Students for Clean Elections
W   e   s    l   e    y   a   n        U   n   i   v   e   r   s   i   t   y

This project would not have been possible without the cooperation and support of countless activists and public officials around the country, who took time out of their busy schedules to answer my questions. In particular I would like to mention Ed Davis of Common Cause and Janice Fine of Northeast Citizen Action who helped steer me to many of their associates and contacts.

I would like to extend special thanks to my thesis advisor, Professor Richard Boyd, who provided insightful comments and support throughout this inquiry.

I would also like to note two professors, David Morgan and Peter Rutland, who played a significant role in shaping my intellectual development during my time at Wesleyan. In addition, my appreciation goes out to Professors Richard Adelstein, Barbara Craig, Marc Eisner, Giulio Gallorotti, and Peter Kilby, whose inspiring courses expanded my knowledge of politics, law, economics, and history, and gave me tools important to the success of this project.

I would like to thank my fiancee, Allison Rodman, for providing emotional support throughout my years at Wesleyan and for insisting that sometimes I take time to have fun, even when deadlines loom large. I am also grateful to my parents, who taught me to value the thrill of discovery.

Finally, I extend my heartfelt appreciation to Randy Kehler, who first steered my interests toward Campaign Finance Reform many years ago. Little did he suspect then that he had awoken a lifelong passion.

For all the others who have provided answers, questions, advice, and encouragement during my four years at Wesleyan, I thank you all.

The election process lies at the very heart of republican democracy. As such, election reform has been a consistent theme of the last one hundred years of American political dialogue. Early reforms, such as the introduction of the Australian ballot, were followed by experiments with direct primaries, non-partisan elections, and regulation of campaign finances through the Corrupt Practices Act of 1925 and the Hatch Act of 1940.

Arguably the most fundamental change in the administration of American elections has been the institution of government financing of many election campaigns. Public campaign financing is distinct from other campaign reforms because it substitutes an alternative funding mechanism for political campaigns rather than seeking to regulate the two existing mechanisms; fundraising and candidate personal financing. Despite the attention devoted to campaign finance reform efforts on the federal level, it is in American states, over half of which employ some form of public financing, where sufficient diversity exists to make conclusions concerning the origins of public financing systems.

Limited studies have been conducted assessing the effect of specific state public financing measures. However, there are no comprehensive nation-wide surveys of the origins of the state systems. This study seeks to profile how American states came to focus on campaign reforms, and explain why individual states chose to implement specific forms of public financing when they did.

Democratic Representative William Bourke Cockran of New York introduced the first proposal for public financing of elections in the United States in 1904, in response to controversy surrounding President Theodore Roosevelt’s extensive fundraising in the heated election campaign of that year. Fueled by revelations of large corporate donations to both political parties, pressure continued to build through 1907, at which time President Roosevelt endorsed public financing for party organizations. However, after much negotiation, the Tillman Act, banning contributions by corporations to political candidates, passed as a broad political compromise and diffused the pressure behind public financing proposals. Although significant campaign finance reform measures continued to be considered throughout the succeeding decades, the next significant public financing proposals were not advanced until the 1960’s.

Overseas, the first modern public funding of political entities was instituted with the advent of the television age. Interestingly, Costa Rica in 1954 and Argentina in 1955 were the first modern countries to formally provide public funds for political parties. European public financing of political parties began in West Germany in 1959, and spread to Austria in 1963. As the German political scientist Karl-Heinz Nassmacher explains, the "political reasoning [for the early European public financing systems] went along an almost classical line: ‘Parties had found that costs had been increasing, particularly in the field of communications, and existing sources of income were proving to be insufficient.’" In specific instances, such as that of Italy in 1974, public financing adoption was facilitated by corruption scandals resulting from this increased fundraising pressure. The pressures for public subsidies of political parties continued to build during the 1960’s. By 1970 France, Israel, and all four Scandinavian countries introduced public financing of political parties.

As in Europe, it is likely that the re-consideration of public financing by American policy makers was a reaction to the rising cost of televised campaigns and the resulting increased pressure on the political fundraising system. The percentage of American households with television sets soared from 9% to 87% between 1950 and 1960, and for the first time in the 1960 election, more Americans received their political news from television than newspapers.

The earliest post war American proposals for public financing did in fact draw heavily from the overseas party-funding model. In 1962, President Kennedy’s Commission on Campaign Costs, charged with examining political campaigns in the new television age, proposed public matching funds in presidential campaigns. In 1966, in response to frenzied fundraising by the Johnson Administration, Congress passed into law but swiftly repealed a measure providing block grants to political parties for campaign use.

However, American policy makers soon realized that the structure of the American political system required distinctive policy approaches. With the exception of the minimal systems of Latin America, the international public financing provisions implemented before 1974 existed in a parliamentary context where party-based electoral structures and unpredictable election cycles meant that political financing was inevitably synonymous with party financing. By contrast, the institutionalized election cycle and candidate-centered nature of the American federal system meant that problems of political money were more likely to occur in the campaign financing of specific candidates directly before the election. Such a system initially presented serious hurdles, as parties tend to be more standardized and predictable than individual candidates.

However, campaign costs were rapidly rising on both sides of the Atlantic. The cost of presidential elections more than doubled between 1968 and 1972, from $37 million to over $90 million dollars. Inspired by the fundraising scandals in Richard Nixon’s 1972 campaign uncovered by the Watergate investigations, in 1974 Congress significantly amended the 1971 Federal Election Campaign Act (FECA) to include an innovative system of matching public funds for presidential primaries and full public financing of presidential general elections. In 1976, public money was used to fund a U.S. presidential election for the first time. No significant federal campaign public financing measures have been seriously considered since the last revision of FECA in 1979.

Largely lost in the attention paid to national campaign reforms, a simultaneous and far more comprehensive public financing revolution was beginning on the state and local level. In 1957 Puerto Rico implemented the first public financing within the jurisdiction of the United States in the wake of the Costa Rican and Argentinean reforms. The Puerto Rican Election Fund Act, which provided significant block grants to the island’s two major political parties, was designed to reduce the dependence of the territory’s political parties on the powerful sugar industry and end the informal practice of requiring civil servants to contribute 2% of their income to the ruling political party. Internationally, twenty German and Austrian laender and the Canadian provinces of Quebec and Nova Scotia had instituted public financing of local elections by the 1980’s. However, there are no indications that public financing in American states was directly modeled on any of the local overseas measures.

In 1973 four states, Iowa, Maine, Rhode Island, and Utah, enacted the first public financing in the United States. The movement continued as the Watergate scandal heated up, with five more states passing public financing measures in 1974, three in both 1975 and 1976, two in 1977, and one in 1978. In 1979, Seattle, Washington implemented the first municipal public campaign finance system. State public financing of elections hit something of a plateau in the 1980’s, with only six states and two municipalities implementing systems over the course of the decade. In the early 1990’s a number of states which had implemented the most comprehensive measures during the 1970’s reformed and expanded their systems. The adoption of full public financing measures in Maine in 1996 and Vermont in 1997, heralded in a subsequent reform movement dubbed Clean Elections, which eventually spread to Arizona, Massachusetts, and the municipalities of Austin, Boulder, Cincinnati, Oakland, and San Francisco.

Between 1972 and 2001 fourteen local governments and twenty-seven state governments instituted public financing for at least some of the political offices or parties under their jurisdiction:

John W. Kingdon’s 1984 book Agendas, Alternatives, and Public Policies has become a seminal work discussing the nature of problem awareness, issue definition, and agenda setting. Kingdon uses case studies in two national issue areas, health and transportation policy, to explore how specific topics came to dominate policy agendas. Kingdon argues that three process streams largely determined of the federal agenda: problem identification, policy generation, and political shifts. According to Kingdon’s rationale, these three streams exist independently of each other but are periodically coupled around successful policy initiatives.

According to Kingdon’s theory, the problem and political streams elevate broad issue areas to the national consciousness and are largely dominated by visible clusters of governmental actors such as elected officials and political appointees. In contrast, specific policy alternatives are usually generated by hidden clusters of policy entrepreneurs, including career bureaucrats, interest groups, and other elements of civil society. Significantly, these policies are often developed based on policy entrepreneurs’ personal or institutional interests, desire to promote values, or wish to affect the system and arise independently from the problem or politics streams.

To summarize, Kingdon’s hypothesis argues that the problem and political streams periodically create windows of policy opportunity by focusing the attention of government actors on specific issue areas. Into these windows rush policy entrepreneurs trying to link their pre-existing policies into the specific issues on the agenda.

A recent example might help to illustrate Kingdon’s theory. Following the September 11^th terrorist attacks and the subsequent war on terror, American dependence on Middle Eastern energy suppliers became an important concern for American policy-makers. Consequently, numerous bureaucrats, interest groups, and academics stepped into this breach pushing policies ranging from heightened hydrogen fuel cell development, to increased ethanol production, to greater cooperation with the Russian government. Most of these policy entrepreneurs had developed these proposals far before the focusing events of September 2001. However, in the aftermath of 9/11, many of these constituencies attempted to tie their policy proposals into the policy window created by the problem of the attacks.

Windows of opportunity are brief due to the short attention span of politicians, the media, and the public. Therefore, as Frank Baumgartner and Bryan Jones assert in Agendas and Instability in American Politics, the evolution of American government policy is typified by a state of punctuated equilibrium, where the traditional policy monopolies of interlocking alliances of elites are periodically altered by the spotlight of specific problems.

In this study, I turn Kingdon’s work around, by using his theories as a starting place for exploring how a particular issue area, reform of the campaign finance system, reached the public agenda of specific states. There are two significant ways in which state campaign finance reform legislation provides an interesting test of Kingdon’s theory. More than any other issue, campaign finance directly affects the status of elected officials within the government, and thus combines the problem and politics streams to a greater degree than other issue areas. The second challenge of this study lies in my focus on the state rather than the national political arena. States provide an interesting test of Kingdon’s conclusions, both because of their varied political culture and because of their distinctive mechanisms of governance.

The relevance of Kingdon’s ideas may change when they are used to explain how a single distinctive issue reached the agendas of the political classes in a geographically and culturally distinct group of states rather than a single national political bureaucracy. Furthermore, state government mechanisms, including peculiarities of interest group interaction and the frequent use of ballot measures, may change the relative balance of power between specific groups and constituencies in the agenda setting and alternative specification roles.

Kingdon restricts his analysis to the agenda setting and alternative specification aspects of policy determination, while leaving the factors influencing the choice of a specific alternative and the implementation of this choice unaddressed. However, when examining a precise policy outcome, it is necessary to assess the decision-making element of policy formation. Over two-thirds of states have enacted at least one major new campaign finance law in the last thirty years. However, only a small percentage of these laws contain public financing. Thus, in addition to explaining how state political actors come to perceive a problem with their democratic process, one must also clarify why, and in what cases, a consensus is generated around the specific policy of public financing of elections.

David Nice presents the only significant previous inquiry into the reasons for the adoption of public financing as a chapter in Policy Innovation in State Government. Nice argues that three main factors influence the adoption of specific policy innovations: problems, resources, and (what I will call) civic culture. Nice’s first two factors are mainly negative. As Kingdon acknowledges, the nature of the specific challenge inspiring action necessarily eliminates those policy alternatives which are unlikely to solve the given problem. The resources available to government actors, often in the form of budgets or constitutional powers, further limit the range of existing choices. Nice places the most emphasis on his third factor, civic culture. Nice argues that civic culture could play a significant role in determining state policy outcomes, and that culture varies between states. This cultural factor could play both a positive and negative role in the decision-making process between policy alternatives.

It is helpful to separate Nice’s third factor into two general categories: political culture and party ideology. To generalize, political culture revolves around the values of citizens concerning change. Party ideology could be defined as the attitude of the political classes toward government.

Nice divides political culture into three broad archetypes loosely based on Daniel Elazar’s seminal 1966 study American Federalism; the traditionalist, the individualist, and moralist archetypes. Nice measures party ideology along the simple scale of Liberal vs. Conservative. He describes these archetypes in terms of views concerning government action as defined by party platforms and major party policy initiatives. Conservatives are generally reluctant to use government to institute change, whereas Liberals are generally eager to use government to institute change. To quantify these values, Nice uses a 1978 study by Eugene McGregor which uses decades of presidential convention role call votes to place the Democratic and Republican Parties in every state on a –1 to 1 scale of liberalism and conservatism.

Nice applies his theories of state innovation to eleven issue areas, including public financing. Nice analyzes the political cultures and party ideologies of the sixteen states which instituted public financing by August 1981. He does not specifically address emergent problems or resources available to these states, but theorizes that these may have been important factors.

Nice asserts that public financing is more likely in states that have highly moralistic political cultures. Furthermore, he concludes public financing is more likely in states that have relatively liberal Republican ideologies and that have experienced Democratic Party dominance during the period of implementation.

Nice’s study is incomplete for four major reasons:

1. Nice’s dependence on categorizations of the cultural and ideological identities of states in simple, quantifiable terms can come under significant criticism.
2. Nice’s sample of states with public financing is now over twenty years old.
3. Nice fails to distinguish between distinct methods and systems of public financing, whose effects and origins may vary greatly.
4. Nice’s study does not survey the views of the actual entrepreneurs and decision-makers in the implementation of public financing. The distinctive nature of campaign finance reform, due to its direct effect on politicians, casts into doubt the applicability of traditional trends of political culture or party ideology when explaining campaign reforms.

Chapter 2 explores many of the difficulties inherent to quantifying political culture. However, before proceeding further, it may be beneficial to address the means by which this study attempts to rectify the other three problems listed above:

The number and variety of public financing measures has greatly increased since Nice’s study. Since 1981, the number of states employing public financing measures has nearly doubled to twenty-seven, while the number of local measures has more than tripled to fourteen. A small number of sources contain profiles of state public financing systems, though up to now no single source presents a truly comprehensive and up to date database of all public financing measures in the United States.

The U.S. Federal Election Commission’s National Clearinghouse on Election Administration puts out a periodic summary of state campaign finance laws. The Council of State Government’s in Lexington, KY also produces two publications, The COGEL Blue Book and The Book of the States, which often present data on state campaign finance laws. In addition, there have been a small number of academic studies which contain public financing information. Herbert Alexander, under the auspices of the Citizens’ Research Foundation, produced a study of statewide public financing measures in 1986, which was updated in 1992. CRF also summarized six public financing systems in local elections in 1999. Michael Malbin and Thomas Gais compile a profile of state campaign finance laws in The Day After Reform. In addition, Joel Thompson’s work Campaign Finance in State Legislative Elections contains a limited number of references to public financing systems.

To update these studies, I compiled a comprehensive database of all state and local public financing laws with a brief profile of the distinctive characteristics and origins of each of these measures. I also contacted election officials, academics, and democracy interest groups in every state and examined the actual laws in question. A list of the states which have adopted public financing is contained in Table 1.2 above. The full database profiling every public financing measure is included in Addendum 1.

Malbin and Gais’ analysis of state disclosure regulations and spending limits shows a definite evolution of these reforms. The earliest state reforms during the 1970’s focused mainly on disclosure requirements and restrictions of atypical behavior such as excessively large campaign contributions or potentially corrupting practices. However, later reforms "were often designed to alter the more typical flows of private money into politics" through lower contribution limits and extensive reporting requirements. Many experts account for this shift as a reaction to the increased cost and complexity of campaigns during this period. For example, in the decade between 1980 and 1990, average spending for gubernatorial candidates increased 44 percent after adjusting for inflation. Simultaneously, the average number of interest groups per state nearly doubled.

An examination of the progression of state and local public financing measures reveals some similar trends. Sixteen significant state public financing measures were implemented by 1980. These systems were made up exclusively of partial matching and block grant mechanisms and were usually very limited in scope. Implementation of reforms continued at a trickle through the 1980’s, perhaps because of resource restraints due to budgetary and anti-tax pressures. In the late 1980’s and early 1990’s, a number of states made attempts to reform and strengthen their existing public financing systems. In the latter half of the 1990’s a small wave of public financing initiatives created systems in four states and at least four municipalities. Significantly, these later initiatives created the first non-federal full public financing systems. In addition, five of these later state and municipal measures were passed by direct democracy ballot initiatives. For whatever reason, the evolution of public financing legislation seems to indicate increasing distrust of private money in the campaign finance system over the last thirty years.

Because of the wide variety of public campaign finance measures, it is problematic to define state financing of elections. The most important key, or common element, is the dispersal of "public money." Thus, a public financing system must disburse money (rather than in-kind contributions of materials or services such as airtime). Furthermore, this money must pass through the public sphere. Thus, measures allowing tax deductions or tax credits for contributors (such as in Alaska and a number of other states) are not considered public financing for the purposes of this study, since state governments never actually possess these campaign funds.

Another characteristic common across all public financing systems is their voluntary nature. In light of Supreme Court decisions equating spending money with free speech, all public financing measures have necessarily been designed to compliment or run parallel to traditional private financing patterns, without inherently replacing them.

Public financing systems can further entrench or greatly weaken existing parties, politicians, and policies. The effect of public financing depends the design of specific systems. In her 1981 article "State Public Campaign Finance: Implications for Partisan Politics," Ruth Jones outlines a number of public financing policy choices available to lawmakers. I have expanded on her list to illustrate the varieties of campaign financing systems:

Public financing funds can be allocated through direct budget appropriations, specific charges, or taxpayer check-off or add-on mechanisms. States which provide simple block grants to political parties often use the tax code to allow citizens to direct a prescribed amount to a specific political party, rather than to the system as a whole. Such systems could aptly be described as institutionalized contribution mechanisms, and are generally designed to achieve very different purposes than impartially disbursed systems. Presently, seven states allow public funds to be dispersed to partisan entities without regulating the spread or ratio of this allocation.

Public funds can also be distributed impartially to political parties or directly to candidates themselves. Some of these systems are consciously crafted to either strengthen parties or to strengthen candidates in the electoral process. Furthermore, the regulations determining the eligibility of third parties and third party candidates for public funds differ significantly across states. Third party participation also significantly influences the effect of public financing measures and may often be associated with different conceptions of the political problems to be addressed. Currently, eight states appropriate money to political parties, while fifteen states appropriate funds directly to candidates, and four states provide money to both. Barriers to third party entry tend to be significant in most systems, but vary greatly.

Public financing can range from small block grants to full financing of elections and can be dispersed using a variety of mechanisms. Nearly all dispersal ranges and mechanisms can be categorized into three types:

1. Full: Full public financing consists of block grants to candidates with strings attached. By taking public money candidates agree to forego raising or spending any additional money. Effective full public financing systems generally appropriate sufficient public money to run a campaign for the given office and usually contain additional matching fund mechanisms for candidates whose non-participating opponents outspend them.

A truly full public financing system becomes the only significant funding source for participating candidate’s campaigns, and thus is designed to exclude private money from the campaign system. Currently, four states have full public financing systems for one or more offices.

2. Partial: Partial public financing generally consists of grants disbursed on a set ratio to participating candidates. The majority of these partial systems provide matching grants, though some systems use flat grants. In a one to one matching system, one dollar of public financing is given to a candidate for every dollar of private contributions this candidate raises. Some systems only match contributions up to a certain amount, thus increasing the importance of smaller donations. Often this effect is multiplied by additional provisions lowering contribution limits. Sometimes partial financing includes stipulations requiring certain actions during the campaign, such as participation in debates. In many cases partial public financing includes overall spending caps for participating candidates.

Most partial public financing measures fund a ratio rather than a set amount of a candidate’s campaign. Partial public financing systems are designed to augment and reduce, but not eliminate private campaign financing. Currently, there are ten states with partial funding systems.

3. Simple Grants: Simple public financing grants disperse a set amount of public money, generally without any strings attached. Unlike full public financing systems (where grants must be large enough to fund a whole campaign), simple grant systems can disperse relatively insignificant funding amounts. Unlike full and partial public financing, public grants are almost exclusively appropriated for a party rather than to a specific candidate or election cycle.

Simple public grants commonly do not provide the only significant funding for a candidate or party and generally do not constrain the behavior of participating candidates or parties in any way. Their main effect is on the distribution ratios of money in the campaign system. As such, simple grants are designed to increase the amount of money in the system rather than reduce it. At present, there are eleven states with simple grant systems.

As mentioned above, many public financing systems include responsibilities imposed upon the receipt of public funds. The nature of these requirements (or the lack thereof) greatly influences political behavior and thus often defines the impact of the system. Therefore, these conditions can serve as evidence when determining the intentions of the framers of public financing systems.

All public financing systems that disburse money directly to candidates allocate funds for the general election. However, a sub-set of public financing systems allocate funds for primary campaigns as well. Funding of primaries greatly increases the cost and complexity of public financing measures, but also, assumedly, expands their effect. Currently eleven state systems fund both primary and general election campaigns while three systems fund only general election campaigns.

Public financing measures vary greatly in range of offices covered. They can include the governor, the lt. governor, other constitutional officers, the state legislature, parties, and local officers. Public financing systems usually distribute different amounts to candidates for each office (though partial public financing systems often maintain a set ratio across all races). Currently, five states cover just the governors race, three states cover the governor and some other statewide offices, and seven states cover the governor, statewide offices, and legislative elections.

In designing public financing systems, drafters must decide whether to create a new agency to administer the system or to confer this authority to existing agencies. In addition, sponsors must determine the enforcement powers to be given to this agency and the degree of discretion the agency will be allowed in administering the program. Because issues of administration and enforcement are primarily relevant for the implementation phase of policy innovation, they are largely beyond the scope of this study.

An adequate explanation for the adoption of state public financing measures has never been advanced. David Nice presents brief but interesting theories concerning the nature of states which implemented public financing. However, his analysis takes for granted the specific reasons that public financing measures may have been implemented, and is now very out of date. Malbin and Gais advance hypotheses concerning the inspirations of public financing, but their arguments are based largely on speculation. An examination of the history and variety of state public financing measures indicates the likelihood that their supporters may have been addressing entirely different problems or intending starkly different outcomes.

Any attempt to understand the intentions of the framers of public financing laws must include evidence concerning the process of diffusion of ideas across states. It is also important to understand the relationship between federal and state agendas concerning campaign finance issues. Significant numbers of state laws were passed at times of heightened awareness of federal campaign finance problems, such as the 1972 funding scandals. However, there has not been a comprehensive study concerning the nature of this relationship.

To determine the intentions of the initial decision-makers who chose public financing, I have gone directly to the source. I drafted a short set of survey questions which I administered mainly via email to individuals (including politicians, academics, and activists) who were active, influential, or closely observant during the passage of public financing measures in their state.

The pool for these surveys was expanded through a snowball sampling scheme whereby everyone I found gave me further names in addition to answering my questions. The final tally was sixty-two surveys completed by individuals in twenty-one states. To gain further information, telephone interviews were conducted with twenty-nine relevant individuals in twelve states. There were significant drawbacks to this method, since it was inevitably piecemeal. Only a tiny percentage of individuals contacted returned completed surveys. Interviews and surveys could not be administered with equal frequency to decision-makers in every state with public financing and a certain handful of states, especially those whose systems were more recently implemented, generated the majority of responses. However, the wide variety of state measures meant that surveys and interviews were conducted with representatives of most of the major geographic regions and public financing archetypes. The survey and interview results were further supplemented by additional information available through existing government resources and academic works. Data and interviews from specific municipalities with public financing measures were used in certain cases to supplement the state data.

Chapter 2 surveys a variety of predictive mechanisms to determine patterns and similarities in state adoption of public financing. A clear correspondence is found between adoption of partial and full public financing and Democratic Party dominance, as well as high relative levels of liberal political ideology, economic and social development, civic culture, and white ethnic diversity. In addition, using classifications of state political culture developed by Daniel Elazar, a clear correspondence is confirmed between more moralistic political culture and adoption of partial and full public financing. The chapter then uses theoretical and empirical data from a variety of authors to construct a general model of state characteristics which may serve to explain adoption of more comprehensive varieties of public financing. Throughout the chapter, a clear difference is shown between the characteristics of states adopting simple grant party public financing systems and states adopting more comprehensive partial and full public financing systems.

Chapter 3 presents the major findings of the survey and interview study concerning Kingdon’s model of agenda generation. It explains why election reform rose to the agenda of specific states at specific times and why public financing was considered a viable alternative. In particular, it highlights the role that scandals and other focusing events in raising election reform to the agenda, and the role that Common Cause and other public interest groups often play in advancing public financing of elections as an attractive policy alternative. In addition, Chapter 3 outlines patterns in the intentions of the policy entrepreneurs and eventual decision-makers during the window surrounding the implementation of public financing legislation. Specifically, it highlights the occasional confluence between politicians trying to improve their standing with the public, good government advocates attempting to reduce the cost of campaigns, and progressive activists wanting to increase the influence of less wealthy segments of the population.

Chapter 3 also briefly discusses the agenda generation relationship between states and the federal government and outlines the major players in state agenda determination. Finally, it confirms Kingdon’s hypotheses and the findings of Chapter 2 by presenting case studies of state processes which eventually resulted in public financing measures. The case studies include examples of full, partial, and simple grant public financing measures from a wide regional, cultural, and demographic variety of states. Particular emphasis is placed on states which typify patterns of public financing or break from established norms and expectations. Specific observations and examples from other states and municipalities are integrated into the chapter to round out the analysis.

Chapter 4 recaps the basic determinations of the study and proposes some lessons that should be drawn from these conclusions. It also places the findings in the larger context of election reform and makes predictions about the future course of public financing policy determination.

Addendum 1 is a database listing and profiling all state public financing measures in the United States. Readers are urged to refer to this database whenever they have questions about the nature of public financing systems in specific states.

Addendum 2 is the survey questionnaire distributed to individuals connected to the adoption of public financing in their states.

Chapter 2 isolates a number of characteristics that typify states adopting public financing measures. In one of the major conclusions of the study, this chapter shows that the characteristics of states adopting only simple grant systems are starkly different from those of states adopting partial and full public financing systems. Possible explanations for this dichotomy are advanced in Chapter 3. Chapter 2 begins by examining party control of state legislatures and concludes that adoption of partial and full public financing measures through traditional legislative mechanisms occurs only in states largely under Democratic Party control. The chapter then goes on to examine five major sets of measurable characteristics, partisanship and ideology, regionalism, socioeconomic levels and social modernization, civic culture and democratization, and racial and ethnic diversity. Finally, this chapter considers issues of state political culture, as first articulated by Daniel Elazar over thirty-five years ago.

The studies and indexes used to examine the more quantifiable factors come from a variety of different sources and methods. To facilitate easy communication of the major findings of the study, I have standardized most of the indexes to a mean of zero and a standard deviation of one. If the resulting indices are normally distributed, 68% of the variation occurs between plus and minus one. Virtually all variation occurs within three standard deviations of the mean. Thus, differences of even half of a standard deviation are substantively important.

Nearly all public financing measures have been adopted through the traditional legislative route requiring passage through both houses of the legislature and the governor’s signature. Thus, it is important to determine whether the party in control of the state government significantly affects the chances of the implementation of public financing measures. Table 2.1 indicates the party in control of both houses of the legislature and the governor’s office at the time of the implementation or significant reform of every public financing initiative. The table is organized by date within the type of initiative adopted.

Party control has a significant effect on the chances of public financing being adopted by states. Sixteen states have implemented and/or reformed simple grant public financing provisions. Six of these measures, making up 38% of the total, were implemented in state governments totally controlled by the Democratic Party. Implementation of simple grant public financing measures by split governments turned out to be relatively symmetrical. Three measures, or 19% of the total, were implemented by split governments with the Democrats controlling both houses of the legislature and by split governments with Republicans controlling both houses of the legislature. Two simple grant measures, or 13% of the total, were adopted by states where Republicans and Democrats each controlled one house of the legislature. However, only two states, Iowa and Idaho, implemented public financing measures when the Republican Party controlled the entire state government. Overall, nine simple grant measures, or 56% of the total were implemented by states with fully Democratic controlled governments or Democratic controlled legislatures. In contrast five simple grant measures, or 31% of the total were implemented by states with fully Republican controlled governments or Republican controlled legislatures.

Party control is an even more striking determinant with the more comprehensive partial public financing systems. Fifteen states have implemented or significantly expanded partial public financing systems. Of these, the Democratic Party controlled the entire government in nine states, or 60% of the total. The Democratic Party controlled both houses of the legislature while a Republican was governor in four states, comprising 27% of the total. The legislature was split between Democratic and Republican control in one state, making up 7% of the total. Finally, one state, Nebraska, has a unicameral non-partisan assembly, but a Democratic governor was in office when the partial public financing system was adopted. Significantly, no states where the Republican Party controlled the entire government or both houses of the legislature has ever implemented or expanded partial public financing measures. Thus, states where the Democratic Party controlled the entire government or both houses of the legislature made up 87% of the cases where partial public financing measures have been adopted or reformed.

Party control is less relevant in states which have implemented full public financing measures, since three out of four of these measures were adopted by direct democracy ballot measures. In fact, policy entrepreneurs in these three states cited the unwillingness of the legislature to pass significant campaign finance reform as the main reason they chose to pursue reforms through the ballot measure process. In the session preceding the public financing ballot measures, Republicans and Democrats split control of the Maine state legislature, Democrats dominated the Massachusetts legislature but faced a Republican governor, and Republicans ran the entire government of Arizona. Democrats controlled the government of Vermont, the only state to adopt a full public financing measure through the legislative process. Thus, Republican opposition may have played a significant role in determining the direct democracy strategy of public financing entrepreneurs in Maine, Arizona, and Massachusetts.

In summary, public financing measures were significantly more likely to be adopted when Democrats enjoyed significant or sole power within the government. More comprehensive public financing measures, including partial and full public financing have never been implemented through the legislative mechanism when Republicans controlled the entire state government or even both houses of the legislature. Thus, to use Kingdon’s terminology, Democratic control of the legislature seems to be a necessary condition for the creation of a policy window where public financing legislation is seriously considered. Reasons for Democratic support of public financing legislation may include an ideological belief in limiting the influence of wealthy individuals, a political interest in minimizing the pervasive Republican advantage in fundraising, and a willingness to use government to solve perceived problems. It is likely that Republican opposition derives from similar, though opposite, motivations.

Party control of state legislatures can only serve as a partial predictive mechanism. Ideology within the same political party varies greatly across states. The composition and issue positions of the Democratic Party in many southern states during the 1970’s may have diverged greatly from Democratic Parties in the Northeast or Far West. To correct for these differences one must attempt to determine the underlying ideologies of the parties and electorates in the American states.

David Nice, the only political scientist to directly address the relationship of party ideology to public financing, used data generated by Eugene McGregor in 1978 to examine the nature of the parties in states which implemented public financing. Mcgregor’s findings are relatively dubious and outdated since they are based mainly on Democratic convention roll calls from the 1950’s through the early 1970’s.

In their 1993 study Statehouse Democracy, Robert Erikson, Gerald Wright, and John McIver combined over a decade of polling and survey data to establish the partisanship and ideology of state electorates and elites. In a set of earlier studies, Erikson, Wright, and McIver show a significant correlation between public opinion, political elite opinion, and the ideological policy tendencies in American states. Thus, the Statehouse Democracy data may be one of the most accurate indicators available of state party ideology in the 1970’s and 1980’s.

Erikson, Wright, and McIver use data from CBS News-New York Times polls conducted between 1976 and 1988 to determine the self-identified partisan and ideology profiles of citizens in the lower forty-eight states. In these surveys, respondents identified their party affiliation as Democratic, Independent, or Republican and described their ideology as Liberal, Moderate, or Conservative. Assuming that Erikson’s data proving a strong correlation between public opinion and political positions and initiatives of state political actors is correct, it is possible to form a rough chart establishing the political center of gravity in given states.

There does not seem to be a significant correlation between a state’s self-identified partisanship and adoption of public financing. As Figure 2.1 illustrates below, there this only about 1/10 of a standard deviation between the percentages of citizens in states with partial or full public financing identifying themselves as Republican and Democratic and the percentages in states without any public financing. Simple Grant states trend more significantly Republican, which distinguishes them from both of the other categories. However, the variation is again relatively minor, so that overall Figure 2.1 seems to illustrate that Democratic control of the legislature, rather than the national partisan tilt of the state’s electorate plays a more direct role in determining public financing adoption.

In contrast to partisan affiliation, there is a significant correspondence between a state’s ideology and the implementation of partial and full public financing. By standardizing the overall self identification of citizens within every state on a Liberal-Conservative scale, Figure 2.2 illustrates that there is a very significant difference of nearly 2/3 of a standard deviation between states with no public financing and states with partial or full public financing. The general correspondence between partisan ideology and a positive view of government action may account for the tendency of states with partial or full public financing systems to be significantly more liberal than states with no public financing. Strikingly, the populations of states with simple grant public financing systems are more conservative than either of these other two categories. The structure of simple grant systems themselves may very well account for the tendency of states adopting these measures to be more Republican and conservative. This factor is elaborated on further later in this chapter and in Chapter 3.

It is also possible to examine state party ideology by measuring the ideology of respondents who described themselves as Democrats and Republicans. States that have adopted full or partial public financing have more liberal Democratic and Republican parties. However, Figures 2.3 and 2.4 show that contrary to Nice’s hypotheses, relative Democratic liberal ideology shows a greater correspondence to public financing adoption than relative Republican liberal ideology. The difference between Democratic ideology in partial and full public financing states approaches half of a standard deviation. In contrast, the Republican ideologies of these same states are separated by only 1/5 of a standard deviation, a relatively insignificant difference. The correspondence between liberal Democratic parties and public financing adoption is likely explained by the earlier determination that comprehensive public financing is only adopted when Democrats control state legislatures.

Figures 2.3 and 2.4 also shows that the political party ideologies in Simple Grant states are far more polarized than in states with comprehensive or no public financing. Figure 2.5 illustrates this finding more clearly by calculating and standardizing the difference between respective party ideologies. Positive correspondence indicates higher than average levels of polarization. Thus, in states with simple grant public financing, Republican parties tend to be more conservative, and Democratic parties tend to be more liberal than the national average. This political polarization may partially account for the emphasis on strong political parties implied by the institution of simple grant party public financing.

States within a given region often have similar values, policy norms, and political structures. Thus, a number of political scientists have argued that "region may serve as a surrogate for political culture in most cases." The advantage of looking for patterns through simple regional location is that the data, through its very simplicity, is more reliable and less subjective than more abstract conceptions of political characteristics or culture. Figure 2.6 shows all the states in the US which have adopted public financing systems, with darker shades indicating states which have implemented partial or full public financing.

Table 2.2 sorts this data by the United States Census classification of the country in nine regional divisions:

The distribution of generic public financing adoption according to Census Bureau divisions does not show many striking patterns. Between 50%-70% of states in six of the nine divisions have implemented some sort of public financing system. The region with the highest proportion is the East North Central, where Illinois is the only state to not have adopted some form of public financing system. In contrast, no state in the West South Central region of Arkansas, Louisiana, Oklahoma, and Texas has implemented public financing.

The distribution patterns for states which have implemented more comprehensive partial or full systems of public financing are significant. The New England states are the clear leader in this category, followed by the East North Central region. In contrast, the West South Central and Pacific regions have no statewide partial or full public financing measures, followed closely by the Mountain states where only Arizona has a recently implemented a full public financing measure. However, significant intra-regional patterns are evident as well. The four regions with the highest percentage of full or partial public financing are the four contiguous North East and North Central regions, which spread from New England across the Great Lakes to Minnesota, Nebraska, and the Dakotas. Further exploration will be undertaken later to determine why this cross-regional pattern of public financing exists.

A major question that arises from the regional patterns of public financing implementation is whether significant diffusion of information or ideas has occurred within regions during the adoption process. The survey data presented in Chapter 3 shows considerable testimonial evidence that state entrepreneurs were aware and affected by of the actions of neighboring states in the years proceeding the New England full public financing initiatives of the 1990’s. However, the data is less conclusive concerning earlier measures instituted in other regions of the country. Thus, before delving into the specific cultures and characteristics of the twenty-seven states that have implemented public financing of elections, it is useful to examine any similarities within regions in type and timing of public financing measures. Figure 2.7 shows four regional groupings of three or more states, each of which instituted or reformed similar public financing systems within two election cycles of each other.

The earliest of these groupings is a set of New England states, which includes Maine, Massachusetts, and Rhode Island. All three states implemented roughly analogous simple grant funding provisions for political parties between 1973 and 1975. In a separate but similar grouping, Maine and Massachusetts joined with Vermont between 1996 and 1998 in establishing structurally similar full public financing systems for political candidates through the ballot measure mechanism. In 1992, Rhode Island established a partial public financing system for statewide candidates, but interviews and surveys from the New England states (as outlined in Chapter 3) indicate that this reform was not a major impetus for the subsequent full public financing initiatives to the north.

The second major grouping consists of three Great Lakes states, Michigan, Minnesota, and Wisconsin, which established partial public financing systems for statewide candidates between 1974 and 1977. Indiana and Iowa also established public payments to political parties during this time period, but the significantly different nature of these public financing systems (both from Michigan, Minnesota, and Wisconsin and from each other) indicate that their origins may not have come from the same concerns or innovations as the systems of their neighbors.

The final potential area of regional diffusion of public financing initiatives is in the mountain states of Idaho, Montana, and Utah. All three of these states established simple grants to political parties funded by similar tax check-off mechanisms between 1973 and 1975. Although very little supplemental evidence from surveys or interviews exists for these systems, their extreme similarity would indicate that conscious diffusion did occur.

Finally, implementation of full public financing measures seems to be closely related to the state ballot initiative process. Figure 2.8 shows that the majority of states allowing statutory ballot measures lie west of the Mississippi River. This regional concentration may hold the historical key to ballot initiative adoption. The first non-constitutional citizen ballot initiative took place in 1904 and the practice became popular only in the 1910’s and 1920’s, propelled forward by rising education levels and the popularity of the national constitutional amendment process. During this period, the statutory tradition of Eastern states had been developi