Patience and Crime

Introduction

Imagine you’re selling a car on Craigslist. The market price for the car is $15,000, but you’re antsy to sell it. You don’t like the car any more, the insurance premiums are too high, and you want cash right now. You could sell the car for $10,000 today to a buyer with cash in hand. But you recognize that foregoing $5,000 is not wise and decide to wait for what the market should pay you. You eventually sell the car for $15,000. During the waiting period, your desire to sell was high. That desire pressed on you from the moment you listed it. You just didn’t act on it. You had self-control but little patience.

Or, imagine you have a friend who drinks three beers a day and smokes marijuana every weekend. S/he coasted through school, doesn’t really have any long-term goals, and has a dead-end job. S/he lives day-to-day and saves nothing for retirement. S/he likes to ride motorcycles, hang glide, and eat fast food. S/he never settled down or got married because s/he likes his/her time for him/herself. In short, s/he is present-oriented, simplistic, physical, self-centered, and risk-seeking. Oddly, s/he has forbearance for annoyances that pester just about everyone else: Being stuck in traffic, waiting on hold for customer service, standing in line at the DMV, and so on. S/he has lots of patience but little self-control.

Patience and self-control are different constructs. Whereas self-control is the degree to which a person is “vulnerable to the temptations of the moment” (Gottfredson and Hirschi 1990:87) and “the tendency to consider the negative consequences of an act” (Gottfredson and Hirschi 2019:43; see also, Hirschi 2004:542), patience is defined as “the capacity to accept or tolerate delay, trouble, or suffering without getting angry or upset” (“Patience” n.d.). Self-control is an act, a capacity, or an eroding resource (Burt 2020), but patience is a phenomenological orientation that is not necessarily revealed in behavior. ¹ Patience implies self-control but does not require it (Schnitiker and Emmons 2007). Persons high in trait self-control can be quite impatient (e.g., the Type-A overachiever) ² , while persons low in trait self-control can display interminable patience (e.g., the laid-back idler/slacker). Although patience does appear to be an embedded feature of self-control in many cases (Peterson and Seligman 2004), this embeddedness is more a tendency than anything else, and there are significant exceptions (which we shall explore in the next section).

The present paper argues for a more nuanced recognition of patience’s role in crime’s expression and contends that its operation is both underspecified and suggestive of larger conceptual processes that advance understanding of oscillating time preferences in offender decision-making, and among persons high in criminal propensity. This more nuanced role, we believe, offers new insights into decision-making research and additional avenues for future inquiry—which we address at the paper’s conclusion.

Patience Contrasted with Self-control and Its Constituent Elements

Research suggests that self-control is a trans-situational, domain-general construct that is more stable than not (Casey et al. 2011)—although analysts have vigorously debated about the stability thesis (Baumeister et al. 1998; Burt et al. 2014; Gottfredson and Hirschi 1990; Hay and Forrest 2006). Patience, by contrast, varies dramatically from circumstance to circumstance and can be quite mercurial (Schnitker 2012). Thus, patience can be domain-dependent—with the same persons able to exercise it in one area of challenges (e.g., finding the perfect outfit for an important social event) but not another (e.g., waiting in line in traffic). Or, patience can be target-dependent—with the same persons able to exercise it for one person (e.g., an exacting boss) but not another (e.g., an irreverent teenager). Finally, patience can be state-dependent—with rested persons more able to withstand delay than the fatigued.

Although trait-based self-control does allow within-person variation, that variation refers more to change within persons over time than change within persons across situations at or near one given time (see, e.g., Hay and Forrest 2006; Higgins et al. 2009; Ray et al. 2013). And even that research concludes that trait is more important than state (see, for example, Crescioni et al. 2011) and that within-person stability is more common than change (see Forrest et al. 2019:4; Turner and Piquero 2002; but see Burt, Sweeten, and Simmons 2014).

The distinction between patience and self-control can also be examined vis-à-vis self-control’s constituent sub-traits. As a domain-general construct, self-control subsumes a constellation of attributes: impulsivity, temper (i.e., frustration intolerance), preference for simplicity, risk-seeking, self-centeredness, and physicality (Gottfredson and Hirschi 1990; Grasmick et al. 1993). Although one might presume that patience is simply the obverse of impulsivity—impulsivity being defined as the “inability to delay gratification,” “spur of the moment” acts (Steinberg 2010:218), and rapid action “without deliberation or consideration” (Pickering and Gray 1999:278)—this is not necessarily true. Time orientation and impulsivity are separate theoretical constructs (Mamayek, Paternoster, and Loughran 2017; Nagin and Pogarsky 2004). Impulsive acts can follow patient deliberation, while impatience does not necessarily lead to impulsivity.

On the first point, research suggests that impulse control is finite and depleted through exertion (Baumeister, Vohs, and Tice 2007). The more restraint one exercises in one situation, the more difficult it is to exercise restraint in another. Insofar as persons in depleted states are also less able to engage in rational calculation (Baumeister et al. 2007), decision-making will be particularly prone to impulsive bursts after exercising patience (Tittle, Ward, and Grasmick 2004). On the second point, if impulse control operates like a muscle, it is capable of strengthening after exertion through rest (see, e.g., Baumeister et al. 2006). When the gap between exertion and evocative stimuli is sufficiently long, patience may not only be possible, but expected.

It should also be underscored that patience is the capacity to accept or tolerate delay while impulsivity is the inability to delay. The capacity to tolerate something and the inability to do it are entirely different (see, e.g., Tittle et al. 2004; see also Mamayek et al. 2017). One can tolerate a bad meal but be unable to avoid eating it; one can tolerate poor hygiene in a friend but be unable to recommend him or her for a job; one can tolerate sleep deprivation but be unable to complete tasks with customary acuity. The present orientation of impulsivity is not necessarily incompatible with patience either. Mindfulness—the consummate expression of present orientation—produces patience in great degree. Focusing only on the present frees up cognitive space and permits decision-makers to operate with extraordinary clarity (Ott 2004). That focus can promote careful deliberation and the patience that goes with it. The abiding tendency for present orientation is an important, albeit paradoxical, dimension of what we believe makes patience possible—which we explore later.

If patience is not the opposite of impulsivity, then it is not the equivalent of frustration tolerance either—another constituent sub-trait of self-control. Whereas frustration tolerance refers to the amount of negativity one can handle, patience refers to how long one can take that frustration before seeking escape. Frustration tolerance and patience typically co-vary—hedonists, for example, are low in both while ascetics are high in both—but not always. Masochists, for example, have low frustration tolerance but high patience; they may not like the negativity they confront, but they will sit there and take it anyway (see, e.g., Widiger 1987). Entrepreneurs, meanwhile, have high frustration tolerance but low patience. Undeterred by multiple failures, they quickly move to the next project, with impatience being an adaptation to setbacks (see, e.g., Organ, Flaherty, and Bogue 2016).

Patience, Crime, and Time Preferences

It is widely assumed that patience and street offending do not coexist readily. Time pressure, uncertainty, desperation, impairment, and affect powerfully undermine deliberation and promote bounds on rationality (Jacobs and Wright 2010). Boundedly-rational conduct is typically portrayed as rash. Summarizing findings from several prominent studies on violent and property crime, Clarke and Cornish (2001:38) thus report that offenders frequently “seize opportunities with little forethought [and] frequently seem content to improvise solutions as the events unfold, without thinking much about alternative courses of action…” Likewise, Wright and Decker (1994:211) underscore how offenders make “hurried, almost haphazard, decisions to offend while in a state of emotional turmoil.” Pogarsky (2009:249) similarly observes that, “offenders are often intoxicated, angry, or otherwise viscerally aroused when they commit crimes…[promoting] states of mind [that] undercut and even supplant ordinary deliberative processes.” Hochstetler (2002:51) agrees, claiming that “crime is congruent with a lifestyle and outlook that prevent careful attention to consequence.” In their critique of Gottfredson and Hirschi’s (1990) views of the impulsive offender, Felson and Osgood (2008:162) compare offenders to the family dog, whose bounded rationality encourages them to “think only of the stimuli they face and the immediate payoff.” Clarke and Felson (1993:5) justifiably conclude that the portrayal of a “…self-maximizing decision maker, carefully calculating his or her advantage, [does] not fit the opportunistic, ill-considered, and even reckless nature of much crime.”

The inordinate focus on the present by crime-prone individuals such as in the portrayals above gives rise to time discounting. Time discounting is the idea that rewards obtained now are far more valuable than those for which one has to wait. That tendency is called the discount rate (Piquero, Farrington, and Jennings 2018). Criminal propensity varies directly with the discount rate, and that rate is particularly steep among serious offenders (Loughran, Paternoster, and Weiss 2012; Nagin and Pogarsky 2004). Discounting, however, may not be uniformly linear, and it is conceivable for offenders to reveal “a small discount rate when considering rewards delayed for longer periods, yet…a much larger discount rate when considering immediate rewards or those with small time delays” (Loughran et al. 2012:614; see also Mamayek et al. 2017). In this so-called “hyperbolic discounting,” offenders behave “impulsively with respect to a choice at one point in time but much more patiently when that choice is delayed in the future” (Loughran et al. 2012:614). Rewards must be reasonably certain for this to happen (Mamayek, Loughran, and Paternoster 2015), lest the delay become intolerable. Nonetheless, the possibility of inconsistency in inter-temporal preferences (that is, some offenders being able to wait a lot longer for rewards than others) implies that the perceived certainty of rewards, coupled with a longer time horizon, can separate patience from criminal propensity and bring into focus its unique role in delaying criminal action (see also Pogarsky, Roche, and Pickett 2018; see also van Gelder [2013] on dual systems processing).

Given that “dimensionality” (i.e., exploring the components of criminal propensity separately rather than treating them as one omnibus construct) is an issue of growing significance in criminology (Forrest et al. 2019), detaching patience from crime propensity (i.e., low self-control) permits improved specification of a theoretical domain that remains underdeveloped. As Mamayek et al. (2017:220) cogently argued on the need for this uncoupling, time preferences and self-control have clear conceptual synergy, but:

…researchers should be hesitant or cautious about [making self-control equivalent to] discounting because there is relatively limited criminological literature on time preferences. [Moreover,] Gottfredson and Hirschi [1990] consider self-control to be a relatively time-stable individual trait; this conceptualization is not consistent with hyperbolic time discounting because hyperbolic discounting allows the discount rate within an individual to vary over time and…can differ absolutely based on whether the outcome is a gain or a loss, whether it is large or small in magnitude,…and the framing of the outcome as either a delay or an acceleration.

Providing that specification through persons (i.e., serious offenders) in whom one should least expect either to find patience or patience variability simultaneously provides the chance to build analytic rigor through the all-important negative case (Mahoney and Goertz 2004; see also Sullivan 2011).

Target Selection and Patience

In no domain is patience more criminologically consequential than the selection of predatory crime targets. Offenders confront a massive amount of data. These data are raw, unfiltered, and rife with noise. Offenders must prioritize competing inputs, determine what is relevant and what is not, and figure out how to act on those inputs in a meaningful way. Cognitively speaking, this process is taxing. “What information consumes,” Simon (1971:40) once mused, “is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.” The process also is time pressured, which increases both the degree of difficulty and the consequences of mistakes. Mistakes include misidentification of targets, exploitation of targets that are not worth it, or selection of targets that bring more harm than good. At best, such outcomes can mean going home empty-handed. At worst, such outcomes can jeopardize one’s liberty or personal safety.

Although the study of target selection is well-trodden intellectual territory (see, e.g., Averdijk 2017; Nee and Taylor 2000), the role of patience in its enactment has been broadly disregarded. Prior research has examined the role of opportunity (Wright and Decker 1994), structural-choice (Miethe and Meier 1990), serendipity (Jacobs 2010b), discriminative cues (Brantingham and Brantingham 1978), microstructure (Shover and Honaker 1992), gender (Mullins and Wright 2003), foraging (Brantingham 2013), standing decisions (Cromwell, Olson, and Avary 1991), and scripts (Copes, Hochstetler, and Cherbonneau 2012), but patience is at best a residual category. Given the centrality of self-regulation to offender decision-making, this oversight should not be surprising: Criminologists widely presume that criminal propensity erodes patience from the supply side while bounded rationality undermines it from the demand side. Yet, hyperbolic discounting reveals that patience is possible, and the resulting non-linear inter-temporal preferences cry out for conceptual explanation. Providing this explanation improves specification of the situational foreground of crime and clarifies inter-temporal choices with regard to present gains and future gains (i.e., what I could get now relative to what I could get by waiting) as opposed simply to between present gains and future losses (i.e., what I could get now relative to what I could lose by waiting). The former is a conceptual break from many time preference studies, which often focus on how actors choose between gains and losses (for an overview of this literature and exceptions to it, see Loughran et al. 2012; Mamayek et al. 2017).

We pursue those objectives here, using the target selection processes of active auto thieves as our empirical vessel. We present two distinct patience modalities that guided respondents’ target selection decision-making: Anticipatory and forage-and-return targeting. Due to the way in which patience is ostensibly conflated with self-control, an examination of its role in low self-control (i.e., crime) outcomes is an exercise in negative case analysis integral for extending and refining theory (see Sullivan 2011). Yet, to understand patience properly as a situated state, instances in which patience is not shown must also be examined. Although this so-called negative case of a negative case (i.e., impatience) may be the default expectation among high-propensity offenders, examining it permits assessment of the working hypothesis that patience is parabolic among such persons. That is, high-propensity offenders show both an inclination to wait and a dispreference for delay that varies independently from trait self-control. Identifying the nature and reasons for this variation permits improved specification of oscillating time preferences in offender decision-making and highlights the conceptual demarcation between patience and self-control.

Methods

Data for this study come from in-depth, qualitative interviews with 35 auto thieves who operated out of St. Louis, MO. Respondents ranged in age from 17 to 49, with a median age of 26 years (the mean age was 27 years). Twenty-seven were male and eight were female. All respondents were African-American. On average, the sample completed 11 years of education. Although twelve were high school dropouts—typically leaving school by the tenth grade—the majority (n = 21) had finished high school/GED equivalency certificate, including six who completed a vocational program or some college (generally stopping after Freshman year). Two respondents were married, and more than half (n = 20) had one or more dependent children (n = 55 children, or 2.75 dependents for each with children). Only eight respondents held a legitimate job at the time of the interview; the overwhelming majority of respondents relied on crime for their income, with drug sales and auto theft being the most commonly mentioned offenses. The sheer number of respondents who reported multiple, out-of-wedlock children indicates a pattern of “analogous” conduct that is emblematic of low self-control. Intermittent employment and being unmarried also are robust indicators of low trait self-control (Gottfredson and Hirschi 1990; Wright et al. 1999).

Respondents typically committed their first auto theft by age 15 (onset ranged from 11 to 18 years), though many reported a period of riding around in cars that were stolen by someone else (e.g., a family member, neighborhood peers) before stealing one themselves. This initial, albeit second-hand, involvement meant that many respondents had some familiarity with the crime prior to committing their first offense. While it was not uncommon for respondents to steal cars from all over the metro area of St. Louis, the home base of operation for nearly all respondents entailed neighborhoods characterized by concentrated disadvantage and its attendant ills (e.g., social isolation, racial segregation, economic inequality, joblessness, and serious crime and violence) that are most endemic to north St. Louis City, and inner-ring suburbs of adjacent north St. Louis County (see Gordon 2008).

Respondents were active (non-incarcerated) at the time of data collection and were located by way of a specially trained fieldworker who employed chain referral methods to identify and recruit them. The fieldworker—an African-American male in his late twenties and intermittently-active offender—assisted the authors and other colleagues on multiple other projects in the research setting over many years. As a long-time offender and trusted participant in the local street scene, he was adept at finding hard-to-reach persons involved in predatory crime. The current project was no exception.

For this study, the fieldworker began by recruiting suitable participants from his immediate network of street offenders (e.g., typically friends or associates), on whom he then relied to provide introductions (i.e., referrals) to active auto thieves who were less familiar with him at the time. He explained our research objectives and screened prospective participants for inclusion in the study. The fieldworker then accompanied eligible and cooperative participants to and from the interview, at which point he was compensated for his efforts ($75 per referral). To encourage (but not unreasonably influence) participation, and to offset opportunity costs from giving their time to talk to us, respondents were offered a monetary sum of $50. The fee seemed prudent given how active offenders are typically not inclined to do anything for nothing. Recruitment fees and all other research procedures complied with ethical standards and guidelines for the protection of human subjects and were approved by the Principal Investigator’s University Institutional Review Board.

Sampling was purposive as the fieldworker was instructed to recruit both males and females, younger offenders (aged 16 at minimum) and older ones, and offenders from different social networks—to avoid a homogeneous, atypical catchment of respondents (cf. Heckathorn 1997). Specific criteria—based on recent, historical, and projective involvement in auto theft—were interwoven into this process to guide recruitment for the study. Our goal was that would-be respondents had: (1) committed at least one auto theft in the month prior to being interviewed; (2) committed five or more auto thefts in their lives; and (3) considered themselves to be actively involved in auto theft by an expressed readiness for continued involvement in auto theft. As far as the fieldworker knew, these were strict criteria to follow when identifying and recruiting participants to meet with us ³ ; what we knew however, is that these criteria were more guidelines than anything else that in practice would be flexibly adhered to so as not to turn away hard-to-reach respondents who could provide potentially valuable insights. Despite the occasional exception, most of the respondents met all three criteria. ⁴

In numerical terms, our sample of 35 was guided by the concept of theoretical/data saturation which, in qualitative research, is more important than overall sample size. Theoretical saturation occurs after responses to questions about a general area become broadly repetitious (Strauss 1987). Achieving saturation is relevant in qualitative inquiry because it indicates that additional (and new) data are not emerging, that appropriate empirical ground has been covered, and that typology building is reasonably complete. Generally speaking, saturation can be reached long before 35 interviews (Guest, Bunce, and Johnson 2006), and that is what we found here. After about 20 interviews, responses on many higher-level themes (including target selection) started to become broadly redundant; sampling continued, however, so as to focus on underrepresented groups (namely female auto thieves; see also Mullins and Cherbonneau 2011:283), and lower-level themes. ⁵

Interviews were conducted in a private room and lasted about 75 minutes on average. Questions were semi-structured and open-ended in design, and focused on the immediate phenomenological context in which crime involvement and event decisions were made (see Clarke and Cornish 2001; Katz 1988; Wright and Decker 1994, 1997); it was here that discussion pertaining to target selection and inter-temporal choice—either implied or direct—arose. ^6,7 Respondents’ gave us permission to record their interview. The recordings were transcribed verbatim by hired transcriptionists, and afterward checked for accuracy and edited as necessary. To ensure anonymity and to further help put respondents at ease, we asked them to provide an alias and made sure they understood that no legally identifiable information would be specifically sought or recorded. ⁸

Data were coded manually using traditional domain analytic techniques (Spradley 1980). Grounded theoretic techniques (Strauss 1987) were then used to link the data to broader criminological principles within the offender decision-making literature. Although grounded theory traditionally is thought to be exclusively inductive, it can vacillate between inductive and deductive. As Cherbonneau and Jacobs (2019:1128) observe, “the process of constant comparison—going from the data to theory and then back to the data—permits identification of thematic patterns that extend and refine theory. Although extant theory…is engaged, that engagement is made possible only because of the identification of emergent patterns in the data that are then juxtaposed against the relevant concepts of inquiry….” The “relevant concepts” in the present study revolved around time preferences and delay tolerances in target selection.

To be sure, it was through this process—and only after the data were already collected—that the conceptual import of patience was recognized. As a result, more granular information about the thought processes of respondents between the time they first spotted the target and the time they stole it is not always available. While we have endeavored to provide as much backstory on these processes as possible (both in the main text and in respondent quotations), additional granularity is not necessary to meet the study’s current analytic objectives: We are addressing patience as it relates to deferring gratification (or not) more generally, so even without comprehensive information on respondent’s interim thought processes, the delay between reward exposure and seizure is the broader conceptual issue that our data permit us to examine.

Throughout the research we were sensitive to the role of positionality, and that sensitivity began with the interviews themselves and carried over to the analytic process. Positionality in social research is the idea that a researcher’s background, voice, upbringing, demography, or culture can influence what questions are asked, how interviewees respond, and the manner in which responses are analyzed (see, e.g., Bucerius 2014; St. Louis and Calabrese Barton 2002). During interviews, our goal was to make respondents the experts; they would teach us and convey that knowledge in a cooperative and nonjudgmental setting. Respondents were the most integral part of the research process, and interviews were structured with this in mind. In distilling the data into thematic areas and then linking the data to theory, we were similarly mindful of respondents’ voice. Our goal was for our findings to speak through those who provided the data. Although positionality can be a challenge in interview-based research, sensitivity to it can counteract some of its pitfalls.

One final note before presenting our analysis concerns the external validity of our findings and conclusions. As with all non-probability samples, we make no claims of generalizability. A sampling frame for offender populations most certainly does not exist and the parameters of the population are not known and not reasonably knowable (cf. Glassner and Carpenter 1985). Although our sample drew from uncaught, non-incarcerated offenders, we also do not claim that they are demonstrative of “successful” or “professional” auto thieves (a number of respondents had been arrested and convicted of auto theft). ⁹ Nor do we claim that respondents were auto theft “specialists,” given that many participated in a variety of offenses aside from auto theft (including, drug sales, robbery and carjacking, burglary, and theft more generally). ¹⁰ We are confident however, that our purposive sampling strategy and application of recruitment guidelines generated a sample of committed auto thieves evidenced by past, present, and (intended) future involvement in auto theft.

Patience in Target Selection

The world of street crime attracts persons with high criminal propensity and then amplifies those tendencies with norms that place enormous emphasis on decisive action. Patience is broadly disfavored, as it means deferring gratification or sacrificing rewards altogether. If exercised strategically, however, patience can bring dividends—particularly in the selection of crime targets. Patience can mean the difference between securing a great reward and a mediocre one, or between securing something and nothing at all. Offenders on some level realize this (Wright and Decker 1994) even if they don’t faithfully pursue the strategy on a regular basis. Two target selection modalities emerged with patience in mind: Anticipatory and forage-and-return.

Impatience in Target Selection

The above data indicate that persons high in criminal propensity can, and in fact do, demonstrate patience. But if patience is more situational than static, consideration of cases in which patience is not shown permits assessment of the working hypothesis that patience is a situated capacity that varies independently from trait self-control. Given that trait self-control is neither necessary nor sufficient for state patience, demonstrating how, when, and why offenders disprefer delay clarifies patience as a construct and improves specification of oscillating time preferences in offender decision-making. No situated force undercuts the inclination to delay more than opportunity, whose time-limited nature puts a premium on immediate action.

Discussion

In criminology, the role of patience in crime has been underspecified because street criminality and patience are not thought to coexist readily and because its conceptual counterpart, impatience, has long been equated with impulsivity and analogous crime-producing forces (Gottfredson and Hirschi 1990, 2019). Using the target selection processes of active auto thieves as our empirical vessel, this paper has argued for a distinction between patience and self-control and explored the variation of state patience among persons with low trait self-control (i.e., high criminal propensity). Patience and self-control are distinguishable at the conceptual level, while patience is distinguishable from self-control’s constituent elements (e.g., frustration tolerance, impulsivity) as well. Our data suggest that patience oscillates upward and downward and is parabolic among high-propensity offenders. ¹⁶ Need, opportunity, ambient risk, and discrete reward cues mediate this oscillation.

Our findings revealed two distinct patience modalities that guided respondents’ target selection decision-making: anticipatory and forage-and-return targeting. In anticipatory targeting, respondents chose particular theft venues, went to those venues, waited for targets to appear, and seized specific ones after certain criteria were met. In forage-and-return targeting, respondents circulated within and around diffuse geographic areas to identify vehicles they would seize later; the gap between target identification and seizure was sizable. Patience of this kind was unexpected given that deferred gratification is broadly inconsistent with criminal propensity. Time discounting is particularly common among committed offenders such as ours (see, e.g., Cromwell et al. 1991; Jacobs 2000; Shover 1996; Wright and Decker 1994): They want what they want, and they usually want it then and there (Nagin and Pogarsky 2004). Allowing suitable targets to pass—or at least, delaying their seizure—is akin to foregoing opportunity that would otherwise translate into fast and appreciable rewards.

Among street-level offenders, foregoing immediate gains is a potential loss because of the anomic nature of motivation—no amount of money ever is enough (Jacobs et al. 2003)—and because there is no guarantee that a non-stationary target banked for later (such as a vehicle or person) will remain available at the moment the offenders want it. Ambiguity of this kind is strongly disliked because decision-makers in general, and offenders in particular, rarely want to give up what is otherwise perceived to be a sure thing (Ellsberg 1961; Kahn and Sarin 1988). For offenders, ambiguity aversion for gains is so powerful that their vaunted impulsivity may, on some level, be uncertainty reduction in disguise: The need to act quickly produces behavior that is rushed or reckless when the real reason might be to avoid the cognitive dissonance associated with foregoing a certain gain in light of uncertain future prospects.

And, in fact, our data also highlight a strong situated dispreference for delay, as we demonstrated through cases of impatient targeting. This dispreference was primarily a function of opportunities too good to pass up or opportunities kindled by need. The dispreference, however, was rooted in time sensitivity, not frustration intolerance. Recall that frustration tolerance refers to the amount of negativity one can handle, while patience refers to how long one can take that negativity before seeking escape. Rapid-acting respondents were less frustrated than they were impatient. Purely opportunistic thefts were often serendipitous and involved no frustration at all. In opportunistic thefts where needs put opportunity into focus, the “needs” were far from truly frustrating. Stealing a car because one was stranded or simply did not feel like walking is not frustration intolerance. It instead reflects the snap recognition that a perceived imbalance can be remedied with ease. ¹⁷ Recognition of a negative state is distinct from intolerance to it, and intolerance is neither necessary nor sufficient for that recognition. Indeed, our respondents typically gave no clear indication that frustration intolerance was the central force behind these thefts. Auto theft was a coolly rational response to the realization that delayed gratification was unnecessary. It may have been impatient, but it was not reckless. It prevented frustration rather than was a response to it.

Impatience should not be mistaken for impulsivity for the same reason. If impulsivity is the inability to delay, impatience is the disinclination to accept delay. Respondents who acted opportunistically to sate needs (real or imagined) were not unable to delay. Rather, they smartly recognized that delay was unnecessary, foolish, or even counterproductive. Consciously overriding delay to remove a deficit or extend a surplus is not knee-jerk or lacking in calculation (Tittle 1995)—the hallmarks of impulsivity. It was instead quite rational and reflective (Paternoster and Pogarsky 2009). Thoughtfully reflective decision-making need not be slow, and quick conduct is not necessarily rash. Whereas impulsivity would trigger a basic inability to recognize the existence of alternative solutions (e.g., taking the bus or getting a ride) or a reckless carjacking, neither modally happened here. The offenders’ response was rapid, but it was calibrated to the moment.

Just as impatience is not equivalent to impulsivity, patience remains possible against a general backdrop of present-orientation. The patience expressed here was near-term—measured in hours (or occasionally days) as opposed to weeks, months, or years—but it was evident nonetheless. Impulsivity is thought to be binary (you’re either impulsive or you’re not), ¹⁸ which makes it difficult for crime propensity to co-exist with time delays of almost any kind. Patient target selection shows that gratification can be delayed yet still reasonably quick in coming. Although anticipatory and forage-and-return targeting did not involve reflexive action (as impulsivity seems to require), the time horizon was short-term and rooted in perceptibly certain rewards. Relaxing impulsivity’s assumption that action be instant permits reconciliation of offenders’ global present-orientation (which is broadly illustrative of a general dispreference for delay that is at times expressed through purely impulsive conduct) with their strategic enactment of patience.

Criminal efficacy, which our respondents had in ample amounts, ¹⁹ puts both poles of our findings (patient and impatient) into focus and helps explain the pattern we found. On the one hand, efficacy promotes three tendencies that create the opportunity structure for time delay: automaticity, superoptimism, and versatility. Automaticity allows offenders to absorb, process, and act upon environmental cues rapidly (for an overview, see Nee and Ward 2015). Superoptimism engenders confidence that things will ultimately go one’s way even if they don’t go that way initially (Walters 1990). Versatility makes it possible for offenders to adapt tactically, spatially, or temporally when circumstances call for it (Jacobs 2010a). Together, these tendencies engender comfort that there is, in fact, a way out if one is unable to act right now (which is what permits delay). On the other hand, because efficacy dramatically accelerates the conversion rate between cue recognition and target selection, delayed gratification is unnecessary when circumstances suggest that rapid action is proper. Efficacy, in other words, makes short-term time preferences feasible even as it provides a cognitive rationale for delay when immediate action is not possible.

Our findings have important implications for dual process models of criminal decision-making (van Gelder 2013). It appears that an important (but not exclusive) factor distinguishing patience from impatience is whether respondents were actively looking for targets or whether they stumbled into them while engaged in other pursuits. The former encouraged offenders to bide their time for the right moment, while the latter—particularly when combined with situational provocations—evoked hot responses to stimuli, real-time. Research on dual process models of decision-making suggests that cool and hot modes of decision-making operate simultaneously and that when crime occurs, it typically is because hot triumphs over cold (van Gelder 2013:753-55), but our data offer a more nuanced alternative. The hot choice to be impatient (seize opportunities now) was paradoxically volitional and quite rational, while the calculated choice to wait (despite having targets in sight or in hand) was demonstrably cool in the face of hot circumstances. Temporal preferences flowed from orientation toward the target, that is, whether the offenders pursued the target or the target came to them. Although prior research suggests that environmental stimuli are implicated in both hot and cold behavior (van Gelder 2013:753) and that the effects on decision-making are thought to move in one direction (that is, hot affect produces reflexive behavior, cool affect promotes contemplative choice), our data suggest that it may be the decision-making posture that determines whether offenders’ orientation is hot or cold, not vice versa. If that posture is, in fact, independent from hot/cold orientations and even shapes those orientations, then assumptions about what directs both rational choice and “hot” reactions have been imprecisely specified.

Our research raises but does not answer the question of whether patience operates differently depending on whether time preferences relate to obtaining pleasure (i.e., rewards) or avoiding pain. Our data have focused primarily on securing rewards. But is, for example, the patience required of someone sitting next to a unruly child on a three-hour plane ride qualitatively different from the patience asked of someone like Jewels Santana to wait at a gas station until the “right car” arrives? Research on reinforcement suggests that decision-makers are more impatient to avoid negative stimuli than they are to secure rewards. Prospect Theory specifically shows that people are substantially more averse to losses than they are eager to realize gains (Kahneman 2011). Yet, there are decision-makers who dread bad outcomes and want the negativity to come earlier to “get it over with” (Mamayek et al. 2017). For such persons, time preferences reverse when facing negative stimuli. Future research must disentangle whether temporal preferences operate differently for pleasure vs. pain and what individual and situational factors mediate this process.

Finally, our research suggests that the “now versus later” phenomenon central to conceptualizing self-control failures and successes (Burt 2020) warrants a second look. Our data reveal that it is not necessarily a self-control success to forego rewards, nor is it a self-control failure to act impatiently. To illustrate, waiting to eat a cupcake until after dinner might maximize one’s pleasure and/or satisfaction from dessert, but if one is 20 pounds overweight and trying to cut down on sweets, one ate a cupcake s/he should not have eaten. Patience, in other words, optimized one’s subjective utility but did nothing to advance his/her self-control. Auto thieves waiting to seize the right car (such as Jewels Santana, Lil’ Bunny, or Capone) technically “gave in” by stealing (a “self-control failure”), but since they stole what they wanted and when, patience maximized their criminal utility. At the same time, delay-averse offenders like Killa and D-Cuz were not necessarily self-control failures given the background work they undertook to make their opportunistic thefts possible. Such efforts appear to be purely hedonistic (e.g., hanging out with friends, going places, doing things, being mobile), but they require penetration in multiple activity nodes over time, persistent alertness, and no promised payoff—all of which require discipline. Rapid reaction to opportunities that do, in fact, arise allows offenders to cash the checks that such efforts write. This is not a self-control failure but a calculated, pragmatic response to time-limited reward cues against a backdrop of pressing need.

In the final analysis, (im)patience represents the fulcrum between impulsigenic and volitional control processes whose joint consideration enhances understanding of the “impelling and restraining forces” that structure offender decision-making itself (Burt 2020:57). This is the same fulcrum implicated in studies of decision-making under risk and which potentially provides an answer to one of the most vexing crime control questions in all of criminology: Why crime reductions in small hot spots of crime are so meager even when the police are, for all intents and purposes, “right there.” ²⁰ Sensitivity to time permits offenders to figure out where risk is in order to defeat it—which causes crime to persist in hot spots, just not when the police are around. The displacement literature broadly corroborates this argument. ²¹ Such temporal flexibility maps well with the current findings and reveals the decision-making agility that criminal propensity provides. The notion that serious offenders are “all gas and no brakes” (cf. Bell and McBride 2010) fails to account for how the horsepower of crime propensity allows offenders to leverage “engine drag” to slow themselves down simply by easing off the pedal. But since the pedal remains under foot, resumption of speed is always close at hand—allowing offenders to toggle between patience and impatience real-time.

Patience remains an understudied concept in criminology and is distinct from trait self-control and two of its core constituent elements—frustration tolerance and impulsivity. Although patience and self-control certainly are related (Mamayek et al. 2017), and self-control may even give rise to patience in some circumstances (Peterson and Seligman 2004), the two constructs are far from equivalent. Future research promises to expand this nascent but growing line of inquiry by detaching patience from existing dimensions of self-control and by continuing to explore oscillating time preferences in offender decision-making and their broader consequences for criminal behavior.