Printer Friendly

The effect of HAVA on late-to-innovate states: external influence on election reform in Arizona and Illinois.

In the wake of the Florida recount debacle of 2000, policymakers in Arizona and Illinois examined their election frameworks and pursued reform. Chief among the concerns of voting methods experts was the existence of punch-card voting systems in a vast majority of counties in both states. The detrimental effects of punch-card systems (most notably in the case of undervotes) (1) became legendary during the mess that was the Florida recount. The outcome in Florida determined the winner of the presidency; hence, all eyes turned to the Sunshine State. In that case, the number of undervotes was alarming, given that the total number of undervotes dwarfed by far the 537 vote margin between George W. Bush and Al Gore. (2)

In terms of Arizona and Illinois, both states were similarly culpable. In fact, Illinois had a larger overall percentage of voters using punch cards than did Florida, while Arizonans voted with punch-card technology in nine of fifteen counties. Hence, it seemed quite obvious to many observers that Arizona and Illinois needed to adopt reforms. However, serious reform was stymied in both states for well over two years. In the meantime, certain other states passed ambitious reforms, and many others approved significant, though not landmark, changes to their election voting functions. (3) Late in the summer of 2003, on the heels of federal innovation, state leaders in Arizona and Illinois finally enacted legislation to reform election administration in their states.

Why did efforts to innovate in Arizona and Illinois fail in 2001 and 2002, while efforts elsewhere progressed? To gain insights into questions such as these, public policy scholars have developed the policy-diffusion literature. (4) This literature probes the why and when of policy innovation. However, as a whole, policy-diffusion scholars are better able to explain why certain states innovate early in the process than why the late innovators finally do so. That is, traditional (and typically internal) explanations of state innovation help us to understand the why and when of innovation in the early-going states, but these explanations lack explanatory value for late-innovating states. This circumstance leads us to look to unique, sometimes external, explanations for the late-innovating states. Indeed, in Arizona and Illinois, election reform was eventually spawned in part by an external force: the federal government in the form of the Help America Vote Act (HAVA).

This article explores the Arizona and Illinois experiences with election reform, including consideration of why the states innovated, when they did, the subsequent impact of HAVA, and how they differed from each other. First, I discuss the diffusion literature as it pertains to election reform. Second, I provide background information on both states pertaining to their election traditions and lateness in innovating. Third, the specific impact of HAVA on each of the two states is covered. Next, the cases are compared and contrasted. The article finishes with conclusions and prospects for future research on policy innovations and late innovators.


Decisions such as those of the American states to initiate election methods reform have been termed policy innovations in the public policy literature. The goal of such research is to answer the question: "What causes a government to adopt a new program or policy?" (5)

The main focus of state policy-innovation research is on the relationships between broad political and social characteristics of states and regions and the date when new policies or programs are adopted. Beginning with Jack Walker, state policy-innovation studies have regularly appeared in the public policy literature. (6) This literature has been concentrated in two main and mostly separate areas. One area studies the effect of internal determinants on the decision to innovate. These internal-determinants models posit that the factors which lead state governments to innovate are political, economic, and social characteristics unique to the state in question. (7)

A second and larger area of state policy-innovation literature examines the diffusion of policies among the states. These studies emphasize the importance of other states and neighboring states in the decisions of states to innovate. (8) Research in both of these areas generally finds that large, well-funded states innovate before smaller, less wealthy ones. That is, the former are progressive leaders, and the latter are reactive followers. At some point, almost all states jump on board what essentially becomes a positive feedback process. These processes involve a somewhat slow start, a fast-moving stage where the idea takes off and many states want to get involved, and then a slowing down as the innovation reaches the final few states. During the diffusion, innovation becomes a fad, and many of the states tend to mimic one another.

If plotted graphically, the pattern of innovation diffusion across the American states typically appears as an S-shaped curve. Figure 1 demonstrates this shape for the diffusion of statewide voter registration databases. Based on state data from, we see that in 1968, South Carolina launched the first such database. (9) The subsequent spread of innovation was quite gradual, until diffusion hit a critical mass of states by 2003. At that time, helped along by HAVA, a large group of states jumped on board as the innovation diffused quickly. (10)


What the classic aggregate frameworks ignore are the interesting dynamics of predecision processes and the forces affecting them. Along these lines, one shortcoming of the state policy-innovation literature is that the role of agenda setting is either underplayed or left out. "Most research on policy innovation assumes earlier agenda setting, but it rarely inquires into that process. But to understand how particular issues emerge, evolve, and are adopted it is essential to begin to forge links between agenda setting and innovation." (11)

Clearly, examining broad social and political characteristics and/or diffusion patterns to predict innovation in a discrete policy area largely ignores the predecision phases of policymaking. There are notable exceptions to the classic approach. One such exception is the work of Henry Glick on right-to-die policy adoptions. (12) In an extensive work that includes in-depth case studies of legislation in three states and consideration of aggregate measures in all fifty states, Glick attempts to link state policy innovations with agenda setting. He finds that "the tendency of most states to adopt right to die legislation is shaped by the rising fide of litigation and publications in the mass media, but they are affected too by the power of the Catholic church in state politics." (13) Such intricate causes of policy innovation are left out when--as was the case in much of the early state policy-innovation literature--scholars consider only broad state characteristics and aggregate issues. (14)

To gain analytical leverage in the area of election methods reform in the states since 2000, Daniel Palazzolo developed a theoretical framework that like Glick's work, explicitly considers predecision processes and carefully models unique, microlevel state factors. (15) To understand variance in reform across the states prior to HAVA, Palazzolo modeled: (1) the threat of a close election, (2) election law capacity, (3) state political culture, (4) partisan control, (5) the budgetary situation, (6) the influence of interest groups and election officials, (7) recommendations from special commissions, and (8) leadership. States were expected to innovate if they had close elections, low election law capacity, a moralistic political culture, unified party control, slack budgetary resources, change-oriented groups and officials, active commissions, and leaders who wanted change. States such as Florida exhibited many of these factors, whereas others such as New York did not. Hence, this framework helps us understand why Florida innovated early and New York dragged its feet.

Other than the lack of attention paid to predecision processes, a second shortcoming of the state policy-innovation literature is the tendency, with a few notable exceptions, to ignore the potential influence of the federal government on state innovations. While state factors can be helpful in explaining why some states innovate and others do not, they are less helpful in explaining why late-innovating states finally come on board. At this point, internal factors have typically run their course and external factors become an important consideration in explaining the innovation. The diffusion of state policy innovations may be helped along when the federal government is active in a policy area and/or institutes programs that require or encourage the states to innovate in order to receive funding. This may be accomplished by offering federal funding contingent on the establishment of a state program with matching funding, which has occurred across a wide swath of policy areas in recent decades, such as the State Student Incentive Grant (SSIG) program. The SSIG program encouraged states to develop needs-based financial aid programs for college and university students by providing a one-to-one dollar match to establish programs. This approach yielded a plethora of new programs in states that had not seriously considered establishing one. In summary, states respond to federal influence and dollars. This can especially be the case late in the diffusion of innovations.

Recent scholarship engages this hypothesis of federal influence on state innovations. Michael Rich and Craig Volden explore the impact of federal cues on state levels of welfare assistance. (16) They find that decreased federal assistance makes welfare assistance less likely and that increased federal assistance makes it more likely. In a thorough test of three policy areas, Lawrence Grossback, Sean Nicholson-Crotty, and David A. M. Peterson explicitly model federal influence on state innovation. They find strong evidence for such an effect, even in the absence of funding, in the areas of state lottery adoptions, academic bankruptcy laws, and criminal sentencing guidelines. (17)

Palazzolo's framework of election reform diffusion also improves the classic literature's status quo in this regard. He explicitly includes a role for external forces, such as the federal government, in precipitating policy change. As noted above, the preponderance of factors in the framework was helpful in explaining why some states innovated in major ways, other states did so more incrementally, and still others gridlocked. What these factors are less helpful in explaining is why the states that gridlocked initially ultimately did innovate. It is Palazzolo's last factor, external forces, that picks up the slack. The federal government, through HAVA, strongly influenced the remaining states to innovate.


The various factors utilized to explain why other states innovated help us understand why Arizona and Illinois did not. However, other explanations (external ones, namely, HAVA) must be considered to fully explain subsequent innovation by these two states.

Case Selection

In order to utilize in-depth case analysis to study the unique dynamics of the late-to-innovate states, I first needed to select particular states. To do this, I consulted the Palazzolo typology. (18) He divides the fifty states into three patterns of election reform: leading major reform states (three states), incremental change states (thirty-eight), and late-developing reform states (nine). Arizona and Illinois were selected for this analysis because they are indicative of late-to-innovate states. Between the two states, we obtain an interesting combination of population size (i.e., Illinois being much larger), geographical location (i.e., Illinois in the Midwest, Arizona in the Sunbelt), growth rate (i.e., Illinois growing incrementally, Arizona showing fast growth), demographics (e.g., the cultural diversity of Chicago and the Hispanic and Native American influence in Arizona), and political culture (i.e., traditionalist in Illinois, antistate tradition in Arizona).

Overview of Framework Applied to Arizona and Illinois

As mentioned above, the election methods reform movement that was sweeping the country in 2001-2002 did not snowball across either Illinois or Arizona. Both states looked at their systems with some element of concern, but the issue did not become especially salient in either state. A variety of factors explain the initial decisions to postpone reform. Table 1 applies key variables of the Palazzolo framework to the two states in the wake of the Florida recount of 2000 and prior to HAVA in 2003.

Arizona--Various Cross-Pressures Prevent Reform (19)

In the wake of election 2000, many people in Arizona wondered about the quality of their voting systems. Would the Arizona elections structure withstand a similar problem to that which transpired in Florida? Voters in the Phoenix area and in Tucson had long used optical scan equipment, but ten rural counties in Arizona (out of fifteen counties total) still utilized punch-card systems. Although a minority of voters used punch cards, the Florida situation had demonstrated how just a few hundred votes can matter. Hence, many who participated in postelection discussions on the two flagship university campuses in Arizona were concerned. The data bear out their concerns. In the 2000 elections, 1.6 percent, or over 24,000, of the more than 1.5 million ballots Arizonans cast did not register a choice for president. (20) While 1.6 percent may sound like a low rate of error, 24,000 votes can matter in a close election like the one in Florida in 2000.

Why, then, given the apparent need, were no major voting methods reforms passed in Arizona until summer 2003? The answer is multifaceted and begins with the "leadership" factor immediately after election 2000. On 7 December 2000, Secretary of State Betsey Bayless released a carefully worded white paper containing rough comparisons of the Florida and Arizona systems. Bayless acknowledged that like Florida, Arizona featured discontinuity of voting methods across counties. Five counties containing 80 percent of the voters used optical scanning election equipment, while ten counties, comprising the remaining 20 percent of voters, still utilized punch-card ballots. The secretary of state reviewed the advantages of optical scanning equipment, including accuracy and speed of counts, but fell short of an all-out condemnation of punch cards or an explicit endorsement of optical scanning statewide. (21)

A second, countervailing theme expounded by Secretary Bayless was the superiority of the recount process in Arizona compared with what had occurred in Florida. The secretary sought to put the best light on the Arizona system as she highlighted how a recount would have worked differently. Although the fifteen counties operate the election, initially count the ballots, and then report to the secretary of state, recounts kick in only when an election is especially close, as determined by state law; and when they do, the state conducts the recounts. Recounts are not conducted at the request of candidates, they are done statewide (not just in some counties, as in Florida) for statewide races, and they are never done manually.

In summary, the white paper contained some contradictions. Although Secretary of State Bayless hinted at the need for optical scanning equipment statewide, she spent much of the memorandum assuring Arizonans that recounts of punch cards would have proceeded much more smoothly in Arizona than in Florida. Hence, she suggested reform at the same time that she suggested reform might not be necessary. Therefore, depending on where one was sitting in the Arizona governmental structure (optical scanning versus punch-card counties, state government versus rural county government), the white paper could have been read as a call for a reform or a defense of the status quo. Indeed, leaders from several rural counties that utilized punch cards would later employ similar logic to argue that if the system is not broken, why should we fix it?

After the release of the white paper, discussions soon broadened, and a critical mass of policy leaders became involved. Upon prompting from two state senators, a statewide task force (the Modern Election Practices Task Force) was assembled to tackle the issue. The membership of that task force was similarly of two minds; many wanted change, while many others opposed it. The task force could never agree even to release recommendations. Thus began a split between the large population centers and state leaders in Arizona who favored modernization and the smaller counties that favored the status quo. This split endured until change was inevitable in the wake of the passage of HAVA.

According to the Palazzolo framework, interested groups can play a supporting or retracting role in reform. Indeed, a constellation of interested groups formed around the reform discussions in Arizona. Several groups supported voting equipment reform, including the Arizona Association of Counties, Maricopa County (the largest and most influential county in Arizona), and state election officials. However, pushing in the other direction were local officials from the punch-card counties, including the Arizona County Recorders Association, which did not believe that widespread reform was necessary and was concerned about the cost and the implementation of new voting systems in a short period of time.

Considered in terms of raw numbers, the pro-reform groups should have held more sway. They represented approximately 80 percent of the state's population. However, there was a high degree of mobilization from reform opponents, and they were intense in their opposition. Mobilization and unity among local officials in the punch-card counties was precipitated in part by the carefully worded white paper of the secretary of state and by the opportunity to participate in the informal task force. The "interested groups" factor, therefore, helps explain gridlock in Arizona until 2003.

Partisanship is also helpful for explaining why Arizona was initially considered a gridlock state in elections methods reform. Although statewide elections have leaned heavily toward Republican candidates in recent years, the Democratic party has made significant gains in recent state legislative elections. During the period when reforms were being considered (2001-2002), governing circumstances were especially vexing because there was a mixed model of party control, and majorities (to the extent they existed) were quite narrow. Essentially, Arizona did not feature unified government in 2001-2002 because the Republicans controlled only the governor's mansion and, narrowly, the state House of Representatives. This factor is an important one in the Palazzolo framework owing to the fact that under unified government, it is easier to make policy because one party, with the same or similar policy preferences, controls the institutional levers of power. Under divided government, different parties hold power and can block the other party's initiatives, which leads to stalemate and gridlock.

The Senate had a fifty-fifty split between Republicans and Democrats in 2001-2002. This presents an especially uncertain environment for making policy as neither party is sure it will be able to enact its policy preference; each wants to prevent the other party from securing legislative victories that it might parlay into votes in the next state legislative elections. Along these lines, when the House appropriations bill containing the $3.5 million for optical scanning equipment was considered, a Democrat (Harry Mitchell) proposed an amendment to remove the funding. Hence, Arizona fits the party control hypothesis for reform quite well. Lacking unified government, we would expect a greater likelihood of stalemate in policymaking; gridlock is indeed what occurred in the area of election methods reform in Arizona.

Illinois--A Bottom-Line Failure by Leaders to Acknowledge the Problem

According to David Kimball's profile of Illinois' process of election reform, (22) leaders were in full denial of their election methods problems in the wake of the 2000 recount in Florida. With a few notable exceptions, leaders stepped forward not to highlight the need for reform but to explain that the system was not broken and, hence, needed no repair. These leaders assured voters that the chad problems so salient in the Florida recount simply did not happen in Illinois. Of course, evidence suggests otherwise. In 2000, Illinois had one of the highest rates of undervotes in the country, a staggering 3.9 percent. Statewide, that translates to about 190,000 votes! To put that in perspective, one need only think back to John F. Kennedy's narrow win over Richard M. Nixon in 1960 and Bush's narrow win over Gore in Florida.

In a close election, the election process in Illinois could become quite ominous, and it actually did become problematic in 1982. In that year, there was a very close gubernatorial election in which both major party candidates (Jim Thompson and Adlai Stevenson III) declared victory. Recounts began sporadically in various counties, and legal challenges mounted. In what became a perfect foreshadowing of Florida 2000, the state supreme court in a divided vote of 4-3 certified Thompson the winner without additional recounts that were demanded by Stevenson's camp. That leaders in Illinois would have some level of awareness about the 1982 case and Florida 2000, together with an undervote total of 190,000 produced in a punch-card system, supports the idea that they had no desire to analyze their election methods problems.

In addition to a lack of leadership, there are significant institutional impediments in Illinois. The structure of elections is well entrenched in local areas, including the state's 102 counties. This tradition of local control is even stronger than in Arizona, where state election officials serve an important role on election day. Local control by definition may not mean that local officials will necessarily be disposed to oppose reforms. However, in Illinois, it is precisely the case. Local election jurisdictions have long used punch-card technology. In fact, as of election 2000, Illinois had the most widespread use of punch-card technology in the country. Although reforms were often introduced in the state legislature, they had never been seriously considered for action until HAVA passed. The setup of the state's election board does nothing to challenge this local autonomy. That board has equal representation from the two parties in a competitive party state. This reality means that the two parties have to compromise to make state election law. This leads to a situation where it is easier to keep delegating decision making to the localities, rather than pursuing a strong state role. (23) By extension, lacking a strong state role, Illinois is not predisposed to passively accept federal mandates in this area of policy.

These two factors (a lack of leadership to acknowledge problems and a weak state governance setup in elections) provide part of the explanation for election reform gridlock in Illinois. Another factor is the constellation of interests. In terms of interested groups, no group championed the cause of election reform. Although many organizations were involved with a state panel on election reform, they were mostly concerned with campaign finance laws. To the extent that election methods were a topic, they were couched in the context of helping people with disabilities. Only the League of Women Voters specifically raised the topic of systemwide election reform. The state panel also included many local election officials, who not surprisingly tended to oppose state-led reform of their local voting systems. In summary, the constellation of interested groups in Illinois provided no critical mass of support for reform. If anything, they served to hold reform back.

Party control of the elected branches provides yet another important explanation of the gridlock over election reform in Illinois. The state is frequently a battleground in presidential elections and has featured many close statewide races over the years. Although this might logically lead to more focus on election methods reform to assure that close races are decided fairly, it also serves to raise partisanship and deliver divided government in many election cycles. Indeed, during 2001-2002, Democrats controlled the legislative branch in Illinois while Republicans held the governor's mansion. When different parties control the institutional levers of power, it presents a particularly challenging set of governing circumstances. Policymakers in that case are best set up to block their opponents' initiatives rather than to pass their own. In the area of election reform, this led to the failure of many bills to move forward in the legislative process. Neither party wanted to pass reform, for fear it would benefit the other party at the polls and would give that party the upper hand on the issue. In summary, for much of 2002, the prospects for election reform in Illinois looked bleak. It seemed all the factors lined up against reform.


With its breadth and depth, the passage of HAVA in 2002 affected all American states to some degree, but its influence was especially potent on late-to-innovate states such as Arizona and Illinois. In this section, I will revisit the key factors from the Palazzolo framework and flesh out how HAVA impacted the process and the vote in both states. Table 2 shows the status of various factors in 2003-2004. We see that even before considering the impact of HAVA, certain conditions had changed that made reform more likely. In Arizona, an energetic leader emerged; in Illinois, the Democrats achieved unified party control.

HAVA and Arizona

Arizona attempted reform in 2001 and 2002, but the confluence of factors discussed above led to gridlock. However, passage of HAVA helped to take Arizona from seriously considering major reform to passing it. Arizona leaders clearly wanted to take advantage of federal funding ($51.7 million) and were also mindful of the prospective mandates in HAVA. To many in state government, HAVA communicates that states must do something. So, why not go ahead and get the funding while doing it? Arizona leaders in particular saw the 5 percent state share and 95 percent federal share for purchasing voting equipment, for example, as a particularly good deal, with mandates coming later anyway.

Once this realization was made, Arizona moved quickly to comply with HAVA and implement its provisions well ahead of the 2006 deadline. Leading the way was a particular entrepreneur, Secretary of State Jan Brewer, who won her position in the November 2002 state election. Brewer made election methods reform her main issue. Recall that Brewer's predecessor, Betsey Bayless, had initially "waffled" in discussing the need for reform while also saying that Arizona was better prepared than Florida for a recount. In summary, Brewer was insistent on reform at the outset. This clear and unmistakable leadership was important in moving Arizona toward reform. In this respect, Brewer can be compared to entrepreneurs for reform in other states, such as Secretary of State Cathy Cox of Georgia. Like Cox, Brewer was steadfast in her belief that Arizona should lead on the issue and not merely defend the status quo.

With the leadership of Secretary of State Brewer, within a year of signing on to the HAVA provisions (summer 2003), Arizona had discarded the remaining punch-card systems, implemented a statewide voter registration base, and pilot-tested special voting equipment (direct recording electronic with a paper trail) for disabled voters in a sample of precincts. (24) All of this was in place in plenty of time for the fall 2004 elections. Indeed, Arizona leaders take pride in the fact that they implemented HAVA provisions quickly and without asking for waivers from the federal government.

In a self-titled and prominent report dubbed the "Brewer Voting Action Plan," Secretary of State Brewer highlighted all that Arizona had accomplished in the wake of HAVA and leading up and through the 2004 elections. The state initially contracted for almost $2.5 million with Diebold Election Systems of North Canton, Ohio, to provide a first phase of optical scanning equipment for the nine Arizona counties that used punch cards. Unlike many other states, Arizona was suspicious about DRE technology without a paper trail and never seriously considered anything but optical scan technology. This concern may flow from the disposition toward "open" government that is rooted in Arizona's political culture. Arizona practiced "government in the sunshine" before it was popular, and the political culture of Arizona is very democratic. This tradition is exemplified in state elections, which regularly feature dozens of ballot questions. (25)

The 2004 elections were the first test of the technology, and all fifteen counties used the technology successfully. Arizona did opt to go with a different pair of companies from Diebold, Chicago-based AutoMark Technical Systems and Election Systems and Software, Inc. of Omaha, Nebraska, for the voting software for disabled access. Regarding the DRE machines (with a paper trail), Brewer aims to buy 2,000 such machines (one for each precinct) to comply with the HAVA mandate on disabled voter access by the 2006 deadline.

The above chronology demonstrates the progress Arizona made in the wake of HAVA and reads somewhat like a list of good news. Has all this accomplished anything? What about the effect; specifically what about the undervote issue? The well-publicized liability of punch-card systems was the tremendous number of ballots on which a candidate was not selected. Had the undervote issue been solved with the move to optical scanning equipment and away from the punch-card approach in the majority of Arizona counties? Given that the large majority of voters in the two most densely populated counties (i.e., Maricopa County in the Phoenix area and Pima Country in the Tucson area) were already voting with optical scan technology while the more rural counties had used punch cards in 2000, evidence suggests that there was a substantial effect with the change from 2000 to 2004. The percentage of undervoting fell from 1.6 percent to 1.3 percent. This may sound slight, but the percentage drop is more dramatic if the rural counties are separated out of the analysis. In the ten original punch-card counties, the undervote rate dropped from 3.8 percent in 2000 (using punch cards) to 0.9 percent in 2004 (using optical scanning).

The discussions and passage of election methods reform in the wake of HAVA also induced consideration of a further, non-HAVA topic: the goal of reducing the possibility of voter fraud in the form of noncitizens, namely illegal aliens, voting in the state. Arizona has a tradition of using direct democracy. In election 2000, voters were greeted with more than sixty ballot measures in the voting booth! The 2004 election brought Proposition 200, a broad and aggressive measure aimed at curbing the negative effects, and public expenses, of the influx of illegal aliens into Arizona. This measure passed with a 56 to 44 percent majority. (26) The proposition's election aspects will require proof of citizenship for someone to register to vote and aggressive ID checks on election day. The U.S. Department of Justice will have to clear these provisions under the Voting Rights Act.

In summary, in part because of the leadership of Secretary of State Jan Brewer, Arizona met most of HAVA's requirements and then some by the end of calendar year 2004. In contrast, Illinois policymakers moved at a less urgent pace.

HAVA and Illinois

As described above, Illinois lacked a sense of urgency for election methods reform, despite having widespread punch-card technology and significant undervote problems. Many leaders were not convinced of a need for major change. Indeed, much attention was instead dedicated to improving public education about the proper use of punch cards. This type of thinking carried over into the post-HAVA phase.

Although Illinois passed laws to comply with HAVA by 2006, the 2004 election still featured punch-card technology prominently, as more than half of the state's precincts used them. Of the nearly 12,000 precincts in Illinois, about 7,000 utilized punch cards, while the remaining 5,000 utilized some form of optical scanning equipment. Although a minority of jurisdictions used advanced technology, this did reflect significant improvement in technology adoption from the 2002 elections, when a vast majority utilized punch cards. Hence, the positive side of this reality is that nearly half of the counties are using more modern, HAVA-compliant technology when just four years earlier almost all polling places used punch cards.

Not surprisingly, the percentage of undervoting decreased dramatically in concert with this change. In the 2000 election, Illinois had one of the worst undervote percentages in the United States: 3.9 percent (about 190,000 ballots). In 2004, when close to half of the counties switched to modern technology (optical scan like Arizona), the undervote rate fell to 1.4 percent. This improvement, one would think, would inspire Illinois leaders to finish the job.

However, rates improved in punch-card counties as well, in part because of a voter education campaign about how to vote properly with punch cards. The undervote percentage fell from 4.3 percent to 3.6 percent in punch-card counties. The battle that modern technology faces for relevance in Illinois continues. Many other states, like Arizona, had moved more aggressively to change their voting systems, and Illinois will arguably be cutting it close to be compliant by 2006 on the voting technology dimensions of HAVA.

Yet, Illinois did pass statutes to accept election methods reform and to implement all the required provisions of HAVA. Why did the state do so? Major reform passed because of two factors: HAVA and party control. In the case of HAVA, the lure of $180 million was too great, even for a state with a strong tradition of local control. With mandates coming anyway in terms of election technology, it made little sense to say no to a 95 percent to 5 percent federal-to-state match. Even so, Illinois moved at a slow pace in adopting a HAVA framework. This more delayed approach is clearly reflected in Illinois's HAVA state plan, produced out of the executive director's office of the Illinois Board of Elections in October 2004. The plan reads very much like a "forced" update to the federal government, including a mere four-sentence introduction. To say it lacks enthusiasm would be an understatement. In nearly all sections of the plan, language such as "continues to work toward..." is used repeatedly. Illinois has requested waivers from the federal government on virtually all the prominent provisions of HAVA. The Arizona HAVA state plan is fundamentally different. An enthusiastic six-paragraph introduction from Secretary of State Brewer frames HAVA as a civil rights act of sorts. At many points in the Arizona plan, points are made with pride concerning early compliance with HAVA. The sheer length of the report also bears out the different approaches: Illinois at 5,921 words, Arizona with 12,790 words. Finally, Arizona to date has not requested waivers.

Party control was the second important factor in Illinois. During the gridlock phase, the elected institutions in Illinois featured split party control. This circumstance led to policy disagreement and a general fear that the other party would gain the upper hand on election reform. Hence, policymakers had more incentive to block their opponents' reform proposals than they did to pass their own. In the 2002 elections, Illinois voters delivered unified party control for the Democrats. The outgoing Republican Governor, George Ryan, while not speaking vociferously against it, had not made election reform a stated priority. Newly elected Democratic Governor Rod Blagojevich did not exactly carry the banner for election reform either, but neither was he motivated to block the reform efforts of Democratic state lawmakers. This served to make the final approval of legislation more predictable and allowed the legislature to move forward.

Comparing and Contrasting the Two Cases

Much is similar about these two states. Both have traditions of local control of elections and a general disdain for being told what to do by the federal government. Both states gridlocked initially on reform as a result of similar factors of partisanship, disagreement among stakeholders, and a lack of a particularly forceful leader (although Arizona got some leadership late in the game from the secretary of state). Both states also finally innovated when federal influence and financial incentives came into play. In this way, they were unique in their profile from earlier innovating states that were driven toward reform by traditional, internal state factors.

While there are similarities, much may be contrasted between the two states in terms of reform and response to HAVA. Arizona moved farther in the reform process initially (before HAVA) than did Illinois. Hence, HAVA was all that was needed to fully embrace election reform and at a comparatively blistering pace. Arizona was punch card free within one year and HAVA compliant on all but the disabled voter requirement. On that latter issue, as was detailed above, Arizona tested new DRE software for disabled voters in a sample of precincts during the 2004 presidential election and has contracted with two companies to provide one for every precinct by 2006.

By way of comparison, Illinois leaders are still not entirely convinced that they have a problem. It is commonsensical, then, that they were not as far along as Arizona in considering reform when HAVA was passed. This is reflected in their slowness to implement HAVA-compliant voting methods. Another difference between the two states is that the decision to innovate in Illinois, though largely driven by HAVA, was also impacted by a changing internal factor: partisan control. The year 2003 brought unified party control to the governor's office and both state houses. The failure of reform in 2001 and 2002 was in part the result of partisan wrangling; each party was fearful that the other party would try to win ownership of the issue and also use it to tilt election structures toward the other party's preferences. In contrast, Arizona innovated despite having divided party control.


This article demonstrates several things. First, traditional (internal) explanations of innovation were helpful in explaining the initial failure by Arizona and Illinois to innovate but were relatively unhelpful in explaining subsequent decisions to innovate. That is, the classic literature is better at helping us understand why some states do not innovate on the front of an innovation wave based on endogenous factors, but it is less helpful in providing explanations for why late states eventually do innovate. The apparent but not yet fully developed answer is that exogenous factors matter more in these states and/or endogenous factors change (as is the case in Illinois with unified government and as in Arizona with leadership). Recalling the S-shape taken by policy diffusion, we observe a cascading, positive feedback process. A few leading states innovate, and the idea then catches on like wildfire until the enthusiasm fades with a handful of "holdout" states that do not engage in the mimicry of interactive positive feedback.

In the case of election methods reform, the internal factors of the Palazzolo framework explain early innovation. Many of the lead states had favorable governing circumstances. For example, the early innovating states of Maryland, Florida, and Georgia had unified party control of the elected branches (governor and legislature) and bold leaders who criticized the status quo. In contrast, the late-innovating states of Arizona, Illinois, and New York had divided party control and lacked a critical mass of leaders early on who were willing to address reform. These factors explain in a fairly deductive manner why the former states innovated and the latter did not, but they are less helpful in explaining subsequent innovation by the initially gridlocked states.

For that, one must turn inductively to simply observe why they did innovate. In the case of election reform, that observation led to the power of HAVA (an external factor) to precipitate innovation. In this respect, the findings of this study are consistent with recent work by Rich, Volden, and Grossback and associates, which emphasized external determinants. (27) However, although recent literature points toward critical external determinants, it examines such effects across all states, rather than separating out the unique late-to-innovate states, as was done in this study.

In conclusion, the innovation literature, while making recent strides to incorporate external explanations, needs to develop more theory about late-innovating states (or other innovating organizations, such as diffusion across countries). The governments that do not innovate on the front or steep slope of the S-curve deserve more analysis. Too often, scholars study the innovation and generalize about the other states without careful consideration of the unique factors of the end-of-the-innovation distribution. In the case of election reform, these late-to-innovate states were unique in the way they went about innovating.

AUTHOR'S NOTE: I wish to thank Justin LeBeau for his excellent research assistance. The Carl Albert Center at the University of Oklahoma provided support in the form of a research fellowship for Justin during these endeavors.

Glen S. Krutz

University of Oklahoma

(1) An undervote is sometimes called an unrecorded vote because it is defined as the case where a voter has cast a ballot but no vote is shown for the office in question (in this case the president). There is a working assumption employed that persons showing up to vote in a presidential election intend to vote for a presidential candidate.

(2) Susan A. McManus, "Goodbye Chads, Butterfly Ballots, Overvotes, and Recount Ruckuses! Election Reform in Florida, 2001 to 2003," Election Reform: Politics and Policy, ed. Daniel J. Palazzolo and James W. Ceaser (Lanham, MD: Lexington Books, 2005), pp. 37-58.

(3) Daniel J. Palazzolo, "Election Reform After the 2000 Election," Election Reform, ed. Palazzolo and Ceaser, p. 4.

(4) See, e.g., Jack L. Walker, "The Diffusion of Innovations among the American States," American Political Science Review 63 (September 1969): 1186-1191; Virginia Gray, "Innovation in the States: A Diffusion Study," American Political Science Review 67 (December 1973): 1174-1185.

(5) Frances Stokes Berry and William D. Berry, "State Lottery Adoptions as Policy Innovations: An Event History Analysis," American Political Science Review 84 (June 1990): 395.

(6) Ibid.; Gray, "Innovation in the States"; Walker, "The Diffusion of Innovations"; Bradley C. Canon and Lawrence Baron, "Patterns of Adoption of Tort Law Innovations," American Political Science Review 75 (September 1981): 975-987; Patricia K. Freeman, "Interstate Communication among State Legislators regarding Energy Policy Innovation," Publius: The Journal of Federalism 15 (Fall 1985): 99-111; Henry R. Glick, The Right to Die: Policy Innovation and Its Consequences (New York: Columbia University Press, 1992); Fred W. Grupp Jr. and Alan R. Richards, "Variations in Elite Perceptions of American States as Referents for Public Policy Making," American Political Science Review 69 (September 1975): 850-858; James L. Regens, "State Policy Responses to the Energy Issue," Social Science Quarterly 61 (March 1980): 44-57; Lee Sigelman and Roland Smith, "Consumer Legislation in the American States: An Attempt at Explanation," Social Science Quarterly 61 (March 1980): 58-70; Lee Sigehnan, Phillip W. Roeder, and Carol Sigelman, "Social Service Innovation in the American States," Social Science Quarterly 62 (September 1981): 503-515.

(7) Regens, "State Policy Responses."

(8) Canon and Baum, "Patterns of Adoption"; Gray, "Innovation in the States"; Grupp and Richards, "Variations in Elite Perceptions"; Walker, "The Diffusion of Innovations."

(9), "Assorted Rolls: Statewide Voter Registration Databases under HAVA," 17 July 2005;

(10) Note that the cumulative number of states in the graph does not reach fifty, even with future projections. This is due to the fact that a handful of states have idiosyncratic registration processes that make a statewide database unworkable. For example, stone states have same-day registration.

(11) Glick, The Right to Die, p. 41.

(12) Ibid.

(13) Ibid., 202.

(14) See also Michael Mintrom, "Policy Entrepreneurs and the Diffusion of Innovation," American Journal of Political Science 41 (July 1997): 738-770; Michael Mintrom and Sandra Vergari, "Policy Networks and Innovation Diffusion: The Case of Education Reforms," Journal of Politics 60 (February 1998): 126-148.

(15) Palazzolo, "Election Reform After the 2000 Election," pp. 7-14.

(16) Michael J. Rich, Federal Policymaking and the Poor: National Goals, Local Choicies, and Distributional Outcomes (Princeton, NJ: Princeton University Press, 1993); Craig Volden, "The Politics of Competitive Federalism: A Race to the Bottom in Welfare Benefits?" American Journal of Political Science 46 (April 2002): 352-363.

(17) Lawrence J. Grossback, Sean Nicholson-Crotty, and David A. M. Peterson, "Ideology and Learning in Policy Diffusion," American Politics Research 32 (September 2004): 521-545.

(18) Palazzolo, "Election Reform After the 2000 Election," p. 4.

(19) Glen S. Krutz, "Arizona: Concerted Effort, Gridlock, and Then Breakthrough," Election Reform, ed. Palazzolo and Ceaser, pp. 177-189.

(20) Charles Stewart III, "Residual Vote in the 2004 Election," Caltech/MIT Voting Technology Project (Cambridge, MA: Massachusetts Institute of Technology, 2005).

(21) Betsey Bayless, White Paper on Arizona Election Law, Office of the Secretary of State for the State of Arizona, 7 December 2000, p. 2.

(22) David Kimball, "Illinois: Ending the Gridlock," Election Reform, ed. Palazzolo and Ceaser, pp. 190-204.

(23) In contrast, Arizona elects a secretary of state, who is the designated leader for elections in the state.

(24) Beth DeFalco, "'Experts Praise Experimental Voting Equipment Tested in Arizona," Associated Press, 23 September 2004;

(25) David R. Belman, Arizona Politics and Government: The Quest for Autonomy, Democracy, and Development (Lincoln: University of Nebraska Press, 1998), pp. 194-197.

(26) Ananda Shorey, "Arizona's Proposition 200: Immigration Restrictions Analyzed," Associated Press, 4 November 2004;

(27) Rich, Federal Policymaking and the Poor, Grossback et al., "Ideology and Learning in Policy Diffusion"; Volden, "The Politics of Competitive Federalism."
Table 1
Selected explanatory factors in
Arizona and Illinois, 2001-2002

Factor               Arizona   Illinois

Leadership?          No        No
Groups for change?   Mixed     No
Party control        Divided   Divided
Close elections?     Mixed     Yes

Table 2
Selected explanatory factors in Arizona
and Illinois, 2003-2004

Factor               Arizona   Illinois

Leadership?          Yes       No
Groups for change?   Mixed     Mixed
Party control        Divided   Unified
Close elections?     Mixed     Yes
COPYRIGHT 2005 Oxford University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Help America Vote Act of 2002
Author:Krutz, Glen S.
Geographic Code:1U8AZ
Date:Sep 22, 2005
Previous Article:Rational and pluralistic approaches to HAVA implementation: the cases of Georgia and California.
Next Article:The implementation of HAVA in New York: from antiques to high tech.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters