![[back]](../../left.gif){kind=small}
Now that we are all conservatives, we naturally divide ourselves into various kinds of conservatives. There are traditional conservatives and neoconservatives, radical conservatives and moderate conservatives (probably no liberal conservatives), fiscal conservatives and social conservatives, isolationist or nationalist conservatives and internationalist conservatives. Then, importantly, there are the libertarians, many of whom deny that they are conservatives. Indeed, one of their heroes, the late Friedrich Hayek, wrote a famous essay entitled "Why I Am Not a Conservative." But if you ask almost anyone whether Milton Friedman and William Buckley, probably our two most famous libertarians, are conservatives, the answer will be "Yes."
I want to speculate a little here about what being a libertarian might mean and what the policy implications of libertarianism might be. That is partly to demonstrate how hard it is, in my opinion, to get from any of the prevailing philosophies or ideologies reliable answers about the issues confronting us today. But mainly I write about libertarianism to recall the thoughts of a remarkable man, Henry Simons, who was, I believe, the first to use the word to describe a political-economic philosophy. Simons was one of my teachers at the University of Chicago in the 1930s and died, at an early age, in 1946. My essay here will consist mainly of variations on themes from Simons, not because I think he answered all the relevant questions but because he raised many of them. Although I shall pick at what seem to me to be some loose threads in his story, he remains for me, 60 years after I sat in his class, the most congenial of economists.
According to the Oxford English Dictionary, the original definition of libertarian is a person who believes man has free will (as opposed to a necessitarian, who believes he doesn't). Examples of that early usage go back to 1789. The OED next refers to a libertarian as one who "approves of or advocates liberty;" hence, "libertarianism" is "the principles or doctrines of libertarians." There is one example from Henry Sidgwick, an economist and philosopher who derisively referred in 1886 to an attempt "to make [Plato] talk modern Libertarianism in a quite unwarrantable way." The OED has no example that comes close to current usage, so I feel entitled to claim priority for Simons.
He first came to prominence in 1934 when he published a pamphlet with the title "A Positive Program for Laissez Faire: Some Proposals for a Liberal Economic Policy." This was written near the bottom of the Great Depression, at the beginning of the New Deal, and concerns itself with specifics of economic policy that I do not wish to go into here. What is most interesting is the title. Simons is arguing for laissez faire, for economic freedom. But he also recognizes, or asserts, that laissez faire would not exist, survive or succeed without positive action by government.
In his title Simons is thus using the word "liberal" in the old-fashioned sense, which it had then and still has in Europe, as meaning the belief that individual liberty was the best route to economic prosperity and social progress. The origins of this idea in economics are usually traced to Adam Smith. When Simons in the text refers to New Dealers or other 20th-century "planners" who claim to be "liberals," he uses quotation marks to indicate what he regards as the illegitimacy of their claim.
In early 1945 Simons wrote "A Political Credo," which he described as "an endeavor to formulate specifically the political predisposition" implicit in his work. This later became the introduction to a collection of his essays, published in 1946 under the title \ for a Free Society. By that time in America the word "liberal" had been appropriated by those who favored a massive expansion of the role of the state, and Simons felt the need for some other description of himself and the policies he favored. The description he decided upon was "libertarian" (although he also from time to time, and somewhat confusingly, reverted to "liberal").
That he regarded his position as not only "libertarian" but "severely" so may be a little surprising to modern readers who are more likely to find it "moderate" or even "compromising." I think Simons might reply that rigorous pursuit of the libertarian goal, which by definition precludes impinging on the liberties of others, logically leads to compromise, just as the pursuit of laissez faire logically led to the need for a positive program of government. Indeed, Simons' 1945 essay quite explicitly repeats the position he took in 1934: "The libertarian policy prescription calls essentially for planning to preserve freedom."
The key paragraph of the 1945 essay begins on the first page:"The distinctive feature of [the libertarian tradition] is emphasis upon liberty as both a requisite and a measure of progress. Its liberty or freedom, of course, comprises or implies justice, equality and other aspectual qualities of the "good society.' Its society, however, is a living, functioning organization or "organism'; and its good society is no static conception but is essentially social process whose goodness is progress-and progress not only in terms of prevailing criteria but also in the criteria themselves. "Libertarianism," Simons continues, "is thus largely pragmatic as regards the articulation or particularization of its values; but its ethics, if largely pragmatic, also gives special place to liberty (and nearly coordinate place to equality) as a "relatively absolute absolute."
The many qualifiers Simons here imposes on "liberty" are interesting. Liberty "of course" comprises or implies justice and equality. The good society is adaptive. The application is pragmatic. Liberty is a "relatively absolute" absolute. When all this is said, being "for liberty" loses its clarity and distinctiveness as a guide to policy.
The use of the word "equality" as a value nearly coordinate with liberty rings strangely in our ears today when someone who demands equality is likely to be accused of fomenting "class warfare." Yet Simons is very cautious about equality: "Save as the bride of liberty, equality is pale and deadly dull, if not revolting. But the ultimate liberty obviously is that of men equal in power."
Justice is the key. The word is not heard much these days, except as the name of a federal cabinet department or of a member of the Supreme Court. But Americans do pledge allegiance to a republic "with liberty and justice for all," and, however much we may value freedom, we do not really mean that people should be free to do whatever they wish. A person is free to do only what he may justly do. To the extent that he cannot do what he may justly do, he is not free. One hundred and fifty years ago, one might have said-and some probably did say-that slaves were as free as everyone else to exercise their property rights, which only happened to be negative. But it was decided that slavery was unjust and therefore a deprivation of the freedom to which the individual was entitled.
Simons distinguishes between two kinds of justice-commutative and distributive justice. Commutative justice "dictates that each shall receive according as he (or it) contributes to organized, cooperative, joint production or, in technical economic language, according to the productivity of his property, capital or capacity (including personal capacity)." Why Simons calls this "justice" is unclear. It is presumably the result that would obtain in voluntary exchange in free, competitive markets. If A and B exchange their product each gets the value of his product as measured by what the other is willing to pay for it.
This has intuitive appeal as being fair. But it implies, in the first place, that these exchange values are the true values of what each produces. That we do not always intuitively accept this is shown by our fascination with the story of Babe Ruth. When asked why he had a higher salary than the president he replied, "Well, I had a better year." We smile and accept the logic, but we also recognize that it is a little odd because we think that what the president does is really more valuable than hitting home runs, even if the president is Herbert Hoover.
There is also intuitive appeal in the idea that a person should get what he contributes. He seems to deserve it as a reward for his efforts. But his contribution may not be the result of his efforts, or not mainly that. The value of his contribution depends on inherited personal capacity and other wealth, family environment, education and the nature of the society in which he lives-none of which is "his" contribution. Players much inferior to Ruth now earn incomes much larger than his was partly because the society is richer, but also because TV has made them more productive in the sense that far more people can watch them at the same time.
Simons' case for commutative justice is really a case for two other things. One is the case for liberty again, because any radical alteration of distribution from the results of voluntary exchange among persons would require a degree of centralized power that would be inconsistent with liberty. This raises the question of what possibilities exist to alter the distribution of income without a dangerous concentration of power. Simons does not deny that there are such possibilities.
The other part of the case for commutative justice is a case for efficiency. Redistributing income away from the system in which each gets the value of his product impairs incentives, disorganizes production and so reduces total output. As Simons puts it, "No large group anywhere can possibly gain enough from redistributing wealth to compensate for its probable income losses from the consequent disorganization of production." Also: "However, our primary problem is production. The common man or average family has a far greater stake in the size of our aggregate income than in any possible redistribution of income. Large and efficient production requires close approximation to the norm of commutative justice." Thus, "large and efficient production" is added to the list of values-liberty, equality and justice-with which a libertarian must be concerned.
What is striking about these statements is that they are quantitative propositions. That is, they imply a quantitative relation between the size of the change in distribution and the size of the consequent change in the rate of growth of total output. But at the time these statements were made we knew little about the distribution of income or about the rate of economic growth, and especially about the sources of economic growth-about which we still know little.
The nature of the quantitative relation between distribution and growth may be illustrated by some arithmetic. Suppose that the lowest 20 percent of the population receives 4 percent of the national income. Redistribution to raise that 4 percent to 5 percent will presumably slow down the growth of the national income because it will reduce the incentive to work and save of the other 80 percent of the population, from whom income would have to be taken. In time, the 5 percent of the now more-slowly growing income will be no more than the 4 percent of what the more- rapidly growing income would have been. But one needs to be cautious about drawing conclusions from such calculations. Suppose that when the lowest 20 percent of the population received 4 percent of the income the rate of growth would have been 2.5 percent per annum but that the redistribution cuts the growth rate to 2.4 percent. How long will it be before the 5 percent is less than the 4 percent would have been? The answer turns out to be 228 years.
In this example, redistribution is assumed to reduce the growth rate from 2.5 percent per annum to 2.4 percent. That may strike some as an absurdly low assumption. But it is a reduction of 4 percent in the annual growth rate, whereas the income share of the other 80 percent of the population is declining only from 96 percent to 95 percent, which is a decline of a little over 1 percent.
In any case, the numbers used in this example are of no importance. What is important is that to say that no large segment of the population can gain from redistribution implies knowing something specific about the determinants of growth that we do not know. The example does suggest, however, that the gain from redistribution could plausibly last for a long time.
Simons' other kind of justice is "distributive justice," meaning simply equality, including equality of income. He is deeply concerned about that. He is not satisfied with the equal right of the rich and the poor to sleep under the bridges of Paris. Nor is he satisfied with the comfortable way-station of equality of opportunity although he gives much weight to it. He is concerned with progressive reduction of inequality of outcomes.
He felt no need for an extensive argument on the importance of equality. In "A Positive Program for Laissez Faire," he discussed the question of the equitable distribution of power as well as the equitable distribution of economic goods or income. "Surely," he wrote, "there is something unlovely, to modern as against medieval minds, about marked inequality of either kind. A substantial measure of inequality may be unavoidable or essential for motivation; but it should be recognized as evil and tolerated only so far as the dictates of expediency are clear." The use of the word "unlovely" comes, in my opinion, as close as possible to saying that the issue is one about which you can agree or disagree but about which there is nothing more to say.
His warning about redistribution is really a warning against a certain way of reducing inequality. He wanted redistribution but wanted it to be achieved in a particular way. He envisaged a two- level process. At one level, production is organized in a free- market system that yields a primary distribution of income that is highly unequal. At the other level, this primary distribution is changed by transfer programs, taxation and investment in the education of those who would otherwise be least productive. He believed that "progressive mitigation of inequality" could be achieved by policy at the second level while the free-market system could be preserved to yield both liberty and efficiency.
Fifty years later one can say that this vision has been supported by history. Societies that tried to achieve greater equality by suppressing the free-market system have failed, their economies have collapsed and there has been poverty for everyone. Those societies that have combined a substantial welfare system for redistribution while preserving the free market have on the whole succeeded in reducing poverty while sustaining economic growth.
Distinguishing between these two levels of activity reduces the conflict between liberty-cum-growth on the one hand and redistribution on the other hand. But the two levels cannot be entirely isolated from each other. Simons was very interested in "neutralizing" the tax system so that it did not distort the pattern of production that would emerge in a free market. That was one reason for his devotion of so much effort to describing a comprehensive nondistortive income tax. But the redistribution system will affect incentives to work and save and thus will affect the rate of growth. The question of whether these effects are desirable, and to what extent, will remain.
Simons placed great emphasis on education, rather than on transfers. His idea was to raise the incomes of the poorest people by raising their productivity. Although he was not explicit about this, one reason was probably his belief that education was cheaper, imposing fewer burdens on the rest of society, than the transferal of income. There is the homely story about teaching a man to fish, which will keep him fed for the rest of his life, rather than giving him a fish, which will leave him hungry the next day. But the story is not compelling. The costs of teaching him to fish may exceed the costs of giving him a fish a day for the rest of his life.
That is, there is a cost-benefit question here, as almost everywhere else. Spending on education is a form of investment. That investment may be, for some students at least, less productive than other investments in the society. In that case it may be cheaper to make the more-productive investment and share the proceeds with underachieving students rather than trying to educate them up to a higher level of productivity.
There may be a reason to prefer education to transfer-which we now call "welfare"-as a way of dealing with inequality. That is because the society would rather see people studying, learning and working instead of receiving welfare, even if the welfare process is cheaper. Whether a "severe" libertarian can take that position is unclear. because libertarians generally deny any right to tell other people how to live their lives.
The problem of controlling the lifestyle of others arises even more acutely in the consideration of welfare policy. The most libertarian version of welfare policy is the negative income tax, associated with Milton Friedman. The basic idea is that people should be given money in proportion as their income falls below a certain level, without any question or requirement as to their own behavior. In other words, the system was to maximize the liberty of its recipients. Eventually, two main defects in the system were perceived. First, it was expensive because it created an incentive for people to remain poor enough to qualify for benefits. Second, and probably more important, it encouraged or permitted people to behave in ways that the society did not like-that is, not to study, not to work and, apparently most disliked of all, for young unmarried women to have babies. Concern over this has led to the current interest in welfare "reform"-which means to give or withhold welfare benefits to encourage or discourage certain kinds of individual behavior. Whether a libertarian can endorse such "reform" is a question.
There is also a great deal of talk these days about "values" and about the need to base conduct on a more elevated set of values than is believed now to prevail in our society. This creates a problem for libertarians. Simons says: "A moral order imposed by force or fraud, by authorities, or by threats of punishment in this world or the next is a contradiction in terms. Moral individual conduct is meaningful only within a range of responsible freedom; and social morality is, like truth, a matter of voluntary consensus. The libertarian recognizes no test of moral truth or moral wisdom save such consensus. Society is always right-provided it is the right kind of society."
But there is no answer here for dealing with the possibility that a significant number of people do not share the consensus and do not do what the society considers right.
I do not propose to try to parse every sentence in Simons' remarkable essay. The important point is that many problems arise when a libertarian says, as Simons does and as I believe he reasonably should, that he values not only liberty but also justice, equality, economic growth and moral behavior. These problems are of two kinds. First, there are many relevant, factual things that we do not know. For example, we do not know to what extent redistribution of various kinds affects economic growth, to what extent investment in education pays off in reducing inequality, or to what extent changes in welfare programs affect behavior. Second, even if we knew the size of the tradeoffs-how much redistribution has how much effect on growth, for example-we have no objective way of deciding how much of one goal it is worthwhile to abandon in the name of achieving another goal.
These problems are not peculiar to libertarianism. They bedevil any general approach to policy. Simons' answer-to take recourse to process-is the only one possible. The democratic process in a free society will give better answers than any other and will give better answers as time passes. That is probably a proposition on which not only all present varieties of conservatives, but also any remaining liberals, can agree.
Return
to COSMOS 1996 Table of Contents Return
to COSMOS Journals Return
to COSMOS Home Page