On socialized preferences

I recently wrote approvingly of SImone de Beauvoir’s claim that “one is not born, but rather becomes, a woman”. “Feminine” preferences such as a lack of pushiness and tendency not to study maths and science might, I said, be socialized ones rather than innate. This prompted some to ask how much evidence we have that preferences are shaped by social pressures.

Quite a bit. Some more evidence in support of de Beavoir’s theory has come from recent research in China. Alison Booth and colleagues say:

Women in Beijing who grew up during the communist regime, when gender equality was emphasised, are more competitively inclined than their female counterparts who grew up during the post-1978 reform era. These women are also more competitively inclined than their counterparts in Taipei.

In the same vein Uri Gneezy and colleagues show that women are no less competitive than men in matrilineal societies (pdf). Also, their willingness to compete can be increased by the presence of role models, and by raising the salience of career concerns relative to their gender.

All this is evidence that “acting girly” is at least to some extent a social rather than genetic construct.

In truth, though, this is only the tip of the iceberg of evidence that preferences are the product of social influences rather than exogenous data.

We have (at least) four other bodies of evidence.

One is that people tend to adapt to their situation. We might look forward to getting married, or dread widowhood, but within a few years of both events we are as happy as we were before them. This suggests that what we want rises or falls with circumstances. As G. K. Chesterton said: “No man demands what he desires; each man demands what he fancies he can get." 

An important sense in which this is the case is that people adapt even to abject poverty. Amartya Sen has written:

Deprived people tend to come to terms with their deprivation because of the sheer necessity of survival, and they may, as a result, lack the courage to demand any radical change, and may even adjust their desires and expectations to what they unambitiously see as feasible (Development as Freedom, p62-63)

It's also the case that people adapt to inequality: the higher is inequality the greater is our preference for inequality. Today, for example, centrist supporters of the status quo believe it fair that the richest 1% get over 10% of income. In the 1980s, support for the status quo entailed believing the 1% should have had only around 7%.

Our second body of evidence is that preferences are shaped by our formative years. Ulrike Malmendier has shown that people who experienced recessions when they were impressionable prefer to hold fewer (pdf) equities even years later than those who experienced booms. Similarly, CEOs who grew up in recession prefer (pdf) their companies to have lower debt and more cash than CEOs who enjoyed better conditions.

Thirdly, there’s the obvious fact that economic preferences such as willingness to trust others (pdf) or to pay tax (pdf) differ from country to country. This might be because even quite distant history shapes today’s preferences: a country’s history of oppression by an imperialist power might leave people willing to dodge tax even centuries later, for example. And it’s also because culture matters. As David Hirshleifer and Siew Hong Teoh say in a new paper:

Economic attitudes are culturally transmitted, and folk economic ideas are often linked together as ideologies, such as socialism or free market ideologies. This means we need to understand how culture evolves, and an explicit focus on the cultural, not just the genetic, evolutionary process to understand the evolution of economic issues.

Our fourth body of evidence is simply that our preferences are shaped by peer pressure. How much we spend, how we invest, how hard we work and what (pdf) and how assiduously we study are all influenced by those around us.

Now, this evidence does not mean that preferences are wholly socially influenced and that (say) genes are irrelevant.  It does, however, have some important implications.

One is that economists’ assumption that preferences are just given is wrong to at least some degree. Of course, it’s reasonable for some analytical purposes to taken them as given. But a useful working assumption must not be mistaken for reality.

There is, though, a much more radical implication. If preferences are shaped by social forces rather than by a wholly rational assessment of our interests then satisfying preferences might not have much if any ethical value. Amartya Sen has written (pdf):

Consider a very deprived person who is poor, exploited, overworked, and ill, but who has been made satisfied with his lot by social conditioning (through, say, religion, or political propaganda, or cultural pressure). Can we possibly believe that he is doing well just because he is happy and satisfied? Can the living standard of a person be high if the life that he or she leads is full of deprivation? The standard of life cannot be so detached from the nature of the life the person leads. As an object of value, happiness or pleasure (even with a broad coverage) cannot possibly make a serious claim to exclusive relevance.

If we merely satisfy preferences, we’ll not give anything to such a deprived person. After all, she (the right pronoun mostly) doesn’t want anything. To this extent, satisfying preferences can perpetuate and even increase injustice: we give too little to the satisfied poor and too much to the over-entitled rich.

Partly for this reason, Daniel Hausman (among others) has argued that we have no ethical reason to satisfy preferences when these are a bad guide to people’s interests. Which, of course, calls into question the value of (actually existing?) democracy.

Comments

[Luis Enrique]

I don't think I have ever encountered an economist who thinks "preferences are just given", as opposed to socialised (at least in great part). The ethical implications are tricky but granting yourself the licence to decide that people's preferences are a bad guide to their interests strikes me a dangerous territory.

I don't quite understand the last part of this blog. Wouldn't Sen's deprived person, content thought they may be, still prefer to be less poor, exploited, overworked, and ill? Is Sen really saying that the impact on this person's welfare from "giving them anything" is zero because they are satisfied by their lot?

IMO this all just reinforces my belief that crude utilitarianism with diminishing marginal utility of consumption is the least bad approach to welfare economics. If you start from there, you don't get into any silliness like saying, we should help this unhappy rich person more than this content poor person.

[Emma]

I can't believe I'm living in a moment in which a bunch of khaki-trousered nerds on the internet have decided they're Nature's truest and most legitimate predators, and that everyone else (especially icky girls) has to be formally prevented from stealing their authority. It's amazing. But I'm probably only angry about it because I haven't had any recent contact with an infant, I suppose. (FYI, contact with an infant is likely to make you more angry, not less. And also exhausted.)

I think the idea with the satisfied poor person is that people can become such slaves to their cultural context that they become untrustworthy reporters of their own misery. I think this is true, actually; people will actively fuck themselves over to maintain compliance with narratives about religion, or staying/getting married, or having kids, or keeping a "good" job that destroys them so they can maintain or further acquire a mountain of consumer debt. I think the suggestion is that we shouldn't rely on self-reporting in these cases, where some Catholic lunatic might refuse birth control despite recently having a 15th baby they can't feed, but on an assessment of people's material circumstances (just as we rely on laws and ideals about equality when it comes to public policy, rather than the arbitrary, reactionary ideas of any given person or population).

Which somehow turns into a value-judgement about the utility of democracy for people so heavily socialized against economic justice that they perpetually vote against their own interests.* Except that democracy is a contingent social construct as much as a terrible political opinion is, and I haven't seen any better system yet devised by Nature's prime predators.

*I distinctly remember multiple posts on this very blog about that very phenomenon, which stipulate that "research" supports the idea that people don't vote *against* their own interests specifically, but *for* what they imagine to be society's best interests. Which means that Marxists just need a better propaganda machine and more cultural clout, so they can convince all the dumb people that Marxism is the actual cure for what ails them. Rather than Jesus, a return to the norms of the 1950s, or khaki trousers.

[Jim]

Its odd that whenever people make choices that the Left don't like then those choices are declared 'invalid' as the result of 'social pressures', yet choices people make in favour of things the Left approve of are always totally valid and not to be in any way gainsaid.

[G]

You could have at least mentioned the Scandanavian thing....

[Handy Mike]

My own, rather snarky phrasing "maybe there are some other 'examples' you could look at that might tell against the idea that preferences are socialised." was too crude.

I should have asked for examples showing that preferences and other dispositions relevant to the reported 'gender pay gap' are driven more by socialisation than by innate factors.

And not just examples, but some sense of the balance of such findings on the issue, instead of call-outs to a few links that sort-of align with the point you wanted to make.

But, cheap BTL snark notwithstanding, that was obviously the critical point to answer from the previous post, and I can't see how the examples in this follow-up post do answer it.

[Nick Drew]

You've given a pretty comprehensive list of relevant factors that make the issue(s) complex (including the last difficult point on democracy: we've always known the parents of chimney-sweeps would have voted to allow them to continue that employment, and colliers to fill their lungs with coal dust)

But you haven't followed the logic all the way.

(1) If socialisation is a major factor in defining preferences, the society in question will tend to select for excellence in the 'socialised' desideratum (e.g. "girly behaviour"), which over time will lead, Darwin-wise, to genetically-based (/reinforced) "girly behaviour" - a classic positive feedback loop. This muddies the waters still further

Going back still further along the causal chain ...

(2) ... I take it nobody disagrees that there is a significant natural disparity in physical attributes (e.g. upper body strength), on average, beween men and women (primarily a matter of testosterone). If outright hormonal differences cause clear, significant physical differences (and why wouldn't they?), then unless you wish to assert a very strong form of Cartesian dualism - i.e. that the mind etc is completely, metaphysically different stuff to the body - why shouldn't hormonal / physical differences lead to preferential differences that, (even if also reinforced by socialisation) have their root cause in purely physical, "innate" factors?