The Transmission of U.S. Election Cycles to International Stock Returns

Abstract

This paper examines the international pervasiveness and importance of the previously uncovered four-year U.S. election cycle whereby U.S. stock returns are significantly lower, and negative, in year 2 following U.S. presidential election relative to years 1, 3 and 4. All eighteen countries examined over the 1957 to 1996 time period possess lower local currency stock market capital gains returns in year 2 (−0.66%) relative to the average capital gains of years 1, 3 and 4 (11.68%). These predominately lower year 2 returns are shown to be robust in conditional expected return regressions which include both local macroeconomic variables as well as U.S. macroeconomic, fiscal and monetary policy variables. In addition, we find that the U.S. dollar trends to depreciate more in year 2 of the election cycle. We conclude that the U.S. election cycle variable is either proxying for information variables not included in our model, or the U.S. election cycle variable is capturing some form of U.S. and international market sentiment. That is, the U.S. election cycle may be an important nondiversifiable political factor in the determination of international conditional expected stock returns.