‘Eat Out to Help Out’ scheme drove new COVID-19 infections up by between 8 and 17%, new research finds.

The government initiative, which cost around £500 million, caused a significant rise in new infections in August and early September accelerating the pandemic into its current second wave. The economic benefits of the scheme, meanwhile, were short-lived.

In a new paper, ‘Subsidizing the spread of COVID-19: Evidence from the UK's Eat Out to Help Out scheme’, Dr Thiemo Fetzer of the CAGE Research Centre in the Economics Department at the University of Warwick analyses the causal impact of ‘Eat Out to Help Out’ on COVID-19 infections. Key findings are:

· Participating restaurants saw an increase in visits of between 10 and 200% compared to 2019.

· Areas with a higher rate of uptake (both from restaurants and consumers) experienced a sharp increase in the emergence of new COVID-19 infection clusters a week after the scheme began.

· Between 8 and 17% of the newly detected COVID-19 infection clusters can be attributed to the scheme.

· Areas with high uptake saw a decline in new infections a week after the scheme ended.

· As the ‘Eat Out to Help Out’ scheme ended, visits to restaurants started to decline – indicating that its positive economic impact was short-lived.

The research leverages data from HMRC’s own restaurant finder app which was the go-to platform for people searching for participating restaurants in their neighbourhood, together with weekly data on new COVID-19 infections measured at the level of Middle Layer Super Output Areas.

To demonstrate the causal connection between ‘Eat Out to Help Out’ and increased infections in areas of high uptake, Dr Fetzer analyses rainfall data and granular mobility data from Google’s Community Mobility Reports. He finds that higher rainfall around lunch and dinner time during the scheme’s period of operation (Monday to Wednesday throughout August) saw both a drop in visits to restaurants and subsequently lower new infection rates compared to areas that enjoyed good weather. Rainfall during lunch and dinner hours did not drastically affect time spent in other locations.

Dr Fetzer said ‘This strongly suggests that the link between ‘Eat Out to Help Out’ and new COVID-19 infections is causal: when people were not dining out as part of the scheme there were fewer new cases of the virus. ‘Eat Out to Help Out’ may in the end have been a false economy: one that subsidised the spread of the pandemic into Autumn and contributed to the start of the second wave. Alternative policy measures, such as extending the furlough scheme, increasing statutory sick pay and supporting low income households through expanding free school meals may well prove to be far more cost effective than demand-stimulating measures that encourage economic activities which actively cause COVID-19 to spread.’

Notes to editors

The Government’s Eat Out to Help Out scheme was conceived to shore up demand for the hard hit restaurant sector in the UK, which employs around 1 million people directly. The scheme cut the cost of meals and non-alcoholic drinks by up to 50% across tens of thousands of participating restaurants in the UK from 3 to 31 August 2020. The discount was capped at a maximum of £10 per person, but there was no limit on how often it could be claimed per individual. Early release statistics suggest that during the four weeks in which the program was active, a total of nearly 100 million covers were claimed, at a total cost to the taxpayer of around £500 million.

Read the Research

Subsidizing the spread of COVID-19: Evidence from the UK’s Eat-Out-to-Help-Out scheme. Thiemo Fetzer, CAGE Working Paper, University of Warwick, https://warwick.ac.uk/fac/soc/economics/research/centres/cage/manage/publications/wp.517.2020.pdfLink opens in a new window