Two Wrongs

The Secret of Meritocracy

The Secret of Meritocracy

Meritocracy: it would have been a good idea if we could reward merit.

In practise, meritocracies reward luck and political skill. Are there better alternatives? I don’t know. Are there worse alternatives? Yeah, lots of them.


Take sales. It’s a very common misconception that the difference between good sales and bad sales is in the total size of contracts closed. This is like measuring how good a software engineer is by how many lines of code they write. The sales department, in choosing which prospects to target, are effectively driving the future technological direction of the organisation. Good sales is about targeting prospects that will steer the organisation in a desired direction. This has no relation at all to the revenue brought in.

But even before we get to the technicalities of prospects – is a salesperson cooperative? Do they take the last coffee and leave the pot empty? Are they responsive to internal support requests? There are many dimensions along which one can be a good employee, before bringing in specialisation. No single person has a full view on all of these qualities; everyone sees a fraction of the person. It’s impossible to meaningfully condense all this down to some sort of merit ranking.

Aside from the impossibility of measuring what’s good, there’s an implicit question that’s left unanswered: good… on what time scale? Tell me to save a lot of money this quarter and I will cut all maintenance and research. We will be rich! For a year, maybe, because I transferred future savings into today. Anyone can do that. But briefly, I looked meritorious. Doing good on longer time scales can have the reverse effect: learning something difficult entails making lots of mistakes. If I am judged on how good I look today, I will avoid learning anything new.


So clearly, meritocracies cannot be based on actual merit. It’s too inscrutable. Instead, they are based the appearance of merit. I think there are two ways to appear meritorious: politics and luck.


Luck, or chance, is what can make good decisions have bad outcomes, and bad decisions have good outcomes. Oddly, people tend to judge a decision by its outcome. This means in a meritocracy, we don’t have to make good decisions – all we need to do is make crazy bets and hope that one of them just happens to pay off. This will look like skill to most people we meet.


Hoping for luck is slow and, clearly, no guarantee of success. There’s an easier option: if we can simply convince people that we’re capable, we don’t have to be. As long as we can maintain the image of our merit, we’re good. Of course, the higher we climb, the harder it will be to maintain. Eventually, we’ll spend all our working hours maintaining the image of our merit. I have met some people like that; they really do exist.


The effect of the above is that meritocracies tend to have relatively incompetent people at the top. It’s easier to make a bad decision and be lucky than a good decision and be lucky, simply because bad decisions are so much easier than good ones. Thus, of the lucky, we find more people who are incompetent. Of the political … well, the more time you spend on politics the better you’ll be at it, but the worse you’ll be at everything else.

Note that this is not a call to abandon meritocracy. I don’t know of any better alternatives, and I know plenty of alternatives that are strictly worse, much worse. The reason I write this is as a reminder that even structures that are ostensibly about one thing can easily optimise for something else if we aren’t very careful.