Today, the Joseph Rowntree Foundation publishes a systematic review of evidence of effects of benefit sanctions in the UK, US , EU, Australia and Canada.

It finds that the most common use of sanctions is to withdraw benefit from those who left work voluntarily or through misconduct. These sorts of sanctions disqualify people from receiving unemployment-related benefits for a period in many countries, but no evidence was found on the efficacy or effects of their use. There is the potential for such sanctions to increase in importance in the UK as the Welfare Reform White Paper seeks to encourage more workless people to take up short-hour and low-paid jobs that have a higher chance of ending quickly.

Most benefit sanctions target behaviour while on benefits. The clear evidence is that these sanctions work from effects based on:

• Anticipation – people change their behaviour before being sanctioned or while under threat of sanction.
• Imposition of sanctions – people change their behaviour having received the sanction.

Most evidence from Europe shows that the anticipation effects are high. The huge majority of evidence comes from sanctions linked to looking for work when people are unemployed – what the White Paper refers to as 'active job search'. The evidence was both from people claiming 'unemployment benefits' (Jobseeker's Allowance equivalent) and people claiming 'welfare' (Income Support equivalent, primarily lone parents in the US evidence).

The clearest and largest body of evidence shows that sanctions:

1. Reduce the time spent on out-of-work benefits.
2. Increase numbers of people leaving benefits for employment.
3. Increase numbers of people leaving, for reasons other than employment.
4. 'Second order' effects, leading to reduced caseloads and so reduced benefit spending.

But the evidence base is limited in both scope and coverage. This means that not all potential effects are looked at in the existing research and some are only covered by a small number of studies.

• There are no studies on the deterrent effect of sanctions on claiming the benefit in the first place (take-up effect).
• Few studies look at 'quality of outcomes' rather than just simple measures of duration and moving off benefits.
• Qualitative studies of impact on people who are sanctioned tend to be selective (covering only those who are sanctioned and do not leave benefits).

There is also strong evidence, though from fewer studies, on the 'unintended effects' of sanctions:

• Widespread problems of information about and communication of sanctions.
• A punitive sanction approach linked to large numbers of people 'disconnecting' from the system (20 per cent of US lone parents).
• Poor job quality and job progression of sanctioned job entrants (Swiss evidence).
• Raised levels of property crime in areas with higher sanction rates (UK evidence).

http://www.jrf.org.uk/blog/2010/12/benefit-sanctions-evidence-review