Research

Working Papers:

What Makes Hiring Difficult? Evidence from Linked Survey-Administrative Data (with Antoine Bertheau and Birthe Larsen)

We design a survey that asks firms about the obstacles that discourage them from hiring despite having potential needs. Using Danish administrative data and subjective beliefs elicited from our survey, we show how hiring obstacles vary across firms. Over two-thirds of employers agree that skill shortages are a hiring obstacle. One-third of employers consider labor costs, the time to find candidates, and the time to train new recruits as hiring obstacles. High-wage firms are less deterred by labor costs, but they are equally affected by skill shortages. Younger and smaller firms are more discouraged by search and training time. Around thirty percent of employers prefer to hire the already employed over the unemployed, because they believe that unemployed workers have lower abilities due to negative selection or skill depreciation during unemployment. Firms with such preferences are more likely to report hiring obstacles. We also find that firms' misperceptions of their wage in the wage distribution could affect their hiring behaviors.

Working Environment and Corporate Outcomes (with Mario Daniele Amore, Morten Bennedsen, and Birthe Larsen)

We use a large-scale survey and register data from Denmark to construct firm-level measures of the working environment, which allows us to capture its physical, psychological, and social dimensions. We document three stylized facts about firms with better working environments. First, they exhibit higher profitability. Second, they have higher innovation quality, as captured by patent citations and survey data. Third, their employees are healthier and have longer tenure. Overall, our findings are consistent with the belief that both shareholders and employees would benefit significantly from improvements in the working environment.

The Political Impacts of Athlete Activism: Evidence from the NBA

How does activism by athletes, which is becoming increasingly common over time, affect political outcomes? This paper provides evidence that National Basketball Association (NBA) players' activism between 2016 and 2020 has affected the political outcomes in the US. More specifically, I find that if a county has an active NBA player between the 2016 and the 2020 U.S. presidential election, and the player is the only player from the county in recent years, the percentage of people voting for the Democratic Party would increase by around one percentage point. This result is obtained by using a difference in differences approach that compares these counties with a control group of counties that have an active minor league basketball player over the same period. I interpret this as the result of the role model effect by the local NBA athletes. To support this hypothesis, I show that these counties also have an increase in protests of more than 30 percent one or two months after the police killing of George Floyd. Despite the increase in protests, polarization does not rise in these counties as a result. These results support the role model effect channel and provide evidence that public leaders play a significant role in social movements.