Thursday, June 14, 2012

How much inequality?

How much inequality is too much?  Answers range from Gracchus Babeuf (all inequalities are unjust) to Ayn Rand (there is no moral limit on the extent of inequalities a society can embody). Is there any reasoned basis for answering the question?  What kinds of criteria might we use to try to answer this kind of question?

John Rawls offered a fairly simple criterion of the extent of inequalities that a just society can tolerate in A Theory of Justice.  His background assumption is that the wealth of a society is a joint product of all members of society.  The rules of distribution create more or less inequality among citizens.  Citizens are concerned to "protect" their interests in case they wind up being in the worst-off positions in society.  So justice requires that institutions should create the least degree of inequalities subject to maximizing the position of the least-well-off position in society.  (He also stipulated an equality of opportunity principle: positions need to be open to all through a fair system of competition.)  If the empirical theory is true that economic inequalities sometimes create more wealth (through incentive effects), then it is just to increase inequalities up to the point where any further increase would leave the position of the least-well-off member of society unchanged.  This is the least system of inequalities subject to maximizing the position of the least-well-off.  Rawls justifies this principle (the difference principle) on the ground that it expresses an important sense of fairness: everyone is fairly treated when higher incomes are assigned to some individuals only when doing so benefits all individuals.

By this principle, current inequalities in the United States and Great Britain are demonstrably unjust: it would obviously be possible to lower the incomes of the 1 percent without lowering the income of the bottom forty percent.  So current inequalities are plainly more extensive than they need to be for the purpose of increasing overall output and improving the condition of the least-well-off.

Joseph Stiglitz offers a different kind of theory of inequality in his recent The Price of Inequality: How Today's Divided Society Endangers Our Future. (Here is the anchor essay as it appeared in Vanity Fair; link.) His argument is a pragmatic one based on a theory of social stability.  Essentially he argues that when inequalities become too extreme in a given society they breed social conflict and instability. This happens perhaps because of raw deprivation -- the bottom 40% really do have pretty desperate lives -- but more because of what Stiglitz identifies as erosion of the social contract.  Here is how he puts the point in the Vanity Fair article:
Of all the costs imposed on our society by the top 1 percent, perhaps the greatest is this: the erosion of our sense of identity, in which fair play, equality of opportunity, and a sense of community are so important. America has long prided itself on being a fair society, where everyone has an equal chance of getting ahead, but the statistics suggest otherwise: the chances of a poor citizen, or even a middle-class citizen, making it to the top in America are smaller than in many countries of Europe. The cards are stacked against them. It is this sense of an unjust system without opportunity that has given rise to the conflagrations in the Middle East: rising food prices and growing and persistent youth unemployment simply served as kindling. With youth unemployment in America at around 20 percent (and in some locations, and among some socio-demographic groups, at twice that); with one out of six Americans desiring a full-time job not able to get one; with one out of seven Americans on food stamps (and about the same number suffering from “food insecurity”)—given all this, there is ample evidence that something has blocked the vaunted “trickling down” from the top 1 percent to everyone else. All of this is having the predictable effect of creating alienation—voter turnout among those in their 20s in the last election stood at 21 percent, comparable to the unemployment rate.
So for Stiglitz, inequality is not simply a moral issue, but is also an issue of social stability and cohesion.

Here is another way of answering the question: perhaps extreme inequalities are actually bad for a population's health.  There is one sense in which this is obviously true: extremely poor people have worse health outcomes than affluent people.  But maybe the simple fact of inequalities is itself a toxic influence on public health, for poor and rich alike.  This is the argument that Richard Wilkinson and Kate Pickett make in The Spirit Level: Why Greater Equality Makes Societies Stronger.  Their argument is a complex one that suggests several causal explanations for why this might be the case; but their core idea is that great inequalities create widespread "stress" that is harmful to each individual's health.  Wilkinson and Pickett are public health experts and their argument is a cross-national statistical one.

So we might adapt their analysis into an answer to the question posed above: inequalities should be reduced until there is not measurable harm to public health.  (These arguments are considered in earlier posts.)

There is even a utilitarian argument for greater equality of income.  If we accept the plausible point that income makes a diminishing marginal contribution to happiness as income rises (perhaps after some inflection point), then under many realistic circumstances a more equal society with lower GNP will have a higher level of happiness than a more unequal society with higher GNP.  Lower-income people have more to gain from an additional $1000 of income than higher-income people.  So if we think it is a good thing to maximize happiness, then we ought to regulate inequalities accordingly.  Benjamin Page makes some of these arguments in an Institute for Research on Poverty position paper, "Utilitarian Arguments for Equality" (link).

Are we forced to choose among these frameworks in order to conclude that our economy has generated vastly too wide a set of inequalities of wealth and income?  No, we aren't, because these arguments are mutually supportive.  A system of greater equality, regulated by a tax system designed for that purpose, would be more fair, more supportive of equality of opportunity, more harmonious, likely healthier, and likely happier.  We have lots of good reasons for preferring greater equality of wealth and income and lots of good reasons to reject the "anything goes" philosophy that currently drives much of the political discourse on the right.


zopolan said...

"If we accept the plausible point that income makes a diminishing marginal contribution to happiness as income rises"

The problem is that this requires interpersonal comparison of utiltiy between individuals, that is, cardinal utility.

Dan Little said...

Zopolan, that's true. But rejection of interpersonal comparisons is part of the baggage of behaviorism and skepticism about other minds that led several disciplines astray in the first place. Isn't this the heart of Sen's critique of revealed preference theory? Doesn't the point made in the post about the relative beneficence of $1000 for the millionaire and the pauper seem obviously true?

Noni Mausa said...

Another effect of inequality can be seen now -- it creates two separate economies, like a dumb-bell with one tiny and one huge weight on each end.

In one, consumption and the use of credit (money, goodwill, or a credible promise of future effort) is constrained, and building up savings is difficult. Deflation, formal or informal, will take place.

Informal? In the sense of abandoning some items in favor of lower priced equivalents. These informal strategies have been on the move for at least 20 years.

For instance, thrift shops are booming even among the middle class, yet in the 70s, they were the preserve of the poorest of the poor. So, though prices in each shop may be sticky, people deflate by stepping down to other shops.

By contrast, at the high end inflation will appear to be flat (because the cost of essentials is held down by the purchasing capacity of the majority) but will show up in unique, artisanal and boutique items like land and art and high-end professional services.

This would be okay, if the two economies weren't linked. But in fact the apparently bigger economy is wholly dependent on the smaller one, couldn't exist without it. The smaller could easily exist without the larger.

The wealthy, being few, cannot contribute a share of the effort needed to sustain the whole system in proportion to their benefits from that system. Their goodwill, skills, training and work-ethic are nothing compared to the weakness of their low numbers.

Intelligent persons in both economies can't help but figure this out, (it's not rocket science, for heaven's sake.)

So increased inequality must result in increased conflict and apprehension.

Beyond some tipping point, a large disparity will have certain predictable, destructive results. At present the US is only showing a tinge of these, compared to fully developed plutocracies, but they are unmistakable.


Zlati Petrov said...

Which inequality though?

Social status inequality? Wage inequality? Cash income inequality? Consumption volume inequality? Consumption quality inequality? Capital ownership inequality? Political power inequality?

Are we sure that we can talk about the justice (and injustice) of inequality without first defining the concept?

And are the various types of inequality similarly linked? For example, Gary Becker has argued that social status and consumption volume are adjacent complements.

Others may argue that social status results from income, but that social status is a social construct. And if social status matters more than income inequality, we can have income inequality but reduce social inequality by affecting change in the way society perceives and assigns social status.

And what if we cannot reduce all types of inequality at the same time? At least some commentators seem to presume that reducing one type of inequality (income) may reduce all others. But can they prove that?

For example, what if reducing income inequality via tax transfers increases social status inequality because society cleaves between those that receive tax transfers and those that provide them?

Finally, what if we cannot reduce all inequalities at the same time?

Which should we reduce first? Political power inequality perhaps? Maybe that will allow us to reduce other inequalities more easily.

Or maybe not...

Anonymous said...

As everyone knows, we need some inequality in order to provide incentives. This suggests that incentives are fundamental, and inequality is a means to an end.

If we think about the need for incentives, it becomes apparent that incentives should be available to everyone. For example, a higher future salary is an incentive to go to college, but that incentive is available only to those who can afford to forgo four years of wages.

So beyond a certain point, inequality reduces incentives for a large fraction of the population. This refutes the naïve view that more inequality implies greater incentives, which implies greater productivity.

If this argument has merit, it has been said before. If someone knows of other sources of this argument, please reply to this comment.

Dan Helphrey said...

I would argue that it is rapid change in the level of inequality, rather than the particlular level at any one time, that creates a social problem. People have, throughout history, adapted to living in societies under a broad range of inequality levels; it is when we see all the benefits of economic advace accruing to the group at the top, so that they are increasing the gap between themselves and the rest of us, that we begin to question why we should keep rowing their ship for them.

Magpie said...

I find it a bit odd, in discussing the subject of inequality in a social-sciences oriented blog, the assumption that income/wealth/consumption inequality is a social given (something inevitable, part of “human nature” or even possibly positive, “as everyone knows”).

That assumption is quite clear in this discussion, while remaining apparently imperceptible to the discussants.

But let’s considers the history of human societies. Our species has existed for between 150K to 200K years, most of that time as hunter-gatherer bands. During this hunter-gatherer stage, people worked and made a living without having to sell their labour or, indeed, essentially without possessions.

By comparison, the wage relationship is much newer. As the commonest form of making a living it is no older than two to three centuries. Before, there was only sporadic wage work for a couple of thousand years (for instance, mercenary troops). And that’s the history of wage work, from where most of our income comes.

(Incidentally, the very word “salary” has a Roman origin: it was the pay of professional Roman legionnaires, often made in species: salt).

How can income inequality be a fixture of human society? Or, to use the universal expression: human nature?

Similar happens with wealth. The notion of private property, as we know it now, hardly goes beyond a thousand years. And even if we are extremely flexible in what we consider private property, it can hardly be found beyond 3000/4000 years ago. Certainly, not before the Neolithic.

In fact, well-known hunter-gatherer societies, in historical times, have hardly had any property: the land, if it could be said that it was a property, was a communal property; it certainly did not belong to any individual. As hunter/gatherer bands are general nomadic, they lack permanent structures; when a temporary structure actually exists, it is often also for communal use.

I am afraid there is a lot of ideologizing taking place in this discussion.