"If people don't come out to the ballpark,
The 2000 presidential election was one of the most remarkable elections in the history of the United States. The election was decided by 537 votes, the margin of victory (disputed of course) in the state of Florida. In fact, the outcome of the election was not known until 36 days later when the Supreme Court ruled in Bush v. Gore that vote totals certified by the Florida Secretary of State were, in effect, final. In reality, we will never know with certainty which candidate actually won the election. George Bush won the Electoral College by two votes based upon the disputed Florida vote, and Al Gore won the popular vote by roughly 550,000 votes (one-half of one percent), which is simply too close to call. One can be sure that the inadequacies of the Florida electoral system exist in almost all states (30,000 votes on ballot initiatives were initially overlooked in Boston and a substantial proportion of under and overvotes have now been identified in Chicago). Because of this, all we can really say with confidence about the election is that it was a dead heat.
One aspect of the election, however, was remarkably similar to past elections. As with every presidential election since 1972 only about half of the eligible citizens actually cast a ballot. While final vote totals have yet to be confirmed by the Federal Election Commission, most estimates indicate turnout nationwide was somewhere between 50 and 52%, which is a modest increase over the 49% rate in 1996 (the lowest since 1924). While turnout indeed increased, this by no means suggests American democracy is on the mend. 2000 is simply another in a long line of low turnout elections since 1972. Given the closeness of the election, the low turnout is surprising. With so much in the balance and the intense focus of the media on the closeness of the election one would expect citizens to show up and participate in order to make a difference. Once again, however, students of political participation are left asking, why is turnout in American elections so low?
In order to understand participation rates in the United States one must compare our turnout with that of other comparable democracies. Between 1960 and 1995 turnout in 24 democracies without mandatory voting averaged 80%. Ten of these 24 states had turnout rates ranging between 85% and 90%. Turnout in the United States over the same period, on the other hand, averaged 54% (in American midterm elections the average turnout rate is below 40%). Clearly, American turnout is very low when compared to other industrialized democracies. Why?
For instance, one of the central differences in electoral participation between the United States and other industrialized democracies is that we make registering to vote voluntary. If an individual wishes to vote in an election, that citizen must first place his or her name on the list of registered voters. Contrast this to the system of automatic registration used in almost all-European democracies in which the citizen's name is automatically placed on the voter list upon reaching voting age. In effect, we increase the costs of participation by adding the registration requirement and thereby making the relatively simple of act of voting a two-step process.
The individual registration requirement has a particularly strong impact on those least likely to participate such as the young and those of lower socio-economic status. With the responsibility for registration placed on the individual those most likely to participate (higher levels of education, income, and older individuals) are the ones most likely to register to vote. These individuals place their names on the registration rolls because they want to vote. Individuals who may lack resources or attitudes conducive to participation (or both) have very little ability or incentive to overcome the obstacle of registration. When election time comes around individuals who are registered are allowed to vote while the unregistered are not. This is in part why the voting population in the United States is skewed toward the middle and upper classes and older Americans.
Most students of participation agree that the implementation of automatic registration in the United States would increase overall turnout by approximately ten percentage points, while also modestly reducing the skew toward the advantaged. However, reform of this magnitude would still leave turnout as much as 16 points lower than the average turnout in other democracies. While restrictive registration requirements are a significant depressant of turnout they cannot completely explain the lower turnout. We must therefore look for other factors that can help explain the difference.
The nature of the party system in the United States also works to constrain participation. One of the key differences in the political system of the United States and other industrialized democracies is that the United States is a two-party system in its purest form. On Election Day, American citizens have a choice between two centrist parties with a legitimate chance to gain a place in government. Contrast this to nations with proportional representation systems in which voters usually have a choice between several parties with a legitimate chance to gain seats in the legislative body. In the United States individuals whose ideological preferences lie to the left or right must either choose the party closest to their preferences (as many do) or not vote at all. While to some voting may be a habit or a reaffirmation of the political system, most citizens cast a ballot in an election hoping to gain a voice in the government. In a two-party system like the United States many individuals see very few differences between the two major parties and simply cannot make a choice between what some have characterized as two sides of the same coin. Faced with this dilemma many citizens simply choose to abstain from voting altogether.
The two-party system in the United States is a natural function of the structures of American electoral politics. The United States' electoral system is based on single-member, winner-take-all districts, which means that our system of representation is based on one representative per geographic district. In any single-member district election the winner of the election gets to participate in government, and the losing side must wait until another day. This type of electoral system tends to suppress the formation of minor parties for two main reasons. First, rather than expend valuable political resources on fighting a losing battle, most politicians will join one of the two main parties closest to their policy preferences. Second, citizens seeking a voice in government do not wish to expend their valuable vote (they only have one) on a losing candidate so they too choose the major party closest to their policy preferences rather than cast a ballot for a minor party. Thus, the structure of electoral politics (i.e. single member districts) constrains the number of choices voters have at the polling place.
Single-member districts also shape mobilization efforts by the parties and their candidates. This is important because when the parties and their candidates expend resources to reach out to voters they make the act of voting easier through increased visibility, the provision of information, or even a ride to the voting precinct. More importantly when a party or candidate contacts a citizen or reaches out to a particular group they are telling citizens "your vote matters." As one political scientist has argued: "people will vote when they are asked."
The problem lies in the fact that strategic candidates (those that seek to maximize their chances of winning) must expend their resources where they will have the most effect and that is in competitive races. Thus, in competitive elections the parties and candidates expend resources reaching out to potential voters in an attempt to win. In non-competitive races, on the other hand, voters are left to their own devices to make sense of the campaign and to participate. Across the nation, then, our single-member district elections tend to produce uneven mobilization efforts leaving many citizens behind.
A good example of how single-member districts influence mobilization efforts is the Electoral College. All states, with the exceptions of Maine and Nebraska, award all of their electoral votes to the winner of the presidential contest in the state, and thus candidates must focus on building an Electoral College majority by winning enough states to reach the required 270 votes needed to win the presidency. Some have argued that the strength of the Electoral College is that it forces the candidates to pay attention to small states because they may need one or a few of these states to win. The logic behind this is very strong. For instance, by simply giving New Hampshire to Gore rather than Bush, the outcome of the election would be different. This argument, while logical, ignores the fundamental premise of campaigning. Use your resources where they are needed most, which as mentioned is in the most competitive electoral contexts. Candidates may focus on a small state if that state is highly competitive but only when such an effort does not detract from campaigning in a larger state which is also competitive. After all, where does it make sense to spend money: in Maine, which is competitive and has four electoral votes, or Florida, which is competitive and has 25 electoral votes? In the end, the name of the game is winning 270 electoral votes and the candidates will not expend resources to win a handful of electoral votes. They simply have no incentive to visit or spend money in those states that do not have much to offer in the way of electoral votes.
A brief analysis of campaign visits to states by the two major campaigns provides an example of how the Electoral College forces the two major campaigns to focus on the most competitive states. For instance, in the 2000 campaign 15 states were not visited at all by the presidential and vice-presidential candidates of the two major parties. Not surprisingly, the average number of electoral votes for these states is a little less than six. The two major campaign organizations had no incentive to expend resources visiting states that provided little in the way of building an Electoral College majority. This is not to say within this environment that the candidates will focus their energy completely on those states with the most electoral votes. New York and Texas, for example, have the 2nd and 3rd most Electoral College votes respectively, yet neither state received any serious attention from the two major campaigns because they were not competitive. Where the campaigns did focus their energy and resources was in the most competitive states. For example, between Labor Day and Election Day (November 7) there were 264 total campaign visits to 35 states (and the District of Columbia) by the two major presidential campaigns. Of these 264 visits 208, or 79%, were to the 16 so-called battleground states (the most competitive). (See Figure 1) Clearly, the two major campaigns were focusing their energy and resources where they could help them the most in attempting to win the White House.
If mobilization matters in increasing turnout we would expect turnout to be higher in the battleground states because of the greater mobilization attempts on the part of the major campaigns. (See Figure 2).
The average of turnout in battleground states this past election was 56%. In non-battleground the average level of turnout was 51%. In fact, turnout in the 16 battleground states increased from 53% in 1996 to 56% in 2000, while in the non-battleground states turnout remained unchanged between these two elections. It appears that the intense attention paid to the battleground states by the two major campaigns resulted in higher levels of turnout.
The point here is that the Electoral College, a constitutionally mandated feature of the American political system, creates an electoral context in which politicians are forced to focus their resources where they will do the most good, and citizens in the rest of the country are left to fend for themselves. Voting is much more likely when everyday people are aware of where the candidates stand. In the sixteen battleground states one can hazard a guess to say any citizen who was marginally engaged with the political world was at least minimally aware of where Bush and Gore stood, or they were at least exposed (relentlessly) to campaign ads and mailings. Citizens in the remaining 34 states (and the District of Columbia) were left to their own devices to decipher where the candidates stood and, as would be expected, turnout was lower.
The final institutional culprit responsible for contributing to the low turnout rates in the United States is the central feature of the structure of government created by the Constitution: the separation of powers. The notion that the government should be divided into three individual branches (legislative, executive, and judicial) has a long philosophical history and is, in fact, deeply ingrained into the American psyche. Most states mandate that college students take government courses developed around this structure and many citizens equate this form of government with democracy itself. The key argument in support of the separation of powers is that it provides stability through incremental decision making. And indeed, across the history of our nation, with the exception of the Civil War, the constitutional arrangements have provided political stability.
Unfortunately, a by-product of this stability is a fragmentation of power and the blurring of accountability. Citizens tend to cast ballots in an election in order choose leaders who will then pursue some broad ideological agenda or at least policy preferences similar to their own. In a sense citizens vote because they want a say in governmental outputs. In nations where the main policymaking power is divided between the legislative and executive branches there is a reduction in the ability of the government to produce substantive policy and also a reduction in the ability of citizens to either assess credit or blame for governmental outputs. A citizen's inability to clearly
determine what the government has produced creates the perception that elections have very little chance of affecting policy outputs. Thus many citizens drop out of the political arena because they come to believe that elections simply don't matter. In fact, across 29 democracies turnout is lower in those countries with separation of powers systems than those with parliamentary systems.
Further, one of the enduring characteristics of the American political system in the post-war era is divided government in which control of the legislative and executive branch is divided between the Democratic and Republican parties. In these periods (34 of the last 48 years) policy is even more difficult to craft, accountability is completely blurred and many citizens are left angry and confused. Across time, then, turnout should decrease during these periods as many citizens come to believe that their vote doesn't matter because government has not dealt with the problems facing the nation. One scholar has estimated that since 1840 for every presidential election held under divided government (this means four previous years of divided control), there is on average a 2% decrease in turnout. We may gain stability through the separation of powers, but we do so at the expense of widespread participation.
The question of why turnout is so low in American elections is fairly easy to answer. The institutional arrangements (electoral and governmental) in our nation create an environment which is not conducive to widespread participation. For those who believe that increased participation in electoral politics is a worthwhile goal, the central question then, is how do we increase turnout in the United States?
Without a complete restructuring of the constitutional arrangements of governmental and electoral politics, single-member districts, the Electoral College, and the separation of powers will be with us for some time to come. Given the difficulty of changing institutional structures, we are left with reforming the laws that regulate citizen access to the voting booth, such as registration requirements at the state level. Many states have experimented with various reforms aimed at increasing participation in elections. North Dakota for instance comes closest to automatic registration with no registration requirement at all. On Election Day citizens can simply show up and vote. Six states (Idaho, Maine, Minnesota, New Hampshire, Wisconsin, and Wyoming) have election-day registration, in which citizens wishing to cast a ballot in an election can show up at the polling place and register before they vote. With registration systems that are very accessible these seven states have turnout rates substantially higher than the rest of the nation. Since 1980 average turnout in these states is 10 points higher than states with more restrictive registration systems.
The success of election-day registration is fairly easy to explain. In states without election-day registration, citizens must place their name on the registration list a specified period of time prior to the election. Most states have a thirty-day "closing date" (Massachusetts is twenty days) after which citizens are not allowed to register and thus not allowed to vote. Closing dates can have a depressive effect on turnout because the height of intensity and publicity of a campaign takes place during the last thirty days prior to the election. In those states with long closing dates, then, the state is cutting off access to the voting booth just as the campaign has the greatest potential to mobilize potential voters. Contrast this to election-day states (and North Dakota) in which citizens can simply show up on Election Day to register and cast a ballot. The opportunity to register on Election Day can have a particularly strong effect on marginal voters who may not become engaged with the campaign until the last couple of weeks prior to the election. In election-day states a citizen mobilized by the politicized environment of the campaign can simply show up, register, and vote. If we want to increase turnout in American elections implementing election-day registration on a national scale is simply the easiest and most effective way to do this. In fact, the reform has been proposed to Congress several times since 1978, but each time it has been defeated.
The most well known registration reform in recent years is the National Voter Registration Act (NVRA) more commonly known as "motor voter." The reform is a federal mandate that requires all states without election-day registration (or North Dakota) to allow citizens to register to vote (or update their registration) at drivers license bureaus, through the mail, or at public agencies such as AFDC, disability, or unemployment offices. The logic of the reform is fairly straightforward. By placing the opportunity to register in front of as many people as possible, the registered electorate should expand and by extension overall turnout rates should increase. Since the implementation of the law in 1995 the registered electorate has expanded, and it appears the proportion of individuals from low participation groups such as the young, the poor, and racial minorities has increased among the registered electorate. In the two elections since, however, (1996 and 1998; the 2000 election data are not yet available) it does not appear that individuals who are registering via the reform are voting once registered.
As mentioned, voting in the United States is a two step process in which the potential voter is required to register before he or she is allowed to vote in an election. The NVRA addresses the first step in the process and at this point appears to have been successful at creating a more representative registered electorate. However, the reform does nothing to motivate the newly registered citizen to vote in the next election. Whatever individual level obstacles, such as socio-economic resources or personal attitudes, that existed prior to registration still stand between the citizen and the voting booth. The NVRA is a positive step in the direction of creating a more equitable electoral politics; however registration reforms such as the NVRA only open the door to the voting booth a little wider. They do nothing to help the person through that door.
There are other reforms aimed at increasing participation either being implemented or at least part of the reform debate. For instance, Oregon conducted the 2000 election completely through the mail, and Arizona ran a limited experiment with Internet voting in 2000. Some have argued that by making election-day a holiday (or at least combining it with an existing holiday such as Veteran's Day) we can increase turnout because 1) people will have time to vote, and 2) it would make choosing our government a celebration rather than something we do the first Tuesday after the first Monday in November. Another reform with some merit is the suggestion that we hold elections over the course of a weekend. This way citizens can vote at their convenience rather than having to squeeze it into their busy schedules between specific hours on a particular day.
While the impact of any reform aimed at making registration and voting easier is likely to be limited due to the institutional constraints inherent to the American system, they are nonetheless worthwhile efforts at creating a healthier democracy through widespread participation. Some have argued that government works best when only the most informed and engaged citizens participate in the political process. Indeed, the structures of government in the United States are based on the belief that widespread participation should be limited to prevent tyranny imposed by an uniformed majority. There is another side to this debate, however. Participation is a matter of voice, or having a true say in the choosing of governmental leaders who may craft policy which has a dramatic impact upon your life. Widespread participation creates greater democratic legitimacy because all groups in the polity have a say in choosing elected officials.
The presidential election of 2000 was one of the most remarkable elections in the history of this nation. However, when we clear away media hype surrounding lawyers, pundits, and state and federal Supreme Court rulings we are left with an election in which a little more than half of the eligible voting population cast ballots. Are the citizens of the United States that much more lazy and ignorant than citizens in other nations? Most comparative public opinion work suggests Americans are as engaged and knowledgeable (if not more so) than citizens in other industrialized democracies. Is the problem, as Jim Hightower put it, that the 2000 election was a choice between "Miller Lite and Bud
Lite...either way you end up with mighty weak beer?" The lack of substantive differences between the two major parties is certainly a good part of the turnout problem; however, this is only a symptom of a larger problem. What I have tried to do here is place American turnout in a larger perspective. Electoral politics takes place within an institutional context and in the United States the institutions that govern our electoral and governmental politics work to depress participation in elections. Will turnout in American elections ever reach that of European elections? No, not as long as the current rules of the game are in place. Can electoral politics in the United States be improved? Certainly, all of the reforms discussed here are positive steps toward creating a more inclusive democracy in which citizens come to believe they have true voice in government.