Real Wage Rate And Productivity Relationship In The Declining U.S. Steel Industry

Abstract

This study examines the relationship between the real wage rate and productivity in the U.S. steel industry in the critical period of 1963-1988. This period witnessed a declining steel output and employment, increasing productivity, and a slight increasing real wage rate. The severity of the decline was felt in the 1980s. The popular explanation focuses on the nominal wage rate relative to productivity (non-nominal value). The study is based on high-frequency monthly data set on output, employment, productivity, wage rate, factor prices, and national unemployment rate. Also control factors are constructed for the steel import protection and non-protection regimes. Some econometric modeling issues are addressed. Recognizing that productivity is stochastic and is potentially an endogenous variable, it is instrumented with a set of productivity-related variables including controls for various steel protection and non-protection regimes. Third, the wage in the industry is modeled as a function of exogenous productivity, price of steel products, national unemployment rate, and real interest rate. Serial correlation characterizes the data, and this is corrected with inter-temporal effect of the real wage rate, and with a differencing model. The main results of the study are threefold. First, OLS and Instrumental Variable (IV) estimates show that productivity is the key variable for explaining the real wage rate. Second, like in the literature, the study finds that heavy and autonomous capitalization has an impact on the rising productivity. Third, the study identifies an inter-temporal high real wage rate as the driving factor for explaining the short run real wage rate. These results are somewhat sensitive across specifications.