Benefit delivery Contents

2Overpayments and underpayments

The scale of incorrect benefit payments

6.In 2014–15, the DWP paid out £168 billion in benefits to support a wide range of claimants including pensioners, working people, jobseekers and disabled people. The Department has the largest overall budget in government, accounting for 23% of the Government’s £735 billion of Total Managed Expenditure.3

7.Official estimates showed that, in 2014–15, £4.5 billion of the DWP’s total benefit spending was incorrectly paid as a result of fraud or error on the part of the Government or claimants, equivalent to 2.7% of total benefits spending. Of this, £3.0 billion was incorrectly overpaid (1.8%) and £1.5 billion was underpaid (0.9%).4 Of this 0.9%, official error accounted for 0.3% and the remaining 0.6% was due to errors by claimants. Overpayments increase costs to taxpayers and reduce public resources available for other purposes. Underpayments mean households do not receive the support they are entitled to.5 This section of our report examines the impact of such errors on claimants.

A significant and long-standing issue

8.Each year, the UK Parliament sets the total amount that the DWP can spend on benefits. The Comptroller and Auditor General (C&AG) determines whether DWP’s spending is in line with Parliament’s intent, and issues an opinion on DWP’s accounts.6

9.In 2014–15, the C&AG qualified his opinion on the regularity of benefit spending, other than on the state pension, because of the “unacceptably high” level of fraud and error.7 The DWP’s accounts have received similar qualified audit opinions in each year since 1988–89. The C&AG reported that “the fact that such audit qualifications have been in place for such a long period of time does not lessen that seriousness”.8

10.Rt Hon Iain Duncan Smith MP, Secretary of State for Work and Pensions, was frustrated by the ongoing accounts qualification, arguing it was illogical and said that such levels of fraud and error would be tolerated in a commercial context:

If this was outside and it was a company, I do not think this would be happening like this. I suspect what is going on here is that nobody wants to be the first person not to qualify the accounts of this Department […] nobody quite wants to break ranks.9

11.In contrast, Robert Devereux, Permanent Secretary of the DWP, recognised the C&AG’s conclusion, acknowledging that “what Parliament wanted is not what is happening” and saying that he aimed to be the accounting officer to have the qualification lifted.10 He told us that the DWP is working with the National Audit Office (NAO) to understand fraud and error and develop a tailored reduction plan for each major benefit.11

12.The C&AG told the Committee of Public Accounts (PAC) in September 2015 that if he is satisfied that the DWP has the level of fraud and error “down to the lowest level feasible” in a specific benefit, he will consider lifting the qualification.12 The PAC concluded, however, that the Department had so far shown little ambition in tackling fraud and error.13

Error in Housing Benefit

13.The Department spent £23.1 billion on housing benefit (HB) in 2014–15, 14% of total benefit spending.14 HB, which is administered by local authorities, helps people on low incomes pay rent. Eligibility is determined by income, rent level and circumstances. Overpayments due to fraud and error within HB increased from 4.2% to 5.3% of expenditure between 2010–11 and 2014–15.15 Underpayments in HB also rose, from 0.9% of total HB spending in 201011 to 1.4 % in 2014–1516 (Figure 1).

Figure 1: Estimated Housing Benefit over and underpayments due to fraud and error, 2010–11 to 2014–15

Note: All numbers are based on the same method for “netting” underpayments from overpayments as the DWP does for other continuously measured benefits. When a claimant had an underpayment and an overpayment, the DWP used to count both. From 2014–15 onwards, the DWP cancels them out and only reports the difference.

Source: Analysis of DWP data on fraud and error in the benefit system: financial year 2014/15 estimates, November 2015.

14.The Permanent Secretary told us that the principal cause of increases in HB fraud and error was claimants failing to inform the Department of a change in circumstances in a timely manner.17 Mike Driver, Finance Director of the DWP, noted that local authorities could now check their systems against real-time information (RTI) income tax data.18 We were, however, cautioned of the limits of RTI. The system only covers earnings data for those in employment, excluding those in self-employment.19 Furthermore, earnings are only one part of the picture. For example, the PAC concluded that the DWP continues to struggle to tackle overpayments caused by uncertainty over claimants’ living arrangements.20 It recommended that DWP should jointly explore with HM Revenue and Customs (HMRC) how to exploit third party data to identify high risk claims.21

15.Mike Driver told us that the Department planned to make better use of data and was working with the NAO to improve data-matching.22 For example, DWP is beginning to access data from a “wide range of public and private sector organisations”,23 to help address fraud and error from undeclared partners, income and capital.

16.Fraud and error in housing benefit have increased in the past five years, despite DWP’s use of real-time information on claimant incomes. We recommend the Government set out in detail how it will exploit other sources of data to gain a better picture of claimant circumstances, including living arrangements, and what reductions in fraud and error it expects these to achieve. We expect the DWP’s 2015–16 annual report and accounts to show progress in this area.

More focus on underpayments

17.In 2010, DWP set a target to reduce the level of benefit overpayments24 to 1.7% of benefit expenditure by 2014–15, an aim it narrowly missed.25 It did not, however, set a target for reducing underpayments.26 The level of underpayments has stayed largely flat over time, compared with the level of overpayments (Figure 2).

Figure 2: Estimated over and underpayments due to fraud and error, 2005–06 to 2014–15

Source: Analysis of DWP data on fraud and error in the benefit system: financial year 2014/15 estimates, November 2015.

18.The Permanent Secretary told us that many of the measures the DWP was taking to improve efficiency were “almost blind to whether it is an overpayment or an underpayment”27 and should result in reductions in both. These measures include the automation of benefit payment processing and sharing best practice to ensure that staff “do things once rather than do them multiple times”.28 Lord Freud, the Minister for Welfare Reform, said that by using RTI income tax data, UC would act to reduce underpayment levels.29

19.Reducing underpayments was also a matter of resources. Lord Freud told us “at one level, this is about costs […] if you spend a lot of money fixing it, you can”.30 The Permanent Secretary stated that the amount of money available to address underpayments would be assessed in the light of the Spending Review.31

20.Advice and support organisations told us that benefit underpayments left individuals vulnerable.32 Shelter told us that “benefit problems often led claimants to face difficult decisions over whether to pay their rent or provide essentials such as food, gas and electricity for their household”.33 Citizens Advice reported that errors can push those less able to support themselves towards expensive borrowing and had “clear consequences for physical and mental health, with stress and anxiety creating or exacerbating existing health conditions”.34 Others found that individuals can become reliant on food banks as a result of underpaid benefits.35

Box 1: Illustrative case example of the human impact of a benefit underpayment.

A claimant faced significant financial hardship from underpayment of their Jobseekers Allowance.

Nathan was a Jobseekers Allowance (JSA) claimant who came to Citizens Advice in Watford. He had previously secured part-time seasonal work, so had been on a reduced rate of JSA.

When this ended, he informed his local Jobcentre Plus of his change in circumstances. However, nearly two months later no changes to his benefit had been made.

Citizens Advice advisers helped him contact the Jobcentre, who informed him that his JSA would be backdated once approved.

Yet this meant that Nathan was struggling to survive on a very low income; he had already fallen into rent arrears. He was trying to reduce all other outgoings, but had been told he could not cancel his mobile phone contract without a penalty.

He now risked falling into more serious debt.

Source: Citizens Advice written evidence submission.36

21.We also heard that benefit underpayments transfer the financial strain to other areas of public spending. We were told that benefit errors were having “knock-on effects” which increased demand for health, social care, children’s and housing services as well as “creating a huge demand for advice services, which are generally funded by the council”.37 Milton Keynes Council drew our attention to benefit errors adding to demand for discretionary welfare schemes which are provided by local authorities.38 These schemes are the subject of another of our inquiries.39 Durham University said that when underpayments occur, claimants go “without sufficient food and/or energy required to maintain health of themselves and their family”.40

22.Lord Freud told us that UC would “dramatically” improve the DWP’s ability to reduce the incidence of underpayments.41 Both he and the Secretary of State invited the Committee to suggest a target for a reduction in the rate.42

23.The DWP has a target for reducing benefit overpayments but not for reducing benefit underpayments. This indicates their relative priorities as the Department seeks to reduce costs. Underpayments, however, can have an enormous effect on individuals and can be a false economy as they transfer the burden to other services. Real time earnings information in Universal Credit is expected to improve the accuracy of benefit payments and should result in a reduction in underpayments over the next few years. We welcome the Department’s invitation to hold them to that expected ‘dividend’ from Universal Credit.

24.Underpayments totalled £1.5 billion, or 0.9% of benefit expenditure, in 2014–15. We recommend the DWP introduce a target to reduce underpayments to less than 0.5% of expenditure by the end of this Parliament. The Cabinet Office’s Fraud, Error and Debt Steering Group should monitor the Department’s progress towards achieving the target.

7 See reference 5, p. 91 and p. 95.

8 See reference 5, p. 95.

9 Oral evidence HC 507, Q16.

10 Oral evidence HC 507, Q17 and Q20

11 Oral evidence HC 507, Q17 and Q20

15 DWP, Fraud and error in the benefit system: financial year 2014/15 estimates, November 2015, table 13. Consistent series using the updated methodology used in 2014-15.

16 DWP, Fraud and error in the benefit system: financial year 2014/15 estimates, November 2015, table 14. Consistent series using the updated methodology used in 2014-15.

17 Oral evidence HC 507, Q28.

18 Oral evidence HC 507, Q28.

19 Work and Pensions Committee, Fraud and error in the benefit system, Session 2013-14, HC 1082, 7 May 2014, para 48. In their additional written evidence, DWP told us that they have a “variety of initiatives underway to combat non-PAYE fraud and error”.

22 Oral evidence HC 507, Q29.

24 The calculation for underpayments only includes claimants that are getting less than they are entitled to and does not include those people who are entitled to benefits but who do not apply, or whose applications are not awarded when they should be.

27 Oral evidence HC 507, Q37

28 Oral evidence HC 507, Q12, Q37.

30 Q192.

31 Oral evidence HC 507, Q37.

32 See, for example, Zacchaeus 2000 Trust (BFD0148).

33 Shelter (BFD0163)

34 Citizens Advice (BFD0173).

35 See, for example, St Mungo’s Broadway (BFD0171); Durham University (BFD0121).

36 Citizens Advice (BFD0173).

37 Lambeth Council (BFD184);

38 Milton Keynes Council (BFD0160)

39 Work and Pensions Committee, Local welfare safety net inquiry

40 Durham University (BFD0121)

41 Q220.

42 Oral evidence HC 507, Q42; Q192

© Parliamentary copyright 2015

Prepared 18 December 2015