The Best Democracy You've Never Properly Thanked

I've spent a career being briefed on systems designed to fail gracefully.

Aircraft have redundancies. Command structures have fallbacks.

Good design assumes something will go wrong and plans for it anyway.

Australia's electoral system is one of the best pieces of political engineering most Australians have never actually understood. We complain about it every three years; usually, it's about the two main parties, or "the Prime Minister is …..", (ironic that every single one seems to elicit that same response), and then, historically, we go back to not thinking about it.

That's a shame, because it's quietly one of the most sophisticated voting systems on earth, and understanding why matters if you actually care about the health of the country's democracy and political future.

I think it's worth setting out clearly. Given the work I do through the Odyssey Leadership Foundation, getting all Australians, not just young Australians, to understand this properly is crucial: rather than absorbing it as vague civics trivia.

It's also one of a relatively small club of countries running a full Westminster system, most of them former British colonies or dominions, and what's even more unique is how Australians vote within that system.

It is not perfect. But it is very, very close, and the places where it goes wrong are worth naming honestly rather than glossing over.

Three ways to run an election

First, we need to understand how we elect someone into power.

Strip it back, and most democracies choose between three broad families of voting systems.

For perspective, just over 20 of the roughly 193 recognised countries worldwide have some form of compulsory voting on the books. Australia is one of them, and one of the few that actually enforces it properly at every election.

I'll explain each one with a thought experiment. There are five candidates or parties running for election: Green, Yellow, Orange, White, and Black.

First-past-the-post (plurality voting). Whoever gets the most votes wins, full stop, regardless of whether that's 20% or 75%.

In this example, Green is elected leader, even though 75% of the people don't agree.

The UK and the US both run on this for their main legislative elections.

It's simple and legible, but it produces some brutal distortions. A party can win a majority of seats on a minority of the national vote.

Votes for anyone other than the top two candidates are effectively wasted, which is exactly why voters feel pressured to vote "strategically" rather than for who they actually want.

It also entrenches two dominant parties almost by mathematical necessity, a pattern political scientists call Duverger's Law.

For our thought experiment, that means:

Party	Vote
Green	25%
Yellow	20%
Orange	20%
White	12%
Black	23%
Total	100%

Green wins. Not because most people wanted Green, but because Green had the single largest slice of a badly fractured pie. Three out of every four voters backed someone else, and it makes no difference at all.

Proportional representation (list PR). Instead of one winner taking the seat, seats are handed out across a whole electorate roughly in line with each party's share of the vote it receives.

Countries like Germany and the Netherlands run variants of this.

Take our same result and imagine it's a 20-seat proportional parliament instead of a single seat:

Party	Vote	Seats (out of 20)
Green	25%	5
Yellow	20%	4
Orange	20%	4
White	12%	2
Black	23%	5

Nobody reaches the 11 seats needed for a majority. Green and Yellow together only get to 9, still short. So the parties now have to go into a room and negotiate; maybe Green, Yellow, and Orange team up to scrape together 13 seats and form a government together.

Nobody voted for that specific combination. It gets decided after election day, by the parties, not by the voters.

It's fairer in the sense that no vote is "wasted"; every party gets seats roughly matching its support. But you've traded a clear result on election night for a negotiation you have no vote in.

Preferential voting (ranked-choice/instant-runoff, with a proportional variant for multi-member seats). This is Australia's system, and it's built to avoid both of the problems above.

Instead of picking one candidate, you number all five in order of preference. If nobody gets an outright majority (50%+1) on first preferences, the last-placed candidate is eliminated, and their vote goes to whoever that voter ranked second. Repeat until someone crosses the line.

Same starting numbers as before: Green 25%, Black 23%, Yellow 20%, Orange 20%, White 12%.

Round 1: Nobody has 50%. White is last, so White is eliminated. White's voters' next preferences split across the field: Yellow picks up 6, Orange picks up 4, Green picks up 2.

Party	Running total
Green	27%
Yellow	26%
Orange	24%
Black	23%

Round 2: Still nobody at 50%. Black is now last, so Black is eliminated. Black's voters split: Yellow picks up 14, Green picks up 9.

Party	Running total
Yellow	40%
Green	36%
Orange	24%

Round 3: Orange is last, so Orange is eliminated. Orange's voters split: Yellow picks up 15, Green picks up 9.

Party	Running total
Yellow	55%
Green	45%

Yellow crosses 50% and wins, with a genuine majority of the electorate ranking them ahead of Green somewhere on their ballot, even though Yellow started the count tied for third, not first.

That's the whole point of the system in one example. Under first-past-the-post, Green would have won this election outright on a 25% primary vote, with three-quarters of voters getting a result they didn't want. Under preferential voting, the count continues until whoever wins actually has broad support, not just the largest single splinter.

For the House of Representatives, this is exactly how we count votes: instant-runoff, one member per seat. For the Senate, we use the proportional method (single transferable vote, spread across multiple seats per state), which is why the Senate looks so different in composition to the House, and why minor parties and independents have a real foothold there.

The case for Australia's model being the strongest of the three isn't sentimental nationalism. It simultaneously solves the two biggest structural failures of the other systems. It kills the spoiler effect that cripples first-past-the-post, because you can vote for a minor party or independent first and have your vote flow to a major party if your favourite is eliminated, so voting for who you actually believe in is never "wasted" in the way it is in the UK or US.

And unlike a pure list PR system, the House still delivers local, named representatives and typically a workable government on election night, while the Senate's proportional design gives smaller parties and independents a real, structurally guaranteed voice rather than a token one. Add compulsory voting and an independent, non-partisan electoral commission drawing boundaries (the AEC, not politicians), and you get a system with unusually high legitimacy: turnout north of 90%, and boundaries that aren't gerrymandered by whoever happens to hold power.

We are not a two-party system. We stopped being one a while ago.

Here's the bit that surprises people, including plenty of politically engaged Australians: we talk about "the two-party system" out of habit, but the data says otherwise, and has for some time now.

At the 2025 federal election, the combined primary vote for Labor and the Coalition fell below two-thirds of the national vote for the first time in the modern era, with roughly a third of first-preference votes going to "others." Compare that to Malcolm Fraser's 1975 landslide, where the Coalition alone polled 53% and minor parties and independents scraped together barely 4%. The drift has been steady across decades: Hawke in 1983, Howard in 1996, and now Albanese in 2025; each landslide has been won with a smaller primary vote share than the last, because more of the electorate is voting outside the major-party binary from the outset.

Look at the crossbench. The 2025 House of Representatives seated the Greens, Centre Alliance, Katter's Australian Party, and ten independents, including the so-called "teals," alongside Labor and the Coalition. The Senate is even more clearly multi-party: after 2025, no single party holds a majority, and the chamber includes Labor, the Coalition, the Greens on eleven seats, a resurgent One Nation on four, and a scattering of other independents and minor-party senators. That is not a two-party parliament with some noise around the edges. That's a genuine multi-party legislature; it's just that our voting method for the House tends to compress the final result into a two-candidate contest in any given seat, which creates the optical illusion of a two-party system even as the underlying vote has fragmented.

This matters, and it's a big part of why the Foundation puts civic education front and centre. A citizenry that thinks it's choosing between two options is a citizenry that under-uses the tools it actually has. Independents and minor parties are not a protest vote or a wasted vote in Australia's system; they are a structurally viable path to representation, and the numbers prove more people are using that path every election cycle.

Why multi-party matters, properly argued

The case for multi-party systems generally is that they force negotiation, dilute the ability of any single faction to capture the whole machine, and give genuinely different constituencies (regional, ideological, single-issue) an actual seat at the table rather than forcing them to hold their nose and vote for whichever major party is marginally less bad on their issue.

Critics of multi-party systems argue the opposite: that they fragment accountability (who do you blame when a coalition of five parties makes a bad decision?), can produce unstable government, and let small parties with a narrow base wield disproportionate leverage in a hung parliament.

Australia's version threads that needle better than most because the House still usually delivers a clear majority government even with a fragmented primary vote, thanks to preferential flows consolidating behind a winner, the same mechanism that handed Yellow the win in our thought experiment. But the Senate's proportionality means that the government still has to negotiate with a genuinely diverse upper house to pass legislation. You get accountability and majoritarian clarity in one chamber, and a check on power through proportional representation in the other.

That's the design that most first-past-the-post systems and most pure-PR systems each fail to achieve on their own.

Where it actually goes wrong

None of the above is an excuse to be starry-eyed about it, and I'd rather argue the case honestly than sell it as flawless. It isn't.

Safe seats breed complacency on both sides. A huge number of House seats are "safe" in the AEC's own classification, held by margins over 60% after preferences. In safe seats, the incumbent party has little electoral incentive to actually engage with voters between elections, and preselection battles inside the party matter more for who represents you than your actual vote does.

Genuine contestability, the thing that's supposed to keep politicians honest, is concentrated in a shrinking number of marginal seats while most of the country watches from the sidelines.

Preference deals and "how-to-vote" cards distort what voters think they're doing. Because preferences flow according to how a voter numbers their own ballot, many people simply follow the party's how-to-vote card rather than making an independent choice, preference by preference, which hands parties real influence over where votes ultimately land, especially in tight contests decided by preference flows rather than primary votes.

It's not fraud, it's legal and transparent, but it means the "instant runoff" ideal of every voter making a genuinely independent ranked choice is softer in practice than in theory.

The Senate's old group voting ticket system was a genuine scandal until it was fixed. Before the 2016 reforms, micro-parties could strike opaque preference deals with each other and get candidates elected off primary votes in the low single digits through preference harvesting nobody could see or understand.

That's been cleaned up with optional preferential voting above and below the line, but it's a useful reminder that a good system on paper can be quietly gamed until someone actually audits it and fixes it.

Structural advantages still favour the majors, even in a multi-party landscape. Public funding formulas, media access, name recognition, and party machine resources all still tilt heavily toward Labor and the Coalition, meaning the "multi-party" reality at the ballot box doesn't yet translate into equal footing in the contest itself.

Independents and minor parties are winning more seats, but they're doing it while running uphill against structural incumbency advantages the majors don't have to think about.

Donkey votes and informality are a real, measurable leakage. Some voters, forced by compulsory voting to turn up, simply number the ballot in the order candidates are printed, and informal (invalid) votes run at a non-trivial percentage every election, disproportionately affecting seats with large multicultural or lower-literacy populations where the ballot's complexity becomes a genuine barrier rather than a neutral inconvenience.

None of this breaks the case for the system. It's a case for maintenance. Every one of these flaws has either already been partially fixed (the 2016 Senate reforms), is fixable with better civic education (informal votes, how-to-vote literacy), or is a genuine ongoing debate worth having honestly (safe seats, major-party funding advantages) rather than pretending it doesn't exist because the underlying architecture is sound.

The actual point

Australia's electoral system is closer to getting the balance right than almost anywhere else in the world, not because it's flawless, but because it was built to absorb failure modes that cripple other democracies: the wasted vote, the spoiler effect, the winner-takes-all majority built on a minority of support, and it mostly succeeds.

The fact that we're a genuine multi-party democracy wearing a two-party costume is a story most Australians don't know they're living inside.

That gap, between what the system actually is and what most people think it is, is exactly the kind of civic literacy problem worth spending a career on.