• Chart of the day: Medicare pays for less than half of spending by its beneficiaries

    In his book The American Health Economy Illustrated, Chris Conover’s charts are prettier than the one below. He shared this pre-pub version with me for posting purposes. At first glance, don’t worry about the left-hand bar in this figure. (Below the chart I’ll get into details.)

    Look at the right bar. See the blue area at the bottom, labeled 42.4%? That’s the percent of total medical and long-term care spending that Medicare pays for. The rest is covered by supplemental plans (employer wraparounds, individual Medigap, Medicaid), beneficiary premiums, and cost sharing. Medicare itself is far less generous than most people probably believe. Note the date, 2005. That’s before the drug program kicked in. So, this is just Medicare Parts A, B, and C.

    Mcare bene hlth cost

    And the details? Here’s the figure caption:

    The program perspective counts all Medicare spending under Medicare and all bills paid through public and private third party payers under third party payments. Beneficiary premiums include only administrative costs related to Medicare and such payers. The beneficiary perspective includes the full amount of premiums paid to Medicare and third parties under beneficiary premiums and deducts these amounts from Medicare and third party payers.

    If that’s still not clear to you, don’t worry about it. I don’t consider it important. The right-hand bar is good enough.

    Here are the indicated sources: KFF, BCBS,  Medicare Trustees, and author’s calculations.


    • In a follow-up post, it would be good to see how this compares to other nations like Canada, Great Britain, Australia, Swenden, etc. Thanks.

    • Marilyn Moon’s book “Medicare: A Policy Primer”, which I’ve unforgivably not yet reviewed, is written from the beneficiary’s perspective. She makes exactly this point about how large a fraction of their care beneficiaries are covering. It gets even worse when you consider what fraction of their *income* beneficiaries are paying, versus the fraction that non-retirees pay; retirees have very low incomes (see, e.g., https://www.socialsecurity.gov/policy/docs/chartbooks/income_aged/2010/iac10.html#table2 ); the denominator is very small and the numerator higher than one might expect.

    • Medicaid is the middle class program for paying for long-term care. That’s why the yellow block is so large.

    • That’s a very enlightening chart.

      I’d be curious to know how much you think these numbers would change from 2005-2013 given that in that time period Part C enrollment has more than doubled (13% to 27%).

    • Key point is that long-term care is *included*. Medicare was never designed to pay for long-term care, for better or worse. Only a minority of Medicare beneficiaries will incur paid long-term care expenses.

      I am not sure how much higher Medicare’s contribution would be if this were just medical care. It would also be quite a bit less than 100 percent, but I’d guess more than 60.

      • It was also designed to have cost-sharing that was not insignificant. High deductibles and costs at point of service were really the only mechanism that was available at the time to control costs and utilization.

    • A more useful chart on this subject — but roughly making the same point — can be found at Chart 5-3 in the June 2012 (most recent) MedPAC Databook. It strips out institutionalized seniors and therefore most of the Medicaid factor but it also strips out the 20% to 30% of us seniors on Part C Medicare. Carefully read the Note on Chart 5-3 which says that the estimated OOP spending by FFS beneficiaries does not include beneficiary payments for supplemental premiums. Add them to the 16% OOP number in the MedPAC chart and you get something close to or higher than the 28% OOP number in the above chart.

      By making the same point, I mean that I assume the author of the book mentioned wants to make the point that Original FFS Medicare is really bad insurance. He doesn’t have to use the fact that it does not include LTC to make that point.

      On the other hand, Part C Medicare — which is not illustrated in the MedPAC chart — overcomes many of these limitations by providing catastrophic coverage not part of FFS Medicare, as well as lower co-pays and many more health services (e.g., annual physicals, drugs usually) than FFS Medicare. A pie chart on Part C Medicare would look more like someone buying healthcare insurance would expect to see (but that is no good to you if your doctor doesn’t take Part C).

      It would probably be stating the obvious for readers of your blog to point out that Part C Medicare is the very successful 1997 Democratic-Party Medicare Voucher Plan, which grew out of an idea by former CBO staffer and leader (just before he came up with the idea apparently — see Health Affairs, 8/30/1995) and now Medicare trustee Robert .Reischauer.

      • just to comment on Part C….

        I think it is hard to say “very successful” when describing Part C. It surely is very expensive (though now attempts are being made to bring costs down), but it seems that in terms of quality, definitive conclusions are hard to come by.

        Also the 1997 version was quite unsuccessful, in regards of enrollment, seeing plans and seniors leave. If we define success by enrollment, it wasn’t very successful until the overpayments begun by the Bush administration with the MMA 2003.

        • @Andrew

          As a beneficiary of Medicare Part C (a Part C HMO to be specific) I take a broader and non-political view relative to enrollment success than you. The Democrats came up with the proposal pretty much as it exists today in 1995. The president signed the law in 1997. It became effective (I dunno exactly) in 1998 or 1999. Now 14 or 15 years later (which seems like a reasonable, maybe even minimum, period of time to analyze such a thing) it has become the second most popular way for seniors to get healthcare insurance (30%-plus market share) whereas previously it did not exist (0% market share; there were demonstration projects prior to the passage of the law so the 0% market share point was actually about 20 years ago ).

          By comparison 5% or less of seniors depend only on FFS Medicare for their healthcare insurance. (See Chart 2-5 in June 2012 MedPAC Databook but factor out the people on FFS Medicare only that are in that categroy because they depend on the VA or a spouse’s employer or employer retiree insurance for their healthcare insurance. Or who still work themselves and still get employee healthcare insurance but also sign up for Medicare.).

          Maybe initially the politicians set the incentive payment too low, then during the 15 years some other politicians tweaked the incentive payment a little too high, and then later some other politicians tweaked it down again. Hopefully this time they got it right .If not, they’ll tweak it again. Whatever, you’re nibbling around the edges with your comment. You’re not seeing the big picture.

          (The major overpayments were made for the stupid idea of having capitated FFS and that idea has already fallen by the wayside. I know the idea was to help the poor but it just didn’t work because it was intellectually flawed. Similarly the big fluctuation in enrollment seems to have been in capitated FFS plans too. HMOs, the sorts of coordinated-care arrangements with all kinds of preventive benefits so highly promoted for non seniors and seniors alike by PPACA — and by Mr. Reischauer back in 1995, seem to have grown in popularity consistently, including almost 20% since the 2010 tweak of incentives down. The incentives do not appear to be a big motivator. At least in Massachusetts, even with the projected 7% increase in premiums predicted for 2014 by AHIP recently, Part C plans are still going to be a much better deal than any combination of Medigap and FFS Medicare.)

          As for your comment on the quality of Part C Medicare, what do you mean? The quality of the insurance is certainly no better or worse than the quality of any other healthcare or fire or auto insurance. In general, Part C insurance companies honor their obligation to the insured in the same way that the FFS Medicare insurance companies do and that Medigap insurance companies do and that the insurance companies hired by former employers do. In fact, in many states, the same insurance company administers all four types of insurance and get involved in administering Medicaid contracts as well.

          • Medicare Part C

            If the program were truly working (part of the goal was to eventually save money) then the per capita payment to Medicare Advantage would not have to exceed the per capita payment to Medicare. Additionally adverse selection would not exist.

            • @ al

              From your comment/question it appears you might be referring to the risk adjustment payment formula in Part C rather than the alleged incentive overpayments in Part C discussed earlier in this thread. (I’m not sure what you mean otherwise because I believe the only per-capita payments in Medicare are the ones made to Medicare Advantage; I don’t believe there is any “per capita payment to Medicare.”) Both the incentive bidding process and the risk adjustment process are extensively discussed in separate MedPAC reports on Part C and there is an overview in the Databook I mentioned above.

              I use the words ‘alleged overpayments’ because if you read the MedPAC reports and other history, it appears the incentives were purposeful–to help the rural and inner- city poor. So calling them overpayments seems misleading.

              But it’s all just history anyways because all or most of these incentives are going away. For 2014, the proposed changes are all explained in this link. Good reading material if you have trouble sleeping (see . In the next few years, the difference is supposed to disappear.altogether. Such incentives would have disappeared almost entirely already except that in 2011 the Obama administration (illegally according to the GAO) devised a Part C Medicare demonstration bonus program for 2013 and 2014 that gave just about everyone a bonus. (See MedPAC chairman’s testimony on March 15, 2013 to the relevant House subcommittee for more details. This everyone-wins bonus program is discussed both in the opening statement of the chairman and in the Q and A of the hearing.)

              As for “adverse selection,” that gets a lot of publicity but the experts from Harvard looked hard and couldn’t find any evidence that it affected the incentive/bidding process. See AJMC, September 2012, Song et. al. I personally believe there probably was some cherry picking or reverse cherry picking back when only about 5% of Medicare beneficiaries were in Part C. But now that over 30% of us are in it, it’s hard to see how you could hide it.

            • Dennis, Medicare calculates the per capita rate to Medicare Advantage so that it is greater than the amount that would be given to comparable patients under FFS Medicare. Additionally those on Medicare Advantage are given a second bite of the apple. If their insurer refuses to pay they can always return to Medicare. That generally involves high cost treatments that alleviates the Medicare Advantage insurer from some very high expenses. The adverse selection phenomenon was finally documented by the GAO a number of years ago when they did a risk analysis of the payments to Medicare HMO’s and found that they were overpaid by about 1/3.

              Risk adjustment is very much in its infancy and can be easily gamed. The EHR makes gaming even easier. I agree that it is likely some rural areas were getting more funding due to need, but that extra funding extends to rich areas as well along the Gold Coast of Florida. As you say the extra funding is disappearing and when it is gone one will be able to better see the efficacy of this program. Adverse selection, however, I believe remains despite your citation which I will look at if you provide an http. The second bite at the apple remains untouched.

              We saw all of these mechanisms in action with the initial Medicare HMO’s and they were discounted by government officials, but a decade later these mechanisms were proven to exist and cost a lot of money.

            • “Additionally those on Medicare Advantage are given a second bite of the apple. If their insurer refuses to pay they can always return to Medicare.”

              Can you explain this in detail and provide a link to something that substantiates it?

            • Austin

              I think I know what Al means but it is not accurate to say “return to Medicare.” No one on a Part C Medicare health plan “leaves Medicare.” Part C is Part of Medicare; hence the word “Part.”. You must be on Medicare Parts A and B to sign up for C. I think he means one can switch back and forth between A/B and A/B/C every year (every month if your are low income); however, you risk not being able to buy Medigap insurance in some states if you make this switch.

              Also it is not accurate (but I don’t think this is what Al means) to say you can switch on the fly and in arrears if an insurer does not pay for a certain procedure or service. This should never happen except if someone erred in the process because Part C at a minimum covers everything A and B covers (plus usually a lot more)


              As I said in my earlier comment, I believe — but cannot find any research (please cite the years-old research source you mention) — that there probably was some cherry picking or reverse cherry picking years ago when only 5% of the base chose Part C. But I intuitively doubt it to be the case with 30%-plus now choosing Part C (it would be too obvious) and the recent AJMC research I cited ( http://www.ajmc.com/publications/issue/2012/2012-9-vol18-n9/Competitive-Bidding-in-Medicare-Who-Benefits-From-Competition ) said it wasn’t there in 2009.

              In general, your argument seems to be that the Original Democratic-Party Medicare Voucher Plan of 1995 was flawed, the 1997 law from President Clinton that partially implemented the 1995 proposal was flawed, the subsequent incentive payment plan (which has since been repealed) was flawed, and attempting to adjust for risk as outlined in last month’s proposed regulation is flawed, so let’s just keep the flawed 1965 system and make the 45,000,000 seniors who have made other arrangements go back to that flawed 1965 system. You don’t really mean that, do you?

            • This may be too far down in the wonk weeds to post but my March 16 AM reply to you that a senior could switch from A/B to A/B/C and back again every month if low income may be incorrect. A low income senior can switch among different A/B/C plans in his or her country every month if he or she so chooses (it would be a very odd thing to do but it is possible). But I am not sure whether, a low income senior that goes from an A/B/C plan to A/B during a given year can go back to that or another A/B/C plan during that year. Possibly Medicare would consider that to be a special enrollment period but I think Medicare would tell the low income senior to wait until the annual A/B/C enrollment period.

              The typical reason to want to do this as often as the law allows relates to drug coverage and not to the medical services covered by the A/B/C plan. The Medicare feature that allows low income seniors (130% of FPL I think) to do this is part of the LIS/Extra Help feature of Part D. If your Part D is included in your A/B/C plan and you are low income, you get the same flexibility as someone on a Part Standalone Plan

            • “Can you explain this in detail and provide a link”

              Austin, Things have tightened up from the old Medicare HMO rules, but if one is unable to get certain services from a Medicare Advantage plan or has high costs from one of the Medicare MSA type plans one can always switch during the enrollment periods to FFS Medicare which is more likely to permit the desired care. There was less of a time delay in the past, but it can still be done today and that can switch costs from MA to FFS Medicare.

              I don’t think you need a link because all of this is on the government web sites that describe the services Medicare has to offer, enrollment periods etc., But, here it is.


              You probably would like this report, but as with most discussions there is an agenda so I would be careful quoting any conclusions or data from it (though I believe it mostly true) without checking the data source. Additionally this is a bit old, 2007.

              Take note. I have disagreements with both sides and when I say something it is not taken out of thin air. As I have discussed with Aaron there is more than one side to the discussion. Nothing is black and white.

            • “during the enrollment periods”

              Yes, that’s my understanding as well. I thought you meant more frequently than that, which is why I asked for clarification. Dennis Byron added additional detail.

            • No, I think Medicare got a bit smarter this time around, but the dynamics are similar, only the bites are further apart.

              If I had made an error I would have hoped you would have corrected me. I have serious interests in healthcare, but like anyone else I can make mistakes. I don’t mind people pointing them out, but I do note that some are unwilling or unable to do so which might mean that one is unable to defend their position. I hope you realize after your test that I am reasonably knowledgeable about many of the problems under discussion. We have some agreement and we have some disagreement. It would be unfortunate to consider only those points where agreement exists discarding those where we might reasonably disagree.

              David added some additional detail, but is it exactly correct? I haven’t had time to respond to David, but I will try to now or later. I am on the road.



            • Al, chill out. I thought you knew something I didn’t. I asked a question. You answered it. I wasn’t “testing you.” We’re cool.

            • ” No one on a Part C Medicare health plan “leaves Medicare.”

              Byron, actually people are leaving Medicare though I delineated Medicare FFS earlier which is the traditional plan. Once one accepts the rules of MA they give up certain benefits of Medicare though they remain under its umbrella.

              ” however, you risk not being able to buy Medigap insurance in some states if you make this switch.”

              I am not sure of all the state regulations so I can’t definitely say that you are right or wrong. What happens I believe in most states is that the fee might be different from the fee that would have been offered at age 65. Since I am not exactly clear on this issue can you tell me in what state(s) one might be denied gap insurance and if convenient a citation for the law?

              “Part C at a minimum covers everything A and B covers (plus usually a lot more)”

              Insurance and care represent two different things. This issue was heavily litigated in the past. What is offered in the contract and actually given can mean two entirely different things because there are all sorts of things that occur between coverage and the provision of care. Think medical necessity.

              If you wish to rehash the old HMO plans we will have to do so on another space as it would be too lengthy. Look up the Patient Bill of Rights that was brought before Congress. There was a reason for that bill and many things that occurred then can occur now. Cherry picking was proven by the GAO with regard to the first attempt at Medicare HMO’s. They came up with a number around 67%, but I cannot recall the exact metrics behind that number. They have tried risk adjustment, but risk adjustment is in its infancy and opens the flood gates to all sorts of manipulations that have recently been reported.

            • “Al, chill out. I thought you knew something I didn’t. I asked a question. You answered it. I wasn’t “testing you.” We’re cool.”

              Sorry. If we are cool then that’s good enough. Medicare is extremely confusing so I don’t hold it against anyone for making an error with regard to their rules. Even Medicare help employees make routine mistakes and not infrequently at least in the past the rules and regulations had to be litigated to find out what was appropriate.

    • This chart would be much more useful if it explicitly showed how much of the yellow box is Medicaid…

    • Can someone explain more about what the Third party payments are and who they come from? Also, how would a chart like this potentially influence the “cost-shifting” debate? Thanks!

      • It appears that the third party payments refer to the fact that around 95% of Medicare beneficiaries depend on some other form of insurance (40% of us on retiree insurance; 30% of us on Part C Medicare, 20% of us on Medigap; 20% of us on Medicaid) to make up for the extensive limitations of Original Medicare

        (If you noticed that the percentages added up to more than 100%, that’s not a typo. Many dual eligibles — people who get Medicare and Medicaid — use their Medicaid “voucher” to buy private Medigap or public Medicare Part C insurance. Also the percentages are all just round numbers.)

    • Al —

      You seem to be dealing with very old information about Medicare or are misreading current information about Medicare.

      1. You say:

      “Byron, actually people are leaving Medicare…”

      Again your wording is confusing but — I repeat — no one leaves Medicare IF they choose Part C Medicare (see page 68 of “Medicare and You, 2013” and multiple other places in CMS documentation and its web pages). This is a common past misunderstanding and CMS began with its 2013 material released in mid 2012 to emphasize that no one leaves Medicare if they join a Part C plan. I believe this has been true since the inception of Part C in 1997 but cannot be sure; it has certainly been the case since 2010 when I first became involved with Medicare (as a beneficiary and a CMS-certified SHIP volunteer).

      Medicare Part C is Part of Medicare (hence the word “Part”). In order to choose Part C, a Medicare subscriber must be on both Parts A and B (not just A as is the case in being a minimal member of Original Medicare).

      ((If — on the other hand — you mean people are “leaving Medicare” altogether, I don’t believe that to be any kind of trend but I am sure there is an oddball that makes that decision every once in a while (e.g., if they were paying for Part A and decided to stop paying for it? I don’t think this is what you mean.))

      2. You say:

      “Once one accepts the rules of MA they give up certain benefits of Medicare though they remain under its umbrella.”

      No one gives up ANY Medicare benefits by choosing a Part C Medicare health plan. No one on a Part C Medicare health plan is less — or more — under the Medicare umbrella than anyone just on Medicare Part A or anyone on a combination of Parts A and B. See page 68 in “Medicare and You, 2013” and also see the Evidence of Coverage document for any Part C Medicare health plan.

      The wording has been similar in all versions of “Medicare and You” at least back to 2010. As far as I know this has been the case since the inception of Part C in 1997 but there were related demonstration projects prior to the passage of the 1997 law. Perhaps what you say was true 20 years ago but it is not true today. It is not true today and anyone making that statement to seniors causes a lot of problems for them.

      Of course, the rules are different depending on what Parts of Medicare a senior picks; that’s the point of having different Parts. Having different rules for different Parts of Medicare is no different than having different rules in different plans offered by employers or unions when working or in insurance bought direct prior to turning 65. Most of us on Medicare now began choosing among such different rules in healthcare insurance 30 years ago, during the 1980s.

      — One of the major rules differences (the most important in my opinion) and one that works strongly in the Part C beneficiary’s favor is that all Part C Medicare plans must provide catastrophic coverage. Original Medicare does not provide such coverage and in many states private Medigap plans do not provide such coverage. Without catastrophic coverage, seniors face medically caused bankruptcy. In other words, Original Medicare does not provide the most important thing that insurance should provide: upside protection.

      — One of the major rules differences that some seniors think is a negative is that for most but not all Part C Medicare plans, the senior has to always work within a provider network (a Part C HMO) or — alternatively — work within a network to get the lowest prices (a Part C PPO). In all counties in the U.S. — except possibly the most rural — the senior has a choice of Part C networks prior to signing up for Part C so this restriction is typically not an issue, especially for the millions of us who have been in HMOs since our 30s..Still the restriction should be well understood before selecting a Part C plan. On the flip side of the network characteristic (which is highly touted now for those under 65, particularly in PPACA), many types of health-care providers and services are included and covered in the Part C network that are not paid for at all in Original Medicare.

      3. You say:

      “I am not sure of all the state regulations (concerning Medigap)… (but I think it’s just a matter of premium cost by age rating).

      As to the risk that a senior might not be able to get on a Medigap plan after leaving a Part C plan (or get on one right away, or get on one without a pre-existing-condition clause, or…), also see “Medicare and You,”, page 66 and surrounding pages, as well as your own state’s rules. The rules for Medigap and types of Medigap plans available vary widely state to state. This issue is not simply one of premium cost to the senior, as you seem to think. On the other hand, in some states there is none of the age rating of which you write (for example, in Massachusetts a senior pays the same for a given Medigap policy whether he or she is 65 or 95).

      4. You say:

      “Insurance and care represent two different things.”

      I can’t agree more. This is actually quite a different position than most people who study health insurance reform. But I don’t understand your making this statement in the context of this thread.

      5. You say:

      “Cherry picking was proven by the GAO with regard to the first attempt at Medicare HMO’s.”

      As I said, apparently you are dealing with very old information. I’d like to read that old data if you can supply a reference. The first attempts at Medicare HMOs were over 20 years ago under demonstration projects and there have been three major legislative changes since then and at least two minor changes (not to mention yearly regulatory changes).

      In summary, returning to the original point of the post and my original comment. The author of a book being reviewed says Original Medicare pays less than 50% of an average (FFS implied) subscriber’s healthcare costs, including his or her long term care. My comment was that Original Medicare pays less than 50% of an average FFS subscriber’s healthcare costs even if you don’t take LTC into account and don’t include institutionalized subscribers with LTC expenses (and I referenced the data source). That is, the situation is even worse than the book’s author portrays. But I opined that this is not the case for Medicare Part C subscribers because Part C includes catastrophic coverage and coverage of many services not included in Original FFS .Medicare, both of which substantially lower Part C subscribers out of pocket healthcare expenses.

      • Dennis writes:”You seem to be dealing with very old information about Medicare”

        1) You are playing the game of semantics. I stated to Austin “return to Medicare”. It is common to call traditional Medicare (FFS Medicare) ‘Medicare’ while designating Part C programs by the individual program name such as Medicare Advantage (MA). One can go back and forth between Part B and Part C though it is limited to enrollment periods.

        This is neither new nor old. It just is. I don’t know what old information you think I am dealing with except for the fact that many of the studies were done on the original Medicare HMO’s. It takes time and money to produce studies and much can be extrapolated based upon incentives.

        2)”No one gives up ANY Medicare benefits by choosing a Part C Medicare health plan.”

        Again you might be dealing with semantics or perhaps might not understand certain concepts. You confirm my feeling when later you state: “Of course, the rules are different depending on what Parts of Medicare a senior picks;”

        Medicare (traditional Medicare) permits one to see any Medicare participating physician and if seen Medicare will reimburse for those thin.g covered by traditional Medicare. If one is in a MA plan they lose this ability. That ability is a valuable benefit. I don’t want to go into a debate between FFS Medicare and HMO Medicare which is what some of the Part C plans are. If you wish to read a study that discusses the differences see Luft’s megastudy written many years ago. Graphically one can look at the different forms of insurance like this:
        pure FFS—–PPO—–hybrids——HMO

        Dennis:”The wording has been similar in all versions of “Medicare and You” at least back to 2010.”

        Be careful of over reading Medicare’s pamphlets for the general public. It is not made to provide the general public with expertise in Medicare law.

        Dennis: “– One of the major rules differences (the most important in my opinion) and one that works strongly in the Part C beneficiary’s favor is that all Part C Medicare plans must provide catastrophic coverage.”

        Yes, and in part that is made possible by paying MA more than it should get per capita. Here, however, it appears you are trying to defend the Medicare HMO concept. There are pros and cons. Here is one of the cons from John Ware’s extremely well respected study. Differences in 4-Year Health Outcomes for Elderly and Poor, Chronically Ill Patients Treated in HMO and Fee-for-Service Systems “Conclusions.–During the study period, elderly and poor chronically ill patients had worse physical health outcomes in HMO’s than in FFS systems” If this is the subject you wish to debate I would be more than willing to oblige where it is appropriate.

        3) Your knowledge is too generalized. I was concerned about the states where one risks “not being able to buy Medigap insurance in some states if you make this switch.” You didn’t answer the question of which states. We all know what the federal government states. What we or at least I don’t know is what all the individual states are doing. That was my question and that is a question you left unanswered.

        Additionally you generalized my statement and distorted its meaning. “What happens I believe in most states is that the fee might be different from the fee that would have been offered at age 65.” Note how I said *most* states not all states.

        4) Insurance is not health care. The meaning is clear. If an HMO denies care (rightly or wrongly) one doesn’t get the care.

        5) The incentives have not changed and thus our expectations should not change either. Yes, there have been changes over time, but it is up to you to prove those changes have ended the problem. That is something you haven’t done.

        You are welcome to your conclusions.