Revisiting the crime funnel

Chapter 3 of this book introduced the crime funnel, the generalisation that for every 1,000 crimes that occur, about 530 are reported to police, but only 429 recorded. Of these, 99 are detected and offenders will have a court appearance in 60 cases. Fifty of these will be found guilty, and four will receive a custodial sentence (reproduced in the first column of Table 8.1 but now with the decimal value for custodial sentences shown). We can continue the generalisation and explore a few scenarios to understand better the impacts on criminal justice policy.

The main point is to understand that any impact generally flows down the funnel rather than up. Therefore, we know that 99 offences are detected, but that number is constrained by the 429 incidents recorded by police. If the number of offences recorded by police increases, perhaps as a result of a change in crime recording policy, it is likely that the detected number may increase. However, increasing the number of offences recorded by police does not mean that the number reported to police necessarily increases. Impacts flow down the funnel, but not necessarily upwards.

With this top-down effect flow in mind, let us explore what would happen if prosecutors decided to charge more offenders. With a 10 per cent increase in prosecutions, the second column of numbers in Table 8.1 shows that the work of the already-stretched courts increases (to about 66), guilty findings would increase, and there would be a need for some extra jail space (the custodial sentence figure increases from 4.0 to 4.4 for every 1,000 offences). However, because the impact on the criminal justice system flows downward and not necessarily up, there is little evidence that this would have a meaningful impact on the 1,000 offences that affect the public.

Flood and Gaspar argue that a central precept of intelligence-led policing is the necessity to tackle and incapacitate the ‘criminal leaders’ and ‘criminal specialists’ (2009). A more intelligence-led and offender-focused approach would allow the police to focus on the prolific and persistent offenders committing much of the crime (Audit Commission 1993). This means that to have a meaningful impact on the 1,000 crimes that affect the community, police enforcement practices are going to have to concentrate on arresting and incarcerating the right people who have earned a place at the bottom of the crime funnel, people who are the criminal elite rather than the low-hanging fruit. Chapter 3 explored what research exists on the ability of police to perform this activity, but for now we can see what would happen with a 25 per cent increase in the clearance rate for reported crime. Given the resources available to police, this is probably quite unrealistic; however, Table 8.1 shows that even with this overly-optimistic view, the result would be increased arrests and paperwork, and greater pressures on the court and jail system.

The hope with intelligence-led policing therefore rides on the possibility that any increase in the custodial sentence rate manages to target the 6 per cent that cause about 60 per cent of the crime (see Chapter 3). Only then is it realistic to suggest that a policy of improved detections would have a chance of reducing the 1,000 actual offences at the top of the funnel. Unfortunately, little research has been conducted to quantify the impact of targeted arrest strategies or more intelligence-driven tactics, and of course this assumes that (1) the increased detections will target the right offenders, and (2) the cost-benefits of this approach – one that places increased stress on the whole criminal justice system – outweigh other crime prevention or reduction tactics. Some research that does examine the impact of a general policy of increased arrests on the crime rate is discussed later in this chapter.

Original crime funnel (decimal values shown)10% increase in prosecutions25% increase in detections

Actual offences1000.01000.01000.0

Reported to police530.0530.0530.0

Recorded by police429.0429.0429.0

Detected offences99.199.1123.9

Prosecuted in court60.566.575.6

Found or pleaded guilty50.255.262.7

Custodial sentence4.04.45.0

Note: decimal values added.

The challenge in targeting the criminal elite is that they tend to be very elusive. With colleagues (Rengert, Ratcliffe and Chakravorty 2005, see also Reuter 1990), I have previously argued that one of the side effects of greater enforcement on the lower levels of the drug importation community has been inadvertently to develop a ‘corps of experienced and expert smugglers who have been able to remove the inexperienced competition’. Even though many police officers profess to wanting to catch the criminal elite, they are constrained by an organisational system that rewards them for the volume of arrests rather than the quality of their captures, a system that returns police again and again to nicking whichever crook was unlucky enough to stick his head above the parapet just when police were looking for their next target.

As a result, the smarter criminals have developed elaborate mechanisms to avoid capture. Drug dealers employ a range of tactics to minimise the impact of police interdiction, including paying lookouts, cooperating between sellers and buyers to identify undercover police officers, using a constantly shifting linguistics code to ‘screen’ potential customers, using stash-spots when in residential areas, or loitering near transit stops to give the impression of waiting for transport (Johnson and Natarajan 1995; St. Jean 2007; Piza and Sytsma 2016). Johnson and Natarajan estimate that experienced and higher-level dealers can use a variety of strategies to minimise the risk of arrest to one for every thousand drug transactions or more.

These strategies are immensely harmful to local communities, and stress the need to target the criminal leaders who organise in this manner. My work with FBI gang units across the US has highlighted the harm to communities when gangs operate across generations of families, corrupt youth, engage in intimidation, or take their activities into schools.

Even when the tools for targeting the top of the criminal food chain are provided, they are sometimes ignored. Levi (2002) laments that anti-money laundering legislation is used too modestly, conviction rates are low, and that adoption of intelligence-led policing targeting in the areas of the criminal elite responsible for fraud and financial crime is uneven at best (Levi and Maguire 2012).

Estimating prevention benefits

An alternative to arrest and prosecution is to consider the benefits of a crime prevention approach. As opposed to addressing crime problems at the lower levels of the crime funnel, crime prevention is designed to work at the top of the funnel. While the number of crimes reported to police is only a small proportion of the actual offences, their geographic dispersion is a reasonable indication of the likely concentration of the larger ‘dark figure’ of crime (Wisnieski et al. 2013; Chainey and Ratcliffe 2005). Therefore, the true value of offences reported to police may be seen as indicators to the bigger problem.

If it were possible to initiate a policy that had a 10 per cent reduction in actual offences that affect the public, the crime funnel could be used to estimate the impact across the criminal justice system. As shown in Table 8.2, the impact of a reduction in crime flows down the crime funnel. Therefore, reducing the number of offences has the corollary of reducing the crimes reported to police, the number recorded and detected by them, the number of people going through the courts, and the size of the jail population. A cost-effective outcome all round.

Crime prevention is therefore the key to long-term crime control, but greater crime prevention can also improve our ability to bring serious and prolific offenders to justice. The scenario above paints a picture of an approach to the crime problem that suggests smaller jail populations and fewer court delays. Swifter justice and greater availability of prison space increases capacity to remove prolific and serious offenders from the general population, and with fewer arrests police could concentrate on these serious, recidivist offenders.

Of course, there are potential threats to this model. For example, one current measure of success in policing is the number of arrests. A preventative focus would reduce the number of offenders from which an officer can demonstrate worthiness for promotion. A change in police managerial practice may be required to resolve this issue; however, a greater threat comes from an absence of political will.

Lab (2004) describes four succinct reasons why there is a general lack of enthusiasm for prevention from policymakers. Politicians like programmes with:

immediate results;

focus on outcomes that can be counted;

a sensationalist streak;

an eye to the immediate problems of the day.

The challenge for crime prevention is therefore significant: ‘Policies to address serious and heinous offences will grab the attention of the media, while community prevention programmes or projects to invigorate the job prospects in struggling communities will receive less attention. Crime prevention simply does not play well in a 15-second sound bite’ (p. 685).

The theoretical foundation of intelligence-led policing is that a focus on the right prolific offenders who are the same ones responsible for a significant proportion of the 1,000 crimes committed against the public will prevent crime. This may be through incarceration; however, further prevention benefits may flow from disrupting the activities of prolific offenders. On the plus side, there is evidence that this approach can be successful (Groff et al. 2015); however, the incarceration component is much lower on the crime funnel. This more intelligence-led approach to tackling the criminal leaders and specialists and learning what opportunities they exploit demands more of police analysts and executives, a challenge to which only some are now rising.

Table 8.2 Revised crime funnel with 10 per cent crime prevention.

Original crime funnel10% decrease in actual crime

Actual offences1000900

Reported to police530477

Recorded by police429386

Detected offences9989

Prosecuted in court6054

Found or pleaded guilty5045

Custodial sentence4.03.6

Note: decimal values shown for custodial sentence rates only.

he challenge in targeting the criminal elite is that they tend to be very elusive. With colleagues (Rengert, Ratcliffe and Chakravorty 2005, see also Reuter 1990), I have previously argued that one of the side effects of greater enforcement on the lower levels of the drug importation community has been inadvertently to develop a ‘corps of experienced and expert smugglers who have been able to remove the inexperienced competition’. Even though many police officers profess to wanting to catch the criminal elite, they are constrained by an organisational system that rewards them for the volume of arrests rather than the quality of their captures, a system that returns police again and again to nicking whichever crook was unlucky enough to stick his head above the parapet just when police were looking for their next target.