Britain Paid Restaurants to Discount 160 Million Meals in August 2020 to Get People Eating Indoors Together Again. Nobody Asked the Chief Medical Officer, Who Called It Eat Out to Help Out the Virus.

On 8 July 2020, the Chancellor of the Exchequer, Rishi Sunak, presented his Plan for Jobs to Cabinet and announced it to the public the same day. Among its measures was a scheme with a name built for a slogan. Throughout August, on Mondays, Tuesdays and Wednesdays, anyone eating in at a registered restaurant, pub or café would get half off their food and soft drinks, up to £10 a person, and the business would claim the difference back from HM Revenue & Customs. It ran from 3 to 31 August 2020 and applied across all four nations of the United Kingdom.

The scheme was developed by Mr Sunak and the Treasury from May 2020 and agreed by the Prime Minister, Boris Johnson. Its stated purpose was to rescue a hospitality sector that had spent months shut by the first lockdown. According to the UK Covid-19 Inquiry, which later examined the internal paperwork, it had "a second, equally significant objective": to "incentivise consumers to return to the hospitality industry … by reassuring people that [it] is once again safe to consume and altering social behaviours that have become entrenched." A one-off payment had been considered and rejected, because an incentive lasting a month was more likely to achieve "breaking entrenched behaviours and encouraging the return to social consumption". The entrenched behaviour in question was not going indoors with other households during a pandemic.

The Uptake

The public responded as the Treasury had hoped. HMRC's official statistics, published on 25 November 2020, recorded that more than 160 million meals had been discounted under the scheme. By 30 September, 49,353 businesses had made at least one claim, covering 78,116 outlets, and they had claimed £849 million. The average discount was £5.24 a meal. Restaurants took 54.8 per cent of the money and pubs 27.6 per cent. The number of meals and the value of claims rose every week the scheme ran, as more people took up the offer before it ended.

The Consultation

The people whose job it was to say what indoor dining would do to the virus heard about the scheme when everyone else did. "Eat Out to Help Out we didn't know about until it was announced," Sir Patrick Vallance, the government's chief scientific adviser, told the inquiry on 20 November 2023, "and I think our advice would have been very clear on that." The Scientific Advisory Group for Emergencies was not asked about it. Nor, the inquiry found, were the Health Secretary, Matt Hancock, the Cabinet Office, the Department of Health and Social Care, the Covid-19 Taskforce or the devolved governments, all of which were "deprived of the opportunity to raise any concerns or to influence the scheme."

Mr Johnson initially told the inquiry that the scheme had been "properly discussed, including with Chris [Whitty] and Patrick [Vallance]". He later confirmed that no scientific advisers had been present at the meetings where he and Mr Sunak discussed it. Professor Sir Chris Whitty, the chief medical officer, had by then been warning about indoor hospitality for months. "I do not think Treasury officials would have needed to consult me to know what I would have said," he said in evidence. His private view surfaced in the notebook of Imran Shafi, Mr Johnson's private secretary. On 30 October 2023, counsel to the inquiry, Hugo Keith KC, put it to him that the chief medical officer had described the scheme to him as "Eat out to help out the virus". "That's correct," Mr Shafi said.

Sir Patrick summarised the policy for the inquiry. Up to that point, he said, the message had been that mixing indoors with people from other households was a high-risk activity. "That policy completely reversed it to saying: we will pay you to go into an environment with people from other households and mix in an indoor environment for periods extended over a couple of hours or more." Asked whether, having increased infections, it must also have increased deaths, he replied: "It's highly likely to have done so."

The Clusters

On 30 October 2020, the economist Thiemo Fetzer of the University of Warwick's CAGE research centre published a paper titled "Subsidizing the spread of COVID-19: Evidence from the UK's Eat-Out-to-Help-Out scheme". It found that areas with higher take-up "experienced a sharp increase in the emergence of new COVID-19 infection clusters a week after the scheme began", and estimated that "between 8 and 17% of the newly detected COVID-19 infection clusters can be attributed to the scheme." To separate cause from coincidence, Dr Fetzer used the weather. Areas where it rained around lunch and dinner time on scheme days saw fewer restaurant visits and, subsequently, lower new infection rates than areas that enjoyed good weather. The evidence was not unanimous. Counsel to the inquiry noted an HMRC paper that had found no correlation between take-up and new cases in local authority data.

Mr Sunak, giving evidence on 11 December 2023, maintained that the scheme had operated entirely within the reopening of hospitality that ministers had already agreed. "This was a micro-policy," he said, designed to ensure that capacity already judged safe "was actually used". He noted that a month had passed between announcement and launch, during which the chief medical officer and chief scientific adviser attended three meetings about forthcoming risks without raising it. "The onus is surely on the people who now believed that it was a risk to have raised it at the time," he said. Asked why the scheme had not been extended, he said it had always been designed to be temporary, because a behavioural response requires an incentive to be "credibly temporary".

The inquiry's Module 2 report, published by Baroness Hallett on 20 November 2025, did not decide the epidemiology. "Whether or not the scheme had a significant impact on the transmission of the virus (which Mr Sunak denied)," it concluded, "it presented, at face value, a risk of increased transmission and was devised in the absence of any scientific advice." It also found that the scheme "undermined public health messaging about the importance of social distancing and being cautious in respect of socialising indoors", and "provided financial incentives to engage in such activity."

The Claims

The restaurants, too, responded to incentives. In its annual report for 2021 to 2022, HMRC estimated that £71 million of the money paid out under the scheme had gone in error or fraud, within a range of £43 million to £99 million. That is 8.5 per cent of everything the scheme cost.

The first person HMRC convicted of fraud against any of its Covid schemes was an Eat Out to Help Out claimant. Mohammed Ikram, a former Keighley town councillor, used his own business, Khan's Café and Chai Point, and seven fake food outlets set up with hijacked company details to submit 20 claims worth £434,073 between 7 August and 6 September 2020. HMRC rejected £244,865 and paid £189,208. On 31 March 2023, Bradford Crown Court jailed him for two and a half years. HMRC's accounts record £2.1 million of scheme payments repaid in 2020 to 2021 and a further £0.7 million the year after, when repayments exceeded payments; sums recovered through tax charges are booked separately.

The Treasury designed a scheme to change the way people behaved, priced it at half a meal and made it last long enough to break an entrenched habit. More than 160 million meals later, the inquiry found that something had indeed been undermined. It was the public health messaging.

Filed from public reporting:

• HM Revenue & Customs — "Eat Out to Help Out statistics: commentary"
• UK Covid-19 Inquiry — "Module 2 full report: Core decision-making and political governance"
• UK Covid-19 Inquiry — "Transcript of Module 2 public hearing, 11 December 2023"
• UK Covid-19 Inquiry — "Transcript of Module 2 public hearing, 20 November 2023"
• UK Covid-19 Inquiry — "Transcript of Module 2 public hearing, 30 October 2023"
• University of Warwick, CAGE — "'Eat Out to Help Out' scheme drove new COVID-19 infections up by between 8 and 17%, new research finds"
• HM Revenue & Customs — "HMRC annual report and accounts 2021 to 2022"
• HM Revenue & Customs — "West Yorkshire man jailed for Eat Out to Help Out fraud"