Topic
Enrollment: three forecasts compared
Why do the enrollment forecasts disagree, and which one does the base case use?
The base case uses the forecast by the district's own consultant, NESDEC, updated for the births that have since been counted. The School Department plans with a second forecast, by Cropper-McKibben, which expects enrollment to rise again; the difference is almost entirely how many young children each expects Brookline to have, not how classes move through the grades.
- All three start from about 6,900 students this fall. By FY2034 NESDEC has 6,127, the base case 5,941 and Cropper-McKibben 7,444.
- NESDEC estimated about 500 births a year from 2023, before those births were counted; the counts came in at 422 in 2023 and 443 in 2024, and the base case updates NESDEC's forecast with them. The 2025 count, 423, is a little below what the base case assumes.
- Cropper-McKibben forecasts the whole population first and projects far more births and young families, so its kindergarten classes rise from about 500 to 607 a year.
- On how classes grow or shrink as they move up a grade, the two published forecasts are close to each other and to Brookline's recent record.
- Measured from their 2005–09 average, births have fallen about twice as fast in neighboring towns as in Massachusetts as a whole, and nearly three times as fast in Brookline; Brookline differs from the family suburbs mainly in how many families move in with young children.
- Four numbers, updated each fall, show which forecast enrollment is following: births, kindergartners per birth, the October count, and the room that leaves for children from new homes.
- For births not yet counted, the base case assumes 449 a year, the 2022 to 2024 average: it continues the level Brookline has actually had rather than assuming a rebound or a further fall. It keeps that average for 2025, which came in at 423, rather than resetting on one year. Births are the main reason the forecasts differ; measured against their own births, all three assume similar rates of families moving in.
Every enrollment forecast answers two questions: how many children start kindergarten each year, and what happens to each class as it moves up a grade. This page sets out how the three forecasts used on this site answer them, where they agree and where they do not. The enrollment workbook has every number below and a version you can change.
The three forecasts
| Item | NESDEC | Cropper-McKibben | The base case |
|---|---|---|---|
| Who | The New England School Development Council, the district's consultant | Cropper-McKibben, for the School Department | This site |
| When | October 2023 | 2023, revised April 2024 | October 2026 |
| Used by | The School Department's budget plans | The forecast on this site | |
| Method | Cohort survival: births become kindergarten, then each class moves up a grade by Brookline's historical ratios | A population model: forecasts every resident by age, area by area, then converts school-age children into enrollment | NESDEC's forecast, with the births it estimated replaced by the births that were counted |
| Births assumed | About 500 a year from 2023 | About 616 a year for 2025 to 2030 | Actual through 2024 (482, 422, 443), then 449 a year (2025 has since been counted at 423) |
| FY2027 (this fall) | 6,774 | 6,917 | 6,833, the October 1 count |
| FY2030 | 6,464 | 7,050 | 6,402 |
| FY2034 | 6,127 | 7,444 | 5,941 |
| FY2040 | 5,415 |
Enrollment is falling below what the schools are staffed for
Public school students, PK–12. Actual through FY2026, then NESDEC’s projection updated for actual births (the base case) and, dashed through FY2034, the School Department’s forecast and NESDEC’s forecast as published, before the births update (both scaled to this fall’s count). The shaded area is the empty seats if classes stay at the FY2027 level, as the base case assumes.
The two assumptions that matter
Every forecast rests on two assumptions: how many children are born to Brookline residents, and how many families move in or out as those children grow up. The chart shows both. On the left, births: Brookline's actual count fell from about 690 a year in 2008–13 to 422 in 2023, 443 in 2024 and 423 in 2025; NESDEC's 2023 forecast estimated about 500 a year, Cropper-McKibben's population model about 590 to 630, and the base case holds 449. On the right, students per birth: total enrollment divided by the births of the 13 years whose children are in K-12, a single measure of families moving in. Measured against each forecast's own births, all three expect it to rise from about 0.84 today to around the 0.90 of the 2016–19 peak; the base case reaches 0.93 by FY2040.
The forecasts' two key assumptions
Left: births to Brookline residents, actual and as each forecast assumes (Cropper-McKibben publishes five-year totals, shown as yearly averages). Right: students per birth, enrollment divided by the births of the 13 years whose children are in K-12; for each forecast, measured against its own births, so it shows what the forecast assumes about families moving in. Fiscal years on the right.
Because the births behind enrollment are smaller, even a return to the families-moving-in rate of the 2010s boom would leave enrollment lower than today. Counting births through 2024 and assuming births from 2025 on match the 2022–24 average, about 449 a year (2025 came in at 423), the pool of Brookline-born children falls about 31% by FY2040:
| FY2040 enrollment if enrollment per birth… | Students | Per birth |
|---|---|---|
| stays where it is today | about 4,880 | 0.84 |
| returns to the peak of the 2010s boom | about 5,250 | 0.90 |
| follows the base case (NESDEC, updated for actual births) | about 5,410 | 0.93 |
| keeps enrollment level with today | 6,833 | 1.17 |
The base case's 0.93 is already a little above the 2010s peak. Keeping enrollment level would take more families moving in than during the 2010s boom, by a wide margin.
So the forecasts agree more than their totals suggest. They differ mainly in how many children they expect to be born, and the births of 2022 to 2024 are now counted.
Is the drop in births unique to Brookline?
No. Between the 2005–09 period and the 2019–23 period, comparing five-year averages, births fell 12% across Massachusetts and faster in the towns around Brookline: 20% in the neighboring cities and inner suburbs and 14% in the family suburbs as a group. Brookline's 26% was the 5th largest fall of the 99 towns with at least 200 births a year, close to Newton (26%), Lexington (25%) and Belmont (24%).
Births have fallen across the region, fastest in the inner suburbs
Births to residents, each group's 2005–09 average = 100. Neighboring cities and inner suburbs: Cambridge, Somerville, Newton, Watertown, Arlington, Belmont. Family suburbs: Lexington, Wellesley, Needham, Winchester, Concord, Wayland, Sudbury, Weston. Source: Department of Public Health birth reports.
Where Brookline differs from the family suburbs is not how many children are born there but how many families arrive with young children. In Lexington, the fall 2025 kindergarten was 2.07 times the town's births five years earlier; in Brookline it was 0.93. Why families settle elsewhere looks at what goes with that difference. For the forecast, it means a rebound in Brookline's births would most likely be part of a regional one, and the counts so far, 422 in 2023, 443 in 2024 and 423 in 2025, show a level rather than a rise.
How a kindergarten class is set
Kindergarten is births five years earlier times a kindergarten-per-birth ratio: some children born here move away or go to private school, and some born elsewhere move in. Brookline's ratio has run from 0.81 to 1.01 since FY2014, leaving out the COVID year, averaging about 0.90 before COVID and 0.85 since. This fall it was about 0.84: 410 kindergartners from 491 births in 2021.
NESDEC uses about 0.85, close to the recent record. What has changed since its forecast is births: it estimated 518 for 2023 and 507 for 2024, before those births were counted, and the counts came in at 422 and 443. Updating for them makes the FY2029 and FY2030 kindergarten classes 12% to 19% smaller than NESDEC's, and later ones about 9% smaller, which is what the base case does.
Cropper-McKibben does not use a fixed ratio. Its population model forecasts births and families moving in together, area by area, and its kindergarten rises from 507 this fall to 607 by FY2034. Most of that rise comes from births: the model projected about 588 a year for 2020 to 2025 (the counts averaged 465) and about 616 a year for 2025 to 2030. Measured against its own births, its kindergarten per birth rises from about 0.86 to about 0.99, within Brookline's historical range. If births stay near the level counted so far, about 449 a year, reaching 607 kindergartners would take about 1.35 per birth.
How a class moves up the grades
Each year a class is multiplied by a grade-to-grade ratio. Above 1, more families arrive than leave at that age; below 1, more leave. Brookline's actual record, against what each forecast assumes:
| Move | Before COVID | Since FY2022 | NESDEC | Cropper-McKibben |
|---|---|---|---|---|
| K to 1 | 1.00 | 1.03 | 1.03 | 0.99 |
| 1 to 2 | 0.99 | 1.02 | 1.02 | 1.01 |
| 2 to 3 | 1.02 | 1.00 | 1.00 | 1.00 |
| 3 to 4 | 0.98 | 1.01 | 1.01 | 1.02 |
| 4 to 5 | 0.99 | 0.99 | 1.00 | 1.02 |
| 5 to 6 | 0.99 | 0.97 | 0.95 | 0.98 |
| 6 to 7 | 0.98 | 1.00 | 1.00 | 1.02 |
| 7 to 8 | 0.98 | 1.00 | 1.00 | 1.00 |
| 8 to 9 | 0.99 | 1.01 | 1.00 | 0.99 |
| 9 to 10 | 1.02 | 1.02 | 1.01 | 1.03 |
| 10 to 11 | 0.98 | 1.00 | 1.00 | 0.99 |
| 11 to 12 | 1.01 | 1.00 | 1.01 | 1.01 |
| Kindergarten to grade 12, multiplied | 0.93 | 1.06 | 1.02 | 1.05 |
Before COVID a class finished grade 12 about 7% smaller than it started; since FY2022 classes have grown about 6%, partly families returning after COVID. Both published forecasts sit between the two eras, and close to each other. Running the forecast on Brookline's post-COVID ratios instead of NESDEC's adds about 80 students by FY2040, against more than 2,000 between the forecasts.
One class, followed through the grades
The children who started kindergarten this fall were mostly born in 2021, when Brookline residents had 491 babies:
| School year | Grade | NESDEC | Cropper-McKibben | Actual |
|---|---|---|---|---|
| 2026–27 | K | 415 | 507 | 410 |
| 2027–28 | 1 | 427 | 505 | |
| 2028–29 | 2 | 437 | 511 | |
| 2029–30 | 3 | 437 | 513 | |
| 2030–31 | 4 | 441 | 523 | |
| 2031–32 | 5 | 441 | 535 | |
| 2032–33 | 6 | 419 | 530 | |
| 2033–34 | 7 | 417 | 544 |
NESDEC sets the class at 491 × 0.845 = 415 and moves it up by its ratios; the base case follows the same path, about 0.9% higher to match this fall's total count. Cropper-McKibben starts the same children 92 higher and keeps the class growing. The actual kindergarten came in at 410.
How each has done so far
Both forecasts were made in 2023, so two school years can be checked:
| Item | Actual | NESDEC | Cropper-McKibben |
|---|---|---|---|
| Total, FY2026 (2025–26) | 6,948 | 6,880 (−1.0%) | 6,929 (−0.3%) |
| Total, FY2027 (2026–27) | 6,833 | 6,774 (−0.9%) | 6,917 (+1.2%) |
| Kindergarten, FY2026 | 453 | 412 | 497 |
| Kindergarten, FY2027 | 410 | 415 | 507 |
| Kindergarten to grade 8, FY2027 | 4,436 | 4,436 | 4,588 |
| Kindergarten to grade 8, FY2028 (the School Department's October 2026 projection) | 4,338 | 4,337 | 4,635 |
Cropper-McKibben was closer on total enrollment last year, because classes already in school grew a little faster than NESDEC assumed. This fall NESDEC was closer on the total (59 low against 84 high) and exactly right on kindergarten through grade 8, and it has been closer on kindergarten, the part that depends on births. The School Department's own projection for next fall, made in October 2026, is within one student of NESDEC's for kindergarten through grade 8. The two forecasts only pull far apart from FY2029, as the smaller birth years reach kindergarten, so the next few kindergarten counts are the real test.
Indicators to watch
Births are a leading indicator: they say five years ahead how many children will start kindergarten, and the October count shows how many families moved in since. Read together each fall, four numbers show whether enrollment is following the base case or the School Department's forecast, and how much room the schools have for children from new homes.
Births to Brookline residents
423 in 2025
Up from 443 in 2024. The base case assumes 449 a year from here; NESDEC had estimated 492 for 2025, and the School Department's forecast about 616.
Why it matters: each year's births set the kindergarten class five years later. Counts near 500 would point back toward NESDEC's original forecast, and near 600 toward the School Department's.
Next reading: 2026 births, in the Town's annual report.
Kindergartners per birth
0.84 this fall
410 kindergartners from 491 births in 2021. Since FY2014 it has run from 0.81 to 1.01, leaving out the COVID year.
Why it matters: it measures young families moving in. Next fall's class comes from 482 births in 2022: about 411 kindergartners in the base case, 388 in NESDEC's forecast and 525 in the School Department's.
Next reading: the October 2027 count.
October enrollment
6,833 students this fall
Down 115 from last year's 6,948. For FY2028, the base case expects about 6,662, NESDEC 6,585 and the School Department 6,886.
Why it matters: next fall the base case and the School Department's forecast are about 220 students apart. The gap widens from FY2029, when the smaller birth years of 2023 and 2024 reach kindergarten, so a few more falls of decline would confirm the base case.
Next reading: the October 2027 count.
Room for children from new homes
about 1,800 homes by FY2030
In the base case, enrollment falls to 6,402 by FY2030, about 430 students below this fall, with homes still being built at the recent pace. Those seats could take the children of about 1,800 more average new homes, and about 5,800 by FY2040.
Why it matters: this is the schools' capacity to absorb new housing at little added cost. It counts every seat the decline opens, which holds only if staffing stays at this year's level. With staff falling half as fast as enrollment, as the base case assumes, about half of them stay open. Seats also open school by school, not evenly.
Recalculated each fall from the three readings above.
Why the base case uses NESDEC, updated for actual births
Births are the one input known five years ahead: every child who will start kindergarten through fall 2030 has already been born. NESDEC's method turns those births into enrollment in a way anyone can check, and once its estimated births are replaced with the counted ones, its kindergarten classes match this fall's. UMass Donahue Institute population projections point the same way, with Brookline's 5–14 age group falling about 28% from 2020 to 2035. Cropper-McKibben's higher enrollment rests on more births than have been counted so far. If births stay near the counted level, reaching its FY2034 enrollment would take about 1.15 students for every birth, against a twenty-year high of 0.90: drawing families the way Newton or Belmont do, a change in what kind of town Brookline is rather than a continuation of its record. Why families settle elsewhere compares Brookline with its neighbors.
What it means for the budget
In the base case, school staff fall half as fast as enrollment, as they did from FY2019 to FY2025 (enrollment fell 10.6% and staff 5.0%). The School Department's own plan holds staff at this year's level. Because staff move with enrollment, at half speed as it falls and in full as it rises, the choice between NESDEC and Cropper-McKibben moves FY2040 by $47.9 million. The difference shows up in two choices:
- Staffing. If school staff follow enrollment down fully, FY2040 has $9.5 million more room than the base case; if they are held at this year's level, as the School Department plans, it has $27.8 million less, and the room runs out for good in FY2036. If enrollment instead rebounds as Cropper-McKibben forecasts and staff follow it up, FY2040 has $47.9 million less. Will Brookline need another override? sets both against the other decisions.
- Empty seats. Enrollment falling faster than staff opens seats that children from new homes can fill at little cost. New homes and Brookline's budget works through that case, and it disappears if Cropper-McKibben is right.
The Forecast Explorer's School enrollment forecast setting switches the whole forecast between the base case, flat enrollment and Cropper-McKibben.
Go deeper
- Schools: staffing, costs and open seats: How is Brookline's school system staffed, what does an added student cost, and how much room do the schools have?
- Why families settle elsewhere: Why does Brookline draw fewer young families than towns like Lexington?
- The base case to 2040, in full: If costs and revenue keep behaving as they have, where is Brookline's budget headed?
- New homes and Brookline's budget: What does a new home cost the Town, now that enrollment is falling?
- Assumptions register: What does the forecast assume, and why?
Last updated 2026-10-27. What has changed.