Poll: The Peter Threshold

The "acceptable" level is, of course, zero. You don't accept incompetence, you attempt to resolve or eliminate it. But that doesn't mean that an org with 1% incompetents will accomplish nothing.

Basically, there's some function from x% competence to y% productivity. I think that it's a power law, something like y=x^2. For low percentages of incompetents, that would mean that each 1% incompetence will mean a 2% hit to productivity.

Especially competent people might manage to reroute around the incompetence, thus reducing the exponent to 1.

Yeah, "acceptable" is bad choice.

It's more fun to estimate what the actual average of incompetents is.

I say this as an incompetent teacher.

homunq-

That's problematic. If the acceptable level is zero, and you are a manager with an incompetent subordinate, you're obviously incompetent, which means your boss is incompetent, etc.

That means that any company with a single incompetent employee has an incompetent overall leader.

On a more serious note regarding "acceptable":
Acceptable is the correct term here. It takes time and money to determine competency so you are weighing resources spent on weeding out the incompetent versus the productivity lost by incompetency. In any venture of a significant size, there will be incompetence that is tolerated (acceptable) because it would be too wasteful to find it or too costly to find a replacement.

The question is ill posed, because it matters not only what fraction of the people in the organization are incompetent, but what their roles are. An incompetent low-level flunky, or even several, can be worked around provided there are enough competent people around to get the job done. The cost is a bit of gratuitous inefficiency, but not necessarily large. An incompetent at the top level can bring down the whole organization, even if everybody else in the system is competent. Likewise in the military: an incompetent buck private is a problem for his platoon which a competent sergeant can solve, but an incompetent general is a problem for the entire army. That is one major reason for civil service protections (a category which broadly speaking includes teacher tenure)--it provides a means by which an organization which is likely to fall under the leadership of political hacks (e.g., Executive Branch departments or school boards) can continue to operate semi-normally even when the people nominally in charge are clueless dolts.

Doesn't it depend on qualities other than being just stupid or clever?

Each officer possesses at least two of these qualities. Those who are clever and industrious are fitted for the highest staff appointments. Use can be made of those who are stupid and lazy. The man who is clever and lazy however is for the very highest command; he has the temperament and nerves to deal with all situations. But whoever is stupid and industrious is a menace and must be removed immediately!â
â General Kurt von Hammerstein-Equord

I think Andre has the right idea. I might put a slightly different spin on it: given the reality of having incumbent incompetent employees, what level of incompetence must an organization be able to withstand in order for it still to operate effectively?

For instance, an organization that cannot withstand even one incompetent employee is an organization that will be out of existence in very short order.

I think Andre has the right idea. I might put a slightly different spin on it: given the reality of having incumbent incompetent employees, what level of incompetence must an organization be able to withstand in order for it still to operate effectively?

For instance, an organization that cannot withstand even one incompetent employee is an organization that will be out of existence in very short order.

Is this a poll about elected bodies?

I once knew of an agricultural research station where none of the researchers understood statistics. They routinely violated some or all of the fundamental assumptions of parametric statistics when it came to plot layout and sampling, yet analyzed their data with ANOVA & regression methods. They had never heard of or didn't understand nonparametric stats, either. Everything they published was essentially meaningless but no one realized this or cared. This organization functioned at the level of 100% incompetence and got away with it because funding agencies assumed that the researchers knew what they were doing. If any actual farmers read the results of this 'research' they were misled but chances are that none ever did. This organization functioned for decades in this manner, at taxpayers' expense, and no one ever realized, cared, or attempted to ameliorate the situation.

Many schools operate in this same manner.

this is an example of what we could call organizational entropy.

it isn't even theoretically possible to have a perfect crystal. there will be dislocations or other defects simply because of entropic effects.

the same with organizations. there will always be incompetance.

I object.

Can't we call this be Leroy Jenkins principle, or something? After all, Peter (stone) is the bedrock on which all organizations are built.

The binning on the poll options is a bit too fine for me, but my guess is that it's on the order of 50%. Here I'm understanding "acceptable" to mean that the organization functions sustainably and recognizably (i.e. does not fall apart or do something completely different than what it's supposed to). In all human endeavors people who do a better job get more work and more serious responsibilities, and shifting things around can mask an enormous amount of incompetence, but above a certain level this becomes impossible.

A good test case for this is the US Congress, which swings back and forth between total and near-total dysfunction. Many or most of the members are essentially incompetent at legislating, and only understand fundraising and campaigning. If I had to guess a fraction of the congress to call incompetent, I would guess 1/2.

Of course if you want a highly functioning organization the fraction is lower, but I would still guess that it's not much lower, maybe 30% or so. As other commenters have noted, rooting out incompetence is very difficult, and often more trouble than it's worth.

Per the comments above, I assume you mean something like "tolerable," rather than (morally) "acceptable." There's still the question, "tolerable in what sense?" Tolerable before the enterprise collapses? Reasonable question. Tolerable because it costs too much to get better? Also a reasonable question.

They still can't be answered simply, though, because not all enterprises are created equal.

The 30% number struck me as ludicrously high, and if someone where I work quoted it to me where I work, it would reflect badly on them and imply to me that they didn't actually understand what was going on. If I had to tag a number at work, it would be something like 5% among the management staff and possibly lower because I don't work closely enough with a hundred different managers to bring it down to 1 or 2%.

But, that commenter doesn't work where I work. Maybe his organization can get by with that.

I would agree with the 30% number, but I think that is mostly because I follow more of a Scott Adams philosophy and believe that people within a business tend to be promoted until they reach their level of incompetence. So, I don't believe 30% of business managers are incompetent because they have always been incompetent, just that at some point they were promoted to a position they could no longer handle, even if they excelled at a 'lower' position.

Let us remind ourselves the obvious that incompetency is a qualitative not a quantitative word and therefore open to interpretation. To me it doesn't mean a bad person, but simply a bad fit between person and job. People get sad, let their personal lives interfere with their professional jobs, and over time all that can change.

Also, 30% is a gross general average, which is unacceptable in some industries, more acceptable in others.

Also, not only does the figure change from industry to industry (Supermarkets for example have extremely low tolerance levels for incompetency, as they only make one percent on what they sell, but they sell a lot ... yet Governments can survive with 80% incompetence. I didn't say they survived well.), withIN a single industry two different companies can have vastly different corporate cultures and the incompetency threshold can vary quite a bit.

Then there's incompetency based on wage level. An incompetent factory shop floor worker will be fired within two weeks, an incompetent CEO who is brilliant but expert in hiding his flaws can in the extreme case take down his company or department. Happens all the time.

Finally, a lesson from Representation Theory, complete with morphism:

NIXXON (arrow) No matter how you spell it, it's still the same old gas.
... MAD Magazine, 1971 or something

I think of the tolerance level as 'how many fairly conscientious people does it take to make up for one incompetent?' It isn't just about the work load. There is also an emotional load from the frustration, etc.

As for business managers, 30% is low in my experience. The best (and one of a very few good) manager I ever had became mediocre when he was promoted from a 'working' Manager to a sub-Director. His competence was unchanged, but he appeared incompetent because his job became 2 hours of meetings, 4 hours of writing reports, 2 hours on the telephone, 2 hours handling emails, 1 hour to work during lunch, and 1 hour of interaction with his department. Almost none of my managers could fire the incompetent person, but they could fire the guy whose daughter had medical problems or whose project wasn't profitable in the first quarter of sales.