Testing the (S, s) Model
The (S, s) model has enjoyed tremendous popularity over the past decade. It has been employed almost everywhere that discrete adjustment is observed. Today microeconomic rigidities are seen as an important influence on aggregate dynamics. In this paper we quickly characterize the microeconomic evidence for the model. To narrow the scope of our discussion, we will focus our attention on real variables, and we will comment on price inertia where appropriate. We conclude that, in spite of its popularity, the evidence for the importance of the (S, s) adjustment is surprisingly weak. We argue that discrete adjustment is, in and of itself, of little macroeconomic interest. To be important these frictions must coordinate agents to act together, thereby exacerbating deviations from the neoclassical benchmark. To date there have been few attempts at empirically identifying such interactions. In the last section, we present some results of our own. We test one of the main implications of (S, s) adjustment, that a greater variance in the forcing process leads to more frequent adjustment. Using data on automobiles from the Consumer Expenditure Survey, we find that more variable income leads to less frequent adjustment. We speculate that this correlation is indicative of a link between discrete adjustment and imperfect capital markets. This interaction could provide an important role for (S, s) frictions.