Exchange Rate Risk and Export Performance in Pakistan: An ARDL Bounds-Testing Analysis
Although the behavior of the exchange rate has been seen as an important factor in export performance of the small open developing economies, there is conflicting evidence in the literature and the available studies are concentrated on shorter sample periods and more recent data for Pakistan. In this study, the relationship between the real exchange rate and exports of Pakistan is examined using a long period time series (1986–2015) along with gross domestic product (GDP), interest rate and import-to-GDP ratio. Secondary data were collected annually from World Bank database, Ministry of Finance of Pakistan, Federal Bureau of Statistics (now Pakistan Bureau of Statistics) and Economic Survey of Pakistan. The Augmented Dickey-Fuller (ADF) unit-root test supported the mixed order of integration result (I(0) and I(1)) and thus the use of the bounds-testing approach (Autoregressive Distributed Lag (ARDL)). The Pesaran, Shin and Smith (2001) bounds test suggested that there was a long run relationship between the variables at the 5 percent level and the Error Correction Model (ECM) showed that the speed of adjustment was slightly negative, but not statistically significant at 5 percent. The residuals were normally distributed and there was no serial autocorrelation for the diagnostic checks. The overall model diagnosis showed the F-statistic insignificant at conventional levels which led to the conclusion that the results should be interpreted as suggestive rather than conclusive. Overall model diagnosis indicated that F-statistic was not statistically significant at conventional levels, so the results should be interpreted as suggestive rather than conclusive. The study does not build a formal statistical volatility index (such as a GARCH index), so it does not address econometrically speaking how volatile the exchange rate is, but rather the link between the level/movement of the exchange rate and exports. Implications for exchange rate policy and export promotion in the case of Pakistan are discussed, and methodological limitations and directions for further research are noted.