The Social Mobility and Child Poverty Commission this week reminded us of an old fact - that privately educated people are disproportionately represented among top jobs. Why is this?

It's not solely because prviate schools give people a better chance of going on to get good degrees. This paper shows that, even controlling for university, the privately educated have a slight advantage in their chances of getting top jobs. And this one shows that, controlling for degree, there's an earnings premium for the privately-schooled. This is consistent with anecdotal evidence. Sajid Javid's experience of getting job offers from foreign but not British banks, I suspect, chimes with the experience of many of us from the wrong side of the tracks.

Equally, though, I'm not sure it is solely due to outright overt class discrimination. Call me naive, but I doubt that Alan Rusbridger - to take one example of someone who employs lots of folk from expensive schools - wants to grind the faces of the poor or faints at the thought of proximityto ghastly oiks.

Instead, I suspect that what's going on is an unintentional form of discrimination. The privately educated give off more competence cues; they are more likely to fit in; and because they seem familiar, they are less of a risk to prospective employers.

What we have here, then, is an example of emergence.Discrimination against oiks can occur as the unintended outcome of reasonable behaviour. Social phenomena are not simply individual behaviour writ large. As Alan Kirman said in his important book, Complexity Economics, "Aggregate behaviour does not always have a counterpart in the microeconomic data.

Although complexity and emergence seem like new ideas they are in fact at least as old as modern economics. When Adam Smith wrote that "it is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest" he was expressing the idea of emergence; benign social outcomes can result from nasty motives.

Now, sometimes, emergence is a good thing - as in Smith's invisible hand or in this example of how stupid traders can generate rational markets. But there are other cases where it's not so benign, for example:

- The idea of rational bubbles shows that intelligent traders can generate irrational markets.

- Patriarchy is, in part, an emergent process - it arises not just because men are consciously sexist, but because of socio-psychological mechanisms.

- Markets and institutions can select for idiots and against the competent, resulting in an anti-meritocracy even though few people really intend this to happen.

- As Marx pointed out, markets might seem like a realm of freedom and equality, but in fact they are places in which the worker is "bringing his own hide to market and has nothing to expect but - a hiding."

Only a silly fanatic would regard emergence as always benign or always nasty. How it works varies from context to context according to quite subtle conditions. My point is simply that combatting inequality might take more than simply changing attitudes.

Herein, though, lies a link with that other big story of the week, the Rotherham child sex exploitation affair. One response to this has been to demand Shaun Wright's resignation as PCC. However, social problems occur not (just) because we have weak or corrupt individuals in positions of power but because of complex social forces. The notion that these can be tamed by the right people represents an unthinking rejection of the notion of emergence.