Common law, statute law, and the theory of legislative choice: an inquiry into the goal of the Sherman Act.

Kleit, Andrew N.

I. INTRODUCTION

The Sherman Antitrust Act is now more than a century old, yet debate still continues about its original goals. Previous authors, focusing on the substance of the 1890 debate, have reached various conclusions about these goals, each of which provides different implications for antitrust policy. Currently, the debate focuses on whether the purpose of the Sherman Act is to maximize economic efficiency or the welfare of consumers. My aim is to reach beyond the rhetoric and discuss the institutional context of the Sherman Act to discern between these two hypotheses. I conclude that the primary goal of the Sherman Act was to promote economic efficiency.(1)

The weight of the evidence, however, suggests that, at least in the later years of the Reagan and the Bush Administrations, the Federal government has applied a welfare of consumers standard.(2) Scholarly support for this position is provided by Lande [1982], who uses the context of the Congressional debates to assert that preventing transfers of wealth from consumers to producers was the primary goal of the Sherman Act. In contrast, Bork [1966] infers from the same evidence that economic efficiency was the goal of the Sherman Act.

Section II reviews the difference between an efficiency and a welfare-of-consumers standard. It also discusses the methodology used here to discern between Bork's and Lande's hypotheses. Instead of concentrating on the congressional debates, I examine the structural context of the Sherman Act. Section III argues that the Act is best viewed as a modest statutory extension of the common law, an extension that fits well into the Law and Economics framework.

Section IV reviews the political support for the Act and the manner in which Congress chose to have the act administered. The modern theory of interest groups and legislative choice yields additional insight into the goals of the Act. The common law origins of antitrust, the support for, and implementation of the Act all support the conclusion that the primary goal of the Sherman Act was to maximize economic efficiency.

II. THE DIFFERENCE BETWEEN ECONOMIC EFFICIENCY AND THE WELFARE OF CONSUMERS

Bork, Lande, and the Williamsonian Trade-off

Of course, no government intervention can be expected to generate mathematical optimality. The question addressed here is whether one particular law, the Sherman Act, was designed to reach toward economic efficiency, seeking to maximize the total wealth of society, or simply to maximize the welfare of consumers without considering the effects on producers. This trade-off between market power and economic efficiency was first formally described by Williamson [1968, 21]. Figure 1 is a slightly modified version of Williamson's Figure 1. Assume that there are only two (identical) firms in an industry and that they vigorously compete so that each is selling at price equal to (marginal and) average costs. Each firm has average costs as denoted by the line [AC.sub.1]. Given the demand curve D, industry price equals [P.sub.1] and output equals [Q.sub.1]. Now suppose that the two firms merge. The merger generates efficiencies that lower the combined firm's average costs to [AC.sub.2]. Due to the lack of competition, however, the combined firm raises price to [P.sub.2] (and lowers quantity to [Q.sub.2]).

As a result, consumers lose the rectangle A, which is wealth transferred to producers. They also lose the triangle B, which is the deadweight loss to society resulting from the allocative inefficiency of monopoly. The monopoly firm gains directly from the pockets of consumers the rectangle A. It also gains the rectangle C, which represents the costs savings due to the merger-related efficiencies. If C (the efficiency gain) is of greater size than B (the deadweight loss due to market power) then the merger would increase economic efficiency while decreasing the welfare of consumers.(3)

Following Bork, Williamson assumes [1968, 21-22] that efficiency is the goal of the Sherman Act. Given this assumption, he generates broad measures of classes of mergers that generate market power but should be not opposed by the antitrust authorities because they increase economic efficiency. In general, he determines that relatively small percentage levels of cost savings are necessary to offset the distortion arising from the exercise of market power. For example, Williamson [1968, 32] shows that the welfare loss associated with a 20 percent price increase would be offset by efficiencies of 4 percent for a price elasticity of two and 2 percent for an elasticity of one.

A welfare-of-consumers standard, on the other hand, would focus solely on the gains and losses to consumers. It is thus concerned, not with the relative sizes of the triangle B and the rectangle C, but with whether or not the rectangle A (and, by implication, triangle B) have size greater than zero. The difference between which rectangles and triangles to compare has significant implications for antitrust policy. For instance, as Fisher, Johnson, and Lande [1989] illustrate, under a welfare-of-consumers standard, the market concentration levels that would trigger a merger challenge are significantly lower than under an efficiency standard. A welfare-of-consumers standard also generates a stronger rationale for enforcement of provisions against "tying" and price discrimination, practices with ambiguous welfare effects that benefit producers and thus may harm consumers. The narrow welfare- of-consumers standard has thus become an argument for a more stringent and activist antitrust policy.

Testing between the Two Hypotheses

In his review of the Congressional debates over the Sherman Act, Bork [1966] concludes that Congress was primarily concerned with promoting economic efficiency. Lande [1982] in his review of the Congressional debates, concludes that Congress was primarily concerned with promoting consumer welfare. In fact, Bork and Lande largely agree on the content of the Congressional debate. The major difference between the two is that Bork [1966, 7] equates consumer welfare with economic efficiency. He does this by implicitly arguing that producers are also consumers. This led Lande to redefine consumer welfare as "the welfare-of-consumers," a term equivalent in economic jargon to "consumers' surplus."

Examining the Congressional debates to determine whether Congress was interested in consumer welfare as the welfare of consumers or consumer welfare as economic efficiency would appear to be a difficult exercise. Any conclusion that could be reached would be based on nuances of rhetoric. Given that examining the debates is not an adequate method of testing the hypotheses in this context, I take another approach: I seek to combine the Law and Economics and public choice approaches to examine the issue. I ask four questions. First, is the Sherman Act an extension of the common law? Second, did the pre-Sherman Act common law seek to move towards economic efficiency? If the answer to both of these questions is yes, the Law and Economics approach would imply that the pre- and post-Sherman Act law on restraint of trade should be seen in the general Law and Economics framework of common law enhancing economic efficiency.

Third, were there any major interest groups behind the passage of the Sherman Act? If there were consumer-oriented interest groups supporting the Sherman Act, the public choice approach would imply that the act was indeed designed to enhance the welfare of consumers. If there was general support for the act, it would be consistent with the passage of a law designed to reach towards economic efficiency.

Fourth, was the administration of the Sherman Act designed to be conducive to rent-transfer? Modern public choice implies that legislation designed to encourage rent-transfers requires an agency committed to that particular task. On the other hand, the Law and Economics approach implies that placing the implementation of a law in the hands of the judiciary is consistent with a goal of economic efficiency.

III. THE ANTITRUST LAWS AND THE COMMON LAW

The Sherman Act did not mark a revolutionary change in competition law. Rather, scholars are clear that the Sherman Act was a logical extension of the centuries-old common law in restraint of trade. The restraint-of-trade caselaw is consistent with how the Law and Economics paradigm predicts the common law will evolve towards economic efficiency. I will show that the legal origins of the Sherman Act support the hypothesis that the goal of the Sherman Act is to promote economic efficiency. In addition, Lande's analysis with respect to the "welfare-of-consumers" will be shown to have a number of logical difficulties.

The Sherman Act as an Extension of the Common Law

The Common Law Background of Antitrust. Antitrust law did not begin in 1890 with the Sherman Act. As numerous writers discuss, the common law opposition to restraint of trade dates back several centuries.(4) For instance, the rule of reason outlined by Chief Justice White in U.S. v. Standard Oil, 221 U.S. 1 (1911) is an amalgamation of several common law cases, the most important being the 1711 English case Mitchel v. Reynolds, 1 P. William 181. White's decision explained how both per se and rule-of-reason cases evolved under the common law and how those common law precedents fit naturally into Sherman Act antitrust enforcement. White wrote that both the common law and the Sherman Act prohibited, "all contracts or acts which were unreasonably restrictive of competitive conditions, either from the nature or character of the contracts or acts [this refers to per se offenses] or where the surrounding circumstances were ... of such character as to give rise to the inference or presumption that they had been entered into with the intent to do wrong to the general public...". This approach is not significantly different from the one used by Chief Justice Parker in Mitchel v. Reynolds: "[I]n all restraints of trade where nothing more appears, the law presumes them bad; but if the circumstances are set forth, that presumption is excluded, and the Court is to judge of these circumstances, and determine accordingly ..."

Perhaps the most important antitrust tenet generated from the common law was the unenforceability of contracts that created restrictive arrangements. The seminal discussion of the common law's opposition to collusive contracts is Judge William Howard Taft's opinion in U.S. v. Addyston Pipe, 85 F. 271 (1898), modified and aff'd. 175 U.S. 211 (1899). In supporting the decision that led to the per se rule under the Sherman Act for "naked" restraints such as price-fixing, Taft refers to more than one hundred common law restraint-of-trade cases in his decision.

The common law can be seen as a framework for giving consumers property rights to "competitively" priced goods to generate efficient economic outcomes. Unfortunately, the common law, by itself, does not appear to have been sufficient for the task of dealing with the anticompetitive problems generated by the Industrial Revolution.

How the Sherman Act Strengthened the Common Law. There were instances prior to the Sherman Act where the common law was used to fight anticompetitive actions. In general, however, it was not well suited to this task. In the age of industrialization, the common law in the United States faced three serious problems.

First, as Oppenheim, Weston, and McCarthy [1981] discuss, industrialization created barriers to entry and economies of scale, increasing the opportunity for the exercise of market power. Coupled with the advent of the railroad, industrialization made interstate commerce more frequent and therefore more important. No federal restraint-of-trade common law existed in the U.S. prior to the enactment of the Sherman Act. Common law at the state level did not protect interstate commerce from anticompetitive practices that the Industrial Revolution had made more likely. While there were a number of state cases, in general the states had difficulty reaching broad areas of commerce.(5)

Second, the existing law was not comprehensive in its handling of antitrust matters. The general unenforceability of anticompetitive agreements among producers under the common law discouraged their use. But where such agreements could succeed without enforcement through the courts, consumers had no cause of action to challenge their use.

Third, private antitrust enforcement may suffer from a public goods problem. While a few sellers may gain handsomely from a successful cartel, the losses may be spread among perhaps millions of consumers. No one consumer, or even perhaps a coalition of thousands of consumers, may have sufficient incentives to bring legal action, because they would bear all of the costs and gain only part of the benefit, even after compensating for the treble damages available in private suits. Even many firms victimized by cartels may not have the appropriate incentives to carry out litigation. Thus, legal challenge by individuals or groups of individuals to monopoly would be likely to be undersupplied.

The Sherman Act deals with these problems in perhaps the most concise manner possible. First, it consists of a brief but vaguely worded statute that creates a federal common law subject to interpretation by the judiciary to deal with problems of "restraint of trade," a common law concept.(6) Bork [1966, 35-36, 46] makes it clear that at least Senator Sherman felt that the Act should be administered in the same fashion as the common law, thus enabling the judiciary to determine which practices should and should not be allowed. Second, it gives consumers a right to challenge restraint- of-trade contracts in court. Instead of being merely unenforceable, it makes such contracts subject to prosecution. In Taft's words (Addyston Pipe at 279)

The effect of the [Sherman Act] is to render such [restraint of trade] contracts unlawful in a positive sense, and punishable as a misdemeanor, and to create a right of civil action for damages in favor of those injured thereby, and a civil remedy by injunction in favor of both private persons and the public against the execution of such contracts and the maintenance of such trade restraints.

Third, the Sherman Act paved the way for the establishment of a public prosecutor to address the public goods problem. The Department of Justice, and later the Federal Trade Commission, could act as a public agent to stop or prevent producers from capturing the property rights of consumers through anticompetitive actions. Thus, the Sherman Act can be viewed as a modest innovation to the common law on restraint of trade.

Deriving the Goal of the Sherman Act from the Goal of the Common Law

The Goal of the Common Law. The Law and Economics school of the past twenty years has argued that the goal of the judicially- written common law is to reach toward economic efficiency.(7) The law and economics paradigm implies that the judicial process will tend to replace less efficient with more efficient rules. Litigation is a competitive process, with each party to a case spending money to present its side. Over time, the parties advocating the more efficient rule will spend more resources advocating that rule. As a result, a relatively inefficient rule will be challenged more often by parties negatively affected by it, giving courts additional opportunities to overturn such a rule. In addition, if a rule is inefficient, parties are more likely to attempt to have the dispute resolved in court, rather than reach a settlement that would have to deal with the deadweight loss in the existing rule. Further, as Rubin [1977] and Posner [1992, 519-24, 534-36] describe, the process gives judges (especially appellate judges) the incentives to view parties not in personal terms, but as representatives of a particular activity or part of society. Finally, Anderson, Shughart, and Tollison [1989] indicate that legislatures give judges important incentives to seek efficient outcomes that maximize wealth.

While others(8) argue that there are goals besides efficiency, the efficiency theory would seem to go a long way towards describing the evolution of common law. Further, there does not appear to be any competing positive theory of common law. Thus, if one believes that the Sherman Act is a logical extension of the common law (which seems generally accepted) and that the goal of the common law is economic efficiency (which is somewhat more disputed), one has sufficient evidence to at least suspect that the goal of the Sherman Act was to promote economic efficiency.(9)

While it is an outgrowth of the common law, the Sherman Act is a product of legislative action. The Law and Economics school, as described in Posner [1992, chapters 13 and 19], often distinguishes between the goals of common law (efficiency) and statutory law (wealth transfers or rent seeking). Rubin [1982; 1983] argues that this distinction between common and statutory law is misleading. He contends that laws created both by courts and legislatures prior to the systematic rise of well-organized interest groups (about 1930) are more likely to be devoted to efficiency enhancement, while laws after that period are more likely to be devoted to rent seeking.(10) According to this theory, legislators have a basic general interest in economic efficiency that can be overcome by the efforts of interest groups.

Rubin further contends that what the Law and Economics school calls the "common law" is really a combination of common law with efficiency-enhancing statutes in torts and property law passed largely during the eighteenth and nineteenth centuries. Certainly the antitrust laws are consistent with this description. Thus, using Rubin's theory, efficiency as the goal of antitrust is consistent with the goal of efficiency in a number of other legal areas.

There still remains the question of why Congress would choose to act in the restraint-of-trade area rather than wait for the common law to evolve in the efficient manner. It is clear that the common law, with its respect for precedent and its largely decentralized decision-making apparatus, is a slowly moving evolutionary force. In times of technological change, it might well be more efficient to change the relevant property rights relatively quickly, rather than wait for the judiciary to take action. Thus, the Sherman Act can be seen as an attempt to speed up the common law process. This may also explain why the individual states passed their own antitrust laws around the time of the Sherman Act (see Stigler [1985, 5-6]) and why Canada passed its own federal competition law in 1889, one year prior to the Sherman Act.

The Evolution of Common Law Restraint-of-Trade Cases

A full discussion of how the common law of restraint of trade reached towards economic efficiency is well beyond the scope of this work. A brief examination of the evolution of common law rules regarding covenants not to compete and naked restraints such as price fixing, however, shows how the pre-Sherman Act restraint-of- trade caselaw fits into the Law and Economics framework of judge- written law generating legal rules that reach towards economic efficiency.

The common law on covenants not to compete can be seen to have evolved steadily towards enhancing economic efficiency, balancing the anticompetitive and the efficiency potential of such restraints. Prior to the seventeenth century, English courts appear to have invoked a per se rule against covenants not to compete.(11) In Rogers v. Parrey, 2 Bulst. 136, 80 Eng. Reg. 1012 (1613), however, the Court of the King's Bench distinguished between general restraints (involving the entire kingdom) and partial restraints (relating to a particular town), with the latter being permitted. In Mitchel v. Reynolds (1711) Chief Justice Parker explained the reasons for the law's general hostility towards restraint-of-trade contracts:

First. The mischief which may arise from them (1) to the party by the loss of his livelihood and the subsistence of his family; (2) to the public by depriving it of a useful member. Another reason is the great abuses these voluntary restraints are liable to; as, for instance, from corporations who are perpetually laboring for exclusive advantages in trade and to reduce it into as few hands as possible.

The efficiency reasons for allowing such covenants was expressed by Baron Parke in Mallan v. May, 11 Mees. & W. 652 653 (1843), who wrote about what now would be termed a "free-rider" defense:

Contracts for the partial restraint of trade are upheld ... because it is for the benefit of the public at large that they should be enforced.... Such is the case of disposing of a shop in a particular place with a contract on the part of the vendor not to carry on a trade in the same place. It is, in effect, the sale of a good will, and offers an encouragement to trade by allowing a party to dispose of all of the fruits of his industry.... In such a case the public derives an advantage in the unrestrained choice which such a stipulation gives to the employer of able assistants and the security it affords that the master will not withhold from the servant instruction in the secrets of his trade, and the communication of his own skill and experience, from the fear of his afterwards having a rival in the same business.

Mitchel v. Reynolds used the distinction in Rogers v. Parrey to separate reasonable "partial" restraints that applied to a particular town from unreasonable "general" restraints that applied to the entire kingdom. Such a distinction made sense in a time when "goodwill" was likely to extend only across a relatively small region. As discussed below, later cases attempted to apply this distinction when the relevant trading areas were larger. Using this early rule of reason, a series of English cases from 1711 to 1880 extended the permissible area of restraint available in such contracts from a ten mile radius to worldwide.(12) The clearest articulation of the approach for such cases comes from Chief Justice Tindahl's decision in Horner v. Graves, 7. Bing. 735 (1831):

We do not see how a better test can be applied to this question whether this is or not a reasonable restraint of trade than by considering whether the restraint is such only as to afford a fair protection to the interests of the party in favor of whom it is given, and not so large as to interfere with the interests of the public. Whatever restraint is larger than the necessary protection of the party can be of no benefit to either. It can only be oppressive. It is, in the eye of the law, unreasonable. Whatever is injurious to the interests of the public is void on the ground of public policy.

In the United States, a similar line of cases evolved.(13) Innovations in the U.S. rule of reason continued past the Sherman Act to Chief Justice White's adoption of a market power test for mergers in Standard Oil.

The common law on naked restraints seems to have taken a different path than the law on covenants not to compete. Prior to 1800, price-fixing agreements appear to have been unenforceable no matter what their circumstances.(14) Two cases in the early part of the nineteenth century, Hearn v. Griffin, 2 Chitty. 407 (1815), and Wickens v. Evans, 17 Q.B. 652 (1827), however, applied the "rule of reason" of Mitchel v. Reynolds and found the relevant contracts valid. Both decisions noted the importance of competitors outside the relevant contractual arrangements.

This line of analysis was rejected, however, in the next important English case, Hilton v. Eckersley, 6 E. & B. 47 (1855), where the court opposed all such "naked" restraints as "unreasonable" because they served no positive public purpose. In doing so, the court accepted the position of the defense that "[t]he doctrine laid down in Mitchel v. Reynolds, and other cases, that a restraint of trade may be upheld when there is a good consideration for it, is entirely inapplicable to a case where the restraint is itself the consideration."(15) Yet the court in this case implicitly tempered its per se rule by explaining (at 76) the economic circumstances of the restraint in question. Not until Mogul Steamship Co. v. McGregor, 23 Q.B.D. 598 (1889), aff'd A.C. 25 (1892), was the unenforceability of naked restraints made clear in the English common law.

The American common law followed a similar path. In some cases, naked restraints were upheld, while in other cases they were struck down.(16) The most important common law decisions in this field appear to have been the Ohio Supreme Court decision in Salt Co. v. Guthrie, 35 Ohio St. 366 (1880) and the New York decision in People v. Sheldon, 139 N.Y. 251, 34 N.E. 785 (1893). According to Judge McIlvaine in Guthrie, at 672.

The clear tendency of such an agreement is to establish a monopoly, and to destroy competition in trade, and for that reason, on the ground of public policy, courts will not aid in its enforcement. It is no answer to say that competition in the salt trade was not in fact destroyed, or that the price of the commodity was not unreasonably advanced. Courts will not stop to inquire as to the degree of injury inflicted upon the public. It is enough to know that the inevitable tendency of such contracts is injurious to the public.

Similarly, Chief Justice Andrews in Sheldon (at 264-5) stated:

If agreements and combinations to prevent competition in prices are or may be harmful to trade, the only sure remedy is to prohibit all agreements of that character. If the validity of such an agreement was made to depend on actual proof of public prejudice or injury, it would be very difficult in any case to establish the invalidity, although the moral evidence might be very convincing. The analyses in Guthrie and Sheldon would serve as important underpinnings for Taft's Appeals Court decision in Addyston Pipe.

There are two ways of reviewing this line of cases in the Law and Economics framework. First, while the common law may evolve towards economic efficiency, this does not imply that every innovation in the common law will individually enhance economic efficiency. But it does imply that mistakes in common law decision making (like the decision outlawing vertical restraints in U.S. v. Arnold, Schwinn & Co., 388 U.S. 365 [1967]) are eventually likely to be corrected by further innovations (such as Continental T.V. Inc. v. GTE Sylvania Inc., 433 U.S. 36 [1977], which overturned the Schwinn decision.).

The recent analysis of Grady [1992] implies another way of looking at the cases that upheld naked restraints. Grady posits several efficiency arguments for these decisions. In particular, Grady suggests that the courts were either implicitly using a market power screen (similar to that explicitly used later in Standard Oil) or allowing agreements that solved "core" problems that may exist in some competitive markets. In this context, the line of cases on naked restraints can be seen as a conflict between a per se rule which would minimize court costs but eliminate any efficiencies arising from horizontal restraints and a rule of reason that would allow such efficiencies, albeit with higher litigation costs.

In this context, the rule of reason for naked restraints can be seen to have failed because, unlike the rationale presented in Mallan v. May, public policy arguments made for upholding such covenants were not seen as sufficiently compelling to counterbalance their clear anticompetitive potential. Given this, and the administrative difficulties in enforcing any "reasonableness" criteria, Taft's Addyston Pipe decision has stood, with only a brief interruption, for almost a century.(17)

"Law and Economics" and Lande's Analysis. In light of this background of common law and Law and Economics, it is important to review Lande's analysis. In particular, Lande presents three arguments why the welfare of consumers, rather than economic efficiency, was the congressional goal of the antitrust laws.

First, Congress spent the bulk of the debate discussing consumer welfare (or, as Lande puts it, the "welfare-of-consumers"). No individual Congressmen expressed explicit interest in having the Act promote the goal of economic efficiency (Lande [1982, 94-95]). Second, Congress was generally motivated to pass redistributive, rather than pro-efficiency laws (Lande [1982, 77, 88]). Finally, Lande [1982, 88] argues that Congress was unfamiliar with the concept of economic efficiency. Lande asserts that, since the term "economic efficiency" was known to very few in 1890, Congress could not have been trying to achieve it.

It may be true that the term economic efficiency was unknown to Congress in 1890, but it does not appear to be of great relevance. In general, economists do not assume that economic actors (either private or public) understand the nuances of economic theory before that theory can be used to describe their actions.(18) In the context of the Law and Economics approach, scholars and judges like Bentham and Holmes were striving for hundreds of years to generate economic efficiency without employing that particular terminology. In effect, they knew what efficiency was, and were unwilling to wait for economists to define it. As Landes and Posner [1987, 23] put it, "[p]eople can apply the principles of economics intuitively--and thus "do" economics without knowledge they are doing it."(19)

Take, for example, the famous common law limited liability rule laid down in Hadley v. Baxendale, Ex. 341, 156 Eng. Rep. 145 [1854]. Bebchuk and Shavel [1991, 309] conclude that this rule avoids unlimited liability situations which "might have a significant efficiency cost whenever it is desirable for sellers to use differential precautions for low and high valuation buyers--for the unlimited liability law would result in either high transactions costs for the low valuation buyers and sellers, or in sellers' failing to exercise differential precaution." While it seems unlikely that the court in Hadley v. Baxendale would have recognized Bebchuk and Shavel's statement, that does not reduce the validity of the analysis. Instead, it would appear that the court applied its own (perhaps less precise) terminology to the situation. Indeed, the common law restraint of trade cases discussed above often refer to such ideas as "the public good," or "the ground of public policy," concepts that may well be good proxies for the modern idea of economic efficiency. The fact that Congress did not use the term "economic efficiency" in the debate on the Sherman Act does not imply that economic efficiency was not the underlying goal of the Sherman Act.

Lande's second point is that Congress generally passes laws that do not enhance economic efficiency. In a narrow sense this may be true, but even in the modern era Congress passes laws for economic efficiency, and Rubin gives us reason to believe that this was more likely to occur in 1890. Indeed, Wittman [1989] argues that in general policies in democratic countries tend to reach towards economic efficiency.

This still leaves unanswered the question of how a nineteenth century Congress would have debated a measure designed to promote economic efficiency. That is, how would they have articulated such a goal in the lexicon of contemporaneous political rhetoric? From a common law framework, they would have noted that the economy had evolved so that consumers could no longer adequately protect their property rights. They would then discuss how the new measure would restore these rights. This is precisely what the main focus of the debates (as presented by both Bork and Lande) was, although naturally the actual rhetoric was somewhat more heated. Thus, an examination of the congressional debates cannot be expected to discern between the efficiency and the welfare-of-consumer hypotheses.

Of course, if Congress were intent on redistributing rents to consumers that would have accrued to producers under the common law, the debate would have also been on these lines. But consider the modern debates over trucking and airline deregulation. (See, for example, Robyn [1987, 26-56], and Behrman [1980, 96-103].) They focused on aiding consumers. Yet those laws were clearly designed to enhance economic efficiency. Had Congress wanted to redistribute rents to consumers, they could have arranged to subsidize air travel and truck shipments. Instead, Congress merely restored to consumers the property rights that would generate effective competition. Similarly, an expressed concern for consumers in the debates over the Sherman Act does not imply that Congress was uninterested in the rights of producers. (For a similar argument, see Rule and Meyer [1988, 689].)

To summarize, the Sherman Act can be viewed as a logical and modest extension of the common law. The common law can be seen as an instrument for promoting economic efficiency. Therefore, Congress likely intended for the antitrust laws to enhance efficiency rather than facilitate wealth transfers. The debate on the Sherman Act can thus be viewed as part of attempts by lawmakers to recapture for consumers the rights to which they were entitled under common law in order to generate efficient outcomes.

IV. SUPPORT FOR AND IMPLEMENTATION OF THE SHERMAN ACT

Political Support for the Sherman Act

According to the theory of wealth-distributing legislation (for example, Olson [1965; 1982], some type of strong interest group lobbying effort is necessary to enact legislation that redistributes rents. Conceptually, an interest group such as the "consumer activists" or "consumerists" loosely associated with Ralph Nader, which arose in the late 1960s and early 1970s, could have promoted a consumer rent-seeking antitrust measure. No such group, however, appears to have been a crucial supporter of the Sherman Act. The closest and most important consumer-type group that scholars (DiLorenzo [1985], Stigler [1985], and Thorelli [1955, 58-60]) record from the 1880s and 1890s are the Grangers, a populist movement that was devoted largely to lowering railroad rates for farmers.

As Stigler [1985] points out, however, it is difficult to conclude that the Grangers were the primary force behind the Sherman Act. The Grangers had already obtained their desired legislation in 1887, the Interstate Commerce Act, which established the Interstate Commerce Commission (ICC) to reallocate rents to farmers.(20) It is reasonable to believe that the Grangers approved of the Sherman Act, since it was not inconsistent with their interests. Yet the Grangers did not have a larger stake in its passage than any other group, and it was not certain at the time of its passage whether the Sherman Act would apply to railroads.(21) Lande [1982, 70] appears also to believe that populist sentiment was not responsible for the Sherman Act, as he is clear in his view that the Act was not a measure for distributing wealth from richer to poorer segments of society.

The support for the Sherman Act came from a great many sources and was widespread. As Stigler [1985, 5-6] describes, a number of states across the country passed their own antitrust measures during the same time period in a pattern unrelated to Granger activity across states. This is consistent with a broad-based desire for economic efficiency achieved by making a moderate change in public policy through amending the common law, as discussed by Stigler [1985, 7], or by an efficiency generating compromise among interest groups, as described in general by Becker [1983] and Wittman [1989] and in particular by Buchanan and Lee [1992]. The Act does not seem to have been generated by the activity typically associated with rent-seeking legislation.

Implementation

Modern political economy also posits that the goals of a particular policy will affect how Congress chooses to implement that policy. Congress' designation of the judicial system to interpret the antitrust laws suggests a stronger likelihood that Congress desired economic efficiency to be the goal of the Sherman Act.

Two basic methods were available to Congress in implementing the Sherman Act.(22) First, it could have entrusted the law to the judiciary, as with the common law. Under this arrangement, decisions would be made by judges under the "preponderance of the evidence" or "greater weight of the evidence" standard generally used in civil cases, as described by Cleary [1972, 793-96]. As discussed above in section II, the Law and Economics approach indicates that decisions under this regime will tend to reach towards economic efficiency. Thus, granting courts the authority to determine the meaning of a vaguely worded law such as the Sherman Act is entirely consistent with the precepts of a common law approach promoting economic efficiency.(23)

Alternatively, Congress could have entrusted the enforcement of the statute to an administrative agency such as the ICC. An administrative agency, as defined here using the general approach of Stewart [1975], would create and enforce its own law by making decisions and creating rules. Its actions would be subject to review by the judiciary, but only on a "reasoned consistency" or "arbitrary and capricious" standard.(24) Under such a standard, a court would generally uphold an agency's decision if due process procedures were followed, if there were a reasonable basis to support the agency's decision, and if the agency was acting in a consistent manner.(25)

The early academic theory of administrative agencies argued that such agencies would be more efficient administrators of one part of the law than judges, who have to deal with a wide variety of matters. (For a summary of this rationale see Mitnick [1980, 31].) This idea has been replaced by the "capture" theory, along the lines of Stigler [1970], Posner [1972], Fiorina [1986], and McCubbins, Noll, and Weingast [1987]. According to the capture theory, Congress establishes an administrative agency to implement the political "deal" it has enacted. The agency then adopts a set of administrative procedures to enforce the political contract.

Under this arrangement, should an agency's future political leadership attempt to undo the original congressional political arrangement, it would have to overcome the institutional arrangements already in place. Existing administrative procedures would require a large amount of both time and agency resources to surmount, so that the future leadership of the agency would find it difficult to depart from the mission Congress intended. Similarly, as Shepsle [1979], and McCubbins, Noll, and Weingast [1987] explain, the political deal would be protected from the courts by the "reasoned consistency" standard and from a new legislative political equilibrium by complicated legislative procedures. In effect, the administrative procedures create a bias towards the client interest group in the administrative agency's decisions.

The rise of administrative agencies in the twentieth century is consistent with Rubin's thesis on the goals of law in an era of interest groups. The "capture theory" explains why Behrman [1980, 115-16] found that administrative procedures constituted a significant obstacle to the Civil Aeronautics Board's (CAB) internal deregulation effort in the late 1970s. It is also consistent with the abolition of the CAB as a result of the Airline Deregulation Act of 1978 and the elimination of almost all of the ICC's trucking responsibilities in the Motor Carrier Act of 1980 (Behrman [1980, 75] and Robyn [1987]).

Congress already had administrative agencies in its legislative arsenal in 1890, having created the first federal one (the ICC) three years earlier in 1887. Indeed, the recent research on the Interstate Commerce Act indicates that Congress in 1887 acted in exactly the manner that the theory of legislative choice outlined above indicates. In that debate, the pro-consumer House of Representatives preferred a measure that would outlaw pooling in order to reduce railroad prices. The House also wanted the measure to be enforced by the judiciary because an administrative agency would be likely to be captured by railroad interests. The pro- railroad Senate, on the other hand, desired an administrative agency with no anti-pooling mandate. The final compromise between the two houses created the ICC and banned pooling.(26) Thus, if Congress sought to redistribute rents to consumers through the Sherman Act, both modern theory and the events three years earlier in 1887 imply that it would have set up an administrative agency to enforce its goal. Instead, Congress created the right of private and public action to allow judicial enforcement of the Sherman Act.(27)

Conceivably administrative agencies could have been considered an oddity in 1890 (though Fiorina [1986, 36] points out that by 1887 they were common at the state level) and Congress may have been reluctant to create another one without first evaluating the ICC's performance. Twenty-four years later, however, in 1914, the Congress created the Federal Trade Commission to also enforce the antitrust laws, as well as to handle consumer protection matters. At first glance, the FTC, with its Commissioners and administrative law judges, looks like an administrative agency. Yet when it comes to antitrust matters, FTC cases use the same body of law as Justice Department cases and are also reviewed by appeals courts on a "preponderance of the evidence" standard.(28) By itself, the FTC, like the Department of Justice, does not have the legal authority to stop a merger, declare an industry trade practice anticompetitive, or create law contrary to established precedent without substantial reason. As Posner [1970, 71] once described the agencies' role in the judicial process, "[i]n both cases, the agencies merely propose and the courts dispose." This is in contrast to the FTC's consumer protection authority under the Magnuson-Moss Act of 1975. Under this law, which is generally credited with "revitalizing" the agency, the FTC can pass consumer protection rules subject to judicial review only under a "reasoned consistency" standard.(29)

To summarize, Congress had a choice in 1890 of whether to implement its antitrust policy through either the judiciary or an administrative agency. The entrustment of antitrust decisions to the judiciary in 1890, and again in 1914, suggests that Congress intended economic efficiency to be the goal of the Sherman Act. Had wealth transfers been the goal of the Act, the modern theory of administrative agencies suggests that the Congress would have acted as it did when it regulated the railroads in 1887 and embodied antitrust authority in an administrative agency.

V. CONCLUSION

Theory and evidence indicate that the primary goal of the Sherman Act of 1890 was to enhance economic efficiency. This type of statute was not uncommon before the modern rise of interest groups, nor unknown afterward. The Sherman Act is a logical and modest extension of the common law, which reaches towards economic efficiency. Unlike the Magnuson-Moss Act of 1975, no "consumerist" lobby appears to have exerted enough influence over Congress in 1890 to pass a law that would redistribute wealth via antitrust proceedings. Further, enforcement authority for the Sherman Act was given to the judiciary, rather than to an administrative agency subject to capture by special interests. Thus, the weight of the evidence suggests that the primary goal of the Sherman Act was to maximize economic efficiency.

1. This paper will therefore be a positive, rather than normative, analysis of the goal of the Sherman Act. For a normative analysis, see Areeda and Turner [1978, 29-31].

2. For example, then-FTC Chairman Oliver [1988] endorsed a "price test" by identifying the role of the FTC as preventing price increases or output reductions, which is equivalent to a welfare of consumers standard. Current FTC Chairman Janet Steiger [1989] has stated that maximizing the welfare of consumers is the appropriate goal of antitrust policy. The 1992 Department of Justice and Federal Trade Commission Horizontal Merger Guidelines describes the adverse results of the exercise of market power to be "a transfer of wealth from buyers to sellers or a misallocation of resources" (emphasis added, the "or" representing a change from the "and" in the 1984 Guidelines). Constantine [1990, 168-9] asserts that the welfare of consumers standard is used by the state attorneys general. For a discussion of Lande's influence on this debate, see Kovacic [1990, 1462-3].

3. Because the monopoly firm raises price, it increases its own profits. Posner [1975] contends that these profits will be dissipated in rent-seeking activities as colluding firms seek to capture profits. (Posner does not seem to address anticompetitive actions by dominant firms, which is the example used by Williamson.) Williamson [1977, 713] argues that "rent seeking" in the form of competition would generate entry and thus only part of these profits should be counted as social costs of monopoly.

4. A short list of such scholars would include Areeda and Turner [1978], Easterbrook [1986], Easton [1890], Letwin [1954], Oppenheim, Weston, and McCarthy [1981], and Thorelli [1955], as well as Judge (later Chief Justice) Taft (whose Addyston Pipe decision is discussed below).

5. For specific state cases, see Attorney General v. American Sugar Refinery Co., 7 RY. & CORP. L.J. 83 (Cal. Super Ct. 1890) and People v. North River Sugar Refining Co. 121 N.Y. 582, 24 N.E. 798 (1889). For a general discussion of both the scope and the limits of state antitrust enforcement during this period see May [1987, 507-20].

6. Section 1 of the Act states, "Every contract, combination in the form of trust or otherwise, or conspiracy in restraint of trade or commerce among the several states, or with foreign nations, is declared to be illegal," while Section 2 states "Every person who shall monopolize or attempt to monopolize, or combine or conspire with any other person or persons, to monopolize any part of the trade or commerce among the several states, or with foreign nations, shall be deemed guilty of a felony." 15 U.S.C. 1 and 2. As Lande [1982, 81] points out, "[t]he antitrust laws are among the least precise statutes enacted by Congress."

7. The thinking of this school is perhaps best represented in Posner [1992], although the idea of the law promoting what is now termed economic efficiency is often traced back at least to Holmes [1963, first published in 1881]. Commons [1925] traces this idea back to the early 19th century writings of Jeremy Bentham. See also the symposium "Changes in the Common Law: Legal and Economic Perspectives," Journal of Legal Studies, March 1980, 189.

8. See, for example, "Symposium on Efficiency as a Legal Concern," Hofstra Law Review (8)485, 1980.

9. Under this theory, in the long run antitrust law will seek to achieve economic efficiency no matter what scholars write today on the subject. The long run, however, could conceivably last several decades or even centuries.

10. Of course, there are exceptions to Rubin's rule, such as the establishment of tariffs and the Interstate Commerce Act in the 19th century (redistributing economic rents), and the Airline Deregulation Act of 1978 and the Motor Carrier Act of 1980 (generating economic efficiency). Rubin's theory is similar to Olson's [1965; 1982] thesis that over time legislation becomes less efficiency-enhancing as more and more interest groups affect the political process.

11. See, for example, The Dyer's Case, Y.B. Pasch. 2 Hen. 5, f.5, pl 26 (1415), and The Blacksmith's Case, 2 Leo. 210, 3 Leo. 217 (1587). Alger v. Thacher, 19 Pick. 51, 52 (Mass., 1837), refers to the per se rule as being "old and settled law" by 1415.

12. See, for example, Davis v. Mason, 5 T.T. 118, 101 Eng. Reg. 69 (K.B. 1793), Bunn v. Guy, 4 East 190, 102 Eng. Rep. 803 (K.B. 1803), Mallan v. May 11 Mees. & W. 652 (1843), Harms v. Parsons, 32 Bey. 328, 55 Eng. Rep. 129 (R.C. 1862), and Rousillon v. Rousillon, 14 Ch. D. 351 (1880).

13. See, for example, Pierce v. Woodward, 23 Mass. (6 Pick.) 206 (1828), Chappel v. Brockaway, 211 Wend. 157 (Sup. Ct. N.Y. 1839), Oregon Steam Nav. Co. v. Windsor, 87 U.S. 22 (1874), and Watertown Thermometer Co. v. ool, 51 Hun. 157, 4 N.Y.S. 861 (1889). Kintner [1980, 371-377] also cites a handful of cases where such constraints were held unreasonable. The distinction between partial and general restraints continued in some form in English law until Nordenfelt v. Maxim-Nordenfelt Guns and Ammunition Company, L.R. 1 Ch. 630 (C.A.) (1893) aff'd A.C. 535 (1894), which upheld a world- wide covenant in the sale of armaments. According to Kintner [1980, 72], in the United States this struggle continued until Langit v. Sefton Mfg. Co., 184 Ill. 326 (1900).

14. See, for example, King v. Norris, 2 Keny, 300 (1758), and King v. Eccles, 1 Leach 274 (1783).

15. Summary of argument of Mellish for the defense, Hilton v. Eckersley, 6 E.&B. 72.

16. See, for example, Commonwealth v. Carlisle Bright., N.P. 36 (Pa. 1821), Lee v. Louisville Pilot Benevolent & Relief Ass'n, 65 Ky. 254 (1867), and Skrainka v. Scharringhausen, 8 Mo. 522 (1880).

17. From Appalachian Coals Inc. v. U.S., 288 U.S. 344 (1933) to U.S. v. Socony-Vacuum Oil Co., 310 U.S. 150 (1940).

18. As Friedman [1953, 21] once put it in a famous essay, "[c]onsider the problem of predicting the shots made by an expert billiard player. It seems not at all unreasonable that excellent predictions would be yielded by the hypothesis that the billiard player made his shots as if he knew the complicated mathematical formulas that would give the optimum directions of travel, could estimate accurately by eye the angles, etc., describing the location of the balls, could make lightning calculations from the formulas, and could then make the balls travel in the directions indicated by the formulas." (italics original)

19. Similarly, Liebermann [1986, 387] argues in his analysis of Jewish law dating back over 2500 years that "the evidence is sufficient to support the proposition that ancient lawyers had an implicit but conscious understanding of the role transactions costs play in the formation and execution of voluntary contracts."

20. See Hovenkamp [1988] and Gilligan, Marshall, and Weingast [1989]. As several scholars (for example, Fiorina [1986], Gilligan, Marshall, and Weingast [1989], and Prager [1989]) have noted, the Grangers were aided by the railroads themselves, who also served to benefit from the legislation.

21. This question was not resolved until U.S. v. Trans-Missouri Freight Association, 166 U.S. 290 (1897).

22. The analysis in this section, of course, simplifies somewhat the nature of the implementation question open to Congress.

23. Conceptually, Congress could also have given specific instructions for rent-distribution in a statute and then have the judiciary enforce the statute. As discussed above, however, supra note 6, the antitrust laws contain no such instruction.

24. See, for example, Greater Boston Television Corporation v. F.C.C., 444 F.2d. 850-853 (1970), Chevron U.S.A. Inc. v. National Resource Defense Council, Inc., 467 U.S. 837 (1984) and Stewart [1975, 1680], as well as 5 U.S.C. Section 706(2)(A).

25. A good example of this difference can be seen in Judge Richard Posner's decision for the Court of Appeals in United Air Lines v. CAB, 766 F.2d 1107 (1985). (This case concerned the Civil Aeronautics Board's restrictions on the use of display preference in airline computer reservation systems on competitive grounds.) In upholding the CAB, the Court indicated that while it had substantial doubts about the CAB's conclusions on the competitive implications of display preference, all the law required was that the CAB make a finding based on an "arbitrary and capricious" standard, not on the preponderance of the evidence.

26. See Fiorina [1986, 38] and Gilligan, Marshall and Weingast [1989, 48]. The results in Prager [1989] indicate that stock market participants also understood the implications of these actions.

27. This argument also refutes the position of DiLorenzo [1985] and Hazlett [1992] that Congress intended the Sherman Act to protect small business. It seems unlikely that if Congress meant to reallocate rents to this class they would have given enforcement of a vague law to the judiciary.

28. Compare, for instance, the decision in FTC v. Indiana Federation of Dentists, 745 F.2nd 1124 (1984), where the FTC was overruled by an Appeals Court, to the discretion shown to the Environmental Protection Agency in Chevron, supra note 24, which was decided four months earlier.

29. For a discussion of the narrowly focused "consumerist" interest groups that supported the Magnuson-Moss Act and the effects of the administrative procedures of that Act, see Kleit [1992, 28-30].

REFERENCES

Anderson, Gary M., William F. Shughart II, and Robert D. Tollison. "On the Incentives of Judges to Legislate Wealth Transfers." Journal of Law and Economics, April 1989, 215-28.

Areeda, Phillip, and Donald F. Turner. Antitrust Law, vol. 1. Boston: Little, Brown and Company, 1978.

Bebchuk, Lucian Ayre, and Steven Shavell. "Information and the Scope of Liability for Breach of Contract: The Rule of Hadley v. Baxendale." Journal of Law, Economics, and Organization 7(2), 1991, 284-312.

Becker, Gary S. "A Theory of Competition Among Pressure Groups for Political Influence." Quarterly Journal of Economics, August 1983, 371-400.

Behrman, Bradley. "Civil Aeronautics Board," in The Politics of Regulation, edited by James Q. Wilson. New York: Basic Books, 1980, 75-120.

Bork, Robert H. "Legislative Intent and the Policy of the Sherman Act." Journal of Law and Economics, April 1966, 7-48.

Buchanan, James M., and Dwight R. Lee. "Private Interest Support for Efficiency Enhancing Antitrust Policies." Economic Inquiry, April 1992, 218-24.

Cleary, Edward W., ed. McCormick's Handbook of the Law of Evidence. St. Paul: West Publishing Co., 1972.

Commons, John R. "Law and Economics." Yale Law Journal, February 1925, 371-82.

Constantine, Lloyd. "Antitrust Federalism." Washburn Law Journal 29(2), 1990, 163-84.

DiLorenzo, Thomas J. "The Origins of Antitrust: An Interest-Group Perspective." International Review of Law and Economics, June 1985, 73-90.

Easterbrook, Frank. "Workable Antitrust Policy." Michigan Law Review, August 1986, 1696-713.

Eaton. "On Contracts in Restraint of Trade." Harvard Law Review, October 1890, 128-35.

Fiorina, Morris P. "Legislator Uncertainty, Legislative Control, and the Delegation of Legislative Power." Journal of Law, Economics, and Organization 2(1), 1986, 33-51.

Fisher Alan A., Frederick I. Johnson, and Robert H. Lande. "Price Effects of Horizontal Mergers." California Law Review 77(4), 1989, 777-827.

Friedman, Milton. Essays in Positive Economics. Chicago: University of Chicago Press, 1953.

Gilligan, Thomas W., William J. Marshall, and Barry R. Weingast. "Regulation and the Theory of Legislative Choice: The Interstate Commerce Act of 1887." Journal of Law and Economics, April 1989, 35-61.

Grady, Mark F. "Toward A Positive Economic Theory of Antitrust." Economic Inquiry, April 1992, 225-41.

Hazlett, Thomas W. "The Legislative History of the Sherman Act Re- examined." Economic Inquiry, April 1992, 263-76.

Holmes, Oliver Wendall, Jr. The Common Law, edited by Mark DeWolfe Howe. Cambridge: Harvard University Press, 1963.

Hovenkamp, Herbert. "Regulatory Conflict in the Gilded Age: Federalism and the Railroad Problem." Yale Law Journal, May 1988, 1017-72.

Kintner, Earl W. Federal Antitrust Law, vol. 1. Cincinnati: Anderson Publishing, 1980.

Kleit, Andrew N. "Beyond the Rhetoric: An Inquiry Into the Origins of the Sherman Act." Federal Trade Commission Working Paper No. 195, August 1992.

Kovacic, William E. "The Antitrust Paradox Revisited: Robert Bork and the Transformation of Modern Antitrust Policy." Wayne Law Review 36(4), 1990 1413-71.

Lande, Robert H. "Wealth Transfers as the Original and Primary Concern of Antitrust: The Efficiency Interpretation Challenged." Hastings Law Journal, September 1982, 65-131.

Landes, William M., and Richard A. Posner. The Economic Structure of Tort Law. Cambridge: Harvard University Press, 1987.

Letwin, William L. "The English Common Law Regarding Monopolies." University of Chicago Law Review 21(3), 1954, 355-85.

Liebermann, Yehoshua. "Economic Efficiency and Making of the Law: The Case of Transactions Costs in Jewish Law." Journal of Legal Studies, June 1986, 387-404.

May, James. "Antitrust Practices and Procedure in the Formative Era: The Constitutional and Conceptual Reach of State Antitrust Law 1880-1918." University of Pennsylvania Law Review 135(3), 1987, 495-593.

McCubbins, Matthew D., Roger G. Noll, and Barry R. Weingast. "Administrative Procedures as Instruments of Political Control." Journal of Law, Economics, and Organization 3(2), 1987, 243-77.

Mitnick, B. M. The Political Economy of Regulation: Creating, Designing, and Removing Regulatory Forms. New York: Columbia University Press, 1980.

Oliver, Daniel. "Statement of Chairman Oliver." Antitrust Law Review, Spring 1988, 235-44.

Olson, Mancur. The Logic of Collective Action. Cambridge: Harvard Press, 1965.

-----. The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities. New Haven: Yale University Press, 1982.

Oppenheim, S. Chesterfield, Glen E. Weston, and J. Thomas McCarthy. Federal Antitrust Laws: Cases, Text, and Commentary, 4th ed. St. Paul: West Publishing Co, 1981.

Posner, Richard A. "Do We Really Need an FTC?" Antitrust Law and Economics Review 3(3), 1970, 65-98.

-----. "The Behavior of Administrative Agencies." Journal of Legal Studies, January 1972, 305-47.

-----. "The Social Costs of Monopoly and Regulation." Journal of Political Economy, August 1975, 807-27.

-----. Economic Analysis of Law, 4th ed. Boston: Little, Brown, and Company, 1992.

Prager, Robin A. "Using Stock Price Data to Measure the Effects of Regulation: The Interstate Commerce Act and the Railroad Industry." Rand Journal of Economics, Summer 1989, 280-90.

Robyn, Dorothy. Breaking the Special Interests. Chicago: University of Chicago Press, 1987.

Rubin, Paul H. "Why is the Common Law Efficient?" Journal of Legal Studies, January 1977, 51-63.

-----. "Common Law and Statute Law." Journal of Legal Studies, June 1982, 205-23.