Random thoughts

Overarching - Free people with free communication and discourse will, in the face of novel uncertainties, more likely and more quickly reach a better solution sooner. They do this by synthesizing mass information across conflicting objectives, priorities, interpretations and risk tolerances through the marketplace of ideas and the free market of risk-based commerce. Crucially, they make a balance of trade-offs across multiple legitimate systems. Statists, central planners, journalists, bureaucrats, and academics tend to be single-issue focused, maximizing rather than optimizing outcomes. They are system switchers rather than system optimizers.

The Absence of Expertise in the face of Novel Complexity Across Multiple Knowledge Domains

One of the consequences of the global pandemic is that it undermines many status systems and institutional credibilities. In the span of five months we have seen groups of the Mandarin Class shift from "Masks don't help" to "Masks are critically necessary". We have seen the switch from "It won't be worse than the flu" to "It will be existentially destructive" and then beginning to edge back towards "It's worse than the flu but not greatly worse." We have seen the switch from "We will reopen as soon as the curve has flattened" to "We will reopen when there is a vaccine." We have seen the switch from "Viruses don't respect borders, no need for travel restrictions" to "Don't you dare move beyond the borders of your home."

The only thing that has been consistent has been the excessive degree of confidence with which these statements are made; the total absence of humility or self-awareness; the extreme disdain for alternate views; and the apocalyptic expression of these empirically thread-bare opinions.

In some cases, the opinion of one group of Mandarin Class advocates displaces the opinion of another group of Mandarin Class advocates. In others, the Mandarin Class simply changes its mind (and then denies that ever held the former opinion.)

And all along the broad public has indicated a substantial and generally sustained support for common sense measures - physical distancing, hand-washing, personal hygiene, mask-wearing, importance of protecting the most vulnerable, not overwhelming hospitals, avoidance of mass-gatherings, and targeted and narrow quarantines for those of proven contagion or risk. All in the face of the ebbs and flows of experts and policy makers.

The reality is that in the face of the unknown, opinions are far more equal than they usually are. Experts in narrow domains facing novel conditions across broad domains of complexity have no greater authority or pertinent knowledge than most people.

Obscure Model Forecasts Are Not Equivalent to Future Facts - A Reduction in Opinion Inequalities

This morning, listening to NPR, they were trumpeting the recent results of a Harvard-NPR joint study which found that people should still be required to remain in isolation for some unspecified further time. On what grounds?

Well, they had built a model based on unspecified assumptions using uncertain data and applying undescribed algorithms and it had forecast dangerous consequences.

A nothing-burger.

You need to specify the model design, the key assumptions, and the quality of the data. You also should run some forecast sensitivity tests to the assumptions. Without that information it is impossible to glean whether the model forecast has merit. And given the dreadful performance of most expert forecast models so far, the default assumption has to be that the model will overstate dreadful outcomes.

In a circumstance of novelty and uncertainty, there is little room for unsupported opinions. It is driving the Mandarin Class crazy. They demand status and respect and yet their opinion now has near equal validity to everyone elses.

I am not saying that there is no value to expertise. I am saying that experts over-claim knowledge and certainty into conditions where it does not exist and thereby undermine their own credibility, status, and prestige. And this is especially common in novel complex systems which cross multiple knowledge domains.

Monomaniacalism is Achilles Heel of the Mandarin Class

What often appears as intolerant, postmodernist, critical theory, statist marxism is, I suspect, frequently really just a manifestation of Mandarin Class Monomaniacalism. A singular focus on maximizing a single attribute.

At the beginning, we focused on bending the curve. The curve is now well bent or well on its way to being bent. Now the goal posts are being shifted and people are people told they need to remain in isolation at the threat of jail (which is now open for new prisoners since all the violent felons have been released). Is this statist power grabbing? Sure looks like it.

But I think that frequently there is something else going on. Not all academics, journalists, social justice advocates, government deep staters or politicians are inherently bad marxists. No matter how much their actions can make it seem that way.

I think, rather, that they share a common attribute of monomaniacalism which is common among intolerant, postmodernist, critical theory, statist marxists without it being necessarily determinative. Being monomaniacal will make you appear similar to an intolerant statist without your necessarily being an intolerant statist.

ADD-like, the Mandarin Class switch from fad to fad in each case treating them as isolated from one another and engineering an ideal and maximizing outcome instead of recognizing emergent order and seeking optimum outcomes which arise from make optimum trade-offs across goals as opposed to single goal maximization.

The NPR experts referenced above, when queried on whether there was a trade-off between beating the Covid-19 virus and restarting the economy, denied that there was a trade-off. People are the economy, we have to defeat the virus before we can reopen the economy.

As Nassim Nicholas Taleb describes them - Intellectual Yet Idiots. They are monomaniacally focused on beating the virus. Good goal. But not the only goal. People have to eat. The economy has to function. Kids need to be educated. Laws need to be enforced. And there are further trade-offs. Quarantines are driving up overdose deaths and suicides. You cannot maximize one goal without taking other legitimate goals into account.

Which is in part why the experts and Mandarin Class come off as arrogant anti-democratic Marxists. If you are intellectually focused on a single success goal (defeat the virus) you appear to be disregarding all the other legitimate opinions held by other experts and by the common wisdom of the public. And there is a common public wisdom which is too frequently disregarded based on class and credentials.

Academics, journalists, government bureaucrats and social advocates tend to monomaniacal switchers. They form a fixed opinion on how to achieve one set of measured outcomes on one single issue. Then they turn their attention to another single isolated issue and form fixed opinions on that as well. And then another.

Meanwhile, everyone else has to constantly juggle the synthesis and integration of all those desires and goals and outcomes and cost and achieve some sort of balance or harmony of trade-offs which might be optimal but can never be maximal. You can't have everything. All systems with limits entail trade-offs.

The public, effective leaders, business people - these categories tend to be much more accustomed to making unpleasant trade-offs. There is even a tacky term for the process - satisficing. Not everyone will be delighted with an outcome from changes to a complex system but you need to achieve a minimal percentage of the stakeholders who can be satisfied with the outcome.

Where No One Can be Certain of Anything, Revealed Preference is King

There are the experts and the Mandarin Class whose opinions in the face of the unknown and uncertain count for not much more than those of the public. Which prompts the question, how do we know the will of the public?

Surveys are virtually useless. Like forecasting models, they will say what you want them to say based on how you word the questions and exactly whom you sample. And even when you are playing a straight hand, your own unrecognized biases and/or ignorance can shape the questions in unexpected ways and your sample can end up being unrepresentative in unknown ways for unknown reasons.

Crowds, and especially mobs, are rightly to be respected even to the point of feared. But like any beast, the collective creature also has strengths not present among the constituent components. There is a wisdom of the crowd which has to be acknowledged and respected even though sometimes it will be wrong. But as Tetlock and others have shown, experts will, on average, be more frequently wrong than a sample of informed non-experts.

It has likely been appropriate to buy time through mandated social-distancing and quarantining. But now that we are approaching the demonstrated limits of the willingness of the population to continue to pay the large, real, and unevenly distributed costs of those restrictions, how do we determine the will of the public as to what to do next?

Do what we should have been doing more of all along. We are a federal republican democracy. Quit trying to dictate to citizens and instead learn from them. Broadcast (rather than all the efforts to select or repress) information and lessons learned. Then let people decide. Throw things open and see what they do. What are their revealed preferences? Crucially, what the public does is not ipso facto what should be done. It is more likely to be right on average than the opinion of experts in unknown and uncertain systems. But it is not necessarily right. Regardless - it is important to discover the revealed wisdom of the public. If what is being asked by the experts and Mandarin Class is too divergent, then a bridge must be discovered, otherwise the system comes awry.

In the Swedish experiment, they have consciously elected to minimize the number of forced lockdowns. School is not cancelled, restaurants not closed, etc. They have left the economy open to function. The logic is, in part, that the virus will exact a fixed toll. If the healthcare system can absorb the spike, then they have elected to sprint towards herd immunity. The faster the better. The same number of people will die but over or shorter or longer duration. And the shorter the duration, the less catastrophic damage to the economy.

It is like the cumulative pain of removing a bandaid. One quick yank or a slow painful pull?

It is too early to tell whether the Swedish yank or the New York pull will prove superior. Too early to know but not too early to have passionate opinions.

One thing that has been overlooked in all the arguments is that the Swedish laissez faire approach was not compulsive. No one was forced to shut down. But no one was forced to stay open. The Swedish government provided the facts and then allowed people to make their own decisions. There has been no lockdown but a lot of businesses have closed. Many people have elected to self-quarantine. It is an experiment in emergent order and the wisdom the crowds.

We don't know what the results will be but at least we need to know what the experiment actually is.


Related Articles

Treat the patient, not the virus by David Goldhill

Who is the “public” in “public choice”? by Henry Farrell

On reopening, Robin Hanson suggests a political economy hypothesis by Tyler Cowen

Why Openers Are Winning by Robin Hanson