direction should thus be geared towards correcting these structural problems.
From the import side, it is necessary to look into the demand structure of our imports. Rather than an overvalued exchange rate, the rising government capital expenditure, which has high import content, may be the most important driving force of imports and hence the rising trade deficit. When this is the case, devaluation increases government budget deficit, which may be inflationary or debt creating depending on its financing. If inflationary financing is used, the RER may appreciate due to high domestic inflation rate. As such the nominal depreciation may miss its target, leading to undesirable macroeconomic instability with a continuous inflation-devaluation spiral. The latter not only discourage investment but also leads to government policy credibility problems as well as dwarfing of poverty reduction efforts through low investment and growth as well as high inflation The appropriate policy stance is to limit government’s import demand to be in line with the targeted level of trade balance.
Unfortunately, against all the empirical evidences, the policy makers appear to be convinced that devaluation is a panacea for improving the trade balance. Probably, this may be due to the fact that such policies are not informed by rigorous economic studies and existing empirical evidences. If this is also the case in the crafting of the growth and transformation plan, which seems highly likely, there will be a clear and present danger in terms of macroeconomic instability in the coming five years. It is not, however, too late for the policy makers to make informed decisions based on rigorous studies as long as they have the courage to do so.
1 KIBROM, T. (2010): The Sources of Inflation in Ethiopia: Modeling Price Formation and Dynamics. Saarbrücken, Germany: Lambert Academic Publishing.