If your 2026 renewal notice was a shock, the filings for 2027 suggest it was not a one-off. Insurers selling on the ACA Marketplaces had to submit proposed premiums for next year by 15 July, and the early analysis of those filings points to another double-digit increase — arriving on top of an increase that has already happened.
This is a cost and coverage explainer, not medical advice and not a plan recommendation. What it tries to do is make your own renewal letter legible: where the number comes from, which part of it is subsidy policy rather than healthcare, and what is still adjustable.
The short version: the median proposed increase for 2027 is 14%, across 77 insurers in 16 states plus DC, with individual filings ranging from 1% to 52%. For 2026, insurers proposed a median 18% and regulators finalised 20% — so last year the proposal came in under the eventual outcome. If 2027 lands near the filings, premiums will have risen by more than a third across the two years.
What the filings actually say
The 14% figure is a median of preliminary proposals, not a decision. State insurance regulators review filings and can push back, and the final numbers usually land somewhere near — but not exactly on — what was asked for. The spread matters more than the median here: proposals run from 1% to 52%, which means the state you live in and the insurer you happen to be with will matter far more to your bill than any national average.
The one useful precedent is last year. For 2026, the median proposal was 18% and the finalised figure was 20%. Regulators did not trim it; the outcome exceeded the ask. That is worth holding in mind before treating 14% as a ceiling.
The two-year arithmetic, in one household's numbers
Averages are hard to feel. The Peterson-KFF analysis works a single example that makes the compounding concrete: a 40-year-old in Indianapolis on an Anthem Silver plan, earning $65,000 a year.
| Year | Monthly premium they pay | Change on the year |
|---|---|---|
| 2025 (with enhanced credits) | $316 | — |
| 2026 (credits expired) | $477 | +$161 |
| 2027 (projected) | $546 | +$69 |
On what actually leaves their bank account, that is $230 a month more than in 2025 — a 73% rise in two years. Annualised, roughly $2,760 of extra after-tax spending that has to come out of something else.
One note on figures you may see quoted elsewhere. Peterson-KFF describes this same example as a cumulative rise of "$158, or 41%", which looks like it contradicts the table above. It does not: that comparison is measured against this person's unsubsidised 2025 premium of $388, not against the $316 they actually paid once enhanced credits were applied. Both numbers are correct, they are just answering different questions. The $230 is the one that matters to a household budget, because it is the change in money actually paid.
Why insurers say it is happening
The filings give reasons, and they are not all the same kind of thing. Roughly in order of weight:
- Medical trend of about 10% — the underlying growth in what care costs, running above the historical 8% average. Hospital, physician and pharmaceutical prices are all cited.
- The expiry of enhanced premium tax credits, which insurers estimate contributed around 4 percentage points on its own. This is the part that is policy rather than healthcare.
- A sicker risk pool. When subsidies shrank, healthier enrollees left. The people who remain cost more per head to cover, which raises the premium for everyone still in.
- GLP-1 medication costs, which several insurers name explicitly — though some have responded by dropping weight-loss coverage rather than pricing it in. We looked at what those drugs actually cost across countries separately.
- Provider consolidation reducing pricing competition, and dispute-resolution costs under the No Surprises Act.
The second and third of those are worth separating out, because together they are roughly the share of your increase that exists because of a policy decision rather than because medicine got more expensive.
The subsidy cliff is back, and for some households it is the whole story
The enhanced credits did two things: they made subsidies larger, and they removed the hard cut-off at 400% of the federal poverty level. Their expiry at the end of 2025 restored that cut-off.
So there is now a cliff, not a slope. For a single person in 2026, 400% of the federal poverty level is $62,600. Below that figure you receive a premium tax credit. One dollar above it you receive nothing at all and pay the full unsubsidised premium. The difference between $62,600 and $62,601 of income can be thousands of dollars a year.
This is why the 2026 increase in what people actually paid — 58% on average, on top of roughly $1,000 more per person in deductibles — was so much larger than the 20% rise in headline premiums. Headline premiums went up 20%; the subsidy absorbing them shrank or vanished.
The lever most people still have: the income the cliff is measured on
Subsidy eligibility is not assessed on gross salary. It is assessed on modified adjusted gross income — and MAGI is reduced by pre-tax contributions such as a traditional 401(k), a deductible traditional IRA, and an HSA.
That makes the cliff unusually actionable compared with most tax thresholds. Someone whose MAGI would land at $66,000 is $3,400 over the 2026 single-filer line. Contributing that $3,400 to a traditional 401(k) brings MAGI to exactly $62,600 — back under the threshold, with subsidy eligibility restored, and the $3,400 still theirs inside a retirement account rather than spent.
Why this is worth checking rather than ignoring: because the cliff is a cliff, the return on the last few thousand dollars of pre-tax contribution can be far larger than the tax saved on them. You are not saving your marginal rate on $3,400; you are potentially restoring an entire year's premium tax credit. Run your own numbers before assuming it applies — the 2026 contribution limits are in our IRS limits guide, and the 401(k) calculator will show what the contribution itself does over time.
Two honest caveats. This only helps if you are near the line — it does nothing for someone at $120,000. And the poverty-level figures are updated annually, so the 2027 threshold will differ from $62,600; check the current number for the year you are enrolling in rather than reusing this one.
Dates worth putting in the calendar
Open enrollment for 2027 coverage is expected to open on 1 November 2026, with 15 December 2026 the cut-off for coverage starting 1 January 2027. There is live litigation over how long the federal window runs — a 2025 rule shortening it was vacated in June 2026 and is under appeal — so the safe approach is to treat 15 December as the deadline regardless of how that resolves. State-run exchanges set their own dates and several run longer. Confirm yours on HealthCare.gov or your state marketplace rather than on this page.
The other date that matters is whenever your own renewal notice lands, usually in October. That letter, not any national median, is your actual number — and it is also the moment when switching plans is cheapest to do.
What to do with this
Three things are worth the time. Work out what a premium at 2027 levels does to your monthly position before the notice arrives, rather than after. Check whether you are within a few thousand dollars of the subsidy cliff, because that is the one threshold where a modest pre-tax contribution can pay for itself many times over. And when the renewal letter comes, actually compare plans rather than auto-renewing — in a year when the spread of proposed increases runs from 1% to 52%, the insurer you are with is doing a lot of the work.