The Flexibility of the Workweek in the United States: Evidence from the FIFA World Cup

Abstract

In this paper I explore the flexibility of the work week in the United States, using the FIFA Soccer World Cup as a natural experiment. My empirical strategy exploits the exogenous variation that arises due to which country hosts the World Cup, as this will determine the time games are broadcast across different time zones in the United States. The hour of the day when games are broadcast differentially affects hours of work across different time zones. Further, the calendar timing of the World Cup allows me to compare labor market outcomes in June/July for a worker in World Cup year t, with the outcomes in June/July for a worker in non-World Cup years t + 1, t + 2 and t + 3. My results highlight the importance of the worker's pay frequency in their work week flexibility, as all differences in hours of work due to the World Cup are concentrated among salary paid workers, while hourly paid workers do not change their market hours during the World Cup. Also, my results show that after controlling for observable demographic characteristics as well as year and month fixed effects, a worker reduces on average his weekly number of hours of work during the World Cup by statistically significant estimates that range from 9 weekly minutes to 28 weekly minutes, depending on specification choice and time of the day during which World Cup games are broadcast live in the U.S.