Boston Housing Tax Abatement: $22,500 a Unit to Build

Earlier this year a Boston developer asked the Wu administration a simple question. Would the city go back to the 13 percent affordable requirement it used under Marty Walsh, instead of the 20 percent it requires today. The answer was no. Banker & Tradesman reported that exchange on September 20, and the refusal was firm.

Now look at the four projects the same administration is proposing to subsidize. One Mystic Avenue, 22-24 Pratt Street, 83 Leo M. Birmingham Parkway and Allston Yards Building D. Together they carry 1,400 homes and 185 income-restricted units. That is 13.2 percent.

The number Boston refused to put in its policy is already sitting in its term sheets. That is the whole story of this proposal, and it is worth a lot more of your attention than the $31.5 million headline.

Today's Deals Updated daily Today's Deals: Home & Tools Fresh markdowns on tools, home upgrades, and décor — new deals drop every day. Shop Today's Deals → As an Amazon Associate, BMN earns from qualifying purchases.

The number the city refused in public is the number in the deals

On September 15 the city proposed roughly $31.5 million in temporary property tax relief to restart four apartment projects in Charlestown, Allston and Brighton. About $22,500 per home. Every one of these projects has already cleared the BPDA board. They have their approvals. They have their zoning. They have their design review and their community process behind them. None of them has a shovel in the ground.

Two of the four, One Mystic and Pratt Street, are separately negotiating to buy their way out of half of their inclusionary obligation in cash. WBUR reported that those two buyouts would generate over $20 million for deeper affordability elsewhere. That is why One Mystic shows 45 income-restricted homes out of 408 and Pratt Street shows 27 out of 318. Eleven percent and eight and a half percent, on sites where policy says 20.

I want to be fair to the city here. Cashing out part of an inclusionary requirement is legal, it is written into the policy, and $20 million buys more deeply affordable units elsewhere than 50 on-site units at 60 percent of area median income would. That is a defensible trade. But you cannot make that trade on two projects, refuse a 13 percent policy in public, and then tell me the 20 percent number is load-bearing. The deals say otherwise.

What is actually on the table

Here is the round one list, with the city’s own unit counts and the on-site share each project carries.

Three of the four sit in Allston and Brighton. The fourth is in Charlestown, a 16-story building across from Sullivan Station that the BPDA board approved back in September 2023. Nordblom’s Brighton project, on the old Audacy radio studios site, was approved in April 2025 at an estimated cost of about $115 million. That works out to roughly $345,000 a home before a dollar of profit, which tells you how thin $22,500 of tax relief really is. It is about six and a half percent of the build.

What a Boston tax abatement actually is

This is not a grant. Nobody is writing a check. The mechanism is a freeze on assessed value.

Today these four sites are taxed on what they are, which is a parking lot, an auto parts warehouse, a radio studio and a junkyard parcel. Under the proposal the owner keeps paying that same amount through roughly three years of construction, even as a 300-unit building goes up on top of it. Then the tax steps up one of two ways. A short track raises it a flat 2.5 percent a year for five years after completion. A long track phases in over ten years, starting at about 20 percent of the eventual obligation and reaching roughly 80 percent by year ten. Kairos Shen, the city’s chief of planning, put the city’s case plainly. “We are not reducing the tax rolls at any point.”

He is right on the narrow point. You cannot lose revenue you were never going to collect, because an unbuilt building generates no incremental tax. Sheila Dillon added that the city is “only giving projects the relief they need.” What both statements skip is that Boston’s residential rate rose to $12.40 per $1,000 for fiscal 2026, up from $11.58, and the commercial rate went to $26.96 from $25.96. The city is asking the rest of us to absorb rate increases in the same year it is freezing assessments on four specific parcels. That is a real choice, and it deserves to be discussed as one.

Forty-seven approved projects, and almost nothing started

Here is the fact that should end the zoning conversation for a while.

Boston has 47 projects of 100 units or more in its pipeline that could qualify for this kind of relief, roughly 10,000 units of approved housing. In 2025, the Globe counted two projects of that size starting construction. The Dorchester Reporter counted three built. The two outlets are measuring slightly different things and I am not going to pretend the discrepancy away. Either way you are looking at low single digits against a pipeline of 47.

Approvals were not the constraint. The approvals happened. Then they sat.

The approval machine itself is now slowing down too, which is the second-order effect people miss. The BPDA board green-lit 5.8 million square feet of development in 2025, worth about $4.8 billion. In 2024 it approved 11.6 million square feet. Cut in half in a year. Greg Vasil, who runs the Greater Boston Real Estate Board, described it to the Globe as a slow crawl out of COVID five years on. Developers stop filing when they can see that filing does not lead to building.

This is the same pattern I wrote about when tariffs started outrunning the state’s zoning reforms. Massachusetts spent four years fixing permission and almost no time on cost. Permission turned out to be the cheap half.

An approval is an option, not a commitment

The mental model that makes all of this click is an options model, and I think it is the honest way to read Boston’s pipeline.

When the BPDA approves a project, the developer does not receive a building. They receive the right to build, with no obligation to exercise it. An approval is an option. Like any option it only gets exercised when it is in the money, which here means when projected rents at stabilization clear the cost of construction, the cost of capital, the insurance, the taxes and the return the equity demands.

Boston spent five years writing options and counting them as housing supply. Then the strike price moved. Greater Boston new construction often runs $500,000 to $600,000 per unit to finance and build, financing costs repriced hard, and a large share of those 47 options went out of the money at once. Nobody exercises an out-of-the-money option no matter how many public meetings it survived.

What the abatement does, stripped of the politics, is move the strike price by about $22,500 a door. Whether that is enough to push a specific deal back into the money is a question only the developer’s model can answer, and it will be different for all four.

The inclusionary argument, taken seriously in both directions

Tamara Small, who runs CREDA, the commercial real estate development association in Massachusetts, is blunt about the cause. “Inclusionary is the thing we hear over and over from developers as being a significant barrier to market rate housing development in Boston,” she told Banker & Tradesman. “This is why housing development is happening in Revere, Everett, Chelsea, etc., but not in Boston.” She has floated cutting the requirement to 5 percent on a trial basis.

The policy gradient is real and you can look it up. Boston requires 17 percent income-restricted units plus a 3 percent voucher set-aside, at an average of 50 to 60 percent of area median income, and it triggers at seven units. That structure took effect in October 2024, replacing the old 13 percent standard. Chelsea, by contrast, is considering 10 percent of units at 80 percent AMI or 7 percent at 60 percent, and raising its trigger from 10 units to 50. Seven units versus fifty. That is the gap a developer is pricing when they choose Second Street in Everett over Pratt Street in Allston.

Now the other side, because I do not think the developers get to have this one for free. Chelsea already has an inclusionary policy, and the Chelsea Record reported that it has produced almost no affordable units in nine years. A weaker mandate is not a production machine. It is just a weaker mandate. And Jesse Kanson-Benanav made the sharpest criticism of the whole proposal from the housing advocacy side, that doing this project by project “results in really elevating the voices of those who are opposed to new housing.” He is right. A deal you negotiate four buildings at a time is a deal you have to renegotiate every time, in front of whoever shows up.

My position is narrower than either camp’s. I do not know that 13 percent gets Boston building again. I do know that the city is already accepting 13.2 percent on its own flagship round while telling the industry 20 is untouchable, and that the gap between the stated policy and the executed deal is where the real number lives.

What the sales record in these three neighborhoods actually says

This is where I can bring something no national outlet has, which is the MLS PIN closing record for the exact ZIP codes involved. I pulled every recorded closing in 02134, 02135 and 02129 for the twelve months ending August 31, 2026.

Read the bottom two rows together. Allston recorded zero condo sales of buildings constructed in 2015 or later across an entire year. Not a low number. Zero, out of 45. Allston’s entire for-sale condo market is old stock, and 22 of those 45 sales were in buildings that predate 1940. Yet Allston is getting 659 of the 1,400 units in round one. Whatever gets built at Pratt Street and Allston Yards will be the newest product anyone in that ZIP code has been able to transact on in years, and it will be rental.

Tools & DIY Laser Level & Stud Finder Hang shelves and art straight the first time. View on Amazon → As an Amazon Associate, BMN earns from qualifying purchases.

Charlestown is the tight one at 3.6 months of supply and 99.2 percent of original ask. Allston and Brighton are already soft on the for-sale side, at 7.7 and 6.8 months. That matters for how you should read the new supply, and it cuts against the intuition most people have.

New construction here does not rent cheaper. It rents higher

This is the part I most want investors to sit with, because the popular version of the supply argument gets the direction right and the timing badly wrong.

I pulled every closed lease recorded in the same three ZIP codes over the same twelve months, 1,255 of them, and split the two-bedroom medians by building vintage.

In Allston and Brighton the new stuff rents for $800 to $1,200 a month above the older stock. So when 1,400 apartments deliver in 2028 and 2029, they arrive at the top of the local rent distribution, not the bottom. They compete with each other and with the newest buildings already standing. The 1920s two-family on Gardner Street is not the comp.

Charlestown inverts, and I find that the more interesting read. There, the pre-2015 stock rents higher than the new construction, because Charlestown’s existing inventory includes a lot of renovated rowhouse product on the Hill that prices above a Sullivan Square elevator building. New supply in Charlestown lands under the top of the market, not over it. The sample is small and I am flagging that, but the direction held across every cut I ran.

If you own in 02134, 02135 or 02129

Four things I would actually do.

Do not assume 1,400 units is a threat to your value. They are rentals, at the top of the rent band, delivering three or four years out. If you own an older two-family in Allston or a 1920s condo in Brighton, your competitive set does not change. If you own a 2019-vintage rental unit inside a half mile of Pratt Street or Boston Landing, it does.

If you are selling in Allston or Brighton, the clock is already running slower than you think. Median days on market is 71 in Allston and 68 in Brighton, against 57 in Charlestown, and Allston sellers are landing at 96.3 percent of their original ask. That is a market where the first price is the price that matters. Start from a real valuation rather than a neighbor’s list price.

Disclose the construction, do not editorialize about it. If you are selling within sight of one of these four sites in 2027, buyers will find it. Hand them the BPDA file. Do not offer a prediction about what it does to value, because you do not know and neither do I.

Watch Charlestown separately. At 3.6 months of supply it is the only one of the three that is genuinely tight, and One Mystic is the single largest project of the four. Charlestown has the most to absorb and the least slack to absorb it with.

If you are buying or investing here

The honest investor read is that this round changes the 2028 and 2029 rent picture in Allston and Brighton and changes almost nothing before then.

If you are underwriting a purchase in these ZIP codes today, I would model the new supply as a cap on rent growth at the high end starting in 2028, not as a rent decline. Deliveries land at $3,800 to $4,000 for a two-bedroom on current evidence. That sets a ceiling for renovated units chasing that price point. It does nothing to a $2,800 two-bedroom in an older building, which is where most of Brighton’s volume actually clears.

For small multifamily specifically, Allston and Brighton two-family and three-family sales both carried a median right around $1.25 million over the last twelve months. That is the entry ticket. The student bed math in these neighborhoods is already priced into those numbers, and 1,400 market-rate apartments a mile away does not unwind it.

One more thing on the geography. I have written before about why Beacon Park Yard will not reprice Allston for a decade. Pratt Street and Allston Yards are the near-term story in that neighborhood, and they are eight to ten years ahead of anything at Beacon Park. If you are betting on Allston, bet on these, and bet on the timeline the BPDA board sets in October rather than the one the site plans imply.

The three dates that decide this

Forget the next zoning headline. These are the dates that actually determine whether 1,400 apartments get built.

Monday, September 28, 2026, 6:00 to 7:00 PM. The Boston Planning Department holds a virtual public meeting on the three Allston and Brighton projects, Allston Yards Building D, 22-24 Pratt Street and 83 Leo M. Birmingham Parkway. Registration runs through bosplans.org, meeting ID 165 508 1546.

Wednesday, October 14, 2026. The 30-day public comment period closes. It opened September 14. Comments go on the record for each project on the BPDA site.

October 2026, BPDA board meeting. The board votes on designating all four as demonstration projects and moving to formal negotiation. Kairos Shen has been direct about the goal. “The mission is to get more housing under construction in 2027.” If that vote passes, watch for foundation permits next spring. If it fails or gets continued, the 2028 delivery assumption moves out a year and so should your model.

Behind round one there is a second round already sketched. About a dozen more projects are in discussion, and the mayor has signaled support for up to $100 million in total abatement agreements, roughly 4,000 additional units. That sits alongside the separate $110 million Housing Accelerator Fund the city stood up in December 2024. If round one clears in October, round two is where the real inventory number gets set for Allston, Brighton and Charlestown.

What I actually think

I am glad the city stopped pretending this was a permitting problem. Spending $31.5 million to convert approvals into buildings is a more honest policy than another rezoning announcement, and 185 income-restricted homes plus $20 million in cash beats zero of each.

But I would rather Boston price its inclusionary requirement out loud than negotiate it one building at a time behind a tax abatement. Round one already lands at 13.2 percent on site. That is the market-clearing number this administration found by itself, on its own four deals, after saying no to exactly that number in public. A policy set at what the deals actually carry would produce the same housing without 47 separate negotiations, and without handing every project’s fate to whoever attends a Tuesday night meeting.

If you own or you are shopping in Allston, Brighton or Charlestown and you want to talk through what a 2028 delivery does to your specific block or your specific underwriting, reach out. I would rather walk you through the actual closing record for your street than have you trade on a headline.

Sources

• City of Boston, “Mayor Michelle Wu Proposes Targeted Tax Abatements to Jumpstart Construction on 1,400 New Homes,” September 15, 2026
• The Boston Globe, “In bid to boost Boston housing supply, Wu proposes tax relief,” September 14, 2026
• WBUR, “Boston explores new tax breaks to jump-start stalled market-rate rental projects,” September 15, 2026
• Banker & Tradesman, “Wu’s Tax Breaks Not Enough to Restart Housing, Critics Say,” September 20, 2026
• Banker & Tradesman, “Tax Breaks Could Jump-Start 1.4K Units in Stalled Developments,” September 14, 2026
• Dorchester Reporter, “Boston explores new tax breaks to jump-start stalled market-rate rental projects,” September 14, 2026
• The Boston Globe, “In Boston, new development approvals sank in 2025,” January 2, 2026
• City of Boston, Inclusionary Zoning policy
• Chelsea Record, “Inclusionary zoning policy referred to full council,” April 23, 2026
• City of Boston, FY26 Property Taxes
• City of Boston, “Mayor Wu Announces Housing Accelerator Fund,” December 2024
• Boston Globe Spotlight, “How sky high construction costs drive the housing crisis”
• Bisnow, “BPDA Greenlights Redevelopment Of Radio Station Building In Brighton,” April 2025
• BPDA project file, 22-24 Pratt Street (comment period through October 14, 2026)
• MLS PIN closed sale and closed lease records for ZIP codes 02134, 02135 and 02129, twelve months ending August 31, 2026. Pulled directly by BMN Boston, September 20, 2026.

Smart Home Wi-Fi Smart Plugs (4-Pack) App and voice control for lamps and appliances. View on Amazon → As an Amazon Associate, BMN earns from qualifying purchases.