Reply to: Richard L. Hasen & John G. Matsusaka, Some Skepticism About the "Separable Preferences" Approach to the Single Subject Rule: A Comment on Cooter & Gilbert, 110 Colum. L. Rev. Sidebar 35 (2010), http://www.columbialawreview.org/sidebar/volume/110/35_Hasen.pdf.
The single subject rule, a widespread and oft-litigated state constitutional provision limiting ballot initiatives to one "subject," has confounded judges, lawyers, and scholars for decades. The problem grows from the inability to define "subject" with precision. In A Theory of Direct Democracy and the Single Subject Rule,1 we attempt to solve this problem. We propose a democratic process theory of the rule, which interprets "subjects" in terms of voters' preferences. Our theory yields a precise, objective test for determining if an initiative complies with the rule. Proper application of our test would achieve the rule's purposes of eliminating logrolling and riding.2
Professors Richard Hasen and John Matsusaka, experts in election law and direct democracy, are skeptical of our approach.3 We appreciate their thoughtful comments, which have contributed helpfully to the debate. However, we think their skepticism misses the mark. They seem to confuse opposition to the single subject rule itself with opposition to our test.
This distinction matters. We support the rule, while Hasen4 and perhaps Matsusaka oppose it. That debate is important, but it is also academic; we know of no state contemplating repeal. Separate from that debate, the heart of our paper takes the rule and its purposes for granted and develops a framework rooted in economic and political theory for applying it. We believe that Hasen and Matsusaka, while debating the merits of the rule, give our framework short shrift.
I. Logrolling in Initiatives
We support the single subject rule for initiatives because we believe that logrolling-roughly, the combining of unpopular proposals into one omnibus initiative that will command majority support-in this context is likely to be socially harmful.5 Hasen and Matsusaka challenge this position, arguing that a particular logroll may be socially beneficial rather than harmful.6 They are correct. The question, however, is not whether any particular initiative logroll is beneficial, but whether the average initiative logroll is beneficial.7
The Coase Theorem posits that parties will bargain to an efficient outcome when transaction costs of bargaining are low.8 As transaction costs of bargaining rise, parties are likely to reach inefficient outcomes. Transaction costs of bargaining in direct democracy are high; tens of thousands of voters cannot compromise with one another. We therefore conclude that the average initiative logroll is inefficient. An inefficient logroll fails to maximize and may decrease aggregate utility. We therefore have a reasonable belief that the average initiative logroll is socially harmful.
This logic does not imply that few logrolled initiatives will appear on the ballot. Initiative sponsors often can combine policy proposals freely.9 However, those sponsors do not represent citizens. Consequently, bargains that benefit them may not benefit society.
We can summarize our argument as follows. Socially beneficial logrolling requires faithful agents of the citizenry who can bargain with one another. Initiative sponsors are not faithful agents of the citizenry, and citizens themselves cannot bargain.
Political theory matters too. Direct democracy aims to empower majorities. A logrolled initiative involves multiple policy dimensions. As we explain in our paper, majority will is incoherent in the context of multidimensional policy proposals.10 Forbidding logrolling confines initiatives to single policy dimensions where majority will is meaningful.11
Hasen and Matsusaka do not explicitly reject these reasons for opposing logrolling in initiatives. Instead, they raise the possibility that logrolling in initiatives generally is beneficial. Without a theory to justify it, we do not think this possibility sustains their position.
This debate is important but theoretical. In practice, judges accept that preventing logrolling is the primary purpose of the single subject rule.12 The heart of our paper takes that purpose for granted and develops a test that would accomplish it. In challenging that purpose, Hasen and Matsusaka challenge the single subject rule itself, not our test.
II. Ideology and Enforcement
The democratic process theory encourages judges applying the single subject rule to focus on voters' ability to make independent judgments about the elements of a challenged initiative.13 Hasen and Matsusaka fear this approach would lead judges to "be more aggressive" in applying the rule and to engage in ideological decisionmaking.14
Our approach does not direct judges to apply the rule aggressively. Rather, it provides a test that, given sufficient information about voters' preferences, will achieve the rule's purposes. Proper application of our test would only increase the rate at which courts invalidate initiatives if more initiatives contain logrolls or riders than traditional tests-rightly or wrongly-identified.
Even if courts applying our test invalidated a higher fraction of challenged initiatives than before, the amount of litigation should fall. Our test has objective characteristics, while traditional single subject tests have only subjective characteristics. Consequently, initiative sponsors in states that use our test could more easily draft their measures to comply with the rule. Rational litigants would not challenge those measures in court.
With respect to litigated cases that rely on our test, if judges lack sufficient information about voters' preferences, or if they have private information, they could inject ideology into their decisionmaking. The question is: Would our approach lead to more ideological decisionmaking than traditional single subject tests? For two reasons we think the answer is no. First, because it has objective characteristics, our test should shrink the universe of indeterminate cases in which ideology matters. Second, research by Gilbert and separate work by Hasen and Matsusaka suggests that judges employing traditional single subject tests are influenced by their personal ideologies.15 We have no reason to believe adjudication under our test would make matters worse.
Given these arguments, why do Hasen and Matsusaka advocate "a restrained approach" to single subject adjudication using traditional tests?16 We surmise that it is because they oppose the rule, and lenient enforcement of it is almost as good as repeal. Again, they focus on the rule, not on our test for applying it.
Conclusion
Our paper seeks to overcome indeterminacy in single subject adjudication by developing an objective framework for applying the rule that would achieve its purposes. We do not believe Hasen and Matsusaka address the merits of that framework. The single subject rule is here to stay, and we believe our paper provides a workable method for adjudicating it.
* Herman F. Selvin Professor of Law, University of California, Berkeley.
** Associate Professor, University of Virginia School of Law.
1 110 Colum. L. Rev. 687 (2010).
2 For a mathematical proof, see id. at 727–30.
3 Richard L. Hasen & John G. Matsusaka, Some Skepticism About the "Separable Preferences" Approach to the Single Subject Rule: A Comment on Cooter & Gilbert, 110 Colum. L. Rev. Sidebar 35 (2010), http://www.columbialawreview.org/sidebar/volume/110/35_Hasen.pdf.
4 See Richard L. Hasen, Ending Court Protection of Voters from the Initiative Process, 116 Yale L.J. Pocket Part 117, 117 (2006) (calling for repeal of rule).
5 See Cooter & Gilbert, supra note 1, at 697-703 (arguing logrolled initiatives are socially harmful because net social utility may be negative, such initiatives are unstable, and such initiatives reflect preferences of a random majority). For a more precise definition of logrolling, see id. at 706, 729 ("Logrolling occurs when two proposals each supported by a minority are combined into one ballot proposition supported by a majority, and the two minorities support the combination of policies but respectively prefer to enact one policy and not enact the other.").
6 Hasen & Matsusaka, supra note 3, at 37-38.
7 See Michael D. Gilbert, Single Subject Rules and the Legislative Process, 67 U. Pitt. L. Rev. 803, 850 (2006) ("If, on average, logrolling produces a social benefit, . . . then courts should presume that every instance of logrolling is allowable, since they cannot tell which logrolls produce a benefit and which do not. If, on average, logrolling causes harm, . . . then logrolling should be banned . . . .").
8 See Ronald H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1, 2-6, 15-17 (1960); see also Robert D. Cooter, The Strategic Constitution 53 (2000) (summarizing Coase Theorem).
9 We therefore agree that "groups with different interests may routinely package . . . rather disparate measures together in order to pool resources to qualify their measures for the ballot." Hasen & Matsusaka, supra note 3, at 38. However, the ability and willingness of some sponsors to package does not imply that such packages increase social welfare.
10 See Cooter & Gilbert, supra note 1, at 701–03 ("‘[M]ajority will' has no meaning when aggregate preferences run in circles.").
11 See id. at 702–03, 703 & n.70 (noting that limiting ballot proposals to single dimension of policy choice animates median voter theorem, thus making "majority will" a concrete concept).
12 See, e.g., Slayton v. Shumway, 800 P.2d 590, 593 (Ariz. 1990) ("Our previous cases have adopted a single subject rule that prevents the ‘pernicious practice of ‘log-rolling' . . . ." (quoting Tilson v. Mofford, 737 P.2d 1367, 1370 (Ariz. 1987))); League of Women Voters v. Eu, 9 Cal. Rptr. 2d 416, 428 (Ct. App. 1992) ("[T]here is no basis for the claims of confusion and logrolling. . . . The single-subject rule is designed specifically to guard against these concerns." (citing Kennedy Wholesale, Inc. v. State Bd. of Equalization, 806 P.2d 1360, 1366 (Cal. 1991))); In the Matter of the Title, Ballot Title, Submission Clause, and Summary Adopted Apr. 5, 1995 v. Hamilton, 898 P.2d 1076, 1079 (Colo. 1995) ("[T]he single subject requirement now embodied in Article V, Section 1(5.5), [of the Colorado Constitution] would prevent proponents from engaging in ‘log rolling' . . . ."); In re Funding of Embryonic Stem Cell Research, 959 So. 2d 195, 197 (Fla. 2007) (advisory opinion) ("The single-subject requirement . . . prevents ‘logrolling' . . . ."); In re Initiative Petition No. 382, 142 P.3d 400, 405 (Okla. 2006) ("The purposes of the single subject rule are: 1) to ensure that the legislators or voters of Oklahoma are adequately notified of the potential effect of the legislation; and 2) to prevent ‘logrolling' . . . ."); Amalgamated Transit Union Local 587 v. State, 11 P.3d 762, 781 (Wash. 2000) ("The purpose of this prohibition [the single subject rule] is to prevent logrolling . . . .").
13 See Cooter & Gilbert, supra note 1, at 712–22 (describing democratic process theory).
14 Hasen & Matsusaka, supra note 3, at 40.
15 John G. Matsusaka & Richard L. Hasen, Aggressive Enforcement of the Single Subject Rule, 9 Election L.J. (forthcoming October 2010); Michael D. Gilbert, How Much Does Law Matter? Theory and Evidence from Single Subject Adjudication (Sept. 29, 2008), available at http://works.bepress.com/michael_d_gilbert/9/ (unpublished manuscript) (on file with the Columbia Law Review).
16 Hasen & Matsusaka, supra note 3, at 41.
Preferred Citation: Robert D. Cooter & Michael D. Gilbert, Reply to Hasen and Matsusaka, 110 Colum. L. Rev. Sidebar 59 (2010), http://www.columbialawreview.org/assets/sidebar/volume/110/59_Cooter.pdf.
