The Astronomical Cost of Lockdown

To laugh or to cry. It is difficult to know which response is more appropriate to the Government’s attempt to quantify the impact of lockdown, published last night with the title “Analysis of the health, economic and social effects of COVID-19 and the approach to tiering“.

If you wanted a chuckle, then imagine you had the job of the unfortunate civil servant who had been given the job of cobbling together this strange hotch-potch of information. The document is clearly a rushed job, published with the political aim of persuading the growing number of Conservative MPs who are sceptical about the need for tighter restrictions that they are, in fact, necessary. (There was once a time, not so long ago, when the Civil Service would have demurred from being involved in such a blatantly political operation.) A futile effort, for no self-respecting MP could be persuaded by such a flimsy document.

Take its estimates of additional deaths from other diseases. Table 9 of the report looks at the effect on morbidity and mortality of certain conditions – alcohol misuse, road injuries, depressive disorders, and the like. But instead of trying to estimate the actual numbers, the report simply uses up and down arrows to describe the general direction of change that social distancing measures might produce. Is that really the best that our Rolls Royce Civil Service can do?

Or take the report’s attempt to take a sectoral approach of the impact of lockdown on the economy. Here again in some cases, all the report does is provide a pre-COVID-19 assessment of Gross Value Added (GVA) output of each sector – it seems to be afraid of making any calculation of the likely impact. If estimates are made, they are drafted as general changes in GDP, not reported as actual costs on deeply impacted industries.

There is another way of doing this, one which Jim McConalogue and I have used in our new report, The Cost of the Cure, published today by Civitas.

For we, unlike the Government paper, do at least try to provide a range of actual estimates of the costs. Consider the following: the UK government routinely quantifies the value of a life saved – a QALY – which it uses when considering licensing a new drug treatment or assessing whether it is worth making an investment in altering a particular road to improve its safety. If it does so when considering spending relatively small sums of a few million quid, then surely it should do so when considering spending many hundreds of thousand times as much?

But not in this latest Government report. It does not even mention QALYs. Why not? Well, maybe one reason is that even on the most optimistic assumptions, our estimates suggest that the Government’s measures have been completely disproportionate. Say you used Professor Neil Ferguson’s now widely questioned forecast in March that there would be 510,000 deaths from the pandemic if nothing changed. (It is worth noting that Ferguson acknowledged that this forecast assumed that there would be no mitigating change at all, an assumption he accepted was “unlikely”.) Then take the average age of those who have actually died (83 as it happens). Then estimate their average life expectancy if Covid had not happened – another seven or so years – and you have about 3.3 million QALYs or quality life years saved ((510,000 x ~7)-~50,000 = ~3.3 million). Compare this with the OBR forecast that the Treasury will spend at least £280 billion in alleviating the impact of lockdowns and you can see that the Government is spending the equivalent of £96,000 for each QALY saved. As it happens, this is over three times the figure that the NHS routinely uses of £30,000 when assessing whether a particular course of action is worthwhile.

More worryingly still is that this is an extremely cautious estimate: we show that the Government could be spending 80 times the maximum threshold that the NHS is usually prepared to spend per QALY. A wide range of estimates is inevitable given that they rely on certain counterfactual assumptions which are by their nature unprovable. But in all cases, in our report the assumptions and estimates have been set out; and in all cases these assumptions and estimates have been cautious.

And it is not just the financial costs of lockdown that need to be properly assessed. These should include items such as:

The 20,000 lives that could be lost from delayed treatment for cancer and other diseases
The 16,900 additional domestic violence cases that were recorded between March and June 2020
The significant increases in depression (64% recording common depressive symptoms), anxiety (69% report increases) and loneliness (reports of loneliness of parents of under-fives up by 1.4 million)
The significant increases in substance abuse, including high-risk drinking among adults up by 3.7 million, 20% increase in opiate addictions, 39% increase in number of relapses among addicts
The 25% to 30% reduction in learning among primary and secondary pupils, respectively
The hugely differential impact of lockdowns where the less well-off have suffered greatly while the better off have tended to enjoy their extended gardening leave (sometimes literally so)

So much pain, so much economic destruction but with so little scrutiny.

That is why the UK Government should now undertake a proper assessment of the financial, economic and wider impact of the measures it has and is taking, not the cut-and-paste job it has just published. For how else can Parliament, future inquiries and the wider public make an informed judgement on whether the measures are proportionate?

Tim Knox is the author with Jim McConalogue of The Cost of the Cure, published today by Civitas.