Rational and pluralistic approaches to HAVA implementation: the cases of Georgia and California.
Part of the reason for this decentralized model of election administration is constitutional because Articles I and II of the U.S. Constitution largely delegate election procedures for federal offices to the states. (1) The federal government has periodically sought to provide some uniformity in election administration, for example, by pushing for more uniform protection of voting rights (e.g., the Voting Rights Act of 1965) or for more consistent voter registration procedures across the states (i.e., the National Voting Rights Act of 1993). However, before the 2000 presidential election, election administration was largely a county or subcounty responsibility.
In the wake of the 2000 election, though, this decentralized American election administration came under criticism. Some of this concern arose after the careful examination of the Florida election process in 2000 found substantial differences across counties in how administrative procedures were implemented; this is one way to view the U.S. Supreme Court's Bush v. Gore decision that stopped the Florida recount in December 2000. (2) Criticism deepened in other states, for example in California, where concern developed about the use of punch-card voting machines in counties with large minority-voter populations, when more reliable and accurate voting technologies were in use in counties with fewer minority voters.
By 2002, the federal government acted, in the form of the Help America Vote Act (HAVA). This legislation provided a stronger federal role in election administration, mainly by establishing a new federal entity, the Election Assistance Commission (EAC), and by mandating that states develop statewide voter registration databases and eliminate inferior voting technologies. Although HAVA did not vastly expand the federal role in election administration, it did open the door for a stronger state role in election administration relative to the historical strength of local government entities in conducting elections. It is this potential transformation in the relative role of the state and local election officials that we examine here.
Specifically, we examine the recently published state HAVA plans to determine how state officials are using HAVA to reshape elections. We are especially interested in evaluating the aspects of state HAVA plans where states are planning to exert control and where they are leaving matters in the control of local officials. Examining three specific areas-voter registration, voting systems, and voter education--we are able to cluster states and evaluate the level of control they are asserting over the election process. After discussing the effect of HAVA generally on the states, we examine two approaches to HAVA implementation--a rational approach and a pluralistic approach--and how these approaches are being used in two states to show how HAVA has been used to change election administration. The stories of how the secretaries of state in Georgia and California have attempted to use the 2000 election and the subsequent passage of HAVA as means of changing the balance of power between state and local officials are illustrative of how election processes and procedures are likely to change in the post-HAVA world. (3)
THE ROAD TO HAVA
The brunt of the criticisms about the conduct of the 2000 election in Florida fell on local election officials. Local officials in various jurisdictions used the butterfly ballot, deferred maintenance on punch-card machines, and implemented other procedures that made voting difficult. (4) Additionally, some election officials made extreme comments to the media after the election expressing a lack of concern about discrimination against certain classes of voters, comments that resulted in the U.S. Department of Justice sanctioning three Florida counties. (5)
The postelection policy environment spawned an array of election reform efforts. Almost all of the major national associations representing state and local election officials (e.g., the National Association of Secretaries of State and the National Association of County Recorders, Auditors, and Treasurers) and state and local governments (e.g., the National League of Cities and the National Conference of State Legislatures) created election reform committees. There were also three major independent election reform efforts: the National Commission on Federal Election Reform (NCFER), the Caltech/MIT Voting Technology Project (VTP), and the Constitution Project's Forum on Election Reform. These independent groups initiated studies of election reform that combined public hearings and input from experts with independent research on various issues associated with elections. Their reports noted that there would be benefits in shifting certain responsibilities from localities to the states, especially in the area of voter registration, and in creating uniform statewide standards in other areas. (6)
This movement to consolidate and reorganize certain aspects of the election process is a typical response in the United States to both real and perceived management and implementation problems in both the private and the public sectors. (7) Hannah Sistare notes that there are four rationales for government reorganization: (1) making government work better, (2) saving money, (3) enhancing power, and (4) addressing a pressing problem. (8) Reorganizing the management of elections was clearly intended to address the first and fourth rationales. However, the election reform proposals also had the potential to address the third rationale for government reorganization--enhancing or shifting power. Governmental reorganizations are often undertaken with the intent of enhancing executive or legislative power, or enhancing the control that one branch or the other has over a policy area, program, or agency. Reorganization can also be undertaken to enhance the standing of certain groups within government, such as when cabinet departments are created and these groups gain a seat at the cabinet table with the president. Reorganization can also be conducted to achieve the more focused goal of shifting decision-making power within a given policy area from one entity to another. (9)
In the case of election reform, the reorganization of elections was initiated by the federal government with congressional passage of HAVA. The final provisions in HAVA reflect the intersection of two political and policy dynamics. First, there was a fundamental difference in views between the House--controlled by Republicans--and the Senate--controlled by Democrats--over the level of ongoing authority the federal government should have in election reform. This conflict reflected obvious differences in ideology between the two houses of Congress and differences in interpreting historical and legal norms. The House was more sensitive to the historical norm that elections are primarily state activities carried out by local governments; the Senate relied on legal views that the federal government can establish rules and standards for all federal elections because they involve members of Congress. (10)
Interest groups played an active role throughout the HAVA legislative process. These groups included civil rights and disability rights organizations, state and local intergovernmental organizations, and corporations (e.g., voting machine vendors). Outside actors, especially academics and prominent members of the various election reform commissions, also were involved in HAVA's development. The NCFER vigorously presented its recommendations to the congressional committees and met with various legislators during the drafting process to provide technical assistance. The Caltech/MIT VTP also provided technical assistance and input. This effort of providing input often occurred in private meetings that included other interest groups, especially interests representing local election officials. (11)
CHANGING STATE AND LOCAL CONTROL VIA HAVA
In the end, the HAVA legislation created a modest role for the federal government. HAVA empowered the EAC to allocate funding to the states to purchase new voting equipment, as well as to finance voter registration and other election-related activities. This design is quite typical of federal intergovernmental programs created to influence state or local activities and induce state and local officials to change specific behaviors. (12) The EAC also is required to study various aspects of election administration and collect data from the states about their election activities.
The EAC did not receive the power to create binding regulations on states regarding standards to be met in any given aspect of election administration. Section 209 of HAVA explicitly states: "The Commission shall not have any authority to issue any rule, promulgate any regulation, or take any other action which imposes any requirement on any State or unit of local government." For example, the EAC cannot create a single national standard for what constitutes a countable provisional ballot. Instead, its major powers are holding hearings and disseminating information--especially related to best practices--that raise awareness of specific election administration issues. The EAC also certified the state plans that allowed for federal funds to flow to states for the purchase of new voting equipment, voter registration systems, and related expenditures. Considered broadly, HAVA does reflect a national consensus for uniform and nondiscriminatory administration of elections. As such, the federal law explicitly requires state governments to ensure uniformity within states for voter registration and other aspects of the election process. It is up to the states to determine how this will be accomplished because the EAC cannot dictate standards for these activities.
The modest role for the federal government in HAVA means that the reform powers in HAVA were largely delegated to state election officials. This decision is interesting in part because most states had played a relatively small role in election administration prior to 2000. State laws typically make a state election official the final arbiter of election results and give the power to promulgate rules to that individual, but the responsibility for election management rests largely with local election officials. Prior to HAVA, in most states, local governments decided which voting systems and voter registration system would be used in their community. HAVA empowered states to accomplish politically what all could accomplish legally--to reclaim control over certain aspects of the elections process they may have historically granted to local governments.
The potential for a shift in the power dynamic between state and local governments required by HAVA can be seen quite clearly in the process that was developed for states to complete prior to receiving federal funds under this law. Section 254 is the centerpiece of the HAVA process; it outlines the requirements for a state plan to guide the HAVA implementation in a given state. The first twelve of the thirteen planning requirements explicitly discuss "how the state" will carry out a given activity. These include how the state will spend or distribute funds; provide programs such as voter education, poll worker training, and election official training; and develop performance goals and measures to determine whether HAVA's goals are being achieved. Although Congress was not requiring that these activities be implemented by the state governments, the language in HAVA clearly puts state actors in control of the planning process.
Section 254 puts the state in a dominant position for implementing HAVA. Section 255 affects the control each state has over the planning process based on its power to determine the composition of its state planning committee. Although the two largest election jurisdictions in a given state have to be included, the state is able to select all other members. Local election officials have a voice in this discussion; given their expertise in actual election implementation, they are obvious candidates for inclusion in a state HAVA committee. However, they are not agenda setters in the process. The state election officials can easily stack the HAVA committee with their allies to achieve the outcome they desire. The state officials are also in a position to control the drafting of the certification document, further increasing their control. Equally as important for states, the scope of activities for which states can use HAVA funds is quite broad. As long as the state controls the HAVA plan, it can determine whether monies should be spent on voting technology, voter registration, voter education, programs for recruiting and training election officials and poll workers, improving the quantity and quality of polling places, or assuring access for voters with physical and cognitive disabilities.
The ability of states to control the process vis-a-vis local election officials can be seen in the composition of the state planning committee membership listed in the HAVA reports published in the Federal Register. On average only 26.9 percent of state HAVA committee members in 2003 were local election officials. On only three state committees--in Vermont, Georgia, and Pennsylvania--did the percentage of local election officials equal or exceed 50 percent. At the other extreme, in four states--Rhode Island, West Virginia, Alaska, and California--local election officials accounted for less than 15 percent of the membership of the state HAVA committee. Table 1 categorizes states based on the percentage of local election officials on the state HAVA committee. The modal percentage of local election official involvement is between 25 and 30 percent, and more than half of the states have a local election official participation rate of less than 30 percent. At this level, local officials have a say in the process, but the state is in a dominant position to control the outcome of the debate.
HAVA ACTIVITIES AND STATE CONTROL
Given the state's control over the planning process, the plans can be viewed as reflecting the level of control that the state desires to exercise over elections. Again, although state laws govern the power possessed by local election officials, the HAVA planning process provided political cover to state officials who might want to change this dynamic. States are in a position to take control over certain aspects of elections but can leave local governments in control over others. The variability in the level of control a state might want to assume over a given component of elections can be seen in the area of voter registration. Section 303 of HAVA requires that
each State, acting through the chief State election official, shall implement, in a uniform and nondiscriminatory manner, a single, uniform, official, centralized, interactive computerized statewide voter registration list defined, maintained, and administered at the State level that contains the name and registration information of every legally registered voter in the State.
Section 303 follows from both the NCFER and Caltech/MIT VTP reports, which recommended the development of statewide voter registration systems. However, there are many ways to implement a statewide voter registration system. One method--used by Michigan, for example--is to have the state play a large role in the operation and management of the system. By contrast, Utah is implementing a statewide voter registration system whereby the state is developing and will manage certain aspects of the system, but the local governments will maintain much of the registration control they had prior to HAVA.
A similar pattern can be seen in the procurement of voting technologies that meet Section 301 voting system standards, especially standards for voters with disabilities. Some states have used HAVA as an impetus to engage in statewide voting system procurements that put all local jurisdictions on the same voting platform. (13) The advantage to this effort is that it allows the state to conduct uniform voter education campaigns on how to use the voting system; it also allows the state to purchase the equipment in a larger quantity, which can improve the price for the overall purchase. However, many states have continued to allow local election officials to determine which Section 301-compliant equipment will be used by citizens.
Table 2 categorizes all states across three dimensions. The first dimension is whether local election officials will play a strong role in the maintenance of the state voter registration system. Although HAVA requires the state to implement a statewide system, the role that local governments can or should play in this process is unstated. The second dimension is whether the state is planning to adopt a uniform statewide voting system. The third dimension is whether the state is planning to dominate the voter education process. Voter education has both political and educational purposes, especially for the individual who sends out the information. Control over this can be seen as an issue both of state versus local control and of political versus policy interests in voter education.
In Table 2, control is defined through the language used in the HAVA state plan. For example, state control over voting system procurement is noted whenever a HAVA plan requires a statewide uniform plan. With voter registration, local control is noted when the statewide voter registration system overlays on local systems or local governments retain almost complete control over system management. In the area of voter education, most state plans explicitly discuss how the state election office will fulfill voter education. Shared control exists here in several states, where HAVA plans discuss collaboration or combined implementation between state and local election officials. Many states are using election reform to control important aspects of elections, especially voter education and voting system choice. This shift in control from historical precedent provides state election officials with greater influence over polling place activities because they can control both the information that voters get before they go to vote and the type of voting technology that is used by voters in the voting booth.
APPROACHES TO HAVA PLANNING
The decisions regarding the role of the state government and of local governments stated in a HAVA plan reflect the outcome of certain decision processes. States can take several different approaches in developing a HAVA plan. Using case studies, we focus on two dimensions of the planning process that come from different theoretical traditions. The first dimension is the choice of decision-making process to use, and here we contrast rational policy analysis with a political decision-making process. (14) The rational policy analysis process involves a careful process of defining the policy problem, collecting information, defining the proposed policy solution, and communicating that solution. The second approach comes from pluralism and the political interest group process, where rational considerations play a diminished role in defining and achieving a specific policy outcome. (15) Instead, political considerations are at the forefront of the policy process, with interest group pressures and the political interests of the principals determining the solution selected. Interest group elites typically dominate this process, and although corporate interests are often seen as being dominant elites in pluralist debates, the more defining attribute is that the interests of traditionally disenfranchised groups are discounted.
Second, we examine the level of conflict that frames the environment in which decision making occurs. Here, we contrast the high-conflict environment in California with the consensual environment in Georgia. In practice, the low-conflict environment was conducive to a participatory management process. This allowed for effective citizen and interest group involvement in various aspects of the policy process, which can increase public support and acceptance for the selected policy choice. By contrast, the high-conflict environment in California was marked by litigation and a failure to negotiate among interests. (16)
Thus, the two contrasting cases of Georgia and California fall into opposing categories in this typology. California's HAVA process became conflictual and political, whereas Georgia's was consensual and rational. Below, when we consider each of these cases in detail, we provide more analysis as to how we can categorize each case along each of these dimensions.
Of course, these are not the only approaches that states could adopt to implement their HAVA (and other) election reforms. First, states could (and did) adopt approaches that were a mixture of pluralistic and rational; pluralistic approaches could be applied to some domains of reform (for example, state decisions about voting technology acquisitions in the wake of HAVA's passage could be pluralistic, whereas their approach to the development of the statewide voter list could be more rationalistic). Second, some states appear to be implementing HAVA in a "muddling-through" approach, where they are tackling HAVA reforms only incrementally and in what appear to be minimal ways. (17) Here we focus on the pluralistic and rational approaches, as they are quite different approaches to implementation of a reform package like HAVA.
In the case studies that follow, the approach chosen largely reflects the political choice of the state's election officials. HAVA created a large opening in the policy window that provided these state officials with a broad choice of approaches. The following case studies illustrate two of these approaches. In Georgia, a rational and participatory approach was used to determine how to reform elections. The state involved primarily local election officials in reform, but it also included voters and groups interested in reforming elections. By contrast, in California, a more pluralistic approach was used, with substantial involvement of many stakeholders throughout the state, but where the role of local election officials was much less significant than in Georgia and where the secretary of state's office attempted to use HAVA to centralize some power (e.g., voter education initiatives and possibly the statewide voter registration database). However, the very pluralism used in the early stages of HAVA implementation in California produced a highly confrontational situation between interest groups, local election officials, groups representing the traditionally disenfranchised (especially individuals with disabilities and language minorities), and the secretary of state's office. These choices reflect very different political and policymaking choices made by each secretary of state to achieve his or her election reform goals, and they possibly reflect the influence of other variables, such as the existing relationships between state and local election officials, the political and partisan context of each state, and concerns in the state about possible litigation and federal scrutiny of their electoral process.
THE RATIONAL PROCESS APPROACH--HAVA IN GEORGIA
Although the focus of the nation was on Florida after the 2000 election, the first legal action related to election administration occurred in Georgia in the case Andrews v. Cox. (18) The American Civil Liberties Union (ACLU) summed up the electoral problems in Georgia in a press release that accompanied its lawsuit:
voters in some Georgia counties were ten times more likely than others to lose their right to vote because of a "fatally flawed" system that disproportionately affects people of color ... the state's voting system [is] "a hodgepodge consisting of antiquated devices, confusing mechanisms, and equipment having significant error rates even when properly used." The state's failure to accurately record votes deprives its citizens of equal protection and due process as guaranteed under state law and by the United States Constitution, and violates the federal Voting Rights Act. (19)
Although this legal action could be seen as providing an impetus for subsequent state action, in fact the state was already acting. Georgia's secretary of state, Cathy Cox, a Democrat, did not dispute the claims made by the ACLU because her office was the source of the lawsuit's data. Cox saw the 2000 election debacle as an opening for her to improve election administration in Georgia and to secure resources for this purpose. In John Kingdon's language, Cox took advantage of the policy window opening and acted as a policy entrepreneur to get the result she desired. (20) This effort started with a comprehensive analysis of the problems that had occurred in the election and then the creation of a proposed solution to the problem.
The problems in 2000 in Georgia were outlined in report entitled "The 2000 Election, A Wake-Up Call for Reform and Change." (21) The report focused on problems associated with the antiquated voting systems used in the state. Two critical problems identified were large variations in over-vote and under-vote rates across voting systems and variation in these rates across demographic groups--especially white and African American voters--within given platforms. This report and subsequent studies found that African American voters were statistically less likely to have their votes counted on paper-based voting systems--punch cards and optical scan ballots--than white voters. (22)
Cox proposed addressing this problem through a clearly delineated, structured, and inclusive reform process. Specifically, she drafted and had legislation introduced into the Georgia General Assembly that would promote election reform by having the state purchase a uniform electronic voting system to be implemented across the state. The legislation did not empower the secretary of state to select the equipment but instead created the 21st Century Voting Commission to conduct a pilot test of various voting technologies to determine which one best suited the needs of Georgians. The commission had eighteen members, and Cox used all eight of her appointments to select local election officials, bringing important expertise and credibility among the other 151 county election officials to the process. The commission then held hearings to allow for greater civic involvement and to educate the public about the state's election reform efforts. The commission then pilot-tested various voting systems in local elections in the fall of 2001 and conducted exit polling as a part of this effort to determine public support for each voting system. This effort was augmented by survey data collected by the University of Georgia's Carl Vincent Institute of Government regarding statewide attitudes about the move to electronic voting. The decision by the state to move to a single electronic voting technology statewide reflects the outcome, therefore, of a process that identified a problem, examined various alternatives for addressing the problem, tested specific alternatives, and allowed for public input into the process.
The work of Georgia's 21st Century Voting Commission presaged in many ways the work many states did in their HAVA planning. Georgia's HAVA committee membership and HAVA plan largely reflect this previous work and illustrate a rational approach to planning and analysis. The rationalist approach begins with information gathering. (23) This feeds into both problem analysis (developing an understanding of the problem) and solution analysis (determining what solution might best address the problem). Georgia engaged in two sets of information gathering, starting with its initial evaluation and report on the problems in the 2000 election and again with the conduct of its pilot-testing. This rational process paid dividends for the secretary of state. It allowed her to develop buy-in and support from other state officials, especially the governor and legislators. This was critical to the success of her efforts because it gave her access to the resources she needed to make the voting system transition a success. For example, in Georgia's 2003 budget, the secretary of state received thirteen positions and $3.4 million to conduct voter education and other activities related to the transition to electronic voting. These resources made garnering support among local election officials much easier because the state could exchange money, resources, and training to localities for greater state control over elections, including determining the voting system to be used statewide.
HAVA was passed just as Georgia was holding its 2002 elections, the first using electronic voting statewide. The analyses that were carried out immediately after the election found that the reform had been a success. The statewide residual vote rate went from 4.8 percent statewide for the top-ticket race under the myriad voting systems used in 1988 to 0.88 percent in 2002. (24) When the residual vote rates in 2002 were compared with the residual vote rates for the 1998 gubernatorial and 1996 U.S. Senate races, it was determined that approximately 77,000 more ballots were counted using electronic voting than would have been counted under the previous systems. More recent analyses have shown that the greatest reductions in residual votes occurred among low-income and African American voters, groups which have been historically disenfranchised in the state. (25) These outcome data reflect the achievement of the rational policy goals that were established at the outset of this process.
In HAVA, Georgia was able to build on the process used by the 21st Century Voting Commission. Many of the same participants were involved in both commissions, and Cox saw to it that the HAVA commission had a strong representation by local election officials. In fact, Georgia had the highest rate of local election official involvement in HAVA, along with Pennsylvania, with 60 percent of the HAVA planning members in the state being local election officials. The state again used a participatory approach for developing the HAVA plan and was able to leverage the success of the previous reforms to promote activities such as uniform statewide poll worker and election official training and voter education. This effort was made easier by the state having a uniform statewide voting system.
THE PLURALISTIC APPROACH--HAVA IN CALIFORNIA
Although there is a stark contrast in their responses to the 2000 election, in many ways the starting points for reform in both states were similar. California, like Georgia, was sued after the 2000 election for using punch cards. In California, the outcome of the Common Cause v. Jones lawsuit was a consent decree that banned the use of punch-card voting systems as of March 2004. (26) Unlike the situation in Georgia, where Secretary of State Cox avoided litigation after the 2000 election, this court case in California was divisive and helped to frame the discussion over election reform in conflictual, not consensual, terms. Also, in March 2002, California voters approved Proposition 42, the "Voting Modernization Act," which allocated $200 million in state funds to replace old voting systems. Proposition 41, while allowing for a strong state role in the acquisition of new voting technologies by California counties, established the Voting Modernization Board to administer these funds, with three board members appointed by the governor and two appointed by the secretary of state. Proposition 41 passed in March 2002, the same election in which Kevin Shelley was chosen as the Democratic candidate for secretary of state--a race he would win in the fall, thereby replacing term-limited Republican Bill Jones.
Here we encounter two very important ways in which California differs from Georgia, differences that have led to two distinct HAVA implementation paths. First, California's election administration environment is vastly more heterogeneous than Georgia's, with California's counties representing a broader spectrum of administrative problems and practices after 2000 than Georgia's counties. Consider the simple demographics of the two states: Georgia's largest county, Fulton County, had 440,291 registered voters in the most recent (2005) enumeration; California's largest county, Los Angeles County, had 4,075,352 registered voters in the most recent (2005) enumeration, more voters than exist in the entire state of Georgia. (27) Although ten California counties have more registered voters than Fulton County, Georgia, California also has some very small election jurisdictions, including Alpine County (831 registered voters) and Sierra County (2,307 registered voters).
Not only is there a broader size range of election jurisdictions in California than in Georgia, but California is also more ethnically diverse. Although 28.7 percent of Georgia's population is African American, California has a smaller white population than Georgia and has a number of language minorities. For example, Los Angeles County must provide election materials in six languages as well as English. This diversity in California set the stage for a potentially more conflictual HAVA process, as a wider diversity of administrative practices and problems needed to be reflected in the development of the HAVA state plan.
The second major distinction between California and Georgia lay in state politics and process. California's term-limit law covers the secretary of state's office; individuals are allowed to hold the office for only two terms. Bill Jones, a Republican who had been secretary of state since 1994, was term-limited in 2002 and was planning a run for higher office just as the post-2000 election debate and HAVA development began. A major upheaval in the secretary of state's office occurred in 2002, when Democrat Shelley took over this office. Shelley, widely seen in California as a Democratic star in the making, came into statewide office looking to make his mark to support a likely run for another statewide office, possibly governor in 2006. As noted below, Shelley's political ambitions led to a much more political HAVA development and implementation process than seen in Georgia. Cathy Cox had only recently taken office--winning her seat in 1998--and operated in a more conservative political environment, both ideologically and in regards to decisions to run for higher office. Statewide officials in Georgia often do not run for other positions, or they wait very long periods before running for another office. The defeat of Governor Roy Barnes in 2002 and the rise of Republican Party competition has begun to change this dynamic as the state's political culture has changed.
Like most states, California created a state plan advisory committee after the passage of HAVA. The committee held five public hearings and released a draft plan in June 2003. However, California's HAVA planning was very different from Georgia's on several levels. First, Georgia acted faster than California, enacting many of its reforms before HAVA was even enacted by Congress. Second, California's planning process was much more pluralistic and political, something that began with the decision to appoint a wide range of interest groups to the committee and appoint very few election officials. In fact, California has the smallest number of local election officials on its HAVA planning committee of any state. Finally, as became very problematic later, the secretary of state used HAVA to achieve his own political goals at the expense of the state's interests.
During the implementation of HAVA in 2003 and 2004, two different storms blew into California's election reform efforts. One of these storms was the gubernatorial recall election, which began building in the spring of 2003 and erupted into a chaotic, controversial, and expensive election held on 7 October 2003. Before this election could be held, the Democratic secretary of state's office made many controversial decisions about the process and procedures under which this novel recall election could be held, and many of these decisions (and decisions by local election officials) were contested in a series of lawsuits in the summer and fall of 2003. (28) The logistics of the recall election, the litigation it spawned, and the growing controversy over what some observers saw as partisan election administration in California effectively stopped election reform well into the late fall of 2003.
The second storm, still raging today, is the debate over electronic voting machines. Partly because some of the strong advocates for the use of the so-called voter-verified paper audit trail (VVPAT) are located in California, and partly owing to the fact that it was clear that some California counties were poised in 2002 and 2003 to transition to electronic voting machines for precinct voting, the fight erupted in 2002 and 2003 in California--both in Sacramento and in a number of counties--over electronic voting systems. Secretary of State Shelley, sensing the building controversy, took a number of steps to try to resolve the debate. By 2004, he had decertified the electronic voting machines of one manufacturer and had issued a whole slate of short- and long-term regulations for the use of electronic voting machines produced by other manufacturers. In response, several counties and disability rights groups sued the secretary of state to have electronic voting equipment allowed in the 2004 election, which resulted in lengthy negotiations between the state and localities in order to arrive at a resolution. (29) At the same time, disability rights advocates were suing the state and four counties because they did not provide access to accessible voting technologies, such as electronic voting machines. This controversy consumed considerable time and resources and kept the HAVA reform process from moving forward quickly.
However, the pluralistic, and possibly partisan, evolution of the HAVA reforms in California slowed down at the end of 2004 as allegations arose that HAVA voter education funds (allocated to statewide voter education efforts as a result of the formula developed in the state's HAVA implementation plan) may have been used for partisan or political purposes by the secretary of state's office. A review of the use of HAVA voter education funds by Shelley's office was conducted by the state auditor, who found a number of questionable grants and questionable partisan activities conducted by contractors using federal HAVA funds. (30) This audit provided additional evidence that the HAVA effort in California was not focused on developing a uniform statewide voting process; rather, the HAVA effort was being used by state election officials in a pluralistic and possibly political manner. This controversy led to the decision by Shelley to resign on 1 March 2005.
Additionally, California's development of a statewide voter registration database is progressing slowly. Again, the difficulties in California with this component of its HAVA reforms are partly exogenous, affected by the delays imposed by the recall election, but they are also partly endogenous. Specifically, delays in HAVA planning occurred because of the secretary of state's heavy involvement of resources and energy in the national VVPAT debate and the allegations regarding improper use of HAVA funds. The substance of this debate centers on whether the current CalVoter voter registration system--which has been used since 2000 in all California counties and was developed under Shelley's predecessor, Bill Jones--is HAVA compliant and can serve the state's voter registration needs. The secretary of state supports developing a new system at a cost of between $40 million and $60 million.
The local registrars, led by Los Angeles County Registrar of Voters Conny McCormack, support modifying the existing system and have referred to Shelley's desire to build a new system as "folly." (31) McCormack, whose own county constitutes 25 percent of the registered voters in California, argues that CalVoter is either HAVA compliant or close to being so. As she told the Los Angeles Times, "Why not simply upgrade CalVoter to meet HAVA compliance, rather than attempt to procure and implement a new [system]?" (32) The state audit of HAVA singled out the failure of the secretary of state to address this procurement in a timely manner and noted that the state risks an enforcement action by the U.S. Department of Justice if it fails to develop this system before 1 January 2006. Although the exact approach that California will follow for the statewide database is currently unclear, what is clear is that the struggle over whether the system will be state-centric or locally autonomous is far from being decided.
The debate over CalVoter highlights the confrontational relationship that has been the trademark of California's HAVA process and the environment in which it has been implemented. Members of the HAVA planning committee have literally been suing each other over various aspects of HAVA implementation, and there is no clear set of goals for everyone to move toward. In fact, it often seems that much of the debate over election reform in California has occurred in the media and in the courts, as the various interest groups jockey to achieve their own goals.
The 2000 election created the impetus for states to rethink their laws governing elections; HAVA planning processes provided a clear mechanism for rationalizing these changes. States vary widely in the level of participation they allow local election officials in the HAVA planning process. Some states, such as Georgia, provide local election officials with a large say in how the state moves forward with HAVA. In Georgia, the state was able to build on a previously established relationship between the secretary of state and local government officials. In California, the state has pursued a much more pluralistic approach and attempted to involve a broad array of interest groups in the HAVA planning process, with local election officials playing a much more modest role.
We have also illustrated how state HAVA plans can vary in the level of control they are actually moving into state hands. Not surprisingly, states have made efforts to keep voter education--and the concomitant publicity it provides state officials--under their control. The movement of voter registration and voting equipment procurement to state control, however, is a much more fundamental change in the previous dynamic of local control over elections. These systems often shape the way work is conducted locally, and because elections are still implemented locally, the state having control over system procurement can change election management locally.
The Georgia and California cases also illustrate differences in approaches for identifying and addressing key issues within the state relative to election reform. In Georgia, there was a rational approach used for addressing election-related problems, with the state conducting a rigorous analysis of problems that existed with voting technologies. State officials then addressed these problems through the procurement of new voting technologies and were able to conduct policy evaluations to show the efficacy of this effort. By contrast, in California, the state has engaged in several studies related to voting equipment, but the studies have focused largely on the issue of VVPATs, in part because the leading advocates of VVPT are located in California. Other issues--such as improving the enfranchisement of language minorities and the disabled, and the reduction of the residual vote rate--have not been the focus of evaluation or action by California. Instead, a pluralistic, and somewhat political, approach has been used for HAVA implementation. The process has been conflictual, with the result being several lawsuits, a harsh state audit of the spending of HAVA funds, and a delay in the implementation of HAVA.
Other states have also followed more rational or pluralistic approaches in the development of state HAVA plans. As these plans are implemented, it will be possible to determine which approach resulted in more effective election reform for voters. The pluralistic model, which in practice can have a conflictual aspect, may lead to slower implementation of reforms and cause state and local election officials to be overly cautious in order to avoid lawsuits and other forms of pressure. The rational model, by contrast, has the potential to build public and interest group support that may provide election officials with greater confidence to move forward with more reforms to improve the election process.
Soon, it will be possible to examine how power has shifted between states and localities and whether states are up to the challenges that HAVA has placed at their doorstep. A key question is whether these changes will benefit voters, the ultimate customer of election services. The HAVA plans are only documents. It is up to the states and localities to implement these plans. It is likely that some reform models and some planning models will build the political capital and support needed to turn these plans into reality.
AUTHOR'S NOTE: We wish to especially thank the editors of this special edition for their helpful comments on earlier drafts of this article.
R. Michael Alvarez
California Institute of Technology
Thad E. Hall
University of Utah
(1) See Kenneth A. Gross, "Constitutional Restrictions on Federal and State Regulation of the Election Process," To Assure Pride and Confidence in the Electoral Process, ed. Jimmy Carter, Gerald R. Ford, Lloyd N. Cutler, and Robert H. Michel (Washington, DC: Brookings Institution, 2002), pp. 222-235, and Pamela S. Karlan and Daniel R. Ortiz, "Congressional Authority to Regulate Elections," To Assure Pride and Confidence, pp. 235-245, for a complete discussion of the constitutional issues surrounding federal governance of elections.
(2) Bush v. Gore, 531 U.S. 98 (2000).
(3) Georgia's and California's election reform efforts have been evaluated using different criteria. See, e.g., Randall Strahan and Mathew Gunning, "Entrepreneurial Leadership and Election Reform in Georgia, 2001 to 2003," Election Reform: Politics and Policy, ed. Daniel J. Palazzolo and James W. Ceaser (Lanham, MD: Lexington Books, 2005), pp. 59-73; Karin McDonald and Bruce Cain, "California: Low-Tech Solutions Meet High-Tech Possibilities," Election Reform, pp. 123-140.
(4) Richard B. Schmitt, "Gore's Suit Is a Duel of Experts As Lawyers Contest Every Point," Wall Street Journal, 4 December 2000, p. A26.
(5) Thad E. Hall, "Public Participation in Election Management: The Case of Language Minority Voters," American Review of Public Administration 33 (December 2003): 407-422.
(6) Caltech/MIT Voting Technology Project, "What Is, What Could Be," July 2001; http://www.vote. caltech.edu/Reports/2001report.html (accessed 8 August 2005); Carter et al., To Assure Pride and Confidence; The Constitution Project's Forum on Elections, "Building Consensus for Election Reform," and "Recommendations for Congressional Action"; http://www.constitutionproject.org/evi/ (accessed 8 August 2005).
(7) See, for examples in the private sector, Christopher Bartlett, "MNCs: Get Off the Reorganization Merry-Go-Round," Harvard Business Review 61 (February 1983): 138-136; Richard Daft, Organizational Theory and Design (Cincinnati, OH: South-Western College Publishing, 1998), pp. 253-258. See, for examples in the public sector, Ronald Moe, Administrative Renewal: Reorganization Commissions in the 20th Century (Lanham, MD: University Press of America, 2003); Hannah Sistare, Government Reorganization: Strategies and Tools to Get It Done (Washington DC: Center for the Business of Government, 2004).
(8) Sistare, Government Reorganization, p. 7.
(9) Thomas H. Hammond, "Agenda Control, Organizational Structure, and Bureaucratic Politics," American Journal of Political Science 30 (April 1986): 379-420.
(10) Carter et al., To Assure Pride and Confidence.
(11) Both authors worked with one of the two commissions mentioned in this article and either participated in these consultations or were apprised of these activities by their colleagues.
(12) See, e.g., Martha Derthick, "Preserving Federalism: Congress, the States, and the Supreme Court," Brookings Review 4 (Spring 1986): 32-37.
(13) Another factor that might motivate states to move toward standardization and uniformity in voting system use across counties is fear of litigation on equal protection grounds, based on the logic contained in Bush v. Gore (2000) and in other recent court decisions in the states, such as Common Cause v. Jones, 213 F Supp. 2d 1110 (C.C. Cal. 2002).
(14) See, e.g., David L. Weimer and Aidan R. Vining, Policy Analysis: Concepts and Cases, 4th ed. (Upper Saddle River, NJ: Prentice Hall, 1999); John C. Thomas, Public Participation in Public Decisions: New Skills and Strategies for Public Managers (San Francisco: Jossey-Bass, 1995); Robert Backoff, Barton Wechsler, and Robert E. Crew Jr., "The Challenge of Strategic Management in Local Government," Public Administration Quarterly 17 (Summer 1993): 127-144.
(15) See, e.g., Theodore Lowi, The End of Liberalism: Ideology, Policy, and the Crisis of Public Authority (New York: Norton, 1969); E. E. Schattschneider, The Semisovereign People: A Realist's View of Democracy in America (New York: Holt, Rinehart and Winston, 1960); David Truman, The Governmental Process: Political Interests and Public Opinion (New York: Knopf, 1951).
(16) In his discussion of policy and social change, Michael Hayes also identifies conflict and complexity as key issues in decision making. See Michael T. Hayes, The Limits of Policy Change (Washington, DC: Georgetown University Press, 2001).
(17) Ibid., 39-46.
(18) Andrews v. Cox, Civil Action File No. 2001CV32490, Superior Court of Fulton County, 2001.
(19) American Civil Liberties Union, "ACLU Seeks to Remedy Georgia's 'Fatally Flawed' Voting System," 1 December 2004; http://www.ach.org/VotingRights/VotingRights.cfm?ID=7000&c=166 (accessed 8 August 2005).
(20) John W. Kingdon, Agendas, Alternatives, and Public Policies, 2nd ed. (New York: HarperCollins College, 1995).
(21) Georgia Secretary of State, "The 2000 Election: A Wake-Up Call for Reform and Change," January 2001; http://www.sos.state.ga.us/acrobat/elections/2000_election_report.pdf (accessed 8 August 2005).
(22) Charles S. Bullock III and M. V. Hood III, "One Person, No Vote; One Vote; Two Votes: Voting Methods, Ballot Types, and Under-Vote Frequency in the 2000 Presidential Election," Social Science Quarterly 83 (December 2002): 981-993.
(23) Weirner and Vining, Policy Analysis, p. 257.
(24) Georgia Secretary of State, "Analysis of Undervote Performance of Georgia's Uniform Electronic Voting System," 1 December 2004; http://www.sos.state.ga.us/pressre1/undervoteanalysis.htm (accessed 8 August 2005).
(25) See, e.g., Charles Stewart III, "The Reliability of Electronic Voting Machines in Georgia," Voting Technology Project Working Paper, October 2004; http://www.vote.caltech.edu/Reports/georgiastewart.pdf (accessed 8 August 2005), for an analysis of Georgia's residual vote rates. See Arnold Fleischmann and Carol Pierannunzi, Politics in Georgia (Athens: University of Georgia Press, 1997), for a discussion of the state's political culture.
(26) Common Cause v. Jones.
(27) Data on California's demographics can be found from the Census Bureau's QuickFacts; http:// quickfacts.census.gov/qfd/states/06000.html (accessed 8 August 2005); data on Georgia's can be found at http://quickfacts.census.gov/qfd/states/13000.html (accessed 8 August 2005).
(28) See Maura Dolan and Jean Guccione, "The Recall Campaign: Court Dates Approach for Challenges to Recall," Los Angeles Times, 14 August 2003, p. A21, for a summary of the major lawsuits.
(29) Seema Mehta, "Secret Vote at Risk, Suit States," Los Angeles Times, 7 May 2004, p. B5.
(30) California State Auditor, "Office of the Secretary of State: Clear and Appropriate Direction Is Lacking in Its Implementation of the Federal Help America Vote Act," December 2004; http:// www.bsa.ca.gov/pdfs/reports/2004-139.pdf (accessed 8 August 2005).
(31) Tim Reiterman, "Shelley Reconsidering Voter Registry Plan," Los Angeles Times, 24 December 2004, p. B6.
Table 1 Percentage of local election officials on HAVA planning committee, 2003 10-15 RI, WV, AK, CA 15.1-20 IA, LA, IN, NE, NC 20.1-25 MN, NH, ND, WY, IL, WA, MT 25.1-30 SD, KS, TX, FL, NY, DE, SC, ME, NV, KY, MA 30.1-35 UT, ID, AZ, MD, MI, MS, TN 35.1-40 AL 40.1-45 OR, AR 45.1-50 WI, OK 50.1-55 VT 55.1-60 GA, PA Table 2 State versus local control of various election reform components Maintenance Voting of voter system registration State AK, CA, GA, HI, AL, AK, AR, CT, DE, HI, control IL, KY, LA, MI, GA, MD, MI, MN, MS, MT, NV, NC, NE, NV, OK, RI, SC, UT OK, PA, SC Shared CT, DE, IN, NM, WY CT control Local AL, AZ, AR, CO, FL, CA, CO, FL, ID, IL, IN, control IA, KS, MA, ME, MD, IA, NKS, KY, LA, ME, MA, MN, MO, MS, NH, NJ, MO, MT, NE, NJ, NM, NY, NY, OH, OR, RI, SD, NC, ND, OH, OR, PA, SD, TN, TX, UT, VT, VA, TN, TX, VT, VA, WA, WV, WA, WV, WI WI, WY Voter information initiation State AL, AK, AR, AZ, CA, control CO, CT, GA, HI, ID, IN, IL, IA, KY, LA, ME, MD, MN, MT, NE, NV, NH, NM, NC, ND, OH, OK, RI, SC, SD, TN, TX, UT, VT, VA, WV, WI, WY Shared DE, FL, MA, MI, MS, MO, control NJ, NY, OR, PA, WA Local KS control
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Help America Vote Act of 2002|
|Author:||Alvarez, R. Michael; Hall, Thad E.|
|Date:||Sep 22, 2005|
|Previous Article:||Implementing HAVA's voter education requirement: a crisis and a federal mandate improve state-local cooperation in Florida.|
|Next Article:||The effect of HAVA on late-to-innovate states: external influence on election reform in Arizona and Illinois.|