Wednesday, December 17, 2008

Class Mobility and the Original Position

I had this thought at 3 AM last night, so forgive me if it is utter gibberish.

One of John Rawls key contributions to American political theory was the idea of the "original position". Basically, it imagines citizens under a "veil of ignorance" -- who have no idea about their actual social position, identity, or even capabilities -- deliberating about how they want their society to be constructed. Rawls argues that persons in such a society would be risk-averse (they'd be worried they would be amongst the least well-off members of society), and hence they would evaluate their options based on the minimax rule: basically, trying to get the best possible outcomes for the least well-off. Specifically, Rawls believes that citizens would demand a system wherein:
1. Each citizen is guaranteed a fully adequate scheme of basic liberties, which is compatible with the same scheme of liberties for all others;

2. Social and economic inequalities must satisfy two conditions:

a) they are to the greatest benefit of the least advantaged (minimax rule);

b) they are attached to positions and offices open to all.

One can argue about what particular social arrangements would reflect these standards, but for Rawls they are pretty closely tied to the modern liberalism (with somewhat of a democratic socialist bent).

Of course, the original position is just a thought experiment: in the real world, people do know their capabilities and do know their social position, and their political demands are reflected accordingly. But (and this is what I thought of last night), we can at least somewhat approximate the conditions of the original position along axes where there is significant social mobility. For example, think of class. In a society where there is a lot of class mobility (both upwards and downwards), people kind of are under a veil of ignorance, because even if they know their current position, they can credibly claim ignorance as to where they will be in a year, or two, or five. If Rawls is correct, we'd expect persons who meet that description to push politically for policies that roughly approximate those he believes will flow from the original position.

The problem is that most societies and most identities aren't that mobile. I'm pretty secure in my identity as White. There isn't much of a possibility that next year I'll become Black, and so I have no incentive to be risk-averse as to what happens to Black people, since I know it will happen to me. Even identities which are more amenable to change, such as religion, wouldn't fit, because most religious adherents -- even those who eventually convert -- don't imagine themselves as being in a permanent state of flux moving between religious affiliations.

An exception might be class. In the United States, there is a solid opportunity for upward class mobility amongst the White middle class, but very little downward class mobility. But the Black middle class is a different story. It is notoriously vulnerable to economic shocks -- "one paycheck away from disaster" -- and thus can credibly claim to be "ignorant", in a Rawlsian sense, to their permanent economic status.

Hence, we construct a thesis: If Rawls is correct, we would expect that the politics of the Black middle class, at least along class issues, to be relatively close to what Rawls would expect to flow from the original position, compared to the "control" of the relatively less mobile White middle class. And if we think Rawls is right that the original position provides a good template for ordering society, then it would also follow that we should give substantial weight to the political preferences of the Black middle class when making decisions about economic distribution.

The second claim is of course a matter of opinion. But the first I suspect would be empirically defensible (though I'd be interested to see someone actually to the legwork here). I'm pretty sure that the Black middle class is significantly more liberal than its White peer group, particular amongst economics, and particularly in terms of assistance to the poor. This would imply that the risk-aversion Rawls predicts is shining through -- knowing that they could become poor with frightening ease, the Black middle class seeks to create social structures that minimize that disadvantage.

Now, there are problems with this model. First, Whites are less downwardly mobile, but are they more upwardly mobile than the Black middle class? And if so, what does that mean (I'm not sure they are more upwardly mobile, but even if they are "poor + middle class" represents a bigger proportion of the American economic experience than "middle class + rich"). Second, Blacks might be more prone to protect the poor for reasons other than risk-aversion -- they might (and in fact are) more likely to have friends and relatives amongst the ranks of the poor, which could affect preference formation. Third, there is an obvious difference between "ignorance of one's social position" and "knowledge that one's social position is prone to change", which could exert an impact -- most notably, the latter position might over-privilege policies which either protect against downward mobility or provide greater resources to families undergoing class transitions.

But even still, I think there might be a there here. Any thoughts?


Matt said...

It seems to me we're largely a nation where almost everyone deludes themselves into thinking they're middle class (or working class when the option is available, which is rarely) while also deluding themselves into thinking they'll be rich someday. It's rather the opposite of Rawls' veil of ignorance.

David Schraub said...

Well that's why I think the Black middle class makes an interesting test case -- my readings indicate they are actually far more precariously situated than their White peers, and I'm assuming they're aware of that fact as well.

Jack said...

So the claim you want to test by looking at the black middle class is whether people under the veil of ignorance are really risk averse? You pick the black middle class because unlike other group they're economic status is legitimately at risk. But if theres no upside it doesn't seem like a fair experiment. Obviously if you're far more likely to drop down in economic status then you are to move up then you'll be likely to give political support to the poorest Americans. The only way to test the original position would be to have a group in which the average member's chances for any given position in society is roughly equivalent to the chances of a random assignment.

I see what you're trying to do but I don't think its really necessary. There is strong experimental evidence that people are generally risk averse and absolutely no reason to think they wouldn't also be risk averse when it comes to betting on their socio-economic status.

RFYork said...

It would be very helpful if someone actually performed a true double-blind experiment on risk aversion and class and ethnicity. I can't argue that your position is wrong simply because there is no empirical evidence to either support or attack your position.

I think Matt's observation is the most important issue underlying the failure of large segments of the population to support their own self interest. Thomas Frank had some good arguments, but I think he missed the most important one, emotion.

For at least all of the last century, the Horatio Alger myth has been nurtured by the right. A large plurality in this country (if not a majority) truly believes that they have equal access to success and wealth as the people who already have them. How else can one explain the antipathy toward the Estate Tax?

Risk aversion probably plays a strong part in people's social and cultural attitudes, but I think Horatio Alger is at least as important.

Good blog, quite thought provoking.

Mark said...

Another issue is that people generically are abysmally poor at doing realistic "reasonable" risk assessment (largely due to innumeracy). Which means you have to deal with people internal notions of risk vs realistic/scientific ones.

Anonymous said...

I wonder if as opposed to American society, a society undergoing periods of extreme flux in socio-economic alignments would be a far better case study to advance this hypothesis?

I'm thinking of what could be described as "Rawlsian" socialist trends in say Argentina in the mid-late nineties. Though this could be another situation where you have significant risk of downward mobility but minor hope of upward.

- Adam

PG said...

a society undergoing periods of extreme flux in socio-economic alignments would be a far better case study to advance this hypothesis

I'd also be curious about using students in a system with high-stakes testing that is less tied to socioeconomics than our own. (Though I don't know how successfully any test writer can ensure that an exam will measure "intelligence" rather than cultural capital.) For example, in India doing well on the high school exam can get you into government-funded colleges where your entire room, board, tuition etc. are covered. (My dad even used to send money home when he was in medical school.) These exams seem to be a potential engine of social mobility at least for the lower-middle class: the kids who can afford to finish their primary and secondary schooling but couldn't afford to go further without a full ride.