How quickly should we reverse the present national lockdown and return, gradually or wholly, to normal? I have no idea.
One reason I don’t know is that I’m neither a medical professional nor an epidemiologist. So I can’t tell you what the consequences would be of removing restrictions this month, next month or the month after. The truth is nor can anyone else. The experts have models and estimates and their state of knowledge is far better than mine, but the uncertainties are pervasive. That’s not to say we shouldn’t use the estimates or models. They are the best we’ve got.
The real reason I don’t know, however, is much more fundamental than my lack of epidemiological knowledge. It’s because I don’t know how to make the trade-offs and huge ethical choices that face us. The uncertainties make those choices harder, but even if we had certainty about how coronavirus spreads, they would still be staggeringly tough choices to make. We shouldn’t even be talking yet about what is the right choice. We should be working out how to make the choice. We need some framework for decision-making, not a set of opinions about the right decision.
I am not an ethicist. I do hope that there are some advising government right now. But I am a social scientist — and social science in general and my branch of it, economics, do have something to say about how to make such choices, as well as something to say about the consequences.
Perhaps I need to start by persuading you that choices and trade-offs need to be made and that they are difficult choices. Maybe they are not so hard right now. With a thousand or so deaths a day even with the existing restrictions, and with the real risk that the health service would be overwhelmed without them, the choice looks easy. But it is still a choice over who lives and dies.
More people will die of other conditions over the coming months because the NHS is prioritising coronavirus patients. A lot of them will be elderly, the same group most at risk from coronavirus itself. They also will be disproportionately poor. More people will die in the years ahead as a result of the poverty, unemployment and mental health problems created by the lockdown.
There is a huge body of literature on the effects of economic and social dislocation on health and mortality. The effects are big. Recent work by, among others, Carol Propper, professor of economics at Imperial College and a colleague of mine at the Institute for Fiscal Studies, suggests that the relatively modest increases in unemployment associated with the 2008-09 financial crisis may have resulted in 900,000 more people of working age suffering from chronic health problems. Work in the United States shows huge spikes in deaths in areas blighted by industrial decline and unemployment. Such illness and death is less visible and less identifiable as being a direct result of choices we make, but it is not any less real for all that.
It is already clear that the economic dislocation caused by the coronavirus crisis will be much bigger than that suffered in 2008-09. An analysis published last week by the respected and normally rather staid National Institute of Economic and Social Research was simply entitled GDP could contract by 15 to 25 per cent in second quarter. I never thought I would read anything like it. Merely writing the words scares me. There’s no point looking for historical parallels. There are none, at least not in this country. And the long-term effects will grow more than proportionally with the period of the lockdown. Four months will be (probably much) more than twice as bad as two.
We are also making distributional choices. The lockdown is hitting the young much, much harder than the rest. That’s true of those missing out on education, of those leaving school or graduating and looking for a first job and of those already in employment. The under-25s are two and half times as likely as those aged over 25 to work in a sector that has been shut down. The lowest-earning 10 per cent of workers are fully seven times as likely to work in such sectors as are the highest-earning 10 per cent.
All of this somehow will need to be weighed against the immediate risks of loosening policy, whenever that decision is taken. In taking that decision, ministers will need to be aware of two strong biases that we as human beings, and they as policymakers, are prone to. The first is a bias towards the present and an underweighting of the future. That’s one reason why we continue to do too little to tackle climate change and invest far too little in prevention — whether that’s prevention of disease, crime or poverty — relative to what we spend on dealing with the problems when they occur.
The second and related bias is towards salience. We accept far more people dying, at least in part, as a result of air pollution than we ever would dying in traffic accidents. The latter, like coronavirus deaths, are here, immediate, obvious. The former are mere statistical deaths. We know exactly who was killed in a traffic accident last year; we know only that more people died than would otherwise have been the case as a result of air pollution, not who they were.
It sounds callous, but it is not, to warn against the same biases when it comes to dealing with coronavirus. Just because we can’t point to those who will be mentally or physically ill, or who will die early — but not directly from coronavirus — as a result of the current disaster, doesn’t mean that they shouldn’t count. The question is how they should count in the grim calculus that faces us.
This article originally appeared in The Times and is used here with kind permission.