Voting Restriction Politics in Maine

Voting Restriction Politics in Maine Amy Fried University of Maine Emily Shaw Thomas College Abstract In recent years, many states passed laws restricting voting. Maine followed suit in 2011 but, unlike those states, the law—which ended election-day registration—was reversed at the ballot box. This paper explains what happened in Maine by pointing to a series of cultural and political factors, some of which interacted. Key elements were Maine’s civic culture and pride in high levels of civic participation, the existence of the referendum option, the creation of an effective coalition and organizational structure to restore the practice, messaging choices, and extensive publicity about unproven fraud allegations. Supporters of the law were less wellfunded and organized and received their strongest support in areas of the state where Governor LePage had received his highest proportions of the vote in the 2010 election. Future analyses of activities involving state-level voting laws should take account of political culture, organizational efforts, and political dynamics. New England Journal of Political Science Voting Restriction Politics in Maine Maine has long had high levels of voter participation and typically joins Wisconsin and Minnesota in the top three states for presidential election turnout. One reason that Maine voters have voted so much is that the registration process is easy and accessible. Since 1973, Maine people 1 could show up at their polling place on election day and register to vote there and then; many thousands have done so over the years, 50,000 in 2008 and 20,000 in 2010. But Maine was not immune from attempts to make voting more difficult that intensified after the Republican sweep of the 2010 elections. As in some other states, Maine voting law was changed after Republicans took power in January 2011. Unlike them, Maine’s new law, which would have ended election-day voter registration, never went into effect. Maine voters overturned it in a November 2011 referendum by a sixty-forty percent landslide in which each and every state legislative district voted to restore election-day registration. What happened in Maine? How and why was election-day registration restored? We suggest that a set of cultural and political explanations best explains why Maine people voted overwhelmingly in favor of reinstating election-day registration. Moreover, culture and politics interacted, as political figures and strategists took account of long-standing elements of Maine’s political culture and built campaigns and generated messages they believed would best resonate with the state’s citizenry. Structurally, the very presence of a referendum option made the campaign possible and indicates that the Progressive movement touched Maine and generated this mechanism of direct democracy. How did political culture matter? The adoption and use of Maine’s election laws is rooted in the state’s political values and patterns of political life, as voting practices are rooted in 1 Residents of Maine usually describe themselves as “Maine people” or “Mainers.” 294 Volume VI, Number 2 cultural norms such as political engagement and pragmatism. At the same time, Republican rhetoric pointing to students and other outsiders as capable of influencing elections was consistent with another strain in Maine culture that separates long-time state residents from what Mainers call “people from away.” Political factors linked to broader Maine political contexts and strategic decisions also played a role in restoring election-day registration (EDR). Voters and progressive activists responded to the Republican electoral victories of 2010 and its policy agenda of 2011. Political organization and strategy mattered, with pro-EDR forces more successful in using existing groups and developing organizational structures. At the same time, numerous strategic and rhetorical choices by opponents of election-day registration, both from within and outside of the state, undermined their effectiveness. While this paper analyzes one state, it also contributes to knowledge about the politics of voting laws, political culture, and political organization and strategy. Our analysis relies on a mix of methodologies, including an examination of campaign finance reports and voting patterns, interviews of people involved in the campaigns, and exploratory discussions with citizens. Before this, we discuss literature on the impact of registration laws on turnout and consider critical dynamics of Maine politics and political culture. How Local Voting Laws Matter The United States is unusual among modern democracies in having so many of its voting laws administered by regional subunits, such as states and municipalities. However national legislation, such as the National Voter Registration Act of 1993 (also known as Motor Voter) and the Help America Vote Act (2002) has had some impact on assisting citizens to vote and introducing greater access to voter registration across states (Ewald 2009, Rhine 1996). 295 New England Journal of Political Science Research has found that the state-level variations in registration practices affect overall turnout as well as on who votes, and thus the composition of the electorate. The strongest effects are on young voters and voters who have moved recently, with some impact related to socioeconomic status and level of education (Ansolabehere and Konisky 2006; Brown, Jackson, and Wright 1999; Highton 1997; Knack and White 2000; Mitchell and Wlezien 1995; Nagler 1991; Rosenstone and Wolfinger 1978). While studies vary, most find that EDR is linked to 3-5% higher turnout, although some studies find twice that effect. Moreover, voter registration rules are influenced by and influence political culture, contexts and outcomes. Political culture affects turnout both directly and indirectly, influencing citizens and the voting laws that enable or restrict access to the vote (King 1994). The composition of the electorate, affected by registration rules, affects the restrictiveness of welfare eligibility rules (Avery and Peffley 2005). States with liberal party control have higher levels of voter registration (Brown, Jackson, and Wright 1999). Following the 2010 elections, nearly every state legislative effort to change voting laws in ways that made voting more difficult came from Republicans, and nearly every effort to resist such changes came from Democrats (Weiser and Norden 2011). Throughout the nation, state legislators introduced and passed laws creating tight regulations on voter registration drives, shorter periods of early voting, requirements for voter photo identification, and an end to election-day registration. In Maine, the newly-elected Republican state legislature passed a bill ending EDR, which was then signed into law by the new Republican governor. This sparked a referendum that put the change on hold until election day and then overturned the law, enabling EDR to continue. 296 Volume VI, Number 2 Institutional Context: Maine’s Initiative and Referendum Provisions On November 8, 2011, Maine voters had an opportunity to weigh in on election-day registration. That Maine voters could vote directly on Maine’s 2011 law ending election-day registration was relatively unusual. Most states in which voting reform legislation was passed do not have provisions for popular referenda on legislative-made law (described in Maine statute as “the people’s veto.”) Restrictive voting legislation passed in two states with the popular referendum (Ohio 2 and Maine) was suspended pending the results of referenda, meaning that only one state out of the 22 states with referenda enacted restrictive voting legislation in 2011. Nearly half of the states without the popular referendum did so. Scholarly opinion on the utility of direct democracy provisions like the referendum and the citizens’ initiative has been mixed. On the positive side, the ability to call a vote to overturn a law provides a critical opportunity for voters who oppose specific laws but not entire partisan or legislative agendas. This is a useful way for voters to interact with representatives, as the referendum helps keep representatives faithful to the wishes of their constituencies and enables voters to convey their preferences to representatives (Matsusaka 2004). Arising from a movement that disliked political party machines and created primaries for selecting party nominees, “Progressive era reforms at ballot reform and direct democracy instruments like initiatives, recall, and referenda reflected and reinforced norms friendlier to conceptions of direct democracy” (Fried and Harris 2010, 339–40). On the other hand, direct democracy has been identified with processes that create laws hostile to minority interests (Gamble 1997). For example, recent studies have shown that states with direct democracy provisions are more likely to enact same-sex marriage bans (Lewis 2011). Direct democracy has also been criticized on the 2 Ohio was to hold a referendum in November 2012 on a law that limited early and absentee voting and moved primary voting, but most of the law was changed by legislation and no referendum went forward. Litigation continues on the elimination of voting on the three days before the election for all but those in military service. 297 New England Journal of Political Science grounds that fiscal limitations placed through citizen initiatives can have unexpected adverse effects on state functioning (Martell and Teske 2007, Smith 1998). For an issue like election reform, direct democracy provides public benefits beyond the development of publicly popular law. Referenda do not simply change laws; they also provide an opportunity to learn how an electorate judges a specific issue. Popular referenda allow citizens to convey an extremely precise message to legislators about a policy direction in a way they are unlikely to convey in any other way, short of extensive and reliable polling. However, where a ballot measure campaign has not been mounted, it is unlikely that polling organizations will choose to do the expensive work of performing a survey where there is no immediate and obvious electoral consequence. Compared with states that heavily use the initiative and referendum, Maine does not employ these processes often. However, Maine’s people’s veto has been used five times in the past six years, a much greater frequency than in any time period since their inception in Maine. As has been true for states that have used referenda, the people’s veto is used to challenge laws on social, political-institutional and fiscal issues. 3 With Democratic majorities in Maine’s State House, people’s vetoes were organized and supported by conservative organizations to overturn the state’s same-sex marriage law, a tax to support the state-run insurance plan, and a major Democratic tax reform initiative. 4 After the 2010 election in which control of the legislature and state executive shifted to Republicans, the 2011 people’s veto was organized and supported by liberal groups. Nonetheless, in the case of all referendum campaigns, organizations mounted vigorous efforts that focused on attracting voters beyond partisan constituencies. 3 The list of popular referenda placed on the ballot in Maine is available at Maine.gov, “People’s Vetoes 1909.” http://www.maine.gov/legis/lawlib/peoplesveto.htm. 4 One people’s veto from this period—to overturn the state’s law prohibiting discrimination on the basis of sexual orientation—was unsuccessful. 298 Volume VI, Number 2 Maine Politics and Political Culture Maine’s political culture values civic life and the state is highly participatory. Using Elazar’s analytical framework for political culture (Elazar 1966), Maine is classified as a moralistic state, with orientations toward good government and care for the community (Harris and Fried 2010, Palmer et al. 2009, Potholm 1998). Moreover, Maine people tend to be pragmatic and independent and expect “good government” in the form of well-run town, city and state agencies. Negativity, personal attacks, and misleading statements can backfire, leading citizens to punish candidates and campaigns that engage in such practices. Maine has elected two independent governors (James Longley, 1975-1979, and Angus King, 1995-2003) and a number of its Republican U.S. Senators, such Margaret Chase Smith, William Cohen, Susan Collins, and Olympia Snowe, are known and lauded for a civil, bipartisan approach, and a tendency toward moderate policy views. With increasing polarization nationally both among elites and citizens, there may be a “certain Maine exceptionalism” (Harris and Fried 2010, 124). Political culture can shift, but as a state with a small population and low residential mobility, Maine’s cultural attributes are relatively well-established and stable. Many Maine people have long-standing relationships with others in their localities and in different areas of Maine, a situation which some believe has made politics more personal and less polarized than elsewhere. Economic development projects face the burden of a population wary of change and interested in protecting traditional land use patterns. Census data indicate the state is the least urban in the nation, has an old population, and has the lowest level of college degrees in New England. While a neighborly ethic predominates, this highly racially homogenous state has had its share of ethnic and racial divisions. In the 1920s, the Ku Klux Klan had a sizeable presence in 299 New England Journal of Political Science the state, with “about 20,000 members; [and] its main targets were Catholics and immigrants working in paper mills” (“Cities in Maine Rally Against the Klan” 1987). Racial tensions arose in the wake of the shut-downs of some paper mills in the 1980s and the arrival of substantial numbers of Somali refugees in Lewiston at the start of the twenty-first century (Cullen 2011, Huisman et al. 2011). State political divides imperfectly follow regional lines, with the Portland area in the south the most liberal and the most rural counties generally the most conservative. The Lewiston-Auburn area, west of Portland, includes many Catholics of Franco-American origin and tends to vote Democratic, but is a swing area that has been an important part of the coalitions that elected the independent governors (Potholm 1998). Maine is sometimes characterized as a highly Democratic state, since it has given its four electoral votes to Democratic candidates in every election from 1988-2008. However, given the state’s proportion of independent voters and candidates, this is too simple a picture. As of February 2012, “unenrolled” (independent) voters constitute a plurality (36.5%) of Maine’s registered “active” electorate. Of the remainder, Democrats enjoy an advantage: Democrats make up 32% and Republicans 28% of the state’s voter registrations. This Democratic advantage is particularly important in electoral choices with just two options. However, in elections for office this edge in registrations is not necessarily determinative since Maine’s statewide elections frequently feature one or more strong independent candidates. As demonstrated by the Republican victory in the 2010 gubernatorial election, where there are strong independent candidates, a more unified Republican electorate can prevail over a Democratic electorate which splits its vote between the partisan and independent candidates. Meanwhile, without the confounding factor of the three-candidate races, the current state-wide Democratic advantage is 300 Volume VI, Number 2 illustrated by the fact that in the last two-person statewide choice—the 2008 electoral decision between Barack Obama and John McCain—only Piscataquis county recorded a majority for John McCain. Based on this advantage, where a policy has a strong partisan association, as was the case with the 2011 electoral reform law, we should predict that where the election is essentially a forced choice between a Democratic and a Republican-favored policy the Democratic registration preference should lead to an outcome favoring the Democratic-favored policy. The total partisan transformation of Maine’s political institutions occurring as a result of the 2010 elections revealed the limits of Democratic influence within the state. Before 2011, the last time Republicans controlled both houses of the Maine Legislature was 1974 and the last time Republicans controlled both the legislature and the governorship was 1964. After winning the House in 1974, Democrats held it until 2011, but the Senate picture is more mixed. Democrats gained control of the state senate in the 1982 elections and held control until after the 2010 elections, with the exception of the 1994-96 period, when the Republicans had a majority, and the 2000-02 period, when there was an even division. The last Republican governor in Maine before its present office-holder, Paul LePage, was John "Jock" McKernan who held the office from 1987 to 1995. Maine Governor Paul LePage came into office eager to make change and, prompted by incidents during the campaign, with a reputation for, depending on one’s perspective, being plain-spoken or rude (“LePage’s Temperament Becomes Campaign Issue” 2010; Taylor 2010). Having won with 39% of the vote in a five-person field in which the runner-up received 37%, LePage and his party quickly rolled out an aggressive legislative agenda. At the same time, the governor sustained his combative style, in January 2011 telling an NAACP group that criticized him for not attending their event in honor of Martin Luther King they could “kiss my butt,” and 301 New England Journal of Political Science in March 2011 removing a labor history mural (Greenhouse 2011; “LePage to NAACP: ‘Kiss my butt’—David Catanese” 2011). By early April, LePage’s style had caused upset within his own party, which includes a larger share of moderates than the national Republican Party. Eight Senate Republicans wrote him a letter rejecting “government by disrespect” and saying “By demeaning others, the governor also discourages people from taking part in debating the issues of the day—worrying if not only their ideas, but they themselves as people, will be the subject of scorn,” which itself caused dismay from LePage’s Tea Party supporters (Bell 2011b; Russell 2011d). 5 Governor LePage was to keep a low profile during the legislative debate regarding voting rules and through the referendum campaign, but his repute may have affected how people saw voting issues and the Republican Party. The Campaigns: Election Day Registration in Maine Election Day registration was established in Maine in 1973, in a noncontroversial, bipartisan vote of the state legislature. Before 2011, Maine Republicans occasionally tried to change voting laws, motivated by concerns with college students voting where they attended school but the issue came to the fore after the election of Republican Governor Paul LePage and the turnover of the Maine Legislature from Democratic to Republican control. In June 2011, the Maine Legislature passed and the governor signed LD1376; the vote was nearly along party lines. During the same legislative session a second bill on voting, LD 199, which required voters present photo identification approved by the Secretary of State, was tabled until 2012. Some key aspects of the referendum campaign had their roots in what was said and done before the bill’s passage. Those in favor of repealing registration on Election Day made arguments they would repeat through the campaign. Proponents of election-day registration 5 See also Fried and Melcher (2012). 302 Volume VI, Number 2 began to join forces and based an element of their strategy on the position taken by Maine’s town clerks. The bill, LD 1376, "An Act to Preserve the Integrity of the Voter Registration and Election Process," limited absentee balloting and eliminated election-day registration. During its consideration by the Maine Legislature, Secretary of State Charles Summers testified that the bill grew out his concern with the “integrity of the system” and “the stability of our election process” (Summers 2011). While Summers merely implied that fraud could be a problem, Charlie Webster, Chair of the Maine Republican Party was anything but subtle. Webster said, "If you want to get really honest, this is about how the Democrats have managed to steal elections from Maine people. . . Many of us believe that the Democrats intentionally steal elections." Speaking about college students and others, Webster said, “Buses. They bring them in in buses . . . Job Corps people—they move 'em around to wherever they have a tough seat and they want to win an election”(Nemitz 2011). 6 Webster’s statements caused an uproar (Bell 2011a); it would not be the last involving him nor surrounding both busses 7 and fraud. Secretary Summers also pointed to town clerks’ problems handling absentee ballots right before the election and noted, “they’ve identified this trend as one of their biggest concerns” as they lacked resources “to simply bring in more staff to handle the final days of an election.” LD 1376, said Summers, would not inhibit the ability to vote. However, others—including the town clerks—were concerned with the bill’s impact on turnout. Town clerks supported changing the absentee-ballot deadline but not the bill, for as the head of their state association said, "Same-day 6 A Republican with direct knowledge of the matter told one of us that Mr. Webster strongly pushed for this issue to be addressed by the Maine Legislature because Webster was surprised by some Republican losses of legislative races in the vicinity of the University of Maine-Farmington. 7 Referendum opponents made claims about fraud and potential for fraud throughout the campaign. One twist involved Mr. Webster stating that in November 2010 College Republicans at the University of Maine at Farmington had reserved and then parked college vans in order to prevent Democrats from using them to transport voters to the polls (Thistle 2011). 303 New England Journal of Political Science registration wasn't the issue for us. Ultimately, we want to facilitate voting—and it does that" (Nemitz 2011). Others testifying against the bill, such as leaders of the Maine League of Women Voters and the Maine Civil Liberties Union, focused on election-day registration, and did not dispute shifts in the absentee ballot window. Protect Maine Votes, the coalition that worked on restoring election-day registration, included but was not limited to groups whose leaders had testified against its repeal, and those with members who were mobilized to contact legislators. Originally composed of eighteen groups, others joined over the course of the campaign, to a total of twenty-three. Seeing the bill was likely to pass, group leaders decided to mount a people’s veto campaign. As one leader noted, “There’s a floating coalition of groups that comes together on some issues” and it was able to make the decision quickly.” 8 The day Governor LePage signed the bill, Protect Maine Votes announced their plans, with the Maine League of Women Voters and Maine Civil Liberties Union as coalition leaders and its public face. (See Appendix A for the full list of members.] Because the Maine Town and City Clerks Association had testified against ending election-day registration, but supported limiting early voting, the coalition decided the referendum would only involve the former issue, enabling them to point to the clerks’ support. The first step was to gather at least 57,277 signatures from registered voters, a task made difficult by the lack of a coming election, since often signatures are gathered at the polls. Moreover, the coalition wanted the vote to take place during the usual Election Day in November, leaving only about three weeks to gather signatures. Republican Party Chair Charlie Webster 8 Confidential interview by Amy Fried. March 2012. 304 Volume VI, Number 2 suggested that it was unlikely that the requisite number of signatures could be gathered in time (Miller 2011). With an effort involving over 1,500 volunteers and staff from coalition members, in just twenty-three days over 70,000 signatures were turned in and certified. 9 Some groups were deeply involved in signature gathering, while others brought knowledge about the correct process and organized the petitions by town so they could be quickly certified. Although Protect Maine Votes had not yet conducted a single poll, the rapidity with which this was done and the large contingent of volunteers suggested a reservoir of public support. In response, Republican leaders cast doubt on Mainers’ interest, with “Republican House Speaker Bob Nutting describ[ing] the Republican House Speaker Bob Nutting the people's veto supporters as ‘extreme left-wing groups and individuals’ who are backing an initiative that will be rejected by Maine voters” and Webster saying ‘I don't think there's any ground swell among regular people for this’” (“Election Day Registration Supporters Deliver Signatures for People’s Veto Campaign” 2011) However, as one referendum leader stated, this was an “incredible logistical feat” that carried a political message: “if you can do it once, you can do it again.” 10 In terms of campaign mechanics, the effort built a list of committed volunteers and voters to contact for the election. Coalition partners coordinated through a coalition board. The coalition raised money, held press conferences, responded to opponents, advertised, and engaged in mobilization and get-out-the-vote activities. Groups devoted staff time to the effort. Protect Maine Votes outraised 9 Indirectly referencing Republican Chair Charlie Webster and others, a memo from the Protect Maine Votes Coalition stated, “Many people familiar with the referendum process believed that this result would not be possible through a grassroots effort and voting rights opponents publicly declared that they believed we would fail to gather the necessary signatures in the allotted time.” (Chin 2011) 10 Confidential interview by Amy Fried. 305 New England Journal of Political Science its opponents and controlled how resources were used to organize and message. Groups had different skills, foci, activities and membership bases. They cooperated, in part through a division of labor. Staff from Engage Maine, a group focused on technical assistance, strategy and collaboration for progressive organizations were involved in communication and leadership. Maine People’s Alliance, Opportunity Maine, Equality Maine, and unions were involved in signature gathering and/or mobilization and turnout activities. As one leader told said, “Everyone did what they do.” 11 The core message of the Protect Maine Votes campaign was: Same day voter registration has worked for more than thirty years, has helped Maine to have a high level of voting by making it easier to vote. Taking it away would make it harder for people to vote and there is no reason to change. 12 This message appealed to pride in the state’s civic engagement and pragmatism, values in Maine’s political culture. The message strategy was to consistently convey those messages and to explain that the referendum “protects voting rights and restores same day registration.” As many People’s Veto questions had confusing wording, the group stressed that those who wanted Maine to have Election Day registration need to vote “yes.” Its television ad, funded by money raised locally and nationally and which used Maine people, delivered that message. 13 In addition, Protect Maine responded to their opponents’ messages about possible and potential fraud. Although the legislative vote was partisan and the coalition well-aware of the national push to undermine access to the vote, they decided to avoid those messages and to focus on Maine and their core message. While op-ed writers and others involved in the public dialogue 11 Confidential interview by Amy Fried. As articulated to Amy Fried in a confidential interview. 13 This ad can be seen at http://protectmainevotes.com/tv-ad. 12 306 Volume VI, Number 2 referred to “disenfranchisement” and partisan aspects of voting laws and practices, the coalition did not. The group was also thoughtful about timing its message. For example, when the legislature held a special session, Protect Maine Votes released information about elected officials who opposed Election Day registration but had personally used it themselves, while stressing that everyone should be able to do so and asking them to correct their error and restore the practice. Protect Maine Votes communications director David Farmer said, “We’re asking these members to reconsider the elimination of same-day registration and the move of the registration deadline. Same-day registration has worked for nearly 40 years, and it has worked for many of the people who voted to kill it. We would like for those lawmakers to consider their own circumstances and work to restore same-day registration during today’s special session” (Russell 2011c). The main opposition groups were the Maine Republican Party and a conservative advocacy group called the Maine Heritage Policy Center (MHPC). These groups were central to the messaging and provided a portion of the financing. MHPC worked closely with Secretary Summers to broadcast the perspective that Maine’s same-day registration system was inherently flawed, sponsoring two luncheons in September in which Summers was the main speaker. During these luncheons, Summers argued that the state was facing an epidemic of errors in the voter registration database, in particular that there was an “84 percent error rate in the registrations that were reviewed” (Metzler 2011); these claims were repeated weeks later in an MHPC report (Maine Heritage Policy Center 2011), despite the fact that Summers’ initial claim was that 84 percent of the problems he found were due to clerical error (Summers 2011). Certainly at the core of Summers’ effort was to demonstrate that fraud was a problem in Maine’s election system, although he could not identify that fraud. Nonetheless, several high- 307 New England Journal of Political Science profile campaign incidents involved claims regarding voter fraud and potential fraud. Charlie Webster, the Chair of the Maine GOP, was a prominent figure early on, with his charges that Democrats brought “busloads” of people to polling sites to register and vote. In late July 2011, Webster asked Secretary of State Summers to investigate students paying out of state tuition but voting in Maine. Webster also alleged that the Democratic Party and others involved in the referendum were “special interests,” explaining, “The difference is we [Republicans] represent regular people, how’s that? We represent working class people, people who drive a truck. We don’t represent the far left of Maine” (Russell 2011a). Through the course of this issue campaign, Webster and other Republican leaders were to repeat the idea that only a small group of Maine people cared about restoring Election Day registration. The next day, Summers stood held a press conference and to discuss “potential identification fraud and a cover-up within the Bureau of Motor Vehicles. That investigation, he said, would be folded into a probe of Webster's vague allegations against college students.” While there was no proof presented or any evident tie to election-day registration, one reporter contended, “[I]t was still a blow to the veto coalition. By simply tossing out the words "fraud" and "voters," Summers and Webster had commanded the news cycle for a week” (Mistler 2011) Nonetheless, some reporters also clearly conveyed the sense that Summers was a strongly biased investigator on a fishing expedition. As one television report noted of Summers’ news conference, “Secretary of State Charlie Summers says just because he isn’t forwarding any voter fraud cases to the attorney general’s office as a result of THIS investigation doesn’t mean there isn’t a problem with Maine’s election registration system” (Cornish 2011). In early September 2011 Webster also claimed fraud involving voters living in a hotel; they turned out to be medical students displaced by a hurricane (Russell 2011b). 308 Volume VI, Number 2 The Secretary of State’s late September 2011 report on students indicated no fraud, but Summers sent out letters asking students who voted in Maine whose cars were registered out of state to remove their names from rolls (Tice 2011) With no actual fraud, efforts to activate fear of outsiders was undermined. As one advocate for restoring election-day registration put it, fraud claims “could have done a lot more damage to the People’s Veto.” Since they came out “proven wrong,” they “provided inoculation to the last minute attacks.” Moreover, while the media initially treated such claims without much skepticism, when, “after a while they didn’t have facts to back things up, media coverage turned.” Another said, while Webster had promised that once people “learned the truth,” they would support his position, “In the end, it turned people in the other direction.” 14 Toward the end of the campaign, in mid-October, the anti-election day registration coalition Secure Our Votes announced its formation. 15 No coalition members were identified, its materials echoed the MHPC, and campaign finance reports filed after the election showed that $250,000 of its $313,000 in contributions came from the Michigan based American Justice Partnership, a group with ties to Charles and David Koch. In the run-up to the election, this group developed several pieces of advertising which were noted state-wide. The first was a television advertisement that ran in the last weeks of October; a radio version was also broadcast. 16 Interestingly, the ad didn’t mention same-day voter registration. Instead, it featured music conveying a sense of urgency while a graphic cartoon showed New York and New England states encroaching on Maine; a voice asks, “Who should decide Maine’s elections? Mainers, or outsiders from other states? Outside interests want 14 Interviews by Amy Fried One advocate for restoring election day registration contended that the debunking of charges from Webster and Summers delegitimized them to such an extent that they needed to create another campaign vehicle and spokesperson (Interview by Amy Fried). 16 See http://www.youtube.com/watch?feature=player_embedded&v=QcOwaVeKYtY 15 309 New England Journal of Political Science to get rid of Maine’s election ethics law.” By capitalizing on Maine’s close-knit political culture and emphasizing the need for Maine’s political autonomy, the message probably was not only intended to resonate with Maine’s general sense of political independence but also to recall the recent battles over the influence of “outside interests” in the 2009 people’s veto of the new Maine law authorizing same-sex marriage. However, the ad was treated skeptically because no “ethics law” was at issue, the language to refer to outsiders was different from what Mainers use, and the leader of Secure Our Votes could not answer questions about who created and paid for it (Carrigan 2011). The ads further muddled election-day registration opponents’ shifting message, which had included everything from saying fraud was self-evident to saying they would support allowing people to register on election day if Maine could only someday develop a better and more robust voter registry. For these reasons and also Mainers’ dislike of negative ads, the publicity it received likely had negative consequences for its cause. The relationship between the 2011 same-day registration people’s veto campaign and the 2009 people’s veto of same-sex marriage was drawn again more strongly by the second major piece of advertising on the “No” side to attract statewide attention. Rather than a widelydisseminated broadcast ad, however, this ad was in print and narrowly targeted. Running in more than 25 free weekly papers in rural locations around the state, the ad was headlined, “FACT!” and went on to link “EqualityMaine (advocacy group for gay/lesbian marriage)” to the people’s veto, asking “Why is this special interest group so interested in repealing Maine’s election laws?” Although no sponsor was listed, the Maine Republican Party later acknowledged running the ad. 17 Charlie Webster, who called the ad “micro-targeting,” argued of the party’s belated ownership of the ad, “it wasn’t some sinister attempt [to deny sponsorship of the ad], we just 17 Image of the ad available at http://thinkprogress.org/lgbt/2011/11/08/363649/gop-defends-maines-gay-baiting-adon-same-day-voter-registration/ 310 Volume VI, Number 2 didn’t do it because it wasn’t required.” Shenna Bellows of the Maine ACLU noted that the ad “appeared to be ‘scapegoating a particular person or group of people.’” (Mistler 2011). In general, public messaging efforts by groups opposing the people’s veto were controversial, appeared factually and ethically questionable, and received substantial press criticism. Secure Maine Votes, the organization apparently intended to be the center of opposition to the people’s veto, never attained political significance. The putative head of the group could not answer questions about the ads and it appeared to be an organization in name only. The head of the Maine Republican Party and the Secretary of State remained outspoken advocates for the opposition throughout the campaign but their advocacy was compromised by perceptions of unethical practices, a perception that was enhanced by the Maine Ethics Commission’s levying of a fine on Secure Maine Votes for their failure to file campaign expenditure statements in a timely way (Russell 2011). Press coverage of opponents to the restoration of Election Day registration became increasingly skeptical over the course of the campaign. Protect Maine Votes raised more money than Secure Maine’s Ballot—$815,000 to $312,000—and the groups spent their money in different ways. Both bought advertising, but efforts to activate and mobilize voters were especially strong from proponents of election-day registration. As with the signature-gathering necessary to put the referendum on the ballot, an extensive field operation was mounted. The final campaign finance report from Protect Maine Votes, the coalition to restore election-day registration, included 1500 hours in paid staff time from coalition partners, with at least two-thirds of the time dedicated to volunteer recruitment, phone banks and get out the vote activities. Of the approximately $32,000 in in-kind expenses included in this report, 87% went to staff. In their last campaign finance report, Secure Maine’s 311 New England Journal of Political Science Ballot reported no in-kind contributions, but spent about $10,500 on recorded and automated telephone calls. 18 Analyzing the Vote on Question One: Looking at Statewide Results Statewide, Question 1 passed with approximately 60% of the vote (237,024 (59.9%) voting yes; 155,156 (39.2%) voting no; 3,627 (0.9%) ballots with no vote on this question). The level of positive response to this ballot measure could be seen as somewhat surprising: The measure had been strongly opposed by the Maine Republican Party, which had made its historic electoral gains in just the previous year’s elections. On the other hand, the state’s electorate had never had an opportunity to vote on election reform legislation before, so there was no track record to draw on in order to predict a state response to this particular question. The vote margin was substantial enough that Question 1 passed in every Maine county. Evaluating the state-level response to 2011’s Question 1 in the context of other state-wide elections allows us to characterize this particular vote as demonstrating a median amount of statewide consensus on the outcome. To achieve a 60% outcome, the winning side needed not only to retain Democratic support but also to capture most of Maine’s substantial independent “middle,” as well as to appeal to at least some Republican voters. However, the fact that the winning side did not get more than 60% suggests that the decision on how to vote still retained some partisan elements. This observation is further borne out by the fact that county and municipal partisan registration generally predicted variation in the degree of local approval. Votes on people’s vetoes in Maine show that the degree of consensus on popular referenda can vary a good deal. The distribution of winning votes shows a slightly bimodal trend, with a greater number of very close votes, in which the winning side gained less than 52% of the 18 Although not identified as an in-kind contribution, the Maine Heritage Policy Center listed $5100 in costs for “campaign workers salaries,” with funds going to MHPC staff. 312 Volume VI, Number 2 vote, and a greater number of votes with high consensus, in which the winning side gained between 68 and 70% of the vote. People’s vetoes have sometimes received overwhelming statelevel support, like the extremely popular veto of a 1931 law to increase the gasoline tax that garnered 89% of the vote. Some vote margins on people’s vetoes are extremely close, suggesting a highly contentious issue; among the closest was the failed 1965 effort to overturn a law permitting the Sunday sale of liquor in Maine. 2011 Question 1’s result of a 60/40 electoral margin placed it in the lower middle of the range of winning margins for state-level popular referenda since the advent of Maine’s “People’s Veto” process (Figure 1). The winning side won slightly less than the median winning margin of 62.3%. A result that was neither among Maine’s closest, nor among Maine’s most overwhelming, the 2011 people’s veto demonstrated a solid amount of agreement—an interesting result that demands further investigation. Figure 1. 2011 Question 1 Win in the Context of Winning People’s Veto Percentages, 19092011 6 50% 5 52% 54% 4 56% 58% 3 60% 2 62% 64% 1 66% 68% 0 50%52%54%56%58%60%62%64%66%68%70%72%74%76%78%80%82%84%86%88% Maine’s high proportion of officially non-partisan, independent (“unenrolled”) voters makes it challenging to perform a traditional analysis on the relationship between a vote and 313 New England Journal of Political Science local partisan makeup. Nonetheless, in some places, the level of Democratic-to-Republican registrations helps to explain variation in the vote on 2011’s Question 1. For example, the counties with the five highest proportions of Republican voter registrations all voted against the referendum at levels higher than the statewide average (though still less than 50%). Further supporting the concept that a core of Republican voters voted with their party on Question 1, it appears that county-level votes for Paul LePage in the 2010 gubernatorial elections also strongly predict levels of opposition to the election-day registration referendum. County-level vote for Paul LePage in 2010 and county-level opposition to the election-day registration referendum correlated quite strongly and demonstrated a strong positive linear relationship (Figure 2). Figure 2. No on Question 1 by Support for Paul LePage 50.0% Voted "No" on 2011 Q1 45.0% 40.0% 35.0% 30.0% 25.0% 25.00% 30.00% 35.00% 40.00% 45.00% 50.00% 55.00% County-level % Vote for Paul LePage in 2010 Examining the relationship between support for LePage and opposition to the electionday registration referendum provides initial support for the hypothesis that self-identified Republicans opposed the return of election-day registration—in other words, supported the additional restrictions on voter registration which were created by the Republican legislative 314 Volume VI, Number 2 majorities. However, there does not appear to be quite as strong a relationship between opposition to the 2011 referendum and support for John McCain in the 2008 presidential elections, which we might have expected to be a stronger predictor of the vote. Similar to a “yes” or “no” vote on the referendum, the choice between Barack Obama and John McCain in 2008 did not have a strong third alternative. Nonetheless, the county-level vote for McCain performed more poorly as a predictor of county-level opposition to the same-day registration referendum than did the LePage vote. To better specify the relationships between several likely variables and municipal-level opposition to the referendum (which is the same as support for the legislature’s restrictive electoral reform), we performed a series of OLS regressions. Our dependent variable was the percent of 2011 referendum voters opposing the same-day registration referendum. To examine the relationship between different aspects of Republican identification and opposition to the referendum, we included variables for the percentage of voters within a municipality supporting LePage in 2010, the percentage of voters within a municipality supporting McCain in 2008, and the percent of voters within a municipality who registered as Republican. To look at the relationship with other recent referenda in Maine—issues which similarly required voters to choose between only two options, yet which are not strictly related to partisan preference—we included a variable for the percentage of voters within a municipality supporting 2009’s referendum to overturn Maine’s same-sex marriage law. Finally, considering that much of the rhetoric from the “No on 1” campaign focused on the danger from “outsiders” —including references to students and immigrants—we considered that municipalities in which there was either a higher-than-average proportion of college-aged 315 New England Journal of Political Science residents or a higher-than-average proportion of non-white residents 19 might be expected to more strongly oppose the referendum. This hypothesis rested on the reasoning that non-immigrant and non-student voters living in those municipalities who were inclined to be convinced by those arguments might feel particularly concerned about the impact of immigrant or student voting on local races. In order to evaluate the effects of greater-than-usual racial and ethnic diversity, we created a dummy variable to identify municipalities in which 5 percent or more of town residents identified as something other than solely white in the 2010 US Census. In order to evaluate the effects of college-aged residents, we created a dummy variable to identify municipalities in which at least 15 percent of the residents were between the ages of 20 and 29, according to the 2010 US Census. 20 Table 1 displays a series of bivariate and multivariate models which consider the effect of these variables on 2011 referendum opposition. Of all of the variables we examined, only the variable to measure the effects of high diversity was never significant; we did not include it in the reported models. As we found in examining votes at the county level, the level of support for LePage was highly predictive of a municipality’s level of opposition to the 2011 referendum; in looking at the LePage vote alone, each additional percentage point for LePage in a municipality predicted an additional 0.44 percentage point opposing same-day voter registration. The bivariate model evaluating the effect of votes for McCain was also significant and the effect substantial, although less so than looking at LePage votes alone. Rates of Republican registration in a municipality demonstrated a strong effect, but with larger variation, making it a poorer predictor of the 2011 vote. Finally, college towns had a substantively small but statistically significant 19 Maine has the second-highest proportion of white residents out of all US states. The last decade has witnessed a slight increase in Maine’s levels of ethnic and racial diversity due mainly to the state’s positive stance towards refugee resettlement. While overall numbers of people of color are still low in Maine, there has been a substantial change in the state’s percentages of non-white residents because of its low starting point. 20 This method successfully captured all of the state’s college towns. 316 Volume VI, Number 2 negative effect in the bivariate model (in other words, collectively demonstrated greater-thanaverage support for election-day registration). Table 1 Variables highlighted in bold are significant at the .001 level or above. Intercept LePage % Model 1 .225 (.012) .440 (.028) McCain % % Reg. Republican Yes on 1, 2009 % Above 15% people aged 20-29 LePage x Reg. Repub McCain x Reg. Repub Adjusted R2 Model 2 .230 (.013) Model 3 .266 (.013) Model 4 .254 (.015) Model 5 .414 (.004) Model 6 .202 (.013) .350 (.034) Model 7 .178 (.015) .257 (.047) Model 8 .312 (.034) .081 (.085) Model 9 .283 (.035) .008 (.088) .394 (.027) .472 (.040) .198 (.045) .258 (.024) .231 (.046) .090 (.030) -.179 (.118) -.129 (.118) .081 (.030) .892 (.257) .845 (.256) Model 10 .194 (.035) .124 (.081) .205 (.120) .131 (.031) -.052 (.019) .33 .30 .22 .20 .02 .35 .36 .37 .38 .113 (.219) .34 Using the observation that votes for LePage strongly predicted a higher percentage of opposition, we considered multivariate combinations which would improve our model. Adding both the municipal proportion of Republican voters improved our explanatory power, as did adding the level of municipal support for the 2009 referendum overturning same-sex marriage. Meanwhile, adding an interaction effect between the LePage vote and the percent of registered Republican voters added still more to our model’s explanatory power. The significance of the interaction effect suggests that the municipalities most likely to strongly oppose the 2011 referendum were locations that had both high levels of registered Republicans and high levels of support for LePage—and far more so than places that supported LePage but did not have high 317 New England Journal of Political Science proportions of registered Republican voters (like Lewiston) or which were heavily Republican but did not strongly support LePage. Adding the additional variable—municipal-level support for the 2009 referendum—created another small improvement in the model’s fit, though at the expense of its parsimony. Taken together, the factors of support for LePage, Republican identity, and opposition to same-sex marriage suggest a voting bloc that is consonant with the preferences of the Tea Party movement. This combination did a far better job of predicting the vote than did a simple analysis of Republican registration. Indeed, when looking at the final model on Table 1 we can observe that the interaction effect between Republican registration and support for McCain was not significant, suggesting that the “No” campaign’s message was in fact much less powerful among more moderate Republicans, or those who supported McCain more strongly than LePage, than it was among Tea Party Republicans, who supported LePage more strongly than they did McCain. Given the fact that opposition to the 2011 was strongest among people who we could identify as belonging, in fact, to only one subset of the Republican party, it is quite evident that the “Yes” campaign was ultimately more successful in reaching out to a broader array of voters. Though many voters beyond this subset of the Republican party supported the 2009 referendum to end same-sex marriage, the “No” campaign was not successful in swaying these voters to oppose the 2011 referendum, despite the effort to link the two referenda in their “micro-targeted” ad campaign in the rural free weeklies. However, how was it that voters outside of this subset understood the issues at stake in the 2011 referendum? A closer look at voters’ own expression of their voting rationales will help to answer that question. 318 Volume VI, Number 2 Analyzing the Vote on Question One: A Qualitative Assessment of Message Effectiveness A second kind of way of analyzing the vote concerns examining how people explain their reasoning for voting the way that they did. By understanding voters’ rationales, it would be possible to understand how successful the advocates on both sides of the people’s veto had been in persuading voters, particularly those in Maine’s substantial political middle. We went to polling places in Oakland and Sidney, two locations that allowed us to oversample Republican voters, in order to conduct qualitative exit polling that would allow voters to express their main reason for voting the way they did on the people’s veto. 21 Our initial expectation was that voters would express rationales that were consistent with the two main political “frames,” or rhetorical arguments, that dominated the media coverage of voting reform in the run-up to the election. One major frame frequently featured in media stories was the “voter fraud” frame, a perspective that was also heavily emphasized by opponents of the 2011 referendum. The “voter fraud” frame, used heavily by the opponents of the people’s veto, revolves around the idea that existing laws are insufficient to prevent people from voting illegally. The simple repetition of the word “fraud” is an effective rhetorical strategy in that the phrase “voting fraud” succinctly evokes the image of a criminal, intentional, coordinated act performed with the intention of changing the otherwise valid outcome of an election. While voting fraud (or election fraud) can have a number of meanings, including behavior by election officials to prevent valid voters from 21 It was important to seek out locations that had high ratios of Republicans to Democrats because of the nature of the campaign. There was a great deal more positivity associated with the Yes on 1 campaign than the No on 1 campaign; it seemed reasonable to expect that opponents of Question 1 would be more reluctant to share their voting rationale since it asked them to cast doubt on the legality of their fellow voters. First, a large number of people refused to be interviewed. Second, more voters who volunteered that they voted “No” would not give their reasons for doing so than was the case for people reporting a “Yes” vote. This provides some anecdotal support to our initial assumption. 319 New England Journal of Political Science voting (Donsanto 2008), the mention of voter fraud in in 2011 nearly always referred to fraud perpetuated by people who cast illegal ballots by misrepresenting themselves as legitimate voters. The strength of the voter fraud frame is that it aims beyond a partisan base, casting the legislation as a matter of nonpartisan common sense, a matter of common safety and security for the civic body. While individuals using the “voter fraud” frame may acknowledge that the new legislation will make voting harder for legitimate voters, this difficulty is cast as an acceptably small individual sacrifice on behalf of a great public good, the few extra steps people will have to take in order to vote help to save the civic body from corruption. Thus adherents of this frame implicitly morally elevate the people who are willing to make this sacrifice on behalf of the country and implicitly (or explicitly) cast moral aspersions on people who are not “willing” to make this sacrifice. However, the fraud frame can backfire if there is no actual fraud to demonstrate. Secretary Summers had a number of opportunities to demonstrate voter fraud but he was unable to find instances of what was claimed to be a serious problem; this undermined the “No” campaign’s credibility. Furthermore, when Republican advocates present versions of the voter fraud frame which emphasize partisanship, the argument loses its bipartisan appeal. Arguments that Democrats steal votes (e.g., “Many of us believe that the Democrats intentionally steal elections” (Nemitz 2011)) project criminality onto a mainstream political party, limiting the framer’s credibility and undermines efforts to convince audiences that voter fraud is an authentic problem. The other major frame featured in media stories was the “voter disenfranchisement” or “voter suppression” frame. This frame, which occupied a prominent role in the national debate about voting changes, points to how the legislation will prevent people from voting who legally have done so. Usually referencing past episodes of voter suppression, the “disenfranchisement” frame describes the illegitimate use of state power to remove voting rights from legitimate voters, 320 Volume VI, Number 2 particularly from marginalized or disempowered social groups. 22 Like the voter fraud frame, the disenfranchisement frame has a partisan component; unlike the voter fraud frame, this partisan component is highlighted rather than de-emphasized. The frame connects voting legislation with Republican efforts to reduce the size of the Democratic electorate by pointing to the fact that nearly all of the legislation was proposed by Republicans and opposed by Democrats, by highlighting the cross-state coordinated nature of the legislation and by noting the common membership of many voting reform legislation sponsors in the American Legislative Exchange Council (ALEC). Where individuals using the voting fraud frame focus on the individual-level “common sense” of the voting reform proposals, individuals using the disenfranchisement frame focus on the partisan impact of the voting reform legislation on categories of individuals who are more likely to vote for Democrats. Although some commentary and letters-to-the-editor in Maine used the disenfranchisement frame, it was not adopted by the coalition working to restore election-day registration. Given the strong partisan element that is generally present in the voter disenfranchisement frame, this was probably a decision that was made with an eye to persuading independent and Republican voters who might like election-day registration but not otherwise see themselves as allied with Democratic voters. We performed brief interviews with 94 voters leaving the polls in Oakland and Sidney. 23 Voters were asked, with open-ended questions, to share their votes and the main reason why they chose to vote the way they did. Of the group interviewed, 59 reported voting “Yes” and 35 reported voting “No,” a proportion of support-to-opposition that was slightly higher than the actual vote percentages in those locations (56% support in Oakland and 59% support in Sidney.) A majority of these “No” voters were male (N=23, 66%) while a majority of the “Yes” voters were female (N=36, 64%). Among respondents who reported their partisan identification, none reported voting against 22 In other states, advocates employing the disenfranchisement frame made frequent reference to the mass disenfranchisement of African-Americans during under “Jim Crow” laws. This was a less prominent element of the debate in Maine because of the state’s low number of African American residents. 23 Confidential interviews by Emily Shaw. 321 New England Journal of Political Science his or her expected party preference (Democrats supporting the people’s veto and Republicans opposing it); while those who reported being registered as independent supported the people’s veto. Over half of the respondents identifying their age were over 60, while only three respondents were under 30. (The timing of the interviews, conducted between mid-day and 5:00pm, undoubtedly affected the distribution of respondent age.) Responses from people who reported voting “No” can be grouped into five main categories. 24 One of the most popular responses (N=12) was indeed the fraud frame, arguing that ending same-day registration was important for ending or preventing fraud. An equally popular response (N=12) was that people had plenty of time to register without election-day registration. Secondary frames offered by some who answered this way suggested that people who felt that there was plenty of time to register also viewed voting as a privilege, and asking for more effort from voters was appropriate. Several respondents described voting as a privilege and implied that extra effort on the part of voters was warranted to merit that privilege. The next largest groups (N=4, N=5) reported that they voted “no” but didn’t want to give a rationale and that they voted “no” to “keep it the way it was.” Among those who voted to “keep it the way it was,” around half were confused by the wording of the referendum. The rest of their answer revealed that they believed they were supporting the reinstatement of election-day registration. The other half was more ambiguous and may have been voting to keep the new law the way it was, in line with the framing of the “No on 1” advertisement. Responses from people who reported voting “Yes” could be broken into four main categories. By a very large margin, the most popular rationale expressed by “yes” voters was that the existing system works well (N=35). Four people independently used the word “broken” or “broke” in their response, suggesting they may have adopted the Protect Maine Votes “ain’t broke, don’t fix it” advertising language. Similarly to “no” voters arguing that “people have plenty of time,” the “Yes” voters’ argument in favor of keeping election-day registration may have hidden additional rationales. 24 A prototypical example of each response category is available in Appendix B. 322 Volume VI, Number 2 Some voters who said that there was no reason to change it reinforced that message with the secondmost popular message. The second-largest group of people (N=15) used a civic participation frame, stating that it was good for voting to be convenient – either as a good in itself or because they valued Maine’s high levels of voter participation. Four respondents stated that they supported the people’s veto because of personal experience—that they had themselves used same-day registration in the past. Only three respondents made an explicitly partisan reference in their response, including just one person who used the word “disenfranchisement.” While it was not surprising to discover that opponents of the people’s veto made no partisan references, given that the success of the “fraud” frame depends on the appearance of party-neutral “common sense,” it was striking that few supporters of the people’s veto explicitly emphasized partisan messaging. The “disenfranchisement” frame was not commonly employed by these respondents, but the explicit argument on behalf of the status quo (“ain’t broke, don’t fix it”), supported by an argument that broader civic participation is a generally good thing, seems to have been a resonant public message. In their emphasis on promoting civic participation, “Yes” voters exemplified Elazar’s description of moralistic culture. “Yes” voters’ rhetoric was consistent with campaign messaging, which stressed pragmatic rationales and Maine’s civic engagement. Meanwhile, voters opposing the people’s veto appeared generally to be driven by concern about fraud, whether actual or potential. The people that respondents were concerned would vote fraudulently ranged from “terrorists,” to “students from away,” to people who would steal names from “graveyards” with no agreed-upon perpetrator of fraud. A secondary response suggested that voting was a privilege and people should put forth more effort in order to merit that privilege. Rather than emphasizing broader participation as a moral value, this group felt that voters’ willingness to make small sacrifices to vote demonstrated their moral legitimacy as voters and that participation should not be extended beyond this group. 323 New England Journal of Political Science Concluding Comments The final act—at least so far—involving Maine voting law came more than four months after the referendum, when the Maine Legislature declined to require voters to present photographic identification. LD199 was amended to strip its core provision and to instead ask the Secretary of State to study the voting system and advise legislators; the denuded bill passed overwhelmingly, indicating the landslide win of the People’s Veto’s political impact. Maine’s pushback of an agenda to restrict voters’ access to the polls grew out of cultural and political phenomena. As a state with strong civic and participatory rules and values and a propensity to support traditional practices, election-day registration was in synch with the state. The coalition in favor of the practice had a nonpartisan message that spoke to Mainers’ pride about its civic tradition and to pragmatic concerns about “hardworking” people having to register one time and vote another, affecting people with multiple jobs and those who live in small towns with town offices open few and unusual hours. While Maine’s culture includes concerns about outsiders, this message did not resonate widely. In part this may have been since the advertisements from opponents of election-day registration sounded like they were, as Mainers say, generated by people “from away.” Political impacts ranged from the structural, with the referendum option providing the opportunity for repeal, to aspects of political organization and messaging. The effort to restore election-day registration benefited from successful coalition formation that started early and included groups like the League of Women voters that have a history of nonpartisan activity. This made it harder to paint the effort as partisan. With its focus on restoring Election Day registration, Protect Maine Votes was on the same page as the town clerks; this enabled them to counter those who said that it was too difficult to administer. Progressive groups had existing organizational infrastructures for linking organizations and an infrastructure for community 324 Volume VI, Number 2 organizing. Their messaging was focused and frequently repeated and came from a professional spokesperson who responded rapidly to the opposition. Fundraising enabled them to run television ads that used Maine people and spread the basic components of the message. In contrast, the campaign opposing the referendum was closely linked to the right flank of the state’s Republican Party. Despite its putatively nonpartisan message focused on fraud and potential fraud, statements by the Republican Party Chair regarding the purported positions of “regular people” and the practice’s supposedly inappropriate use by Democrats gave the referendum opposition a partisan stamp, as did the actions of the Republican Secretary of State and the party’s organizational associate, the Maine Heritage Policy Center. Governor LePage was quiet through the campaign, but he had signed the bill and did not have strong public support beyond his base. Without an organized coalition to communicate and mobilize over the course of the campaign, referendum opponents’ messaging was inconsistent and grassroots activity rather limited. Some political elements should be understood as linked to Maine’s political culture and practices. For instance, the Protect Maine Votes coalition’s decision to use a nonpartisan, pragmatic message reflected the numbers of independent voters along with the degree to which bipartisanship and political participation is prized in the state. Unproven intimations of voter fraud, coupled with student intimidation and deceptive advertising cut against the state’s traditions and hurt referendum opponents. Success in turning back this effort to restrict voting was based in Maine’s political culture and structures, along with its patterns of political and partisan support and the referendum supporters’ highly effective campaign operation. 325 New England Journal of Political Science Appendix A Members of the Protect Maine Coalition Engage Maine, Maine People’s Alliance, Maine Civil Liberties Union, Maine League of Conservation Voters, League of Women Voters of Maine, Maine League of Young Voters, Opportunity Maine, Maine Equal Justice Partners, EqualityMaine, Maine Women’s Lobby, Maine Education Association, MSEA-SEIU, Speaking Up for Us, Disability Rights Center, Preble Street Resource Center, Homeless Voices for Justice, Maine AFL-CIO, Maine Chapter of the Sierra Club, AARP Maine, Teamsters Local 340, NAACP of Maine, OneMaine, and Planned Parenthood of Northern New England. 326 Volume VI, Number 2 Appendix B Examples of Answers to the Question, “What was the Main Reason You Voted the Way You Did on Question 1?” Verbatim examples of popular rationales for voting “No” • • “Too much fraud.” “Because they're able to get out before election day—the new law allows them plenty of time.” • “Don't think it's too much to ask from someone who wants to vote.” Verbatim examples of popular rationales for voting “Yes” • • “It's not broken” “I don't see the problem with registering on the same day we're here to vote—I've used it before myself, I've moved a few times.” • • “Because I think it's disenfranchising to voters.” “Because people should be able to register anytime they want.” 327 New England Journal of Political Science Bibliography Ansolabehere, Stephen, and David M. Konisky. 2006. “The Introduction of Voter Registration and Its Effect on Turnout.” Political Analysis 14 (1): 83–100. Avery, James M., and Mark Peffley. 2005. “Voter Registration Requirements, Voter Turnout, and Welfare Eligibility Policy: Class Bias Matters.” State Politics & Policy Quarterly 5 (1): 47–67. Bell, Tom. 2011a. “Bill to End Same-Day Registration Approved.” Portland Press Herald 11 June. http://www.pressherald.com/news/bill-to-end-same-day-registrationapproved_2011-06-11.html (27 September 2012). ———. 2011b. “Pleas from within Maine GOP Stir up Tea Party Activists.” Portland Press Herald 5 April. http://www.pressherald.com/news/pleas-from-within-gop-rile-tea-partyactivists-_2011-0405.html?searchterm=Pleas+from+within+Maine+GOP+Stir+up+Tea+Party+Activists (27 September 2012). Brown, Robert D., Robert A. Jackson, and Gerald C. Wright. 1999. “Registration, Turnout, and State Party Systems.” Political Research Quarterly 52 (3): 463–79. Carrigan, Don. 2011. “Question 1 TV Ad Raises Campaign Controversy.” 2011. WLBZ2 News Center. http://www.wlbz2.com/news/article/178453/3/Question-1-TV-ad-raisescampaign-controversy (April 5, 2012). Chin, Ben. 2011. “Protect Maine Votes.” Scribd. http://www.scribd.com/miketipping/d/61862243-Protect-Maine-Votes-Signature-DetailMemo (April 8, 2012). 328 Volume VI, Number 2 “Cities in Maine Rally Against the Klan.” 1987. New York Times 27 September. http://www.nytimes.com/1987/09/27/us/cities-in-maine-rally-against-the-klan.html (March 31, 2012). Cullen, Andrew. 2011. “Struggle and Progress: 10 years of Somalis in Lewiston.” Sun Journal 18 December. http://www.sunjournal.com/news/city/2011/12/18/struggle-and-progess10-years-somalis-lewiston/1127846 (27 September 2012). Elazar, Daniel Judah. 1966. American Federalism: A View from the States. New York: Crowell. Higgins, A. J. 2011. “Election Day Registration Supporters Deliver Signatures for People’s Veto Campaign.” Maine Things Considered 8 August. http://www.mpbn.net/Home/tabid/36/ctl/ViewItem/mid/3478/ItemId/17530/Default.aspx (27 September 2012). Ewald, Alec C. 2009. The Way We Vote: The Local Dimension of American Suffrage. Nashville: Vanderbilt University Press. Fried, Amy, and Douglas B. Harris. 2010. “Governing with the Polls.” Historian 72 (2): 321–53. Gamble, Barbara. 1997. “Putting Civil Rights to a Popular Vote.” American Journal of Political Science 41 (January): 245-69. Greenhouse, Steven. 2011. “Gov. Paul LePage Takes Aim at Mural to Maine’s Workers.” The New York Times 23 March. http://www.nytimes.com/2011/03/24/us/24lepage.html (March 20, 2012). Harris, Douglas B., and Amy Fried. 2010. “Maine’s Political Warriors: Senators Snowe and Collins, Congressional Moderates in a Partisan Era.” New England Journal of Political Science 4 (1): 95–129. 329 New England Journal of Political Science Highton, Benjamin. 1997. “Easy Registration and Voter Turnout.” The Journal of Politics 59 (2): 565–75. Huisman, Kimberly A., Mazie Hough, Kristin M. Langellier, and Carol Nordstrom Toner, eds. 2011. Somalis in Maine: Crossing Cultural Currents. Berkeley, California: North Atlantic Books. King, James D. 1994. “Political Culture, Registration Laws, and Voter Turnout among the American States.” Publius 24(4): 115–127. Knack, Stephen, and James White. 2000. “Election-Day Registration and Turnout Inequality.” Political Behavior 22 (1): 29–44. Lewis, Daniel C. 2011. "Direct Democracy and Minority Rights: Same-Sex Marriage Bans in the U.S." Social Science Quarterly 92 (2): 364–83. Catanese, David. 2011. “LePage to NAACP: ‘Kiss my Butt.'” 2011. Politico 14 January. http://www.politico.com/blogs/davidcatanese/0111/LePage_to_NAACP_Kiss_my_butt.h tml (April 2, 2012). Sharon, Susan. 2010. “LePage’s Temperament Becomes Campaign Issue.” Maine Things Considered September 29. http://www.mpbn.net/Home/tabid/36/ctl/ViewItem/mid/3478/ItemId/13704/Default.aspx (27 September 2012). Martell, Christine, and Paul Teske. 2007. "Fiscal Management Implications of the TABOR Bind." Public Administration Review 67 (4): 673–87. Miller, Kevin. 2011. “Coalition to Wage People’s Veto Campaign to Revive Election Day Voter Registration Law.” Bangor Daily News 23 June. 330 Volume VI, Number 2 http://bangordailynews.com/2011/06/23/politics/coalition-to-wage-people%E2%80%99sveto-campaign-to-revive-election-day-voter-registration-law/ (27 September 2012). Mistler, Steve. 2011. “Election Day Registration Supporters Battle Opponents’ Stagecraft.” Sun Journal 1 August. http://www.sunjournal.com/state/story/1066495 (April 1, 2012). Mitchell, Glenn E., and Christopher Wlezien. 1995. “The Impact of Legal Constraints on Voter Registration, Turnout, and the Composition of the American Electorate.” Political Behavior 17 (2): 179–202. Nagler, Jonathan. 1991. “The Effect of Registration Laws and Education on U.S. Voter Turnout.” The American Political Science Review 85 (4): 1393–1405. Nemitz, Bill. 2011. “Ballot-Box Fraud? GOP Fixes Problem That Isn’t There.” Portland Press Herald 10 June. http://www.pressherald.com/news/Bill-Nemitz-Ballot-box-fraud-GOPfixes-problem-that-isnt-there_2011-06-10.html (27 September 2012). Palmer, Kenneth T., G. Thomas Taylor, Jean E. Lavigne, and Marcus A. LiBrizzi. 2009. Maine Politics and Government. 2nd ed. Lincoln, Nebraska: University of Nebraska Press. Potholm, Christian. 1998. An Insider’s Guide to Maine Politics. Aurora, Ontario: Madison Books. Rhine, Staci L. 1996. “An Analysis of the Impact of Registration Factors on Turnout in 1992.” Political Behavior 18 (2): 171–85. Rosenstone, Steven J., and Raymond E. Wolfinger. 1978. “The Effect of Registration Laws on Voter Turnout.” The American Political Science Review 72 (1): 22–45. Russell, Eric. 2011a. “GOP Chairman Says University Students behind Voter Fraud.” Bangor Daily News 25 July. http://bangordailynews.com/2011/07/25/politics/gop-chairman-saysuniversity-students-behind-voter-fraud/ (27 September 2012). 331 New England Journal of Political Science ———. 2011b. “GOP Chairman Targets Voters Who List Hotel Rooms as Residence.” The Bangor Daily News 9 September. http://bangordailynews.com/2011/09/09/politics/gopchairman-targets-voters-who-list-hotel-rooms-as-residence/ (April 5, 2012). ———. 2011c. “Legislators, LePage Registered to Vote on or Near Election Day, Group Says.” Bangor Daily News 27 September. http://bangordailynews.com/2011/09/27/politics/legislators-lepage-registered-to-vote-onor-the-day-before-election-day-group-says/ (27 September 2012). ———. 2011d. “Maine Senate Republicans Criticize LePage’s Comments, Actions in Upcoming OpEd.” Bangor Daily News 1 April. http://bangordailynews.com/2011/04/01/politics/senate-republicans-blast-lepagecomments-actions-in-op-ed/ (27 September 2012). Smith, Daniel A. 1998. Tax Crusaders and the Politics of Direct Democracy. New York: Psychology Press. Summers, Charles. 2011. “Testimony in Support of LD 1376: ‘An Act to Preserve the Integrity of the Voter Registration and Election Process.’” Maine.Gov 9 May. http://www.maine.gov/sos/news/2011/ld-1376-testimony.htm (27 September 2012). Taylor, Jessica. 2010. “LePage: I’d Tell Obama to ‘Go to Hell.’” Politico 29 September. http://www.politico.com/news/stories/0910/42886.html (April 3, 2012). Thistle, Scott. 2011. “Maine GOP Leader Says UMF College Republicans Parked Vans on Election Day.” Bangor Daily News 30 July. http://bangordailynews.com/2011/07/30/ politics/maine-gop-leader-says-umf-college-republicans-parked-vans-on-election-day/ (27 September 2012). 332 Volume VI, Number 2 Tice, Lindsay. 2011. “State to Students: Register Your Car in Maine or Don’t Vote Here.” Bangor Daily News 4 October. http://bangordailynews.com/2011/10/04/politics/state-tostudents-register-your-car-in-maine-or-dont-vote-here/ (April 5, 2012). Weiser, Wendy, and Lawrence Norden. 2011. Voting Law Changes in 2012. Brennan Center for Justice. http://brennan.3cdn.net/9c0a034a4b3c68a2af_9hm6bj6d0.pdf (27 September 2012). 333