After about a year and a half of working in health care, and at least some casual reading of the relevant literature, I’m increasingly skeptical that a libertarian free market can provide an adequate basis for the provision of health care. The longer I work in the field, the more convinced I become of the essential truth of Kenneth Arrow’s famous insight about the economics of health care:
[T]he special economic problems of medical care can be explained by adaptations to uncertainty in the incidence of disease and in the efficacy of treatment (emphasis added).*
That’s an understatement. I’d be more inclined to say that the economics of health care is like economics done under conditions beset by the Evil Demon of Descartes’ Meditations. I’ve also come to think that Arrow’s insight applies even more broadly than he suggested, not just to medical care per se, but to medical research, to the provision of ancillary services, to preventive care, to administration, and so on. In general, I would say that Arrow’s insight lends credence to the thought that health care differs fundamentally from most of the other goods that we buy and sell in the market, which leads in turn to the (tentative) inference that the market is ill-suited to its provision. Anyway, consider this post another expression of my skepticism on this score (this one being the first).
As with my last blog post on this topic, I have a particular case in mind, one that (once again) involves the funding mechanism for medical research. Consider a thought-experiment based on an actual case.**
Imagine that a medical researcher has promising preliminary research for the cure of a serious, chronic, widespread disease that currently lacks a cure. Suppose further that the research satisfies the following description:
It’s expensive enough to demand a substantial amount of long-term funding.
The researcher’s hypothesis about the etiology of the disease is true.
Given (2), the researcher has a corresponding hypothesis/proposal about a possible therapy, which also turns out to be true.
The therapy cures the disease in a safe, decisive way at very low cost.***
The envisioned therapy is so cheap that there is little profit in administering it–so little that providers of the therapy would be unlikely to recoup investment costs from it in any reasonable amount of time.
The envisioned therapy has the further effect of wiping out the entire industry devoted to treating the disease as a chronic condition. (I say “the industry,” but realistically, that probably means “several different but interrelated industries.”)
Though not an absolute, it seems to me that a free market is biased against any research program with the preceding six features. Very few entities would, at step (1), have a profit-making incentive to fund the research. And some of the succeeding steps only make things worse. Left unfunded, such a research program might well wither on the vine despite the truth of its claims and/or the promise of the therapies it proposes.
Even if there were non-profit-driven entities in the marketplace willing to fund the research, precisely because they weren’t profit-driven, they might lack the funds to invest in an optimal version of the research program. And so, the research program might consistently be tested in a methodologically sub-optimal way. A predictable consequence would be chronic, self-fulfilling uncertainty about the merits of the research. An underfunded series of studies might (precisely through underfunding) fail to demonstrate the merits of the research despite the truth of the hypotheses involved, merely reproducing and intensifying unfounded doubts through defects in the testing protocol–defects explained by lack of funding, not by the supposed demerits of the hypothesis.
Given this, a series of inconclusive studies might easily (however unfairly or prematurely) be taken to indicate the wholesale failure of the research program. Hence the research program might fail–might even be entirely lost–despite its merits, and despite the fact that, if properly funded, it might have demonstrated the truth of its claims.
This is, to be sure, a single thought-experiment involving a single disease. And it might legitimately be objected that a single thought-experiment, even if unanswerable, could hardly be thought to sink the idea of free market medicine as such. A thought-experiment is an exercise of imagination, not an assertion of truth. And the results of such an exercise point to a possibility, not an actuality. Beyond that, as stated, this particular thought-experiment points at best (or at worst) only to a single possibility.
Fair enough. So consider another thought-experiment, or less grandly put, an addendum to the first one: imagine that the preceding case can be generalized across a range of similar cases. And imagine that it can be generalized past medical research to other parts of medicine–preventive medicine/ primary care, surgical services, the marketing of pharmaceuticals and medical devices, medical administration, etc. If so, it’s a structural defect of any free market system, whether explicitly “capitalist” or not, that it lacks a profit-making motivation to fund this research. That would, if true, indicate a serious defect in any free market approach to health care as such.
Some obvious objections come to mind at this point. Am I not assuming, falsely, that everyone in the market is driven exclusively by a desire for profit, and isn’t that obviously false? The free market may incentivize profit-maximization, but it permits other motives, as well. So we can’t assume that profit is the only possible motive behind action in a free market. Indeed, most of the most famous medical research organizations in the United States arose for charitable reasons, and were (or are?) driven by charitable, not monetary considerations. So we have historical evidence to suggest that a great deal of medical research is undertaken in defiance of the imperative to maximize profit. Why then assume that if a research program isn’t profitable, it wouldn’t be funded within a free market? Many have been, and many could be.
There’s a great deal of truth in this objection, but it still, I think, misses the mark. I’m not assuming that literally everyone in the free market is driven by a desire for profit. I’m simply assuming–with most, if not all, of its defenders–that the free market has an incentivizing bias in the direction of profit: i.e., that the market differentially favors profit-making enterprises as against enterprises that don’t.
The standard free market claim is that free market profit-seeking and competition track desired (hence desirable) outcomes via the invisible hand, and that tracking desirable outcomes via the free market is both morally and economically superior to any other option at our disposal. It’s morally superior because it relies entirely on voluntary action without recourse to first uses of force, and it’s economically superior because market-revealed preferences are a more effective route to socially desirable outcomes than any route that bypasses them.
The fact remains that while a free market permits non-profit-oriented action, it favors profit-oriented action, and does so at the expense of enterprises that don’t make a profit. So research that doesn’t promise a profit is bound to be disfavored relative to research that does. That’s still a problem, and it’s all that my thought-experiment requires.
In any case, since profit is the fundamental and charity only a secondary or relatively marginal force in the market, profit-maximization is the fundamental driver and fundamental ideological selling point of the free market. If something can’t be done at a profit, it can perhaps be done through philanthropy (true), but if there is a basic mismatch between the desired outcome and the imperatives of profit-maximization, there is still bound to be a systematic mismatch between that outcome and the fundamental mechanism behind a free market.
Philanthropy is a more or less effective way of patching up the shortfall between profit-maximization and unprofitable-but-desirable social objectives (how effective is a hotly contested issue), but there’s no way of denying the existence or systematic character of the shortfall. Given that, advocates of a free market are either obliged to deny the desirability of the unprofitable outcome, and/or (if they admit it) to insist that profit-maximization and/or philanthropy can rise to the challenge of meeting it.
Reflection on the particulars of my thought-experiment should suggest why the preceding gambit is implausible. If research is not profitable, indeed, would wipe out profits of an existing industry, there is bound to be a strong, indeed visceral aversion to it among players in the health care market. And these players need only have garden-variety motivations; they need not be stereotypical villains of any kind, just ordinary profit-maximizers of the kind we entertain at parties or hang out with at the bar. The research they overlook or reject may save lives, but does so at the cost of an industry more powerfully dedicated to other things.
Advocates of the free market tend to underplay the trade-off involved between the internal and external goods of medicine–between the pursuit of desirable clinical outcomes and the pursuit of the material rewards of participation in a lucrative profession. But the trade-off is very real, and only too obvious to anyone in the field.
As for the fact that the American market has given rise to philanthropy-based research, it’s undeniable that a free market makes that possible. But it’s also undeniable that nothing about a free market makes philanthropy a scientifically reliable source of funding–by which I mean, nothing about a free market directs funds to the most scientifically promising research programs, especially in cases where those programs have little profit-making potential (or market-discernible potential) about them. “Market discipline” can (and does) discipline and punish scientifically worthy enterprises to death. Facing stringent market discipline, such organizations can either fold, or else survive by turning themselves into profit-oriented businesses. Arguably, this has in fact happened to American medical research. If it happens often enough, at the expense of meritorious research programs of the kind described in my thought-experiment, we back into the problem that the thought-experiment serves to depict.
Put differently, the thought-experiment serves to highlight the problematic move from “desired” to “desirable” in the standard free-market account. Desired outcomes as filtered through a market process are not necessarily desirable by other normative standards, e.g., by the imperative to reward moral desert, or to satisfy medical needs, or to achieve positive clinical outcomes, or (frankly) even to show minimal kindness in cases where displays of kindness are not remunerated (as they often aren’t). Nor can the divergence between market outcomes and the satisfaction of these other standards be finessed by insisting that the market can inevitably be imagined into tracking them. That just seems like imaginative hand-waving–hand-waving vulnerable to counter-displays of imagination involving thought-experiments intended to disconfirm them.
What we seem to be left with is the adamant assertion that any aim pursued voluntarily is morally superior to any aim pursued by first-uses of force, regardless of the magnitude of the divergence between what the market delivers, and the satisfaction of any other normative aim. Better to preserve economic freedom at the price of the perpetuation of a chronic disease than to fund the cure by a mechanism that violates economic freedom. Maybe, but it’s a hard sell–certainly a hard sell if you have the disease in question, and also a hard sell if the decision is to be made behind a Rawls-type veil of ignorance.
All of this suggests that Robert Nozick was right to argue that free markets resist patterned distributions–including patterned distributions involving moral desert, need satisfaction, positive clinical outcomes, and counterfactually stable displays of non-remunerated kindness.**** Indeed, if progress (including medical progress) exhibits a patterned distribution, then the workings of a free market seem at odds even with the stable display of this pattern. That’s why opponents of the market call themselves “progressives.” Libertarians find this maddening, but it shouldn’t be. If “liberty upsets patterns,” it upsets progress. It seems natural enough for those who favor progress over markets to choose a name that reflects their commitments and priorities.
Where Nozick saw the market’s tendency to “upset patterns” as a morally neutral (or perhaps even morally groovy) feature, others might be inclined to see it as a bug. Taking Nozick at face value, it’s an understatement to say–with Yeats–that under a free market, “the center does not hold.” Nothing holds, except the desire for profit, or put differently, the desire to satisfy desire. As far as the market is concerned, the satisfaction of desire is a proxy for a summum bonum: it takes on a life of its own that bypasses any other normatively salient end or pattern. At a certain point, the divergence is bound to seem or become intolerable. And for obvious reasons, health care is the place where the intolerability will seem most obvious. Once you see its workings up close, you begin to see the truth in Marx’s dictum that the capitalist desire for profit is a fetish. If you’re in health care, you also begin to wonder what this fetish has to do with medicine.
Many of my thoughts on this subject are variations on a theme I’ve blogged here before: to the extent that a free market in anything is driven by the profit motive, it serves widely desired ends, but there are domains of life where the satisfaction of desire on a market model can be systematically at odds with otherwise desirable outcomes. In and of itself, that’s not a conclusive objection. But once the examples pile up, the normative strikes against the free market start to add up as well. This isn’t to say that there are no good ideas to be borrowed from market-based approaches to health care. Nor does it point unequivocally in the direction of any single policy prescription. But there’s a real problem here that needs a better solution than adherents of free market solutions have (as far as I can see) so far generated. If there is a market for better solutions, I hope it kicks in soon.
*From Victor Fuchs, “Kenneth Arrow’s Legacy and the Article that Launched a Thousand Studies,” in Health Affairs, citing Kenneth Arrow, “Uncertainty and the Welfare Economics of Medical Care,” American Economic Review, 53:5 (1963).
**The actual case is Denise Faustman’s research on Type 1 diabetes, as described in Elisabeth Rosenthal’s An American Sickness: How Healthcare Became Big Business and How You Can Take It Back, pp. 182-84. Faustman’s research is highly controversial, but evaluating it is well beyond my competence, so I don’t mean to be taking a position on its merits. For a summary, see “Why Denise Faustman Isn’t Deterred by Skeptics…” Healthline, January 2021.
***Recouping the investment becomes somewhat more likely if patents are involved, but if we take patents out of the equation, there’s no reason to think that an otherwise cheap therapy would be a money-making venture, or could without force or fraud be turned into one. And there are two ways of taking patents out of the equation: we could either suppose that the therapy in question relies entirely on generic drugs and/or medical devices; or we could reject the legitimacy of patents as such. Take your pick.
****Robert Nozick, Anarchy, State, and Utopia, pp. 160-64. Nozick’s objection to patterned conceptions is that maintaining them requires “continuous interference in people’s lives” (Anarchy, 163). I agree. What he doesn’t show is that the interferences in question are necessarily rights violations on a normative conception of freedom of the kind he himself adopts (in, e.g., his famous paper, “Coercion,” in Socratic Puzzles). That said, I reject the gambit taken by Nozick’s critics to the effect that we can have patterning without continuous interference (e.g., Stephen Nathanson, Economic Justice, pp. 76-78). Within health care, at least, there’s no two ways about it: a commitment to a patterned conception of justice does require continuous interference of some kind. The question is not whether interference is required, but whether the required interference violates the rights of those interfered with.
Let’s ask a preliminary question about the thought experiment: why does the research needed to prove the theory correct, and the treatment effective, require substantial long-term funding? In most Western countries, the main expense of medical research by far is not finding treatments, but proving to the satisfaction of the authorities that a treatment is indeed effective, and moreover, that it doesn’t do harm to the patient in some way.
Now there are good reasons why this expensive gauntlet of tests exists. But it isn’t a failure of the market mechanism that its existence prevents some cheap, effective medical treatments from being used. It’s a cost of regulation – a deliberate judgement of Western states, that the danger of poisons being falsely sold as medicines is greater than the danger of sound treatments not being used.
There is the further point, counterbalancing 6), that nearly all physicians and patients would far rather cure a chronic disease than treat its symptoms. In the absence of friction (in the sense Coase used that word – transaction costs) this strong desire not to suffer from a curable disease would override the interests of those who provide treatments of the symptoms – particularly if, as is normally the case, the latter’s expertise transfers to treatments of other diseases. Again, there are good reasons why the cost of proving the validity of medical treatments is kept artificially high, but it isn’t a structural defect in the free market when a well-meant regulation prevents the socially optimal result.
LikeLiked by 1 person
The point you make in your first two paragraphs is inarguably true if one accepts the in-principle legitimacy of FDA-type regulations. Your view–common enough among conservatives–treats regulations as though they were part of the market, or as a necessary adjunct of the market. That’s a reasonable view, but it’s not compatible with libertarianism. Libertarians don’t accept the legitimacy of the FDA, or of FDA-type regulation. They regard such regulations as pre-emptive, first uses of force, hence as violations of economic rights, and as incompatible with a free market. (Nozick himself has a complicated, equivocal view on risk and regulation, but the view I’ve just described is pretty common about libertarians.)
The thought-experiment was primarily geared toward raising questions about libertarian defenses of the free market, not conservative (or garden-variety liberal) ones. The libertarian defense involves the application of a single univocal proscription on first uses of force to the entire social system and economy, one that rules regulation out from the start. The conservative one is a more eclectic affair. I agree that my thought-experiment has comparatively little bite against the latter view.
As for your further point, that “nearly all physicians and patients would far rather cure a chronic disease than treat its symptoms,” ‘I’d make two points. One is that it assumes that physicians and patients call the shots. Where they do, or when they do, perhaps you’re right. But when and where they don’t, that claim won’t cut much ice. Patients and physicians want all kinds of things they can’t get. They’re constrained by other actors. There’s no way of knowing, a priori, which situation would obtain in a free market–physicians/patients calling the shots, or both of them being beholden to some third or fourth parties who do.
Second point: take the cases in which physicians call the shots. (Arrow’s point was that given the uncertainties and duress of illness, physicians generally have asymmetric power over patients, so cases in which “physicians and patients” call the shots are primarily cases in which physicians do.) In the situation I’ve described, the therapy rests on true hypotheses that go unproven for lack of the resources to demonstrate their truth. The physicians, meanwhile, are embedded within an industry (or several interconnected ones) dedicated to treating what they and most of their peers in the field regard as a chronic condition. So the physicians are simultaneously in a situation of uncertainty, of conformity, and of path dependency. They can’t be certain that the experimenter’s therapy is sound; most of their peers are speaking and acting as though it isn’t sound; and they’ve invested a great deal, financially and emotionally, in the infrastructure dedicated to treating (what they regard as) a chronic disease.
To describe the choice they face as a problem of “transaction costs” is an understatement, to say the least. It’s more natural to say that the incentives here militate sharply against the optimistic view you’ve taken. In the circumstances I’ve described, it would take rare independence, integrity, and far-sightedness to opt for the cure rather than to ignore it in favor of “chronic treatment.” Taken separately, very few individual physicians have a strong financial incentive to do so. The relevant incentives to choose the cure over perpetual treatment are moral, not financial. But neither an Adam Smith nor a Karl Marx would gamble on those.
I have no a priori way of knowing how many physicians would go one way, and how many would go another. But I doubt anyone else does, either. It’s not the kind of thing that’s knowable a priori. I’d leave it at this: to the extent that defenders of a free market in health care rely for the soundness of their case on the benevolence and economic selflessness. of physicians, that’s a weakness in their argument. They either need to ditch it or work around it. Relying on it won’t get them very far.
If your thought experiment was meant to challenge libertarian defenses of free markets, not conservative ones, it misfired. A libertarian can – and does – cite such cases as evidence that the FDA should not exist, because even as a piece of pragmatism it does more harm than good. Such arguments are his stock in trade; they’re of the same kind as Bastiat’s famous parable of the broken window. It’s the more eclectic conservative who grants that institutions like the FDA are justified in principle who has a real difficulty with your thought experiment.
As for the other point, defenders of free markets in health care don’t rely on the altruism of physicians, but on the self-centered desire of patients to be healthy instead of sick. Patients who learn of possible cures to their chronic diseases will, if determined enough, set aside physicians who refuse to attempt them. Assuming that, because patients generally follow physicians’ advice, the physicians’ incentives will override the patients’ when they are in conflict, is a mistake – such things can happen, but they often don’t. Arrow’s point is basically an instance of the principal-agent problem, with the physician as the better informed agent of his patient. It is a genuine problem, but not a decisive argument against free markets.
Incidentally, if the benevolence and economic selflessness of a physician treating his patients in person can’t be relied on, then the benevolence and economic selflessness of government regulators considering millions of patients in the abstract, working only with statistical reports, is surely even less reliable? You can’t solve the principal-agent problem with agents further removed from the principals – the distance between agents and principals is where the problem comes from in the first place.
Now that’s something I can disagree with. You say:
Yes, libertarians do that, but if that’s the best they can do, their best won’t do. The inference from “the research won’t get funded in a market” to “the FDA should be abolished” is a non-sequitur. What gives it plausibility is the very claim under scrutiny, whether the market needs supplementation by an entity that is not driven by market forces. The libertarian claim is “no, we don’t need the FDA.” Well, even if we don’t need the FDA per se, the further question arises: do we need something like FDA-type regulations, i.e., regulations that require clinical trials for safety and efficacy of medications and medical devices prior to their administration to the ordinary patient population? Either the answer is “yes” or “no.”
If it’s “yes,” then we need to be shown a viable libertarian-market-consistent alternative to the FDA, and need to be shown that its regulatory costs are such that the therapy in the thought-experiment can be put through clinical trials, then administered, without recourse to redistribution. The thought-experiment doesn’t rule that possibility out, but simply asserting that libertarians can invoke Bastiat doesn’t demonstrate anything of great significance. Libertarians invoke Bastiat at the drop of a hat. One reason for doing so is that they have nothing better to say.
If the answer is “no,” then the cost of a purely libertarian approach to health care is to sacrifice one of the basic principles of public health–that clinical trials should be distinguished from administration of a drug to the general population. Even apart from public health issues, given libertarian scruples about informed consent, I’d think that this distinction is well worth maintaining. It’s doubtful you can get patients to consent to experimental treatments by ordinary administration of the treatment in the doctor’s office. It’s not impossible, but it would require a fundamental revision to the way that primary care physicians practice.
Could a libertarian bite this latter bullet? Of course. But is the bullet worth biting? That’s the question that the thought-experiment leaves open.
On your second comment: I didn’t mean for the thought-experiment to be decisive. I simply meant for it to identify a problem in need of a solution. There are more and less sanguine ways of describing the principal agent problem in medicine. I’ve identified a case in which there are strong reasons to adopt a pessimistic view. You’ve moved to a more general level and described things in an optimistic way. If so, we’re simply talking about different things. The basic empirical issue is this:
It’s an empirical question how often things go one way rather than the other, but a common sense consideration to bear in mind is that the physician has asymmetric resources and power that the patient lacks, especially in the case described in the thought-experiment. The patient is suffering a chronic and debilitating disease; the physician is not. The patient is in a state of anxiety-producing uncertainty; the doctor, much less so. The physician has less need of the patient than the patient of the physician. The patient incurs real costs in visiting a doctor; the doctor incurs few costs in having a patient visit him, or replacing that patient with someone else. If we add to this that the physician has mixed financial and clinical motives, and can easily (relatively costlessly) refuse a patient, it seems natural to infer that “such things not only can happen, but happen a lot.”
I would just reiterate that if the libertarian case for free markets in medicine really turned on your sense that “such things don’t often happen,” they’d be gambling on a rather implausible empirical hypothesis (one that flouts the point of Arrow’s paper). Maybe it’s true, but it’s very common to find libertarians speaking and writing as though certain empirical facts must be true because their truth is required by consistent adherence to libertarianism. The point of the thought-experiment is to expose that move when people are tempted to make it.
As for your third point:
I didn’t say that physicians’ benevolence couldn’t be relied on as a general proposition. I said that it couldn’t be relied on in the scenario described in the thought experiment, where the incentives line up against it. I’m the last person to dispute that regulators face a version of the same problem: my wife committed suicide because she had trouble accessing pain medications, a predicament created by regulators’ so-called “war on opioids.” The CDC has only now, a year later, considered revising the regulations that arguably pushed her over the edge.
But the bad incentives involved in the thought-experiment are primarily financial, and are primarily a problem for providers (physicians and pharmaceutical manufacturers, plus everyone involved in the industries having to do with chronic treatment of the disease), not regulators or government funding agencies. The latter have no strong incentive to demand that a generic therapy that cures a chronic disease not be funded. If anything, they have the reverse incentive: they come out looking like heroes if they dump on the providers, and fund the salvific cure.
In any case, your rhetorical question makes two questionable assumptions. For one, it seems to assume that only government regulators work at a distance from the clinical action (or research action), armed only with statistical reports. Actually, that describes the entire administrative side of modern medicine. It even describes people in utilization management (or other middle managers) who work in the hospital (or in patient financial services, on a campus a few miles away from the hospital). A UM person who never leaves her office is in exactly the same epistemic situation as a bureaucrat in Washington. She’s just reading statistical reports. Same with the people making determinations of “medical necessity” in an insurance office, or in a company doing denials management. I work for a company that does the latter. If “working at a distance from the clinical action armed only with statistics” is an objection, it cuts every which way.
But second, it’s. hardly obvious that “working at a distance from the clinical action armed only with statistics” puts you in a suboptimal epistemic condition relative to the average clinician. Most physicians know less about statistics than the average holder of a master’s degree in public health or public administration. Lacking that, they have no idea how to make sound statistical inferences about the safety or efficacy of a therapy, or by implication how to read and process epidemiological findings in scientific journals. Often, they rely on seat-of-the-pants anecdotal information and hunches. There’s no reason to think that proximity plus lack of reliance on statistics is a more reliable source of clinical information than statistical inference over large populations or samples.
“If the answer is “no,” then the cost of a purely libertarian approach to health care is to sacrifice one of the basic principles of public health–that clinical trials should be distinguished from administration of a drug to the general population.”
The typical libertarian would bite that bullet, though – it wouldn’t even be difficult for them. After all, the reason for that rule is to prevent the sale of useless or harmful drugs to the general population, and the libertarians don’t think that ought to be prevented. In their minds, if a drug does prove to be harmful after it’s been in use for a while, lawsuits can compensate the people harmed and force the manufacturers to stop making it.
And, well, we allow such lawsuits already, because clinical trials don’t always catch the harm a putative medicine can do. It isn’t obvious a priori that a massive effort to keep bad treatments from the public, which can’t guarantee either that all bad treatments are blocked or that no good treatments are, is better than just monitoring the results of treatments when the public uses them and issuing warnings when problems appear, as the libertarians would prefer.
“It’s doubtful you can get patients to consent to experimental treatments by ordinary administration of the treatment in the doctor’s office.”
I’d think it would be exactly as difficult as it is now to get patients to participate in clinical trials. You wouldn’t expect primary care physicians to use experimental treatments on their patients – even in a free market the average clinician is going to be more risk averse than the average patient, for reasons you yourself enumerated.
“The latter have no strong incentive to demand that a generic therapy that cures a chronic disease not be funded. If anything, they have the reverse incentive: they come out looking like heroes if they dump on the providers, and fund the salvific cure.”
Ah, no, this claim is contradicted by the facts. The FDA, and regulatory bodies like it, are famously very risk-averse. It’s much worse, for them, if they approve a therapy that ends up harming patients than if they reject a therapy that would cure a chronic disease, because if the cure is rejected nobody will ever know it works, but everyone knows when a drug turns out to be poisonous. (Another case of Bastiat’s point: everyone sees the patient killed by his medicine, nobody can see the patients not cured by medicine that can’t be sold.)
Your first comment is simply a concession. If libertarians bite the bullet you mention, they’ve conceded the point I’m making through the experiment. The idea of dealing with medical harm entirely through malpractice lawsuits is a non-starter. Most people harmed lack the resources to initiate them, and even in the most successful cases, the time involved to collect is prohibitive, as are the contingencies involved along the way. My late wife had a failed laminectomy in which unsecured equipment fell onto her spine during the procedure, crippling her for life. She was too busy dealing with the aftermath to sue anyone. Having set up spinal cases myself in an OR, it is patently obvious to me that the OR was at fault. But fault didn’t matter. She lived with it, and died from it. The OR didn’t suffer a scratch.
One of the ironies of Nozick’s libertarianism is that he claims to reject a “utilitarianism of rights” in which some rights take precedence to others in terms of enforcement. Rights violations are not merely to be minimized but eliminated in a libertarian regime. But it should be obvious that if iatrogenic injury is a rights violation (or often is), a regime that lacks pre-emptive measures against unsafe medicines, devices, or procedures, will fall far, far short of Nozick’s supposed ideal. Ours does, and so would his.
As for your second point, you’re ignoring that the regulatory process requires animal model testing before it involves testing on humans. Animal testing creates a baseline for further evaluations of safety and efficacy, so that one isn’t just testing willy-nilly without having any idea of what one is doing. It’s outlandish to imagine primary care physicians engaging in animal model testing. Further, a clinical trial has to have a control arm of healthy people. But the ad hoc testing that you’re describing would lack one, and in lacking one, would be severely compromised, scientifically. In general, a clinical trial has to follow fairly precise, rigid scientific protocols that we couldn’t expect primary care physicians to follow in ordinary administration of a drug. Beyond that, the samples involved in primary care administration of the drug would be tiny, non-random, and ultimately scientifically uninformative. The testing protocol would not be double blind. Etc.
In short, there’s no way of avoiding the fact that if you don’t run formal clinical trials, you’re compromising the scientific bona fides of your therapy. If libertarians want to bite that bullet, fine. But it’s a serious cost of their view. What they can’t do is claim that if we abolish the FDA, and do away with any market-based alternative to the FDA, we somehow reduce the costs associated with regulation and manage to get the same safety and efficacy that we currently get out of clinical trials. That claim marks a descent into a fantasy world.
It also is not true that it’s just as easy to recruit people into a clinical trial as it is to spring the idea of having a patient become a trial subject from within a primary care setting. Those are two different populations of people in two different settings. Yes, they may have the same medical condition, but there is a big difference between inviting someone with medical condition X into a clinical trial, and springing the idea of a clinical trial on a patient who’s gone to a primary care physician expecting conventional treatment for X. The distinction between the two is not hard-and-fast, but it’s not imaginary, either, and can’t be treated as though it was dispensable. That said, it’s a question worth asking people in the field, something I intend to do. (And to repeat: the clinical trial requires healthy people as well, people not likely to show up at the office of a primary care physician to sign up for a half-assed pseudo equivalent of a “clinical trial.”)
As for your third point: the FDA and regulatory bodies like it are famously risk averse, true. They can also be flexible when they see the need for it. The Emergency Authorizations for the COVID vaccinations are a clear case in point. The standard objection made to them was not that the regulators took too long to authorize the vaccinations, but that they were too quick about it. It is possible that the FDA et al would be overly risk averse in authorizing the medication in my thought-experiment, but it’s possible they might be expeditious. In the latter case, a regulated and redistributive market would do a better job than a market lacking an FDA, an NIH, or their equivalents. Broad generalizations only take us so far in medicine, or for that matter, in social science.
Well, I’ve nothing further to add on the first two points. On the third, though – in my opinion the response to COVID has been one of public regulators screwing up by the numbers, from first to last. The vaccines are a particularly clear instance of politics-based decisions. Faced with several possible treatments, some based on antivirals in common use with well-understood side effects, and others based on truly experimental mRNA techniques, never previously used in any licensed treatment, the FDA allowed the most experimental and riskiest methods and forbade the ones based on proven drugs – because the mRNA vaccines were backed by pharmaceutical companies, as far as I can tell, while the antivirals are generic drugs with no advocates. In other words, when put under pressure to authorize something, the FDA went with the choice that had the most political backing, not the ones with the least actual risk of medical complications.
Given the hypotheses of your thought experiment – that the treatment is cheap, once the cost of proving it’s safe and effective has been met – it’s clear to me that it’s exactly the kind of treatment the FDA would never authorize, as it would be impossible to collect enough political backing to overcome the FDA’s risk aversion. I don’t think a legal regime that refuses to take risks unless driven to it by panic, and thus incapable of assessing risks rationally at any time, has much to recommend it. But is the FDA capable of being anything else?
That doesn’t strike me as a factually plausible description of the process of the EUA process for mRNA vaccines, or of the efficacy of any generic therapy. I’ll just say if that’s what it takes for a libertarian to respond to my thought-experiment, point noted.
Irfan, do you have thoughts about a) Boldrin and Levine’s arguments about pharmaceutical patents —
Click to access ip.ch.9.m1004.pdf
— or b) Kevin C.’s arguments about the healthcare industry?
Click to access C4SS-The-Healthcare-Crisis-A-Crisis-of-Artificial-Scarcity-by-Kevin-A.-Carson.pdf
I had read Kevin’s piece before, but not Boldrin and Levine’s. I have thoughts on both, but don’t see the relevance of either to this post.
I agree with a lot of what Kevin says, and disagree with some, but one over-arching problem I have is: even if you accept all of what Kevin says, how far does it go? How much of what we call “the health care problem” is explained by his explanations, or resolved by his prescriptions? That would require a discussion devoted specifically to the strengths and weaknesses of his paper, which this post was not designed to do. That said, my claim at the very end–“this isn’t to say there are no good ideas to be borrowed from free market approaches to health care”–is implicitly a reference to Kevin’s paper, among others.
Unless Kevin would want to argue that my thought-experiment could only arise within the context of a licensing cartel, my thought-experiment side-steps virtually everything he says in his paper. My contention is that the situation described in the thought-experiment could arise within our current mixed system, or within a pure free market. If libertarians would want to argue that it could not arise within a free market, I’d like to see the argument. The thought-experiment is designed to elicit one. But there is no such argument in Kevin’s paper.
As for Boldrin and Levine, my view is that patents (and copyrights) are, in principle, a legitimate form of property. So I am prima facie opposed to any view that literally rules them out entirely. That said, I find the current IP regime extremely problematic, so I’m sympathetic to many of their arguments. But the fact remains that the thought-experiment explicitly does an end-run around the entire issue of patents, leaving it to the reader to adopt her own favorite rationale for subtracting them from the equation. See the third footnote.