How did some of Massachusetts’s buzziest ballot campaigns get tripped up in court?

LAST DECEMBER, Secretary of State Bill Galvin said his office had never seen so many ballot questions submitted at once. A total of 11 initiative petitions ultimately cleared the signature-gathering threshold, joining a gun law referendum already on track for the 2026 contest. Groups that had spent years frustrated by Beacon Hill ignoring calls to pass legislation opted to take their case directly to voters.

Six of them landed before the Supreme Judicial Court. Three of them hit the wall.

As a result of legal challenges and an advisory opinion requested by lawmakers, the two highest-profile ballot campaigns and a third question, which would have reworked the payment system in the Legislature, are off the table.

Rent control? Booted from the ballot. A major cut in the income tax? Also gone. A third measure, aimed at reforming how legislative leaders hand out stipends, was ruled out of bounds by the attorney general before advocates could even finish gathering signatures.

The initiatives all failed in some way to comply with rules for ballot questions set out in the Massachusetts Constitution.

Their court-ordered demise stands as a cautionary tale for enthusiastic advocates looking to find a route around recalcitrant lawmakers. They learned the hard way that the ballot process may be quicker, but it can also prove to be a more perilous method for lawmaking than working through normal legislative channels.

First things first: How do questions get to the ballot?

Article 48 of the Massachusetts Constitution, adopted in 1918, established the citizen initiative petition process — the mechanism by which residents can propose laws and constitutional amendments, and put them directly before voters.

It includes a set of guardrails, with certain subjects explicitly off-limits for a ballot question, such as measures that relate to religion, the judiciary, or specific municipalities. Ballot questions also can’t include provisions that are not closely related to each other.

That makes the rules for ballot measures fundamentally different from those governing the normal legislative process. While lawmakers can jam multiple unrelated policies into one omnibus bill, and can pass legislation that implicates courts or religion or particular cities and towns, ballot measures must find a way to pose a straightforward question to voters without tripping over any of the excluded matters.

As the SJC made clear in its decision on the income tax petition, Article 48 “does not create unconditional access to the ballot.” The initiative process, it said, was designed “in a carefully prescribed manner and with certain precisely defined safeguards” to “prevent hasty action” and make sure voters know what they are voting for or against.

The attorney general’s office is tasked with reviewing and certifying petitions as eligible to move forward.

But certification does not mean the measure marches straight to voters, even if the campaign then collects enough signatures. Opponents can sue, which sends the matter to the SJC to determine whether or not the petition should have been certified. And the AG’s review focuses only on a narrow set of ballot eligibility criteria, not every possible angle by which a proposal’s constitutionality can be judged.

There is also a rarely used parallel track available, in which the Legislature can ask the justices for a non-binding advisory opinion on whether a measure is constitutional.

Why are we facing so many ballot questions?

Ballot measures are booming. Even with the tossed petitions, Massachusetts is still on track to tie its record number of nine ballot questions last set in 1994.

Most years see three to five measures come before voters. But a few things have ramped up the ballot enthusiasm in recent years.

Boston Mayor Michelle Wu this week said that “frustration about inaction” at the State House “has reached a boiling point.”

Major policies have recently broken past legislative inertia to win at the ballot box. A surtax on high incomes passed in 2022, and has funneled billions in new revenue toward education and transportation. The next cycle, the high-stakes 10th grade MCAS graduation requirement was wiped out by ballot vote.

The threat of a ballot question also recently proved a powerful negotiating tool, fueling a settlement between the attorney general and Uber and Lyft over whether their workers are employees or independent contractors. That ended a pricey ballot campaign that was poised to decide the matter. Under a different 2024 ballot measure, those ride-share drivers got the right to unionize.

This is also the first ballot cycle since a notorious pile-up of major legislative business kept lawmakers working deep into the summer of 2024, increasing public frustration with the pace and transparency of Beacon Hill’s work.

“It can’t go worse” than last term, House Ways and Means chair Aaron Michlewitz quipped in mid-June.

The 2024 cycle also included a much-debated ballot measure, passed with 72 percent of the vote, allowing the state auditor to audit the Legislature. Legislative foot-dragging and ugly court fights over its implementation have followed, which has also fueled Auditor Diana DiZoglio’s push in the current cycle for a ballot measure that would subject legislators and the governor to public records law.

Advocates see ballot measures as a vital way around a Legislature resistant to act on issues that enjoy strong popular support. But legislators and a fair share of political watchers say ballot questions are a bad way to make law, viewing them as blunt instruments that cut out the deliberative process of lawmaking and tempt special interests to jam poorly considered ballot measures into law.

What happened to the income tax cut?

A business-backed coalition proposed gradually stepping the state’s income tax rate down from 5 percent to 4 percent over three years, pitching it as relief from the state’s high cost of living and a check on outmigration.

The SJC tossed it June 18. The problem was three words in the attorney general’s summary: capital gains income.

Under Article 48, the attorney general’s office is required to produce a “fair, concise” description of each petition, which appears on the forms voters sign when petitioning to put a question on the ballot. Justice Serge Georges, Jr., writing for a unanimous court, said the summary produced by Attorney General Andrea Campbell’s team “misstates the petition’s impact” by incorrectly stating the rate cut would not affect taxes on long-term capital gains.

Long-term capital gains tax rates are pegged by law to the income tax rate, so the measure would, in fact, also have reduced capital gains taxes and significantly benefited high-income earners.

“The summary’s contrary statement is not a minor imprecision. It is significantly misleading and likely to influence voters,” Georges wrote. And because the flawed summary was already on tens of thousands of signed petition forms, the court said allowing a corrected summary at that stage would call into question the validity of all the signatures already gathered.

How did incorrect language make it into the summary?

There is a period of time in which the AG’s office communicates with the registered proponents and opponents of a measure and gives them a chance to offer feedback on the summary.

According to emails between proponents and the attorney general’s office in August 2025, the summary suggested by tax-cut backers made no mention of any effect on capital gains. But the final version of the AG’s summary, reviewed by proponents, did.

It said the petition would “lower the tax rates on (1) personal taxable income consisting of interest and dividends, and (2) personal taxable income other than interest, dividends or capital gain[s] income, such as wages and salaries.”

The proponents’ attorney told Campbell’s office that the summary “works well for proponents, thanks so much.”

Opponents never offered input on the summary.

Then, in early February — more than two months after the deadline for gathering signatures using the summary had passed — opponents filed a challenge with the SJC, saying the attorney general’s summary was inaccurate. The coalition opposing the measure said they had not noticed the error during the initial review of the summary.

What does that mean for income tax cut proponents?

It means they’re out of this ballot cycle, with a multi-year reset ahead if they want to try again.

The lesson for future campaigns is to scrutinize the AG’s summary as carefully as the petition language itself and act early if something looks wrong, rather than waiting to see if the opposition or the court catches it.

What happened to the rent control measure?

After years of urging the Legislature to rein in rising housing costs by stabilizing rents, tenant advocates seemed like they finally found momentum. A ballot question pushing to restore rent control for the first time since the practice was banned by ballot in 1994 was poised to go before voters this fall. The prospect of the sweeping measure, which would have imposed statewide limits on rents, led some real estate groups and Beacon Hill leaders to begin discussion of compromise legislation that might lead proponents to drop the ballot campaign.

The SJC killed the measure — and that momentum — on Tuesday morning, again based on just a few words.

The petition would have mandated all 351 cities and towns impose a rent control limit of inflation or 5 percent, whichever is lower. Drafters included a provision, however, excluding from rent control any housing operated solely for educational, religious, or nonprofit purposes.

Justice Frank Gaziano wrote that because of this, the petition “relates to religion, religious practices or religious institutions” — an excluded matter under Article 48 — is therefore not eligible to appear on the ballot.

Campbell defended the petition’s certification, arguing that the court should consider whether regulating religious institutions was the “main purpose” of the ballot measure, and saying the petition did not discriminate in favor of religious institutions because it also exempted secular facilities.

The court disagreed, seeming to lay out a strict new standard under which any mention of religion in a petition would be problematic.

The posture of the court clarifies a previously fuzzy area of ballot law — how much religion is too much religion in an initiative. It also sends a signal about the SJC’s interest in playing referee in a ballooning ballot environment. Simpler rules make it easier for the AG’s office to give a measure an up-or-down ruling during certification, and reduce the likelihood that the high court will have to spend its time on how controversially religious a ballot measure may be.

“When we certified it, we thought we were obviously on the right legal footing,” Campbell said of the rent control decision during an appearance on GBH’s Boston Public Radio Tuesday.

What does the rent control ruling mean for the negotiations that were underway?

It ends them, for now. The opposition campaign said the decision “puts the issue to rest,” the real estate group NAIOP said it was no longer interested in negotiating, and the other two major opponent groups expressed continued hostility toward the idea of any form of rent control.

Gilbert Winn, CEO of WinnCompanies, who helped lead the compromise negotiations, said the ruling was “ultimately more of a postponement” and that the two sides had been “fairly close on a workable plan.” But the campaign’s leverage was always the ballot. Without it, there is no particular reason for the real estate industry to negotiate, and no particular pressure on the Legislature to act during the last legs of their busy season.

The Senate has been warmer to the idea of rent stabilization than the House, historically, but without clear buy-in from both chambers there is little incentive for an end-of-session rush when other housing measures are still in limbo. House Speaker Ron Mariano’s office said Tuesday that Mariano “has long argued that rent control, and specifically the proposal that was slated to go before voters in the fall, will stifle housing production.”

What happened to the legislative stipend measure?

The measure, backed by a coalition of government accountability advocates, would have restructured the system of added pay that some lawmakers receive for serving as committee chairs or in other roles assigned by legislative leaders. The petition would have tied those stipends to performance goals like holding public hearings on bills and approving committee reports by majority vote at a public meeting.

Supporters framed this as a check on the use of leadership power to reward allies with extra pay, while opponents saw it as an intrusion into the Legislature’s internal operations.

Rather than wait for a lawsuit, the Senate took the unusual step of requesting an advisory opinion from the SJC. The justices determined in April that the stipend measure was not appropriate for the ballot because it proposed internal legislative rule changes — another excluded matter — rather than a law and should not have been certified.

The advisory opinion was technically non-binding, but Campbell’s office rescinded its earlier certification, sending a letter to Galvin a letter to Galvin on May 7 saying the measure “may proceed no further in the Article 48 process,” and Galvin’s office stopped providing signature papers to the campaign.

Proponents had gathered tens of thousands of signatures and had more to collect before the July deadline. They never got the chance.

They have vowed to return in a future election cycle with a more streamlined ballot question — one that simply gets rid of the stipend system rather than setting up rules for legislative operations — that they think won’t run afoul of the Article 48 guardrails.

“The next question will not be how to reform the stipend system, it will be to end it,” said John Lippitt, chair of the Legislative Effectiveness and Accountability Partnership. “We would have preferred reform. The Legislature has forced us to move to eliminate it.”

What does it mean that three campaigns are out and the Legislature passed none of the 11 petitions?

The Legislature, which had until the first Wednesday in May to pass any of the 11 initiative petitions outright, recommended no action on any of the measures. This is the same approach it took two years ago.

Beacon Hill was facing its largest ballot question field in three decades, driven substantially by advocates who had tried and failed to move their issues through the normal legislative process. The Legislature’s response was to wait and see.

On rent control, income taxes, and stipend reform, the answer from the high court was a gift to lawmakers: time, and relief from the pressure to act.

Despite the SJC rulings that nixed them from this year’s ballots, the rent control proposal and income tax cut both seemed broadly popular, according to early polling, so proponents could try again if they find the energy and funds for another round.

Because the two high-profile court decisions turned on procedural failures and technicalities, the proponents now have some clear guidance on how to tweak the language of their proposal.

The SJC rulings offered a harsh lesson not to treat ballot measures like legislation.

They are laws like any other, once passed, but the Article 48 process is designed to keep the impacts of a “yes” or “no” vote as simple as possible, while heading off the risks of an agitated populace deciding to wade into matters that the framers of the state Constitution thought would be best left to deliberative elected officials. Each court decision makes the scope of what to include and what to omit a little clearer.

Campbell’s office was at the center of all of these opinions. How is the attorney general responding?

It didn’t take long after the income tax ruling for conspiracy theories to begin flying that the attorney general’s office — which told the SJC that if the summary was unfair the measure should be tossed — deliberately tanked the petition that was irritating Beacon Hill.

The AG has taken a battering from op-ed pages and angry campaigns, noting that her job was to describe the measures fairly and certify them accurately. In three cases, the SJC said her team failed at that task.

Campbell is defending her team publicly, facing prodding during her monthly “Ask the AG” segment on GBH News radio.

“No one should expect us to get a perfect result,” she said. “No organization could. But that being said, we take our jobs seriously and we hold ourselves accountable.”

A record number of 47 ballot measures came through her office this year, she said. Six made it to the court. Batting .500 is not a bad average, in her telling.

“We had six challenges, we got three wrong. I think that’s a great record,” Campbell told reporters Tuesday. “That just tells me we have more to do to be better. Any institution, whether it’s media outlets or any industry, if they can get 100 percent right every time — wow. That doesn’t happen. We own these mistakes.”