How much subsidy should each person get for health insurance?

For all the bubble and boil the past 4.5 years on health reform, there are basic questions that have not been asked clearly, much less answered. An important one is how much subsidy should people get from government for their health insurance?

The table below is a quick estimate of the per capita subsidy that different groups now receive, or will receive in health exchanges under the ACA. The magnitudes range from a maximum of $10,720 per capita for Medicare, to none for the uninsured or persons purchasing health insurance in the individual market currently, or in exchanges in the future if they make above 400% of poverty. Most people get some government subsidy for their health insurance.

ScreenHunter_02 Nov. 16 11.16

Link to table, including sources: insurancesubsidy.11.16.13

A few points:

  • Medicare. You might object to the word subsidy in the case of Medicare. People did pay taxes in the past, but they were of course paying for old people’s health care then. This is the magnitude of actual government flows required to fund the program today (and does not include out of pocket costs). If you want to call it something other than subsidy, fine.
  • Employer Provided (ESI). Underneath per capita figures there are big variations. For example, I receive about double the average subsidy shown for ESI. My federal tax bill is around $3,000 less than it would be if ESI did not receive tax preferences because: (1) the amount Duke pays for my insurance is not subjected to taxation (and thus tax free compensation); (2) My premium share is pre-tax, and thus reduces my income that is subjected to taxation. Further, there are 5 people who receive coverage via my employment at Duke and this subsidy (me, wife, 3 kids). This subsidy, combined with the pooling power of (especially) large employers mean that there are a very large number of people who have it quite good under the current system. So, the size of the per capita subsidy understates the political power for the status quo that exists in employer sponsored coverage. Paul Starr calls this plus the Medicare subsidy above the policy trap.
  • Individual purchase v uninsured. Speaking of political power, the force behind the “President lied” meme leading to the administrative “fix” is a subset of those currently in individual coverage plans who make over 400% of poverty and were therefore going to be forced to buy employer-sponsored levels of benefits without the benefit of an employer paying 75-80% of the premium cost with tax preferenced spending (a true disadvantage for the self employed). A better fix would have been to add some tax preference short term while not messing with the risk pooling aspects of the law, and answering the question posed in this post longer term. Note the crushing political silence on behalf of the uninsured, and especially on behalf of those purposefully left out by their state’s decision to not expand Medicaid, while their income falls below 100% of poverty so they don’t qualify for exchange subsidies.
  • Speaking of Medicaid, the per capita expenditure of $7,000 shown above does not characterize any group of Medicaid beneficiaries well. Children and adults are numerous and relatively inexpensive; there are fewer disabled and elderly, but they have high per capita costs. The most important thing to understand about Medicaid reform is who is covered by the program. In a state like North Carolina that has not expanded Medicaid for 2014 (seven part series of NC Medicaid reform), most of the expansion effect would be in groups with relatively low per capita costs, and who would therefore likely improve the state’s Medicaid risk pool (in North Carolina, childless adults can never be eligible for Medicaid, and working parents only up to 49% of poverty; non-working to 39%).
  • Exchange subsidies. There is a sliding scale from 100-400% of poverty ($11,500-$46,000 for singles, $23,500-$94,000 for family of 4) that determines how much subsidy is provided to purchase private health insurance in the exchanges. The $5,510 shown in the table is an average, with the amount of subsidy falling with income. Note that the tax benefits of employer sponsored insurance are the opposite, and rise with income. The different slopes of these curves summarize how politically hard it would be to modify/end the tax preference for ESI that now exists, to be replaced by a fixed amount of money to purchase private insurance.

What is the correct amount of subsidy? Above is what we have now and coming in the near future. I am not telling you what to think, but your thinking needs to account for all these groups.

cross posted at freeforall

Author: Don Taylor

Don Taylor is an Associate Professor of Public Policy at Duke University, where his teaching and research focuses on health policy, with a focus on Medicare generally, and on hospice and palliative care, specifically. He increasingly works at the intersection of health policy and the federal budget. Past research topics have included health workforce and the economics of smoking. He began blogging in June 2009 and wrote columns on health reform for the Raleigh, (N.C.) News and Observer. He blogged at The Incidental Economist from March 2011 to March 2012. He is the author of a book, Balancing the Budget is a Progressive Priority that will be published by Springer in May 2012.

24 thoughts on “How much subsidy should each person get for health insurance?”

  1. Medicare. You might object to the word subsidy in the case of Medicare. People did pay taxes in the past, but they were of course paying for old people’s health care then. This is the magnitude of actual government flows required to fund the program today (and does not include out of pocket costs). If you want to call it something other than subsidy, fine.

    Don — it is a subsidy, but I suspect that CMS or an independent researcher could provide the necessary correction to the estimate, i.e., how many taxes paid over the years fall short of Medicare spend today.

  2. What is the correct amount of subsidy?

    To that hard question, I would add this simple personal one:

    What is the correct amount of my subsidy?

    It is truly amazing that Democrats and this Administration have buried this bloody essential detail for 4 years.
    I mean really: If you are going to require people to buy insurance shouldn’t the help you will get be trumpeted long and loud?
    And get this: this essential fact is still buried when you try to navigate the shame that is

    Look children this is how it is done:

    Show me my damn subsidy up front so that I may use it to do my shopping.


  3. One of the really frustrating things about balance sheets is that while they make it easy to back-figure into means (like the per capita subsidies in your table, they do nothing to help us understand the variation about those means. The information is simply not recoverable.

    One of the problems with that reality is that it drives us back to anecdata, like Don’s note about his subsidy being substantially higher than the average.

    I wish I had a solution, but the math says to me that the information about variation is not available.

    1. The new requirement to put on W-2 the amount employer paid for insurance benefits means IRS should be able to put out detailed info on range in the future if they wish to do so.

      1. The W-2 number is the total amount of the insurance premium (employer portion plus employee portion). So it’s not as easy as it might be for the IRS to figure out the subsidy: the employer portion is subsidized, but the employee portion is not.

        1. What are you talking about? I don’t know about where you work, but where I work my portion of the insurance premium is paid with pre-tax dollars. Neither the employer portion nor the employee portion is subject to the income tax.

          1. As of the 2012 W-2, employers are required to put the entire cost of the employee’s health premium on the W-2. It’s line 12, marked with DD.

          2. I don’t understand how your comment is responsive to mine. You said the employee portion of the health-insurance premium isn’t subsidized in the tax code; it is.

          3. The employer portion of your employer-supplied insurance is subsidized: Your employer can deduct it as a business expense. If your employer just gave you the money as salary rather than buying insurance for you, that money would be subject to tax as your income. So you are getting your insurance bought with pre-tax dollars, but someone who buys insurance on the individual private market is buying her insurance with post-tax dollars.

          4. employee portions can be pre-tax – but does it count for the SSI taxes? I bet it’s murky

          5. Cardinal Fang: If it’s really that important to you to pretend that you weren’t wrong when you claimed that the employee portion of the premium of employer-sponsored health insurance isn’t subsidized, far be it from me to rain on your parade.

      2. The key phrase being, “… if they wish to do so.” The GOP has been so busy starving the IRS that I don’t think it’s likely they’ll wish to publish anything they aren’t already publishing. In addition, ranges are helpful but not adequate measures of variation.

        As to premium payment with pre- vs. post- tax dollars, doesn’t the pre-tax payment only hold for those working for approved not-for-profits? Or am I thinking of some other tax provision.


  4. How did you compute the average amount of the subsidy on the exchange? Is this assuming that everyone who is eligible to buy on the exchange does so? What assumption did you make about adult children? Where did you find the data about the ages, incomes and regions of residence of the people buying on the exchange?

    1. From CBO, # to be enrolled in 2014 x avg subsidy. Reality depends on who signs up. There is a citation in the second worksheet of the excel table that shows what CBO source used and it lists means subsidy over next few years per their projections

      1. Thanks! Sorry, I should have clicked the link in the first place. So it’s the CBO estimate, a respectable estimate but one that might not be right. In particular, the data is from 2011. The subsidies are based on the premiums for the second-lowest Silver plan, numbers that weren’t known at that time but are known now. The number of currently eligible people and their incomes might also be different than the 2011 estimate, though that would probably vary less than the premiums. I wonder if the CBO has updated their estimate now that the premiums are known.

        I didn’t mean to doubt your data. I was just curious.

  5. Thanks for this post. I will have fun sharing it with my libertarian friends.

    Here is a probably dumb question: you speak of expansion as improving Medicaid risk pools, but what does that actually mean in practice? Doesn’t the state government just kind of declare what they’re going to pay, and then the providers sue them because it’s too low? (That’s what seems to happen here in Cal.) If this risk pool were improved, would anything actually change as a result?

    1. but what does [improving Medicaid risk pools] actually mean in practice?

      Basically, all the old people in nursing homes with no assets are already in the Medicaid pool. Adding anyone else who is eligible (almost all of whom are under 65) will make the average pool member healthier, simply because “elderly and in a nursing home” is a REALLY unhealthy population.

    2. It means the average subsidy per Medicaid user would go down. Right now the subsidy averages $7K per person. Averaging in more non-disabled non-elderly adults, at $4.4K a pop (or probably less, because that $4.4K includes pregnant women), would bring the average down. But the total would still go up, so I’m not sure why bringing down the average would be a good thing per se.

    3. The primary source of expansion in North Carolina is childless adults, who can never qualify (there are about 30 states in which this is the case). In NC parents are eligible if income under 49% poverty if they have a job, 39% if unemployed. So, newly eligible via expansion would be adults, who have lower per capita costs than avg.

      1. Thanks. I get the idea of a risk pool and an average subsidy, but I had not realized before, I have little idea of how Medicaid administrators decide how much to pay for care and/or procedures. (I have an idea that in Medicare, they pay flat fee per service?)

        Though I guess at least in some places they were/are trying to switch Medicaid to a managed care model anyway, more of a whole person thing? I was just wondering how all these younger people in the pool would change the actual payments, and when I wondered that, I realized I have no idea how they do it now anyhow.

Comments are closed.