Revisiting Crime Rates as a Measure of Crime Prevention Effectiveness: Does the "Crime Drop" Reveal a Policy Effectiveness 'Outcomes' Gap?

Measuring Crime Prevention ‘Outcomes’

Evidence-based crime prevention is defined "not by its intentions, but by its consequences" (Sherman, Farrington, Welsh, & MacKenzie, 2002, p. 3), typically assessed by fewer offences, fewer offenders, fewer victims, or harm prevented. All are valid, and which is preferred may reflect the prevalent political or public discourse. Policy effectiveness and outcomes literature indicates the common thread is a focus on the impact or effects of crime prevention initiatives. Although some may be difficult to quantify, many of the social benefits of crime reduction strategies are capable of measurement, and may be significant. For example:

a) One study assessed the aggregate annual burden of crime in the United States at US$1.7 trillion in 1998. This figure included the burden on the criminal justice system, deaths and injuries, economic losses, costs of private deterrence, and the opportunity cost to victims (and criminals), with time diverted from (or to) productive, lawful, activities (Anderson, 1999). Conservative, ‘lower bound’ estimates that exclude many intangibles place the costs of crime in developed countries at 3, 4 or 7 percent of GDP, with a 13% upper bound mooted “probably accurate” (Entorf & Spengler, 2002, p. 91). Domínguez & Raphael report Anderson’s updated 2012 estimates up to $3.2 trillion or 9.7-19.5% of GDP (2015, p. 592).

b) Australia’s first attempt to systematically record the financial costs of serious and organised crime (A$36 billion) was “more than double” the previous “conservative” estimate of 1-2% of GDP, and itself excluded many costs currently considered too difficult to assess (ACC, 2015, pp. 2, 11) and other crimes beyond its scope.

c) Another study assessed economic consequences of violent crime, including medical costs, lost productivity, physical and psychological damage, the risk of death and reduced quality of life (Reiss Jr & Roth, 1993, p. 60).

Advertisement

d) Other studies assessed the value of illicit drug markets, perhaps the biggest source of illicit funds (UNODC, 2011, p. 29, estimated it at US$350 billion annually, or a fifth of all crime proceeds).

An extensive body of research has established many ways to estimate the costs of crime, including property value, market-based, willingness-to-pay, quality-of-life and numerical crime-ranking methods. These approaches are outlined by McCollister et al (2010) and extensively reviewed by Cohen (2005) and Czabanski (2008). McCollister et al (2010) distilled four ‘fundamental components’ of the cost of crime to society: victim costs (direct economic losses); criminal justice system costs (police, criminal justice and corrections); crime career costs (the opportunity cost of not engaging in productive/legal activities); and intangible costs (indirect victim costs, pain and suffering).

England’s Home Office estimates the social and economic costs of crime along a criminal activity spectrum: costs in anticipation of crime (such as expenditure on security); costs consequent on crime (such as the cost of stolen property and physical and emotional impacts), and costs in response to crime (such as costs of the criminal justice system) (Home Office, 2011 ; H. Mills, Skodbo, & Blyth, 2013, p. 17). To help evaluate the economic effects of crime prevention programs, a series of unit-cost studies also estimate the costs of individual crimes (Aos, Phipps, Barnoski, & Lieb, 2001; McCollister et al., 2010; Miller, Cohen, & Rossman, 1993; Miller, Levy, Cohen, & Cox, 2006; Rajkumar & French, 1997; Roper & Thompson, 2006; Smith, Jorna, Sweeney, & Fuller, 2014).

Many such estimates, however, “do not even begin to measure the full impact of corporate price-fixing and other criminal activities,... [or] the social and psychological costs to society and crime victims. Fear, mistrust, a curtailing of public activity, and a decline in the quality of life are but a few of the inestimable impacts of crime on society" (Hagan, 2011, p. 16). The United Nations adds that criminal funds also distort resource allocation, ‘crowd out’ legitimate businesses and hamper economic growth. Amongst many “negative socio-economic consequences" one of "the most severe consequences of criminal funds is the further perpetuation and promotion of criminal activities" (UNODC, 2011, pp. 8, 10). Moreover, early ‘cost-of-crime’ studies should be considered in light of the November 2015 edition of Criminology & Public Policy, starting with Tonry’s cautionary overview which optimistically opines that “the fog around cost-of-crime studies may finally be clearing” (2015).

The value of such studies is particularly compelling in New Zealand because the manifold harms caused by crime go to the core of NZ Police’s Prevention First operating strategy, with "reduced victimisation" a central objective. At an organisational level police also know what staff see daily: that "crime, and fear of crime, can cause profound harm and distress", with "significant social and economic costs to society" (NZ Police, 2014d, pp. 19, 22). An operational awareness of harm prevention, reflecting that "police are in reality harm-fighters, not just crime-fighters" (Ratcliffe, 2015a, p. 7), is also evident in practice. For example:

a) In an operation intercepting the flow of methamphetamine between the North and South Islands, Detective Inspector Graham confirmed Police commitment to "interrupting the supply of illegal drugs in New Zealand and preventing the social harm this illegal activity supports"(NZ Herald, 2015).

b) After the seizure of packs of methamphetamine each valued at $28,000, Detective Sergeant McNeill referred to "20 bags of utter misery and by that I mean the flow-on effects that the sale of this drug would have had on the users, their families, friends and the rest of the community" (Stuff, 2015).

c) Detective Senior Sergeant Brent Murray spoke of the “significant harm within communities [and sometimes] fatal consequences” of Alpha-PVP in the context of stopping a group responsible for its importation and distribution. “Removing this drug from the community is preventing further serious harm and saving families the anguish and heartache” (NZ Police & NZ Customs, 2015).

d) Even in the context of falling crime figures, Assistant Commissioner Nicholls noted a rise in sexual and other serious assaults and affirmed a police focus "on directing resources towards the offences that cause the most harm to our communities" (2014).

e) Also regarding property crime, police know that victims’ "terror and fear" when their "right to feel safe in their own home has ... been taken away from" them (Hutt News, 2015) is a significant harm that preventive policing ameliorates.

Advertisement

f) The operational awareness of the societal benefits of crime reduction beyond simple crime rate measures goes to the highest levels. The Police Commissioner's response to drug seizures was a determination "to stop those who deal in this misery" (Bush, 2015b).

Similarly, police recognise the harm- and crime-prevention benefits of asset forfeiture, as noted for example by Detective Senior Sergeant Craig Hamilton after the successful forfeiture of three properties and a marina berth from a convicted drug dealer. “The stripping of assets which have been illegally gained is a vital preventative measure because it sends a message to those on the cusp of offending that crime ultimately does not pay” (NZ Police, 2013a).

In practice, therefore, police clearly understand the harms caused by crime, and the value of efforts to reduce such harm, yet this collective 'coalface' experience is neither reflected nor quantified by traditional crime rate measures.

Ultimate Policy Effectiveness ‘Outcomes:’ A Path Out of the Fog?

At a policy level, there is some evidence of more explicit recognition of ‘outcomes’ in terms of the effects or impact of crime reduction strategies. For example, regarding the social effects of offending such as "[diverted] expenditure from other consumption and investment,… physical harm and fear" from burglary and theft, and the "physical and mental ill-health,... violence,... [and] lost productive potential and direction of public resources" from illegal drug use (MOJ, 2011, p. 12). Similarly, that "reducing drug-related harm in New Zealand communities is of paramount importance" (MOJ, 2011, p. 17, emphasis added).

Notwithstanding crime cost analyses in New Zealand and Australia (ACC, 2015; Roper & Thompson, 2006), much public discourse nonetheless continues to be conducted at a relatively basic level, with simple quantification of loosely defined societal costs and benefits, such as estimated costs of crime to society "in excess of $11 billion per annum" (NZ Police, 2012, p. 3), and crime rate reductions resulting in "nearly 90,000 fewer crimes and tens of thousands of fewer victims" (NZ Police, 2014c, p. 4).

It might be said that sweeping approximations are sufficient because common sense tells us that reduced crime rates and "nearly 90,000 fewer crimes" is a good thing. It is, of course, but resting at a common sense assumption leaves much unsaid. There is obvious value applying common sense to policy development and implementation processes, but "sometimes common sense is nonsense" (Hagan, 2007, p. 6). What seems or is asserted to be common sense may prove to be neither sensible nor commonly perceived. This is particularly relevant in criminology, where politicians, media and commentators frequently espouse opposing views, all purportedly based on common sense.

Any divergence between perceived or expressed common sense and tested evidence may also have an impact on policy effectiveness. As New Zealand's chief science adviser observed, programs based on advocacy rather than evidence risk not meeting intended results (Gluckman, 2011, p. viii), and even when ‘successful’ may fail to achieve better outcomes. Sound application of research methods using objective data therefore offers a useful test of perceived common sense beliefs. Moreover, if a scientific approach confirms common sense assumptions, it may also help more accurately define what the common sense value of fewer crimes and fewer victims really means in terms of the societal benefits of crime prevention policies.

Although crime rates have long been useful as a proxy indicator, particularly when measuring ultimate policy goals is difficult, it is now possible to "distinguish the harms of different crime events based on a financial assessment of societal impact" (Ratcliffe, 2015a, p. 3). However, whether it is appropriate for policymakers to do so, and if so how, is not without debate. Drawing from Kelman’s ethical critique of cost-benefit analysis in policymaking (1981), some argue that the work of Cohen and others, which influenced an “out-of-control emotionalism and mean-spiritedness” of extreme sentencing in the 1980s and 1990s, represents the “dead end of a road that should never have been taken,” because program evaluation using “Cohen-type” analysis produces “fundamentally misleading” results (Tonry, 2015, pp. 654, 658).

Others regard the rehabilitation of early cost-of-crime studies possible, with some refinement; such as transforming cost estimates into relative orderings (Domínguez & Raphael, 2015). New cost-of-crime estimates also continue to be produced, such as Australia’s first Costs of Serious and Organised Crime report, released in December 2015 (ACC, 2015). The complexities of conducting criminal justice cost-benefit analyses are, however, substantial, and some of their underlying assumptions viewed with the clarity of a modern context “[make] many policy analysts squeamish” (Black, Solow, & Taylor, 2015, pp. 643-644).

Economic metrics and crime-harm indices each have theoretical appeal and apparent objectivity, yet operationalising them is not without difficulty. Cost-of-crime measures are typically based on broad categories of crime and do not specifically reflect the manifold perceived social harms caused by crime (McMahon, 2008). Ratcliffe noted that measures based on sentencing guidelines or economic costs also often differ quite dramatically; a homicide, for example, may be rated 2, 12, or 128 times more severe than a robbery (2015b, p. 176). Uneven ‘steps’ between each offence category (the relative harms of theft, burglary, rape and murder are not of course evenly spaced) also renders the apparent linearity of many indices largely illusory for operational purposes.

For any new measure to be effective, it should extend beyond the policy evaluation debate currently reflected in much of the literature. In particular, it must be capable of being ‘operationalised’ into ‘real-time’ policing, and sufficiently robust to be sustainable. The following two sections offer practical perspectives in which a stronger evidence base in this regard may respectively help extend outcome and operational effectiveness, and policy success longevity.Continued on Next Page »