This paper reports the results of a detailed examination of the hypothesis that improved inventory management and production techniques are responsible for the decline in the volatility of U.S. GDP growth. Our innovations are to look at the data at a finer level of disaggregation than previous studies, to exploit cross-sectional heterogeneity to obtain clearer identification of this hypothesis, and to provide a complete accounting of the change in GDP volatility. Changes in inventory behavior can account directly for only up to half of the total reduction in GDP volatility. Cross-section evidence from the manufacturing and trade sector indicates that change in the covariance structure among industries accounts for most of the remaining portion of the reduction in GDP volatility. Sales have become less correlated among industries and inventory investment has become more correlated. These distinctive changes in co-movement of industries suggest that development and management of supply chains may be an indirect channel through which changes in inventory management and production techniques have influenced GDP volatility.