<%BANNER%>

Efficacy of Early Voting Systems in the United States and Switzerland

HIDE
 Title Page
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Election reforms
 Swiss government and elections
 Data and hypotheses
 Results
 Discussion
 References
 Biographical sketch
 

PAGE 1

EFFICACY OF EARLY VOTING SYSTEMS IN THE UNITED STATES AND SWITZERLAND By JOHN STUART RABON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by John Stuart Rabon

PAGE 3

iii TABLE OF CONTENTS page LIST OF TABLES...............................................................................................................v LIST OF FIGURES...........................................................................................................vi ABSTRACT......................................................................................................................vii CHAPTER 1 INTRODUCTION........................................................................................................1 2 ELECTION REFORMS...............................................................................................7 Registration Reforms....................................................................................................7 Voting Reforms............................................................................................................8 In-Person Early Voting........................................................................................10 Postal Voting.......................................................................................................12 3 SWISS GOVERNMENT AND ELECTIONS...........................................................16 4 DATA AND HYPOTHESES.....................................................................................20 United States...............................................................................................................20 In-Person Early Voting........................................................................................21 Political Competition...........................................................................................22 Unemployment....................................................................................................22 National Macropartisan Strength.........................................................................23 Election-Day Registration...................................................................................24 United States Regressions...........................................................................................24 Switzerland.................................................................................................................24 Postal Voting.......................................................................................................26 Unemployment....................................................................................................26 Political Competition...........................................................................................26 Swiss Regressions.......................................................................................................27 5 RESULTS...................................................................................................................28 In-Person Early Voting...............................................................................................28 Postal Voting..............................................................................................................29

PAGE 4

iv 6 DISCUSSION.............................................................................................................32 LIST OF REFERENCES...................................................................................................36 BIOGRAPHICAL SKETCH.............................................................................................43

PAGE 5

v LIST OF TABLES Table page 1 Introduction of In-Person Early Voting.........................................................................11 2 Introduction of Simplified System of Early Voting.......................................................18 3 Effect of In-Person Early Voting in the States (VAP)...................................................28 4 Effect of In-Person Early Voting in the States...............................................................29 5 Effect of Postal Voting in the Cantons..........................................................................30 6 Effect of Postal Voting in the Cantons (Schaffhausen excluded)..................................30

PAGE 6

vi LIST OF FIGURES Figure page 1 International Voter Turnout.............................................................................................6 2 U.S. Voter Turnout..........................................................................................................6

PAGE 7

vii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts EFFICACY OF EARLY VOTING SYSTEMS IN THE UNITED STATES AND SWITZERLAND By John Stuart Rabon August 2006 Chair: Michael Martinez Major Department: Political Science Of all the industrial democracies, the two nations which exhibit the lowest levels of voter turnout, due to certain in stitutional factors, are the Un ited States and Switzerland. These two nations, partially as a result of low participation, have recently begun to implement policies meant to augment participa tion: in-person early voting in the United States and postal voting in Switzerland. The purpose of this thesis is to analyze the effectiveness of these policies on increasing turnout, utilizing OLS regressions. In the United States, I use the state-year interaction as the unit of analysis measuring the effect of in-person early voting acro ss six separate midterm electi ons. In Switzerland I conduct a similar analysis, using the canton-year inte raction as the unit of analysis over eight legislative elections. For both nations, I c ontrol for both political and economic factors within each state-year and canton-year intera ction, as well as fixed effect for districts (states in the U.S., cantons in Switzerland) a nd years. I find that in-person early voting does not work to increase turnout in the American states, but postal voting provides a

PAGE 8

viii significant boon to aggregate turnout levels in Switzerland. I also analyze positive and negative consequences of each system, and a ttempt to ease any fear s associated with voting by-mail.

PAGE 9

1 CHAPTER 1 INTRODUCTION The purpose of this thesis is to analyze policies that theoretically reduce the costs of voting to the individual and lead to an in crease in aggregate tur nout. According to the rational choice model, an individual will vote as long as the benefits of voting outweigh the costs (Downs 1957; Riker & Ordeshook 1968). In theory, the decision to vote is a marginal one, as both the costs (of information, as well as the actual ac t) and benefits of voting are low (Aldrich 1993; Bl ais 2000). Any slight change in the costs or benefits of voting could result in the altering of the voting decision, according to the theory of rational choice. I note that, since Downs, two separate models of participation have been developed: the resource (Brady et al. 1995) and mobilization (Rosenstone and Hansen 1993) models. Both models overlap the rati onal choice model somewhat, as they focus on how socioeconomic status and group memb ership diminish the costs of voting, respectively. For the purpose of this study the rational choi ce and resource models are the same, as I focus on reducing the costs of voting to citizens re gardless of economic class. The mobilization model is not applicable in this thesis either, as its creators note it is better at predicting when one participates rather than why (Rosenstone and Hansen 1993, p.20) Henceforth, I will use the rational choice model as the framework for the individual decision to vote. Since the seminal work of Wolfinger a nd Rosenstone (1980), analysts have believed education to be the most important de terminant of an individuals propensity to overcome the costs of voting and participate in the political sphere. One would naturally

PAGE 10

2 expect that the United States and Switzerland, two nations with literacy rates close to 100% (Directorate of Intelligence, CIA 2006a), to exhibit high rates in the easiest form of participation voting. But as Figure 1 demonstrat es, this is not the case. As a percentage of the voting-age population, voter turnout in all other original G-7 countries is significantly higher than turnout in the United States and Switzerland. (I also point out the fact that turnout in the U.S. as a percentage of the voting-age populat ion is declining, as evidenced by Figure 2.) Out of 58 parliamentary elections in those six countries from 1960 to 2000, only one (1994 Japan) displayed tu rnout lower than any U.S. midterm or Swiss parliamentary election. U.S. and Swiss voters are conspicuously averse of going to the polls, and present a conundrum to rese archers and analysts of voter turnout. In theorizing the determinants for tur nout rates across industrial democracies, including the U.S. and Switzerland, po litical scientists have posed an institutional model for voter turnout. In this model, researcher s propose that forces related to government structure play a greater role in shaping turnout rates than individual attributes. In a comparative study of seven nations touting the impact of institutions, Verba, Nie, and Kim (1978) assert aggregate turnout is de rived from the juxtaposition of individual attributes working against national institutio nal factors. Powell (1986) concludes that U.S. turnout is advantaged slightly by indivi dual political interest and education levels. However, institutional factors such as compulsory registration, nationally competitive election districts, and strong party-group linka ge result in lower rates of American turnout. (Party-group linkage reflects the consistency of cues given by overlapping groups such as labor unions, religious organi zations, etc. to parties to determine the strength and volume of political cues.) Jackman discovers unicameralism provides a

PAGE 11

3 clearer link between elections and legislation, increasing turnout (1987, p. 405). Not surprisingly, nations with compulsory vo ting experience higher levels of turnout. (Tingsten 1937; Franklin 1996; Jackman 1987; Franklin 1999; Hirczy de Mino 1994) The institutional model of turnout, however, does not fully account for outlying U.S. and Swiss turnout rates. In analyz ing cross-national differences in turnout, researchers often have to control for the two states through an electoral salience variable (Franklin 1996, 1999) or by du mmies, per se (Jackman and Miller 1995). Franklin (1996) goes as far as conducti ng an analysis withou t the two states. Franklin defines electoral salience as the linkage be tween legislative electoral outcomes and government complexion (1996, p .224). In the context of Europe, the most salient elections are those whose outco me determines the a llocation of government power (Franklin & Hirczy de Mino 1998, p.317). Switzerland, then, exhibits exceptionally low salience, as the composition of its executive will remain stable regardless of election results. The low elect oral salience displayed by the United States stems partially from divided government. (i.e ., different parties c ontrol the Presidency and Congress) Franklin and Hirczy de Mino th eorize that if one part y controls both the executive and legislative branches of gove rnment, it becomes easier for the American electorate to give blame or credit to the contro lling party. Turnout, therefore, will increase under unanimous control. The rese archers find that, controlling for changes in closeness of elections and el ectoral legislation, divided gove rnment does in fact depress turnout. For each election held under conditions of divided government, turnout declines, on average, about 1.96% (p.321). After three consecutive Presidenti al elections under divided government, turnout would be depres sed 5.88% (p.321). Franklin also mentions

PAGE 12

4 that low electoral salience is often correla ted with the possibility of voter fatigue. Because the United States and Switzerland have more elections than any other democracy in the world, it is likely voter fatigue plays a role in depressing turnout. Both the U.S. and Swiss, as well as a ll governments elected by a small proportion of citizens, share cause for concern, as the negative effects produced by low levels of voter turnout are numerous. Researchers a nd political pundits have long been troubled by low turnout, and much of the literature associat ed with voter turnout has been devoted to analyzing the effects of minimal participation. As a result of this research, experts have developed a number of reasons for finding met hods to improve turnout. For one, political participation is stratified in terms of socioeconomic status (Verba & Nie 1972; Wolfinger & Rosenstone 1980; Brady et al. 1995) and de cline in voter turnout can lead to an exacerbation of the discrepancy between the belie fs and preferences of active participants and the entire population (Rosenstone & Hansen 1993; Burnham 1980). (I note that research by Leighley and Nagler (1992) disagrees with this assessment.) As a result of this skew, discord often appears between public policy and pub lic opinion as the politically active exhibit greater influence over the legislative and policy-making process (Key 1949; Burnham 1987). Specifi cally, research has shown that states with low turnout among the poor exhibit significantly lower we lfare spending (Hill, Leighley, and HintonAndersson 1995). Lijphart c ites especially poor turnout in less salient but no less unimportant elections (1997, p.1) as a problem as well, due to Verba and Nies assertion that participation is at the heart of democra tic theory and at the heart of the democratic political formula in the United States (1972, p.3). Classic literatur e also points to the benefits to the individual of voting. The voti ng act could provide positive externalities to

PAGE 13

5 the individual, such as lear ning civic virtues like knowle dge and responsibility (Mill 1958). Lawmakers in both the United States and Switzerland, partially as a result of low turnout and its consequences, have recently enacted policies to simplify the process of voting. The purpose of this study is to anal yze the effect of two of these reforms on turnout: in-person ea rly voting (IPEV) in the Unite d States, and postal voting in Switzerland. Unfortunately, a limitation of this study is the differences in the institutional and political structures of the tw o nations do not allow a true test of which system better works to increase turnout. Most notably, the existence of elec tions by proportional representation leads to a multiparty system in Switzerland. Other factors, such as differences in the composition of the executive, will be discussed in further detail later.

PAGE 14

6 0.3 0.4 0.5 0.6 0.7 0.8 0.9 119 60 1963 19 66 1969 1972 1 975 1978 1981 1 984 1987 1 990 1 993 199 6 1 999 USA Swiss UK France Germany Japan Italy Canada t Figure 1 International Voter Turnout ( VAP) in Parliamentar y Elections Data provided by the International Institute for Democracy and Electoral Institute Voter Turnout Website (International In stitute for Democracy and Electoral Assistance 2006) 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.651960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 Presidential Midterm Linear Trend Figure 2 U.S. Voter Turnout (VAP) Data provided by the United States Election Project (McDonald 2006)

PAGE 15

7 CHAPTER 2 ELECTION REFORMS In the United States voting is a two-step process; in order for a citizen to cast a ballot one must be registered. While the anal yses presented here focus on the actual act of voting, the importance of registration wa rrants mentioning, as registration reforms have been the main vehicle state legislatures have employed to increase levels of turnout. Registration Reforms The idea that registration laws form a ba rrier to the act of voting permeates the literature (Wolfinger & Rosenstone 1980; S quire et al. 1987; Highton 1997, 2004; Piven & Cloward 1988; Rosenstone & Wolfinger 1978). Brown et al. conclude group registration level is the primary determinan t of group turnout in both presidential and midterm elections (1999, p.474). Burnham belie ves that the abolition and replacement [of registration laws] by automatic state-enrollme nt procedures is the first step to solving the dilemma of low turnout (1980, p.68). Also of note, states have hi storically utilized registration laws to depr ess turnout (Highton 2004). As a result, in 1993 Congress passed the National Voter Registration Act (Motor Voter) mandating the availability of voter registration via ma il and in various government agencies. Research conducted on policies mandated by Motor Voter provides differing results, (Knack 1995; Highton & Wolfi nger 1998; Martinez & Hill 1999; Brown & Wedeking 2006) but no experts have concluded that the policies dramatically boost turnout. Since the implementation of Motor Voter, turnout has remained remarkably stagnant in the face of burgeoning regist ration rates (Neeley & Richardson 2001).

PAGE 16

8 States have also begun to implement Electi on Day registration (EDR) in an effort to assuage the hindrance of registering, and si nce 1990 the number of states with EDR has nearly doubled (Fitzgerald 2005). The signifi cance of EDR to increasing turnout is constant. Highton (1997) concludes that EDR am plifies turnout levels ten points, while Brians and Grofman (2001) predict a seven percent boost. In their classic work, Wolfinger and Rosenstone conc lude election-day registration is the most effective legal change to registration procedures, deduc ing turnout would increase by about 6.1 percentage points (1980, p.78) Fitzgerald (2005) finds a lesser but still significant impact an augmentation of more than one and three points in presidential and congressional electi ons, respectively. Voting Reforms The crux of this study, however, is the eff ect on turnout of post-registration reforms dealing with the act of voting. While I focu s on state-level and cant onal-level reforms, I note a recent modification in American voting law at the federal level: the Help America Vote Act (HAVA). Created in 2002 as a respon se to the havoc caused by the Presidential Election two years prior, the bi ll calls for the disbursement of funds to states with the purpose of improving the administration of el ections. Specifically, the act mandates the elimination of punch-card ballots and lever m achines in order to modernize and ease the voting process. It also calls for further training of election offi cials and compulsory institution of a provisional voting procedure, allowing thos e who believe they are missing from the registration rolls by mistake to cast a ballot and dispute the error. At the state level, legislat ures have begun to institute reforms outside of electionday registration. Increasingly, state policymaker s are turning to early voting in an effort to augment turnout. Reasons exist for the employment of early voting outside of the

PAGE 17

9 general benefits of increasi ng turnout (Gronke et al. 2005). First, early voting systems provide a more convenient method to vote, es pecially for those who work for an hourly wage, have long commutes, or have heavy tim e constraints on a November Tuesday (p. 3). Research has shown that ballot box accessi bility is can have a significant effect on election participation (Gimpel & Schuknecht 2003; Haspel & Knotts 2005). Second, advocates of early voting claim the procedure would augment the effectiveness of democratic decision-making (Davis 2005). Th ird, Gronke et al. (2005) theorize citizens like early voting because campaigns can focus their mobilization and persuasion efforts on those yet to particip ate, leaving partisan voters the oppor tunity to continue their daily lives free of campaign activity. Theoretically, early voting increases tur nout through providing greater availability for a registered voter to cast a ballot. This increases the probability of individual voting, especially for those unlikely to vote on Election Day. However, early voting may also increase turnout through a social cont ext, as nontraditional voting sites may communicate cues to voters about candidate choices (Stein & Garcia-Monet 1997). As a result, strategic politicians may be able to exploit early voting for their own electoral advantage through efficient mobilization techniques. Three methods currently exist for casting a ballot before Elect ion Day: in-person early voting (IPEV), voting-bymail or postal voting, and abse ntee balloting. Many states have recently begun to liberalize absentee ba lloting, as twenty stat es allow no-excuse absentee balloting and California permits vot ers to be placed on a permanent absentee voter list (Gronke 2005). In the 1996 Presiden tial election, twenty percent of votes from

PAGE 18

10 those states came from absentee ballots (Kar p & Banducci 2001). The analyses presented here, however, will focus only on in-per son early voting and voting-by-mail. In-Person Early Voting In-person early voting grants registered vot ers the ability to vote at a designated location in the days leading up to the electi on. Texas became the first state to institute IPEV in 1991, and twelve other legislatures have adopted statewide in-person early voting since (Election Reform Informati on Project 2006; along with various state elections officials). (Table 1) Two other st ates, California and Kansas, have in-person early voting procedures but only in select counties. The process of in-person early voting varies across the states a nd counties. For the most part, state policies allow approximately fifteen days for prospective in-person early voters to cast a ballot. Arizona policy allows the most days, as in-person early voting begins 33 days before Election Day and ends the Friday prior. Georgia voters, on the other hand, only have the five business days the week before the election to cast an early ballot. Unlike the number of days states permit IPEV, the locati on of voting sites is determined by the counties. Most counties ma ke available in-pers on early voting only at the county elections office. For more popul ous counties however, such as Miami-Dade in Florida and Johnson County in Kansas, in-person early voting occurs at traditional locales such as shopping malls, airports and libraries (Sola 2006; Newby 2006). The empirical evidence supporting the positiv e effect of in-person early voting on turnout is slim. Previous research on IPEV focuses as much on its effect on the composition of the electorate as its impact on turnout. In a study of in-person early voters in 1994, no party advant age existed between the two groups, and education levels were similar (Stein 1998). Interestingly, Stein does not expect early voting to mobilize a

PAGE 19

11 Table 1 Introduction of In-Person Early Voting STATE YEAR OF INTRODUCTION Texas 1991 Colorado 1992 Arizona 1993 Nevada 1993 New Mexico 1994 Tennessee 1994 Arkansas 1996 Kansas* 1996 California* 1998 North Dakota 2003 Florida 2004 Georgia 2004 Illinois 2006 Georgia 2006 West Virginia 2006 In-Person Early Voting is not statewide significant number of new voters (p.68). In an analysis of Oregon and Florida, Gronke and Galanes-Rosenbaum contradict Stein by conc luding early voters are likely to be older and more educated than election day partic ipants, and African-Americans, on average, tend to refrain from early voting (2005). Gr onke and Stein both agree that early voters are much more partisan and ideological than precinct voters, conf irming the idea that partisans decide on a candidate well befo re political moderates (Campbell et al. 1960; Berelson et al. 1954). In a study of the 1994 elections in Tennessee, Richardson and Neeley (1996) note in-person early voting po ssesses the possibility to increase turnout, but focus their analysis on the effect of different ballot types and voting locations. A study of the 1992 Presidential election reveals that in-p erson early voting slightly increased turnout in Texas counties (Stein & Garcia-Monet 1997). Analysis shows that counties with a higher proportion of the vot e cast early entaile d marginally, but significantly higher participati on rates. In the election, a one percent incr ease in early

PAGE 20

12 votes as a proportion of the total vote led to an increase of voter turnout by sevenhundredths (0.07) of one percent (p.665). The researchers also conclude that in-person early voting effectually brings vot ers with a history of electoral participation to the polls, rather than new voters, as Stein and Garcia -Monet actually find a negative relationship between change in voter regist ration and turnout (p.665). It is interesting to note that the authors discovered that the number of early vot ing sites at non-trad itional locations was positively and significantly correlated with turnout. For every ten nontraditional voting locations in a county, turnout increased fift een-hundredths (0.15) of one percent (p.665). Fitzgerald (2005) concludes in-person early voting has a ne gligible effect on increasing turnout. Using the state-year inte raction as the unit of analysis, the author evaluates the impact of va rious voting reforms during general elections from 1972 to 2002. Controlling for political but not economic variables, Fitzgerald finds in-person early voting, while insignificant, is negativ ely correlated with turnout. As it is highly unlikely that early voting drives citizens from the polls, the author not es the possibility of self-selection, admitting alternativ e voting methods are most likely to exist in states that experience high rates of voter turnou t regardless of the reforms (p.856). Postal Voting The expansion of postal voting across the country has been much more protracted than in-person early voting. In 1977, the county of Monterey, California, was the first to exclusively use voting-by-mail for anything ot her than a very small, special district election (Magleby 1987). Since, the state of Oregon has come to the forefront in postal voting, and in 2004 conducted a general, stat ewide election solely through VBM after first experimentation in 1981. Oregonians have had the opportunity to vote-by-mail in special statewide elections since 1993, a nd in 1998 voters overwhelmingly passed an

PAGE 21

13 initiative expanding all-mail voting to primary and general elections (Oregon Secretary of State 2006). Research on postal voting, un like in-person early voting, has consistently shown a significant, positive effect on turnout. In a st udy of all-mail ballots of elections in seven states, Hamilton (1988) finds increased turnout in every analysis. He also cites reasons to implement voting-by-mail outside of general early voting the incr eased integrity of elections. The utilization of all-mail balloting provides election administrators as nearly foolproof a way as can be de vised to purge registration ro lls of non-eligible voters (p.864). As registration rolls become more accurate, the likelihood of the deceased casting a vote declines precipitously. Magleby (1987) focuses on the effect of voting-by-mail on participation in seven local, special elections. Of the seven, onl y one did not produce higher turnout than a typical citywide, polling place election for th e given locale. In San Diego in 1981, the voters participated at a highe r rate (61%) than the preceding Presidential primary (58%) (p.82). Magleby also utilizes an OLS regression of polling-place and vote-by-mail elections in five locations from 1980 to 1984 (N=43). He concludes converting from polling-place elections to all-mail balloting wi ll increase participation nineteen points, controlling for locale, year, and the type of election. He cauti ons however, that the effect of all-mail balloting in a general election w ill be significantly reduced, due to the lower baseline of turnout in refe rendum elections. Magleby also cites the possibility of a novelty effect, and admits the effect of vote-by-mail elections on turnout may decrease over time. Not surprisingly, he finds the bes t predictor of participation in mail ballot elections is education (p.89).

PAGE 22

14 In January 1996, the state of Oregon carried out the first statewide election solely in vote-by-mail format to select a replacemen t for Senator Bob Packwood. In a survey conducted after the election, Southwe ll and Burchett (1997, 1998) find voters overwhelmingly prefer voting-by -mail to casting a ballot at a polling place. Of those who favored voting-by-mail, the most comm on reasons cited were convenience, more time to read and complete the ballot, and th e ability to avoid possible weather problems and work responsibilities. The researcher s find distinct demographic distinctions between vote-by-mail and traditional polling-place voters, but mitigate the criticism that voting-by-mail provides a partis an advantage as they conclu de no significant ideological difference exists between the two groups. Sout hwell reiterates their previous findings in a survey conducted in December 2002 and January 2003 (2004). When analyzing the data from the 1996 special Senate electi on, Southwell and Burchett confirm their conclusions by emphasizing that the electorate might change demographically in terms of age and levels of partisanship, but election outcomes will remain stable (2000b). Using regression analysis on 48 statewide elections, three of which were administered solely through the post, Southwell and Burchett conc lude all-mail balloting boosts turnout over ten percentage points (2000a). Like Southwell and Burchett, Karp a nd Banducci (2000) st udy Oregon election data to determine if all-mail elections augmen t turnout. They disagree in their results, however, by concluding the effect of all-mail balloting differs in low salience versus high salience elections. Corresponding with Maglebys prediction, the authors find the greatest effect in low-salienc e elections, specifica lly local and special statewide votes. In high-salience elections (Presidential, midterm general elections, and Presidential

PAGE 23

15 primaries), the researchers determine the e ffect of all-mail balloting on turnout is insignificant.

PAGE 24

16 CHAPTER 3 SWISS GOVERNMENT AND ELECTIONS Like citizens in the United States, Swiss voter s seem averse to the act of voting. In Switzerland the executive is determined desp ite election results, as a seven-member board represents the four majo r parties in the Federal Counc il. The chairperson of the Federal Council is elected annually by its members. While the Swiss constitution calls for the supervision of the Council by the legisl ature, the Council has gradually assumed a role in directing the legislat ive process (Linder, 1994). The power of the Swiss federalist structure is highly concentrated in the canton. The Swiss constitution states that in the ab sence of an amendment by the people, all future powers should be delegated to the canton. The rule has been effective, due to the lack of an implied powers clause and the im plication of direct democracy. (In order to confer any new federal power, the majority of the people and the cantons must approve via referendum, as well as both chambers of th e legislature.) As evidence of the weakness of the central government, federal tax shar e and expenditure as a proportion of total receipts and expenditure pales in comparis on to other industrial democracies (Linder 1994, p.43). Every four years voters elect a national legislature di vided into two equal bodies: the Council of States, elected through fixed representati on, and the National Council, elected through List PR. (Bef ore World War I, the Nationa l Council was elected through a First-Past-the-Post system.) The nation is divided into twenty cantons and six halfcantons, giving a total of twenty -six official districts through which legislative seats are

PAGE 25

17 apportioned. The most recent addition to the federation was the canton of Jura in 1979. Part of the canton of Bern since the C ongress of Vienna, Jura seceded due to longstanding linguistic and religious differences The canton became official member of the federation after the approval of the Swiss population through referendum. Switzerland is one of the clos est examples of direct democracy in the world. On every bill decided before Parliament, the at tainment of 50,000 signatures provides for the calling of a popular referendum. The signature s of 100,000 citizens call for the vote for a constitutional amendment. When the con cept was introduced in the 19th century, the founders of Switzerland believed the referendum would be used more for innovation and less for inhibition. The opposite has turned ou t to be true, as Swiss voters tend to mobilize essentially to reject constitutional revisions and laws rather than accept them (Eschet-Schwarz 1989, p.255). The best example of the Swiss inclin ation for obstruction is womens suffrage, which wasnt implemented until 1971. Partially due to the high number of elec tions, Swiss cantons have intermittently begun to implement postal voting, as displa yed in Table 2 (Initiative & Referendum Institute 2004). Two systems for postal voting exist in Switzerland: the simplified system and the system of voting on request. In the latter, voters must request a ballot for every vote. Under the simplified system, each Swiss ci tizen receives a ballot in the mail in the weeks leading up to the election or referendum. For the purpose of this analysis I will only focus on the simplified system since it pr ovides the greatest redu ction in voting costs and the easiest way to cast a ballot. The can ton of Basel-Landschaft became the first to institute a simplified system in 1978, and all bu t two of the twenty-six had introduced it

PAGE 26

18 Table 2 Introduction of Simplified System of Early Voting CANTON YEAR OF INTRODUCTION Basel-Country 1978 St. Gallen 1979 Appenzell Inner-Rhodes 1979 Solothurn 1980 Thurgau 1985 Appenzell Outer-Rhodes 1988 Bern 1991 Aargau 1993 Luzern 1994 Nidwalden 1994 Zurich 1994 Basel-Town 1995 Friburg 1995 Geneva 1995 Glarus 1995 Graubunden 1995 Obwalden 1995 Schaffhausen 1995 Uri 1995 Zug 1997 Jura 1999 Schwyz 2000 Vaud 2002 Neuchatel 2003 by 2003. Unlike in-person early voting in the United States, the implementation of postal voting in Switzerland was pervasive through each canton. To an extent, Switzerland serves as an ex ample to display the possibility of postal voting in the U.S. due to the low levels of voter turnout, government structure (bicameral legislature), and decentralized federalism sh ared by both nations. The nations are also fairly heterogeneous, but in different ways. Where America is a melting pot in terms of ethnicity, Switzerland comprises people group s partitioned by four official languages. I also note similarities in terms of GDP pe r capita (Directorate of Intelligence, CIA 2006b). However, it is also necessary to docum ent some of the differences that exist in

PAGE 27

19 the nations as well, as Switzerland is not a perfect proxy for the effect of U.S. election reforms on voter turnout. While the United St ates elects the Hous e of Representatives utilizing a First-Past-the-Post system, the Sw iss National Council is elected by List PR. As a result of PR elections, the party struct ure is much more volatile in Switzerland, as evidenced by the rapid ascension of the Green Party. Also, the possi bility of referendum on any proposed legislation provides the Swi ss people a much greater political voice than their American counterparts. Demographi cally, Switzerland is more densely populated even though more citizens liv e in rural area. Despite th ese differences, however, the parallels between Switzerland and the United St ates can provide us some information on the efficacy of electoral reforms.

PAGE 28

20 CHAPTER 4 DATA AND HYPOTHESES The purpose of this paper is to analyze the effects of in-per son early voting and postal voting on turnout in the U.S. states a nd Swiss cantons, respectively. These studies examine the effects of these sy stems of early voting across bo th space and time. In order to measure the effects I utilize an OLS regression. United States In the investigation on the United States the unit of analysis is the state-year interaction in every general midterm el ection from 1982 to 2002. As Texas first implemented IPEV in 1991, using those six elections provides a symmetry three national elections with in-person voting, and three without Due to state law regarding unchallenged seats, coupled with the absence of a senatorial or gubernatorial election, turnout data are not availabl e from the 1982 Louisiana electio n, leading to a total sample size of 299. Turnout is regressed on in-p erson early voting w ith dummy variables representing the proportion of th e population in each interacti on with the ability to vote early, as well as control variables for une mployment and political competition. Fixed effects for individual states and years ar e also included and represented by dummy variables. Voter turnout is calculated by two different me thods: as a proportion of the voting-age population (VAP) and as a proporti on of the voting-eligib le population. (VEP) Voting-age population is chosen due to the prevalence of studies utilizing VAP in analyzing turnout rates, and recent res earch (McDonald and Popkin, 2001) has given

PAGE 29

21 precedent for analysis by VEP.1 Data used to calculate turnout are obtained from the United States Election Project (McDonald 2006). Along with the succeeding controls I include state and year dummy variables. I note, however, differences between liberal ized absentee voting laws and in-person early voting. In most states, early voti ng . place[s] fewer demands on voters than participation by absentee, (Hansen 2001, p.59) as voters must apply for an absentee ballot well before the day they will actually co mplete their ballot. For the most part, the costs of voting absentee are much greater th an IPEV. In the few states which allow voters to apply and fill out absentee ballots on the same day, research has shown turnout is not affected unless accompanied by massive mobilization efforts by the parties. Even with substantial mobilization, turnout is s hown to increase only two percentage points (Oliver 1996). In-Person Early Voting Of the nine states which had impl emented early voting by 2002, seven had instituted the procedure statewide. For these states, a dummy variable equal to one is set to represent IPEV in the corresponding interact ions. The states of California and Kansas, however, have in-person early voting procedures but only in select counties. In these states the variable represen ting in-person early voting is the proportion of the total population exposed to IPEV. The population data are provided by the Census Bureau website (United States Census Bureau 2004, 2005) In line with previous research, I expect the effect of this variable to be insignificant. 1 I admit that calculating turnout as a percentage of the population of registered voters would provide better analysis due to the aforementioned two-step process of voting. Howe ver, the incongruence of state and county policies in purging registration rolls wo uld necessarily lead to inaccurate results.

PAGE 30

22 Political Competition I control for political competition by crea ting two separate variables accounting for the presence of gubernatorial and senatorial elections and their respective margins of victory. For these elections, the data values represent the negative log of the margin of victory. A greater value, therefore, repres ents a closer election. For the state-year interactions without a gubernat orial or senatorial election, th e value is zero. The data for the gubernatorial elections are obtained from the Almanac of American Politics (Barone, et al. 2004). The data for the senatorial elections are obtained from election results provided by the Clerk of the House webs ite (Carle 1995; Dendy 1987, 1991; Ladd 1983; Trandahl 1999, 2003). In Texas in 1994, the ballot included two Senate elections. For this case the logs are taken for the two margins of victory, and the datum is equal to the negative of the sum of those two values. I also control for the closeness of Congre ssional elections. I represent this by the number of elections in a state decided by five percentage points or less, and setting that number as a proportion of the total number of state Congressional districts. These data are also found from the Clerk of the House website. Corresponding to classic literature (D owns 1957; Riker & Ordeshook 1968), I expect the Gubernatorial Competition Senatorial Competition and Congressional Competition to have a positive and significant relationship with turnout. Unemployment The Unemployment variable represents the economic environment across the individual states. Research has shown that economic adversity can l ead to falling out from the political system, and effectively reduces an individuals propensity to participate (Wolfinger and Rosenstone 1980; Rosenstone 1982). Rosenstone and Hansen conclude

PAGE 31

23 the money, time, and energy spent comba ting extreme economic adversity provide payoffs that are more immediate and valuable than the benefits that might be gained from investing in electoral politi cs (1993, p.135). They find the unemployed are 8.5% less likely to vote in midterm elections, confir ming earlier research on the effect of unemployment. With data provided by the Bure au of Labor Statistics website (Bureau of Labor Statistics 2006), this variable is expected to have a negative, significant relationship with turnout. National Macropartisan Strength Previous research also indicates the importance of partisanship in analyzing turnout. Bartels (2000) comes to two conc lusions about partisanship: since the 1970s, voters have become more partisan, and the impact of partisanship on vote choice has significantly increased. While partisanships effect on vote choice is not germane to this paper, I am concerned with the overall increase in partisanship in among American citizens. Verba and Nie (1972) conclude that strong party identifiers are more likely to vote than weak identifiers, who in turn are more likely to vote than independents. Therefore, the increase in macropartisansh ip beginning in the 1970s provided a positive bias to aggregate turnout rates in the United States. From the National Election Studies, as a m easure of partisanship I have created a national-level variable representing National Macropartisan Strength. Utilizing the zeroto-six values for partisanship from the NES, I folded the data, so a Strong Republican and Strong Democrat would both be represente d by a value of . I did the same for Moderate (2) and Weak (1) identifiers, as we ll as Independents (0). From those, I calculated the mean value of each year and used it to represent macropartisanship at the

PAGE 32

24 national level. I expect National Macropartisan Strength to be positively correlated with turnout. Election-Day Registration As mentioned earlier, previous research has concluded the strong and significant impact of registration reforms on aggregate turn out. Using data from Fitzgerald (2005), I control for election-day registration (EDR) in th e model. For the purpose of this analysis I count North Dakota, which has no registra tion laws, as a state with election-day registration. As EDR provide s those not registered on th e day of the election the opportunity to participate, so does th e registration policy of North Dakota. United States Regressions The following equations summarize the model for turnout in the United States as proportions of both the voting-age populati on and the voting-eligible population. I control for economic and political variables, as well as fixed effects for states and years through dummy variables. Equation 1 Turnout(VAP) = + 1( IPEV ) + 2( Gubernatorial Competition ) + 3( Senatorial Competition ) + 4( Congressional Competition ) + 5( National Macropartisan Strength ) + 6( EDR ) + ( State Dummies ) + ( Year Dummies ) + e Equation 2 Turnout(VEP) = + 1( IPEV ) + 2( Gubernatorial Competition ) + 3( Senatorial Competition ) + 4( Congressional Competition ) + 5( National Macropartisan Strength ) + 6( EDR ) + ( State Dummies ) + ( Year Dummies ) + e Switzerland In the Swiss investigation the unit of analys is is the canton-year interaction in the parliamentary elections in each canton from 1975 to 2003. I choose to use elections back to 1975 for two reasons: it was th e second election after women were granted the right to

PAGE 33

25 vote (accounting for a novelty effect), and the method of calculating Swiss unemployment data changed between and 5 (as to make constant the effect of unemployment on turnout levels). The unavailability of turnout statistics for three of the interactions, along with missing unemployment da ta for two, leads to a final sample size of 202. (The establishment of Jura in 1979 n ecessarily leaves only 25 data points in the 1975 cycle.) Turnout as a percen tage of VAP is regressed on the effect of postal voting. (Swiss citizens were not allowed to vote unt il the age of 20 before 1991, when legislation changed the age to 18. This modification was accounted for in the creation of the votingage population.) Calculating turnout as a f actor of the voting-eligible population is unnecessary, as the number of in eligible people is small.2 The voting-age population and vote totals were acquired from emailcorrespondence with employees of the Swiss Federal Statistical Office. The data can be found in French and German from their website (Swiss Federal Statistical Office 2006). Similar to the U.S. analyses, I control for both economic and political factors and include fixed effects for canton and year dummy variables. Since 1904 compulsory voting has existed in the canton of Schaffhausen. While the penalty of nonvoting is only three Swiss francs (or $2.44), I expect turnout in Schaffhausen to be comparatively higher than the other cantons. As a result, I conduct two analyses, one excluding Schaffhausen. Along with the succeeding controls I include canton and year dummy variables. 2 Information obtained from email correspondence with Elisabeth Willen of the Swiss Federal Statistical Office, April 26, 2006

PAGE 34

26 Postal Voting As of 2003, all but two of the twenty-six cantons had implemented a simplified system of postal voting. When these policies were establis hed, the ability to vote by mail permeated the entire populations of the resp ective cantons. As a result, the variable representing postal votin g is a simple toggle dummy of either one (existence of postal voting) or zero (absence). As previous ly noted, these data are obtained from SwissWorld. Unemployment As in the U.S. analysis, I include cant onal levels of unemployment to account for the economic environment. I expect this eff ect to be negative. The data were obtained from email correspondence with the represen tatives of the Swiss Federal Statistical Office. Political Competition Characteristic of a List PR parliamentary system, many parties exist and hold some power in Switzerland. One measure utilized to represent political competition is the difference between the vote shares for the two leading parties in the election at the cantonal level. Previous research in a cr oss-national study indicat es that a ten point difference between the leading and second pa rties reduces turnout 1.4 points (Blais 2000, p. 30). I use the same measure in my analys is with data obtained from Swiss Politics (Swiss Broadcasting Company 2006). Another m easure used is the number of parties in each canton winning seats in an election. A pr evious cross-national study revealed that a greater number of parties is positively correlated with higher turnout, as electors have a greater selection from which to choose (B lais & Carty 1990, from Blais 2000). However, party strength is not disseminated equally throughout the cantons, and as a proxy I use the

PAGE 35

27 number of parties in each canton which won le gislative seats, as obt ained from the Swiss Federal Statistical Office. In the regression, I expect both Vote Margin and Number of Winning Parties to be positively correlated with turnout. Swiss Regressions The following equation summarizes the m odel for turnout in Switzerland as a proportion of the voting-age population. I contro l for economic and political variables, as well as fixed effects for cantons and y ears through toggle dummy variables. Equation 3 Turnout(VAP) = + 1( Postal Voting ) + 2( Vote Margin ) + 3( Number of Winning Parties ) + 4( Canton-Level Unemployment ) + ( Canton Dummies )3 + ( Year Dummies ) + e 3 One regression will include Schaffhausen, one will exclude

PAGE 36

28 CHAPTER 5 RESULTS In-Person Early Voting Corresponding to previous literat ure, Tables 3 and 4 show that the existence of inperson early voting is not posit ively correlated with voter tu rnout. As a percentage of both VAP and VEP, turnout is actually ne gatively correlated w ith IPEV, although the coefficient is insignificant. The variable s representing political competition, along with EDR, all have a positive relationship with tu rnout, as expected. However, none of the variables are significant. The partisanship variable is also positive and insignificant in both analyses. Unexpectedly, the unemployme nt variable is positively correlated with turnout. But like the others, th e effect is insignificant. Table 3 Effect of In-Person Ea rly Voting in the States (VAP) Coefficient SE Significance IPEV -.030 .020 .129 Gubernatorial Competition .012 .009 .183 Senatorial Competition .000 .000 .749 Congressional Competition .009 .023 .715 National Macropartisanship Strength .068 .245 .781 State-Level Unemployment .003 .003 .273 EDR .018 .027 .508 R-square=0.594; N=299; Estimated by OLS Fixed Effects for State and Year Dummies not included in Table

PAGE 37

29 Table 4 Effect of In-Pers on Early Voting in the States Coefficient SE Significance IPEV -.023 .020 .259 Gubernatorial Competition .013 .009 .155 Senatorial Competition .000 .000 .707 Congressional Competition .009 .024 .702 National Macropartisanship Strength .069 .251 .783 State-Level Unemployment .004 .003 .163 EDR .012 .028 .660 R-square=0.566; N=299; Estimated by OLS Fixed Effects for State and Year Dummies not included in Table The results are very surprising. While I am not astonished to find that in-person early voting does not catalyze aggregate turnout rates, th e nonnegative correlation is frustrating. I am also disappointed with th e ineffectiveness of th e political and economic variables. While the sign of those coefficien ts are all in the exp ected positive direction, the high standard errors associated with the co efficients indicate their ineffectiveness in predicting turnout in these models. Due to importance of registration reforms on aggregate turnout levels, I was also surprise d to discover the insigni ficance of electionday registration. Combined with the fixe d effects for state and year dummies, the weakness of all of the variable s included in the regressions leads to low R-square values (.594 and .566 for the VAP and VEP models, re spectively), indicati ng the feebleness of the model in predic ting turnout rates. Postal Voting Much like voting-by-mail in the United States Tables 5 and 6 indicate the effect of postal voting in Switzerland is positive and subs tantial. At the cantonal level, I expect

PAGE 38

30 the incidence of postal voting to augment tur nout approximately six percentage points. Vote Margin the variable representing the differen ce in vote share between the leading and second parties has an enormous effect on turnout. Every ten point decline in margin leads to increased tur nout of 2.3 percentage points. Th e other variable measuring party competition the number of parties in each canton securing legislative seats in the National Council is actually negative. It is insignificant in both regressions, however. The Unemployment variable, as in the U.S. regressions, is surprisingly positive and insignificant. Table 5 Effect of Postal Voting in the Cantons Coefficient SE Significance Postal Voting .057 .014 .000 Vote Margin -.234 .029 .000 Number of Winning Parties -.006 .006 .296 Canton-Level Unemployment .003 .006 .589 R-square=0.810; N=202; Estimated by OLS Fixed Effects for Canton and Year Dummies not included in Table Table 6 Effect of Postal Voting in the Cantons (Schaffhausen excluded) Coefficient SE Significance Postal Voting .060 .014 .000 Vote Margin -.233 .030 .000 Number of Winning Parties -.006 .006 .285 Canton-Level Unemployment .003 .006 .579 R-square=0.768; N=194; Estimated by OLS Fixed Effects for Canton and Year Dummies not included in Table Unlike in-person early voting in the U.S ., I find the effect of postal voting on aggregate turnout levels in Switzerland to be strong. I cannot say I am surprised with the value of the coefficien t corresponding to the Number of Winning Parties As mentioned previously, the number of competitive parties in a district has a positive effect on turnout.

PAGE 39

31 While it is obvious that the number of winning parties is not equivalent to the number of competitive parties, I believed that the hi gh correlation between two would allow the number of winning parties to se rve as a suitable proxy. This proved not to be the case. The variable representing Vote Margin however, resulted in being positive and significant, so I believe political compet ition was successfully controlled in the regression. Unfortunately, the coefficient corresponding to the variable representing Cantonal-Level Unemployment is positive, yet insignificant. It appears that the Unemployment variables in all four regressions do an inadeq uate job in representing the economic environment. Unlike the U.S. re gressions, the corresponding R-square values in the Swiss regressions (.810 and .768 for th e regression including all cantons and the regression excluding Schaffhausen, respectiv ely) indicate the models strength in predicting turnout. One interesting finding from the Swiss anal yses involves the eff ect of compulsory voting. In the regression including th e canton of Schaffhausen, the corresponding coefficient is strong, positive, and significant. (b =.192) It is the highest coefficient for all canton dummies, with the next-strongest positive coefficient for a canton (Tessin) coming in almost seven points lower. (b=.124) This finding indicates that a government does not need to institute a forceful penalty for nonvo ting to produce higher turnout. The desired effect can be accomplished by a weak penalty, as evidenced by a fine of only three Swiss Francs in the canton of Schaffhausen. Consid ering the existence of a compulsory-voting policy, it should come as no surprise that turnout in Schaffhausen is almost twenty points higher than a given canton.

PAGE 40

32 CHAPTER 6 DISCUSSION The most salient finding from this study is the strong effect of postal voting on turnout. Previous research on postal voting in the United Stat es has indicated it provided a boon to turnout, but analyses were conduc ted on a limited scope. Here, I study the effect of postal voting across time and space in a nation, like the U.S., with a heterogeneous culture and comparatively low turnout. In line with previous resear ch on in-person early voting, I find no significant effect of IPEV on aggregate turnout levels. The natural question, then, is why postal voting increases turnout yet in-person early voting does not. I believe the answer stems from the extent by which each system reduces the costs of voting. In an IPEV system, a prospective voter must still e ndure the process of traveling to the polls. Depending on the early voting location, voting before Election Da y could actually be more time-consuming than voting on Election Day, depending on how early voting sites are designated and the manner in which voters are assigned to t hose locations. A populous county which only opens IPEV at one location could actually resu lt in increasing the time needed to cast a ballot, due to such factors as longer waiting lines and increased driv e time. Under postal voting, however, all transportati on costs are obviated as all th at is required (outside of completing the ballot) to vote by post is a visit to ones mailbox. An obvious worry of postal voting is the perceived possibility of voter fraud. Specifically, opponents of postal voting fear t hat someone other than the addressee will use undelivered or duplicate ballots (Sout hwell & Burchett 1998, p.348). However,

PAGE 41

33 committing voter fraud through postal voting woul d require forging the registered voters signature. In Oregon, Southwell and Burchett note poll workers are required to compare the signature on the outside of the envelope to the signature on the voter roll. I believe similar rules regarding postal voting would render the possibility of voter fraud through the mail no greater than the possibility of fraud at a standard Election Day polling location. Another concern associated with voter fraud is the possibility of vote influence. As citizens fill out their ballots at home, critics believe the presence of an individual close to the voter can unduly influence the voters choice. However, research by Southwell and Burchett showed this is simply not the case. In a survey of 1225 respondents, researchers found only three (0.3%) vote-by-ma il voters said the presence of a person in the room made them feel pressured to vot e a certain way. Of t hose three, only one indicated the pressure caused him/her to vote differently (Southwell and Burchett 1997, p.54). Previously mentioned, one of the dilemm as of low turnout is the associated socioeconomic skew between voters and nonvoter s. Piven and Cloward (1989) theorized that increasing turnout would aid in ameliora ting the problem. Un fortunately, research on various forms of voting reforms indicates th at increased turnout does nothing to solve this predicament. Karp and Banducci (2001) co nclude that early voters are more likely to be educated, active in politics, and partisan th an traditional voters. In an analysis of liberalized registration laws, Mitchell and Wlezien (1995) discover the change to the electorate in terms of income is menial. Another study finds that postal voting works to retain registered nonvoters, therefore stimul ating individuals bel onging to groups already likely to participate (Berinsky et al. 2001) Berinsky (2005) actually concludes that

PAGE 42

34 voting reforms will exacerbate the problem of socioeconomic skew by escalating, rather than lessening the biases in the electorate. Nevertheless, benefits associated with high turnout outside of socioeconomic grounds sti ll provide ample reasons to find ways to induce more citizens to vote. Future research should focus on the effec tiveness of postal voting in ameliorating the problem of socioeconomic skew in political participation. It w ould be interesting to see research that utilizes Swiss survey da ta and demonstrates the effect of all-mail balloting on the socioeconomic composition of the electorate While I have not found any evidence of SES skew specific to Switzerland, previous research on international turnout has indicated the importance of resources (V erba et al. 1978), which gives cause to believe in the existence of bias. Also, future research should concentrate on the administrative side of in-person early vo ting. Gronke, et al. (2005) analyze the long waits and voter intimidation that occurred ofte n in early voting locat ions in Florida in 2004, but focus more on the demographics of early voters compared to election-day voters. Researchers should look to determine the optimal locations for early voting sites, (e.g. county elections offices vs. shopping ma lls) as well as the pr oper distribution of voting machines, in order to completely minimize the costs associated with IPEV. This will aid in depleting the learning curve asso ciated with enacting in-person early voting. Currently, it is difficult to ascertain the effect of all-mail balloting in the United States, as only one genera l election (Oregon 2004) has been conducted by postal voting. However, I hope Switzerland has provided a fr amework to demonstrate the benefits to turnout rates of conducting all el ections through the mail. As a result of this analysis, I believe instituting a postal voting system across the United States population would

PAGE 43

35 induce many more citizens to cast a ballot. Along with other in centives stemming from of all-mail balloting, such as the significantl y reduced costs of administering elections, I pose that converting all general elections in the U.S. to vote-by-mail would be beneficiary to all parties i nvolved, and look forward to the day all American citizens cast a ballot through the mail.

PAGE 44

36 LIST OF REFERENCES Aldrich, John H. 1993. Rational Choice and Turnout. American Journal of Political Science 37 (February):246-78 Barone, Michael, Grant Ujifusa, and Douglas Matthews. 2004. Almanac of American Politics New York: E.P. Dutton. Bartels, Larry. 2000. Partisansh ip and Voting Behavior. American Journal of Political Science 44 (January):35-50 Berelson, Bernard, Paul Lazarsfeld, and William McPhee. 1954. Voting: A Study of Opinion Formulation in a Presidential Campaign Chicago: University of Chicago Press. Berinsky, Adam J. 2005. The Perverse Conseque nces of Electoral Reform in the United States. American Politics Research 33 (July):471-491. Berinsky, Adam J., Nancy Burns and Michael W. Traugott. 2001. Who Votes by Mail? A Dynamic Model of the Individual-Le vel Consequences of Voting-By-Mail Systems. Public Opinion Quarterly 65 (Summer):178-97. Blais, Andre. 2000. To Vote or Not to Vote: The Meri ts and Limits of Rational Choice Theory Pittsburgh: University of Pittsburgh Press. Blais, Andre, and Kenneth Carty. 1990. Doe s Proportional Representation Foster Voter Turnout? European Journal of Political Research 18:167-81. Brady, Henry E., Kay Lehman Schlozman, and Sidney Verba. 1995. Beyond SES: A Resource Model of Polit ical Participation. American Political Science Review 89 (June):271-294. Brians, Craig Leonard, and Be rnard Grofman. 2001. Election Day Registrations Effect on U.S. Voter Turnout. Social Science Quarterly 82 (March):170-83. Brown, Robert D., Robert A. Jackson, a nd Gerald C. Wright. 1999. Registration, Turnout, and State Party Systems. Political Research Quarterly 52 (September):463-79.

PAGE 45

37 Brown, Robert D., and Justin Wedeking. 2006. People Who Have Their Tickets But Do Not Use Them: Motor Voter, Registration, and Turnout Revisited. American Politics Research 34 (July): 479-504. Bureau of Labor Statistics. 2006. Local Area Unemployment Statistics. Accessed April, 2006. http://data.bls.gov/cgi-bin/surveymost?la Burnham, Walter Dean. 1980. The Appearance and Disappearance of the American Voter. in Electoral Participation: A Comparative Analysis ed. Richard Rose. Beverly Hills, CA: Sage. Burnham, Walter Dean. 1987. The Turnout Problem. in Elections American Style ed. A. James Reichley. Washi ngton, DC: Brookings Institution. Campbell, Angus, Philip Converse, Wa rren Miller, and Donald Stokes. 1960. The American Voter Chicago: University of Chicago Press. Carle, Robin H. 1995. Statistics of the Congressional Election of November 8, 1994. Office of the Clerk U.S. House of Representatives Accessed April, 2006. http://clerk.house.gov/member s/electionInfo/1994/94Stat.htm Davis, Susan. Voting by Mail Could Improve American Democracy. Roll Call June 22, 2005. Dendy, Dallas L. Jr. 1987. Statistics of th e Congressional Election of November 4, 1986 Office of the Clerk U.S. House of Representatives Accessed April, 2006. http://clerk.house.gov/member s/electionInfo/1986election.pdf Dendy, Dallas L. Jr. 1991. Statistics of th e Congressional Election of November 6, 1990 Office of the Clerk U.S. House of Representatives Accessed April, 2006. http://clerk.house.gov/member s/electionInfo/1990election.pdf Directorate of Intellingence, CIA. 2006( a). Field Listing Literacy. in World Factbook Accessed April, 2006. http://www.cia.gov/cia/publica tions/factbook/fields/2103.html Directorate of Intellingence, CIA. 2006(b). Rank Order GDP pe r capita (PPP) in World Factbook Accessed April 2006. http://www.cia.gov/cia/publicati ons/factbook/rankorder/2004rank.html Downs, Anthony. 1957. An Economic Theory of Democracy New York: Harper & Row. Election Reform Information Project. 2006. Election Reform: Whats Changed, What Hasnt and Why? Accessed April, 2006. http://www.electiononline.org

PAGE 46

38 Eschet-Schwarz, Andre. 1989. Political Partic ipation in Swiss Referenda at Federal and Cantonal Levels: 1879-1981. Political Behavior 11 (September):255-272. Fitzgerald, Mary. 2005. Greater Convenience but not Greater Turnout: The Impact of Alternative Voting Methods on Electoral Participation in the United States. American Politics Research 33 (November):842-67. Franklin, Mark N. 1996. Electoral Participation in Comparing Democracies: Elections And Voting in Global Perspective ed. Laurence LeDuc, Richard G. Niemi, and Pippa Norris. Thousand Oaks, CA: Sage. Franklin, Mark N. 1999. Electoral Engineer ing and Cross-National Turnout Differences: What Role for Compulsory Voting? British Journal of Political Science 29 (January):205-216. Franklin, Mark N., and Wolfgang P. Hirczy de Mino. 1998. Separat ed Powers, Divided Government, and Turnout in US Presidential Elections. American Journal of Political Science 42 (January): 316-326. Gimpel, J.G., and J.E. Schuknecht. 2003. Politic al Participation and the Accessibility of the Ballot Box. Political Geography 22 (June):471-488. Gronke, Paul. 2005. Ballot Integrity under Or egons Vote by Mail System. Report Prepared for the Commission of Fe deral Election Reform, Washington DC. Gronke, Paul, Benjamin Bishin, Daniel Stevens, and Eva Galanes-Rosenbaum. 2005. Early Voting in Florida, 2004. Paper pr epared for the Annual Conference of the American Political Science Associ ation, Washington DC. Accessed June, 2006. http://www.reed.edu/~gronkep/docs/GronkeBishinStevensGalanesRosenbaum.APSA.2005.pdf Gronke, Paul, and Eva Galanes-Rosenbaum. 200 5. Getting out the Early Vote: Lessons for Progressives. Paper presented at the annual meeting of the American Political Science Association. Accessed July, 2006. http://earlyvoting.net/r esources/GronkePTC.pdf Hamilton, Randy. 1988. American All-Mail Balloting: A Decades Experience. Public Administration Review 48 (September October):860-866. Hansen, John Mark. 2001. To Assure Pride and Confiden ce in the Electoral Process Final Report of the National Commissi on on Electoral Reform, University of Virginia. Accessed June, 2006. http://millercenter.virginia.edu/progr ams/natl_commissions/commission_final_re port/task_force_report /task_force_complete.pdf

PAGE 47

39 Haspel, Moshe and H. Gibbs Knotts. 2005. Location, Location, Location: Precinct Placement and the Costs of Voting. Journal of Politics 67 (May):560-573. Highton, Benjamin. 1997. Easy Regi stration and Voter Turnout. Journal of Politics 59 (May):565-75. Highton, Benjamin. 2004. Voter Registration and Turnout in the United States. Perspectives on Politics 2 (September):507-15. Highton, Benjamin, and Raymond E. Wolfinger. 1998. Estimating the Effects of the National Voter Registration Act of 1993. Political Behavior 20 (June):79-104. Hill, Kim Quaile, Jan E. Leighley, and Angela Hinton-Andersson. 1995. Lower-Class Mobilization and Policy Linkage in the U.S. States. American Journal of Political Science 39 (February):75-86 Hirczy de Mino, Wolfgang. 1994. The Impact of Mandatory Voting Laws on Turnout: A Quasi-Experimental Approach. Electoral Studies 13 (March):64-76. Initiative & Referendum Institute Europe (IRI), Amsterdam. 2004. Postal Voting in SwissWorld Accessed April, 2006. http://www.swissworld.org/dvd_rom/e ng/direct_democracy_2004/content/votes/p ostal_voting.pdf International Institute for Democracy and Electoral Assistance. 2006. Voter Turnout Accessed April, 2006. http://www.idea.int/vt/ Jackman, Robert W. 1987. Political Institu tions and Voter Turnout in the Industrial Democracies. American Political Science Review 81 (June):405-424. Jackman, Robert W., and Ross A. Miller. 1995. Voter Turnout in the Industrial Democracies During the 1980s American Political Science Review 81 (June):405-423. Karp, Jeffrey A., and Susan A. Banducci. 2000. Going Postal: How All-Mail Elections Influence Turnout. Political Behavior 22 (September):223-39. Karp, Jeffrey A., and Susan A. Banducci 2001. Absentee Voting, Participation, and Mobilization. American Politics Research 29 (March):183-95. Key, V.O., Jr. 1949. Southern Politics in State and Nation New York: Vintage Books. Knack, Stephen. 1995. Does Motor Voter Work ? Evidence from State-Level Data. Journal of Politics 57 (August):796-811.

PAGE 48

40 Ladd, Thomas E. 1983. Statistics of the C ongressional Election of November 2, 1982. Office of the Clerk U.S. House of Representatives Accessed April, 2006. http://clerk.house.gov/member s/electionInfo/1982election.pdf Leighley, Jan E., and Jonathan Nagler. 1992. Socioeconomic Class Bias in Turnout, 1964-1988: The Voters Remain the Same. American Political Science Review 86 (September):725-36. Linder, Wolf. 1994. Swiss Democracy London: Macmillan. Lijphart, Arend. 1997. Unequal Participati on: Democracys Unresolved Dilemma. American Political Science Review 91 (March):1-14. Magleby, David. 1987. Participation in Mail-Ballot Elections. Western Political Quarterly 40 (March):79-91. Martinez, Michael D., and David H ill. 1999. Did Motor Voter Work? American Politics Quarterly 27 (July):296-315. McDonald, Michael. 2006. Voter Turnout. United States Election Project Accessed April, 2006. http://elections.gmu.e du/voter_turnout.htm McDonald, Michael, and Samuel L. Popkin. 2001. The Myth of the Vanishing Voter. American Political Science Review 95 (December):963-74. Mill, John Stuart. [1861] 1958. Considerations on Representative Government New York: Liberal Arts Press. Mitchell, Glenn E., and Christopher Wlezien. 1995. The Impact of Legal Constraints on Voter Registration, Turnout, and the Com position of the American Electorate. Political Behavior 17 (June):179-202. Neeley, Grant W., and Lilliard E. Rich ardson, Jr. 2001. Who is Early Voting? An Individual Level Examination. Social Science Journal 38:381-392. Newby, Brian D. 2006. Vote in Advance In Person in Johnson County Election Office Accessed June, 2006. http://www.jocoelection.org/ AdvanceVotingPrimary2006.htm Oliver, Eric. J. 1996. The Effects of Elig ibility Restrictions and Party Activity on Absentee Voting and Overall Turnout. American Journal of Political Science 40 (May): 498-513. Oregon Secretary of State. 2006. A Brief Hi story of Vote by Mail. Accessed April, 2006. http://www.sos.state.or.us/elections/vbm/history.html

PAGE 49

41 Piven, Frances Fox, and Richard A. Cloward. 1988. Why Americans Dont Vote New York: Pantheon Books. Piven, Frances Fox, and Richard A. Cl oward. 1989. Government Statistics and Conflicting Explanations of Nonvoting. PS: Political Sc ience and Politics 22 (September):580-588. Powell, G. Bingham., Jr. 1986. American Vote r Turnout in Comparative Perspective. American Political Science Review 80 (March):17-43. Richardson, Lilliard E. Jr., and Grant W. N eeley. 1996. The Impact of Early Voting on Turnout: The 1994 Elections in Tennessee. State and Local Government Review 28 (Fall):173-179. Riker, William H., and Peter C. Ordeshook. 1968. A Theory of the Calculus of Voting. American Political Science Review 62 (March):25-42. Rosenstone, Steven J. 1982. Econom ic Adversity and Voter Turnout. American Journal of Political Science 26 (February):25-46. Rosenstone, Steven J., and John Mark Hansen. 1993. Mobilization, Participation, and Democracy in America New York: Macmillan. Rosenstone, Steven J., and Raymond E. Wolf inger. 1978. The Effect of Registration Laws on Voter Turnout. American Political Science Review 72 (March):22-45. Sola, Lester. 2006. Early Voting Sites in Miami-Dade County Elections Accessed June, 2006. http://elections.metro-dade.com/ voting_absent_earlyvoting-sites.asp Southwell, Priscilla L. 2004. Five Years Later: A Re-Assessment of Oregons Vote by Mail Electoral Process. PS: Politics and Political Science 98 (March):89-93. Southwell, Patricia L., and Justin Burc hett. 1997. Survey of Vote-by-Mail Senate Election in the State of Oregon. PS: Political Science and Politics 30 (March):53-57. Southwell, Priscilla L., and Justin Burc hett. 1998. Vote-By-Mail in the State of Oregon. Willamette Law Review 34 (Spring):345-56. Southwell, Priscilla L., and Justin Burchett 2000(a). The Effect of All-Mail Elections on Voter Turnout. American Politics Quarterly 28 (January):72-79. Southwell, Priscilla L., and Justin Burche tt. 2000(b). Does Changing the Rules Change the Players? Vote-by-Mail and th e Composition of the Electorate. Social Science Quarterly 81 (September):837-45.

PAGE 50

42 Squire, Peverill, Raymond E. Wolfinger, a nd David P. Glass. 1987. Residential Mobility and Voter Turnout. American Political Science Review 81 (March):45-66. Stein, Robert A. 1998. Introduction: Early Voting. Public Opinion Quarterly 62 (Spring):57-69. Stein, Robert A., and Patricia Garcia-M onet. 1997. Voting Ea rly but not Often. Social Science Quarterly 78 (September):657-671. Swiss Broadcasting Company. 2006. Geopolitics in SwissPolitics Accessed April, 2006. http://www.swisspolitics .org/en/politnavigator/ind ex.php?page=geopolitics_suche Swiss Federal Statistical Of fice. 2006. Accessed April, 2006. http://www.bfs.admin.ch/bfs/portal/en/index.html ` Tingsten, Herbert. 1937. Political Behavior: Studie s in Election Statistics London: P.S. King & Son. Trandahl, Jeff. 1999. Statistics of the C ongressional Election of November 3, 1998. Office of the Clerk U.S. House of Representatives Accessed April, 2006. http://clerk.house.gov/member s/electionInfo/1998/98Stat.htm Trandahl, Jeff. 2003. Statistics of the C ongressional Election of November 5, 2002. Office of the Clerk U.S. House of Representatives Accessed April, 2006. http://clerk.house.gov/member s/electionInfo/2002/2002Stat.htm United States Census Burea u. 2004. Intercensal Estimates: Ti me Series of Intercensal Estimates by County. Accessed April, 2006. http://www.census.gov/popest/archives/2000s/vintage_2001/CO-EST2001-12/ United States Census Bureau. 2005. Annual Esti mates of the Population for the United States and States, and for Puerto Rico : April 1, 2000 to July 1, 2005. Accessed April, 2006. http://www.census.gov/popest/s tates/NST-ann-est.html Verba, Sidney, and Norman H. Nie. 1972. Participation in America Chicago: University of Chicago Press. Verba, Sidney, Norman H. Nie, and Jae-On Kim. 1978. Participation and Political Equality: A Seven-Nation Comparison Cambridge: Cambridge University Press. Wolfinger, Raymond E., and Steven J. Rosenstone. 1980. Who Votes? New Haven, CT: Yale University Press.

PAGE 51

43 BIOGRAPHICAL SKETCH A 2004 graduate of the University of Alab ama, I enrolled in the Department of Political Science at the University of Flor ida in August of that year. I soon became intrigued with the subject of political behavior, and in my second year embarked on a study of voting reforms, which eventually blosso med into my masters thesis. In August I will attend the University of North Carolina, where I will begin study toward a Ph.D. in economics, focusing on political economy.


Permanent Link: http://ufdc.ufl.edu/UFE0015772/00001

Material Information

Title: Efficacy of Early Voting Systems in the United States and Switzerland
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0015772:00001

Permanent Link: http://ufdc.ufl.edu/UFE0015772/00001

Material Information

Title: Efficacy of Early Voting Systems in the United States and Switzerland
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0015772:00001


This item has the following downloads:


Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
    List of Tables
        Page v
    List of Figures
        Page vi
    Abstract
        Page vii
        Page viii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
    Election reforms
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
    Swiss government and elections
        Page 16
        Page 17
        Page 18
        Page 19
    Data and hypotheses
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
    Results
        Page 28
        Page 29
        Page 30
        Page 31
    Discussion
        Page 32
        Page 33
        Page 34
        Page 35
    References
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
    Biographical sketch
        Page 43
Full Text












EFFICACY OF EARLY VOTING SYSTEMS IN
THE UNITED STATES AND SWITZERLAND














By

JOHN STUART RABON


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA


2006



























Copyright 2006

by

John Stuart Rabon















TABLE OF CONTENTS



L IS T O F T A B L E S ................................................... ............................... ..................... v

LIST OF FIGURES .................................................. ............................ vi

ABSTRACT ...................................................................... vii

CHAPTER

1 IN T R O D U C T IO N ................................................. .............................................. .

2 ELECTION REFO RM S .................... ...............................................................7......

Registration Reforms .................. .. ........... ...............................7
V oting R eform s ............... ........................... ................................................8..
In-P erson E arly V oting ..................................... ........................ ............... 10
Postal V voting .......................................................................................................12

3 SWISS GOVERNMENT AND ELECTIONS ......................................................16

4 D A TA A N D H Y PO TH E SE S ....................................... ...................... ................ 20

U n ite d S ta te s ...............................................................................................................2 0
In-P erson E arly V oting ..................................... ........................ ................ 2 1
Political Com petition ....................................................................... 22
U n em p loy m en t ............................................. .. ........................ ...... ........ ... ..2 2
N national M acropartisan Strength.................................................... ................ 23
Election-D ay Registration ........................................................ 24
U united States R egressions.......................................... ......................... ................ 24
Sw itzerland .............. ....................................................................... . ......24
P ostal V voting ................................................................................................. 26
Unemployment ................................... ......... ...... ...............26
Political Com petition ....................................................................... 26
Swiss Regressions ...................... ........... ............................... 27

5 R E S U L T S .......................................................................................... ..................... 2 8

In-Person E arly V oting ................................................................... ................ 28
P o stal V o tin g .............................................................................................................. 2 9









6 D ISCU SSION ............................................................................... . .................32

L IST O F R E FE R E N C E S ... ......................................................................... ................ 36

BIOGRAPH ICAL SKETCH ................... ............................................................... 43















LIST OF TABLES


Table page

1 Introduction of In-Person Early V oting .................................................... ............... 11

2 Introduction of Simplified System of Early Voting................................................. 18

3 Effect of In-Person Early Voting in the States (VAP)..............................................28

4 Effect of In-Person Early Voting in the States..........................................................29

5 Effect of Postal Voting in the Cantons ...................................................30

6 Effect of Postal Voting in the Cantons (Schaffhausen excluded)...............................30
















LIST OF FIGURES

Figure page

1 International V other Turnout ..................................................................... ...............6...

2 U .S V other T u rn out .................................................. ............................................... .. 6















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts

EFFICACY OF EARLY VOTING SYSTEMS IN
THE UNITED STATES AND SWITZERLAND

By

John Stuart Rabon

August 2006

Chair: Michael Martinez
Major Department: Political Science

Of all the industrial democracies, the two nations which exhibit the lowest levels of

voter turnout, due to certain institutional factors, are the United States and Switzerland.

These two nations, partially as a result of low participation, have recently begun to

implement policies meant to augment participation: in-person early voting in the United

States and postal voting in Switzerland. The purpose of this thesis is to analyze the

effectiveness of these policies on increasing turnout, utilizing OLS regressions. In the

United States, I use the state-year interaction as the unit of analysis, measuring the effect

of in-person early voting across six separate midterm elections. In Switzerland I conduct

a similar analysis, using the canton-year interaction as the unit of analysis over eight

legislative elections. For both nations, I control for both political and economic factors

within each state-year and canton-year interaction, as well as fixed effect for districts

(states in the U.S., cantons in Switzerland) and years. I find that in-person early voting

does not work to increase turnout in the American states, but postal voting provides a









significant boon to aggregate turnout levels in Switzerland. I also analyze positive and

negative consequences of each system, and attempt to ease any fears associated with

voting by-mail.














CHAPTER 1
INTRODUCTION

The purpose of this thesis is to analyze policies that theoretically reduce the costs

of voting to the individual and lead to an increase in aggregate turnout. According to the

rational choice model, an individual will vote as long as the benefits of voting outweigh

the costs (Downs 1957; Riker & Ordeshook 1968). In theory, the decision to vote is a

marginal one, as both the costs (of information, as well as the actual act) and benefits of

voting are low (Aldrich 1993; Blais 2000). Any slight change in the costs or benefits of

voting could result in the altering of the voting decision, according to the theory of

rational choice. I note that, since Downs, two separate models of participation have been

developed: the resource (Brady et al. 1995) and mobilization (Rosenstone and Hansen

1993) models. Both models overlap the rational choice model somewhat, as they focus

on how socioeconomic status and group membership diminish the costs of voting,

respectively. For the purpose of this study the rational choice and resource models are

the same, as I focus on reducing the costs of voting to citizens regardless of economic

class. The mobilization model is not applicable in this thesis either, as its creators note it

is better at predicting when one participates rather than why. (Rosenstone and Hansen

1993, p.20) Henceforth, I will use the rational choice model as the framework for the

individual decision to vote.

Since the seminal work of Wolfinger and Rosenstone (1980), analysts have

believed education to be the most important determinant of an individual's propensity to

overcome the costs of voting and participate in the political sphere. One would naturally









expect that the United States and Switzerland, two nations with literacy rates close to

100% (Directorate of Intelligence, CIA 2006a), to exhibit high rates in the easiest form of

participation voting. But as Figure 1 demonstrates, this is not the case. As a percentage

of the voting-age population, voter turnout in all other original G-7 countries is

significantly higher than turnout in the United States and Switzerland. (I also point out

the fact that turnout in the U.S. as a percentage of the voting-age population is declining,

as evidenced by Figure 2.) Out of 58 parliamentary elections in those six countries from

1960 to 2000, only one (1994 Japan) displayed turnout lower than any U.S. midterm or

Swiss parliamentary election. U.S. and Swiss voters are conspicuously averse of going to

the polls, and present a conundrum to researchers and analysts of voter turnout.

In theorizing the determinants for turnout rates across industrial democracies,

including the U.S. and Switzerland, political scientists have posed an institutional model

for voter turnout. In this model, researchers propose that forces related to government

structure play a greater role in shaping turnout rates than individual attributes. In a

comparative study of seven nations touting the impact of institutions, Verba, Nie, and

Kim (1978) assert aggregate turnout is derived from the juxtaposition of individual

attributes working against national institutional factors. Powell (1986) concludes that

U.S. turnout is advantaged slightly by individual political interest and education levels.

However, institutional factors such as compulsory registration, nationally competitive

election districts, and strong party-group linkage result in lower rates of American

turnout. (Party-group linkage reflects the consistency of cues given by overlapping

groups such as labor unions, religious organizations, etc. to parties to determine the

strength and volume of political cues.) Jackman discovers "unicameralism provides a









clearer link between elections and legislation, increasing turnout" (1987, p. 405). Not

surprisingly, nations with compulsory voting experience higher levels of turnout.

(Tingsten 1937; Franklin 1996; Jackman 1987; Franklin 1999; Hirczy de Mino 1994)

The institutional model of turnout, however, does not fully account for outlying

U.S. and Swiss turnout rates. In analyzing cross-national differences in turnout,

researchers often have to control for the two states through an "electoral salience"

variable (Franklin 1996, 1999) or by dummies, per se (Jackman and Miller 1995).

Franklin (1996) goes as far as conducting an analysis without the two states.

Franklin defines "electoral salience" as "the linkage between legislative electoral

outcomes and government complexion" (1996, p.224). In the context of Europe, "the

most salient elections are those whose outcome determines the allocation of government

power" (Franklin & Hirczy de Mino 1998, p.317). Switzerland, then, exhibits

exceptionally low salience, as the composition of its executive will remain stable

regardless of election results. The low electoral salience displayed by the United States

stems partially from divided government. (i.e., different parties control the Presidency

and Congress) Franklin and Hirczy de Mino theorize that if one party controls both the

executive and legislative branches of government, it becomes easier for the American

electorate to give blame or credit to the controlling party. Turnout, therefore, will

increase under unanimous control. The researchers find that, controlling for changes in

closeness of elections and electoral legislation, divided government does in fact depress

turnout. For each election held under conditions of divided government, turnout declines,

on average, about 1.96% (p.321). After three consecutive Presidential elections under

divided government, turnout would be depressed 5.88% (p.321). Franklin also mentions









that low electoral salience is often correlated with the possibility of voter fatigue.

Because the United States and Switzerland have more elections than any other democracy

in the world, it is likely voter fatigue plays a role in depressing turnout.

Both the U.S. and Swiss, as well as all governments elected by a small proportion

of citizens, share cause for concern, as the negative effects produced by low levels of

voter turnout are numerous. Researchers and political pundits have long been troubled by

low turnout, and much of the literature associated with voter turnout has been devoted to

analyzing the effects of minimal participation. As a result of this research, experts have

developed a number of reasons for finding methods to improve turnout. For one, political

participation is stratified in terms of socioeconomic status (Verba & Nie 1972; Wolfinger

& Rosenstone 1980; Brady et al. 1995) and decline in voter turnout can lead to an

exacerbation of the discrepancy between the beliefs and preferences of active participants

and the entire population (Rosenstone & Hansen 1993; Burnham 1980). (I note that

research by Leighley and Nagler (1992) disagrees with this assessment.) As a result of

this skew, discord often appears between public policy and public opinion as the

politically active exhibit greater influence over the legislative and policy-making process

(Key 1949; Burnham 1987). Specifically, research has shown that states with low turnout

among the poor exhibit significantly lower welfare spending (Hill, Leighley, and Hinton-

Andersson 1995). Lijphart cites "especially poor" turnout in "less salient but no less

unimportant elections" (1997, p. 1) as a problem as well, due to Verba and Nie's assertion

that participation is "at the heart of democratic theory and at the heart of the democratic

political formula in the United States" (1972, p.3). Classic literature also points to the

benefits to the individual of voting. The voting act could provide positive externalities to









the individual, such as learning civic virtues like knowledge and responsibility (Mill

1958).

Lawmakers in both the United States and Switzerland, partially as a result of low

turnout and its consequences, have recently enacted policies to simplify the process of

voting. The purpose of this study is to analyze the effect of two of these reforms on

turnout: in-person early voting (IPEV) in the United States, and postal voting in

Switzerland.

Unfortunately, a limitation of this study is the differences in the institutional and

political structures of the two nations do not allow a true test of which system better

works to increase turnout. Most notably, the existence of elections by proportional

representation leads to a multiparty system in Switzerland. Other factors, such as

differences in the composition of the executive, will be discussed in further detail later.

















'A- ---S wiss


Ain2France
< *-- Le rTmarny












Figure 1 International Voter Turnout (VAP) in Parliamentary Elections Data
provided by the International Institute for Democracy and Electoral Institute
Voter Turnout Website (International Institute for Democracy and Electoral
Assistance 2006)







-- ^ t




r* """, --- Residential

SA"", A -A- r Mdterm
-- Linear Trend


Figure 2 U.S. Voter Turnout (VAP) Data provided by the United States
Election Project (McDonald 2006)


0.3

4P


0.65

0.6

0.55

0.5

0.45

0.4

0.35

0.3


- -.A. -













CHAPTER 2
ELECTION REFORMS

In the United States voting is a two-step process; in order for a citizen to cast a

ballot one must be registered. While the analyses presented here focus on the actual act

of voting, the importance of registration warrants mentioning, as registration reforms

have been the main vehicle state legislatures have employed to increase levels of turnout.

Registration Reforms

The idea that registration laws form a barrier to the act of voting permeates the

literature (Wolfinger & Rosenstone 1980; Squire et al. 1987; Highton 1997, 2004; Piven

& Cloward 1988; Rosenstone & Wolfinger 1978). Brown et al. conclude "group

registration level is the primary determinant of group turnout in both presidential and

midterm elections" (1999, p.474). Burnham believes that the "abolition and replacement

[of registration laws] by automatic state-enrollment procedures" is the first step to solving

the dilemma of low turnout (1980, p.68). Also of note, states have historically utilized

registration laws to depress turnout (Highton 2004).

As a result, in 1993 Congress passed the National Voter Registration Act (Motor

Voter) mandating the availability of voter registration via mail and in various government

agencies. Research conducted on policies mandated by Motor Voter provides differing

results, (Knack 1995; Highton & Wolfinger 1998; Martinez & Hill 1999; Brown &

Wedeking 2006) but no experts have concluded that the policies dramatically boost

turnout. Since the implementation of Motor Voter, turnout has remained remarkably

stagnant in the face of burgeoning registration rates (Neeley & Richardson 2001).









States have also begun to implement Election Day registration (EDR) in an effort to

assuage the hindrance of registering, and since 1990 the number of states with EDR has

nearly doubled (Fitzgerald 2005). The significance of EDR to increasing turnout is

constant. Highton (1997) concludes that EDR amplifies turnout levels ten points, while

Brians and Grofman (2001) predict a seven percent boost. In their classic work,

Wolfinger and Rosenstone conclude election-day registration is the most effective legal

change to registration procedures, deducing "turnout would increase by about 6.1

percentage points" (1980, p.78). Fitzgerald (2005) finds a lesser but still significant

impact an augmentation of more than one and three points in presidential and

congressional elections, respectively.

Voting Reforms

The crux of this study, however, is the effect on turnout of post-registration reforms

dealing with the act of voting. While I focus on state-level and cantonal-level reforms, I

note a recent modification in American voting law at the federal level: the Help America

Vote Act (HAVA). Created in 2002 as a response to the havoc caused by the Presidential

Election two years prior, the bill calls for the disbursement of funds to states with the

purpose of improving the administration of elections. Specifically, the act mandates the

elimination of punch-card ballots and lever machines in order to modernize and ease the

voting process. It also calls for further training of election officials and compulsory

institution of a provisional voting procedure, allowing those who believe they are missing

from the registration rolls by mistake to cast a ballot and dispute the error.

At the state level, legislatures have begun to institute reforms outside of election-

day registration. Increasingly, state policymakers are turning to early voting in an effort

to augment turnout. Reasons exist for the employment of early voting outside of the









general benefits of increasing turnout (Gronke et al. 2005). First, early voting systems

provide a more convenient method to vote, especially for those "who work for an hourly

wage, have long commutes, or have heavy time constraints on a November Tuesday" (p.

3). Research has shown that ballot box accessibility is can have a significant effect on

election participation (Gimpel & Schuknecht 2003; Haspel & Knotts 2005). Second,

advocates of early voting claim the procedure would augment the effectiveness of

democratic decision-making (Davis 2005). Third, Gronke et al. (2005) theorize citizens

like early voting because campaigns can focus their mobilization and persuasion efforts

on those yet to participate, leaving partisan voters the opportunity to continue their daily

lives free of campaign activity.

Theoretically, early voting increases turnout through providing greater availability

for a registered voter to cast a ballot. This increases the probability of individual voting,

especially for those unlikely to vote on Election Day. However, early voting may also

increase turnout through a social context, as "nontraditional voting sites may

communicate cues to voters about candidate choices" (Stein & Garcia-Monet 1997). As

a result, strategic politicians may be able to exploit early voting for their own electoral

advantage through efficient mobilization techniques.

Three methods currently exist for casting a ballot before Election Day: in-person

early voting (IPEV), voting-by-mail or postal voting, and absentee balloting. Many states

have recently begun to liberalize absentee balloting, as twenty states allow no-excuse

absentee balloting and California permits voters to be placed on a "permanent" absentee

voter list (Gronke 2005). In the 1996 Presidential election, twenty percent of votes from









those states came from absentee ballots (Karp & Banducci 2001). The analyses presented

here, however, will focus only on in-person early voting and voting-by-mail.

In-Person Early Voting

In-person early voting grants registered voters the ability to vote at a designated

location in the days leading up to the election. Texas became the first state to institute

IPEV in 1991, and twelve other legislatures have adopted statewide in-person early

voting since (Election Reform Information Project 2006; along with various state

elections officials). (Table 1) Two other states, California and Kansas, have in-person

early voting procedures but only in select counties.

The process of in-person early voting varies across the states and counties. For the

most part, state policies allow approximately fifteen days for prospective in-person early

voters to cast a ballot. Arizona policy allows the most days, as in-person early voting

begins 33 days before Election Day and ends the Friday prior. Georgia voters, on the

other hand, only have the five business days the week before the election to cast an early

ballot. Unlike the number of days states permit IPEV, the location of voting sites is

determined by the counties. Most counties make available in-person early voting only at

the county elections office. For more populous counties however, such as Miami-Dade

in Florida and Johnson County in Kansas, in-person early voting occurs at traditional

locales such as shopping malls, airports, and libraries (Sola 2006; Newby 2006).

The empirical evidence supporting the positive effect of in-person early voting on

turnout is slim. Previous research on IPEV focuses as much on its effect on the

composition of the electorate as its impact on turnout. In a study of in-person early

voters in 1994, no party advantage existed between the two groups, and education levels

were similar (Stein 1998). Interestingly, Stein does not expect "early voting to mobilize a









Table 1 Introduction of In-Person Early Voting
STATE YEAR OF INTRODUCTION

Texas 1991
Colorado 1992
Arizona 1993
Nevada 1993
New Mexico 1994
Tennessee 1994
Arkansas 1996
Kansas* 1996
California* 1998
North Dakota 2003
Florida 2004
Georgia 2004
Illinois 2006
Georgia 2006
West Virginia 2006
In-Person Early Voting is not statewide

significant number of new voters" (p.68). In an analysis of Oregon and Florida, Gronke

and Galanes-Rosenbaum contradict Stein by concluding early voters are likely to be older

and more educated than election day participants, and African-Americans, on average,

tend to refrain from early voting (2005). Gronke and Stein both agree that early voters

are much more partisan and ideological than precinct voters, confirming the idea that

partisans decide on a candidate well before political moderates (Campbell et al. 1960;

Berelson et al. 1954). In a study of the 1994 elections in Tennessee, Richardson and

Neeley (1996) note in-person early voting possesses the possibility to increase turnout,

but focus their analysis on the effect of different ballot types and voting locations.

A study of the 1992 Presidential election reveals that in-person early voting slightly

increased turnout in Texas counties (Stein & Garcia-Monet 1997). Analysis shows that

counties with a higher proportion of the vote cast early entailed marginally, but

significantly higher participation rates. In the election, a one percent increase in early









votes as a proportion of the total vote led to an increase of voter turnout by seven-

hundredths (0.07) of one percent (p.665). The researchers also conclude that in-person

early voting effectually brings voters with a history of electoral participation to the polls,

rather than new voters, as Stein and Garcia-Monet actually find a negative relationship

between change in voter registration and turnout (p.665). It is interesting to note that the

authors discovered that the number of early voting sites at non-traditional locations was

positively and significantly correlated with turnout. For every ten nontraditional voting

locations in a county, turnout increased fifteen-hundredths (0.15) of one percent (p.665).

Fitzgerald (2005) concludes in-person early voting has a negligible effect on

increasing turnout. Using the state-year interaction as the unit of analysis, the author

evaluates the impact of various voting reforms during general elections from 1972 to

2002. Controlling for political but not economic variables, Fitzgerald finds in-person

early voting, while insignificant, is negatively correlated with turnout. As it is highly

unlikely that early voting drives citizens from the polls, the author notes the possibility of

self-selection, admitting "alternative voting methods are most likely to exist in states that

experience high rates of voter turnout regardless of the reforms" (p.856).

Postal Voting

The expansion of postal voting across the country has been much more protracted

than in-person early voting. In 1977, the county of Monterey, California, was the first to

exclusively use voting-by-mail for "anything other than a very small, special district

election" (Magleby 1987). Since, the state of Oregon has come to the forefront in postal

voting, and in 2004 conducted a general, statewide election solely through VBM after

first experimentation in 1981. Oregonians have had the opportunity to vote-by-mail in

special statewide elections since 1993, and in 1998 voters overwhelmingly passed an









initiative expanding all-mail voting to primary and general elections (Oregon Secretary of

State 2006).

Research on postal voting, unlike in-person early voting, has consistently shown a

significant, positive effect on turnout. In a study of all-mail ballots of elections in seven

states, Hamilton (1988) finds increased turnout in every analysis. He also cites reasons to

implement voting-by-mail outside of general early voting the increased "integrity" of

elections. The utilization of all-mail balloting "provides election administrators as nearly

foolproof a way as can be devised to purge registration rolls of non-eligible voters"

(p.864). As registration rolls become more accurate, the likelihood of the deceased

casting a vote declines precipitously.

Magleby (1987) focuses on the effect of voting-by-mail on participation in seven

local, special elections. Of the seven, only one did not produce higher turnout than a

typical citywide, polling place election for the given locale. In San Diego in 1981, the

voters participated at a higher rate (61%) than the preceding Presidential primary (58%)

(p.82). Magleby also utilizes an OLS regression of polling-place and vote-by-mail

elections in five locations from 1980 to 1984 (N=43). He concludes converting from

polling-place elections to all-mail balloting will increase participation nineteen points,

controlling for locale, year, and the type of election. He cautions however, that the effect

of all-mail balloting in a general election will be significantly reduced, due to the lower

baseline of turnout in referendum elections. Magleby also cites the possibility of a

"novelty" effect, and admits the effect of vote-by-mail elections on turnout may decrease

over time. Not surprisingly, he finds the "best predictor of participation in mail ballot

elections is education" (p.89).









In January 1996, the state of Oregon carried out the first statewide election solely

in vote-by-mail format to select a replacement for Senator Bob Packwood. In a survey

conducted after the election, Southwell and Burchett (1997, 1998) find voters

overwhelmingly prefer voting-by-mail to casting a ballot at a polling place. Of those

who favored voting-by-mail, the most common reasons cited were convenience, more

time to read and complete the ballot, and the ability to avoid possible weather problems

and work responsibilities. The researchers find distinct demographic distinctions

between vote-by-mail and traditional polling-place voters, but mitigate the criticism that

voting-by-mail provides a partisan advantage as they conclude no significant ideological

difference exists between the two groups. Southwell reiterates their previous findings in

a survey conducted in December 2002 and January 2003 (2004). When analyzing the

data from the 1996 special Senate election, Southwell and Burchett confirm their

conclusions by emphasizing that the electorate might change demographically in terms of

age and levels of partisanship, but election outcomes will remain stable (2000b). Using

regression analysis on 48 statewide elections, three of which were administered solely

through the post, Southwell and Burchett conclude all-mail balloting boosts turnout over

ten percentage points (2000a).

Like Southwell and Burchett, Karp and Banducci (2000) study Oregon election

data to determine if all-mail elections augment turnout. They disagree in their results,

however, by concluding the effect of all-mail balloting differs in "low salience" versus

"high salience" elections. Corresponding with Magleby's prediction, the authors find the

greatest effect in "low-salience" elections, specifically local and special statewide votes.

In "high-salience" elections (Presidential, midterm general elections, and Presidential






15


primaries), the researchers determine the effect of all-mail balloting on turnout is

insignificant.














CHAPTER 3
SWISS GOVERNMENT AND ELECTIONS

Like citizens in the United States, Swiss voters seem averse to the act of voting. In

Switzerland the executive is determined despite election results, as a seven-member

board represents the four major parties in the Federal Council. The chairperson of the

Federal Council is elected annually by its members. While the Swiss constitution calls

for the supervision of the Council by the legislature, the Council has gradually assumed a

role in directing the legislative process (Linder, 1994).

The power of the Swiss federalist structure is highly concentrated in the canton.

The Swiss constitution states that in the absence of an amendment by the people, all

future powers should be delegated to the canton. The rule has been effective, due to the

lack of an "implied powers" clause and the implication of direct democracy. (In order to

confer any new federal power, the majority of the people and the cantons must approve

via referendum, as well as both chambers of the legislature.) As evidence of the weakness

of the central government, federal tax share and expenditure as a proportion of total

receipts and expenditure pales in comparison to other industrial democracies (Linder

1994, p.43).

Every four years voters elect a national legislature divided into two equal bodies:

the Council of States, elected through fixed representation, and the National Council,

elected through List PR. (Before World War I, the National Council was elected through

a First-Past-the-Post system.) The nation is divided into twenty cantons and six half-

cantons, giving a total of twenty-six official "districts" through which legislative seats are









apportioned. The most recent addition to the federation was the canton of Jura in 1979.

Part of the canton of Bern since the Congress of Vienna, Jura seceded due to

longstanding linguistic and religious differences. The canton became official member of

the federation after the approval of the Swiss population through referendum.

Switzerland is one of the closest examples of direct democracy in the world. On

every bill decided before Parliament, the attainment of 50,000 signatures provides for the

calling of a popular referendum. The signatures of 100,000 citizens call for the vote for a

constitutional amendment. When the concept was introduced in the 19th century, the

founders of Switzerland believed the referendum would be used more for innovation and

less for inhibition. The opposite has turned out to be true, as Swiss voters "tend to

mobilize essentially to reject constitutional revisions and laws rather than accept them"

(Eschet-Schwarz 1989, p.255). The best example of the Swiss inclination for obstruction

is women's suffrage, which wasn't implemented until 1971.

Partially due to the high number of elections, Swiss cantons have intermittently

begun to implement postal voting, as displayed in Table 2 (Initiative & Referendum

Institute 2004). Two systems for postal voting exist in Switzerland: the simplified system

and the system of voting on request. In the latter, voters must request a ballot for every

vote. Under the simplified system, each Swiss citizen receives a ballot in the mail in the

weeks leading up to the election or referendum. For the purpose of this analysis I will

only focus on the simplified system since it provides the greatest reduction in voting costs

and the easiest way to cast a ballot. The canton of Basel-Landschaft became the first to

institute a simplified system in 1978, and all but two of the twenty-six had introduced it









Table 2 Introduction of Sim lified System of Early Voting
CANTON YEAR OF INTRODUCTION
Basel-Country 1978
St. Gallen 1979
Appenzell Inner-Rhodes 1979
Solothurn 1980
Thurgau 1985
Appenzell Outer-Rhodes 1988
Bern 1991
Aargau 1993
Luzern 1994
Nidwalden 1994
Zurich 1994
Basel-Town 1995
Friburg 1995
Geneva 1995
Glarus 1995
Graubunden 1995
Obwalden 1995
Schaffhausen 1995
Uri 1995
Zug 1997
Jura 1999
Schwyz 2000
Vaud 2002
Neuchatel 2003

by 2003. Unlike in-person early voting in the United States, the implementation of postal

voting in Switzerland was pervasive through each canton.

To an extent, Switzerland serves as an example to display the possibility of postal

voting in the U.S. due to the low levels of voter turnout, government structure (bicameral

legislature), and decentralized federalism shared by both nations. The nations are also

fairly heterogeneous, but in different ways. Where America is a "melting pot" in terms

of ethnicity, Switzerland comprises people groups partitioned by four official languages.

I also note similarities in terms of GDP per capital (Directorate of Intelligence, CIA

2006b). However, it is also necessary to document some of the differences that exist in









the nations as well, as Switzerland is not a perfect proxy for the effect of U.S. election

reforms on voter turnout. While the United States elects the House of Representatives

utilizing a First-Past-the-Post system, the Swiss National Council is elected by List PR.

As a result of PR elections, the party structure is much more volatile in Switzerland, as

evidenced by the rapid ascension of the Green Party. Also, the possibility of referendum

on any proposed legislation provides the Swiss people a much greater political voice than

their American counterparts. Demographically, Switzerland is more densely populated

even though more citizens live in rural area. Despite these differences, however, the

parallels between Switzerland and the United States can provide us some information on

the efficacy of electoral reforms.














CHAPTER 4
DATA AND HYPOTHESES

The purpose of this paper is to analyze the effects of in-person early voting and

postal voting on turnout in the U.S. states and Swiss cantons, respectively. These studies

examine the effects of these systems of early voting across both space and time. In order

to measure the effects I utilize an OLS regression.

United States

In the investigation on the United States the unit of analysis is the state-year

interaction in every general midterm election from 1982 to 2002. As Texas first

implemented IPEV in 1991, using those six elections provides a symmetry three

national elections with in-person voting, and three without. Due to state law regarding

unchallenged seats, coupled with the absence of a senatorial or gubernatorial election,

turnout data are not available from the 1982 Louisiana election, leading to a total sample

size of 299. Turnout is regressed on in-person early voting with dummy variables

representing the proportion of the population in each interaction with the ability to vote

early, as well as control variables for unemployment and political competition. Fixed

effects for individual states and years are also included and represented by dummy

variables. Voter turnout is calculated by two different methods: as a proportion of the

voting-age population (VAP) and as a proportion of the voting-eligible population. (VEP)

Voting-age population is chosen due to the prevalence of studies utilizing VAP in

analyzing turnout rates, and recent research (McDonald and Popkin, 2001) has given









precedent for analysis by VEP.1 Data used to calculate turnout are obtained from the

United States Election Project (McDonald 2006). Along with the succeeding controls I

include state and year dummy variables.

I note, however, differences between liberalized absentee voting laws and in-person

early voting. In most states, "early voting places] fewer demands on voters than

participation by absentee," (Hansen 2001, p.59) as voters must apply for an absentee

ballot well before the day they will actually complete their ballot. For the most part, the

costs of voting absentee are much greater than IPEV. In the few states which allow

voters to apply and fill out absentee ballots on the same day, research has shown turnout

is not affected unless accompanied by massive mobilization efforts by the parties. Even

with substantial mobilization, turnout is shown to increase only two percentage points

(Oliver 1996).

In-Person Early Voting

Of the nine states which had implemented early voting by 2002, seven had

instituted the procedure statewide. For these states, a dummy variable equal to one is set

to represent IPEV in the corresponding interactions. The states of California and Kansas,

however, have in-person early voting procedures but only in select counties. In these

states the variable representing in-person early voting is the proportion of the total

population exposed to IPEV. The population data are provided by the Census Bureau

website (United States Census Bureau 2004, 2005). In line with previous research, I

expect the effect of this variable to be insignificant.



1 I admit that calculating turnout as a percentage of the population of registered voters would provide better
analysis due to the aforementioned two-step process of voting. However, the incongruence of state and
county policies in purging registration rolls would necessarily lead to inaccurate results.









Political Competition

I control for political competition by creating two separate variables accounting for

the presence of gubernatorial and senatorial elections and their respective margins of

victory. For these elections, the data values represent the negative log of the margin of

victory. A greater value, therefore, represents a closer election. For the state-year

interactions without a gubernatorial or senatorial election, the value is zero. The data for

the gubernatorial elections are obtained from the Almanac of American Politics (Barone,

et al. 2004). The data for the senatorial elections are obtained from election results

provided by the Clerk of the House website (Carle 1995; Dendy 1987, 1991; Ladd 1983;

Trandahl 1999, 2003). In Texas in 1994, the ballot included two Senate elections. For

this case the logs are taken for the two margins of victory, and the datum is equal to the

negative of the sum of those two values.

I also control for the closeness of Congressional elections. I represent this by the

number of elections in a state decided by five percentage points or less, and setting that

number as a proportion of the total number of state Congressional districts. These data

are also found from the Clerk of the House website.

Corresponding to classic literature (Downs 1957; Riker & Ordeshook 1968), I

expect the Gubernatorial Competition, Senatorial Competition, and Congressional

Competition to have a positive and significant relationship with turnout.

Unemployment

The Unemployment variable represents the economic environment across the

individual states. Research has shown that economic adversity can lead to "falling out"

from the political system, and effectively reduces an individual's propensity to participate

(Wolfinger and Rosenstone 1980; Rosenstone 1982). Rosenstone and Hansen conclude









"the money, time, and energy spent combating extreme economic adversity provide

payoffs that are more immediate and valuable than the benefits that might be gained from

investing in electoral politics" (1993, p. 135). They find the unemployed are 8.5% less

likely to vote in midterm elections, confirming earlier research on the effect of

unemployment. With data provided by the Bureau of Labor Statistics website (Bureau of

Labor Statistics 2006), this variable is expected to have a negative, significant

relationship with turnout.

National Macropartisan Strength

Previous research also indicates the importance of partisanship in analyzing

turnout. Bartels (2000) comes to two conclusions about partisanship: since the 1970s,

voters have become more partisan, and the impact of partisanship on vote choice has

significantly increased. While partisanship's effect on vote choice is not germane to this

paper, I am concerned with the overall increase in partisanship in among American

citizens. Verba and Nie (1972) conclude that strong party identifiers are more likely to

vote than weak identifiers, who in turn are more likely to vote than independents.

Therefore, the increase in macropartisanship beginning in the 1970s provided a positive

bias to aggregate turnout rates in the United States.

From the National Election Studies, as a measure of partisanship I have created a

national-level variable representing National Macropartisan Strength. Utilizing the zero-

to-six values for partisanship from the NES, I "folded" the data, so a "Strong Republican"

and "Strong Democrat" would both be represented by a value of "3". I did the same for

"Moderate" (2) and "Weak" (1) identifiers, as well as "Independents" (0). From those, I

calculated the mean value of each year and used it to represent macropartisanship at the









national level. I expect NationalMacropartisan Sn engili to be positively correlated with

turnout.

Election-Day Registration

As mentioned earlier, previous research has concluded the strong and significant

impact of registration reforms on aggregate turnout. Using data from Fitzgerald (2005), I

control for election-day registration (EDR) in the model. For the purpose of this analysis

I count North Dakota, which has no registration laws, as a state with election-day

registration. As EDR provides those not registered on the day of the election the

opportunity to participate, so does the registration policy of North Dakota.

United States Regressions

The following equations summarize the model for turnout in the United States as

proportions of both the voting-age population and the voting-eligible population. I

control for economic and political variables, as well as fixed effects for states and years

through dummy variables.

Equation 1
Turnout(VAP) = a + Pi(IPEV) + 32(Gubernatorial Competition) + 33(Senatorial
Competition) + 34(Congressional Competition) + 35(National Macropartisan
Strength) + P6(EDR) + (State Dummies) + (Year Dummies) + e

Equation 2
Turnout(VEP) = a + P31(IPEV) + 32(Gubernatorial Competition) + 33(Senatorial
Competition) + 34(Congressional Competition) + 35(National Macropartisan
Strength) + P6(EDR) + (State Dummies) + (Year Dummies) + e

Switzerland

In the Swiss investigation the unit of analysis is the canton-year interaction in the

parliamentary elections in each canton from 1975 to 2003. I choose to use elections back

to 1975 for two reasons: it was the second election after women were granted the right to









vote (accounting for a "novelty" effect), and the method of calculating Swiss

unemployment data changed between '71 and '75 (as to make constant the effect of

unemployment on turnout levels). The unavailability of turnout statistics for three of the

interactions, along with missing unemployment data for two, leads to a final sample size

of 202. (The establishment of Jura in 1979 necessarily leaves only 25 data points in the

1975 cycle.) Turnout as a percentage of VAP is regressed on the effect of postal voting.

(Swiss citizens were not allowed to vote until the age of 20 before 1991, when legislation

changed the age to 18. This modification was accounted for in the creation of the voting-

age population.) Calculating turnout as a factor of the voting-eligible population is

unnecessary, as the number of ineligible people is small.2 The voting-age population

and vote totals were acquired from email-correspondence with employees of the Swiss

Federal Statistical Office. The data can be found in French and German from their

website (Swiss Federal Statistical Office 2006). Similar to the U.S. analyses, I control for

both economic and political factors and include fixed effects for canton and year dummy

variables.

Since 1904 compulsory voting has existed in the canton of Schaffhausen. While

the penalty of nonvoting is only three Swiss francs (or $2.44), I expect turnout in

Schaffhausen to be comparatively higher than the other cantons. As a result, I conduct

two analyses, one excluding Schaffhausen. Along with the succeeding controls I include

canton and year dummy variables.





2 Information obtained from email correspondence with Elisabeth Willen of the Swiss Federal Statistical
Office, April 26, 2006









Postal Voting

As of 2003, all but two of the twenty-six cantons had implemented a simplified

system of postal voting. When these policies were established, the ability to vote by mail

permeated the entire populations of the respective cantons. As a result, the variable

representing postal voting is a simple "toggle" dummy of either one (existence of postal

voting) or zero (absence). As previously noted, these data are obtained from

SwissWorld.

Unemployment

As in the U.S. analysis, I include cantonal levels of unemployment to account for

the economic environment. I expect this effect to be negative. The data were obtained

from email correspondence with the representatives of the Swiss Federal Statistical

Office.

Political Competition

Characteristic of a List PR parliamentary system, many parties exist and hold some

power in Switzerland. One measure utilized to represent political competition is the

difference between the vote shares for the two leading parties in the election at the

cantonal level. Previous research in a cross-national study indicates that a ten point

difference between the leading and second parties reduces turnout 1.4 points (Blais 2000,

p. 30). I use the same measure in my analysis with data obtained from Swiss Politics

(Swiss Broadcasting Company 2006). Another measure used is the number of parties in

each canton winning seats in an election. A previous cross-national study revealed that a

greater number of parties is positively correlated with higher turnout, as electors have a

greater selection from which to choose (Blais & Carty 1990, from Blais 2000). However,

party strength is not disseminated equally throughout the cantons, and as a proxy I use the









number of parties in each canton which won legislative seats, as obtained from the Swiss

Federal Statistical Office. In the regression, I expect both Vote Margin and Number of

Winning Parties to be positively correlated with turnout.

Swiss Regressions

The following equation summarizes the model for turnout in Switzerland as a

proportion of the voting-age population. I control for economic and political variables, as

well as fixed effects for cantons and years through "toggle" dummy variables.

Equation 3
Turnout(VAP) = a + 13i(Postal Voting) + 32(Vote Margin) + 33(Number of Winning
Parties) + 34(Canton-Level Unemployment) + (Canton Dummies)3 + (Year
Dummies) + e


3 One regression will include Schaffhausen, one will exclude















CHAPTER 5
RESULTS

In-Person Early Voting

Corresponding to previous literature, Tables 3 and 4 show that the existence of in-

person early voting is not positively correlated with voter turnout. As a percentage of

both VAP and VEP, turnout is actually negatively correlated with IPEV, although the

coefficient is insignificant. The variables representing political competition, along with

EDR, all have a positive relationship with turnout, as expected. However, none of the

variables are significant. The partisanship variable is also positive and insignificant in

both analyses. Unexpectedly, the unemployment variable is positively correlated with

turnout. But like the others, the effect is insignificant.

Table 3 Effect of In-Person Early Voting in the States (VAP)
Coefficient SE Significance
IPEV -.030 .020 .129
Gubernatorial .012 .009 .183
Competition
Senatorial .000 .000 .749
Competition
Congressional .009 .023 .715
Competition
National .068 .245 .781
Macropartisanship
Strength
State-Level .003 .003 .273
Unemployment
EDR .018 .027 .508
R-square= 0.594; N 299; Estimated by OLS
Fixed Effects for State and Year Dummies not included in Table









Table 4 Effect of In-Person Early Voting in the States
Coefficient SE Significance
IPEV -.023 .020 .259
Gubernatorial .013 .009 .155
Competition
Senatorial .000 .000 .707
Competition
Congressional .009 .024 .702
Competition
National .069 .251 .783
Macropartisanship
Strength
State-Level .004 .003 .163
Unemployment
EDR .012 .028 .660
R-square= 0.566; N 299; Estimated by OLS
Fixed Effects for State and Year Dummies not included in Table

The results are very surprising. While I am not astonished to find that in-person

early voting does not catalyze aggregate turnout rates, the nonnegative correlation is

frustrating. I am also disappointed with the ineffectiveness of the political and economic

variables. While the sign of those coefficients are all in the expected positive direction,

the high standard errors associated with the coefficients indicate their ineffectiveness in

predicting turnout in these models. Due to importance of registration reforms on

aggregate turnout levels, I was also surprised to discover the insignificance of election-

day registration. Combined with the fixed effects for state and year dummies, the

weakness of all of the variables included in the regressions leads to low R-square values

(.594 and .566 for the VAP and VEP models, respectively), indicating the feebleness of

the model in predicting turnout rates.

Postal Voting

Much like voting-by-mail in the United States, Tables 5 and 6 indicate the effect of

postal voting in Switzerland is positive and substantial. At the cantonal level, I expect









the incidence of postal voting to augment turnout approximately six percentage points.

Vote Margin, the variable representing the difference in vote share between the leading

and second parties has an enormous effect on turnout. Every ten point decline in margin

leads to increased turnout of 2.3 percentage points. The other variable measuring party

competition the number of parties in each canton securing legislative seats in the

National Council is actually negative. It is insignificant in both regressions, however.

The Unemployment variable, as in the U.S. regressions, is surprisingly positive and

insignificant.

Table 5 Effect of Postal Voting in the Cantons
Coefficient SE Significance
Postal Voting .057 .014 .000
Vote Margin -.234 .029 .000
Number of -.006 .006 .296
Winning Parties
Canton-Level .003 .006 .589
Unemployment
R-square 0.810; N=202; Estimated by OLS
Fixed Effects for Canton and Year Dummies not included in Table

Table 6 Effect of Postal Voting in the Cantons (Schaffhausen excluded)
Coefficient SE Significance
Postal Voting .060 .014 .000
Vote Margin -.233 .030 .000
Number of -.006 .006 .285
Winning Parties
Canton-Level .003 .006 .579
Unemployment
R-square 0. 768; N 194; Estimated by OLS
Fixed Effects for Canton and Year Dummies not included in Table

Unlike in-person early voting in the U.S., I find the effect of postal voting on

aggregate turnout levels in Switzerland to be strong. I cannot say I am surprised with the

value of the coefficient corresponding to the Number of Winning Parties. As mentioned

previously, the number of competitive parties in a district has a positive effect on turnout.









While it is obvious that the number of winning parties is not equivalent to the number of

competitive parties, I believed that the high correlation between two would allow the

number of winning parties to serve as a suitable proxy. This proved not to be the case.

The variable representing Vote Margin, however, resulted in being positive and

significant, so I believe political competition was successfully controlled in the

regression. Unfortunately, the coefficient corresponding to the variable representing

Cantonal-Level Unemployment is positive, yet insignificant. It appears that the

Unemployment variables in all four regressions do an inadequate job in representing the

economic environment. Unlike the U.S. regressions, the corresponding R-square values

in the Swiss regressions (.810 and .768 for the regression including all cantons and the

regression excluding Schaffhausen, respectively) indicate the models' strength in

predicting turnout.

One interesting finding from the Swiss analyses involves the effect of compulsory

voting. In the regression including the canton of Schaffhausen, the corresponding

coefficient is strong, positive, and significant. (b=. 192) It is the highest coefficient for all

canton dummies, with the next-strongest positive coefficient for a canton (Tessin) coming

in almost seven points lower. (b=. 124) This finding indicates that a government does not

need to institute a forceful penalty for nonvoting to produce higher turnout. The desired

effect can be accomplished by a weak penalty, as evidenced by a fine of only three Swiss

Francs in the canton of Schaffhausen. Considering the existence of a compulsory-voting

policy, it should come as no surprise that turnout in Schaffhausen is almost twenty points

higher than a given canton.














CHAPTER 6
DISCUSSION

The most salient finding from this study is the strong effect of postal voting on

turnout. Previous research on postal voting in the United States has indicated it provided

a boon to turnout, but analyses were conducted on a limited scope. Here, I study the

effect of postal voting across time and space in a nation, like the U.S., with a

heterogeneous culture and comparatively low turnout.

In line with previous research on in-person early voting, I find no significant effect

of IPEV on aggregate turnout levels. The natural question, then, is why postal voting

increases turnout yet in-person early voting does not. I believe the answer stems from the

extent by which each system reduces the costs of voting. In an IPEV system, a

prospective voter must still endure the process of traveling to the polls. Depending on the

early voting location, voting before Election Day could actually be more time-consuming

than voting on Election Day, depending on how early voting sites are designated and the

manner in which voters are assigned to those locations. A populous county which only

opens IPEV at one location could actually result in increasing the time needed to cast a

ballot, due to such factors as longer waiting lines and increased drive time. Under postal

voting, however, all transportation costs are obviated as all that is required (outside of

completing the ballot) to vote by post is a visit to one's mailbox.

An obvious worry of postal voting is the perceived possibility of voter fraud.

Specifically, opponents of postal voting fear "that someone other than the addressee will

use undelivered or duplicate ballots" (Southwell & Burchett 1998, p.348). However,









committing voter fraud through postal voting would require forging the registered voter's

signature. In Oregon, Southwell and Burchett note poll workers are required to compare

the signature on the outside of the envelope to the signature on the voter roll. I believe

similar rules regarding postal voting would render the possibility of voter fraud through

the mail no greater than the possibility of fraud at a standard Election Day polling

location. Another concern associated with voter fraud is the possibility of vote influence.

As citizens fill out their ballots at home, critics believe the presence of an individual close

to the voter can unduly influence the voter's choice. However, research by Southwell

and Burchett showed this is simply not the case. In a survey of 1225 respondents,

researchers found only three (0.3%) vote-by-mail voters said the presence of a person in

the room "made them feel pressured to vote a certain way. Of those three, only one

indicated the pressure caused him/her to vote differently (Southwell and Burchett 1997,

p.54).

Previously mentioned, one of the dilemmas of low turnout is the associated

socioeconomic skew between voters and nonvoters. Piven and Cloward (1989) theorized

that increasing turnout would aid in ameliorating the problem. Unfortunately, research

on various forms of voting reforms indicates that increased turnout does nothing to solve

this predicament. Karp and Banducci (2001) conclude that early voters are more likely to

be educated, active in politics, and partisan than traditional voters. In an analysis of

liberalized registration laws, Mitchell and Wlezien (1995) discover the change to the

electorate in terms of income is menial. Another study finds that postal voting works to

retain registered nonvoters, therefore stimulating individuals belonging to groups already

likely to participate (Berinsky et al. 2001). Berinsky (2005) actually concludes that









voting reforms will exacerbate the problem of socioeconomic skew by escalating, rather

than lessening the biases in the electorate. Nevertheless, benefits associated with high

turnout outside of socioeconomic grounds still provide ample reasons to find ways to

induce more citizens to vote.

Future research should focus on the effectiveness of postal voting in ameliorating

the problem of socioeconomic skew in political participation. It would be interesting to

see research that utilizes Swiss survey data and demonstrates the effect of all-mail

balloting on the socioeconomic composition of the electorate. While I have not found any

evidence of SES skew specific to Switzerland, previous research on international turnout

has indicated the importance of resources (Verba et al. 1978), which gives cause to

believe in the existence of bias. Also, future research should concentrate on the

administrative side of in-person early voting. Gronke, et al. (2005) analyze the long

waits and voter intimidation that occurred often in early voting locations in Florida in

2004, but focus more on the demographics of early voters compared to election-day

voters. Researchers should look to determine the optimal locations for early voting sites,

(e.g. county elections offices vs. shopping malls) as well as the proper distribution of

voting machines, in order to completely minimize the costs associated with IPEV. This

will aid in depleting the "learning curve" associated with enacting in-person early voting.

Currently, it is difficult to ascertain the effect of all-mail balloting in the United

States, as only one general election (Oregon 2004) has been conducted by postal voting.

However, I hope Switzerland has provided a framework to demonstrate the benefits to

turnout rates of conducting all elections through the mail. As a result of this analysis, I

believe instituting a postal voting system across the United States population would






35


induce many more citizens to cast a ballot. Along with other incentives stemming from

of all-mail balloting, such as the significantly reduced costs of administering elections, I

pose that converting all general elections in the U.S. to vote-by-mail would be

beneficiary to all parties involved, and look forward to the day all American citizens cast

a ballot through the mail.















LIST OF REFERENCES


Aldrich, John H. 1993. "Rational Choice and Turnout." American Journal of Political
Science 37 (February):246-78

Barone, Michael, Grant Ujifusa, and Douglas Matthews. 2004. Almanac ofAmerican
Politics. New York: E.P. Dutton.

Bartels, Larry. 2000. "Partisanship and Voting Behavior." American Journal of Political
Science 44 (January):35-50

Berelson, Bernard, Paul Lazarsfeld, and William McPhee. 1954. Voting: A Study of
Opinion Formulation in a Presidential Campaign. Chicago: University of
Chicago Press.

Berinsky, Adam J. 2005. "The Perverse Consequences of Electoral Reform in the United
States." American Politics Research 33 (July):471-491.

Berinsky, Adam J., Nancy Burns and Michael W. Traugott. 2001. "Who Votes by Mail?
A Dynamic Model of the Individual-Level Consequences of Voting-By-Mail
Systems." Public Opinion Quarterly 65 (Summer): 178-97.

Blais, Andre. 2000. To Vote or Not to Vote: The Merits and Limits of Rational Choice
Theory. Pittsburgh: University of Pittsburgh Press.

Blais, Andre, and Kenneth Carty. 1990. "Does Proportional Representation Foster Voter
Turnout?" European Journal of Political Research 18:167-81.

Brady, Henry E., Kay Lehman Schlozman, and Sidney Verba. 1995. "Beyond SES: A
Resource Model of Political Participation." American Political Science Review 89
(June):271-294.

Brians, Craig Leonard, and Bernard Grofman. 2001. "Election Day Registration's Effect
on U.S. Voter Turnout." Social Science Quarterly 82 (March): 170-83.

Brown, Robert D., Robert A. Jackson, and Gerald C. Wright. 1999. "Registration,
Turnout, and State Party Systems." Political Research Quarterly 52
(September):463-79.









Brown, Robert D., and Justin Wedeking. 2006. "People Who Have Their Tickets But Do
Not Use Them: "Motor Voter," Registration, and Turnout Revisited." American
Politics Research 34 (July): 479-504.

Bureau of Labor Statistics. 2006. "Local Area Unemployment Statistics." Accessed
April, 2006. http://data.bls.gov/cgi-bin/surveymost?la

Burnham, Walter Dean. 1980. "The Appearance and Disappearance of the American
Voter." in Electoral Participation: A Comparative Analysis, ed. Richard Rose.
Beverly Hills, CA: Sage.

Burnham, Walter Dean. 1987. "The Turnout Problem." in Elections American Style, ed.
A. James Reichley. Washington, DC: Brookings Institution.

Campbell, Angus, Philip Converse, Warren Miller, and Donald Stokes. 1960. The
American Voter. Chicago: University of Chicago Press.

Carle, Robin H. 1995. "Statistics of the Congressional Election of November 8, 1994."
Office of the Clerk- U.S. House of Representatives. Accessed April, 2006.
http://clerk.house.gov/members/electionInfo/1994/94Stat.htm

Davis, Susan. "Voting by Mail Could Improve American Democracy." Roll Call, June
22, 2005.

Dendy, Dallas L. Jr. 1987. "Statistics of the Congressional Election of November 4,
1986" Office of the Clerk- U.S. House of Representatives. Accessed April, 2006.
http://clerk.house.gov/members/electionInfo/1986election.pdf

Dendy, Dallas L. Jr. 1991. "Statistics of the Congressional Election of November 6,
1990" Office of the Clerk- U.S. House of Representatives. Accessed April, 2006.
http://clerk.house.gov/members/electionInfo/1990election.pdf

Directorate of Intellingence, CIA. 2006(a). "Field Listing Literacy." in World
Factbook. Accessed April, 2006.
http://www.cia.gov/cia/publications/factbook/fields/2103.html

Directorate of Intellingence, CIA. 2006(b). "Rank Order GDP per capital (PPP)" in
WorldFactbook. Accessed April 2006.
http://www.cia.gov/cia/publications/factbook/rankorder/2004rank.html

Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper & Row.

Election Reform Information Project. 2006. Election Reform: What's Changed, What
Hasn 't and Why? Accessed April, 2006.
http://www.electiononline.org









Eschet-Schwarz, Andre. 1989. "Political Participation in Swiss Referenda at Federal and
Cantonal Levels: 1879-1981." Political Behavior 11 (September):255-272.

Fitzgerald, Mary. 2005. "Greater Convenience but not Greater Turnout: The Impact of
Alternative Voting Methods on Electoral Participation in the United States."
American Politics Research 33 (November): 842-67.

Franklin, Mark N. 1996. "Electoral Participation" in Comparing Democracies: Elections
And Voting in Global Perspective, ed. Laurence LeDuc, Richard G. Niemi, and
Pippa Norris. Thousand Oaks, CA: Sage.

Franklin, Mark N. 1999. "Electoral Engineering and Cross-National Turnout Differences:
What Role for Compulsory Voting?" British Journal of Political Science 29
(January):205-216.

Franklin, Mark N., and Wolfgang P. Hirczy de Mino. 1998. "Separated Powers, Divided
Government, and Turnout in US Presidential Elections." American Journal of
Political Science 42 (January): 316-326.

Gimpel, J.G., and J.E. Schuknecht. 2003. "Political Participation and the Accessibility of
the Ballot Box." Political Geography 22 (June):471-488.

Gronke, Paul. 2005. "Ballot Integrity under Oregon's Vote by Mail System." Report
Prepared for the Commission of Federal Election Reform, Washington DC.

Gronke, Paul, Benjamin Bishin, Daniel Stevens, and Eva Galanes-Rosenbaum. 2005.
"Early Voting in Florida, 2004." Paper prepared for the Annual Conference of the
American Political Science Association, Washington DC. Accessed June, 2006.
http://www.reed.edu/-gronkep/docs/GronkeBishinStevensGalanes-
Rosenbaum.APSA.2005.pdf

Gronke, Paul, and Eva Galanes-Rosenbaum. 2005. "Getting out the Early Vote: Lessons
for Progressives." Paper presented at the annual meeting of the American Political
Science Association. Accessed July, 2006.
http://earlyvoting.net/resources/GronkePTC.pdf

Hamilton, Randy. 1988. "American All-Mail Balloting: A Decade's Experience." Public
Administration Review 48 (September- October):860-866.

Hansen, John Mark. 2001. To Assure Pride and Confidence in the Electoral Process.
Final Report of the National Commission on Electoral Reform, University of
Virginia. Accessed June, 2006.
http://millercenter.virginia.edu/programs/natl_commissions/commission_final_re
port/task_force_report/task_force_complete.pdf









Haspel, Moshe and H. Gibbs Knotts. 2005. "Location, Location, Location: Precinct
Placement and the Costs of Voting." Journal of Politics 67 (May):560-573.

Highton, Benjamin. 1997. "Easy Registration and Voter Turnout." Journal of Politics 59
(May):565-75.

Highton, Benjamin. 2004. "Voter Registration and Turnout in the United States."
Perspectives on Politics 2 (September): 507-15.

Highton, Benjamin, and Raymond E. Wolfinger. 1998. "Estimating the Effects of the
National Voter Registration Act of 1993." Political Behavior 20 (June):79-104.

Hill, Kim Quaile, Jan E. Leighley, and Angela Hinton-Andersson. 1995. "Lower-Class
Mobilization and Policy Linkage in the U.S. States." American Journal of
Political Science 39 (February):75-86

Hirczy de Mino, Wolfgang. 1994. "The Impact of Mandatory Voting Laws on Turnout: A
Quasi-Experimental Approach." Electoral Studies. 13 (March):64-76.

Initiative & Referendum Institute Europe (IRI), Amsterdam. 2004. "Postal Voting" in
SwissWorld. Accessed April, 2006.
http://www.swissworld.org/dvd_rom/eng/direct_democracy_2004/content/votes/p
ostal_voting.pdf

International Institute for Democracy and Electoral Assistance. 2006. "Voter Turnout"
Accessed April, 2006. http://www.idea.int/vt/

Jackman, Robert W. 1987. "Political Institutions and Voter Turnout in the Industrial
Democracies." American Political Science Review 81 (June):405-424.

Jackman, Robert W., and Ross A. Miller. 1995. "Voter Turnout in the Industrial
Democracies During the 1980s" American Political Science Review 81
(June):405-423.

Karp, Jeffrey A., and Susan A. Banducci. 2000. "Going Postal: How All-Mail Elections
Influence Turnout." Political Behavior 22 (September):223-39.

Karp, Jeffrey A., and Susan A. Banducci. 2001. "Absentee Voting, Participation, and
Mobilization." American Politics Research 29 (March): 183-95.

Key, V.O., Jr. 1949. Swuthw iin Politics in State and Nation. New York: Vintage Books.

Knack, Stephen. 1995. "Does 'Motor Voter' Work? Evidence from State-Level Data."
Journal of Politics 57 (August):796-811.









Ladd, Thomas E. 1983. "Statistics of the Congressional Election of November 2, 1982."
Office of the Clerk- U.S. House of Representatives. Accessed April, 2006.
http://clerk.house.gov/members/electionlnfo/1982election.pdf

Leighley, Jan E., and Jonathan Nagler. 1992. "Socioeconomic Class Bias in Turnout,
1964-1988: The Voters Remain the Same." American Political Science Review 86
(September):725-36.

Linder, Wolf. 1994. Swiss Democracy. London: Macmillan.

Lijphart, Arend. 1997. "Unequal Participation: Democracy's Unresolved Dilemma."
American Political Science Review 91 (March): 1 -14.

Magleby, David. 1987. "Participation in Mail-Ballot Elections." Western Political
Quarterly 40 (March):79-91.

Martinez, Michael D., and David Hill. 1999. "Did Motor Voter Work?" American
Politics Quarterly 27 (July):296-315.

McDonald, Michael. 2006. "Voter Turnout." United States Election Project. Accessed
April, 2006. http://elections.gmu.edu/voter tumout.htm

McDonald, Michael, and Samuel L. Popkin. 2001. "The Myth of the Vanishing Voter."
American Political Science Review 95 (December):963-74.

Mill, John Stuart. [ 1861] 1958. Considerations on Representative Government. New
York: Liberal Arts Press.

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 (June): 179-202.

Neeley, Grant W., and Lilliard E. Richardson, Jr. 2001. "Who is Early Voting? An
Individual Level Examination." Social Science Journal 38:381-392.

Newby, Brian D. 2006. "Vote in Advance In Person" in Johnson County Election
Office. Accessed June, 2006.
http://www.jocoelection.org/AdvanceVotingPrimary2006.htm

Oliver, Eric. J. 1996. "The Effects of Eligibility Restrictions and Party Activity on
Absentee Voting and Overall Turnout." American Journal of Political Science 40
(May): 498-513.

Oregon Secretary of State. 2006. "A Brief History of Vote by Mail." Accessed April,
2006. http://www.sos.state.or.us/elections/vbm/history.html









Piven, Frances Fox, and Richard A. Cloward. 1988. Why Americans Don't Vote. New
York: Pantheon Books.

Piven, Frances Fox, and Richard A. Cloward. 1989. "Government Statistics and
Conflicting Explanations of Nonvoting." PS: Political Science and Politics 22
(September):580-588.

Powell, G. Bingham., Jr. 1986. "American Voter Turnout in Comparative Perspective."
American Political Science Review 80 (March): 17-43.

Richardson, Lilliard E. Jr., and Grant W. Neeley. 1996. "The Impact of Early Voting on
Turnout: The 1994 Elections in Tennessee." State and Local Government Review
28 (Fall): 173-179.

Riker, William H., and Peter C. Ordeshook. 1968. "A Theory of the Calculus of Voting."
American Political Science Review 62 (March):25-42.

Rosenstone, Steven J. 1982. "Economic Adversity and Voter Turnout." American
Journal of Political Science 26 (February):25-46.

Rosenstone, Steven J., and John Mark Hansen. 1993. Mobilization, Participation, and
Democracy in America New York: Macmillan.

Rosenstone, Steven J., and Raymond E. Wolfinger. 1978. "The Effect of Registration
Laws on Voter Turnout." American Political Science Review 72 (March):22-45.

Sola, Lester. 2006. "Early Voting Sites" in Miami-Dade County Elections. Accessed
June, 2006. http://elections.metro-dade.com/voting_absent_earlyvoting-sites.asp

Southwell, Priscilla L. 2004. "Five Years Later: A Re-Assessment of Oregon's Vote by
Mail Electoral Process." PS: Politics and Political Science 98 (March): 89-93.

Southwell, Patricia L., and Justin Burchett. 1997. "Survey of Vote-by-Mail Senate
Election in the State of Oregon." PS: Political Science and Politics. 30
(March):53-57.

Southwell, Priscilla L., and Justin Burchett. 1998. "Vote-By-Mail in the State of
Oregon." Willamette Law Review 34 (Spring):345-56.

Southwell, Priscilla L., and Justin Burchett. 2000(a). "The Effect of All-Mail Elections
on Voter Turnout." American Politics Quarterly 28 (January):72-79.

Southwell, Priscilla L., and Justin Burchett. 2000(b). "Does Changing the Rules Change
the Players? Vote-by-Mail and the Composition of the Electorate." Social Science
Quarterly 81 (September):837-45.









Squire, Peverill, Raymond E. Wolfinger, and David P. Glass. 1987. "Residential Mobility
and Voter Turnout." American Political Science Review 81 (March):45-66.

Stein, Robert A. 1998. "Introduction: Early Voting." Public Opinion Quarterly 62
(Spring):57-69.

Stein, Robert A., and Patricia Garcia-Monet. 1997. "Voting Early but not Often." Social
Science Quarterly. 78 (September):657-671.

Swiss Broadcasting Company. 2006. "Geopolitics" in SwissPolitics. Accessed April,
2006.
http://www.swisspolitics.org/en/politnavigator/index.php?page=geopolitics_suche

Swiss Federal Statistical Office. 2006. Accessed April, 2006.
http://www.bfs.admin.ch/bfs/portal/en/index.html

Tingsten, Herbert. 1937. Political Behavior: Studies in Election Statistics. London: P.S.
King & Son.

Trandahl, Jeff. 1999. "Statistics of the Congressional Election of November 3, 1998."
Office of the Clerk- U.S. House of Representatives. Accessed April, 2006.
http://clerk.house.gov/members/electionInfo/1998/98Stat.htm

Trandahl, Jeff. 2003. "Statistics of the Congressional Election of November 5, 2002."
Office of the Clerk- U.S. House of Representatives. Accessed April, 2006.
http://clerk.house.gov/members/electionInfo/2002/2002Stat.htm

United States Census Bureau. 2004. "Intercensal Estimates: Time Series of Intercensal
Estimates by County." Accessed April, 2006.
http://www.census.gov/popest/archives/2000s/vintage_2001/CO-EST2001-12/

United States Census Bureau. 2005. "Annual Estimates of the Population for the United
States and States, and for Puerto Rico: April 1, 2000 to July 1, 2005." Accessed
April, 2006. http://www.census.gov/popest/states/NST-ann-est.html

Verba, Sidney, and Norman H. Nie. 1972. Participation in America. Chicago: University
of Chicago Press.

Verba, Sidney, Norman H. Nie, and Jae-On Kim. 1978. Participation and Political
Equality: A Seven-Nation Comparison. Cambridge: Cambridge University Press.

Wolfinger, Raymond E., and Steven J. Rosenstone. 1980. Who Votes? New Haven, CT:
Yale University Press.















BIOGRAPHICAL SKETCH

A 2004 graduate of the University of Alabama, I enrolled in the Department of

Political Science at the University of Florida in August of that year. I soon became

intrigued with the subject of political behavior, and in my second year embarked on a

study of voting reforms, which eventually blossomed into my master's thesis. In August

I will attend the University of North Carolina, where I will begin study toward a Ph.D. in

economics, focusing on political economy.