1 Hereafter “F&F.” A brief pedogogical critique of the book is in order. Students in my public choice seminar had difficulty with the book, partly because the authors omit any discussion of the basics of public choice theory. They do not take the time to explain ordinal and cardinal preference rankings; constant-sum, cooperative, and mixed-motive games; or the nature of solutions to games and how solution concepts differ. Professors thinking of using the book should either explain matters like these in a preliminary lecture or assign introductory readings. It may also be helpful to have students read some of the primary materials on which Farber and Frickey rely, especially an article or two on civic republicanism.Google Scholar

2 Actions can be irrational in the narrow economic sense but rational in a larger sense. A person acts in an economically irrational manner by choosing an inefficient means of advancing his or her self-interest. A person may be said to act rationally in a wider sense “if, after considering all of [his or her] concerns—moral, altruistic, familial, narrowly self-interested, and so forth—[the person] then chooses coherently in trading each off against the others, or even in refusing to make certain trade-offs.” Russell Hardin, Collective Action 10 (Baltimore: Resources for the Future, Johns Hopkins University Press, 1982). Obviously, acts that are broadly rational may be narrowly irrational.Google Scholar

3 In an election of any size, there is only a razor's-edge possibility that one's vote will influence the outcome, and there is only a probability that a winning candidate will enact policies that make one better off. Voting therefore has an expected private payoff that is virtually nil. Moreover, the election of good candidates, “good” being defined from the perspective of one's self-interest, is an outcome on which one can free-ride. One can benefits from the election of good candidates even if one stays at home instead of going to the polls. Finally, voting is costly. It takes time to register and vote. It takes even more time to learn where candidates stand, time that could be spent doing other, more profitable things. Voting is therefore an inefficient means of pursuing one's self-interest. See Dennis C. Mueller, Public Choice 120–24 (Cambridge: Cambridge University Press, 1979). Glen Robinson discusses the rationality and importance of voting in an interesting way in American Bureaucracy 132–34 (reviewed here) (hereafter “Bureaucracy”.).Google Scholar

4 Ideology probably explains a lot of voting behavior, including the tendency of some people to vote when they are angry but not when they are content and the fact that people sometimes use their votes to “send a message” rather than to elect a candidate. Whether or not one agrees that ideology can explain conduct, it clearly is extremely difficult to explain lots of voting behavior on narrow economic grounds. For example, in the recent presidential election, several major television networks refrained from predicting the outcome until the polls closed on the West Coast. They did so in response to a widely held fear, based on conduct in a prior election, that voters on the West Coast would leave the polls in droves if the networks called the election on the basis of results in other regions. Given the minuscule impact of any person's vote in a presidential election, and given that no West Coast voter could change the results in other regions, it seems impossible to explain the feared reaction on narrowly economic grounds.Google Scholar

5 An example may help the reader understand how, in theory, an officeholder can secure majority support for an outcome by rigging the order in which alternatives are presented and the way they are compared. Suppose an imaginary House of Representatives contains members 1, 2, and 3. Member 1, the officeholder who controls the agenda, prefers policies A, B, and C in that order. Member 2 prefers the policies in the order B, C, A, and member 3′s preference ranking is C, A, B. Member 1 can ensure that the House will choose A, even though members 2 and 3 prefer C to A, by requiring the House to decide between policies B and C first in a pair-wise contest, with the winner to be placed in a runoff with A. Members 1 and 2 will vote for B, thereby eliminating C in a pair-wise contest. Then, when A and B are paired, the House will choose A, because members 1 and 3 prefer A to B. Similar analyses show that, by pairing A and B first, member 1 could assure the selection of policy C, and by pairing A and C first, Member 1 could assure the choice of policy B. The example also shows how members 2 and 3 could manipulate the outcome by voting strategically, that is, against their true preferences. If, in the initial contest between policies B and C, member 2 were to vote for C instead of B, C would win. Then, in the contest between C and A, C would win because members 2 and 3 both prefer C to A. By voting strategically in the first round, member 2 can secure the selection of C rather than A. For Riker's views, see William Riker, Liberalism against Populism: A Confrontation between the Theory of of Democracy and the Theory of Social Choice (San Francisco: W. H. Freeman & Co., 1982).Google Scholar

6 “Rents” are “payments for the use of an economic asset in excess of the market price.”Macey, Jonathan R., “Promoting Public-regarding Legislation through Statutory Interpretation: An Interest Group Model,” 86 Colum L Rev. 223, 224 n.6 (1986) (quoted in F&F at 15 n.10). “Rent-seeking” is any attempt to acquire rents by government interference with a market.CrossRefGoogle Scholar

7 Obviously, this is a nutshell account of a work of broad scope. Readers who want a fuller description can consult William Dubinsky, Review of Law and Public Choice , 90 Mich. L Rev. 1512 (1992).CrossRefGoogle Scholar

8 I found no reviews of Bureaucracy and only one reference to it, on a minor point, when I searched the Lexis and Westlaw data bases in September 1992. A phone call to the University of Michigan Press uncovered a glowing two-paragraph review by W. P. Browne in Choice 643 (1991). A subsequent search by Mary Ann Nelson, Associate Director of the Tarlton Law Library at the University of Texas, bore no more fruit.Google Scholar

9 F&F at p. 93 (quoting Easterbrook, “Foreword: The Supreme Court and the Economic System,” 96 Harv. L Rev. 4, 60 (1984).Google Scholar

10 Other scholars, including some who care little for public choice theory, share this view. See, e.g., Martin H. Redish, The Federal Courts in the Political Order: Judicial Jurisdiction and American Political Theory 19–20 (Durham, N.C.: Carolina Academic Press, 1991).Google Scholar

11 Justice Scalia has stated his views on the use of legislative history in several cases. For a list, see F&F at 89 n.3. He also presented an address on the subject at the University of Texas School of Law. Antonin Scalia, “Legislative History: Judicial Abdication to Fictitious Legislative Intent,” Fourteenth Annual Tom Sealy Law and the Free Society Lecture, 20 Jan. 1986 (videotape on file at the Tarlton Law Library, University of Texas School of Law). Because Justice Scalia is unwilling co circulate the text of his lecture, 1 encourage readers who are interested in his views on statutory interpretation to request copies of the videotape. Judge Easterbrook's views on statutory interpretation can be found in Frank Easter-brook, “The Role of Original Intent in Statutory Construction,” 11 Harv. & Pub. Poly 59 (1988); and id., “Statutes' Domains,” 50 U. Chi. L Rev. 533 (1983).Google Scholar

12 Judge Posner offers the military analogy in The Problems of Jurisprudence 269–73 (Cambridge, Mass.: Harvard University Press, 1990). Interestingly, he uses the analogy to show that “[e]ven in a system that is a model of deference, obedience, and hierarchy—the military—orders must at times be interpreted, which imposes on subordinates a duty of creativity and imagination.”Id at 271 (italics omitted).Google Scholar

13 Hamilton's Essay No. 78 in The Federalist suggests that “the independence of the judges” is a structural feature of government that enables federal judges to blunt the effects of constitutional but “unjust and partial laws” and that discourages Congress from enacting such laws. The Federalist No. 78 at 470, ed. C. Rossiter (New York: Penguin Books, 1961). This bit of history runs contrary to the notion that the Constitution places federal judges in the role of Congress's agents on questions of statutory construction.Google Scholar

14 Members of Congress can submit amicus curiae briefs to federal courts but federal judges are not obligated to decide cases as these briefs recommend.Google Scholar

15 Congress can refuse to raise judges' salaries and it can impose budgetary restrictions on the courts. But these are weak control mechanisms. For one thing, they sweep too broadly. They do not enable Congress to punish and reward judges selectively, even though judges differ in their willingness to respect Congress's intent. For another thing, battles between Congress and the federal judiciary over salaries or budgetary matters appear to be more nearly disputes between equals than disputes between superiors and subordinates.Google Scholar

16 For statistics on the tendency of presidents to appoint members of their own party to federal judgeships, see Goldman, Sheldon, “The Bush Imprint on the Judiciary: Carrying on a Tradition,” 74 Judicature 294 (1991).Google Scholar

17 Eskridge, William K. Jr., “Overriding Supreme Court Statutory Interpretation Decisions,” 101 Yale LI. 331, 335–39 (1991) (since 1975 each Congress has overridden an average of 12 Supreme Court statutory decisions and 23 statutory decisions of lower courts).CrossRefGoogle Scholar

18 Overrides may also occur when the sentiment in Congress changes. An override due to a change of sentiment does not indicate that a federal judge ignored the wishes of the enacting Congress. To the contrary, the judge may have adhered to the intent of the enacting Congress surpassingly well.Google Scholar

19 The reforms Farber and Frickey recommend include a safe harbor under the takings clause for laws that benefit diffuse interest groups at the expense of compact ones (at 71–73); limitations on campaign expenditures by business and labor political action committees (at 132–35); tax-favored status for contributions to political parties (at 135); heightened scrutiny of congressional delegations of power to private interests and subgroups of legislators (at 136–39); continued evolution of common law liability rules in favor of disorganized and diffuse interests (at 140–41); and rules of statutory construction that deny “well-organized, political powerful, and wealthy” interests the benefit of any exemptions from liability that are not expressly conferred (at 141–42). The discussion of due process of lawmaking, which emphasizes the need to hold Congress to procedural formalities, appears at 118–43.Google Scholar

20 Public goods theory and public choice theory are not the same thing. Public choice theory is the economic analysis of nonmarket decision making. Public goods theory is the economic analysis of goods like military defense, police protection, and pollution control that are unlikely to be produced efficiently in markets and that are often provided collectively instead. One might use public choice theory to study a collective decision to produce a public good. For discussions of public goods, see Robin W. Boadway & David E. Wildasin, Publlc Sector Economics 85–95 (2d ed. Boston: Little, Brown & Co., 1984); Duncan Snidal, “Public Goods, Property Rights, and Political Organizations,” 23 Int'l Stud Q. 532 (1979).Google Scholar

21 There is a mistake in Robinson's characterization of public goods. At 30 he writes: “If goods are collectively enjoyed, then the marginal cost of supplying the nth unit must be zero.” That is not what Robinson means to say. He means to say that the marginal cost of extending enjoyment of the nth unit to the ith person must be zero, as an example he provides makes clear. He writes that “in the ordinary mode of over-the-air broadcasting the cost of supply for each extra viewer is zero.”Id. Thus, although it is costly to add a new station to the airwaves (a positive marginal cost of supplying an additional unit), any number of viewers can receive the station (zero marginal cost of supplying additional viewers within the broadcast area). Even this characterization can be contested, however, because public goods can be undersupplied even when the marginal cost of supplying additional consumers is positive.Google Scholar

22 At 62, Robinson writes, “The ambition of constitutional review must be to prevent the ambiguity of public welfare from being exploited by particular interest groups seeking purely private gain at public expense. Regulatory laws ostensibly designed to correct market failures should be judged by whether the legislature has a coherent story, a theory in the broad sense of the term, of what the failure is and how its regulatory program will correct it.” And at 66–67 he writes, [C]oerced wealth transfers are rightly treated by economists as a social waste …, but treating [them] as only waste would miss the point of moral interest. We do not need a course in welfare economics to teach us that theft is wrong. There are, of course, ambiguities when we seek to apply this simple moral intuition to the public realm, and it is probably too provocative to identify legislative wealth transfers as simple theft. But … the essential lesson is the same: power that does not seek the justification of principle has no claim to our loyalty or our respect. The purpose of the rule of law is to ensure that some plausible justification is forthcoming.Google Scholar

23 The obvious weakness of this standard of review is that it is too easy to meet to be of much use. For possible responses to this objection, see text accompanying note 33–35.Google Scholar

24 Robinson at 63. The “could” in the quotation should be read as “actually did,” in the sense that the “convincing justification” must actually be presented to a court. I am not sure whether Robinson takes the further step of requiring that the reasons offered in court be the reasons that actually led the legislature to enact a statute, but I think he does not. Nor, I think, would Robinson require Congress to state purposes and aims in statutory preambles, although when such preambles appear, he would encourage the courts to take Congress at its word.Google Scholar

25 Robinson at 63 does not greatly care “what specific constitutional clause is invoked …, the equal protection clause …, the takings clause …, or the due process clause …. The critical point [for him] is the character of the legislation and the nature of the justification required to sustain its constitutionality.”.Google Scholar

26 Lochner v. New York, 198 US. 45 (1905).Google Scholar

27 Williamson v. Lee Optical Co., 348 U.S. 483 (1955).Google Scholar

28 United States v. Carolene Products Co., 304 U.S. 144 (1938).Google Scholar

29 On (e), see Robinson at 155 (“It is in the case of broadly delegated legislative power that courts have the broadest freedom to determine the rationality and legitimacy of agency action”).Google Scholar

30 463 U.S. 29 (1983).Google Scholar

31 463 U.S. 29, 47 (1983).CrossRefGoogle Scholar

32 Judges may have conformed to (d) when construing various antitrust acts. Although Congress's desire to protect small businesses from competition by larger, more efficient operations may have provided the impetus for these acts, judges have recently declined to construe the acts as protectionist measures. Instead, they have used the acts to prevent anticompetitive combinations that threaten consumers. See Lino A. Graglia, “One Hundred Years of Antitrust,”Pub Interest No, 104, p. 50 (Summer 1991). Judges have thus applied a public interest or market failure standard.Google Scholar

33 Farber and Frickey identify Richard Epstein, Cass Sunstein, and Jerry Mashaw as their targets.Google Scholar

34 My colleague Douglas Laycock points out that a classic takings case—confiscating one person's land for use as a public park—appears to fall within Farber and Frickey's safe harbor.Google Scholar

35 I derived this insight from Elhauge, Einer R., “Dm Interest Group Theory Justify More Intrusive Judicial Review 101 Yale L.J. 31, 51–52 (1991). For readers interested in the topics discussed in the review, Elhauge's article is a must read.CrossRefGoogle Scholar

36 Farber and Frickey may yield on this point. They “concede that at least some … legislation may be hard to justify based on anybody's view of social justice” (at 35; italics in the original). They thus appear to agree that some attempts by interest groups to seek rents may be impossible to disguise.Google Scholar

37 Obviously, this is an empirical proposition, not a theoretical one, and it may be inaccurate. However, as I explain in the text, the important point for present purposes is that it appears to be consistent with Farber and Frickey's empirical beliefs.Google Scholar

38 This argument reflects the belief, embodied in a number of “government in the sunshine” and “open meetings” laws, that publicity encourages accountability on the part of public officials. Robinson's proposal can thus be thought of as substituting judges for the public, which often fails to pay attention even when political proceedings occur in plain view.Google Scholar

39 As used in the text, the word “desirable” implies only that a successful collective action would yield a net gain for the participants. It does not mean that a collective action would necessarily be desirable from any other point of view.Google Scholar

40 This conclusion follows even though access to the store is potentially a private good. If a collective action did form, handicapped persons who contributed to it could be issued cards and the merchant might exclude all handicapped persons without cards from the store. I do not mean to imply that this arrangement would be permissible under existing laws, only that access could be a private good. Farber and Frickey ignore the collective action problem in their discussion of handicapped access. After observing that “it may well cost society more to give the access than the handicapped would be willing to pay to obtain it,” they note parenthetically that “this is probably true; otherwise the market would already offer access to the handicapped” (at 35). The point of this paragraph is that, because of the need for collective action, the premise does not imply the conclusion.Google Scholar

41 It is tempting, but ill-advised, to respond to this point by wondering why, if handicapped people have difficulty acting collectively in markets, they are able to do so in politics. The answer is simply that different conditions attend different collective actions. Of particular interest in the political setting is the ability of a single organization to lobby on behalf of an unlimited number of persons and to leverage a small budget into a large wealth transfer. For a general discussion, see Hardin, Colletive Action (cited in note 2).Google Scholar

42 A brief but very thoughtful discussion of the circumstances in which it is better to provide goods than cash can be found in Russell Hardin, Morality within the Limits of Reason 160–65 (Chicago: University of Chicago Press, 1988).Google Scholar

43 Id at 134.Google Scholar

44 One may question judges' competency or ability to assess the soundness of the reasons offered in support of policies with redistributive effects. Robinson's discussion of the State Farm case suggests that he expects a fair degree of sophistication from judges. At the very least, they must be able to understand the facts sufficiently well to follow and to spot weaknesses in the justifications offered for the policies. Robinson may overstate the ability of judges to do that.Google Scholar

45 See also Bureaucracy at 181 (an “attempt … to formulate very specific guidelines for review [of administrative decisions] is unlikely to prove successful …. Some very general guidelines allocating decision-making responsibility and burdens of proof are appropriate, but the quest for some kind of algorithm of review seems to me a waste of time”).Google Scholar

46 Consider 42 U.S.C. 1988, the Civil Rights Attorneys' Fees Awards Act, which authorizes federal judges to grant “reasonable” awards in favor of prevailing parties. Although the statute provides no standards for measuring awards, judges have supplied so many standards that the law is now thick with them. See, e.g., City of Burlington v. Dague, 112 S. Ct. 2638 (1992) (no contingency multipliers); and Kay v. Ehrler, 111 S.Ct. 1435 (1991) (no fee awards in favor of plaintiffs, including lawyers, who represent themselves). One can say the same of the due process clause of the Constitution as it applies to deprivations of property. The clause says nothing about the amount of process a person is due, but the courts have nonetheless provided considerable guidance. See, e.g., Connecticut v. Doehr, 111 S.Ct. 2105 (1991) (invalidating statute authorizing prejudment attachment of real estate without notice or a prior hearing).Google Scholar

47 With respect to judicial errors, Robinson says only that one must expect mistakes of commission whenever judges exercise the power of constitutional review, but that one must also expect mistakes of omission when they do not (at 65). The point is correct, but it neither responds to Farber and Frickey's concern nor establishes that judges will decide cases in a principled way, as Robinson wants.Google Scholar

48 With respect to facts, substantive laws supply criteria of relevance but procedural laws regulate the manner in which facts are proved and the standard to which trial judges are held on review.Google Scholar

49 See Kornhauser, Lewis A. & Sager, Lawrence G., “Unpacking the Court,” 96 Yale L.J. 82 (1986) for an argument to the effect that multijudge panels may encourage accuracy and consistency in adjudication and coherence in law.CrossRefGoogle Scholar

50 A better example of this effect would involve an attempt by legislator A to “buy” legislator B's support for a good bill by offering to support a bad bill legislator B wants. If both legislators knew that the courts would invalidate the bad bill, A would be unable to “purchase” B's vote.Google Scholar

51 See Bureaucracy at 65 (urging tolerance for review of economic legislation similar to that accorded review under the First Amendment).Google Scholar

52 My understanding of legal positivism, the jurisprudence of which H. L. A. Hart was the leading exponent, is that it is a descriptive account of the law. It contains normative elements, including the internal aspect or sense of legitimacy to which Robinson refers, because a legal system is a system of norms. But one can be a legal positivist without endorsing any norms a legal system contains. The point of legal positivism is to make one aware that law is a system of norms, not to convince one to endorse particular norms. To the contrary, a fundamental point of legal positivism is that a possible legal system could be devoid of morally attractive norms. See Coleman, Jules L., “Negative and Positive Positivism,” 11 J. Legal Stud 139 (1982); Silver, Charles, “Elmer's Case: A Legal Positivist Replies to Dworkin,” 6 Law & Philosophy 381 (1987).CrossRefGoogle Scholar

53 The principle “Congress can do whatever it wants” would not free Congress from the requirement of following recognized procedures when making laws. Nor would its use remove properly enacted statutes from the class of things properly called laws, as long as the formal requirements of the concept of law, e.g., publicity, were met.Google Scholar

54 I have some doubts as to whether I have correctly understood the view I describe in this paragraph because I found Robinson's presentation difficult to follow. He develops his view, which in the end seems straightforward, by critiquing moral and social theories advanced by others and by sifting the wheat from the chaff. As a result, although I followed the critical discussions individually, I had difficulty sizing up Robinson's position and seeing how he derived it.Google Scholar

55 See Sunstein, , “Lochner's Legacy,” 87 Colum. L Rev. 873 (1987); id., “Naked Preferences and the Constitution,” 84 Colum L Rev. 1689 (1984); id“Beyond the Republic Revival,” 97 Yale LJ. 1539 (1988). Sunstein believes that market outcomes are suspect because they reflect political allocations of ownership rights that cannot be justified in market terms and because preferences that drive market transactions are often suspect. For an excellent critique of Sunstein, see Jerry Mashaw, “As If Republican Interpretation,” 97 Yak L.J. 1685 (1988).CrossRefGoogle Scholar

56 An excellent source of readings on public choice theory with comments from the perspective of moral theory is Brian Barry & Russell Hardin, eds., Rational Man and Irrational Society? (Beverly Hills, Cal.: Sage Publications, 1982).Google Scholar