select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to preprint by an economist looking at Eat Out to Help Out scheme and COVID-19 spread

A preprint, an unpublished non-peer reviewed paper, by an economist at the University of Warwick looks at the Eat Out to Help Out scheme and COVID-19 spread.


Dr Simon Clarke, Associate Professor of Cellular Microbiology at the University of Reading, said:

“This is report is further evidence of the push and pull relationship between the economy and controlling the virus.  Eat out to help out was successful in getting people out into restaurants and money in the tills, but a contribution to the huge spike in infections among young people appears to be the cost.  The boost the scheme gave to the hospitality sector seems unquestionable, but it will be more than wiped out if we are forced to close everything again and follow other nearby countries in announcing a second full lockdown.

“While the conclusions in this paper appear sound, the authors concede that their findings are based on “a back of the envelope calculation”.  We currently live in a world that is already saturated with supposed evidence which has not been peer-reviewed and where theories are often presented as facts.  While this is often well-intended, it may be flawed or false.  We should remember that studies that have gone through a rigorous scientific process are the best sources to inform decisions and commentary at this time.”


Dr Flavio Toxvaerd, University Lecturer at the Faculty of Economics, University of Cambridge, who specialises in the economics of infectious diseases and economic epidemiology, said:

“While this paper is not yet peer reviewed, its results appear to confirm what many in the economics profession have long thought and expressed publicly; the Eat Out to Help Out scheme was an ill-conceived policy that helped spread the disease at taxpayers’ expense.

“The scheme was introduced to support the hospitality sector, an important source of employment that had suffered greatly because of the Spring lockdown.  But while supporting this sector may have had merit, this could have been achieved via direct transfers from government.  This would have supported the sector while helping stem the spread of the disease.

“This back-firing policy initiative shows the urgent need to better coordinate economic policy with the effort to control the disease.  Until a vaccine or effective treatments become available, we will need to rely on social distancing and similar measures to protect the health, economic and social well-being of the population.  This means that economic policy is health policy and both must be formulated together.”



Prof Paul Hunter, Professor in Medicine, UEA, said:

“I can certainly follow the argument that the impact of the eat out to help out was to increase restaurant visits and concentrate them onto those days when the scheme was operational.  I accept that this would likely have had the unintended consequence of meaning that on certain days restaurants were more crowded then otherwise and such crowding may well have increased transmission of COVID-19.  I think the author has shown the impact of eat out to help out on restaurant visits quite well.

“I am not convinced however by how the author linked this to actual cases.  From what I gather the main measure was a cluster which was defined as 3 or more cases within a week in a local authority area.  This is a binary outcome measure (i.e. Yes/No) and so would not distinguish between authorities with many cases that were declining and ones with a few cases that were increasing.  In my view this approach is fraught with difficulties.  The problem is that in areas with more population with have more restaurants and also other things being equal more cases and so more likely to be classed as a cluster.  Even though the author included population and population density in several models I am not convinced that this will have adequately controlled for such a correlation effect.  Also even though the author controlled for a number of other confounding factors there are likely to other unknown confounders that were not taken into account.  Finally I think the author did not adequately include all lags between getting an infection and having a test taken.

“The hypothesis that eat out to help out may have increased transmission of COVID is plausible and in my view likely to be true to some extent.  Nevertheless, I am not convinced that this study provides strong evidence for the hypothesis.  In particular, any estimate of the size of the impact of eat out to help out on case numbers should be treated with some caution.”



Preprint (not a paper): ‘Subsidizing the spread of COVID19: Evidence from the UK’s Eat-Out-to-Help-Out scheme’ by Thiemo Fetzer.  This work is not peer-reviewed.



All our previous output on this subject can be seen at this weblink:



Declared interests

None received.

in this section

filter RoundUps by year

search by tag