Jump to ContentJump to Main Navigation
For the Many or the FewThe Initiative, Public Policy, and American Democracy$

John G. Matsusaka

Print publication date: 2004

Print ISBN-13: 9780226510811

Published to Chicago Scholarship Online: March 2013

DOI: 10.7208/chicago/9780226510873.001.0001

Show Summary Details
Page of

PRINTED FROM CHICAGO SCHOLARSHIP ONLINE (www.chicago.universitypressscholarship.com). (c) Copyright University of Chicago Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in CHSO for personal use. Subscriber: null; date: 23 October 2019

Majority Tyranny and the Constitution

Majority Tyranny and the Constitution

(p.113) Chapter 8 Majority Tyranny and the Constitution
For the Many or the Few
University of Chicago Press

Abstract and Keywords

The initiative process is a controversial part of American government, and has been for more than a century, yet the terms of debate have hardly changed. Opponents argue (as they did one hundred years ago) that the initiative allows narrow interests to subvert the political process, while defenders maintain (as they did one hundred years ago) that it allows the majority to counteract special interest subversion of the legislature. This chapter begins by considering the problem of majority tyranny and then examines how the initiative meshes with the Constitutional principles of American democracy.

Keywords:   majority tyranny, American government, American democracy, Constitutional principles, initiative, narrow interests

It may well happen that the public voice pronounced by the representatives of the people, will be more consonant to the public good, than if pronounced by the people themselves convened for the purpose. On the other hand, … men of factious tempers, or local prejudices, or of sinister designs, may by intrigue, by corruption or by other means, first obtain the suffrages, and then betray the interests of the people.

—James Madison, Federalist no. 10

The initiative process is a controversial part of American government, and has been for more than a century, yet the terms of debate have hardly changed. Opponents argue (as they did one hundred years ago) that the initiative allows narrow interests to subvert the political process, while defenders maintain (as they did one hundred years ago) that it allows the majority to counteract special interest subversion of the legislature. The main purpose of this book is to sort through the huge amount of data available after one hundred years of initiative lawmaking, and make an objective, factual assessment of whether the initiative brings about policies favored by the many or the few. I focus on fiscal policy, perhaps the most important function of government, and certainly the one that attracts the most attention, and utilize a century of data for all fifty states and almost five thousand cities. These are the main findings:

  1. (p.114) 1. Over the last three decades, the initiative significantly changed state and local policy (chapter 3):

    1. a. It cut expenditure and taxes.

    2. b. It decentralized spending from state to local governments.

    3. c. It shifted financing away from taxes and into user fees and charges for services.

  2. 2. Opinion surveys show that a majority of people favored the changes brought about by the initiative (chapter 4).

  3. 3. In the early part of the twentieth century, the initiative increased state and local spending, and there is some reason to believe that a majority favored these spending increases (chapter 5).

  4. 4. The initiative had the biggest impact in periods where legislatures were slow to respond to changes in public opinion (chapters 6 and 7).

The natural interpretation of these facts is that the initiative pushed policy in the direction favored by a majority of people. Not a single piece of evidence links the initiative to nonmajority policies as the special interest subversion hypothesis would predict. There are, of course, a number of caveats that must be kept in mind with this and any other empirical work, and I have tried to note them in the preceding chapters. Yet the fact that a comprehensive examination of fiscal policies from the beginning to the end of the twentieth century reveals not a shred of evidence for the special interest subversion view makes as compelling a rejection of a hypothesis as we ever get in empirical research. This is the main message of the book: based on the facts, the initiative serves the many and not the few.

Beyond this conclusion, the interpretation of the evidence is fairly open. In particular, I do not think the evidence in itself tells us whether the initiative is a good or bad thing for American government. An overall assessment of the initiative requires understanding not only whether it serves the many or the few, but also how it affects the rights of minorities, how it impacts the Constitutional structure of American democracy, how it affects the responsiveness of legislators to constituent interests, how it affects the innovativeness of public policy, and a host of other issues.

I want to conclude the book by discussing some of the more important issues that have been skipped over so far. This and the next chapter identify what I see as the key unresolved questions, bring out points of contention, and review the evidence such as it is. My goal is to give the reader a sense of where things stand on these issues and to suggest to scholars some interesting research directions. Since little evidence is available on these issues, the reader is forewarned that what follows is leavened with a (p.115) healthy measure of my personal reflections and intuitions. In this sense, the last two chapters represent a departure from the rest of the book, which tries not to stray far beyond the bounds of what can be empirically documented.

This chapter begins by considering the problem of majority tyranny and then examines how the initiative meshes with the Constitutional principles of American democracy. The final chapter contains some reflections on the initiative process as an approach to delegation problems and as a way to increase political competition.

The Danger of Majority Tyranny

As we have seen, the initiative process promotes majority rule, a core principle of American democracy. Yet an equally important principle is that all people have certain fundamental rights that cannot be violated even if a majority wishes to do so. The institutions of American government are designed to prevent the many from “tyrannizing” the few. To repeat an oftcited passage from Federalist no. 51:

It is of great importance in a republic, not only to guard the society against the oppression of its rulers; but to guard one part of the society against the injustice of the other parts. Different interests necessarily exist in different classes of citizens. If a majority be united by a common interest, the rights of the minority will be insecure.

An important question is then: does the initiative, by empowering the majority, undermine the rights of numerical minorities?

Much has been written on this question, mostly in the last thirty years, and many opinions have been expressed both for and against the initiative (see Cronin 1989, and Bowler, Donovan, and Tolbert 1998 for a summaries). Perhaps the most important thing to note is that these opinions rest on intuition, not on convincing, systematic evidence. No one has gone to the data and actually measured how the initiative affects minority rights.1

Let me briefly describe what has been done. By far the most common approach is anecdotal. Typically, an author singles out a handful of laws involving the civil rights of minorities, studies how they came about in some detail, and then generalizes from his or her specific examples to the initiative process as a whole. There is nothing wrong with this approach to a point; anecdotes certainly provide useful information and help flesh out (p.116) the picture. The problem is that we do not know if a particular anecdote is representative of a general pattern or is simply a special case, an exception to the rule. There is little justification for inferring general tendencies from a handful of anecdotes.

The other complication is that the performance of the initiative has to be compared to the alternative, which is lawmaking by elected legislatures. It is not enough to show that initiatives have occasionally undermined minority rights, since no form of government known to man provides ironclad protection of minorities. Legislatures also have violated the civil rights of minorities. Indeed, for every blatant instance of initiative failure, we can readily identify a similar example where minority rights were undercut by a legislature. Take Jim Crow laws, for example: in Oklahoma, an initiative was approved in 1910 that disenfranchised black citizens; in the other southern states, similar laws were passed by state legislatures. Another example: in California, an initiative was approved in 1920 that restricted the property rights of Japanese; the internment of Japanese-American citizens during World War II came about through the actions of elected officials.2 A listing of anecdotes like this is useful to remind of us the dangers of democracy, but it only tells us that no form of government is perfect. It does not tell us if the overall record of initiatives is worse than that of legislatures.

Several recent studies attempt to go beyond anecdotes by collecting broad samples and subjecting them to statistical analyses. These studies compare the passage rate of civil rights measures to the passage rate of all other measures. They assume that if antiminority initiatives pass more often than other initiatives, the initiative process is a threat to minority rights. The evidence points in no particular direction. Gamble (1997) found that antiminority measures were more likely to pass than other measures in a sample dominated by local initiatives, while Donovan and Bowler (1998) and Frey and Goette (1998) found lower passage rates for antiminority measures in samples of state ballot propositions in the United States and national and local ballot propositions in Switzerland (respectively).

Yet none of these studies compared initiative lawmaking to legislative lawmaking, and we know that legislatures sometimes pass antiminority laws, too. Even if their findings had agreed, these studies would not tell us whether the initiative increases or reduces the risk to minority rights. The impact of the initiative on minority rights cannot be discerned by comparing how often voters approve antiminority initiatives to how often they approve other initiatives.3 If, for example, voters approve 60 percent of the antiminority initiatives that come before them, while city councils approve 80 percent of the antiminority measures that come before them, then one (p.117) could argue that city councils are the more serious threat to minority rights. The ratio of the passage rate of antiminority to other initiatives has no bearing on the conclusion.4 The unavoidable fact is that the only way to compare how the initiative process treats minorities with how legislatures treat them is to look at data from initiatives and legislatures. None of the existing studies, however, examined the performance of legislatures on civil rights issues; they only looked at initiatives. I believe these studies are to be commended for moving the discussion beyond the level of anecdotes, and their evidence does give a sense of how often initiatives threaten minority rights. But they shed no light on the question of interest: how much of a threat does the initiative process pose to minority rights compared to the legislative process?

Another unresolved issue in the literature is how to distinguish “majority rule” from “majority tyranny.” It will not do to classify every issue where the majority prevails as an instance of tyranny. If we did, then majority-rule democracy would be tyranny by definition. What we need is to identify a set of basic human rights that are essential to a free society, such as life, personal liberty, and political equality. Laws that violate these basic rights are out of bounds even if favored by a majority, and their approval would constitute “majority tyranny.” Most people would put slavery, Jim Crow, and the internment of Japanese Americans into this category. Laws that work against the interest of minorities but do not infringe on their basic human rights should be considered the normal stuff of politics, examples of majority rule. The problem is that the line between a “right” and an “interest” is gray and highly subjective. Consider, for example, two California initiatives from the 1980s. Proposition 63 made English the state's official language, without spelling out exactly what that meant. Proposition 96 mandated AIDS testing for persons charged with sexual offenses or assaults if bodily fluids were transferred. According to Gamble (1997), both measures were instances of “majority tyranny” (the first violated the civil rights of language minorities and the second the civil rights of people with AIDS) rather than legitimate exercises of majority rule. Not every reader would agree with her classification, I suspect. The point here is not to criticize Gamble's classification—she had to draw a line somewhere—but to highlight a fundamental difficulty in evaluating the initiative.5 We want to assess how the initiative affects minority rights, but we do not know exactly what these rights are.

In short, we simply do not have compelling evidence yet on whether initiatives or legislatures pose a greater threat to minority rights, or even if there is a difference. The most defensible position is one of agnosticism: pending more evidence, we just don't know. Fortunately, the question (p.118) seems amenable to empirical analysis. It should not be too difficult to collect data on civil rights legislation and compare the outcomes under both forms of government. Given the increasing interest in the issue of majority tyranny, I feel reasonably optimistic that in the next few years we will start to see evidence that addresses the limitations of previous work.

Now let me move from what is known to what I suspect. If I had to guess what the evidence ultimately will show, it would be that the initiative process is not a greater threat to minority rights than legislatures. I say this primarily because opinion data consistently show that racial and ethnic minorities themselves do not consider ballot propositions dangerous. According to several Field polls, large majorities of blacks, Latinos, and Asians believe that “statewide ballot propositions are a good thing” (see table 8.1).6 The numbers vary somewhat over time, but even in 1997 after a series of racially charged initiatives, strong majorities of each minority group still supported the process. And the number that considered the initiative process to be a “bad thing” is minuscule, typically less than 5 percent. Evidently, most members of these minority groups do not view the initiative process as a threat to their well being. It is possible that people are ignorant of the danger they face; that is, they mistakenly approve of the initiative process (and that is one reason we need data on the actual effects). But given a century of experience with the initiative process, and the constant attention paid to civil rights issues by elites of these groups, I would find it surprising if a significant threat existed that the survey respondents had simply failed to appreciate.

Now in a way, this creates something of a puzzle. The evidence in this book shows that the initiative promotes the interests of the many not the few. I studied fiscal policies not civil rights laws, but I cannot see an obvious reason to expect the pro-majority feature of the initiative to be confined to taxes and spending. Thus, I would expect that when the interests of the majority collide with those of a minority on social policy, the majority

Table 8.1 Support for Ballot Propositions by Racial/Ethnic Groups

“Do you think that statewide ballot proposition elections are a good thing for California, a bad thing, or don't you think they make much difference?”

Percentage responding “good thing” and “bad thing”









































(p.119) will win more often when the initiative is available. Why then would minorities express such strong support for the initiative process?

One possible explanation is that blacks, Latinos, and Asian Americans do not view themselves as minorities on most of the important policy questions that have to be decided. Rather they perceive the “minority” (the group that loses to the majority) to be someone else. We have some evidence on this. Hajnal, Gerber, and Louch (2002) studied fifty-one California ballot propositions and found that black, Latino, and Asian American voters were only about 1 percent less likely than white voters to be on the winning side of the vote. Essentially the same pattern emerged even when they focused on measures where minorities voted cohesively or on issues that minorities said they cared most about. Perhaps minorities expect to be disadvantaged by the initiative when it comes to issues that divide along racial lines, but they anticipate making up for these losses and then some by being on the winning side of most other issues. English as the official state language might be an acceptable price to pay for having lower taxes, for example. Such a calculus does not seem entirely unreasonable. Racial and ethnic minorities enjoy significant Constitutional protection stemming from the Equal Protection Clause, and the courts have been fairly aggressive in enforcing these rights for the last thirty years. I find it hard to imagine an initiative that seriously undermined minority rights surviving judicial scrutiny given the current state of jurisprudence.

Finally, this line of reasoning suggests to me that the literature might not be looking for the right kind of majority tyranny. Majority tyranny may be more of a threat to numerical minorities who do not enjoy clear Constitutional protection and the favor of the courts, such as business groups, people who live in certain geographic areas, people who enjoy activities the majority disapproves of, such as smoking, drinking, and hunting, and so on. Although it has become a mantra in the literature to cite the Federalist papers on the dangers of majority tyranny, it is seldom noted that the concern there was primarily about the rights of economic interests(Federalist no. 10):

But the most common and durable source of factions, has been the various and unequal distribution of property. Those who hold, and those who are without property, have ever formed distinct interests in society. Those who are creditors, and those who are debtors, fall under a like discrimination. A landed interest, a manufacturing interest, a mercantile interest, a monied interest, with many lesser interests, grow up of necessity in civilized nations, and divide them into different classes, (p.120) actuated by different sentiments and views. The regulation of these various and interfering interests forms the principal task of modern Legislation.

We might do well to take this argument more seriously and investigate how the initiative process impacts the interests of unpopular, numerically small economic groups. For example, are cigarette taxes higher—that is, are tobacco companies asked to provide a larger share of government revenue—with initiatives or legislatures? This aspect of the majority tyranny issue, so far ignored in the literature, constitutes yet another important unresolved question concerning the initiative and minority rights.

The Initiative, the Constitution, and the Foundations of American Democracy

The U.S. Constitution established a representative democracy, not a direct democracy. This was not by accident. The Founders were deeply skeptical of direct democracy. As James Madison put it in another oft-cited passage of the Federalist no. 10:

A pure democracy, by which I mean, a society, consisting of a small number of citizens, who assemble and administer the government in person, can admit of no cure for the mischiefs of faction … Democracies have ever been spectacles of turbulence and contention; have ever been found incompatible with personal security, or the rights of property; and have in general been as short in their lives, as they have been violent in their deaths. Theoretic politicians, who have patronized this species of government, have erroneously supposed, that by reducing mankind to a perfect equality in their political rights, they would, at the same time, be perfectly equalized and assimilated in their possessions, their opinions, and their passions.

A republic, by which I mean a government in which the scheme of representation takes place, opens a different prospect, and promises the cure for which we are seeking.

The initiative process is not exactly what Madison had in mind by a “pure democracy,” but it is surely a step away from the “pure” representative government he envisioned. The institutions of American government have served the country well for more than two hundred years and should not be tampered with lightly. An important question is whether the (p.121) initiative departs so far from the Constitutional scheme that it undermines the foundations of American democracy.

Much has been written on this question but the answer is elusive. As a matter of law, the answer is clear: the initiative process is constitutional. Article IV, section 4 of the U.S. Constitution “guarantee[s] to every state in this union a republican form of government.” In the early twentieth century, the initiative process was challenged as a violation of this so-called Guarantee Clause. The Supreme Court declined to consider the merits of the case, holding that it was for the Congress not the courts to decide whether a state's government was “republican” or not.7 When Congress admits a state with the initiative process to the Union (as happened with Oklahoma and Alaska) or accepts the representatives of an initiative state, or when the President recognizes a government under powers delegated to him by Congress, that state is thereby presumed to have a republican form of government. The initiative has thus been accepted as constitutional by the proper authorities for more than one hundred years now.

Although the constitutionality of the initiative process is settled for practical purposes, the underlying issue is not. Serious critics still contend that the initiative is incompatible with the principles of American government. In the dire words of David Broder (2000, 1), the process is “alien to the spirit of the Constitution” and a threat to “subvert the American system of government.” What follows is a brief overview of the main arguments, and my view of where things stand.

What Would the Founders Say?

A natural starting point for understanding how the initiative process might affect the Constitutional machinery is the writings of the Founders. Unfortunately, nothing like the initiative process was on the table when the Constitution was framed, so the best we can do is speculate what the Founders would have thought had they considered it.

One approach has been to focus on the meaning of the phrase “republican government,” since it is clear that the Founders sought to establish a “republic.” “Republican government” clearly entails some form of popular participation, but exactly what forms are included is vague, and it is doubtful the Founders would have agreed on the particulars. In Federalist no. 10, as just noted, Madison defined a republic narrowly as government by representatives, contrasting it with “pure democracy,” by which he had in mind something like a town meeting. It would be quite a stretch to fit the initiative process into his notion of a republic. Thomas Jefferson's (p.122) definition, on the other hand, was broader, and could easily accommodate the initiative process:

Action by the citizens in person, in affairs within their reach and competence, and in all others by representatives, chosen immediately, and removable by themselves, constitutes the essence of a republic…. All governments are more or less republican in proportion as this principle enters more or less into their composition.8

John Adams simply noted that he “never understood” what a republic meant, and “no other man ever did or ever will.”9

A second approach to determining the Founders' intent is to examine the initiative process in light of the underlying theory of government in the Constitution. The emphasis here again has been on the celebrated Federalist no. 10, in which Madison laid out the argument for representation and against pure democracy. It is worth quoting at length:

The two great points of difference between a democracy and a republic are, first, the delegation of the government, in the latter, to a small number of citizens elected by the rest: secondly, the greater number of citizens, and greater sphere of country, over which the latter may be extended.

The effect of the first difference is, on the one hand to refine and enlarge the public views, by passing them through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice, will be least likely to sacrifice it to temporary or partial considerations. Under such a regulation, it may well happen that the public voice pronounced by the representatives of the people, will be more consonant to the public good, than if pronounced by the people themselves convened for the purpose. On the other hand, the effect may be inverted. Men of factious tempers, or local prejudices, or of sinister designs, may by intrigue, by corruption or by other means, first obtain the suffrages, and then betray the interests of the people….

The other point of difference is, the greater number of citizens and extent of territory which may be brought within the compass of republican, than of democratic government; and it is this circumstance principally which renders factious combinations less to be dreaded in the former, than in the latter. The smaller the society, the fewer probably will be the distinct parties and interests composing it; the fewer the distinct (p.123) parties and interests, the more frequently will a majority be found of the same party; and the smaller the number of individuals composing a majority, and the smaller the compass within which they are placed, the more easily will they concert and execute their plans of oppression. Extend the sphere, and you take in a greater variety of parties and interests; you make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens; or if such a common motive exists, it will be more difficult for all who feel it to discover their own strength, and to act in unison with each other.

Modern commentators have tended to focus on the first of Madison's arguments, the potential for representatives to “refine and enlarge” the public views. However, that was not the decisive point for Madison. Indeed, he observes that representatives can “betray the interests of the people.”10 Thomas Jefferson expressed an even greater skepticism toward delegated power: “The mass of citizens is the safest depository of their own rights; … the evils flowing from the duperies of the people are less injurious than those from the egoism of their agents.”11

In fact, the key advantage of a republic over a democracy from Madison's perspective is that a representative democracy can function over a wider expanse of territory and a larger population than a pure democracy. The strength of representative democracy comes from the counterbalancing interests that are encompassed by a government that extends over a large population and territory, not from the superior virtue of representatives or the give and take of their deliberations.

It helps to understand the context of Madison's argument. Madison's immediate purpose was to persuade the citizens of New York to adopt the federal Constitution. As such, his purpose was not so much to argue against direct democracy—there were no serious proposals to move the United States in that direction. Rather, he was trying to defuse the antifederalist argument that the new Constitution concentrated too much unchecked power in the hands of the central government. It was generally accepted in the late eighteenth century that popular rule had been a failure throughout history (the classical Greek city states were favorite examples). Given this, the antifederalists argued that the power of the central government had to be limited (as in the Articles of Confederation, which gave veto rights to individual states) or the people's rights would be endangered. Madison accepted the premise of the antifederalist argument—that democracies historically were unstable—but then argued that the problem could be cured by extending the government over a sufficiently large (p.124) population and territory. Previous attempts at popular government failed, in his view, because they had not been applied on a large enough scale. The Constitution would cure the problem of pure democracy—its limited scope—by working through representatives instead of directly through the people. The important point here is that the main problem with pure democracy is its limited scope, not its lack of representation. Madison was not alone in this view: “A democracy [is] the only pure republic, but impracticable beyond the limits of a town,” wrote Jefferson.12

Federalist no. 10 never reaches the question: Would pure democracy be acceptable if it could be established over a large population and territory? As such, it is not clear how far the argument against “pure” democracy extends to the initiative process. The initiative process can be applied over a large population and area (the population of California alone is now more than ten times the population of the entire United States in 1787), so it should enjoy the same benefits from encompassing multiple competing interests that Madison used to justify representative democracy. The case against the initiative process, then, would seem to rest primarily on the alleged ability of representatives to refine the public views, but as we have seen, even Madison viewed this argument with skepticism.

Another important ingredient of the Constitution is the system of checks and balances. The Constitution deliberately fragmented power between the legislative, executive, and judicial branches, following principles laid down by Montesquieu (1989, book 10, chap. 6) and widely accepted by the Founders: “All would be lost if the same man or the same body of principal men, either of nobles, or of the people, exercised these three powers: that of making the laws, that of executing public resolutions, and that of judging the crimes or the disputes of individuals.” In broad terms, separation of powers is preserved when the initiative process is available. The voters have the power to legislate, but they do not administer, execute, or judge the laws. And as Gerber et al. (2001) have shown, the power of the other branches to check direct legislation is real: many initiatives approved by the voters are subsequently blocked by judicial decisions or the failure of the executive to enforce them.

However, the Constitution incorporated two additional counterbalancing features that the initiative process overrides: the division of legislative power into two houses, and the executive veto over legislation. Both of these devices again followed Montesquieu, although he envisioned an upper house of hereditary nobles and a monarchical executive. While a bicameral legislature and executive veto were considered desirable by most of the Founders, it is not clear whether they viewed these checks as (p.125) fundamental to the new Constitution. Three American states in the late eighteenth century did not have bicameral legislatures (Georgia, Pennsylvania, Vermont), and all but three states (Massachusetts, New York, and South Carolina) excluded the executive veto from their first constitutions.13 These states were not considered to be in violation of republican principles as far as we know.14

Some time ago, Charles Beard, who was generally sympathetic to the initiative process, concluded from his survey of the Founders' views that “no one has any warrant for assuming that the founders of our federal system would have shown the slightest countenance to a system of initiative and referendum applied either to state or local affairs. If some state had possessed such a system at that time, it is questionable whether they would have been willing to have compromised with it, as they did with the slave states, in order to secure its adherence to the Union” (Beard and Shulz [1912] 1970, 29). I think the case is not so cut and dried, as I have tried to explain here. The Founders were deeply distrustful of legislatures, much like modern-day initiative proponents, and at least some of them had faith in the public at large to vote its interest. Moreover, the Founders were not uniformly hostile to direct democracy in the forms that existed at the time, such as New England town meetings, nor were they opposed to periodic popular lawmaking via referendums. Massachusetts held a statewide referendum in 1778 on a post-independence constitution, and New Hampshire held a series of constitutional referendums (working through town meetings) in 1778, 1781, and 1783. It seems clear the Founders would have objected to a national town-meeting form of government as proposed by some modern direct democracy advocates. How they would have felt about the initiative process—a grafting of periodic citizen-initiated referendums onto a predominantly representative government—is not so obvious.15

It Comes Down to Evidence Again?

The writings of the Founders give valuable insight into how the pieces of the Constitution fit together, but they do not incorporate the experiences of the last two hundred years. Yet it is the historical record that holds the most promise for understanding how the machinery of American government works. Indeed, the Founders probably would have viewed the parsing of their words in search of answers (as I just did above, and many others have done elsewhere) somewhat skeptically. “Forty years of experience in government is worth a century of book-reading,” wrote Thomas (p.126) Jefferson in 1816, and this the Founders “would say themselves were they to rise from the dead.”16 The men who set up the American federal system were thoughtful empiricists. They sought practical solutions for governance problems from historical experience, not in the writings of authorities. Our approach should be the same: the answer to whether the initiative undermines American government must be found in the data.

Empirical research has only begun to seriously engage this question. The difficulty is not in determining whether the Constitution has served the country well. It has. Rather, the problem is identifying what parts of the Constitutional scheme are important and which are not. It is clear that not everything established in 1787 is essential. The Constitution has been modified without damaging its essential features many times over the last two centuries. Most would agree that some changes have improved on the original plan and brought American democracy closer to its true principles, such as the ending of slavery, the granting of voting rights to blacks, women, and those without property, and the direct election of U.S. senators. Yet we can also conceive of modifications that would have a pernicious effect. There is no shortage of opinions about why American government works, but these are at best informed hunches. The truth is that we simply do not have much of an empirical basis at present for saying that one part of the Constitutional scheme is important and another is not. Consequently, there is not much factual ground to stand on when assessing whether the initiative is harmful to the core principles of American government.17

Although definitive answers await serious empirical research, an impressionistic look at the historical record does not give much reason to be concerned about the initiative process. The initiative has been part of American government for more than a century now. While it is not obvious exactly how to measure “quality of government,” casual observation suggests that initiative states have done just as well as noninitiative states using the more obvious metrics: initiative states seem to be just as free, prosperous, and nice places to live as noninitiative states. There is no heavy migration of citizens from initiative states to noninitiative states as would be expected if the initiative was a serious threat to life, liberty, and the pursuit of happiness. Indeed, many states that are typically viewed as attractive places to live—for example, California, Florida, Massachusetts, and Washington—happen to be initiative states. I am not claiming that the initiative caused any of this (although I would not foreclose the possibility, either), only pointing out that the initiative has not caused things to go (p.127) horribly awry. Whatever its demerits may be, the initiative does not cause the Constitutional sky to fall.

The few pieces of modern empirical research that are designed to assess how the initiative affects the quality of governance point in the same direction. One purpose of government is to promote (or at least not hinder) economic prosperity. Evidence from two recent studies suggests that if anything the initiative process is good for the economy. One study documented that output per worker was higher in initiative states than in non-initiative states, and imputed that noninitiative states “wasted” 20 percent more of their government spending than initiative states (Blomberg, Hess, and Weerapana 2001). The other study found that output per capita was higher in Swiss cantons (roughly the same as American states) with the initiative than in cantons without the initiative, after controlling for the physical and human capital stock (Feld and Savioz 1997). One interpretation of this evidence is that the initiative process causes governments to choose policies that are more favorable for economic growth. For example, the initiative might be used to direct public funds into valuable infrastructure instead of pork barrel projects. This interpretation is reinforced by another study of Switzerland: an analysis of municipal trash collection found greater efficiency in cities with the initiative than cities without the initiative (Pommerehne 1983). A more general measure of government performance might be the “happiness” of the citizens. Happiness is notoriously difficult to measure, but a recent study of Switzerland found that citizens who lived in cantons where the initiative is available reported higher levels of subjective well-being (“happiness”) than those living in cantons where the initiative is unavailable, after controlling for a large number of demographic and economic variables (Frey and Stutzer 2000). I should say that I have some methodological misgivings about much of this evidence, so I take it as only suggestive. Robustness needs to be established and other metrics of government performance need to be investigated. It is notable, however, that to this point there does not appear to be a single empirical study that finds the initiative process reduces the quality of government, however measured.


(1.) This may be about to change. As I write this chapter, a working paper by Gerber and Hug (2002) is circulating that attempts a systematic comparison of civil rights laws in initiative and noninitiative states.

(2.) Cronin 1989 contains a good discussion of these cases.

(3.) Curiously, Donovan and Bowler (1998) and Frey and Goette (1998) both seem aware of the problem, the former describing the methodology as “a potential logical fallacy” (1021), and Gamble (1997, 262) admits that “direct democracy must be compared with the representative system.”

(4.) Actually, even this inference is problematic since we have no information on how often antiminority measures come before voters and city councils. For example, suppose that both voters and city councils approve 60 percent of the antiminority measures they face, but that such laws are proposed much more often in noninitiative cities. Then the noninitiative cities would end up with more “majority tyranny” even though approval rates are the same.

(5.) I don't mean to entirely absolve previous research, however. I think there has been a tendency to define majority tyranny far too broadly. For example, if California's Proposition 38 for English-only ballots was really majority tyranny, why did 52 percent of Latinos support it (Hajnal, Gerber, and Louch 2002)? A partial solution would be for researchers to report their results using several different definitions. Every study should also explicitly state what definition is being used, as Goette and Frey (1998) did.

(6.) This table was inspired by a similar table in Hajnal, Gerber, and Louch 2002.

(7.) The key decision is Pacific States Telephone and Telegraph Company v. Oregon, 223 U.S. 118 (1912), which essentially applied the principle established in Luther v. Borden, 42 U.S. 1 (1849), that the Guarantee Clause is nonjusticiable.

(8.) Letter from Thomas Jefferson to Pierre Samuel Dupont de Nemours, 1816. Thomas Jefferson Digital Archive, University of Virginia, etext.lib.virginia.edu/jefferson. Other letters cited below are also available through the archive.

(9.) As quoted in Wood 1998, 48.

(10.) In his letter to Jefferson on October 24, 1787, explaining the Constitutional Convention, Madison did not even bother to list the “refine and enlarge” argument for representative government, focusing entirely on the scope argument.

(11.) Letter from Thomas Jefferson to John Taylor, 1816.

(12.) Letter from Thomas Jefferson to Isaac H. Tiffany, 1816.

(13.) Moreover, the New Jersey Plan introduced at the Constitutional Convention by delegates from Connecticut, Delaware, Maryland, New Jersey, and New York provided for a unicameral legislature. Executive veto information comes from Fairlie 1917.

(14.) A recent trend in political economy has been to view legal institutions in terms of creating “veto players,” individuals or bodies who can slow down or (p.185) prevent an action from taking place. For example, Hammond and Miller (1987) develop a model of the U.S. Constitution that interprets the checks and balances as creating veto players in order to induce stability. I believe it is incomplete and somewhat misleading to think of checks and balances in this way, especially when it comes to the Constitution. Stability—too many veto players in the Articles of Confederation—was the problem the Founders were trying to solve with the Constitution. Instead, I would argue that fragmentation of power creates a competitive environment where many ideas can flourish and bad ideas rejected. This involves the idea of agenda control, but also information provision and innovation (see chapter 9.)

(15.) This section draws on Beard and Shulz [1912] 1970, Oberholzer 1912, Ketcham 1957, Diamond 1959, Cronin 1989, and Wood 1998, in addition to the more specific references throughout. The Federalist papers and Madison's Notes of Debates in the federal convention are essential in understanding how the Founders perceived the Constitution. I also found the Anti-Federalist helpful.

(16.) Letter from Thomas Jefferson to Samuel Kercheval, 1816.

(17.) Empirical researchers in economics and political science are increasingly interested in understanding how institutions affect the performance of government, so we can hope that a clearer picture will soon emerge. Recent work on the checks and balances of bicameralism and the executive veto is particularly relevant for assessing the initiative process. See, for example, Bradbury and Crain 2001a on bicameralism and McCarty 2000 on executive veto.