We examine the ability of existing and new factor models to explain the comovements of G10- currency changes, measured using the novel concept of “currency baskets”, representing the overall movement of a particular currency. Using a clustering technique, we find a clear two-block structure in currency comovements with the first block containing mostly the dollar currencies, and the other the European currencies. A factor model incorporating this “clustering” factor and two additional factors, a commodity currency factor and a “world” factor based on trading volumes, fits currency basket correlations much better than extant factors, such as value and carry, do. In particular, it explains on average about 60% of currency variation and generates a root mean squared error relative to sample correlations of only 0.11. The model also fits comovements in emerging market currencies well. Economically, the correlations between currency baskets underlying the factor structure are inversely related to the physical distances between countries. The factor structure is also related to the exposure of the corresponding pricing kernels with respect to the global pricing kernel and is apparent in cross-country retail sales growth data.
We would like to thank Pasquale Della Corte, Bruno Gerard, Robert Hodrick, Peter Nyberg, Angelo Ranaldo (discussant), Dagfinn Rime, Florent Rouxelin (discussant), and Hakon Tretvoll and the participants at Norges Bank’s workshop on Financial Determinants of Foreign Exchange Rates, the University of Missouri’s finance seminar, the Sabanci Center in Istanbul, and the European Financial Management Association meeting in Milan for helpful comments. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.