This paper studies the driving forces of business cycles within and across G-7 countries over the period 1961 to 2005. We use the term driving forces in the same manner it is employed in dynamic rational expectations models of the international business cycle variables taken to be exogenous in the model employed by the researcher. Most published work focuses on a single driving variable in isolation with the world divided into home and foreign locations. The most common driving variables are total factor productivity and monetary and fiscal policy shocks.
The reason quantitative equilibrium models have focused on small dimensional models in both the number of countries and number of driving variables is tractability. Moreover, there are two dimensions along which tractability isa concern. The first is related to the ability of the researcher or the researchers audience to sort out an expanding list of theoretical interactions as the dimensions of the model increase. To some extent this is alleviated by tracing out the time profile reaction of themacroeconomy to one shock in one location at a time, the comparative statics counterpart to an impulse response function intime series econometrics.
The second constraint is empirical. Even if the researcher decides to calibrate preferences and technology using existing micro studies, thereby economizing on the number of parameters to estimate using macroeconomic series, the remaining parameters necessary to capture the dynamic covariance structure of the shocks grows rapidly with the dimension of the model. This is a consequence of the need—in dynamic rational expectations models—to develop forecasts of the future evolution of exogenous variables and their interdependence.
We mitigate the curse of dimensionality by employing dynamic factor models to characterize the stochastic processes for both endogenous variables and exogenous variables. We follow the Bayesian Factor Model Approach of Otrok andWhiteman (1998), which was first employed to study international business cycles by Kose, Otrokand Whiteman (2003). Each variable is modelled as the sum of three unobserved factors: a common factor, a nation-specificfacto rand an idiosyncratic factor. Aside from parsimony, this approach lends itself to the interpretation of shocks as arising from common or nation-specific sources; a distinction which is central for a number of international business cycle theories.
Download
What Arethe Driving Forces of International Business Cycles?
