Deterrence Theory and the Death Penalty

Deterrence theory and the death penalty concerns one of the longest-running and most methodologically contested debates in criminology: whether the threat of execution deters homicide beyond what less severe sanctions, including life imprisonment, would achieve. Unlike most deterrence questions, which can be studied through natural variation in enforcement or moderate sentencing reforms, capital punishment research faces distinctive obstacles, including the rarity of executions relative to homicides, the long and variable delay between sentencing and execution, and profound uncertainty about how potential murderers actually perceive capital punishment as a threat. This article traces the econometric and criminological literature on capital punishment’s deterrent effect, examines the National Research Council’s influential 2012 assessment of that literature, and considers the theoretical objections that have shaped the debate since Beccaria’s original eighteenth-century argument against the death penalty.

Introduction

Capital punishment occupies a distinctive position within deterrence theory, since it is simultaneously the most severe sanction available to any legal system and among the least certain and least swift in practice, given the lengthy appeals processes most capital cases undergo before an execution, if one occurs at all, is actually carried out. This combination makes the death penalty a uniquely difficult test case for classical deterrence theory’s predictions, since severity is maximized while certainty and swiftness, the dimensions with the strongest empirical support elsewhere in the deterrence literature, are often minimized.

This article is part of the broader treatment of Deterrence Theory, which examines how the threat of legal punishment shapes offending decisions across historical, theoretical, and applied dimensions. Beccaria’s original contractarian objection to capital punishment is addressed in Classical Roots of Deterrence; this article focuses specifically on the empirical question of whether capital punishment deters homicide, a question that has generated one of the most methodologically contentious literatures in all of criminology.

The Death Penalty as a Deterrence Question

Capital Punishment’s Special Place in Deterrence Debate

The death penalty’s deterrent effect has attracted disproportionate research and public attention relative to its practical scope, since capital punishment is applied to a small fraction of homicides even in jurisdictions that retain it, and executions themselves are rare events even in the states most active in carrying them out. This disproportion reflects the moral and political stakes attached to the question rather than its statistical centrality to overall crime control, since even a substantial deterrent effect of capital punishment would prevent only a small number of homicides relative to the crime-control potential of broader certainty-focused policing or sentencing reform.

The theoretical case for a capital deterrent effect follows directly from the classical severity dimension: if increasing a sanction’s severity increases its deterrent power, then the most severe possible sanction should, in principle, produce the strongest possible deterrent signal, and Beccaria’s proportionality argument aside, this straightforward severity logic has motivated much of the econometric literature attempting to detect capital punishment’s marginal deterrent effect relative to life imprisonment.

Beccaria’s own influential objection to capital punishment, developed on contractarian rather than empirical grounds, held that no rational individual would consent in advance to a social contract permitting the state to take his own life, making the death penalty illegitimate regardless of any demonstrated deterrent effect, a position that has continued to inform abolitionist argument even as the empirical deterrence debate has proceeded largely independently of Beccaria’s original contractarian reasoning.

Absolute vs. Marginal Deterrence Applied to Capital Punishment

The distinction between absolute and marginal deterrence, discussed at greater length in Absolute vs. Marginal Deterrence, is directly relevant to the capital punishment debate, since the empirically and policy-relevant question is not whether punishing homicide at all deters homicide, a form of absolute deterrence essentially no one disputes, but whether capital punishment deters homicide more effectively than the most severe available alternative, typically life imprisonment without parole, a marginal deterrence question entirely.

This marginal framing sharpens the empirical challenge considerably, since it requires comparing outcomes under a capital sentencing regime against outcomes under a counterfactual regime in which the same offenders instead faced life imprisonment, a comparison that real-world data, in which capital and non-capital jurisdictions differ in innumerable other ways, can only approximate rather than directly observe.

Much of the confusion in public debate over capital punishment deterrence research stems from a failure to maintain this marginal framing, with studies purporting to show that the death penalty deters homicide sometimes conflating evidence that severe punishment in general deters homicide with evidence that capital punishment specifically outperforms life imprisonment as a deterrent, two claims that require different comparisons and different data.

The Econometric Literature and Its Controversies

Ehrlich’s 1975 Study and Its Reception

Isaac Ehrlich’s 1975 econometric analysis, which concluded that each execution prevented approximately seven to eight homicides, remains the most cited and most contested study in the capital punishment deterrence literature, and its publication in a leading economics journal at a politically consequential moment, shortly before the Supreme Court’s 1976 decision permitting states to resume executions, gave it outsized influence on both academic and public debate.

Ehrlich’s methodology applied Gary Becker’s economic model of crime to national time-series data on execution rates and homicide rates, treating the relationship between the two as evidence of a deterrent effect operating through potential murderers’ rational responses to the probability and severity of capital punishment, an approach methodologically consistent with the broader economic deterrence tradition but one that drew immediate and sustained criticism regarding its specific empirical execution.

Subsequent reanalyses using the same or similar data found that Ehrlich’s results were highly sensitive to the specific time period examined, the functional form of the statistical model, and the inclusion or exclusion of specific control variables, with several replications finding no significant deterrent effect once these specifications were altered, a pattern that became a recurring theme throughout the following decades of capital punishment econometric research (National Research Council, 2012).

Fagan’s (2006) detailed methodological review of the post-Ehrlich literature catalogued a further set of recurring problems, including the difficulty of distinguishing a genuine deterrent effect from a brutalization effect, in which state-sanctioned killing normalizes lethal violence and thereby increases rather than decreases subsequent homicide, a pattern some studies of specific state execution histories have reported even as others report the opposite (Cochran, Chamlin, & Seth, 1994). This theoretical ambiguity, in which the same underlying mechanism, exposure to state killing, could plausibly produce effects in either direction, has made capital punishment research considerably harder to interpret than deterrence research involving less severe and less symbolically freighted sanctions.

Methodological Critiques: Donohue, Wolfers, and the Panel Data Debate

A second wave of econometric studies in the early 2000s, employing panel data methods that exploited variation in execution rates across states and over time rather than relying solely on national time-series data, produced a range of conflicting findings, with some studies reporting substantial deterrent effects and others finding no effect or even a counterintuitive brutalization effect in which executions were associated with subsequent increases in homicide.

Donohue and Wolfers’s influential 2005 methodological critique systematically examined this second wave of panel studies and concluded that their results were extraordinarily sensitive to seemingly minor specification choices, such as which control variables were included or which time period was analyzed, to the point that essentially any conclusion, a strong deterrent effect, no effect, or a brutalization effect, could be generated from the same underlying data through different but individually defensible modeling choices (Donohue & Wolfers, 2005).

This specification sensitivity became the central methodological concern animating the National Research Council’s subsequent formal review of the literature, since it suggested that the econometric capital punishment literature as a whole lacked the kind of stability that would normally be required before drawing confident causal conclusions from observational data, regardless of which particular direction any individual study’s point estimate happened to point.

The National Research Council’s Assessment

The 2012 NRC Report and Its Findings

The National Research Council’s 2012 report, chaired and co-edited by Daniel Nagin and John Pepper, undertook a comprehensive review of the post-1975 econometric capital punishment literature and concluded that none of the existing research was sufficiently credible to determine whether capital punishment increased, decreased, or had no effect on homicide rates, a notably stronger and more categorical conclusion than a simple finding of mixed or inconclusive results (National Research Council, 2012).

The report identified three specific methodological flaws common across the literature it reviewed: existing studies failed to account for the deterrent or incapacitative effects of the noncapital sanctions, typically lengthy imprisonment, that would apply to the same offenders in the counterfactual absence of capital punishment; studies relied on incomplete or implausible models of how potential murderers actually perceive and respond to the risk of execution; and the statistical models used to estimate capital punishment’s effect rested on assumptions that the committee did not find credible given the available data (National Research Council, 2012).

The report’s conclusion that existing research should not be used to inform capital punishment policy proved controversial among some economists who had produced the panel studies under review, but it has been broadly influential within criminology and has become the standard reference point for subsequent scholarly and journalistic discussion of what the evidence does and does not show about capital punishment’s deterrent effect (Nagin, 2014).

Implications for Policy and Research Standards

The NRC report’s methodological critique extends well beyond capital punishment specifically, illustrating a broader concern that has shaped deterrence research standards across the discipline: observational studies relying on aggregate time-series or panel data are highly vulnerable to specification sensitivity, and confident causal claims require either genuinely exogenous variation in the sanction being studied or considerably more caution in interpreting correlational findings than much of the earlier capital punishment literature exercised.

The report also highlighted a specific gap that remains largely unaddressed even in more recent research: essentially nothing is known with confidence about how potential murderers actually perceive their risk of execution, a gap that is particularly consequential given the perceptual deterrence literature’s broader finding, discussed in Perceptual Deterrence Theory, that perceived rather than objective sanction risk is what actually enters the deterrence calculus.

Perceptual and Behavioral Considerations

Do Potential Murderers Perceive Capital Punishment as a Distinct Threat?

The perceptual deterrence framework raises a question the econometric literature has largely been unable to answer directly: whether potential murderers form any meaningfully distinct perception of capital punishment risk, as opposed to a general perception of severe punishment risk that does not meaningfully distinguish between a lengthy prison sentence and execution. Given that the large majority of homicides are committed in the heat of interpersonal conflict rather than through the kind of extended rational calculation deterrence theory assumes, some scholars have questioned whether marginal severity increases at the very top of the sanction scale are likely to be perceived, let alone weighed, by the population actually committing the offense in question.

This concern connects to a broader critique of severity-based deterrence discussed throughout the deterrence literature: severity increases require that offenders possess reasonably accurate knowledge of the applicable penalty, an assumption perceptual deterrence research has substantially undermined even for far more common and less extreme sanctions than capital punishment, raising particular doubt about whether the distinction between capital and non-capital sentencing is salient to the population of potential offenders in the way the classical severity argument requires.

Certainty and Swiftness Problems Specific to Capital Cases

Capital punishment’s practical administration undermines exactly the two dimensions, certainty and swiftness, that the broader deterrence literature identifies as the strongest and most reliable deterrent levers, discussed at length in Certainty, Severity, and Swiftness of Punishment. Only a small fraction of death-eligible homicides result in an actual death sentence, and only a fraction of death sentences are ultimately carried out, given the extensive appeals process capital cases undergo, meaning the objective certainty of execution, conditional on committing a death-eligible homicide, is extremely low even in jurisdictions that actively use capital punishment.

The delay between sentencing and execution, frequently spanning a decade or more in American capital cases, similarly undermines the swiftness dimension that Beccaria’s original framework identified as essential to punishment’s deterrent force, since an execution carried out many years after the offense bears little of the cognitively vivid, temporally proximate connection between crime and consequence that the classical swiftness argument requires for maximal deterrent effect. This combination, high severity paired with low certainty and minimal swiftness, is precisely the profile the broader deterrence literature would predict to produce a comparatively weak deterrent effect, independent of any econometric evidence bearing on the question directly.

Applications in Criminal Justice Policy

Capital Punishment’s Legal and Political Status

Capital punishment’s legal status in the United States has continued to narrow since the 2012 NRC report, with a declining number of states retaining and actively using the sanction and death sentences and executions both declining substantially from their late-twentieth-century peak, a trend that has proceeded largely independent of the deterrence debate and has been driven more by concerns about wrongful conviction, cost, and the sanction’s disproportionate application than by any resolution of the deterrent-effect question.

Policy debate over capital punishment has increasingly shifted away from deterrence-based justification specifically, reflecting the NRC report’s influence and the broader recognition that even a resolved deterrence question would speak to only one of several considerations, including retribution, cost, and the risk of executing an innocent person, that inform capital punishment policy.

Alternatives and the Life-Without-Parole Comparison

Life imprisonment without the possibility of parole has increasingly become the explicit comparison point for capital punishment policy debate, reflecting the marginal deterrence framing discussed above: since virtually no jurisdiction is genuinely debating whether to punish murder at all, the live policy question is whether capital punishment deters more effectively than this severe but non-capital alternative, a comparison the existing econometric literature has not credibly resolved according to the NRC’s assessment.

This life-without-parole comparison also raises a specific-deterrence consideration largely absent from the capital punishment literature’s traditional focus on general deterrence: because execution categorically forecloses any possibility of the sanctioned individual reoffending, capital punishment achieves complete specific deterrence and incapacitation by definition, a consideration entirely separate from, and sometimes conflated with, the general deterrence question of whether the threat of execution deters other potential murderers.

Surveys of leading criminologists conducted over the past several decades have consistently found that a substantial majority do not believe the empirical evidence supports a deterrent effect of capital punishment beyond that of long-term imprisonment, a professional consensus that has remained relatively stable even as the underlying econometric literature has continued to generate conflicting individual studies (Radelet & Akers, 1996; Radelet & Lacock, 2009). This expert consensus has itself become a feature of the policy debate, cited by reform advocates even as individual econometric studies purporting to find a deterrent effect continue to be published and cited by proponents of retaining capital punishment.

Critiques, Limitations, and Current Research Directions

Beccaria’s Contractarian Objection Revisited

Beccaria’s original argument against capital punishment, discussed at length in Classical Roots of Deterrence, rested on social contract grounds rather than empirical deterrence grounds, holding that no rational individual would consent in advance to a state authority permitting the taking of his own life, regardless of whatever deterrent benefit that authority might produce. This contractarian objection has continued to inform abolitionist argument independent of the empirical deterrence debate, since it implies that capital punishment could be illegitimate even if a strong deterrent effect were someday credibly demonstrated.

This separation between the empirical and normative arguments against capital punishment is often lost in public debate, which frequently treats the deterrence question as though it were dispositive for the underlying policy question, when in fact a resolved deterrence finding in either direction would leave Beccaria’s contractarian objection, along with concerns about wrongful conviction, cost, and proportionality, entirely unaddressed.

Current Research Directions

Contemporary capital punishment deterrence research increasingly incorporates the methodological standards the NRC report established, favoring quasi-experimental designs that can more credibly isolate causal effects over the aggregate time-series and panel methods that dominated the literature the report reviewed, though the practical rarity of executions continues to limit the statistical power available for any such design.

A second research direction examines perceptual questions directly, attempting to measure whether individuals at elevated risk of committing homicide form any distinct perception of capital punishment risk relative to life imprisonment risk, addressing the perceptual gap the NRC report identified as one of the literature’s most significant blind spots.

A third direction situates the capital punishment question within the broader marginal deterrence framework, examining whether any measurable deterrent difference between capital and non-capital severe sanctions exists once certainty and swiftness are properly accounted for, rather than treating severity as the sole relevant dimension, an approach consistent with the broader certainty-severity-swiftness framework developed throughout the deterrence literature.

Conclusion

The deterrence question surrounding capital punishment remains, more than four decades after Ehrlich’s original 1975 study, substantially unresolved on the National Research Council’s own authoritative assessment, with the existing econometric literature suffering from specification sensitivity severe enough that the 2012 NRC report concluded it should not inform capital punishment policy at all. This conclusion reflects a pattern familiar throughout the broader deterrence literature: severity-focused sanctions, of which capital punishment represents the most extreme example, have consistently proven far harder to link credibly to crime reduction than certainty-focused interventions.

Capital punishment’s practical administration, characterized by low certainty of an actual execution and substantial delay between sentencing and execution, undermines precisely the dimensions the broader deterrence literature identifies as most consequential, a theoretical consideration that predicts a weak deterrent effect independent of the econometric debate’s specific findings. Beccaria’s original contractarian objection to capital punishment, developed on grounds entirely separate from the deterrence question, continues to inform policy debate regardless of how that empirical question is eventually resolved.

Future progress on this question likely depends less on refining the aggregate econometric methods that have dominated the literature since Ehrlich and more on directly addressing the perceptual and specification gaps the National Research Council identified, an agenda that connects the capital punishment literature to the broader perceptual and methodological developments reshaping deterrence research throughout criminology.

References

Beccaria, C. (1995). On crimes and punishments and other writings (R. Bellamy, Ed.; R. Davies, Trans.). Cambridge University Press. (Original work published 1764)
Becker, G. S. (1968). Crime and punishment: An economic approach. Journal of Political Economy, 76(2), 169–217.
Bessler, J. D. (2009). Revisiting Beccaria’s vision: The Enlightenment, America’s death penalty, and the abolition movement. Northwestern Journal of Law and Social Policy, 4(2), 195–328.
Blumstein, A., Cohen, J., & Nagin, D. S. (Eds.). (1978). Deterrence and incapacitation: Estimating the effects of criminal sanctions on crime rates. National Academy of Sciences.
Cochran, J. K., Chamlin, M. B., & Seth, M. (1994). Deterrence or brutalization? An impact assessment of Oklahoma’s return to capital punishment. Criminology, 32(1), 107–134.
Donohue, J. J., & Wolfers, J. (2005). Uses and abuses of empirical evidence in the death penalty debate. Stanford Law Review, 58(3), 791–846.
Durlauf, S. N., Fu, C., & Navarro, S. (2013). Capital punishment and deterrence: Understanding disparate results. Journal of Quantitative Criminology, 29(1), 103–121.
Ehrlich, I. (1975). The deterrent effect of capital punishment: A question of life and death. American Economic Review, 65(3), 397–417.
Fagan, J. (2006). Death and deterrence redux: Science, law and causal reasoning on capital punishment. Ohio State Journal of Criminal Law, 4(1), 255–320.
Kleck, G., & Sever, B. (2018). Deterrence and the assumed relationship between severity and certainty of punishment. Justice Quarterly, 35(1), 1–29.
Land, K. C., Teske, R. H. C., & Zheng, H. (2009). The short-term effects of executions on homicides: Deterrence, displacement, or both? Criminology, 47(4), 1009–1043.
Mears, D. P., & Stafford, M. C. (2024). A theoretical critique of deterrence-based policy. Journal of Criminal Justice, 95, Article 102305.
Mocan, H. N., & Gittings, R. K. (2003). Getting off death row: Commuted sentences and the deterrent effect of capital punishment. Journal of Law and Economics, 46(2), 453–478.
Nagin, D. S. (2013). Deterrence in the twenty-first century. In M. Tonry (Ed.), Crime and justice in America: 1975–2025 (Vol. 42, pp. 199–263). University of Chicago Press.
Nagin, D. S. (2014). Deterrence and the death penalty: Why the statistics should be ignored. Significance, 11(2), 9–13.
National Research Council. (2012). Deterrence and the death penalty (D. S. Nagin & J. V. Pepper, Eds.). National Academies Press.
Paternoster, R. (2010). How much do we really know about criminal deterrence? Journal of Criminal Law and Criminology, 100(3), 765–824.
Radelet, M. L., & Akers, R. L. (1996). Deterrence and the death penalty: The views of the experts. Journal of Criminal Law and Criminology, 87(1), 1–16.
Radelet, M. L., & Lacock, T. L. (2009). Do executions lower homicide rates? The views of leading criminologists. Journal of Criminal Law and Criminology, 99(2), 489–508.
Sorensen, J., Wrinkle, R., Brewer, V., & Marquart, J. (1999). Capital punishment and deterrence: Examining the effect of executions on murder in Texas. Crime & Delinquency, 45(4), 481–493.
Yang, B., & Lester, D. (2008). The deterrent effect of executions: A meta-analysis thirty years after Ehrlich. Journal of Criminal Justice, 36(5), 453–460.
Zimring, F. E., & Hawkins, G. (1973). Deterrence: The legal threat in crime control. University of Chicago Press.
Chalfin, A., Haviland, A. M., & Raphael, S. (2013). What do panel studies tell us about a deterrent effect of capital punishment? A critique of the literature. Journal of Quantitative Criminology, 29(1), 5–43.
Berk, R. (2005). New claims about executions and general deterrence: Déjà vu all over again? Journal of Empirical Legal Studies, 2(2), 303–330.
Katz, L., Levitt, S. D., & Shustorovich, E. (2003). Prison conditions, capital punishment, and deterrence. American Law and Economics Review, 5(2), 318–343.
Shepherd, J. M. (2004). Murders of passion, execution delays, and the deterrence of capital punishment. Journal of Legal Studies, 33(2), 283–321.
Von Hirsch, A. (1976). Doing justice: The choice of punishments. Hill and Wang.