Image credit: The Daily Star
“Poverty bothers me. Inequality does not. I just don’t care.” This terse statement from Willem Buiter, chief economist at Citigroup, nicely captures the prevailing sentiments of neoclassical economists. Talk of disparities in the distribution of wealth and income wreak of politics and power, of envy and conflict. Admitting that distribution is a concern might justify re-distribution, which would bring into question the neoclassical fantasy of self-regulating markets. But poverty is different. Focusing on the plight of the poor speaks to our altruistic side. And most importantly from the neoclassical viewpoint, poverty alleviation detracts attention from uncomfortable questions about how people at the top managed to make so much more than everyone else.
Yet the growing consensus outside of neoclassical economics is that inequality matters. And the reason it matters is because inequality is linked to social dysfunction. The grave consequences of inequality were explored most famously by two epidemiologists, Kate Pickett and Richard Wilkinson, whose book The Spirit Level provided systematic evidence that more unequal societies suffer from severe health and social problems, including higher rates of obesity and mental illness, higher crime rates and incarceration, lower educational performance, trust and social mobility. Perhaps the most important finding of Pickett and Wilkinson is that in rich countries these grave consequences cannot be reduced to concerns with poverty. In other words, beyond a certain level of development, there is no correlation between average standards of living and these same health and social problems.
The research of Pickett and Wilkinson should be enough to dispel neoclassical assertions about the unimportance of inequality. But unfortunately, if there’s one thing we’ve learned since the Global Financial Crisis, it’s that, despite its fundamental flaws, neoclassical dogma is remarkably resilient in the face of challenges. So I was also heartened to recently stumble across another critique from Branko Milanovic. In a journal article published well before his ‘elephant chart’ fame, Milanovic dissects the neoclassical arguments about inequality. What makes Milanovic’s approach so unique is that he develops a purely logical critique, one that uses the assumptions of neoclassical economics to undermine neoclassical arguments about inequality.
Milanovic starts with an example that the Harvard economist Martin Feldstein gave in a speech to try to justify the irrelevance of inequality. If we imagine that economists are each given $1,000 to attend an academic conference, this will, ceteris paribus, increase inequality in America. But this increase in inequality is of no concern because it’s Pareto Optimal: no one has been made worse off by this re-allocation of resources. Everyone’s income (and therefore their utility) has either increased or stayed the same. But then Milanovic offers a more extreme example: What if one of the conference participant was paid $20,000 and every other participant received 25 and 75 cents? In these circumstances, the income of all participants increases. But the other participants are likely to be deeply offended by the fact that one participant received such an astronomical amount, while they received a pittance. The outrage will likely run so deep that the disutility they feel from the discrepancies in honoraria will outweigh the utility gained from the 25 to 75 cent increase in their income. The lesson? Inequality matters.
But what if we introduce further nuance and assume that widely skewed conference honoraria are determined by position in the profession and the “expected quality of the paper” to be delivered? Milanovic maintains that under these circumstances, conference participants will inevitably compare their own honoraria with other participants to gauge whether their pay is similar to those at the same career level and with similar publishing records. Those at the bottom of the scale “might” be upset or even offended by their position. In this case, the pleasure derived from the modest income gain will be more than offset by the pain and humiliation of being at the bottom. So whenever someone else’s income enters into our own utility function we have to admit that we’re concerned with inequality. And as social creatures, we are never completely satisfied to judge our well-being in the absolute terms of neoclassical utility maximization. We always judge our own well-being relative to others (i.e. our “peer group”).
The social (relative) nature of our behaviour has been consistently demonstrated through experimental tests known as ultimatum games involving two people. In ultimatum games, one person is given a certain amount of money and then that person gets to decide how to divide the money with the other person. The second person then gets to decide whether to accept or to reject the first person’s offer on how to divide the money. If the second person accepts the offer, then the two players receive the money. But if the second person rejects the offer, then no one gets any money at all. And what these studies find is that most people tend to reject the offer when the first person decides to give themselves a share significantly greater than 50 percent. So when the second person is offered less, they will often reject the offer, so that neither participant receives any money at all.
Now, according to standard economic reasoning, this doesn’t make any sense. Even if the second person is offered only a penny, then they should accept the offer because it means that their income has increased in absolute terms. But the reason why people reject the offer is because they care about inequality. They feel a deep sense of injustice whenever the first person in the experiment gives themselves more, because they’ve really done nothing to earn a greater share. The disutility of the injustice outweighs the utility of the case payment (as a side-note, this disutility of injustice felt by inequitable distribution extends beyond humanity to other species).
With reference to neoclassical assumptions, Milanovic manages to undermine the neoclassical economists’ own arguments about inequality. But ultimately, this turns out to be a rather soft critique. With reference to the marginal theory of productivity, neoclassical economists are likely to counter that the above-examples of pay discrepancy boil down to the marginal contribution of the individuals involved . Give $20,000 to one conference speaker and 25 cents to another? The discrepancy must be because of the vast differences in their respective differences in their skills, education, talents, and experience. What matters is not the individuals own self-perception of justice or injustice, but the objective criteria of marginal productivity that governs the distribution of wealth and income. Unless the edifice of marginal productivity theory is dismantled — something that was already achieved decades ago during the Cambridge Controversies — then it’s quite easy for neoclassical economists to dismiss Milanovic’s criticisms.
But what is most persuasive and valuable about Milanovic’s logical critique is that it completely undermines neoclassical arguments about poverty. If we adopt the neoclassical arguments about the irrelevance of inequality, we have to inevitably concede that poverty is also irrelevant. If other peoples’ incomes don’t enter into our utility function from above (concerning inequality), then there’s no reason why they should enter our utility function from below (concerning poverty). Of course a neoclassical economist could always counter by suggesting that there’s some arbitrary threshold beneath which we start to care about the income of others, but anything above that threshold leaves us indifferent. But for a theory that prides itself on its supposed objectivity, introducing caveats of this type seem rather implausible. We’re more likely to be concerned about those that have more than us than those that have less.
My sense is that Milanovic’s logical critique still won’t be enough to change the minds of neoclassical economists. Nor will this logical critique provide much to those who already reject the neoclassical approach. But it might be enough to convince others, tempted by but not yet converted to the neoclassical faith, that there’s something inherently wrong with the dismissal of inequality within mainstream economics.