Double Curse of Incompetence

The Bush Administration's stunning managerial incompetence (at least on anything remotely related to effective governance) seems so broad and deep that it raises its own research questions:

1. Is it possible for such massive incompetence to arise in a natural manner?

2. Why do incompetents hire so many other incompetents? Why don't they mask their weakness by hiring competent people to cover for them?

Yes, Virginia, there is a science to incompetence. Like any other natural phenomenon, scientists can study it - and they have discovered a few interesting things about the way that incompetence actually works. Follow over the fold for the deeper truths.

First the science.

The study of incompetence as a formal area of research seems to have been kickstarted in a paper by Dunning and Kruger in 1999. The article is viewable online at the journal's website. They mention some earlier observations and comments, but theirs is frequently cited as the basis for the empirical research in this field. There is even a wikipedia page on the 'Dunning-Kruger' effect.

If you are inclined to read well-written research papers, you will doubtless want to review the paper for yourself. However, to cut to the chase, the researchers conducted 4 experiments with Cornell university students and drew the following conclusions:

1. The less competent the individual, the more they tended to overestimate their own performance compared to others. Conversely, competent people tended to underestimate their performance compared to others.

2. Incompetent people could not distinguish others' good work from bad work, while competent people could tell the difference quite accurately.

3. When competent people had a chance to compare their work to those of others, they revised their self-estimates upward. This suggested that their original error was primarily due to their generous assumption that others were as competent as they were.

4. When incompetent people received training to make them competent, they began to make more accurate self-assessments. This suggested that the only way to make incompetent people aware of their incompetence is to teach them to be competent.

One of the authors' most important inferences concerns the inability of incompetent people to learn through feedback. Incompetent people may be unable to learn from their own experience because their incompetence prevents them from recognizing the lessons that their own experience contains. Worse, since incompetent people have difficulty comparing good and bad work or behavior, they are less able to learn from others' successes or failures.

While a number of subsequent studies have tended to reinforce these findings, a recent study by Burson, Larrick and Klayman (2006) suggests that the results may depend on the perceived difficulty of the task or subject. Incompetents tend to be overconfident when tasks are widely considered easy and overly pessimistic for tasks that are generally considered hard. Of course, you have to be competent to know whether a task is really easy or hard.

Note: If this subject really piques your interest, you can view this literature review or you may want to go to a library or online academic database (subscriber only) or track down papers like the following:

1. Metcalfe, J. (1998). Cognitive optimism: Self-deception or memory-based processing heuristics. Personality and Social Psychology Review, 2, 100–110.

2. Ehrlinger, J., & Dunning, D. (2003). How chronic self-views influence (and potentially mislead) estimates of performance. Journal of Personality and Social Psychology, 84, 5–17.

Then the Implications

George Lakoff has argued that the Bush administration's failures are not due to incompetence. He blames the administration's adherence to a conservative philosophy that, by definition, cannot work.

This explanation makes a lot of sense if one is looking at the panoramic scope of their big-ticket mistakes. However, I find it less convincing as an explanation for the massive litany of smaller failures and idiocies. The inherent inconsistency of Conservative philosophy doesn't explain why anyone would be stupid enough to hire the head of the Arabian Horse Association to run FEMA. Or pick a wet-behind-the ears graduate of Regents University over multiple Yale Law school applicants for a senior Justice Dept. position. It's like they don't even understand what a resume is supposed to tell them.

This is where the research of Kruger and Dunning might plausibly come into play. Let me posit two hypotheses:

1. You have to be pretty incompetent to buy into a management/governing philosophy that has as many internal inconsistencies as Republican Conservatism.

2. If you are that incompetent, you will have great difficulty in telling competence from incompetence in others.

Let us examine the Bush Administration's ability to hire competence. On a naive level, one might think that the inability to tell competence from incompetence would lead to hiring a mixture of competent and incompetent administrators for political government positions. OK, what sort of mixture would one expect?

One possibility is that the hiring decisions of incompetents is essentially random. That would mean that the number of hired competents would be approximately equal to the fraction of competents in the resume pool. If 40% of the applicants were competent and 60% incompetent, then 40% of the hires should be competent and the rest dolts.

Unfortunately, that ignores the fact that hirings typically involve interviews. During interviews, incompetents are asking candidates questions and judging the answers. This creates a different set of options:

1. Some competent candidates (the honest ones) will give answers that the incompetent questioners don't understand and presumably eliminate themselves from contention.

2. Other competent candidates will lie to tell the incompetent interviewers what they want to hear. If they are sufficiently good liars, they may be selected because the incompetent interviewers won't be able to spot the subterfuge.

3. Some incompetent candidates will be lucky enough to tell the incompetent interviewers something that the interviewers mistakenly credit as competence. Their odds will increase if there is widespread publication of the preferred answers (i.e., the party platform). Nothing like having a cheat sheet when you don't know the answer.

4. Some incompetent candidates will say things that offend the equally incompetent interviewer and will be eliminated.

I may be wrong, but this suggests to me that that a selection process administered by incompetents is not likely to be random. Rather, it is likely to strongly favor the hiring of lying scoundrels and sycophantic incompetents. In their defense, it is barely possible that the Bush Administration may have upheld the Republican mantra and succeeded in excluding people who are too stupid even to cheat.