Payroll Taxes Cover About a Third of Medicare Costs

I get the impression that many Americans believe Medicare is financed like Social Security. They know that a portion of payroll taxes goes to Social Security and a portion goes to Medicare. So they conclude workers are paying for Medicare benefits the same way they are paying for Social Security benefits.

That isn’t remotely true, as new data from the Congressional Budget Office demonstrate.

In 2010, payroll taxes covered a little more than a third of Medicare’s costs. Beneficiary premiums (and some other earmarked receipts) covered about a seventh. General revenues (which include borrowing) covered the remainder, slightly more than half of total Medicare costs.

If you prefer to focus on just the government’s share of Medicare (i.e., after premiums and similar payments by or on behalf of beneficiaries), then payroll taxes covered about 40% of the program, and other revenues and borrowing covered about 60%.

In contrast, payroll taxes and other earmarked taxes covered more than 93% of Social Security’s costs in 2010, and that was after many years of surpluses.

The difference between the two programs exists because payroll taxes finance almost all of Social Security, but only one part of Medicare, the Part A program for hospital insurance. Parts B and D (doctors and prescription drugs) don’t get payroll revenues; instead, they are covered by premiums and general revenues. But that distinction often gets lost in public discussion of Medicare financing.

As recently at 2000, general revenues covered only a quarter of Medicare’s costs. That share has increased because of the creation of the prescription drug benefit in 2003 and because population aging and rising health care costs have pushed Medicare spending up faster than worker wages. Over the next decade, CBO projects that premiums will cover a somewhat larger share of overall costs, while the general revenue share will slightly decline.

Note: For simplicity, I have focused on the annual flow of taxes and benefits. The same insight applies if you want to think of Social Security and Medicare as programs in which workers pay payroll taxes to earn future benefits. That’s approximately true for workers as a whole in Social Security (but with notable differences across individuals and age cohorts and uncertainty about what the future will bring). But it’s not true at all for Medicare.

8 thoughts on “Payroll Taxes Cover About a Third of Medicare Costs”

  1. What difference does it make if Medicare is financed from a dedicated tax or general revenues?

    The CBO has also written that moving from current Medicare to a system in which patients pay more would increase national healthcare spending (reasons include that Medicare’s size gives it more buying power and that it has lower overhead than private insurance). That seems a worse problem than possible misperceptions in the details of Medicare financing.

  2. This article might give the impression that these statistics apply to a senior citizen’s entire health care costs. In fact these statisics primarlly relate to those senior-citizen healthcare costs paid for by the government for the senior.

    But 94% of seniors pay a considerable extra increment above these numbers for their health care. In addition to the Part B premium noted in the article and out of pocket costs primarily for annual physicals, vision and dental services (which are mostly not covered by Medicare), many seniors pay for an employer sponsored retiree healthcare insurance plan, a large group pay extra for a Part C Medicare plan, about 15%-20% buy a private Medicare supplement policy (commonly called Medigap), a small percentage are in the VA system, and about 10%-20% of us have to apply for welfare.

    And I’m not even beginning to describe the premium/deductible/OOP complexity of Part D and the fact that about 6% of us fall into the infamous donut hole and about 1% of us kick over into catastrophic Part D coverage.

    Your bar graph needs about eight colors, not three. 🙂

  3. Very interesting article. I was quite curious about the bar graph- after reading Dennis Byron’s comment, I realised why that was. Like he said, it looks like there are other factors that needed to be included in that graph.

Comments are closed.