Thanks again for your thorough response (particularly your reminder on point #2 — I should have thought of that!)
After taking some time to think more extensively about your proposal, I think my biggest problems with it are (1) that it loses its major selling points if it accounts for adverse selection in the insurance pool for the remaining 80% of coverage and (2) that it’s not the best way to address the problems of cost and overutilization that you have identified.
First, regarding the private insurance of the other 80%: In your response to my first comment about pre-existing conditions or other expensive health problems, you said that you expect people to be able to get private insurance to cover this. As you probably know, though, without additional government intervention, private insurance companies will either screen these people out or offer them only exceedingly expensive plans. This seems to me like a bad outcome, socially speaking. (Do you disagree?)
The ACA addressed this by prohibiting the denial of coverage based on pre-existing conditions and expanding the risk pool with the individual mandate and insurance subsidies. Since the individual mandate is satisfied by many different types of coverage, people are free to choose a high deductible health plan if they have low risk or low risk aversion, or more expensive plans if they have higher risk or higher risk aversion. In other words, they can choose their level of trade-off between risk and expense. Under your plan, you suggested that the government could instead solve the adverse selection problem by selectively covering some expensive and persistent conditions. But this actually gives patients as little or less choice than the ACA does, because it picks a single (minimum) level of coverage for everyone.
To provide any level of coverage that makes it non-catastrophic to treat an expensive condition (because remember, people with expensive conditions have potentially unaffordable private insurance until the government coverage is very generous), you would need premiums on the whole population to be relatively high. And actually, at this point we have almost replicated the ACA — bringing healthy, low-risk people into the risk pool to subsidize the higher-risk ones. Because most of the expensive treatments are covered by the government (and therefore most spending is not subject to deductibles), we also have not really addressed costs, except for the single-payer aspect of the theory. I’m agnostic on this, because I acknowledge that it could potentially reduce costs, but I think there are still issues with it. I don’t think you satisfactorily addressed my previous point #1 about keeping providers, particularly drug and device companies, in business, because patents have nothing to do with it. Even if a company holds a patent, if it doesn’t earn enough selling the drug within the patent lifetime it won’t have money to pay for new R&D. So I think single-payer could work, but it’s definitely complicated and as I explain below, I don’t think it should be paired with a high deductible framework.
The other point I wanted to make is that you seem to be predicating your plan on the assumption that most of the costs of the US healthcare system come from people choosing overly expensive, low-value treatments that they would not choose if they were responsible for paying. However, I really don’t think that is the case. The top 5% of spenders in the US account for almost 50% of health care spending, while the bottom 50% account for just over 3% (source: https://meps.ahrq.gov/data_files/publications/st354/stat354.shtml). Putting the bottom 50% on high deductible health plans (HDHPs) would barely move the needle on total costs, and HDHPs would also be ineffective for the top 5% because they would blow past their deductible very quickly. HDHPs might have some effect on the middle of the distribution, but I certainly don’t think they’re a panacea.
Finally, I want to point out that it is difficult for any type of health insurance scheme to do very much on its own to reduce costs and overutilization, because we still don’t know what exactly is necessary or unnecessary. Don’t you think that private insurers right now have a strong incentive to find and eliminate inefficiencies? Health care is complicated, and if insurers have a hard time figuring out what is unnecessary then I think it would be even harder for patients.
I believe that it’s extremely important to continue studying all aspects of the health care system to identify particularly inefficient areas. For example, a major source of inefficiency is post-acute care (https://www.bostonglobe.com/opinion/2015/03/30/hospital-care-saves-money-and-lives/3QEgafleCxuxzqV6rcdliK/story.html), but HDHPs would do nothing to address this because patients have usually used up their deductibles by the time they get there. Also, how is a patient in this situation supposed to know whether it would be better for them to stay in the hospital or be discharged to post-acute care? As another example, many states have laws prohibiting nurse practitioners from providing standard services because of lobbying by doctor’s associations. No type of health insurance will allow patients to get a cheaper service if it isn’t being offered.
To sum up, I think the idea of a monopsony to reduce costs is interesting, but pairing it with high-deductible plans would probably be ineffective and, if unmitigated by additional coverage, cruelly expensive for the unlucky. If mitigated by additional coverage, I think it has a chance of being effective, but I think any reasonable level of coverage would probably reduce choice and I’m skeptical that it justifies overturning the ACA.