In a recent contribution to an excellent volume on histories of global inequality (edited by Christian Olaf Christiansen and Steven Jensen) Eli Cook in a piece entitled “Historicizing Piketty: The Fall and Rise of Inequality Economics” discusses the reasons why inequality has been practically exiled from the mainstream economics during the entire second half of the 20th century. Cook’s is a very nice exercise in the history of economic thought that shows how a topic that was once at the center of economists’ interests (it suffices to read only the first paragraphs of Ricardo’s Principles to see that) was gradually shifted to the margins, so much so that Martin Feldstein, one-time president of the American Economic Association, could declare that all concerns with income or wealth inequality are the product of “envy” and “spiteful egalitarianism”.
According to Cook there were three developments that led to the loss of centrality that the questions of distribution originally had in economics: theory of marginal productivity of the factors of production, the turn toward utility, and the Pareto optimality.*
What pervades these three developments and is, in my opinion, the key issue is whether production and distribution are seen together as one process, or are seen as two, loosely related, processes. The latter approach is characteristic of neoclassical economics. Production is considered as a prior and economics is regarded as a science that ensures maximization of production. That output can be later, using political decisions, redistributed to help those who are poor but one has to do it carefully in such a way that the next round of production is not adversely affected by the wrong incentives coming from too much redistribution. In such a view of the world, almost any redistribution is seen as injurious to the process of maximization of output.
But perhaps more fundamentally, redistribution is regarded as lying outside of economics, in the political domain. This is, as Cook shows, clearly enunciated by Samuelson in his Economics. Economists thus appear as seemingly modest in their claims. Like the engineers of societal production they are in charge of output maximization under conditions of given endowments and technology. Self-denyingly they leave the task of redistribution to those who are more qualified than them: to the politicians.
Yet as economics has gradually come to dominate social sciences and government decision-making, this seeming self-restraining had come to be seen from what it is: an attempt to ignore as many of distributional issues as possible. If redistribution is the province of the politician, and the politician should be above all else an economist, then clearly redistribution should be an exception to be used sparingly. Cook rightly emphasizes the role played in this by the so-called First and Second Welfare Theorems. The First shows that, under idealized conditions of perfect competition, market produces the outcome that cannot be improved without making somebody else’s condition worse. This is economics. But it is then argued, in the Second Welfare Theorem that, if, for some reason, we change endowments or resort to lump-sum taxation, the distribution of incomes can be altered (although it would still remain Pareto optimal). But that second part, it is averred, is not the topic of economics but of political scientists or politicians. So the economists can cheerfully ignore it and make all distributional concerns secondary or marginal.
The classical view of economics was rather different. It saw production and distribution as a single process. If endowments were differently distributed, the structure of production would be different, and the power of various classes would be different. It is very clear in Ricardo that landowners receive their income solely based on a monopoly over land, not thanks to any useful activity they perform. So changing their endowment or taxing them cannot be bad. (Of course, the same view is present in Henry George.) It is also clear in Marx that under different modes of production, the structure of production, relative prices, and individual incomes would be different. Distribution of endowments and the way production is organized are thus organically linked.
While neoclassical economics envisages the economic world as:
Production => distribution => production
classical economics sees it as:
Distribution of endowments => production => re-distribution of endowments.
It is for that reason that in Marx (and of course among the Neo-Ricardians from Sraffa onwards) factor prices are seen as being determined prior to production (say, through relative power of labor vs. capital). This in turn makes the composition of output different under different systems: if workers are more powerful, wage rate will be higher relative to interest (profit), and labor-intensive commodities will be more expensive etc. Or similarly, as Marx says in his famous paragraph, relations of production become forces of production: if relations of production (basically, the distribution of endowments) are, at a given stage of development, inefficient (say, slavery leads to a waste of effort), maximum output that can be achieved under such a system will be less than what can be achieved under a more efficient social system. Production is thus seen as fully interdependent with distribution. The neoclassical idea that production and distribution can be neatly separated and considered almost in isolation from one another is exploded. This is the crux of the matter, and I think the reason for the divergence between the classical and neoclassical schools in their views of, and interests in, inequality.
Luckily, things are changing—but not as fast as they should. We still do not have textbooks or courses that deal with inequality of income or wealth as such. Inequality is often seen as an anomaly or a problem that may be relevant for the “Third World” societies only. This is so obviously wrong that insisting on how wrong it is, is almost superfluous. But so long as this is not fixed economics will continue to remain divorced from real life.
* It is interesting, and perhaps a bit ironic, that that the Pareto optimum which strictly speaking rules any redistribution (since any redistribution of received incomes must make somebody worse off) was defined by the same person who introduced the empirical study of interpersonal inequality.