Who Paid the Most to Write State Law at the Ballot Box, and Whether It Worked
The most expensive ballot measure in American history cost $418 million, and it lost with 82 percent of voters saying No.
In 2022, online sportsbooks and California tribes poured that much into a single proposition to legalize online sports betting. The sportsbooks spent about $169 million to pass it. The tribes spent about $249 million to kill it, and Prop 27 failed in every county in the state.
I ranked the most expensive state ballot measures ever by the total money raised on both sides, then looked at who paid, what they wanted and whether the money worked.
The pattern is hard to miss. Money is very good at stopping a ballot measure and surprisingly bad at passing one.
The Most Expensive Ballot Measures in American History
Companies, unions, tribes and billionaires have spent hundreds of millions of dollars trying to write state law directly at the ballot box. The biggest spenders usually get what they want, but mostly when what they want is a No.
Total contributions to committees supporting and opposing each measure, in millions of dollars, not adjusted for inflation. Florida measures needed 60% to pass. Packages of linked measures (such as California's four 2008 tribal compacts) are not ranked as one measure, and the 2012 California tax and union measures are left out because their committees overlapped. Florida's 2024 abortion amendment, about $135 million, is left out because a final split between the two sides was not published, and so is California's 2005 Prop 79, whose committees overlapped with a rival drug-discount measure. Sources: Ballotpedia, using California Secretary of State filings where Ballotpedia has no total (Prop 56).
The green part of each bar is money raised to pass the measure. The red part is money raised to defeat it. Of the 13 measures on this list, 11 were in California, and only two passed.
Does the money win?
On the surface, yes. In 10 of these 13 campaigns, the side that raised more got the result it wanted. But that number hides the real story, which is which side the money was on.
The one big-money Yes campaign that won was Prop 22, the gig-driver measure. Every other time the heavier spender was pushing for a Yes, it lost, including the tribes' own measure, Prop 26, and Florida's marijuana amendment, which won a majority but not the 60% Florida requires.
This matches what political scientists have found for decades. Elisabeth Gerber's 1999 research, drawn from 168 ballot campaigns in eight states, found that in a set of 57 measures from 1988 to 1992, initiatives backed mainly by economic interests passed just 31% of the time, against 50% for those backed by citizen groups. Those economic interests put about two-thirds of their resources into defeating measures rather than passing them. A more recent analysis by New America of initiatives from 2013 to 2024 found that when the No side outspent the Yes side, about 71% failed.
Part of the reason is that No is the easier sell. A No campaign doesn't have to convince anyone the measure is good, just raise enough doubt that voters stick with the way things are. Only about 42% of all statewide citizen initiatives from 1904 to 2021 passed, according to the Initiative and Referendum Institute. In California it is about 36% since 1910, by Ballotpedia's count.
The ballot, case by case
Allow licensed operators, partnered with tribes, to offer online and mobile sports betting, taxed at 10%, with most of the money for homelessness programs.
The sportsbooks' pitch was that online betting would raise "hundreds of millions" a year for homelessness. The state's own analysts said that was possible, but likely not more than $500 million a year. FanDuel put in $35 million and DraftKings $34.1 million. The tribes, who wanted to keep sports betting in their own casinos, answered with an even bigger war chest, led by the San Manuel Band of Mission Indians at $103.1 million. Fact-checkers found the Yes campaign's ads leaned on tribal support from just three of California's roughly 110 tribes. Prop 27 lost in every county. So did the tribes' own measure, Prop 26. Together the two campaigns raised about $463 million.
Keep ride-hailing and delivery drivers as independent contractors, with a minimum earnings guarantee and some benefits.
Prop 22 is the exception that proves how much it takes. Uber, Lyft, DoorDash, Instacart and Postmates put in more than $200 million, outspending labour unions by more than ten to one. The companies also campaigned inside their own apps and on delivery bags. It passed with 58.6%, and in July 2024 the California Supreme Court unanimously upheld it.
2018: limit clinic profits. 2020: require a doctor on site. 2022: require a doctor, nurse practitioner or physician assistant on site. Both later measures also required infection reporting and state approval to close.
The kidney dialysis fight is the clearest example of an industry using the ballot as a shield. A health-care workers' union, SEIU-UHW West, put a dialysis measure on the ballot three elections in a row. DaVita and Fresenius, which run most of California's dialysis clinics, spent more than $300 million combined over the three campaigns to defeat them, against about $36 million from the union. Each one lost by a wider margin than the last.
Bar state agencies from paying more for a drug than the U.S. Department of Veterans Affairs pays.
Drugmakers including Merck, Pfizer and Johnson & Johnson each gave more than $9 million against Prop 61, the most raised for or against any of California's 17 measures that year. Their ads featured veterans warning they could end up paying more for their medicine. Early polls had shown the measure ahead. It lost 53% to 47%.
Repeal or loosen the 1995 state law that limits local rent control.
The AIDS Healthcare Foundation funded three attempts to expand rent control, in 2018, 2020 and 2024. Landlords, apartment owners and realtors answered each time, raising about $280 million to the foundation's roughly $117 million across the three. All three lost with about 60% voting No.
Legalize adult possession of up to 3 ounces, and let existing medical marijuana companies sell it. Needed 60% to pass.
Florida's 2024 marijuana amendment is the most expensive measure on this list from outside California, and almost all of its money came from one company. Trulieve, a medical marijuana company, gave about $145 million. Its signature drive alone cost about $23 million, the most expensive on record. The amendment got about 56% of the vote, a clear majority, but Florida requires 60% for constitutional amendments.
Tax California oil production to fund a $4 billion alternative-energy program.
Before the 2020s, the record belonged to a 2006 fight over oil. The film producer Steve Bing put tens of millions of his own money into Prop 87, which would have taxed California oil to fund alternative energy. Oil companies led by Chevron and Aera Energy raised about $93 million against it and won.
Require the state to allow casino gaming on tribal land.
Prop 5 shows that big money can pass a measure and still lose. California tribes spent about $68 million to pass it, then a national record, and won 62% of the vote. The California Supreme Court struck it down the next year. The tribes and the governor went back to voters with a constitutional amendment, Proposition 1A, which passed in March 2000 and is the basis of tribal casinos in California today.
How the record kept breaking
Total raised by both sides for each measure that set a new record for a single measure. Packages of linked measures are not counted: California's five 1988 insurance measures together drew more than $80 million, and its four 2008 tribal compacts together slightly more than Prop 87.
Inflation doesn't change the top of the list. In 2024 dollars, Prop 27 comes to about $449 million, still far ahead. Prop 87 from 2006 moves up to third, at about $235 million.
Why it is almost always California
Two Supreme Court decisions explain why a single company can spend $100 million on one ballot question. In 1978, in First National Bank of Boston v. Bellotti, the Court ruled that states can't stop corporations from spending on ballot measures. It reasoned that the risk of corruption that justifies limits in candidate races "simply is not present in a popular vote on a public issue." In 1981, in Citizens Against Rent Control v. City of Berkeley, it struck down a $250 limit on contributions to ballot committees. Since then, contributions to ballot-measure campaigns have been effectively unlimited, apart from narrow exceptions such as the ban on foreign money.
California adds size. It has the biggest electorate in the country, some of the most expensive television markets, and a signature requirement based on a very large vote total. For the 2026 cycle, a new law needs 546,651 valid signatures to qualify and a constitutional amendment needs 874,641, gathered in 180 days. Paid signature drives have become expensive in their own right. Across the country, the average cost per required signature hit a record $14.87 in 2024, and Prop 27's signature drive alone cost $18.8 million.
And California votes on a lot of measures. Industries that win once often have to win again: dialysis three times, rent control three times. That repeat business is how one state ends up with 11 of the 13 most expensive ballot measures ever.
The most expensive ballot measure in American history raised $418 million and lost with 82% of the vote against it. On the list of the costliest measures ever, only two of 13 passed.
Money matters at the ballot box, but it works far better as a veto than as a pen. When the bigger spender wanted a No, it got one in 9 of 10 campaigns. When it wanted a Yes, it got one in 1 of 3.
Sources: Ballotpedia and Ballotpedia News, ballot measure campaign finance analyses (2016 to 2025); California Secretary of State, measure contribution totals and qualification requirements; CalMatters; Florida Phoenix and WUSF; Los Angeles Times; NPR and AP; KPBS fact-check (2022); Elisabeth Gerber, The Populist Paradox (1999); New America, Direct Democracy analysis (2013 to 2024); Initiative and Referendum Institute at USC; First National Bank of Boston v. Bellotti, 435 U.S. 765 (1978); Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290 (1981); Castellanos v. State of California (2024); Bureau of Labor Statistics, CPI-U.
Filed under: General Knowledge
