It is a common belief that whether a person votes is largely determined by the costs and benefits associated with voting. Downs (1957) said “[i]f the returns outweigh the costs, he votes; if not, he abstains.” Is there a significant increase in voter turnout when steps are taken to lower the costs of voting? The major goals of convenience voting are to increase turnout, save money and increase the accuracy of vote tabulation. The effect on voter turnout is what will be looked at in this paper. It is important to note that there are two distinct ways in which the electorate can grow: mobilization and retention (Berinsky, Burns, and Traugott 2001). Mobilization – the recruitment of new voters coming to the polls – is the type of growth that most commonly comes to mind when thinking of increasing voter turnout. However, it is important to differentiate mobilization from retention, defined as the ability for a certain voting method to keep the voters who have already come to the polls. If a voting method only increases turnout via retention and new voters are not mobilized the electorate will not expand to include voters from groups with lower participation levels such as minorities, those with low education levels and low socioeconomic statuses.
Often times voting reformers will argue that an important step towards increasing voter turnout is the reduction of barriers between voters and the ballot. Based on the rational choice theory this may seem to be a good enough reason to assume that voter participation would increase. However, most scholarly literature on the subject of convenience voting agrees that the use of these methods does not have a great impact on voter turnout. Any gains in turnout are minimal and are seen in those who are already predisposed to voting.
Studies of the 2000 Arizona Democratic primary by Solop (2001) and Gibson (2001) looked at the demographic and attitudinal differences between those voters who used an electronic voting system and those who used a more traditional method to cast their votes. In Solop’s study he looked at three telephone surveys conducted by the National Science Foundation. The first was a 1,200-person cross sectional survey of Arizona adults, the second was a 1,200-person survey of Arizona’s registered Democrats, and lastly a 783-person post election panel of which 318 people participated in the primary election. These studies showed that electronic voting was best received by better educated voters, those coming from high income households and middle aged voters. It was also shown that voters who used the electronic vote had a higher overall efficacy than the entire group of registered Democrats. This was found by looking at three standard efficacy questions1from the National Election Study and comparing the mean rating of all registered Democrats and the mean rating of electronic voters.
Also looking at the use of an electronic voting system, Roseman and Stephenson (2005) studied the relationship between the change in turnout and the presence of elderly, poorly educated and minority voters. Their study focused on the statewide equipment upgrade in Georgia after the 2000 election. The examination of the data surrounding the turnout rates between the1999 and 2002 gubernatorial elections showed that there was a negative effect on the turnout of elderly voters.
It is clear by looking at previous research in this field that convenience voting has not had as positive an impact on voter turnout that many thought it would. In an attempt to prove that convenience voting alone will not increase voter turnout I will be looking at turnout data from the National Annenberg Elections Survey as well as from other sources such as Oregon’s Secretary of State Website and the United States Elections Project. I will also be comparing turnout data between states with different methods of voting as well as comparing data before and after states implement certain voting reforms.
Convenience voting can take on many different forms ranging from a citizen going to a polling place a month before Election Day to cast their ballot, to a citizen voting remotely from a Personal Computer during their lunch break on Election Day. There are three major types of convenience voting that will be covered in this paper: early in-person voting, voting by mail, and electronic voting. The first two listed have been used for many years in a number of binding elections where as electronic voting is still in the experimental phase in the United States. No matter the subtle differences between these methods it is clear that the problem of low turnout is not procedural, it is motivational. Turnout will not be significantly raised and sustained at a higher level by simply relaxing administrative rules, the electorate must be motivated into having a personal interest in politics and government.
A study by Stein, Owens and Leighley (2005) find that individuals with greater opportunity to participate in early voting are no more likely to vote unless such opportunities are exploited by partisan mobilization efforts. Most research shows that parties focus on known voters and their own party members when attempting to mobilize the electorate. This focus on the core constituencies of a given party may explain why research shows limited effects of convenience voting reforms on voter turnout. The coupling of convenience voting reforms and mobilization efforts by political parties and other entities may prove to be an effective way to increase turnout if the elite’s mobilization strategies broaden to include non-voters and independents.
1. Early In-Person Voting: The largest distinction between early in-person voting and other convenience voting methods is that those who partake in early in-person voting will have to go to a polling place during specified hours to cast their ballot. It is a common belief that getting to the polls is a significant cost associated with voting which in turn lowers turnout. Therefore an in-person system may only partially lower this burden (Gronke, Galanes-Rosenbaum, & Miller 2007). It is important to note that not all early voting systems were created equally. In many cases early voters have to go to a local elections office to cast their votes, but it is becoming more common to have satellite locations available to further increase the convenience to the voters. The availability of multiple satellite locations over just one elections office per county is not the only difference between the different early voting systems. Some districts allow their citizens to vote up to forty-five days before Election Day whereas some areas only allow early voting one week ahead of time.
Voting early is becoming a popular trend, according to the NAES survey conducted between October 14, 2008 and October 27, 2008 23.4% of participants had already cast their vote with an additional 38.2% reporting that they plan on voting early (National Annenberg Election Study 2008). Early estimates suggest that nearly one third of ballots were cast early in 2008, up from approximately 23 percent in 2004 and 14% in 2000. These numbers become more impressive when looking at the fact that not all states offer early voting. As of the 2000 election only thirty-one states permitted in-person early voting at election offices (National Conference of State Legislatures 2008) leaving nineteen states without an in-person early voting option.
In looking at turnout statistics between states with early in-person voting and those without there is not a significant difference in the turnout percentage. As shown in Table 1, the 2008 election states with an early in-person voting option had an average turnout rate of 58.8% where those with no early in-person option was 60.5% (United States Elections Project 2009). The percentage of low income early voters (12%) proves to be slightly higher than the percentage of high income early voters (11%) in the 2008 NAES survey.2 This finding however is only true when looking at the percent of low or high income voters that voted early compared to the total number of voters in their income range. If you look at the percentage of low or high income voters compared to the total number of early voters in the sample high income voters are seen as turning out at a much higher percentage. Looking at the data in this way low income voters turnout at about 26% while high income voters turnout at 40%. A similar trend was found in the turnout rates related to education. Those with low education did turnout to a lesser degree than those with a high education,3however the spread was narrower when looking at the percent of low or high educated voters compared to others in the same category. In this survey 10% of those with low education came to the polls early and 13% of those with high education voted early. Looking at this data against all early voters, low and high educated voters had a turnout rate of 25% and 46% respectively.
2. Vote by Mail and Absentee Voting: Voting by mail is the process by which voters receive their ballot papers through the mail and return their ballot papers through the mail, or they may drop-off their ballot at a local elections office. There are two types of mail in voting, absentee voting and voting only by mail (VOBM). The main difference between the two types of voting is that voters who live in a VOBM district will automatically receive all their ballot paperwork through the mail, whereas absentee voters have to sign up to vote absentee.
VOBM has been used by nineteen states in at least one election, with Oregon being the only state to utilize this method for every election within every district. Currently thirteen states have at least one jurisdiction that conducts their elections strictly via the mail. A poll following Oregon’s Special Election in 19964 found that 79% agreed that VOBM is more convenient than going to a polling place, and 77% preferred VOBM elections over elections with polling places (Karp, and Banducci 2000).
It has been found that elections using VOBM may further disadvantage the mobile populations within a state. Those who do not own their own homes may find the cost of keeping electoral officials up to date with their current address too high. Related to this is the impact of purging registered voters from the rosters who do not have a valid mailing address because of its higher impact on low-income and homeless voters (Kousser, Mullin 2007). The effectiveness of this method of voting also seems to be skewed towards the white population, as it has been found that as the percent of non-white voters raises the increase in turnout decreases. Contrary to many assumptions, in Oregon participation did not increase among rural voters who generally have to travel far distances to get to the polling place. Those living in Portland and the suburbs were more likely than those living in rural areas to say that VOBM is more convenient (Karp, and Banducci 2000).
In the general election VOBM most benefits those who would already vote: older, well-educated and politically active registered voters. For this reason this method is most likely to increase voter turnout by means of retention rather than mobilization. The most significant increase in turnout occurs in low stimulus elections such as local elections and primaries. When turnout is generally low, 20%-50% of registrants, such as turnout for special and local elections, the voter may be more inclined to participate in such a race when a convenience voting method is used. Generally these types of election receive less press coverage than general elections, so receiving a ballot in the mail may be a more important reminder for voters than it is in a presidential race (Kousser, Mullin 2007).
A state law in California allows county registrars to designate any precinct with fewer than 250 registrants as a mandatory VOBM precinct. A study by Kousser & Mullin (2008) which studied these vote by mail trends in California in 2000 and 2002 general elections found that mail voting increased local special elections by an average of 7.6% (Gronke , Galanes-Rosenbaum, Miller, and Toffey 2008). Also, in the election and primary studies for the special election in Oregon 1.4 million ballots were mailed and 820,000 (57.9%) were returned for the primary and 1.2 million (66.3%) were returned for the election (Berinsky, Burns, and Traugott 2001, pg 180). This is a large increase when compared to the turnout in the non-presidential primaries of 1994 and 1990 where the average turnout was 45% (Oregon Secretary of State 1995).
Absentee voting is the oldest most common form of convenience voting. In the past absentee voting was strictly for those living overseas, students, military members and the infirm, but more recently it is becoming commonplace for districts to allow no-excuse absentee voting. This form of absentee voting can be used by any citizen who would prefer to cast his vote via an absentee ballot rather than go to the polls. Some states even allow their absentee voters to request to become a permanent absentee voter which essentially makes them a vote by mail voter because they will automatically be mailed their ballot papers before every election. In states where this permanent status is possible a substantial portion of the electorate has taken advantage of this opportunity, with California and Washington state both reaching 20% of their voting via the mail in the 1996 Presidential Election (Karp, and Banducci 2000).
Voting absentee is much like voting by mail with the exception that in absentee voting there is a choice between mailing in the ballot and going to a polling place. Much like mail-in voting the voters have a loose timeline for returning their ballots. An absentee ballot may be cast as early as forty days before the election in Iowa, Wyoming and Maine and in many jurisdictions the ballot can be hand delivered on Election Day to an elections office (Gronke et al 2008). Also like those who participate in vote by mail elections absentee voters tend to be older, better educated and more politically active than most of the electorate (Kousser, Mullin 2007). The main difference between vote by mail and no excuse absentee voting is the wide availability of no-excuse absentee voting. Twenty-eight states offer no-excuse absentee voting as opposed to one state and various districts within thirteen states which offer a vote only by mail system. Half of the twenty-eight states that do offer no excuse absentee voting will require a witness or notary signature on each ballot (National Conference of State Legislatures 2008) making the symbolic costs of voting more prominent with voting absentee as compared to a mandatory vote by mail election.
3. Internet Voting: Electronic voting sometimes referred to as Internet or I-voting is not strictly defined as voting from a remote location via the Internet. Some other iterations of this notion is voting at a polling place or satellite location via kiosk using the Internet or an Intranet, voting via the telephone or fax machine as well as voting through a television via cable or satellite.
While electronic voting has not been widely used in American elections, there have been some large-scale natural experiments with this method of voting. The most notable was the 2000 Democratic Primary in Arizona where 46% of voters opted to use the I-vote rather than a normal ballot in the polling station (16%) or voting with an absentee ballot (38%). Voters were able to use a kiosk at the polling place (5%) as well as use a PC connected to the Internet (41%) from a remote location. Within their study of this primary election Solop (2001) and Gibson (2001) found that voters from the highest income bracket were almost three times as likely to cast their ballot through this method over voters from the lowest income bracket. It was also found to be more popular among males, the well educated and liberal democrats as opposed to moderate or conservative democrats. Following the election 34% of those who voted via the electronic methods stated that the availability of Internet voting would encourage them to vote more often in the future, 9% of those who voted through the more traditional methods also said Internet voting would encourage them to vote more often. On the flip side 2% of those who voted via the Internet said that the availability of Internet voting would encourage them to vote less often in the future, as opposed to 8% of those who voted through the traditional means.
While the I-vote may have an opportunity to increase the turnout among the young and computer savvy this type of vote, especially if it were used as an all I-vote system, would negatively impact those with a lower SES as well as the elderly as seen in a study by Roseman and Stephenson (2005). They found that changing voting technology to a more advanced touch screen system such as used in ATM’s negatively impacted the turnout of elderly voters. Although this negative impact was found after Georgia changed their voting system to an electronic kiosk it is not expected to have a long lasting effect on voting turnout. One explanation given for this was that the future elderly voters will be more comfortable with computers having used them when they were young.
Reducing many of the direct costs of electoral participation by applying different methods of convenience voting does not significantly increase voter turnout. Any increase in turnout appears to be related to keeping voters in the electorate rather than bringing in new voters. This is troublesome because the electorate will not grow to include disengaged citizens. Convenience voting allows the habitual behavior of voters and non- voters to take center stage. Those we would expect to vote, the politically savvy, are more likely to stay in the electorate over the long haul, and those we would expect to abstain are more likely to stay out of the electorate over the long haul. Overall convenience voting reforms do not draw in new citizens and do not seem to appeal to disempowered segments of the population. Due to this convenience voting laws seem to offer campaigns little incentive to expand their efforts beyond their base to disengaged citizens. (Gronke et al 2008).
Berinsky, Adam J, Nancy Burns, and Michael W Traugott. 2001. “Who Votes By Mail?” The Public Opinion Quarterly Volume 65(2): 178-197
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper & Row.
Gibson, Rachel. 2001. “Elections Online: Assessing Internet Voting in Light of the Arizona Democratic Primary.” Political Science Quarterly Volume 116(4): 561-583
Gronke, Paul, Eva Galanes-Rosenbaum, and Peter A. Miller, 2007. “Early Voting and Turnout.” PS: Political Science & Politics Volume 40: 639-645
Gronke, Paul, Eva Galanes-Rosenbaum, Peter A. Miller, and Daniel Toffey, 2008. “Convenience Voting” Annual Review of Political Science Volume 11: 437-455
Karp, Jeffrey A, and Susan A. Banducci. 2000. “Going Postal: How All-Mail Elections Influence Turnout.” Political Behavior Volume 22(3): 223-239
Kousser, Thad, and Megan Mullin 2007. “Does Voting by Mail Increase Participation? Matching to Analyze a Natural Experiment.” Political Analysis Volume 15: 428-445
Kousser, Thad, and Megan Mullin 2008. “Will Vote-by-Mail Elections Increase Participation? Evidence from California Counties”
National Annenberg Election Survey. 2008. “National Annenberg Election Surveys Data Sets” Retrieved April 22, 2009, from http://www.annenbergpublicpolicycenter.org/
National Conference of State Legislatures. 2008. Absentee and Early Voting. Retrieved Mar. 10, 2009, from http://www.ncsl.org/programs/legismgt/elect/absentearly.htm
Oregon Secretary of State. 1995. “Special US Senate Primary” Retrieved March 18, 2009, from http://www.sos.state.or.us/elections/dec595/dec595.htm
Roseman, Gary H. JR, and E. Frank Stephenson. 2005. “The Effect of Voting Technology on Voter Turnout: Do Computers Scare the Elderly?” Public Choice Volume 123: 39-47
Solop, Frederic I. 2001. “Democracy Comes of Age: Internet Voting and the 2000 Arizona Democratic Primary” Political Science and Politics Volume 34 (2): 289-293
Stein, Robert, Chris Owens, and Jan Leighley. “Early Voting in Texas.” Retrieved April 22, 2009 from http://www.allacademic.com/meta/p41479_index.html
United States Elections Project. 2009. 2008 General Election Turnout Rates. Retrieved Mar. 13, 2009, from http://elections.gmu.edu/Turnout_2008G.html
1 The three efficacy statements used in this analysis were: (1) Public officials don’t care much about what people like me think. (2) People like me don’t have any say about what the government does. (3) Sometimes politics and government seem so complicated that a person like me can’t really understand what’s going on.
2 In this study low income voters are classified as those with an annual family income of less than $35,000, and high income voters are classified as those with an annual family income over $75,000. In this survey 30% of participants were low income, and 46% were high income.
3 In this study low education is classified as those with high school or less, and high education is classified as those with a college degree. In this survey 30% of participants had low education, and 42% had high education.
4 After the resignation of Senator Packwood in September 1995, Oregon held the first statewide special mail election to replace him in January 1996.