How Classical Liberal Principles Address Cultural, Social, and Economic Issues: Richard Epstein's Reply

My Liberty Forum essay, Freedom of Association and Antidiscrimination Law: An Imperfect Reconciliation has provoked three thoughtful responses. Those by Marc DeGirolami and Paul Moreno are supportive of my approach and may be best described as intramural disputes among individuals who agree on the relationship between freedom of association and basic antidiscrimination laws. Andrew Koppelman’s response, notwithstanding his stated sympathy toward the contract at will, offers a defense of the antidiscrimination law on a mixture of moral and consequentialist grounds.  It is useful to consider DeGirolami and Moreno in tandem before turning to Koppelman. Needless to say I cannot cover all the points they raise in their essay. But I do hope this cover the essentials.

Marc DeGirolami.  The initial response by DeGirolami welcomes the effort to expand the scope of freedom of association, but chides me because of my inability to understand that my excessive reliance on a “marketized and contractualized conception of First Amendment freedoms” ultimately undermines freedom of association. But I plead innocent to this charge. My point was to indicate that the notion of contractual and associational freedoms were not tied to any narrow economic conception of what people value. The operative notion is one of voluntary association, and that association can be directed toward whatever ends the parties to it chose. If they wish to have affective relations, they can, and in so doing achieve, as DeGirolami suggests, Aristotle’s highest form of friendship: “an association in which each member ‘loves the other for what he is’ and which endures ‘so long as they remain good.’” There are obvious limits as to the size of groups in which these affective arrangements can last. And even within the smaller groups in which they do take place, there is no reason why the parties must have any intention to create legal relations.  Their arrangement lasts only so long as they both choose to do so, and they make their choice as moral agents.

Even under the current law, there is a general consensus that certain limited intimate associations remain outside of the scope of the antidiscrimination laws, so that the legal action takes place on the broader canvass in which there are larger organizations like the Jaycees who in Roberts v. United States Jaycees were ordered to admit women on the grounds that their organizations were not sufficiently intimate to be governed by freedom of association principles. No one has lost sight of the role of affective arrangements. The battle is over the proper role of antidiscrimination law, and my position remains that in the absence of monopoly power we do not have to worry about the type of organization, its size, or its objectives. All are defended by the same principle, which will not work if we hitch the doctrine of freedom of association to affective relationships that cover only a small part of the spectrum.

In making this point, I do not agree with DeGirolami’s observation that politics originates in family and not in markets. But the question is what to make of this claim. Historically, the dominance of the family antedates the risk of the state. Hunter-gatherers did not have permanent interests in land, and they did not have exchange relations among members of the same immediate and extended families. The interdependence of their welfare eased the conflicts of interest. But once groups put down roots and need to expand their population base, these affective ties tend to weaken. Formal property rights and voluntary exchange become ways to link strangers, even as the nuclear family continues to operate. It is wrong to think that the market drives out the family or other close-knit associations. It is simply that the principles of trust that govern family partnerships or other sharing arrangements cannot work with the bills of exchange of the 15th century or the credit card networks of the 21st. While legal enforcement is rarely an issue within families it becomes a key backstop for reputation in other contexts, and in so doing permits gains from trade to cover intertemporal transactions: loans or sales today for repayment tomorrow. DeGirolami is wrong to see the rise of complex voluntary arrangements as leading to the “demotion” of simpler ones. The problem here is not with the synthesis. The problem from the progressive wing is their disbelief that transactions between large and small players can ever result in mutual gain. And on that point, progressives are surely wrong. People do not make the same mistake time and time again. Voluntary transactions are neither exploitation nor some form of theft. They should, subject to the limits for monopoly, be praised not scorned.

Paul Moreno. Moreno’s comments address the way this skeptical Progressive attitude has shaped the modern conception of monopoly power, which goes beyond a single seller of a particular good or service. Moreno is surely right to point to the enormous influence of both Robert Hale[1] and Friedrich Kessler[2] (the latter who taught me antitrust law at the Yale Law School). Hale, painting on a large canvas, took as his major premise that coercion was necessarily everywhere in social relations so that it was pointless to try to defend any notion of freedom of association. Kessler insisted that this point was especially true with respect to standardized contracts, which left consumers and employees with “no choice” but to accept the terms that were offered to them, without noting that it is a lot easier to say no to a big seller from whom you can walk away than it is to the tax collector who can seize all your assets. Nonetheless, as Moreno notes, this Progressive worldview was later reflected in Lyndon Johnson’s famous remark that no one whose ancestors suffered centuries of discrimination could be expected to have a fair shot in the competition of life.

Historically and intellectually, these three writers weave together three elements: the ubiquity of coercion, the absence of real consumer choice, and the significance of historic discrimination. Thereafter, the potent combination of these three attitudes leads to the dubious conclusion that monopoly conditions are omnipresent, thereby requiring antidiscrimination laws, affirmative action programs, and even reparations to correct the social inequities.

The important point to stress is that all each of these three points, taken separately, is incorrect. To start with Hale, it is just incoherent to think that the refusal to deal in competitive markets constitutes an illicit form of coercion. By that standard coercion is the routine feature of every failed negotiation, where each party may rightly say that it has been coerced because the other side has refused to see the merits of his proposal. Coercion ceases to be a rarity and becomes ubiquitous. Yet that conclusion has to ignore the social welfare consequences of these two definitions of coercion. A competitive equilibrium yields the highest obtainable output for any given set of resources. It is rarely attainable in pure form, but many markets develop techniques by which prices and wages can be readily determined. To compare this situation with the Hobbesian war of all against all is a genuine travesty. To be sure, the position of the monopolist is different because the control over price or wages can lead to systematic economic distortions for which a system of rate regulation, seeking a fair and reasonable rate of return, could be a sensible alternative. Hale just misses the importance of market structure in his analysis of refusals to deal.

Kessler makes the same mistake in a more limited context. Standardization of contract terms has several key efficiencies. The first is that it reduces the cost of transacting, which increases gains from trade over a wide range of deals, many of which could not take place at all if transactions costs were high. In addition, the clarity of the terms gives individuals some confidence that they will receive terms that others have found efficient, and thus will reap similar benefits. Finally, in modern markets, standardization actually makes it easier to change prices and wages where circumstances call for it. Keeping the terms the same in all cases means that customers do not have to worry that a particular price decrease may be offset by some concealed inefficient terms. Yet by the same token, it is hard to see how standardization is evidence of monopoly power when firms in competitive markets use the same devices. Indeed, no monopolist will, if left undisturbed, change the terms found efficient in the competitive market. Instead the rational monopolist is better able to reap his excess profits by increasing price, in some cases utilizing price discrimination.  So the standardized contract is a good thing, and the law of regulation and antitrust should be directed toward any monopoly power, not the method of contracting.

Lyndon Johnson’s point presumes that the past determines the shape of the future, and there is of course no doubt that lower levels of wealth and education can result in lifetime disadvantages.  But no matter how bad past circumstances are, market institutions help the least fortunate. Just like other people they will seek out the best deals that they can get, which means that they won’t enter into contracts that make them worse off. What they need is a maximum set of choices to develop their human capital and then to exploit it in voluntary transactions. The race metaphor makes it appear that only one party can win and that the others must lose, so that markets are at best zero sum games. But in a voluntary exchange both parties win, which is why they are encouraged by sound social policy.

Andrew Koppelman. The simple proposition that two parties of unequal wealth can both benefit from voluntary exchange explains the key errors in Koppelman’s one-sided account of the civil rights laws. He begins with an attack on dominance and subordination. Slavery was such a system. But the proper response is to end slavery and encourage freedom, so that individuals can make use of their talents in whatever way they see fit. At this point, the applicable legal rules are those governing the contract at will whose general efficiency Koppelman concedes. One reason why they are successful is that in competitive markets all individuals can gravitate to persons who will give them the best deals. The bigot will not attract formerly disenfranchised minorities, and he cannot block them from taking their best deals, which is why individual refusals to deal in competitive markets carry no systematic sting, while the threat or use of force constitutes a deadly threat in both competitive and monopoly markets.

Given these fundamental truths, why depart from this system by adding an antidiscrimination law? Koppelman suggests that the civil rights laws can lead to “the amelioration of persistent economic equality.” But this is a makeweight argument at best. His major support for that proposition is that the median income of black males rose from 57 percent of white males in 1964 to 66 percent in 1985, and likewise that the number of black professionals and managers increased from 32 to 64 percent. But he gives away the game when he notes that the most dramatic gains occurred in the first decade after passage. Why? Because during that period the major activity in the civil rights laws was to eliminate the formal system of state coercion and segregation, chiefly in the Southern states. Thereafter progress came very slowly, if at all, because there were few barriers of entry left to dismantle. Instead, in that time the civil rights law themselves became a barrier to entry. Thus if the law makes it more difficult to fire a black worker, it makes an employer more reluctant to hire one. In addition, the early colorblind version of the 1964 Act stifled the use of private affirmative action programs that might well have changed the climate of opinion on race relations. The doctrines of freedom of contract and association, it should be recalled, do not require firms to maximize profits or to adopt a colorblind strategy. The pressures today to introduce affirmative action programs are permanent and intense, and they are driven as much by widely shared norms of social justice as by any narrow economic model.

Once that simple fact is understood, it should become clear that Koppelman, not I, misunderstands, the role of culture in these many social settings. Of course, there were fierce forces in opposition to racial equality, but they did not rely on individual refusals to deal. They depended on the use of public and private force to keep out those individuals and firms who understood the defects of the then-dominant culture and sought mightily to change it. The racists well understood that systemic segregation could only survive if backed by state force, which is why they sought to exclude, by all means fair or foul, any business that did not play by their rules.

Ironically, the antidiscrimination laws do not do much to reshape culture. Indeed in some instances they create a strong backlash as sanctions, especially under disparate impact theory, are imposed against firms that did not discriminate, and, indeed in some cases even had affirmative action programs. It takes a heroic assumption to assume that the disparate impact cause of action has done much good, given the incentives of firms to make all sorts of decisions to avoid the antidiscrimination law: locate in places where they are not exposed to threats of liability, reconfigure jobs to reduce the need for low-income labor. The simple point here is that private voluntary affirmative action, which is fine under classical liberal ideals, does far more to combat diffuse social discrimination than the unspecified form of coercion that Koppelman champions.

Conclusion The role of antidiscrimination laws in modern society will continue to be one of the flashpoints of intellectual, legal, and political debate. The stakes are so high that it is important that the normative debate proceed on an accurate foundation of the behavioral assumptions of human behavior in both nonmarket and market settings. All too often the critics of the antidiscrimination laws misstate or misunderstand how markets operate, and often attribute to them an all-consuming power that drives out other forms of social arrangements. But that is not now and has never been the case. So long as young children have to be nurtured, affective ties will continue to matter in the lives of everyone, be they market skeptics or market participants. The combinatorial plasticity of voluntary arrangements has this desirable property. Social patterns are not built up by once-and-for-all transactions. Different forms of social arrangements will be tried, and those that fail will be jettisoned. But that flexibility is lost with any antidiscrimination law that places governmental power at the center of social relations. The basic Hayekian point remains: Never substitute a system of state monopoly control for a decentralized mix of familiar, social, and market transactions. People are better at organizing their own lives than is the state.  The law should let them do it.

[1] Robert L. Hale, “Coercion and Distribution in a Supposedly Non-Coercive State,” Political Science Quarterly 38 (1923), 470.

[2] Friedrich Kessler, “Contracts of Adhesion—Some Thoughts About Freedom of Contract,” Columbia Law Review 43 (1943), 629.