Risky individuals, risky families or risky societies?

In a telling moment in his speech to the Labour Party Conference in September 2009 Gordon Brown turned his fire on families ‘playing by different rules or no rules at all’. Most parents, he observed, ‘do a great job – but there are those who let their kids run riot and I'm not prepared to accept it as simply part of life.’ The Prime Minister went on to promise that ‘every one of the 50,000 most chaotic families will be part of a family intervention project’, a ‘tough love, no nonsense approach with help for those who want to change and proper penalties for those who don't or won't’ (Brown, 2009).

Mr Brown's remarks reflect a long-standing preoccupation of the Labour government with the family as a potential source of criminality in later life. His predecessor Tony Blair made a similar point three years earlier in a speech in Bristol:

We need far earlier intervention with some of these families, who are often socially excluded and socially dysfunctional. That may mean before they offend; and certainly before they want such intervention. But in truth, we can identify such families virtually as their children are born.
(Blair, 2006)

This faced the government and country with‘some unpalatable choices about liberty and security’, Mr Blair argued, where the state's ‘power to intervene’ in the lives of ‘dysfunctional families’ needed to be enhanced. ‘The “hardest to reach” families,’ he observed, ‘are often the ones we need to reach most.’ (ibid).

Family intervention projects are but one of a number of initiatives to have emerged from the government's commitment to intervene early in the lives of troubled children and their families. The government claims intellectual credibility by appealing to recognised and significant research, notably the ‘risk factor prevention paradigm’ championed by David Farrington (Farrington, 2000). But does this research really justify the government's ‘tough love’ approach?

The ‘risk factor prevention paradigm’

At the heart of the risk factor prevention paradigm, Farrington argues, is a ‘very simple’ idea: ‘Identify the key risk factors for offending and implement prevention methods designed to counteract them’. Such an approach ‘can be used not only to identify variables to be targeted but also to identify persons to be targeted in an intervention programme,’ he argues. Typically things such as low intelligence, low empathy, impulsiveness, family problems, abuse and neglect are identified as factors likely to put a person at greater risk of getting into trouble. Targeted and intensive work with ‘problem’ children and their families is a logical policy outcome.

Much that is nuanced in academic research does tend to get lost and mangled by policy makers in search of simple ideas and implementable policy. Blair apparently consulted an essay by Farrington before making his confident claims about identifying problem families and children from birth (Downing Street, 2006). But as Farrington points out in his essay, risk factor analysis tends to be much better at explaining links and associations after the event than it is at predicting future behaviour: ‘Typically, prospective prediction … is poor but retrospective prediction … is good’.

The weak predictive power of risk factor analysis is a theme developed by Derrick Armstrong. At best, he argues, it can account ‘for a statistically significant proportion of the variance in respect of the antecedents correlating with offending’ (Armstrong, 2004). But as a predictive tool risk factor analysis has limited utility. It might identify an increased probability of committing crime and/or being captured by the criminal justice system among certain population groups with shared characteristics. But at the level of the individuals themselves false positives and false negatives abound. The margin for error is very high.‘This sort of evidence,’ Armstrong notes,‘is much more suited to generalizations about groups rather than predictions about individuals’ (Armstrong, 2006).

Ideology, research and policy making

This mismatch between the limits of research knowledge and current policy agendas is striking. But perhaps government policy in this area is not simply based on the ‘best’ research evidence. As Armstrong points out, all sorts of political and ideological factors are at stake when it comes to the formulation of policy. Research forms ‘only part of the context in which policy is formulated’. It is ‘pure idealism’ to think otherwise. Rather than thinking about ‘evidence based policy,’ Armstrong suggests we should be thinking in terms of ‘evidence informed policy’. Research might help to guide, sometimes justify, particular policy initiatives. It will rarely if ever be the single determining factor (Armstrong, 2004). Beyond research lies high and low politics, moral judgements and ideological commitments.

Armstrong develops the point about ideology in relation to what he claims is the tendency of the risk factor paradigm to present ‘social problems in terms of individual and micro-social risks which occur and are perpetuated within those domains’ (Armstrong, 2006).Take the example of growing up in poverty, which risk factor research identifies as a factor increasing a child or young person's likelihood of committing crime or coming to the attention of the police and courts. In a simple sense, poverty could be considered the property of particular individuals – the poor – and not of other individuals – the rich and comfortably off. Moreover, living and growing up in poverty has a profound impact on a range of life chances. In this sense, living in poverty is a particular risk factor associated with particular individuals. Yet there is clearly a social and institutional dimension to poverty and income inequality that is beyond and outside of the control of those individuals who live ‘in’ poverty.

In Armstrong's view, risk factor research drives policy making down the road of individual adaptation rather than more far-reaching social reform: ‘poverty, although recognized as a factor associated with high risk, is countered not by economic redistribution but by interventions aimed at supporting individuals at a micro level with the management of their own risk’ (Armstrong, 2006). The policy challenge then becomes, in essence, a ‘crime management’ challenge. Denying ‘any social structural contribution to the construction and reproduction of offending behaviour’, policy makers can focus their attention ‘not on the causes of crime but rather on a policy of containment’.

Beyond policies of containment

Risk factor analysis has proven useful to policy makers eager to find a basis in research for policies that might otherwise be considered dubious, if not downright oppressive. The nuances and qualifications have at times got lost in translation from the academy to Whitehall. But the focus on individual and micro-social risks has also chimed with the priorities of policy makers.

As an approach, risk factor analysis was imported into criminology from medicine and public health (Farrington, 2000). It is worth noting that within the public health discipline there are approaches to risk analysis that offer an important corrective to mainstream criminological risk factor analysis. In turn they provide the basis for a more grounded and holistic set of policy responses to ‘problem’ children and their families.

Within public health it is possible to identify certain individual risk factors associated with particular diseases. Smokers are at greater risk of developing various diseases than non-smokers. Cardio-vascular disease is associated with raised blood pressure, and so on. Of course, such biological markers do not generally develop spontaneously. Behavioural factors – the decision to smoke or eat unhealthy food for instance – interact with these biological factors in a significant manner. But as Sir Michael Marmot points out in a recent collection of essays on the social determinants of health, the behavioural and biological risk factors are themselves affected by macro social processes:

It is not an accident that people consume diets high in saturated fat and salt. It represents the nature of the food supply, culture, affordability, and availability, among other influences. These are the causes of the causes. For example, given that smoking is such an important cause of premature disease and death, we need to understand the social determinants of smoking. In particular, in many rich countries now, there is a social gradient in smoking: the lower the socio-economic position, the higher the rate of smoking.
(Marmot, 2006)

British citizens are not compelled to smoke cigarettes or consume excessive amounts of unhealthy food. But the choices they make are partly determined and conditioned by the circumstances they find themselves in. The distribution of health and illness is, in other words, socially determined. This explains why there is a clear social gradient in relation to a large array of health outcomes, despite the apparently ‘free’ choices individuals are able to make about their own lives.

Marmot's essay, and the others in the same volume, places in sharp relief the micro-level preoccupations of criminological risk factor research. The concrete policy implications in relation to crime and ‘problem’ families would be to direct attention away from identifying and intervening in the lives of ‘risky’ individuals and families and towards addressing the socio-structural factors that shape their ‘riskiness’: poverty, inadequate housing and mental distress, to name but three. The work of Marmot and colleagues, in other words, challenges the researcher and the policy maker to engage in thinking through and articulating a clear theoretical perspective – informed by empirical research – on the forces and structures that shape individual agency, institutional structures and the political process.

This article is adapted from Risky people or risky societies: rethinking interventions for young adults in transition published by the Centre for Crime and Justice Studies in November 2009.

References

Armstrong, D. 2004. ‘A Risky Business? Research, Policy, Governmentality and Youth Offending’. Youth Justice, 4(2): 100–116.

Armstrong, D. 2006. ‘Becoming criminal: the cultural politics of risk’. International Journal of Inclusive Education, 10(2): 265–278.

Blair, T. 2006, ‘Our Nation's Future’. Speech on June 23 www.number10.gov.uk/Page9737 accessed on January 27, 2009.

Brown, G. 2009, ‘Speech to Labour Conference’, Sept 29 www.labour.org.uk/gordon-brown-speech-conference, 2009-09-29 Sept 29 accessed on October 2, 2009).

Downing Street 2006, ‘David Farrington Report’. www.number10.gov.uk/Page10035. accessed on January 27, 2009.

Farrington, D.P. 2000. ‘Explaining and preventing crime: the globalization of knowledge – the American Society of Criminology 1999 presidential address’. Criminology, 38(1): 1–24.

Marmot, M. 2006. “‘Introduction’”. In Social Determinants of Health, Second Edition, Edited by: M Marmot, M and Wilkinson, R.G. Oxford: Oxford University Press.