Who needs experts?

During the Brexit referendum campaign, Michael Gove, who at the time went by the Orwellian job title of Secretary of State for Justice, said in a TV interview with Sky News that “I think the people in this country have had enough of experts.” When quizzed on this, he went further and suggested that “these people are the same ones who got consistently wrong what was happening.” It was a somewhat off the cuff remark but it instantly caught the mood of the times. Recall the summer of 2016 was the time when British politicians exaggerated the benefits of Brexit. It was also the first time that the world became aware of just how little regard Donald Trump has for the truth – those inconvenient pieces of evidence that suggest that one’s prejudices may not always be right. But this attitude has had a hugely damaging effect on the quality of policy debate, and nowhere is this more important than in the debate over Covid-19.

As an economist who does not always get every forecast right, this argument is sometimes thrown in my direction. After all, what use are experts if they are not infallible? I have explained on numerous occasions that economics is not a predictive discipline – economists cannot foretell the future – but in the public mind that is what we do. When it comes to matters scientific, the public holds the view that there is a single body of evidence which represents the truth and anything that is not inside this envelope of perceived wisdom must be false. But just as with economics, the public perception of science is not wholly true. Scientific conclusions on issues such as epidemiology depend on a host of input assumptions, which if changed can result in very different outcomes.

There has been much debate in recent weeks about the size of the R (or reproductive) number associated with Covid-19. As we are all now aware, an R number in excess of one implies the rate of infection is rising. It is extremely difficult to measure R in real time and estimates for the UK in the range 0.7 to 0.9 imply a margin for error that puts it dangerously close to one. It also varies by geography so if it is higher than one in some places, this runs the risk of a second wave of Covid-19 cases. The R value is calculated using data such as hospital admissions, intensive care unit admissions and deaths. However if the cause of death is wrongly attributed, this will impact on estimates of R. Since age is also a factor in deaths from Covid-19, the overall R value may be biased upwards if we do not take sufficient account of this.

Those responsible for making these calculations are acknowledged experts in their field, and are aware that their estimates are subject to a margin of uncertainty. The problem then becomes one of deciding whether the estimates form a sufficiently strong basis for the decisions made by policymakers, who ultimately have to carry the can. Or to put it another way, is the science sufficiently robust to support some of the recent policy actions? 

With the UK having suffered the second highest recorded death rate from Covid-19, questions are increasingly being asked of the policies adopted over recent months. Quite how an island has significantly more deaths than other European countries which share land borders suggests that there have been policy mistakes. The government has consistently stated that it is “following the science,” therefore either the advice was flawed or the policy implementation was. 

The most obvious question is why the UK did not impose some form of border controls – after all, they were employed by nearly all other European countries? As it happens, the minutes of the Scientific Advisory Group for Emergencies (SAGE) for 23 March suggested that “closing borders would have a negligible impact on spread.” Yet on Monday the UK introduced a quarantine regime in which people entering from overseas are now expected to self-isolate for 14 days. This appears to be somewhat self-defeating since the rates of infection are now lower in other European countries than the UK, and it would have made more sense to implement these restrictions in March. Nor is it clear how the policy is enforceable since there is no guarantee that the address people give on the form which they are legally required to fill in is necessarily where they intend to stay. 

The policy on schools closure has been similarly muddled. The SAGE view on 16 March was that “school closures constitutes one of the less effective single measure to reduce the epidemic peak.” Two days later, “SAGE reviewed available evidence and modelling on the potential impact of school closures. The evidence indicates that school closures, combined with other measures, could help to bring the R₀ number below 1.” On the basis that the government believed the epidemic to be under control, it announced that primary school pupils would return to school at the start of June. But many local authorities, and indeed parents, questioned whether the policy was safe and many children simply did not show up at school. With attendance rates last week running at just 7% the government this week conceded that its plan was not workable and backtracked on its school reopening policy. 

Then there is the vexed question of the lockdown. On 18 March, SAGE concluded that there was a case for a lockdown in London but “measures such as restricting public transport … would have minimal impact.”  Five days later, there was a stronger case for “reducing contact with friends and family outside the household, and contact in shops and other areas.” One of the attendees at this meeting was the epidemiologist Professor Neil Ferguson, who yesterday told a committee of MPs that “had we introduced lockdown measures a week earlier, we would have reduced the final death toll by at least a half.” Yet the SAGE minutes do not suggest that the scientific consensus was pushing for an earlier lockdown. Nor is there much evidence during the early stages of the debate that they paid much attention to the problem of shielding the older, more vulnerable members of society despite the fact that some estimates suggest “more than half of England’s coronavirus-related deaths will be people from care homes.”

It is easy to be critical of a government which has presided over the highest number of Covid-19 deaths in Europe, and its communication strategy has been muddled and inconsistent (viz. the Dominic Cummings affair). However, its claim to be following the science does appear to stand up to scrutiny – at least to some degree – as a cursory glance of the SAGE minutes suggests. The government has made errors and will ultimately be held accountable for them (we hope). But the scientific advice has also flip-flopped. This is not to say that the SAGE committee was wrong – it was acting on the best information available at the time, and like all good scientists members changed their views in the face of new evidence.

Whilst the experts may not get everything right, they do get more right than they do wrong. Deductive failures do not mean that we can do without experts – Michael Gove was wrong about that. But next time you hear the media calls suggesting that economists’ forecasts are always wrong remember that the so-called hard science disciplines do not always get it right either