Productivity and the labor market: comovement over the business cycle*

TLDR: In this paper, the authors show that the discrepancy between the data and the model becomes substantially smaller if employment data from the Current Population Survey (CPS) is used in measuring productivity instead of the commonly used CPS data from Current Employment Statistics (CES).

Abstract: The productivity-driven Mortensen-Pissarides model predicts that labor productivity, dened as the ratio of output to employment, is strongly correlated with employment, unemployment, vacancies and wages whereas these correlations were argued to be much weaker in the data, especially since the mid 1980s. We rst document that the size of these discrepancies between the data and the model becomes substantially smaller if employment data from the Current Population Survey is used in measuring productivity instead of the commonly used employment data from the Current Employment Statistics. Second, we show that incorporating time to build and a stochastic value of home production helps reconcile the quantitative performance of the model with the data with stochastic productivity being the key determinant of the business cycle dynamics of the model.