Academic journal article Journal of Money, Credit & Banking

International Interest Rate Linkages in the Term Structure

Academic journal article Journal of Money, Credit & Banking

International Interest Rate Linkages in the Term Structure

Article excerpt

Assets denominated in different currencies but otherwise equivalent are perfect substitutes if those assets generate equal expected returns. This paper considers one particular aspect of the equalization of international returns. The difference between a component of the term premia in the Canadian dollar term structure of interest rates and the same component of the term premia in the term structure of American dollar interest rates is measured at two, ten, and twenty years to maturity. This measurement compares ex ante excess returns earned on Canadian dollar long-term bonds to ex ante excess returns earned on American dollar long-term bonds. This comparison yields substantial evidence finding differences in term structure premia associated with currency of denomination. The results in this paper strongly suggest international bond markets in which term premia differ by currency of denomination for significant periods of time. This provides one important route through which long-term bond returns in Canada and the United States are not equalized.

A very large literature studies the behavior of the foreign exchange premium, the difference in expected return by currency of denomination at a single term to maturity, usually between one month and one year. Theoretical models of the foreign exchange premium are surveyed in Adler and Dumas (1983) or Branson and Henderson (1985). In these models, the sign and magnitude of foreign exchange premia depend on relative asset supplies, preferences of both home and foreign consumers over home and foreign goods as well as risk and, of course, the joint distributions of all variables in a given model. No general conclusions are possible. The resulting theoretical uncertainty has meant measurement of the foreign exchange premium occupies much of the literature; surveys are found in Cumby and Obstfeld (1984) or Levich (1985). Equivalently this paper measures the difference between the same component of the term premia in each term structure and does not consider the specific sources of such differences.(1) Several studies [Hakkio (1981), Hakkio and Leiderman (1986), Baillie and McMahon (1985), and MacDonald and Taylor (1990)] do study the behavior of foreign exchange premia at more than one maturity over the same time period. These studies reject the null hypothesis that, for example, the foreign exchange premium at three months to maturity is the sum of the three expected one-month foreign exchange premia over the same period of time. In this example, even if the foreign exchange premium on a one-month bond is expected to be zero for each of the next three months, it would still be possible that the expected return on a three-month bond in the United States is not equal to the expected return on a three-month bond in Canada. In that case term premia on Canadian dollar and American dollar three-month bonds are not equal. This paper considers differences in term premia by currency of denomination for maturities up to twenty years.

One branch of the literature that tries to consider such differences in the international term structure begins with Porter (1971), continues with Beenstock and Long-bottom (1981), and includes Bisagnano (1983), Krol (1986), Murray and Khemani (1989), and Boothe (1991). These authors regress domestic long-term bond yields or the slope of the domestic currency yield curve (or the first differences of such variables) on their foreign currency counterparts. Large explanatory power in such regressions is interpreted to mean a close link between the domestic and foreign term structure. These results are difficult to interpret because they are unable to present or measure explicitly the role of the expected rate of change in the nominal exchange rate over the term to maturity of the relevant long-term bond.(2)

A second approach to the international term structure uses factor analysis. Boothe and Glassman (1988), in the Canada-U.S. case, and Logue and Sweeney (1984), for a set of European interest rates, find a single factor dominant in explaining the joint term structure of interest rates. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.