Markups and the seasonal cycle

Monthly, non-seasonally adjusted data are used to study markup behavior at the seasonal and business cycle frequency. The shift in focus to the monthl y frequency provides an identification mechanism which helps to rule out se veral competing models of markup variation that would have been plausible a t lower frequencies. Considerable support for collusive pricing is found by examining the monthly behavior of markups in 2-digit U.S. manufacturing in dustries. In particular, the hypothesis that given current demand, markups should increase when demand is expected to increase in the future is confir med by the data.