South Shore News | Ballot Question Series, Week 5 of 9
This is the fifth in a nine part series looking at each of the ballot questions coming to Massachusetts voters in November.
Please take the poll in the middle of the article.
On or before Sept. 15, the State Auditor is required to report whether Massachusetts collected more money last year than a 1986 law allows it to keep.
The answer will almost certainly be no, and it will not be close. This has been the answer in every year but two since 1986. Fiscal 2026 came in $43.11 billion against a ceiling of $48.27 billion, $5.15 billion under. The new formula would change not just the ceiling but what to include in the revenue as well.
Question 5 on the Nov. 3 ballot would rewrite the formula that produces that number. Quick math on the year just closed with the proposed formula and the answer flips. Under Question 5, the ceiling for fiscal 2026 would have been about $46.48 billion, and the state would have collected about $46.49 billion against it — over the line by roughly $11.7 million, or one part in 4,000. The same collections that came in $5.15 billion under the current cap would have triggered a rebate.
The rebate would have been small: about $11.7 million returned to taxpayers, on the order of a dollar or two for a typical filer, against the roughly $3 billion returned in 2022. And it takes both halves of Question 5 to get there. Counting the millionaire’s surtax alone, the state would still have finished $1.8 billion under. Resetting the base to last year’s actual collections alone, it would have finished $260 million under. Only together do they land the state a hair over the line.
That is the whole of the question: not whether to cut taxes, but whether a ceiling that is currently missed by billions should be reset each year to something the state might actually hit.
What Chapter 62F does now
The law voters passed in 1986 caps how much tax revenue the state may keep. Excess must go back to taxpayers.
The mechanism is everything. This year’s ceiling is last year’s ceiling multiplied by the three-year average growth in Massachusetts wages and salaries. It compounds forward from a base set in fiscal 1986. And it ratchets — if the formula produces a lower number than last year’s ceiling, last year’s number carries forward. The cap can rise. It cannot fall.
Each September two officials run the comparison. The Commissioner of Revenue prepares it “on or before September first” and hands it to the State Auditor, who determines the result “annually on or before the third Tuesday of September.” The Auditor’s finding is conclusive.
In forty years that has produced a rebate twice.
In 1987 the excess was $29,221,675, claimed as a line on that year’s tax return rather than paid as a check. In 2022 it was $2,941,499,731 — FY2022 collections of $41.81 billion against an allowable $38.87 billion. About three million taxpayers received a credit worth 14.0312 percent of their 2021 state income tax bill, beginning Nov. 1, 2022.
The case that the current law is broken
The strongest argument for Question 5 is not in the campaign’s official statement, and it is not seriously contested.
Forty years of compounding off a 1986 base, with a floor that prevents the ceiling from ever falling, has produced a limit with little relationship to what the state collects. The FY2025 gap was $4.7 billion. FY2026 collections came in at $46.49 billion, $11.7 million above the state’s own benchmark, and the gap on the current formula’s basis appears wider still. A limit missed by that margin is not restraining anything.
Tufts researchers, whose analysis is the most critical published assessment of the measure, concede the point: tying the cap to prior-year collections “would eliminate drift over time, ensuring that the 62F cap stays pinned to actual revenue,” and “clearly provides a more vigorous check on revenue collection, regularly returning money to taxpayers when revenues are strong.”
Whatever else Question 5 does, it would make 62F a cap that occasionally binds — which is what voters were told they were getting in 1986.
The disagreement is about what happens next.
What Question 5 would change
The petition runs three sections. Section 1 rewrites two definitions in Chapter 62F.
The first is the engine. Where current law sets the year’s maximum at the prior year’s maximum times the growth factor, Question 5 sets it at the prior year’s actual collections times the growth factor. The cap stops compounding off itself and resets annually to what the state actually took in. Because collections have run well below the drifting ceiling, the ceiling drops sharply — MTF calculates that FY2025’s cap would have been $41.47 billion instead of $46.38 billion — and lands close enough to real revenue that ordinary growth pushes through. The cap can also now fall.
The second folds in the 4 percent surtax on income above $1 million, writing surtax revenue into the definition of “State Tax Revenues.” Section 2 then strikes the subparagraph of state law that currently exempts that revenue from the calculation. Section 3 is a severability clause; Question 3 carries a similar one.
Everything takes effect for fiscal years beginning after June 30, 2027 — fiscal 2028. Nothing changes on either side of a yes vote until then.
How much more often
Two organizations have modeled it, and they broadly agree on direction.
The Tufts Center for State Policy Analysis, in a March report titled “A Hair Trigger for Automatic Tax Refunds,” ran thousands of simulations and found the measure would trigger “3–5 times as many refunds and returning 5–15 times as much money to taxpayers,” equivalent to cutting the income tax about 0.4 percentage points. Backcast over 39 years, 62F would have fired 22 times instead of twice, returning more than $30 billion.
cSPA has since revised those figures downward in its own 2026 voters’ guide, to 2–4 times as many refunds, 2–7 times as much money, and 0.15 to 0.2 percentage points — with taxpayers receiving roughly $200 to $250 a year averaged across refund and non-refund years.
The Massachusetts Taxpayers Foundation ran a ten-year window: the proposed formula would have triggered four times, refunding $7.9 billion, or about $10.1 billion counting the surtax change. Its per-filer figures for those years were $269, $935, $503 and $1,095 — an average of about $701 in a triggering year.
Both organizations independently identify the same design flaw, and it is the sharpest technical objection available.
Rebates are subtracted from “net” revenue. A refund paid in one year lowers the following year’s measured collections — which, under Question 5, lowers the following year’s ceiling. cSPA calls this “a risky feedback loop in the tax system, encouraging refunds every other year and triggering 62F in periods of weak revenue growth.” A recession sets a low base; the recovery clears the low cap and pays out. The money leaves precisely when the state is trying to restore what it cut.
The Legislature’s own majority report, printed in the voter guide, describes the same mechanism: “a bad economic year would likely result in lower revenues, and a resulting economic bounce back the following year would more likely trigger a tax refund.”
MTF adds the reserve-fund arithmetic: in the four years the proposed cap would have produced $10.1 billion in refunds, the state deposited $4.6 billion into its Stabilization Fund. Refunds would have roughly doubled the deposits. MTF concludes only that the current $8.3 billion balance “would be significantly lower.”
One caveat belongs with all of these estimates. If the Legislature’s own pending rewrite of 62F survives conference — see below — both tests would have to be met before any rebate is owed. Andrew London, who has analyzed the interaction, says reading the two together “will likely reduce the frequency of rebates” and adds “ambiguity to the policy impact.” Nobody has modeled by how much.
And there is a live question about what counts as revenue at all. In October 2022 the Massachusetts Budget and Policy Center argued the certified $2.94 billion excess overstated FY2022 collections by roughly $1.4 billion, because the calculation counted $2.26 billion in pass-through entity excise payments while ignoring the credits those filers were entitled to claim later. Under current law that kind of timing artifact affects one year’s rebate. Under Question 5 it would also set the following year’s ceiling.
Who would get the money — and this is where the measure is misunderstood
Most voters thinking about Question 5 are thinking about the 2022 check. That check was 14 percent of what you owed, so it tracked income closely. South Shore News’s estimate for a household at Brockton’s median income is about $436; for one at Hingham’s median, about $1,164 — roughly 2.7 times as much.
Question 5 does not change how the money is divided. The Legislature already did, in the October 2023 tax package: Chapter 50, Section 26 replaced proportional distribution with an equal share for every taxpayer who filed in both the prior and current year, with a married couple filing jointly counted as two taxpayers. Where the credit exceeds liability, the balance is paid out — so a filer with no tax liability, who received nothing in 2022, now receives a full share.
Both analysts who have modeled Question 5 flag the change, and both read it the same way. Tufts calls the new method “progressive.” MTF, which opposes the measure, says the benefit would be “equal for all filers, but proportionately larger for lower-income residents.”
Put those together and the local picture inverts. On MTF’s average of about $701 per filer in a triggering year, a married couple at Brockton’s median would collect roughly $1,400 — about three times what our estimate says they got in 2022 — and, on either set of Tufts figures, considerably more often. A Hingham couple would collect about the same as before, more often.
So on the rebate side, Question 5 is more favorable to lower-income households than the rebate voters actually received. That is the opposite of what most coverage implies, and it is the case for the question that the Yes campaign has not made.
One thing to hold onto: equal-per-filer is itself only a statute, passed in 2023. The Legislature could change it again — including back to proportional — without returning to voters.
It is the aid side where the asymmetry runs the other way, and there it does not narrow at all.
What it means for towns
The state has made this argument for us. The Statement of Fiscal Consequences, written by the Executive Office for Administration and Finance and printed in the voter guide, does not use the boilerplate sentence A&F applied to Questions 3 and 4:
“The proposed law would put stricter limits on allowable annual revenue growth available for budgeting. As a result, the proposal would reduce the amount of money available to support the state budget, which includes local aid for schools and municipal budgets. The proposal would also reduce the amounts available to build the state’s Stabilization, or ‘rainy day,’ Fund — a reserve of money set aside to help Massachusetts manage economic downturns or emergencies without raising taxes or making major spending cuts. Depending on future conditions, the proposal could change the frequency of statutorily-required refunds to taxpayers.”
Adam Chapdelaine, executive director of the Massachusetts Municipal Association, put the mechanism plainly in May, arguing against a different tax question: “When municipalities lose state aid, there are only a few options available: significantly reduce services, delay investments, or shift more onto local property taxpayers.” MMA’s December report, “Navigating the Storm,” found that unrestricted general government aid — restored to its 2008 nominal level only by this year — has eroded roughly 49 percent in real terms since 2010.
The scale of exposure is wildly uneven. Brockton receives $288.0 million in Chapter 70 education aid and $26.7 million in unrestricted aid this year — state money on a scale that dominates the city’s budget. Hingham receives $10.0 million and $2.1 million; Duxbury $7.8 million and $1.2 million.
Bridgewater-Raynham is the clearest local illustration of what a tighter cap would do, because it has already been through the compressed version.
The district’s Chapter 70 aid rose 8.5 percent in fiscal 2026 and 2.6 percent in fiscal 2027 — a growth rate cut by more than two-thirds in a single year. Its FY2027 operating budget of $103.8 million was up 4.92 percent, so the gap went to the towns: Bridgewater was asked for 9.87 percent more against a town revenue forecast of 2.5 to 3 percent, Raynham for 7.8 percent against 1.52 to 1.82 percent. Both towns had already rejected overrides the previous June — $8 million in Bridgewater, $3.9 million in Raynham. The district’s FY2024 in-district per-pupil spending was $16,125 against a state average of $21,256.
That is the transmission line: state aid growth slows, the district’s costs don’t, the assessment lands on towns whose levy Proposition 2½ caps at 2.5 percent growth plus “new growth,” and the choice becomes service cuts or an override.
The recent override record here is not encouraging. Whitman rejected a $2 million operating override 1,678 to 677 in May 2025. Hanson rejected $3 million, 1,172 to 741, the same day. Easton rejected $7.3 million. Whitman-Hanson shows the same pattern, with Chapter 70 up 2.08 percent for FY2027 against assessment increases of roughly 4.9 percent for both towns.
Abington is the exception, and instructive: a seven-part menu override in May 2026 let voters fund most departments while rejecting only one. Six of seven passed. Silver Lake cut $586,000 and still raised Plympton’s assessment 8 percent. Hanover and Duxbury each passed overrides in their second year of asking, with each smaller than the first ask.
None of this was caused by Question 5, which does not take effect until fiscal 2028. That is the point. A tighter state cap would arrive in districts that have already spent their slack.
What the bond market says
The opposition’s official argument warns that Question 5 threatens the state’s credit rating. The rating agencies have been asked, and they have not gone that far.
Massachusetts carries Aa1 from Moody’s and AA-plus from S&P and Fitch. S&P analyst Ladunni Okolo told The Bond Buyer in August that the firm already treats Massachusetts as having “some revenue limitation” because of the existing cap, and that the rating reflects it. Her concern about the ballot question was timing rather than creditworthiness: if the state hits a recession, revenue “could plunge, then return to normal the following year,” and the measure “would prevent the commonwealth from collecting that revenue” — the same feedback loop cSPA and the Legislature’s majority report describe.
Deputy Treasurer Sue Perez said rating agencies “feel like we have enough tools to be able to react to different things,” and that it is “too early to say” whether Question 5 would affect the state’s borrowing. She also called the policy “confusing.”
The surtax provision, and a disagreement worth reporting
The surtax change is the most misunderstood part of the question, and the two analysts disagree about it.
cSPA says the effect is nearly neutral. Counting surtax revenue raises measured collections — but because the new cap is built from prior-year collections, it raises the cap by almost the same amount a year later. “The impact here would be quite small,” the report concludes. On that reading, the frequency of rebates is driven entirely by the base-year change.
MTF’s two scenarios differ by about $2.2 billion over the decade, which it attributes to the surtax inclusion.
The legal question underneath is untested rather than unresolved. Massachusetts voters dedicated the surtax to education and transportation by constitutional amendment in 2022, and the amendment’s operative language is that the revenue “shall be expended, subject to appropriation, only for these purposes.” Whether that constrains a statutory rebate has not been litigated — nobody has sued over Question 5, and the Attorney General has not opined. Proponents can argue a rebate reduces revenue rather than spending it; opponents can argue the effect is identical. The Legislature’s majority report notes that the petition “has not been evaluated by her or the courts for its constitutionality at this time,” which is true of every initiative petition.
