In a webinar on 8 October, the government’s chief medical officer, Chris Whitty, informed MPs that the number of Covid-19 patients in intensive care in northern England might surpass the April peak – and that one of the main culprits was the hospitality industry.
Whitty showed the MPs unpublished figures that suggested pubs, bars, restaurants and cafés accounted for 30 per cent of common exposure settings in England, with the figure rising to 41 per cent for those under the age of 30.
The figures were based on data from a sample of just 98 pubs and 67 cafés and restaurants. The Institute of Economic Affairs has since said that the evidence for closing pubs is “very weak”.
The debate isn’t purely theoretical: under Boris Johnson’s three-tier Covid lockdown system, pubs will have to close in the areas under the toughest restriction level unless they serve food. These new restrictions follow those in Scotland, where restaurants in the central belt have closed for at least two weeks.
The leader of Manchester City Council, Richard Leese, has claimed that according his data “very little” Covid-19 transmission takes place in pubs and that closing them en masse could lead to more cases as people instead meet in each other’s homes. Significant numbers of jobs hang on the issue.
Publicly available figures from Public Health England (PHE) stand in contrast to Whitty’s slides, which warn of the disproportionate role the hospitality industry plays in the spread of Covid-19 .
PHE data shows that there was only a modest rise in the number of coronavirus outbreaks recorded in food outlets and restaurants, with the bulk of the new cases originating in educational settings, workplaces and care homes. Of the 634 Covid-19 infections linked to particular institutions registered in England in the week to 4 October, just 22 are believed to have occurred in food outlets and restaurants – 3.5 per cent of the total. Pubs and bars that aren’t classed as “food outlets” are not counted as an individual category, but the number of infections linked to “other” settings, while rising, is still lower than in care homes, workplaces and educational establishments.
There is an important caveat, which is that this data isn’t stating that 3.5 per cent of all cases started in restaurants and food outlets. The figures instead represent outbreaks where more than one person had or is suspected to have had Covid-19. A large outbreak – in a university, perhaps – is still just one outbreak. The data gets even trickier: a previous PHE report showed that eating out, along with shopping, was one of the most common activities those who tested positive did before they experienced Covid-19 symptoms.
These figures don’t necessarily imply that those activities are disproportionately responsible for the spread of the virus. It may be the case that shopping and eating out are the most common activities among those who didn’t test positive for Covid-19 too.
Bookings data from OpenTable shows a dramatic rise in restaurant footfall in August on the days the Chancellor Rishi Sunak’s Eat Out to Help Out scheme was in effect. (The initiative offered diners 50 per cent off food up to £10 every Monday, Tuesday and Wednesday.)
Eat Out to Help Out temporarily shifted the usual weekend spike in pubs and restaurants to the beginning of the week, meaning the average outlet experienced up to a three-time increase in customers when compared to the same day and week in 2019.
Once the scheme ended, the number of diners leveled off in September and the beginning of October to rates similar to last year, indicating that eating out was as common as before the pandemic.
However, restaurants bookings sharply fell by 20 per cent on 12 October and 24.4 per cent on 13 October, although it is too early to say whether this is the beginning of a larger trend.
While the publicly available figures in the UK are contradictory, the Centers for Disease Control and Prevention in the US has published a study that found those who tested positive for Covid-19 were around twice as likely to have dined in a restaurants compared to those who tested negative. There are some general epidemiological considerations, too. Covid-19 outbreaks are generally thought to be more likely to start in crowded indoor settings. Pubs and restaurants pose an additional risk as people can’t wear masks while they’re dining. Alcohol can also reduce inhibitions and make it more likely that people will not respect social-distancing guidelines.
“Bars are a really important place of spreading of infection,” said Dr Anthony Fauci, the US’s top infectious diseases specialist.
“And that becomes particularly important if you happen to be in an area where there’s a high degree of community spread.”
The government has a difficult task in stopping the spread of the virus while also sustaining economic growth. Its options are limited and the data, as we have seen, is incomplete. While the bulk of the spread still happens in homes, authorities can’t do much to curb that beyond banning household mixing. Schools and universities are also playing an increasing role in the rising number of infections, but the government has insisted they will remain open in all circumstances.
That leaves restaurants, pubs and cafés as outlets that ministers feel they can – and perhaps must – close, short of imposing a full national lockdown. But that comes with significant economic costs. Better data – a better ability to track and trace individual cases, a more comprehensive picture of how and where people are contracting the virus – would make it easier to work out whether pubs and bars are the right lever to pull, rather than merely the one that is most politically convenient.