Alcohol and rape: An “economics-of-crime” perspective

https://doi.org/10.1016/j.irle.2007.09.002Get rights and content

Abstract

Using a panel of state-level data over the years 1982–2000, this study explores the potential relationship between alcohol policy, alcohol consumption, and rape by considering a number of theoretical and methodological issues. First, the potential relationship is examined in the context of an economics-of-crime model, controlling for various deterrence and opportunity cost variables. Second, unlike most studies, consumption of liquor and wine are considered as well as consumption of beer. A third focus is on the potential endogeneity of alcohol consumption. A fourth consideration dictates the focus on rape-rates rather than other violent crime rates. Several studies have discovered a tendency for relatively high alcohol consumption by both offenders and victims. Since rape victims are virtually all female, and at least some alcohol policy variables appear to have differential impacts on females and heavy-drinking males, an examination of policy impacts for rape-rates, unlike other types of violent crime, can suggest the nature of the alcohol-consumption–violence relationship. The empirical results support an “alcohol-increases-potential-victims-vulnerability” hypothesis. In the context of the economic theory of crime, this lowers the expected cost of rape to potential offenders and raises rape-rates. Tests for endogeneity of alcohol consumption and resulting simultaneous equation estimates reinforce this implication because they suggest that potential victims respond to relatively high rape-rates by reducing alcohol consumption.

Introduction

A large academic literature explores the alcohol–crime relationship, but for the most part, economists were not involved in this literature until the 1990s.1 Cook and Moore (1993) explain why, noting that the rationality assumption naturally leads economists to be skeptical of an alcohol-to-violence causal relationship, and that economists also recognize the difficulty in determining the direction of causality in empirical models of the alcohol–crime relationship. They go on to argue that the economic model can be applied to the study of drinking-related violence, however,2 and since then a number of studies by economists have appeared.3 The empirical analysis presented below augments this growing literature with a careful examination of the direction-of-causality issue while simultaneously exploring the path of causality. The focus is on the relationship between alcohol consumption and rape, as this facilitates exploration of the path through which alcohol consumption may influence violence.

The general consensus in the academic literature is that alcohol consumption and aggressive behavior (criminal or otherwise) are correlated. However, whether the observed alcohol–aggression relationship is truly causal remains controversial. Consuming alcohol might make an individual act more aggressively through some physiological, psychological, and/or sociological process. On the other hand, the acts of consuming alcohol and committing violence may simply be behaviors arising simultaneously from a latent behavioral factor such as an individual's preference for risk. Alcohol consumption also may make individuals less wary or reduce their ability to defend themselves, and therefore make them more vulnerable to victimization.4 In the context of the economic theory of crime, where potential offenders rationally weigh costs and benefits, this vulnerability lowers the expected cost of committing violent crimes for potential perpetrators (these expected costs may be relatively low for other reasons too, as explained below), and as the economic model predicts, these relatively low expected costs should increase the level of crime, all else equal.5 The following analysis attempts to determine whether this victim-vulnerability path is an important source of the correlation between alcohol consumption and rape-rates.

If, as many previous studies suggest,6 there is a significant relationship between alcohol use and rape-rates, then the crime of rape is particularly well suited for further exploration of the direction and path of causality for two reasons. First, alcohol use appears to be prevalent among both offenders and victims. Koss (1998, pp. 62–63) reports that 55% of the victims of sexual assaults and rapes in her survey of college students admit having been under the influence of intoxicants at the time of their victimization, for instance, while 75% of the offenders report that their victims were either drinking or using drugs when the offense was committed.7 Second, victims are virtually all female while offenders are virtually all male,8 and at least some alcohol policy variables appear to have differential impacts on females and those males for whom heavy drinking might lead to increases in aggressive behavior and violent offenses. In particular, there is evidence that young heavy-drinking males are relatively non-responsive to price changes for beer (at least over price ranges represented in available data), while young females tend to be much more responsive. Saffer and Dave (2003) find that both price and advertising effects are larger for females than for males, for instance, using data from both the Monitoring the Future (MTF) Survey and the National Longitudinal Survey of Youth. Indeed, their coefficient estimates in a series of female regressions for participation (consumption of alcohol) during the past year, participation during the past month, and bingeing in the past 2 weeks suggest that females clearly respond to alcohol price (coefficients are negative and significant in all three regressions). On the other hand, price coefficients are insignificant in their past month and binge-in-past-2-weeks regressions for males. Thus, Saffer and Dave (2003, p. 16) conclude that: “Overall, the results indicate that … male participation is explained more by demographics than public policy.” Chaloupka and Wechsler (1996) find similar relationships in their data on college students: males in their sample do not significantly change the quantity of beer consumption in the face of price changes while females do. Kenkel (1996) also reports that the price elasticity of the frequency of heavy drinking by males who have relatively little information about the dangers of drinking is not statistically significant.9

Men who are sexually aggressive typically are heavy drinkers (Schwartz & Dekeserdy, 1997, p. 100; citing Lisak & Roth, 1988; Schwartz & Nogrady, 1996). Thus, the heavy-drinking males examined in the Saffer and Dave (2003), Kenkel (1996), and Chaloupka and Wechsler (1996) studies may be representative of the population from which alcohol-consuming sexual offenders are drawn. This in turn suggests that policies that increase the price of alcohol might not have a substantial impact on rape, assuming that a causal relationship runs from alcohol to violence through its impact on potential offender (male) behavior. On the other hand, if a causal relationship arises because alcohol lowers the expected cost of rape by making potential victims (females) more vulnerable, then policies that raise the price of alcohol may reduce rapes, assuming that the females in the Saffer and Dave (2003) and Chaloupka and Wechsler (1996) studies are representative of a substantial portion of the potential rape-victim population.10 Findings of policies that raise the “full” price of alcohol and also appear to reduce rape-rates may not be definitive evidence in support of the “alcohol-increases-potential-victims’-vulnerability” hypothesis, of course, but another focus of this study provides corroboration for the implications of such results.

Simultaneity bias can be introduced into a rape-rate model by controlling for alcohol consumption. For instance, it has been suggested that individuals chose to consume alcohol after they have decided to commit rapes because alcohol lowers the severity of potential punishment for rape (or perhaps offenders’ psychological costs) by allowing the offender to blame his behavior on alcohol. It has also been contended that some offenders encourage their targeted victims to drink. Either possibility implies that as rape-rates rise, alcohol consumption simultaneously rises, so the alcohol coefficient in an OLS estimate is biased upward. On the other hand, if females believe that alcohol consumption makes them relatively vulnerable to rape, and if they perceive a relatively high risk of rape (i.e. if rape-rates rise), then one of their rational self-protection actions may be to reduce alcohol consumption.11 In this case, the OLS estimate is biased downward. Tests for the endogeneity of alcohol consumption in an OLS rape-rate model are, therefore, performed, and given that simultaneity appears to apply, corrections to the model are made, thus allowing for this second test of the victim-vulnerability causal connection between alcohol consumption and rape-rates.

Section 2 discusses hypotheses regarding the relationships between alcohol consumption, alcohol control policies, and the incidence of violence. Empirical exploration of the relationship between alcohol consumption and rape begins in Section 3 where OLS estimates are presented. Tests for endogeneity of alcohol consumption measures imply that these specifications may suffer from simultaneity bias, so two-stage-least-squares (2SLS) models are also estimated and reported. Section 4 examines the resulting estimates of the effects of alcohol policy on rape-rates. The results support a relationship between alcohol and rape, but they suggest that a substantial portion of the relationship runs through the “alcohol-increases-potential-victims’-vulnerability” channel. Section 5 contains concluding remarks.

Section snippets

The alcohol–violence relationship

Most of the focus in the literature is on how the consumption of alcohol by offenders may be related to offender behavior.12 A number of non-formal hypotheses have been proposed that are consistent with a positive correlation between offender alcohol consumption and violence. One is the “direct cause” hypothesis, wherein alcohol is assumed to

The effects of alcohol consumption on rape-rates

Developing an empirical model of criminal participation requires the identification and specification of those factors that are most likely to be both empirically important and testable given the nature of the available data. Empirical studies of the link between alcohol consumption and the subsequent violent activities comes from studies that can be broadly classified as individual-specific or aggregate-level (Pernanen, 1981), but aggregate-level studies are of particular interest in the

Estimates of the effects of alcohol consumption on rape

The equilibrium rape-rate is estimated by the simultaneous interaction of the supply of offenses and the consumption of alcohol. All variable definitions are the same as discussed previously. The basic crime rate equation is of the form presented in Eq. (2) above, except that the relevant per-capita alcohol consumption variables are treated as endogenous determinants. As such, separate structural equations must be specified and estimated for these variables in the first-stage regressions.

The

Conclusions

This study explores the potential relationship between alcohol consumption and violence by considering a number of theoretical and methodological issues. First, the potential relationship is examined in the context of an economics-of-crime model, controlling for various deterrence and opportunity cost variables. The statistical relationship between alcohol consumption and violence, in this case measured by reported rape-rates, survives inclusion of these control variables. Second, unlike most

Acknowledgements

This paper draws from a large study on “Alcohol Control Policies and the Incidence of Violent Crime” supported by a grant from the National Institute of Justice, Grant No. 1999-IJ-CX-0041. We wish to thank Doug Young for providing some of the data used, and for his helpful comments on the research. We also want to thank David Rasmussen, Tom Zuehlke, Rich Feiock, Dave MacPherson, Tim Sass, and Ron Johnson, as well as those who made valuable comments and suggestions during a session at the

References (65)

  • B.L. Benson et al.

    Implicit taxes collected by state liquor monopolies

    Public Choice

    (2003)
  • Chaloupka, F. J., & Saffer, H. (1992). Alcohol, illegal drugs, public policy and crime....
  • F.J. Chaloupka et al.

    Binge drinking in college: the impact of price, availability, and alcohol control policies

    Contemporary Economic Policy

    (1996)
  • P.J. Cook et al.

    Economic perspectives on reducing alcohol-related violence

  • A. Cordilia

    Alcohol and property crime: explaining the causal nexus

    Journal of Studies on Alcohol

    (1985)
  • R. Davidson et al.

    Estimation and inference in econometrics

    (1993)
  • J.M. Doyle et al.

    The effects of labor markets and income inequality on crime: Evidence from panel data

    Southern Economic Journal

    (1999)
  • R.T. Dull et al.

    The impact of local alcohol ordinances on official crime rates: The case of Tennessee

    Justice Quarterly

    (1987)
  • I. Ehrlich

    Participation in illegitimate activities: A theoretical and empirical investigation

    Journal of Political Economy

    (1973)
  • I. Ehrlich

    Crime and punishment, and the market for offenses

    Journal of Economic Perspectives

    (1996)
  • T. Ensor et al.

    Modeling the interaction between alcohol, crime and the criminal justice system

    Addiction

    (1993)
  • J. Fagan

    Intoxication and aggression

  • R.B. Freeman

    Why do so many young American men commit crimes and what might we do about it?

    Journal of Economic Perspectives

    (1996)
  • P.J. Goldstein

    The drugs/violence nexus: A tripartite conceptual framework

    Journal of Drug Issues

    (1985)
  • Gould, E. D., Weinberg, B. A., & Mustard, D. B. (1999). Crime rates and local labor market opportunities in the United...
  • Greenfeld, L. A. (1997). Sex offenses and offenders: An analysis of data on rape and sexual assault. Washington DC:...
  • M. Grossman et al.

    Alcohol regulation and violence on college campuses

  • R. Halvorsen et al.

    The interpretation of dummy variables in semilogarithmic equations

    American Economic Review

    (1980)
  • S.D. Johnson et al.

    Alcohol and rape in Winnipeg, 1966–1975

    Journal of Studies of Alcohol

    (1978)
  • E.J. Kanin

    Date rape: unofficial criminals and victims

    Victimology

    (1984)
  • D.S. Kenkel

    New estimates of the optimal tax on alcohol

    Economic Inquiry

    (1996)
  • M.A.R. Kleiman

    Against excess: drug policy for results

    (1992)
  • Cited by (16)

    • Alcohol abuse, personality disorders, and aggression: The quest for a common underlying mechanism

      2017, Aggression and Violent Behavior
      Citation Excerpt :

      On the other hand, chemical effects of heavy alcohol consumption may interfere with normal cognitive and motor processes contributing to the escalation of aggressive behavior and ultimately leading to sexual assault (Abbey et al., 2004). Similarly, other scholars found increased alcohol consumption among both perpetrators and victims of sexual assaults (Zimmerman & Benson, 2007). Importantly, however, societal alcohol regulation is likely to impact the association between alcohol use and alcohol-related violence (Markowitz, 2005; Zimmerman & Benson, 2007).

    View all citing articles on Scopus
    View full text