A 34.9% drop in property crime is the newest number in the fight over Massachusetts cannabis. It comes from one outlet's comparison of 18 towns, not from a government report or a peer-reviewed study. Voters decide on Nov. 3 whether to repeal retail sales, and the outlet behind the number says its analysis cannot show that the stores caused the decline.
Eighteen towns, two years, one gap
Anthony Martinelli of The Marijuana Herald compared FBI crime records for 2017 and 2025 in 18 municipalities that opened recreational stores during the market's first year. Their combined property crime rate, weighted by 2017 population, fell from about 1,798 to 1,171 offenses per 100,000 residents. That is a 34.9% decline.[1]
Statewide, the rate fell from about 1,428 to 1,019 per 100,000, or 28.6%, over the same years. The gap between the two declines is 6.3 percentage points. The article says property crime fell in all 18 towns.
The towns are Brockton, Brookline, Chicopee, Easthampton, Fall River, Gardner, Greenfield, Hudson, Lowell, Lynn, Marlborough, Nantucket, Newton, Northampton, Pittsfield, Salem, Somerset and Worcester.
Violent crime shows a much smaller difference. The 18 towns went from about 546 to 452 violent offenses per 100,000, down 17.2%, while the statewide rate fell from about 354 to 296, down 16.4%. Twelve of the towns had lower violent crime rates in 2025 than in 2017. Six had higher rates.
Reported offenses per 100,000 residents, 2017 vs 2025
Source: The Marijuana Herald analysis of FBI data
Why two snapshots cannot settle cause
The comparison uses two endpoints eight years apart. It does not track crime in each town before and after its store opened. It does not match host towns against similar towns that never allowed a store. And the statewide benchmark includes the 18 host communities themselves, so it is not a clean control group.
A lot changed between 2017 and 2025. The period covers the pandemic and any shifts in how local police record and report crime. Towns chose whether to allow stores, and companies chose where to open. Host towns may differ from nonhost towns in ways that have nothing to do with cannabis.
Fixing each town's population weight at 2017 levels solves one problem: it keeps population shifts from moving the combined rate. It does not answer the selection or timing questions. Municipal totals also cannot show whether an offense happened next to a dispensary or on the other side of town.
The Marijuana Herald does not claim the stores caused the decline. It says it excluded towns that lacked complete, comparable records.
Leicester is missing from the list
Leicester hosted one of the first two stores to begin sales in Massachusetts, on Nov. 20, 2018. It does not appear among the 18 towns.
The outlet has not published a town-by-town account of which early host communities were left out and why. Until the municipality-level inputs, opening dates and calculations are public, the 6.3-point gap stands as one outlet's descriptive finding that has not been independently reproduced.
Voters were already leaning no before the numbers arrived
A University of New Hampshire Survey Center poll released Sept. 24 found 71% of likely voters plan to vote no on Question 8, 24% plan to vote yes and 5% are unsure.[2] The survey ran Sept. 17 to 21, before the crime analysis was published on Sept. 23, so it cannot measure any reaction to it.
Photo: VapeExperts/AI
The poll surveyed 590 Massachusetts residents online with a margin of error of roughly 4 points, according to the Telegram & Gazette.[3]
One wording detail matters for anyone reading the results. The UNH release describes the proposal as prohibiting recreational use. The secretary of the commonwealth's ballot summary describes a repeal of retail sales that keeps limited adult possession legal.
A criminologist says legalization was oversold as crime control
Alex del Carmen, a criminologist and dean of Westfield State University's College of Professional Studies, told the Las Vegas Review-Journal on Sept. 26 that the research does not support treating legalization as a public safety tool. He was speaking about the broader literature, not the Massachusetts analysis.[4]
"Legalization is a regulatory and civil-liberties reform. It was oversold as a crime-control measure, and it has not delivered as one," del Carmen said.
He pointed to a different effect as the clearest one. "What legalization did accomplish, and this is the most robust finding in the literature, is a reduction in arrests," he said.
What a yes vote would change
Question 8 would repeal the laws that authorize, regulate and tax adult-use retail sales, according to the secretary of the commonwealth's ballot summary. Limited adult possession and the medical program would remain. Possessing more than one ounce and up to two would carry a $100 civil penalty and forfeiture.[5]
Existing adult-use businesses could apply for expedited medical licenses. The repeal would take effect Jan. 1, 2028, not on election day. A no vote keeps the current adult-use framework in place.
The market at stake is large. The Cannabis Control Commission announced on Sept. 9 that cumulative adult-use sales crossed $10 billion on Sept. 5, with $1.65 billion sold in calendar 2025 alone.[6]
The commission attributes about $336 million in sales, local and excise taxes to fiscal 2026. We covered the campaign and its backers in our earlier report on the repeal vote.
Commission chair Chris Harding tied the milestone to revenue. "Every adult-use sale generates tax revenue that goes back into vital programs and services at the state and local levels," he said. The statement addresses tax dollars, not crime.
Voters authorized adult use in 2016. The first two stores opened Nov. 20, 2018, and 33 retailers were operating by the end of the market's first year, according to the crime article. Ten years after that vote, the same voters will say whether the stores stay.
The ballot is Nov. 3. A yes vote closes the retail market on Jan. 1, 2028, and lets existing shops seek medical licenses. A no vote leaves the market standing. Whether the 18 towns got safer because of their stores, or simply alongside them, is a question the analysis does not claim to answer, and one the ballot will not settle either way.
