Response to Holt, Havighurst, and Cohn


Matthew Holt raises a number of issues. His premises tend to be true. However, his conclusions do not follow.

1. Our system favors providers, not consumers

I agree that our system favors providers, but the question is what government can do to help. Almost every government intervention that I can think of, from insurance mandates to licensing restrictions to tax policy, has the effect of either restricting supply or subsidizing demand, all to the benefit of providers. This is not surprising, given what Holt himself calls “the political strength of the health care system actors combined with the disaggregated weakness of the consumers and those paying the bill.” It’s sort of Public Choice 101.

So I agree with Holt’s diagnosis, that health care providers have too much political power. But his prescription—to put more of the health care system under the purview of government—seems perverse. It is like telling someone with adult-onset diabetes to go eat more doughnuts.

2. Consumers cannot make good health care decisions on their own

Holt writes, “It seems to me that most people faced with any kind of significant medical decision will either be too ill , be too susceptible to the influence of providers, or unable to get good enough data about their individual situation, to be able to make a rational choice.”

Again, I agree. However, it does not follow from this that insulating consumers from the cost of health care will lead to better decisions. Instead, it will simply cause them to err in the direction of choosing the most expensive medical services—the overuse of premium medicine.

3. Insurance and Market Failure

Holt writes,

[Kling] tentatively admits “that if health insurance were relatively unregulated and unsubsidized, then many people would opt to do without health insurance.” Anybody observing American health insurance today knows that this is true. Of course, that means that in order to cover everybody with health insurance, the market for it does need to be heavily regulated and subsidized.

Here, I believe that I was misunderstood. Let me spell out what I mean by using an example. Suppose that a healthy 30-year-old has a 5 percent chance over the next ten years of contracting an illness that would require more than $25,000 in treatment. One insurance policy that would be appropriate in this case would be a policy that costs $1000 up front, is in force for ten years, and a policy that pays $20,000 if the expensive illness strikes, but otherwise pays nothing.

Consider three reasons why the thirty-year-old might not have such a policy.

  1. Government regulations distort the insurance market.
  2. The insurance market fails on the supply side, because of too much risk sorting and not enough risk pooling.
  3. Consumers are offered fair policies, but some choose to turn them down. Some of them

    just would prefer to gamble that they will not need help paying for medical expenses.

What I was referring to is the third possibility, that even if the market works, some consumers will have so much risk tolerance that they will turn down catastrophic insurance.

However, let us suppose that Holt’s premise is correct, and that catastrophic insurance will not emerge in the private sector because of risk sorting. It still does not follow that government should provide insulation. Instead, what the government ought to step in with is a transfer mechanism that mimics catastrophic insurance. It should pay large benefits to a small percentage of unlucky individuals who have serious health problems. Yet when we observe single-payer health care in other countries, it makes small payments to everyone. Single-payer means insulation, not insurance.

4. Single-payer in America

In Crisis of Abundance, I point out that single-payer can take three forms. One form I call “doctor-friendly,” in which government picks up the check, but otherwise plays no role in medical decision-making. Medicare mostly works this way, with the predictable result of out-of-control spending and fiscal unsustainability. Another form I call “doctor-hostile,” in which government forces providers to offer the same services at lower prices. The final form I call “doctor-limiting,” in which the government acts like a managed-care company, controlling cost by limiting access to medical services.

My sense is that the version of single-payer that most closely resembles Holt’s preference is doctor-limiting, with perhaps a bit of doctor-hostile thrown in. In any event, doctor-hostile single-payer health care probably would look a lot like doctor-limiting health care, because many doctors would refuse to participate in a government program that severely limits fees.

Holt sees the public turning to government because of concerns about family financial security. I do not pretend to have my finger on the consumer’s pulse. However, I would suggest that the bad reception that America gave to managed care in the 1990’s seems to indicate some unwillingness to sacrifice the freedom to make extravagant use of premium medicine in exchange for financial security with regard to health care. In that regard, I could throw Holt’s words back at him and assert that in today’s America, single-payer health care “ignores the political and cultural realities of the health care system, and probably wouldn’t even work in theory.”


Clark C. Havighurst makes an eloquent case for getting rid of the tax subsidy for employer-provided health insurance. I endorse his view.

Havighurst writes,

something approaching their goal of universal health coverage could be achieved by ending the current tax subsidy and offering refundable tax credits of, say, $6000 to families that spend at least that amount in health plan premiums or contributions to a health savings account (HSA). Although citizens would be free to go uninsured, those who do so would forgo the tax credit…

I can endorse this, also, However, moving to this approach and abolishing the tax subsidy for employer-provided health insurance may be less acceptable politically than a compromise. The compromise would be to cap the deductibility of employer-provided health insurance premiums at, say, $5000 per family, in order to encourage employers to offer high-deductible plans. At the same time, individuals who buy their own health insurance could be offered Havighurst’s capped refundable tax credit. It seems to me that such tax reforms are relatively easy to implement, particularly in comparison with the proposals for expanding government-provided health care.


Jonathan Cohn correctly criticizes both my essay and my book for being evasive on the issue of dealing with the very sick (e.g., someone with diabetes) and the near-poor. I believe that there is a role for government for providing a safety net for both groups. I am confident that such a safety net can be provided without government providing full-bore insulation for everyone. I am less confident about the best approach. Perhaps “catastrophic reinsurance” (having the government pay your insurance company if your annual bill exceeds $50,000) would facilitate risk pooling that covers the chronically ill. Perhaps a tax credit that phases out gradually with income could provide support for the near-poor. But taking the entire category of consumer spending that we call health care and declaring it a government responsibility strikes me as the least efficient approach.

But evasiveness cuts both ways. Where the Left gets evasive on its proposals for health care is on how it will handle the trade-off among insulation, access, and affordability. For example, Jacob Hacker wrote recently,

If one word captures the essence of Health Care for America, it is “guaranteed.” Health Care for America would guarantee coverage; it would guarantee a generous package of benefits; it would guarantee greater choice; and it would guarantee real savings and improved quality.

To me, this sort of “guarantee” is inconsistent with having an intelligent, grown-up discussion of health care policy. Instead, it is an exercise in evasion.

When asked how they will lower costs, the Left will point to countries that do not offer the same autonomy and availability of services as Americans enjoy. But when asked how they will offer the choice that Americans are used to, the Left will say, “it will work like Medicare.” Neither Jacob Hacker nor anyone else offers a credible way to achieve lower expenditures without restricting supply.

The Left is happy to use Medicare as an example of a government-run health program that is liked by its beneficiaries. But the Left is evasive on the topic of Medicare’s financial problems. What are we going to do about the tens of trillions of dollars in unfunded liabilities in Medicare, meaning the gap between promised benefits and expected taxes in the future? What are we going to do about the extravagant use under Medicare of procedures with high costs and low benefits, as documented by the work of John Wennberg and colleagues at Dartmouth?

Finally, the Left is evasive on how it will preserve America’s leadership role in developing new diagnostic procedures and treatments. For example, in the United States, five-year survival rates for various forms of cancer tend to be higher than in other OECD countries. Suppose that we take the least generous interpretation of this data for the United States, which is to presume that we are no better at treatment. That is, all we are able to do better than other countries is diagnose cancer sooner.

Even under this least-generous interpretation, my guess is that when better treatments are developed, it will be in the United States. My guess is that some of the more effective treatments will work better when diagnosis is early rather than late, and then we will start to see differences in cancer mortality relative to other countries. Finally, even if we gain no advantage from earlier diagnosis, cutting back our diagnostic efforts to European or Canadian levels is something that would be resisted by American consumers and physicians.

Another complaint I have about the Left is that they are always citing the same handful of studies that appear to support their views, while ignoring many more studies that go the other way. See Anna Bernasek or the aforementioned Jacob Hacker for classic examples of what I call the high-investment strategy for avoiding truth. By simply reading my work with an open mind, Cohn is standing head and shoulders above many of his colleagues.

Also from this issue

Lead Essay

  • In this month’s lead essay, Cato Institute adjunct scholar Arnold Kling draws from his book, Crisis of Abundance, to argue that the health coverage most Americans enjoy is not insurance at all, but what he calls “insulation.” “The problem with insulation,” Kling argues, “is that it is not a sustainable form of health care finance… Insulation leads people to over-consume health care services. Americans make extravagant use of services that have high costs and low benefits.” Kling explains how real health insurance would work, and how it would help solve the crisis in health care, and explores how we could transistion to a system over time institutionally and culturally in order to resolve the inconsistent demand for insulation and affordable, effective care.

Response Essays

  • According to health care strategist Matthew Holt, Arnold Kling is correct that consumer insulation from the costs of “premium medicine” is partly responsible for the rising cost of health care, but Holt dissents from Kling’s solution. Holt examines what he takes to be the three main strategies for dealing with “the insulation and overuse of medical care in the U.S.”: a nationalized “single payer system; a system of “managed competition”; and “individual consumer control of spending at the point of service.” Holt argues that the latter two options face deep problems, and that a nationalized single-payer system “is the likeliest outcome in perhaps a decade or so,” even it is not politically feasible at present. “Kling has provided a decent analysis,” Holt argues, “but has proposed a solution that both ignores the political and cultural realities of the health care system, and probably wouldn’t even work in theory.”

  • Clark C. Havighurst agrees with Kling’s “diagnosis of what’s wrong with health care” in the U.S. “as far as it goes.” Havighurst goes further and digs into the reasons the U.S. health system “has evolved into an entitlement program under which everyone expects nothing less than the very best that ‘modern medicine’ has to offer.” Havighurst lays the blame at the feet of the government’s choice to subsidize the purchase of health care by “excluding the cost of employer-sponsored coverage from employees’ taxable wages and income” and lucidly details three different mechanisms by which the tax subsidy insulates workers, consumers, and voters from the costs of health care. Havighurst proposes that “something approaching [liberals’] goal of universal health coverage could be achieved by ending the current tax subsidy and offering refundable tax credits of, say, $6000 to families that spend at least that amount in health plan premiums or contributions to a health savings account.”

  • Jonathan Cohn, a senior editor at the New Republic, agrees with Kling that our current health care system doesn’t function according to the widely understood principles of individual insurance, but he doubts we’d do better at fighting rising costs and maintaining quality if citizens with “real” insurance were free to take price into account in their choice of care. “We have precious little evidence to believe that people can distinguish good care from bad care,” Cohn writes. And the notion that consumer choices will improve over time is, according to Cohn, “a lovely idea, but one that seems highly dubious.” Cohn argues that we need a broader notion of insurance – social insurance – to shield people not only against unexpected illness and harm, but against “genetic and economic bad luck.” Cohn argues that many nations do just fine in managing the cost/quality tradeoffs inherent in a state-controlled system of universal coverage, and that Americans would be happy with such a system “if only they knew how those systems really worked.”