The District Ledger

Is your school district thinning?

Every school district in the twenty metros this site tracks, measured the same way: whether teaching staff is falling faster than enrollment. 3,165 districts, 17,617,531 students.

NCES Common Core of Data2020-21 → 2024-25Published August 2026

What this measures

A district that loses students and loses teaching staff in proportion is absorbing — it is resizing to the population it actually serves. A district where teacher staffing falls faster than enrollment is thinning: fewer adults per child in the classrooms that remain. That distinction does not appear in any budget document, and it is not the same thing as a deficit. A district can run a large deficit and still be absorbing; a district can be growing, well-rated and financially sound and still be thinning.

3,165

Districts covered

2,559

With two-year data

40%

Thinning

17.6M

Students

Four things this data says

The metros where staffing fell behind are the growing ones, not the shrinking ones. Austin, Charlotte, Dallas–Fort Worth, San Antonio and Houston all added students faster than they added teachers. The pressure there is an inability to hire, not a decision to cut.
The big shrinking metros protected their teachers. Los Angeles lost 7.7% of its students and only 2.4% of its staff — students per teacher improved from 22.9 to 21.6. Philadelphia, Boston, New York and Chicago show the same shape. They are running deficits and closing buildings, and the classroom ratio still moved in the students' favour.
A budget deficit does not tell you what happened in the classroom. Several districts with the largest deficits on this site are staffing ahead of enrollment. The fiscal story and the staffing story are separate measurements, and conflating them is the most common error in coverage of school budgets.
Texas is the concentration. Four of the seven metros where staffing fell behind are Texan, and about 60% of Texas districts are thinning against roughly 35% nationally — the clearest state-level divergence in the file.

The national picture, stated first because it cuts against the alarming reading. Across every US district with comparable data, students fell 0.2% while teaching staff rose 1.3%. Nationally, staffing has been protected, not cut. Roughly 35% of individual districts are thinning and 56% are absorbing. This page exists to show where the exceptions are and how large they are — not to argue that schools everywhere are hollowing out, because the data does not say that.

What this means in an actual classroom

Take a district with 1,000 students and 60 teachers — about 17 students per teacher. Four years later it has 950 students and 52 teachers. Enrollment fell 5%; teaching staff fell 13%. The ratio is now about 18.3. Same buildings, same course catalogue on paper, but roughly two more children in every room, every period, all year. That is the whole of what this measurement describes. It is arithmetic, not a verdict.

What the change tends to look like from inside a school, in the order it usually arrives:

Class sizes rise first, because it is the adjustment that requires no announcement.
Sections get consolidated. Two sections of the same course become one, and students who need it during a particular period can no longer take it. Scheduling conflicts do the work that a formal cut would have done visibly.
Specialist and advanced courses go earliest, because they have the smallest enrollments and the hardest-to-replace teachers. Physics, chemistry, computer science, upper-level mathematics and the arts sit at the front of that queue.
More classes are taught out-of-field — by a teacher certified in something else. In high-poverty schools nationally this reaches roughly 45% in the physical sciences and 58% in computer science, against about 22% in mathematics.

A counter-signal, because the picture is not uniformly bad. Over the same period, students per school counsellor improved nationally, from about 440 to 395 — still well above the 250-to-1 that the American School Counselor Association recommends, but moving in the right direction. Districts are making choices about where to protect staffing, and counselling is one place many of them protected.

What thinning is not

It is not a quality rating, and a thinning district is not a failing one. Most districts on this page are rated acceptable or better by their own states. It is not a judgement of teachers, who are the people absorbing the change. It is not the same as a budget deficit — a district can run a large deficit while staffing ahead of its enrollment, and several on this page do. And it is not necessarily a failure of management: a district losing students has to resize eventually, and doing so deliberately is competent, not negligent. What the measurement isolates is narrow — whether the resizing is falling harder on staffing than on enrollment — and it is published because no one else publishes it, not because it settles anything.

Why this reaches the job market

Parents ask a version of this question: does what happens in my child's school now show up in their working life later? The honest answer is that four things are measured, in sequence, and the links between them are not proven — but they are worth seeing next to each other.

Districts are thinning — the measurement on this page.
Advanced and STEM instruction thins first. Where the course survives it is more often taught out-of-field; in rural districts it frequently is not offered at all — only about 30% of remote rural high schools offer calculus, and roughly half offer physics, against 72% and 79% at the smallest suburban schools.
The entry point into professional work has narrowed. Recent college graduates are unemployed at 5.63% against 3.01% for all college graduates — a reversal of the postwar pattern that began around 2018.
Employers are buying experience over potential. Entry-level job postings fell 7.5% over the past year while senior postings rose 14.7%.

What is not claimed. That step one causes step four. These are four separate measurements from four separate sources, and no analysis on this site demonstrates a causal chain between a district's staffing ratio and a graduate's job search a decade later. What can be said is narrower and still worth saying: the pathway from childhood into working life is narrowing at several points at once, and this page measures the earliest of them — the one that happens while a child is still in school and while a family can still respond to it. The later points are documented in The Degree Still Works. The Pipeline Doesn't. and Why Companies Stopped Hiring Entry-Level.

The twenty metros — and the pattern nobody expects

The single most useful number is not how many districts are thinning. It is the gap: how much faster teaching staff grew or shrank than the student body it serves, across the whole metro. A negative gap means staffing fell behind students. A positive gap means staffing was protected relative to students.

Nationally, students fell 0.2% while teaching staff rose 1.3% — a gap of +1.5 points. Across the country as a whole, staffing has been protected. That is the baseline every row below should be read against, and it is the opposite of what most coverage of school budgets would lead you to expect.

Where staffing fell behind the students

These are, almost entirely, the fastest-growing metros in the country. Austin added 7.6% more students and only 2.5% more teachers. Charlotte added 4.3% more students and lost teachers outright. The problem here is not decline — it is that districts could not hire fast enough to keep pace with the families arriving.

Metro	Students Δ	Teachers Δ	Gap	Students per teacher	Districts	% thinning
Austin	+7.6%	+2.5%	-5.1	14.0 → 14.7	50	64%
Charlotte	+4.3%	-0.2%	-4.5	15.3 → 16.0	60	54%
Miami	-1.9%	-5.3%	-3.4	18.5 → 19.2	6	40%
Dallas–Fort Worth	+3.8%	+0.8%	-3.0	14.5 → 14.9	150	65%
Seattle	+0.5%	-2.0%	-2.5	17.7 → 18.1	51	69%
San Antonio	+4.7%	+3.0%	-1.7	14.9 → 15.1	56	69%
Houston	+3.4%	+2.2%	-1.2	15.2 → 15.4	92	62%

Where staffing was protected as students left

The mirror image, and mostly the large coastal and Northeastern metros. Los Angeles lost 7.7% of its students and only 2.4% of its teachers — students per teacher actually improved, from 22.9 to 21.6. These districts are shrinking, often painfully, and running large deficits while doing it. But the classroom ratio moved in the students' favour.

Metro	Students Δ	Teachers Δ	Gap	Students per teacher	Districts	% thinning
Las Vegas	-4.4%	+14.3%	+18.7	21.9 → 18.3	1	0%
Los Angeles	-7.7%	-2.4%	+5.4	22.9 → 21.6	343	28%
Atlanta	+1.1%	+5.0%	+3.9	14.8 → 14.3	58	19%
Washington DC	+1.6%	+5.4%	+3.8	13.9 → 13.4	55	26%
Philadelphia	-0.3%	+3.1%	+3.4	13.5 → 13.0	281	34%
SF Bay Area	-4.9%	-2.8%	+2.2	21.7 → 21.2	244	34%
Chicago	-3.3%	-1.2%	+2.1	17.0 → 16.6	336	33%
Boston	+0.6%	+2.6%	+2.0	12.2 → 12.0	275	34%
Jacksonville	+5.4%	+6.8%	+1.3	18.0 → 17.8	6	17%
New York	-2.4%	-1.1%	+1.3	12.4 → 12.2	813	39%

Roughly in step

Metro	Students Δ	Teachers Δ	Gap	Students per teacher	Districts	% thinning
Phoenix	-1.9%	-2.4%	-0.6	17.9 → 18.0	244	47%
Denver	-2.0%	-2.6%	-0.6	16.6 → 16.7	25	44%

Read the last two columns against the gap, not instead of it. The share of individual districts thinning can point the other way from the metro total — a metro can have many small districts falling behind while its largest district protects staffing, and the metro total follows the large district because that is where the students are. Both numbers are true and they answer different questions: the gap tells you what happened to the average child in the metro, the district share tells you how widely the pressure is distributed.

An important caveat about the starting year. The baseline here is 2020-21 — the height of pandemic disruption, when enrollment was artificially depressed, especially in the youngest grades. Some of the enrollment growth in these figures is recovery from that low base rather than genuine expansion, and the effect is strongest in the metros that lost the most students in 2020. This does not affect the gap between staffing and enrollment, which is the measurement here, but it does mean the enrollment column alone should not be read as a growth rate. The next release of federal data will allow a cleaner pre-pandemic comparison.

Coverage gaps, named rather than hidden. Nashville reports no comparable teacher data at all in the federal file — no reading is possible and its row is omitted from these tables. Chicago (36 of 336 districts) and Phoenix (64 of 244) are partial. Las Vegas is a single countywide district, and its unusually large positive gap reflects one district's reporting rather than a metro-wide trend. Districts under 300 students are excluded throughout.

How to read a district row

Enrollment change and teacher-FTE change are both four-year figures, 2020-21 to 2024-25. Four years is deliberately longer than a single year: annual staffing counts move on hiring cycles and one-year noise, and the four-year read is the more stable signal. Service direction compares the two: thinning where FTE falls more than one point faster than enrollment, absorbing where enrollment falls more than one point faster than FTE, holding where they move within a point of each other.

This page is descriptive. It does not rank districts by quality, does not evaluate any district's decisions, and offers no advice to parents, teachers, districts or officials. A district that is thinning may be managing an unavoidable contraction well; a district that is absorbing may be doing so through closures that families experience as loss. The measurement is narrow on purpose.

Source: National Center for Education Statistics, Common Core of Data, district-level (LEA) universe — total students all grades and full-time-equivalent teachers, school years 2020-21 and 2024-25, retrieved via the ELSI table generator, August 2026. Author's computation. Metro membership assigned by county. Federal collection runs roughly two years behind the current school year, so these figures describe trend, not present-day status. Budget, closure and board-decision detail for individual districts appears on each metro page and is separately sourced.