Band 4, Heft 2

We investigate business cycle dynamics for 26 countries, estimating a set of nonlinear models for real GDP where appropriate. We find a great deal of heterogeneity in the dynamics of real output growth across countries. At the same time, a common feature is the asymmetric response of real GDP to shocks and the tendency for negative real GDP shocks to be less persistent than positive real GDP shocks.

Several recent studies show that seasonal variation and cyclical variation in unemployment are correlated. A common finding is that seasonality tends to differ across the business cycle stages of recessions and expansions. Since seasonal adjustment methods assume that the two sources of variation can somehow be separated, the present study examines the impact of seasonal adjustment on the analysis of cyclical patterns. Seasonally adjusted quarterly unemployment data for five G-7 countries are modeled by a Smooth Transition Autoregression (STAR), whereas the corresponding unadjusted data are modeled by a so-called Seasonal STAR (SEASTAR). A comparison of the implied estimated peaks and troughs shows that there is substantial agreement on the business cycle chronologies, albeit that for seasonally adjusted data, recessionary periods tend to last longer.

There is now a great deal of empirical evidence that business cycle fluctuations contain asymmetries. I focus on a theoretical model intended to capture the nonlinear behavior of aggregate output following a large negative shock. Nonlinearity introduced by Bayesian updating and an information externality produces an economy in which the response to large negative shocks is an increase in future output. The expansionary effect is produced by the negative shock imparting information about what not to do.