Direct-purchase health coverage rate, 2025

Direct-purchase coverage is the Census-side counterpart to marketplace enrollment. It tests whether agents can reconcile administrative sign-ups with survey-reported insurance categories.

The forecast holds the rate near 2023-2024 despite strong marketplace sign-ups because Census direct-purchase coverage includes classification noise and because some Medicaid transitions replace, rather than add, covered people.

The resolver is the U.S. Census Bureau CPS ASEC first print for calendar-year 2025 direct-purchase health insurance coverage among all persons, reported as a percent. This is an annual coverage-for-all-or-part-of-year concept, not an end-of-year enrollment count.

Base-rate/reference-class anchor: the recent CPS ASEC direct-purchase rate moved 9.9 to 10.2 to 10.8, a three-year level near 10.3 percent and recent annual changes of +0.3 and +0.6 percentage point. A neutral continuation anchor is roughly 11.1 percent before current-year policy adjustment.

Level and momentum: the 2024 level already incorporated part of Medicaid unwinding and marketplace growth. Momentum into 2025 remains positive because marketplace selections rose again, but CPS annual coverage will translate enrollment into the all-person annual coverage category imperfectly.

One-off and policy mechanisms: enhanced ACA subsidies were still active for 2025 and should keep direct-purchase take-up elevated. Some Medicaid unwinding transition into marketplace coverage likely persists, but the largest unwinding shock was earlier, so I do not project another 0.6 point jump.

Counter-consideration: CPS classification noise, off-exchange attrition, employer/public coverage substitution, and churn can cause marketplace growth to show less strongly in the direct-purchase all-person annual rate; a flat 10.8 to 11.0 print is plausible if selection counts overstate sustained coverage.

Base 2024 rate 10.8 + expected 2025 net increase 0.5 = 11.3. The 80 percent interval is judgmental rather than a clean long-history volatility estimate: recent annual changes of +0.3 and +0.6, survey/category noise, and policy churn support a +/-0.6 point band, giving 10.7 to 11.9.

Upside scenario: marketplace enrollment persistence and continued subsidy take-up push direct-purchase coverage above 11.9 percent. Downside scenario: churn, Medicaid/public coverage retention, or CPS classification effects keep the rate near 10.7 percent or below. Outside-interval downside would require little translation from the 24.3 million marketplace selections into annual CPS coverage.

Review disposition: accepted the resolver critique by grounding September 8, 2026 in the official Census 2026 schedule and distinguishing the timing URL from the HHI-01 resolving table; accepted the interval critique by labeling the +/-0.6 band as judgmental 80 percent uncertainty rather than a precise long-history volatility estimate; retained the central forecast because the source evidence and mechanism balance did not change.