2

# Data Description

The criteria for choosing the countries in our sample are data availability for the yields on risk-free government bills, whose returns are the most important determinant for the carry trade. The variables in this data-set include nominal exchange rates expressed in foreign currency units per U.S. dollar; yields on 3-month government bills; the consumer price index (CPI); and the implied volatilities, derived from 3- month FX options for the nominal exchange rate. The exchange rate of the Euro against the U.S. dollar is used to construct the nominal exchange rate of Germany. All data except for the yields on U.K. government bills are obtained from Bloomberg (www.bloomberg.com). The yields on U.K. government bills are collected from the Bank of England. The sample period runs from December 2000 to March 2009, and this range is mainly determined by the data availability for the implied volatilities, which are the main focus of this paper. The nominal exchange rates and the CPIs are logged and the CPIs are seasonally adjusted. All data are observed at monthly frequency and represent the average values of the designated month. Table 1 provides sample mean statistics for the main variables broken down by country.

Several features of the data deserve comment. First, the interest rate differential against the Japanese Yen, often used as the funding currency, is positive (note that the interest rate differential is defined as the difference between the U.S. interest rates and other countries' interest rates), which indicates that there is a profit opportunity in borrowing Yen and investing into U.S. dollar. However due to the depreciation of the U.S. dollar over the sample period, which exactly offsets the interest rate differential, carry traders would not have made any profits. On the other hand, the New Zealand dollar, a typical investing currency, has a sizeable interest rate differential and this profit gain is enforced by the appreciation of the New Zealand dollar thus generating a 1.5% average quarterly return.

Second, although not explicitly reported in Table 1, it is easy to see that there is a positive cross-sectional correlation between changes in exchange rates and carry trade returns. This correlation reinforces the importance of predicting the right direction of exchange rate movements in order to realize the carry trade returns. Third, there is an

7