• Lower spending is the wrong rationale for private health insurance

    The proposal to raise the Medicare eligibility age from 65 to 67 has legs. Earlier in the week, the Health Leadership Council, a consortium of health industry leaders, endorsed the idea. About that Matt Yglesias wrote,

    [S]hifting marginal people out of Medicare is lucrative for the health care industry. By the exact same token, if you’re interested in reducing health care costs, what you want to do is listen to executives of health care companies and then do the reverse. Rather than raising the Medicare eligibility age, we ought to be lowering it. […] Both domestically and internationally, it’s clearly the case that the way to reduce health care costs is to adopt more statist payment systems. UK health care is cheaper than Canadian health care is cheaper than American health care. In America, Medicare is cheaper than private insurance.

    Interestingly, this used to be well-understood. From the end of World War II up through the Hillarycare episode, the main conservative argument against increased government provision of health care was precisely that by reducing costs it would reduce profits and thereby reduce innovation.

    More recently, conservatives have tried to take up the reverse banner and claim that health care prices are high because of government subsidies and that therefore free market solutions will reduce health care prices. This is, however, contradicted by all the available evidence along with the voluminous history of conservative whining that government-funded health care would be too stingy.

    I’ve got a lot to say about the issues Yglesias raises, so this will be a long post. Before I get to the issues, though, I want to make one thing perfectly clear. You should not care what my opinion is about the virtues of public vs. private spending on health care. What I might find preferable should not be important to you. In fact, I will not state my preferences in this post, and you should not infer what they are from what I write here (because it doesn’t matter). What I think you should care about is what I care about: facts, evidence, and truth, to the extent it is knowable.

    Now, about facts: here are some, with links that back them up.

    All of that is consistent with the hypothesis that public health financing programs can control costs at least as well, if not better, than private ones. In addition to that and as Yglesias wrote, there are the decades of statements by health industry leaders and proponents of private health care that suggest they all know this to be true. You can’t argue low public payments stifle innovation if public payments aren’t below those of the private sector. You can’t argue that low public payments lead to cost shifting if public payments aren’t lower than private ones.

    By the way, I have asked, more than once and publicly, for those who claim public health spending outpaces private health spending to provide anything like a body of evidence I’ve given above. (Aside: I didn’t even work that hard to come up with the list above. There may be a lot more out there I’m not thinking of.*) So far, I’ve seen nothing other than ideas that badly miss the mark. What I tend to see is a lot of creative arguments about why the evidence I’ve cited above is not illustrative of lower levels or growth in public spending, about how they leave out this factor or that consideration or don’t adjust for thus and such. All of this may be true. But a theoretical argument about a hypothetical omitted factor is nothing compared to cold, hard facts. As for those that show private beats public on cost control, I’m still waiting.

    However, even if I take as proven that public health insurance programs control costs better than private ones, that alone does not not mean that public programs are to be preferred. There is value in choice. Having more insurance options increases people’s ability to find a product that suits them best. Leemore Dafny and colleagues estimated the value of the ability to choose plans in an exchange-like setting to be worth over $2,000 more for a family of four relative to typical, constrained choice of employer-based options.

    There is also value in innovation. Prescription drug coverage options were available from private plans decades before Medicare implemented a drug program. Medicare still has no long-term care benefit, though one can buy private long-term care insurance. So, it is eminently plausible that the private markets adjust to consumer demand more quickly than public programs.

    Thus, the right argument for private options is not that they reduce health care spending. The evidence I’ve seen and the statements by the health care industry itself illustrate otherwise. The right argument is that private options offer more choice, greater flexibility, and rapid innovation. Those things are good, but they cost something. A well-functioning, competitive private market can drive that cost to a minimum, but that cost may still be above what we’d pay for a government health plan (albeit one with less or no choice and relatively slow to adjust to the changing needs and demands of Americans).

    Private health plan advocates are right. Choice, innovation, and flexibility are all valuable. They’re so valuable, people may be willing to pay more for them. There’s nothing wrong with that. But paying more means, well, paying more. If you think you can get all that for less, show me the evidence. Until then, I don’t see any reason to, on faith, waive the fundamental law of economics: there’s no such thing as a free lunch.

    * See the Health Affairs post by Diane Archer for more evidence.

    • –“The right argument is that private options offer more choice, greater flexibility, and rapid innovation.”

      Is choice a good thing if its only purpose is to provide choice. That is to say, the chosen option costs more and adds nothing (“nothing” is relative i suppose–that is, if the enrollee thinks they are in a better place so to speak, have they maximized their utility?)

      –Also, on Medicare x last 40 years–the linked post provides no references for the actual % difference relative to private. Whats the delta?

      Finally, I disagreed with Yglesisia’s assertion re: Intl cost growth and “cheaper.” Those countries he mentioned and others, started at a much lower baseline decades ago.

      If annual increases in health spending are any indication, the statist approach is as unsuccessful as any. Years of underinvestment is catching up, and OECD nations are experiencing annual increases in excess of US. They are succumbing to all the same forces.


    • Whether or not I believe they are true, I think the two market criticisms are not as opposed as presented. They both come down to, if there is a single organization, it does not have to worry about costs (because it can’t go bankrupt and there are no alternatives) or attracting customers (because there are not alternatives and no-one can opt out) , then it has no incentive to either 1) change the mix of goods and services it provides or 2) reduce costs for those goods and services.

      Which half of that claim is more important/interesting depends on what people are discussing at the time, but the main point is the same.

      • Exactly. The benefit level is crucial in all this. It’s conceivable, even likely in some areas (depending on provider market power), that private entities can provide a fixed benefit at lower cost than can the government (your point 2). However, it is also conceivable that a private entity, responding in part to consumer demand, will increase the benefit level (and therefore the cost and price of that more generous set of benefits) beyond that which the government would provide (your point 1). There is evidence for both these things within Medicare. It’s precisely the product (features, choice, innovation, flexibility) vs. price tradeoff my post is about. The two are correlated and government and private entities have different incentives for each. The bottom line is, as you say, what matters more to you. That is subjective.

        • Yeah, I was really arguing (clarifying?) against Matt’s comments, but in reply to your post. I should have been more clear about that.

    • Any reactions to this news item, Austin? http://www.kaiserhealthnews.org/Stories/2011/September/15/medicare-advantage-premiums-fall-next-year.aspx?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+khn+%28All+Kaiser+Health+News%29

      Why do you think premiums are falling for MA? I have no idea how the CMS administrators can attribute the premium drop to “to the agency’s strong negotiations with plans” given that the plan payments are based on the bid system.

      • Could it be that MA prices are declining because of the underwriting cycle’s waxing and waning prices? Looking at the product over the long term would be more informative.

        • Interesting explanation (recession, risk pool, and pharma). Not to beat a point to death, but the impact of the underwriting cycle would mean that the plans were budgeting a lower margin for a given year. So all things recession, risk pool, pharma and lower margin expectations could be influencing premium.

          Thanks again for the reference.

    • What’s the evidence that the cost containment within Medicare is the result of increased operational efficiency at the clinical level vs cost shifting onto the private sector? Medicare just pays less to have its patients treated by the same doctors, hospitals, etc that private payers use, no?

      Unless Medicare operated as an entirely separate entity that financed every penny of capacity that it’s patients use, this could easily be nothing more than cost-shifting masquerading as efficiency. If that’s the case, all of these cost containment gains are voided when you eliminate the entity that you’ve been shifting the costs onto – e.g. single payer.

      Forcing grocery stores to sell food to old people at a lower price than everyone has to pay is something rather different than achieving efficiencies that actually lower the price of food for everyone.

    • Done. Lots of debate about the significance, existence, and magnitude about “cost-shifting,” but I didn’t see evidence that the cost containment achieved by public payors has been driven *primarily* by something other than paying doctors and hospitals less money for the same care.

      The data from the VA are probably the strongest, and it would be interesting to hear from folks who have worked in both VA and non-VA settings to see if they actually believe that the efficiencies were transferrable and scalable enough to serve as the basis for a single payer system.

      If there were evidence that the overwhelming majority of Medicare/Medicaid patients were treated in fundamentally different ways than private patients – variations in the diagnostic regimes, treatment regimes, provider inputs/intensivity, etc that generated real cost reductions then the empirical cost data that you cite would certainly support your claims. 1) Is this evidence out there, and 2) How would private insurers simply adopting Medicare/Medicaid payment schedules fundamentally differ from a single-payer system predicated on deploying Medicare/Medicaid cost control achievements system wide?

      • I think you’re confusing cost of care with spending. Sometimes “cost” means one or the other (cost to provider vs. cost to taxpayer). Even if Medicare/Medicaid doesn’t reduce cost (to the provider), it might spend less (reducing taxpayer cost) due to lower payment and different (lower) generosity of benefits.

        However, even this is not correct thinking. It isn’t the level of spending that is debated, but the rate of change. Of course Medicare spends more per beneficiary than private payers. The elderly and disabled are sicker. But the massive pile of information I provided points to the idea that public payment systems lead to lower rates of increase in spending.

        There is one piece of evidence that comes closest to suggesting private payers can have lower costs (to the payers), but it comes with a several big asterisks. We discuss it on our forthcoming podcast. The background info is here: http://theincidentaleconomist.com/wordpress/competitive-bidding-faq/ . Be careful not to over interpret it. Results vary by market and interpretation depends on what you believe about favorable selection and risk adjustment.

    • The correct or right argument to make is that by abolishing public health insurance, the federal government will reduce taxation in a progressive system, eliminate the need for private employers to provide such health coverage in a weak global labor market, and require individual citizens to budget – or not – for their own medical requirements.

      The Conservative reaction to the statements of Ron Paul during the most recent Conservative presidential debate epitomize this position in full.

    • “The right argument is that private options offer more choice, greater flexibility, and rapid innovation.”

      Why is this the right argument? There’s zero reason a public option, especially if it’s just public insurance and not care, can’t offer exactly the same things… it could even charge for them.

    • Thanks for the reply – I’ll listen to your podcast with interest when I get a moment.

      I agree that costs and spending are different*, and that it’s important to make a distinction between the two. My argument was that all we are seeing in your data is constraints on spending via price controls, rather than fundamental improvements in the way care is delivered. Same care, same delivery – the people delivering the care just get paid less.

      If you impose the same, lower payment schedule globally (single payer, single rate) without a corresponding increase in efficiency, you lose capacity in a manner that varies with the magnitude of the cuts in total spending. Per capita capacity loss looks much worse under the above conditions if you assume demand growth.

      If there’s data that shows that the way care is delivered when it’s payed for by public payers actually uses X% fewer resources, then you’ll have a convincing basis to support the claim that a public payment system could spend X% less without the quantity or quality of care going down – but not before then.

      Have a nice weekend.

      *For the reason that prices are different than costs. It’s easy to control how much someone can legally sell a given output for by regulatory fiat. It’s much more difficult to control how much it costs them to produce the output, since that depends on a dynamic matrix of interconnected variables that is beyond anyone’s capacity to control precisely enough to coordinate supply and demand. The inevitable result is a producer price windfall followed by a glut, or a price below the cost of production followed by a scarcity. IMO we’re seeing both as a consequence of the Dr. Hsioa’s system of price controls (Windfall = Ortho, scarcity = Primary Care).

      • You are asking me to substantiate a claim I did not make. The data show per person spending on health care increases at a lower rate in public than in private systems. That is as far as I went in the post.

        If you follow all the links they will take you to things like: If Medicare provided its private plan-based drug benefit in a manner equivalent to the VA’s the taxpayer savings would be higher than the loss in welfare due to reduced choice. In other words, according to standard welfare economics, a Pareto improving step could be taken, one that is toward a public and away from a private system we have.

        Elsewhere on this blog, though perhaps not linked to anything to this post, you will learn that the vast majority of additional payments Medicare Advantage plans receive over traditional Medicare (per beneficiary) is valued at a rate of 14 cents on the dollar by the beneficiaries themselves. This too is based on neoclassical welfare analysis. We’d be better off by just paying beneficiaries 15 cents in cash, saving the 85 cents for other, more efficient uses.

        These results may not be convincing to you. I’ll leave you with this: rather than presume the government is not actually providing more efficient care and suggesting I substantiate that it is, you could equally presume that care is quite efficiently provided and I could ask you to show me that it is not. Either way, it is outside the scope of this post.

    • Thanks for the thoughtful reply – definitely above the call of duty and I appreciate the pointers to the literature, which I do intend to pursue.

      On the topic of things that are beyond the scope of the post, whenever the topic of neoclassical equilibrium analysis comes up, I can’t help but link to the following:

      “What is the problem we wish to solve when we try to construct a rational economic order? On certain familiar assumptions the answer is simple enough. If we possess all the relevant information, if we can start out from a given system of preferences, and if we command complete knowledge of available means, the problem which remains is purely one of logic. That is, the answer to the question of what is the best use of the available means is implicit in our assumptions. The conditions which the solution of this optimum problem must satisfy have been fully worked out and can be stated best in mathematical form: put at their briefest, they are that the marginal rates of substitution between any two commodities or factors must be the same in all their different uses.

      This, however, is emphatically not the economic problem which society faces. And the economic calculus which we have developed to solve this logical problem, though an important step toward the solution of the economic problem of society, does not yet provide an answer to it. The reason for this is that the “data” from which the economic calculus starts are never for the whole society “given” to a single mind which could work out the implications and can never be so given….”

      I’m sure that you can guess the author.


    • I think you are confusing the issue with an either-or proposition. Why not do as France does and have every citizen be covered by the public system and permit citizens to purchase private, supplementary insurance. If choice is so important, this seems like a reasonable compromise. If you’re rich and you want better coverage, you’re free to buy it. But we wouldn’t have people dying because they can’t afford insurance (45,000 per year: , and children wouldn’t be included in this tally either (http://www.washingtonpost.com/wp-dyn/content/article/2007/02/27/AR2007022702116.html). And we could manage chronic conditions, like hypertension or heart disease, that wind up being costlier the longer they go untreated. Skip to the France section here: http://prospect.org/cs/articles?article=the_health_of_nations

      • Who is “you” in your first sentence? If I presume it is me, then your comment astounds me. I’ve not made any such proposition, nor made any normative statements in the post.

        To be sure, one can conjure up all manner of compromise or conclude any number of things. But I assert it cannot be done without adding more to the mix than just the evidence. To reach such conclusions you have to weigh the evidence somehow. What value do you give to lower spending? What value do you assign to more choice, flexibility, innovation? How much do your own valuations matter relative to others?

        I have not answered those questions in this post.

    • Chris:

      It’s been a while since I looked into the methodology behind the claim that 45,000 people that wouldn’t otherwise die, do, because they lack insurance.

      Didn’t that study presume that the only difference between the insured and uninsured was the lack of insurance? In reality having insurance is highly correlated with education, income, marital status, wealth, education, etc, etc, etc. Ergo in the aggregate people that lack insurance are different than those who do have insurance in lots of ways that are highly correlated with poorer health. IMO it’s much more meaningful to look at the associations between poverty and poor health, which are much more robust than the association between insurance status and health.

      Just to take a quick example, I’m confident that I could run a regression comparing uninsured people who make over $75K a year and people covered by Medicaid to justify a bogus argument that enrolling in Medicaid is far more hazardous to your health than going without insurance.

      • FYI, these issues have been explored here (the effects of insurance on health outcomes and Medicaid in particular). If they’re not in the FAQ I will add them. In the meantime, search for a post called “Our unpublished response to McArdle” (or similar) and for posts under the “Medicaid-IV” tag.