Conditional growth volatility and sectoral comovement in U.S. industrial production, 1828–1915

Abstract

This article investigates conditional growth volatility for industrial production in the U.S. during 1828–1915, taking as a reference sectoral and aggregate indexes constructed in connection to Davis (Quart J Econ 119:1177–1215, 2004). The period includes the major shock represented by the Civil War with the associated resource allocation distortions. The evidence mostly suggests high persistence in conditional volatility as would be found in later studies for the U.S. on GDP growth volatility. However, the evidence of asymmetric volatility appears to be more localized and salient examples of a stronger role of negative shocks on volatility can be identified in the cases of the textile, machinery and metals sectors that might have been more vulnerable to the Civil War. As for interindustry linkages, a complementary factor analysis suggests that the communality changes between the antebellum and postbellum eras. The relative importance of the aggregate shocks increased considerably after the Civil War. This indicates that the Civil War had significant effects in raising the cross-correlation between most sectors, suggesting substantial changes in the basic productive relationships in U.S. 19th century economy.