A couple of days ago, Brandon Berg posted on Catallarchy a response to a common hypothetical: what happens when a large percentage of a population becomes superfluous given much cheaper labor? Berg’s posting was in response to Michael at Half Sigma, who raised the possibility that cheaper sources of labor would cause a segment of the population to become permanent unemployable at almost any wage, rendering them "economically worthless." Michael and Brandon take this example to its logical conclusion by assuming that a class of "worker" exists which can beat *any* worker on price and management costs — in this case, robots. But of course we need not imagine actual robots here, the analysis is equally useful if one simply imagines very large relative differences between groups of people in their expected wages and working conditions.
This may seem like a bizarre science-fiction scenario, but of course it’s also a thought experiment for possible "halting states" of a globalized economy. For example, progressives in the U.S. wonder whether it’s even possible to ensure that we can roughly match the supply of jobs (let alone well paying jobs) to the domestic demand among American workers, or whether additional technological advances (in communications, supply, energy efficiency, etc) will further decrease the "friction" that continues to make globalized workforces a difficult thing to manage for most companies. Were that "friction" to lessen, it’s an open question as to whether we can count on there being a rough balance between labor demand and job supply. Massive domestic unemployment, or at the very least drastic "underemployment" at low wages, may result. The end result of such a scenario is a "third-world" economic structure here in the U.S. — a small population of wealthy property-and-company owners, and vast numbers of people who are chronically under- and unemployed, dependent for subsistence and survival. Clearly, this is a problem of political economy most progressives would wish to avoid, or if not, to remedy.
Not everyone agrees that such a situation is problematic, however — there is a strain of libertarian economics which seems to view such a situation as almost desirable if the economy is productive enough in aggregate. For example, examine Brandon’s response to such a situation:
We live in a world of scarce resources and unlimited desires….But let’s suppose for the sake of argument that scarcity is eradicated. Robots make everything, including other robots. They perform all manner of services as well as or better than humans, and no feasible desire, however trivial, goes unfulfilled. There’s no work left to do except to maintain the robots, and a small minority will be sufficient to handle that. I don’t see the problem. A world without scarcity would, by definition, be a world so fantastically wealthy as to allow the productive minority to support the idle majority at virtually no cost.
Note that a welfare state would not be necessary to achieve this. With no scarcity to temper the benevolence of the wealthy, pure charity would be far beyond sufficient to provide for the needs of the rest.
I suspect this argument is deeply flawed, even though it seems to be fairly common in libertarian circles. I see at least two flaws, although there may be more.
First, it ignores the role of social preferences in whether a small population of productive wealthy (or even non-productive wealthy) actually would be benevolent towards a large population of non-productive poor. Even if we assume that the wealthy in our experiment are wealthy enough that even large outlays of pure charity have a negligible impact on their individual assets, we run into the problem that "altruistic" behavior doesn’t seem to be governed purely by asset availability.
Indeed, the emerging subdiscipline of experimental economics has shown pretty conclusively that (1) most people in advanced industrial economies are much more generous (even to strangers in one-shot interactions) than rational choice theory would predict, but (2) most people are not unconditionally altruistic. Instead, by far the most common type of preference measured in experiments (running the Ultimatum, Public Goods, and Dictator games) is "strong reciprocity." In strong reciprocity, individuals are likely to cooperate (e.g., share the wealth, split a hypothetical payout, etc) when they believe the people with whom they interact will cooperate in return, and are likely to not just defect, but punish, those who do not (or have not in the past) cooperate in return. Interestingly, the exact nature of the preferences, and how such preferences are distributed within a population, depends critically on the nature of the economy within which people participate. Experiments among groups that participate in markets, whether advanced industrial economies or localized and primitive market structures, display a high level of "fairness" whenever subjects believe others will be fair in return or are "deserving." Experiments among groups that do not participate in markets and are largely reliant on individual effort for subsistence display much more self-interested (i.e., "rational" in the parlance of microeconomics) behavior. I’m summarizing a lot of literature very quickly here, so perhaps in future posts I’ll unpack this argument a bit. In the meantime, Heinrich’s book is an excellent place to start, as is the work of Herbert Gintis, Samuel Bowles, and Ernst Fehr.
What does this mean in terms of our thought experiment? Brandon states that a welfare state "would not be necessary" to achieve redistribution from a small population of haves, to a large population of have-nots. The experimental evidence, however, strongly suggests that whether Brandon is correct depends critically upon whether the "haves" can be made to see the "have nots" as "deserving cooperators" rather than free-loaders. What’s more, the security of the "have nots" will only last as long as the rich feel "reciprocity" with the poor. In the absence of economic roles which the poor can perform to demonstrate common cause and productive effort, it seems difficult to imagine that "pure charity" can be a stable long-term equilibrium, capable of sustaining a whole population. My personal conclusion from this is that Brandon and others who deploy this argument are very likely to be wrong in practice. The difference in public perception between "welfare" (which was commonly perceived as a haven for "freeloaders" in the U.S.) and Social Security (in which everyone reciprocates to everyone else, rich or poor) would seem to bear out this conclusion as well.
Second, even if we assume that the wealthy do "give big" and are unconditionally altruistic rather than simply strong reciprocators, we fall afoul of Hayek’s analysis of centrally planned economies. For what is a "pure charity" economy other than a centralized economy based on gifting? How are the small number of wealthy in this model to know how much each poor person needs, and when? How are the wealthy donors to coordinate with each other to ensure that each poor individual receives needed help, but not a double share, while leaving others without help? Obviously, a coordination mechanism is needed, and once we reach that point in the argument, we’ve fallen right into Hayek’s hands — from here, much of his analysis in "The Road to Serfdom" applies. (see note 1 below)
In other words, a world without scarcity doesn’t make problems of planning and redistribution irrelevant. A world without scarcity doesn’t make social preferences about cooperation, redistribution, punishment, and reciprocity irrelevant. We don’t escape the problems of contemporary political economy simply by "ratcheting up" the general level of wealth.
I see this as a core difference between "classical liberals" (e.g., libertarian believers in free-market dynamics over collective action), and "New Deal liberals" (e.g., progressives, modern liberals, those who believe in some type of "welfare state."). The former want to see market mechanisms and economic growth as the solution to problems of political economy and fairness; the latter don’t see such problems going away no matter how much we produce and grow — such problems are solved only through cooperative, collective action and incentive structures which take into account the real social preferences that characterize us as individuals (rather than applying rational choice theory uncritically).
In other words, a well-constructed welfare state is a sociopolitical structure which creates incentives for an entire population to act as generous strong reciprocators with respect to their countrymen. And the need for such a construct will be with us as long as major variability exists in the well-being of our fellow citizens, if we are to rely on more than just fantasies of "pure charity" and robot workers who will free us all from drudgery.
(more soon on how such a stance can be factored into institutional design for a "well-constructured" welfare state)
Note 1: (Those, in the comments to Brandon’s post, who mentioned Star Trek and "replicator technology" ending scarcity, should remember that the Federation is a quasi-military, somewhat authoritarian organization whose missions of "scientific discovery" are accomplished using heavily armed warships to enforce galactic "peace." Hayek would have hated Roddenberry’s Federation, likely preferring to live among the decidedly market-oriented Ferengi.