By Arnold C. Harberger, UCLA
This paper is based on a 2-week visit to Ethiopia in September 2010, under the auspices of USAID. I have framed it as an observer’s report, with the purpose of conveying, especially to younger USAID economists, some aspects of how one incorporates economic analysis into the process. The paper should be read keeping this underlying purpose in mind.
The single overarching guideline to good economic observation is to work within, and to constantly keep in mind, a sound theoretical framework. Dealing piecemeal with each separate element that one observes, and coming up with a separate plausible ad hoc explanation for each one, is the opposite, indeed the enemy of professionalism. What makes economic analysis such a powerful tool is its capacity to fit the separate pieces one observes into a coherent whole.
An Expectation of Dutch Disease
The particular feature of the Ethiopian economy that first struck my mind was the fantastic excess of imports over exports. In fact, exports in recent years have accounted for less than a third of the annual flow of foreign exchange available to the country. Starting right there, it would be wrong for the observer to conclude that this “trade imbalance” was a problem that somehow was in urgent need of correction. Rather, this imbalance is the consequence of two separate factors — large inflows of foreign aid and emigrant remittances, each of which provides the country with more foreign exchange than its exports. In recent years these non-export flows of foreign exchange have been supplemented by significant and increasing amounts of foreign direct investment.
All this places Ethiopia in that group of countries whose economies have had to adapt to huge inflows of foreign exchange. This syndrome, as is well-known to economists, typically leads to the phenomenon that we call “Dutch Disease” — a sharp appreciation of the domestic currency, rendering the dollar quite cheap, in real terms. Obviously, this is a natural, supply and demand result. When the supply of anything experiences a large and rapid expansion, it is perfectly natural that its price (in real terms) should fall. If the nominal exchange rate is fixed, the fall in the real price of the dollar takes place via a rise in the internal price level, predominantly of nontradable items (house prices, rents, local wage rates, restaurant meals, local transportation costs, etc.)
Tradables prices are held in check by the fixity of the exchange rate. With a flexible exchange rate it would be possible for the full adjustment to be reflected in a fall in the local-currency price of the dollar (the nominal exchange rate). But this rarely occurs in practice. Often, the authorities in a flexible-exchange-rate country will take steps to keep the nominal price of tradables from falling, so that in those cases as well, the appreciation of the real exchange rate comes from a rise in the internal price level. Sometimes, the adjustment is split, coming partly from a fall in the nominal exchange rate and partly from a rise in the internal price level.
The phenomenon of Dutch Disease has been widely observed in countries experiencing massive inflows of foreign exchange. Notable cases include The Netherlands (at the time of the discovery of oil and natural gas under the North Sea, from which Dutch Disease got its name), El Salvador (from foreign aid and remittances), Russia in recent years (from oil and gas discoveries), several Latin American countries in the late 1970s and early 1980s (from large inflows of capital), others in the 1960s and later (from a boom in the world price of coffee and other export commodities).
Thus there were good reasons to expect a similar result in Ethiopia. But — and this was our first surprise — the data showed no evidence of the sort of real appreciation of the currency that we associate with Dutch Disease. Moreover, day-to-day observation of prices in the local economy tended to confirm the judgment that Ethiopia was not suffering from the case of Dutch Disease that we had good reason to expect that we would find… [continue reading]