Intimate Partner Violence and Football

We study the question how and when football causes intimate partner violence. Using individual crime and incident data reported in the period from 2008 to 2019 in Greater Manchester (UK), we study how the timing of home (Manchester United and Manchester City) Premier League games and national games shapes the cycle of intimate partner violence (IPV). We find that a relevant football game significantly increases intimate partner violence in the twelve hours following the game.

We test a number of hypothesis on the nature of this effect. We do not find evidence that a turn in pregame predicted probabilities (an unexpected loss) significantly increases IPV (Card and Dahl (2011)). We do confirm that a loss of the home team emphasizes the effect. Using variation in the timing of football games following a decision to increase the number of early games and exogenous assignment to group game timing in national games, we find that games early in the afternoon cumulatively increase IPV significantly more than later games. During the game and immediately after the game, domestic abuse levels are lower, while the largest percentage increase in domestic abuse is recorded around the 8th hour after the earlier game ends most likely as a result of a longer period of drinking. The unintended consequences of earlier kick-off times, with the aim of easing the immediate policing of games, have caused an increase in IPV.

We run a series of placebo tests such as using the British-specific rule of a blackout of matches on live television on Saturday between 2:45pm and 5:15pm. Moreover, we contribute to the literature by ruling out increased policing on specific match days by complementing the analysis with police movement data, the exact location of the intimate partner violence and whether the violence occurs in the home of the victim.

Finally, our unique contribution speaks to the nature of the victims and perpetrators of this intimate partner violence. We test if football related IPV is a pattern of ongoing behavior for specific couples, whether these perpetrators are likelier to have a violent criminal record and whether there is a substitution effect between intimate partner violence and violent anti-social behavior. We also test whether football triggers first time 'football' IPV perpetrators, whether they are likelier to reoffend and whether there is a higher chance of violence escalation for these cases.