VoxEU Column Frontiers of economic research

What are the effects of mandatory military conscription on crime and the labour market?

The effects of mandatory military conscription on the education, crime, and labour market outcomes of the draftees are not clear. This column suggests that the heterogeneous nature of the effects could be an explanation for the lack of consensus. The findings show that military service increases the likelihood of future crimes, mostly among males from disadvantaged backgrounds and with a previous criminal history. The only positive effect of conscription for this group is the decrease in disability benefits and the number of sick days. 

Young men in more than 60 countries around the world face the prospect of mandatory military conscription.1 This occurs at a critical juncture in their lives – when they are making decisions about higher education, entering the labour market, and are at the peak of the age-crime profile. It is therefore not surprising that conscription remains a hotly debated topic; a number of European countries have recently abolished it (France in 1996, Italy in 2005, Sweden in 2010, and Germany in 2011), while others have had failed referendums (Austria and Switzerland in 2013).2  This column aims to raise awareness of this important issue and shed more light on how mandatory military conscription may affect crime and labour market outcomes of young men.3

The contemporaneous effect of conscription on crime is ambiguous. Keeping young men engaged and isolated from mainstream society during their most crime-prone years can incapacitate crime, while increased social interactions among young men who serve could increase crimes that are highly ‘social’ in nature. Conscription could also affect post-service crime through a number of channels. An ‘incapacitation’ effect, combined with the persistent nature of crime, could lead to a reduction in post-service crime. The promotion of democratic values, obedience, and discipline may also decrease post-service crime by focusing men at this high risk age. Exposure to weapons and desensitisation to violence, however, could exacerbate criminal tendencies (Grossman 1995).

Conscription may also affect crime through its impact on education and labour market outcomes. Conscription would decrease crime if it is viewed as a positive signal of quality by employers, or improves a young man’s marketable skills, health, or physical fitness. However, post-service crime may increase if conscription interrupts a continuous educational path, delays entry into the labour market, and reduces future labour market opportunities. Intense exposure to new peers during service may have either positive or negative effects, depending on the relative characteristics of the new and old peer groups.

Reconciling a mixed and outdated literature

There is little consensus in the academic literature about the impact of this potentially life transforming event. Angrist’s (1990) seminal study found that Vietnam draftees in the US had lower earnings than non-draftees. Subsequent papers (Angrist and Chen 2011, Angrist et al. 2011) find that this gap closes over time, so that by age 50 draftees are on par with non-draftees. There is some evidence that conscription causes an increase in violent crimes among Vietnam veterans in the US (Rohlfs 2010, Lindo and Stoecker 2012), though this is not seen amongst Australian veterans (Siminski et al. 2016).

The effects of peacetime conscription are similarly mixed: no effect on wages in Britain and Germany (Grenet et al. 2011, Bauer et al. 2012), a negative effect in Holland and for high-ability men in Denmark (Imbens and van der Klaauw 1995, Bingley et al. 2014), and a positive effect for low-educated men in Portugal (Card and Cardoso 2012). Galiani et al. (2011) find that conscription increases crime in Argentina, while Albaek et al. (forthcoming) find that service reduces property crime among Danish men with previous convictions.

What can explain these diverse findings?

  • First, the effect of conscription may change over the lifecycle.

For an outcome like crime which peaks as a young adult, focusing on crime after age 40, as done in some of the previous studies, may skew the results.

  • Second, the conscription ‘experience’ varies greatly across studies.

While peacetime versus wartime conscription is the most obvious example, other differences may emerge as countries approach the end of their mandatory conscription regimes.

  • Third, measured differences may be related to differences in how the causal effect is identified.

Because of the selection process involved in military service, one cannot simply compare outcomes for those who do and do not serve. The above mentioned studies use various quasi-experimental designs to solve this potential omitted variables problem. The most convincing studies rely on random variation in service generated by draft lotteries. But we should also be interested in the effect of service in countries that do not rely upon a lottery to assign service.

Several studies do this by comparing cohorts before and after the abolition of mandatory conscription. This research design can yield different results than the lottery design for a number of reasons: the conscription experience likely differs when it is about to be abolished; it may include general equilibrium effects; and the average and marginal individuals ‘treated’ may not be comparable across studies. If conscription has heterogeneous effects, then it is not surprising if studies with different identification strategies find different effects.

New research

Our new paper (Hjalmarsson and Lindquist 2016) contributes to this debate by utilising individual administrative records and a quasi-experimental research design to identify the causal impact of mandatory military conscription in Sweden on crime (both during and after conscription), legitimate labour market outcomes, and work-related health outcomes. Our paper stands out from the previous literature by:

  • Studying modern-day cohorts;
  • Using a comprehensive set of crime and labour market outcomes;
  • Applying a new identification strategy; and
  • Providing the first clean evidence of an incapacitation effect using information on the exact dates of service.

Mandatory military conscription in Sweden dates back to 1901 and was abolished in 2010, after a gradual decline that began upon the end of the Cold War. For most of this period, Swedish male citizens underwent an intensive drafting procedure upon turning 18, including tests of physical and mental ability. Generally speaking, the tested were positively selected for conscription; those with the highest cognitive and non-cognitive test scores were most likely to serve. Given that such ability measures are also likely correlated with criminality, a naïve comparison of post-service crime rates of those who do and do not serve would most certainly yield biased estimates of the effects of conscription.

Though potential Swedish conscripts are not assigned to service on the basis of a lottery, there is some ‘chance’ involved in service decisions. Namely, each individual’s test results were reviewed by a randomly assigned officiator, with a relatively high or low tendency to assign conscripts to service; we can observe these officiators from 1990 to 1996.  It is this exogenous variation in the likelihood of serving that we use to identify the causal effect of conscription on crime and labour market outcomes. Those who serve are 20 percentage points more likely to have been assigned to a high service rate officiator than those who do not serve.

New findings

Our baseline results are striking:

  • Military service significantly increases both the likelihood of crime and the number of crimes between ages 23 and 30.

These effects are seen across all crime categories, are quite large in magnitude, and are driven by those with a criminal history prior to service or who come from low socioeconomic status households.

  • Given these findings, it is perhaps surprising that we also find large and significant incapacitation effects of conscription, especially for drug and alcohol offences and for traffic crimes.

Unfortunately, our analysis suggests that these effects are not large enough to break a cycle of crime that has already begun prior to service.

This heterogeneous impact of service is also seen with respect to labour market outcomes. Individuals from disadvantaged backgrounds have significantly lower income, and are more likely to receive unemployment and welfare benefits. In contrast, military service significantly increases income and does not impact welfare and unemployment for those at the other end of the distribution. There is no effect of service on the likelihood of higher education.

The only positive effect of service we see, at least for those from disadvantaged backgrounds, is a decrease in disability benefits and the number of sick days; these effects are in fact seen for all subsamples.

Conclusion

Our analysis indicates that mandatory military conscription significantly impacts the life course of young men; the heterogeneous nature of the effects reinforces already existing inequalities in the likelihood of future success. Our results contradict the idea that military service may be a way to straighten out troubled youths and build skills that are marketable in the post-service labour market. These non-monetary costs should be taken into account when deciding whether to reinstate or abolish mandatory conscription or when devising the system through which conscription occurs (e.g. lottery, testing, etc.). Who are the average and marginal conscripts? How will conscription affect these individuals?

References

Albaek, K, S Leth-Petersen, d le Maire and T Tranaes (forthcoming), “Does Peacetime Military Service Affect Crime?” Scandinavian Journal of Economics.

Angrist, J D (1990), “Lifetime Earnings and the Vietnam Era Draft Lottery: Evidence from Social Security Administrative Records," American Economic Review 80(3), 313-336.

Angrist, J D and S H Chen (2011), “Schooling and the Vietnam Era GI Bill: Evidence from the Draft Lottery,” American Economic Journal: Applied Economics 3(2), 96-118.

Angrist, J D, S H Chen and J Song (2011), “Long-term Consequences of Vietnam-Era Conscription: New Estimates Using Social Security Data,” American Economic Review: Papers and Proceedings 101(3), 334–338.

Bauer, T K, S Bender, A R Paloyo and C M Schmidt (2012), “Evaluating the Labour-Market Effects of Compulsory Military Service,” European Economic Review 56(4), 814-829.

Bingley, P, P Lundborg and S Vincent Lyk-Jensen (2014), “Opportunity Cost and the Incidence of a Draft Lottery,” IZA DP No. 8057.

Card, D and A R Cardoso (2012), “Can Compulsory Military Service Raise Civilian Wages? Evidence from the Peacetime Draft in Portugal,” American Economic Journal: Applied Economics 4(4), 57-93.

Galiani, S, M A Rossi, and E Schargrodsky (2011), “The Effects of Peacetime and Wartime Conscription on Criminal Activity," American Economic Journal: Applied Economics 3(2), 119-136.

Grenet, J, R Hart, and E Roberts (2011), “Above and Beyond the Call: Long-term Real Earnings Effects of British Male Military Conscription in the Post-War Years,” Labour Economics 18(2), 194-204.

Grossman, D (1995), On Killing. The Psychological Cost of Learning to Kill in War and Society, Boston: Little, Brown.

Hjalmarsson, R and M J Lindquist (2016), “The Causal Effect of Military Conscription on Crime and the Labour Market,” CEPR Discussion Paper No. 11110.

Imbens, G and W van der Klaauw (1995), “Evaluating the Cost of Conscription in The Netherlands,” Journal of Business & Economic Statistics 13(2), 207-215.

Lindo, J M and C Stoecker (2014), “Drawn into Violence: Evidence on ‘What Makes a Criminal’ from the Vietnam Draft Lotteries,” Economic Inquiry 52(1), 239-258.

Poutvaara, P and A Wagener (2007), “Conscription: economic costs and political allure,” The Economics of Peace and Security Journal 2(1), 6-15.

Poutvaara, P and A Wagener (2011), "The Political Economy of Conscription," in Christopher J. Coyne and Rachel L. Mathers (eds.) The Handbook on the Political Economy of War, Edward Elgar Publishing Ltd, Cheltenham, 154-174.

Rohlfs, C (2010), “Does Combat Exposure Make You a More Violent or Criminal Person? Evidence from the Vietnam draft,” Journal of Human Resources 45(2), 271-300.

Siminski, P, S Ville, and A Paull (2016), “Does the Military Train Men to Be Violent Criminals? New Evidence from Australia’s Conscription Lotteries,” Journal of Population Economics 29(1), 197-218.

Endnotes

[1]See the CIA’s World Factbook (https://www.cia.gov/library/publications/the-world-factbook/fields/2024....) and http://chartsbin.com/view/1887 for a summary of this data.

[2] Though the US moved to an all-volunteer military in 1973, young men ages 18 to 26 are still required to register for the draft. Today, the US is debating extending this requirement to young women (http://www.nbcnews.com/news/us-news/military-officials-women-should-register-draft-n509851). Sweden is also considering reinstating some form of mandatory public service for both men and women.

[3] For a broader discussion of the pros and cons of a conscription army vs. an all-volunteer army from a more general economic and political perspective see Poutvaara and Wagener (2007, 2011).

40,629 Reads