Aggregate Productivity Growth: Lessons from Microeconomic Evidence

TLDR

Empirical studies reveal striking patterns in establishment‑level reallocation and productivity, yet findings on its contribution to aggregate productivity growth vary widely. The study investigates how reallocation of activity across individual producers drives aggregate productivity growth, focusing on selected service‑sector industries. The analysis focuses on 4‑digit industries within the 3‑digit automobile repair sector, examining reallocation patterns in selected service‑sector industries. Our review and sensitivity analysis show that reallocation significantly contributes to industry‑level productivity growth, with net entry disproportionately boosting growth by displacing less productive exiting plants, but the magnitude depends on decomposition methodology, and learning and selection effects also play important roles.

Abstract

In this study we focus on the role of the reallocation of activity across individual producers for aggregate productivity growth. A growing body of empirical analysis yields striking patterns in the behavior of establishment-level reallocation and productivity. Nevertheless, a review of existing studies yields a wide range of findings regarding the contribution of reallocation to aggregate productivity growth. Through our review of existing studies and our own sensitivity analysis, we find that reallocation plays a significant role in the changes in productivity growth at the industry level and that the impact of net entry is disproportionate since entering plants tend to displace less productive exiting plants, even after controlling for overall average growth in productivity. However, an important conclusion of our sensitivity analysis is that the quantitative contribution of reallocation to the aggregate change in productivity is sensitive to the decomposition methodology employed. Our findings also confirm and extend others in the literature that indicate that both learning and selection effects are important in this context. A novel aspect of our analysis is that we have examined the role of reallocation for aggregate productivity growth to a selected set of service sector industries. Our analysis considers the 4-digit industries that form the 3-digit industry automobile repair shops. We found tremendous churning in this industry with extremely large rates of entry and exit. Moreover, we found that productivity growth in the industry is dominated establishment data at Census, the results are quite striking and clearly call for further analysis.