The promise of a “final” end to lockdowns in the spring of 2021 is the kind of hyperbole we have come to expect about new products and policies. The Oxford University vaccine may work; it may even be delivered effectively. Meanwhile, Covid-19 is still around, the UK government is extending lockdown for large parts of the country and effective protections are still being ignored, at grave cost.
From the start of the pandemic, the policy choice in Europe has been presented as a trade-off between lives and livelihoods. Since priority was (rightly) attached to saving lives, the livelihoods of large sections of the population have been sacrificed, with income support for workers in the form of long paid holidays called furloughs, and loans and grants for business prevented from trading. As a consequence of widespread business distress and associated redundancies, European countries face a huge problem in reopening their economies in the wake of the pandemic. Forecasts suggest that the UK economy will be 6 per cent smaller in 2021 than in 2019, and unemployment at 7.5-8.0 per cent – roughly double its pre-crisis level.
The experience of East Asia shows the choice in Europe was, and continues to be, wrongly presented. Countries such as China, Japan, South Korea and Taiwan found a way of protecting both lives and livelihoods. Their death rates per head of population have been much lower than in Europe; their economies have barely contracted; and they are forecast to be larger, not smaller, next year. Their secret was an effective system of testing, tracking and quarantining. The question is why such a system was not adopted as the first line of defence in Europe. This is not just a historical question. If it were technically feasible in March it is even more so today. It is still not too late to avoid future economic and social damage, even though most of the damage already inflicted cannot be repaired.
Virologists identify the nature of an epidemic, while epidemiologists study the way it spreads. It was an almost-forgotten English doctor, Ronald Ross, who first developed a predictive model of malaria transmission, which was later generalised as the SIR (Susceptible, Infected and Recovered) model of contagious disease epidemics. His successors concluded that a viral infection ends when the virus runs out of hosts in which it can reproduce itself; that is, when the population develops “herd immunity”.
Central to the SIR model is the “R rate”, the rate of infection. The statisticians work out a data-based prediction of the rate of infection in a susceptible population. Politicians, advised by medical scientists, as well as by health professionals who tell them about medical capacity, and economists who tell them about strains on the economy, decide policy. Their skill lies in assessing political reactions to their policies.
“Mass protection” and “focused protection” have been the two main political responses to the models produced by these experts. They have been presented as polar opposites, but are essentially two variants of the same response.
Think of them in terms of breaking the S⇀ I chain. If a virus is circulating in a susceptible population, the effect of a mass lockdown (semi-isolation) is to reduce the transmission rate and thus the overload of medical services. As soon as the lockdown is eased, the spread rate picks up, but because the susceptible population has been reduced by recovery or death, R continues to decline (with smaller spikes) to the point when normal life can be resumed. The treatment is effective, but the economic cost is horrendous.
The alternative of focused protection, as proposed by the Great Barrington Declaration and others, is a restricted application of mass protection. It aims to remove the most susceptible population (the old and those with underlying health issues) only from the path of the virus. It thus makes possible a more normal life for most people and, in principle, reduces the pandemic’s economic costs. However, the Great Barrington Declaration’s signatories never made clear how a sizeable section of the population can be securely “shielded” from the rest. For this reason, it is supported by very few public health professionals. Sweden has shown that any exponential growth of the virus in a healthy population is bound to spread to the vulnerable. The country’s death rate was similar to that of mass protection countries; and the effects on its economy also similar.
The bird that never flew in most of Europe (Germany was a partial exception) was “targeted protection” based on digital testing, tracking and self-isolating. Such systems exist, and have been rolled out in the UK as well as in other European countries, but never as an alternative to mass or focused protection. Indeed, the UK government decided early in March to stop contact tracing and impose a mass lockdown.
Digital or targeted protection works not by locking down pre-determined blocks of people but only those individuals and clusters who test positive. Every individual spreader and his or her contact group is isolated straight away. As a result, normal life, with a few sensible precautions (masking, distancing, hand washing) can continue as before. No lockdowns, total or partial, are needed. The South Korean economy contracted by 2.74 per cent between March and September; the UK economy by more than 9 per cent. (Taiwan’s economy actually grew.)
South Korea is the pin-up story for digital protection. Only 289 out of 51 million inhabitants are known to have died from Covid-19 between February and July 2020; in the UK it was 44,600 out of 66 million. South Korea avoided a national lockdown; the UK has had two.
South Korea’s success is generally attributed to early identification and management of cases, clusters and contacts. Every time an individual tests positive in the densely populated country, 100 contacts are identified by location and payment data and then tested. The country has also skilfully controlled its borders (as has Taiwan and China) and it has sensible precautions such as masking and selective social distancing and localised temporary lockdowns. Short of a vaccine, there is no single efficacious system of protection; the difference lies in the weight given by policymakers to the different elements of protection. The national lockdown method is by far the most expensive and, on the evidence, the least efficacious.
Electronic shielding was never central to the European policy response. The reasons are complicated, but essentially due to a mixture of unfamiliar technology, logistics and politics. Consider these in turn.
The technology behind digital shielding is quite simple. It consists of a mobile app on every mobile phone. This is similar to the English NHS Covid-19 app, but without the need for a QR code check-in.
An effective system works by identifying those users who have been exposed to someone else who has tested positive. The system asks them to be tested within 48 hours and for them to quarantine in the meantime. In a lighter version of the scheme they can be asked if they have symptoms and only then will they have to isolate while they wait to get tested.
An efficient test and track system of this kind would obviate the need for either mass lockdowns or continuous mass testing. It would concentrate on the crucial infection-spreading interactions, allowing optimisation of testing and isolating to those who really are at risk. Thus the rate of transmission is curbed not by locking down whole populations for months or testing everyone, but by briefly quarantining small clusters. You would avoid the ludicrous need to lock down a million people because infection had spread in a cluster of a hundred. The economic effects are incomparably milder.
Such a system was available from the start of the pandemic. But it was never central to stopping the spread of the virus in Europe. Indeed, it was presented as a minor separate measure, subordinate to lockdown. There was no effective lobby for its priority. At no point did those familiar with the technology say to politicians: “Why not make this your main weapon?” Nor did the media take up the cause. Two obstacles seemed insuperable to mass implementation: logistical and political.
To obviate or minimise the need for other measures everybody has to have the app. The scale of roll-out depends on medical capacity to test and track and user capacity to understand what is required of them. Lack of medical capacity seems not to have been the decisive constraint, except in the UK.
A more important barrier to roll-out was the large fraction of the “digitally challenged” in the population – those who didn’t know how to install the app or who had obsolete telephones. (The same problem is faced when any service is put online.) This is where a big governmental effort was needed and was not forthcoming. The government might have provided free up-to-date phones and free help in setting them up. Along with video tutorials, FAQs and other online resources every technology shop staff could have been trained and subsidised to offer free assistance to the digitally disadvantaged, while ensuring that the app was designed according to user-centred techniques.
For those with an obsolete mobile phone, who refused to replace it, but still wanted to take part in the test and track scheme, an alternative device such as a digital bracelet or necklace could have been offered to track the users, as have been trialled and deployed in Hong Kong, Bulgaria and Bahrain. Mobile phones or bracelets need to be with the user at all times when outdoors and infringement would need to be pursued by law, just as lockdown rules, mask-wearing and social distancing now are.
Enriching the population technologically in double-quick time would have cost a significant amount of money, though a tiny fraction of the cost of the lockdowns. And it would have offered a valuable opportunity to use the crisis to achieve an enhanced overall level of digital competence, with favourable knock-on effects on people’s employability.
Nor was the technical processing of data quickly the problem it has been made out to be. These are very simple and light data (a few bytes, probably not even kilobytes). The ghastly delays in processing information in the UK, revealed by the Panorama programme “Test and Trace Exposed“, was the result of organisational ineptitude, not technical difficulty.
The central point is that the technology was available from the start and the logistical problem of wiring everyone up and getting up-to-the-minute information could have been solved had attention and resources been directed to solving it. The reason they were not was political.
As European failures have demonstrated, if you opt for digital shielding you can’t be half-pregnant: failing to make sure that everyone follows the rules of the scheme would not be a partial success. It would be a complete failure.
This means two things: authorities need to make sure that the system covers the entire population; second, that every test returning a positive result is synced on to the system, and rapidly processed accordingly. Once this is done, all the new tests performed by the health authority are addressed to the close contacts (as identified above) of those who are positive. The colossal waste of mass lockdown is avoided.
Standing in the way were, and remain, serious concerns about privacy and enforcement.
The following incident in South Korea in May 2020, as reported by the Mail on Sunday, produced national headlines: “Night clubs in Seoul have been linked with 119 coronavirus cases nationwide after a ‘super-spreader’ visited a number of bars in the Itaewon district. [The] 29-year-old man, who is thought to be at the epicentre of the latest cluster of cases, was tracked by authorities… and tested positive for Covid-19.”
It turned out the bars were in the gay district, so, understandably, most of the client names and addresses registered by these clubs and available to the health and law enforcement authorities were false. And the fact that electronic tagging is widely used everywhere for tracing movement of criminals and that CCTV cameras might be used to track people’s movements increases the feeling that Big Brother is watching you.
The charity Privacy International argues persuasively: “Unprecedented levels of surveillance, data exploitation and misinformation are being tested across the world [in the light of Covid-19]. Many of those measures are based on extraordinary powers, only to be used temporarily in emergencies. Others use exemptions in data protection laws to share data. Some may be effective and based on advice from epidemiologists, others will not be. But all of them must be temporary, necessary and proportionate. It is essential to keep track of them. When the pandemic is over, such extraordinary measures must be put to an end and held to account.”
The key question is how to secure the efficient population protection offered by digital shielding while keeping the information it needs anonymous.
The Italian app Immuni provides the answer. Not only is it not reliant on a centralised database to trace Covid-19, it also does not rely on venue check-ins, preferring closer range Bluetooth connection with other devices. This is far more accurate (in that it drastically reduces the risk of false positives) and grants full anonymity. The system doesn’t need to know my name, nor where the interaction took place. It just knows the date and that the length of the interaction was long enough and the distance was close enough for people to be infected by me. My identity is not disclosed at any step of the process, nor the places I have been. My anonymity will only be breached if someone has to visit me because I have not followed the rule of disclosure.
It is hard to evaluate the argument that Asian culture is more hospitable than European to tracing because Asians value personal liberty less, and the “public good”, more. Contemporary Europeans associate liberty with privacy, but as Hannah Arendt pointed out in her 1958 book The Human Condition, the right to be protected from the public gaze cannot ever be more than relative.
Enforcement concerns are easily addressed: anybody can opt out of the scheme and embrace the alternative solution in the form of a personal lockdown. The prospect of this would itself be an incentive to become digitally literate.
In our kind of society, people would probably need some financial incentive to join the scheme (such as free up-to-date systems or cash payments). Those who join the scheme but who don’t follow the rules would face fines or policing through phone calls and other checks.
Digital shielding presupposes some degree of scrutiny and enforcement of its rules. But this has to be weighed against the prohibitions of normal life (as well as enforcement of these prohibitions) entailed by mass lockdowns. The lesson of the lockdowns is that the liberty to do what I want requires some regulation of that liberty to ensure it does not harm others.
Angela Merkel said people “needed beds to be full before they would accept a lockdown”. A further implication might be that people had to experience a full lockdown before they would be ready to accept the degree of intrusion involved in digital shielding. This explains why the test, track, isolate scheme has been implemented at best as subsidiary to full or partial lockdowns. This political calculation may have been right for contemporary Europe; it was clearly not so in many Asian countries.
There are certainly legitimate concerns about limiting personal freedoms. It’s only in the light of much stricter limitations, such as those that come with a full lockdown, that one can put things in perspective and see digital shielding as a desirable alternative. In the end, the aim of this solution is to offer an alternative to general lockdown in order to avoid the ascending economic, social and medical costs for the community.
The world economy was hugely damaged; trillions of pounds, dollars and euros were spent protecting communities, and hundreds of thousands of unnecessary deaths have been caused by want of effective deployment of a system of testing, tracing and quarantining. The technical requirements for such a system existed from the start; its effective deployment would have taken some time in Europe, so that the earliest lockdowns were probably inevitable. But by May or June, there should have been no need for further lockdowns. The reason digital shielding was so slow in coming is not technical or epidemiological, but political. With some notable outliers, most European citizens have preferred the anonymity of lockdown to the individual scrutiny of testing and tracing. Given the absolute priority attached to protecting lives, the policy was therefore set. But, despite optimistic claims for a vaccine, Covid-19 is not over. It is not too late to change.
Robert Skidelsky is the author of a three-volume biography of JM Keynes, a crossbench peer and emeritus professor of political economy at the University of Warwick. His most recent book is “Money and Government: The Past and Future of Economics”.
Massimiliano Bolondi is a technology adviser.