Adam Smith thought there were two economies - meritocratic for the poor and powerless and anti-meritocratic for the powerful:

In the middling and inferior stations of life, the road to virtue and that to fortune, to such fortune, at least, as men in such stations can reasonably expect to acquire, are, happily in most cases, very nearly the same. In all the middling and inferior professions, real and solid professional abilities, joined to prudent, just, firm, and temperate conduct, can very seldom fail of success...The good old proverb, therefore, That honesty is the best policy, holds, in such situations, almost always perfectly true. (The Theory of Moral Sentiments, Part I, Sec 3, Ch 3)

But:

In the superior stations of life the case is unhappily not always the same. In the courts of princes, in the drawing-rooms of the great, where success and preferment depend, not upon the esteem of intelligent and well-informed equals, but upon the fanciful and foolish favour of ignorant, presumptuous, and proud superiors; flattery and falsehood too often prevail over merit and abilities. In such societies the abilities to please, are more regarded than the abilities to serve.

This fits the theory of a friend of mine. He says that further up the hierarchy one goes, the longer it takes to identify merit, and so the more chance shysters and bluffers have of thriving. You can spot immediately whether someone is capable of sweeping a factory floor, but it takes longer to discover whether they can manage the factory.

Which raises the question: how extensive is the anti-meritocratic economy? Karen Bradley's recent admission to being clueless about the province's politics before she became Northern Ireland secretary was merely the latest and most egregious reminder that in politics merit and abilities are not always decisive to preferment.

And since Smith's time, we've seen the rise of celebrity culture - a whole sub-economy of media and entertainment where abilities to please (and dumb luck) produce Adler superstars (pdf), people whose success rests upon simply being talked about. Boris Johnson and Kim Kardashian have much in common in this respect - the difference being that one has a massive arse and the other is a massive arse.

It's easy to believe the same is true near the top of corporate hierarchies. Top managers often prefer to hire men in their own image - partly because they are more likely to trust them. This gives (pdf) us the Peter principle - that people are promoted to their level of incompetence - and the Dilbert principle, that duffers are more likely to be promoted. And their charm and deceitfulness mean that psychopaths are over-represented in top jobs.

All this is compatible with evidence that there is a "long tail (pdf) of extremely badly managed firms" with low productivity (pdf): this is what we'd expect if managers are promoted on factors other than ability. (And no, the wheels of competition don't grind so finely as to swiftly eliminate such laggards.)

In this context, a new paper (pdf) by Andrew Oswald and colleagues challenges my priors. They estimate that only around one-eighth of workers in Europe has a bad boss - fewer than one would expect if ; flattery and falsehood too often prevail over merit.

Perhaps, though, the question here is not merely about numbers. Instead (as is often the case) it is about mechanisms. Are the mechanisms that select for merit and abilities rather than flattery and falsehood really as extensive as they should be? And if not, how can we change this - assuming of course that we want to?