Alternative measures of marginal cost and inflation in estimations of new Keynesian inflation dynamics

• Published in: Journal of macroeconomics 2008-09, Vol.30 (3), p.928-940
• Main Authors: Gwin, Carl R., VanHoose, David D.
• Format: Article
• Language: English
• Subjects: Cost analysis; Economic models; Estimation; Inflation; Keynesian theory; Macroeconomics; Marginal costs; Model testing; New Keynesian inflation dynamics; Phillips curve; Price stickiness; Sticky prices; Studies
• ISSN: 0164-0704; 1873-152X
• DOI: 10.1016/j.jmacro.2007.07.005

In an important and highly influential paper [Gali, J., Gertler, M., 1999. Inflation dynamics: A structural econometric analysis. Journal of Monetary Economics 44, 195–222] develop and estimate a structural new Keynesian Phillips curve model of inflation dynamics based on the [Calvo, Guillermo, 1983. Staggered prices in a utility maximizing framework. Journal of Monetary Economics 12, 383–398] model of sticky prices. To estimate their model, Gali and Gertler measure inflation as the quarterly percentage change in the GDP deflator, and they consider a specific calculation of the deviation of labor-income share from its mean value as the only proxy measure of aggregate marginal cost. Most other researchers have followed their lead. In this paper, we closely replicate Gali and Gertler’s results and compare these replicated results to those obtained when alternative measures of marginal cost and inflation are utilized. We find that estimations of the structural new Keynesian model of inflation dynamics are not particularly altered by using a marginal-cost measure based on actual industry costs but are sensitive to using a PPI-inflation measure that arguably is more appropriate for testing a theory of firm pricing. Furthermore, we find that utilizing a detrended measure of marginal cost yields results that are at least as plausible as those of Gali and Gertler, yet which generally suggest greater aggregate US price flexibility.