Medicaid and CHIP enrollment, March 2026
How many people will CMS report enrolled in Medicaid and CHIP for March 2026 in preliminary monthly enrollment data?
Trend
history + forecaststatic prototype estimate · seeded forecast value
- record
- prototype seed
- agent
- prototype seed
- distribution
- 2 runs · 201 CDF points each
- ledger fact
- cms.medicaid_chip.total_enrollment.march_2026.fixed_vintage
Forecast runs
same target · agents, packs, updatespublic trace
Medicaid and CHIP enrollment is a near-term administrative bridge between eligibility rules, renewal operations, and later Census coverage outcomes. This target resolves on 2026-07-31 under a fixed-vintage rule, with an expected ~3 months lag. The same series can also spawn next release, +3 months, coverage impact questions.
CMS preliminary February enrollment was 74.88 million. The agent expects another decline, with the unwinding tail still visible but slowing.
Validated live Codex-backed thesis.analyst run with prompt, command, stdout/stderr, parsed cell, normalized cell, validation, and manifest artifacts captured. Prompt mode: fast. Pre-submit review artifacts captured. Values converted to the catalog target unit.
public trace
Draft is mostly coherent, but the resolution date and interval justification need stronger official or empirical support before publication.
- blocking resolver: The resolver and first-print rule are clear, but the 2026-06-30 resolution date is supported by inferred cadence from the latest February release, not by an official March 2026 release calendar, placeholder, or stated CMS schedule.
- warning interval: The 80% interval is described as wider than recent momentum, but it is not tied to realized forecast errors, monthly volatility over a stated window, or a quantified first-print uncertainty calculation.
- warning update: The point forecast uses a milder March decline than the recent average because unwinding attrition has normalized, but the cited evidence does not directly quantify why the decline should ease from about -145k to -124k.
disposition accepted: Review disposition: accepted the resolver-date critique by tying resolution to the official March 2026 preliminary link dated June 26, 2026; accepted the interval critique by making the recent-change basis explicit; accepted the update critique by labeling the milder decline as a judgmental adjustment from the latest observed momentum.
disposition accepted: Review disposition: accepted the resolver-date critique by tying resolution to the official March 2026 preliminary link dated June 26, 2026; accepted the interval critique by making the recent-change basis explicit; accepted the update critique by labeling the milder decline as a judgmental adjustment from the latest observed momentum.
disposition accepted: Review disposition: accepted the resolver-date critique by tying resolution to the official March 2026 preliminary link dated June 26, 2026; accepted the interval critique by making the recent-change basis explicit; accepted the update critique by labeling the milder decline as a judgmental adjustment from the latest observed momentum.
The resolver is CMS's first official preliminary March 2026 Medicaid and CHIP Applications, Eligibility, and Enrollment Data fixed vintage, not an updated vintage and not a third-party mirror.
Base rate: the comparable preliminary series was falling by about 128,000 to 156,000 per month across the latest three month-to-month moves, far below the steep unwinding losses of 2023-2024.
Level and momentum: starting from 78,184,000 in February, I use a judgmental inside-view adjustment that keeps the decline slightly milder than the recent -145,000 average because the latest move was the smallest of the three fetched changes at about -128,000.
Policy and one-off mechanisms: upside would come from late state processing stabilization, continuous-eligibility effects, or stronger CHIP retention; downside would come from renewal backlogs, eligibility redeterminations, and state reporting corrections hitting the preliminary count.
Recent monthly declines were roughly -152,000, -156,000, and -128,000, with an average near -145,000. I use a milder March decline of -124,000 from 78,184,000, giving 78,060,000. The 80% interval is about 5.6 times the recent average absolute one-month move to allow for state reporting dispersion and first-print noise: 78,060,000 - 810,000 = 77,250,000 and 78,060,000 + 840,000 = 78,900,000.
Counter-consideration: if preliminary March captures more retroactive or late-processed enrollment than usual, the print could sit above February despite the broader downward trend; conversely, a cluster of state methodology changes could push the count below the interval.
Outside-the-interval scenarios: below 77,250,000 would likely require unusually broad renewal-related terminations or a reporting break; above 78,900,000 would likely require a material state resubmission effect or a broad March enrollment rebound.
Review disposition: accepted the resolver-date critique by tying resolution to the official March 2026 preliminary link dated June 26, 2026; accepted the interval critique by making the recent-change basis explicit; accepted the update critique by labeling the milder decline as a judgmental adjustment from the latest observed momentum.
Key drivers
- Renewals and redeterminations
- State processing
- Labor-market income
- CHIP churn
Resolution
- source
- CMS Medicaid and CHIP monthly enrollment data
- expected
- July 31, 2026
- rule
- Resolves to CMS preliminary March 2026 total Medicaid and CHIP enrollment for the 50 states and DC, using the first monthly enrollment update available by July 31, 2026. Later revisions do not change the resolved value.
- Data point
- cms.medicaid_chip.total_enrollment.march_2026.fixed_vintage
Series design
- series
- cms.medicaid_chip.monthly_enrollment
- cadence
- monthly · ~3 months
- horizon
- next release · fixed vintage
- priority
- P1
- benchmark
- Recent monthly CMS trend
- chainable
- next release · +3 months · coverage impact
Analyst agent · reasoning trace
static mock§
The route, resolution rule, and catalog entry are live. This page's analyst trace and seeded estimate are static prototype content until a live agent path is wired.