Research

Three years after the COVID-19 crisis, employment and total hours worked in Europe fully recovered, but average hours per worker did not. We analyze the decline in average hours worked across European countries and find that (i) it is not cyclical but predominantly structural, extending a long-term trend that predates COVID-19, (ii) it mainly reflects reduced hours within worker groups, not a compositional shift towards lower-hours jobs and workers, (iii) men—particularly those with young children—and youth drive this drop, (iv) declines in actual hours match declines in desired hours. Policy reforms could help involuntary parttimers and women with young children raise their actual hours towards desired levels, but the aggregate impact on average hours would be limited to 0.5 to 1.5 percent. Overall, there is scant evidence of slack at the intensive margin in European labor markets, and the trend fall in average hours worked seems unlikely to reverse.

We use firm-level data from 10 European countries to establish several new stylized facts about firms’ labor market power. First, we find the pervasive presence of labor market power across countries and sectors, measured by average and median markdowns above unity. Second, focusing on the dynamics, we find that weighted average markdowns have increased 1.3 percent between 2000 and 2017. However, median and unweighted average markdowns have actually decreased over the same time period, suggesting the existence of divergent paths across the markdown distribution. Third, we show that high-markdown firms tend to have a large footprint in both their product and input (labor) markets, and are most commonly listed and found among services sectors. Finally, a Melitz-Polanec decomposition of the change in weighted average markdown finds that the increase has been driven by a reallocation of resources towards high-markdown incumbents and by the extensive margin via the net entry of high-markdown firms while, in contrast, there was a decline in within-firm markdowns. Our findings highlight the importance of using granular and broad-based data for a thorough analysis of firms’ labor market power.

What are the implications of the needed climate transition for the potential reallocation of the U.S. labor force? This paper dissects green and polluting jobs in the United States across local labor markets, industries and at the household-level. We find that geography alone is not a major impediment, but green jobs tend to be systematically different than those that are either neutral or in carbon-emitting industries. Transitioning out of pollution-intensive jobs into green jobs may thus pose some challenges. However, there is a wage premium for green-intensive jobs which should encourage such transitions. To gain further insights into the impending green transition, this paper also studies the impact of the Clean Air Act. We find that the imposition of the Act caused workers to shift from pollution-intensive to greener industries, but overall employment was not affected.

      This paper proposes a hidden state Markov model (HMM) that incorporates workers' unobserved labor market attachment into the analysis of labor market dynamics. Unlike previous literature, which typically assumes that a worker's observed labor force status follows a first-order Markov process, the proposed HMM allows workers with the same labor force status to have different history-dependent transition probabilities. The estimated HMM generates labor market transition probabilities that match those observed in the data, while the first-order Markov model (FOM) and its many-state extensions cannot. Even compared with the extended FOM with observable (unobservable) heterogeneity, the HMM improves the fit of the empirical transition probabilities by a factor of 30 (6.5). I apply the HMM to calculate the long-run consequences of separation from stable employment and study evolution of employment stability across different demographic groups over the past several decades. The proposed HMM can be a useful benchmark for future research in describing labor market dynamics for all three labor market states (employment, unemployment, and non-participation) by parsimoniously incorporating unobserved heterogeneity and duration dependence.