Asymmetrical Information - Megan McArdle

02.04.13

The Technocratic Dilemma

What experts don't know can kill a program

Washington is in the throes of a rekindled romance with technocracy: the idea that what is most needed to save the economy and the planet is a bunch of really smart people who have spent their lives studying complicated problems like health care finance or renewable energy.  The last time we had this many fresh-faced wonks geting feted for their deep thoughts was the New Deal.

And who could be against having very smart people studying problems?  What am I, in favor of having stupid, ignorant people run things?  

No, of course not.  But there's a danger in the technocratic class.  Funnelled through an ever-expanding system in which they are paid to sit around and think about things, observe things, and write things, but never actually do things, our information mandarins are frequently totally disconnected from the people and processes they are supposed to rule.  Their knowledge comes from sporadic meetings, usually with members of the regulated class who have been carefully selected as spokesmen.  The screening process by which these people are found selects heavily for personal contacts with the mandarin class, or an ability to speak to a specific agenda (single mother with cancer who wants to buy health insurance; small business owner who will go broke trying to provide it under Obamacare).  This creates problems, as Arnold Kling points out:  

I have many concerns with the technocratic mindset. The one I wish to raise here is that the technocrat has no experience working in the context of an organization. To the technocrat, everything boils down to setting the right parameters and imposing the right rules. Implementation is taken for granted.

Within a business, someone with the technocratic mindset does not get very far. The people who get ahead are the people who can negotiate, build organizational capital, manage projects, and sell.

If you want to see how ineffective technocrats are, I give you as Exhibit A the attempts by the government to prevent foreclosures. The rules and the parameters looked good to the technocrats. The results in the real world were abysmal. That is because nobody had any idea what was involved in actually implementing these policies on the ground.

Before he was a commentator, Kling worked for Freddie Mac, doing things like forecasting default risk.

I think about this often with health care and education.  All of the health care experts that I interview are quite sure that there is a lot of waste and nest-feathering inside of hospitals, waste that could be cut if we could hit on the right combination of regulations and payment reform.  They make a very compelling case.  But none of them can describe how it happens, because they are all academics, not hospital administrators.  As far as I know, no hospital administrator has so far come forward to explain how we could cut costs by 30% without causing quality to suffer.  No one even really understands what is going on inside hospital finance departments: who is cross-subsidizing who, how and when "upcoding" is used to pad billings, and so forth.  We are trying to regulate behavior that we can't even really describe.

One hears almost the same claims about education, with the same lack of understanding of where, exactly, all the extra money we spend has gone.  Yet many of the critiques come from people who work at universities. Professors don't even understand the administrative processes of their own industry.  

But an administrator who came out and said "I can cut costs by 30% would run into three problems".  The first is that this is a hard cognitive leap to make; any turnaround expert will tell you that the folks who run money-losing companies are genuinely, sincerely convinced that there is no cheaper or better way to do it.  The second is that they would be viewed as a traitor to their professional class--I know that Paul Campos, for example, has taken quite a bit of flack from fellow law professors for his scathing critiques of the law school cost structure.  And the third is that they would not, quite, be believed.  That is, people would be happy to hear that there were all these great ways to cut costs.  But when the expert looked at their pet proposal and said, "No, really, you haven't thought this all the way through; we need all those staff and procedures", they would be dismissed as self-interested.

The benefit of technocracy is that it avoids the petty mercenary self-interest of industry players.  (That doesn't mean that it avoids all self-interest, of course; just the particular desire to enrich yourself by directing government contracts, or regulatory benefits, to firms that you work or have worked for).  The problem of technocracy is that it has no understanding of how the industry actually works, in the boring, you-personally-grinding-it-out day by day. Technocrats understand what they can measure, or be told in interviews. But they do not know how the firms actually work, or decisions actually get made.  

The core complaint about the financial crisis was that it was too nice to bankers, because the bailout was run by bankers.  I don't think that the ultra-cynical view (bankers giving goodies to their friends for a payoff later) is correct.  But it's hard to deny the more nuanced claim that people who spent their lives in the banking industry are bound to be excessively sympathetic to the idea that the banking industry is uniquely vital and must be saved at all costs.  It's the regulatory version of the old Woody Allen joke: "I think my brain is the most important organ--but look who's telling me that!"  

As in the Woody Allen joke, the problem is inescapable.  If you want information about how things work, you have to get it from the people in the industry.  But those people will naturally carry with them the belief that their industry is incredibly important, and that its needs are the needs of the American people.  

When you complain about the "revolving door" between government and industry, what you are saying is that government and industry should be completely separate classes, staring suspiciously at each other over a gaping cognitive divide.  Is that better than letting the fox take over the henhouse?  Morally, my heart sings "Yes! yes!".  But if we care about practical results, the answer is less clear.