• Priceless: Chapters 8-9

    The 8th chapter of John Goodman’s book Priceless covers a lot of important concepts, most of which I think John explains well and gets right. He covers some of the history of the US health system, a lot more of which you can learn by reading Paul Starr’s The Social Transformation of American Medicine. I’ve posted plenty on it as well.

    I have only a few questions and concerns about the rest of the chapter. First, John characterizes managed competition as a market that would otherwise be free were it not constrained by “artificial rules.” Don’t fall for this. All markets have rules and all those rules are “artificial.” I’m not saying we need to like managed competition’s rules. I’m just saying calling them “artificial,” isn’t saying anything.

    Anyway, the main rule John doesn’t like is community rating. He explains the problems with community rating, leading to a seeming take-down of risk adjustment. One problem with risk adjustment is that no methods predict costs all that well. Of course, some of health care, probably most of it, is unpredictable, the very part John thinks we should insure against.

    John’s proposed solution to risk adjustment is that, upon switching plans, an individual’s “original health plan would pay the extra premium being charged by the new health plan, reflecting the deterioration in health condition.” There are two things about this I do not understand. First, how would this extra premium be calculated in a way that is different from risk adjustment payments? If we knew a better way, we’d have better risk adjustment now.*

    Second, this idea seems no different than risk adjustment by another name. Think about it from the new plan’s point of view. Would the plan manager act any differently if the payment is called a “change of health status offset” and paid by the original insurer or a “risk adjustment payment” and paid via a market administrator of some sort (funded, for example, by assessments on low-risk bearing plans)? A dollar is a dollar. The same limitations of risk adjustment apply, don’t they?*

    Dispersed throughout Chapter 9 is an excellent summary of HSAsHRAs,  and FSAs. This can be a confusing area and John explained it well. He breaks it up with a bit of history on their development. I particularly like this quote:

    Good ideas should stand or fall on their own merits. They should not be adopted or disowned, depending on the party of their advocates and opponents.

    I could not agree more. But, I have to add, good ideas do not require exaggerated or selective interpretation of the evidence. The alternatives need not be seen as less than worthless for a good idea to shine. Moreover, even good ideas have limitations, and especially so when they must pass through the political process to achieve implementation.

    What did you think of Table 9.2 and related discussion? Were you impressed with how much could be saved in premium by ratcheting up a deductible? I was. It reminded me of my experience a year or two ago when I considered switching to a high deductible plan. Fortunately, I had access to a good bit of data about the plans available to me and, using it, I learned that I would not likely save anything by switching. I wondered why.

    The answer is that the degree to which premium trades off against deductible has as much or more to do with the differential risk of enrollees in the plans being compared as it does with the amount by which a higher deductible causes more prudent use of health care. That is, if healthier people are more likely to enroll in higher deductible plans (either because of the deductibles or because of other features and benefits of the plans), that also causes a lower premium. Or, if sicker people do, it could cause higher premiums.

    In a prior chapter (or two), John impressed upon us how well plans are able to risk select. In light of this, don’t take the evidence in Table 9.2 at face value. It isn’t risk adjusted. It is highly likely that every dollar saved in premium is not offset by a dollar in deductible in part because healthier people choose the higher deductible plans. The answer to my puzzle of a couple of years ago (per the prior paragraph) is that the risk selection of the plans I was comparing must have offset whatever effect a higher deductible had on spending and premiums. Another possibility is that there was little risk difference across the plans and high deductibles didn’t lead to much resource use reduction. Therefore, for the plans I was comparing, every dollar I could save in premium, I’d expect to pay right back in deductible.

    Also in the chapter, John describes some findings from the RAND health insurance experiment. I’ve written about those too. One thing I’ve written that John left out is that

    The poorest and sickest 6 percent of the sample had better outcomes (including mortality) under free care for certain conditions or types of care (hypertension, vision and dental care) and serious symptoms were less prevalent for poorer people on the free plan (chest pain when exercising, bleeding, loss of consciousness, shortness of breath, excessive weight loss).

    So, there may be heterogeneous effects of higher deductibles, affecting lower income and sicker individuals differently than others. Isn’t that something we should care about or at least mention? I think so because John claims that HSA’s don’t favor the healthy and wealthy. This is an empirical matter and, methodologically, the RAND HIE was a superb, though by no means perfect, test.

    This was nice to see acknowledged:

    The problem with a deductible is that although the incentives are ideal as long as spending is below it, they become completely perverse once you reach it. That is, below the deductible, every dollar you spend is yours, but above the deductible, every dollar you spend belongs to someone else, except for your co-payments.

    It was in the opening paragraph of my favorite section of the book so far, “A Third Simple Way of Controlling Healthcare Spending.” I found every word in this section reasonable, from how to split care that a patient could be responsible for and fund with an HSA vs. care that should be third-party billable to the use of value-based purchasing (reference pricing). About the latter, see also the latest paper by James Robinson. It’s well worth a read.

    They key point that moved my thinking about John’s vision is that he’s not really talking about high deductible health plans. He’s talking about treating two categories of health care differently. There’s a category for which it is reasonable to think standard market theory is applicable (a lot of low-cost, elective stuff) and a category for which it is not (emergency surgeries, for example). For the former, paying more or all of the cost at the point of purchase could motivate valuable innovation. For the latter, catastrophic plans are his proposed solution. Completely reasonable.

    My next post on Chapters 10 will appear on Wednesday. All posts on Priceless are found under the Priceless tag. The book has 18 chapters. We’re half-way done!

    * By email, John says he doesn’t have any issues with what I wrote here. What’s most important to him is that private entities work out the risk-sharing arrangements among themselves, rather than a government-imposed solution.


    • Austin bsically hated the last four chapters. Yet they are my favorite chapters. The reason? Chapters 5,6 and 7 explain why we have the problems of cost, quality and access — to a degree not found in any other industry, other than education — with a single thematic answer: we have suppressed the price system.

      It’s possible readers don’t like that explanation. But I have not seen any other.

      Chapter 4 explained why all the people you may want to trust — doctor, employer, insurance company and government — are unreliable agents, because they face perverse incentives created by the lack of a real price system.

      You may not like those conclusions either. But I haven’t seen readers rebut them and tell us why these entities are reliable.

      • If I understand your theme, it’s that we’ve suppressed the price mechanism by our reimbursement system – consumers don’t make the same sorts of choices on where to spend their dollars, due to various public and private insurance mechanisms, as they do when deciding to buy TVs and toasters.

        If so, how much consideration have you given to the massive distortions on the supply side, such as patent protections for drugs and devices, licensing and immigration restrictions on healthcare providers, etc.? How about the issues raised by Arrow’s classic Uncertainty and the Welfare Economics of Medical Care?

        • In this series I keep asking about what will protect patients from supplier abuses and distortions. So far, my concerns get dismissed, not addressed. That’s very disappointing. It’s the difference between engaging and domineering. I’m only interested in the former.

          • Austin, I how does my comment relate to your concern about supplier abuses and distortions?

            My comments relate to the failure of the market mechanism to work due to one class of government distortion – restricting supply. Expanding supply would not prevent mechanisms to promote quality. You could easily have a functioning FDA without patent protection, for example. State regulation of doctor quality seems miserable from what I can tell, with almost nothing ever done. Relaxing immigration restrictions and building more medical schools does not mean action can’t be taken against the incompetent or worse.

    • While I agree with most of this, I’d like to discuss two points:

      >>There’s a category for which it is reasonable to think standard market theory is applicable (a lot of low-cost, elective stuff)>>

      This stuff is closer to standard market theory, but not close enough.

      Standard market theory, as I understand it, requires competition (lack of restraints on new entrants) and information (ability of customers to make informed decisions). It’s not clear to me that these apply to “a lot of low-cost elective stuff.”

      Competition is limited by licensing restrictions. There are non-market constraints on who can practice medicine. This limits the normal mechanism by which high pricing or profits are competed away. Similarly, patent protection limits competition for drugs, medical devices, etc.

      It is difficult to get pricing information, let alone information sufficient to judge quality.

      >>paying more or all of the cost at the point of purchase could motivate valuable innovation>>

      In addition to the foregoing issues, paying more could motivate people to not seek needed care, leading to to higher costs later. Also, doesn’t routine care represent only a small portion of total healthcare expenditures, reinforcing the “penny wise, pound foolish” nature of this proposal. Isn’t there empirical work on these points?

    • P-HRA accounts solve many of the problems John presents in “Priceless” but he makes no mention of them. Under current tax law, employers can deposit HRA funds into employee-owned bank accounts to be used only for qualified healthcare expenses creating a Portable HRA.

      Some of the benefits of a P-HRA are:

      “Freeing the Employee”
      Today’s workforce is mobile. John writes: “One of the biggest problems in the US healthcare system is that health insurance is not portable . . . pre-existing conditions arise because of a transition from employer-provided insurance.” With a P-HRA, employees purchase insurance that best suits their individual needs and is completely portable.

      “Freeing the Nontraditional Workplace” (pg 167)
      Today’s families usually have multiple wage earners. Why should only one employer carry the burden for a family? P-HRA accounts can accept contributions from multiple or part-time employers to a single account. Even migrant workers can now have healthcare insurance as they move between growers during a season.

      “Freeing the Employer” (pg 166)
      John writes: “Most employers . . . would prefer not to be in the health insurance business, however. In a world of portable insurance, they would not have to be.” Employers simply make a defined contribution to the employee’s P-HRA bank account. These contributions may be based on an hourly labor rate or a set amount specific to that employee. The employer is no longer “in the health insurance business.”

      “Advantages of Self-Insurance” (pg 148)
      John writes: “To the degree that people regard the money in HSA and HRA accounts as their own, they will be more careful, prudent shoppers.” Again, P-HRA accounts belong to the employee where they may elect to purchase low-cost, high-deductible insurance and pay all their health related expenses with their VISA smart-card. A perfect solution to those, like John, who believe in CDHC!

      I can only speculate why John wouldn’t explain P-HRAs in “Priceless” when they solve many of the problems he lists in chapter 9. A good example of a functioning P-HRA can be found at LyfeBank.com.

      • Leon, I thought HSAs were portable too. How is a P-HRA more portable than an HSA?

        • Employer contributions to an HRA are before tax while contributions to an HSA are after tax. They are two different animals. HSA’s are more of an investment tool to shelter earnings like IRA.

        • error: That should be ” more of an investment tool to shelter earnings like a ROTH IRA”.

    • When considering the abuses seen in health care one of the major problems I worry about is when 2 sides team up against the third (government, provider and patient). We can never rid ourselves 100% from abuse so one will always be able to point to abuse no matter what system is in use. Foremost in many minds is patient abuse by providers and this is very scary, but if government exists as a regulator and not a participant it is able to act in a more neutral fashion affording protection to the patient.  Once government gives up its status as a neutral regulator it becomes a participant. As a participant it is unable to act in a neutral fashion and suddenly the patient can find government and the provider teaming up against the patient. That is when what previously was called an abuse suddenly becomes normal behavior.

    • [John’s proposed solution to risk adjustment is that, upon switching plans, an individual’s “original health plan would pay the extra premium being charged by the new health plan, reflecting the deterioration in health condition.”]

      There is a practical matter here that I don’t realyl understand. In a free market, firms should be free to succeed or fail, right? One of the ways a firm can fail is to go bankrupt. Now, let’s put aside (but not forget) games like deliberating looting a company and going bankrupt. Even in the face of honest to goodness “my business model failed unexpectedly” bankruptcy, then who pays the extra premium?

      I think that this piece is utterly and completely critical. If it is the government, then we are getting back to the need to have governments regulate closely (to avoid fraud). If it is re-insurers, the ability to get such contracts will be a serious barrier to entry into the market (just think of how huge these obligations could be). If it is entirely due to the reputation of the firm, then how do small firms with innovative business models compete with large, well-established firms?

      What if the previous company contests the size of the premium increase? What if the new company deliberately overcharges to bleed the “big company” and offers incentives to switchers to make them want to enroll.

      Is the legal system the correct solution? And what happens to bills during the dispute? How is the made seamless?

      And if the answer is tight government regulation, how does this beat the current system?

      • Joseph,

        I don’t think anyone objects to regulating insurance companies for solvency purposes. This is absolutely essential and the state DOIs do a pretty good job of it, to the point of requiring minimum reserve levels and closely regulating how those reserves are invested. They even have state guarantee funds that protect insured when a company becomes insolvent (not “bankrupt” — insurers do not have bankruptcy laws available to them).

        Regulation becomes “over-regulation” when they define required benefits, allowable premiums, underwriting rules, and so on. One of the rationales for imposing such regulations on health insurance (and not on other lines) is because it is not the insured choosing the product, but the employer, and employers are in fact poor agents for their employees.

    • Goodman’s proposed solution to risk adjustment sounds like it comes from John Cochrane’s paper “Time-Consistent Health Insurance” (Journal of Political Economy, 1995). The basic idea is that health insurance is time-inconsistent because it fails to account for how changes in health status in the current period affect costs in future periods. Cochrane’s solution is to include a settling-up provision at the end of the contract in which a lump sum changes hands equal to the present value of the expected change in future premiums.

      In effect, Cochrane takes a conventional health insurance contract and re-packages it as a futures contract. Cochrane makes it very clear that his proposal is based on the dynamic trading concept that serves as the basis for models of financial markets. In this framework, the periodic settling-up process (“marking to market”) solves the time-inconsistency problem.

      Cochrane’s target is the pooling of risk mechanism behind conventional health insurance. He points out that, in this system, the only way to bind healthy individuals to long-term contracts is an individual mandate. However, as he points out, this precludes competition in the market for risk. On the other hand, the time-consistent contracts he proposes do not have this problem. This seems to me the basic idea behind Goodman’s proposal.

      At the end of the paper, Cochrane has a long section in which he addresses possible objections. In particular, he addresses the question of why we do not already observe such contracts in the market. His analysis, like Goodman’s, emphasizes path-dependence. Basically, we are stuck on local maximum because pooling of risk emerged in the market first.

      So, I suspect that Austin’s question, about what mechanism Goodman would propose to replace risk adjustment, misses the core idea which is that a dynamic “no arbitrage” equilibrium will replace a “pooling of risk” equilibrium in a competitive market.

      While I like Cochrane’s paper a lot, I am not convinced that the idea would work in practice. First, it may be the case that we don’t observe these contracts in the U.S. due to path-dependence but that does not explain why these contracts have failed to emerged in markets that have followed different paths. Second, it could be that demand-side imperfections, like hyperbolic discounting, make time-consistent contracts unattractive, in which case, the welfare-maximizing answer would be an individual mandate. Third, in a financial market, the mechanism by which you get to a “no arbitrage” equilibrium involves the smart traders taking advantage of the dumb traders. I’m not sure that’s what we want.

      • Rob’s discussion of Cochrane indicates some of the dangers of unrestricted markets in primary care and chronic care insurance. Cochrane apparently suggests that inefficiencies of insurance, including the need for long-term contracts, can be removed by making insurance contracts more like commodity futures. Maybe so. I’m not able to assess this suggestion.

        But in that proposed market, no one except financial market players would understand who is covered for what. It would be a lot more like our financial markets. Who would regulate this market? How would anyone know whether they’d end up with “insurance” that vanishes in lost financial gambles? What incentives would such markets create for healthcare? As Rob concludes, “smart traders would take advantage of dumb traders.” I’m pretty sure I’m not a smart trader, but that’s just speaking for myself.

        Beyond the information problem, there is a moral problem. Health insurance has to be subsidized for a large portion of the population who otherwise cannot afford it. If these subsidies are in the form of money or vouchers, rather than a guarantee of care, then two possible outcomes can result, in the presence of the information problem.

        One is that people who make bad choices would be judged as responsible for their own problems. I believe that this is how John’s reasoning would be used politically, although that is not his intention.

        The other possible outcome is that doctors, hospitals, other providers, and regulators would rise to the occasion by providing higher quality and better choices, despite consumers’ inability to make that assessment. That is John’s optimistic hope, as I understand it.

    • @ Leon

      When i wrote Priceless, I was unaware that employers could put cash in an account for employees, using the HRA vehicle. I have subsequently learned that this is legal and doable.

      However, I do not like your comparison of HSAs and HRAs. Deposits to both acounts are made with pre-tax dollars.

      @ foosion and Austin

      I thought I made it clear (maybe in later chapters) that I favor eliminating most supply side restrictions. These include CON laws and licensing laws that prevent nurses, pharmacists and other non-doctor personnel from doing things they are perfectly capable of doing.

      Also in a later chapter I argue for replacing the malpractice system with a radical form of no fault liabililty.

      @ Rob

      i am relying on John Cochrane’s model.

      @ Joseph

      When someone switches health plans, the two plans must agree to the terms of the transfer. If they don’t agree, the insurance commissioner can dictate the result. Just as 90% of law suits settle without a jury verdict, i suspect that the commissoner will be rarely needed.

      • @John:

        Thanks for the quick response. I worry that the transition process would have a lot of friciton, as both sides have enormous incentives to misprice risk on older patients.

        I am also interested in the way one handles bankruptcy or poor asset management on the part of the insurance company? We already have this problem in pension plans (where shortfalls happen due to poor management). An 18 year is going to be in a health plan for 47 years (longer if we removed medicare). That is a long time to presume that a firm will survive. How do we handle this?

        Or is it purely PAYGO? If so, what happens if a lot of people leave a firm and drain it of assets?

      • Thanks for correction. Yes, HSA contributions are tax deductible.

        Let’s consider the key benefits of P-HRA vs. HSA:

        1. You can pay insurance premiums with HRA funds (and, the policy need not be limited to a specific HDHP structure). So, an HSA is not a very viable account type to offer contributions to part-time employees, for example.
        2. While HSA contributions are tax deductible, they are subject to annual limits, unlike a P-HRA.
        3. Employers prefer the P-HRA account structure, because HRA contributions remain as employer plan assets until spent. This is different from, an HSA, in which an employee vests immediately in the employer contributions and has a lot of discretion in how to use them. In contrast, P-HRA funds are under the control of the employee ONLY to the extent that they choose to use them for qualified medical expenses for the employee and their dependents.
        4. P-HRA funds are fully protected against any investment risk of loss.

        HSAs do have their places, but they are less flexible and useful in the employee benefit context than P-HRAs. Properly structured, an HSA and P-HRA can coexist, with employer contributions into the P-HRA and the employee contributions into the HSA

        • Leon,

          Two things. 1. Most DOIs are not allowing HRA funds to be used to buy individual health insurance. This is based on a late-Clinton era HCFA regulation on HIPAA that tried to forestall dodging small group laws on guaranteed issue by employers giving money to employees to buy their own coverage. 2. I understand that employers get to deduct HRA contributions only once the money is spent on a health service, not when it is deposited in an account. Do you have information that changes that?

          • Greg,

            1. LyfeBank has met with multiple state insurance commissioners and has received confirmation that their approach is sufficiently different from the traditional HRA model that the rules proscribing use of HRAs to purchase individual policies do not apply to Lyfebank.

            2. In a traditional HRA, their is nothing to deduct until a claim for reimbursement is made because traditional HRAs operate on the “promise to pay” model. However, employers can deduct the contributions to employee LyfeBank accounts at the time of funding because the employer is making a real deposit of funds that can never revert to the employer. Funds not used by the employee roll over from year-to-year and are portable from job-to-job.

    • What I find disappointing in these chapters, being written by an economist, are the missing numbers. I agree that there is probably some ability to affect prices using a deductible or HSA type plan. However, when the insurance kicks in, there is no incentive for the patient to worry about costs. What I would like to see are estimates, using the deductible limits of choice, or HSA limits of choice, of how much of our spending would be affected. This could be augmented out of personal savings, but we also need the numbers to see how feasible that is for different income groups. I think we also need to remember the non-fniancial issues. Time and geography are more limiting than is shown in pure financial models.

      Rob- Has Cochrane’s model worked in other areas outside of health care? Has it been adopted by any other health care system in the world?


    • Thank you Austin for puncturing the libertarian balloon about high deductible plans saving a lot of money. I have sold health insurance, and what you say about the risk pool being dominant is absolutely right on.

      If you put 100 GE retirees into a health insurance policy, even a high deductible policy will cost $700 a month. If the deductible is $6,000, the poor insured will pay out $14,400 a year before getting one cent of coverage. By international standards, that is plain repulsive.

      Whereas 100 young male Microsoft engineers can save hundreds of dollars a month with a high deductible plan.

      So much of our health insurance policy goes back to the 1950’s, when Blue Cross offered community rating — but commercial insurers found that they could peel off the healthier groups and charge them a lower premium.

      These last 60 years of insurance competition may be seen to have accomplished NOTHING for America.

      Bob Hertz, The Health Care Crusade

    • I don’t understand how high deductibles can’t save money. One of the big claims against insurance companies is that their administrative costs are so high. If that is true then at least a large portion of those administrative costs would be saved. Additionally one would think there would be savings based upon moral hazard. Is someone able to explain how those two items don’t save money?

    • I usually ignore the passage Austin quoted from the original RAND experiment. The reason? It is most likely the result of data mining. See Robin Hanson on this here:


      A subsequent RAND study found no harmful effects of consumer directed health plans on vulnerable populations. Further, in the latest RAND study the savings are up to 30%. This is the result of behavioral change, not selection.


    • How do we square the RAND findings of 30% savings with consumer directed health care with Aaron Carroll’s observation that 5% of the patients spend half the money in one year?


      Here is what Mark Pauly posted at my blog:

      Here is my take on why the Carroll/NIHCM analysis is probably not right and the Rand research (both old and new) on the cost containment potential of moderate amounts of upfront cost sharing is probably right. Rand showed that the main effect of cost sharing is to reduce the rate at which people initiate episodes of care. Some of these episodes end up being very expensive, thus plunging someone into the top 5% of all spending. But, not only is that not known at the beginning whether an episode will turn out to be high or low cost, it appears that cost sharing discourages treatment that ends up being high cost (think of the endless spending to track down an allergy). How else explain that it cut inpatient use by 30%? So, while high cost sharing does not affect most people who are high cost (because they blow through the deductible), it need not prevent the start of very many high cost episodes to save a lot of money.


      • I missed this one earlier. Tracking down allergies is not our cost driver. What I think Mark is suggesting (advocating?) is that with large deductibles, people are ignoring chest pain because they know they might need a CABG or stent, and they cant afford it. Ignoring night sweats and 30 lb weight loss because they know it is some kind of cancer, and they cannot afford it. Is this how we want to control costs? I concede it might work, but what are the social costs?


    • @Steve:

      Cochrane’s proposal was directed specifically at the time-consistency problem with health insurance. So, I doubt that it has been adopted outside health care. Also, I’m not aware of any health system that has experimented with it. However, in a 2009 article he wrote for the Cato Institute (“Health Status Insurance: How Markets Can Provide Health Security”), Cochrane points to a new product offered by UnitedHealth Group that includes an option giving the policy holder the right to purchase future health insurance at a premium based on the holder’s current health status. I don’t know how this product has fared in the marketplace.

      The main problem I have with Cochrane’s proposal is how the market for risk is supposed to work. In a financial market, information affects price through the trading process. All the traders in the market have different sets of information about, for example, the future prospects for wheat and, through the buying and selling process, this information gets aggregated into the price.

      Obviously, the health insurance market doesn’t work this way. Instead, health plans aggregate individual risks into risk pools and price the premium based on the expected cost of the pool. Cochrane’s proposal unbundles the risk pool, shifts idiosyncratic risk back to the individual, and includes a risk hedge to keep the policy’s economic value constant.

      Essentially, this structure resembles the Black-Scholes option pricing technology. I see the basic idea but not how health plans manage the risk hedge. The obvious answer would seem to be that plans package all these individual policies into policy pools, slice and dice the risk into derivatives backed by cash flows from the policy pool, and then sell these securities to investors. Of course, we all know how well Countywide made out with this strategy.

      I’m not saying that Cochrane’s proposal won’t work — in fact, I think it’s very interesting — but the idea that a good dose of unrestricted free market competition is all we need leaves a lot of loose ends.. Austin notes that John has some good proposals in the latter chapters, so I’m looking forward to those ideas.

    • Austin,

      Where you agree with John is pretty much where I disagree.

      I don’t have any problem with varying deductibles based on market demand, but this distinction between essential and non-essential services leaves me cold. I know that is how they did it in South Africa, but istm that very often even essential services can be shopped. It may be “essential” that a diabetic take insulin, for instance, but there are many ways to buy insulin. Some are less expensive than others. I can’t think of many areas of life where a price system doesn’t make access better for all — even for essential services.

      And risk adjustment. Insurers have been doing a reasonably good job of risk adjustment for centuries. It is known as underwriting and it works well (not perfectly, but well). Risk adjustment only becomes problematic when you try to do it without underwriting.

      • Wait until the end of the book. What I was able to infer through Chapter 9 turns out not to be the whole story. In the end, I think you’ll find your comment is moot. (Is anyone noticing that I’m defending John here? Indeed, I am!)

    • Greg, you are always one of the best commentators.

      I appreciate your inference that even the “essential services”can be shopped.

      The border seems to stop with heart and cancer treatments. There are no cheap alternatives — I think — when your heart stops or your platelets signal imminent death.

      If we regulated the prices of drugs which have no substitutes (maybe 70 drugs in all) , that would be progress.

      Bob Hertz, The Health Care Crusade