David Ruccio points to labour's falling share of income in the US and says:

We need to talk much more about profits and who owns capital. And, in addition, who appropriates and distributes the surplus and to whom that surplus is subsequently distributed.

This is like saying a man should put his trousers on before leaving the house.It's good advice, but it shouldn't need saying.

A nice new paper by Amparo Castello-Climent and Rafael Domenech at the University of Valencia supports his point.

They point out that there's no correlation between inequality of human capital and inequality of incomes. This is true across time: since 1950 human capital inequality has fallen in most countries but income inequality hasn't. And it's true across countries; many Asian countries have quite high inequality of educational achievement but low income inequality whilst south America has unequal incomes but relatively equal human capital.

This is a challenge for the neoclassical view that income inequality is due to inequality of marginal productivities.

One might try and rescue the marginal productivity story by arguing that the return to primary education is low and that to university education has risen because of skill-biased technical change, so that a rise in human capital equality because of higher basic skills is compatible with rising inequalities of marginal productivity. However, other research suggests this story isn't true; the returns to basic skills are quite high, and changes in inequality are loosely linked to changes in education.

Instead, the more obvious possible reason for the lack of link between human capital and income equality is simply that inequality reflects not differences in productivity but differences in power which themselves arise from institutional differences.Inequality is higher in south America than in Japan or South Korea simply because south America has extractive institutions which enable a small minority to exploit the masses, whereas Japan and South Korea do not.

Institutional differences in power also help explain another fact: why does the return to university education differ so much (pdf) across European nations of similar income? It is higher in the UK than in Germany or Nordic countries, for example. It's hard to explain this by technical change or globalization, as these factors should have affected countries reasonably similarly. A more plausible possibility, surely, is that institutional factors - the power of capital over labour - allow (some) graduates greater access to the economic surplus in the UK than it allows them in the Nordic countries.

Although I'm speaking here in macroeconomic terms, the point holds at a micro level too. Why did Rebekah Brooks get a ''10.9m payoff from Murdoch? It's not because she has obvious greater marginal productivity or technical human capital than the rest of us. It's because (for reasons we needn't consider) she had privileged access to the surplus.

Inequality, then, is better explained by power than by human capital or marginal productivity.