This second (of two) posts evaluates Rational Choice Theory and, by extension, any New Right / Right Realist theories based on the notion of rational cost-benefit analyses of criminal motivation.
A list of all the relevant bits to save you having to read through the rest of the post…
• Rational Choice Theory (RCT) reminds us that an understanding of social action – how and why people make certain choices – is important for an understanding of criminal behaviour. It also focuses our attention on crime as a rational process.
• A cost-benefit analysis of offending fits neatly with a common sense understanding of criminal behaviour. It also underpins a range of contemporary crime theories – such as RCT, Broken Windows and Routine Activities – that can be generally characterised as Right Realist. It has, however, serious limitations related to how offenders receive and process information, particularly in time-limited situations.
• An alternative and, according to Simon (1956), more realistic way to understand the behaviour of offenders, is to see it in terms of a bounded rationality. Interview evidence, for example, suggests burglars evaluate alternative forms of behaviour within what Walters (2015) calls “the limits of their knowledge and abilities”. Offenders, in this respect, seem to make “rational enough” decisions based on a range of “rule-of-thumb” beliefs and experiences.
• If a cost-benefit model of criminal decision-making is invalid, this has important ramifications for both crime-control theories and situational crime prevention techniques and strategies. More specifically it suggests that if offenders do not rationally weigh likely benefits against potential costs any attempt to lower the former and raise the latter will have only a limited long-term effect on crime.
The Longer Read
In thinking about how to evaluate Rational Choice Theory (RCT), it’s useful to begin by noting it had a number of attractions for all kinds of interested parties: politicians, control agencies such as the police and, unusual as it may seem given we’re talking about criminological theories, the general public.
On a general level it had “face validity” – its central argument seemed eminently plausible: offenders are rational individuals who calculate the likely benefits of crime against its possible costs, such as the likelihood of being caught and punished. When an opportunity occurs and the odds are favourable, then offending will occur; if the odds are against, offending is unlikely.
As an added bonus, RCT had a simple underlying clarity that made it publicly accessible in ways that many criminological theories are not. The theory seemed to “make sense” because it explained the process by which individuals became offenders when the odds were in their favour and didn’t offend when they weren’t. If offenders are not demonstrably different, sociologically or psychologically, to non-offenders then what makes the difference is the environment within which behaviour takes place.
If the environment is “conducive to crime” – it’s relatively unprotected and unpoliced, no-one seems to care about what people are doing there, crime is relatively easy, risk-free and painless – then it “makes sense” that some people will take advantage of the environmental opportunities to “act in their own interests” by offending. This recognition of the importance of physical and social environments also meant RCT offered a range of relatively-simple, highly-practical, measures to make “choosing crime” less attractive.
It suggested, for example, clear and effective forms of crime management and control: if you deny potential offenders the opportunity to commit crimes, through practical measures such as target hardening, you both reduce the benefits and increase the costs of crime. As a consequence, crime will inevitably fall.
Which, from the mid-1990s onwards in both the UK and USA, it did.
Statistically at least (because there are a range of factors involved in the social production of crime statistics that make the claim of “falling crime” at best, questionable and at worst, highly unreliable).
Be that as it may, RCT (and related theories such as Broken Windows and Routine Activities) undoubtedly leant itself to a wide range of situational crime prevention strategies and techniques: if crime was a rational choice that involved individuals making a cost-benefit analysis of a particular environment – weighing-up, as it were, the pros-and-cons of offending – then by changing the physical environment in ways that made crime more difficult and more costly you would, at best, deter crime and, at worst, manage it in ways that made it much less likely.
While it’s not difficult to accept that a great deal of, if not necessarily all, crime involves the potential offender making rational choices about their behaviour, we need to dig a little deeper into what we might term “the nature of those choices” in order to expose weaknesses in the RCT argument. One way we can do this is by using an example of a type of crime – burglary – that on the face of things should provide the theory with some of its most-compelling support, mainly because the evidence suggests it’s a crime that involves:
• careful planning
• clear risk assessments
• consciously choosing targets that maximise rewards while minimising costs.
Nee and Meenaghan (2006), for example, suggest burglary generally involves a “sequential decision-making process or offence-chain” with two broad stages:
• a general decision to offend
• deciding upon the right location for the crime
• choosing a specific crime target.
2. A second stage, prior to actually committing the crime, that involves using a range of “learned responses to visual cues to discriminate between potential targets”. These involve assessing four main things:
• the relative wealth of the occupiers based on the location of their property
• the different times of residential occupancy: unoccupied property being preferred for obvious reasons
• ease of access into and out of the property
• the level of security – alarms, cameras, locks and the like – used to protect the property.
This dual process, as Nee and Meenaghan (2006) suggest, is one that reflects an underlying “model of target appraisal in the burglar which is rational and discriminating in nature” – an observation that, on the face of things, provides substantial support for the RCT argument.
The question here, however, is not the extent to which offending is rational, per se. Rather, it’s the extent to which rational offending can be attributed to a cost-benefit analysis of crime, one where the potential offender carefully weighs risks against rewards: and if it can’t, this calls into question any theory – from Rational Choice to Broken Windows – based on this type of argument.
Costs and Benefits
As we’ve previously noted, the classical cost-benefit analysis of offending is one that has potential offenders weighing the desirable features of crime against its less-desirable features and making a rational decision based on this assessment.
While this seems a quite logical calculation – why would anyone offend in a situation where the potential costs far outweighed any likely benefits? – the main problem for offenders is the vast majority will have little or no objective information about actual costs or benefits in advance of committing an offence (after which, of course, they will have such information – an important, if frequently overlooked, idea we will develop in a moment). The question here, therefore, is the extent to which offending behaviour could be considered remotely rational in such a context? If you are not – indeed cannot – be in full possession of all the relevant factors in this decision-making process how “rational” can the decision actually be?
One solution to this problem, as Wilson and Kelling (1982) acknowledge, is to argue cost-benefit analyses may be subjective rather than objective. In other words, what’s important is not the actual, calculable, chances of being caught, for example, but rather what the individual offender believes those chances to be. If someone believes the calculable costs of committing a crime are negligible – even if, in reality, they are extremely high – they will still offend.
While the idea of cost-benefit analyses being subjective resolves the issue of “inadequate information” – and also goes some way to explaining why, if offenders are rational, calculating, individuals they still make “bad choices” (choices that result in their arrest, imprisonment and the like) – it also introduces a new, less easily explained, problem, namely that if risk assessments are subjective, there’s no requirement for them to be rational. Subjective cost-benefit analyses also, as Weinstein (1980) has demonstrated, have a tendency towards optimism bias. That is, they tend to be wildly over-optimistic, with potential benefits over-predicted and possible costs underestimated.
The notion of subjective cost-benefit analyses may, therefore, be misleading in the context of Rational Choice Theory if we accept that irrational beliefs (“There is no risk of being caught”) effectively carry as much weight as rational ones (“If I try to rob a police officer there’s a high probability of being caught”). One way to resolve this problem, therefore, is to retain the notion of rational behaviour in offending, while replacing the idea of a cost-benefit analysis with something more realistic and reflective of offenders’ actual behaviour.
As various studies (Palmer, Holmes and Hollin (2002); Nee and Meenaghan (2006)) have shown, interviews with convicted burglars reveal their crime choices aren’t based on cost-benefit analysis. Rather, these choices are made in the context of a “bounded rationality”: the idea that while the offenders’ behaviour was rational – burglaries were planned, risks were minimised and so on – this was rational behaviour within certain clearly-defined and to some extent controlled, psychological and sociological limits or boundaries. The argument here is that our ability to make rational decisions is limited (“bounded”) by three major considerations:
In terms of criminal situations, an offender’s knowledge is likely to be limited, partial, unreliable or plain wrong and if this is the case the ability to make a rational decision – particularly one supposedly based on a careful cost-benefit analysis – is similarly limited.
Most forms of crime – from opportunistic robberies to carefully-planned burglaries – are time-sensitive. Decisions generally have to be made within a very short time-frame and this is not conducive to carefully-considered, rational, cost-benefit decision-making.
3. Individual Cognition
Simon (1956) argued that rational choice theory is a psychologically unrealistic model of human behaviour, mainly because the variables involved in even a very simple decision-making process are too complex for the individual to accurately assess.
In complex decision-making processes, where many of the major variables are unknown, can’t be adequately determined or their significance comprehended and evaluated, Simon argued it was all-but-impossible for the individual to successfully carry-out a rational cost-benefit analysis.
The argument here, therefore, is that our ability to make rational decisions is limited or bounded by social factors, such as access to information or the time-scale involved in decision-making and psychological factors such as our cognitive ability to process and evaluate complex sources of information quickly and effectively. All three of these factors are, of course, an important feature of offending.
Bounded rationality, therefore, suggests there are limits to our ability to rationally access situations and while this doesn’t mean offending is irrational, it does suggest a different process is at work, particularly in relation to crime. This, Simon suggests, involves “satisficing”: a rational thought process that involves a rapid, quick-fire, search for solutions to potential and actual problems that are simply “satisfactory” and “sufficient” to any given individual in any given context.
In relation to burglary, for example, the argument here is that decision-making is governed by a bounded rationality, based on prior knowledge, learned skills and personal experience. As Nee and Meenaghan (2006) put it: “[Burglars] operate using the same bounded, almost habitual decision-making processes that all experienced individuals use to navigate quickly and effectively around their world. This does not involve a time-consuming cost-benefit analysis of the pros and cons of each action, but relies on the use of speedy rules of thumb based on prior learning”.
In other words, the burglar brings to the situation a range of context dependent, highly-specialised, skills that move them quickly and efficiently through a time-pressured process. Their experience and situational familiarity allows them to make instantaneous, semi-automatic, decisions based on their experience of, or “a feel” for, similar situations.
Looking at offending in this way, bounded rationality has two main advantages over its cost-benefit counterpart:
1. It takes into account the experience of the offender. Naïve offenders are likely to be much less successful not because they have miscalculated the costs but because they have less experience in encountering, minimising and resolving unpredicted risks, such as someone arriving at the house when they’re still inside.
2. It can account for changes in criminal behaviour, both in terms of crime displacement and the rise or fall in different types of crime.
In this respect, a weakness of the cost-benefit hypothesis is the prediction that increasing the costs and reducing the benefits of crime, reduces offending: the harder the target, the less likely the crime. The problem here is that the available evidence doesn’t support this argument.
After an apparently long, slow, decline since the mid-1990s (at least according to increasingly unreliable crime statistics) crime in the UK is rising. This suggests potential offenders, when faced with harder targets, don’t simply “give up”; rather, they find alternative, easier, targets or, as seems to be the case for burglary, they develop new techniques to circumvent crime controls. They may also, given the massive increase in online crime over the past decade, simply move-on to easier, less risky and potentially more beneficial targets…
Wilson and Kelling (1982) Broken Windows
Nee and Meenaghan (2006) Expert Decision Making in Burglars
Walters (2015) The Decision to Commit Crime: Rational or Nonrational?
Simon (1956). Rational choice and the structure of the environment
Weinstein (1980) Unrealistic Optimism About Future Life Events