Perceived Sanction Threats and Projective Risk Sensitivity: Auto Theft, Carjacking, and the Channeling Effect

Abstract

Although sanction threats promote fear, among committed offenders, that fear can become a resource with which to sculpt emerging crime preferences. In such cases, criminality is not deterred but channeled. We explore the channeling process here as it relates to auto theft and carjacking. Our qualitative findings reveal that auto thieves are reluctant to embrace the violence of carjacking due to concerns over sanction threat severity they attributed to carjacking—both formal (higher sentences) and informal (victim resistance and retaliation). Meanwhile, the carjackers are reticent to enact auto theft because of the more uncertain and putatively greater risk of being surprised by victims, a fear that appears to overcome the enhanced long-term formal penalty of taking a vehicle by force. We examine the implications of offenders’ decision-making for the analytic intersection of rational choice and deterrence, offering the notion of projective risk sensitivity to encapsulate the process.

Keywords:

• restrictive deterrence
• rational choice
• offender decision-making
• motor vehicle theft
• carjacking

Acknowledgements

We would like to thank Richard Wright and VolkanTopalli for their assistance and collaboration at various stages of the grant-writing and research process that resulted in the carjacking (Richard and Volkan) and auto theft (Richard) data-sets used in this article. We also would like to thank the anonymous reviewers for their helpful comments and criticisms.

Notes

¹ As Wright et al. (2004, p. 184) illustrate, “Deterrence will best inhibit the criminal activity of those who are actively at risk of offending. Those who are effectively inhibited from crime by other [informal] considerations will be immune to the threat of punishment.” Other researchers have similarly observed that deterrence may be limited to “marginal” populations (Toby, 1964; Zimring & Hawkins, 1968). By contrast and somewhat paradoxically, non-offenders may not be good candidates for deterrence insofar as they fail to contemplate crime in the first place. They are deterred but not necessarily “deterrable” (see Gibbs, 1975, p. 89; Thomas et al., 2013; see also Pogarsky, 2002 on acute conformity).

² We would like to thank an anonymous reviewer for making this point.

³ This argument is logically similar to Sutherland’s (1947, p. 7) ninth proposition of Differential Association which recognizes the futility of any attempt to discriminate among criminal and non-criminal behavior in circumstances where similar dynamics underpin both phenomena: “While criminal behavior is an expression of general needs and values, it is not explained by those general needs and values since non-criminal behavior is an expression of the same needs and values.” The difference of course is one of distinction: Criminal and non-criminal behavior for Sutherland, and auto theft and carjacking for us.

⁴ Due to the technical/logistical problems with two of the transcripts, the final total for the present paper was 26.

⁵ Eligible participants were those who had committed at least two carjackings in the previous year and had not ruled-out the prospect of committing additional carjackings at any point in the future. These criteria were necessarily more liberal than those employed in the auto theft study due to concerns over the crime’s rarity, juxtaposed against the need to build a suitable sample of currently active carjackers.

⁶ These data were collected with regularity based on the often-recited supposition in the auto theft literature that as vehicle security advances and cars become harder to steal, motivated offenders may turn to carjacking (Beekman, 1994; Clarke & Harris, 1992, p. 46). Although auto theft participation is likely among carjackers, similar figures could not be determined from that data-set due to the inconsistency with which such questions were asked. We would be surprised if many of the carjackers had not at some point been involved in auto theft or accumulated auto theft-like experiences (e.g. riding in a stolen car, being present while peers stole cars, etc.) due to the frequency with which such crimes occur in a group context in adolescence (see Cherbonneau & Wright, 2009 for a review of this literature), along with the fact that early auto theft involvement is common in the lineage of “chronic” criminal careers (Svensson, 2002).

⁷ We know of at least one case where this was true and suspect two others. There is no way of knowing precisely how many offenders were interviewed for both projects; anonymity built into the study precludes such knowledge. From personal experience working on these projects and others like it, along with the six-year separation in sample selection, we believe cross-project overlap to be rare (see also Mullins, 2006, p. 35).

⁸ The carjacked vehicle in this incident was equipped with OnStar, which is a built-in security device that provides, among other things, subscription-based stolen vehicle assistance (e.g. GPS tracking and remote ignition disabling). As End Dog described, shortly after completing the carjacking, OnStar personnel interrupted on the vehicle’s communication system to warn that the seized vehicle was being tracked, that law enforcement was en route to intercept it, and that its occupants should abandon the stolen vehicle—which End Dog summarily did.

⁹ Goldie was the only verified member of both the auto theft and carjacking samples. His unfavorable view of carjacking may be an artifact of being shot in retaliation for a prior carjacking several years earlier. After shooting, running over, and taking the drug dealer-victim’s car, Goldie was tracked down and sprayed with bullets by the dealer. The gaping wound in Goldie’s abdomen had healed by the time he was interviewed for the auto theft study, but the auto theft interviewer was not aware of this history and thus could not inquire further.

¹⁰ Although Corleone and Goldie’s comments could insinuate a potential lack of skill vis-à-vis auto theft, it is important to keep in mind that the overriding concern is enactment time. Regardless of skill level, enactment time will almost always be shorter for carjacking than for auto theft because of the efficiency with which target selection, enactment, and escape merge in space and time. The only exception to this rule in auto theft might be instances where offenders seize vehicles left running but vacant (see Jacobs & Cherbonneau, 2016). Even so, the highly opportunistic nature of this enactment style belies the technical and practical demands associated with stealing a parked vehicle off the street—the general approach that Corleone and Goldie clearly emphasize.

¹¹ Note the contrast here with Goldie’s sentiments earlier in the paper about carjacking. The above interview took place several years before the auto theft interview and also before being shot in retaliation for a carjacking.

¹² The quotes by Loco and Low have appeared (either in full or in part) previously in published work (Topalli & Wright, 2012) but not in the context of channeling effects, nor for the purpose of systematically examining how formal and informal sanction threats mediate the auto theft/carjacking decision.

¹³ We are not suggesting that offenders are not engaging in risky behavior by offending but rather, point out the role that risk sensitivity plays once criminal involvement has begun.

¹⁴ We should note that our view of projective risk assessment is distinct from dual-process and -system theories of decision-making (Mamayek, Loughran, & Paternoster, 2015; Van Gelder, 2013). Whereas these models focus on the manner in which “cool” decision-making and/or self-control acts to restrain hot emotions and impulsivity (and thereby inhibit crime), projective risk assessment attends to how threat sculpts criminal propensity to channel crime that is committed.

¹⁵ We thank an anonymous reviewer for this insight.

¹⁶ Once again, we are grateful to an anonymous reviewer for making this point.

¹⁷ We fully recognize the inferential problems of our findings relative to sampling on the dependent variable. Our study does not provide a mechanism by which to gauge when and under what circumstances auto thieves’ and carjackers’ decision-making processes were not channeled by sanction threats, nor do they reveal would-be offenders who were absolutely deterred. Indeed, absolutely deterred offenders would have appeared in neither of our studies and would therefore have no opportunity to “channel.” The exclusion of such offenders could imply that offenders are more risk-adaptive than they really are. We remind readers, however, that findings from both of our samples were serendipitous. We had no expectations that channeling effects would appear in either study, and these findings could not reasonably have emerged without targeted interviews. Targeted interviews may be the ideal way to provide richly detailed information about causal processes relating to narrow phenomenon that might never realistically emerge in large random samples (Forgues, 2012). This is particularly true in active offender research, where large random samples are neither feasible nor practical since the parameters of the population are unknown (Glassner & Carpenter, 1985). Like most qualitative research, our findings should be considered preliminary but provocative; they are intended to stimulate further quantitative inquiry on the structure and process of channeling.

Additional information

Notes on contributors

Bruce A. Jacobs

Bruce A. Jacobs is a professor of Criminology at the University of Texas at Dallas. He studies offender decision-making.

Michael Cherbonneau

Michael Cherbonneau is an assistant professor of Criminology and Criminal Justice at the University of North Florida. His research interests lie in the theoretical and empirical understanding of crime and offenders with an emphasis on foreground influences on offending.