Degree Year


Document Type


Degree Name

Bachelor of Arts




Barbara Craig


J-curve, Latin American, Economy, Depretiation, Peru, Argentina


In this paper, I test whether quarterly data either supports or disproves the existence of such a J-curve in two Latin American countries -- Argentina and Peru. The time periods studied will be: for Peru, 1979 quarter 1 to 1991 quarter 2 and for Argentina, 1977 quarter 1 to 1990 quarter 4. Before completing the empirical part of this paper, I expected to find, if any, only a short deterioration in the trade balance following a depretiation of the real exchange rate in these two countries. With prices as well as other economic indicators changing daily, the reaction times of economic actors who deal with Latin America, whether from the inside or the outside, have grown ever shorter. Many contracts are short-term or are indexed so the real terms of the contract will not change as the nominal indicators do. Short reaction times should mean that the import/export market will adjust quickly to any change in the real exchange rate or other indicators. Contracts will be renegotiated quickly because of their short-term nature. This quick adjustment is not only important for the first possible cause of a J-curve, namely stickiness caused by contracts outstanding, but could also lead to a prediction about the second factor involved, that of trade elasticities. Quick renegotiations of contracts as well as a general attitude among actors that are expectant of quick changes would seem to favor a high elasticity of demand. For example if a foreign importer suddenly sees the price of imports from Peru go down, she may be more anxious to take advantage of this fall in price than if the imports were from a more stable economy. Since in the Peruvian economy, things are likely to change quickly, the foreign importer will want to take advantage of the low prices before they are eradicated by, for example, a sudden rise in inflation in Peru. Because of this expectedly short duration of a J-curve, should one exist, I define a significant J-curve as anything longer than two quarters. In other words, if the trade balance does not start to improve before the second lagged quarter after the depretiation (after the quarter of the depretiation and the first lagged quarter), a J-curve exists. Any deterioration shorter than two quarters can be interpreted only as the result of the impossibility of instantaneous reactions on the part of the actors. While this may be a brief J-curve by any standards, I think this definition is appropriate for a study of Latin American countries.

Included in

Economics Commons