On September 1, 2026, one hundred and fourteen leases closed across the five ZIP codes that ring Boston’s biggest campuses. Allston, Brighton, Mission Hill, the Fenway and Kenmore. In one day.
All of January closed 85. All of February closed 86. I pulled those numbers out of MLS PIN myself, and the shape of them is the whole story. This market does not run on twelve months. It runs on one week in the fall, and the reason it runs that way is a bed count that never got built.
I want to make an argument that cuts against how most people read this. Boston’s universities have been under a reporting ordinance since 2013, the city publishes exactly how far short they fall, and the shortfall is real and durable. But if you are underwriting a three-family in Brighton on the theory that a broken housing promise is free money, the market has already read the same report you did. It priced it. My proprietary numbers say the campus premium costs about 2.2 points of gross yield, and this fall is the first year in a decade that anyone has gotten to find out whether it was worth paying.
The city has been keeping score since 2013, and the score is public
Boston’s University Accountability Ordinance makes every college in the city file an annual report on who is enrolled and where those students sleep, down to the address. The Mayor’s Office of Housing turns that into the Student Housing Report. The current edition covers Fall 2024 filings. It is the single most useful document in Boston real estate that almost nobody in my business reads.
The headline numbers: 162,458 students enrolled at Boston institutions. Of the students living in Boston, 39,012 are off campus and not at home, meaning they went into the private rental market specifically to attend school here. Citywide, schools house 51 percent of their full-time undergraduates.
Then there is the production number. Between 2014 and 2024, Boston’s colleges added 7,751 dorm beds and removed 1,196, for a net gain of 6,555. That is a 17 percent increase over eleven years. Against enrollment, it is not close. The city’s own arithmetic puts the gap at 24,255 full-time undergraduates who could remain unhoused by their institutions even after every bed currently under construction, board approved, and in the pipeline gets delivered.
What matters for an investor is not the citywide number. It is which school, next to which blocks, and whether anyone has been made to sign a date.
Source: City of Boston Mayor’s Office of Housing, Student Housing Report
Boston College broke the promise. It is not the binding shortfall.
BC’s 2009 Institutional Master Plan pledged a net increase of 1,280 undergraduate beds over ten years. It has delivered about 240. That is 19 percent. In November 2025 the BPDA renewed BC’s plan for two years instead of the four the school asked for, and Allstonia reported City Councilor Liz Breadon calling the renewal “really just kicking the can down the road for two years.” Five of six recent major Brighton campus projects were athletics and recreation. Breadon has since described the Brighton campus as being in a holding pattern while BC waits on a new president who took office in summer 2026.
All true, all fair, and I would still tell a client not to underwrite Allston-Brighton on it.
Here is what the same city report says two tables later. Boston College houses 77.1 percent of its full-time undergraduates, the highest share of any Boston school with 6,000 or more undergrads. BU is 65.2 percent. Northeastern is 44.1 percent. And BC’s off-campus undergraduate count in Boston has gone 1,444, then 1,280, then 1,111 across three reporting years. Down 23.1 percent in two years.
So BC is fifth on the city’s list of schools by undergraduates in private housing, and it got there while shrinking. The 1,111 number that neighbors are angry about is the smallest it has been in years. If you own a three-family off Comm Ave and you think BC undergrads are your rent floor, they are not. Your rent floor in Allston-Brighton is Boston University’s 3,703 off-campus undergrads, plus something people consistently miss: Allston has more graduate students living off campus than undergraduates, 3,002 against 2,513. Graduate demand is quieter, older, and much less tied to any one school’s master plan.
Northeastern signed a date, and a date is a countdown on your rent floor
Northeastern is the actual shortfall. 6,765 full-time undergraduates off campus and not at home. The lowest housing rate of the big schools at 44.1 percent. And it is the only large school whose off-campus population grew, up 1,639 undergraduates in two years, a 36.1 percent increase.
Which is exactly why I think Mission Hill is the weakest long hold of the three neighborhoods, not the strongest.
In November 2025 the city and Northeastern signed a five-year PILOT agreement and a ten-year master plan together. The terms are specific: 1,000 additional student beds within five years, on top of the 1,215-bed tower at 840 Columbus Avenue that broke ground in February 2026 and opens in 2028. A further 2,226 beds are board approved. One million dollars a year for Roxbury housing stabilization, ten million over ten years to city housing stabilization funds, and more than 49 million in cash and community benefits over the five-year term.
This is the part I want investors to sit with. A promise with no delivery date is worth more to a landlord than a promise with one. For fifteen years the Northeastern shortfall had no clock on it, and Mission Hill rents priced accordingly. It has a clock now.
The clock is not airtight. The agreement’s enforcement language only triggers a mitigation plan and a public BPDA hearing if Northeastern increases its off-campus student population. Holding steady at 6,765 triggers nothing. That is a real gap, and it is the strongest remaining argument for Mission Hill. But 3,441 beds between the tower and the board-approved pipeline is a genuine supply event aimed directly at the neighborhood with the highest rent per bedroom in Boston.
One footnote for anyone who opens the report and checks my math. The city’s own table lists Northeastern’s unmet need as 7,424, which is its off-campus count plus 500 planned demolitions and does not net out the 2,226 approved beds. The citywide total does net them. Credit the approved beds on the row and Northeastern’s post-delivery gap is closer to 5,200.
Berklee is the school nobody is watching
If I were buying for a ten-year hold on the strength of a chronic shortfall, I would be looking at the Fenway, and the reason is a school that never comes up in these conversations.
Berklee College of Music houses 27.0 percent of its full-time undergraduates. That is the lowest rate of any midsize Boston institution. It puts 3,853 full-time undergraduates into the private market and ranks second in the entire city for undergraduates in private housing, ahead of Boston University and more than triple Boston College. Against an unmet need of 3,457, Berklee has 450 beds in the pipeline and nothing under construction or board approved.
No PILOT bed commitment. No renegotiated master plan in the news. No councilor holding a hearing about it. Berklee sits in the Fenway, which at 7,249 off-campus students is the largest concentration in Boston by a wide margin, and its shortfall has no date attached to it at all.
That is the position I would rather own. Not the biggest gap, and not the loudest broken promise. The gap nobody has been forced to put a calendar next to.
Where the students actually live is a triple-decker map
The geography is tight. Four neighborhoods hold 21,649 off-campus students, roughly 57 percent of the city’s total.
| Neighborhood | Undergrad | Graduate | Total | Share |
|---|---|---|---|---|
| Fenway | 5,058 | 2,191 | 7,249 | 19.1% |
| Allston | 2,513 | 3,002 | 5,515 | 14.5% |
| Mission Hill | 2,779 | 1,667 | 4,446 | 11.7% |
| Brighton | 1,773 | 2,666 | 4,439 | 11.7% |
| Roxbury | 1,451 | 1,726 | 3,177 | 8.4% |
| Dorchester | 1,671 | 1,253 | 2,924 | 7.7% |
| Jamaica Plain | 392 | 1,731 | 2,123 | 5.6% |
| Back Bay | 1,090 | 665 | 1,755 | 4.6% |
Source: City of Boston Student Housing Report, Appendix C
And the property type is the part that should matter to anyone bidding on a small building. The city geocoded the addresses. 17,517 off-campus students live in properties of three units or fewer, or in condominiums. That is more than half of them, sitting in the exact stock the report describes as “initially built for Boston’s workforce.”
This is why the investor story and the owner-occupant story are the same story told from opposite sides of a bid. There is no separate student housing market in Allston. There is one triple-decker, and either a landlord buys it and rents it by the bedroom or an owner-occupant buys it and lives in it. The city’s data says which one has been winning.
What a bedroom rents for, and why landlords keep adding them
Here is where I can bring my own data instead of somebody’s rent index. These are closed leases out of MLS PIN, January through early September 2026, in Allston 02134.
■ Rent per bedroom
Source: MLS PIN closed leases, BMN Boston analysis. The five-bedroom line rests on 8 closings; treat it as directional.
Read the two columns against each other. As you go from a one-bedroom to a five-bedroom, the rent a landlord collects on the unit climbs from $2,430 to $5,201. The rent each bedroom carries falls from $2,430 to $1,040. Adding bedrooms is worth roughly $700 a month each to the owner and it lowers the per-person price for the tenants, which is why a three-family with nine bedrooms outperforms the same square footage cut into tidy two-bedrooms.
A note on a number that circulates locally. Allston’s median asking rent is not $5,200. Depending on whose index you use it sits somewhere between $2,450 and $3,650. The $5,200 figure is real, but it is specifically what a five-bedroom closed for in Allston this year, average $5,201 across eight closings. Small sample, correctly labeled, and a much more interesting number than a neighborhood median because it is the number that actually drives an underwriting model.
The sharpest comparison in my whole dataset is the four-bedroom line, where the samples are decent. Mission Hill four-bedrooms closed at an average of $5,545 across 26 leases. Allston came in at $4,479 across 20. Brighton at $4,253 across 46. On a per-bedroom basis Mission Hill runs $1,386 against Brighton’s $1,063, a 30 percent spread across about four miles. That is the Northeastern shortfall showing up in a rent roll.
The rent floor is real, and you are already paying for it
Now the part that changes what I would actually advise.
I pulled every closed small multifamily sale in ten Boston ZIP codes from September 2024 through September 2026, computed price per bedroom, then set it against closed lease rent per bedroom from this year’s leasing season. That gives a gross yield built entirely from local closed transactions rather than asking rents or a national model.
| ZIP | Sales | Price / bed | Rent / bed | Gross yield | Days | % of orig. ask |
|---|---|---|---|---|---|---|
| Dorchester 02124 | 152 | $148,716 | $1,185 | 9.56% | 79 | 97.6% |
| Mattapan 02126 | 56 | $153,720 | $1,130 | 8.82% | 106 | 100.0% |
| Roxbury 02119 | 53 | $172,101 | $1,246 | 8.69% | 117 | 94.9% |
| Dorchester 02122 | 54 | $176,131 | $1,270 | 8.65% | 74 | 99.4% |
| Roslindale 02131 | 78 | $193,686 | $1,348 | 8.35% | 64 | 103.7% |
| Hyde Park 02136 | 56 | $161,924 | $1,119 | 8.29% | 103 | 101.4% |
| Mission Hill 02120 | 15 | $234,946 | $1,600 | 8.17% | 103 | 92.8% |
| East Boston 02128 | 93 | $209,464 | $1,419 | 8.13% | 95 | 94.5% |
| Allston 02134 | 26 | $214,297 | $1,359 | 7.61% | 89 | 96.7% |
| Brighton 02135 | 91 | $227,093 | $1,390 | 7.35% | 80 | 96.8% |
2.21 points. That is the gross-yield gap between Dorchester 02124 and Brighton 02135, and it is what the market charges for the student rent floor. Campus ZIPs also take longer to sell and close further under their original asking price.
The three campus ZIPs are the top three in Boston for price per bedroom. Mission Hill at $234,946 a bedroom is 58 percent above Dorchester’s 02124. And they land at the bottom of the yield column. Dorchester 02124 returns 9.56 percent gross. Brighton returns 7.35 percent. That spread, 2.21 points of gross yield, is the market’s price for the student rent floor, and it is charged whether or not the shortfall persists.
The transaction data says the same thing a second way. Roslindale closed at 103.7 percent of original asking price in 64 days. Hyde Park at 101.4 percent. Mission Hill closed at 92.8 percent of original ask after 103 days on market, Allston at 96.7 percent after 89 days. The campus neighborhoods are not where sellers have the leverage. They are where the list prices got ambitious and buyers negotiated them back down.
So here is where I land, and I want to be precise because I do believe the shortfall case. A chronic bed shortfall is a durability signal, not a yield signal. It tells you your vacancy risk is lower and your rent has a floor under it in a soft year. It does not tell you the building is cheap, because it is not. You are buying about 220 basis points of current return in exchange for stability. If you are a ten-year holder with modest leverage, that trade can be right. If you need the deal to cash flow in year one, Dorchester and Roslindale are doing more for you and it is not close.
Two caveats I will not bury. Mission Hill’s sale sample is thin, 15 closings in the two-year window, so I re-ran it over three years and got $227,425 per bedroom across 24 sales, which holds. And the Fenway ZIPs, 02115 and 02215, had too few small multifamily closings to make the table at all, so they are absent rather than reported. If you want to run these ratios against a building you are actually looking at, our investment property analyzer takes rent roll and price and does the same arithmetic.
The stress test is running right now
I would be doing nobody any favors if I published the shortfall case without this section, because 2026 is the first year the rent floor has been genuinely tested.
New international student enrollment fell about 17 percent last fall on visa policy. Massachusetts hosts roughly 82,000 foreign students and China and India together account for nearly half. The Boston Globe found Mission Hill vacancy had more than doubled to 2 percent from 0.81 percent, with 76 units available on Mission Hill alone, which a Boston Pads executive called virtually unheard of. This past move-in, availability more than doubled again year over year and landlords were offering one to two months free.
Now the honest complication, and it is one I run into constantly: Boston’s rent trackers do not agree, and this year they do not even agree on direction.
| Tracker | What it measures | Reading | Change | Caveat |
|---|---|---|---|---|
| Boston Pads | City of Boston average | $3,408 | up 2.62% | Own listed and managed inventory |
| Realtor.com | Boston metro average | $2,930 | down ~4% | National listing model, 13 straight monthly declines |
| Boston Pads | Allston average, Jan 2026 | $3,062 | down 2.08% | Neighborhood cut, all unit sizes |
| MLS PIN (mine) | Allston closed leases, 2026 | $3,496 | n = 156 | Closed 2 to 5 bedroom leases, not asking rents |
Do not let anyone hand you one of those as “the” Boston rent number. Boston Pads counts its own managed and listed inventory. Realtor.com models national listing data. They are measuring different things and this year the gap is about six points of annual change. What all three do agree on is the direction of vacancy, which is up everywhere, and vacancy is the variable the shortfall thesis actually rests on.
My read: this is a demand shock on the graduate and international side, not the collapse of the undergraduate bed gap. The 24,255 number is structural and takes a decade to move. The visa number can move in one policy cycle and can move back. But it means anyone buying a campus-adjacent building this fall is paying a premium set in a tighter market than the one they are stepping into, and that is worth a hard conversation before the offer, not after the first vacancy.
What I would do, on either side of the bid
For investors:
- Underwrite the housing rate, not the broken promise. The number that predicts your rent floor is the share of undergraduates a nearby school actually houses, and whether its off-campus count is rising or falling. BC’s is falling. Northeastern’s is rising. Berklee’s has barely moved in three reporting years at 27 percent housed.
- Treat a signed delivery date as a negative. Northeastern’s 3,441 beds between the 840 Columbus tower and its approved pipeline are aimed at Mission Hill and Roxbury. If you are buying there for a ten-year hold, model a softer 2029 than 2026.
- Price the premium explicitly. Write the 2.2 points down in your model as what you are paying for stability, then decide if you want it. Do not let it hide inside a cap rate you inherited from a listing sheet.
- Count bedrooms, not units. The per-bedroom rent curve is the entire business model, and a legal four-bedroom beats a nicer two-bedroom on gross rent almost every time. Confirm the bedroom count is legal and the egress is compliant before you underwrite it, because Boston enforces this and a paper bedroom is worth nothing.
- Compare against the boring alternative. Run the same building against Somerville or Cambridge, where you get graduate and professional demand without the September concentration risk, or against Dorchester and Roslindale, where the yield is 100 to 220 basis points better today.
For owner-occupants bidding on the same three-family:
You are not imagining it. In Allston, Brighton, Mission Hill and the Fenway you are bidding against someone underwriting the building by the bedroom, and on a four-bedroom unit in Mission Hill that buyer is capitalizing $5,545 a month where you are capitalizing a place to live. That is a structurally harder bid to win in those four ZIP codes than anywhere else in the city.
Two things actually help. First, this is the softest campus rental market in years, which weakens the investor’s number more than it weakens yours. Second, the same triple-decker stock exists in Roslindale and Dorchester at $148,716 to $193,686 per bedroom against Brighton’s $227,093, and the investor competition there is doing math on ordinary tenants instead of a captive September market. If you want the two-family owner-occupant math run properly, including how the rental income affects what you qualify for, that is a conversation worth having before you write an offer rather than after you lose three.
The universities will keep filing their reports every November, and the city will keep publishing exactly how far short they came. That much is dependable. What is not dependable is the assumption that a shortfall you can read about in a public PDF is still an edge by the time it reaches the closing table.
If you are looking at a small multifamily in one of these neighborhoods and want the per-bedroom rent comps and the closed-sale math on the actual address, send it over. I will pull it out of MLS PIN and tell you straight whether the premium is worth paying on that building. You can reach me here or at 617-955-2224.
Sources
- City of Boston, Mayor’s Office of Housing, Student Housing Report (Fall 2024 University Accountability Reports).
- City of Boston, Northeastern five-year PILOT and ten-year Institutional Master Plan announcement, November 2025.
- City of Boston, Council Strengthens Oversight of Off-Campus Student Housing.
- Allstonia, As Boston College renews its Institutional Master Plan, some neighbors question its promises on housing, November 23, 2025.
- Boston Planning Department, Boston College Institutional Master Plan.
- Northeastern University, 840 Columbus Avenue residence hall groundbreaking, February 2026.
- The Boston Globe, An uptick of apartment vacancies in Boston suggests a decline in student enrollment, September 15, 2025.
- Boston Pads, 2026 Boston Apartment Rental Market Report and 2026 Allston Apartment Rental Market Report.
- Boston.com, Boston rents are down again, July 14, 2026 (Realtor.com data via Boston Business Journal).
- NBC Boston, Boston rent situation: more apartments available for Sept. 1, 2026.
- MLS PIN closed sales and closed leases, proprietary BMN Boston analysis. Sales window September 2024 to September 2026; lease window January 1 to September 3, 2026.
