Matt Zwolinski recently discussed arguments for a Basic Income Guarantee (BIG) at BHL and Libertarianism.org.  One of those arguments was that a BIG might be required on libertarian grounds because private charity will be insufficient. However, there are good reasons to believe that private charity will not be insufficient. For those interested in a longer version of the arguments I set out here, see my Is the Welfare State Justified? (Cambridge University Press, 2007) chapter 6, section 6. (An earlier version is “Egalitarianism and Welfare State Redistribution,” available here)

1.    What does “enough” mean?

One thing it doesn’t mean is that the amount of private aid must be equal to the amount presently spent on state welfare. As Matt noted in his post, state welfare programs are an inefficient, byzantine mess. Private charities will be far more efficient than state welfare, and thus will not have to match the quantity of state welfare.

The reasons for this comparative efficiency are no doubt familiar to readers of this blog. Private charities have more freedom than state welfare programs to treat those receiving aid as individuals and to target their aid to specific groups with very specific problems. They are superior in their ability to change policies if need be, since they have to jump through fewer hoops or go through fewer intermediaries than a political system does in order to try a new approach. Private charities are superior to state welfare systems in evaluating and obtaining information about whether its goals are being met. It is easier to determine how a policy is working when it is for a small, specific group rather than for a larger, heterogeneous group. Furthermore, private charities are more closely monitored than state welfare programs; voluntary donors have a stronger incentive to evaluate the charities they fund than rationally ignorant voters have to monitor the programs they are taxed to support.

Not only are private charities more efficient than state welfare, but in a libertarian society it is less likely charity will be needed, since that society will lack  barriers to the poor’s advancement that exist in our crony capitalist or interventionist economic system–occupational licensure, minimum wage laws, zoning laws, oppressive levels of taxation, etc.

But if private charities need not match the quantity of state funds for the former to be ‘enough’ or falsify the claim of ‘insufficiency,’ what should enough mean? Since to be a BHL means, roughly, that a central justification for libertarianism is that it is good for or serves the interests of the poor and vulnerable, then private charities would fail to be enough if they didn’t do that. Of course that’s vague, but at the very least we would know private charities were insufficient if the amount provided were so stingy that would enormous suffering would occur—widespread malnutrition, large numbers of people without shelter, etc.

2.    Empirical considerations

At first glance, it might appear that there is no empirical evidence to evaluate a claim of insufficiency, since even in the US and the UK, there was never a period when state welfare was completely absent or abolished. However, the US in the late nineteenth century provides us with something very close to a natural experiment for testing the claim that private charity would be insufficient absent state welfare, because from the mid-1870s until around the turn of the century, unconditional aid to able-bodied needy people—“outdoor relief,” as it was called—was either abolished or curtailed drastically in large, and some medium-sized, cities. Since organized charities kept fairly detailed records of their activities, we can see whether the claim that private charity alone would be insufficient is historically accurate. It does not appear to be. In almost all of the relevant cities, private giving rose to the occasion, and the amount contributed was roughly comparable to the amount given by outdoor relief.

Of course, that individuals seem to have risen to the occasion a century ago does not prove that this would occur today. However, there is a more general argument that explains why the results of the late-nineteenth century United States should not be too surprising. Government welfare tends to crowd out private giving; so when government welfare is (nearly) abolished, we should expect crowding in, that is, people react to the absence of government welfare by increasing their donations. And that’s what appears to have happened.

A natural response to this argument is that we cannot rely on crowding in. Even if it is plausible that the abolition of state welfare would increase donations, the increase may not be sufficient in today’s circumstances. True, it may not be. The question, though, is this: if history does not support a claim of insufficiency, and if the crowding-out that occurs because of state welfare suggests that some crowding-in will occur when such welfare is absent, why think that private charity will be insufficient? Now we are at the point where we can look at the argument Matt cited for the insufficiency claim, which is from chapter 9 of Milton Friedman’s Capitalism and Freedom:  “[W]e might all of us be prepared to contribute to the relief of poverty, provided everyone else did. We might not contribute the same amount without such assurance.”

3.    Public good arguments

David Friedman, in his reply to Matt, describes the argument of his father as follows: “charity faces a public good problem.” Strictly speaking, his father does not say that in the passage Matt quotes: what Milton Friedman argues is that contributing to charity (relief of poverty) faces an assurance problem. To show the existence of a public goods problem one has to show, first, that there is a genuine public good, and second, that voluntary provision of this good means it will be underfunded. The usual way of showing the latter is to argue that because of free rider and assurance problems, not contributing to the funding of these goods is a dominant strategy.

At first glance, voluntary provision of charity is a terrible candidate for a public good. A public good is nonexcludable and jointly consumed. But if what one values is one’s own contribution to charity and the benefits that accompany this (the psychological benefits from helping others, the sense that one did the right thing, etc.) by definition failure to contribute excludes one from this good.

One can get out of this problem by assuming that what potential contributors value is simply that charity be provided, and that they are indifferent between whether it is themselves or others who provide the contribution. If a significant number of potential contributors feel this way, then it appears that for them contribution to charity is a public good: noncontributors cannot be easily excluded from enjoying the benefits generated when others provide charity, and one’s enjoying those benefits does not seem to diminish others’ enjoyment.

But for such contributors, do we have an assurance problem and a free rider problem? Milton Friedman’s argument that there is an assurance problem is flawed. As Robert Nozick pointed out in Anarchy, State, and Utopia, there is an assurance problem only if a potential donor does not value his contribution (or values it less than its cost) if it fails to produce, in conjunction with others, a sufficient amount of charity. It is this assumption that supports the claim that a potential donor will prefer withholding his contribution if he thinks enough others will not contribute; if he thought that there was some net value in helping reduce poverty or disadvantage even if a sufficient reduction was not achieved, he would contribute even if others did not. It is hard to see, however, why a potential donor of this type would place no or almost no value on the partial reduction of poverty. Even if someone thinks that one should address the whole problem, this implies not that addressing the parts has virtually no value, but only that doing so has less value than addressing the problem in its entirety.

Thus, it seems that there are good grounds for contributing to charity even if one thinks that others will not adequately contribute. If this is so, then there is no dominant strategy here: I will not contribute (that is, I will free-ride) if others give a sufficient amount, but I will give if others do not. In game-theory parlance, we have a game of “chicken” here. There is no settled view about what strategy is rational in a game of chicken, but noncontribution is clearly not a dominant strategy.

Furthermore, a plausible case can be made that just as there really is no assurance problem, there really is no free-rider problem, either. This is because it is not obvious what amount of charity is “sufficient,” and therefore one should probably reason as if providing some sufficient amount of charity is not a real option. In these circumstances, contribution becomes a dominant strategy—one gives because one is never sure that others have given a sufficient amount, and one values the bringing about of a partial reduction of poverty or disadvantage.

4.    The burden of proof

Since neither the historical evidence nor the public-goods argument supports the claim that private charity would be insufficient were state welfare abolished, and since the sensitivity of private aid to the amount of state welfare provided suggests that private aid would rise if state welfare were ended, it is hard to see what basis there is for claiming that private aid would be insufficient in the absence of state welfare. Of course, I haven’t proved that private charity won’t be insufficient. Indeed, I doubt that this is the sort of thing one can prove. But unless there is some obvious argument I am overlooking, I think the considerations offered here put the burden of proof upon those who argue it won’t be sufficient.

***********************************************

A few qualification, caveats, etc.

#1: What occurred in the late nineteenth century in the US was very close to a natural experiment regarding the claim of insufficiency. Although unconditional government aid was abolished or drastically reduced in large and some medium size cities, not all forms of government welfare were abolished. Prior to the rise of an extensive welfare state at the national level, government aid for the poor in the UK and the US consisted of outdoor relief mentioned in the body of this post (cash and in-kind aid such as food or fuel for the winter), and “indoor relief,” i.e., poorhouses. Poorhouses were workhouses that had rather harsh conditions: long hours were mandatory, and whipping and other punishments for infractions of a house’s rules were common. Poorhouses were not abolished in the U.S. cities that abolished outdoor relief in the late nineteenth century. Still, examining whether or not private giving made up for the absence of outdoor relief in those cities is relevant for the issue at hand: a major form of government welfare was abolished or drastically reduced for a quarter-century, and thus these cities relied on private aid for a substantial portion of the provision of aid to the needy and unfortunate.

#2: The empirical evidence I mention above comes from Frederic Almy, secretary of the Buffalo Charity Organization, who in 1899 gathered data on outdoor relief and private charity in forty cities, ten of which had completely abolished outdoor relief. Almy found that the cities with the lowest level of such aid had the highest level of private charity, and vice versa. See Frederic Almy, “The Relation between Private and Public Outdoor Relief—I,” Charities Review 9, no. 1 (1899): 22–33; and Frederic Almy, “The Relation between Private and Public Outdoor Relief—II,” Charities Review 9, no. 1 (1899): 65–71. Almy’s study does have some drawbacks. The relationship he found did not hold very well for cities with intermediate levels of outdoor relief; for these cities, the main observable relationship was that northern cities provided more total aid (public and private) than southern cities did. (Almy thought that the explanation for the regional difference was the harsher winters in the north.) Also, Almy’s study only measured private giving by regularly organized charitable societies; it omitted charity provided by individual churches, mutual aid societies, and the Salvation Army, so it may be that his study systematically underestimated the amount of private charity. Still, Almy’s study seems to refute the claim that when state welfare is abolished or drastically cut back, significant  harm must result since private charity will not pick up the slack.