NBB Working Paper, 2022, 428: Identifying latent heterogeneity in productivity

Productivity is influenced by several firm-level factors, often latent. When unexplained, this latent heterogeneity can lead to the mismeasurement of productivity differences between groups of firms. We propose a flexible, semi-parametric extension of current production function estimation techniques using finite mixture models to control for latent firm-specific productivity determinants. We establish the performance of the proposed methodology through a Monte Carl…

Productivity is influenced by several firm-level factors, often latent. When unexplained, this latent heterogeneity can lead to the mismeasurement of productivity differences between groups of firms. We propose a flexible, semi-parametric extension of current production function estimation techniques using finite mixture models to control for latent firm-specific productivity determinants. We establish the performance of the proposed methodology through a Monte Carlo analysis and estimate export premia using firm-level data to demonstrate its empirical applicability. We apply our framework to assess export productivity premia and their robustness with respect to latent heterogeneity. Our results highlight that latent heterogeneity distorts export premia estimates and their contribution to aggregate productivity growth. The proposed approach delivers robust estimates of productivity differences between firm groups, regardless of the availability of productivity determinants in the data.