The U.S. business cycle, 1867-1995: dynamic factor analysis vs. reconstructed national accounts

This paper presents insights on U.S. business cycle volatility since 1867 derived from diffusion indices. We employ a Bayesian dynamic factor model to obtain aggregate and sectoral economic activity indices. We find a remarkable increase in volatility across World War I, which is reversed after World War II. While we can generate evidence of postwar moderation relative to pre-1914, this evidence is not robust to structural change, implemented by time-varying factor loadings. We do find evidence of moderation in the nominal series, however, and reproduce the standard result of moderation since the 1980s. Our estimates broadly confirm the NBER historical business cycle chronology as well the National Income and Product Accounts, except for World War II where they support alternative estimates of Kuznets (1952).