Blog Home


Is Medicare More Efficient Than Private Insurance?

August 9th, 2011

Of all the issues bandied about in the recent debate over the debt ceiling, none generated more contention, more TV ads and more unseemly rhetoric than potential changes to Medicare.

Health economists generally believe that Medicare is on an unsustainable course and is desperately in need of reform. Yet public opinion polls show that most seniors disagree. They not only resist cuts in Medicare to solve the problem of federal deficit spending, they also resisted the spending cuts and delivery of care innovations envisioned by the Affordable Care Act (ACA), as well as the private insurance innovations envisioned by Rep. Paul Ryan (R-WI) and the House Republicans.

In short, most seniors would like to keep Medicare just like it is.

A similar view is held by a small, but vocal group on the left that favors single-payer national health insurance. The Physicians for a National Health Program, for example, claims that Medicare has lower administrative costs than private insurance and is able to use its monopsony (single-buyer) power to suppress provider fees. The group, which is resistant to managed care, favors “Medicare for all” and endorses a bill to do just that by John Conyers.

Paul Krugman, writing in The New York Times, also argues this way. He points to a chart (see Figure I) which seems to show that Medicare per capita spending is growing at a slower rate than private insurance. Krugman, along with others,  touts the slower rate growth in the Canadian health care system (also called “Medicare”).  In recent editorials, both Krugman and Robert Reich have joined the call for Medicare for everyone.

Are these unconventional critics right?

Let’s begin with a fundamental point that almost everyone tends to ignore. Medicare is not actually managed by the federal government. In most places it is managed by private contractors, including such entities as Cigna and Blue Cross. To argue that Medicare is more efficient is tantamount to arguing that when Blue Cross is called “Medicare” it is more efficient than when it is called “private insurance.” Further, there is nothing particularly special about the way Medicare pays providers. Private insurers tend to use the same billing codes and their payment rates are often pegged as a percentage of Medicare rates.

Claim Of Lower Medicare Administrative Costs Is Based On An Incomplete Comparison

What about the claim that Medicare’s administrative costs are only 2 percent, compared to 10 percent to 15 percent for private insurers? The problem with this comparison is that it includes the cost of marketing and selling insurance as well as the costs of collecting premiums on the private side, but ignores the cost of collecting taxes on the public side. It also ignores the substantial administrative cost that Medicare shifts to the providers of care.

Studies by Milliman and others show that when all costs are included, Medicare costs more, not less, to administer. Further, raw numbers show that, using Medicare’s own accounting, its administrative expenses per enrollee are higher than private insurance. They are lower only when expressed as a percentage – but that may be because the average medical expense for a senior is so much higher than the expense for non-seniors. Also, an unpublished ongoing study by Milliman finds that seniors on Medicare use twice the health resource as seniors who are still on private insurance, everything equal.

Ironically, many observers think Medicare spends too little on administration, which is one reason for an estimated Medicare fraud loss of one out of every ten dollars of Medicare benefits paid. Private insurers devote more resources to fraud prevention and find it profitable to do so.

Are Medicare Costs Growing More Slowly? No.

What about the claim that Medicare’s cost per enrollee is growing more slowly? The problem with the diagram in Figure I  is that it ignores the falling share of private out-of-pocket spending  over the past 40 years, as the insurance share of the bill escalated and the changing demographic structure of the two populations as the private insured market aged. A more accurate picture is provided by the Congressional Budget Office (Figure II), which makes appropriate adjustments and calculates spending in excess of GDP growth for the public and private enrollee populations. As the figure shows, Medicare has been growing faster than the private sector. For that matter, Medicaid has also been growing faster.

Source: Krugman/New York Times; Centers for Medicare and Medicaid Services

Source: Congressional Budget Office

As the CBO acknowledges, its comparison is far from perfect. The “other” category includes the uninsured as well as out-of-pocket spending by Medicare enrollees. Still, there is no reason to believe that overall spending would have been lower if the entire country had been in Medicare for the past 35 years.

Looking forward, the CBO believes that Medicare will grow more slowly than the private sector. It will grow a lot more slowly if the provisions of the Patient Protection and Affordable Care Act (PPACA) are implemented without any changes. Under the act, Medicare will suppress provider fees so immensely that real per capita Medicare spending will grow no faster than real per capita GDP. Yet the rest of the health care system will be growing at twice that rate. As the Medicare Office of the Actuaries has explained, this means greatly reduced access to care for the elderly and the disabled. As we previously explained, a two tiered system will emerge before this decade is out. Because of the political ramifications of all this, almost no one believes that the PPACA will remain unchanged. Like the so-called “doctor fix,” the Medicare cuts in the law are likely to be restored.

The Argument Based On Government Single-Buyer Market Power: Five Problems

What about the argument that government can use its power as a single buyer to suppress providers’ fees? There are five problems with it.

Health care markets are local. First, we don’t buy health care in a national market. We buy locally. And in local markets, private entities are often as big, or bigger, than Medicare (the auto companies in Detroit, for example, or the mine workers and their employers in West Virginia). There is nothing the US government can do that a lot of private companies and unions cannot also do.  Similarly, if Canada is seen as the ideal, nothing is stopping the auto companies and the UAW from creating a global budget and rationing care for auto workers just the way the Canadians do it. That they choose not to do so is telling.

Side effects of suppressing provider fees. Second, there are negative consequences from unduly suppressing provider fees. Doctors can leave the city, the state, or even the country where they live and go elsewhere. Able people can also avoid the profession altogether.  If we paid doctors only the minimum wage, for example, medicine would attract only those people who can earn no more than the minimum wage doing something else. The suppression of provider payments ultimately harms patients as highly qualified providers exit the market. The effects of price controls in health care will be similar to their effects in other markets.

Cost-shifting. Third, the suppression of provider payments shifts costs from patients and taxpayers to providers. Shifting costs, however, is not the same thing as controlling costs. Providers are just as much a part of society as patients. Shifting cost from one group to the other makes the latter group better off and the former worse off. It does not lower the cost of health care for society as a whole, however. In fact, it introduces a cost to society as the supply of providers falls.

Political pressures and lobbying. Fourth, the argument overlooks the fact that public insurance in a democracy is ultimately subject to pressures at the ballot box. Providers get to vote too. They also can make campaign contributions and lobby. Patients can also exert political pressure. Political competition in a democracy constrains public policy in much the same way that economic competition constrains the behavior of private firms in the marketplace.

You can get Medicare price controls without Medicare. Finally, if it really were desirable to have everyone pay low prices we do not need to enroll everyone in Medicare to achieve that outcome. We could instead impose Medicare-type price controls on the entire health care system. In fact, one organization advocates that very thing. Doing so would run into all the problems listed above, however.

Medicare Advantage Plans Deliver More Than Traditional Medicare

What about the observation that private sector Medicare Advantage plans are costing the government more than what Medicare would have paid? Some of the “over payments” represent Congressional interference (e.g., the desire to make sure the plans serve rural areas) and some represent poor administration of the program. A more basic issue, however, is that these plans are not just providing Medicare services. They are also providing a less expensive, more efficient form of medigap coverage — mainly to low-income seniors who could not otherwise afford supplemental insurance. If we are indifferent about whether seniors have this insurance, that is one thing. But if we are going to insist that everyone have it, integrated private plans are probably much superior to more subsidies for the current Medicare/medigap arrangement.

For our purposes, the most interesting characteristic of Medicare Advantage plans is that many of them are already doing what the Obama administration says it wants to do with Medicare as a whole — without any prodding or nudging from the federal government. That is, many of these plans are using coordinated/integrated/managed care systems to achieve fewer admissions, fewer readmissions and fewer hospital days than conventional Medicare. By contrast, the Obama administration’s plan to encourage Accountable Care Organizations is so bogged down in bureaucratic rulemaking that our most cost effective health organizations are refusing to participate.

On Innovation, Medicare Follows Rather Than Leads

What about the argument that Medicare is needed in order to spur doctors to practice medicine efficiently? As in the case of Medicare advantage plans, all too often, Medicare is a follower, not a leader, on the innovation front. It is more likely to slow things down than to speed things up. What spurs private firms to be efficient is competition for consumers, not regulation.

In health care, many private sector entities are already doing what the administration says needs to be done:

  • Concierge doctors are consulting by email and telephone, keeping electronic medical records (EMRs), prescribing electronically and offering same day or next day appointments.
  • Walk-in clinics are posting (transparent) prices, using evidenced-based medicine by following computerized protocols, and keeping EMRs as well.
  • Cosmetic and Lasik surgeons routinely offer “bundled” prices, compete for patients based on price and quality and have lowered the real price of their services over the past decade.
  • There are lots of successful examples of coordinated care, integrated care, managed care, medical-home care and home-based care (see, for example, here) — almost all of it developed despite Medicare incentives not to do so and in some cases saving Medicare millions of dollars without any compensation.

Genuine reform would seek ways to encourage more such private sector activity. The Ryan proposal would allow a more flexible and robust Medicare Advantage program, in which health plans would compete on price and quality for patient dollars.

Email This Post Email This Post Print This Post Print This Post

 to the #1 source of health policy research.

19 Trackbacks for “Is Medicare More Efficient Than Private Insurance?”

  1. Is Government Provision the Answer? (Part II) | The Beacon
    November 5th, 2012 at 1:03 pm
  2. Conservatives: You Lose, and You Lose by old_skul - Page 13 - TribalWar Forums
    June 28th, 2012 at 6:49 pm
  3. If Massachusetts Can Do It. - Page 5
    April 9th, 2012 at 8:56 am
  4. Hypocrisy on Medicare Reform | John Goodman's Health Policy Blog |
    January 11th, 2012 at 9:17 am
  5. Medicare Cost Increases Drop | John Goodman's Health Policy Blog |
    January 5th, 2012 at 11:53 am
  6. Keynesianism, Social Services, and Solvency | Economic Horizons
    November 17th, 2011 at 5:20 pm
  7. Keynesianism, Social Services, and Solvency | US Economic Directions
    November 15th, 2011 at 1:00 pm
  8. The Collectivist Mind | John Goodman's Health Policy Blog |
    October 31st, 2011 at 8:14 am
  9. Keynesianism, Social Services, and Solvency « Public Discourse
    September 18th, 2011 at 8:30 pm
  10. When Health Insurance is Free | The Health Care Blog
    September 16th, 2011 at 12:45 pm
  11. When Health Insurance is Free | John Goodman's Health Policy Blog |
    September 14th, 2011 at 8:21 am
  12. Health Wonk Review: Hurricane Irene Edition - Forbes
    September 9th, 2011 at 9:43 am
  13. Health Wonk Review: Hurricane Irene Edition « 4-Liability
    September 2nd, 2011 at 8:03 am
  14. Health Reform Scam » Is Medicare More Efficient Than Private Insurance?
    August 22nd, 2011 at 10:04 am
  15. Is Medicare More Efficient Than Private Insurance? | John Goodman's Health Policy Blog |
    August 22nd, 2011 at 8:18 am
  16. | PointOfLaw Forum: Douglas v. Independent Living Center of Southern California
    August 11th, 2011 at 8:46 am
  17. Medicare Is More Efficient If You Ignore Its Inefficiencies - Hit & Run : Reason Magazine
    August 10th, 2011 at 10:28 am
  18. Medicare and facts | The Incidental Economist
    August 9th, 2011 at 12:57 pm
  19. Krugman Disposed Of | John Goodman's Health Policy Blog |
    August 9th, 2011 at 11:11 am

22 Responses to “Is Medicare More Efficient Than Private Insurance?”

  1. Thomas Cox Says:


    Sorry nogden – didn’t know you had responded.

    I think the problem is that you are confusing low loss ratios with insurer efficiency. Nothing could be further from the truth.

    Insurer efficiency is having a loss ratio close to the expected loss ratio. From your comments one might think that an insurer with a loss ratio of 0.00 would be your ideal of a perfectly efficient insurer. The problem is that your efficient insurer isn’t paying any claims.

    There is nothing academic about the fact that insurers’ loss ratios are primarily the result of randomness in the health care experiences of their policyholders. One policyholder gets sick and another does not.

    Your whole premise seems to be that insurers, left to their own devices, will find ways to save money by having some keen sense of the health care needs of, and appropriate treatment protocols for, human illness and accident.

    But that isn’t insurance at all, it is medicine, nursing, biology, pharmacology… Insurance is about managing risk by aggregating the random health experiences of large numbers of policyholders. Your ideal insurer is more likely to spend its time and energy developing new ways to exclude coverage.

    It would be good for you to pick up a statistics book if you want to learn about risk management through insurance.

  2. nogden3929 Says:

    “The most efficient insurer, when it comes to keeping the actual costs close to the true average costs for the population would be an insurer of infinite size. Why? Because the standard error in the estimate of the true population average cost is zero for an insurer of infinite size and larger for every insurer smaller than this.”

    This is a great academic definition of efficiency but doesn’t exist in the real world. As size increases efficiency decreases once you get over a population size sufficient to smooth out the variations. The underwriting you describe is all retrospective, based upon historical data if an insurer of infinite size underwrote the population they could hit their mark, a smaller insurer that was able to spot trends and take action before the next annual rate setting will far out perform the infinite carrier. This is born out in the real world comparison of self funded plans vs BUCA. Self Funded plans constantly outperform BUCA because they are small enough to react to changes in market. I can spot a problem and resolve it before BUCA even knows it exist.

    For example, Lipitor came out with a co-pay card to offset the member cost that has been increased by insurers to steer them away from Lipitor. My self funded plans already increased the co-pay on Lipitor to offset the benefit of the co-pay card. BUCA hasn’t and most likely can’t respond, at their size they are unable to deal with small dollar opportunity even when those opportunities collectively add up to serious money.

    There is a reason every large employer is self funded and not fully insured with a carrier, its more efficient not to be, at least outside of academia.

  3. Thomas Cox Says:

    The really disturbing thing is that we ignore efficiency completely.

    So, let’s take a quick look at what it means to be an efficient insurer.

    Insurance is all about managing fluctuations in health care costs across a population. One member of the population has very high costs, many others have no costs or very small costs. The most efficient insurer, when it comes to keeping the actual costs close to the true average costs for the population would be an insurer of infinite size. Why? Because the standard error in the estimate of the true population average cost is zero for an insurer of infinite size and larger for every insurer smaller than this.

    But, we obviously cannot have insurers of infinite size, so let’s do a “thought experiment” where we start with an insurer that is somewhat efficient and then compare the efficiency of other insurers that are larger, or smaller, than the acceptably efficient insurer. I do this in a number of papers (e.g.:

    Cox, T. (In press). The Impact of Size on Success of Health Insurance Companies. Nurse Leader .

    Cox, T. (2011). Exposing the true risks of capitation financed healthcare. Journal of Healthcare Risk Management . 30: 34-41. doi: 10.1002/jhrm.20066

    Cox, T. (2010). Legal and Ethical Implications of Health Care Provider Insurance Risk Assumption. JONA’S Healthcare Law, Ethics, and Regulation, 12(4): 106-116.

    Cox, T. (2009). Commentary: A Common Flaw in Health Care Finance Mechanisms That Adversely Affects Service Quality and Capacity: Professional Caregiver Insurance Risk In The Trenches. Journal of Psychiatric Mental Health Nursing, 16: 938-940.

    Cox, T. (2006). Professional caregiver insurance risk: A brief primer for nurse executives and decision- makers. Nurse Leader, 4 (2): 48-51.

    Cox, T . (2001). Risk theory, reinsurance, and capitation. Issues in Interdisciplinary Care, 3 (3): 213- 218.) but most people do not quite get it yet.

    So, my usual example is a Paradigm Insurer that covers 1,000,000 policyholders, has a loss ratio that fluctuates from year to year by an amount consistent with a standard error for its portfolio loss ratio of 5% of its annual premiums. I assume that 75% of the premiums it collects covers the costs of health care services, 5% is allocated as a profit margin, another 5% is a risk premium for its insurance risk accumulation and management services, and 15% of its premiums cover the costs of non-health care costs.

    PI earns a profit of at least 10% when its actual loss ratio is 0.7500 or less, and this occurs with probability 0.5000. PI earns a profit of at least 5% when its actual loss ratio is 0.8000 or less, and this occurs with probability 0.8413. PI earns a profit of at least 0% when its actual loss ratio is 0.8500 or less, and this occurs with probability 0.9772. PI has a very slight probability (0.00135) of incurring losses of 5% or more at an actual loss ratio of 0.9000 or higher.

    The standard errors of insurers smaller and larger than PI are all different. An insurer 10 times larger than PI has a standard error that is about 1/sqrt(10)th the size of PI’s standard error. An insurer 100 times larger than PI has a standard error that is about 1/10th the size of PI’s standard error.

    On the other hand, an insurer 1/10th as large as PI has a standard error that is about sqrt(10) times the size of PI’s standard error and an insurer 1/100th as large as PI has a standard error that is about 10 times larger than PI’s standard error.

    All insurers larger than PI have smaller standard errors but identical expected costs. This means that all insurers larger than PI have the same probability of profits of 10% or more (0.5000), higher probabilities of profits greater than 5%, much higher probabilities of profits greater than 0%, and virtually no chance of incurring a loss, than PI. THEY ARE MORE EFFICIENT INSURERS SOLELY BECAUSE THEY INSURE MORE POLICYHOLDERS AND HAVE SMALLER STANDARD ERRORS!!!!!!!!!!!!!!!

    All insurers smaller than PI have larger standard errors but identical expected costs. This means that all insurers smaller than PI have the same probability of profits of 10% or more (0.5000), lower probabilities of profits greater than 5%, much lower probabilities of profits greater than 0%, and they are highly likely to incur losses greater than 5% of their premium revenues, than PI. THEY ARE LESS EFFICIENT INSURERS SOLELY BECAUSE THEY INSURE FEWER POLICYHOLDERS AND HAVE LARGER STANDARD ERRORS!!!!!!!!!!!!!!!

    If you do not understand what it means to be an efficient insurer you are highly likely to make unsupported statements about what does, and what does not constitute efficiency.

    In particular, you are highly likely to associate reduced costs for private sector insurers that result from being able to kick policyholders who are too sick or poor to pay their premiums with their efficiencies as risk managers. Transferring the costs associated with canceled policyholders to the public or charitable payment systems is not a sign of insurer efficiency, it is a sign of the depths to which insurers will resort to extract unconscionable profits from healthy policyholders while failing to fulfill their obligations to their sick and injured policyholders.

    The largest insurer for any country is an insurer that covers the entire population. No insurer smaller than this natural size limitation can ever be a more efficient risk manager than the national insurer. Any perceived benefits of competition between large numbers of small insurers is ALWAYS the result of a failure to understand the mathematical and probabilistic theory underlying all sound insurance operations.

    Fortunately, failure to grasp what insurer efficiency really is can be easily corrected – just read the above referenced papers.

  4. John Goodman Says:

    @ Roger Collier
    Adjudicating and paying claims is what insurers do. Medicare contracts out its “insurance functions” because private companies are so much more efficient at it than the government.
    It really doesn’t matter who does the study of administrative cost, once you factor in the fact that the social cost of collecting a dollar of taxes is 25 cents or more, it is impossible for government insurance to have lower administrative costs than private insurance.
    I have no idea why you would want to remove the uninsured from the “private sector” much less how you would even begin to go about doing it. The private sector, including the uninsured, is controlling costs better than Medicare is. Why is that so difficult to understand?
    Don’t confuse what Medicare pays with what Medicare Advantage plans are doing. The former is subject to all kinds of Congressional restrictions due to political pressures. The best MA plans are outperforming Medicare on every measurable dimension.
    Ironically, the best MA plans are doing some of the very things the Obama administration says it most wants done (medical homes, ERMs, team-based care), but the way they are doing it is the result of competition in the marketplace, not bureaucratic rule making. See the post at my blog.

  5. Devon Herrick Says:

    Roger Collier argues that CMS manages Medicare, private insurers merely pay claims. Then you would need to include much of the bureaucratic overhead of running the Centers for Medicare and Medicaid Services to the overhead costs of processing Medicare claims.

    Suggesting funding Medicare is easy because it piggybacks on the current payroll system is a tough sell. As a former accountant for a hospital, the payroll tax system is anything but easy.

    It’s self-evident that it’s easy to make claims processing appear small as a percentage when each claim is large (i.e. seniors hospital charges compared to non-seniors physician bills).

    All Other medical spending rose faster than Medicare spending in only one period, because the CBO had to base projections on draconian cuts in Medicare physician fees. The Medicare chief actuary has stated he doesn’t believe cuts of that magnitude are possible if seniors are to retain access to care.

    Ken Thorpe and others have consistently found that Medicare Advantage plans provider richer benefits that traditional Medicaid.

  6. Roger Collier Says:

    There’s plenty of room for improvements to Medicare, but John Goodman’s misstatements are more likely to create obstacles than solutions.

    Here are a few of the errors in John’s post:

    “Medicare is not actually managed by the federal government…” Wrong. CMS manages Medicare. The role of Blue Cross and other intermediaries is only to process claims in accordance with CMS rules.

    “The problem with this comparison is that it includes the cost of marketing and selling insurance as well as the costs of collecting premiums on the private side, but ignores the cost of collecting taxes on the public side. It also ignores the substantial administrative cost that Medicare shifts to the providers…” John is confused, although the costs of private insurance marketing and premium collection are certainly part of the argument for a single payer system. As any provider will testify, much of their administrative cost burden results from private insurer demands, while the cost of Medicare tax collection is minimal since it piggybacks on the underlying tax system.

    “Studies by Milliman and others show that when all costs are included, Medicare costs more, not less, to administer…” Misstatement of outdated studies. Misstatement because the 2005 Milliman study actually concludes that Medicare’s administrative costs are 5.2 percent of claims costs compared with 16.7 percent for private insurance (including commissions, profit and taxes). Outdated because the study was conducted prior to Medicare D implementation. And John’s per beneficiary comparison is equally misleading since it fails to consider Medicare’s higher percentage of hospital claims.

    “As the figure shows, Medicare has been growing faster than the private sector. For that matter, Medicaid has also been growing faster…” Not exactly, John. If we eliminate the approximately 50 million Americans with no coverage from the “Other” base, we find that Medicare’s costs per capita have grown at about the same rate as private insurance, in spite of the addition of Medicare D benefits in 2006.

    “[Medicare Advantage] plans are not just providing Medicare services. They are also providing a less expensive, more efficient form of medigap coverage…” John apparently believes that the primary reason MA plans cost some 13 percent more than traditional Medicare is that they offer more benefits. In fact, the vast majority of MA plan bids to CMS for basic benefits—before adding additional services—are significantly higher than traditional Medicare costs.

    It’s a pity that John has misstated so many facts about Medicare. If we are to control Medicare costs without slashing benefits, we need accuracy of understanding—not egregious errors.

  7. John Ballard Says:

    @Devon Herrick

    “Cosmetic surgery is one of oue best examples on how a free market in health care might work. ”

    Glad you mentioned that. I would guess dental practices might show a similarly modest inflationary trend line. The working poor have never and will never participate in those or several other medical specialties. Many of the so-called “middle class” also cannot afford the high end of medicine.

    “If the scheduled cuts to Medicare provider fees are allowed to take effect, we will be left with a two-tiered health care system, where those covered by public programs have a difficult time finding doctors who will treat them.”

    Exactly. And I would say ten percent of the roster is approximately correct. Hard to tell. At present 45% of all health care dollars are from tax money and several millions are doing without insurance. When and if that population gets covered (which appears questionable at the moment) that percentage will rise to about half of all healthcare dollars. Cutting reimbursements from the government is nothing more than cutting expenditures, exactly what the Tea Party and other doctrinaire extremists are yelling about. Perhaps you think Medicare beneficiaries will fill the gap? I think not. That’s the population that already has many whose income is too small to even pay tax. (Big piece of that bunch referred to as “half of all Americans don’t pay any income tax.” Ask Mr. Perry about that.)

    I say let the chips fall as they may;. We’ll see how many concierge practices flourish on ten percent of their current patient roster, losing the Medicare revenue stream in the bargain.
    Let’s face it. Barring unforeseen changes, America is on the way to a Medicare program which will more nearly resemble a FEDERALIZED MEDICAID system.

  8. Devon Herrick Says:

    @ jimrecht

    CAHI may have commissioned the work, but the analysis was performed by the actuarial firm Milliman.
    Moreover, there are some administrative tasks that Medicare does not pay for that would probably strengthen the program. For instance, fraud my consume 10% of Medicare expenditures. Economic theory suggests Medicare should invest in fraud detection until the last dollar spent equals a dollar saved. The same is true of other areas where additional administrative costs might improve quality or reduce costs.

    @ John Ballard
    If the scheduled cuts to Medicare provider fees are allowed to take effect, we will be left with a two-tiered health care system, where those covered by public programs have a difficult time finding doctors who will treat them. One likely result will be more physicians converting to concierge practices where only 10% of a physician’s former patient roster is retained (those willing to pay a $1500 retainer).

    Cosmetic surgery is one of oue best examples on how a free market in health care might work. My analysis on cosmetic surgery fees illustrates what happens when doctors compete on price. The average price for cosmetic surgery has climbed more slowly than inflation over the past 15 years. By contrast, the Medical Care component of the CPI has doubled during that period. See:

  9. jimrecht Says:

    @ John Goodman

    Additionally, and regarding your reliance on the arguments of Zycher, here’s a representative excerpt from that jaw-dropping article: “The lower reported administrative costs for Medicare are unsurprising, in that Medicare spends substantially less on such functions as marketing [and] risk evaluation…[t]his does not mean that the higher reported administrative costs of private health insurance are ‘wasteful.'” Correct — unless you happen to consider marketing and underwriting to be wasteful. Most of us do!

    Finally, the unpublished study you referenced was funded (and printed) by The Center for Affordable Health Insurance, described on its own website as an “association of insurance carriers active in the individual, small group, HSA and senior markets.” Get your references straight, man!

  10. jimrecht Says:

    @ John Goodman

    “There is nothing wrong with the Book study.”
    Wow! Roma locuta est, causa finita est.

  11. John Ballard Says:

    Very revealing that concierge practices, walk-in clinics and cosmetic surgery are cited as examples of how much better the private sector is than Medicare. Do tell.

    There will always be a market for the carriage trade and those who follow fashion in all its forms. Medicare, thank you, is a safety net for those who cannot avail themselves of more costly forms of medicine. After the chips fall (cuts to providers and all that) I’m expecting it to look more like a federal version of Medicaid, consisting of what might be called wholesale priced medicine.

    At that point your retail medicine providers (with two or more layers of mark-up for marketing, profits, bonuses, other corporate benefits) will be able to do just fine without tax dollars collected from payrolls and other sources.

    At this point something like forty-five percent of all healthcare dollars come from the government. When ACA kicks in, barring intervention (which is almost sure to happen), that will rise to about half. When “consumers” (that’s what we now call ALL sick people but after Medicare is cut the term will have real meaning) get a taste of the real costs of retail medicine we’ll see how competitive the private sector can really be.

  12. John Goodman Says:

    @ Tom Miller

    The CBO believes Medicare will grow more slowly in the future than ALL other health spending mostly because the ACA suppresses provider fees. But few thoughtful people believe these fee cuts will actually happen.

    The Milliman work is consistence with their earlier work comparing people of the same age in Medicare and the private sector.

    @ Jim Recht

    There is nothing wrong with the Robert Book study. You ignored the Manhattan Institute study by Ben Zycher, which also shows Medicare administrative costs are higher than private insurers. Finally, the unpublished study you reference was not by done by an insurance company but by Milliman, an actuarial consulting firm.

  13. John Goodman Says:

    Dr. Geyman is much better at ad honimem attacks and citing his own work than he is at analyzing data or understanding basic economics.

    Nothing in his comment convinces us to change a single word in our post.

  14. jgeyman Says:


    Yesterday’s blog post by John Goodman and Thomas Saving of the National Center for Policy Analysis (NCPA) is the latest in an avalanche of unfounded assertions and distortions that have characterized the writings from this center for many years. The Dallas-based NCPA, established in 1983, describes itself as a “nonpartisan public policy research organization, with the goal to develop and promote private alternatives to government regulation and control, solving problems by relying on the strength of the competitive, entrepreneurial private sector” (its website). This latest post puts forward, without context and with cherry-picked references, carefully selected statements that might seem to some to support their case—that deregulated markets will solve all of our health care problems. It would take a very long paper, or a number of papers, to respond to the many unfounded claims in their latest post.

    Here are just three of their unfounded claims, together with references from the health policy literature and recent publications that rebut their assertions:

    • Re the alleged advantages of privatized Medicare, see my 2006 book (Geyman, JP. Shredding the Social Contract: The Privatization of Medicare. Monroe, ME. Common Courage Press, 2006), my extensive article in The International Journal of Health Services (Geyman, JP. Privatization of Medicare: Toward
    disentitlement and betrayal of a social contract. Intl J Health Services 34 (4): 573-94, 2004), a 2009 report by the Committee on Energy and Commerce (Committee on Energy and Commerce. New report highlights Medicare Advantage insurers’ higher administrative spending. Washington, D.C., December 9, 2009), a 2010 article in the Wall Street Journal on retrenchment of private Medicare plans (Johnson, A. Private Medicare plans are retrenching. Wall Street Journal, November 19, 2010: B1), and a recent article in The New England Journal of Medicine describing the failures of regulated competition among private insurance companies in the Netherlands and calling into question managed competition as a model for private Medicare plans in the this country. (Okma, KGH, Marmor, TR, Oberlander, J. Managed competition for Medicare? Sobering lessons from the Netherlands. N Engl J Med, June 15, 2011)

    • Re the alleged advantages of private health insurance over single-payer national health insurance, see my 2008 book on the private health insurance industry (Geyman, JP. Do Not Resuscitate: Why the Health Insurance Industry is Dying, and How We Must Replace It. Monroe, ME. Common Courage Press, 2008), my extensive article in The International Journal of Health Services (Geyman, JP. Myths and memes about single-payer health insurance in the United States: A rebuttal to conservative claims. Intl J Health Services 35 (1): 63-90, 2005), and a 2009 report by the Congressional Research Service, The Market Structure of the Health Insurance Industry (Austin, DA, Hungerford, TL. The Market Structure of the Health Insurance Industry. Washington, D.C, Congressional Research Service, November 17, 2009).

    • Re the claimed efficiencies of competition in health care, see a multi-year study by the Community Tracking Study showing the failures of markets to be more efficient or to enhance the quality of health care (Nichols, LM et al. Are market forces strong enough to deliver efficient health care systems? Confidence is waning. Health Affairs (Millwood) 23 (2): 8-21, 2004) and a recent article by Mark Weisbrot, co-director of the Washington, D.C-based Center for Economic and Policy Research (Weisbrot, M. Problems of U.S health care are rooted in the private sector, despite right-wing claims. McClatchy-Tribune Information Services, July 20, 2011).

    Health policy is too important to leave to the biased, well-funded propaganda
    machine of these “research” organizations that keep promulgating policies that have long since been discredited, either by their failing track record or legitimate research studies.

    To read this article with hot links to the quoted publications, please visit:

    John P. Geyman, M.D.
    Professor emeritus of Family Medicine, University of Washington

  15. jimrecht Says:

    This blog entry leans heavily — almost exclusively — on one unpublished study funded by the private insurance industry, and on a 2009 paper by Robert Books published by the explicitly partisan Heritage Foundation. The latter has been debunked so thoroughly and repeatedly over the past 2 years (see, for example, the Kaiser Family Foundation’s 2011 “Primer on Medicare Financing”) that one wonders whether Goodman/Saving are underinformed or simply disingenuous.

  16. JoeBarnett Says:

    I suspect that the notion that seniors want Medicare to stay just like it is an artefact of the alternatives offered and the way questions are worded. Seniors are paying increasingly higher premiums both to Medicare and for Medigap insurance, and yet the coverage is limited and out-of-pocket costs can be high.

  17. Thomas Miller Says:

    There are lots of interesting and valuable points made in this post. However, several assertions raise some secondary questions about the true nature of the statistical comparisons.

    First, the link to a description of the unpublished Milliman study (a recent WSJ op ed) actually compared Medicare beneficiaries over age 65 to individuals under 65 with the same health-care needs (not to seniors who are still on private insurance). Perhaps the risk adjustment was so remarkably sophisticated that it made “everything equal” by accounting for all the differences in spending patterns for those two different types of patients, or perhaps that’s why the ongoing study is still unpublished.

    Second, CBO’s most recent analysis of per capita spending in excess of GDP Growth does not rely on the 1975 to 2008 figure displayed. It concludes, “In CBO’s judgment, the average rate of excess cost growth since 1985 best reflects features of the health care and health insurance systems that are likely to endure.” Hence, its newest long-term projections rely on the average rate of cost growth observed between 1985 and 2008.CBO explains:
    “Excess cost growth was lower, on average, during that period than during the longer 1975–2008 period. That slowing probably stems, at least in part, from two important shifts: Private health insurance moved away from indemnity policies—which generally reimburse enrollees for their incurred medical costs, and which predominated before the 1990s—and toward greater management of care; and Medicare shifted from cost-based payment methods to fee schedules that seem less conducive to spending growth because price increases are constrained. Excess cost growth was even lower, on average, during the shorter 1990–2008 period, but that average gives a good deal of weight to the years in the 1990s when managed care was spreading most rapidly—some of which probably represented a one-time downward shift in health care costs rather than a change in the underlying growth rate. In CBO’s judgment, the average rate of excess cost growth since 1985 best reflects features of the health care and health insurance systems that are likely to endure.

    Here’s what fuller disclosure of the CBO table for different periods of time actually looks like:

    Table 2-1.
    Excess Cost Growth in Spending for
    Health Care

    Medicare Medicaid All Other Total
    1975 to 2008 2.5 2.0 1.8 1.9
    1980 to 2008 2.2 1.7 1.9 1.9
    1985 to 2008 1.5 1.2 1.8 1.7
    1990 to 2008 1.8 1.0 1.4 1.4

    Of course, any of these comparisons “are far from perfect.” But there’s no need to stack the statistical deck further by only using the wider spread between Medicare and all other for the 1975-2008 period.

    Third, the authors’ defense of “overpayments” to MA plans is curiously incomplete. We don’t see any estimates of how much the average Medicare FFS enrollee (not just the minority currently purchasing individual Medigap plans) actually pays in additional unsubsidized private plan premiums – – compared to the marginally higher MA plan reimbursements above comparable Medicare FFS costs. And we don’t know whether the authors would suggest that such supplemental coverage should be subsidized for all Medicare beneficiaries, or just low-income ones (apart from the way in which Medigap coverage leverages basic Medicare FFS subsidies to reduce cost sharing and drive overall Medicare spending higher). I was not aware that “we” or “they” are insisting that everyone have this extra insurance

  18. rz Says:

    The Milliman article you cited indicates that Medicare costs run at 6.2% of payments vs. 16.7% for private carriers.
    It’s only when you take out commission, taxes and profit from private insurance that private carriers come in at 8.9%. But I don’t think any private insurers are going to eliminate commissions and profit, do you?
    Then it waves its hand at higher costs per beneficiary and cost of capital before concluding that Medicare must come in around 8%.
    But Medicare advocates maintain that commissions, profits and cost-of-capital constraints are precisely where Medicare maintains a cost advantage.
    In essence, the paper says if you ignore Medicare’s advantages, it enjoys no advantage. I suppose that’s true.

  19. Virginia Traweek Says:

    Your “five problems” section exceptionally well done. Anyone who believes that controlling Medicare cost inflation is as simple as waving a magic wand and cutting reimbursement rates is divorced from reality. When providers have their rates cut, they make up for it by performing more procedures. They resort to volume when margin is cut. If it goes far enough, they leave the profession altogether.

    I also liked the point pertaining to low Medicare administrative costs and Medicare fraud. While there are still plenty of cases of insurance fraud in the United States, you never hear of the fraud anywhere near the scale of Medicare and Medicaid.

  20. Brian1981 Says:

    I like what Dr. Goodman noted about the single payer supporters ignoring “the substantial administrative cost that Medicare shifts to the providers of care. There are so many ways for them to bend the statistics and in the case of the single payer supporters, they just leave out important factors like that, while including things like marketing, selling of insurance, etc. on the private side. They manipulate the numbers in the process.

  21. Devon Herrick Says:

    Health care is one-sixth of our economy. Depending on your age, each person on Medicare stands to benefit from something like $300,000 in lifetime benefits. This money has to come from somewhere.

    The idea that Medicare should continue using Pay and Chase to recover fraudulent payments is misguided. Medicare may have low administrative costs. But one should consider the addition cost of fraud and abuse. A dollar spent preventing fraud might be an administrative expense, but it may save several more dollars from being spent on wasteful or fraudulent expenditures. Merely looking at the low overhead of Medicare obscures the true cost of running the program.

  22. Richard Walker Says:

    Were it not so pitiable, it sometimes is actually amusing to try and follow the Boolean permutations Krugman uses to twist and contort the facts in his public policy pronouncements. To assert that Medicare, or any other government-run health program, is efficient as currently administered is disingenuous at best, yellow journalism at worst. The heft of evidence to the contrary is simply too overwhelming.

Leave a Reply

Comment moderation is in use. Please do not submit your comment twice -- it will appear shortly.

Authors: Click here to submit a post.