132 times a school line went to zero — and 93 of them came back

What this establishes

How much stopped, year by year

Each bar is the previous year’s funding on every line the book took to zero that year, split by whether the line was ever funded again. The shaded years print so few zeros that nothing can be measured in them either way.

What the money bought

The same $734,469, cut by what the district’s own line name says the money was for. The grouping is ours, read off the label text; the district publishes no category for these lines.

And by the school the line is named for

Read off the district’s own prefix — H.S., M.S., E.S., P.S. Lines naming no school are district-wide or central, and are reported as naming no school rather than assigned to one.

The book changed shape, and that is not a cut

88 line names present in FY2022 are absent from FY2023. That is by far the largest apparent ending in the series, and the two panels below are why it is not one: the count of printed lines falls and the money does not.

FY2022 is restated by fy24-approved-budget.txt. FY2023 is restated by fy27-budget-projections-as-of-2-24-26-with-restorations.txt and fy27-budget-projections-as-of-3-16-26-with-restorations.txt. Different documents, different line lists, the same district.

The rename check, and what it caught

Rule 6 of this project exists because a line that goes to zero and reappears under a new name produces a −100% rate that looks like a finding. So renames are tested for arithmetically rather than by reading the words: two differently-spelled lines carrying the identical value in every fiscal year they share are one line under two spellings. 10 pairs pass that test.

Where no shared year exists, the words are all there is

Three of the lines that stop at the FY2022 boundary reappear the next year with a letter changed, and there is no overlapping year to settle it arithmetically. These are ours and they are candidates: near-identical labels for the same school, the successor first printed after the predecessor’s last year.

And 4 pairs that are not renames at all

The same test also finds pairs where both labels run to the end of the series carrying identical figures. That is not a rename and it is not evidence of one. Nothing in the document says whether it is one line printed twice or two lines that happen to be paid the same.

A second source, and it does not show the money leaving

Everything above this line comes out of the district’s own book, which shows the general fund and nothing else. DESE collects spending from every district in the state to its own definitions and publishes it split by fund — which is precisely the thing the budget book structurally cannot show. So for each family of lines that stops on this page, the state’s figures for the same kind of spending are set beside it. cross-checkedtwo independent sources, at different grains

Lines that fell sharply and did not recover

Going to zero is the loud version. 25 lines are still printed in FY2025 at half or less of their own peak — $1,633,664 of annual funding below where those lines once stood. The test is stated in full below the table.

Recent is the mean of the last 2 years the book prints for that line; a line qualifies if its own peak reached $10,000 and its recent level is at or under 50% of it. Those thresholds are ours.

What the town said about one of these

The meeting archive now runs from 2009-01-05 to 2026-11-17 — 12014 readable documents out of 12065 the town has published — so it reaches back over the whole of the series above. Three things people said sit close enough to a line on this page to be worth setting beside it. None of them explains a figure; each of them is what a figure looks like from the other side.

3 of the 5 guidance counsellor lines stop appearing after FY2022. The other 2 are still printed in FY2025 — including H.S. Guidance Counselor and M.S. Guidance Counselor. In the same meeting, the same speaker said: “Dr. Jodi Fortuna and the other presenters do not suggest cutting social workers.”

The largest permanent ending, and a board meeting five years later

System School Psychologist is the largest line on this page that went to a printed zero and stayed at zero, last funded in FY2016. The meeting archive now reaches back far enough to have something to say near it.

Professional development: five lines to zero, one line up

Every per-school professional development line in the book is printed at zero from FY2021 onward. The district-wide line is not: it carried $25,000 in FY2021 and $32,925 in FY2025. The per-school lines came to $28,813 at their peak in FY2020.

What was searched, and what came back empty

A search that finds nothing prints nothing, and nothing reads as nobody said it. It is not: it means nobody said it in the 12014 documents that can be read. Terms that returned meeting documents: guidance counselor (11), professional development (26), department head (157), tutoring (19), psychologist (4), school psychologist (3), textbook (2). Terms that returned none at all: adjustment counselor.

And the record itself is thin in places. Of 402 School Committee meetings the town lists between 2021 and 2026, minutes are published for 162 — 40%. In 2021 and 2022 there are none at all, and 3 years sit under half. A search finding nothing in a year like that is not evidence that nobody discussed it. 51 of the 12065 documents cannot be read at all, so no search here covers them either.

Every line that went to zero and stayed there

Every zeroing, all 132 of them

Lines that stop appearing altogether

Not the same thing as going to zero, and much weaker evidence. 164 lines stop being printed at some point in the series, carrying $2,931,555 in their last printed year — and 88 of them stop at the single document boundary above.

What a board member could do with this

One thing, and it is small. A zero in this book means the line was printed at zero. It does not mean the money stopped. The three routes that produce an identical zero — consolidation onto another line, a grant picking it up, and a re-presentation of the whole book — are all visible somewhere in the 12 years above, and none of them is a cut.

So the question worth asking when a line shows zero is not why was this cut. It is which line is it on now, and which fund is paying. The second half of that is the one nobody can currently answer from anything published, and it is why what the schools actually spent, per line, in any year before fy2026 sits in the register.

And the thing that would make next year’s version of this page far stronger costs the district nothing it does not already do: a prior-year column printed in the same line shape as the current year. The break at FY2023 is the whole reason 88 lines look like endings and are not.

How this was measured

One query, over one table. budget_figure where stage=‘restated’: 351 lines, FY2014 to FY2025, out of 16 district documents. The same line_key is compared across years and nothing is compared across stages — a budget and a restatement are different quantities and this page never differences one against the other.

Two rows in the table are printed totals that the extractor read as though they were budget lines — “GRAND TOTAL ATHLETICS” and “plus TOTAL Revolving Fund” — and both are excluded by name. 35 rows are flagged in the database as ones where two documents restate the same line and year differently.

The documents behind this

The payload this page draws is published whole at /data/stopped-funding.json, and the rows behind it can be queried directly — see the database.

What we still cannot answer

All four are rows in the register at what we cannot answer, which is what the records request reads — not prose written only here.