Agricultural Trade Bias in Exchange Rate Volatility Effect Estimation: An Application of Meta-Regression Analysis

Econometric estimates of the effect of exchange rate volatility on international trade are mixed and often wildly divergent, resulting in an ambiguous average effect. It is hypothesized that agricultural trade in particular is more sensitive to exchange rate volatility resulting in more consistent effect estimates. Meta-regression analysis is performed on a sample of 351 econometric exchange rate effect estimates, controlling for study specific characteristics including identification of agricultural sector studies. Results indicate that agricultural trade studies report more consistently significant and negative estimates than non-agriculture or aggregate trade studies, revealing a less ambiguous effect in this sector.