Uncovered interest parity and the forward premium puzzle : implications for market efficiency and carry trade

Abstract

Uncovered interest parity is a fundamental concept in foreign exchange and implies that the same deposit placed at home or abroad should yield equal returns. The forward premium puzzle refers to a well known empirical failure of the uncovered interest parity relation. Under the forward premium puzzle, currencies that are expected to depreciate, in fact tend to appreciate. This puzzling fact have been interpreted as a failure of the efficient market hypothesis in the foreign exchange market, and has served as a theoretical foundation for earning excess returns from the currency speculation known as carry trade. According to uncovered interest parity, no excess return from such speculation should be possible. This thesis tests for the appearance of the forward premium puzzle in recent data through the conventional approach of regressing the change in spot prices on the forward premium. In addition, two excess return based trading strategies are analyzed as a more practical and direct approach to testing the efficient market hypothesis and uncovered interest parity. My findings regarding the puzzle are consistent with existing literature in the sense that the forward premium puzzle is identified for all eight currency pairs which are included in the regression. However, the estimated coefficients are statistically insignificant, and it is therefore difficult to draw definitive conclusions from the analysis. On the other hand, results from testing the excess return based strategies shows that the apparent presence of the forward premium puzzle not necessarily indicates that there are excess return possibilities in the foreign exchange market. Excess return is only identified for the Norwegian krone and Australian dollar against US dollar parities, but test results remain inconclusive due to violations of the conditions under the ordinary least squares methodology in regression analysis.