While policy action may not justify the rhetoric, there’s a strong and historic bipartisan consensus that we ought to strive for equality of opportunity, if not equality itself. Liberals think of this as everyone’s right to a fair chance – a shot at the “American Dream”. (Unfortunately, I feel like an idiot typing that phrase without scare quotes). Conservatives write in terms of just deserts – everyone getting what they deserve. But – at least rhetorically – both sides are reaching for the same item: meritocracy.
There’s been some discussion of its inherent contradictions, but the critique stretches as far back as John Rawls Theory of Justice, whose criticism may best be captured from Ben Bernanke’s recent speech at Princeton:
The concept of success leads me to consider so-called meritocracies and their implications. We have been taught that meritocratic institutions and societies are fair. Putting aside the reality that no system, including our own, is really entirely meritocratic, meritocracies may be fairer and more efficient than some alternatives. But fair in an absolute sense? Think about it. A meritocracy is a system in which the people who are the luckiest in their health and genetic endowment; luckiest in terms of family support, encouragement, and, probably, income; luckiest in their educational and career opportunities; and luckiest in so many other ways difficult to enumerate–these are the folks who reap the largest rewards. The only way for even a putative meritocracy to hope to pass ethical muster, to be considered fair, is if those who are the luckiest in all of those respects also have the greatest responsibility to work hard, to contribute to the betterment of the world, and to share their luck with others. As the Gospel of Luke says (and I am sure my rabbi will forgive me for quoting the New Testament in a good cause): “From everyone to whom much has been given, much will be required; and from the one to whom much has been entrusted, even more will be demanded” (Luke 12:48, New Revised Standard Version Bible). Kind of grading on the curve, you might say.
Andrew Gelman’s more (conceptually, not temporally) antecedent criticism of the subject is the instability of meritocratic regimens. That is meritocrats will naturally help their friends and family who will then earn more than they deserve thereby undermining meritocracy’s governing tenet of just deserts. But for sake of discussion, let’s simply assume this point away by stipulating permanently equal opportunity.
John Rawls fundamental concern is that meritocracy suggests the smarter, more attractive, or more charming individuals – deriving these characteristics entirely from genetic luck – deserve more and that this is not a just society. Indeed, the Rawlsian critique is reductionist to an extreme: even a man’s work ethic is luck, not virtue. For this reason, it is alien to the American psyche: we culturally idolize the rags-to-riches story of a boy working two shifts to put his way through college to happen upon a brilliant idea to take risks to become a billionaire. So he famously (though, in my opinion, tautologically and somewhat trivially) argued that the justest society is that in which we constituted a world without knowing our station therein – behind the proverbial Veil of Ignorance. Rawlsians augment said Veil with the “Difference Principle” asserting that inequality ought only to be tolerated to the extent that it benefits society’s poorest members (never mind how that’s defined).
Let’s assume the Rawlsian will is instituted through progressive taxation and social spending: without concerning ourselves with the specific – 39% or 35%? Income or estate? – details. Let’s further introduce IQ as some undefined variable that is a stand in for intelligence, charm, looks, character, work ethic, and basically everything else genetic but relevant. Finally, let’s consider how our policy’s must differ along some heritability factor of IQ between 0 and 1. In the former a child’s IQ is entirely uncorrelated with his parents, and in the latter it is of direct consequence.
At the lower extreme, it’s very clear progressive taxation and spending is morally just. If we assume an initial level of inequality of outcome at some previous point, the natural extension without social programs as an inequality of opportunity for the next generation. However, we believe it is immoral for children of the poor to suffer for being born in the wrong home. Furthermore, because IQ is entirely uncorrelated with parental genetics, we know that kids from poor families are as likely to succeed as those from rich families, therefore we must equalize the environment to the greatest extent possible. We must equalize inputs, but never outcomes.
At the upper extreme, it’s also very clear that progressive taxation and spending is morally just. Because IQ is perfectly predictable and we think in a rich society like ours one must not suffer for his genetic composition. That is to say, we must let someone with an IQ of 90 – low, but by no means disorderly – to suffer for his station no more than we let autistics on the streets. Both, obviously, are the result of genetic forces that conspire beyond individual agency. We must equalize outcomes, but never inputs.
Aside from the fact that optimal economic regime within either is brilliantly untenable (“From each according to his ability, to each according to his needs”, anyone?) the key argument here is the tradeoff between equalizing inputs on one end to outcomes on the other.
But this speaks only of the quality not quantity of redistribution. The crucial point is that if we can somehow, abstractly, quantify “redistribution”, the optimal level would be maximum globally at the upper end, minimum locally at the lower end, and bimodal throughout.
Why? At low levels of heritability, you don’t care about normalizing the outcome – only the initial environment (say we can discretely define childhood). At high levels of heritability, you don’t care about normalizing the opportunity. Note, this is a very different thing from saying we shouldn’t spend on education. All it says is that we shouldn’t care about educational inequality. Other things equal, more education is still better.
But redistributing opportunity naturally redistributes outcome more than redistributing outcome redistributes future opportunity. Therefore, the total quantity of redistribution needed increases from 0, but theoretically may fall in between. Diminishing returns on current opportunity along with activities that mutually redistribute both (public infrastructure, for example) cause a decrease in total redistributive spending necessary before it tends towards absolute socialism at absolute heritability.
The conceptual question then is a) in our society does the heritability-redistribution curve ever fall to zero – in other words is redistribution always required – and if so is heritability of the abstractly defined IQ at that point. (Note that the curve can never really fall below zero since redistribution is defined in terms of some percent of income from the top n% of the country, “top” and n selected arbitrarily).
The interesting result here is that – even within pretty liberal and Rawlsian assumptions of what ought to be – there can exist many states where redistribution isn’t important or necessary. I don’t have any way of judging this framework against empirical reality, but it is still pretty intriguing that the radical normative stipulations of Rawlsian thinking do not necessarily imply redistribution as an optimal policy.