Social Security COLA 2027: The Biggest Raise in Three Years, and What Medicare Takes Back

The Social Security COLA 2027 forecasts have converged on a range of roughly 3.5% to 3.7%, which would be the largest annual adjustment in three years and a clear step up from the 2.8% increase paid through 2026. For the average retired worker that is somewhere between $67.90 and $75 more a month, depending on whose estimate you take.

That is the headline, and it is accurate. It is also the part of the story that matters least, because Social Security does not pay the headline. It pays the headline minus whatever Medicare takes first — and the Medicare figure that every current calculation relies on is a projection rather than a decision.

What the forecasts actually say

The direction of travel is worth noting as much as the level. In June the same independent forecast sat at 4.7%; by September it had come down a full point as inflation moderated. Estimates published in August were lower still before ticking back up. A forecast is a moving object, and anyone budgeting against a figure they read three months ago is budgeting against a number that no longer exists.

How the number is actually set

The COLA is not a policy decision, a negotiation or a political choice. It is an arithmetic output, and the formula is fixed in law.

Social Security compares the average CPI-W — the Consumer Price Index for Urban Wage Earners and Clerical Workers — across the third quarter of the year, meaning July, August and September, against the same quarter a year earlier. The percentage change, rounded to the nearest tenth, is the COLA. Nobody votes on it.

That is why the timing matters. As of mid-September, two of the three months are known and the third is not. The September CPI release completes the calculation, and the Social Security Administration announces the official figure in October.

The 2026 baseline you are measuring against

Note the scale. A 3.5% adjustment on an average benefit of roughly $2,064 produces around $72 a month. That is real money to a household living on it, and it is also barely more than a month’s worth of the $700 annual Part D deductible. Percentages sound larger than the amounts they describe.

What Medicare takes back before you see it

This is the part that determines the answer, and it is the part most coverage of the Social Security COLA 2027 handles in a sentence at the end.

Medicare Part B premiums are deducted directly from Social Security payments before they reach a bank account. A beneficiary does not receive the raise and then pay the premium; they receive the difference. So the meaningful question is not how large the COLA is, but how large it is relative to the Part B increase.

Why the gap between those two rows matters

If the Trustees’ $209.50 holds, then 2027 becomes the first year since 2023 in which the COLA grows faster than the Part B premium that erodes it. That is a genuinely positive result and it is what the more optimistic coverage is built on.

If the private forecasters are closer — and they are projecting $216 to $219 — the arithmetic changes. The Trustees have a documented tendency to project low, and a premium at the top of the private range would consume a substantially larger share of the increase. Taking the projected Part B increase together with the finalised $700 Part D deductible, analysis suggests Medicare costs absorb roughly 20% to 29% of the 2027 raise before a retiree sees any of it.

Five years of adjustments, in context

Read that column downward and the phrase “biggest raise in three years” loses some of its shine. It is the biggest since 2023 because 2024, 2025 and 2026 were all smaller — and it remains less than half the 2023 figure. The Social Security COLA 2027 would represent a return toward normal inflation, not an unusually generous year.

What it actually looks like on a real benefit

Percentages obscure. The table below applies the mid-range 3.6% forecast to three benefit levels and then subtracts the projected Medicare Part B increase, using both the Trustees figure and the top of the private forecast range, so the spread is visible rather than asserted.

At the average benefit, the difference between the Trustees’ Part B projection and the upper private estimate is about $9.50 a month — roughly 13% of the entire raise, decided by a number that has not yet been published.

Why the raise never quite feels like the number

There is a structural criticism of the COLA formula that long predates this year, and it explains a persistent gap between what the adjustment says and what recipients report experiencing.

The index used is CPI-W: the price index for urban wage earners and clerical workers. It measures the spending patterns of working-age employed people. It is applied to a population that is overwhelmingly retired, and retired households spend differently — proportionally more on healthcare and housing, proportionally less on transport to work, education and the categories that move a working-age basket.

An alternative index exists. CPI-E, the experimental index for Americans aged 62 and over, weights medical care and shelter more heavily. Advocacy organisations have argued for years that it should be used instead. It is not, because the statute specifies CPI-W and changing it requires an act of Congress.

The consequence is not that the COLA is calculated wrongly. It is calculated exactly as the law requires. The consequence is that it is calculated against a basket that does not describe the people receiving it — which is precisely why healthcare costs keep absorbing a disproportionate share of each year’s increase.

Who else this affects

Coverage tends to frame the COLA as a retirement story. It is broader than that. The same adjustment flows to Supplemental Security Income, which supports people with disabilities and older adults with very limited income and resources, and it flows to disabled workers and survivors receiving Social Security Disability Insurance.

For SSI recipients the proportional effect is larger, because the base is smaller and because SSI recipients typically have little or no other income to cushion a shortfall. A 3.5% adjustment on the 2026 individual maximum of $994 works out at roughly $35 a month.

There is also a timing quirk worth knowing: SSI payments are made on the first of the month, and when that falls on a weekend or holiday the payment arrives on the preceding business day. That is why SSI recipients typically see their January increase land at the very end of December.

COLA season is also peak scam season

This is the predictable and under-reported consequence of an announcement that reaches nearly 71 million beneficiaries on a known date: October and November are when Social Security impersonation attempts spike, because the caller has a plausible reason to be contacting you and you have a reason to want the call to be real.

The pattern is consistent. A message says your COLA increase needs to be “activated”, your benefit is suspended pending verification, or your new payment amount cannot be released until identity details are confirmed. Each of these describes something that does not exist. The adjustment is automatic and applies to every beneficiary without any action, request, form or fee.

What to do between now and January

For readers outside the United States

The mechanism here is not unusual; the transparency of the arithmetic is. Most countries with contributory pensions index them to something — prices, wages, or a blend of the two — and the choice of index is one of the most consequential and least discussed decisions in public finance.

The United States indexes purely to a price index, and to one built around working-age spending. Systems that index to wages tend to track living standards rather than just costs. Systems using a blended or “triple lock” style rule guarantee a floor and cost far more over time.

None of these is obviously correct. But the American design has one clear property worth recognising: because it is pure arithmetic with no discretion, the adjustment cannot be quietly withheld in a bad fiscal year. The COLA is automatic. That is a real protection, and it is the flip side of the complaint that it is calculated against the wrong basket.

Final verdict

The Social Security COLA 2027 forecasts are credible, they are converging, and 3.5% to 3.7% would genuinely be the largest adjustment in three years. Nothing in the reporting is wrong. The problem is what it leaves out.

Two numbers determine a beneficiary’s 2027 income and only one of them is being reported. The COLA arrives in October and will be covered everywhere. The Medicare Part B premium arrives in November, is covered far less, and is the difference between a raise that outpaces healthcare costs for the first time since 2023 and one that does not. The Trustees project $209.50; private forecasters say $216 to $219; the Trustees have a habit of projecting low. That spread, not the spread between 3.5% and 3.7%, is where the uncertainty actually sits.

The deeper issue is structural and will not be fixed by a good year. An adjustment calculated from a working-age spending basket, applied to a retired population, and then reduced by a healthcare premium that rises on its own schedule, will systematically under-deliver against the costs it is meant to cover. That is not a forecasting problem. It is what the formula does.

Our read: treat the October announcement as half the news. Do not revise a 2027 budget until the Part B premium is published in November, because the gross figure and the net figure have differed by a fifth to nearly a third in recent projections. And read “biggest raise in three years” for what it actually is — a statement about last year’s inflation, not about next year’s prosperity.