Interpreting aggregate fluctuations looking at sectors

A dynamic factor model is used to investigate on the variability in labor productivity and hours across the 2-digit US manufacturing industries. Two kind of shocks emerge as quantitatively relevant during the postwar period. They can be reasonably interpreted as technology shocks to the production of equipment and economy-wide shocks. The former induces a positive correlation between productivity and hours growth rates in the durable-goods roducing sector; the latter spurs a negative correlation in the nondurable-goods producing sector. Such evidence provides a novel interpretation of the aggregate near-zero correlation between the two variables.