2Evidence on the effectiveness of sanctions
10.The DWP says that the aim of conditionality and sanctions is:
to motivate claimants to engage with support on offer to look actively for work and thereby to move into work. It also ensures the system is fair to the taxpayer by reducing the standard rate of benefit from claimants when they fail, without good reason, to meet a conditionality requirement that they have agreed.13
But evidence was mixed on whether sanctions were achieving this. On the one hand, the Department said “evidence shows that sanctions have a positive impact on behaviour”.14 It cited its own research, which found that over 70% of JSA and 60% of ESA recipients said the threat of sanctions made them more likely to comply with their work-related requirements.15 Further research showed that mandatory checks on whether someone was complying with their work-related requirements were “effective at decreasing time on benefits”.16 The National Audit Office (NAO) reported that JSA claimants “spent less time claiming after getting a sanction”.17
11.On the other hand, several studies contradicted, or qualified, these findings. First, UK and international studies showed that while receiving a sanction could increase the likelihood of moving into employment, it was often into low-quality jobs, in terms of pay, conditions and sustainability.18 Moreover, a five-year project (2013–2018) funded by the Economic and Social Research Council, Welfare conditionality: sanctions support and behaviour change (hereafter the ‘Welfare Conditionality Project’), found that “stasis—a lack of significant, sustained change in employment status—was the most common outcome” for sanctioned claimants participating in its study.19
12.Second, the NAO found that while JSA claimants spent less time claiming benefits after getting a sanction, they were just as likely to stop claiming without finding work, as they were to find work.20 Some evidence also suggested that people who came off benefits following a sanction did not necessarily move into work, but might turn to other forms of support, such as family members.21
13.Third, several witnesses told us that the cause of positive behaviour changes was unclear, as it was difficult to untangle the effects of support offered by the jobcentre, the obligation to comply with certain conditions, and the sanctions themselves. Tony Wilson, Director of Policy and Research at the Learning and Work Institute, an independent policy and research organisation, argued that the extent to which the threat of a sanction affected behaviour was “a justification for the conditionality regime”, rather than the sanction itself.22 Furthermore, the Welfare Conditionality Project found that jobcentre’s focus on claimants fulfilling their mandatory conditions, plus claimants’ fear of sanctions, led to “counterproductive compliance”. It explained:
Pressure to achieve more demanding job application/work search requirements (up to 35 hours per week) coupled with benefit recipients’ strong desire to avoid the punitive effects of a sanction resulted in people applying for jobs they had no realistic chance of getting. The threat of a benefit sanction therefore encouraged a culture of counterproductive compliance and futile behaviour that got in the way of more effective attempts to secure employment.23
The 2012 reforms
14.The Welfare Reform Act 2012 (‘the 2012 Act’) is the foundation of today’s conditionality and sanctions regime. It established the rules for UC and amended those for legacy benefits so that they were broadly aligned. In doing so, it increased the length and severity of sanctions and made them applicable to more claimants than ever before.24 The Department’s Impact Assessment said the changes would address problems of inconsistency, lack of clarity and “insufficiently tough” sanctions for those who “repeatedly fail to meet their most important responsibilities”.25
15.Several witnesses argued, however, that mixed evidence on the effectiveness of sanctions did not support the introduction of a tougher regime, including longer sanctions following a similar ‘failure’ within the previous 12 months.2627 At the time, the DWP itself acknowledged that the 2012 changes would “elicit behavioural responses which are difficult to predict with certainty”.28 But it said the policy would be “reviewed from 2013 on an ongoing basis” to “establish its impact and the extent to which the policy objectives have been achieved”.29 Several witnesses to our inquiry criticised the Department for failing to conduct such an evaluation.30 Tony Wilson described it as “a real dereliction of the Department”31 and told us:
This is the only major welfare reform introduced after 2010 that has not been evaluated. There has been no attempt to evaluate the impact of sanctions. There has been no attempt to understand what the impacts are on individuals.32
Longer sanctions
16.Furthermore, Dr David Webster, an expert academic in this field, noted that the “DWP does not publish any information on how many sanctions there are of each duration” nor “record the date of the ‘failure’ for which the claimant is sanctioned”.33 As a result, the Department could not know how many claimants received longer sanctions for second or successive ‘failures’. He described this as “an astonishing gap in the statistics” as it meant the impact of escalated sanctions—a major reform—could not be evaluated.34
17.Alok Sharma MP, Minister of State for Employment (‘the Minister’), confirmed that the Department “does not hold the start and end date for a sanction and thus [does] not hold the length of the sanction either”.35 He said that it would be possible to infer the length of sanctions from payment data and the dates of failures, but “this would take considerable time to extract and analyse”.36 He was also unable to point to any specific evaluation of the more severe sanctions regime introduced in 2012 when giving oral evidence.37 We asked again, in correspondence, “what further evaluation has been done of the regime introduced by the 2012 Act, in particular regarding the impact and effectiveness of more severe sanctions … ?”. He replied:
The Department is building its understanding of how sanctions help to underpin our conditionality regime, in particular how the impact of Universal Credit has changed our engagement with claimants, the effect of sanctions on customer behaviour and how this varies for different customer groups, as well as how we can make the sanctions regime more effective. This work is on-going, especially as we analyse the new data available on Universal Credit systems. As part of this work, we will explore the use of analytical techniques that attempt to isolate the impact of sanctions on transitions into work and of earnings when in work.38
He acknowledged that “this will not provide evidence of the effectiveness of the 2012 Act, compared to the previous system”, but “may provide insight into the effectiveness of the current sanctions system in supporting conditionality”.39 We also asked how the Department expected to evaluate the increased sanction lengths without knowing how many of each length had been imposed. To this, the Minister said:
There is existing academic literature that shows the impact of different levels of sanctions on work search behaviour. We are considering how we might apply some of this to our UC data, to look at sanction durations, but it is complicated as sanction duration is currently identified on our analytical systems using drops in payment amounts, rather than purely identifying individual sanctions at different durations.40
He cited two studies when referring to “existing academic literature”. The first analysed the impact of ‘mild’ and ‘strong’ sanctions applied to young benefit claimants in Germany.41 It found that “both types of sanctions lead to a higher transition rate to work, and that this effect is higher for strong sanctions”.42 The second analysed the effect of sanctions on the exit rate from unemployment in Denmark.43 It concluded that “for both men and women … more severe sanctions have a larger impact on the exit rate from unemployment”.44 It is important, however, to note the length of sanction being considered: In the German study, a ‘strong’ sanction withdrew benefits for up to three months, and in the Danish study, the “more severe” sanction lasted just three weeks. It is therefore a huge leap to see these findings as evidence for the more severe sanctions regime introduced in 2012, under which a higher-level sanction can last up to three years.
Sanction rates under Universal Credit
18.A recent House of Commons Library publication explained that sanction rates “are complicated to assess due to the range of experimental data series available from DWP and the different sanction policies covering each sanctionable benefit”.45 They could be looked at in two ways:
i)‘Prevalence’—a measure of how many people are currently under a sanction, and
ii)‘Incidence’—a measure of how many sanctions are newly issued in a given period.
‘Prevalence’ therefore provides a better idea of how many people are affected by sanctions at any one time, while ‘incidence’ provides a better idea of the actual volume of sanction decisions being made.46 The figure below shows that sanctions are more prevalent under UC than JSA.
19.Dr David Webster described the rate of sanctions under UC as “strikingly high” at around 5% per month—a rate reached under JSA only for relatively short periods in 2010—11 and 2012—14.47 The difference is even more pronounced when looking solely at UC claimants who are required to search for work (a more direct comparison to JSA claimants). In December 2017, around 0.3% of JSA claimants were experiencing a sanction, compared to 8.2% of those UC claimants (this fell to 5.3% in May 2018, but equivalent JSA data was not available at the time of writing this report).48 In addition, data show that the incidence of adverse sanction decisions (i.e. the number of sanctions imposed over a period of time) “has been consistently higher under the UC live service than under JSA” (data on UC sanction decisions was only available for the live service at the time of writing).49 In April 2018, around 7,000 individuals were newly sanctioned under the UC live service, compared to around 2,300 under JSA. This was equivalent to around 5% of all UC live service claimants and around 1% of all JSA claimants.50
20.In written evidence, the DWP said sanction rates for UC appeared higher because actions that resulted in closing a claim under JSA led to a sanction under UC: for example, failing to attend a work coach meeting and not making contact within five days, or “not actively seeking work”.51 Cases could not be closed under UC as claims might include other elements that were not affected by a sanction and needed to continue, such as those for children or housing.52
21.Dr David Webster said this explanation was “rather puzzling” and did not explain fully the higher rate of sanctions under Universal Credit. So we asked the House of Commons Library to test the Department’s hypothesis. To do this, it removed interview-related sanctions from the overall sanction rates for UC live service and JSA (noting that this discounted a considerable proportion of total sanctions). The figure below shows the result. Although the difference between the two rates is smaller, the rate of sanctions under UC was still higher than under JSA in recent months.
22.The House of Commons Library concluded that this result “might seem to affirm DWP’s suggestion that the higher rate of sanctioning under UC is due to the practice of sanctioning claimants when they miss an interview rather than simply stopping their benefit payments entirely”.53 It said, however, that it was still not possible to know why the rate of interview-related sanctions was so high under UC, or what the rate of sanctions would be under JSA if cases were not closed for missed interviews. In addition, the effect of the higher rate of sanctioning on a UC claimant’s likelihood of finding work was still unknown. In a letter dated 23 August 2018, the Minister said:
The Department is yet to undertake robust analysis into the extent of the effects on sanction rates resulting from the differences in policy between UC and legacy benefits. Such analysis will be scheduled into our work plan for the next 12 months.54
23.At best, evidence on the effectiveness of sanctions is mixed, and at worst, it shows them to be counterproductive. The Coalition Government had little or no understanding of the likely impact of a tougher sanctions regime when it was introduced in 2012. It said the policy would be reviewed on an ongoing basis to understand its impact and the extent to which it was achieving its objectives. But six years later, it is none the wiser. The lack of any such evaluation is unacceptable. Furthermore, without evidence to support the significantly longer sanctions introduced, or data to understand the behavioural impact of escalated sanctions for repeated failures, the policy appears to be nothing other than arbitrarily punitive. The high rate of sanctions under Universal Credit only increases the urgency with which the Government must understand fully the effect of the 2012 reforms. Crucially, if such research suggests changes are required, the Government must be prepared to respond accordingly.
24.We recommend that the Department urgently evaluate the effectiveness of reforms to welfare conditionality and sanctions introduced since 2012 in achieving their stated policy aims. The Department should commission an independent review of its methodology for this work. We further recommend that higher level sanctions should be reduced to two, four and six months for first, second and subsequent failures to comply, until the Department can present robust evidence that longer sanctions would be more effective at moving people into work.
The impact of sanctions
25.The Joseph Rowntree Foundation acknowledged that “sanctions by nature have to threaten some level of hardship. But this should not result in destitution”.55 We heard evidence, however, that sanctions could have a profound and long-lasting financial impact. For example, frontline workers at Citizens Advice reported that their sanctioned clients “more commonly borrowed money, cut back spending on food and other essentials, or fell into arrears with bills” rather than increased their income through finding work.56 Samantha, a single parent claiming UC while working part-time, told us how she fell into debt following a sanction. She said:
After three months less £500 … how do you catch up? … you are still getting yourself to work, but the ends are still not being met … when the sanction [ends], you’ve still got that backlog. You have still got bills that are outstanding and you are still being chased.57
26.In addition to borrowing money and seeking help from family members, Samantha turned to food banks.58 We heard evidence of other people in the same situation, suggesting that Samantha’s experience was far from unique.59 Furthermore, Dr Rachael Loopstra’s quantitative study of the relationship between sanctions and the use of food banks found that “in each case … .at the area level, as more Jobseeker’s Allowance claimants received sanctions over 2012 to 2015, food bank usage rose”.60 She said the results of her study likely “understate the true magnitude of the relationship between sanctions and food insecurity”.61 This was because she used Trussell Trust food bank usage as a proxy for food insecurity, which did not capture everyone experiencing hunger.
Box 1: Comments from respondents to online survey question: How did the sanction affect you?
|
“I had to rely on food banks and the kindness of family to get through Christmas. It made me feel like a failure and caused my children stress”—Female aged 31–45 claiming UC “It caused anxiety and anger… I fell into debt with bills and at times was unable to top up my prepayment electric meter so was without heating or lighting”—Male aged 31–45 claiming UC “I got into severe debt that I am still paying off … I borrowed from loan sharks [and] have been served with an eviction notice from the council”—Female aged 45+ claiming ESA |
Source: Between 4 May and 5 June 2018 the Work and Pensions Committee invited views on benefit sanctions via an online survey on its webpage.
27.As well as affecting financial well-being, we heard how sanctions could have a deeply negative impact on people’s physical and mental health. Child Poverty Action Group (CPAG), an organisation that develops and campaigns for policy solutions to end child poverty, found that sanctions “tend to reduce people’s capabilities”, including their physical health, mental health, financial security and coping abilities.62 This could, in turn, increase the burden on the National Health Service. In 2017, Mind, the mental health charity, surveyed 3,000 people with mental health problems and experience of the benefits system. Nine in ten of those who were sanctioned, or threatened with a sanction, said it had led to a deterioration of their mental health.63
Box 2: Comments from respondents to online survey question: How did the sanction affect you?
|
“ I was already depressed, this worsened it, seriously knocking my confidence. I self-harmed”—Male aged 31–45 previously claiming JSA “I was reduced to 1 meal a day for several weeks and lost a stone and a half. Prior to this I was already considered dangerously underweight”—Male aged 31–45 previously claiming JSA “It made my depression worse and I am too scared/anxious to enter a jobcentre” Male previously claiming JSA “It sent my mental and physical health spiralling out of control. I have severe anxiety at the mention of the DWP or sight of a brown envelope”—Female aged 45+ claiming ESA |
Source: Between 4 May and 5 June 2018 the Work and Pensions Committee invited views on benefit sanctions via an online survey on its webpage.
28.Several witnesses, including the Shaw Trust, a national charity helping people into work, Centrepoint, a charity providing housing and support for young people, and CPAG, also highlighted that claimants’ negative experiences of sanctions often eroded their relationship with the jobcentre.64 Others, such as Inclusion London, a London-wide organisation promoting equality for London’s Deaf and Disabled people, told us how “extreme poverty is a barrier to looking for work” as people are forced to prioritise basic needs, like:
using a foodbank, trying to sort out your housing benefit with your local council, trying to sleep for longer so you don’t feel the cold and hunger so badly, looking up friends who may give you a hand-out (who you can’t telephone because you have no credit) as well as appealing the sanction.65
Box 3: Comment from respondents to online survey question: How has being sanctioned changed how you think about your benefits?
|
“I’ve lost all faith and trust in the staff and system”—Male aged 31–45 claiming UC “I think there’s no point even engaging with Jobcentre Plus”—Male aged 18–24 claiming UC “I am now very nervous of JCP staff”—Male aged 45+ claiming JSA |
Source: Between 4 May and 5 June 2018 the Work and Pensions Committee invited views on benefit sanctions via an online survey on its webpage.
The wider implications of sanctions
29.The NAO highlighted that supporting people affected in the ways described above can “lead to extra public spending in areas such as local authority funding welfare support”.66 It noted that while the DWP had estimated costs and benefits for changes to lone parent conditions, “it has not assessed costs and benefits for sanctions as a whole”.67 The Public Accounts Committee said “the Department does not understand the wide effects of sanctions” and recommended two years ago that the Department:
work with the rest of government to estimate the impacts of sanctions on claimants and their wider costs to government and report back to us on progress at the end of 2017.68
In evidence to our inquiry, the DWP said it had explored the possibility of conducting research into the impact of sanctions, including on wider public services, but concluded there were “significant practical issues which make it difficult to quantify potential impacts”.69
30.It is one thing for a sanction to result in short-term hardship as a consequence of breaching an agreed work-related requirement. It is something else entirely for a sanction to affect someone’s physical and mental well-being, drive them into debt and leave them on the brink of destitution. We agree that research to understand the impact of sanctions would be complicated, but we do not agree that this is a reason for the DWP not to try. There is too much at stake not to.
31.We recommend that the Department include in the evaluation we have recommended an assessment—to whatever extent is feasible—of the impact sanctions have on claimants’ financial and personal well-being, as well as on wider public services. It should take expert advice on how to achieve this and consider commissioning external research if necessary.
13 Department for Work and Pensions (ANC 0083)
14 Department for Work and Pensions (ANC 0083)
15 Department for Work and Pensions, The Jobcentre Plus Offer: Final evaluation report, November 2013
16 Department for Work and Pensions (ANC 0083)
17 National Audit Office, Benefit Sanctions, November 2016
18 MRC/CSO Social and Public Sciences Unit University of Glasgow (ANC 0061), Child Poverty Action Group (ANC 0056), Q13, see also Beth Watts, Suzanne Fitzpatrick, Glen Bramley et al, “Welfare sanctions and conditionality in the UK”, September 2014
19 Welfare Conditionality Project (ANC 0079)
20 National Audit Office, Benefit sanctions, November 2016
21 Dr Rachel Loopstra (ANC 0078), Citizens Advice (ANC 0067)
22 Q13
23 Welfare Conditionality Project (ANC 0079)
24 National Audit Office, Benefit sanctions, November 2016
25 Department for Work and Pensions, Impact Assessment for conditionality measures in the 2011 Welfare Reform Bill, October 2011
26 This is measured as the time between the dates of the ‘failures’, not the sanction decisions.
28 Department for Work and Pensions, Impact Assessment for conditionality measures in the 2011 Welfare Reform Bill, October 2011
29 Department for Work and Pensions, Impact Assessment for conditionality measures in the 2011 Welfare Reform Bill, October 2011
30 See for example Dr David Webster (ANC 0019), Q13 [Tony Wilson], DePaul (ANC 0070), Baptist Union of Great Britain, the Church of Scotland, the Methodist Church, the United Reformed Church and the Church Action on Poverty (ANC 0043), MRC/CSO Social and Public Sciences Unit University of Glasgow (ANC 0062), British Psychological Society (ANC 0061)
31 Q13
32 Q13
33 Dr David Webster (ANC 0019)
34 Dr David Webster (ANC 0019)
35 Letter to the Chair from Alok Sharma MP, Minister of State for Employment, 4 September 2018
36 Letter to the Chair from Alok Sharma MP, Minister of State for Employment, 4 September 2018
37 Qq245–8
38 Letter to the Chair from Alok Sharma MP, Minister of State for Employment, 4 September 2018
39 Letter to the Chair from Alok Sharma MP, Minister of State for Employment, 4 September 2018
40 Letter to the Chair from Alok Sharma MP, Minister of State for Employment, 4 September 2018
41 Gerard J. van den Berg, Arne Uhlendorff and Joachim Wolff, “Sanctions for young welfare recipients”, IZA Discussion Paper No. 7630, September 2013
42 Gerard J. van den Berg, Arne Uhlendorff and Joachim Wolff, “Sanctions for young welfare recipients”, IZA Discussion Paper No. 7630, September 2013
43 Michael Svarer, “The effect of sanctions on the exit rate from unemployment: Evidence from Denmark”, Economica, September 2011
44 This finding is itself caveated as being “explorative analysis” due to data limitations
45 House of Commons Library, Why are sanction rates higher under Universal Credit?, accessed 23 October 2018
46 House of Commons Library, Why are sanction rates higher under Universal Credit?, accessed 23 October 2018
47 Dr David Webster, Benefit sanctions statistics briefing July 2018, accessed 23 October 2018
48 House of Commons Library, Why are sanction rates higher under Universal Credit?, accessed 23 October 2018
49 House of Commons Library, Why are sanction rates higher under Universal Credit?, accessed 23 October 2018
50 House of Commons Library, Why are sanction rates higher under Universal Credit?, accessed 23 October 2018
51 Department for Work and Pensions (ANC 0083)
52 Department for Work and Pensions (ANC 0083)
53 House of Commons Library, Why are sanction rates higher under Universal Credit?, accessed 23 October 2018
55 Joseph Rowntree Foundation (ANC 0068)
56 Citizen’s Advice (ANC 0067)
57 Qq125–129
58 Q129
59 See for example Joseph Rowntree Foundation (ANC 0068), Gingerbread (ANC 0080), Shaw Trust (ANC 0086)
60 Dr Rachael Loopstra (ANC 0078)
61 Dr Rachael Loopstra (ANC 0078)
62 Child Poverty Action Group (ANC 0056)
63 Mind, “People with mental health problems made more unwell by benefits system”, accessed 25 October 2018
64 Shaw Trust (ANC 0086), Gingerbread (ANC 0080), Centrepoint (ANC 0060), Child Poverty Action Group (AND 0056), Welfare Conditionality Project (ANC 0079)
65 Inclusion London (ANC 0026)
66 National Audit Office, Benefit sanctions, November 2016
67 National Audit Office, Benefit sanctions, November 2016
68 Committee of Public Accounts, Forty-second Report of Session 2016–17, Benefit sanctions, HC 775
69 Department for Work and Pensions (ANC 0083)
Published: 06 November 2018