Thursday, November 08, 2012

South Africa: Hit and Miss? Ensuring that Prisoner Rehabilitation Programmes Work

Hit and Miss? Ensuring that Prisoner Rehabilitation Programmes Work

Source: ISS

Chandre Gould,  Senior researcher, Crime and Justice Programme, ISS Pretoria

Prisons are often referred to as ‘universities of crime’ because of the substantial number of convicts who re-offend after being released. It is for this reason that the South African Department of Correctional Services (DCS) states that the rehabilitation of inmates is one of its key priorities. However, whether ‘rehabilitation’ is even possible and if so, how to achieve it, is the subject of much debate and scholarly work.

A reintegration programme for prisoners that combines getting them ready to take up jobs and resist the lure of alcohol and drugs, seeing to their mental health and providing them with support and services for two years after they leave prison, sounds like a good idea. And yet, the results from a controlled experiment aimed at measuring the effectiveness of just such an initiative that was developed and implemented in the US showed that prisoners in this programme fared no better, or worse, than those who did not receive the intervention.

Such interventions can prove expensive and resource intensive. The finding therefore reinforces the importance of measuring the effectiveness of ‘rehabilitation’ and reintegration programmes for offenders so that we can avoid spending money and time on those things that don’t work, and focus on those that do.

Professor Sheldon Zhang shared the findings of a randomised control trial at the ISS’s 3rd international conference on ‘National and international perspectives on crime reduction and criminal justice’ at the end of October. The experiment tested whether those who took part in the programme did better than those who did not. Did they find and keep jobs more effectively? Were they less likely to be re-arrested? Were they more successful in finding stable accommodation and resisting drugs? For each of these criteria the difference between those who participated in the programme and those who did not was negligible. This raised the difficult question for those funding and implementing the programme of whether it was worth offering at all. This is the question that South Africa should consider in relation to the ‘rehabilitation’ programmes offered to some offenders in our prisons too.

According to the most recent 2011/12 budget vote for the DCS, 28 ‘rehabilitation’ programmes are available to South African prisoners. Prisoners’ access to these programmes is determined by their sentence plan, which is drawn up by DCS officials after an assessment has taken place. The programmes offered include inter alia interventions aimed at assisting the participant to manage anger, address alcohol and drug dependence, address sexual offending and understand restorative justice, and at providing the skills necessary to cope with life after prison. Most of these programmes are offered by non-governmental organisations (NGOs) and faith-based organisations. The availability of these programmes depends primarily on whether there are organisations that are able to offer them in the geographical area where the prison is based. There is a cost both to the state and to the donors that support the relevant NGOs.

The cost to the DCS is really only the cost of the 75 staff members allocated to the rehabilitation work of the department. It therefore costs taxpayers some R50,7 million a year (of which 94% is for salaries). While this is a small percentage of the Department’s total budget of R15,34 billion, it is important to ensure that it is spent on interventions that can be shown to be effective. It is very difficult to assess the exact cost to the NGO and donor community, that run most of the rehabilitation programmes. Since 68% of the cost of the reintegration programmes offered in correctional services is carried by NGOs, the cost to donors may amount to as much as an additional R107 million a year.

Assessing the effectiveness of these programmes is not only important because we need to make sure that money is well spent, but also because ineffective programmes will not prevent repeat offending. Indeed, there is very little evidence to suggest that anything except the most sophisticated, individually tailored rehabilitation programmes do in fact reduce recidivism. However, if a proper study were to find that the programmes being offered in our prisons do in fact work, then the argument could be made that the DCS and the donor community should dedicate even more resources to making sure all prisoners have access to these programmes.

Zhang offered three good reasons for why randomised control trials are worth the trouble in assessing the outcomes of such programmes:
  • Limited resources for such interventions demand that there is accountability for programme implementation and the outcomes
  • Rigorous evaluation design saves money by avoiding ineffective programmes and therefore wasteful spending on these intervention efforts
  • Randomised control trials (RCTs) are easy to implement and can be done without sophisticated statistical analysis being required
However, Zhang warns that RCTs are inflexible in that you have to have a control group (people who have not been offered the service) in order to test whether there is a difference in outcome between those who have access to the intervention and those who don’t. In other words, you have to do this kind of study right if the results are to be reliable. As the state has moved to measuring outcomes to hold departments accountable for public expenditure, conducting these types of assessments are important to ensure that resources are directed towards those programmes that are most effective.