نوسانات نرخ ارز در آمریکای لاتین و تاثیر آن بر تجارت خارجی

This paper investigates empirically the impact of real exchange-rate volatility on the export flows of eight Latin American countries over the quarterly period 1973–2004. Estimates of the cointegrating relations are obtained using different cointegration techniques. Estimates of the short-run dynamics are obtained for each country utilizing the error-correction technique. The major results show that increases in the volatility of the real effective exchange rate, approximating exchange-rate uncertainty, exert a significant negative effect upon export demand in both the short-run and the long-run in each of the eight Latin American countries. These effects may result in significant reallocation of resources by market participants.

The impact of increased exchange rate variability on foreign trade has been investigated in a large number of empirical and theoretical studies.1 The issue is particularly important for countries that switched from a fixed to a flexible exchange rate regime due to the higher degree of variability associated with flexible exchange rates. While many Latin American countries have moved to a flexible exchange rate regime at some point in the recent past2, it is surprising that there are very few studies that analyze the relationship between exchange rate variability and foreign trade for Latin American countries.3 The purpose of this paper is to close this gap and provide estimates of the short- and long-run impact of exchange rate variability on export flows for eight Latin American economies. In estimating these effects, we follow the approach introduced by Arize, Osang, and Slottje (2000) who examine the impact of exchange-rate volatility on the export flows for thirteen LDCs using both cointegration and error-correction techniques. Based on that approach, we find that the variability of the real exchange rate had a negative effect on export demand for all Latin American countries in our sample, both in the short and the long run. This result is quite surprising given that most countries in this study are middle-income economies according to World Bank classification and thus should have forward markets that would allow traders to hedge transaction exposure. But, as our results show, even fairly developed economies may not be able to completely insulate real economic flows from the fluctuations in international financial markets, and, as a result, these countries have to bear the negative consequences of such fluctuations. Our results are on the whole consistent with the scant evidence on the relationship between exchange rate variability and export behavior of Latin American countries obtained by previous studies. Coes (1981) uses a log-level specification to examine Brazilian exports (annual data for 1965–1974) and concludes that a significant reduction in exchange-rate uncertainty in the country's economy during the crawling-peg era had a positive effect on the country's exports after the crawling peg was adopted in 1968. The study by Brada and Mendez (1988) includes 15 Latin American countries and covers the 1973–1977 period. While their conclusion is similar to ours, namely that exchange-rate uncertainty inhibits bilateral exports, they do not use a measure of exchange-rate volatility, but instead rely on a various dummy variables to account for the effects fixed versus flexible exchange rate regimes. Caballero and Corbo (1989) use a Koyck-type model and real bilateral exchange-rate volatility measure to estimate an export demand equation for six countries, among them Chile, Colombia, and Peru. They conclude that there is a strong negative effect of real exchange-rate uncertainty on the exports of all these countries. Furthermore, the empirical results derived in this paper are also consistent with recent studies showing a significant negative (long-run) impact of exchange-rate volatility on export flows for developing countries outside of Latin America (e.g., Arize et al., 2000 and Bahmani-Oskooee, 2002).4 The remainder of the paper is organized as follows. In Section 2, we examine the specification of our empirical model followed by a discussion of econometric methodology issues. Data sources and variable definitions are described in Section 3. In Section 4, we discuss the empirical results for the eight countries. Conclusions are drawn in the last section.

Our results concerning the effects of exchange-rate volatility on export flows suggest that there is a negative and statistically significant long-run relationship between export flows and exchange-rate volatility in each of the eight Latin American countries. In addition, we also find evidence for a negative short-run effect of exchange-rate volatility on export flows in all Latin American countries studied. Our results have several policy implications. First, and foremost, economic policies that aim to stabilize the exchange rate (of which the establishment of a common currency area would be the most pronounced) are likely to increase the volume of trade among Latin American countries. Second, attempts to extend the North American Free Trade Agreement southward may find little support from Latin American countries, if the potential welfare gains through trade expansion are called into question through reduction in trade due to increased exchange rate variability. Finally, the intended positive effect of a trade liberalization policy may not only be doomed by a variable exchange rate but could also precipitate a balance-of-payments crisis. It is worth noting that the approach we have used here to investigate the relationship between export flows and exchange-rate volatility for eight Latin American countries is characterized by a number of important econometric features typically not found in other empirical studies on this topic. First, the data set for each country covers the current floating exchange-rate era and thus allows us to address the stability over time of the estimated dynamic models during this period. This is essential for appropriate policy conclusions to be inferred from the estimated results. Second, by considering an error correction model, this study provides estimates of the speed of adjustment or the average time lag for adjustment of exports to changes in the explanatory variables as well as the short-run effects of exchange-rate volatility on exports. Third, each estimated model satisfies several recently developed econometric tests in the analysis of time-series data for issues such as cointegration, stationarity, specification errors, residual autocorrelation, heteroskedasticity, residual normality, and structural stability.