Productivity Measurement, Model Averaging, and World Trends in Growth and Inequality


Our paper provides new methods to robustify productivity growth measurement by utilizing various economic theories explaining economic growth and productivity and the econometric model generated by that particular theory. We utilize the World Productivity Database from the UNIDO to analyze productivity during the period 1960-2010 for OECD countries. We focus on three competing models from the stochastic frontier literature, Cornwell, Schmidt, and Sickles (1990), Kumbhakar (1990) and Battese and Coelli (1992) to estimate productivity growth and its decomposition into technical change and efficiency change and utilize methods due to Hansen (2010) to construct optimal weights in order to model average the results from these three approaches.