Why do small places govern themselves better?
Across 2,105 Swiss communes voting on the same questions, the smaller the place, the more people turned up. What does a system lose as it grows?

Art: Omxus Press

Cover of Incentive and Outcome (Omxus Press)
On 16 July 1924 Herbert Payne, a Nationalist senator from Tasmania, brought in a private senator's bill. He was worried: turnout had slid from over 71% in 1919 to under 60% at the 1922 federal election. His bill met little opposition, and it went through Parliament almost without argument (Biographical Dictionary of the Australian Senate, n.d.). Then, without much anybody noticing, a country changed who it was.
At the election of 1922, 59.38% of enrolled Australians cast a vote for the House of Representatives. At the next one, in 1925, it was 91.38% (Australian Electoral Commission, 2023). Nobody became a better citizen in those three years. A law changed, with a stated purpose as plain as a council notice: "to increase voter turnout and reduce party campaign expenditure" (Australian Electoral Commission, n.d.). Today, if you don't vote and don't have a good reason, you pay a $20 penalty.
Ask an Australian why they vote and you'll hear about duty, or the sausage sizzle, or "it's just what you do". Almost nobody says "to dodge twenty bucks". That's the interesting part. A rule became a habit, the habit became an identity, and then everybody forgot it was ever a rule.
Take every federal election from 1901 to 2022, 47 of them, and look around the two big rule changes: compulsory enrolment in 1912 and compulsory voting in 1924.
The jump from 1922 to 1925, 32 points, is the biggest change between any two elections in Australian history. The next biggest is 11.3. Allow for the earlier jump and any slow trend, and compulsory voting added about 24.5 points.
A political scientist reached the same answer a different way. Anthony Fowler (2013) compared states that made voting compulsory at different times and also found turnout rose 24 points. And he found something sharper: before compulsion, the wealthy voted far more than working people. Once everyone voted, Labor's share of votes and seats rose by 7 to 10 points. That rule didn't just raise a number. It changed who got heard, and that's how one bill from a Tasmanian conservative helped shift a century of Australian politics towards working people. Almost certainly not what he had in mind.
Here's the odd thing. Twenty dollars can't explain 94.6%. Plenty of us would pay more than that to get our Saturday back, and enforcement is mild. So what's holding it up?
Everyone else is. You vote because your neighbours do, because the school's got a sausage sizzle, because you'd feel strange telling your mum you didn't. A law made voting normal, and normal does the rest. And incentives only work that well when they say the right thing. In a Swedish experiment, offering women about $7 to give blood nearly halved the number willing to do it (Mellström & Johannesson, 2008). When an Israeli day-care centre fined parents for picking up their kids late, more parents came late, because the fine turned guilt into a price (Gneezy & Rustichini, 2000). Pay someone for a gift and you've told them it's a job. Fine someone for a favour and you've told them it's for sale.
So why didn't the voting fine do that? Maybe because it's so small, and because it arrived with a message: this is what we do here. Australia's law told people voting was who they were, and they believed it. That's our reading, and nobody's run the experiment. But something held for a hundred years.
Australia is one country, and this is a natural experiment rather than a trial. Other election rules changed around the same time, and turnout is counted against the roll, which itself became compulsory in 1912. A hundred years of data can't rule out every coincidence, though Fowler's separate method landing on the same 24 points makes one unlikely. Nor is compulsion free. Some people resent it, and a forced vote isn't an informed one. It's fair to ask what a person's vote is worth if the only reason they showed up was the law, and the honest answer is that nobody knows how many of us that describes.
People who want change usually start by trying to change people. They run a campaign, write a stern column about apathy, and blame the young. Payne didn't lecture anyone. And a third of the electorate went from staying home to voting, the same people, three years apart.
That should give anyone hope who's sick of being told human nature is the problem. If we want more people voting, giving, helping, turning up when a neighbour's heart stops, we don't need different people. We need a different design.
Across 2,105 Swiss communes voting on the same questions, the smaller the place, the more people turned up. What does a system lose as it grows?

Art: Omxus Press
Colorado lawmakers had $120 million of requests and $40 million to give. So they tried a vote that lets you say how much you care. Then a judge stepped in.

Cover of Quadratic Voting (Omxus Press)
Donation caps, fast disclosure, public funding, open diaries, and a voucher every voter spends. Seattle has run one since 2017. Here's what happened.

Photo: Betterkeks, CC BY-SA 4.0, via Wikimedia Commons