Utilization-Adjusted TFP Across Countries: Measurement and Implications for International Comovement

This paper develops estimates of TFP growth adjusted for movements in unobserved factor utilization for a panel of 29 countries and up to 37 years. When factor utilization changes are unobserved, the commonly used Solow residual mismeasures actual changes in TFP. We use a general equilibrium dynamic multi-country multi-sector model featuring variable factor utilization to derive a production function estimating equation that corrects for unobserved factor usage. We compare the properties of utilization-adjusted TFP series to the standard Solow residual, and discuss the implications for international business cycle comovement generated by technology shocks. Unlike the Solow residual, utilization-adjusted TFP is virtually uncorrelated across countries, and as a result its direct contribution to GDP comovement is negligible. A general equilibrium model calibrated to the observed levels of international trade cannot generate much comovement through propagation of these TFP shocks.