Written evidence submitted by Centre for Economic and Social Inclusion (SAN0143)

About the Centre for Economic and Social Inclusion

1. The Centre for Economic and Social Inclusion is the leading UK organisation dedicated to tackling disadvantage and promoting social inclusion in the labour market. We aim to do this through high quality research, thought leadership and by building communities that can learn from each other through events, training and online networks. 
2. Inclusion provides services across the UK to a range of clients in central and local government, key public bodies, the voluntary and the private sectors.  This includes the Department for Work and Pensions – for whom we are evaluating the Universal Support trials, Supervised Jobsearch Pilots and (with others) the Work Programme; local government; international bodies; and trusts and foundations.
3. This research draws on our own analysis of labour market and sanctions data; previous research on conditionality and sanctions regimes; and extensive recent qualitative research on welfare reform impacts – interviewing benefit claimants affected by sanctions and changes to benefit conditions.

Introduction

“The unemployed should be required, as a condition of continued benefit, to attend a work or training centre, such attendance being designed as a means of preventing habituation to idleness and as a means of improving capacity for earnings.”

Beveridge Report, 1942

4. The principle of making unemployment benefits conditional on taking steps towards work is not a new one.  Indeed it pre-dates even Beveridge – with cases dating back more than a hundred years where unemployment insurance was stopped for not taking up a job offer.
5. However the growth in the use of sanctions in recent years – and in particular since the introduction of new rules with the Welfare Reform Act 2012 – is perhaps without precedent.  As Figure 1 below shows, 6% of all claimants of Jobseeker’s Allowance (JSA) are sanctioned every month[1] – the highest rate since records began in 2000, and likely to be higher than at any point before then.

Figure 1 – Proportion of JSA claimants sanctioned each month

Source: Stat X-Plore, NOMIS and Inclusion calculations

6. These figures do not include those cases where claimants are discouraged from claiming or staying on benefit.  Direct data on this is not available, although we have found a number of examples of this in our qualitative research (particularly amongst younger unemployed people).  However it is striking that the share of the unemployed that are not claiming Jobseeker’s Allowance has risen relentlessly since around the time that new sanctions rules were implemented – from 37% of the unemployed in late 2012 to 51% now[2].  So, just over half of all unemployed people are now not receiving Jobseeker’s Allowance, and therefore not receiving support to get back into work.
7. Perhaps of even more concern, in recent years we have seen growing numbers of sanctions being applied to claimants who have a health condition or impairment that limits their ability to work (claimants of Employment and Support Allowance, or ESA).  The number of ESA claimants sanctioned has risen seven-fold since 2012 – from 2,201 in the first quarter of 2012 to 15,955 in the first quarter of 2014.
8. These large increases in the incidence of sanctions should be of deep concern.  The application of a sanction represents a failure – whether that is of an institution, an individual, a wider system, or a combination of all three – and we know that this can have profoundly damaging impacts on those affected.  We should all have an interest in understanding if there are ways that we can improve this system.  For this reason, we warmly welcome the Committee’s inquiries in this area.

Conditionality, support and sanctions

9. In our view, the design and application of sanctions needs to be considered alongside the design of benefit conditions and also employment support.
10. There are strong arguments for placing clear conditions on the receipt of unemployment benefits, with the threat of sanctions where conditions are not met.   However we believe that there is strong evidence from the UK and internationally that the most effective systems combine clear conditions with the right support to meet them.
11. In the UK, this ‘something for something’ approach has proved to be highly effective – leading to more employment, fewer people on benefits and reduced inequality in the labour market.  Probably the best example, given by Paul Gregg in his 2008 review of conditionality[3], is the New Deal ‘gateway’: where intensive adviser support was matched with increased requirements to attend interviews and undertake jobsearch.  Figure 2 below illustrates this – with the likelihood of a claimant leaving JSA increasing sharply around the time that the ‘Gateway’ began (26 weeks – with an apparent ‘deterrent effect’ earlier than this) and then continuing to rise during that period.

Figure 2 – Proportion of young people flowing off Jobseeker’s Allowance each week by length of claim (2006/7)

Source: Gregg Review

12. Looking internationally, there is also evidence that these models are often more effective than those with less conditional support.  In one major study of 137 programmes across 19 countries[4], that grouped interventions into one of six categories, it was found concludes that so-called ‘services and sanction’ models – that combine jobsearch support with the threat of sanction – were substantially more effective, and more cost-effective, than any other type of labour market programme.  These impacts were clear across labour markets and groups, and over the longer term as well as short term.
13. We believe that within this approach, the threat of sanctions is a necessary part of having a conditional system.  Repeated studies find that claimants report that knowledge of sanctions is a motivation to comply with benefit conditions – with recent DWP research[5] suggesting the same for a majority of both ESA and JSA claimants.  However, knowledge of sanctions is not enough – the same DWP research concludes that “there was no evidence ... that knowledge of JSA conditions led to actual movement into work.”  The threat of sanctions needs to matched with the right support.
14. However while in our view there is a generally strong case for a system based on conditionality and support for the unemployed, the evidence is not entirely clear-cut.  First, it is important to recognise that conditionality is generally better at moving people off benefit than it is at moving people into work (a point illustrated in evaluations of the Jobseeker’s Allowance regime).  This points to the need for clearly setting objectives around increasing employment rather than reducing benefit receipt.
15. In addition, importantly, there is generally weak evidence on the effectiveness of conditions and sanctions for other groups.  There is in particular no strong evidence that Work Focused Interviews have led to any positive impacts for those on incapacity benefits or ESA, and a wealth of qualitative evidence of negative impacts on those sanctioned.

The role of sanctions

16. We have set out that the threat of sanctions is a necessary part of the system for the unemployed in the UK.  However, the evidence from a range of studies on the actual imposition of sanctions suggests that the impacts are far from clear cut, and over the longer term are more often than not negative. 
17. As was concluded in a comprehensive review of sanctions impacts in the UK and internationally, conducted for the Joseph Rowntree Fondation[6]: “[sanctions] strongly reduce benefit use and raise exits from benefits, but have generally unfavourable effects on longer-term outcomes (earnings over time, child welfare, job quality) and spill-over effects (i.e. crime rates).”
18. We agree that the right principles for a sanctions system should be those set out by Paul Gregg in his review of conditionality, and expanded by Matthew Oakley in his sanctions review – that the system is fair, proportionate, well understood, and with a clear focus on improving employment outcomes.
19. In addition we would argue that a successful sanctions system should be one that is used sparingly and as a last resort.  A system with high rates of sanction would represent a failure – with individuals not understanding their responsibilities, institutions unable to enforce them properly, and families and communities most likely worse off.  Instead, over the last two years, we have instead seen an unprecedented increase in the use of sanctions and an inexplicable increase in their severity. 

The extent and severity of sanctions

20. In the space of less than two years (October 2012 to June 2014), 834,000 individual JSA claimants have been sanctioned.  And for every one claimant who has been sanctioned, there has been roughly one more referral where a sanction has not been applied or has been overturned – causing stress, anxiety and often a temporary loss of income.
21. Just over half of sanctions applied have been so-called ‘Low Level’ sanctions – leading to a loss of benefit for four weeks, rising to 13 weeks for repeated failures.  However 110,000 claimants have had at least one ‘High Level’ sanction – which leads to loss of benefit for 13 weeks, rising to 26 weeks for a second offence and three years for a third.  So far, 1,800 claimants have had this sanction imposed – a figure that will only rise. 
22. The previous system was largely based on benefits being temporarily stopped until claimants re-engaged, with sanctions of usually less than four weeks where they were applied.  There was and is no clear rationale for the substantial increases in the severity of sanctions since 2012 – in particular, we are aware of no evidence that longer sanctions act as a stronger motivator to engage with support or to find work.

Sanctions and disadvantaged groups

23. The Committee is seeking evidence on the application of sanctions among particular groups, including those disadvantaged in the labour market.
24. Our analysis suggests that there is clear evidence that young people are being disproportionately sanctioned, but that the available evidence does not suggest that disabled people, lone parents or black and minority ethnic groups are disproportionately sanctioned.  However there are some caveats to this below, as our information on disability and health is limited.

• Young people (18-24) account for 24% of the claimant count, but 38% of those sanctioned.  In other words, young people are around 1.5 times more likely to be sanctioned than we would expect.  This was also reiterated in a survey in mid-2013 for DWP of 1,900 young JSA claimants[7], which included the startling fact that fully one third of all young claimants reported that they had had their benefit stopped or reduced. 

• Disabled people account for 21% of all JSA sanctions.  Recent data on the extent of (declared) disability is not available, but data from 2013[8] suggests that around one in five JSA claimants report that they are disabled – so at a very high level (all disabled people) there is no evidence of disproportionate impacts.
• Ethnic minorities make up one in five JSA claimants and one in five of those sanctioned – there are also no variations for different minority groups.
• Finally, lone parents appear to be substantially less likely to be sanctioned than we would expect – they make up around 11% of the claimant count, but 6% of those sanctioned.  This may reflect the fact that many lone parents are not subject to the same worksearch requirements as other JSA claimants.

25. These findings are overall somewhat reassuring, although it is important to note that there is no data available on the incidence of sanctions amongst more disadvantaged groups (for example those with mental health conditions, vulnerably housed, addictions, etc), where we might expect to find claimants more likely to fail to meet their conditions. 
26. Indeed there is some evidence, from DWP research, that some disadvantaged groups may be being disproportionately affected by the new regime.  The evaluation of the Youth Contract[9], in particular, highlighted that “Claimants with a longstanding health condition or disability were more likely than other claimants to say their benefit had been stopped or reduced (42 per cent compared with 30 per cent), and this was also higher among those who had been out of work for longer (32 per cent of those who had been out of work for a year or more).”
27. Our local research has found good, recent examples where DWP and Jobcentre Plus have been developing local approaches to better support those with chaotic lives – for example through closer working with other key workers, and developing ‘early warning’ systems.  Where this is happening it is welcome, and we would urge similar approaches to be adopted across the network.

The impacts of sanctions

28. Other submissions will doubtless bring home the very real impacts that sanctions are having on claimants, their families and on local communities and services.  Our research in local communities has found a range of impacts.
29. Among JSA claimants, we have identified two distinct groups:

• Those who experience only one sanction, are generally motivated to work, and may have increased their jobsearch as a consequence of this; and
• Those who experience a series of sanctions, report very negative experiences of support, often have more complex needs, and often become dislocated, isolated, and in increasing hardship.

30. Among ESA claimants, we have found nothing but negative experiences – claimants with often complex needs, severe health conditions, poor awareness and understanding of the system, and very significant impacts on finances, wellbeing and health.
31. It is important also to recognise the very real, and negative, impacts that the sanctions regime is having on the standing of Jobcentre Plus – and the Work Programme – within communities.  This is sometimes based on individuals’ own experiences of sanctions, but often also on stories from families, friends and networks.  The consequence is more often than not that marginalised and disadvantaged groups are reluctant to seek support, for fear that this will raise issues around their own entitlement to benefits.

Reforming sanctions

32. Our starting point, like the Committee’s, is to pick up where the review by Matthew Oakley left off.  We strongly support his conclusions and urge the government to adopt them in full.
33. We were disappointed that the Terms of Reference for his review were limited to how sanctions are communicated.  We believe that a wider review, looking at the purpose and design of the sanctions regime, should be an early priority for a new government.  A key starting point should be to review the use of sanctions internationally.  This is covered in some depth by Griggs and Evans[10].  Many or most of the systems that they reviewed had shorter and more flexible sanctions rules, often with greater discretion and personalisation.  Many also included access to hardship support that could effectively reinstate most of the lost benefit.  However they also identified more punitive models – notably in Switzerland and in the US (which included the original ‘three strikes’ models in some states, where repeated sanctions could lead to long-term loss of benefit).  There would be value in bringing this analysis up to date – as some countries have made their systems far tougher (e.g. Australia) while others have made them arguably less punitive (some US states) – and reviewing in much more depth how sanctions and support are applied and communicated in different systems. 
34. Any review should also look at how sanctions, support and conditionality fit together.  In our view, this must be viewed as a single system – with clear expectations alongside the right support.  We believe that as part of this, there is a clear case for a more personalised approach that draws on behavioural insights – we support, in particular, the idea of using non-financial sanctions or early warnings before financial sanctions are imposed.
35. As noted, we have a number of concerns around the growth in sanctions among ESA claimants in particular.  There is very little evidence that conditionality has been effective for this group, low understanding of the sanctions system, and overwhelmingly negative impacts where sanctions are applied.  In our view there are strong arguments for suspending the sanctions regime until we have a clearer model for how we support ESA claimants to prepare for and find employment.
36. We also believe that there is a strong case for introducing independent oversight of how sanctions rules are applied.  Sanctions are, in effect, a form of administrative justice.  However they are subject to no independent oversight or monitoring.  We would support the creation of an Ombudsman or Inspector with the power to examine cases and a clear remit to review the system. 
37. Finally, we would urge the government to formally evaluate the impact of its sanction reforms.  This should be a broad evaluation looking not just at the impacts on the likelihood of receiving benefit, or the impacts on the financial circumstances of those sanctioned, but also the wider economic and social impacts of reform.  We believe that the lack of research and evidence behind the 2012 reforms is deeply concerning, and that the early impacts suggest that they risk causing lasting harm to individuals and families.  They are also, in our view, undermining the important role that Jobcentre Plus must play in supporting the most disadvantaged to prepare for and find work. 

15 December 2014