Estimation of Scale Economies Underlying Growth and Productivity: The Empirical Implications of Data Aggregation

Abstract

Estimation of scale economies underlying growth and productivity patterns is typically based on aggregated data, raising questions about the potential for aggregation biases. This paper provides empirical evidence on the existence and patterns of such biases. We use a cost‐based model to estimate short/long‐run and internal/external scale effects for U.S. manufacturing data at different aggregation levels. Our results suggest that aggregation biases in such a model are not substantive. Also, internal scale economies seem more appropriately represented by the aggregate data, whereas more disaggregated data appears preferable for estimation of external or spillover effects that occur between industries or sectors.