Measuring productivity and absorptive capacity evolution

We develop a new way to estimate cross-country production functions which allows us to parametrize unobserved non-factor inputs (total factor productivity) as a global technology process combined with country-specific time-varying absorptive capacity. The advantage of our approach is that we do not need to adopt proxies for absorptive capacity such as investments in research and development (R&D) or human capital, or specify explicit channels through which global technology can transfer to individual countries, such as trade, foreign direct investment (FDI) or migration: we provide an endogenously-created index for relative absorptive capacity which is easy to interpret and encompasses potential proxies and channels. Our implementation adopts an unobserved component model and uses a Bayesian Markov Chain Monte Carlo (MCMC) algorithm to obtain posterior estimates for all model parameters. This contribution to empirical methodology allows researchers to employ widely-available data for factor inputs (capital, labor) and GDP or value-added in order to arrive at policy-relevant insights for industrial and innovation policy. Applying our methodology to a panel of 31 advanced economies we chart the dynamic evolution of global TFP and country-specific absorptive capacity and then demonstrate the lose relationship between our estimates and salient indicators of growth-enhancing economic policy.