The Virtue of Diversity

One odd feature of the Covid pandemic was the uniformity of the response. There are many independent countries in the world and yet almost of them responded with essentially the same policies, some mixture of lockdowns, mask requirements or recommendations, school closures, bans or restrictions on group meetings. France, Italy, and Japan have very different food, although less different than a century or two ago, but they had very nearly the same Covid policies. China and Australia had stronger versions, some US states weaker, but they were the same policies.

That might have made sense if it had been clear at the time what the relevant facts about the disease were and what policies were best to control it, but it wasn’t. Early on many authorities conjectured that contagion was mainly by surface contact and recommended or required extensive precautions against it. Eventually it became clear that transmission was mainly through the air. Masks were recommended or required despite the lack of solid evidence that they prevented the spread of Covid or similar diseases and despite the fact that medical authorities had advised against their effectiveness in previous epidemics and early in this one.¹ Lockdowns had not been recommended in advance planning for dealing with a pandemic but were almost universally adopted, possibly on the principle of “something must be done. This is something.”

The one serious attempt from within the medical profession to propose an alternative approach was the Great Barrington Declaration, authored by three prominent epidemiologists, one each from Harvard, Stanford, and Oxford; they proposed that, since mortality rates were serious only for the most vulnerable part of the population, principally the old, they should be isolated and the disease allowed to spread through the rest of the population until enough had been infected, making them resistant to reinfection, for herd immunity. The proposal was ferociously attacked by supporters of the policies being followed.

Tedros Adhanom Ghebreyesus, the Director-General of the World Health Organization, warned against the idea of letting the virus spread in order to achieve herd immunity at a 12 October 2020 press briefing, calling the notion "unethical". … "scientifically and ethically problematic".

Concerns about the declaration had been issued on behalf of the British Academy of Medical Sciences by its president, Robert Lechler, who similarly described the declaration's proposals as "unethical and simply not possible". Martin McKee, professor of European public health at the London School of Hygiene & Tropical Medicine, compared the declaration to "the messaging used to undermine public health policies on harmful substances, such as tobacco". (Wikipedia)

The uniform approach and the hostility to alternatives are evidence that science, at least medical science, has become a monoculture. Epidemiologists in France, Italy or Japan depend for their professional status, funding and employment on how their work is viewed by the international community of their profession. Government health officials such as Tedros Ghebreyesus, whose previous positions were as health minister and then foreign minister of Ethiopia, get their views, including their view of what medical policies to impose on their citizens, from the same place. Heretics, people in the profession who disagree with current orthodoxy, can expect a hostile reception, as the authors of the declaration discovered.

We do not know for certain whether the policy recommended in the Declaration would have worked better than the policies actually followed. We do not know because no country tried it.

One country came close however: Sweden. While it did not, perhaps unfortunately, follow the Declaration’s policy of isolating the vulnerable, it also did not impose lockdowns or social distancing requirements or require masking.² It was widely predicted that the result would be catastrophe; it wasn’t. Judged by excess mortality, increased mortality during the pandemic relative to its normal level, Sweden suffered less from Covid than most comparable countries.

It also suffered less from policies adopted to deal with Covid. As best we can tell from the Swedish experience the costs countries imposed on themselves, shutting down schools and severely limiting mobility, were an own goal.

A global monoculture, in health and many other things, is probably a bad thing. If it gets the right answer to something the result is a little better than if different countries tried different things, but once it becomes clear which answers work the less successful can imitate the more successful. If it gets the wrong answer, as in the case of Covid it pretty clearly did, it may never find out or, as in that case, find out too late.

For another example of the virtues of diversity, consider the issue of how poor countries are to get rich. The orthodox answer in the post WWII period was central planning. Since it was widely believed that the Soviet Union had done it and was doing it successfully, India and many other poor countries adopted similar policies. If Taiwan, Hongkong, South Korea and Singapore had listened to the advice of economists from Harvard and the LSE they would have tried it too; fortunately they did not. All four are now at developed world standards of living. India is not.

The result was that countries trying to develop by government central planning have now largely abandoned it, including India and even Russia. They are much farther from laissez-faire than I would like, as is the US, but considerably closer than they were seventy years ago.

Contra Subsidized Science

The prospect of domination of the nation's scholars by Federal employment, project allocations, and the power of money is ever present and is gravely to be regarded. (Dwight Eisenhower, Farewell Address)

Donald Trump says that he intends, with the assistance of Elon Musk, to sharply cut federal expenditures. The fact that Trump says he will do something is at best weak evidence that he will do it but Musk has an impressive record of accomplishing difficult things, although usually not as fast as he says he will, so some reduction in government spending might actually happen. One possible target is government subsidy of scientific research, a particularly tempting target for this administration given that much of the subsidized research has been in support of causes, such as gun control and controlling climate change, that conservatives are skeptical of.

The obvious approach to arguing for or against government funding of scientific research is to compare the cost of the funding to the value of the research funded. In my view that is the wrong place to start. The prior question is whether government funding, cost aside, results in better or worse science.

In my view the answer is worse. While some of the research subsidized is doubtless worth doing, the costs of having government subsidy of research are greater than the benefits even ignoring the budgetary cost — because the existence of government subsidies makes science work less well, be less likely to produce true conclusions, more likely to produce false ones. That might not be true of government subsidies were modest, only one among multiple funding sources of similar size, but they are not.

A single source controlling a large fraction of the resources for funding scientific research produces a scientific monoculture. Covid again gives an example. Jay Bhattacharya, one of the authors of the Great Barrington Declaration, had an earlier and related Covid heresy.

Dr. Bhattacharya’s tale begins on these pages with a March 25, 2020, op-ed titled “Is the Coronavirus as Deadly as They Say?” Co-authored by Eran Bendavid, a fellow professor of medicine at Stanford, the article argued that many asymptomatic cases of Covid were likely going undetected, making the disease far less dangerous than authorities were claiming.

“That is when the attacks started,” Dr. Bhattacharya, 56, says in a Zoom interview from his office in Palo Alto, Calif. In April 2020 he and several colleagues published a study that confirmed his hypothesis. The prevalence of Covid antibodies in Santa Clara County, where Stanford is located, was 50 times the recorded infection rate. That, he says, “implied a lower infection mortality rate than public-health authorities were pushing at a time when they and the media thought it was a virtue to panic the population.” His university opened a “fact finding” investigation into him after BuzzFeed made baseless charges of conflict of interest. “This was the most anxiety-inducing event of my professional life,” he says. (WSJ)³

After the declaration came out the attacks on it came from, among others, Francis Collins, head of the National Institutes of Health, which spends about fifty billion dollars a year funding medical research. A bad man to offend if you are a researcher.⁴

My own experience of the problems with a scientific monoculture occurred twice, with nearly fifty years between them, neither involving medical research. The first was the population scare of the nineteen sixties, when virtually all respectable opinion — the notable exception was Julian Simon — held that population growth was a serious threat to human well being, especially in poor countries, an emergency that required drastic action. I wrote a piece at the time trying to estimate the next externality from having one more child, concluded that I could not sign the sum, that if you looked at positive effects as well as negative it was not clear whether it made other people better or worse off. That was not the orthodox opinion.

The second was the current climate change scare. From my standpoint as an economist it raised the same issue: What is the net effect on human welfare? I concluded that, again, there were benefits as well as costs and the size of both was sufficiently uncertain that I could not sign the sum.

That is not the current orthodoxy so I looked into the scientific literature and concluded that it had been badly corrupted by the fact that everyone, including the federal funding agencies, knew what answer they wanted, ignored or minimized benefits, exaggerated costs. For the defense of that claim see my posts on the subject.⁵

P.S. Graphs, from the article quoted in fn 2, of excess mortality by week showing the pattern among Scandinavian countries — higher in Sweden in the first year as Covid spread, much lower thereafter.

---

My web page

Past posts, sorted by topic

A search bar for past posts and much of my other writing

1. The evidence was described in detail by Scott Alexander early in the pandemic: “Face Masks: Much More Than You Wanted To Know.” His conclusion was that while there was no really solid evidence for their effectiveness it was probably worth wearing surgical masks in high risk situations.

2. “In Sweden, restrictions primarily relied on voluntary measures and recommendations, alongside limitations on the size of public gatherings, prohibition of visits to the elderly and the implementation of distance learning for individuals aged 17 years and older. Denmark, Finland and Norway, in contrast, opted for more stringent lockdown measures with closing of schools, kindergartens, work places and many other services.” “Excess mortality in Denmark, Finland, Norway and Sweden during the COVID-19 pandemic 2020–2022”

3. Bhattacharya’s account of the attempts to suppress his views.

4. Jay Bhattacharya has now been proposed by Trump as Collins replacement. For his view on what is wrong with the NIH, see the WSJ article.

5. A summary of the argument and two examples of indefensibly bad work, biased towards exaggerating costs and neglecting benefits, taken seriously in the field.