Inside Story

A complex compulsion to vote

Australians are more likely to vote than their international counterparts — but does that mean the system should stay as it is?

Peter Brent 13 October 2020 1587 words

Compulsory voting certainly increases turnout — but by how much? Australian Electoral Commission


How good is Australia’s compulsory voting system?

No, this is not a prime ministerial affirmation, but an actual question.

Most Australians — 70 per cent or more — support compulsory voting. But no Australian alive today has voted at a federal election under voluntary voting. We don’t know anything else.

Defenders of compulsion (who include a huge majority of academic political scientists) routinely point to the horrors of US electoral practice, on full display at the moment, as if that’s the alternative, as if it’s the only other country on Earth. It’s true that America presents a template for how not to run elections, but voluntary voting is a small part of that mix. The big majority of advanced democracies don’t force people to vote. There’s a little one nearby, across the Tasman, holding one such event in a few days’ time.

The best thing to be said about compulsory voting in Australia is that it does what it is designed to do: increase turnout. A lot. If it were done away with tomorrow, we would probably reach a new, much lower turnout after several elections. How much lower? Well, compulsion’s introduction in 1924 boosted turnout from 59 per cent of those enrolled in 1922 to 91 per cent in 1925. So perhaps we’d drop back to around 60 per cent. Or some other figure. But it would be substantially lower.

The $20 non-voting fine is trivial for a big majority of Australians, and probably less of a deterrent than the paperwork involved. For a large slab of the population compulsion is a gentle prod to swing by that primary school on the way to wherever on a sunny Saturday (or, increasingly, to vote early).

What’s not to like?

Some people object on the grounds of individual rights. But it’s only compulsory to “turn out”; you can always vote informal. Still, it’s a minor infringement of liberty, like many others our society insists on, like stopping at a red light even if the road is empty.

Others reckon compulsion drags the uninformed to the ballot box, but that view involves subjective assessments about what being informed means. It’s not obvious that people who follow the political soap opera are better equipped to decide who governs us than those who have better things to do. You’ve seen the internet, haven’t you? Anyway, all votes are supposed to count equally.

Is high turnout better than low turnout? I believe it is. Our turnout is impressive by international standards. But it’s not as good as we like to think.

New Zealand, the seventh antipodean colony — the one that decided not to join the upcoming federation in the 1890s — gets by okay without it. Its last election in 2017 saw an eligible voter turnout of around 71 per cent. Our numbers are better than that, but the figures we wave around are misleading.

The best way to make comparisons is to calculate what percentage of the estimated number of eligible voters (whether or not they’re on the electoral roll) actually vote. At the 2010 election, Australia’s figure was around 85 per cent; at the next two it was 87; and last year it hit 89. (We are actually bucking the worldwide trend of declining turnout, a function of our improving electoral roll and more early voting availability, particularly pre-poll.)

The estimated figure for the 2016 US presidential election is 59 per cent.

There is also a question of whether to include informal votes in our calculation of turnout. Is simply attending the polling place enough? By international standards we have a very high level of informal votes (around 5 per cent), about half of them deliberate and half accidental. The deliberate ones are nearly all a result of mandated turnout: they are people choosing not to vote despite being forced to attend. I believe they shouldn’t be counted in turnout. The accidental ones flow from our complicated voting system, and it’s not clear that they should be included either.

Making international comparisons is complicated, though, because no accepted practice seems to exist for making the calculation.

From what I can determine, America’s turnout figures don’t include what they call “invalid votes.” (The International Institute for Democracy and Electoral Assistance doesn’t have any post-2006 figures on the number of invalid votes in the United States. And the numbers in this table seem to exclude totally blank ballot papers.) If we take all the informal votes out of Australia’s number, we drop to a turnout of 84 per cent in 2019 (and a touch below 80 in 2010).

Still good, but getting less to write home about.

American advocates of compulsion (there are a lonely few) tend to cite the Australian experience to boost their case. Like, who else are they going to hold up — Brazil and Uruguay? Belgium and Luxemburg? No, good old unpretentious, un-European, almost-American us. This recent Brookings Institution/Ash Center paper cites the Australian Model, particularly our ease of voting, but most of its features have little to do with compulsory voting and can be found (again) across the ditch, and in fact in most of old Western Europe.

But there is one significant problem with compulsory voting in Australia that gets worse as the electoral roll becomes more complete.

Getting on the roll has been compulsory since 1911, but no one has been fined or prosecuted for breaking this law since the 1980s. Instead, a rolling amnesty is designed to encourage people to register. But it creates an incongruity: if you avoid being on the roll, and you don’t vote, the state doesn’t bother you. Enrolled people who don’t vote will get a fine.

Until 2012, simply moving house was enough to get unenrolled. The AEC would find out from other government agencies that you’d left an address and take you off the roll. It would then send you enrolment forms at your new home, several times, but if you didn’t fill in and return them there was nothing it could do. And people, increasingly, weren’t returning those forms.

As a result, the roll was becoming less complete. Lots of people would try to vote and find they couldn’t. So, in 2012, the AEC was finally given the power to do the other bit: put people on the roll at their new address, and find others who had not been on the roll, without the need for a signed form. It’s what most citizens expect in the modern age, and indeed what many assumed was happening. It’s why the roll has improved dramatically over the past eight years. Remember the “missing 1.5 million” the commission publicised (also in 2012)? It’s down to 600,000 or less (out of a bigger population).

Terrific. At the last election the roll was around 97 per cent complete, up from 91 per cent ten years ago. But because of this, non-voter numbers have skyrocketed. Not surprisingly, people who don’t like to vote are over-represented among those new, passive enrollees. And unlike other minor transgressions such as littering, jaywalking or illegally parking, the chances of getting caught and punished are pretty close to 100 per cent.

The number jumped most spectacularly, by 46 per cent, between the 2013 and 2016 elections. In 2016, the commission went after some 970,000 people. (In 2019 it was probably a bit under this.) Now, $20 isn’t much to most people, but it’s a lot to a few. Some people who receive the please-explain letter have an accepted excuse. Others simply fib and pretend they do — they were sick or looking after a sick loved one, or insist that they did vote (and someone else’s name must’ve been ticked off). The commission will accept it, but this doesn’t engender respect for those who run our elections. Ignoring the letters, or not paying the fine, will usually end up in court and a $180 fine plus costs. The numbers clogging up our courtrooms in this manner run into the thousands.

It costs the AEC millions to go after non-voters (and sadly the fines go not to them but to general government revenue). And the letters disproportionately go to low-income (particularly Indigenous) Australians.

What penalties do other countries with compulsory voting use? Australia sits at the more heavy-handed end. Some nations don’t bother to enforce at all; others, counterintuitively, take people off the electoral roll as punishment.

In early 2016 the Turnbull government greatly improved the Senate system by eliminating inter-party preference tickets and bringing in optional preferential voting above the line. The one potential drawback was the prospect of millions of Australians continuing to tick just one box, which would still count as a formal vote but would not see preferences flow to any other party. The solution? Have polling staff tell voters to “number at least six,” and also make that clear on the ballot itself. The implication being that fewer than six and your vote doesn’t count. Not quite a lie, because it’s not actually stated. Anyway, it works (and we have not yet seen a Langer-style campaign alerting voters to the fact that they can indeed number fewer than the six).

Should we adopt that strategy for voting? In other words, maybe we should become one of those countries where voting is compulsory but no one actually gets punished for not voting. Alternatively, should we chase up every, say, tenth, or one-hundredth non-voter?

The compulsory voting stick is too unwieldy. Something more gentle is needed.

Or perhaps we should do away with compulsion all together. •