Capital Input and the Sources of U.S. Economic Growth across Industries: A Comparison of Alternative Approaches

Paper Abstract: While there is broad consensus in the growth accounting literature that capital service measures should reflect the productive capital stock and the implicit annual user cost of capital, conceptual and implementation issues persist. This paper uses detailed source data from the Bureau of Labor Statistics (BLS), which publishes the official productivity statistics, and the Bureau of Economic Analysis (BEA), which publishes the detailed fixed asset data and income measures, to empirically assess two alternative measures of capital services that span many of the conceptual and implementation issues. The first approach uses the methodology of the official capital services estimates that are produced by the BLS. This methodology assumes hyperbolic depreciation and uses either the internal or external (when implicit rental rates are negative) (Harper, Nakamura, Zhang 2012). The second approach follows the methodology in (Jorgenson, Ho, and Stiroh 2007), and assumes geometric depreciation and asset-specific and industry fixes for negative implicit user costs. In addition to comparing the capital measures themselves, we evaluate the broader impact of the difference in capital methodologies by re estimating the BEABLS integrated industry level production account in Rosenthal, Russell, Samuels, Strassner, and Usher (2014).