Should Voting Be Compulsory?
This article first appeared on the Brookings Institution site. The full text can be found here.
When we receive a summons for jury duty, we are required to present ourselves at the court. Should we treat showing up at the polls in elections the same way?
Although the idea seems vaguely un-American, it is neither unusual, nor undemocratic, nor unconstitutional. And it would ease the intense partisan polarization that weakens both our capacity for self-government and public trust in our governing institutions.
It is easy to dismiss this idea as rooted in a form of coercion that is incompatible with our individualistic and often libertarian political culture. But consider Australia, whose political culture may be as similar to that of the United States as the culture of any other democracy in the world.
The Australian Solution
Alarmed by a decline in voter turnout to less than 60 percent in the early 1920s, Australia adopted a law in 1924 requiring all citizens to present themselves at the polling place on Election Day. (This is often referred to as mandatory voting, although Australian voters are not required to cast marked ballots.)
Enforcing the law were small fines (roughly the same as for routine traffic tickets), which increased with repeated acts of nonparticipation. The law established permissible reasons for not voting, such as illness and foreign travel, and procedures allowing citizens facing fines for not voting to defend themselves in court.
It also required citizens to register to vote (much as the United States has draft registration) and the Australian authorities have created systems to make registration easy.
The results were remarkable. In the 1925 election, the first held under the new law, turnout soared to 91 percent. In the 27 elections since World War II, turnout in Australia has averaged 95 percent.
It is hard to doubt that there is a causal connection between the law and the large change in Australians' voting behavior. And there is additional evidence from the Netherlands, which operated under similar legislation from 1946 to 1967. During that time, turnout averaged 95 percent. After the Netherlands repealed this law, turnout has fallen to an average of 80 percent.
The impact of such laws can extend well beyond the act of voting. In Australia, citizens are more likely than they were before the law was passed to view voting as a civic obligation. This norm helps explain why the negative side effects that many feared did not materialize.
For example, the percentage of ballots intentionally spoiled, left blank, or randomly completed as acts of resistance has remained quite low. The Australian experience suggests that when citizens know that they are required to vote, they take this obligation seriously. Their sense of civic duty makes them reluctant to cast uninformed ballots and inclines them to learn at least the basics about issues, parties and candidates.
Why the Australian Model Makes Sense for Democracies—Including Ours
The most straightforward argument for near-universal voting is democratic. Ideally, a democracy will take into account the interests and views of all citizens so that its decisions represent the will of the entire people. If some regularly vote while others do not, elected officials are likely to give less weight to the interests and views of non-participants.
In practice, this might not matter much if non-voters were evenly distributed through the population, so that voters were a microcosm of the people. But that is not the case: In the United States, citizens with lower levels of income and education are less likely to vote, as are young adults and recent immigrants.
Changes in our political system have magnified these disparities. The decline of formal political organizations, including political machines, has reduced mobilizing efforts that were often year-round propositions and frequently gave life to political clubs that served as centers of sociability as well as electoral action.
The sharp drop in union membership since the 1950s has further eroded connections between citizens of modest means and lower levels of formal education to electoral politics. In their heyday, national civic institutions organized along federal lines performed these functions as well, but they too have undergone a relentless decline.
These factors were partly offset by a democratization of the electorate through the Voting Rights Act of 1965 that empowered African Americans, particularly in the South, and by the 26th Amendment to the Constitution that lowered the voting age to 18 throughout the country in 1971.
But with the exception of a few states that provided for registration on Election Day itself, the inclusion of younger voters into the electorate was not matched by changes in voter registration laws to make it easier for younger Americans, who tend to change residencies more frequently than their elders, to be included on the voter rolls.
As it is, registration rules are biased in favor of those with relatively stable residential patterns. The combination of the decline in political mobilization and the rise of a younger electorate mean that turnout in presidential elections has fallen off since the 1950s.
As measured against the voting age population, turnout in 1952 hit 63.3 percent, fell slightly to 60.6 percent in 1956 and rebounded to 62.77 percent in the Kennedy-Nixon election of 1960. The last time turnout topped 60 percent was 1968.
The drop between 1968 and 1972, after the enfranchisement of all 18 year olds, was especially sharp—from 60.84 percent to 55.21 percent. The highest turnout since then (58.23 percent) came with the Obama mobilization efforts in 2008, but even this number was lower than the turnout figures between 1952 and 1964. And turnout fell off again in 2012, to 54.87 percent.
Universal voting would help fill the vacuum in participation by evening out disparities stemming from income, education and age. It would enhance our system's ability to represent all our citizens and give states and localities incentives to lower, not raise, procedural barriers to the full and equal participation of each citizen in the electoral process.
If citizens had a legal obligation to vote, managers of our electoral process would in turn have an obligation to make it as simple as possible for voters to discharge this duty.
The weakening of the Voting Rights Act by the Supreme Court has allowed many states to impose new requirements on voters and to cut back on early and Sunday voting. Universal voting would change the presumptions in favor of broad democratic participation and put states on the side of promoting that goal.
It would also improve electoral competition. Campaigns could devote far less money to costly, labor-intensive get-out-the-vote efforts. Media consultants would not have an incentive to drive down turnout with negative advertising (even though such advertising would no doubt remain part of their repertoire). Candidates would know that they had to do more than appeal to their respective bases with harshly divisive rhetoric and an emphasis on hot-button issues.
This brings us to a benefit of universal voting that goes to the heart of our current ills. Along with many other factors, our low turnout rate pushes American politics toward hyper-polarization. Intense partisans are more likely to participate in lower-turnout elections while those who are less ideologically committed and less fervent about specific issues are more likely to stay home.
Although responding to strong sentiments is an important feature of sustainable democratic institutions, our elections tilt much too far in that direction.
A structural feature of our system—elections that are quadrennial for president but biennial for the House of Representatives—magnifies these ills. It is bad enough less than three-fifths of the electorate turns out to determine the next president, much worse that roughly two-fifths participate in midterm elections two years later.
As Republicans found in 2006 and Democrats in 2010 and 2014, when intervening events energize one part of the political spectrum while disheartening the other, a relatively small portion of the electorate can shift the balance of power out of proportion to its numbers. And with the rise of the Obama Coalition, the midterm electorate is decidedly older and less diverse than the electorate in presidential years.
The vast difference between these two electorates has enshrined new forms of conflict in an already polarized political system. Bringing less partisan voters into the electorate would reduce this instability, and it would offer parties and candidates new challenges and opportunities.
The balance of electoral activities would shift from the mobilization of highly committed voters toward the persuasion of the less committed. Candidates unwilling or unable to engage in persuasion would be more likely to lose. If political rhetoric cooled a bit, the intensity of polarization would diminish, improving the prospects for post-election compromise.
Rather than focusing on symbolic gestures whose principal purpose is to agitate partisans, Congress might have much stronger incentives to take on serious issues and solve problems. To pick up a term of the moment, universal voting might combat the "Trumpification" of politics.
The electorate that turns out is not representative of the country as a whole. After the election of 2014, the Public Religion Research Institute (PRRI) re-interviewed 1,339 respondents who had been contacted in a pre-election survey.
The post-election poll found that Hispanic voters comprised 8 percent of midterm voters but 22 percent of non-voters. Millennials, those of ages 18-to-34, made up 17 percent of voters—and 47 percent of non-voters. Those earning less than $30,000 a year accounted for 26 percent of voters and 44 percent of non-voters.
And the underrepresentation of middle-of-the-road voters was brought home by both the PRRI survey and a Pew Research Center study of the 2012 electorate. In the PRRI study, independents accounted for 33 percent of voters but 42 percent of non-voters. Moderates accounted for 31 percent of voters but 38 percent of non-voters.
Based on the turnout model of the 2012 Pew pre-election study, independents made up 27 percent of likely voters but 44 percent of non-voters; moderates accounted for 34 percent of likely voters but 38 percent of non-voters.
A republic governed under a Constitution that begins with the words "We the people" should want an electorate as broadly representative of the people as possible.
There is a final reason for the country to embrace universal voting, and it may be the most compelling: Democracy cannot be strong if citizenship is weak. And right now, citizenship in America is radically unbalanced: It is strong on rights but weak on responsibilities.
With the abolition of the universal draft, citizens are asked to pay their taxes and obey the law—and show up for jury duty when summoned. That's about it.
Making voting universal would begin to right the balance. And it would send an important message: We all have the duty to help shape the country that has given us so much.
William F. Buckley Jr., who can fairly be thought of as the founder of contemporary American conservatism, wrote a book in 1990 called Gratitude: Reflections on What We Owe to Our Country. Gratitude is personal, but as Buckley made clear, it is also civic, and it is a disposition that transcends ideology.
Participation in self-rule is an expression of gratitude for the freedom we have to govern ourselves.
Take Part in the Debate
If voting and registration rules were made easier, should voting in national elections be universal and mandatory for all eligible citizens?
To voice your opinion, click the image below and vote. We will share the results on social media.
William A. Galston is a senior fellow and the Ezra K. Zilkha Chair in Governance Studies and E.J. Dionne Jr. is a senior fellow and the W. Averell Harriman Chair in Governance Studies at the Brookings Institution.
