Violent crime and incentives in the long-run: evidence from England and Wales

Abstract

This study uses recent advances in time-series econometrics to investigate the non-stationarity and co-integration properties of violent crime series in England and Wales. In particular, we estimate the long-run impact of economic conditions, beer consumption and various deterrents on different categories of recorded violent crime. The results suggest that a long-run causal model exists for only minor crimes of violence, with beer consumption being a predominant factor.

Keywords:

• violent crime
• time series
• co-integration
• weak exogeneity
• error correction model

Acknowledgements

We would like to thank for their helpful comments Wojciech Charemza, Mike Clements, Derek Deadman, Baibing Li, Robin Marris, Steve Pudney, Kavita Sirichand, Mark Tippett and Paul Tracey. In addition, the paper benefited from comments from the participants of seminars and conferences at the University of Leicester, the Institute for Social and Economic Research and the 4th European Society of Criminology Conference. The usual disclaimer applies.

Notes

Co-integration methods have been used in studying crime in the USA 35, as well as property crime in the UK 14.

When the analysis is conducted in levels, the results show that there is a long-run relationship among the variables, whereas when the analysis is conducted in differences the results measure the effect of short-term changes (information about the trend in the level of the series is discarded).

The sole exceptions are the estimated imprisonment rate (−0.097) and poverty rate (0.323) elasticity parameters in the indecent assault on a female model were found to be statistically significant and in line with the theory.

Wolpin used national crime statistics for E&W for the period 1894–1967 and estimated a single-equation model to assess the effect of deterrence and economic conditions on specific types of violent crime. Although Wolpin's study provides an important contribution to the economics of crime literature, it generally lacks statistical testing for reliability of the regression models and the inference which were derived from them.

Ormerod et al. 25 provide, however, an innovative methodological approach to the process by which crime spreads (or contracts) over time, and makes an interesting theoretical contribution to the role of social interaction between agents.

Field 11 also found a small positive and significant relationship between unemployment and violence against the person.