The Australian ballot in America: how a state-printed ballot reshaped parties, candidacy, and design

The Australian ballot in America is the government-printed ballot, listing every qualified candidate and marked in secret, that American jurisdictions began adopting in 1888, with Louisville, Kentucky, and Massachusetts among the first.⁠¹ Its best known effect, secrecy, is covered in the secret ballot. This chapter covers the second effect: once the state printed the ballot, it had to decide who appeared on it and how, and those decisions became a lasting field of law and political strategy.⁠^2,6

In plain words

Before 1888, American voters usually brought a ballot printed by a political party. The party decided which names were on it. The government counted whatever came in.

Then states began printing one official ballot themselves. That sounds like a small change. It was not. If the government prints the ballot, the government has to decide whose name goes on it. It has to decide the order of names, whether party labels appear, whether a voter can pick a whole party with one mark, and whether a voter can write in a name.

Each of those choices helps some candidates and parties more than others. Courts have spent decades deciding which rules are fair. Political scientists have spent decades measuring how layout changes votes. A bad design can even cause voters to pick a candidate they did not mean to choose.

So the lesson is simple. Whoever designs the ballot shapes the result. Any new way of asking the public, including this framework, inherits that responsibility.

Defined terms

• Party ticket. A ballot printed and supplied by a party, used widely in the United States before reform. See the secret ballot.
• Official (Australian) ballot. In John Henry Wigmore’s 1889 description, “an official ballot containing the names of all candidates, printed and distributed under state or municipal authority,” combined with “compulsory secrecy of voting”.⁠¹
• Ballot access. The legal rules that decide which candidates and parties are printed on the official ballot, such as petition signatures or prior vote thresholds.⁠⁴
• Fusion. The practice of more than one party nominating the same candidate, whose votes on each party line are combined.⁠²
• Straight-ticket voting. A ballot option that “allows voters to choose a party’s entire slate of candidates with just a single ballot mark,” in the National Conference of State Legislatures’ definition.⁠⁵
• Office bloc and party column. Two layouts. An office bloc ballot groups candidates by office; a party column ballot lines up each party’s candidates in a column. Layout effects on turnout are discussed in the secret ballot.
• Write-in. A vote for a person not printed on the ballot, where law allows it.⁠³

Evidence

Primary source: what reformers said they were fixing

Wigmore, writing in 1889, compiled the ballot laws of several countries and introduced them with a case for reform. He described British elections before reform as places where “corruption, fraud, and intimidation” “thrived abundantly,” and where “Landlords intimidated their tenants, and marched detachments to the polls to vote in their interests.” He argued the secret official ballot “checks bribery and all those corrupt practices which consist in voting according to a bargain or understanding”.⁠¹

He also pointed to a second American problem: control of nominations. Under the party ticket system, he wrote, “the only avenue to an election is through a nomination by a caucus or a convention,” and for an independent candidate “the road is completely blocked.” He treated open nomination as “an integral and invaluable part of the Australian system of ballot reform”.⁠¹ Reformers, in other words, hoped the official ballot would open candidacy, not narrow it.

Primary source: first adoptions

Wigmore records that Louisville applied the system to its municipal elections by “the Act of Feb. 24, 1888,” that the Massachusetts act “was approved by the Governor May 30, 1888,” that Wisconsin had passed a 1887 law for cities of more than 50,000 inhabitants, and that a New York bill passed both houses in 1888 but “was vetoed, however, by Governor Hill”.⁠¹ The national spread that followed is summarized in the secret ballot.

Law: the official ballot created a law of fusion

The Supreme Court described this transition in Timmons v. Twin Cities Area New Party, decided April 28, 1997. After the 1888 presidential election, “which was widely regarded as having been plagued by fraud, many States moved to the ‘Australian ballot system,’” under which “an official ballot, containing the names of all the candidates legally nominated by all the parties,” was used. “During the same period, many States enacted other election-related reforms, including bans on fusion candidacies,” and in the twentieth century “fusion has become the exception, not the rule”.⁠²

The Court held that Minnesota’s fusion ban “does not violate the First and Fourteenth Amendments.” It accepted that states may enact “reasonable election regulations that may, in practice, favor the traditional two-party system,” and that “States also have a strong interest in the stability of their political systems”.⁠² The opinion is a direct link between nineteenth century ballot reform and modern limits on minor parties.

Law: who gets printed

Because the state prints the ballot, the state must set conditions for appearing on it. In Williams v. Rhodes, decided October 15, 1968, the Court reviewed Ohio laws that required a new party to file petitions signed by electors totaling 15 percent of the ballots cast in the last gubernatorial election, while the two established parties kept their place by winning 10 percent. The Court found that “these various restrictive provisions make it virtually impossible for any party to qualify on the ballot except the Republican and Democratic Parties,” and held that the laws, taken as a whole, were “an invidious discrimination”.⁠⁴

In Burdick v. Takushi, decided June 8, 1992, the Court upheld Hawaii’s ban on write-in voting. It said the purpose of the election process is “to winnow out and finally reject all but the chosen candidates,” and set out the standard still used today: severe burdens must be “narrowly drawn to advance a state interest of compelling importance,” while “reasonable, nondiscriminatory restrictions” are generally justified by a state’s regulatory interests.⁠³ The same official ballot that Wigmore hoped would open nominations became, in the Court’s words, a tool for winnowing.

Current law: straight-ticket options

Whether voters can choose a whole party slate with one mark is a design decision made by state law, and it changes. The National Conference of State Legislatures, in a page updated December 9, 2025, lists six states that offer straight-ticket voting (Alabama, Indiana, Kentucky, Michigan, Oklahoma, and South Carolina, with Indiana excluding at-large races) and records a series of states that eliminated it, including Georgia in 1994, Illinois in 1997, North Carolina in 2013, Texas in 2017 (effective 2020), and Pennsylvania in 2019. Michigan eliminated the option in 2016 and restored it after a 2018 ballot proposal.⁠⁵

Research results: design as political strategy

Erik Engstrom and Jason Roberts, in The Politics of Ballot Design (Cambridge University Press, 2020), draw on data on state ballot laws from 1888 to the present and argue that strategic politicians use ballot features such as the straight-ticket option to influence voting and elections, with significant effects on representation.⁠⁶

Other work cautions against overstating what the reform did. Daniel Moskowitz and Jon Rogowski, in the British Journal of Political Science (2023), revisit the claim that the Australian ballot drove candidate-centered politics. Using legislator behavior, distributive politics, and party influence, they report little evidence that the reform directly changed legislator behavior and suggest it may have been a necessary condition for later candidate-centered politics rather than a transformation on its own.⁠⁷

Research results: layout can change outcomes

The clearest modern evidence that layout matters comes from the 2000 presidential election. Jonathan Wand and five coauthors, writing in the American Political Science Review (2001), argue that the “butterfly ballot” used in Palm Beach County, Florida, led more than 2,000 voters who intended to vote Democratic to vote by mistake for the Reform Party candidate, Pat Buchanan, a number larger than the certified statewide margin. Among 3,053 counties where Buchanan was on the ballot, they report Palm Beach had the largest anomalous excess of his votes, and his share of election-day ballots there was about four times his absentee share.⁠⁸ The finding is an inference from statistical patterns, not a count of individual intentions. The broader history of voting equipment is in election technology.

Competing interpretations

Reform as anti-corruption. Wigmore and the reformers framed the official ballot as a remedy for bribery, intimidation, and caucus control.⁠¹ On this view the later problems are failures of implementation, not of the idea.

Reform as party consolidation. A second reading, supported by the history the Court recounts in Timmons and by the ballot access record in Williams, is that the official ballot let established parties write rules that disadvantaged new ones.⁠^2,4 The Court upheld Minnesota’s ban in Timmons but struck Ohio’s laws in Williams, which shows that the line between legitimate regulation and entrenchment is drawn case by case.

Stability as a legitimate goal. The Timmons majority treated political stability and a “healthy two-party system” as interests a state may pursue.⁠² Defenders argue that crowded ballots confuse voters and that thresholds keep frivolous candidacies off. Critics answer that “stability” can protect incumbents. Both arguments appear in the opinions and both rest on empirical claims about voter behavior that are themselves contested.

How much design matters. Engstrom and Roberts argue design has large strategic effects;⁠⁶ Moskowitz and Rogowski find smaller direct effects of the reform on legislators.⁠⁷ These are different questions, one about voters and one about officeholders, and both can hold.

Interpretation, this site’s reading. The Australian ballot shows that neutrality is not a default. A ballot cannot be printed without deciding inclusion, order, labels, and shortcuts. The honest response is not to pretend those decisions are absent but to publish them, justify them, and test them.

Application to this framework

These are design implications, not claims about the framework’s performance.

1. The ballot is a governance document. In My Vote Forward, the closest equivalent is the list of options in a priorities ranking or a proposal. Who chooses that list is a governance question, covered in governance and in who chooses the ten.
2. Inclusion rules must be public. Ballot access law exists because exclusion from the ballot is exclusion from the choice.⁠⁴ Any rule for adding or removing an option should be published with its reason.
3. Order and layout should be tested. The butterfly ballot shows that layout can move votes.⁠⁸ Option order in a ranking can be randomized per participant, and question wording should follow question design.
4. No shortcuts that encode party. A straight-ticket option is a design choice with known strategic stakes.⁠^5,6 The framework does not organize priorities by party and offers no party-wide selection.
5. Write-in paths. Burdick shows that whether people can add their own choice is a policy decision.⁠³ An “other” or proposal channel lets participants raise issues the list missed; see agenda capture.

Hypothetical example. A city’s priority survey lists ten issues in alphabetical order. “Affordable housing” is always first and “Water quality” always last. If first-listed options get extra picks from hurried respondents, alphabetical order quietly favors some topics. Randomizing order for each participant, and publishing that it was randomized, removes that advantage.

Unresolved questions

• What ballot access threshold separates reasonable regulation from protection of incumbents? Courts apply a balancing test, and results vary by case.⁠^3,4
• How large are layout effects in ordinary elections, as opposed to rare failures like the butterfly ballot.⁠⁸?
• Should ballot design be set by legislatures, which have partisan stakes, or by independent bodies? Engstrom and Roberts’s evidence on strategic design bears on this but does not settle it⁠⁶
• For non-binding priority tools, who should hold the equivalent of ballot access power, and how should it be checked?