Enrollment & Cost

Why the money doesn't follow the child

It is natural to expect that a third fewer students should cost a third less. That expectation would be right if schools bought things one student at a time. Mostly they don't.

Nearly all of a school budget buys capacity: a teacher standing in front of a room, a bus driving a route, a boiler heating a building, a principal running a school. Those costs come in whole units. A class of 24 that becomes a class of 22 still needs one teacher, one room, and one heating bill. Nothing is saved until the loss is big enough to remove the whole unit — one section, one route, one wing of a building.

The arithmetic of a slow decline

Colchester lost 1,032 students over 16 years. That is a big number in total, and a small number per year in each place it lands:

Each grade sits in one of the district's four school buildings, so that loss of about five children lands in a single grade in a single school — it does not get divided again. These are averages, not a grade-by-grade count: some years and some grades lost more, some lost none. But at that scale, a year of enrollment loss usually does not empty a classroom. It makes existing classrooms slightly smaller. That shows up in the student-to-teacher ratio, which fell from about 14.1 students per teacher to about 11.8, but not in the bill, because the same number of rooms stayed open.

Over enough years the losses do add up to whole sections, and staffing has come down: teaching positions fell 18.2% over the same window. But staff reductions arrive in steps, several years apart, while the cost of everything the district keeps rises every year.

Source for the arithmetic above: enrollment endpoints from the enrollments series (3,267 in FY 2008-09 → 2,235 in FY 2024-25); 13 grades = K–12; four school buildings = Colchester Elementary, Jack Jackter Intermediate, William J. Johnston Middle, and Bacon Academy, the schools covered by the district's EdSight staffing files (METHODOLOGY.md). Student-to-teacher figures are enrollments ÷ teacherFTE.

What one year of "changing nothing" costs

Suppose a district changes nothing at all: same staff, same programs, same buildings. The budget still goes up, because raises are set in contracts and prices rise on their own.

Using the FY 2018-19 budget structure above, and two published outside rates — 2.61% a year for school-staff compensation (the BLS Employment Cost Index for state and local K–12 schools) and 2.47% a year for general prices (the compound CPI-U rate across FY 2007-08 → FY 2024-25) — one year of standing still looks like this:

Now add the staff reduction that a year of enrollment loss would justify. At the district's own FY 2024-25 ratio of about 11.8 students per teacher, 65 fewer students is about 5.5 teaching positions. A teaching position costs roughly $107,000 all-in — the FY 2024-25 average teacher salary of $85,500 plus the district's 25.1% benefit load:

That is the mechanism in one line. Even when a district cuts staff fully in proportion to the students it lost, the raises and price increases on everything it kept are larger than the savings. The budget goes up in a year when enrollment went down — not because the savings are zero, but because they are smaller than the increase.

Rates used. 2.61%/yr = BLS Employment Cost Index, state and local government K–12 schools, total compensation (FRED CIS3016110000000I) — the same central rate used in the "do nothing" baseline on the Budget Trends page. 2.47%/yr = the compound annual CPI-U rate across FY 2007-08 → FY 2024-25 (index 100 → 151.3). Benefit load of 25.1% = FY 2018-19 BOE benefit lines ($6.41M) ÷ BOE salary lines ($25.50M). Average teacher salary of $85,500 (FY 2024-25) is a compiled estimate from GovSalaries, Niche and ConnecticutTeach benchmarks, not a district payroll extract — see data-dictionary.md (avgTeachSal).

"So how many teachers would you have to cut?"

That is the fair follow-up question, and it has a number. Holding the school budget flat in dollars for one year means finding the whole $1,046,000 that raises and inflation add. At about $107,000 per teaching position, all-in:

Put in the same terms as the question: breaking even takes roughly one teaching position for every 6.6 students lost. Measured against the district's own staffing ratio — the 5.5 positions that 65 students' worth of enrollment actually frees up — breaking even needs about 1.8 times that many. Measured against the one-teacher-per-20-students rule in the first row, it needs about three times as many.

Five years of doing exactly that

Suppose the district holds its budget flat in dollars for five straight years, takes every dollar of the savings out of teaching positions, and enrollment keeps falling at its recent average of about 65 students a year:

Three things fall out of that table, and they cut in different directions:

• It is a large reduction. 45 teaching positions over five years is about 24% of the district's current teaching staff, and would leave fewer teaching positions than in any year on record here (the low is 184.4 FTE in FY 2017-18).
• It is arithmetically possible. Holding the budget flat does not require anything exotic — it requires cutting roughly 9 to 10 teaching positions a year, every year.
• Students per teaching position would still be lower than in FY 2008-09. The ratio rises from 11.8 to 13.3 over the five years, and FY 2008-09 was 14.1 — because enrollment keeps falling at the same time the staff does.

Method. Same FY 2018-19 structure and rates as the section above: pay and benefits ($31.91M) grow 2.61%/yr, everything else ($8.64M) grows 2.47%/yr, and each year enough teaching positions are removed at the loaded cost of a position ($106,994, itself escalated 2.61%/yr) to return the total to the prior year's dollar figure. Cuts shrink slightly each year because the pay base they are applied to also shrinks. Starting teaching FTE 189.5 and general-education teacher FTE 161.5 are the FY 2024-25 values from EdSight (teacherFTE, genEdTeach); enrollment falls 64.5/yr, the FY 2008-09 → FY 2024-25 average. This is an illustration of one arithmetic path, not a recommendation, a forecast, or a plan anyone has proposed.

What this page does and doesn't say

It says: falling enrollment produces much smaller budget savings than the headcount drop implies, because most school costs are per-classroom, per-route and per-building rather than per-student — and because the annual cost of raises and inflation on everything the district keeps is larger than the savings from the staff a single year of enrollment loss would release.

It does not say: that no savings are possible, that any particular budget was the right size, or that a resident who expected costs to fall was being unreasonable. The expectation is a sensible one; it just runs into how the costs are actually structured.

Both of these numbers describe the same budget, and both are correct:

• The school budget is up 32.9% in dollars since FY 2008-09 while enrollment fell 31.6%.
• The school budget is down 8.4% after inflation over exactly the same years.

Both are arithmetic on the same adopted budgets. The first counts dollars; the second counts what those dollars buy (see nominal vs. real). This site publishes both, and every series used on this page is listed in data-dictionary.md so anyone can recompute them.

Related pages: the Budget Trends page has the town-wide "do nothing" baseline and the staffing charts; the Education Comparison page puts Colchester's per-pupil spending next to peer towns; the Insurance page covers the employee-insurance lines that drive much of the benefits growth; the Glossary defines every term used here.