Tom Palmer, who shares with Brink the role of capo di tutti capo of the Cato Blog Mafia, has been giving a series of lectures on globalization at colleges and universities around the country. Out in that wilderness, capitalism and globalization (often prefixed with “corporate driven”) tend to be dirty words, though only for rather vaguely understood reasons. So it was that Tom encountered (and blogged about) the following challenge—

Objection:Capitalism requires the existence of poverty to function. Without a poor underclass to exploit, the wealth that capitalists brag about cannot be created. Globalization is all about finding new pockets of poor people to take advantage of. If poor people ceased to exist, it would be the end of capitalism as we know it.

Tom’s response, as you might expect from one of the more terrifyingly erudite members of our species, is spot on. Tom focuses mainly on the “exploitation” or “taking advantage of” end of it. But there are really a few distinct theses here. One is that the poor are taken advantage of or “exploited” in the sense that they are literally made worse off by capitalist exchange. We might call that the Strong Exploitation Thesis, and Tom deals with it succinctly. There’s also a less evil sounding version, what we might call the Weak Exploitation Thesis, formulated this way—

Weak Exploitation Thesis: Within the wide boundaries of some mutual benefit, the gains from trade can be distributed in any number of ways, perhaps as a function of relative bargaining power. The vast majority of the gains from trade are therefore captured by the affluent folks. The poor are made “better off” than, say, a condition of near starvation, but not better enough off to ever be anything but poor. There exist alternative mutually beneficial exchanges—different points on the same contract curve—that would leave the poor far better off.

This isn’t actually the one I want to deal with, so I’ll just say a couple of quick things. In principle, this might be an accurate description where you have only one employer, or a small number and more workers than those firms can accomodate—the conditions for real bargaining power. But that describes a situation of insufficient globalization and capital availability, and is remedied as globalization accelerates. There’s also the possibility that the employers have lower transaction costs to collusion (de facto equivalent to monopsony), but that’s a problem with the local legal institutions, not markets per se. And as an empirical matter, as Tom notes, plenty of countries which were sources of cheap labor have grown more affluent… some of them (say, Japan) very much more affluent.

The other part of the objection is what we might call The Instability Thesis, which allows that trade does lift the poor out of poverty, but adds that once we’ve run out of poor, the system can’t sustain itself. We’ll formulate it this way—

The Instability Thesis: Grant that the poor are made better off by trade, maybe even enough that they cease to be poor. Nevertheless, some poor or underclass are required by the capitalist system. If they are ever made better off enough that they all cease to be poor, or that a critical mass cease to be poor, then the capitalist system will come crashing down, like a gleaming plane that’s burned up all its fuel.

To really get at this one, though, we’ll have to tease out in a bit more detail why anyone might suppose The Instability Thesis to be true. I’m more or less just taking a stab at this one, based on vaguely recalled late-night dorm-room talks with socialist acquaintances…

If we take “capitalist” and “worker” to be pure types—such that everyone either makes all their income just by forsaking present consumption of accumulated wealth and collecting returns, or by wages for labor—then not everyone can be a capitalist, at least until we build much better robots. Efficient resource allocation is part of wealth generation, and there is pure gains-from-trade wealth generation, which just means we shift consumption goods around such that the new distribution is Pareto-optimal. But someone actually has to be physically doing something with those efficiently-allocated resources for genuine production to happen. A world in which everyone makes a living farming is conceivable, if unappealing. A world in which literally everyone makes their living like Warren Buffett is not.

This is true, of course, but trivial. Not everyone could be a doctor, novelist, or trucker either, since at minimum you need people who grow food (unless everyone also tends their own little farm, which is clearly pretty inefficient). It proves that capitalism requires workers, though those workers may also be capitalists if they invest part of their wages, as (e.g.) many American workers do via mutual funds and pensions and so on. The leap to the conclusion that capitalism requires poor only works if you assume that being a worker (who may also be part-capitalist) entails being poor. But there’s no justification for that move. Plenty of very wealthy people are “workers” in the sense that their wealth, at least at first, came in the form of wages or (for entertainers and artists) royalties, rather than return on capital investment.

I imagine that the way the proponent of the Instability Thesis gets to the idea that we need, not just workers, but poor workers, probably runs something like this: maybe you could have affluent “workers” who are professionals—lawyers or engineers, say—but there are plenty of industries (think sweatshops) that run on cheap labor. The only reason people are willing to spend their time hand-stitching clothing at the abyssmal wage that lets us consume cheap garments is that they’re in a desparate situation, such that even this raw deal is better than the alternative. If nobody were poor, nobody would be willing to work at the low wages that make companies like Nike and its ilk profitable. Those firms couldn’t stay in business without a supply of people in desperate poverty who are therefore willing to accept low wages. They’ve either got to be kept poor, or the whole thing comes tumbling down.

This is also very wrong, but it’s not as obvious why unless you think in terms of process, rather than in terms of static snapshots. Imagine an oversimplified case, where the cheapest labor in the economy is employed making Widgets at various different firms. Now, in plenty of cases, if wages started rising, you might have all kinds of shifts in production, such that (say) it suddenly became efficient to have fewer skilled workers at higher wages, and a shift to productivity enhancing machinery or something like that. But assume that’s not the case, that widgets just need to be made by hand. Now, the Widget industry might find that as wages are rising across the economy, they can’t afford to keep people on at the same low wages. The death-of-capitalism scenario here is that all the various widget industries promptly go under; fade to black.

But what would it mean for this to happen? How would it happen? Well, for the rising-wages condition to be satisfied, we’d have to presume that through specialization, technological innovation, education, and all that other good stuff trade brings, productivity was rising in other sectors of the economy. The labor of the Widget-makers has a more highly valued use elsewhere in the economy. But demand for Widgets is elastic: as the economy becomes more productive, meaning people are getting richer (producing more stuff with the same inputs), the demand for Widgets could well increase, enabling a higher wage to be paid. Again, to use a simpling example, if we imagine a small, self-contained community, farming technology might get good enough that there was plenty of food. With that surplus, people could express demand for, say, beaded jewelry made by others (ex-farmers, presumably) where there was no such demand before. Supply, as the economists like to say, creates its own demand. The richer and more productive we are, the greater the effective demand for (ability to pay someone to do) whatever we most want done.

In any particular sector, of course, that might not happen. In poorer societies, it’s common to see middle class families with multiple servants. You see fewer servants in more developed economies because higher labor value makes it more expensive to hire people as servants, and demand doesn’t keep pace accordingly, because people find they’d rather spend their limited resources on the ever greater variety of consumer goods the same labor can now produce. By the same token, it might be that people no longer want Widgets at the lowest price they can be sold given the wage required to get workers to make them. But that can only be because there are other burgeoning industries where productivity is high enough that people find its a better deal to spend their money there. Once you actually try to conjure a scenario where there’s some real, plausible process of “running out of poor people,” it becomes clear that the instability thesis doesn’t make any sense. It requires you to believe that everyone is getting much richer—i.e. that the return to labor is growing—but that nobody is willing to pay any more (of their larger income) for anything—i.e. that the return to labor isn’t growing anywhere. Now, call me wacky, but I operate on the principle that if a theory asks you to simultaneously believe a proposition and its negation, it’s time to start shopping for a new theory.

For an entertaining explosion of a related fallacy, by the way—one that became the subject of an absurd book called One World, Ready or Not—check out this article by Paul Krugman.