What we cannot answer

money in

• Bus fees

  charged by published policy · no destination found in any ledger

• Trust fund income

  scholarships, cemetery, stabilisation · extract is check-failed

• Student activity accounts

  held by the school under its own authority · not in the town’s books at all

• Grants received in earlier years

  spent now, booked then · nine funds, no FY26 revenue

• Whether a shortfall in state aid falls on the schools or on the rest of the town

  Chapter 70 is 35% of the school appropriation and the town has no vote on it, so which side of the budget absorbs a miss is the question a resident actually has. Nothing published answers it. The free cash proof states one net cherry sheet figure for the whole town; the general ledger’s state aid accounts share no organisation code with any expense account, 0 of 222, so no aid dollar can be followed to anything it paid for; and the appropriation is voted before the final aid number is known. A year where aid rose and the appropriation rose with it and a year where aid was flat and the town covered the difference are different facts about the world and identical facts on the page. — closes: the Town’s FY-end recap (Form A-1) beside the Cherry Sheet for the same year, which states estimated and actual receipts on one page.

• Why DLS certifies a different amount from the free cash its own proof identifies

  The Division of Local Services’ free cash proof cross-foots exactly: in all 45 town-years of the nine-town set the eleven component lines sum, to the cent, to the workbook’s own Identified Free Cash July 1,. But that is not the figure DLS certifies. Current Year Calculation — the number the Town quotes, the number the Finance Director announced, and the number that carries into the next year — differs from it in all 45, by $361,912 for Lunenburg in 2025 and by 69% for Uxbridge in 2024. The workbook prints no reconciling line, and nothing in this archive states what the adjustment is. — closes: DLS’s free cash certification letter to the Town, which states the adjustments made between the balance sheet the Town submitted and the amount certified.

• Which departments’ unspent appropriations produced the free cash

  Money appropriated and not spent is the largest single component of Lunenburg’s free cash — 47% of five years of it, and 66% of the record 2025 certification. It is published as one town-wide figure across 67 departments. Prudence, a post left vacant, a project that slipped into the next year and a line that was over-budgeted from the start are four different facts about the world and one number on the page. — closes: the period 13 general ledger, showing budget, expended and turnback for each department at the year-end close.

• How much of free cash is the schools’

  Free cash is certified for the town as a whole. Neither the DLS proof nor the Town’s own undesignated fund balance roll-forward attributes any component to any department, so nothing published says how much of it the school department’s unspent appropriation produced — nor how much of any appropriation of it reaches the schools. The projection on this site is built from budget columns and free cash is derived from actuals; they are subtracted once, side by side, and never joined. — closes: a departmental turnback schedule for the year-end close, which would split the unexpended appropriations line by department.

• Whether Lunenburg’s free cash is unusual for a town its size

  The DLS proof carries no denominator — no population, budget, revenue or levy for any of the nine towns, Lunenburg included. So Littleton’s $11.3M against Shirley’s $272K says nothing about which is closer to its own target, and free cash as a share of the operating budget cannot be computed for a peer at all. Composition and volatility are ratios and do compare; the level does not, and is not shown. — closes: the Tax Rate Recapitulation (Schedule A) for each town, which prints the operating budget the ratio needs.

• Why the Town’s undesignated fund balance and DLS’s free cash differ

  Two publishers describe the same 30 June balance sheet and print three different figures. The Town’s annual report roll-forward gives an undesignated fund balance of $4,236,488.24 at 30 June 2025; DLS identifies $3,716,282 of free cash on the same date and certifies $3,354,370. Each worksheet cross-foots exactly to its own printed total, and the FY2024 report’s closing balance is the FY2025 report’s opening one to the cent — so neither is in error, and no document in this archive prints a line between any pair of the three. — closes: the Town’s free cash submission to DLS (the balance sheet and the reconciliation of receivables and deferred revenue that accompanies it).

• Where the PEG Access balance went between FY2019 and FY2021

  The fund's carried balance is continuous in the report from FY2015 to FY2019 -- each edition's Starting Balance is the previous edition's Ending Balance to the cent -- and then stops being a series. FY2019 prints a Total of $573,837.37 (starting balance, plus revenue, less expenses, plus the Comcast Tech Fund of $124,796.01); FY2020 opens at $447,813.36, $126,024.01 lower; FY2021 opens at $272,000.00, which the same page's prose states is the budget rather than a balance. One document explains part of it: Article 16 of the FY2019 Annual Town Meeting established a PEG Access Enterprise Fund under MGL c.44 s53F1/2 and transferred into it the balances of the PEG Access and Cable Related Receipts Reserved For Appropriation Fund and the Comcast Tech Capital Grant Fund as of 30 June 2019. So the fund changed legal form at exactly this point, which is why the statement changes shape. WHAT IS NOT ESTABLISHED: the amounts transferred, and therefore whether the $126,024.01 step is the transfer, a closing entry, a restatement, or something else. There is also a smaller unexplained step earlier, FY2017's Ending Balance $303,909.77 against FY2018's Starting Balance $303,551.54, $358.23 apart, with no fund change to account for it. -- closes: the Town Accountant's opening and closing trial balances for the PEG Access and Cable Related funds for FY2018, FY2019 and FY2020, showing the transfers in and out.

• What quantity the Town’s own published state-aid growth rates of 5.3% and 1.4% measure

  The FY2027 Town Meeting booklet states that "From FY13-FY23 state aid increased at an annual rate of 5.3%, but in the last four years, it has slowed to approximately 1.4%" (3765-town-meeting-booklet-including-warrant, line 338), and does not say which figure it grew. Rebuilding the Town’s own Subtotal State Aid line from its own revenue/expenditure worksheets — 23 consecutive budget years, FY2005 to FY2027, notes/findings/state-aid-budget-series.csv — gives 4.5660% for FY2013→FY2023 and 1.9144% for FY2023→FY2027. Neither published rate is reproducible from the Town’s own published series, so a rate quoted to Town Meeting cannot be checked against anything. — closes: the Cherry Sheets themselves (DLS form CS 1-ER for Lunenburg, one per year), or the Town Manager’s working file behind the FY2027 booklet’s revenue narrative.

• How much of the FY2027 enacted state-aid increase is receipts and how much is offsetting assessments

  model/finance.py grows a FY2027 base of $11,404,917 + $471,121. The first is the Governor’s GROSS cherry sheet Total Receipts, printed by the FY2027 Town Meeting booklet. The second is recorded in fy27-and-the-override.md as NET — "+$471,121 net (Ch.70, charter/choice receiving tuition, Smart Growth, UGGA, vets/elderly exemptions; offset by higher charter assessments)". A gross figure and a net figure are being added, so the gross FY2027 receipt estimate the projection actually grows is not established, and neither is how much of the increase the Town keeps. This archive holds no Cherry Sheet for any year, which is the only document that prints receipts and assessments side by side. — closes: the FY2027 Cherry Sheet for Lunenburg on the enacted budget (DLS form CS 1-ER), which states each receipt and each assessment separately.

• What Chapter 70 aid would actually do if enrollment fell

  MUCH NARROWER THAN IT WAS. DESE’s Chapter 70 trends workbook breaks the aid calculation into its named components, and for Lunenburg they show that this year’s aid is last year’s aid plus named increments — an identity that holds to the dollar in every one of the last seven years. In FY2026 the whole of the increase is the minimum aid increment, $240,450, which is exactly $150.00 for each of the 1,603 foundation pupils, and the foundation aid increment is zero. Over thirty-four years foundation enrollment fell in 17 and aid fell in 3 — FY1997, FY2010 and FY2011. DESE’s component columns show a reduction applied in FY2010 and FY2011; they start at FY2007, so nothing in this archive says why aid fell in FY1997. So the marginal aid attached to one foundation pupil is currently about $150 and not the $14,440 foundation budget per pupil. What is NOT established: that the Legislature sets the same minimum aid rate next year — it has ranged from $30 to $150 a pupil in the years published — nor what happens if a large enough enrollment fall moves the district off the minimum aid track, which no document here models. — closes: DESE’s preliminary Chapter 70 aid calculation for Lunenburg for the year in question, which states the foundation aid, minimum aid and hold-harmless components before the year is set.

• What period the school athletic family cap covers

  The $1,500 family cap was set by the School Committee on 26 February 2025 and the minutes state the amount and no period: "with a family cap of $1500". The cap it replaced did state one — the LHS Athletics FAQ prints "Total Cap per season=$475.00" — so the schedule has moved from a stated period to none. The district workbook obtained by records request carries the per-student rates on three worksheets named for the three seasons and carries no family cap in any cell, so the corroboration that dates the other rates does not reach this one. Read per season and read per year, $1,500 is a different bill for the same family. The arithmetic argues for the annual reading — with the published FY2027 ladder a family of five children playing one sport each in one season reaches $1,375, still under the cap — but that is an inference from the cap being unreachable, not a statement in any document. — closes: the district’s athletic fee schedule as published to families for 2026-27, which is the document the Superintendent’s August 2026 email to families attached or described.

• What the six fees the district’s payment portal sells actually cost

  The district’s RevTrak portal offers an Afterschool Activity Fee, a Chromebook Repair Fee, Extended Day & ELC, Field Trips, an LHS Parking Permit and a Primary Preschool Program. The portal renders in JavaScript and serves no amounts to a plain fetch, and no fee schedule for any of them appears in a budget book, a warrant or a set of minutes in this archive. So the existence of each fee is established and the amount of every one of them is not, and any total of what a Lunenburg family pays the schools in a year is a floor with six named items sitting above it. — closes: the district’s published schedule of student fees for the current year, listing every fee the portal charges with its amount.

• What a fourth child in one family pays for high school athletics

  The FY2027 schedule names a first, second and third child — $400, $300 and $225 — and stops. The FY2024 and FY2025 schedules did the same, and there the cap resolved it: the three published rates sum to exactly the $475 the FAQ prints as the cap, so a fourth child added nothing. That identity does not hold in FY2027, where the three rates sum to $925 against a $1,500 cap, so the fourth child is neither priced nor obviously free. A family of four children — the case residents ask about most often — therefore cannot be costed from anything published. — closes: the district’s athletic fee schedule as published to families for 2026-27, stating the rate for each additional child or the point at which the ladder stops.

• What FY2027 athletic fees are for anyone other than a full-paying high school family

  The only account this archive has of the 2026-27 athletic fees is a Superintendent’s email to families in August 2026 that we do not hold, reported as four figures: a first, second and third child and a family cap. FY2026 published more than that — a middle school rate, a reduced rate for high school and for middle school, and a sibling discount — and none of those has an FY2027 successor in any document here. So a middle school family, a family paying the reduced fee and a family whose fee is waived cannot be priced for the year now running. — closes: the district’s athletic fee schedule as published to families for 2026-27, in full, including the reduced and waived rates and the middle school schedule.

• How the $1,500 athletic family cap was arrived at, and over what period

  The School Committee minutes of 26 February 2025 record the decision and not the derivation: "A slideshow presentation of athletic user fees is presented ... with a family cap of $1500. ... all approve via roll call vote". A slideshow WAS presented, so the arithmetic exists somewhere; this archive does not hold it, and the only athletics presentation material here is two Athletic Advisory Council agendas from the preceding weeks. Minutes summarise, so the absence of arithmetic in them is not evidence that there was none — what is established is that no document we hold derives the figure or states the period it covers. The figure itself is unreachable per season by family size alone: five children at one sport each come to between $1,220 and $1,375 depending on how the sibling discount is carried past the third child, both under $1,500. — closes: the athletic user fee presentation given to the School Committee on 26 February 2025, which is the slideshow the minutes name.

• What the foundation budget pays per pupil for each kind of child

  DESE builds the foundation budget from per-pupil rates that differ by grade and by category, and it publishes the CATEGORIES for Lunenburg — FY2026 foundation enrollment of 1,603, of whom 54 are English learners (3.4%), 452 are low income (28.8%, low-income group 5) and none is vocational, at a wage adjustment factor of 1.000 — without publishing the rate attached to any of them in the same file. So which children leave a district changes the foundation effect and not only how many, and this archive can state that it matters and cannot size it. — closes: DESE’s FY26 foundation budget rate schedule, the table of per-pupil rates by grade and category and the low-income increment by decile group.

• Which Lunenburg budget lines the circuit breaker reimbursement offsets

  dese_circuit_breaker gives the reimbursement the district receives for high-cost special education placements, split into instruction and tuition against transportation, FY2006 to FY2026. It is real money and it is not in the general fund appropriation the town votes. Rule 11: a tuition line that fell because circuit breaker rose is not a line that got cheaper, and the budget document cannot tell the two apart. WHAT IS NOT ESTABLISHED is which lines it lands against, or whether the district budgets net of it — the workbook nets other lines down against expected receipts and says so in its own comments column, so this is not a hypothetical. — closes: the district’s circuit breaker revolving fund cashbook for the year, which shows what the receipt was spent on.

• How much of the SY2025 grant fall was ESSER, as against every other grant and revolving fund

  The meeting record names ESSER as the cause repeatedly and with figures — the district told the School Committee on 15 November 2023 “We have a loss of $650,000 in ESSER funds, this is much more significant than we expected”, and on 17 January 2024 put the loss at $600,000 beside a $500,000 health insurance rise. DESE reports one combined grants_revolving column, which holds ESSER, Title I and the other federal and state grants together with circuit breaker, school choice, athletics, food service and every other revolving fund. The SY2025 fall of $2,023,188 across matched functions therefore cannot be attributed to any one of them. The district's own stated ESSER figures and this fall are different quantities measured at different stages — a forward budget gap against a closed-year fund total — and this project does not difference one against the other. — closes: the district's end-of-year grant expenditure detail by grant code for FY2024 and FY2025, or Schedule 19 of its End of Year Financial Report as filed, which lists each grant separately.

• Why Chapter 70 aid moved by more than the components DESE publishes, in FY2010, FY2011, FY2012 and FY2019

  DESE’s Chapter 70 trends workbook prints the aid build-up as named increments — foundation aid, down payment aid, growth aid, target aid phase-in, minimum aid — and in fifteen of the nineteen differenced years they sum to the change in aid to the dollar. In four they do not. FY2010 and FY2011 carry a separate c70aidreduct column of −$92,416 and −$279,101, and only the first of the two accounts for its year; FY2012 moved +$24,149 with every increment printed as zero; FY2019 moved $13,365 more than its single increment. The workbook prints no residual line, so the difference has no name in the source. It matters because this page reads the components to say what the formula did, and in those four years the components do not add up to what it did. — closes: DESE’s Chapter 70 aid calculation worksheet for FY2010, FY2011, FY2012 and FY2019, which states every adjustment applied to a district between the prior year’s aid and the current year’s.

• What the student activity fee is for 2026-27, and what period the $70 covers

  The School Committee voted a universal student activity fee of $70, up from $55, on 7 May 2025 — the motion carried and the minute records the amount and no period. So whether a family pays $70 a child once a year, once a term, or once for each activity joined is not established, and those are a threefold difference on the same household. Nothing in this archive restates the fee for 2026-27 at all, and it is charged. — closes: the adopted student activity fee schedule for the 2026-27 school year, stating whether the fee is per student per year, per activity, or a base fee with activity fees on top, and any family cap or free/reduced provision.

• Whether the district charges anything for music, and what an instrument costs a family

  A resident told the School Committee on 26 February 2025 that "Instruments cost a lot, you need reeds and that can cost $50 a semester", and that band families "just aren’t paying anything directly to the school". That establishes music as a real household cost and establishes nothing about a district charge: no music fee, instrument rental rate or materials fee appears anywhere in this archive. A cost paid to a shop and a fee paid to the district are different things and neither is currently priced. — closes: the district’s music program fee schedule for 2026-27, or a statement that it charges no music fee and rents no instruments.

• What the processing fee is on every school payment made online

  The Superintendent’s email of 17 August 2026 tells families paying online that "RevTrak charges a small processing fee" and does not say what it is. It is added to every other fee a family pays through the portal, so it scales with the rest of the bill rather than sitting beside it, and a household paying athletics, bus and activity fees online pays it three times. "Small" is not a rate. — closes: the RevTrak merchant agreement or the portal’s own fee disclosure, stating whether the charge is a percentage, a flat amount per transaction, or both.

• What a family spends in the cafeteria when the standard meal is free

  Massachusetts funds universal free school meals and the district takes the programme, so the standard reimbursable meal is not billed to a household — that much is established and it is a real reduction in what families pay. What is outside the programme is not: a la carte items, snacks and second meals are charged, and no price list for them is held here. The School Lunch fund’s own revenue cannot separate the two. — closes: the district’s cafeteria a la carte price list for 2026-27.

• What a single-child family pays for the bus at the reduced rate in 2026-27

  The Superintendent’s May 2025 email states five bus tiers for FY2026 including $50 for a single-child family qualifying for a reduced fee. The August 2026 email for FY2027 states three tiers and does not restate that one. Whether the tier was dropped or simply omitted from the email is not established, and the difference decides whether a qualifying single-child household pays $50, $180, or nothing. — closes: the adopted transportation fee schedule for 2026-27, with every tier and the free/reduced provision, as approved by the School Committee.

• What Lunenburg receives for a child who chooses IN, and what the fall from 51 to 15 is worth

  DESE counts the children arriving in Lunenburg from other towns in every school year from SY2014 to SY2026: 51 at the start, 15 at the end, a fall of 71% while the number leaving stayed flat. The town’s own School Choice fund receipts fell over the ten years the two series share, from $220,058 in FY2014 to $116,976 in FY2023. Neither document states a RATE. The statutory base tuition is set by M.G.L. c.76 §12B, no document in this archive states it, and it is higher for special education; the arriving children also come by three different mechanisms — school choice, tuition waived by local agreement, and foster placement — which do not carry money the same way. So the two series can be published side by side and cannot be divided into each other: a per-student figure produced that way would be an assumption in a measurement’s clothes, and the money value of a 71% fall in arrivals is therefore not established anywhere on this site. — closes: the DLS Cherry Sheet CS 1-ER (estimated receipts) for Lunenburg, which prints School Choice Receiving Tuition as its own receipt line, for the years the DESE counts cover.

• Which fund and account each field-rental receipt from Lunenburg Youth Soccer was posted to, and whether the $15,000 turf contribution is inside or beside them

  The district’s records-request answer lists seven receipts, FY2024–FY2026, $39,073.50 in all, with one MUNIS reference; the town’s FY26 special-revenue report carries fund 1306 School Facilities Use Revolving and fund 1545 Artificial Turf Revolving as balances only — closes: the journal detail export for funds 1306 and 1545, FY2024–FY2026, the same report the archive holds for fund 1301

• What Parks & Recreation’s programme fees and beach-pass sales became on the town’s books — how much of the $38,955.50 of programme fees and $8,828.00 of passes MyRec recorded for FY2025 reached fund 1500 (Park Revolving) or the general fund, and when

  MyRec is the registration system, not the ledger: its FY2025 sales reports foot to their own totals but say nothing about deposits, refunds or the fund each fee was posted to, and the FY2025 special-revenue report is not in the archive — closes: the FY2025 special-revenue report and the journal detail for fund 1500, the same report the archive holds for fund 1301.

• What put $62,231 into the School Facilities Use fund (1306) in FY2025 — the balance went from $3,010.67 at 30 June 2024 to $65,241.82 at 30 June 2025, more than the fund has taken in in any of the thirteen years the annual reports print

  Rent alone cannot produce the rise: the best receipts year on record is $39,944. Either revenue arrived that is not field rent — a transfer, a reimbursement, an insurance recovery — or something that had been charged to the fund stopped being charged to it, and the two balances cannot tell those apart. FY2025 is the one year no special-revenue schedule is published for. THE TEST WHEN THE DATA ARRIVES: read FY2025 line by line, total the receipts that are field or building rent, and see whether the remainder is a transfer in or an expenditure that moved elsewhere — closes: the journal detail for fund 1306, FY2024–FY2026, and the FY2025 special-revenue report.

• Where do the other towns’ ski co-op reimbursements land when they are paid?

  Lunenburg bills Littleton, Narragansett and Leominster for their skiers’ shares — $9,736.57 for 2025-26 — and which account receives the money decides whether it offsets the athletics line, the athletic revolving fund, or general revenue. The invoice is a bill and does not show receipt — closes: the revenue detail for the account the checks are deposited to, or the athletic revolving fund cashbook for the season

• Would regionalising change what the state sends Lunenburg, and by how much?

  This is among the most common arguments made about the deficit and the archive cannot price the whole of it. Regional districts receive regional transportation reimbursement under c.71 s16C that single-town districts do not, and that part IS held: the DLS Cherry Sheet line, FY2010-FY2027, priced in the row 'What would regionalising actually be worth to Lunenburg...'. What is not held is how the Chapter 70 formula would treat a newly regionalised district, or the busing cost a regional district takes on against the reimbursement. Per-pupil SPENDING is held and is a different quantity: it says what a district spends, not what the state sends. — closes: DESE guidance on how the Chapter 70 formula treats a newly regionalised district, and a regional district's own budget stating transportation cost beside the reimbursement.

• What would regionalising actually be worth to Lunenburg, once the transport reimbursement is set against the transport costs it would also take on?

  NARROWED 26 September 2026, and half of it is now answered. The reimbursement side is measured: the DLS Cherry Sheet is in the archive, FY2010–FY2027, and Regional Transportation under M.G.L. c.71 s.16C is $88,183,608 statewide in FY2026 to 84 regional districts and nothing to any single-town district. North Middlesex draws $1,195,321, Ashburnham-Westminster $1,190,725, Groton-Dunstable $1,001,384. What is NOT answerable from any file is the other half: a regional district also INCURS the busing it is reimbursed for, and no published figure separates a district’s transport cost from its reimbursement — DESE’s twelve per-pupil categories carry no transport line at all. Nor is the counterfactual observable: what three towns would spend merged is not what other merged towns spend today. — closes: the regional districts’ own budgets, which state transportation as a line and the reimbursement against it; and a regional agreement, which sets how the net cost is apportioned.

• Which fund pays the teacher salaries a district charges to grants and revolving accounts?

  DESE’s function-code file splits spending into general fund and grants-plus-revolving, and the gap between comparable districts is large: Harvard charges $2,482 per pupil of TEACHER salaries to grants and revolving funds where Lunenburg charges $363 (FY2025, over total FTE pupils). That single line is the largest part, about half, of the difference in what the two draw from outside the appropriation. (Corrected 15 October 2026: this row said $4,964 and $726, each exactly double -- DESE's category subtotal row had been summed with its own line items.) The file does not say WHICH fund — a specific grant, school choice tuition received, a tuition agreement — and the answer decides whether it is something Lunenburg could pursue or a feature of one district’s circumstances. — closes: DESE’s End of Year Pupil and Financial Report for the district, which reports spending by individual fund.

• How do Ipswich, Woburn, Lenox, Medford, Canton, Norwood fund schools in the top quarter of spending while taxing homes in the lightest quarter?

  Of the 20 towns that spend among the most per pupil while asking the least of a home’s value, 6 fit none of the four routes this report can test: an unusual amount of state aid, many houses per child, houses worth a great deal each, or an unusually large business base. Three of them — Woburn, Canton, Norwood — carry business shares of roughly twice the median and fail only the exact 2x test. Whatever the rest are doing is the most directly useful thing on this page and nothing here identifies it. — closes: DLS Schedule A for each town, which reports every revenue source a municipality collects, so local receipts, enterprise funds and reserves can be compared rather than inferred.

• Which of the FY2021 receipts the Town collected are TAXES & EXCISES, which are FEES, and which are NON RECURRING REVENUE?

  The Treasurer's FY2021 receipts page (annual town report PDF p25) prints 100 line items in three free-form label/amount columns with NO column header, and a SUMMARY OF RECEIPTS block of nine categories footing to a GRAND TOTAL of $44,252,049.61. Six of the nine categories are defined by a heading the page prints above the lines and each foots exactly -- licenses and permits, state revenue, special assessments, fines and forfeits, investment income, transfers from other funds. The other three print no heading that separates them, and their 60 lines foot as one block to $31,988,239.19. A residual does fit -- the taxes/interest/demands lines plus prior year excise and prior year interest, LESS supplemental tax on new construction, is $31,200,742.70 to the cent -- but that is arithmetic choosing a grouping rather than the document stating one, and other partitions fit the same three printed totals equally well. So we can say what the town took in, and not how it classified 60 of the lines. -- closes: the Treasurer's own receipts workbook for FY2021, the file the page is printed from, which carries the category each line is coded to. Not published; a records request to the Town Treasurer would obtain it.

• How many Montachusett foundation-enrolment students did Harvard have in FY2018 — 7 or 8?

  The FY2017 annual town report prints the figure twice and the two disagree by one student. Page 144 is DESE's own FY18 Chapter 70 / Regional District Enrollment and Contributions by Member City or Town for LEA 832 and prints HARVARD at 7; page 148 is the district's COMMUNITY ASSESSMENTS table and prints its FOUNDATION ENROLLMENT at 8. The two agree on all 17 other towns and on all 18 Required Minimum Contributions to the dollar. Only page 148's column foots to the District Total of 1,496 that BOTH pages print; page 144's 18 rows sum to 1,495, although its own per-row Change identities all close. Foundation enrolment is the input Chapter 70 runs on, so a member town's count is load-bearing for every town's assessment — and neither page says which of its own figures is the error. — closes: DESE's published FY18 Chapter 70 regional district enrolment file for LEA 832 Montachusett, at doe.mass.edu, which is the source both pages reproduce.

• Which of the FY2020 receipts the Town collected are TAXES & EXCISES, which are FEES, and which are NON RECURRING REVENUE?

  The Treasurer's FY2020 receipts page (annual town report PDF p24) prints 100 line items in three free-form label/amount columns with NO column header, and a SUMMARY OF RECEIPTS block of nine categories footing to a GRAND TOTAL of $43,089,202.54. Both halves of the page were read off its own embedded bitmap on 2 October 2026 and both close to the cent: the 100 detail lines add to the GRAND TOTAL, and so, separately, do the nine summary lines. Six of the nine categories are defined by a heading the page prints above their lines and each foots exactly -- licenses and permits $331,254.75 (19 lines), state revenue $9,790,141.52 (12), special assessments $68,681.50 (3), fines and forfeits $37,182.61 (4), investment income $102,340.41 (1) and transfers from other funds $2,291,492.46 (1). The other three print no heading that separates them, and their 60 lines foot as ONE block to $30,468,109.29, the sum of the three printed category totals. A residual does fit -- moving SUPPL TAX-NEW CONSTRUCTION $10,057.81 out of taxes and $100,180.88 of RESCUE WAGON out of non-recurring would close all three -- but that is arithmetic choosing a grouping rather than the document stating one, and other partitions fit the same three printed totals equally well. So we can say what the town took in, and not how it classified 60 of the lines. This is the SAME gap already registered for FY2021 and it is a property of the form, not of one year. -- closes: the Treasurer's own receipts workbook for FY2020, the file the page is printed from, which carries the category each line is coded to. Not published; a records request to the Town Treasurer would obtain it.

• What was the Town owed in motor vehicle excise, user charges and betterments at 30 June 2016?

  THE FY2016 RECEIVABLES SUMMARY IS PRINTED CLIPPED, AND THE CLOSING COLUMNS ARE THE ONES CUT OFF. Page 47 of the FY2016 annual town report prints the property-tax half of the schedule with EIGHT columns -- FORWARD, COMMITTMENTS, ADJUSTMENTS, REFUNDS, PAYMENTS, ABATEMENTS, TRANSFER, BALANCES -- and every row closes on the identity those columns state. Page 48 (motor vehicle excise, twenty levy years) stops after PAYMENTS. Page 49 (deferred taxes, water and sewer user charges, utility liens, sewer/water/street/septic betterments, apportioned assessments) stops after REFUNDS, and its own title is cut mid-word -- FY2016 COLLECTION OF TAXE. Page 50 is blank but for its folio, so nothing continues. Both pages were rendered and looked at on 2 October 2026: the ruled table simply ends at the right-hand edge of the page. SO THERE IS NO PUBLISHED BALANCE for any of those accounts and no arithmetic on either page can be closed. The combined balance sheet on page 20 does not substitute: it prints one undivided EXCISES $243,852.65 against page 48's five columns netting $303,306.05, which are not the same quantity. This is NOT the same gap as the FY2013/FY2017/FY2020 receivables refusals, which are about an extractor failing on pages that ARE fully printed -- here the document does not carry the figures. — closes: the Collector's and Treasurer's own FY2016 receivables control report, or an un-clipped printing of the same worksheet, which the Town Accountant's system can reproduce.

• What else did the Town receive in FY2016 after CH 70 SCHOOL AID, and what did its receipts add up to?

  THE FY2016 RECEIPTS PAGE IS CUT OFF MID-LIST AND PRINTS NO TOTAL. FY 2016 RECEIPTS - JUNE 2016, PDF page 23 of the annual town report, is a two-column list of about 54 figures. The left column ends at RELEASE FEES in the middle of the FEES block; the right column ends at CH 70 SCHOOL AID in the middle of the STATE REVENUE block. Both stop at the same height on the page, which is what a print that ran out of room looks like. There is no SUMMARY OF RECEIPTS block and no GRAND TOTAL anywhere on it. The comparison is the Town's own later practice: the FY2020 and FY2021 receipts pages print about 100 detail lines AND a nine-category summary footing to a printed grand total, which is what let both of those years be read and checked. FY2016 prints neither, so the page cannot be footed, the receipts cannot be classified into the Town's own categories, and what the Town took in during FY2016 is not stated in this report. — closes: the Treasurer's FY2016 cash receipts report as the accounting system produces it, the same report whose later editions the FY2020 and FY2021 annual reports print in full.

• Which of the FY2011 receipts the Town collected are TAXES & EXCISES, which are FEES, and which are NON RECURRING REVENUE?

  Page 58 of the FY2011 annual town report prints 90 receipt lines under ten block headings and then a nine-category SUMMARY OF RECEIPTS footed to a GRAND TOTAL of $27,388,537.87. The summary ties to the cent, and every detail figure on the page is now transcribed, but the page never says which detail line belongs to which category and the headings are not the categories — ten blocks against nine categories, one block unheaded, and FEES naming both a block and a category without the two agreeing. Five categories can be reproduced exactly (LICENSES AND PERMITS, STATE REVENUE, SPECIAL ASSESSMENTS, INVESTMENT INCOME, TRANSFER FROM OTHER FUNDS); four cannot. TAXES & EXCISES comes out $23,947.19 OVER its printed total, FINES & FORFEITS $2,025.00 SHORT, and FEES ($502,520.65) and NON RECURRING REVENUE ($71,325.29) cannot be reconstructed by any grouping the page states. So the town’s own revenue mix for FY2011 — how much is tax, how much is fee, how much is one-off — cannot be read off the document that reports it. This is the same limit already registered for FY2016, FY2020 and FY2021 and it recurs because the format does. — closes: the Treasurer’s or Accountant’s FY2011 receipts detail as the accounting system prints it, with each receipt account carrying its revenue classification (a MUNIS revenue report by account), which is the only document that states the mapping rather than assuming it.

• What did each of the ~130 individual receipt lines on the FY2019 receipts page bring in?

  PDF page 30 of the FY2019 annual town report (printed folio 28), FY 2019 RECEIPTS - JUNE 2019, prints about 130 named receipt lines in three label/amount column pairs, and separately a nine-category SUMMARY OF RECEIPTS with a GRAND TOTAL. The SUMMARY is now read and closes to the printed GRAND TOTAL of $40,734,004.07 to the cent (sources/data/page-reads/fy2019-p30-receipts.csv). THE DETAIL IS NOT READ. The page is a ~150 dpi photograph with no text layer, and recognition finds only about 75 of the ~130 printed money figures, so the second identity the page states about itself -- that the detail lines also sum to the GRAND TOTAL -- is untested. Until it is, FY2019 has category subtotals and no line-level receipts, so questions like what the town took in from building permits or from ambulance billing cannot be answered for that year from this page. - closes: nothing from the Town is needed and no better scan is required; the page is legible. This closes by transcribing the three column pairs and footing them to the nine category subtotals already proven, which is about 130 figures against ten independent assertions.

• Why the general fund’s starting budget grew more slowly than the levy limit, FY2010 to FY2025

  Measured from the general fund history against the Division of Local Services’ certified levy limits, the whole starting budget grew more slowly each year, on average, than the levy limit did — even with the FY2025 override taken back out of the limit. Three readings fit: the rest of the revenue (state aid, local receipts, free cash) grew more slowly than the levy; the town levied below its limit; or money raised under the limit went outside the general fund operating budget, to capital articles or enterprise subsidies. They imply different things about how much room the levy has left. Measured in /analysis/department-budgets. — closes: the Tax Rate Recapitulation for each fiscal year, which states the levy actually raised, the excess levy capacity and every other source the budget was balanced with.

• Why does DESE's FY2024 figure for out-of-district tuition paid from funds other than the general fund exceed what the circuit breaker account paid by exactly the FY2024 circuit breaker payment?

  DESE's End of Year Financial Report reports FY2024 tuition (functions 9300 and 9400) paid from funds other than the general fund. The town's year-end report for the school special funds shows what the circuit breaker account (fund 2640) paid in tuition that year. The remainder between the two is, to the dollar, the circuit breaker payment DESE's reimbursement file lists for FY2024; in FY2023 and FY2025 the remainder is small. The figures are computed, not typed, in the report's payload under circuit_breaker.unexplained. It may be a coincidence, a second fund, or the receipt reported as an expenditure; nothing in this archive distinguishes them, so /analysis/special-education-costs shows the figure as 'other funds' and says it cannot explain it. Computed in scripts/build_sped_costs.py, which refuses to print the sentence if the equality stops holding. — closes: the district's FY2024 End of Year Financial Report as submitted to DESE (Schedule 1 by fund), or the school special funds year-end report for any fund other than 2640 that paid tuition in FY2024.

• Why local receipts come in above the town’s own estimate

  The state's free cash proof shows receipts above the town's estimate (its line CL#6) as one of the largest components of Lunenburg's free cash in every one of the five years it covers, and above a million dollars in four of them. On 28 May 2026 a Finance Committee member told the committee the town had not collected less than four million dollars in local receipts in ten years, against a current estimate of 3.535 million. Three readings fit the same line: an estimate set low on purpose as a buffer; receipts that grew faster than any forecast could know; or one-time receipts that should not be estimated at all. They imply different things about how much of each year's free cash is predictable, and so about how much the budget could count on. Measured in /analysis/sitting-on-money. — closes: the town's local receipts estimate against actual collections, by category, for each year -- the local receipts page of each year's Tax Rate Recapitulation beside DLS Schedule A.

• How much of the special education circuit breaker account's balance is already committed to the next year's tuition?

  The account (fund 2640) pays out-of-district tuition every year, overrun or not, and its 30 June balance is computed on /analysis/special-education-costs and /analysis/sitting-on-money. Whether that balance is free to meet a mid-year surprise, or is already planned against next year's placements, is not published: the year-end ledgers show what was spent, not what was intended, and in one year the account carried a tuition budget while in the others it carried none. So a resident cannot tell how much of the balance is a cushion beside the new special education reserve. — closes: the district's circuit breaker spending plan for the year, or the budget narrative for the out-of-district tuition line stating how much it assumes the account will pay.

• Which school special revenue funds the period-13 special funds report leaves out

  The FY2023 annual town report's special revenue schedule prints $3,667,636 received and $4,333,280 spent across every school special revenue fund; the MUNIS period-13 special funds report for the same year, delivered 6 October 2026, carries $2,440,883 and $2,962,716. The difference is funds that report does not carry -- among them, the annual report's FY22 ESSER II #115, FY22 ARP Idea #252 and FY22 Idea ARP #264 rows have no fund of that name or number in the MUNIS report. And the FY2026 report books $59,333 of grant revenue against $557,427 of grant spending. A reimbursement booked after the close, a grant receipt held in a fund outside the report, and both together all fit; the report cannot separate them, so whether the schools spend the grant money that comes in can be said only for the funds it carries. Measured in /analysis/spending-what-comes-in. — closes: the MUNIS period-13 revenue and expense report for EVERY school special revenue fund, FY2023 to FY2026, with each fund's opening and closing balance (or the trial balance for funds 2xxx at each close).

• Whether the school choice fund’s SCH. CHOICE BUS FEE account is where the student bus fees are booked

  In the school special funds year-end reports, fund 1308 (school choice) carries revenue account 437601, SCH. CHOICE BUS FEE. It took in nothing in FY2023 and FY2024, began receiving money in FY2025 -- the year the first student bus fees fell due -- and received again in FY2026, while the town general fund's STUDENTBUS revenue account shows nothing. That timing fits the fees and does not establish them: the account is named for school choice, and a fee charged only to school choice students would book the same way. Narrows 'Where bus fee receipts are booked' without closing it. — closes: the Town Accountant's definition of account 1308-437601, or the cash receipts detail behind it for FY2025 and FY2026 showing whether the deposits are the RevTrak and check payments for the student bus fee.

• What the athletic fee fund held at each 30 June, FY2021 to FY2026

  The school special funds year-end reports give fund 1301 (athletics) its receipts and its spending for FY2023 to FY2026, and no balance. Receipts fell short of spending in FY2023 and FY2024 and exceeded it in FY2025 and FY2026, from the year athletic bus costs began to be reclassified onto the general fund line; the district said in its FY2026 budget overview that the fund could not support those costs, and its FY2027 plan draws on the fund again to restore high school buses. The last balance this archive holds is the Finance Committee’s FY2020 special revenue fund workbook, at 30 June 2020. So whether the fund ran below zero in FY2024, and what it holds now, cannot be said; /analysis/transportation sets receipts beside spending and stops there. — closes: the Town Accountant’s fund 1301 balance at each 30 June from FY2021 to FY2026 (beginning balance, receipts, expenditures, ending balance), or the special revenue fund balance report for those years.

• How much of Monty Tech’s FY2026 spending is outside its operating budget, and from which funds

  The district’s FY 2027 Budget Summary sets out its operating (general fund) budget for FY2026 and FY2027 and the five sources that pay for it: Chapter 70, the member towns’ assessments, state transportation aid, its Regional Transportation Fund and its Excess & Deficiency reserve. Grants and revolving funds are not in it. DESE’s expenditures-by-function file DOES split the district’s total by fund, general fund against grants and revolving, but only through FY2025, and that total, which DESE titles Expenditure (Admin, Instruction, Operations), carries no capital or debt service row. So the operating budget on /enrolment/monty-tech-funding and DESE’s per-pupil figure are different years and different scopes, and nothing here reconciles them. — closes: DESE’s End of Year Financial Report for LEA 08320000 for FY2026, which splits spending by fund.

• What part of Lunenburg’s Monty Tech assessment the property tax levy pays, rather than state aid or other receipts

  A member town pays its regional assessment out of one budget that the levy, Chapter 70 and local receipts fund together, and no document assigns one revenue to one appropriation. So /homes-and-taxes/regional-school-share can only ESTIMATE the assessment’s weight in an average bill, in proportion to its share of the levy (our arithmetic on DLS values and rates by class). For the other seventeen member towns only the state-required minimum is held, not the whole assessment. — closes: nothing assigns it; the town’s tax recapitulation sheet sets every appropriation against the levy and every receipt, and each member town’s voted Monty Tech appropriation would replace the minimum.

money out

• School share of the pension

  inside WRRS · town and school staff together, no published split

• Debt service by project

  two accounts for ALL town borrowing · school buildings cannot be separated

• What the capital transfers bought

  two transfer lines · no project detail

• What any special revenue fund bought

  no expense report exists for these funds · purpose is presumed, never observed

• School share of the town’s health insurance before FY2026

  The town’s insurance department prints one undivided Health Insurance CH 32B line in fifteen of the sixteen annual reports · a school share is named only in the FY2023 report and in the FY2026 ledger, so no year between them can be split. — closes: The town’s Chapter 32B enrollment schedule, which counts insured employees and retirees by department and would split every year rather than the two that happened to be printed.

• School share of Medicare, active health insurance and the insurance reserves

  Five of the seven FY2026 insurance accounts — active health insurance, Medicare, life insurance, insurance cost control and Public Employee Committee expenses — carry nothing in their names that says which side of the town they are for, so the school portion of them is not separable in any year. — closes: The town’s Chapter 32B enrollment schedule, which counts insured employees and retirees by department.

• What the four ADJ EXP journal entries moved, and from where

  $254,121.18 — 65% of everything that reached the athletics revolving fund in FY2025 — arrived on four general-journal entries whose entire description is an expense adjustment made per a memo. An entry raising cash in a fund and labelled as an expense adjustment fits at least three readings: expenses charged here and later moved elsewhere, a transfer in from another fund, or a correction of charges posted to the wrong account. Those are different facts about the world and identical facts on the page, and the same shape of entry could be adjusting any fund in the town without either budget document showing a trace. — closes: the five memos the ledger names, dated 08/12/24, 01/30/25, 05/02/25, 07/02/25 and 08/20/2025.

• Whether a general fund athletics line is net of the revolving fund

  In FY2024 the town’s comparable athletics lines came to $124,301 against $351,643 the district’s own sport-by-sport workbook totals for the same categories — so the appropriation covered 35% of the cost. That is measurable for FY2024 and FY2025 only, because those are the only years anybody published both halves. For every other year, and every other programme, a line that rose because the cost rose looks exactly like a line that rose because a fund stopped paying part of it. — closes: the MUNIS Account_Detail export for the general fund athletics orgs S3066672 and S3066671, all object codes.

• What the PEG Access fund spent in FY2020 and in FY2024

  In both years the annual town report prints two totals for the same year's PEG expenses and they differ. FY2020, both tables on printed page 75: FY20 Line Item Expenses by Percentage foots to Total $116,933.22 and FY20 REVENUE vs EXPENSES prints Expenses $116,933.54 -- 32 cents apart. FY2024: the statement on printed page 63 prints Expenses $174,150 and the line-item table on printed page 64 foots to Total $174,151.16, and since that statement is printed in whole dollars throughout, rounding 174,151.16 gives 174,151 rather than 174,150. Each table foots to its own printed total exactly, so neither is internally wrong; they disagree with each other. FY2017 is a third case of the same shape: its ten printed expense lines add to $74,665.74 against a printed TOTAL of $74,215.73, a difference of $450.01, of which $450.00 is the Vendor Expenses line that the facing pie chart also omits. WHAT IS NOT ESTABLISHED: which figure is the year's expenditure in any of the three years. -- closes: the MUNIS year-to-date expense report (glytdbud) for the PEG Access and Cable Related Enterprise Fund for FY2017, FY2020 and FY2024, the same report the town already runs for the general fund.

• What the schools actually spent, per line, in any year before FY2026

  NARROWED 5 October 2026. the Finance Committee files received by records request on 4 October 2026 (sources/budget-workbooks/finance-committee/, see its PROVENANCE note) include the school department's FY2024 MUNIS year-to-date report at PERIOD 13 (glytdbud, FOR 2024 13, generated 08/07/2024, every school fund), which answers FY2024 by ledger, and a general fund budget-vs-actual history, FY2010-FY2025 by account, assembled from MUNIS into one workbook. FY2025 in that workbook is part-year. The district publishes a column headed ACTUAL in its forward budget books, and twelve years of it load as budget_figure where stage='restated' — the stage was called actual until 7 September 2026, and the rename is the point. Those are RESTATEMENTS — a prior year re-presented inside a document written by the party that spent it — and document-basis.csv classifies all sixteen source documents behind them as restatement (14) or forward (2). None is a ledger. The archive holds ten ledger documents in total, and the only ones reaching school general-fund spending are one quarter of FY23 and FY2026 period 12. So the question 'did the schools spend what they budgeted' cannot be settled before FY2026 from anything published: both sides of the comparison come out of the same budget book, and a restatement cannot show money moved between lines mid-year because approved transfers do not appear in a budget document. — closes: the MUNIS year-end expense report (glytdbud, period 13) for the school department for each year wanted, which is the same report we already hold for FY2026.

• What any individual sport actually costs

  Three documents state per-sport costs and they disagree — see “Which of the three published per-sport athletics cost figures is correct”, which supersedes this row's original claim that the district's own athletics workbook was the only place per-sport costs appear. The district has publicly declined to stand behind the figures: School Committee minutes of 24 June 2026 record that community members 'have asked straightforward questions regarding the actual cost of middle school athletics' and that 'the answers have been inconsistent'. The one independent record we hold does not reconcile to the workbook — in FY2024 the revolving fund paid out $317,004 and the general fund carried $285,281 against a stated workbook cost of $351,643, and nothing published explains the difference. And the fund's own disbursements cannot close it either: each carries a warrant number and nothing else, with no vendor, no description, and not one mention of any sport across three years of transactions. So participations, fees, and the fund's total in and out are established; what a sport costs is not. — closes: the accounts-payable detail behind the athletics warrants — vendor and description per payment — which the town's accounting system holds and does not publish.

• Why a budget line came in under plan, as against by how much

  A line under plan can be a post left vacant, a service cut, a grant that covered the cost instead, money transferred to another line, or a line over-budgeted from the start. All five produce the same smaller number in the spent column. The accounting system narrows it — it records mid-year transfers, so money moved can be distinguished from money not spent, and we hold that for FY2026 — but it cannot separate a vacancy from a cancelled service from a funding switch. Every published explanation of a Lunenburg school surplus is therefore a hypothesis about a measurement, and this project must not restate one as the measurement. — closes: payroll detail by position and date for the year in question, which separates a vacancy from every other cause, plus the grant award records for the same year.

• Whether a school line that stops appearing in the district’s book stopped being funded

  A line can leave the book because the money stopped, or because the book changed shape. Both look identical: an absence. Between FY2022 and FY2023 the restated series moves from one district budget book to another, the number of printed lines falls by about a third, and the reported total RISES — so the lines that stop appearing at that boundary cannot have taken their money with them. At least five of the disappearances are established renames, two lines differing only in spelling and carrying the identical value in every year they share (Maint. of Buildings/Contrcted Services and Maint. of Buildings/Contracted Services). Nothing published maps the older line list onto the newer one, so for any individual line, absence is a fact about the document and not about the service. — closes: the district’s own crosswalk from the FY2022 budget-book line list to the FY2023 presentation, or the MUNIS year-end expense report (glytdbud, period 13) for the school department for those years, which is keyed on accounts rather than on printed line names.

• Which school lines stopped being funded in FY2023, FY2024 or FY2025

  The restated figures for those years come from the FY27 budget projection workbooks, which print a shorter line list and almost no zeros. Not one line in them goes from funded in one year to a printed zero in the next — so the measurement this project can make for FY2015 through FY2022 has no evidence at all for the three most recent years. That is a property of which document restates which year, not a finding that nothing stopped. — closes: a district budget book for FY2026 or later printing the full line list with prior-year columns, in the same shape as fy24-approved-budget.

• What Lunenburg pays in school choice sending tuition, and for how many children

  The town pays tuition to every district a Lunenburg child chooses into, and that money is assessed rather than appropriated: it sits inside the single State Assessments (Outside Operating FY27) line of $1,069,349 in the FY2027 Town Meeting booklet, which is never broken out. The archive now holds DESE’s count of the children — 58 leaving under school choice in SY2026 — and still not one figure for what they cost. The statutory base rate is $5,000 a child under M.G.L. c.76 §12B and is higher for special education, and no document in this archive states either, so every scenario of students transferring out is priced on an assumption. — closes: DLS Cherry Sheet CS 1-EB (estimated charges) for Lunenburg, which prints the school choice sending assessment on its own line.

• How much of a school budget stops being spent when a student leaves

  A student leaving removes revenue at once and in full. It removes cost only where a staffing or programme decision is taken, and nothing published says at what enrollment a class section is removed, a bus route is dropped or a teacher is not replaced. So the avoidable share of a per-pupil figure is an assumption in every scenario on this site, and it is the input the answer is most sensitive to: at the FY2027 appropriation per pupil, the difference between assuming none of it is avoidable and assuming all of it is is larger than the entire revenue effect being modelled. — closes: the district’s own staffing and section plan by grade — the document behind the School Committee’s 16 April 2025 discussion of "each grade, their size, and potential seats" — which states the class sizes at which a section is added or removed.

• Which of the three published per-sport athletics cost figures is correct

  NARROWED 5 October 2026: one of the three is now traced to its workbook. The Finance Committee's files hold Athletics Costs (1).xlsx, whose FY24 Programmatic Cost column (sheet ALL, rows 2-26) sums to 275,947.63 -- the published 'Athletic Program Costs by Sport' total -- once cell K26 is counted: Girls Ice Hockey, typed as the TEXT *1100, an estimate. The workbook's own printed total, 255,042.35, comes from =SUM(K3:K26), which leaves out both row 2 (Football, 19,805.28) and that text cell. So that figure was built from these rows, an estimate included; it does not say which of the three is correct. Three documents the district produced state three different costs for the same team in the same year, FY2024. Across all 25 teams they total $185,355.62 (the Athletic Program Funding Overview deck of 1 May 2024), $275,947.63 (“Athletic Program Costs by Sport”, published with the FY26 budget materials) and $349,145.39 (the Total Expenses column of the district’s own by-sport workbook, obtained by records request) — a spread of $163,789.77, 88.4%. Per team it is far wider: Boys' Lacrosse is stated at $612.79 and at $14,189.18, 2215.5% apart, and not one of the 25 teams carries the same figure in every column. The three may be answering different questions — a programmatic cost, an all-in cost and a season’s booked expenses would all differ legitimately — but no document we hold says which each column is, and nothing published reconciles them. So a resident choosing which sport to give up is choosing between figures that disagree by a multiple. — closes: a reconciliation from the district of its own three per-sport cost statements, sport by sport, naming what each column counts.

• What folding one athletics team would actually stop the town spending

  A per-sport figure is not a saving. The district’s own workbook divides district-wide costs into thirds and pushes them into season totals — the notes column headed “Costs for all 3 Seasons” reads, at AX6, “MIAA 1/3 4550. -1516.67”, and at AY10 “MIDWACH - 1/3 7000 - 2333.33” — so league membership, trainer supplies, CPR certification and the online forms subscription sit inside per-sport totals. Whether any of those actually falls when one team folds is untested: a league assessment, a trainer’s hours and an insurance premium may each scale with the number of teams or not at all. This app answers it structurally rather than empirically — the FY27 athletics lines are split into a coaching-and-equipment pool that follows the teams and an overhead that does not, and the published per-sport figures are used only as weights to spread the pool — which bounds the saving without establishing it. — closes: the district’s athletics budget for a year in which a team was actually dropped, line by line against the year before, or a quotation from the MIAA and MIDWACH showing how each assessment varies with the number of teams entered.

• Which budget line each dollar of grant and revolving money paid for

  The fund split now exists at DESE's function level and nowhere below it. Lunenburg SY2025 spent $3,044,898 of grants and revolving money against $26,986,344 of general fund across all functions. But the district's budget book prints one NET figure per printed line (rule 11), DESE prints one figure per function code, and no published document maps the second onto the first — so a line that rose because a grant ended still looks identical to a line that rose because the district grew. This is the budget-line half of Which fund pays which post, which is the same wall approached from the staffing side. — closes: DESE's End of Year Financial Report Schedule 1 as Lunenburg files it, which separates revenues and expenditures by source of funds; or the MUNIS year-to-date expense report run for the school grant and revolving funds rather than the general fund.

• Whether a general fund rise beside a grant fall is the same cost, or a different cost in the same function

  DESE's End of Year Financial Report splits every function code into general fund and grants/revolving, so a function where grants fell and the general fund rose in the same year can be MEASURED — 8 of them in Lunenburg in SY2025, grants -$1,127,647 against general fund +$1,136,013 (dese_function_expenditure). Sorting those functions into “swapped” and “reduced” is OUR label on the signs of two numbers, and the state publishes no such classification. A function code is a bucket: Teachers, Classroom covers every classroom teacher in four schools, and both funds can move inside it for reasons that have nothing to do with each other. So the split establishes that the funding mix moved and cannot establish that the town picked up the thing the grant was paying for. — closes: the district's grant award letters and end-of-grant expenditure reports, naming the posts and object codes each grant paid, set against the general fund payroll and object detail for the same function in the following year.

• What DESE’s per-pupil figure buys

  DESE publishes dollars over a pupil count, by function, for every district. It does not publish what those dollars bought: a Teachers line is not a class size, an Instructional Materials line is not a textbook count, and a Paraprofessional FTE is not a person or a caseload. So Lunenburg spending $226 a pupil on instructional materials, equipment and technology against a statewide median of $626 is a measured difference in money and not a measured difference in what children have. Rule 7’s standing form: dollars are not services. — closes: the district’s own inventory and purchasing records for the materials and technology line, which would say what was bought and when.

• Whether spending less per pupil is a choice or a constraint

  Lunenburg’s total per-pupil spending has been below the statewide first quartile in every one of the seventeen years DESE publishes, and its net school spending is a smaller multiple of the required minimum than the median district’s. Nothing in any of these documents distinguishes a town that decided to spend less from a town that could not raise more: the levy limit, two failed overrides, the district’s proposed budget and Town Meeting’s vote all resolve into the same single number, and the number is the outcome of all of them at once. This matters because the two readings imply opposite remedies. — closes: nothing published settles it; the nearest evidence is the district’s own level-service request against the appropriation voted, year by year, which the archive holds for FY2024 onward only.

• Whether Lunenburg spends the foundation budget the way the state assumes it will, category by category

  Chapter 70 builds its foundation budget from about eleven CATEGORIES — classroom teachers, administration, instructional leadership, materials, and so on — each with a per-pupil rate the state sets. That is the state saying what an adequate education costs and how it decomposes. Comparing it line for line against what the district actually spends on the same categories would show exactly where Lunenburg is above and below the state’s own model, and it is the single most useful unanswered question about the expense side. WE HOLD THE TOTAL AND NOT THE PARTS. dese-ch70-foundation-nss.xlsx carries required net school spending, actual net school spending and the foundation budget as single figures per district per year, and dese-ch70-key-factors.xlsx carries the aid calculation; neither breaks the foundation budget into its categories. So the comparison cannot be made from this archive at all, in any year. WHAT MUST NOT BE DONE INSTEAD: dividing the foundation total by DESE’s statewide category shares to estimate Lunenburg’s. That produces a number for every category and establishes nothing, because the shares are what the comparison is meant to test. — closes: DESE’s foundation budget detail workbook, which prints the per-pupil rates and the resulting category amounts for every district, published annually beside the Chapter 70 aid figures this project already holds.

• How the Monty Tech assessment splits between operating, transportation and capital

  A regional assessment is four parts, and the district names them: required minimum contribution set by the state, transportation and other operating above minimum net school spending, capital outlay, and bonds. The split is published for FY2024, FY2026 and FY2027 in the district’s own budget books, and for FY2023 it was itemised in prose at a Finance Committee meeting. Every other year holds only the total. That includes FY2025, and it includes all eleven annual-report years FY2011 to FY2022 — so for the whole of the series this project charts, the roughly 4-10% of the assessment sitting above the state minimum cannot be attributed to transportation, to capital or to debt. It matters because those three behave nothing alike: one tracks a busing contract, one is voted year by year, and one is fixed until it ends. — closes: the district’s annual assessment certification to each member town, which states the four components separately; the Town Accountant receives one every year and the district retains all of them.

• What Monty Tech spends, and on what

  Lunenburg is assessed $1,334,521 for Montachusett Regional Vocational Technical in FY2026 and that figure is what the town is BILLED, not what educating 101 foundation pupils costs. PART OF THIS ROW HAS CLOSED. DESE’s RADAR district comparison publishes an all-funds per-pupil total for LEA 08320000 for FY2009 to FY2025 — seventeen consecutive years, every one of them summing to the workbook’s own printed in-district total — so the district’s per-pupil spending CAN now be tracked over time and set beside Lunenburg Public Schools’ own, measured the same way, which is what /what-other-districts-spend does. WHAT REMAINS OPEN is the composition and the attribution. RADAR gives ten per-pupil function columns and no dollars, no fund split and nothing specific to Lunenburg’s children; the archive holds two of the district’s own budget books — the FY2024 hearing book and the FY2027 presentation — so a FY24-to-FY27 view by DESE function code exists in dollars, with no year before FY2024 and nothing for the years between the two books. So what the district spends on a Lunenburg child, as against on any child, is still not established. — closes: Monty Tech’s annual budget book for each year, presented to its own School Committee and to the Lunenburg Finance Committee every March, and DESE’s End of Year Financial Report for LEA 08320000, which gives all-funds spending by function in dollars after the year closes.

• What Lunenburg’s net school spending actually was in the two most recent years

  DESE publishes net school spending against the Chapter 70 requirement for thirty-three fiscal years, and the last two of them are BUDGETED rather than actual — the column carries a stage flag saying so. So the most recent MEASURED position Lunenburg has against the state is FY2024, and every statement about where the town sits NOW rests on a budget somebody voted rather than on money anybody spent. The two may not be differenced and are never drawn as one line, which is why /what-the-state-requires-us-to-spend carries them as separate series. — closes: DESE’s Chapter 70 district profile workbook for the year in which FY2025 moves into the actual column, published annually at educationtocareer.data.mass.gov; the district’s own End of Year Financial Report to DESE for FY2025 is the filing that produces it.

• How the district’s own accounts add up to the net school spending DESE reports

  NARROWED 5 October 2026 for FY2024. the Finance Committee files received by records request on 4 October 2026 (sources/budget-workbooks/finance-committee/, see its PROVENANCE note) include the district's FY24 End of Year Financial Report workbook, whose own Schedule 19 states net school spending as 22,496,198 (reports sheet, row 28) -- a working copy, so DESE's filed figure governs. Net school spending excludes debt service, capital, transportation and food service — stated in the town’s own record, to the Finance Committee on 20 March 2024: “This is the minimum spending every school district must spend not including debt services, capital transportation, meals.” Because of those exclusions no figure in the town’s appropriation, the district’s budget book or the MUNIS ledger equals the net school spending DESE reports, and nothing in this archive maps the district’s line items onto DESE’s definition. So a reader cannot get from a budget line they can see to the one number the state enforces. — closes: the district’s End of Year Financial Report as filed with DESE, whose schedules derive net school spending from the district’s own accounts line by line; requestable from the Superintendent’s office or from DESE.

• How much of a receiving district’s per-pupil spending Lunenburg actually pays

  177 Lunenburg children were educated outside Lunenburg Public Schools in FY2026 and DESE publishes what each receiving district spends for every pupil, all funds. It does not publish what LUNENBURG pays towards any of it, and the two are different quantities by construction: a receiving district’s per-pupil figure is its whole spending over its whole pupil count, paid for by its own levy, its own Chapter 70 aid, its grants and its revolving funds as well as by any tuition or assessment. Lunenburg’s side arrives as three unlike mechanisms — a member-town assessment for Montachusett, a school choice tuition set by statute, and a charter tuition set by formula — and none of them equals the receiving district’s per-pupil figure. So the honest statement is that the districts our children go to spend more for each pupil than Lunenburg does; what a Lunenburg child costs the town at each of them cannot be read off it. — closes: the Town’s cherry sheet assessment detail for the fiscal year, which states the charter tuition, school choice tuition and regional assessment by receiving district, set beside DESE’s End of Year Financial Report for each of those districts.

• What share of the schools’ grounds, custodial, heating and utility cost is athletics

  The town’s ledger codes every school account to a function, and athletics is function 3510 — 12 accounts, $518,334 appropriated in FY2026, which ties to the cent to the district’s own budget book. Custodians, heating, utilities, grounds and building maintenance are separate functions in the same printout, $2,035,523 revised across the whole school department in FY2026, and not one of those accounts carries the athletics program segment. The fields are mown, the gym is lit and heated and the lines are painted, so some of that $2,035,523 is a cost of school sports — but nothing published splits it out, no document states an allocation, and the athletics revolving fund’s cashbook does not name a field, a light or a custodian on any of its 277 postings. So every published athletics total, including this project’s, is a floor and not a total. — closes: a cost allocation from the district or the Town Accountant apportioning functions 4110, 4120, 4130, 4210 and 4220 across the programmes that use the buildings and grounds; failing that, the accounts-payable detail behind the grounds-maintenance warrants, which would at least name what was mown, lined or resurfaced and when.

• What each announced school position cut was worth

  Only the FY2027 documents print a dollar figure beside a position. The FY2026 list the School Committee approved names 39 positions and prints no amount against any of them; the FY2025 impact slides print an FTE and a consequence and no money. So the register can say what was cut and, for one cycle in five, what it saved — and a figure beside a post is a budget-line saving, net of grants and fees (rule 11), rather than what the post costs. — closes: the personnel cost worksheet behind each year’s approved school budget, from the Business Office.

• Where was the schools’ FY2024 employee insurance reported, if not in the school department’s End of Year Financial Report?

  NARROWED 5 October 2026, and possibly reframed. the Finance Committee files received by records request on 4 October 2026 (sources/budget-workbooks/finance-committee/, see its PROVENANCE note) include a general fund budget-vs-actual history that carries the school department's active health insurance inside the SCHOOL general fund in every year FY2010-FY2025: account 0100-3-300-5200-99-1-99-2-570001, FY2024 actual 3,212,078.37 (sheet School Only, row 396). A Budget Master note records '$1,522,236 SHIFTED TO SCHOOL DEPT.' at FY2003 to FY2004. Why the EOYR total differs is not established. DESE’s EOYR shows function 5200 Insurance for Active Employees at $485,010 in FY2024 against $3,286,872 in FY2023 and $3,511,566 in FY2025; a difference read across that year looks like a $3M rise and is a reporting gap closing. Whether the year was mis-filed or the town appropriated and reported the insurance outside the school department is not established. — closes: the district’s FY2024 EOYR as submitted, Schedule 1, and the town’s FY2024 appropriation for group insurance by department.

• What each of the town’s 500-odd accounting measures did before FY2026, line by line — the appropriation history of every department and the year-by-year activity of every fund

  The registry (fund-owners.csv) lists every measure the FY26 reports print, but only the special-revenue funds have a read, tied history (FY2011–FY2023); the appropriation lines’ history sits in the annual reports where the extract cannot yet say which printed column is which, and FY2024–FY2025 fund activity was never printed — closes: the year-to-date budget reports and special-revenue reports for FY2023–FY2025 from the Town Accountant.

• What Lunenburg’s Chapter 32B §19 Public Employee Committee agreement actually sets — the premium split, the opt-out payment, and the plan menu

  CLOSED 6 October 2026 for FY27–FY29. The memorandum of agreement (sources/contracts/pdf/pec-agreement-fy27-fy29.pdf, 5 October) sets the plan menu, in-network cost-sharing, the HSA contribution, the $96 Medicare Part B reimbursement and the splits — 75/25 for active plans and Managed Blue for Seniors, 50/50 for Medex — 1 July 2026 to 30 June 2029. Its two blank attachments arrived on 6 October from the Town’s Human Resources Director, who wrote that ""The Benefit Summaries are Attachment A, and the Opt-Out Policy is Attachment B"": four Blue Cross summaries of benefits for the four active plans (pec-fy27-attachment-a-1 to -4) and the FY27 opt-out guidelines (pec-fy27-attachment-b-opt-out-policy.pdf), which pay $3,000 for individual coverage, $6,000 for family and $3,000 for a change from family to individual. Still open: summaries for the two Medicare plans, registered separately. — closes: closed by the documents named.

• What the town’s 2020 post-65 Medicare transfer actually saved

  The town states it cut the active renewal 2% and reduced OPEB liability by $5–6M; both are figures in a budget presentation with no actuarial valuation behind them in the archive, and c.32B §20A means a valuation exists — closes: the OPEB valuation filed with PERAC for the year of the transfer.

• Whether the schools’ stated field and grounds costs match the books — about $126,000 of grounds work in FY2024 and $38,100 of outsourced mowing and field repair, paid, the Superintendent says, from the same revolving fund the field rent goes into

  The figures are the district’s own, read back from FY24 budget requests in an email of September 2026, and nothing in the archive tests them: fund 1306’s FY2024 expenditure is not published and the fund’s journal is not held — closes: the journal detail for fund 1306, FY2024–FY2026, and the FY2024 and FY2025 special-revenue reports.

• How much of each earlier year’s ski team cost was Lunenburg’s, and how much the other co-op towns’?

  The district’s by-sport workbook prints one figure for the ski team and it is the whole four-town co-op’s cost — $12,984 in FY2025, $10,960 in FY2024. The only invoice held is the 2025-26 one, where Lunenburg’s share was 21.4%. Nothing says any earlier year split the same way, or that the co-op had the same members — closes: the co-op invoices for the earlier seasons, which the athletic department issues and holds

• What each town department was appropriated and spent in FY2024 and FY2025

  NARROWED 5 October 2026 for FY2024. the Finance Committee files received by records request on 4 October 2026 (sources/budget-workbooks/finance-committee/, see its PROVENANCE note) include a general fund budget-vs-actual history for every department, FY2010-FY2025, by MUNIS account (original, transfers, revised, actual). Its FY2025 column is part-year, so FY2025 stays open. Special revenue funds are not in it. The town published a General Fund Appropriations / Summary & Classification of Accounts schedule in every annual report through FY2023 — every department, appropriated against expended, with a printed GRAND TOTAL — and STOPPED. Its own contents pages show it: the FY2023 Accountant section lists the schedule and the General Fund Revenue and Expenditures charts; FY2024 lists a balance sheet, receivables, special revenue, capital, enterprise, trust and undesignated fund balance, and no appropriations at all; FY2025 is shorter again. FY2024 still prints the warrant’s FY 2025 Omnibus Budget, which is what was VOTED and not what was spent; FY2025 prints no department figures in any form, only Article 10’s total of $49,374,465.00 split across five funding sources. So for two consecutive years there is no published answer to what any single department was given or used — the question this archive answers for FY2011 through FY2023. — closes: the Town Accountant’s general fund appropriation and expenditure report for FY2024 and FY2025, the same schedule the annual report printed through FY2023, or the MUNIS year-end expense report by department for those years.

• What does Lunenburg pay, per child, for each place a resident child is schooled outside the district’s own schools?

  DESE publishes WHERE Lunenburg children are enrolled — 177 outside the district in FY2026 — and nothing about what any of them costs. A count of children is not a tuition, a charter assessment or a share of a regional assessment, and the general fund budget shows the appropriation rather than the fund each payment leaves. So the two halves of the question a resident actually asks — how many leave, and what that costs us — cannot be joined. — closes: DESE’s End of Year Financial Report for Lunenburg, which separates tuition and assessment spending by fund and by receiving district.

• Which department carried forward the $362.00 the FY2017 appropriations schedule does not print?

  The FY2017 GENERAL FUND APPROPRIATIONS schedule, pages 24-28 of the annual town report, states twice that a $362.00 carry-forward exists and never prints the line. Its APPROPRIATED FORWARD column sums to $37,626,083.44 against a printed GRAND TOTAL of $37,626,445.44, and its footer legend fwd - forward from FY 2016 $637,981.43 exceeds the 26 fwd lines actually printed by the same $362.00 — while the srt and rft legend totals reproduce to the cent and the other three columns tie exactly. The subtotal for Police Fire Medical Expenses on page 25 is $5,362.00 over a single line item of $5,000.00, so the money belongs to that block; which prior-year appropriation it is a carry-forward OF is not stated anywhere in the document. The amount is trivial; the question is not, because the same schedule is the basis of every departmental growth rate this project measures and a carry-forward line changes which year a dollar belongs to. — closes: the Town Accountant's FY2017 year-end MUNIS encumbrance and carry-forward detail, which lists each forwarded balance by account.

• What did Lunenburg owe, by borrowing and by fiscal year, from FY2020 to FY2035?

  FY2019 PDF pages 49-52 print the whole forward debt repayment schedule -- about seventeen named borrowings against sixteen fiscal-year columns, each printed as PRINCIPAL / INTEREST / ADMIN FEE under a TOTAL row, on the order of 800 cells. It is fully self-proving (PRINCIPAL + INTEREST + ADMIN FEE = TOTAL, per issue per year, roughly 270 independent assertions). None of it has been read: debt-repayment-detail-refused.csv records 3 issue rows on p50 and 5 on p52 and no closing block, because the column header band at the top of each page sits inside the top 22.7% that the clipped render discarded. So the town's own forward debt service for sixteen years is published and unread. - closes: nothing from the Town is needed. The pages are legible at 150 dpi; this closes by reading them, after the document is re-OCR'd with the current renderer so the header band is visible. RESOLVED 2 October 2026: read in full, and the sizing in this row was wrong in every figure — it is FIFTEEN borrowings not seventeen, TWENTY-EIGHT fiscal years plus a TOTAL column not sixteen, 879 printed figures not ~800, and 213 issue-year identities not ~270. Measured off the pages. 198 of 213 close exactly and the 15 that miss by $1 are every one of them a Mass Water Pool trust line, which is displayed-dollar rounding and not a misread: bond principal, the column with no cents beneath it, ties exactly for 13 of 15 issues and the fifteen printed issue principals sum to $46,568,549 at ZERO tolerance.

• Why the school department's FY2024 actual spending differs between its period 13 ledger and the general fund history, on 12 of 265 lines

  Measured 5 October 2026 from the Finance Committee files: the period 13 report (glytdbud, generated 7 August 2024) and the FY2010-FY2025 general fund history agree to the cent on 253 of 265 school general-fund lines; 12 differ, by a net 243,636.29, the history higher -- led by a paraprofessional line and a clerical salary line, with some lines moving the other way. Entries posted to FY2024 after 7 August 2024 would produce this pattern; nothing here tests that. — closes: the Town Accountant's FY2024 journal entries posted after 7 August 2024 for fund 0100, department 300.

• What a level-service budget for each town department would have cost, year by year

  The general fund history the Finance Committee supplied holds each department’s original budget, transfers, revised budget and actual spending, FY2010–FY2025, and no request: nothing in it says what a department asked for or what keeping its service level would have cost. The schools present a level-service budget, and its difference from the appropriation is what is reported as the school deficit. For the town’s departments no equivalent figure is in this archive for any year before FY2027 — the year the Town Manager asked department heads for level-service budgets (Finance Committee minutes, 12 February 2026) and the override questions covered town and school alike — so whether they would have shown a gap in earlier years cannot be measured, and the difference between the two sides may be one of process rather than of cost. — closes: each town department’s budget request as submitted to the Town Manager, with the Town Manager’s recommendation beside it, for each fiscal year, held by the Town Manager’s office.

• What the money was that reached departments’ revised budgets outside any recorded transfer

  In the general fund history the revised budget differs from original plus transfers in and out on about one account-year in twenty, almost always upward, in every year FY2010–FY2025, and the workbook’s own Grand Total shows the same gap. By dollars, most of it is no larger than what the same account left unspent the year before, which is what a carried-forward encumbrance would look like; the rest fits a supplemental appropriation at a special Town Meeting. Neither is confirmed, and the difference matters: a carry-forward is last year’s money, a supplemental appropriation is new money. Measured in /analysis/department-budgets. — closes: the MUNIS budget amendment detail for each fiscal year, or the Town Accountant’s appropriations schedule with its APPROPRIATED FORWARD column, which the annual report printed through FY2023.

• Whether a town department’s budget growth is more service or a cost moved into it from another budget

  Several town departments’ starting budgets grew in single steps — general highway maintenance, facilities and grounds, public buildings and recycling each have one year in which the starting budget rose by more than a third — and a step that size fits a service being taken on, or a cost moved in from another line or fund, as well as it fits the same service costing more. The ledger records dollars by account and nothing about why an account moved, so a growth rate measured across such a step is not a trend. Measured in /analysis/department-budgets. — closes: the Town Manager’s budget message, or the Finance Committee’s report to Town Meeting, for each year in which a department’s starting budget stepped.

• Where did the $122,502.44 that moved out of the FY2025 school department mid-year go, and what posted to department 300 after the School Committee's 17 September 2025 surplus statement?

  The closed MUNIS ledger (sources/data/munis-school-ytd.csv, FY2025 period 13, department 300) carries transfers_adjustments as a net figure per account, with no counterparty -- so a line moved in cannot be matched to the line it moved out of. And the ledger's own $627,116.86 available at the close exceeds the $603,885.97 the district reported that day by $23,230.89, with no single account or pair of accounts equal to the gap, which is consistent with activity posted after 17 September 2025 but not established by anything held here. — closes: the FY2025 transfer schedule for department 300 with counterparties and authority, and the journal entries posted to department 300 after 17 September 2025.

• What does one out-of-district placement cost Lunenburg in a year?

  DESE publishes what the district spent on tuition to private schools and collaboratives (functions 9300 and 9400), by fund, and the town publishes how many children were placed on 1 March. Nothing in this archive prices a single placement. /analysis/special-education-costs divides the first by the second and labels the result our estimate; /what-special-education-costs refuses to, and both are right about what the quotient is not: it mixes residential, day and collaborative places, it counts a child placed for part of a year as either a whole year or nothing depending on one day, and transportation is in a different line. A reserve for an unplanned placement needs the price of one, and the only figure available is that estimate. — closes: the accounts-payable detail behind general fund and fund 2640 accounts in functions 9300 and 9400 -- vendor, invoice period and amount, no student names -- from the Town Accountant, or the district's tuition rate sheets for the year.

• How much of the $236,766.94 still encumbered in the FY2026 school general fund at period 13 will be paid, how much released to the surplus, and how much carried into FY2027?

  The FY2026 period-13 MUNIS report for department 300 (sources/data/munis-school-ytd.csv), run on 6 October 2026, shows $482,118.07 unspent and $236,766.94 still encumbered across 27 accounts -- the largest electricity ($62,362.13) and collaborative tuitions ($58,708.07). Between the period-12 run of 1 September 2026 and that run, spending did not move in any account and one encumbrance of $16.95 was released, so the FY26 school surplus is still anywhere from $482,118.07 to $718,885.01. The FY2025 report run the same day shows nothing encumbered. A ledger shows a commitment, not whether it will be paid, cancelled or carried forward, and which of those happens decides the surplus. /analysis/fy26-school-surplus. — closes: the FY2026 period-13 report for department 300 run again after the year-end close, and the Town Accountant's list of FY2026 encumbrances carried forward into FY2027, by account.

• Was the FY2026 school surplus thrift — did discretionary spending slow late in the year?

  A speaker at the School Committee on 7 October 2026 said, per our machine captions, "We were thrifty"; no posted minutes record a decision to hold FY2026 spending. The period-13 ledger shows where money was left (by our category table, 59% of it in supplies, upkeep and equipment) but only as year-end totals, so restraint, a postponed project and a line budgeted high look identical. — closes: the school general-fund journal (MUNIS GL detail) for department 300 by month, FY2023 to FY2026, so discretionary spending in the last quarter can be set against prior years.

• Why FY2026 private special-education tuition came in far under budget while collaborative tuition ran far over

  In the FY2026 period-13 ledger the general fund’s private-tuition line (9300) ended well under budget and its collaborative-tuition line (9400) well over, netting to a small surplus; the circuit breaker fund (2640) paid collaborative tuition that year and no private tuition, the reverse of FY2025. Fewer private placements, placements moved to collaboratives, and the line a payment was charged to all fit the same rows. — closes: the FY2026 out-of-district placement list by setting, with start dates and each tuition invoice and the fund it was paid from.

• What a fee-funded school programme costs that is not booked to its own fund

  A revolving fund that takes in more than it spends covers the costs BOOKED TO IT, and nothing more (rule 11). The ledger can test whether the general fund also carries a programme's costs only where both book to the same programme-specific DESE function code, and among the schools' fee and lunch funds that is true of athletics alone (function 3510, where the general fund booked most of the four years' athletics spending). Extended day, after-school activities, facilities rental and adult education book their spending to function codes that name no programme -- 2000, 4000, 5200 -- so whether the general fund pays any of their custodians, utilities, administration or benefits cannot be shown either way. 'Pays its own way' therefore cannot be stated for any of them beyond their own books. Measured in /analysis/spending-what-comes-in. — closes: the district's indirect-cost allocation for each revolving programme, or a cost study listing the general fund accounts that carry part of each programme's cost.

• What was paid from the school budget lines that spent against a $0 budget, FY2023–FY2026

  The closed MUNIS ledger (sources/data/munis-school-ytd.csv, report gf-school, period 13) gives one spent figure per account per year and nothing else: not the payee, not the warrant, not whether it was salary for a person or a coding correction. /analysis/spent-with-no-budget lists every general fund line that spent against a $0 voted budget; for every one of them, what the money bought is unknown. — closes: a MUNIS general ledger detail (journal) export for department 300, FY2023–FY2026, for those accounts, with the warrant or payroll reference on each entry.

• What the Dee Bus rates and fleet were in FY2025, so the 7.6% FY2026 increase can be split into price and buses

  The FY2026 budget's "7.6% line increase = $69,300" is exactly the new contract's first-year regular-route price ($976,500: 8 buses at $485 and 3 at $515 a day, 180 days) over the FY2025 budget line ($907,200). The new contract fixes every year's price per bus per day, so FY2027 and FY2028 are known; the FY2025 side is one number, $907,200, which fits eleven buses at an average $458.18 a day (all of the 7.6% is price) and ten at $504.00 (the increase is a bus, not price) equally well. A district presentation says 11 buses were in use; that is a statement, not a rate. Requested on 9 October 2026 and not delivered. See notes/findings/DEE-BUS-CONTRACT.md. — closes: the FY2023-FY2025 Dee Bus contract and its bid form (daily price per bus by size, number of buses), or the FY2025 Dee Bus invoices.

• How many athletic bus trips each sport took, and what each trip cost

  The district's sport-by-sport workbook gives transportation dollars per sport for FY2024 to FY2026 as sums of charges, and the posted schedules (ArbiterLive) list each team's away games and venues from FY2022. /analysis/transportation prices the schedules at the Dee Bus contract's per-mile and per-hour rates as our estimate, with the waiting hours an assumption from the bid form. What no document shows is which games actually got a bus, whether teams shared one, how long each bus waited, and what each charge on the sheet was. — closes: the Dee Bus athletic and field-trip invoices for FY2024 to FY2026, with date, team, miles and waiting hours per trip, from the Business Office

• How many vans, monitors and routes special education transportation pays for, and at what rates

  The general fund special education transportation line is the only published figure: voted and spent, by year. The district's FY2026 budget overview attributes the line's rise to a rate increase and to more vans and monitors, and the August 2026 route list for the regular buses excludes the special education vans. The van contract is not in this archive, so price, routes, children and monitors cannot be separated, and no cost per child transported can be computed. — closes: the district's contract with Van Pool Transportation for special education transportation and its rate schedule, with the number of vans, routes and monitors by year; from the Business Office by records request.

• What the athletic fee fund’s high school purchase-of-service account paid for in FY2026

  Account 1301-3-300-3510-06-0-00-2-531006, where athletic buses were charged before they moved to the general fund, books officials, ice time, dues and transportation together. In FY2026 -- the first year the general fund athletic transportation line was budgeted to carry the full cost of athletic buses -- it spent about as much as in FY2024, when it still carried most of them. Fits equally: officials, ice time or other costs rose; some buses are still charged to the fund; reclassifications landed in a different year. Nothing published separates them. — closes: the accounts-payable detail for fund 1301 account 531006, FY2024 to FY2026, by vendor and invoice, from the Town Accountant.

• Why FY2024’s regular-route bus spending was higher than FY2025’s

  The general fund regular-route transportation account spent more in FY2024 than in FY2025, and closed FY2024 past its revised budget even after a transfer in, while FY2025 spent its budget exactly. Both years were under the Dee Bus contract that preceded the current one, which this archive does not hold, so whether FY2024 carried an extra bus, extra routes, a price step that later reversed, or a cost booked in the wrong year cannot be told. A Finance Committee member asked for the bus contracts in March 2024. — closes: the Dee Bus contract in force for FY2024 and FY2025 with its rate schedule, and the FY2024 Dee Bus invoices (the contract was requested on 9 October 2026).

• Will the school accounts that closed FY2026 past their revised budgets be covered by year-end transfers, and from which lines?

  The FY2026 period-13 MUNIS report for department 300 (sources/data/munis-school-ytd.csv, run 6 October 2026) shows 57 accounts past their revised budgets by $1,201,433.81, after the ten year-end transfers the School Committee approved on 29 July 2026 ($148,364.31), whose minutes say more "may still be forthcoming". FY2025 closed with 4 such accounts; FY2023 with 70 and FY2024 with 52. The school appropriation is a bottom-line total, so this is not spending past the appropriation; it is whether the line budgets the next budget is built from say where the money goes. /analysis/fy26-school-surplus — closes: the FY2026 year-end line-item transfer schedule for department 300 as posted, by account, with the vote that authorised each.

• What does the FY2027 out-of-district tuition budget assume: which placements, for which months, and how much the circuit breaker will pay?

  The district's FY2027 workbook (budget-workbooks/fy27-proposals.xlsx) budgets $700,142 across private and collaborative tuition, the same figure in every scenario. The same two general-fund lines spent and committed $1,202,770.94 in FY2026 at period 13, and the circuit breaker fund (2640) paid $333,494.89 more. Placements ending, a larger circuit-breaker share, or FY2027 tuition paid in advance out of FY2026 money all fit, and the ledger cannot tell them apart. /analysis/fy26-school-surplus — closes: the FY2027 tuition worksheet (each placement anonymised: setting, rate, months, and the fund expected to pay it) and the service period of every FY2026 tuition invoice.

• Where did each FY2026 school line-item transfer come from, and where did it go?

  The MUNIS year-to-date report gives one NET transfer figure per account, so a line that gave and received money in the same year shows only the difference. The 29 July 2026 School Committee minutes record $11,000 moved out of natural-gas heating; the ledger shows heating up $11,024.22, net. The 5 November 2025 minutes record $13,500 moved to School Committee dues "to cover superintendent search invoice"; that line closed FY2026 with $13,689 unspent and had spent less than its original budget. /analysis/fy26-school-surplus — closes: the MUNIS budget transfer report for department 300, FY2026, gross, with each entry's date and counterparty account.

• Why has school electricity run past its voted budget in every year from FY2023 to FY2026?

  Account 0100-3-300-4130-99-1-74-2-521011 spent and committed $303,101.82 more than it was voted across FY2023-FY2026, and the FY2027 budget, $316,250, is below the $319,108.72 FY2026 had already paid. Natural-gas heating beside it was voted above its spending in FY2024-FY2026 and received transfers in every year. Usage, rates, load moving from gas to electricity, and the solar panels a Finance Committee member asked about on 26 February 2026 all fit. /analysis/fy26-school-surplus — closes: the school utility bills by building and month, with kilowatt-hours and any solar or net-metering credit.

• Whether the FY2025 spring section of the district’s by-sport athletics workbook is complete

  The district's sport-by-sport workbook prints $0.00 as its FY2025 spring season total, in cell AT25, while its spring rows sum to $18,242.50, 46.1% below FY2024's spring rows; spring is 58.9% of the whole fall in athletic transportation from FY2024 to FY2025. Outdoor track, baseball, softball and lacrosse all look cheaper in FY2025 on that sheet, and /analysis/transportation cannot say whether that is fewer or shorter trips or a section filled in before the season's bills arrived. Every other season total the workbook prints equals its own rows. — closes: the district's confirmation that the FY2025 spring transportation column is final, or Dee Bus's spring 2025 athletic invoices, from the Business Office.

• Which budget lines the School Committee's line item transfers moved, when the minutes record only the vote

  Of the transfers the School Committee approved in the minutes read so far (every set the archive holds, 6 September 2023 to 26 August 2026), some are recorded only as a vote to approve transfers "as outlined" or "as presented", or as one total for a batch, with no from-line, to-line or amount -- 24 June 2026 approved four transfers with no figure printed, and 4 October 2023 approved $127,786 as a single total. The minutes are the record of the VOTE; the lines are on the district's transfer sheet the committee is shown, and the archive holds that sheet for only three: 26 September 2025 (the athletic $20,000), 5-6 October 2026 (FY2027) and one from October 2022. The recordings cannot substitute: a caption model hears figures unreliably. And the town's ledger extracts held here carry no budget-transfer journal for the school department, so nothing can tie a vote to the accounts it moved. — closes: the line item transfer sheets presented at each School Committee meeting (in the meeting packets), or the MUNIS budget transfer journal for the school department, by fiscal year.

• Which games Lunenburg’s athletic teams played in the seasons their posted schedules do not cover

  /analysis/transportation builds its athletic bus model from the schedules the district posts on ArbiterLive, the scheduling site linked from its athletics page. ArbiterLive lists only what a coach posts: cheer posts no schedule at all, some teams post none in some years, and indoor and outdoor track list far fewer meets than the district's by-sport sheet has charges. A team with no listed game is not a team that took no bus, so the model is a floor for those teams and cannot be checked against the sheet for them. — closes: the athletic department's final game and meet schedule for each team, FY2022 to FY2026, with home and away and the venue, from the Athletic Director by records request.

• How the district built its FY2027 athletic transportation estimate

  On 26 August 2026 the Superintendent put restoring high school athletic buses at about $58,880, attributing the lower figure to scheduling competitions closer to Lunenburg, more home competitions, and estimating from actual schedules and prior-year spring costs. /analysis/transportation tests the first two reasons against the posted schedules and prices the schedules with the contract's rates, but most of its result is the hours a bus waits, which is an assumption. Without the worksheet behind the estimate, the two cannot be set side by side assumption for assumption. — closes: the worksheet or calculation behind the $58,880 FY2027 athletic transportation estimate (trips, miles and waiting hours by team), from the Superintendent's office.

• Which accounts Lunenburg's FY2025 health insurance figure on the state's Schedule A counts

  The state's figure for Lunenburg equals the town insurance department less Medicare and life insurance in every year FY2010-FY2023, and the whole department in FY2024 (asserted in scripts/build_health_insurance_topics.py). So any growth rate measured across FY2024 mixes a change in what was counted with a change in spending, and FY2025 cannot be tied because the ledger history's FY2025 actuals are part-year. — closes: the town's FY2025 Schedule A health insurance detail, set against the year-end ledger.

• How much of DESE’s general fund figure for Lunenburg the school budget paid, and how much other town budgets paid

  DESE’s per-pupil figure counts general fund spending on the schools whoever paid it: insurance for retired school staff (function 5250) is in it, and in FY2026 the town carries that insurance in its own insurance department (SCHRETHLTH), not in the school appropriation. DESE’s expenditures-by-function file has one GEN_FUND column and does not say which budget paid which line, so the school appropriation divided by pupils is not DESE’s figure a pupil, and the difference cannot be itemised from it. Found building /school-spending/per-pupil-line-by-line. — closes: the district’s End of Year Financial Report as filed with DESE, whose Schedule 1 reports expenditures by the school committee and, separately, expenditures by city or town (part B, row 1016 of DESE’s FY2025 form), for the same year.

• How much of Harvard’s higher spending a pupil comes from being a smaller district

  Harvard spends more a pupil than Lunenburg on DESE’s all-funds measure, and divides its spending among fewer pupils. A smaller district spreads costs that do not shrink with enrolment — a principal, a building, a superintendent’s office — over fewer pupils, so part of any per-pupil gap may be scale rather than programme, and DESE’s function-code file cannot say how much: it reports dollars by function, not which costs are fixed. The Bromfield shelf states this as a hypothesis and tests nothing. — closes: nothing published settles it; DESE’s staffing by job classification and salary (EPIMS) for both districts, school by school, would bound the share that is fixed posts.

people

• How many people the schools actually employ

  The town publishes per-school staff ROSTERS in every annual report, FY2011 to FY2025 — a list of names. A roster carries no FTE, so a 0.4 music teacher and a full-timer are one row each, and it is a point in time, undated within the year. A count of names the town printed is a real quantity and it is not a staffing level. AND THE DISTRICT’S OWN CONTACT SHEETS DO NOT AGREE WITH EACH OTHER, which this project did not know until it held them: its roll-up of every employee names 257 people, the five listings the office and the four schools keep name 45 the roll-up does not, and the roll-up names people no building lists — Primary leaves its own paraprofessionals off the sheet the school keeps. Some of that difference is one person written two ways (Skye Abreu on one, Skye Abrue on the other) and nothing published distinguishes a second spelling from a second person, because no sheet carries an employee identifier. So there is no count of district staff here that two documents support. — closes: a staffing report stating FTE by position and year.

• Which fund pays which post

  This is the load-bearing one. The rosters name people and the budget names dollars, and nothing published joins them to a funding source. So a paraprofessional line rising cannot be distinguished from a grant that was paying for those posts ending — which is exactly what the special education escalator rests on. — closes: DESE's End of Year Financial Report, which separates spending by fund.

• Whether a budgeted position was filled

  A budget line is money the town has to raise, not a person in a chair. A line funded all year for a post vacant half of it looks identical on the page to one that was filled. Bounded, not settled: the rosters show who was printed, which narrows the question without answering it. — closes: a filled-position list, or payroll by position rather than by name.

• How many children play sports

  Two town documents give 593 and 649 for the same year and count different things: one counts fee categories, the other counts participations by sport, and a participation is not a child — one student playing three seasons is three of them. Some workbook rows are negative, subtracting out-of-district athletes back out, so a season total is a net figure rather than a headcount of bodies. Nothing published is an unduplicated count. — closes: an unduplicated athlete roster, or a participation report that states whether it counts students or participations.

• How many children Chapter 70 is actually paid for

  The aid is calculated on a Foundation enrollment of 1,599 in DESE’s FY27 Chapter 70 summary. DESE publishes three other counts for the same district — Student Headcount, In-District FTE Pupils and Total FTE Pupils — and in the latest year published they are 1,563, 1,568.9 and 1,665.9. Four numbers, one agency, none of them equal, measured on different dates and counting different things. So every per-pupil figure on this project, including aid per pupil, is a division by a count whose definition has to be stated with it, and no published document reconciles the foundation count to the children in the buildings. — closes: DESE’s FY27 foundation enrollment worksheet for Lunenburg, which states the October 1 count by grade and category the formula was run on.

• What the children who leave the district cost, as against how many there are

  The COUNTS are published and this project now holds them: DESE’s enrollment of town residents by district (dataset vxt3-k35x) states, for every school year from SY2014 to SY2026, how many Lunenburg residents attend each other district and under which programme. SY2026: 58 children under School Choice, 22 at charter schools and 97 at Montachusett Regional as members rather than by choice. The companion file (8xyg-59b2) counts the children arriving the other way, 11 in SY2026. So "a fall in enrollment is not a count of departures" is no longer a reason to refuse the question — the departures are counted. What is NOT published anywhere in this archive is the MONEY: not one document states the tuition Lunenburg is assessed for those 58 children, or for the 22 at charters, and the two rates are set differently and are higher for special education. — closes: DLS Cherry Sheet CS 1-EB (estimated charges) for Lunenburg, which prints the school choice sending and charter tuition assessments on their own lines.

• Whether a rise in the paraprofessional line is more posts or a grant unwinding

  DESE now publishes the split at the level of its own FUNCTION CODES, SY2009-SY2025: Lunenburg's function 2330, Paraprofessionals, is $1,338,477 of general fund beside $478,097 of grants and revolving money in SY2025 (dese_function_expenditure). That BOUNDS the question and does not settle it. A function code aggregates many posts and many funding sources; it is DESE's measure of spending rather than the town's appropriation; and nothing in it says which post any dollar paid, or whether the same posts were general-funded the year before. The in-district special education escalator still rests on a line that cannot be distinguished from grant money unwinding. — closes: the district's grant award and payroll charge detail for the paraprofessional posts, by fund and fiscal year.

• What a DESE special education FTE and a district special education budget line have in common

  DESE's dese_sped_indicator publishes special education paraprofessionals per 100 students with disabilities; times the published count that implies about 18 FTE for SY2026 against a district paraprofessional line that has roughly doubled since SY2019, which works out at some $69,000 per FTE. A paraprofessional does not cost $69,000, so the two are not measurements of the same population. The EPIMS Data Handbook gives half the reason: it defines FTE as *“the percent of workday staff are involved in an assignment”* — per ASSIGNMENT, not per person — while a budget line pays whole salaries. The other half is vocabulary: the district calls a line *Special Ed Paraprofessionals* and DESE has its own definition of a paraprofessional on the return, and neither is wrong. WHAT IS ESTABLISHED is that the two figures cannot both describe the same set of people. WHAT IS NOT is which is narrower, or by how much, or whether the line carries anything besides paraprofessionals. — closes: the district’s own crosswalk between its budget line names and what it reports to the state on EPIMS, which may exist nowhere in writing and would then be a records request rather than a download.

• Why DESE special education teacher FTE falls from 18.5 to 2.0 while total teacher FTE holds flat

  dese_teacher_program_area shows Lunenburg’s special education teacher FTE at 18.5 in SY2008 and 2.0 in SY2026 — a 90% collapse — while general education FTE holds between 100 and 113 and the district total is flat. A district that had lost nine tenths of its special education teachers would show it everywhere else in these files and does not. Teachers being RECODED from special education to general education produces exactly this signature. WHAT IS NOT ESTABLISHED is that any recoding happened: a change in how FTE is apportioned when one person works across categories, a change in DESE’s program-area definitions, and a real staffing change all fit the same series. Rule 6, and nothing in the archive tests it. — closes: EPIMS work assignment records for Lunenburg aggregated by job classification and program area by year, which would show whether the same people were reported under a different code.

• How many children Lunenburg has with disabilities, given that DESE publishes three different counts

  NARROWED, 8 September 2026, and the narrowing is itself the finding. The denominator behind the staffing ratios in yamx-769q is NOT unexplained: it equals the published IN-DISTRICT count exactly, in all six years the file covers, and the difference between it and the count printed beside it is the published out-of-district count to the child. Computed on every build by scripts/build_special_education.py, which fails rather than publishing the page if it stops reconciling. WHAT SURVIVES is a THIRD count: dese_sped_movement reports 10 to 26 fewer children on an IEP than dese_sped_program in every overlapping year, on a different enrollment figure, and nothing published states the census date or the inclusion rule behind either. — closes: DESE’s SIMS submission detail for Lunenburg stating the census date and the inclusion rule behind each published count.

• Which of DESE’s special education staff figures can be read at all

  Every row of the SPECIAL EDUCATION STAFF table in yamx-769q prints an FTE, a count of students and a rate per 100 of them, so the rate has to reproduce from the other two. Recomputed for Lunenburg on every build: paraprofessionals reproduce in 8 of 8 years, teachers in 1 of 8, service FTE in 0 of 7, support FTE in 0 of 6. So three of the four rows do not describe what their own column headings say, and /how-many-students-are-on-an-iep publishes the paraprofessional row and refuses to publish the others as figures. WHAT IS NOT ESTABLISHED is which of the three columns is the one that is wrong. — closes: DESE’s data dictionary for dataset yamx-769q, or the EPIMS aggregation behind it, stating what each column is.

• How many of the children who leave Lunenburg under school choice have an IEP

  DESE publishes where every town’s resident children are enrolled, by district and by programme — 58 leaving under school choice in SY2026, 22 at charter schools, 97 at Montachusett Regional as members — and publishes NO disability status with it. So a special education student who transfers out and a student with no plan who transfers out are one row. This is why /where-students-go-instead is a separate report from the other three special education pages rather than a section inside one: merged, “58 children left” becomes “58 special education children left” in a single retelling. — closes: DESE’s SIMS detail crossing enrollment reason with IEP status, which is not published at town level at all.

• How many out-of-district placements are children who arrived already placed

  The trajectory file measures the route through the district’s own settings, and the district told the School Committee on 4 February 2026 that “we have also had students move into the district that require out of district placements”. Such a child enters the town’s placement count without ever appearing in a cohort followed from an in-district setting, so the count and the route measure overlapping but different populations. The quote is evidence that it happens; nothing establishes how often, and the two published counts of placed children — the town’s on 1 March and DESE’s — agree in only 2 of the 5 years they overlap. — closes: the district’s own record of placements by origin, showing how many were referred from an in-district setting and how many arrived already placed.

• Which grades Lunenburg children leave the district in

  DESE publishes where a town’s resident children go to school and by which route — school choice, charter, regional vocational — by receiving district and by year, FY2014 to FY2026. It does not publish that by GRADE. dese_town_enrollment carries no grade column, and the grade counts that do exist (dese_enrollment) are headcounts INSIDE a district rather than an outflow from a town, so they cannot be differenced to recover it. WHAT THIS BLOCKS: the question everyone actually asks, which is whether families leave at a transition year — entering grade 9, or entering grade 7 — or steadily throughout. Those imply completely different things about what the district could change, and the published data cannot tell them apart. IT IS NOT: an absence of evidence that transitions matter. It is an absence of evidence either way, and the plausible story (that choice-outs cluster at grade 9 when families pick a high school) is a hypothesis nothing here tests. — closes: DESE SIMS Report 7 or the district’s own October 1 enrollment submission broken out by grade and residence, which the district files annually and does not publish; or the receiving districts’ school choice enrollment by sending town and grade.

• How many Lunenburg applicants Monty Tech turns away

  Lunenburg’s count at Monty Tech has risen in nine of the last twelve years, and a count is the outcome of two things at once: families applying, and Monty Tech admitting. The lottery exists precisely because more apply than there are places. Asked at the Finance Committee on 5 February 2026 how many Lunenburg students were turned away, the district’s business manager said she was unsure of the exact number. The Class of 2030 lottery result in this archive gives one cohort’s admissions — 24 Lunenburg students of 365 places — and its per-applicant pages are OCR of an image PDF and cannot be counted from. So a rising count cannot be read as rising demand, and the two have opposite implications for what the town should expect next. — closes: Monty Tech’s admissions summary by sending town and year — applications received, offers made, waitlisted — which the district produces for its own School Committee.

• Why DESE’s foundation enrollment for Lunenburg at Monty Tech differs from its own October headcount

  Two DESE counts of the same thing disagree in every year. The Chapter 70 foundation enrollment used to compute the assessment — derived on /monty-tech by subtracting the Lunenburg district’s from the town’s — runs between eight below and fifteen above the October headcount of Lunenburg residents attending Montachusett Regional in DESE’s own residents-by-district file, and the gap has no consistent sign or size. One documented mechanism puts children into the assessment who are not at Monty Tech: the district’s business director told the Finance Committee in March 2022 that four Lunenburg students attending trade programmes at Leominster and Nashoba “become part of the Monty Tech assessment”. That accounts for a difference in one direction and not for the years running the other way. Until it is reconciled, no per-student figure on this page can be quoted without saying which count is the denominator. — closes: DESE’s foundation enrollment detail for the Chapter 70 calculation, which states the collection date and the categories included, published beside the key factors workbook this project already holds.

• How many school choice seats Lunenburg opened, by grade and by year

  Children arriving from other towns fell from 51 in SY2014 to 15 in SY2026. Two readings fit that equally well and the published record separates them in one year only. A receiving district DECIDES each year how many school choice seats to open and at which grades — the School Committee minutes record that decision being taken on 22 January 2025 (elementary seats against high school only, tabled until the budget was settled), on 16 April 2025 (options beyond grade 9, moved contingent on free cash passing) and on 4 February 2026 (“We are looking to only add school choice options to the high school”). So fewer arrivals is consistent with fewer seats offered and equally consistent with fewer families applying. The Superintendent stated both counts to the Finance Committee on 26 February 2026 — 24 seats opened, 5 students applied — which points at applications for that one year and cannot speak for the other twelve. — closes: the School Committee’s annual school choice vote, seats by grade, for every year from SY2014 onward; the minutes hold it in pieces and nobody has compiled it as a series.

• Whether a position the district announced it would cut was removed

  The district publishes a cut list most budget cycles — 158 named reductions across five cycles in this archive — and nothing published says what became of any of them. The only independent series that reaches any of it is DESE teacher FTE by school and subject, one October snapshot a year, and teachers are a minority of every list: paraprofessionals, secretaries, custodians, guidance staff, social workers, maintenance and BCBAs are counted by nothing at school level. A list of intended cuts is not a record of staffing. — closes: the district’s year-end position control report — filled posts by FTE, school and funding source — which the Business Office produces for payroll and does not publish.

• Whether an announced cut was reversed by a hire during the school year

  Residents describe posts being cut in the spring and filled again a month into the school year. The one instrument that reaches teacher staffing is DESE’s EPIMS return, collected on 1 October, so a post cut in June and refilled in November is invisible to it and a post filled on 1 October and vacated in January counts as filled. Nothing in this archive is a within-year series. — closes: the district’s monthly payroll position report, or the School Committee’s personnel action items compiled as a series — the minutes carry appointments one meeting at a time and nobody has assembled them.

• How many Lunenburg children are educated at private or parochial schools

  DESE publishes where a town’s resident children go to school by receiving district, every year since SY2014 — and every row is keyed on the eight-digit code that identifies a Massachusetts PUBLIC district. Private schools do not have one and do not report enrolment to the state, so a Lunenburg child at Cushing Academy, Notre Dame or St Bernard’s generates no record: not a zero, not an “other”, nothing, for any town in any year (checked across the whole table). WHAT THIS BREAKS: the count of children educated outside Lunenburg’s own schools — 177 in FY2026 — is a FLOOR and reads as a total. A resident asked why Cushing was missing before the page admitted the limit. AND IT MATTERS IN THE OPPOSITE DIRECTION FROM THE REST: a child at a private school costs the town no tuition and no assessment, and moves Chapter 70 aid by about $150 a pupil, so those departures are close to free for the school budget. They are the only kind of leaving that does not cost money, and they are the kind nobody can count. IT IS NOT: evidence that the number is large, or small. Nothing here bounds it at all. — closes: the town census, which counts every resident child of school age regardless of where they enrol, or the district’s own October 1 resident count, which is compiled annually and not published.

• How many paraprofessionals Lunenburg needs, against how many it employs

  603 CMR 28.06(6)(d) is now in this archive, and it sets the class-size rule everybody argues around: a substantially separate group may not exceed eight students to one certified special educator, or twelve to a special educator AND AN AIDE. So a paraprofessional is partly what PERMITS a larger group rather than being a ratio in itself — which means paraprofessional counts track PROGRAM STRUCTURE and not student headcount, and would explain the thing /school-staffing measures and cannot account for: paras rising while enrolment fell. WHAT STILL CANNOT BE COMPUTED: the regulation binds INSTRUCTIONAL GROUPS and nothing published says how many groups Lunenburg runs, how large each is, or how its 41 substantially separate students are divided among them. 41 students could be four groups or seven, needing very different staffing, and both are lawful. THREE THINGS THAT WOULD DEFEAT ANY RATIO ANYWAY, all in the regulation itself: the sizes are MAXIMUMS and districts are expected to go smaller for complex needs (28.06(6)(c)); grouping must match each student’s own IEP, which can require one-to-one support no class-size rule predicts (b); and a group at maximum may take two further students mid-year by decision of the Administrator of Special Education (e). IT IS NOT: evidence that staffing is too high or too low. Nothing here supports either. — closes: the district’s own list of special education instructional groups by school and setting, with the students and staff assigned to each — compiled for DESE’s program approval under 603 CMR 28.09 and not published.

• Whether an “aide” in the class-size regulation is a paraprofessional the town budgets

  603 CMR 28.06(6) sets every group size in terms of a certified special educator AND AN AIDE — a group of eight becomes lawful at twelve because an aide is present. 28.02(3) defines the certified special educator in full and the regulation defines no counterpart for the aide anywhere; the word “paraprofessional” appears in the whole of 603 CMR 28.00 only in the staff-training clauses of 28.03(1)(a), one of which names “teachers, paraprofessionals, and teacher assistants” as three separate things. DESE’s staffing return and every Lunenburg budget line say paraprofessional. So the term that DECIDES whether a group of twelve is lawful cannot be matched to any post the town funds, and a count of paraprofessionals is not a count of aides in the regulation’s sense. This sits directly under the gap about how many paraprofessionals the town needs: even with the group list, the units would not match. — closes: DESE’s guidance on what qualifies a person as an aide for 603 CMR 28.06(6), or the district’s own job descriptions mapped to that clause.

• Where 16 of Lunenburg’s students with disabilities are placed, in FY2026

  DESE’s placement breakdown prints four categories — Full Inclusion, Partial Inclusion, Substantially Separate and Separate School in District — and in FY2026 they sum to 242 of the 258 the same file prints as the total. Element DOE034 has nine acceptable values and this breakdown is in-district only, so the other 16 children are in categories the file does not print, and nothing published says which. That bounds every reading of the placement mix: the four printed rows are not a partition, and adding them and treating the result as the whole is the same error as summing a column that was never the total’s parts. — closes: DESE’s full DOE034 distribution for Lunenburg, or the district’s 1 October SIMS placement submission.

• What the special education staffing ratio requires moment to moment

  603 CMR 28.06(6) sets the group sizes everybody quotes — eight students to one certified special educator, twelve with an aide — and residents are asking a further question the regulation does not answer: what happens when the group splits, when students go to recess or specials, or when the teacher leaves the room. THE FULL TEXT WAS SEARCHED. There is no mention of recess, none of "under the direction of", nothing about a group dividing, and nothing about a teacher stepping out. The only supervision clause is general and applies to every student in the Commonwealth: "The school staff shall at all times provide appropriate supervision of students while they are engaged in any school related activity on or off school grounds." And 28.06(6) says nothing about an aide beyond the two group-size clauses — it never defines what an aide does, what qualifications one needs, or whether one may supervise alone. WHAT THIS MEANS: the ratio governs THE SIZE OF AN INSTRUCTIONAL GROUP, not the staffing of a moment. It is not a supervision rule and cannot be read as one, in either direction — it neither permits nor forbids a paraprofessional holding a group while the teacher steps out. This is the second time this regulation has turned out to bind less than people assume. — closes: the district’s own program descriptions filed for DESE approval under 603 CMR 28.09, which describe staffing as operated rather than as capped; an individual IEP, which CAN specify one-to-one support at named times and is the only document that binds a moment; and the paraprofessional provisions of the collective bargaining agreement, for duties and coverage.

• Whether an aide assigned to one child also counts as the aide that raises a group from eight students to twelve

  603 CMR 28.06(6)(c) and (d) raise the maximum group size from eight students to twelve when the certified special educator is "assisted by one aide". The regulation never defines the aide, never states their duties, and — searched on every build for eight phrasings — contains NO rule about one-to-one support of any kind. So where a child’s IEP assigns an aide to that child, nothing published says whether that person also satisfies the tier. THE DIRECTION OF THE ERROR IS THE POINT: assuming they do understates the staffing a group needs, and assuming they do not overstates it, so neither assumption is safe and this project makes neither. It is the same wall as the paraprofessional gap, hit from the other side — the rule counts students and the staffing record counts posts, and nothing converts one into the other. — closes: DESE’s program approval materials under 603 CMR 28.09, which describe how each approved group is staffed; or the district’s own group descriptions naming which adults are counted against which tier.

• Whether Lunenburg’s paraprofessionals moved out of special education, or were recoded, or the state changed what it counts

  Two DESE files count Lunenburg paraprofessionals over FY2020–FY2025 and move in opposite directions. The district comparison sheet’s all-programmes Paraprofessional FTE rises 57.0 to 67.0; the special education staff table in n62c-bx65 falls 44.0 to 20.0, at every one of the five year-steps, while the number of children with a plan goes 244 to 238. The difference between the two therefore goes from 13.0 to 47.0 FTE. WHAT IS ESTABLISHED is that the two published series diverge. WHAT IS NOT is why: paraprofessionals moving onto general education assignments, the same people being recoded, and a change in what the state’s special education staff table counts all produce this shape, and subtracting one file from the other assumes the second sits inside the first, which nothing published says. This matters because the town’s own in-district special education escalator is built on a paraprofessional budget line. — closes: Lunenburg’s EPIMS work assignment detail by job classification and program area, year by year, which would show whether the same people were reported under a different code.

• What Lunenburg’s counsellors, social workers and psychologists do, as a series

  None of them is a teacher on DESE’s teacher returns, so the FTE files that carry fifteen years of teaching staff carry none of these people. The state’s educator retention file reaches them for three years only (FY2021–FY2023), as headcount rather than FTE, and folds them into two residual buckets named Other - Licensed and Other - Non-Licensed. The town’s annual report rosters name them every year FY2011–FY2025 and carry no FTE, no funding source and no date within the year. The district itself holds the missing instrument: it told the School Committee on 6 December 2023, while deciding what to cut, that “Information is broken out to social worker caseloads by bui lding”. That document is not published. Public comment at the same board on 24 January 2024 named MassHealth reimbursement and outside agencies as ways to pay for the same work, which is why a count without a funding source cannot answer the question anybody is asking. — closes: the per-building social worker and adjustment counsellor caseload the district prepared for the FY2025 budget, with FTE and funding source, which is a records request rather than a download.

• What share of a post any non-teaching adult in a school holds

  DESE publishes a full-time equivalent for TEACHERS at school level and for nobody else. There is no published FTE — in any file, in any year — for a paraprofessional, an administrator, a nurse, a counsellor, a custodian or a kitchen post at a named school in this town. The one job-class table DESE publishes (dese_educator_workforce) is DISTRICT-only and carries headcount rather than FTE. So a school panel can say Lunenburg High printed 64 names and that 35.8 full-time-equivalent TEACHING posts sat among them, and it cannot say what the other 28 names amount to. The difference between a roster count and a teaching FTE is not part-timers — it is everybody the FTE file does not cover. — closes: DESE’s EPIMS work-assignment detail by job classification and school, or the district’s own staffing plan stating FTE by building and post.

• Which school year a printed staff roster describes

  The rosters in the annual town reports carry no date, and the town’s own prose suggests the school section may run a year ahead of the cover. Three sentences in fifteen reports name a fall and a student count; matched against DESE’s October enrolment for the same school, two are closer to the year AFTER the report’s fiscal year and one to the year on the cover. So a roster labelled FY2024 may describe the autumn of 2024, which the state calls FY2025 — and any children-per-adult figure drawn from the two may be a year out of step. The FY2024 report makes the hazard concrete: it prints two complete Turkey Hill rosters, 33 names in common, one naming the leadership the town printed in FY2025 and the other the leadership it printed from FY2011 to FY2022, and neither page says which year it is. — closes: a dated staff roster, or the district’s October 1 EPIMS submission, either of which states the day the count was taken.

• Who actually lives in Lunenburg — how many are over 65, how many households have a child in the schools, and what each group can afford?

  ANSWERED IN PART, 11 September 2026, and the part that remains is the part the argument actually runs on. The Census tables named below were fetched and are loaded as census_acs: 11,804 ± 19 residents, 1,979 ± 309 of them 65 or over, 4,529 ± 235 households, 1,477 ± 198 with a child under 18, median household income $109,753 ± $12,275 and $72,935 ± $8,677 where the householder is 65 or over. See /lunenburg-by-the-numbers, where every figure carries its margin. WHAT SURVIVES is the third clause — what each group can AFFORD. Income is not wealth, not savings, not equity and not a tax bill, and nothing published joins a property tax payment to the age of the person who paid it; the two rows below carry the two halves of what is still missing. — closes: the town census, which counts every resident by name, age and household rather than sampling them, held by the Town Clerk.

• How many Lunenburg homes have a child in the LUNENBURG PUBLIC SCHOOLS — as opposed to a child under 18?

  The Census answers the second and not the first. B11005 puts 1,477 of 4,529 households — 32.6% — with a child under 18, which is what people mean when they say "about 30% of taxpaying homes", and on that reading the claim is fair. But that count includes children at Monty Tech, at private and charter schools, and children not yet of school age, so it is a CEILING rather than the figure. The arithmetic suggests the gap is real: 1,477 households against roughly 1,745 school-age children is about 1.2 per household, low enough that a meaningful share of those homes have only pre-school children. Nothing published joins a child to a household to a school. — closes: a count of DISTINCT HOUSEHOLDS with at least one child enrolled, which the district can produce from its own student information system and DESE cannot, because DESE counts children and the Census counts households and neither knows about the other.

• How much of the property tax levy is paid by households headed by someone 65 or over?

  The Census now gives income by age of householder — a household headed by someone 65 or over has a median income of $72,935 ± $8,677, against $129,183 ± $21,150 for a 45-to-64 household — and that is income, not tax. The override argument is about a BILL: what a $3,000,000 override adds to a household is quoted at meetings as a town-wide average, and the town publishes no split of the levy by who pays it, by age or by anything else. Nor does it publish how many households take the statutory elderly exemptions and abatements that would change the answer. So "seniors on fixed incomes cannot carry it" cannot be checked in either direction, and neither can its opposite. — closes: the Board of Assessors’ commitment list with the Clause 41A, 41C and 17D exemptions granted, joined to the town census by age of owner — both held by the Town and neither published.

• How many school employees are enrolled in the district’s active health insurance, and on which plan tiers?

  NARROWED 5 October 2026. the Finance Committee files received by records request on 4 October 2026 (sources/budget-workbooks/finance-committee/, see its PROVENANCE note) include a school department presentation whose second page counts active subscribers by plan for FY26 budgeting: HMO Blue family 110, HMO Blue individual 32, Blue Care Elect family 19, individual 9, Blue Select 1 -- 171 in all. A district slide, not the carrier's census; the per-employee sheet behind it is held privately and not cited. The $3,701,195 FY2026 active line is the largest piece of the schools’ $5.22M health insurance and no document states its headcount. The Town Manager’s FY27 Unclassified presentation (12 March 2026, pp. 6–7) gives the town side to the person — 55 town actives, 67 town retirees, 203 school retirees, November 2025 enrollment — and the school active side only as a method (FY25 Superintendent’s budget update, p. 41: “counting all staff currently subscribing”). Subscribers are not people covered: a family plan is one subscriber. — closes: the MIIA renewal census by plan tier for the school department, or the PEC negotiation packet; both exist and neither is published.

• How many administrators does the district employ, and how has that changed over ten years?

  The annual-report rosters carry names without FTE, classify department liaisons as administrators, and print no central office at all in FY2025, so a series read off them shows administration falling by a quarter in a year in which nothing of the kind happened; DESE publishes administration only as dollars per pupil. — closes: the district’s position control list by FTE and funding source, for any two years ten apart.

• How many students played a school sport in each of the last ten years?

  The district has published season participation counts for three years (FY2024–FY2026), and the archive holds no earlier ones; a ten-year trend in athletes cannot be drawn. — closes: the athletic department’s season participation reports for FY2016–FY2023, which the department compiles each season.

• Are long-tenured homeowners being taxed out — what do households who bought decades ago earn against what they pay?

  The Census publishes tenure (B25038) and income by age of householder (B19049) but never the two together for a town this size; the assessor publishes the bill per parcel and nothing about the household. — closes: Census PUMS microdata for the PUMA containing Lunenburg, which carries tenure, year moved in and income per household, at the cost of covering several towns at once.

• How many people each TOWN department employs, in FTE, in a form that can be counted the same way twice

  CORRECTED. This row previously said the only published series stops at FY2016, when the gross-wages list stopped tagging each name with a department. That is true of the WAGE LIST and it is not true of the reports. Every annual report carries departments describing their own staffing in the prose of their narrative section -- 77 such statements across FY2011 to FY2025, in every year. The Fire Department gives a series on its own: 5 Career and 40-45 On Call in FY2017, 7 Career from FY2018, 10 Career and 30-35 On Call/Per Diem from FY2022. The DPW states its establishment post by post: one Director, one Executive Assistant shared with Facilities, one Highway Superintendent, 5 Heavy Equipment Operators, one Mechanic, one Cemetery Superintendent, two Seasonal Cemetery Laborers, one Sewer Business Manager and one Assistant. The Building Department, the Council on Aging and the Assessor's office do the same. WHAT REMAINS A GAP is that none of it is FTE and none of it is consistent: a range (30-45 on call), a headcount, and a list of titles are three different quantities, stated in whichever form that year's department head chose, and they cannot be summed into a town total or compared like for like across departments. -- closes: the town's payroll register by department, or the personnel schedule behind the omnibus budget, which sets an FTE per line before Town Meeting votes it.

• Whether a department cut SERVICE, as opposed to cutting dollars or people

  A budget line that falls is not a cut and a headcount that falls is not a service reduction -- the thing a resident notices is hours, response times, programmes and waits. The annual reports do carry that: every year's department narratives print activity statistics such as calls for service, items circulated, patron visits, permits issued and inspections performed. None of it is extracted, and it is prose rather than tables, which is the same reason fifteen years of out-of-district placement counts went unread. Until it is read, no claim that a department held its service level while another cut can be tested in either direction. -- closes: nothing external is needed. This one is ours to do -- the figures are already in documents the archive holds.

• How many people work in the offices voted as General Government — the Town Manager, the Town Clerk, the Treasurer/Collector and Information Technology

  Nine town departments now state a headcount somewhere in their own report, but General Government is voted as ONE figure covering several offices and only the Assessing office inside it states a number. Health & Sanitation, Central Purchasing, Solid Waste and Employee benefits state none at all. The gross-wages list named a department beside each employee through FY2016 and stopped — closes: the Town’s gross wages list with its department column restored, or any payroll headcount by department.

• What each town department spends on wages, so departments can be compared by payroll rather than by headcount

  The town prints a Town & School Employee Gross Wages list every year, name by name — but it stopped printing the DEPARTMENT beside each name after FY2016, so the list cannot be split by department. Joining the names to the rosters we hold matches only a third to a half of them, so any departmental total built this way would rank our reading of the scan rather than the town’s payroll. THE REASON RECORDED HERE UNTIL 28 SEPTEMBER 2026 WAS WRONG: it said the two-column layout loses the second column’s given names, which was a measurement of OUR OCR CACHE written down as a fact about the town’s typesetting. The FY2019 pages print both columns in full, and a re-read lifted that year from 44 names to 516 in gross-wages.csv, and a reading of the pages themselves on 2 October 2026 found the pages print 599 -- so 83 names are in no dataset, 65 of them on PDF page 200 alone. 516 was a count of OUR dataset; 599 is a count of the town’s page. Nothing from it is published for that reason — closes: the wage list with its department column restored, or any payroll report by department.

• How much of a department’s headcount is full-time

  A roster counts the Council on Aging's Tuesday van driver and its Director as one each. ONE FTE FIGURE DOES EXIST and this row used to deny it: the FY2025 Finance Committee report states the FY2026 school budget “necessitates eliminating approximately 28 Full-Time Equivalents (39 employees)”, which also measures the gap between posts and people — 39 people in 28 FTE. That is one department in one year; no FTE is published for any other department or any other year — closes: an FTE schedule, or the wage list read in full.

• Whether a fall in a department’s headcount is a post being cut or a post standing empty

  A roster is people IN post; a cut removes a POST. The two come apart both ways — somebody leaves a funded post and the count falls with nothing cut, or a post is eliminated and its holder moves into another vacancy so the count holds. The Assessing office reported being “fully staffed for the first time in over a year” in FY2024, which is a year of posts that existed and stood empty and would have read here as a smaller department. Only the DPW and the Assessing office publish an establishment at all — closes: an authorised-post schedule, or a vacancy report, for any department.

• Who chaired each town board after FY2016?

  the officials listing marked its chairs with ** denotes chairperson and stopped printing the footnote after FY2016; 46 bodies name a chair somewhere in their own report and 53 do not, so 218 body-years have members and nobody at the top — closes: the Town Clerk’s record of board organisational votes, which every board takes annually

• Which town posts are funded but standing empty?

  a roster names who is there and an establishment names the posts, and only the DPW and the Assessing office publish the second — so a department that shrank and a department that could not hire look identical. The Police Department’s FY27 budget presentation marks every post Funded or Funded – Vacant, which is exactly the distinction, for one department in one year — closes: the FY27 budget presentations for every department, if they all mark posts that way

• Who works in Accounting, Human Resources, Land Use and Facilities?

  the town’s own staff directory lists 29 departments and these four are in none of this project’s 61 bodies — a department that files no annual report is absent from the contents page, and one whose head is not an APPOINTED OFFICIAL is absent from the officials listing, so both our sources miss the same four. The Facilities Director is named in the FY2023 report and on the Town Office Hours page and appears in no roster — closes: the town staff directory at lunenburgma.gov/directory.aspx, whose PEOPLE tab names everyone by department

• Who are the ten people the Public Library employs?

  the Library told the town in FY2019 that it employs ten — *“we only employ ten total”* — and it has never printed a roster. Its annual report is signed by the Director and names a colleague only in passing prose (a Young Adult Services Librarian in FY2024, a Digital Services Librarian the same year), so the org chart holds ONE of ten. The town staff directory does not carry the Library at all: it is governed by elected Trustees rather than run from Town Hall, and has no category on lunenburgma.gov — closes: the Library’s own staff page, or the Trustees’ minutes, which record appointments

• Which building a shared school employee actually works in, and for how much of the week

  Twenty-eight people are printed as staff by more than one Lunenburg school on the schools’ own listings, one of them by four of them. THAT REPETITION IS THE MEASUREMENT AND NOTHING MORE: it is consistent with a specialist splitting a week between two buildings, with a district post housed in one school and serving all of them, and with two buildings both claiming somebody neither employs alone — and the documents do not separate those. The district’s roll-up encodes sharing a second way, in a free-text column that reads Primary School & THES, and the two encodings agree about five of the 28. Neither carries a fraction, so Carol Saranich, listed by the high school, the middle school and Turkey Hill, could be a third of a post in each or a full post in one. This is rule 11 pointed at people: a name on a list is not a share of a job, and the budget lines that pay for these posts are netted and per-district, so they cannot be split by building either. — closes: the district’s FTE-by-building assignment schedule, the sheet behind the personnel lines in the school budget.

• When an agenda was actually POSTED, and so whether a meeting had its 48 hours’ notice

  The Open Meeting Law requires 48 hours’ notice excluding weekends and holidays, and NOTHING THE TOWN PUBLISHES SAYS WHEN AN AGENDA WENT UP. Checked rather than assumed: the AgendaCenter listing that every agenda is fetched from was searched in full for posted, published, uploaded, created and modified, and for any date-bearing attribute — zero occurrences of any of them. The row carries the MEETING date and nothing else. Two things stand in for it and neither is the answer. A PDF’s own creation timestamp is a LOWER BOUND on posting, since a file cannot be posted before it is made — the School Committee Policy Sub-Committee agenda of 25 September 2026 was created at 14:29 the previous afternoon, so the most notice possible for that file was 19 hours, and whether that is the notice actually given is not established. The AgendaCenter also allows a file to be replaced in place, so an earlier version could have been posted earlier and overwritten. And our own first_seen is worse, because it measures US. The watcher did not run on 18, 22 or 24 September 2026, so a document first seen on the 25th could have been posted on any of three days. A gap in the crawl silently becomes a gap in what can be said. — closes: the AgendaCenter’s posting log, which CivicEngage records internally and the site does not display; the Town Clerk can produce it.

• Is a rate printed in a town salary schedule the rate that was paid, and is the schedule itself internally consistent?

  The Salary Administration Plan schedule voted at the 2022 annual town meeting -- page 152 of the FY2022 annual report -- prints 20 grades against 8 steps and NO TOTAL of any kind, so nothing on the page tests its own figures. Its 160 hourly rates sit about 3% apart step to step and no single rounding rule reproduces them: round(step 1 x 1.03^(n-1)) misses 50 of the 140 derived cells, and grade 2 step 2 computes 15.38 where the page prints 15.37. The reading is safe because the page carries its own text layer and the characters are exact, but the DOCUMENT has no internal check a reader could apply, and a typo in a published schedule would be invisible. Separately, the schedule says what a grade and step pay per hour and says nothing about who is in which grade, at which step, or for how many hours -- so it cannot be turned into what the town spent on anyone. WHAT IS NOT ESTABLISHED: whether the printed rates are the rates the payroll system actually applied. -- closes: the Personnel Committee’s own rate table as submitted to the Town Clerk, which the warrant says is on file, read against the payroll register for the same year.

• Who is in which of the two lists the gross wage pages print, and is anyone in both?

  The Calendar Year 2022 Gross Wages pages (186-192 of the FY2022 annual town report) print TWO columns, and each is alphabetised A to Z independently all the way through the seven pages -- the left column runs ABRAHAM to ZRATE across 332 names and the right BIERY to YOURK across 329. Neither is a continuation of the other, and the document puts NO HEADING over either column and no total under either, so nothing in it says what separates the two. The obvious reading -- one is the town payroll and the other the school payroll -- is a HYPOTHESIS and nothing on the page tests it. It matters because a resident adding the two columns is adding two populations they cannot name, because several surnames appear in both columns, and because the list carries no department, no position and no hours for anyone: it is 661 names and 661 gross figures. WHAT IS NOT ESTABLISHED: what the two lists are, whether any person is counted in both, and which fund paid any of it. -- closes: the Town Accountant’s payroll register for calendar year 2022 with its department and fund coding, which is the report the wage list is printed from.

• How many children enter or leave an out-of-district placement during a school year, and when?

  Only two snapshots are published, months apart and taken by different bodies: DESE's count of students with disabilities placed out of district in October, and the town's count of placements on 1 March in the Special Services report of each annual town report. Their difference is a NET change -- an arrival and a departure between the two dates cancel, and a child placed and returned between them is in neither -- and some of it may be the two bodies counting differently. DESE's caseload-movement file (8aww-sugs) counts children entering and leaving special education SERVICES, not placements. The district has said at public meetings that children have moved in already needing a placement and that a child entering after 1 April is the sending district's cost; neither is a count. A mid-year reserve for tuition is sized by exactly this number, which is why /analysis/special-education-costs cannot size one from children and sizes it from the ledger instead. — closes: the district's out-of-district placement log, one row per placement with its start date, end date and setting (collaborative, day or residential) and no names, from the Director of Student Services by records request.

• Which school counselor, social-worker and psychologist posts were vacant or on leave in FY2026, and for how long?

  The FY2026 period-13 ledger leaves these salary lines well under budget — one psychologist account spent nothing of its budget — and no posted minutes explain it. A dollar is not a post (rule 11): vacancy, leave, a lower hiring step and contracted cover all produce the same row. — closes: the district’s position-control roster by month for functions 2710 and 2800, each budgeted post with who held it and from when.

• Whether the FY2026 spending on the kindergarten aide lines paid for aides in kindergarten classrooms

  The ledger books FY2026 spending to two accounts titled kindergarten aides and kindergarten paraprofessionals (orgs S2032121 and S2032131), both voted at $0; a Finance Committee minute of 12 January 2026 records the School Committee chair saying kindergarten classrooms had no aides. An account title is not a person in a room: staff coded to the line while assigned elsewhere, a reclassification, or posts moved between lines all print the same ledger, and nothing published joins the account to a building. — closes: the FY2026 payroll distribution for the two accounts, naming position and building (names may be redacted).

• What the corrected paraprofessional salary scale costs, and what was underpaid

  The School Committee approved a corrected paraprofessional scale on 7 October 2026: 107 of 120 hourly rates rose, steps 1-8 by about 4% (FY2026-FY2027) and 4.5% (FY2028), and the top steps did not move. Draft agreements released with it offer a lump sum of FY2027 retroactive pay and one personal day in exchange for a Wage Act waiver covering FY2026 and part of FY2027; the Superintendent's estimate is about 52 people. The cost cannot be computed: the scale is a full-year figure at 6.5 hours a day, and nothing published says how many paraprofessionals sit in each column and step or on what hours, nor what each was actually paid. See notes/findings/PARA-SALARY-CORRECTION.md. — closes: payroll's calculation of the FY2027 retroactive payment (headcount by column, step and appointed hours, with the sums), or the costing put to the School Committee for the 7 October vote.

• How many children ride the school buses, and how many pay the fee

  The only count in the record is the Superintendent's report to the School Committee on 4 June 2025 of bus requests received, and how many were unpaid, for the first fee year. A request is not a rider and an unpaid request is not a refusal. Without riders, no cost per rider and no share of the bus cost covered by the fee can be computed, and whether the fee changed ridership cannot be seen. — closes: the district's transportation registration count for each school year by grade band, fee tier (one child, family, reduced, free) and paid status, with the ridership counts the routes are planned on.

• Which of the people on the school district's master staff list still work for the district

  The district publishes its staff two ways: a sheet per building, and one master list ("Directory of Staff - All Lunenburg Public Schools Staff", a published Google Sheet). They disagree. On the 9 October 2026 fetch, 36 people appear on the master list at a building that keeps its own sheet -- 25 at the Primary School, 10 at Turkey Hill, 1 at the High School -- and that building's own sheet does not list them; a resident who knows the schools reports that at least one of them left years ago. The org chart takes each building's own sheet as the authority and sets those 36 aside, named, on the chart; it keeps 11 people the master list places district-wide or at the shared Middle/High building, marked as on that list alone. Neither list says which is current, and neither carries a start or end date. — closes: the district's current staff roster from Human Resources (or the MUNIS payroll employee list), dated.

held

• What the $87,293.86 is that the FY2023 annual report states twice and differently

  Two tables printed in the same FY2023 annual town report state the money held in special revenue and enterprise funds, and they disagree. The combined balance sheet on printed page 24 gives TOTAL FUND EQUITY of $6,131,303.68 for SPECIAL REVENUE and $3,812,502.24 for ENTERPRISE, $9,943,805.92 together. The special revenue schedule in the same report prints a GRAND TOTAL carried forward of $10,031,099.78. The difference is $87,293.86, where FY2011, FY2013, FY2019 and FY2022 agree to the cent. It is not a misreading: the FY2023 page was transcribed and every column foots to its own printed TOTAL ASSETS, TOTAL LIABILITIES and TOTAL FUND EQUITY, the accounting identity holds in all six columns, and the long-term debt mirror agrees. So FY2023 is transcribed and deliberately NOT written into the dataset, which is why the combined balance sheet here stops at FY2022 while the special revenue schedule runs to FY2023. One explanation has been tested and RULED OUT: the fund 'Sale of Cemetery Lots' carried $87,308.09 in FY2023, within $14.23 of the gap, which suggested a fund counted in one statement and classified elsewhere in the other. It is not that — the same fund appears in the schedule in all thirteen years including the four where the two documents agree to the cent, so it is counted consistently and the near-match is a coincidence. What the difference IS remains unestablished; nothing here says which document is right, nor whether it is connected to the $17,861.24 of grant funds the town restated between the FY2022 and FY2023 reports. Those are two different amounts and nothing joins them. — closes: the Town's FY2023 trial balance or general ledger fund balances as of 30 June 2023, which would say which of its own two published tables the accounting system agrees with.

• What the $102,000.00 is that the FY2024 enterprise balance sheet classifies two ways at once

  The Combining Balance Sheet - Enterprise Funds in the FY2024 annual town report — printed twice, on printed pages 21 and 28, identically — prints 'Reserved for expenditures' as 400,000.00 in the Sewer column and 502,000.00 in the Totals (Memorandum Only) column, and prints 'Unreserved retained earnings' as 2,973,842.15 in the memorandum column where the four fund columns (1,813,817.53 Sewer, 51,333.93 Water, 181,853.82 Solid Waste, 1,028,836.87 PEG Access) add to 3,075,842.15. The memorandum column is $102,000.00 high on the first line and $102,000.00 low on the second, so the two cancel and Total Fund Equity (3,483,633.65), Total Assets, the identity and the sheet's own printed PROOF row all still tie in every column — compensating errors, invisible to any check that reads only totals. WHAT IS NOT ESTABLISHED: which presentation is right, or what the $102,000.00 is. A reservation sitting in a fund column that prints blank, a memorandum column keyed to a different worksheet row, and a reclassification made only in the total column all fit the same signature, and nothing on the page distinguishes them. These are ratepayer funds; nothing here is general-fund money. — closes: the Town Accountant's FY2024 trial balance for the enterprise funds, which would state the reserved and unreserved balance of each fund separately.

• What the town held town-wide at 30 June 2024 and 30 June 2025

  The FY2024 and FY2025 annual town reports print NO combined balance sheet across all fund types and account groups — the schedule the reports carried every year from FY2011 to FY2023. In both editions the table of contents lists 'Balance Sheet for FY Ending June 30, 2024' at Page 21 and 'Balance Sheet for FY Ending June 30, 2025' at Page 21 respectively, and printed page 21 of each holds the Combining Balance Sheet - Enterprise Funds instead; FY2024 prints that enterprise sheet a second time on page 28, which is where its contents page says it should be. So for two consecutive years the only balance sheet the town published covers four ratepayer enterprise funds and says nothing about the general fund, special revenue, capital projects, trust and agency, or long-term debt. The consequence is that the general fund's cash, receivables, reserves and undesignated fund balance — the figures free cash is built out of — cannot be read for either year, and the balance_sheet dataset stops at FY2023 for that reason rather than by choice. WHAT IS NOT ESTABLISHED: whether the town-wide sheet was produced and dropped in layout or never produced; nothing in either report says. — closes: the Town Accountant's year-end balance sheet submitted to the Division of Local Services for FY2024 and FY2025, or the town's audited financial statements for those years.

• What the PEG Access enterprise fund actually held at 30 June 2025

  Three figures for the same fund on the same date appear in the same FY2025 annual town report and no document in this archive prints a line between any two of them. The PAC report's FY25 REVENUE vs EXPENSES table (printed page 60, pdf 64) gives an Ending Balance of $30,632, computed on its own face as Beginning Balance $208,772 less Expenses $178,140 -- and the prose beside it says the $208,772 was "the budget starting balance", so the ending balance is unspent spending authority rather than money. The Treasurer's Cash schedule in the same report prints Bartholomew- PEG Access Enterprise Fund 782,591.31$. The Combining Balance Sheet - Enterprise Funds (printed page 21, pdf 25) prints PEG Access Total Fund Equity of $1,072,661.74, against $1,028,836.87 the year before -- a rise of $43,824.87, where the statement's own revenue less expenses is $262,565 - $178,140 = $84,425. The two do not reconcile and nothing on either page relates them. WHAT IS NOT ESTABLISHED: which quantity any of the three is; a budget balance, a bank balance and a fund equity are three different things and the report labels none of them as such. -- closes: the Town Accountant's FY2025 year-end statement of revenues, expenditures and changes in fund balance for the PEG Access and Cable Related Enterprise Fund, which the annual report does not contain.

• Which cell of DESE’s contribution sheet is mislabelled in FY2020

  In eighteen of the nineteen differenced years the town’s preliminary contribution equals the previous year’s required contribution multiplied by that year’s municipal revenue growth factor, exactly as DESE’s own User Guide defines it, and the required contribution is that figure adjusted toward target. In FY2020 the two are the other way round: dataContribution!O2593 prints $10,427,328, which is FY2019’s required contribution un-grown, and dataContribution!U2593 prints $10,889,259, which is that figure grown by the 4.43% factor. One of the two cells is carrying the other’s quantity and nothing in the workbook says which. The extract records what the sheet holds rather than repairing it. — closes: DESE’s FY2020 required local contribution worksheet for Lunenburg, which prints the preliminary contribution and the required contribution as separate stated steps.

• Why the district and the town state different FY2027 Monty Tech assessments

  Monty Tech’s FY2027 budget presentation puts Lunenburg’s total assessment at $1,452,379, built from a required minimum contribution of $1,378,183 and a transportation and operating assessment of $74,196 with no capital and no bonds. The Town’s own FY2027 operating budget book carries the same line at $1,452,426, in all three of its scenarios. The difference is $47 and neither document acknowledges the other. It is small enough to be a rounding or a late revision and it is not nothing: the two figures are the district’s bill and the town’s appropriation for the same year, and every other year in this archive where both exist ties to the dollar or to eighteen cents. — closes: the district’s FY2027 assessment certification to the Town of Lunenburg, and the Monty Tech line as finally voted at Annual Town Meeting.

• What spending the general Stabilization Fund would do to the town’s borrowing

  The town’s reserves are one of the things a rating agency weighs, and this archive holds NO bond rating for Lunenburg, no rating report, and no official statement from any bond issue — so what the rating is, what any agency has said about the reserves, and what a downgrade would cost per million borrowed are all unknown here. The question is live and has been asked in public: at the Finance Committee on 13 March 2025 a resident asked whether not contributing to OPEB affects the town’s bond rating, and the minutes record the question and no reply. Until this closes, any sentence about "what the rating agencies would think" is a hypothesis, including on /analysis/stabilization-option, which says so. — closes: the official statement for the town’s most recent bond issue, and the rating agency report that accompanied it, both of which the Treasurer holds.

• Whether Lunenburg has a written reserve target, and what it is

  NARROWED 5 October 2026. the Finance Committee files received by records request on 4 October 2026 (sources/budget-workbooks/finance-committee/, see its PROVENANCE note) include a Free Cash Fact Sheet for the FY27 warrant quoting the Town of Lunenburg Free Cash Policy, Section 3.03 (amended 09/02/2014 and 10/18/2018). That is a quotation of the policy, not the adopted policy document, which remains the remedy. Nothing in this archive states a level the town intends to keep in stabilization or free cash — no reserve policy, no fund balance policy, no target as a share of the operating budget. So there is no stated floor any drawdown could be measured against, and "the reserves are healthy" and "the reserves are excessive" are both unfalsifiable here. The Finance Committee has had "Financial Related Policies" as standing Old Business (agenda, 10 September 2026) and funded a financial policies line, so a document may exist or be in progress. — closes: the town’s adopted financial reserve policy, or the Finance Committee’s confirmation in writing that none has been adopted.

• What the Sewer Reserve Capacity and Arts Lottery funds held at 30 June 2023

  The FY2023 annual town report prints each of these balances TWICE, in two different tables, and the two printings disagree. TRUST AND STABILIZATION FUNDS HELD BY OTHER BANKS (pdf page 49) gives SEWER RESERVE CAPACITY (UNIBANK) an ENDING CASH VALUE of $28,144.20 and ARTS LOTTERY (MMDT) $19,120.90. Treasurer's Cash as of 6/30/2023 (pdf page 48) gives Unibank Sewer Capacity Reserve Stabilization $28,144.50 and Mass Municipal Depository Trust (MMDT) Arts Lottery $19,121.07. NEITHER IS A MISREADING: each column foots to its own printed total to the cent -- page 49's eleven funds sum to the printed GRAND TOTALS of $2,363,536.23, and page 48's thirty-nine accounts sum to the printed Total Treasurer Cash of $27,244,162.79 -- so the report disagrees with itself by $0.30 and $0.17. Six other funds appear on both pages and match exactly, which is what makes the two the odd ones rather than the tables a different measure. WHAT IS NOT ESTABLISHED: which of the two each pair is. An ending cash value on a custodian's trust statement and a treasurer's cash balance are produced by different parties on different books and the report relates them nowhere, so the difference may be a posting date, an accrual, or a typing error in one of them, and nothing here can tell those apart. The amounts are small; the pattern is not, because these two tables are the only two statements of these balances the town publishes and they are printed in every annual report. -- closes: the Treasurer's year-end reconciliation of cash to the custodian's statements at 30 June 2023, which the annual report does not contain.

• How much the Special Education Reserve Fund holds, and what the law lets it do

  Special Town Meeting on 18 November 2025 accepted Mass. General Laws chapter 40 section 13E and established a Special Education Reserve Fund for unanticipated or unbudgeted special education, out-of-district tuition and transportation costs (FY2025 annual town report). The Town Manager told the Select Board on 7 October 2025 that spending from it needs a majority of both the School Committee and the Select Board and that its balance may not exceed 2% of annual net school spending. The statute itself is not in this archive, so the cap and the vote are as described at a meeting; and no balance, deposit or withdrawal is in any ledger held here. The School Committee's agenda for 7 October 2026 carries an item to vote an amount to fund it. — closes: the text of chapter 40 section 13E (malegislature.gov, public); and the Treasurer's balance of the fund with every vote to fund or spend it, from the Town Accountant.

• What the town’s own special revenue funds held at 30 June 2024, 2025 and 2026

  The annual town reports print every special revenue fund's balance at 30 June, and this project has read and proved FY2011 to FY2023 -- forward, receipts, disbursements and carried, every row and every printed GRAND TOTAL tying (special-revenue-read.csv). FY2024 and FY2025 print a DIFFERENT table, a balance detail with no receipts or disbursements, no row identity and no stable fund key: it is printed and has not yet been read into a dataset, which is our gap rather than the town's. FY2026's annual report is not yet published. The school side of 'is anyone sitting on money' runs to 30 June 2026 through the MUNIS period-13 ledgers; the town side stops at 30 June 2023, so whether the town's own funds kept growing cannot be said. Measured in /analysis/sitting-on-money. — closes: a MUNIS fund balance report (or trial balance) at period 13 for FY2024, FY2025 and FY2026 covering the town's special revenue funds -- the same system that produced the school delivery of 6 October 2026, pointed at the town's funds.

• What the school grant funds held at 30 June 2024, 2025 and 2026

  Fourteen school funds -- circuit breaker, school choice, lunch, eight fee-funded revolving funds and three gift funds -- are carried from the annual report's 30 June 2023 balance through the MUNIS period-13 ledgers, and the result ties to the cent to the Town's March 2026 special revenue report for every one. The grant funds cannot be carried that way: the annual report names a grant by programme and year, MUNIS numbers it, and nothing published maps one to the other; and a reimbursement grant spends before it is paid, so one year's revenue less expense is not a change in what it holds. So the school total after 2023 covers the fourteen and not the grants. Measured in /analysis/sitting-on-money. — closes: a MUNIS fund balance report at period 13 for the school grant funds (26xx-29xx), FY2024 to FY2026, with each fund's opening and closing balance.

record

• What was said at a meeting whose minutes the town published only as a photograph

  3,116 of the 12,015 agendas and sets of minutes this project holds (26%) contain no text a search can match. 3,096 are image scans, 3 draw their text as vector outlines, 14 are blank, 2 have a text layer our extractor could not read and 1 will not parse; a further 50 are listed by the town and not held here. WHAT THIS IS NOT: it is not evidence that anything was left unsaid. The words are on the page and the page is a picture of a page, so an unreadable document and a subject nobody raised produce the identical empty result. Rule 15a instructs everyone working here to search the meeting archive before concluding a thing was never discussed, and until this was measured that search silently skipped a quarter of the record while printing a coverage line saying otherwise. IT IS NOT EVEN: the coverage is not evenly spread, so an archive-wide figure understates the boards that carry it — Board of Assessors 255 of 488 (52%); Select Board 629 of 1,347 (47%); Lunenburg Housing Authority 110 of 255 (43%). Measured per board and year in sources/data/minutes-searchable.csv, which scripts/build_minutes_searchable.py regenerates and check_generated.py refuses to let go stale. — closes: OCR of the image-only documents, already on the roadmap in notes/QUEUE.md, which recovers the scans and the vector-outline pages but produces OUR rendering rather than the town's own words (rule 13) and so must never be quoted as the document's text; and, for the documents the town lists that we do not hold, a records request to the Town Clerk for the minutes of the named meetings in a text-bearing format.

• When was a meeting recording made, where the title does not say?

  The PEG channel’s video titles are the only chronology there is: a flat channel extraction returns no upload dates, and 652 of the 4,671 titles state no date at all (as indexed 8 September 2026). For those the recording can be placed only by its position in the channel listing, which is not a date. — closes: the channel’s own per-video publication date, which the YouTube Data API returns as snippet.publishedAt. Even that is an UPLOAD date and not the meeting date.

• Which body met, for a recording whose title names none?

  14 of 4,671 videos are undetermined — Informational Symposium, Lunenburg Town Charrette . FULL, Master Plan Public Forum, School Funding Forum, 911 YouTube. The title is all we fetch; the channel also carries a description field on each video and several of these titles say "See Description". — closes: the per-video description text from the channel, which the index does not currently store.

• Was a meeting recorded at all, when no video was posted?

  The video index establishes THAT a recording was posted, which is a fact the written record does not always carry — the town posts minutes for a fraction of the meetings it holds. It cannot establish the reverse: a meeting recorded and not posted, or held and not recorded, looks identical from here. — closes: the PEG station’s own cablecast log, which lists what was recorded rather than what was uploaded.

• How many meetings the town held, and how many of them left no record at all

  This project can count meetings it has EVIDENCE of — 7,014 occasions where an agenda, a set of minutes or a recording says a board met on a day. It cannot count meetings. The town publishes documents rather than a schedule of what took place, so a meeting with no agenda, no minutes and no recording is indistinguishable from a meeting that never happened, and neither appears in sources/data/meeting-register.csv. That the gap is real rather than theoretical is shown by the meetings that survive on ONE thread: 231 occasions have a recording and no paperwork whatever, 162 of them School Committee, and 2,395 have an agenda announcing the meeting and no minutes saying what happened. Each of those would have been invisible if the other source had also been missing. WHAT THIS IS NOT: it is not a finding that the town met in secret. Open Meeting Law notice and the filing of minutes are separate obligations, and nothing here establishes which unrecorded meetings were noticed. — closes: the Town Clerk’s own record of meetings posted under the Open Meeting Law, which is a list of meetings NOTICED rather than a list of documents published, and is the only source that would give this register a denominator it did not derive from its own contents.

• Over what period does DESE’s placement trajectory measure movement

  Dataset 92x3-2qj9 follows a cohort from the setting its placement started in to where the same children are now, and it is the only published route into out-of-district placement. The workbook is a single Data sheet with no notes, and nothing in it or on the portal page states the interval between the start and the destination. Every rate drawn from it — including the finding that Lunenburg children who started in a substantially separate classroom are out of district about fifteen times as often as those who started in an inclusive setting — is a rate over an unstated span, so none of them can be annualised or compared with a rate measured over a known one. — closes: DESE’s methodology note for dataset 92x3-2qj9, stating the cohort rule and the interval.

• Why these six districts and not six others

  Every comparison on this page runs against a set of six districts chosen by this project — Ashburnham-Westminster, Ayer Shirley, Groton-Dunstable, Harvard, North Middlesex and Lunenburg. DESE’s RADAR workbook covers all 421 Massachusetts districts and this archive extracts six because the full file takes the published database past Cloudflare’s per-asset limit. No document here records the criterion by which those six were picked, and a peer set is an argument: whoever chooses it has already made a claim about what Lunenburg is comparable to. That is why every headline figure on /what-other-districts-spend is stated against DESE’s statewide distribution of 318 districts rather than against the six. — closes: DESE’s own DART comparison group for Lunenburg, which names the districts the Department considers similar and states the criteria it used.

• Whether Lunenburg’s MCAS results are high or low for Massachusetts

  Residents have argued the comparison in both directions at School Committee — that the town is “in the top 20% for school performance while being in the bottom 10% for spending”, and that “neighboring districts… have faced similar challenges while achieving stronger outcomes”. The spending half of that can be checked here, because DESE publishes the statewide distribution of per-pupil spending with quartiles and a rank. The results half cannot: this archive holds MCAS for six districts and no statewide distribution of it at all, so no percentile, rank or state median can be computed for any of the four measures. — closes: DESE’s district-level MCAS achievement file for all districts, published per year at profiles.doe.mass.edu, extracted the way the finance distribution already is.

• When the town actually posted an agenda or a set of minutes, as against when we first saw it

  The AgendaCenter lists a document's MEETING date and publishes no posting date anywhere -- not in the listing markup, not in the file's URL, not in the response headers. So scripts/watch_meetings.py can record only first_seen, the day this project's crawler first observed a document, and the gap between a meeting and its minutes appearing is bounded by how often we look rather than measured. Every such field in sources/data/meeting-watch-events.csv and in fy28/public/data/meeting-feed.json is named days_after_meeting_upper_bound for that reason. WHAT IT IS NOT: it is not a measurement of how long the town takes, and a document posted the day after a meeting and one posted three weeks later are indistinguishable if we looked once in between. It is also blind backwards: the 12,065 documents the watch adopted when it was seeded on 8 September 2026 carry no first_seen at all -- an empty cell rather than a guess -- so no lag figure exists for anything published before that date, and the series begins the day the watch did. The separate question of how many meetings the town HELD is registered above, “How many meetings the town held, and how many of them left no record at all”. -- closes: the CivicEngage AgendaCenter's own record of when each document was uploaded, which the platform stores and the public listing does not render; the Town Clerk or the site administrator can export it.

• How much of what was said in a public meeting does a search of the archive actually reach?

  Our own text extraction inserts spaces inside words, at scale -- w ell, s elling, dep artment, B oard are all in the extracted minutes. Nine of the eleven hits for ELL are this rather than English learners. It costs twice: it manufactures tokens that were never printed, and -- the part that matters -- it makes PHRASE SEARCH MISS, because study hall broken as study h all will not match "study hall". So a search that returns nothing may mean the words were said and our copy cannot be read, which is the exact failure the coverage denominator exists to prevent, arriving underneath it. The rate has not been measured and neither has the phrase-search undercount; both are countable against the source PDFs. This belongs to extract_minutes.py, not to the index. -- closes: nothing anybody publishes. A re-extraction with a layout-aware reader, measured against the same PDFs, is the only thing that settles it.

• When the town actually POSTED each agenda and each set of minutes — the two timestamps the Open Meeting Law turns on (48 hours of notice; minutes in a timely way)

  The AgendaCenter publishes no posting timestamp and our watch was seeded on 8 September 2026, so only 24 of 12,089 documents have a first-seen date. The documents’ own creation timestamps (sources/data/meeting-document-timestamps.csv) bound the posting from below and cannot prove when a file went up, and a file replaced in place takes a new date — closes: an AgendaCenter export carrying the posted-on date per document, or the town’s own record of when each was uploaded.

• What is in the PEC agreement that sets the town’s health insurance split and terms?

  NARROWED 6 October 2026. The PEC Agreement of 16 June 2008 is held (sources/contracts/pdf/pec-agreement-2008.pdf, a 16-page scan holding the 8-page agreement twice), received from the Town’s Human Resources Director by email. It names BCBS plans, a 75% Town share on the PPO, 80% then 75% from 1 July 2011 on the HMO and the two senior plans, 50% on Medex, 75% of the Medicare Part B premium, and a term of 1 July 2008 to 30 June 2012 continuing ""until the new agreement is reached"". What sits BETWEEN it and the FY27 memorandum, which amends ""the current in-force health insurance agreement"" without naming it, is not held: the Select Board minutes record ""the amendment dated May 14, 2013"" and a further amendment approved 11 March 2014, an agreement expiring 30 June 2016 (10 May 2016), and a Public Employee Agreement approved 12 April 2022 ""effective July 1, 2022"". Which of these the FY27 memorandum amends is not established. — closes: the complete, current agreement the Human Resources Director wrote on 6 October 2026 she was ""working on producing"", or the 2013, 2014, 2016 and 2022 instruments themselves. DECIDED 6 October 2026 (TJ): the current terms are modelled from the FY27 memorandum and its Attachments A and B, which restate every money term in full -- plans, the 75/25 and 50/50 splits, the $96 Part B payment, the $3,000/$6,000 opt-out, and the term to 30 June 2029. The missing instruments do not change any figure the model uses. They matter for the HISTORY of the split (when it moved, and to what) and for the procedural terms the memorandum carries over unrestated (eligibility, retiree and spouse definitions). Lower priority; ask when convenient.

• What do the two Medicare plans in the FY27–FY29 PEC agreement cover?

  The FY27 memorandum names six plans and puts ""benefit brochures"" in Attachment A; the Attachment A received on 6 October 2026 holds four, all for active plans (Network Blue New England, Network Blue Select, Access Blue New England Saver, Blue Care Elect). Managed Blue for Seniors and Medex — the plans the memorandum sets at 75/25 and 50/50 for retirees — have only their prescription copays stated in the memorandum itself. — closes: the MIIA summaries of benefits for Managed Blue for Seniors with PDP and Medex with PDP, FY27.

• Did the town ever adopt c.32B §§21–23, the expedited plan-design route?

  This decides whether plan design here is bargained in full or changed through 30 days and a binding panel. §21(a) makes it a vote of the board of selectmen in a town, so Town Meeting warrants were never the place to look; no such vote has been found in the Select Board record we hold. The Board was walked through the reform and its "Maximum Possible Savings" on 22 January 2013 and the follow-ups went to executive session. — closes: the Select Board vote adopting or declining §§21–23, or a statement that none exists.

• What was actually said in the 232 executive sessions the Select Board entered citing collective bargaining?

  624 Select Board documents mention executive session and 232 cite collective bargaining; NO executive session minutes are in this archive. Health insurance is bargained there, so the open-session minutes catch the PEC only in passing — a phrase, an agenda line, a date — and never its substance. — closes: released executive session minutes, which the Open Meeting Law requires once the reason for secrecy has passed.

• What did Ayer and Shirley each spend for each pupil BEFORE they regionalised?

  Ayer and Shirley regionalised for FY2012, which is the one local case anybody could learn from. The archive holds Ayer for FY2009-FY2011 and Ayer Shirley School District from FY2012, and nothing for Shirley alone beforehand - so the figure before is one town and the figure after is two. That is a different denominator, not a measured change, and comparing them would produce a confident number about the wrong thing. - closes: DESE per-pupil expenditure for Shirley, FY2009-FY2011.

• Is account 8129 the Zoning Incentive Stabilization Fund or the playground fund?

  The annual reports print 8129 as ZONING INCENTIVE STABILIZATION (TD BANKNORTH) in FY2020 and as Zoning Stabilization Fund in FY2025. This project’s own account registry, sources/data/fund-owners.csv, records 8129 as the playground fund with its owner unresolved. Both cannot be right, and the registry is what several pages read to name an account. The extracted balances keep the document’s name and flag the disagreement rather than resolving it. — closes: the Treasurer’s chart of accounts, or the Town Meeting article that created 8129.

• Who appoints each appointed official, and to whom do they answer?

  the annual report heads six pages APPOINTED OFFICIALS and names nobody as the appointing authority, so a Constable, a Town Clockwinder and a Wellness Coordinator sit in the record with no department and no stated superior. The town charter is the document that says which appointments belong to the Select Board and which to the Town Manager, and this archive does not hold it — closes: the Lunenburg town charter and the general bylaws

• Which figure did the 2018 Annual Town Meeting actually vote where the printed FY 2019 Omnibus Budget does not add up?

  The budget as printed on pages 155-158 of the FY2018 annual town report carries TWO arithmetic misprints and the document contradicts itself about both. Total General Government $1,401,119.25 is 709.00 MORE than the 22 lines beneath it, and the printed Total Omnibus $36,804,408.83 agrees with the subtotal rather than with the lines. Total Assistance $243,493.56 is 2,000.00 LESS than the two subtotals above it, and there the Total Omnibus agrees with the subtotals rather than with the line. Ten of the twelve section totals foot their own line items to the cent, every subtotal foots exactly, and the pages carry a digital text layer, so this is the town's arithmetic and not our reading. What a resident cannot tell from the published report is which number was put to the meeting: the appropriation is voted as one sum, and a 709.00 or 2,000.00 difference changes what the departments beneath it were authorised to spend. - closes: the Town Clerk's certified record of the vote on Article 7 of the 5 May 2018 Annual Town Meeting, which states the sum voted and the schedule attached to it.

• Which fiscal year did Town Meeting vote the left-hand half of the FY 2016 Omnibus Budget for?

  The omnibus schedule on pages 140-141 of the FY2015 annual town report is titled FY 2016 Omnibus Budget and is set as two panels per page. Its right panel is headed VOTED FY 2016 and its left panel VOTED FY 2015. Everything else on the page says it is one column: the line numbers run 1 to 82 continuously across both panels and both pages, the 91 line items sum exactly to the single TOTAL OMNIBUS of $32,025,895.00, and so do the twelve printed section totals. But the report itself never states which heading is wrong, and the arithmetic cannot tell a misprinted heading from a correct one. Until it is settled, no figure from the left panel may be cited as an FY2015 appropriation or as an FY2016 one without saying which. - closes: the Town Clerk's record of the vote at the 2015 Annual Town Meeting, with the warrant article and the amount as moved, which the town holds and does not publish.

• Which capital project is the unnamed $25,000.00 on the FY2016 capital projects schedule?

  THE PAGE PRINTS A RULED ROW WITH A FIGURE AND NO NAME. On CAPITAL PROJECT FUNDS FY 2016, PDF page 36 of the FY2016 annual town report, the row between the column header and 08/02 ATM Development of Cemeteries carries $25,000.00 in APPROPRIATED FORWARD and nothing else -- no purpose, no article reference, no fwd flag, no expended figure, no balance. IT IS NOT A READING ERROR AND IT IS NOT IGNORABLE. All three columns of the page tie to the printed Grand Total to the cent only when that $25,000.00 is counted, so the figure is part of the schedule. And the Cemeteries row immediately below it does not close as printed -- $1,019.01 less $35,229.38 is ($34,210.37) against a printed balance of ($9,210.37), short by exactly $25,000.00. WHAT IS NOT ESTABLISHED: that the $25,000.00 belongs to the Cemeteries project. A label that failed to print, a second article against the same project, and a separate project whose text was lost all fit the same evidence, and nothing in the report tests between them. — closes: the Town Accountant's FY2016 capital project fund detail by article, which names the article every capital appropriation sits under.

• Which figure did the 2019 Annual Town Meeting vote for Central Purchasing, where the printed FY 2020 Omnibus Budget prints two that are a factor of ten apart?

  The budget is printed on pages 161-164 of the FY2019 annual town report (printed pages 159-162) and the document contradicts itself in two places. p162 prints 38 Central Purchasing $ 603,030.00 and, three rows beneath it, Total Central Purchasing $ 60,303.00 -- $542,727.00 apart. And p162 prints four Fire lines (1,093,374.24 + 20,000.00 + 15,000.00 + 10,000.00 = 1,138,374.24) above Subtotal Fire Dept. $1,138,434.24, $60.00 more than its own lines. Sixteen of the eighteen printed group totals foot their own line items to the cent and the twelve section totals sum to the printed Total Omnibus $38,058,164.30 exactly, so this is the town's arithmetic and not our reading: both pairs were read at 400 dpi off the page's own bitmap. The town's grand total is footed on $60,303.00 and on $1,138,434.24, so the printed Total Omnibus cannot distinguish a misprinted line from a misprinted subtotal. A resident cannot tell from the published report what Central Purchasing was authorised to spend in FY2020, and the two readings differ by more than half a million dollars. -- closes: the Town Clerk's certified record of the vote on the FY2020 omnibus budget article at the 2019 Annual Town Meeting, with the schedule attached to the motion as moved.

• Whether a board discussed a thing at a meeting it noticed but left no minutes and no recording of

  1,754 of the 6,990 meetings in sources/data/meeting-register.csv (25%) have a posted AGENDA and neither minutes nor a recording, so the only surviving record is that the meeting was noticed and what it meant to take up. 235 of them are since 1 July 2024. It is worst where a body is small: School Advisory Councils & Committees 229 of 268 (85%), Capital Planning 63 of 182 (35%), Board of Health 101 of 297 (34%). WHAT THIS IS NOT: it is not evidence that nothing was discussed, and a search that returns nothing for one of these dates has not searched the meeting -- it has searched an agenda. The practical cost is the inverse question. Parks Commission can be shown DISCUSSING its budget (44 of 192 meeting dates carry the word, and Budget / Finance Review is a named agenda item from 22 July 2026), and cannot be shown NOT to have, because 40 of its 193 meetings left no account at all -- including 30 September 2026, whose agenda item was Budget Check-in. This is distinct from the register's own limit, that a meeting with no agenda either is indistinguishable from one never held: here the town has published a notice, so the meeting is established and only its content is missing. — closes: the minutes themselves. G.L. c.30A §22 requires a public body to create and maintain accurate minutes of every meeting and to make them available on request, so for each of these dates the document is owed whether or not it was ever posted to the AgendaCenter. Requested by board and date.

• Does segment 51 of a school account string mean special education?

  The town's account string is nine segments; the fourth is the DESE function code. The fifth is 51 on every general fund school account whose own description names special education, except special education transportation, and /analysis/special-education-costs classifies in-district special education by it -- less two accounts the district's own budget book labels as English learner costs, which shows the segment is not purely special education. The meaning of the segment is read from the account names, not from a published chart of accounts, so the in-district total rests on our reading. — closes: the Town's MUNIS chart of accounts, with the definition of each segment and its values, from the Town Accountant.

• What was the School Committee told under "FY26 Year End Budget review" on 2 and 9 September 2026?

  Both agendas carry the item (Old Business, 5 minutes). The archive's catalogue of the Town's AgendaCenter lists no minutes for either meeting, so the only written statement on the FY26 school year-end held here is the 29 July 2026 minutes, in which the business manager was "confident the remaining amount would be well below" the approximately $600,000 returned in FY25. The machine captions of the recordings can locate a moment but are not a record. /analysis/fy26-school-surplus. — closes: the approved School Committee minutes of 2 and 9 September 2026, and any year-end budget document presented under that item.

• Why DESE’s FY2025 transportation spending for Lunenburg differs from the town ledger

  DESE's End of Year Financial Report gives general fund in-district transportation (function 3300) and out-of-district transportation, and their sum equals the town's MUNIS regular plus special education transportation accounts to the dollar in FY2022, FY2023 and FY2024. In FY2025 DESE's sum is higher than the ledger and nothing published explains it. In FY2020 and FY2021 the two differ by equal and opposite amounts, which is what one payment booked in different years by the two would look like. — closes: the district's FY2025 End of Year Financial Report workpapers mapping MUNIS accounts to DESE functions, from the Business Office.

• How many votes each FY2027 override question received on 16 May 2026

  Two records disagree and neither is the certified count. The Town's unofficial tally sheet (town-supplementary/docs/town-2026-election-unofficial-results.pdf) foots precinct by precinct to its own ballots-cast total; the Division of Local Services' override record prints different yes and no counts for both questions, whose yes plus no alone exceed that ballots-cast total. Which is right is not established here. — closes: the Town Clerk's certified results of the 16 May 2026 Annual Town Election.

curriculum

• Which courses a Lunenburg High student could actually choose from, in any year

  DESE publishes sections by SUBJECT AREA and never by course, so an elective ending is invisible inside a subject total that rose — a student told the School Committee on 7 February 2024 that Music Tech would be "non-existent", and the high school's Arts sections went from 24 that year to 25 the next. Both statements are true and this archive cannot hold them together. — closes: The district’s own Program of Studies for each year, which the High School Principal handed the School Committee on 6 March 2024, and the LHS master schedule presented to the School Committee on 4 December 2024. Neither is published.

• What is taught in the fifth of Lunenburg High sections DESE files as Miscellaneous

  51 of the high school’s 247 sections in SY2025, averaging 4.9 students, in a subject area DESE gives no course names for — and the single largest mover in the section count since SY2017. The only public account of one is a student describing a new elective with four on the register and eight on the first day. — closes: The LHS master schedule by year, or the district’s own Student Course Schedule submission to DESE, which holds the course records the published aggregate is built from.

• How many classes ran in each subject at the two elementary schools

  The counts swing by a factor of two between adjacent years with no matching change in enrolment, and two school-years report an average class size no primary school can have — 87.1 at Lunenburg Primary in SY2011 and 36.5 in SY2022, against a maximum of 24.1 everywhere else. The subject coding of a self-contained classroom is not stable enough to trend, so the elementary half of the district cannot be read at all. — closes: DESE’s Student Course Schedule submission detail for Lunenburg, which carries the course records behind the published aggregate.

• Why fewer Lunenburg High students take a history or a language course than before SY2020

  Three DESE series move together for both subjects — teacher FTE, student places and the share of the school taking one — and the fall in each begins in the years the school closed and reopened. A staffing decision, a change in what the timetable allowed, and students choosing differently all fit the same three numbers, and the state files nothing that separates them. — closes: the district’s course-request tallies for each year — what students asked for beside what ran. The High School Principal told the School Committee on 6 March 2024 that courses are whittled away once student choices are entered, so the tally exists at the moment the decision is made and is not published.

• How many classes ran in each subject at Lunenburg Middle School and Lunenburg High before SY2017

  Every trend on /what-courses-actually-ran is read inside an era, because for four years — SY2013 to SY2016 — Thomas C Passios was closed and every remaining school held one grade band higher, with grade 8 in the high school building. Four subjects move by more than ten points of participation into those years and back out of them without anything changing about what a ninth to twelfth grader was offered. So the file cannot say what the schools as they are now offered before SY2017 — the schools as they are now did not exist. — closes: nothing published can. DESE files by school and the schools changed; the question is only answerable by grade, and the state’s class-size collection carries no grade inside a school.

students

• Which school a Lunenburg student who left went to, by grade

  DESE publishes two files that each hold half of this and never the join. dese_attrition gives an attrition rate for every grade, every year, and names no destination. dese_residents_sending names every receiving district and the reason -- Monty Tech, school choice, charter, foster care -- and carries no grade at all. So the single largest measured departure in this town, about one in five of every eighth grade, cannot be attributed to anywhere: a vocational admission, a private school, a charter, a school choice transfer, an out-of-district placement and a family moving out of town are one number. — closes: one column. DESE's residents-sending file with a grade on each row, or the district's own October SIMS submission broken out by exit grade and exit reason, which the district files to the state every 1 October and does not report to its own School Committee.

• Why one in five Lunenburg eighth graders does not return for grade 9

  The rate is measured and stable across seventeen years, two building reorganisations and a pandemic. The reason is not measured anywhere. Monty Tech admits at grade 9 and Lunenburg is a member town there, so a child going there moves part of the town's bill rather than adding to it -- but nothing published links the two quantities, and the same rate is equally consistent with private and parochial admission at grade 9, with charter admission, with an out-of-district special education placement made at the transition, and with families moving. — closes: nothing anybody currently publishes. The nearest thing is an exit survey or an exit-reason tally kept by the district at the point a student is withdrawn, which is a record the district could choose to keep and report and does not.

• How many of the children counted as leaving a Lunenburg school moved out of town

  Attrition is leaving for ANY reason, so a family that moves to another state and a family that keeps its address and enrolls a child elsewhere are the same number. Without the split, the rate cannot be read as a statement about the schools at all -- and it is the reading every resident supplies. The town's own records could bound it: residency is what the district verifies at enrollment and what the assessor's and clerk's records track. — closes: the district's withdrawal records coded by reason (moved out of district / transferred within Massachusetts / private / home school), which is a field SIMS already requires the district to report on every exit.

money

• Why does the town open FY2018 owed $13,300 more than it closed FY2017 owing?

  THE TOWN PRINTS BOTH FIGURES AND THEY DO NOT MEET. A receivable balance rolls forward: what the town was owed on 30 June is what it starts the next year owed. Every other consecutive pair we can read ties to the cent (FY2020 through FY2024, one pair off by 2c). The Collector’s grand total on FY2017 page 45 reads $9,008,228.35 and on FY2018 page 54 reads $9,021,528.35 — checked on the pages, not off our extract. A prior-year adjustment, a reclassified charge and a typo all fit it equally and nothing published distinguishes them. — closes: the Collector’s receivable control reconciliation for FY2018, which is a report the accounting system can print.

• What was Lunenburg's certified new growth in fiscal 2012?

  THE TOWN’S OWN REPORTS GIVE TWO FIGURES AND DISAGREE ABOUT WHICH YEAR EACH BELONGS TO. The FY2013 report prints $19,124,255 against fiscal 2012. The FY2014 through FY2017 reports print $27,729,619 against fiscal 2012 and re-date $19,124,255 to fiscal 2013. Nothing in the archive resolves which labelling is right, and new growth is what the levy limit grows by, so the two readings imply different ceilings. — closes: DOR’s LA-13 new growth certification for FY2012.

• What added valuation produced fiscal 2009's $138,168 of new growth?

  THE ARITHMETIC ON THE PAGE DOES NOT WORK AND FIVE EDITIONS REPRINT IT UNCHANGED. $5,077,276 of added valuation, at the $12.23 rate printed on the same line, gives $62,095 — not the $138,168 beside it. The report’s own percentage-change column corroborates $138,168, so the added-valuation figure is the one that cannot be right. Recorded rather than corrected: which of the two the town meant is not something this archive can say. — closes: DOR’s LA-13 new growth certification for FY2009.

• Why does the same $1,085,674.86 appear twice in the town’s FY2015 opening receivables?

  THE TOWN’S OWN TWO BOOKS DO NOT MEET, AND THE AMOUNT IS EXACT. FY2014 closes MEADOW BETTERMENT at $1,085,674.86. FY2015 opens SEWER BETTERMENT $1,085,674.86 higher than FY2014 closed it, AND carries the same $1,085,674.86 again on a new MEADOW WOODS WATER BETT line. Both years tie to their own printed GRAND TOTAL, and line by line every other balance rolls forward near-exactly. A recoding of Meadow Woods into the sewer betterment line would explain it; so would a restatement; nothing published distinguishes them. — closes: the town’s FY2015 receivable roll-forward, or the MUNIS receivable detail for both years.

• Whose $92,709.54 is the unnamed line in the FY2015 receivables summary?

  THE REPORT PRINTS A FIGURE ON A LINE WITH NO NAME. $92,709.54 in the PAYMENTS column, under UTILITY LIENS ADDED TO TAXES, with an empty name cell — confirmed by rendering page 45 rather than off the scan. Seven of the eight columns tie to the cent without it and PAYMENTS is short by exactly that amount, so the figure is real and the name is what is missing. Carried in the data as (the page prints no name for this line). — closes: the Collector’s detail behind the FY2015 receivables summary.

• Does the town’s printed debt repayment grand total include all of its debt?

  IN THE FY2015 REPORT IT DOES NOT, BY THE EXACT AMOUNT OF TWO BONDS. The GRAND TOTAL PRINCIPAL row is short by the two MS-HS CONSTRUCTION bonds in every year column — $625,000 in 2016, $640,000 in 2017, $665,000 in 2018 — while both bonds are printed in the table directly above it. The school construction bonds are the second largest debt on the page. Nothing here establishes whether the row or the table is the intended figure. — closes: the Treasurer’s debt service schedule for FY2015, or the official statement for the MS-HS bonds.

• What was Lunenburg's certified assessed valuation and tax rate for fiscal 2021?

  THREE DOCUMENTS GIVE THREE ANSWERS AND NONE OF THEM IS THE TOWN'S OWN TABLE ADDING UP. The FY2021 annual report, page 38, prints four property classes totalling $1,742,668,758.00 under a Totals line that reads $1,548,256,450 -- and that Totals line, with its $28,054,406.88 levy, is to the cent the one the FY2020 report prints, where it foots FY2020's own four classes at FY2020's rate of 18.12. DLS's assessed-value series puts Lunenburg's FY2021 total at $1,628,643,589, which matches neither. And the rate is printed as 17.19 in every row of the table while the prose above it says the rate went from $18.12 to $17.74 for FY 2021. Until one of these is settled, no FY2021 valuation, levy or rate can be quoted from the annual report, and the class SHARES -- which /commercial-base is built on -- rest on a base the town does not state. The four class rows are internally consistent and are not in doubt: each one's valuation x 17.19 / 1000 is its own printed levy to the cent. — closes: the Division of Local Services Tax Rate Recapitulation (the certified recap, with the LA-4 assessed values) for Lunenburg, fiscal 2021, which DLS publishes per municipality per year on its Gateway/Municipal Databank.

• Does the FY2018 debt repayment schedule's printed grand total account for every issue's INTEREST?

  IN FY2019 IT DOES NOT, BY THE EXACT INTEREST OF ONE ISSUE. Pages 48-51 of the FY2018 annual town report print sixteen debt issues against FY2019-FY2047 and, separately, GRAND TOTAL rows per year. Read in full, the issues reconcile to those grand totals within $2 in all 29 years -- the schedule shows whole dollars over amounts carrying cents -- except FY2019, where the issues' interest exceeds the printed GRAND TOTAL INTEREST by 272 and TOTAL DEBT by the same 272 while principal ties EXACTLY. 272 is precisely the FY2019 interest of MASS WATER POOL 4*, whose own row closes to the dollar (10,409 + 272 = 10,681) and whose 10,409 of principal IS inside the grand total. So one of the two printed figures is wrong and the document does not say which; the archive publishes both and asserts neither. This is the same defect the FY2015 row in this register names for PRINCIPAL, in a different year and a different column, and the two may be worth merging. - closes: the Treasurer's own debt service schedule for FY2019, or the Massachusetts Clean Water Trust (formerly MWPAT) amortization schedule for Pool 4, either of which states that loan's FY2019 interest and admin fee independently of the town's own summary.

• What were Lunenburg’s fiscal 2011 class valuations, where the town’s own two printings disagree by $7,295,100?

  The FY2011 annual town report prints the same quantity twice on one page (PDF p12, printed folio 9) and the two do not reconcile. Fiscal 2011 Assessments and Revenues by Major Property Class puts Commercial at 64,514,700 and foots its tax levy to the printed 18,006,811.48 to the cent; the Fiscal Year 2011 Abstract of Assessments below it puts 300 - 393 Commercial at 56,208,300 and 600 - 821 Chapter 61, 61A, 61B at 1,011,300, so its twelve classes sum to 1,106,987,789 against its own printed TOTALS of 1,114,282,889. The residual is 7,295,100 and it sits entirely in the commercial class: the abstract’s eight residential lines sum to 1,000,004,500, which is the first table’s Residential valuation to the dollar, and its account count ties exactly at 5,185, so no row is missing from the reading. Both cells were re-read off the page’s own 400 dpi bitmap at 6x. Which printing is wrong is not established, and the report itself settles nothing — the first table’s own TOTALS cell is misprinted 1,114,2828,889, ten digits. — closes: the Assessors’ certified fiscal 2011 Tax Recapitulation (DOR Form LA-4/LA-5) for Lunenburg, which states valuation by class as the Department of Revenue certified it and is the only document that outranks both printings.

• What are the $25,276.56 of fines and $3,110.00 of fees that FY2012 receipts counts but does not name?

  The FY2012 annual town report, PDF p58 (FY 2012 RECEIPTS - JUNE 2012), prints 91 detail lines and a nine-category SUMMARY OF RECEIPTS. The nine categories foot to the printed GRAND TOTAL of $28,471,842.32 exactly; the 91 lines sum to $28,443,455.76, short by $28,386.56. Seven of the nine categories are reproduced to the cent by a partition of those lines, which puts the whole shortfall in two: FINES & FORFEITS, where the page prints only PARKING FINES $2,120.25 and DOG FINES $2,550.00 against a category total of $29,946.81, and FEES, short by $3,110.00. The page was swept end to end and prints no further row, so this is the document omitting detail rather than our reading missing it. Court fines are the obvious candidate for the larger figure and nothing in the report says so. — closes: the Town Accountant's FY2012 general ledger revenue detail (MUNIS), which lists every receipt account by code rather than by the summary categories this page prints.

• Which figure is wrong in the FY2013 Omnibus Budget: the eleven section totals, or the $600 larger Total Omnibus?

  The budget Town Meeting voted under Article 8 is printed in the FY2012 annual town report, PDF pp.81-82. Every one of its eleven sections foots to its own printed subtotal exactly, and those eleven subtotals sum to $27,488,841.00 against a printed Total Omnibus of $27,489,441.00 — six hundred dollars more. Both figures were re-read upright at full scale off the page's own 518 dpi bitmap, so this is the document disagreeing with itself and not a misreading. The $600 could be a line the page does not print, a section total understated, or the grand total overstated, and nothing on those two pages distinguishes them. — closes: the Town Clerk's certified copy of the Article 8 motion and the FY2013 budget handouts it refers to (Refer to Handouts on pages 27 – 30), which the annual report does not reproduce.

• What the $600.00 is between the FY2013 receipts page's own detail and its own GRAND TOTAL

  PDF page 63 of the FY2013 annual town report prints FY 2013 RECEIPTS - JUNE 2013: 91 detail lines in three label/amount column-pairs, and beside them a nine-line SUMMARY OF RECEIPTS footing to a printed GRAND TOTAL of $29,744,547.31. The nine summary categories sum to that GRAND TOTAL EXACTLY. The 91 detail lines sum to $29,743,947.31 -- $600.00 less. The reading is not the problem: the page has no text layer and was read off its own 400 dpi bitmap twice, and three of its printed detail blocks foot to their matching summary line to the cent (LICENSES/PERMITS to 174,193.45, STATE REVENUE to 7,010,803.56, FINES & FORFEITS to 27,405.00), which bounds the difference away from 29 of the 91 lines and leaves it inside the fees and departmental revenue lines. The page never states that the detail is exhaustive of the summary, so this is NOT recorded as the town's arithmetic failing -- it may be one receipt listed only in the summary, a line printed $600 low, or two different universes. Nothing on the page distinguishes them. sources/data/page-reads/fy2013-p63-*.csv. — closes: the Town Accountant's FY2013 year-end revenue report out of the accounting system (the MUNIS revenue ledger for fiscal 2013, which this project already holds for FY2026 as sources/munis-ledgers/revenue/), which lists receipts by account and foots itself.

• Does FY2024's debt service include the Mass Water Pool 9 balloon payment, or not?

  The FY2013 annual town report's debt repayment schedule (PDF pages 81-82) prints, in its 2024 column, GRAND TOTAL PRINCIPAL 846,635 and GRAND TOTAL INTEREST 120,310 -- 966,945 -- against a printed TOTAL DEBT of 548,241. The document is out by 418,704 in that one column, and by the same 418,704 in its own TOTAL column (24,298,664 + 7,307,840 = 31,606,504 against a printed 31,187,800). Every other year of the 34 closes exactly. The figures were read off the page's own 400 dpi bitmap twice. 418,704 is also the principal the FY2014 annual report prints for MASS WATER POOL 9* in the FY2024 column -- the only row in that report's 156-row schedule with a principal and NO interest and NO total beside it (sources/data/page-reads/fy2014-p40-debt-schedule.csv). So the discrepancy is located on one borrowing, and it is still not known whether the town excludes that payment from debt service on purpose because the Trust subsidises it, or whether one of the two printed figures is simply wrong. The difference matters: it is 76% of what the town otherwise shows itself owing in FY2024. The per-issue rows of the FY2013 schedule are NOT read -- that limit is already registered as “What does each individual debt issue cost the town in a given year” and is not refiled here. — closes: the Treasurer's debt service schedule for the Mass Water Pollution Abatement Trust loan numbered 9, showing the FY2024 payment and any Trust subsidy against it.

comparison

• Which towns does Lunenburg consider comparable to itself, for tax and budget purposes?

  This project compares Lunenburg with five peer towns and that set is OURS. Nothing the town has published names a cohort: a 2023 Finance Committee minute cites "the 8 communities that were surveyed" for a salary study and never lists them, and no budget document, capital plan or assessors’ report names a comparison group. So every "compared with similar towns" statement on either side rests on a set nobody has agreed. — closes: the eight-community list behind the 2023 salary survey, which the Town Administrator or the bargaining unit that commissioned it holds.

• How is a regional school district’s assessment apportioned between its member towns, and what does each town actually appropriate for schools?

  NARROWED 26 September 2026. This used to say a regional town could not be compared at all. That was asking at the wrong level: Chapter 70 is computed per MUNICIPALITY whatever district its children attend, so DESE publishes, for all 351 towns, the foundation budget for that town’s own children, the contribution the town is REQUIRED to make and by subtraction what the state adds — Ashby is required to raise $7,819 per child, Groton $10,748, Lunenburg $8,571. What is still unpublished is the step between: how the district divides its assessment among members, and what each member town actually appropriates, which may be well above the required minimum. — closes: each regional district’s annual assessment schedule, voted by its member towns’ meetings and filed with DESE.

• Do two districts put the same purchase on the same side of the line, so that their per-pupil figures compare like for like?

  DESE’s per-pupil figure counts functions 1000-5000 and leaves out capital (7000) and debt service (8000); its expenditures-by-function file carries no 6000, 7000 or 8000 row for any district, so no district’s figure contains capital. But the line between counted and capital is a coding rule, and DESE’s FY2025 End of Year Report form draws it: instructional equipment, “including vocational equipment and science lab equipment irrespective of unit cost”, goes in 2420 and is counted; non-instructional equipment over $5,000 and extraordinary maintenance over $150,000 a project go in 7000 and are not. Whether every district applies that rule the same way cannot be seen: DESE publishes each district’s reported totals, not the purchases behind them. So a difference between two districts on an equipment or maintenance line may be spending, or may be coding. — closes: each district’s own End of Year Financial Report workbook with its 7000 schedule, lease schedule and depreciation schedule, and DESE’s desk-review or audit findings on them.

extraction

• What does each individual debt issue cost the town in a given year, and which of them fall off when?

  FY2022 NO LONGER BELONGS TO THIS GAP: its schedule was read by issue on 29 September 2026 -- 14 borrowings against 25 fiscal years, 184 issue-years, each closing on its own printed TOTAL row and reconciling to the GRAND TOTAL block on pages 46 and 48. The FY2023 AND FY2024 annual reports' debt repayment schedules each print about fifty debt issues against twenty-three fiscal years, roughly 1,200 figures, and only its GRAND TOTAL block has been read: principal, interest, principal and interest, MWPAT admin fees and total debt, FY2025 to FY2047, which foot to the schedule's own TOTAL column. So the town's debt SERVICE by year is established and the COMPOSITION of it is not — a resident cannot see which bond ends in which year, or what the Meadow Woods water general obligation costs on its own. — closes: nothing further is needed from the Town. The figures are printed on page 53 of the FY2023 report and page 37 of the FY2024 report and have simply not been transcribed.

• How many financial pages do the annual town reports actually contain?

  The page tracker's universe is built from what RECOGNITION found: a page is listed only if fifteen money figures were recognised on it. FY2024 page 189 prints ninety-six wages and recognition found ZERO, so it was not unfinished, it was ABSENT -- and five of the eight pages of that wage run were. FY2024 was reported as 22 financial pages and is at least 28. A page a person has now read is added to the map, so the eight are on it; nothing finds the pages nobody has looked at yet, and the same failure is unmeasured in the other fifteen reports. A screen for candidates -- many boxes, almost no recognised money -- returns about 1,800 pages archive-wide, but it catches every page of ordinary prose too and is a finding aid rather than a count. — closes: nothing from the Town. Every page is already held; what is missing is somebody looking at the ones recognition could not read.

• Which of the town's two FY2023 statements of what the special revenue and enterprise funds held is right?

  The FY2023 combined balance sheet gives special revenue fund equity as 6,131,303.68 and enterprise fund equity as 3,812,502.24, together 9,943,805.92. The special revenue schedule in the SAME annual report prints a grand total carried forward of 10,031,099.78. They are 87,293.86 apart. The comparison ties to the cent in FY2011, FY2013, FY2019 and FY2022, and the balance sheet's own arithmetic is complete -- every column foots to its own printed totals and liabilities plus equity equals assets -- so this is two published tables disagreeing and not a reading error. — closes: the Town Accountant's FY2023 trial balance, which would say which of the two the books support.

• What did the town receive in FY2023, receipt line by receipt line?

  The FY2023 receipts page -- annual report page 25 -- cannot be read. It is a photograph embedded at 93 dpi, and it packs about 150 line items across three column-pairs of a landscape page, so the digits are below the resolution of the stored image: 588 and 568, 405 and 465, 009 and 000 are not distinguishable at any magnification, because the detail is not in the file to magnify. Reading the SUMMARY OF RECEIPTS block gives a total $20 from the printed GRAND TOTAL of $45,549,438.56, consistent with exactly one misread digit, and choosing the digit that makes it close would be fitting the answer to the check. For contrast the same report stores page 50 at 183 dpi and page 53 at 156 dpi, and both are legible. RE-CHECKED 1 October 2026 off the page's own embedded bitmap rather than off the recognition cache, with the grid measured by projection and the amount columns rendered at 20x: the labels are legible and the figures are not, so what the town received money FOR in FY2023 is readable and how much is not. And a better scan is not a remedy that exists. Of the two publisher addresses on record, DocumentCenter/View/4131 now returns HTTP 404 and ArchiveCenter/ViewFile/Item/160 returns 18,274,143 bytes -- byte-for-byte the copy this archive already holds. — closes: the Treasurer's own FY2023 receipts file, or a MUNIS revenue report for FY2023; this archive holds neither, and no MUNIS revenue export for any year.

• What was appropriated on the supplementary pages that do not foot to their own subtotals?

  1 page of what Town Meeting appropriated in the annual town reports is READ AND REFUSED: an extractor reached it, could not tie the rows to a total the page itself prints, and published nothing. So a query about what Town Meeting appropriated returns silence for FY2017, and the silence is ours rather than the town’s — every one of these pages IS printed. Refusals and their reasons are in sources/data/appropriations-supplement-refused.csv; the pages are listed in sources/data/annual-report-pages.csv where state is refused. Re-running a model over them returns the same refusal and pays for it. — closes: the subtotal identity those pages state about themselves, applied per column.

• What did the town owe on each bond issue, principal and interest, in the years whose repayment schedule cannot be read?

  1 page of the debt repayment schedule in the annual town reports is READ AND REFUSED: an extractor reached it, could not tie the rows to a total the page itself prints, and published nothing. So a query about the debt repayment schedule returns silence for FY2011, and the silence is ours rather than the town’s — every one of these pages IS printed. Refusals and their reasons are in sources/data/debt-repayment-detail-refused.csv; the pages are listed in sources/data/annual-report-pages.csv where state is refused. Re-running a model over them returns the same refusal and pays for it. — closes: a reader of scripts/extract_debt_tables.py measuring the year header and splitting the merged year boxes by the pitch the page itself sets, then publishing only the columns where PRINCIPAL + INTEREST foots to the printed TOTAL for that issue.

• What cash did the Treasurer hold in the years whose column does not foot to the page’s own printed total?

  1 page of the Treasurer’s cash on hand in the annual town reports is READ AND REFUSED: an extractor reached it, could not tie the rows to a total the page itself prints, and published nothing. So a query about the Treasurer’s cash on hand returns silence for FY2019, and the silence is ours rather than the town’s — every one of these pages IS printed. Refusals and their reasons are in sources/data/extraction-blocked.csv; the pages are listed in sources/data/annual-report-pages.csv where state is refused. Re-running a model over them returns the same refusal and pays for it. — closes: the column ruler in the treasurer’s-cash reader; the page prints the total the column must reach, and sources/data/extraction-blocked.csv records both figures.

• What did the town pay its people, name by name, in the years whose payroll listing produced nothing?

  1 page of gross wages, name by name in the annual town reports is READ AND REFUSED: an extractor reached it, could not tie the rows to a total the page itself prints, and published nothing. So a query about gross wages, name by name returns silence for FY2012, and the silence is ours rather than the town’s — every one of these pages IS printed. Refusals and their reasons are in sources/data/gross-wages-refused.csv; the pages are listed in sources/data/annual-report-pages.csv where state is refused. Re-running a model over them returns the same refusal and pays for it. — closes: a Vision RE-READ of the page, which is free and is what closed FY2019: its seven pages held 871 boxes in our cache against 1,753 on a re-read, and two of them held the surname column and no money at all. The refusals that survive a re-read are scan defects rather than layout — FY2019 p200 loses a 21-row band that produced no box at scale 2, 3, 4, 5, 6 or 8 — and those close only with a better copy of the page from the Town. Note that the town stopped printing the department beside each name after FY2016, so these pages bound the question rather than settling it.

• What is on the pages a model read and returned no rows for?

  2 pages of pages with no extractor of their own in the annual town reports are READ AND REFUSED: an extractor reached them, could not tie the rows to a total the page itself prints, and published nothing. So a query about pages with no extractor of their own returns silence for FY2013, FY2017, and the silence is ours rather than the town’s — every one of these pages IS printed. Refusals and their reasons are in sources/data/read-and-refused.csv; the pages are listed in sources/data/annual-report-pages.csv where state is refused. Re-running a model over them returns the same refusal and pays for it. — closes: reading the refusal note in sources/data/annual-report-reads/ and deciding whether the page holds a table at all — some of them correctly do not.

• What was the town owed, account by account, in the years whose receivables page will not foot?

  3 pages of what is owed to the town in the annual town reports are READ AND REFUSED: an extractor reached them, could not tie the rows to a total the page itself prints, and published nothing. So a query about what is owed to the town returns silence for FY2013, FY2017, FY2020, and the silence is ours rather than the town’s — every one of these pages IS printed. Refusals and their reasons are in sources/data/receivables-refused.csv; the pages are listed in sources/data/annual-report-pages.csv where state is refused. Re-running a model over them returns the same refusal and pays for it. — closes: scripts/extract_receivables.py made to sum to the totals those pages print; the pages are legible and they grade the extractor.

• What did the Town have receivable at 30 June 2020, line by line, across all three pages of the summary?

  receivables.csv holds 43 rows for FY2020 -- 31 on annual-report page 47, 12 on page 48 and NONE on page 49 -- while receivables-reconciliation.csv records the same run as 87 rows and says it ties to its printed GRAND TOTAL on all eight columns. Page 47 alone prints about 44 rows. Page 49 was read by eye on 2 October 2026 and all eleven of its rows (ten detail plus the run's GRAND TOTAL) close on the identity the page states, so the page is readable and the extractor's refusal -- no single sign convention proves the page (2 of 5 rows at best) -- is a statement about the instrument. Summing the 43 published rows plus the ten read from page 49 gives $375,657.49 forward against a printed GRAND TOTAL of $9,108,033.04, so the published extract is short of what the pages print by most of the table. WHAT THIS DOES NOT ESTABLISH: whether the missing rows were dropped by the reader or were never published -- nobody has read pages 47 and 48 by eye. -- closes: nothing from the Town is needed. The remedy is ours: read pages 47 and 48 of the FY2020 annual town report the way page 49 was read, and foot all three pages to the printed GRAND TOTAL.

• RESOLVED 2 October 2026 — PDF page 51 of the FY2019 debt schedule was never absent from the tracker, and it is now read

  CLOSED, AND BOTH HALVES OF THE ORIGINAL CLAIM WERE WRONG. The tracker held 2019,51,debt,3,unread,... — the page was unread, not absent, so this row was a statement about our reading of the tracker rather than about the archive (learning-loop entry 12). And it is not "the missing middle": p51 carries the TOTAL column every issue is proved against, plus the second half of the fiscal years. Its figures=50 was a RECOGNITION count, low by 2.7x — the page prints 136. All four pages (49-52) were read off their own JPEG2000 bitmaps at 150 dpi on 2 October 2026: 879 figures, 34 files, 17 groups, four nested identities, and the printed corner closing on itself exactly — 46,568,549 + 18,922,925 = 65,491,474 and + 25,861 = 65,517,335. — closes: nothing further; the question was ours and it is answered.

• Can we trust what Town Meeting appropriated, article by article enough to chart what the town voted to spend, year over year?

  311 of 5,345 rows in report_appropriations are reconciled to a total the document itself prints — 5,034 failed that check and 0 were never checked. The figures are read and published, so a query returns them and they look confident; only the status column says otherwise. Every page was surveyed before anyone tried: of 30 years, 1 need real PDF geometry rather than text, 16 are messy and 13 are clean — mirrored layouts, rows offset from their own names, a missing fund-name column, and column counts that change between years. 30 of 30 years do print a grand total to reconcile against, so the anchor exists. — closes: a geometry-aware extractor for this table family, tied to the printed total on each page. The per-year survey is in sources/data/extraction-plan.csv.

• Can we trust the capital projects and what they cost enough to chart what the town has built and what it paid?

  48 of 618 rows in report_capital_projects are reconciled to a total the document itself prints — 544 failed that check and 26 were never checked. The figures are read and published, so a query returns them and they look confident; only the status column says otherwise. Every page was surveyed before anyone tried: of 19 years, 2 need real PDF geometry rather than text, 9 are messy and 8 are clean — mirrored layouts, rows offset from their own names, a missing fund-name column, and column counts that change between years. 16 of 19 years do print a grand total to reconcile against, so the anchor exists. — closes: a geometry-aware extractor for this table family, tied to the printed total on each page. The per-year survey is in sources/data/extraction-plan.csv.

• Can we trust the debt schedule enough to chart what the town owes and when it falls due?

  75 of 366 rows in report_debt are reconciled to a total the document itself prints — 99 failed that check and 192 were never checked. The figures are read and published, so a query returns them and they look confident; only the status column says otherwise. Every page was surveyed before anyone tried: of 43 years, 11 need real PDF geometry rather than text, 13 are messy and 19 are clean — mirrored layouts, rows offset from their own names, a missing fund-name column, and column counts that change between years. 40 of 43 years do print a grand total to reconcile against, so the anchor exists. — closes: a geometry-aware extractor for this table family, tied to the printed total on each page. The per-year survey is in sources/data/extraction-plan.csv.

• Can we trust gross wages, name by name enough to chart what the town pays its people in total?

  0 of 3,547 rows in report_gross_wages are reconciled to a total the document itself prints — 0 failed that check and 3,547 were never checked. The figures are read and published, so a query returns them and they look confident; only the status column says otherwise. Every page was surveyed before anyone tried: of 16 years, 6 need real PDF geometry rather than text, 9 are messy and 1 are clean — mirrored layouts, rows offset from their own names, a missing fund-name column, and column counts that change between years. NO year of this family prints a grand total, so there is no arithmetic on the page to foot a row against and no extractor can supply one. — closes: nothing we can build. What stands in for a total here is the PAIRING and the order the list prints itself in, both of which prove the assignment and say nothing about the digits; a reconcilable figure needs a total from the Town. The per-year survey is in sources/data/extraction-plan.csv.

• Can we trust the stabilization and trust fund balances enough to chart how much the town holds in reserve, and how that has moved?

  12 of 644 rows in report_trust_funds are reconciled to a total the document itself prints — 418 failed that check and 214 were never checked. The figures are read and published, so a query returns them and they look confident; only the status column says otherwise. Every page was surveyed before anyone tried: of 17 years, 10 need real PDF geometry rather than text, 5 are messy and 2 are clean — mirrored layouts, rows offset from their own names, a missing fund-name column, and column counts that change between years. 16 of 17 years do print a grand total to reconcile against, so the anchor exists. — closes: a geometry-aware extractor for this table family, tied to the printed total on each page. The per-year survey is in sources/data/extraction-plan.csv.