Fraud and error in the benefit system: financial year 2019 to 2020 estimates
Estimates of fraud and error levels in the benefit system in Great Britain in the financial year 2019 to 2020.
Applies to England, Scotland and Wales
Documents
Details
This report provides estimates of fraud and error levels in the benefit system in Great Britain for the financial year 2019 to 2020.
The main points from the report are:
- 2.4% of total benefit expenditure was overpaid due to fraud and error
- the estimated value of overpayments was £4.6 billion
- 1.1% of total benefit expenditure (or £2.0 billion) was underpaid due to fraud and error
- the net government loss, after recoveries, was £3.6 billion, or 1.9% of benefit expenditure
Send any comments or questions about the fraud and error estimates to: [email protected].
Revisions made to the estimates of fraud and error in the benefit system 2019 to 2020
On 14 May 2020 the fraud and error estimates for 2019 to 2020 were first published.
On 20 May 2020 we temporarily removed Table 10 from the tables, due to errors in the data.
On 29 May 2020 we made revisions to the publication and data, and republished the following:
1. a revised main statistical release with a small change to text on page 3, clarifying the impact of change to Universal Credit expenditure and overpayment rates on the overall overpayment rate.
The text now reads:
“The total rate of overpayments has risen due to the increase in expenditure on Universal Credit (which has more than doubled between 2018/19 and 2019/20) and because Universal Credit has a higher rate of overpayments than any of the other benefits we measure. The increase in the overpayment rate for Universal Credit from 2018/19 has a marginal impact; if the overpayment rate for Universal Credit had remained at 8.7%, the total rate of overpayments for 2019/20 would have risen to 2.3%.”
2. a revised background and methodology document, including additional information on the percentage of expenditure reviewed in 2019 to 2020 (incorporating only the Housing Benefit client groups that have been reviewed in 2019 to 2020) in section 3, and additional information on omissions from the estimates in section 5.
3. a number of changes to the published tables have also been made.
Changes made to the published tables
After identifying a number of errors in the tables, these have now been corrected. The headline estimates of fraud and error overpayments and underpayments have not been affected.
A number of errors related to a cell incorrectly displaying an ‘x’ (denoting no data), a ‘z’ (denoting not applicable) or a zero (where data was present, but the rounded rate or monetary value of overpayments or underpayments was zero), when it should have displayed a different value.
Figures within the tables which were greater than zero (and should have been displayed) have now been corrected, and are published in the table of revisions made to the 2019 to 2020 estimates, attached to this page.
Other changes made include updating the *s to denote significance for 2019 to 2020 (where necessary in light of the changes to the data), and removing *s from the revised 2018 to 2019 data.
Interactive information and charts
Use DWP’s online tool to explore the fraud and error statistics.
Updates to this page
Published 14 May 2020Last updated 29 May 2020 + show all updates
-
Revisions have been made to the statistical publication and data tables - full information about the revisions is published on the release page.
-
Published a revised version of the data tables. Table 10 has been temporarily removed from the tables, due to errors in the data. These are being corrected, and a revised version of table 10 will be published when this work is complete. We believe this will be by the end of May 2020. The other tables have not been changed.
-
First published.