Arnold Kling  

Cognitive Hubris

PRINT
John Cochrane's Talk at Hoover... Random Family and Inequality...

My latest essay weaves together Daniel Kahneman and Jeffrey Friedman.


Suppose you were to ask yourself how well you understand the world around you. How accurate is your map of reality?

If you interrogate System Two, it might reply, "There are many phenomena about which I know little. In the grand scheme of things, I am just blindly groping through a world that is far too complex for me to possibly understand."

However, if you were to interrogate System One, it might reply, "My map is terrific. Why, I am very nearly omniscient!"

Kahneman talks about the "planning fallacy," in which projects always seem to cost more and take more time than expected. I would say that this affects public policy. Look at how often policies are less effective and have more adverse consequences than expected.



COMMENTS (12 to date)
Randy writes:

I would add to this what I think of as "manager time". That is, managers have a tendancy to think of something as done once they have done their part - which generally consists of putting something on paper and handing it over to a tech. They just don't get "problems", and the time it takes to handle problems, because the problems aren't in the plan.

david writes:

I am always surprised you don't get more explicit pushback from economists who believe more in incentives.

a different Randy writes:

It is a great essay, and I intend to convey parts of it to my students.

Here is my challenge to Kling and to libertarians more generally, though: suppose "spontaneous order" or something else has generated institutions which, for hundreds or thousands of years, have been foundational for civilization. Even if there are individuals pushing for the destruction of those institutions (or minority groups which harness the power of government to destroy those institutions), perhaps these pushes should be resisted. If we should have humility with respect to grandiose schemes to improve society, we should have equally well have humility with respect to the institutions which have arisen that, so far, have been helpful to society. (To make the point more sharply, we may not understand, for example, the role that traditional marriage plays in civilization.) How do you *know* that these institutions are optional, or that their destruction won't doom civilization?

Is there a double standard here?

John T. Kennedy writes:

@different Randy

"If we should have humility with respect to grandiose schemes to improve society, we should have equally well have humility with respect to the institutions which have arisen that, so far, have been helpful to society."

Individuals should bear the costs of their own biases. Unfortunately some institutions have arisen which allow allow some to impose those costs on others.

M. Flood writes:

Don't forget the other factor in projects taking longer than estimated: Lying. People who want to get something done will always provide the lowest feasible estimate of completion time whether they believe it or not.

Randy writes:

@M. Flood,

Yes indeed. I deal frequently with contractors where my tasks are dependent on the completion of theirs. I have learned from experience to take the technicians aside after the meeting and ask them, "So, when do you really expect to be done". The answer is nearly always a grin, and a fairly accurate answer - and almost never the same answer as was given in the meeting.

Glen Smith writes:

Randy,

The biggest form of lying I've seen is a result of not appropriately costing staff. To the specific project manager, a project that becomes obsolete before it is complete is better than a project that fails due to bad hires or one that is rejected because it is too costly relative to its returns. Anyway, even when the project comes in over budget, the project results may still be considered acceptable. I even gave it a name: "Teaching the Horse to Sing".

Steve Roth writes:

"Kahneman talks about the "planning fallacy," in which projects always seem to cost more and take more time than expected. I would say that this affects public policy. "

But Kahneman's key example -- which I think he even repeats in the book -- is people's projections of the costs of kitchen remodels, which are typically half (as I remember) of the actual cost.

I have no idea whether there's any decent data on the prevalence of this effect in private businesses. Do you? Suppositions and assumptions don't qualify.

James writes:

Different Randy,

Who is pushing for the destruction of traditional marriage? I know some people are pushing for the law to permit non-traditional marriage, but that's not the same thing as pushing for the destruction of traditional marriage.

If you mean to say there are people pushing for policies which you forecast will ultimately destroy traditional marriage, fine. But please realize that emphasizing the value of traditional marriage doesn't do anything to address the lynchpin issue of whether or not your forecast is reliable.

mike shupp writes:

But this "planning fallacy" is NOT a problem of government per se, it's built into American cultural traditions, and you'll never get rid of it.

Simple example: fill an auditorium up with middle and low level managers, and ask them "How many of you have run a program where you were suddenly told that you would have to let people go, that your budget would fall, that your time schedule had to be shortened, and that you should accomplish what you origially planned regardless? Hold up a hand." Most of the managers would hold up a hand. Then ask, "How many of you were unexpectedly given more people, more time to work with, and a larger budget?" Do you think many people would hold up a hand? Do you think, as a general rule, that it makes a difference whethe managers work for IBM or the DMV?

Think again: You're interviewing applicants for a manager's job. You describe the task -- setting up a customized payroll system, let's say. One applicant very confidently tells you "Ten people, six months, half a million." The other applicant says, "Maybe 15 people, maybe 20. Somewhere between 8 and 12 months, and I don't think it cost more than two million dollars." That's all you have to go on. Make your fine economical, managerial choice. Do you pick differently when you work for the FAA rather than Boeing?

Sure, government managers make bad choices on occasion. So do business men; it's a reason a lot of businesses fail.

Randy writes:

@mike shupp,

That's a good point, but I'd say that though the root problem is the same, it has worse consequences in political operations, because political problems are generally more complex, and because political incentives are more often perverse.

Xerographica writes:

My Pragmatarian Manifesto weaves you, Jeffrey Friedman, Roger Koppl, David Brin, Peter Boettke, Hayek, Spencer, Bastiat, Adam Smith, Buddha, Socrates and Pokemon all together.

If you get a chance I'd appreciate hearing your critique of my critique of liberalism/libertarianism.

Comments for this entry have been closed
Return to top