3 November 2020

Mismanagement of the COVID crisis has left the country with no good options. What does an evidence-based approach look like from here?

The difference between good and bad management of the COVID crisis was brought into stark relief this week. While Taiwan passed 200 days without a confirmed COVID case, the UK Government imposed a new nation-wide lockdown. At the same time, Taiwan’s GDP is expected to grow in 2020, as is China’s, which compares with a UK fall now expected to exceed 10%.

The best pathway to dealing with COVID now seems clear when comparing those countries that have coped well with those, like the UK, that have not:

·      Impose restrictions early

·      Build societal consensus around clearly and consistently communicated measures

·      Reduce infections below the rate that can be dealt with by track and trace

·      Build track and trace capacity to allow identification and management of outbreaks

Successful countries have shown that there should be no trade-off between health and economic outcomes. In fact, health and economic outcomes are correlated. The economic costs of suppressing the virus until a vaccine or treatments arrive needn’t be prohibitive.

The UK Government has been criticized for not locking down early enough, and for wasting the grace time we were given by being able to observe what was happening in Asia and then Italy. This criticism smacks of being with the benefit of hindsight as the UK’s much vaunted ‘world-beating’ scientific establishment was not speaking with one voice on the issue at the time. (Although this may have been the fault of the mechanisms by which scientific advice is provided, which may have encouraged group-think, as opposed to the quality of the scientists themselves.) The delay in lockdown looks more like a system failure across UK institutions than one that can be put solely at the door of the Government. 

Instead, our biggest error came in a premature lifting of lockdown at the end of June, before the infection rate and the effectiveness of track and trace had reached the cross-over point where the UK could move to an Asian (or even German) style monitoring approach. This is not with the benefit of hindsight and was widely discussed contemporaneously.

The failure to reach the cross-over point between infection levels and the capability of track and trace meant that it was only a matter of time before exponential growth in the virus would return as the economy reopened, universities returned, and winter approached. What is particularly tragic about the wasted opportunity is that the Government had been pleasantly surprised by the level of lockdown compliance. A further period of lockdown at that point, if the time had been well-used on building track and trace, could have led to us being closer to the German position today. As it is, the pain incurred during the first lockdown has to a significant degree been incurred in vain. 

Was lockdown ever worth it? 

Those arguing for early easing of lockdown restrictions were concerned that the economic price being paid was too high. I disagree. Early estimates suggested that 500,000 COVID deathscould have occurred for letting the virus rip. NICE has an established framework for assessing the efficacy of NHS treatments. In essence this deems treatments costing less than £20,000 per Qualy (Quality Life Year) to be effective. This sounds callous, but a moment’s thought tells us that there must be some resource criterion to decide what health actions are worthwhile. 

What does this mean in practice in relation to COVID? Data is still emerging but one early study indicated that on average a COVID death results in ten lost years of life, confounding the idea that most people killed by COVID didn’t have long left anyway. This would suggest that we should pay £100bn to save 500,000 lives. We also know that there are knock-on consequences of high COVID infection rates. For every person killed, another suffers longer term health consequences. An overwhelmed NHS would be forced to cancel non-essential treatments and screenings leading to missed diagnoses. It is hard to quantify these indirect impacts, but it seems reasonable to assume both that they are non-zero but also less than the direct impact of COVID deaths. For the sake of argument let’s add 50% to what we would be prepared to pay to avoid these consequences. The resulting cost is £150bn, or around 7% of 2019 UK GDP. 

Note that this is the incremental cost that we should have been prepared to pay compared with the scenario of 500,000 lives lost. We know that wide virus prevalence itself affects consumer behaviour, so there would also have been an economic cost to letting the virus circulate unchecked. It is hard to know what this would be, but one datapoint is that in the US, where the economy has remained much more open than in the UK, GDP is projected to fall by a little under 4%. Taking into account the stronger underlying growth trajectory in the US, a reasonable guesstimate is that in the absence of any lockdown action, UK GDP would have shrunk by 5% simply due to changes in UK consumer and company behaviour. This can be viewed as the unavoidable economic cost of COVID.

So adding what we should be prepared to pay to mitigate the health impacts to what the unavoidable cost of COVID might be, we should have been prepared to tolerate a cost of 12% of GDP for an effective lockdown resulting in a V-shaped recovery. This is broadly in the range of what the Bank of England was suggesting in May the economic costs of dealing with COVID would be, before the economy regained its prior growth trajectory.

Overall, the decision to lockdown in spring seems reasonable to me given the information available then and since. It does seem that the problem was not with the action taken, but in the failure to see it through. 

No good answers

So which way should we go now? As the old joke goes, I wouldn’t start from here. But as the cliché says, we are where we are. What does an evidence-based approach look like from here?

COVID is undoubtedly soaring back in the UK. And while better practices and health management mean that we can expect lower mortality rates than in the spring, COVID deaths are already within two doublings of the April peak. It seems, therefore, that we are very much back where we started. There are three options from here.

1.     Seek to eradicate the virus

Evidence from the UK’s prior experience, but also that of Victoria in Australia, suggests that crushing the virus from this point so that it comes within the capacity of the UK’s stuttering track and trace system would require a fairly extreme lockdown of the order of three months or more. From this starting point, with an increasingly jaded and economically suffering population, this seems hard to imagine. Moreover, given the cumulative stress already placed on business, the economic costs of such a lockdown now would inevitably be higher than first time round, perhaps not in terms of GDP but in terms of business failures and permanent scarring. Still it should not be ruled out, given that the costs already incurred are now in effect a sunk cost that cannot be recovered whatever we do. This approach would at least have the benefit of outlining a clear strategy and objective. 

2.     Stop-start

This appears to be the Government’s current approach, whether by accident or design. Under this approach periodic lockdowns are undertaken as infection rates cause hospitalisations to rise to the point at which the ability of the NHS to cope is called into question. Lockdowns are then released once the NHS is able to cope again, but before the point at which the virus is within the capacity of track and trace to control. This continues until a mixture of vaccine, treatment, and natural burnout of the virus allows some level of normality to return – probably at least 12 months. The problem with this is that it is easy to imagine the UK being locked down for a month or more every few months given the background level of infection rates. The run-rate economic cost increases as businesses become ever more stressed and workers lose experience and skills, not to mention the compounding mental health toll. It also seems highly likely that public tolerance for yo-yoing levels of restriction with no clear underlying strategy will result in reduced compliance in the periodic lockdowns.

3.     Tolerate higher virus levels on a managed basis

This is in essence the approach behind the self-importantly titled Great Barrington Declaration (lucky they didn’t sign it in Whynot, North Carolina). Shield the vulnerable and let the virus circulate amongst the rest of the population in order to build herd immunity. The problem is that, especially in more economically vulnerable communities, the safe and the vulnerable can’t be separated so easily. Moreover, we don’t yet know at what level herd immunity will start to apply, if at all. From a public health point of view, this seems a high risk strategy and one that seems likely to result in leakage of the virus from low risk to high risk populations, and thereby to lead to a level of hospitalisation that causes the NHS to creak at the seams.

So what next?

My own view is that we should’ve locked down for longer first time round to get the infection rate right down, done a proper job of building test and trace, and in essence gone for the successful East Asian (or German) approach. This would have cost us more than Germany, because of the Government’s tardiness in locking down in the first place, which meant circulation of the virus was allowed to get to a higher level. But we could’ve got there.

But that opportunity is now gone. My fear is that the approach of largely eradicating the virus is no longer realistically open to us. Trust in Government is too low, economic stress too high, and public weariness too great to go through the required effective three month lockdown. Any attempt to do so would end up with political pressure pushing us towards stop-start. We need a different approach.

What about stop start? The idea that the economy can function for two months and then take a month off seems to me to be for the birds. Many businesses are not viable on that basis. The cumulative human cost for business owners and workers, and especially for the young, is just getting too high. It is a truism that the young well bear all the costs of COVID response despite reaping few of the benefits. We cannot continue to blight their futures with no end in sight. A month’s lock-down, as announced this weekend, seems likely to end with us being in roughly the situation we’re in today, given the embedded momentum in infection rates already in the system.

This leads me reluctantly to conclude that we are left with no option but to tolerate higher adverse health outcomes from COVID than should’ve been the case with competent management of the pandemic. With extreme discomfort I come to a version of the Barrington approach of sheltering the vulnerable as far as possible, and focussing on behaviour changes to deal with the virus longer term, putting more emphasis on personal responsibility, while maintaining fewer restrictions to enable the economy to function as much as possible. Alongside this, as Ben Yeoh has suggested, we need to focus on accelerating vaccines (including roll-out plans), therapeutics, and rapid testing. Let us be clear though that this is an approach that still comes with high costs and risks. There will still need to be some restrictions on activity. The supposed shields for the vulnerable are highly likely to be porous. Deaths will rise and there will be a need urgently to build more NHS surge capacity to deal with the greater level of hospitalisations that will inevitably occur without crippling the rest of the NHS. Widespread circulation of the virus will negatively affect behaviour and so the potential economic benefits may be less than hoped. We may find ourselves becoming more callous as a society in the face of suffering and death. 

The change in approach also feeds through into economic policy. The furlough scheme was predicated on a short, sharp shock followed by a V-shaped recovery. Instead, economic disruption will be with us for the long-term. Zombie companies kept alive by the state are not healthy for the economy. Employees kept on furlough for months will lose skills, confidence, and employability. Economic support should move from companies and jobs to individuals, to enable resources to be reallocated and the economy to adapt. There will be hard transitions, but we cannot freeze the economy for another year. At the same time, there will need to be much better economic support for those who need to shield from the virus. Mental health support for an already isolated elderly population will be at a premium. 

We’ve been put us in a position where we have to choose between the welfare of the young and the lives of the old. It’s going to be extremely painful from here, and it didn’t have to be this way. Other countries have done much better and only a few have done worse. We must not forget that.