In his testimony to the Health and Science select committees, Dominic Cummings heavily criticised the Government’s handling of the pandemic. One of the biggest mistakes, he argued, was the failure to impose border controls:
Obviously, we should have shut the borders in January. We should have done exactly what Taiwan did… Yes, that has some disruption, but the kind of cost-benefit ratio is massively, massively out of whack, and at least it is worth a try, like lots of things. At least you try it … If it doesn’t work, you still have the whole nightmare to deal with, anyway.
However, Cummings somewhat absolved the Government of blame on this score – at least with the respect to the period before April – insofar as all the scientists were advising against border controls:
He was told, and we were all told repeatedly, that the advice is not to close the borders, because essentially it would have no effect… you cannot blame the Prime Minister directly. That was the official advice. The official advice was, categorically, that closing the borders will have no effect.
Cummings’ testimony is consistent with the evidence from SAGE meetings in January and February of last year. For example, the minutes of a meeting on January 22nd record that “NERVTAG does not advise port of entry screening”.
Another factor Cummings mentioned, as to why the Government didn’t impose border controls, is political correctness:
At this time, another group-think thing was that it was basically racist to call for closing the borders and blaming China, the whole Chinese new year thing and everything else. In retrospect, I think that was just obviously completely wrong.
What should we make of Cummings’ argument that border controls were at least “worth a try” in January? On the face of it, the argument seems very reasonable. In the best-case scenario, we could have achieved the same outcomes as New Zealand – zero excess mortality and just a small decline in GDP. And in the worst-case scenario, we’d have been in the same situation as otherwise.
However, the latter outcome – being no worse-off – isn’t necessarily the worst-case scenario. A potentially even worse scenario is if we’d contained the virus until the autumn, and then experienced a major epidemic at the same time as the NHS came under its normal winter pressures.
This was in fact one of the reasons why scientists were initially advising against both border controls and lockdowns. Cummings was apparently told:
Even if we therefore suppress it completely, all that you are going to do is get a second peak in the winter when the NHS is already, every year, under pressure … If you try and flatten it now, the second peak comes up in the wintertime and that is even worse than the summer.
This argument should not be dismissed out of hand. Several of the European countries with the highest death tolls – Poland, Bulgaria, Czechia – are ones that escaped the first wave, only to get clobbered in the second. (Of course, there may be several reasons for the high death tolls in these countries; I’m not suggesting the epidemic’s timing is the only one.)
Deciding whether to impose border controls therefore represents a trade-off between the benefits of buying time and/or achieving containment versus the risks of postponing the epidemic until the winter.
Comparative evidence suggests that achieving containment would have been a tall order for the U.K. Although some Western countries have successfully contained the virus, they are all geographically peripheral nations with relatively low population densities and few ports of entry. Britain has the advantage of being an island, but it’s larger, denser and more connected than most of its peers.
Indeed, there are several countries that failed to contain the virus despite imposing early border controls. To identify when countries introduced restrictions on international travel, I will rely on the Oxford Blavatnik School’s COVID-19 Government Response Tracker. And for charting the epidemic’s trajectory, I will use the daily number of deaths per million people, as reported by Our World in Data.
Canada imposed a total border closure on March 18th (see chart below). This was too late to prevent the first wave. Yet the policy has been in place ever since, and it did not stop a second wave from emerging in the autumn.
Hungary imposed a total border closure on September 1st (see chart below), but this did not prevent two major epidemics from burgeoning in the autumn and the spring. Note: the Blavatnik School rates countries’ travel restrictions on a 0–4 scale (with 4 corresponding to a “ban on all regions or total border closure”) and Hungary’s rating hasn’t gone below 3 since March 9th 2020. (By comparison, South Korea has never had a rating of 4.)
Argentina imposed a total border closure on March 16th (see chart below), but this did not prevent a major epidemic from burgeoning in the country’s spring. Note that the policy was put in place 199 days before the peak of daily deaths, which would be the equivalent of the U.K. having closed its borders on September 27th 2019.
Colombia imposed a total border closure on March 25th (see chart below), but this did not prevent a major epidemic from burgeoning in the country’s spring. Note that the policy was put in place 131 days before the peak of daily deaths, which would be the equivalent of the U.K. having closed its borders on December 4th 2019.
Although several East Asian countries have apparently used border controls to great effect, it’s unclear whether the U.K. would have been able to replicate their success. In a recent article, the Economist suggested that people in East Asia may “benefit from ‘cross-immunities’ – a level of protection against SARS-CoV-2 conferred by past infection by other viruses circulating in the region”.
Assuming Britain wouldn’t have been able to contain the virus, deciding whether to impose border controls then becomes a trade-off between the benefits of buying time versus the risks of postponing the epidemic until the winter. I’m not sure what the correct answer is here, but given the risks of postponement, Cummings is wrong to suggest that imposing border controls was “obviously” the right thing to do.