What, Me Worry… about Inequality?

Ezra tries to explain to me why “liberal” economists are worried about inequality as opposed to some people's not having enough to lead a good life. Maybe it's a good sociological explanation. I don't know. But I'm afraid Ezra's explanation vexes me. Here's Ezra:

What concerns liberal economists is the relative apportionment of income. Inequality is something of a proxy for this. Take the so-called Krugman calculation which, in the early '90s, showed that 70 percent of the post-1973 rise in incomes had gone to the wealthiest 1 percent. As he put it, “when incomes at the top of the scale are rising faster than the average, incomes farther down must correspondingly grow less rapidly than the average. In an arithmetic sense, we can say that most of the growth in productivity was “siphoned off” to high-income brackets, leaving little room for income growth lower down. “

What we're worried about is what's called the “inequality wedge.” The gap in incomes is so vast and the pool of money at the top so great that growth, which once would have rushed all the way down the income ladder (rising tides, all boats), isn't making it down the distribution, instead clogging up at the top percent or two (rising tide, only yachts).

Sadly, this makes almost zero sense to me, no doubt due to a Hayek-Buchanan-Nozick gestalt switch about “distribution” that I underwent years ago, forever crippling my ability to think like, well, the staff of the American Prospect.  Ezra's picture seems mysterious to me. It looks as if, first, there is something called “the economy.” Then there is something called “growth,” which is the size of the economy getting bigger. And then there is a matter of how to divide or share this year's slightly larger pool of wealth (a gigantic pile of doubloons in the basement of the Treasury?) through the “apportionment” of income. I'm sorry, no.

What there is is an ongoing dynamic pattern of exchange, within and across national boundaries, that creates new wealth. For each exchange among the billions that create the overall pattern (labor for money, money for interest, money for capital goods, money for consumption goods, etc.) there is a surplus from exchange and a distribution of that surplus among the parties to the exchange. We can count what everybody within some semi-arbitrary geographic region gained or lost in money terms from all their exchanges between period one and period two. But there is no further question about distribution. That was all already settled when the exchanges were completed. If there was no worry for the interested parties about the distribution of the surplus of each exchange, then it is hard to see why there is a worry later about the aggregates. 

Here's why I might worry about intra-national income inequality.

Nation-states aren't completely arbitrary regions from an economic point of view because the policies of national jurisdictions change the relative price of various forms of exchange for the people living within those jurisdictions, and this partly (sometimes largely or completely) determines what particular exchanges will and won't get made. If you see a lot more exchange, or a lot more complex exchanges, going on within a national jurisdiction, that probably has something to do with the rules that govern the jurisdiction. If certain jurisdictions include a lot of people involved in a lot of inter-jurisdiction exchanges, then that says something good about the rules there.

Now, the rules that help determine what exchanges will and won't be made—policy, “institutions,” etc. — will have a definite effect on the relative size of incomes within the jurisdiction.

Let me say that I'm not going to say anything about the “distribution of income” here, because people too often equivocate between the statistical and disbursement senses of “distribution.”  Here's how I think about it. Distribution is a matter between parties of an exchange, not between co-members of a political jurisdiction. Further, your money income in a period is the sum of your money distributional shares from all the exchanges closed in that period. It is a straightforward fallacy of composition to think of that sum as itself a distributional share. So, when we're comparing incomes of co-members of a political jurisdiction I'll say we're talking about  the disposition, not the distribution, of incomes in that jurisdiction.  You are of course free to think about the distribution of income just as long as you promise to think of “distribution” in a strict statistical sense, and not as the “apportionment” of individual shares from some mysterious pooled “national income,” since that is voodoo gibberish on par with “intelligent design.”

OK! A good example of how policy affects the disposition of incomes is the disemployment effect of a high minimum wage. People whose labor is worth less than the price floor will be excluded from exchanges of labor for money, and so will have very little money income at the end of the measured period. Or, to take another example, a big business-backed regulation or legal action that pushes smaller competitors out of the market, is generally meant to ensure that the big business will be party to a larger portion of the exchanges in that domain (and if they achieve monopoly-ish status, a larger take of the surplus), so that its owners and officers will have a bigger income than they would have had in a competitive environment.       

Now, it seems pretty likely that greater inequality in the rate of income growth (a different thing than income inequality per se) for people at the top and middle of the income scale might reflect a change in exchange opportunities and the distributional terms of those opportunities for people at different points of the scale. I am worried just in case this is a consequence of rules like price controls or anti-competitive regulation that create a pattern of exchange that is distorted compared to the pattern that would have emerged in the absence of such exchange-restricting rules.

Now, I don't find it totally implausible that the decline of unions as an agent of political predation for a good chunk of the middle class has decreased the rents accruing to many middle class workers. But that's good. And I don't find it implausible that some corporations have become more efficient at gaming the regulatory and government contracting processes to increase their political rents. That's bad. So a combination of that good thing and that bad thing could be part of the inequality story. But in that case, I'm not worried about the inequality per se; the justice or injustice of the rules generating that kind of inequality doesn't flow from the inequality itself, but from interference or non-interference with the moral right to enter into voluntary exchanges with others on mutually acceptable terms.

If inequality in the rate of income growth is just a function of a change in the market value of capital, human or otherwise, then I can't see the normative upshot. If the payoff t
o investment and education has gotten bigger, then that's just a good incentive for people to get more edu
cation and save more. Now, if some people don't have access to an adequate education (and the public school system ensures that many people—urban minorities especially—don't), then they don't have enough, and that worries me plenty. And if unwisely anticipated Social Security and Medicare benefits suppress savings, such that people are investing too little, and missing out on potential returns to capital, that would ensure that they have enough in retirement, then that's a big problem with the jurisdictional rules of the game, and we ought to worry. But the issue just isn't inequality.  

Well, that's a mouthful.

Philosophical wrap up… My view is that materially egalitarian liberalism is not really a form of liberalism at all. It is a socialist corruption of liberalism. Liberal egalitarianism properly concerns the equality of political power—none has a natural right to rule, and unequal political power requires special justification and special limitation. In terms of the material resources we command, liberalism is concerned that we are equal in the sense that everyone's liberties have genuine value—are not “merely formal”—and this is a matter of people having enough to develop their capacities and to realize their meaningful ends. Property, rule of law, civil society, and free exchange in a well-functioning price system—backed up if necessary by minimal means-tested welfare and educational assistance—is the best way to make sure everyone has enough.  

So, I still don't understand why welfare economists should worry about inequality, much less liberals.      

[Note: By the way, the Krugman quote Ezra provides is extra-mysterious. Yes, if somebody is above average, then somebody is below. Logic! But I am not familiar with this “siphoning” in an “arithmetic sense.” Suppose height is growing fastest in the top 5% of the height distribution. Are those tall people “siphoning off” height-growth lower down the distribution? What is Krugman even trying to say? I don't know, but I suspect he is managing to confound different senses of “distribution” without even using the word.]