COVID-19 and the Road to Serfdom

The art of economics is in part to calculate the trade-offs of various policy decisions. This involves speculating about opportunity costs and counterfactual scenarios. For example, when a company considers opening a factory in, say, Leeds, it has to consider what opportunities it would miss out on by not opening up in Manchester.

In response to the (ultimately successful) calls for a lockdown in response to COVID-19, many quite sensibly questioned what the economic costs involved would be. Pundits and armchair experts the Twittersphere all over decried such shameless materialism as heartless at best and eugenicist at worst. Most drew a false dilemma between lives saved and economic performance. They failed to realise that without a healthy economy, the citizenry will have inadequate access to resources that make them healthier. Privation does in fact cost lives. Indeed, Philip Thomas from the University of Bristol suggested that the economic costs of lockdown may cost more lives than it saves.

This has received a fair share of criticism from the likes of Sam Bowman, whose rebuttal of Toby Young’s article was characteristically rigorous and impressively well researched. However, his argument to a large extent rested on the supposition that there is no evidence that short-term reductions in income can increase mortality.

Unfortunately, the British Medical Journal disagree. They estimate that an excess of 7,000 suicides took place the year immediately following the 2008 financial crisis. And it may be reasonable to expect these numbers to be larger in a situation where the economic contraction is more dramatic and accompanied by a crippling reduction in social interaction. Bowman’s argument also rested on the assumption that a total lockdown is the least intrusive way of saving lives; a proposition for which there is little evidence.

In a similar line of thought, professional economist Jonathan Portes wrote an article in The Guardian protesting that mortality is actually reduced in times of recession. This relies on empirical evidence from the 2008 crisis which shows that lives were saved in this recession due to reduced automobile accidents, reduced alcohol consumption, and reduced CO2 emissions, among other factors. He then went on to assert that the government must do whatever it can to save lives, whatever the cost.

It is my contention that there are several fatal flaws with this analysis. They are, as I see it, as follows: that the 2008 crisis is largely incomparable to the current situation for his purposes; that his sample size is inadequate to make a generalized claim about the negative causal relationship between recessions and fatalities; and that the deaths saved by reduced choice and freedom should be discounted from the analysis.

The first objection is relatively straightforward. It is that the human cost of 2008 was lower than what we currently face. This is because reduced income and access to resources, when accompanied by enhanced loneliness and isolation, can cause notably pernicious health outcomes. A meta-analysis co-authored by Julianne Holt-Lunstad, PhD, a professor of psychology and neuroscience at Brigham Young University, found that a lack of social connection heightens health risks as much as smoking 15 cigarettes a day or having an alcohol use disorder. She also found that loneliness and social isolation are twice as harmful to physical and mental health as obesity.

Furthermore, as the Financial Times have pointed out, “the economic impact of the lockdown has been devastating. The UK’s unemployment rate has doubled to 2m”. Alongside evidence from the NCBI which suggests that unemployment may cause increases in alcohol and tobacco misuse in the young, it seems clear that the weight of evidence is not entirely in Portes’ favour even if we put to one side the situational incongruences. This also says nothing about the negative impact that the lockdown is having on patients presenting at hospitals with symptoms of other morbidities.

Recessions are not one-in-a-lifetime occurrences. Indeed, they are fairly frequent. Yet, Portes presents evidence about the impact that the recession of 2008 had on mortality as conclusive evidence about the causal relationship between all recessions and mortality. It is not particularly controversial that this is not a sufficiently large sample to justify such a claim.

The final criticism I have of Portes’ analysis is that, in my view, many of the deaths avoided during a recession should be discounted. This is because there is a clear normative distinction between a death caused by lifestyle choices that are made voluntarily, and deaths caused due to the involuntarily imposed and unexpected consequences of a recession. When I make the decision to smoke, drink, or drive, I do so aware of the risk. But a person who commits suicide due to the consequences of an economic contraction has given no such consent. To make such a comparison is in my view unsympathetic and misleading.  Of course, deaths from CO2 emissions do not fall under this category, as emissions are externalities. But the evidence cited by Portes’ indicates that only a small number of lives are saved in this way.

Furthermore, physicians are familiar with the perennial tension between quality and quantity of life. Indeed my father, a practicing physician who works primarily with drug dependent patients, constantly reminds me of this. Is it unreasonable to apply this notion on a macrosocial level? Is it worth saving a few months or years for the few at the expense of misery for the many? Of course, I do not claim to have the knowledge or expertise to make the claim that this is what is happening, but I know of no evidence that this question has been given extensive attention.

This article should not be taken as a call to lift the lockdown that we endure in Britain and most of the West. It should merely serve as a reminder about how difficult the decision before us and how limited the evidence is. There are no quick and easy answers, yet we in Britain seem to act with such haste and turbulence. Up until the lockdown, the UK government was following expert advice compiled beforehand to the letter. They were sensitive to concerns about isolation fatigue which now seem to have been vindicated in light of protests across the United States and in Germany.

And yet, in response to work done by Professor Neil Ferguson of Imperial College London, who has received heavy criticism for his flawed work on the foot and mouth crisis, we have suspended all that nuance alongside our economy and our freedom. Less drastic measures have barely received attention, despite there being little apparent difference between outcomes of subtler approaches in Sweden and the Netherlands compared to the UK. As Daniel Hannan has pointed out, the law of diminishing returns indicates that tighter social restrictions are less likely to yield proportionate gains.

My concern is that in months, perhaps years, from now we will look back with the deepest regret. The liberal tradition has always accommodated for emergency powers. Milton Friedman said in an interview with Peter Robinson that the state should have powers to reduce contagion; and stimulus to protect unsuspecting firms from exogenous shocks, though unpreferable compared to voluntary contingencies, are palatable from a libertarian view. But our conception of an emergency must be rigid and require the strongest of evidence and deliberation from all experts, not just virologists; lest we pave the road to serfdom from which it is difficult to return. As Jonathan Sumption put it, tyrants rarely take power; we give it to them.