Sometimes we face agonising dilemmas between terrible alternatives. We should try to avoid getting into such situations. In the UK at present we should avoid an uncontrolled second wave of Covid-19 which kills tens of thousands and also a second general lockdown knocking billions off the economy.
ow might we evaluate the current lockdown? First of all, there is the undoubted success in terms of lives saved. A death toll of about 850 in Northern Ireland (NI) was grievous but contrasts to the projection made in March that 15,000 might be lost.
What about the cost? Given all the ethical and philosophical difficulties relating to trying to put a cash value on a human life I avoid placing a monetary value on the "benefit" of the lockdown in number of lives saved. What can be done is to measure the implied economic cost per life saved and compare that to the standard NHS/National Institute for Health and Care Excellence (NICE) guideline that the cost of treatment should not exceed £30,000 per year of life.
It seems safe to assume the upper limit on the number of lives saved is 14,000. We adjust from lives saved to years of lives saved, to Quality Adjusted Life Years (QALYs), by assuming an average of eight additional years per person. That adjustment is based on demographic data: the average age of Covid victims and the normal life expectancy at that age. By implication, lockdown saved 112,000 QALYs: 14,000 times eight.
As for the costs, we have to make a series of bold assumptions including that we can confine those costs to the economic ones. There have of course been other costs such as disruption to wider healthcare. We also assume NI's GDP, which was estimated to have been £47.3bn in 2019, will decline by 10% in 2020. We then assume a recovery follows albeit it takes four years to return to the level of output achieved in 2019. From the mid-2020s the trend rate of growth is a "moderate" 1.5% per annum. There is obviously no certainty about what would have happened to the economy in the absence of Covid. So, we end up with two paths for the economy.
First, what we assume the level of output would have been without the virus and lockdown and, second, what we assume will actually happen. We further assume that half of the difference can be attributed to the lockdown policy during March-July 2020. The argument here is that even without a lockdown consumers and investors would have cut back spending making a substantial recession unavoidable.
Given this background here is the actual calculation. If we consider just the cost in 2020, this year's recession, a GDP shortfall of £2.5bn compared to 112,000 QALYs saved, that is £22,400 per QALY. This falls below the £30,000 threshold used to judge other health procedures. However, the economic cost of the Covid-related recession will almost certainly last much longer than one year.
The economic cost in 2020 values over five years 2020-24 rises to £7bn and hence the cost per QALY is £62,100. Over 20 years, discounted back to 2020, the cost is £18.7bn or £166,700 per QALY. In other words, once allowance is made for the fact that output over many years will be less than would otherwise have been the case then cost per QALY escalates to several times the £30,000 limit.
In reacting to these figures we should always remember that one life lost is a human tragedy. It is also the case that we are dealing with very large but also uncertain numbers where all the assumptions can be questioned.
It looks quite likely that economic cost per QALY has been considerably higher than the £30,000 maximum normally allowed for other health care procedures. Importantly, I have chosen assumptions which will if anything tend to under-estimate that cost.
It could, of course, be argued that £30,000 is a somewhat arbitrary test figure in any case.
In trying to conduct a cost-benefit analysis of the UK lockdown the London economics Professor David Miles considered evidence from the US suggesting a QALY could be valued at perhaps two or three times that level. He was, however, doubtful that it should be increased many times over.
This further seem a pretty macabre exercise.
I am very sympathetic to the argument that in certain contexts a human life is of infinite value.
However, in the world of government policy health spending is not allowed to consume 100% of GDP and even that GDP is not limitless.
Policy makers have to be aware that resources committed to one outcome must mean less is available for something else.
Was the March 2020 lockdown decision justified? We should always remember policy decisions are made given the information available at the time. Given what was known at the time and given the absence of an alternative the policy probably was justifiable.
What this exercise in measuring the cost of lockdown confirms is what we probably already knew: we really cannot afford another general lockdown which costs the economy billions and hence reduces wellbeing in all sorts of ways.
We do not know when a vaccine will be available but the priority must be alternative ways of containing a second wave such as test and trace, track and trace and very localised lockdowns.