Arnold Kling  

When Experts Fail

PRINT
New Year's Blogging Resolution... Courts Fail. Use Regulators...

David Brooks writes,


For better or worse, over the past 50 years we have concentrated authority in centralized agencies and reduced the role of decentralized citizen action. We've done this in many spheres of life. Maybe that's wise, maybe it's not. But we shouldn't imagine that these centralized institutions are going to work perfectly or even well most of the time. It would be nice if we reacted to their inevitable failures not with rabid denunciation and cynicism, but with a little resiliency, an awareness that human systems fail and bad things will happen and we don't have to lose our heads every time they do.

Read the whole column. I agree that it is counterproductive to react to institutional failure by denouncing individuals and piling on new layers of bureaucratic irrationality. However, he seems to imply that there is no constructive way to channel anger and frustration at expert failure.

Instead, I believe that a central issue for our time is excessive concentration of political power. One of the causes of this concentration is the ideology of "trust the experts." To the extent that denunciation of expert failures serves to undermine this ideology, I think that is a good thing.

Many in the mainstream media, probably including David Brooks, tend to believe that a loss of trust in experts would be a bad thing. Trust in the experts is a key component of Progressive ideology. It goes along with the belief in social science, and the belief in elite education as providing a natural aristocracy.

As you know, I am very skeptical of experts. I think that experts undertook misguided financial regulatory policies that contributed to the crisis, experts over-rate their ability to steer the health care system toward higher quality and lower cost, and experts are overly confident that the stimulus improved the economy.

What experts are really good at is circling the wagons to protect themselves. Talking points get repeated. Evidence that appears to cast doubt on experts gets picked apart, while evidence that appears to favor the experts is uncritically passed on. Ben Bernanke enjoys protected status. The global warming believers enjoy protected status. Stimulus advocates enjoy protected status. Above all, the label "expert" is carefully restricted to those who can be trusted to respect other experts and to denounce skeptics.

As I have said before, this is the syndrome that David Halberstam identified in The Best and the Brightest as leading us into the Vietnam War. When experts fail, it may be immature to decry their imperfections. However, it is not mature to shrug helplessly. What would be truly mature would be to work for decentralization of power and to work against the concentration of power in the hands of experts.


Comments and Sharing





COMMENTS (9 to date)
Buzzcut writes:

This incident has made it very clear to me that the whole "anti-terrorism apparatus" is Hansonian signalling.

The no-fly lists, the airport security, CIA, FBI, all of it is there not to stop terrorists, but to make passengers feel better about flying.

Also, Brooks is incredibly wrong. All that was needed to stop this guy is to follow some basic rules that are, allegedly, already supposed to be followed. We've all heard that paying cash for a ticket and not having any baggage are red flags, right? Well, this guy paid cash and had no bags.

So why didn't he get any extra scrutiny?

Great Britain denied this guy a visa, so some bureacracy somewhere did its job.

I don't expect perfection from government because when mistakes are made, no one is fired. But it would be beneficial to can Napolitano. She seems to be an extreme example of political correctness in action. Which I think was Cheyney's point.

And, you know, it would make me feel better. ;)

Paul D writes:

It seems Brooks lays the blame on the people for asking too much from government. Yet, he never mentions that it is the politicians--especially when as candidates--who promise to be able to solve each and every problem if elected. They, the elected politicians, deserve the blame for promising what they cannot deliver.

What Brooks fails to realize is that decentralized systems fail far, far, far less often than do centralized systems. The existence of redundancy and failsafes inherent in decentralization make catastrophic collapse far less likely. He is right that we should not rail against the particular people put in charge, but at the system that puts people in charge. The problem is that there is a centralized system, not who is in charge of it.

What people don't seem to understand is the proper role of experts. An expert is good at giving you a big picture, putting things in historical, systematic perspective, etc. They provide big picture information that should supplement the local information each person has and can work with. The expert, lacking the latter information, is inherently incapable of making a good decision -- they do not and cannot have the knowledge to do so.

A good example of this is book, film, and theater reviewers. They are literature and performance experts who are able to provide the average person with information about the work in question. Then those people make an informed decision, based on recommendations from reviewers and friends and family and from viewing ads (the last being the reason I went to see Sherlock Holmes, which, as a literature expert, I highly recommend). Movies are often as good as they are because of this system and its informational feedback loops.

But imagine what would happen if we had some panel of literature, film, and theater experts. What kinds of garbage would we be seeing? No doubt, most of it would be a quite conservative avant-garde (the avant-garde of the first half of the 20th Century is very conservative now -- and is the foundation of the boring, unimaginative postmodernist works now out there) that nobody would want to watch or read.

The literary equivalent of the Federal Reserve would have the same devastating effects on literature as the Fed has had on money and the economy as a whole.

agnostic writes:

Oh it's worse than that -- the trust the experts view goes much farther back into history and pre-history. Witch doctors, priests, etc.

Trying to stamp it out of ideology or psychology will be impossible, so we should focus on institutional changes, like you are, that will dampen its impact on decision-making.

David C writes:

You had me up until the last sentence. If experts cannot be trusted, then you cannot be trusted to have the expert knowledge to know that experts cannot be trusted. This uncertainty about who is right and who is wrong leads me to the conclusion that people should not be so quick to judge others as wrong. The experts who tell me decentralization or markets will work in this particular case are just as likely to be wrong as the experts who tell me central planning will work.

The "mature thing" to do then is to advocate change whenever the current system doesn't work. The "mature thing" is to eliminate possibilities that are certain to fail based on previous experiences, and be wary of changes that are very similar to previous failures. The people who advocate policies whose outcomes are less certain should be promoted even if those individuals advocate policies which are significantly different from your own.

For instance, in health care, it is reasonable to believe that increasing the amount of money paid for by patients and decreasing the amount paid for by insurance will lower costs to the health care system overall. However, it is not reasonable to argue with great certainty that government paying for costs instead of insurance companies will not lower costs. Both options can result in lowering costs without significantly harming benefits depending on which assumptions a person makes. They can also lower costs while simultaneously reducing benefits based on other assumptions. I have no more reason to trust your assumptions than, say, Paul Krugman's.

Joey Donuts writes:

This discussion opened my eyes to something that should stick out as obvious but, at least for me, has not. Back in 2001 after the Sept. 11 attacks, we absolved the airlines of any responsibility (Too important, too regulated to fail?) for the attacks. Furthermore, we established this huge centralized bureaucracy to assure safety for airline passengers and for the citizenry in general. Of course, the government didn't establish any direct responsibilities.

My question now "Would the panty bomber incident have occurred if the airlines still had the responsibility of passenger safety?" In Addition, "Would information about possible terrorists have more availability and integration if the airlines had more responsibility?"

Alex J. writes:

I think the governments policies are preferred by the voter at the margin. To say the same thing, the current policies are what the various politicians and bureaucrats have selected as the policies most likely to keep them in office, given the voters as they are. Better to inconvenience a million travelers to "protect" us against a threat that has already occurred than to allow even a single repeat attack. Even if politicians understand Type I and Type II errors, the voters don't.

Les writes:

It seems to me that centralized decision-making will always fail, because the central experts lack the necessary knowledge and the necessary incentives to make efficient decisions.

Therefore the only acceptable centralized decision-making should be where private experts cannot function - namely where essential public goods are involved, such as:

a)our protection from foreign attacks by our military, and

b)our protection from domestic crime by our justice system (police, courts, jails).

ERIC writes:

David C...

I think in your comment you are missing the point and what is known from history which supports markets in most (all?) cases over central planning.

We know that politics are corrupt, this leads to some sort of failure where we can see how things could be improved but are not, and we are generally left wanting, right?

I don't think you could say the same about markets -- that they are corrupt, or even anywhere near as corrupt as politics are. It might be easier to "hate" markets when they crash because the mispricing is so obvious (after the fact), but doesn't the correction happen a lot faster?

When have you ever seen something worked out quickly once political opportunists get their hands on it?

Comments for this entry have been closed
Return to top