Aggregate uncertainty and sectoral productivity growth: The role of credit constraints. (November 2018)

• Record Type: Journal Article
• Authors: Choi, Sangyup; Furceri, Davide; Huang, Yi; Loungani, Prakash
• Is Part Of: Journal of international money and finance. Volume 88(2018)
• Page Start: 314
• Page End: 330
• Publication Date: 2018-11
• Subjects: Productivity growth -- Financial dependence -- Uncertainty -- Information and communication technology investment; E22 -- F43 -- O30 -- O47; International finance -- Periodicals; Foreign exchange -- Periodicals; Finances internationales -- Périodiques; Change -- Périodiques; Foreign exchange; International finance; Periodicals; 332.04205
• DOI: 10.1016/j.jimonfin.2017.07.016
• Languages: English
• ISSNs: 0261-5606

Abstract

Highlights: We study the effect of aggregate uncertainty shocks on sectoral productivity. The effect is stronger in industries that depend heavily on external finance. Uncertainty induces such industries to switch the composition of investment. The mechanism is stronger during recessions, when credit constraints bind more. Abstract: We show that an increase in aggregate uncertainty—measured by stock market volatility—reduces productivity growth more in industries that depend heavily on external finance. The mechanism at play is that during periods of high uncertainty, firms that are credit constrained switch the composition of investment by reducing productivity-enhancing investment—such as on ICT capital—which is more subject to liquidity risks (Aghion et al., 2010). The effect is larger during recessions, when financing constraints are more likely to be binding, than during expansions. Our statistical method—a difference-in-difference approach using productivity growth of 25 industries from 18 advanced economies over the period 1985–2010—mitigates concerns with omitted variable bias and reverse causality. The results are robust to the inclusion of other sources of interaction effects, instrumental variable approaches, and different datasets. The results also hold if economic policy uncertainty (Baker et al., 2016) is used instead of stock market volatility as a measure of aggregate uncertainty.