An Analysis of the Seasonal Cycle and the Business Cycle

Abstract

Robert Barsky and Jeffrey Miron (1989) revealed the seasonal cycle of the U.S. economy from 1948 to 1985 was characterized by a “bubble-like” expansion in the second and fourth quarters, a “crash-like” contraction in the first quarter, and a mild contraction in the third quarter. We replicate, in part, their seasonal cycle analysis from 1946 to 2001. Our results are largely in line with theirs. Nonetheless, we find the seasonal cycle is not stable and can evolve across time. In particular, the Great Moderation affected both the business cycle and the seasonal cycle. Robert Barsky and Jeffrey Miron also found real aggregates, like the output, move together in the seasonal cycle across broadly defined sectors, similar to a phenomenon observed under the conventional business cycle. They posed a challenge question concerning why “the seasonal and the conventional business cycles are so similar.” To answer their question, we focus on a number of aggregate variables with a recursive application of the HP filter and find that aggregates, such as the GDP, consumption, the S&P 500 Index, and so forth, have a “bubble-like” expansion and a “crash-like” contraction in their cyclical trends in business cycle frequencies. Although preference shifts and production synergy are the two major forces that drive the seasonal cycle, we find the time-varying stochastic discount factor is the main cause of the business cycle and plays a more important role in macroeconomic fluctuations in business cycle frequencies than other factors.