pexels photo 1550337

Unlike Australia, compulsory voting won’t work everywhere

pexels photo 1550337

This article was originally published on Election Watch

The introduction of compulsory voting in Australia has helped drive the development of integrated, effective and inventive electoral management that is both extremely well managed and trusted with unusually high and socially-even rates of electoral inclusion.

Compulsory voting (CV) exists where the state imposes a legal requirement to vote. The term is somewhat misleading because CV does not involve the requirement to mark the ballot formally. Due to the secret ballot, technically, it is only attendance at a polling place that can be legally required.

In 1911 compulsory registration was introduced at the Federal level to improve the accuracy of the electoral roll and CV voting at the Federal level became law in 1924. Aside from regularising the electoral roll, compulsory voting was introduced to address the problem of low voter turnout. It proved to be an extremely decisive and successful remedy causing turnout to surge dramatically from 58.7% (RV) in 1922 to 91.4% (RV) in 1925.

For the decades since 1924 turnout in Australia has consistently hovered in the 91-95% (RV) range.

Although Australia was not the first nation to use compulsory voting (Switzerland and Belgium got there first) it is one of the few advanced democracies and the only English speaking country to use it. It is also a good example of a successful regime that is constantly experimenting with enhancements to electoral inclusiveness, accessibility and integrity.

Thirty democracies or so worldwide claim to use compulsory voting but a much smaller number use it –or have used it– with reasonable levels of support and enforcement.

Vanuatu and Samoa are currently working towards introducing it as a means to fight electoral corruption and improve the representation of women.


For compulsory voting to be effective, in terms of assuring high turnout, accepted, in terms of enjoying high levels of public tolerance, and appropriate, insofar as it is not being used as a mechanism to forge consent or legitimize one-party contests, a number of conditions should be met.

Foremost is a well-established system of democratic institutions with adequate civil and political rights protection (including universal suffrage). The political structure should also have the apparatus of constitutionalism; limits on political power; rule of law; and free, competitive and fair elections. It should be what I call an ‘authentic democracy’. The following are also desirable:

  • Fairly well-developed national infrastructure so that elections can be conducted properly with high levels of integrity and adequate follow-up for non-compliance.
  • There should also be reasonable levels of co-operation between regional and central governments, particularly within Federal structures and electoral offices should be professional, well-funded, independent and accountable.
  • In order to ensure that the obligation is not burdensome, voting should be relatively easy with few opportunity and transactions costs to voters (see below).
  • There need to be sanctions but they should be applied consistently yet without zealotry. This helps to minimise the coercive character of the compulsion.
  • It is also preferable if there is some degree of genuine choice reflected in the political offer and the voting system in use. Proportional representation (PR) and preferential voting are good ways of ensuring that voters can achieve at least some degree of meaningful representation. This undoubtedly enhances public acceptance of the compulsion.

Some critics say there are less coercive ways of raising turnout: these methods include  Saturday voting, absentee voting, lowering the voting age, proportional representation, offering incentives to vote, improving registration, placing voting booths in malls and online voting. But none of these work nearly as effectively as compulsory voting (Louth and Hill, 2005) which increases turnout from around a minimum of 12 to a maximum of 30+ percentage points. Further, when it is administered properly, it is the only institutional mechanism that can achieve turnout rates of 90 percent and above on its own (Lijphart, 1998). And its effect on turnout is immediate and lasting.

Levelling Effect

Empirical studies have shown that CV reduces the gap in voter turnout between low-and high-income voters (Fowler, 2013; Singh, 2015), educated and less educated citizens (Gallego, 2015; Dassonville and Hooghe, 2017) and younger and older individuals (Irwin, 1974; Singh, 2011, 2015). It also closes gender participation gaps (Cordova and Rangel 2017). This was shown vividly in the Australian case where the introduction of compulsory voting immediately closed the 10 percentage point participation gap between men and women (see Brett, 2019: 136).  This tendency towards closing voter turnout gaps between advantaged and marginalised groups happens wherever CV is administered properly and with reasonable sanctions.


This figure has remained fairly stable for decades.

This is probably related to a number of things that are done quite well in Australia:

  • Voting is easy in Australia because the state handles most of the opportunity and transactions costs involved. Due to the fact that people are required to vote, ‘politicians and electoral officials have gone to considerable lengths’ to make voting as simple, accessible and pain-free as possible (Mackerras and McAllister, 1999: 223).
  • The state provides more than just a universal, formal entitlement to vote; it also practically enables almost everyone to be present on election day. Groups that are particularly targeted for assistance are precisely the same groups that tend to abstain under a voluntary system:  aging and immobile people, the homeless, those living in remote regions, prisoners, people with a disability, those who are ill or infirm, housebound, hospitalised, living abroad, approaching maternity, have literacy and numeracy problems or are from a non-English speaking background.
  • There are also special provisions for ‘silent enrolment’ (for those concerned that having their name on a public roll endangers themselves or their families) and itinerant enrolment (‘for homeless people, or people who travel constantly and have no permanent fixed address’).
  • Australian electoral commissions actively facilitate registration; provide electoral education; offer absent voting, mobile polling and postal voting; ensure that elections are held on a Saturday and that polling booths are generally close at hand. Over the course of any given Federal election up to 500 mobile teams will visit 2000 special hospital locations; mobile teams will visit 300 or so remote outback locations and over 40 prisons; there will be hundreds of pre-poll voting centres and around 100 overseas polling places to which approximately 3 tonnes of election-related and staff training material will be air-freighted immediately prior to polling. Even Australian citizens living in the Antarctic and based on Antarctic supply ships will be supplied with voting material and facilities.


Failure to vote cases are usually dealt with through ‘please explain’ letters provided by abstainers themselves. This is really an honour system: if the reasons are deemed by the relevant electoral commission to be ‘valid and sufficient’ no penalty is applicable. It is notable that well below 1 percent of the Australian electorate is ever faced with a fine or court attendance in any given election period (Mackerras and McAllister, 1999: 224). Voting is so stress-free in Australia that it is just as easy and less time-consuming to attend a polling place than it is to deal with the administrative consequences of non-compliance.


Despite occasional problems, Australian elections tend to be free from corruption and rigging scandals. Further, Australian electoral offices are organised, reasonably integrated (despite Federalism), professional, well-funded, independent, accountable and apolitical. Australians have faith in the mechanics of their electoral process with the effect that the accuracy and procedural legitimacy of outcomes is rarely—if ever—disputed.

When asked what they like most about their democracy, most Australians rated highly that “Australian elections are free and fair” (Evans et al, 2018). They also report being content with their electoral choices. According to the Voter Choice Project (2018), which is investigating how voters decide who to vote for, 80 per cent of Australians are happy ‘with how they voted in 2016, and would not have voted differently if they had known how close the result would be’ (6 per cent were undecided and 12 per cent were unhappy).

Australians also ‘place more value in obeying laws’ than do citizens in comparable voluntary-voting settings, possibly because they all voted for those who make the laws, leading to a sense of investment in them. In a sense then, Australians feel that they have signed a social contract by voting. Australians are also more likely to see voting as more of a duty than a right. Overall, then, most Australians regard compulsory voting as an acceptable imposition on their individual autonomy.

Satisfaction, Trust and Looming Problems

CV is also correlated with higher levels of ‘satisfaction with democracy’ both in Australia and elsewhere (Birch 2009: 132−3). Historically, the Australian Election Study has found that a significant majority of voting-age Australians (between 70 and 80 per cent) are ‘satisfied with the way democracy is working in Australia’ (Cameron and McAllister 2017). However, these levels have recently started to decline (Evans et al 2018). The Social Research Institute at Ipsos finds considerably lower levels of trust and satisfaction with Australian democracy within the electorate, particularly among the young (Evans et al 2018). Declining trust is not, however, associated with the legitimacy of the electoral process itself but with concerns like: lack of accountability for ‘broken promises’; failure to prioritise the issues most deserving of attention; and politicians’ perceived lack of independence from either ‘big business’ or unions (Evans et al 2018).

Declining engagement among the young, despite compulsory voting, is also a concern in the Australian setting. In recent decades there has been a worrying drop in the number of eligible young Australians who either do not register to vote or who abstain from voting. The Australian Electoral Commission estimates that at the 2016 federal election almost 700,000 eligible voters were not enrolled to vote. It has been suggested that this is attributable to the ‘current combination of light sanctions (which is not indexed [to] economic growth)’ and the ease of being able to have the sanction waived (Sheppard 2018).

Australia’s political parties have done little to address declining voter satisfaction, trust and engagement. This is likely related to compulsory voting; whereas in comparable voluntary-voting democracies parties are incentivised to mobilise potential supporters to cast a ballot, in compulsory-voting Australia parties can afford to be more complacent (Sheppard 2018).

On the positive side, though, because compulsory voting mobilises voters, there is less corruption associated with campaign finance in CV settings (Birch 2009). Therefore, Australians continue to have a reasonable level of faith in the mechanics of their electoral process.

Further, citizens in CV regimes are not more apathetic or less politically sophisticated as Jason Brennan (Brennan in Brennan and Hill 2014) has claimed. In fact, they are more attentive to the behaviour of their governments than are citizens in voluntary settings and are therefore more politically knowledgeable (Shineman 2009, 2010; Sheppard 2015; Córdova and Rangel, 2017).

Governments within Compulsory Voting settings also tend to be more responsive

Because compulsory voting delivers high and socially-diverse turnout it makes governments more representative of the people as a whole. In general, governments are more attentive to the demands of habitual voting groups at the expense of those who abstain (usually people from low SES groups). Voting in voluntary systems is concentrated among the more prosperous members of society; it therefore tends to help people who are already better off. As the eminent political scientist Walter Dean Burnham once put it: “if you don’t vote, you don’t count”.

There is no shortage of studies detecting a strong relationship between electoral participation rates and implementation of public policies that affect spending in areas like health services, education and public amenities. This relationship is particularly true of redistribution that affects those on lower incomes (see Hill in Brennan and Hill, 2014, Chapter 6 passim). The bottom line here is that governments in compulsory voting settings are responsive to more and different kinds of citizens.  Attention is more evenly distributed, as it should be in a properly functioning democracy.

An Unjustified Imposition?

Some claim that CV is an unjustified imposition on personal autonomy. Yet we compel citizens to do certain things all the time. Compulsory voting imposes a relatively minor restriction on personal freedom compared to other problems of collective action such as paying taxes, jury duty, and compulsory school attendance.

The autonomy objection is only fatal if one insists that choice to abstain is more important than other key democratic values that CV can—and does– serve, namely:

  • Legitimacy
  • Representativeness
  • Political equality (one-vote, one-value)
  • Inclusiveness
  • Minimisation of elite power
  • Uncorruption
  • Responsive government