Unemployment and Crime: Explaining the Apparent Nonrelationship
33 Pages Posted: 18 Jun 2020
Date Written: October 10, 2001
The majority of macro-level research has found no significant positive association between unemployment rates and crime rates, even though theory suggests that there should be one. This paper considers two possible explanations for the absence of the expected relationship. The first is that public assistance expenditures may cushion the impact of unemployment so that it has no impact on crime. The second is that much prior research has controlled for poverty rates or income levels, which may partial out much of the indirect crime-increasing effect of unemployment on crime, creating the appearance that crime and unemployment are unrelated.
Analyses of 1990 metropolitan area data fail to support the welfare explanation. Specifically, the effect of unemployment rates on crime rates is just as high where welfare payments are more generous as where they are less generous. In support of the second explanation, a review of past unemployment-crime research indicates that 45% of all findings, and 65% of cross-sectional findings, were based on models that controlled for poverty or income, and that studies with such controls were less likely to find a significant positive association than those without the controls. Moreover, analyses of the metropolitan data also support the partialling hypothesis in that the coefficient for unemployment decreases once poverty is controlled, for all crimes except rape.
Suggested Citation: Suggested Citation