Posts tagged ‘Dunning Kruger’

Government, Arrogant Ignorance, and the Power of Incentives

As most of you know, my company operates parks on public lands, so I work with government agencies a lot.  Years ago, from this experience, I coined a term called "arrogant ignorance."  It comes from numerous times when government employees will be completely ignorant of some process, perhaps even their agency's own rules and procedures, but will fight to the death any suggestion that I might be able to enlighten them or that they are doing something wrong.

For a while, people had me believing that I had just rediscovered the Dunning–Kruger effect.  But I am now convinced that this is not the same as my "arrogant ignorance".  And the difference between the two highlights a key point about failure of government I have made for years, which is that government does a bad job not because the people are bad, but because it hires good (or at least average) people who have terrible incentives and information.

First, here is Dunning-Kruger per Wikipedia:

The Dunning–Kruger effect is a cognitive bias in which relatively unskilled persons suffer illusory superiority, mistakenly assessing their ability to be much higher than it really is. Dunning and Kruger attributed this bias to a metacognitive inability of the unskilled to recognize their own ineptitude and evaluate their own ability accurately.

Like most people, I see Dunning-Kruger all the time.  But I see it equally frequently in private and public settings.  I don't think it is necessarily unique to the public sphere, and may be over-represented there only to the extent that it is much harder to eliminate under-performers from public rather than private jobs, so they may tend to concentrated more in public positions.

But my concept of arrogant ignorance is not really a cognitive effect, I think, but rather a symptom of incentives.   The problem with most government jobs is that they have no service or output metrics so that they are instead judged mainly on conformance to procedure.  And even that is not quite correct, because most agencies I work with do not even have formal standards or quality review processes for their employees, at least below the executive level.

I want to take an aside here on incentives.  It is almost NEVER the case that an organization has no incentives or performance metrics.  Yes, it is frequently the case that they may not have clear written formal metrics and evaluations and incentives.  But every organization has informal, unwritten incentives.  Sometimes, even when there are written evaluation procedures, these informal incentives dominate.

Within government agencies, I think these informal incentives are what matter.  Here are a few of them:

  1. Don't ever get caught having not completed some important form or process step or having done some beauracratic function incorrectly
  2. Don't ever get caught not knowing something you are supposed to know in your job
  3. Don't ever say yes to something (a project, a permit, a program, whatever) that later generates controversy, especially if this controversy gets the attention of your boss's boss.
  4. Don't ever admit a mistake or weakness of any sort to someone outside the organization
  5. Don't ever do or support anything that would cause the agency's or department's budget to be cut or headcount to be reduced.

You ever wonder why government agencies say no to everything and make it impossible to do new things?  Its not necessarily ideology, it's their incentives.  They get little or no credit for approving something that works out well, but the walls come crashing down on them if they approve something that generates controversy.

So consider the situation of the young twenty-something woman across the desk from me at, say, the US Forest Service. She is probably reasonably bright, but has had absolutely no relevant training from the agency, because a bureaucracy will always prefer to allocate funds so that it has 50 untrained people rather than 40 well-trained people (maintaining headcount size will generally be prioritized over how well the organization performs on its mission).  So here is a young person with no training, who is probably completely out of her element because she studied forestry or environment science and desperately wanted to count wolves but now finds herself dumped into a job dealing with contracts for recreation and having to work with -- for God sakes -- for-profit companies like mine.

One program she has to manage is a moderately technical process for my paying my concession fees in-kind with maintenance services.  She has no idea how to do this.  So she takes her best guess from materials she has, but that guess is wrong.  But she then sticks to that answer and proceeds to defend it like its the Alamo.  I know the process backwards and forwards, have run national training sessions on it, have literally hundreds of contract-years of experience on it, but she refuses to acknowledge any suggestion I make that she may be wrong.  I coined the term years ago "arrogant ignorance" for this behavior, and I see it all the time.

But on deeper reflection, while it appears to be arrogance, what else could she do given her incentives?  She can't admit she doesn't know or wasn't trained (see #2 and #4 above).  She can't acknowledge that I might be able to help her (#4).  Having given an answer, she can't change it (#1).

You may think I am exaggerating -- how could people react so strongly to seemingly petty incentives.  But they do.  In my example above, this is probably her first job.   The government is the only employer she has known.  The confidence you might have to ignore these incentives to do the right thing likely come from jobs and experience that this woman has not had.

I will give you a real example.  One government contract manager asked us to spend $10,000 to do something, promising that the agency would reimburse me.  I told her that I had never heard of this type of spending being reimbursable, but she said we would be reimbursed.  So we did it.  Later, her boss's boss heard about the reimbursement and said it was not correct under the rules.  Eventually, our contract manager was challenged on it.  You know what she said?  She said our company spent the money without permission and that we were never promised reimbursement.  She sacrificed her honor out of the fear of #1 and #3 - the incentives were that powerful for her.  She knowingly lied and -- by the way - cost me personally $10,000 and a reprimand in our contract file.  When I called her afterwards and asked her, "what the hell?" -- she apologized to me in tears and said she just would be in too much trouble once her boss's boss was involved to admit she had authorized the expense.

So, I try to learn from this.   One thing, for example, I always do is ask myself when someone who works for me screws up, "Is this really my fault, for not training them well."  A surprising number of times, the answer is a reluctant, "yes".

Fargo, But with Lawyers and Porn

This is a little dated, but Ken White has a mega-update on the Prenda Law case he has been following.

The ins and outs of this case are complicated beyond belief (likely purposefully by the key players in a bid to obfuscate what they were doing), but the basic facts appear to point to this:  Prenda and a series of related entities were buying copyrights to porn, uploading electronic versions of these videos to known pirating sites, and then suing folks who downloaded the files  (knowing that most folks, embarrassed that they downloaded "Chubby Nurses in Heat" or whatever, will fold and pay a settlement rather than get in a public legal fight).  One reason for the complexity and obfuscation is that the porn companies (AF holdings and many other shells) have to pretend on the one hand that they didn't upload the files themselves in a "honeypot" operation, and on the other hand that they have no relation to Prenda Law.  By the way, the scheme apparently brought in about $2 million in 2012 alone of which at least two thirds, and likely more, ended up in the pockets of the key principles.

What makes the case so fun to read about is the just idiotic antics and evasions by the key players, the hapless lawyers, the "dog ate my homework" excuses in front of senior Federal judges, etc.   All this combined with an arrogance among the principles that could be a case study in the Wikipedia entry on Dunning-Kruger effect.   The bad guys remind me of nothing so much as the William Macy character in Fargo.   This, for example, is a hilarious article with examples of one principle after another offering absurd testimony to various Courts.

Since the post above, Ken has an update here.

Arrogant Ignorance

Years ago I coined a term for a number of people I deal with in business -- "arrogant ignorance."  I don't mind running into folks who are young and inexperienced and admit such -- in fact, I like educating and training people and sharing what limited knowledge I have accumulated.  But what really sets me off are folks who have no idea about the subject on which they are making decisions but act as if their judgment is beyond question.  This tendency seems to be reinforced by organizations that have few real performance metrics and where, as such, looking like one knows what he or she is doing is more important than actually doing anything.

More recently, I found that this effect already has a name - Dunning-Kruger, though I think my term is much more evocative.  Anyway, this is an interesting article on Dunning-Kruger and its legitimacy in describing actual human behavior.

Via Tyler Cowen