Compulsory Voting Research Paper

CV raises questions that strike at the heart of democratic theory. For example, is it ‘democratic’ to compel citizens to turn out to vote via the threat of financial or social sanction? In what sense are elections conducted under such circumstances ‘free and fair’? On the other hand, what legitimacy attaches to elected representatives and their policies when large proportions of the electorate abstain? Responding to this latter question, political scientists have long considered CV an effective mechanism to increase voter turnout (e.g., Gosnell 1930). Scholarly interest in CV was piqued by Arend Lijphart’s 1996 Presidential Address to the American Political Science Association, providing a detailed review of CV and a compelling case for its adoption. CV raises numerous issues for students and practioners of electoral politics, which are the focus of this research paper. Real-world implementations of CV are considered first, followed by an assessment of the consequences of CV for electoral politics, policy, campaigning, and the administration of elections.

1. Implementations Of CV

Western democracies utilizing CV for at least some of their elections include Australia, Austria (in selected Lander), Belgium, Greece, Italy, Liechtenstein, Luxembourg, and Switzerland ( just one canton). The Netherlands also had CV up until 1970. In almost all of these Western democracies CV was instituted in the early twentieth century, shortly after the expansion of voter suffrage and the political organization of labor movements in the late nineteenth and early twentieth centuries. At the same time there were some experiments with CV in American states: North Dakota (1898) and Massachusetts (1918) actually amended their constitutions to permit CV, but the respective legislatures did not implement CV in their statutes (Gosnell 1930, pp. 206–7). Abraham (1955) also reviews American flirtations with CV. CV is widespread throughout Latin America; for instance, Argentina, Brazil, Costa Rica, Ecuador, Guatemala, Honduras, Panama, Peru, Uruguay and Venezuela have CV. Cyprus, Egypt, Fiji, Turkey, Thailand, Singapore and the Phillipines also have CV.

1.1 Inducing Compliance

Compliance with CV is most commonly induced with a system of small to moderate fines (e.g., Australia and Belgium), sometimes in conjunction with other penalties, although a substantial number of countries do not impose fines. Other penalties include ineligibility for elected office for a prescribed period (e.g., Argentina), or ineligibility for certain types of government employment (e.g., Venezuela), or disenfranchisement (e.g., Thailand). In other countries, CV has more the status of a norm, with sanctions for noncompliance taking the form of social embarrassment, with non-voting noted on official documents (e.g., Italy). In another class of countries CV appears as a constitutional provision or in statute, but with no mention of any penalty for non-compliance (e.g., Bolivia, Chile, Liechtenstein, Egypt). Many countries also have large classes of exemptions. For instance, Australia’s federal CV statute exempts non-voters with ‘valid and sufficient’ reasons for not turning out; Australia’s courts have rigorously denied non-voters’ claims of indifference between the candidates or alienation from politics as valid and sufficient reasons (AEC 1999). Age exemptions are also common: for instance, Brazil makes voting optional for citizens between the ages of 16 and 18, citizens over the age of 70, and for illiterates.

2. Consequences Of CV

CV has a direct impact on voter turnout, evident in both aggregate and individual-level analyses. The logic of this is extremely straightforward, with the sanctions of CV offsetting the costs of going to the polls; that is, CV makes the option of not turning out costly, overcoming the fact that turning out is a low benefit activity for many citizens (Lijphart 1997, p. 9).

2.1 Voter Turnout: Cross-National Evidence

Even a casual inspection of compendiums of aggregate turnout statistics reveals higher turnout among countries with CV. For instance, a recent collection of data on voter turnout in 171 countries finds turnout roughly six or seven percentage points higher in the 24 countries with CV than in countries without CV (IDEA 1997, p. 32). Multivariate statistical analyses typically find CV to have larger impacts on turnout, by controlling for other institutional and political variables that affect turnout. Lijphart’s (1997) review finds CV associated with a boost in turnout rates of seven to sixteen percentage points (for examples of the studies reviewed, see Powell (1981), Jackman (1987), Jackman and Miller (1995), and Franklin (1999), the latter study being distinctive for including an individuallevel analysis, exploiting survey data from European Union countries). Among Latin American countries, the estimated turnout boost associated with CV is roughly eleven to seventeen percentage points (Fornos 1996).

These results are striking considering (a) large cross- national differences in institutional and political characteristics of these countries that impact turnout (e.g., Jackman’s 1987 study considered competitive-ness of elections, electoral disproportionality, number of political parties, unicameralism vs. bicameralism), and (b) considerable variability in the implementation of CV among those countries that nominally have CV. In addition, of all the institutional variables found to impact voter turnout, CV has far and away the largest effect.

2.2 Voter Turnout: Evidence From Within-Country Comparisons

Hirczy (1994, p. 65) makes a compelling argument that cross-national analyses provide ‘no causal proof that mandatory voting actually produces high turnout,’ and indeed, the causal arrow may be reversed; i.e., a country that adheres to a norm of high turnout simply ‘enshrines its civic norm in law.’ A research design that overcomes this threat is to compare turnout within countries, before and after the implementation or repeal of CV, or across sub-national units with and without CV. An additional strength of this research design is that, within countries, many of the factors affecting turnout remain constant, regardless of CV.

Studies of this type find CV to have large effects on aggregate turnout. Prior to the implementation of CV in 1924, turnout in elections to Australia’s House of Representatives averaged 64.2 percent; in the nine elections following the introduction of CV turnout averaged 94.6 percent, an increase of 30.4 percentage points (t = 8.7; author’s calculations, using data in Hughes and Graham (1968)). In The Netherlands, the abolition of CV in 1970 caused a drop of roughly 10 percentage points to around 84 percent (Irwin 1974, Hirczy 1994). In addition, the removal of fines for nonvoting in Venezuela in 1993 saw turnout fall by roughly

30 percentage points (Lijphart 1997, p. 9). In Austria, cross-provincial and longitudinal variation in the use of CV permits a powerful assessment of the impact of CV. Turnout in eleven federal parliamentary elections between 1953 and 1987 averaged 92.7 percent in provinces without CV; among provinces with CV turnout averaged 95.7 percent, to yield a treatment effect of 3.0 percentage points (t = 3.4), this smaller treatment effect reflecting a ‘ceiling effect’ with turnout rates bounded from above at 100 percent. A reasonable conclusion based on these studies is that CV’s effects are conditional on baseline levels of electoral participation; i.e., CV is likely to have bigger impacts on turnout when other factors predispose a country to low turnout, and vice versa (Hirczy 1994). This insight can be applied in cross-national studies of voter turnout, using interaction terms or multilevel statistical models to make the effects of compulsory voting conditional on other variables.

Public opinion surveys have also been used as proxies for the ‘natural experiment’ of removing CV. For instance, researchers in a number of countries have asked respondents to engage in thought experiments, in which respondents report their likely behavior in a counter-factual scenario of voluntary turnout. Results from studies of this type suggest modest falls in voter turnout in Australia—to about 88 percent from the current 96 percent figure (Mackerras and McAllister 1996)—but larger falls in Belgium (Hooghe and Pelleriaux 1998), Brazil (Power and Roberts 1995), and Venezuala, as large as 30 percentage points. Jackman (1999) urges caution in interpreting these estimates, arguing that (at least in the Australian case) survey response bias produces an overestimate of voluntary voter turnout, since wouldbe nonvoters are less likely to respond to a survey on politics in the first place.

2.3 Political Consequences

To gauge the political consequences of CV, most scholars start by noting the near universal connection between political participation and socioeconomic status. Noting that ‘low voter turnout means unequal and socioeconomically biased turnout’ (Lijphart 1997, p. 2), many scholars see CV as going a long way to removing this ‘bias.’ Comparative studies of turnout note that the relationship between socioeconomic status and voter turnout weakens as turnout increases (e.g., Powell 1986). Thus to the extent that CV generates higher voter turnout (and the research summarized in the previous sections suggests strongly that it does), CV also removes socioeconomic differences in electoral participation. Quite simply, when almost everyone votes, there can be no socioeconomic ‘biases’ in turnout.

It also follows that the high turnout rates engendered by CV also have partisan and ideological implications. Since support for parties of the left is correlated with socioeconomic status (notwithstanding a lengthy debate about a decline in ‘class voting’), as socioeconomic based differentials in turnout diminish, support for parties of the left increases (Pacek and Radcliff 1995, Nagel 1988, McAllister and Mughan 1986). In turn, increased support for parties of the left results in higher welfare spending and more state interventions in the macroeconomy and labor markets (Hicks and Swank 1992, Castles and McKinlay 1979). Again, to the extent that CV increases turnout, and turnout increases support for left parties and their policy agendas, then CV can be shown to have policy implications. Put simply, CV helps mobilize voters who would otherwise abstain, generating a distribution of voters’ policy preferences further to the left than would otherwise result, and so party competition takes place over an interval of the ideological spectrum shifted to the left from where it would otherwise be. CV is thus causally quite remote from outcomes such as left-party success or particular policy outcomes. Indeed, this causal link could even be reversed: i.e., it is only in countries with commitments to social democratic outcomes that CV is adopted, and in fact, this accords with the pattern of adoption of CV among Western democracies in the early twentieth century.

2.4 Other Consequences

One criticism of CV is that it compels the participation of disinterested and hence poorly informed citizens who would otherwise abstain. Lijphart’s (1997, p. 10) response to this is criticism is that CV ‘may serve as an incentive to become better informed’; a cross-national study by Gordon and Segura (1997) finds a small though statistically significant increase in political sophistication in countries with CV, but otherwise the evidence for CV promoting greater civic awareness is scant. Another untested argument along these lines is that CV leads to ‘higher quality’ political campaigns, since the parties are less concerned with mobilizing voters and more with the conversion of voters that CV guarantees will go to the polls. Lijphart (1997, p. 10) speculates that this diminishes (a) the role of money in political campaigns, and (b) incentives for candidates to use attack advertising.

Students of Australian politics have speculated whether CV leads to higher levels of party identification than might otherwise result (Mackerras and McAllister 1999), consistent with a view of voters as ‘cognitive misers,’ looking for heuristics to help them deal with a forced choice among parties. On the other hand, Australia’s minor parties are obvious beneficiaries and keen supporters of CV, providing an alternative for voters dissatisfied with major parties, but legally compelled to turn out. Another longstanding feature of CV is a higher rate of invalid ballots (e.g., Tingsten 1937) and ‘donkey ballots’ (where voters simply select the first name they see on the ballot); some instances of these phenomena are protests against CV itself.

Two final points. First, CV places an onus on citizens, but the state typically reciprocates with institutional mechanisms that make it easy to comply (e.g., weekend voting, ease of registration, widespread use of postal ballots): it would be rather odd if a state made voting compulsory but did not employ these other turnout-enhancing institutional mechanisms. According to Gosnell (1930, p. 209), ‘fines and penalties under a system of compulsory voting are a minor matter. The important feature of the system is that voting is regarded as a civic duty and the government does everything to impress upon voters this point of view.’ Second, the more seriously a country enforces its CV provisions, the more bureaucratic resources are required to maintain registration records and ensure compliance. In Australia, these two functions—the ‘carrot and stick’ of CV, as it were—are administered by the Australian Electoral Commission, a large and highly professional bureaucracy responsible for all aspects of Australian federal elections. Ensuring compliance with CV is just one of many AEC functions, and the bulk of its activities are to do with other aspects of election administration (e.g., redistricting, voter registration, public financing of campaigns, ballot design, staffing polling stations, vote tallying). Thus one (perhaps unintended) consequence of an aggressive system of CV is the centralization and professionalization of election administration. In turn, this may mitigate the problems that accompany decentralized and nonprofessional election administration, clearly evident in the aftermath of the 2000 US presidential election.

Bibliography:

1. Abraham H 1955 Compulsory Voting. Public Affairs Press, Washington, DC
2. AEC 1999 Compulsory Voting. Number 8 in Electoral Backgrounder. Australian Electoral Commission, Canberra
3. Australia, Joint Standing Committee on Electoral Matters 2000 The 1998 Election: Report of the Inquiry into the Conduct of the 1998 Federal Election and Matters Related Thereto. Parliament of the Commonwealth of Australia, Parliament House, Canberra
4. Castles F, McKinlay R D 1979 Does politics matter: An analysis of the public welfare commitment in advanced democratic states. European Journal of Political Research 7: 169–86
5. Fornos C 1996 Explaining voter turnout in Latin America. Master’s thesis, Lousiana State University
6. Franklin M 1999 Electoral engineering and cross-national turnout differences: What role for compulsory voting? British Journal of Political Science 29: 205–24
7. Gordon S B, Segura G M 1997 Cross-national variation in the political sophistication of individuals: Capability or choice. Journal of Politics 59: 126–47
8. Gosnell H 1930. Why Europe Votes. University of Chicago Press, Chicago
9. Hicks A M, Swank D H 1992 Politics, institutions, and welfare spending in industrialized democracies, 1960–82. American Political Science Review 86: 658–74
10. Hirczy W 1994 The impact of mandatory voting laws on turnout: A quasi-experimental approach. Electoral Studies 13: 64–76
11. Hooghe M, Pelleriaux K 1998 Compulsory voting in Belgium: An application of the Lijphart thesis. Electoral Studies 17: 419–24
12. Hughes C, Graham B D 1968 A Handbook of Australian Government and Politics, 1890–1964. Australian National University Press, Canberra
13. IDEA 1997 Voter Turnout from 1945 to 1997: A Global Report on Political Participation. International Institute for Democracy and Electoral Assistance, Stockholm
14. Irwin G 1974 Compulsory voting legislation: Impact on voter turnout in the Netherlands. Comparative Political Studies 7: 292–314
15. Jackman R W 1987 Political institutions and voter turnout in the industrial democracies. American Political Science Review 81: 405–23
16. Jackman R W, Miller R A 1995 Voter turnout in the industrial democracies during the 1980s. Comparative Political Studies 27: 467–92
17. Jackman S 1999 ‘Non-compulsory voting in Australia? What surveys can (and can’t) tell us. Electoral Studies 18: 29–48
18. Lijphart A 1997 Unequal participation: Democracy’s unresolved dilemma American Political Science Review 91: 1–14
19. Mackerras M, McAllister I 1999 Compulsory voting, party stability and electoral bias in Australia. Electoral Studies 18: 217–33
20. McAllister I, Mughan A 1986 Differential turnout and party advantage in British general elections, 1964–1983. Electoral Studies 5: 143–52
21. Nagel H 1988 Voter turnout in New Zealand general elections, 1928–1988. Political Science 40: 16–38
22. Pacek A, Radcliff B 1995 Turnout and the vote for left-of-centre parties: A cross-national analysis. British Journal of Political Science 25: 137–43
23. Powell G B Jr. 1981 Party systems and political system performance: voting participation, government stability and mass violence in contemporary democracies. American Political Science Review 75
24. Powell G B Jr. 1986 American voter turnout in comparative perspective. American Political Science Review
25. Power T, Roberts J 1995 Compulsory voting, invalid ballots, and abstention in Brazil. Political Research Quarterly 48: 795–826
26. Tingsten H 1937 Political Behavior: Studies in Election Statistics. P. S. King and Son, London