The other day, I quoted Steven Lukes to the effect that one manifestation of capitalists' power is that voters do not challenge that power. This claim needs substantiating.

Let's start from the reasonable premise that voters do seem content with inequalities of power and income. Why might this be? One possibility is that it reflects a rational, autonomous preference. Maybe inequality is a price worth paying for economic freedom and prosperity, and that if we had a more socialist society people would rebel and choose managerialist capitalism. The other possibility is that contentment with inequality is endogenous - that capitalism generates preferences which help sustain inequality, against the interest of many people.

What's the evidence for this latter theory? A lot is indirect and suggestive:

- We know from the work (pdf) of people such as Christopher Hsee and (pdf) Dan Haybron that people's preferences don't necessarily promote their interests; this is the basis of "Nudge". But if our choices of consumer goods and savings can be irrational, isn't it more likely that our voting behaviour will be, given that inattention and ignorance towards politics can be rational?

- Cognitive biases are ubiquitous, and many of these might bias people towards a greater acceptance of the existing order than is rational.

- We know that peer pressure can influence people's behaviour - even, for example, about which shares they buy. But if our preferences are malleable by our social networks, mightn't they be malleable by other social forces too?

- The function of social norms is to get us to behave in ways that are good for society, but not in our direct material interest. They explain why we don't steal even if we won't get caught, or why we tip in restaurants we'll not revisit. But if norms can restrain the pursuit of our interests in this way, mightn't other norms restrain hostility to capitalism and managerialism?

- We know that capitalism can generate irrationality in one respect; hunter-gatherers are less prone to the endowment effect than modern men. So, could it generate other types of irrationality, such as support for even damaging forms of inequality?

On top of this indirect evidence, we have some laboratory evidence. Jeffrey Butler got people to play an urn-guessing game, and found that people who were paid well for correct guesses thought they were better at guessing than people who were paid badly. This suggests that arbitrary inequality can boost the confidence of the well-off and depress that of the worst off, which in turn can entrench that inequality by giving the rich a sense of entitlement and the poor a sense that they shouldn't upset the apple-cart.

Another experiment was done at the University of Rennes. Researchers split subjects into two groups - advantaged and disadvantaged - and gave the disadvantaged group the chance to destroy some of the advantaged's wealth at a cost to themselves. They found that the subjects did so when inequality was modest, but not when inequality was greater. "The level of conflicts significantly declines with the extent of inequality", they concluded - probably because the disadvantaged become resigned to their fate.

All this is consistent with the Marxian hypothesis that capitalism generates an ideology which helps sustain inequality. And you don't need conspiracy theories or a belief in the power of mass media to believe this.

Granted, it's not proof. And there are other things that explain the weakness of leftist parties - not least of which is the stupidity of so many on the left. All I'm saying is that we should take seriously the possibility that preferences for the existing order are endogenous and not fully rational.