CRIME REDUCTION: Are Government policies likely to achieve its declared aims?
David G. Green
The Government has set itself some policy targets, but has it adopted policies that are likely to achieve them?
1. The Government’s aims
The Home Office has seven official aims, of which five relate to crime. The first is to reduce crime and the fear of crime. It’s favoured measure of crime is the British Crime Survey (BCS). It has specific targets for three offences: domestic burglary, vehicle crime and robbery. In its departmental report for 2003 (May) the Home Office says that Public Service Agreement (PSA) Target 1 is to reduce recorded vehicle crime, domestic burglary and robbery.(1) However, the technical notes on the PSA targets published in March 2003 say that vehicle crime and domestic burglary will be measured by the BCS, whereas robbery will be measured by police records. These were the sources used by Professor Michael Barber, head of the Prime Minister’s Delivery Unit when presenting the Government’s progress in July 2003.(2) We follow the approach described in the technical notes, not the Home Office Departmental Report.
The baseline for vehicle crime is the 2000 BCS (crime in 1999): 2,941,927 crimes. The target is a reduction of 30% to 2,059,349 crimes by 2004/05. The 2002/03 BCS found 2,366,070 vehicle crimes. In 2004/05 there were 1,886 vehicle crimes. (Crime in England and Wales 2004/05, Table 2.01.)
The baseline for domestic burglary is also the 2000 BCS: 1,261,364 crimes. The target is a 25% reduction to 946,023 by 2005/06. The 2002/03 BCS found 973,522 domestic burglaries. In 2004/05 there were 756,000. (Crime in England and Wales 2004/05, Table 2.01.)
The baseline for robbery is police records for 1999/2000: 68,782 crimes in the ten street crime initiative areas. The target is a reduction in those areas of 14% to 59,153 crimes. In 2002/03 the police recorded 83,661 robberies in the ten areas, well above the baseline. The Departmental Report for 2004/05 accurately reports that there were 76,776 robberies in 2003/04, an increase of 12% (p. 8).
2. The Government’s policies for achieving its aims.
There are a number of statements describing the policies the Government intends to pursue, but they do not always correspond to what is happening on the ground. Policy towards the use of prison provides an example. The white paper of July 2003, Justice for All, contained some tough talking. The Government boasted that it had already increased prison capacity by 18%(3) and said that its sentencing policy would protect the public, punish offenders, and encourage them to make amends for their crime.
So far so good, but it then went on to say that: ‘Our aim is not to increase the prison population’ but to make sure that people receive the right punishment.(4) Legislation, it said, ‘will make it clear that custody should be used only when no other sentence would be adequate’.(5) The cost of prison appeared to be the main concern:
‘Custody has an important role to play in punishing offenders and protecting the public. But it is an expensive resource which should be focused on dangerous, serious, seriously persistent offenders and those who have consistently breached community sentences.'(6)
The use of the term ‘seriously persistent’ rather than just plain ‘persistent’ suggests that the Government is prepared to tolerate a certain amount of reoffending. And similarly, use of the phrase ‘consistently breach community sentences’ suggests that criminals will be allowed to ignore community sentences on more than one occasion before serious action is taken.
In addition to wanting to avoid the cost of new prisons, the Government also appears to believe that prison can make matters worse. Justice for All repeated some common mistakes about prison.
Short prison sentences increase offending:
Justice for All noted that prisoners given short sentences were reconvicted at a higher rate than those who served longer sentences and concluded that short spells in prison, ‘increase the chances of reoffending’.(7)
It is true that short sentences are associated with a higher re-conviction rate than longer sentences, but to claim that short sentences are the cause of this re-offending implies that a prisoner’s subsequent conduct is determined by his short time in jail. Consider someone aged 20 who has been a regular offender and only just been caught. How likely is it that 3 months in jail will become the main cause of his later conduct? It is more likely that the attitudes acquired in the previous 20 years continue to exert a powerful influence.
It would be more true to say that the short prison sentence only provides the public with a short respite from persistent offenders. The pattern of continued offending is not the result of prison, but rather the consequence of a failure to imprison habitual offenders for long enough to protect the public, not to mention the failure to discourage offending at an earlier stage. In taking this line the Government disregarded the findings of the Social Exclusion Unit’s (SEU) report on the consequences of repeat offending. Failure to jail repeat offenders in the hope of reducing expenditure on prison could be a false economy. The SEU estimated that an unrepentant offender would cost the criminal justice system an additional £65,000 per year. To try an offender again at a Crown Court would alone cost on average £30,500.(8) The full-year cost of keeping an adult offender in a local prison is on average £23,700.(9)
Prison causes family break-up:
Justice for All also claimed that prison can make matters worse in another respect. It can ‘break up families, impede resettlement and place children at risk of an inter-generational cycle of crime’. Over 40% of sentenced prisoners claim to have lost contact with their families since entering prison, yet, according to the white paper, ‘Research shows that prisoners are six times less likely to reoffend if contact with their families is maintained’.(10)
There are two main problems with this statement. First, it is not true that simply reinstating family contacts will reduce offending. The SEU reported in 2002 that 43% of prisoners had other family members who had been convicted (compared with 16% of the general population) and 35% had a family member who had been in prison. In such cases the family may be a bad influence. The Government’s manner of reasoning has been called the ‘ecological fallacy’ by statisticians. This fallacy assumes that the average or predominant characteristics of a group apply to all the individual members. Thus, while it is true that people with strong family ties are less likely to be criminals, it does not follow that all people with strong family ties are law abiding. They are not. Moreover, to leave criminals in the community in the belief that they will automatically have stronger family ties will often lead to no change in their family circumstances, while leaving the public exposed to crime.
Second, it is not true that prison always causes the breakdown of family contacts. Many criminals had few, if any, close family ties before admission to prison. The SEU report shows that 47% of male prisoners had run away from home as a child, and 27% had been in care (compared with 2% of the general population).(11) Some 81% were unmarried prior to imprisonment, nearly 5% were sleeping rough before admission, and 32% were not living in permanent accommodation prior to their imprisonment.(12) Moreover, when their family disowns them or a wife leaves them, it is often because they disapprove of the prisoner’s self-chosen conduct. In such cases, the prisoner’s law-breaking is the cause of the breakdown, not prison as such.
To summarise: Government policy reflects two potentially contradictory lines of thought. Many in the Home Office are hostile to prison and want to reduce its use, but their influence has been tempered by acknowledgement that community sentences do not adequately protect the public. This realism has led the Government to the search for ‘tough community sentences’ that are a ‘credible alternative to custody’, including community sentences with multiple conditions like tagging, reparation and drug treatment and testing. It is imperative, according to the Government, that ‘we have a correctional system which punishes but also reduces reoffending through the rehabilitation of the offender’.(13) Consequently, it says, a genuine third option is needed in addition to custody and community punishment.
The most important of these third options to be implemented so far is the Intensive Supervision and Surveillance Programme (ISSP) which began in April 2001 with an investment of £45m over three years.
3. Have the Government’s policies worked?
The Intensive Supervision and Surveillance Programme (ISSP)
According to Justice for All, ISSP is the most rigorous, non-custodial intervention available for young offenders. It initially targeted 2,500 of the most serious and prolific young offenders (aged 10 to 17) per year. They were thought to be responsible for a quarter of all youth crime. Young offenders on ISSP can be subject to intensive monitoring for up to 24 hours a day, seven days a week, if necessary (although usually for a far shorter period). Electronic tagging and voice verification (telephone checking of an offender’s ‘voice print’) can be used to monitor offenders, as well as intelligence-led policing and ‘tracking’ of their movements by probation officers. The minimum requirement is for two surveillance checks per day.
They are also subject to a structured programme of activities for 25 hours a week for 3 months. Core elements include education and training, interventions to tackle offending behaviour and reparation to victims. It is available for convicted young offenders and to prevent persistent young offenders on bail from committing more crimes while awaiting trial. According to the Youth Justice Board, however, ISSP is restricted to the most serious offenders, defined as those who have been charged or convicted of an offence and have either been charged or warned for an imprisonable offence on four or more separate occasions within the past 12 months, or previously received at least one community or custodial penalty.
Most young people will spend six months on ISSP. The most intensive supervision (25 hours per week) lasts for the first three months of the programme, after which the supervision continues at a reduced intensity (a minimum of five hours per week) for a further three months.
The YJB claimed that ISSP was based on the best evidence of what will reduce the frequency and seriousness of offending. It would bring structure to offenders’ lifestyles, and tackle the factors contributing to their offending behaviour, lack of educational qualifications, weaknesses in thinking skills, or drug misuse. But when the final report was published in October 2005, the reconviction rate was 91% and the programme had been less effective than ordinary probation in reducing the frequency and seriousness of offending. This result was not surprising, given the weak evidence originally relied on.
The Rotherham pilot scheme:
In July 2003 the main evidence drawn from England and Wales was based on a pilot scheme in Rotherham conducted from August 2000 to June 2001. The government web site www.crimereduction.gov.uk claims that 27 young offenders in Rotherham had been convicted for 160 offences in the 9 months before the scheme, but during the programme they committed only 47 offences. The final evaluation of the Rotherham Intensive Supervision, Support and Advocacy Programme (RISSAP), however, was never published, but we managed to obtain a copy. The report does conclude that RISSAP was successful but a careful reading reveals why it was never published in full. It found that 7% more of the programme group did not offend by comparison with the control group.
However, the programme group was 35 strong and the control group 18. Seven out of 18 members of the control group did not offend (39%) compared with 16 out of 35 members of the programme group (46%). If one person had changed sides in each group, so that 8 out of 18 (44%) control-group members and 15 out of 35 (43%) programme-group members had not re-offended, the result would have been reversed. This is hardly a scientifically valid study on which a multi-million programme ought to be based. Yet in the summer of 2003 it remained the best evidence available to the Government.
In July 2003 this scheme provided the only evidence from England and Wales about the impact of ISSP on reoffending. Pilot schemes had been established in April 2001 and more in July 2001 on the flimsiest of evidence. In April 2003 another variation was introduced, the Intensive Control and Change Programme for 18-20 year-olds. It was similar to ISSP with a stronger focus on a scheme for compensating victims.
Tagging improves compliance with rehabilitation programmes:
A Youth Justice Board ‘briefing’ about ISSP is available on their web site, but in 2003 provided no systematic information about reoffending. This online briefing is said to based on ‘research collated by Oxford University and PA Consulting’. Among its claims is this: ‘Research from Canada has shown that electronic monitoring can aid effective rehabilitation by improving compliance with more rehabilitative community interventions.’ The source for this claim is given as ‘Bonta, Rooney and Wallace-Copreta, Electronic Monitoring in Canada, 1999.'(14) This sentence from the Youth Justice Board briefing, however, reports only part of the Canadian findings, and the least significant at that. Moreover, closer examination reveals it to be a carefully worded sentence that manages to be highly misleading without being completely untrue.
The main finding of the study, Electronic Monitoring in Canada, was that recidivism rates of offenders did not change significantly as a result of electronic monitoring (EM). The study considered three methods of punishment and control: EM, probation and prison. The crude scores initially suggest a benefit from EM. Reconviction rates were 26.7% for EM participants, 33.3% for probationers and 37.9% for offenders who had been imprisoned.(15) However, the three groups were not equal. They had all been assessed on a ‘risk-needs’ scale previously found to be a reliable indicator of future offending. The report concluded that the ‘lower recidivism rates found with EM participants could be explained by the differences in risk-needs levels’. Consequently, the reduced re-offending was not the result of the type of sanction. The researchers concluded that adding electronic monitoring to the supervision of offenders had ‘little effect on recidivism’.(16) Moreover, they found no lasting post-programme effect on criminal behaviour.(17)
Offenders also took part in an intensive treatment programme, based on cognitive-behavioural psychology (to be explained below) consisting of 2.5 hours per day for 4 days a week. Those on EM and probation were combined to form a treatment group and compared with a control group of offenders who had been sentenced to prison. The overall result showed slightly worse results for the treatment group (32.4% re-offended compared with 31.0% of the control). However, when the groups were divided according to their risk-needs score it was found that recidivism for high-risk offenders was lower (31.6% compared with 51.1%). The researchers conjectured that the educational scheme was more effective for high-risk offenders, although they were unable to explain why. Treatment resources, they said, were ‘wasted’ on low-risk offenders.(18)
The only benefit of EM was the one mentioned by the Youth Justice Board, namely that offenders subject to EM who attended the educational programme were more likely to complete it than the probationers (87.5% for those on EM and 52.9% for those on probation). However, the probation sample was small (17 people) and the researchers thought that the additional requirements of EM, including ‘the threat of a return to prison for non-cooperation’ might explain the difference.(19)
Thus, the Youth Justice Board is claiming that this scheme is evidence of the efficacy of tagging under a community sentence, when it is really evidence of the deterrent effect of prison. A more impartial summary of the Canadian findings would have been that EM made no difference to re-offending, that the rehabilitation programme was assoicated with reduced offending by high-risk offenders but made no overall difference (when all risk groups were combined), and that the threat of a return to prison encouraged those on EM to complete the rehabilitation course. Altogether, these findings do not add up to a very powerful case for an ISSP scheme which combines EM and rehabilitation in the community.
What happened when rehabilitation schemes based on the same cognitive-behavioural principles as the Canadian programmes were attempted here?
A central feature of ISSP is the use of cognitive skills courses to encourage offenders to alter their attitudes. Cognitive skills courses were first introduced in 1992 and have been stepped up under the Blair Government. They are based on the belief that criminals carry out crimes because of mistaken beliefs. They might tell themselves that no one gets hurt (they are all insured) or interpret innocent actions as aggressive (demanding to know ‘what are you looking at’ if you catch their eye in the street ) or they may simply be unable to put themselves in other people’s shoes. Psychologists claim to know how to alter these attitudes and the Home Office has been stepping up the number of offending behaviour programmes inspired by their theories.
Two evaluations have been published by the Home Office. Findings 161 claimed that the courses were effective. However, a closer look reveals that cognitive-behavioural treatment made no real difference. The first bullet point says: “Reconviction fell considerably after cognitive skills treatment. For example, two-year reconviction rates for treatment groups were up to 14 percentage points lower than matched comparison groups.” It claims that this represents 21,000 crimes prevented.
However, it turns out, if you read on that the 14% reduction was for those classified as ‘medium-low’ risk. There were four groups: low risk of conviction, medium-low, medium-high and high. But if you add up the reconvictions for all four groups, the rate is 44%, not the 18% for the medium-low group. And if you compare this proportion with the control group, the overall reconviction rate of the treatment group was worse than that of the control group (44% compared with 40%). The Home Office has pointed out in private correspondence with Civitas that the treatment group had a higher risk of offending because, inexplicably, the samples were not matched. After allowing for differences in the predicted risk of re-offending the results of the scheme were neutral. That is, the treatment made no difference.
Perhaps the treatment works well with medium-low risk offenders and not with high-risk offenders, in which case it would be legitimate to break the treatment group into such categories. However, in a journal article (in Legal and Criminological Psychology) the authors tested the possibility that the treatment might be more effective for medium-risk offenders. They found that there was no ‘statistically significant interaction’ and concluded that ‘this suggested that treatment impacts on reconviction rates despite the offenders’ prior level of risk of reconviction’. In other words, there was no apparent justification for breaking the sample down into groups. In any event, previous studies in Canada (above) suggested that the schemes worked best on high-risk offenders.
It is, perhaps no surprise that a subsequent Home Office publication in July 2003, Findings 206, acknowledged that the schemes failed. The summary states: ‘This evaluation found no differences in the two-year reconviction rates for prisoners who had participated in a cognitive skills programme between 1996-1998 and a matched comparison group. This contrasts with the reduction in reconviction shown in the previous evaluation of cognitive skills programmes for prisoners, delivered between 1992-1996.'(20)
Tagging reduces offending:
The Government not only believes that tagging increases attendance of rehabilitation courses. It is also relying heavily on tagging to protect the public from offenders serving community sentences. However, the only study of tagging so far to look at the impact on re-conviction rates after tagging ceased found that it made no significant difference. Another study has looked at the impact on offending while the tagging was being carried out and found that offending was reduced, but the tagging only lasted from 2-4 months. The Government points out that only 2.1% offended while tagged, and this is undoubtedly a gain over the normal rate of reoffending, but it only lasts for a short time, and to claim that it is an adequate substitute for prison is highly misleading.
The best evidence we have so far is from the home detention curfew (HDC) scheme. Usually called ‘tagging’ the scheme was introduced throughout England and Wales in January 1999. The Home Office has reported on the results after the first 16 months of the scheme.(21) Prisoners could be released for up to 60 days (now 90 days) before the normal end of their sentence. The maximum period of tagging was 60 days and minimum 14. The curfew applied for a minimum of 9 hours and a maximum of 12.
Mr Blunkett has claimed that tagging is an important part of his strategy for cutting reconviction rates. In a press release (21 March 2002) about the home detention curfew he said:
‘Reconviction rates are dramatically lower for those who have been released under HDC than those who have served the final weeks of their sentence in prison. It is an important part of our crime reduction package and our drive to cut reconviction rates.’
This statement implies that tagging will lead to a reduction in reoffending, but a closer look reveals that the lower re-conviction rate reflects the selection process. Prisoners chosen for early release under HDC were considered by prison governors to be less likely to reoffend, and were released for that reason. The study confirms that prison governors made reasonably reliable assessments of the risk of re-offending, but there was no lasting effect on re-offending once the tag had been removed.
Some 21,400 prisoners were released under the scheme, of which 1,100 (5%) were recalled to prison because they failed to comply with the conditions. The re-conviction rate during HDC was 2.1%.(22) The reconviction rate for those subject to HDC during the six month period after their automatic release date was 9.3 per cent, compared with a rate of 40.5 per cent of those who were refused HDC.(23) As Home Office researchers pointed out, this is evidence that prison governors made accurate assessments of the risk posed by offenders.
Reasonable people could disagree about whether or not early release under tag has been justified. Tagging is cheaper and releases prison space. These benefits have to be set against the cost in human suffering and money of the crimes that were committed but which would have been prevented by imprisonment. A trade-off of this sort can be made when comparing how best to deal with offenders during the last 60 days of their prison term. But what is the evidence that tagging has a more lasting effect on reoffending? Moreover, under ISSP, tagging is applied to serious offenders, not those hand-picked by prison governors because they present a low risk to society.
To gain more insight into the enduring effect of tagging, a short-term reconviction analysis was carried out on a sample of prisoners who were eligible for discharge under HDC in May and June 1999, some of whom were released on HDC and some of whom were not. This ‘programme group’ was compared with a ‘control group’ of similar prisoners discharged in October and November 1998 who would have been eligible for consideration for HDC had it been in force at the time.
In the six months after the normal discharge date, offenders eligible for HDC had very similar reconviction rates to the control group (30.5% and 30.0% respectively, or 30.8% and 30.0% if offences during the curfew period were included). The rates of reconviction were much higher for some groups, defined by their original offence. Of those serving a sentence for violence against the person 6% were reconvicted within only 4 months, of those sentenced for theft 11% had been reconvicted within 4 months and of those sentenced for burglary, 22%.(24) The researchers concluded that this suggests that the impact of HDC was ‘broadly neutral’ in terms of reoffending when compared with the results for the control group.(25)
There are two main elements of the new community sentences, summed up by the name of the latest variant, the Intensive Control and change Programme (ICCP). This scheme is for 18-20 year-olds but similar principles are applied to younger age groups, under the name Intensive Supervision and Surveillance Programme (ISSP).
The ‘control’ element includes curfew or home detention and monitoring by probation officers. HDC is at best neutral in its impact on reoffending after tagging is terminated. While they were tagged for up to 60 days it had a significant restraining effect on low-risk offenders, though not as powerful as prison, which prevented any offences being committed against members of the public. To apply tagging to the 2,500 most serious offenders in the 10-17 age-group based on evidence of its effectiveness among hand-picked low-risk offenders is a highly risky venture.
The ‘change’ element includes teaching educational skills, both general and vocational, and providing ‘offending behaviour programmes’ inspired by cognitive-behavioural therapy. What we know about the effectiveness of these programmes suggests they are highly imperfect substitutes for prison.
In July 1998 the Home Affairs Select Committee concluded that: ‘The absence of rigorous assessment of the effectiveness of community sentences is astonishing. Without it confidence in them must be limited and sentencing policy a matter of guess-work and optimism’.(26) By the second half of 2003 the position had hardly altered, although more evidence about ISSP should start to emerge in 2004.
1. Home Office Departmental Report 2003, p. 16.
2. Home Office, SR2002 Public Services Agreement, Technical Notes, March 2003.
3. Justice for All, p. 18.
4. Justice for All, p. 33.
5. Justice for All, p. 88.
6. Justice for All, p. 87.
7. Justice for All, p. 102.
8. Social Exclusion Unit, Reducing Re-offending by Ex-prisoners, 2002, p. 5.
9. SEU, p. 31.
10. SEU, p. 85, 106.
11. SEU, p. 18.
12. SEU, pp. 19, 21.
13. Justice for All, p. 87.
14. ‘Intensive Supervision and Surveillance Programmes’: http://www.youth-justice-board.gov.uk/NR/rdonlyres/6C269732-E2FF-4DFD-9265-552A68990702/398/ISSPvsshortDTObriefing1Jan03.doc
15. Bonta et al., Electronic Monitoring in Canada, pp. 46-7.
16. Bonta et al., pp. 47-48.
17. Bonta et al., p. 54.
18. Bonta et al., p. 52.
19. Bonta et al., p. 16.
20. Falshaw, L. et al, Searching for ‘What Works’: an evaluation of cognitive skills programmes’.
21. Dodgson, K., et al., Electronic monitoring of released prisoners: an evaluation of the Home Detention Curfew scheme, Home Office Research Study 222. Home Office, March 2001.
22. HORS 222, p. 52.
23. HORS 222, p. 54.
24. HORS 222, p. 60.
25. HORS 222, p. 54, p. 61.
26. Alternatives to Prison Sentences, Session 1997-98, 3rd Report, HC486, July 1998, para 249.