By Giulio Fella and Giovanni Gallipoli
Crime is a hot issue on the policy agenda in the United States. Despite a significant fall in crime levels during the 1990s, the costs to taxpayers have soared together with the prison population. The US prison population has doubled since the early 1980s and currently stands at over 2 million inmates. According to the latest World Prison Population List (ICPS, 2013), the prison population rate in 2012 stood at 716 inmates per 100,000 inhabitants, against about 480 in the United Kingdom and the Russian Federation – the two OECD countries with the next highest rates – and against a European average of 154. The rise in the prison population is not just a phenomenon in the United States. Over the last twenty years, prison population rates have grown by over 20% in almost all countries in the European Union and by at least 40% in one half of them. The pattern appears remarkably similar in other regions, with a growth of 50% in Australia, 38% in New Zealand and about 6% worldwide.
In many countries – such as the United States and Canada – this fast-paced growth has occurred against a backdrop of stable or decreasing crime rates and is mostly due to mandatory and longer prison sentencing for non-violent offenders. But how much does prison actually cost? And who goes to jail?
The average annual cost per prison inmate in the United States was close to 30,000 dollars in 2008. Costs are even higher in countries like the United Kingdom and Canada. Punishment is an expensive business. These figures have prompted a shift of interest, among both academics and policymakers, from tougher sentencing to other forms of intervention. Prison populations overwhelmingly consist of individuals with poor education and even poorer job prospects. Over 70% of US inmates in 1997 did not have a high school degree. In an influential paper, Lochner and Moretti (2004) establish a sizable negative effect of education, in particular of high school graduation, on crime. There is also a growing body of evidence on the positive effect of education subsidies on school completion rates. In light of this evidence, and given the monetary and human costs of crime, it is crucial to quantify the relative benefits of policies promoting incarceration vis-à-vis alternatives such as boosting educational attainment, and in particular high school graduation.
When it comes to reducing crime, prevention may be more efficient than punishment. Resources devoted to running jails could profitably be employed in productive activities if the same crime reduction could be achieved through prevention.
Establishing which policies are more efficient requires a framework that accounts for individuals’ responses to alternative policies and can compare their costs and benefits. In other words, one needs a model of education and crime choices that allows for realistic heterogeneity in individuals’ labor market opportunities and propensity to engage in property crime. Crucially, this analysis must be empirically relevant and account for several features of the data, in particular for the crime response to changes in enrollment rates and the enrollment response to graduation subsidies.
The findings from this type of exercise are fairly clear and robust. For the same crime reduction, subsidizing high school graduation entails large output and efficiency gains that are absent in the case of tougher sentences. By improving the education composition of the labor force, education subsidies increase the differential between labor market and illegal returns for the average worker and reduce crime rates. The increase in average productivity is also reflected in higher aggregate output. The responses in crime rate and output are large. A subsidy equivalent to about 9% of average labor earnings during each of the last two years of high school induces almost a 10% drop in the property crime rate and a significant increase in aggregate output. The associated welfare gain for the average worker is even larger, as education subsidies weaken the link between family background and lifetime outcomes. In fact, one can show that the welfare gains are twice as large as the output gains. This compares to negligible output and welfare gains in the case of increased punishment. These results survive a variety of robustness checks and alternative assumptions about individual differences in crime propensity and labor market opportunities.
To sum up, the main message is that, although interventions which improve lifetime outcomes may take time to deliver results, given enough time they appear to be a superior way to reduce crime. We hope this research will advance the debate on the relative benefits of alternative policies.
Giulio Fella is a Senior Lecturer in the School of Economics and Finance at Queen Mary University, United Kingdom. Giovanni Gallipoli is an Associate Professor at the Vancouver School of Economics (University of British Columbia) in Canada. They are the co-authors of the paper ‘Education and Crime over the Life Cycle‘ in the Review of Economic Studies.
Review of Economic Studies aims to encourage research in theoretical and applied economics, especially by young economists. It is widely recognised as one of the core top-five economics journal, with a reputation for publishing path-breaking papers, and is essential reading for economists.