In early March 2020, it seemed that the COVID-19 virus in the UK was spreading exponentially with no clear sign of imminent slowing; the fatality rate was unknown and the ability of the National Health Service to deal with rapidly rising numbers of seriously ill people was unclear. Estimates made at that time by Professor Ferguson’s team at Imperial College put the possible level of UK deaths if there were no change in behaviour at 500,000 (Ferguson et al. 2020). Based on that (and other) assessments, the UK government followed the example of several other European countries in introducing severe restrictions, backed by law and enforced by police, on individual movement. The key message was to stay at home; this was a blanket lockdown. But how long should a lockdown last and when has its benefits fallen below its costs?
In recent research undertaken with health researchers at Manchester University and at health consultancy company RES Consortium (Miles et al. 2020), we estimated costs and benefits of the UK lockdown. We find that the costs of carrying on with such a lockdown beyond three months are likely to have become greater – perhaps far greater – than its benefits.
Ornelas (2020) provides a useful conceptual framework for thinking about how long a lockdown implemented to halt the spread of the COVID-19 pandemic should be kept in place.
Any such assessment has to place a value on possible lives saved. There is no way to do this that is clearly ethically justifiable, empirically reliable and widely accepted. One approach is to focus on quality-adjusted life years (QALYs) that may have been saved. The NHS has established guidelines about how much should be spent on medical treatments that on average yield benefits in life years saved. This effectively values a QALY at £30,000. One can use that rule with estimates of lives that might have been saved due to the restrictions and compare those with the costs of lockdown.
Those costs go well beyond a simple focus on GDP lost. Health costs from telling people to stay at home are likely to be large and long-lasting. Referrals for cancer investigations were 70% down in April 2020; there were hardly any follow-up routine appointments for long-term conditions in UK primary care between mid-March 2020 and the beginning of June; outpatients seen were 64% down and elective admissions to hospitals down 75%; attended appointments in general practice were down 35%. The impact of the stress of the lockdown on anyone with a pre-existing mental health condition, let alone the population as a whole, is yet to be determined.
Suppose one took a very favourable set of assumptions about a widespread lockdown that lasts three months.
Suppose that without this lockdown that required people should not leave their home, there would have been no change in behaviour at all (wildly unlikely) and that the estimate of 500,000 deaths with no change in behaviour made by the team of Professor Ferguson is accurate. The lockdown could then have saved around 440,0000 lives – the 500,000 deaths net of 60,000 estimated excess deaths that have occurred. Based on the ages of those who have died with the virus in the UK, and using average residual life expectancy for UK citizens of the relevant ages, the average number of years of life lost is around 10. Assume those who would have died lose, on average, 10 QALYs (although in fact those who might have died are likely to have had substantially lower life expectancy because of the prevalence of co-morbidities). Using the £30,000 figure then generates a value of potential years of life saved of £132 billion.
What is the lowest plausible estimate of the cost of the lockdown? Suppose we only count lost GDP. The Office for Budget Responsibility and the Bank of England put this at 13-14% of GDP if restrictions are eased now and the economy bounces back. If we ascribe only two thirds of this lost GDP to the lockdown (since behaviour would clearly have changed with no government restrictions), this generates a cost of around £200 billion. This cost ignores all future lost GDP beyond 2020, it excludes all medical side effects (not treating cancer patients, stopping screening for serious conditions, etc.), and it ignores the future damage of huge disruption to education.
This is a macabre calculation, but one which estimates the cost of the extended lockdown to be high relative to the benefits. It is also a calculation which is hugely pessimistic about what would have happened in terms of lost lives with no blanket lockdown, since people would have (and had already) taken precautions, short of simply not leaving home, absent the UK lockdown. Table 1 shows other scenarios for costs and benefits (the latter a reflection of potential lives saved by lockdown), all of which show costs of extending the lockdown for as long as three months exceeding costs.
Table 1 Benefits (+), costs (-) and net benefits of a March to June lockdown in the UK (converted to an index of £ billions)
Notes: Each life saved is estimated to resulted in 10 more quality-adjusted years of life. The NICE resource threshold of £30,000 is applied to each of these quality adjusted years. The benefit number (+) is the product of assumed lives saved and £300,000. The money value of GDP losses are taken as the only cost (-) of lockdown and percentage falls in GDP are applied to the UK 2019 level of GDP of £2.2 trillion
But why use the figure of £30,000 per QALY as a guide to how much we should pay for policies that might save years of life? The £30,000 figure is used in decisions within the UK health system; it is not an arbitrary number. It is not based on likely future earnings lost or the value of future consumption – calculations that are open to the moral objection that they reduce the value of human life to how much people would have spent on commodities. Instead, it is what is considered the highest level of spending that should be used in the UK health system to generate extra quality adjusted years of life – and it is the saving of lives that the lockdown is for. In using this yardstick, one is treating decisions on how to face COVID-19 in the same way as decisions are made in the UK about treatments for cancer, heart disease, dementia, diabetes and many other health conditions. On that basis, it would seem as though the benefits of continuing with the lockdown are lower than its costs.
But a case can be made for using much higher values for potential good years of life saved. In the US, figures for the value of statistical life of around $10 million dollars are often used (Viscusi 2018, 2020). For someone with 50 years to live, that gives a value of a year of life of $200,000 – which is five times the NICE figure, though lower if one allows for differences in average incomes in the UK relative to the US. Layard et al. (2020) also use vastly higher figures for the value of a year of life saved, putting it at £750,000 – some 25 times the guidelines used for the NHS. The difference is immense. If the higher figure is anywhere near being reliable, it suggests that NHS spending should be massively larger – not by 10% or 20%, but by an amount that would raise it to a huge share of GDP. The question then is: Why do we not see anywhere near that level of public health spending in the UK (or in any country)? Does the political system, decade after decade, generate health spending that is too low by such an enormous amount? If that is so, why does it happen? And if it does generate far too little public health spending, why isn’t private health spending in the UK 15% of GDP rather than 1.5%?
Figures twenty or more times the NICE value seem implausible. But figures two or three times as great might not be. Layard et al. note that a value of potential extra years of life of £60,000 (and not the NICE value of £30,000) is often used in the UK. Yet our analysis suggested that even using values three times as high as the NICE guidelines mean that maintaining the blanket lockdown for as long as three months is still likely to have taken costs beyond benefits.
A movement away from blanket restrictions that bring large, lasting and widespread costs and towards measures targeted specifically at groups most at risk is now imperative in the UK.
Editors’ note: This is an extended version of an article that appeared in The Times newspaper on 25 June 2020.
Ferguson, N M et al. (2020), “Report 9: Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand”, Imperial College COVID-19 Response Team.
Ornelas, E (2020) “Managing economic lockdowns in an epidemic”, VoxEU.org, 28 March.
Layard, R, A Clark, J-E De Neve, C Krekel, D Fancourt, N Hey and G O’Donnell (2020) “When to release the lockdown: A wellbeing framework for analysing costs and benefits”, Center for Economic Performance Occasional Paper 49.
Miles, D, M Stedman and A Heald (2020), “Living with COVID-19: balancing costs against benefits in the face of the virus”, forthcoming in the National Institute Economic Review, August (available at http://www.imperial.ac.uk/people/d.miles)
Viscusi, W K (2018) Pricing Lives: Guideposts for a Safer Society, Princeton University Press
Viscusi, W K (2020) “Pricing the lives saved by coronavirus policies”, New IZA World of Labor Commentaries on the Coronavirus Crisis, 27 March 27.