LABOR FORCE TRANSITION DYNAMICS

Abstract

Job posting counts (JPCs) are emerging as indicators of employment dynamics, yet their validity requires assessment. This study evaluates the effectiveness of big-data-based JPCs compared to the traditional survey-based unemployment rate in capturing labor market transitions in the United States. Using the Current Population Survey, our comparison focuses on their ability to predict individuals' transitions between employment and unemployment. We explore not only monthly national JPCs but also four additional versions that measure labor demand at various granularities. Our findings suggest comparable predictive power for JPCs and the unemployment rate, with each capturing different aspects of the variation in these transitions. Coefficients for both metrics remain statistically significant when considered together. Notably, the unemployment rate's correlation with transitions changes signs when adding year fixed effects, a phenomenon not observed for JPCs. Among more granular levels of JPCs, the most refined measure - those by state, occupation, and industry - exhibits the strongest predictive capabilities. Furthermore, our main results remain robust when applying alternative JPC measures (total unique and total JPCs). Show more