Recommendation 12

We asked HM Treasury what its current assessment of fraud and error across government was, given the availability of the new data. HM Treasury told us that its upfront estimates on losses due to fraud and error were broadly in line with HMRC’s most 25 Committee of Public Accounts, Fraud and Error, Ninth Report of Session 2021–22, HC253, 24 June 2021 26 HM Revenue & Customs, Annual Report & Accounts 2020–21, 4 November 2021 HMRC Annual Report and Accounts 2020 to 2021 (Web) (publishing.service.gov.uk); Department for Work & Pensions, Annual Report & Accounts 2020–21, 15 July 2021 HC 422 – Department for Work and Pensions – Annual Report and Accounts 2020–21 (publishing.service.gov.uk); Department for Business, Energy & Industrial Strategy, Annual Report & Accounts 2020–21, 25 November 2021 BEIS Annual report and accounts 2020–21 (publishing.service.gov.uk) COVID-19 cost tracker update 13 likely scenario in its annual report and accounts for three of its measures: the Coronavirus Job Retention Scheme (CJRS), the Self-Employment Income Support Scheme (SEISS) and Eat Out to Help Out (EOHO). It explained that this had helped HM Treasury validate its methodology for estimating fraud. For example, it had initially estimated that fraud within CJRS and EOHO would be between 5% and 10%. It told us that the most likely estimated of fraud within the schemes were 8.7% and 8.5% respectively. It similarly told us that this had allowed it to target funding in a more effective way when investing in counter-fraud activities.27