Eating out, helping out

Running the sums

In the last few days there’s been a resurgence of interest in the UK’s Eat Out to Help Out scheme of August 2020, in connection with the UK’s imminent public inquiry into COVID. For example, Sunday’s Observer headline was

Sunak under fire as ‘stupid’ Eat Out to Help Out scheme to be focus of Covid inquiry

I ended up discussing this on BBC Radio 4 More or Less just now. Of course, you can’t share graphs on the radio, so here for me is the main one.

This is a plot of the admissions ratio through summer and autumn 2020. This is roughly an estimate of weekly growth rate: if the figure is below 1 admissions are shrinking, if they are above 1 they are growing. So, it’s clear that for most of the Eat Out to Help Out period (August 2020), admissions were shrinking, and the return to serious growth only came after the scheme ended. (You can even see the small return to admissions decline that followed the second lockdown).

As I said on the radio, the 8-17% case increase attributed to Eat Out to Help Out pales into insignificance compared with this kind of sustained exponential growth. If you plot that number on the graph, it’s between the solid line and the first dotted line. It perhaps corresponds to a few days of the kinds of sustained autumn growth we saw, and perhaps at most caused lockdown to take place a few days earlier as a result.

But more importantly, I think this whole thing is an example of a foundational myth, of the kind that the inquiry should really be questioning. In the same way, despite Christopher Snowdon more or less demolishing the idea that “scientists wanted us to lock down in March 2020, but politicians and Dominic Cummings overruled them” here, that seems to have been accepted as part of the story in some quarters.

In the same way, I’d hate it to be the case that (as the Observer headline implied) Eat Out to Help Out became the focus of the story about the disasters of autumn 2020. Of course, it’s part of the story, and should be examined, but the picture is much more complicated than this one factor might suggest.

Indeed, I think it’s important that the inquiry avoids a particular kind of bias (as I discussed in Numbercrunch). We should definitely look at times when we under-reacted to COVID and ignored voices of warning. But equally, it’s important to remember that there were times when warning voices were ignored and things didn’t end in the disaster that was prophesied in some quarters. For example

1. We were warned that reopening schools in March 2021 would end in disaster:

Without additional mitigations, increases in transmission are likely, this time with more infectious and possibly more virulent variants, resulting in further lockdowns

2. We were warned about the July 2021 unlocking:

We believe the government is embarking on a dangerous and unethical experiment, and we call on it to pause plans to abandon mitigations on July 19, 2021

3. We were warned in December 2021 that:

The situation is so urgent we must take emergency action now and that means it is imperative to reduce contacts … Accordingly we now call for an immediate circuit break … No indoor gatherings between households of any size

On all these occasions the Government did not heed these warnings, and we didn’t return to a situation where deaths became very high or admissions numbers returned to levels last seen in January 2021.

So in my view, it’s important to remember that you can’t just construct a narrative of “the Government ignored advice and things went wrong” without at least considering the other times as well, or there is a serious danger that the result of the inquiry will be recommendations that would embed future overreactions into the system.