Risk,responsibilities and rights: Reassessing the ‘economic causes of crime’ thesis in a recession

Abstract

This article explores competing accounts of an apparent inversion of the previously prevailing relationship between young people’s unemployment and the incidence of youth offending at a time of economic recession. It begins by highlighting the faltering association between unemployment and offending, and considers the paradoxical implications for risk-based methodologies in youth justice practice. The article then assesses explanations for the changing relationship that suggest that youth justice policies have successfully broken the unemployment–offending link; and alternatively that delayed effects of recession have yet to materialize, by reference to the work of four inter-governmental organizations and to youth protests beyond the UK. In place of ever more intensive risk analyses, the article then focuses on the adverse effects of unemployment on social cohesion, and proposes a rights-based approach to youth justice that recognizes the growing disjuncture between the rights afforded to young people and the responsibilities expected of them.

Keywords

mutual obligations risk rights social cohesion unemployment–crime relationship

Introduction

The academic study of the relationship between unemployment and offending amongst young people is supported by both a feast and a famine of literature. There is an apparently limitless self-renewing supply of panel and cohort studies that map the patterns of statistical association between these two phenomena, at a great range of levels of aggregation and variables across many countries. The economists, psychologists and others who pursue these analyses (see for example Agell, 2003; Baron, 2008; Britt, 1997; Buonanno, 2006; Bynner, 2009; Bynner et al., 2004; Entorf and Spengler, 2000; Farrington et al., 1986; Grönqvist, 2011; Narayan and Smyth, 2004; Phillips and Land, 2012; Pouget, 2006) invariably identify a positive statistical relationship between unemployment and offending amongst young people, often over extended periods. For all the abundance of data they draw upon, these analyses offer no clear indications of direct causality in the links between unemployment and offending. In contrast, a very small number of researchers using predominantly qualitative methods intermittently investigate this vexed relationship. Some achieve only glancing analytical engagements with the elusive links (see Furlong and Cartmel, 2007; Mizen, 2006). A very few others, in the form of local studies using relatively small samples, aim to understand the relationship between unemployment and offending in terms of context, meanings, causes and the choices made by young people (notably MacDonald et al., 2001; Webster et al., 2006; White, 1989), but are limited by the difficulty of transferring and applying context-specific evidence and interpretations elsewhere.

This article reassesses the connections between young people’s unemployment and offending in exceptional recessionary times, using recent data. In doing so, it charts the apparent dissipation of the dominant positive pattern of connection between the two in the contemporary UK, and considers whether the present pattern can be explained by recent risk-based policies. In the absence of clear evidence that this is so, the article explores other interpretations of an apparently changing interface between unemployment and crime, setting them in the context of major cross-national increases in youth unemployment, and the serious concerns of some major inter-governmental organizations about its impact on social cohesion. The article concludes with some reflections on the ‘responsibilities and rights’ discourse at a time of major demand failure in youth labour markets.

The Fragmenting Association of Youth Unemployment and Crime

The ‘economic causes of crime’ thesis (originally Barnes, 1895; more recently Jones, 2007; Reiner, 2007) anticipates that the archetypal linkage between a weak economy, rising youth unemployment and increased offending should never be more evident in the UK than at present. Remarkably, however, the current profile appears to be almost the inverse. At no time since the 1930s have economic recession and unemployment impacted so greatly on young people in the UK. As Figure 1 shows, the sharp and sudden upturn in youth unemployment rates at the onset of the 2008/9 recession reversed a long-declining trend. Unemployment amongst young people reached its highest rate since records began, climbing to 1.2 million in 2011. Four of every ten 16 to 17 year-olds were classified as not economically active (which includes 15 per cent of this age group who were not in education or training, as well as being economically inactive). Almost one in five 16–24 year-olds were known to be economically inactive.

Figure 1.

Youth unemployment rates 1992 to 2012 (percentages).

At the same time, youth offending is in sharp and sustained decline. Contrary to historical patterns, trends in the numbers of young people aged 10–17 in the youth justice system show a sharp decline while trends in youth unemployment rise. As Figure 2 shows, data from both the Police National Computer (PNC) and the Youth Justice Management Information System (YJMIS) show that a downward trend that began in 2007/8 continued apparently unaffected by the sudden sharp rise in youth unemployment in 2008, and continued further after the 2009 recession was formally declared.

Figure 2.

Trends in the number of young people in the Youth Justice system: comparing data from YJMIS and the PNC.

This data is even more clearly reflected in longer-term data for adults aged 18 and above. Recessionary conditions were becoming clear in the late 1980s, and recession began in 1991. The continuous line in Figure 3 shows victim-reported crime levels from the Crime Survey for England and Wales (CSEW – previously known as the British Crime Survey); the bar charts show police recorded crime (PRC). Both sets of data capture the steep rise in the number of recorded offences after 1991, though they differ in the duration of the rising trends and the onset of the decline in offending. By comparison, the 2009 recession coincides with a decline in numbers of recorded offences which begins in 2008, and then continues (according to PRC data) or flat-lines (according to CSEW data) until 2012, coinciding with the run-up to the second ‘dip’ of the recession.

Figure 3.

Trends in recorded crime and CSEW, 1981 to year ending June 2012.

On the face of it, the worst recession in modern times has been associated with a marked decline in offending. And while attempts to demonstrate historically consistent indices of association between unemployment and crime have foundered (Britt, 1997), this inversion is not only unprecedented and counterintuitive, it is a potential threat to the paradigm upon which 15 years of risk-based practice in youth justice has been founded – an observation to which we return later. Analysts now face the perverse proposition that recessionary conditions appear to suppress the incidence of offending.

Unemployment and Crime: A Brief History

The historical relationship between unemployment and crime rates for all ages is complex and contested. Analyses using regression modelling to produce precise quantifications of the shifting pattern have a long, disputatious and fraught history in the literature (see, for example, Becker, 1968; Hale, 1998, Scorcu and Cellini, 1998 amongst many other). Meta-studies have sought to weigh competing claims. Tarling’s (1982) overview identified a majority of studies that found no evidence of a (significant) relationship between offending and unemployment. Box (1987) observed that 32 of 50 studies found a significant relationship between the two. Knepper (2007) argues that most analysts agree that crime rates reflect unemployment rates over the long term, but that these vary by type of crime. Bell and Blanchflower’s (2010) review identifies numerous studies that find significant relationships between crime and unemployment. These two more recent meta-studies therefore suggest one way forward.

Differentiating types of crime

Differentiating types of crime may help explain why some recorded crime rates sometimes rise in times of high unemployment. Bell and Blanchflower (2010) report that ‘as unemployment rates increase, crime rates tend to rise, especially property crime’ (R9), (see also for example Hansen and Machin, 2002 in support of the ‘economic causes of crime’ thesis). Wu and Wu’s (2012) study used estimates based on UK regional data for the effects on crime rates of variability in earnings inequality and unemployment between 2002 and 2007. They find that ‘crimes motivated by economic gain are highly correlated with income inequality and unemployment’ and that ‘there is a clear distinction between economic and non-economic crimes’ (p.3774).

But taking into account different types of crime, recent analyses in the USA suggest that the patchy match between rising unemployment and rising crime continues, with strong evidence that since the onset of recession violent crime and property crime fell by about 5 per cent, while robberies – often most associated with recession – fell by 10 per cent in a year from 2009 ( The Economist, 2012 ). While some of the most recent data for England and Wales reflects these counter-typical trends, as Figure 4 shows, the total aggregated flat-lining CSEW data masks some contrary data for crimes that are claimed to be those that are most causally associated with unemployment and financial duress. The incidence of opportunistic crimes like ‘other household theft’ along with thefts of personal property rose markedly soon after 2009, and the long-standing continuous decline in recorded burglary and vehicle-related theft ceased in the same year.

Figure 4.

CSEW acquisitive crimes indexed to 1995, 1995 to 2011/12.

It is clear that descriptive statistics alone show that differentiating types of crime moderates claims that the current trend breaks prevailing historical patterns of association between unemployment and offending.

Who and what counts?

In addition to differentiating types of crime, further refinements need to be considered. Firstly, one of the effects of welfare reform and the shift to conditionalized unemployment benefits has been to alter significantly the respective proportions of working age people who are registered as unemployed and who are assigned to welfare-to-work programmes. It is not only changes in the unemployment rate that might correspond to increases in property crime: more stringent benefit regimes may also have a more pronounced effect. Machin and Marie (2006) analysed these effects and concluded that ‘the benefit cuts and sanctions embodied in the stricter JSA regime shifted people off the benefit system and raised crime’ (Machin and Marie, 2006: 163) . Confining analyses to crime rates thus gives only a partial picture.

Secondly, there are fundamental methodological difficulties with studies which estimate the links between unemployment and crime using panel data. Aggregate data gathered for one purpose are placed alongside aggregate data gathered for another to examine the relationship between shifts in the rates of both over time. The data used show the incidences of both measures in particular periods at aggregate levels, not multiple identifiable individual cases (see Carmichael and Ward, 2000; Wu and Wu, 2012).

Overall, aggregate studies do nothing to show that unemployed people are more likely to offend, or that offenders are more likely to be unemployed. If the purpose of analysis is to understand causation and reassess the ‘economic causes of crime’ thesis during an economic recession, isolating a wide range of influences on changing crime rates is essential. Levitt’s (2004) renowned studies suggested that the strong US economy over a decade of sustained growth in the 1990s played little or no role in the decline in rates of offending, compared to other more specific variables during that period, notably increases in police numbers, prisoner numbers, and the control of crack-cocaine alongside the legalization of abortion. Most correlational studies lack this breadth of scope and depth of contextual analysis, and are thereby considerably limited in the deductions that they can safely make.

On the basis of the research and evidence reviewed here, it is possible neither to dismiss a significant connection between economic conditions/unemployment rates and crime, nor to demonstrate a secure relationship, nor to assert demonstrable causal connections.

Youth Crime and Unemployment

Age differences can be expected to exert a significant influence on the unemployment–crime relation, because of the wide range of conditions under which young people experience unemployment. Specific studies of this relationship amongst young people aged 16–24 are, however, remarkably scarce, and with the exception of those noted above, are almost exclusively concentrated on large-cohort aggregated data sets over time. Dickinson (1994), however, reports an (unreferenced) small-scale qualitative study of 399 18 year-olds dating from 1971–2. Those who were unemployed were about three times as likely to offend. Property crime was more common. Dickinson also argues that young men in recessionary times are more likely than other groups to commit crimes when they are unemployed. It is nonetheless difficult to find clear data that demonstrates this. Carmichael and Ward’s (2000) large scale aggregate data study in England and Wales found that youth unemployment is positively related to burglary, robbery and criminal damage rates, but observes ‘whether it is the unemployed themselves that are committing the crime is another question’ (Carmichael and Ward, 2000: 570). The sole large-scale study of young people that makes any claims of a direct causal connection concludes that ‘youth unemployment has a positive and robust causal effect on most property crimes’ (Fougère et al., 2007: 933, emphasis added).

The statistical bifurcation of youth as an age category

These three analyses strike a notable consensus on the strong positive correlation between acquisitive crime and unemployment amongst young people. But each takes a different view of the nature of the relationship identified. Dickinson (1994) conjectures a causal link, Fougère et al. (2009) claim to demonstrate such a link, while Carmichael and Ward (2000) are explicitly noncommittal both on the issue of causality, and on the significance of the data at the level of actual individual cases. Together they provide a reasonable prima facie case in support of a significant association between unemployment and offending, but they are far from conclusive.

No detailed analytical study of a similar kind focused on young people has been undertaken in the UK since the beginning of the 2008/9 financial crisis. Some of the recent official statistical data is nevertheless indicative. The data from the Youth Justice Board (YJB) cited earlier (Figure 2) offers striking accounts of the reduction in recorded crime over the past five years. The YJB’s most recent report calculates that the 2010/11 figure of PNC-recorded numbers of young people in the youth justice system is a reduction of 19 per cent since 2009/10 and a reduction of 46 per cent since 2006/07 (Youth Justice Board / Ministry of Justice, 2012: 19). But this is a limited approximation of the number of offences committed by young people.

A more direct indicator is the number of young people sentenced – although this too is restricted to reported crimes. As Figure 5 shows, a steady state in numbers sentenced (c.90,000 to 100,000 per annum since 2000/01) began a series of sharp progressive falls coinciding with onset of the recession – from 78,000 in 2008/9 to 71,000 in 2010/11.

Figure 5.

Trends in disposal given to young people at court, 2000/01 to 2010/11.

In interpreting this data it is important to remember that the reduction in first time entrants closely reflects the revision of the police ‘offences brought to justice’ target, which had previously incentivized the prosecution of high numbers of minor offences (Home Office, 2010: paragraph 1.10). Furthermore, for 10–17 year-olds the only official data is derived from police crime records, which are acknowledged by the Home Office to undercount the total number of crimes committed. The CSEW’s record of self-reported crime by victims offsets this limitation, and so – as we have noted –gives a considerably higher estimate of the overall incidence of crime. But its data cannot differentiate offences committed by 10–17 year olds from those of adults. The crime count for young people must therefore be assumed to underestimate the incidence of actual offences.

Data disaggregated by offence types also shows some significant distinctions that may also signal the risk of over-stating the decline in young people’s recorded offending during the recession. Between 2007/8 and 2010/11 all categories of proven offences showed reductions, but these reductions were sharply differentiated. Figure 6 shows that the greatest reductions over this period were in motoring offences (–56%), criminal damage (–48%) and breach of a statutory order (–39%).The reductions in some forms of acquisitive crime were markedly less: robbery (–11%) and burglary (–20%). Amongst acquisitive crimes, only theft and handling stolen goods had declined more sharply (–37%).

Figure 6.

Reduction in proven offences by young people between 2007/08 to 2010/11.

Overall, then, once again, the data is ambiguous. Amongst adults aged 18+ there are some indications that despite a marked continued fall in the number of recorded offences, acquisitive crime rose during or after the recession, whilst a similar overall trend in reduced recorded offences amongst 10–17 year olds shows only some slow-down in the rate of comparable offences for some acquisitive crimes. The significance of this data is further complicated because it is based on two different methodologies. The adult data is based on CSEW reportage which is prone to errors of sampling and respondent recall, but has the benefit of recording a high proportion of unreported (and unprosecuted) crimes. The data on young people captures only proven offences. The former source is equally likely to overestimate or underestimate the actual number of offences, for a variety of reasons (see Muncie, 2001); the latter is certain to underestimate actually occurring offences.

The statistical invisibility of offending amongst 18–24 year olds

Rates of recorded offending amongst 10 to 17 year-olds have continued to decline less sharply than first appeared, alongside increasing levels of youth unemployment. One explanation is that the age of peak offending has been delayed.YJB data shows that Youth Offending Teams (YOTs) work with more than twice as many 17 year-olds as 14 year-olds (YJB, 2012: Chart 3.2). The peak age of offending is usually estimated at around 18 years (Flood-Page et al., 2000; Youth Justice Board / Ministry of Justice, 2010), continuing the progressive rise in age-related rates from 14 to 17 years. Although Figure 7 shows that 17 is the age of peak offending, the rate of decline in recorded offending by age cohort above that age is relatively slow up to the age of 20, and the 21–24 age group still offends more than older groups.

Figure 7.

The peak rate for for offending is at age 17.

A number of plausible interpretations of the emergence of the mooted delay in the age of peak offending make this an important line of enquiry. One would suggest that YJB tactics of preventative intervention have succeeded in containing, controlling or diverting offending within the 10 to 17 year-old age range. As a result, young people who desist or who avoid reoffending may experience the transition to the adult criminal justice system at 18+ as a relaxation of the regime and so offend more than 17 year-olds.

A second interpretation similarly emphasizes a number of intensive interventionist measures to minimize the numbers of 16–18 year olds who can be classified as unemployed through extended compulsory schooling and other forms of post-16 participation. While such measures fall far short of protecting this age-group from the depredations of worklessness they serve both to delegitimize claims of ill-effects, and to place considerable pressures on families and households to absorb such effects.

A third interpretation is that the sequential removal of entitlement to benefits and allowances for almost all 16 to 18 year-olds between 1988 and 1996 (see Muncie, 2009: 249–251) has imbued a sense of the inevitable dependency of this interstitial age-group in the minds of young people and their parents alike. This serves to normalize ‘self-selected’ economic inactivity in ways that ameliorate resentment about the lack of jobs and other opportunities. Difficult as it is to demonstrate the presence and reach of such responses, some recent studies of the intra-familial effects of recession point to similar adaptations (see Henwood et al., 2010; Thomson et al., 2010).

None of the published data allows this line of enquiry to be pursued. Published CSEW and PRC data do not routinely differentiate 18–59 year olds by age. In effect, this means that the highest profile age-specific impacts of economic effects on offending rates amongst 18–24 year olds are concealed within the data for the 18–59 age group. This major omission has particular significance for understanding offences by young people. In-depth independent research has shown shifting patterns of particular offence types that vary significantly by age. Soothill et al.’s (2002) study of the 1953 birth cohort in England and Wales shows that more robberies, burglaries and thefts (including handling stolen goods) are committed by 16–20 year olds than by any of six age categories across the 16–40 years span. Indeed this age-group is by far the most prolific in all three categories of offending.

The absence of more recent corroborative analyses obscures important insights into the variable links between offending and unemployment. This has considerably more significant implications. There is clear evidence that the 18–24 age group experiences the first, the most extreme and the most long-lasting effects of recessions upon the conditions of labour markets. The cessation of recruitment of young people by employers as the first, easiest and least costly economy measure manifests itself in extreme rates of unemployment in their age group. This near-universal phenomenon (see International Labour Organisation, 2010) often results in youth unemployment rates at least double the adult rate. In the UK the peak count of 1.2 million unemployed 18–24 year-olds is widely believed to substantially underestimate actual youth unemployment rates. In all, efforts to test the relationship between recession and offending amongst young people are hamstrung by lack of differentiated data.

These observations might be taken to imply that the persistence of acquisitive crimes amongst under 18year-olds shown in Figure 6 is continued or exaggerated in contiguous age cohorts. This would also reflect the observed recent increases in adult personal theft and ‘other household theft’ (against falling overall recorded adult offending rates) noted above. Exceptional data derived from police records by the YJB (Figure 8) shows an increase in the total number of people sentenced for indictable offences since the onset of the 2009 recession. However, consistent with the other data, the steady progressive reduction in the numbers of under18-year-olds sentenced since 2007/8 is reflected here. But it is also notable that a marginal reduction in recorded offending amongst 18–20 year-olds began at the same time.

Figure 8.

People sentenced for indictable offences by age, 2000/01 to 2010/11.

Although the age ranges covered in Figures 8 and 9 are not an exact match, the divergence of ascending unemployment rates alongside declining rates of offending is reinforced by two sets of data which straddle the 17-18 years age divide. Thus, over the same ten-year period, these tables respectively show the steady decline in rates of offending at the same time as rates of unemployment and other non-participation begin their ascent. In terms of absolute numbers, the totals of young people ‘not in education, employment or training’ (NEET) show a relatively regular trajectory of increase from 2001 to 2011. However, this occurs in the context of a demographic decrease in the size of these age cohorts over this period such that, proportionally, the numbers of young people who are NEET remains fairly steady for the first eight years of the decade, and then begins its exceptional rise in the third quarter of 2009.

Figure 9.

Estimates of 16–24 year olds Not in Employment Education or Training (NEET), 2001-2012, third-quarter: totals and percentage of total age cohort.

Most recently, using freedom of information data England and Wales, the Howard League found a dramatic 34 per cent decline in the number of arrests of young people.

Despite due caveats about the interpretation of data noted aboveconcerning changes in police targeting policy and variations according to types of crime, taken in conjunction, the data in Figures 2, 4, 5 and 10 contributes substantial cumulative evidence of a decline in the number of recorded offences committed by young people. Seen in the round, an apparently major decline in offending throughout a period of sharply increased unemployment represents a serious challenge to the ‘economic causes of crime’ thesis.

Figure 10.

Number of arrests of under-18-year-olds, England and Wales, 2008-2011.

The Youth Justice Board and Reduced Offending: Cause and Effect?

One change in the conditions under which the apparent inversion of the relationship between unemployment and offending has occurred is that it followed the establishment of the YJB, albeit several years later. The intensive, highly regulated modes of preventive intervention based on risk assessments managed through YOTs might well account for the decline in recorded youth offending. Not only did YJB strategies reach maturity at the time of a decline in the numbers of recorded offences from 2008/9 onwards: one key factor in the risk profiling calculus which guides intervention is non-participation in employment, education and training. This approach targets the very young people whose lack of economically or educationally gainful activity may have led them to offend, and provides a plausible basis for inferring a causal connection between YJB practices and reduced offending.

The YJB’s own self-assessment does not make any claim to credit for the change. Its assigned performance measures are benchmarked to reductions in first-time entrants to the criminal justice system, reoffending, and the use of custody. YJB was able to demonstrate reductions in all three (YJB/MoJ, 2012). However, none of these criteria have any necessary connection with an overall reduction in rates of recorded offending. All could in principle be the product of selective targeting and priorities alongside an overall increase in recorded offending. Notably, the YJB itself quite deliberately eschews claims to have reduced offending (YJB/MoJ, 2012: 16).

Certainly, there are many evaluations of the work of the YJB that would lend support to claims of its broad effectiveness and that of YOTs (Burnett, 2004; Earle and Newburn, 2001; Earle et al., 2002; Stephenson et al., 2011). Others have been considerably less confident of the association between interventions and reductions in crime (see for example Smith D, 2010; Smith R, 2003; Phoenix, 2010; Solomon and Garside, 2008). The cautious balance struck by both an independent report on the YJB (Department for Education, 2010) and by the House of Commons’ Public Accounts Committee’s report (PAC, 2011) areperhaps the most measured and telling assessments of the effectiveness of the YJB. The circumlocutious wording of the DoE Report’s opening recommendation makes clear its caution about attributing reduced offending to the JYB in any direct way:

The Youth Justice Board should now build on its strengths and re-invigorate its role, having contributed to encouraging reductions in youth crime (Department of Education, 2010, 1.11. emphasis added)

A year later, the PAC’s principal conclusion that the youth justice system has been successful in reducing the number of criminal offences over the last ten years is immediately qualified by querying whether these are genuine reductions in crime or displacements dealt with by non-criminal sanctions. The Committee notes that YJB has ‘limited evidence of what interventions work’ (Recommendation 4) and that it ‘lacks sufficient knowledge of what interventions are being used by Youth Offending Teams and how well they are working’ (Recommendation 6).

At this point it is important to recall two caveats. Firstly, overall rates of offending across the adult age range have also continued to decline since 2008, in part of the criminal justice system over which YJB has no jurisdiction. The supposedly counter-dominant trend for the number of recorded offences to fall as unemployment rises can therefore reasonably be assumed to have other causes which are likely to be independent of the YJB’s distinctive interventions.Secondly, as was noted earlier, because of the limitations of PRC data and the lack of age-differentiation in CSEW data, the crime count for young people must be assumed to underestimate the incidence of actual offences.

It is clear that attempts to associate YJB practices with reduced offending in any direct causal way lack a sufficiently clear empirical foundation, and that data on the incidence of offences committed is an underestimate. It is not safe to attribute a perceived decline in youth offending to the work of the YJB.

Re-calibrating the calculus of risk: The paradox of success

Attributing reductions in offending to the YJB also leads towards an uncomfortable paradox. Unemployment and non-participation are prominent risk factors in the YJB’s risk assessment methodology. What, then, would be their significance in circumstances in which risk-based preventive interventions had succeeded in breaking the claimed connection between unemployment and the offending to which it gave rise? If aggregate levels of association between unemployment and offending had been demonstrably weakened by the YJB’s practices, what would be the rationale for retaining unemployment as a factor in the case-specific calculus of risk of offending? Revised models would surely have been needed that demoted or excluded unemployment as a risk factor. If the current pattern of weakened association is sustained and the data concerning types of crime and over-18s are resolved, it is clear that, whatever the causes of the change, the models too should change.

This apparent‘paradox of success’ whereby targeted interventions may have helped render redundant the very risk calculus that guided their targeting suggests that risk assessments are as inherently limited in their capacities to adapt to the exigencies of volatile economic conditions as are other analyses of connections between unemployment and offending – whether causal or spurious. Put differently, the uncertainties highlighted in the foregoing analysis would surely no longer provide a credible basis for including unemployment as a factor in risk profiling. Its connection to offending in current times is at best highly tenuous, and at worst a way of legitimizing the intrusive management of vulnerable and disadvantaged categories and classes of young people.

At this juncture it is salutary to recall that aggregate patterns of statistical association between unemployment and offending are insufficiently strong to support claims of causal connections; but that the claimed association may nevertheless be defensible if sufficient offence-specific and age-differentiated data were available. The uncomfortable tension between these two statements makes for a less-than-compelling case for persisting with a faltering if not flawed mode of analysis. This constitutes something of a ‘parting of the ways’. The arguments rehearsed here come to a head in a stark and difficult choice: to pursue more reliable and better age-differentiated data and refine and extend the calculus of risk and its associated policies and practice; or to find alternative explanations for and responses to the apparently shifting relationship between unemployment and offending in times of major socio-economic change.

Social Cohesion and the Delayed Effects of Unemployment

One prospectively crucial alternative explanation draws attention to the limitations and ambiguities of current statistical data and analyses, and adds other reasons to be cautious in assuming that the unemployment-crime nexus has changed significantly. Unemployment is a well-established ‘lagging indicator’ of economic recession; so also some of the most adverse manifestations of the effects of unemployment show a relatively systematic delay. There is an extensive literature on a range of so-called ‘scarring’ effects of unemployment on young people too large to cite here in detail. While it is not consensual, some significant tendencies are evident. For econometricians, periods of unemployment in youth have quite reliably strong adverse long-term effects on earning power (see Bell and Blanchflower, 2010 for an overview). Social-psychological studies reliably demonstrate reduced self-esteem, self-confidence and sociality (e.g. Goldsmith et al.,1997; Jahoda, 1979). Other contextual and individual psychological variables associate unemployment with depression and suicidal tendencies (Hagquist, 1998; Pritchard, 1992). More broadly-based sociological studies tend to associate unemployment with manifestations of social exclusion and alienation.

Many of the relevant effects are indeterminate, discernible only through attitudes, dispositions, degrees of withdrawal and so-called ‘disengagement’ that make claims of delayed effects difficult to substantiate. Suicide and attempted suicide are among the few finite junctures against which delayed effects can be measured. And here, there is some general agreement that, in young men in particular, this extreme outcome is mostly likely to occur between one and two years after the onset of unemployment (Hagquist, 1998; Pritchard, 1992).

The international rise of social unrest

Far more important than the delay interval is the clear evidence that, at the level of the individual, it takes time for the damage of inactivity and exclusion from labour markets and other modes of productive activity to be realized. It would therefore be unremarkable if the time-lag between the individualized experience of being unemployed and a public and/or collectivized expression of frustration or anger at this condition was commensurately longer. The evidence from the Mediterranean countries that experienced major youth protests in 2011 and subsequently supports this hypothesis. In most of North Africa and in Greece and Spain extraordinarily high levels of youth unemployment have long been endemic (ILO, 2010), but the emerging effects of global recession in combination with country-specific factors prompted mass collective protest in 2011.

Unsurprisingly, the significance of these often quite extreme collective responses on the part of young people has been widely observed. It has raised sufficient anxiety about its social and economic consequences to prompt a strikingly diverse range of Inter-Governmental Organizations (IGOs) to respond. It is a mark of the levels of international concern about the now-globalising problem of youth unemployment that three IGOs as diverse as the Organisation for Economic Cooperation and Development, the International Monetary Fund, and the International Labour Organization have all issued guidance, cautionary comments and direct warnings to national governments about the social consequences of failure to address youth unemployment (see also Fergusson and Yeates, forthcoming).

In some instances, the responses constitute significant shifts in position. Until 2010, the International Monetary Fund (IMF) had been silent on youth unemployment. Remarkably, its first intervention on the subject concluded that […] ‘the labour market experience of today‘s youth will have deep adverse impacts on the faith in public institutions of future generations. (IMF/ILO, 2010:.21). This is a surprising departure for an IGO concerned with economic not social stability. Even more striking is the shift of position at the Organisation for Economic Cooperation and Development (OECD). Until the global financial crises and the sharp rise in youth unemployment, the OECD’s position had been heavily invested in its predominant strongly neo-liberal stance and its assertively anti-welfarist position (see McBride, 2011; Fergusson and Yeates, forthcoming). By 2011, its main annual employment report commented:

Investing in youth and giving them a better start in the world of work should be a key policy objective. Otherwise, there is a high risk of persistence or growth in the hard-core group of youth who are left behind, facing poor employment and earnings prospects. In a context of ageing populations, OECD economies and societies simply cannot afford the large economic and social costs that such an outcome would entail (Martin/OECD, 2011: 14)

Awareness of the social implications is particularly significant here, sufficient to prompt the OECD Directorate to urge member nations to target intensive assistance, to make temporary extensions to the safety net as ‘vital to prevent poverty’, and to provide income support, and effective, reliable social protection (Martin/OECD, 2011:13).

The International Labour Organization’s (ILO’s) attention to social unrest and its implicit connections with crime take the debate a significant step further. One report (ILO, 2012a) envisages an explicit causal link:

Unemployment breeds social exclusion that, in turn, gives rise to high welfare dependence, crime and anti-social behaviour… (ILO, 2012a: paragraph 22)

The depth and breadth of the current youth employment crisis, and the social unrest that is unfolding, underscore how much is at stake in international and national efforts to avert a second and potentially more dangerous phase of the global crisis… (ILO, 2012a: paragraph 24)

These powerful statements are founded in substantial research undertaken by the ILO’s Institute for International Labour Studies. Its annual report (International Institute for Labour Studies, 2011) shows that social unrest is increasing worldwide especially in the advanced economies. Globally, unemployment is the largest and most significant determinant of risk of unrest (see IILS, 2011: Figure 1.12). Other studies observe an historical association between poor job prospects amongst young, educated people and public protests (Jenkins and Wallace, 1996), and between austerity measures and social protests (Ponticelli and Voth (2011). Recent evidence from southern Europe and North Africa powerfully supports this thesis (Jawad, 2012; LaGraffe, 2012; Papoudopolos and Roumpakis, 2012; Ramos-Diazand Varela, 2012).

A number of recent speeches by the Director General of the ILO have underscored this connection. At the Davos Summit he drew on unreferenced data that counted more than a thousand cities across 82 countries that had experienced youth protests during 2011 (ILO, 2012b). Subsequently, he applauded what he described as ‘a powerful reassertion of activism from young people who refuse to accept a future of unemployment, marginal work and expensive, poor quality education’ (Somavia/ILO, 2012a); and supported this by stating his belief that ‘there is a growing conviction of the need to change course stimulated by social movements and protests in many countries, themselves fuelled by the growth of inequality and intolerable levels of youth unemployment’ (Somavia/ILO, 2012b).In the same year the ILO’s crisis report on youth unemployment, Time for Action, concluded:

The situation of youth employment thus deserves urgent attention. Not only does it threaten to violate the principle of equality and solidarity between generations, which is an important aspect of social justice, but any further prolonging or deepening of the crisis will also increase the likelihood of political and social unrest as more and more young people lose faith in the current economic paradigm. (ILO, 2012a: 1)

This is a stark warning of the potential consequences of inaction. In case the ILO’s words be thought alarmist and exaggerated, many commentators would agree that faltering faith in the current economic paradigm motivated many during the uprisings of the Arab spring, in which the key role of young people and the prominence of unemployment and an endemic absence of economic opportunities have been identified as important ingredients (Jawad, 2012; Weddady and Ahmari, 2012). The increase in some key indicators of social unrest (including poor confidence in government and high unemployment) has been as great in the ‘advanced economies’. The responses of young people in Greece and Spain offer an alarming insight into the ways in which weakening confidence in an economic paradigm begin to manifest themselves as the beginnings of classic crises of legitimacy. In a prescient speech to the prestigious Brookings Institution two years earlier the ILO Director-General was warning that:

The crisis has again put before our eyes something that we all know: good jobs, quality jobs, decent work are, everywhere, central to the lives of women and men. Decent work is a source of personal dignity. Stability of family and households. Peace in the community. Trust in government and business and overall credibility of the institutions that govern our societies. (Somavia, 2010)

Unemployment, crime and social cohesion

Importantly, what connects these commentaries from leading IGOs is an emerging focus on threats to political legitimacy and social cohesion posed by the exclusion of young people from labour markets and wider social and economic participation. Indeed, the most thorough, evidenced recent analysis of social cohesion in this context that is available comes from the World Bank. It dedicated a significant strand of its World Development Report 2013 to the subject. The chapter on jobs and social cohesion states:

…cross country analysis of values surveys finds that jobloss or lack of access to jobs is associated with lower levels of trust and civic engagement. …In extreme cases, if people, particularly youth, lack jobs and hope for the future, they may turn to violent or criminal activity to compensate for the absence of self-esteem and sense of belonging that a job might otherwise provide.(World Bank, 2012: 126)

The text is concerned with the correlates of unemployment and a range of attributes and dispositions of people who are unemployed, focusing on social trust, community and political participation. But the explicit conception of social cohesion is grounded in social–psychological categories and individual behaviours. As such it lacks a fully social – or economic – conception of cohesion that takes into account structural factors. By comparison, Green et al.’s (2000) comprehensive study is commendable. It is a major comparative modelling project across 28 countries with advanced economics using a typology of ‘regimes of social cohesion’ acknowledged to be heavily derived from the work of Esping-Andersen (1990). For all the limitations of this approach, it provides some well-analysed indicators. One of the assigned indices of social cohesion is the level of state management of labour markets and employment protection alongside levels of income inequality. Unremarkably, this marker finds the Social Democratic Nordic countries and to a lesser extent the Social Market countries of most of the rest of Europe to have high levels of institutionally embedded state-driven social cohesion. Characterized as a Liberal regime, in contrast, Great Britain (the analysis does not include data for Northern Ireland) exhibits lower levels of social cohesion on this indicator: less employment protection and greater income inequality, but also compensatingly higher levels of civic participation, also taken to be indicative of a more socially cohesive societies. More significantly for this discussion, across the board, greater extremes of income inequality are found to be closely associated with higher levels of violent crime, and taken together these are presented as significant markers of weak social cohesion. This is the profile that characterizes Great Britain, in greatest contrast to Social Democratic regimes where incomes are considerably less unequal and violent crime rates significantly lower.

Positivist high-volume, high-aggregation studies like those of Green et al. (2009) and the World Bank (2012) are open to the same methodological criticisms that applied to risk analyses and to the studies based on inferential statistical association considered earlier in this article. Furthermore, many other definitions, indices and interpretations of social cohesion are available. Nevertheless, their sheer international scale, the robust strengths of the associations they report, and their contextualizing differentiation of regime types carry some weight. Many of their findings are echoed in different ways in more nuanced, qualitative analyses that associate exclusion from labour markets with tendencies towards lower levels of social and economic integration of a kind that evidence the ‘scarring’ effects of unemployment (for example Hannan,1998; MacDonald et al., 2001; Webster et al., 2006).

The sharp distinctions which Green et al.’s (2009) study identifies between Great Britain’s model of social cohesion, and that of most of the rest of Europe inevitably prompts speculation about the different ways in which young people have responded to mass unemployment across all three social cohesion regime types. In the social democratic Nordic countries, street protest by young people about unemployment has been minimal and overall crime rates remain low. In the social market countries of Mediterranean Europe where public protests have been most frequent and voluble, overall violent crime rates are amongst the lowest in Europe, but income inequality is much greater than in the social democratic countries. And in Great Britain, violent crime is the second highest of all the countries analysed, far in excess of other European countries, but income inequality is broadly the same as that for the Mediterranean social market regimes. These observations make a strong prima facie case that the youth crime and unemployment link is weakest in the Nordic social democratic countries, and strongest in liberal Great Britain. In the Nordic countries this profile is associated with minimal protest; in the Mediterranean countries with recurrent and mostly peaceful protest that directly articulates objections to the economic injustices of unemployment and inequality; and in England with a single explosive episode of violent rioting in August 2011 involving extensive criminal damage, which was not directly articulated with a specified unified set of injustices.

It would be rash to draw conclusions from these observations without thorough assessment of the age-specific incidence of crime, inequality and unemployment in these countries, and without a comprehensive review of the forms of protest and other mass reactions. But even from this brief overview it is clear that the concept of social cohesion should be axial to any analysis that purports to explain increased levels of recorded crime or mass protestin relation to endemic unemployment. But the insights which a full analysis could be expected to bring forward are capable of providing an entirely different interpretation of the inversion of the historical patterns of association between unemployment and offending. On this reading, social protest is a manifestation of delayed effects of extended unemployment which, as some IGOs have begun to infer, result from weakening social bonds and diluted social solidarity. Whether protest is an alternative to crime – and thus in some contexts might explain the weakening link between unemployment and offending – or whether it takes criminal forms that re-configure that link is far from clear. Either way, delayed protest or delayed offending as responses to being unemployed may mean that the full effects of mass youth unemployment have yet to manifest themselves.

Once again, this line of interpretation leads to a prospective methodological ‘parting of the ways’. One response is to persist with the apparently fast-diminishing returns from aetiologically-inspired cohort studies that aim to ‘see inside’ the points of articulation between unemployment and offending – or make better calculations of the determinants of weakening social cohesion. Alternatively, this may be a critical moment at which to consider some of the fundamental principles of social and economic organization that underpin social cohesion. Here, the words of the ILO report quoted earlier, that the principle of equality and solidarity between generations is an important aspect of social justice, seem particularly pertinent to an analysis that is cognisant of the centrality of social cohesion to this debate.

Responsibilities and the Principle of the Right to Work

In the multiple layers of policy revision and re-steering of the youth justice system in England and Wales over more than six decades, it is easy to lose sight of the principles which should inform every system of youth justice, especially if it is also committed to securing social justice. Goldson and Muncie (2006) outline a ‘critical anatomy’ of such principles. They take the International Human Rights Framework (IHRF) as providing the founding precepts for any such system, to argue that:

Ethically legitimate, rights-compliant and effective approaches to youth crime and justice must be located within a broad corpus of social and economic policy rather than the narrower confines of youth/criminal justice policy. (Goldson and Muncie, 2006: 222, emphasis in original)

There could be no better illustration of the importance and relevance of this assertion than the preceding discussion. The current socio-economic context fundamentally conditions the operating circumstances of the youth justice system of the UK. Understanding this policy context is essential for ensuring an appropriate alignment between social justice and criminal justice of a kind that has special significance for responding to norm infraction amongst young people. Of fundamental relevance to a system of youth justice that is located within its social and economic policy contextare young people’s rights in labour markets. One such right concerns the protection of children up to the age of 18 from economic exploitation (Article 32 of the United Nations Convention on the Rights of the Child, 1989). The first clause of Article 23 of the UN Declaration of Human Rights reads:

Everyone has the right to work, to free choice of employment, to just and favourable conditions of work and to protection against unemployment.

Everyone who works has the right to just and favourable remuneration ensuring for himself and his [sic] family an existence worthy of human dignity, and supplemented, if necessary, by other means of social protection.

For that overwhelming majority of young people who have nothing to sell but their labour, the failure of the basic universal chartered rights to employment, to the free choice of employment, to freedom from economic exploitation, and to just and favourable remuneration between the ages of 16 and 24 is clear.

Responsibilities, rights and ‘mutual obligations’

It is highly significant that one of the most penetrating discourses of recent times straddles the fields of welfare/unemployment policies, and criminal justice policies. New Labour’s insistence throughout the previous decade that rights carry responsibilities has a distinctively multi-valent capacity to cross these policy fields. Young people in particular were enjoined to take responsibility for their own actions or accept the consequences of not doing so. In the context of labour markets, this entails accepting participation in work programmes in lieu of contracts of employment when no work is available. In welfare and social protection policy, it entails readiness to work. And in youth criminal justice policy it entails meeting conditions of conduct, participation in behaviour management and other programmes, and observing restrictions on movement and association in lieu of custodial or other sentences. It is thus against the responsibility to meet prescribed norms of participation and conduct that rights to employment or to social protection are to be honoured.

This ‘mutual obligations’ discourse has always implied a balance between the inputs through which responsibilities are discharged, and the outputs each individual claims ‘by rights’. Its credibility depends on the commensurate nature of inputs and outputs. The great purchase of this discourse was achieved in the highly conducive conditions of the longest boom of the post-war era. It made sustainable the underlying bargain of maintaining tolerable minimum living standards in exchange for actual jobs or simulated work.

This publicly-sanctioned ‘mutual obligations’ approach was one of the transformative features of New Labour’s consolidatory normalization of neoliberal market economics and neoliberal social governance (see Heron and Dwyer, 1999). Their faltering and contested introduction under previous Conservative governments gave way to the infinitely more effective managerialized and governmentalized structures and processes of New Labour’s centralist forms of governing (see for example Clarke et al.,1994). As the thresholds of young people’s increasingly conditionalized entitlement to unemployment benefit, housing benefits, supplementary benefits and other now-defunct means of financial support were raised, young people’s tolerance for the ‘rights/responsibilities’ bargain was maintained by generally buoyant youth labour markets.

The conditions for the maintenance of this bargain began to falter with the onset of the 2008/9 recession. With the election of the Conservative–Liberal Democrat Coalition government in the UK in 2010, the political–economic context shifted much more significantly. The removal of the safety valve of buoyant labour markets began to throw into much sharper relief incommensurate rights ‘outputs’in return for responsible ‘inputs’ – and the eroded mutuality that this implies. The absence of sufficient employment opportunities are the essence of the emerging disjuncture. In addition, unreasonable pressures upon young people to accept places on programmes which provide no skills or development, to work without pay and to accept levels of weekly remuneration that could not sustain independent living for a day, grossly exaggerate reasonable expectations of responsible participation. An emergent discourse of ‘disengagement’ attributes failure to compete in labour markets to unwillingness to gain qualifications and accumulate skills in circumstances in which it is beyond question that the failure to gain employment is the result of demand failure in a large majority of local labour markets (Fergusson, 2013).

Such conditions so powerfully exaggerate what should constitute reasonable individual responsibility for worklessness as to risk destruction-testing the credibility of the mutual obligations discourse. The palpable erosion of the rights to work, to free choice of employment and to just and favourable conditions and remuneration now threatens the legitimacy of the ‘rights and responsibilities’ discourse. This, it is crucial to remember, is the discourse which became the axis of settlement which disabled and mollified protest and resistance in the 20-year transition from broadly social democratic-inclined market and welfare conditions in the post-war UK, to a heavily non-liberal set of economic and social conditions. The weakening of the discourse has exactly the potential to threaten faith in the dominant paradigm noted earlier.

Concluding Comments

Across the piece, quantitative analysis has demonstrated weak predictive and explanatory powers and a poor capacity to account for atypical trends in rates of unemployment and the incidence of offending amongst young people in the UK. Despite this, the same correlational, inferential and actuarial statistical techniques as have been used in such studies continue to provide the core models for the methods of risk profiling which have shaped the dominant approach of youth justice practice in England and Wales for the past 15 years. Some puzzling paradoxical and potentially contradictory issues arise. As we have noted, it is unclear why techniques that have had such a limited success in identifying a significant relationship between unemployment and offending are deemed adequate contributors to the processes of risk profiling which help determine who will be the object of intensive surveillance and intervention in the current practices of youth justice.

Rather than revise the parameters for the correction and refinement of techniques that have, at best, been only partially successful, other explanations for the counter-typical divergent trend of unemployment and offending can lay strong claim to being afforded new levels of priority for research and policy alike. One approach briefly outlined here is to consider the evidence that the unique characteristics of the present recessions may have the effect of delaying responses to extended periods of economic inactivity on the part of young people. In the absence of clear and reliable evidence, events beyond the UK have highlighted very real possibilities that their responses can take quite different forms. Modes of social unrest and manifestations of extreme discontent abroad already have partial parallels with the English Riots of 2011. Once again, the reflex response of research and policy analysis leans towards the proto-aetiological, focusing on efforts to profile the characteristics and circumstances of rioters and establish parameters of causality (e.g. National Centre for Social Research, 2011). However, the surprisingly congruent responses of four powerful IGO shave drawn attention to the threats to social cohesion posed by extremes of extended or endemic levels of youth unemployment – in one case to issue stern warnings of the risks of radicalization and loss of confidence in the dominant economic paradigm. By raising serious questions about the continued credibility of the ‘mutual obligations’ and ‘rights and responsibilities’ approaches as imbalances intensify between what is provided for and what is demanded of young people in increasingly adverse conditions, this article has emphasized the priorities and principles that underpin fundamental human rights that should surely frame all systems of youth justice.

Translating such rights into workable practicable youth justice policies deserves serious consideration of a kind that it has not received for many decades. There has been a resurgence of rights-based initiatives – for example the Decent Work Initiative and the Global Social Protection Floor spearheaded by UN agencies. These merit careful exploration and costing on premises that have considerably higher aspirations than the minimalistic, disciplinary and exploitative modes of job creation and diversion from labour markets that currently characterize provision for young people without work. An essential part of such calculations would be more thoroughgoing research into the crime reduction effects of such measures than have currently been available. The benefits to social cohesion, to remaking a flawed economic paradigm and to the reassertion of chartered human rights are potentially great indeed. Some substantial diversion of research and policy endeavour away from modelling and aetiological approaches that have poor and unproven returns, and towards lines of enquiry that include an appropriate recognition of the human and social as well as economic costs of unrest and weakening social cohesion has rarely been more warranted in modern times.

Footnotes

Author biography

Dr Ross Fergusson is Senior Lecturer in Social Policy in the Department of Social Policy and Criminology at the Open University, UK.

References

Agell

(2003) Crime, Unemployment and Labor Market Programs in Turbulent Times.Uppsala: Institute for Evaluation of Labour Market and Education Policy.

Barnes

(1895) Economic Causes of Crime. Madison, WI: University of Wisconsin-Madison.

Baron

(2008) Street youth, unemployment, and crime: Is it that simple? Using general strain theory to untangle the relationship. Canadian Journal of Criminology and Criminal Justice 50(4): 399–434.

Becker

(1968) Crime and punishment: An economic approach. Journal of Political Economy 76: 169–217.

Bell

Blanchflower

(2010) UK unemployment in the great recession. National institute economic review, 214(1), R3–R25.

Box

(1987) Recession, Crime and Punishment. London: Macmillan.

Britt

(1997) Reconsidering the unemployment and crime relationship: Variation by age group and historical period. Journal of Quantitative Criminology 13(4): 405–428.

Buonanno

(2006) Crime and labour market opportunities in Italy (1993–2002). Labour 20(4): 601–624.

Burnett

Roberts

(2004) What Works in Probation and Youth Justice: Developing Evidence-based Practice. Cullompton: Willan Publications.

10.

Bynner

(2009) Lifelong Learning and Crime, IFLL Public Value Paper 4: A Life-course Perspective. Leicester: National Institute for Adult and Continuing Education.

11.

Bynner

Londra

Jones

(2004) The Impact of Government Policy on Social Exclusion among Young People. London: Social Exclusion Unit.

12.

Carmichael

Ward

(2000) Youth unemployment and conditions and the labor market outcomes of young men in crime in the English regions and Wales. Applied Economics 32(5): 559–571.

13.

Clarke

Cochrane

McLaughlin

(eds) (1994) Managing Social Policy. London: Sage Publications.

14.

Department for Education (2010) Safeguarding the Future: AReview of the Youth Justice Board’s Governance and Operating Arrangements. London: Department for Education.

15.

Department of Education (2012) Statistical Release NEET Statistics Quarterly Brief, Third Quarter 2012 November. Available at: http://www.education.gov.uk/researchandstatistics/datasets/a00215766/neet-statistics-quarterly-brief-quarter-3-2012 (last accessed 6 December 2012).

16.

Dickinson

(1994) Criminal benefits. New Statesman, 7, 285, 20–1.

17.

Earle

Newburn

(2001) Creative tensions? Young offenders, restorative justice and the introduction of referral orders. Youth Justice 1(3): 3–13.

18.

Earle

Newburn

Crawford

(2002) Referral orders: Some reflections on policy transfer and ‘what works”. Youth Justice 2(3): 141–150.

19.

Entorf

Spengler

(2000) Socioeconomic and demographic factors of crime in Germany: Evidence from panel data of the German states. International Review of Law & Economics 20(1): 75–106.

20.

Esping-Andersen

(1990) The Three Worlds of Welfare Capitalism.Oxford: Wiley.

21.

Farrington

Gallagher

Morley

St Ledger

West

(1986) Unemployment, school leaving, and crime. British Journal of Criminology 18(1–2): 13–29.

22.

Fergusson

(2013) Against disengagement: Nonparticipation as an object of governance. Research in Post-compulsory Education 0,0,00.

23.

Fergusson

Yeates

(forthcoming) ‘The Emergence of Youth Unemployment As an Object of Global Policy Analysis’.

24.

Flood-Page

Campbell

Harrington

Miller

(2009) Youth crime: Findings from the 1998/99, Youth Lifestyles Survey, London, Home Office.

25.

Fougère

Kramarz

Pouget

(2009) ‘Youth unemployment and crime in France’, Journal of the European Economic Association, 7, 5, 909–938.

26.

Furlong

A and

Cartmel

(2007) Young people and social change (Second Edition), Maidenhead, Open University Press/ McGraw Hill.

27.

Goldsmith

A. H

Veum

J.R

Darity

(1997) ‘Unemployment, joblessness, psychological well-being and self-esteem: Theory and evidence’, Journal of Socio-Economics, 26, 2, 133–158.

28.

Goldson

Muncie

(2006) ‘Critical Anatomy: Towards a Principled Youth Justice. In: Goldson

Muncie

(eds) Youth Crime and Justice: Critical Issues. London: Sage Publications.

29.

Green

Janmaat

Han

(2009) Regimes of Social Cohesion. Londo:, Centre for Learning and Life Chances in Knowledge Economies and Societies, Institute of Education.

30.

Grönqvist

(2011) Youth Unemployment and Crime: New Lessons Exploring Longitudinal Register Data.Stockholm: The Swedish Institute for Social Research.

31.

Hagquist

(1998) Youth unemployment, economic deprivation and suicide. International Journal of Social Welfare 7(4): 330–339.

32.

Hale

(1998) Crime and the business cycle in post-war Britain revisited. British Journal of Criminology 38(4): 681–698.

33.

Hannan

(1998) Beyond Networks: ‘Social Cohesion’ and Unemployment Exit Rates. Colchester: Institute for Labour Research, University of Essex.

34.

Hansen

Machin

(2002) Spatial crime patterns and the introduction of the UK minimum wage. Oxford Bulletin of Economics and Statistics 64: 677–699.

35.

Henwood

Shirani

Coltart

(2010) Fathers and financial risk-taking during the economic downturn: Insights from a QLL study of men’s identities-in-the-making.21st-Century Society: Journal of the Academy of Social Sciences 5(2): 137–148.

36.

Heron

Dwyer

(1999) Doing the right thing: Labour’sattempt to forge anew welfare deal between the individual and the state. S ocial Policy & Administration 33(1): 91–104.

37.

Home Office (2010) Policing in the 21st Century: Reconnecting Police and the People. London: Home Office.

38.

House of Commons Committee of Public Accounts (2011) ‘The Youth Justice System in England and Wales: Reducing Offending by Young People, Twenty-first Report of Session 2010–11, HC 721. London: The Stationery Office.

39.

House of Commons Library (2012) Youth Unemployment Statistics, SN/EP/5871, London: House of Commons.

40.

Howard League for Penal Reform (2012) Police in England and Wales arrest a child every two and half minutes, News Release, 3 December. Available at: http://www.howardleague.org/police-child-arrests/ (last accessed 6 December 2012).

41.

International Institute for Labour Studies (2011) World of Work Report 2011: Making Markets Work for Jobs. Geneva: ILO.

42.

International Labour Organization (2010) Global Employment Trends for Youth.Geneva: ILO.

43.

International Labour Organization (2012a) The Youth Employment Crisis: Time For Action. Geneva: ILO.

44.

International Labour Organization (2012b) Somavia Puts the Accent on Youth Employment in Davos, press release, 12 January. Geneva: ILO.

45.

International Monetary Fund/International Labour Organization (2010) The Challenges of Growth, Employment and Social Cohesion, Joint International Monetary Fund/International LabourOrganisation Conference, Oslo, IMF/ILO.

46.

Jahoda

(1979) The impact of unemployment in the 1930s and the 1970s. Bulletin of the British Psychological Society 32(2): 309–314.

47.

Jawad

(2012) Fromblack hole to screen: The coming of age of social policy in the Arab countries? In: Kilkey

Ramia

Farnsworth

. (eds) (2012) Social Policy Review, 24: Analysis and Debate in Social Policy. Bristol: Policy Press.

48.

Jenkins

Wallace

(1996) Thegeneralized action potential of protest movements: The new class, social trends, and political exclusion explanations. Sociological Forum 11(2): 183–207.

49.

Jones

(2007) The governance of security: Pluralisation, privatisation and polarisation in crime control. In: Maguire

Morgan

Reiner

(eds) The Oxford Handbook of Criminology. Oxford: Oxford University Press.

50.

Knepper

(2007) Criminology and Social Policy. London: Sage Publications.

51.

La Graffe

(2012) The youth bulge in Egypt: An intersection of demographics, security, and the Arab Spring. Journal of Strategic Security 5(2): 65–80.

52.

Levitt

(2004) Understanding why crime fell in the 1990s: Four factors that explain the decline and six that do not. The Journal of Economic Perspectives 18(1): 163–190.

53.

Mc Bride

(2011) Global Governance and ‘Post-crisis’ Employment Policy. Canadian Political Studies Association Conference, Wilfrid Laurier University, Waterloo, June 2011.

54.

Mac Donald

Mason

Shildrick

Webster

Johnston

Ridley

(2001) Snakes & ladders: In defence of studies of youth transition. Sociological Research Online 5(4).

55.

Machin

Marie

(2006) Crime and benefit sanctions. Portuguese Economic Journal 5(2): 149–165.

56.

Martin

/ OECD (2011) Unfinished business: Investing in youth. Editorial introduction to: Organisation for Economic Cooperation and Development (2011) Employment Outlook, 2011. Paris: OECD

57.

Mizen

(2006) Work and social order: The New Deal for the young unemployed. In: Goldson

Muncie

. (eds) Youth Crime and Justice: Critical Issues. London: Sage Publications.

58.

Muncie

(2001) The construction and deconstruction of crime. In: Muncie

McLaughlin

(eds) The Problem of Crime, 2nd Edition. London: Sage Publications.

59.

Muncie

(2009) Youth and Crime.London: Sage Publications.

60.

Narayan

PK and

Smyth

(2004) Crime rates, male youth unemployment and real income in Australia: Evidence from Granger causality tests. Applied Economics 36(18): 2079–2095.

61.

National Centre for Social Research (2011) The August Riots in England: Understanding the Involvement of Young People. London: National Centre for Social Research.

62.

Office for National Statistics (2012) Crime in England and Wales, Year Ending June 2012. London, Office for National Statistics, 18 October 2012. Available at: http://www.ons.gov.uk/ons/dcp171778_283456.pdf (last accessed 28 November 2012)

63.

Papoudopolos

Roumpakis

(2012) The Greek welfare state in the Age of Austerity: Antisocial policy and the politico-economic crisis. In: Kilkey

Ramia

Farnsworth

(eds) (2012) Social Policy Review, 24: Analysis and Debate in Social Policy. Bristol: Policy Press.

64.

Phillips

Land

(2012) The link between unemployment and crime rate fluctuations. Social Science Research 41(3): 681–694.

65.

Phoenix

(2010) Pre-sentence reports, magisterial discourse and agency in the youth courts in England and Wales. Punishment & Society 12(3): 348–366.

66.

Ponticelli

Voth

H-J

(2011) Austerity and Anarchy: Budget Cuts and Social Unrest in Europe, 1919–2009, Discussion Paper Series No. 8513. London: Centre for Economic Policy Research.

67.

Pouget

(2006) Youth Unemployment and Crime in France. Bonn: Institute for the Study of Labor.

68.

Pritchard

(1992) Is there a link between suicide in young men and unemployment? Acomparison of the UK with other European Community countries. The British Journal of Psychiatry 160(6): 750–756.

69.

Ramos-Diaz

Varela

(2012) From opportunity to austerity: Crisis and social policy in Spain. In: Kilkey

Ramia

Farnsworth

. (eds) (2012) Social Policy Review, 24: Analysis and Debate in Social Policy. Bristol: Policy Press.

70.

Reiner

(2007) Political economy, crime, and criminal justice. In: Maguire

Morgan

Reiner

(eds) The Oxford Handbook of Criminology. Oxford: Oxford University Press.

71.

Scorcu

Cellini

(1998) Economic activity and crime in the long run: An empirical investigation on aggregate data from Italy, 1951–1994. International Review of Law and Economics 18(1): 279–292.

72.

Smith

(2010) (ed.) A New Response to Youth Crime. Cullompton: Willan Publications.

73.

Solomon

Garside

(2008) Ten Years of Labour’sYouth Justice Reforms: An Independent Audit. London: Centre for Criminal Justice Studies.

74.

Somavia

(2010) Achieving and Sustaining an Employment-based Recovery: United States and Global Strategies for Governments, Businesses, Workers and Families, address at the Brookings Institution, Washington, DC, 26 February.

75.

Somavia

/ International Labour Organization (2012a) Director-General’s opening address to the101st International Labour Conference, Geneva, ILO, 30 May.

76.

Somavia

/ International Labour Organization (2012b) Director-General’s Address to the Plenary of the 101st International Labour Conference,Geneva, ILO, 6 June.

77.

Soothill

Francis

Fligelsone

(2002) Patterns of Offending Behaviour: A New Approach, Home Office Research Findings, No.171. London: Home Office.

78.

Stephenson

Giller

Brown

(2011) Effective Practice in Youth Justice. Cullompton: Willan Publications.

79.

Tarling

(1982) Unemployment and Crime, Research Bulletin. London: Home Office.

80.

The Economist (2012) Good news is no news, 9 July.

81.

Thomson

Hadfiled

Kehily

M-J

Sharpe

(2010) Family fortunes: An intergenerational perspective on recession. 21st-Century Society: Journal of the Academy of Social Sciences 5(2): 149–158.

82.

Webster

MacDonald

Simpso

(2006) Predicting criminality?Risk factors, neighbourhoodinfluence and desistance. Youth Justice 6(1): 7–22.

83.

Weddady

Ahmari

Steinem

(2012) (eds) Arab Spring Dreams: The Next Generation Speaks Out for Freedom and Justice from North Africa to Iran. New York: Palgrave Macmillan.

84.

White

(1989) Making ends meet: Young people, work and the criminal economy. Australian & New Zealand Journal of Criminology 22(3):136–150.

85.

World Bank (2012) World Development Report 2013: Jobs.Washington DC: World Bank.

86.

(2012) Crime, inequality and unemployment in England and Wales. Applied Economics 44(29): 3765–3775.

87.

Youth Justice Board / Ministry of Justice (2012) Youth Justice Statistics 2010/11 England and Wales. London: Ministry of Justice. Available at: http://www.justice.gov.uk/downloads/statistics/youth-justice/yjb-statistics-10-11.pdf (last accessed 1 September. 2012)