Locking down too late but ending lockdown too early

The major reason we have one of the highest death rates as a result of the coronavirus pandemic is that Johnson/Cummings had an intense aversion to imposing a lockdown, and an unusual disregard for human life. Any decent politician, after being told at the end of February that 500,000 of their citizens might die, would have moved heaven and earth to stop that happening. Anyone watching the recent Despatches documentary would have heard about scientists worrying about how to stop the pandemic, with little pressure from politicians to do so. It may have been an ultimatum from French president Macron that finally forced Johnson/Cummings to enact a full lockdown on March 23rd.

Unfortunately the same factors that delayed a lockdown have led Johnson/Cummings to relax the lockdown too early, when the number of infections was still pretty high. This is partly because in the circles that Johnson/Cummings move, the recession that the pandemic has created is associated with the lockdown rather than the pandemic. Free the economy by ending the lockdown, they cry. Some practice what they preach, in the sense that they ignore the lockdown rules. One of these people was Cummings himself, which is the second reason why the UK lockdown is being relaxed too quickly.

This is a tragic error, not just because it will lead to yet more deaths but also because it will delay any economic recovery. As I explained in a Guardian article, any recovery will be severely limited if new infections per day remain high. While we often focus on the irresponsible minority, the majority of people are cautious, and do not want to risk catching the virus. They are going to stay away from shops as much as possible, and will certainly not go back to pubs and restaurants, or public transport if they are able to avoid it.

The idea that there is a trade-off between protecting the economy and protecting people’s health is not only wrong, it is also dangerous. It encourages politicians to relax the lockdown too early, which risks reducing the speed at which the number of new infections fall, or even stabilising infection rates at too high a level. Below is data for the number of COVID-19 admissions to hospital, that is shown in the government’s daily briefing.

It is probably the best indirect measure we have for the path of new infections over time, with a lag of 2-3 weeks. (The problem with data on the number of people tested is that it depends on how easy it is to get tested, which has varied greatly over time.) This seems reasonably consistent with the Cambridge/PHE Joint Modelling Team’s estimates that currently there are around 17,000 new infections every day.

The worrying aspect of this data is that it seems to be levelling off, which is another way of saying that R is getting close to 1. This also accords with Cambridge/PHE Joint Modelling Team’s estimates of regional R values. They note “There is some evidence that Rt has risen in all regions and we believe that this is probably due to increasing mobility and mixing between households and in public and workplace settings”.

All this is important not because we might see ‘a second peak’. It is important because it means that the number of new infections is declining very slowly, which in turn means that most people will not return to previous patterns of ‘social consumption’. That in turn means that there cannot be a complete recovery. We do not know at what level of daily infections people will be happy to resume social consumption, but it is bound to be well below 17,000. The difference between R=0.8 and R=0.9 in getting to that much lower number of infections is measured in months, as is the difference between R=0.9 and R=0.95. We are relaxing lockdown at much higher levels for daily new infections compared to Italy, France and Germany.

Relaxing the lockdown might (I stress might) be justified if there was a tried and tested alternative mechanism to suppress R. That mechanism does exist: a well functioning and comprehensive track, trace and isolate (TTI) infrastructure. Yet the government still attempts to gaslight journalists with a launch of the new Serco led, “world beating” TTI regime at the beginning of June, that we now learn will not be fully operational until September or October. Quite how Serco sold that to ministers/Cummings we can only guess. Scaling up existing local authority teams would have been both quicker and more effective, but is contrary to this government’s ideology and the interests of those who fund it.

It seems clear that many/most of the scientists advising the government also think lockdown is ending too quickly. The alert level remains at 4, despite Johnson/Cummings’ wishes. As Rafael Behr put it, “Johnson's relationship with science has gone the way of most of his relationships.” Yet this divergence does not seem to worry him and those around him at all, which is a bit odd for a government that kept claiming they were following the science.

I should resist the temptation to suggest that all this is obvious. When I modeled the economic impact of a pandemic I was surprised at how much of aggregate consumption was social. It isn’t just pubs, restaurants and tourism, but large parts of recreation, culture and transport. These sectors make up over a third of consumption. Even the demand for clothing may decline if there are no parties to go to. The pandemic creates a huge demand shock even without any lockdown measures like school closures.

That is why many better-off households have been saving much more during the pandemic. The certain way to get a recovery is to release those savings, by creating the conditions for social consumption to resume. That in turn means getting daily infections down substantially by not relaxing the lockdown too soon. As the Faculty of Public Health writes “Like everyone else we are longing for restrictions on our lives to be lifted. But evidence from around the globe shows that the way to achieve this is not to merely suppress covid-19, but to systematically reduce its incidence.”

In other words there is no trade-off between public health and the economy: better public health (less COVID-19 infections) is the sure way to a substantial recovery. The idea that we have to lift the lockdown for the sake of the economy is the new austerity. With austerity it was about how we had to get the deficit down, in order to have a sound economy. Now it is that we have to end the lockdown, in order to free the economy. In both cases it was the opposite of the truth. In both cases lives were unnecessarily lost. In both cases the recovery was blunted.

Could we get a similar recovery by some other means, such as a large fiscal stimulus? The short answer is no. Because social consumption is such a large proportion of the total, you would need a ridiculously large increase in spending in other sectors even to come close to substituting for that loss. The only reason why you would contemplate not doing the first best option, getting infections down, is because your ideology is screwing your common sense. Which is a pretty good description of how this government has dealt with this pandemic so far.