Wednesday, January 25, 2023

Casually Dismissing the Thesis of Burning Down the House

 In a recent links roundup post at Astral Codex Ten, Scott Alexander links to this piece with the following commentary:

One of the most common objections to libertarianism, right after “but who would fund the roads?”, is “wouldn’t a private fire department leave your house to burn if you hadn’t paid?” Here is a very long investigation by someone who has investigated the history of private fire departments and says that - at least in early modern England - the answer was no. I don’t want to argue with this detailed historical scholarship, but I notice I am confused - if the private fire department would save your house whether or not you paid, what was the incentive to pay? [update: see here for answer, the companies sold fire insurance]. Related: government fire department lets man’s house burn because he hadn’t paid a $75 fee, and there was no procedure for allowing him to pay on the spot.

Emphasis mine, and here's the link about the incident in question. I found this amusing, because Andrew Koppelman made a big deal out of this event as if it were a black mark upon libertarianism in his book Burning Down the House. (My earlier posts on his book are available here and here.) I think Scott Alexander's reaction is what a normal person would think if they'd heard this story without any explicit prompting that it reflects on free markets or libertarianism. (I don't see any reference to Koppelman or his book in the comments, so as far as I can tell this is an unprompted reaction to the story.)  Alexander has dismissed Koppelman's thesis, without intending to or even knowing that he did so. 

If you don't know about the story, this was a government fire department, refusing to service a house in an area that previously got no fire department services at all. This is more a reflection of the DMV model of customer service than it is or untrammeled free markets. It is typically governments that decline to provide service even when there is an obvious deal to be made that would benefit both parties. In contrast, markets usually find a way. 

Sunday, January 8, 2023

Reflections on Burning Down the House

Andrew Koppelman's book Burning Down the House deserved a little more scrutiny than I gave it previously. I wrote an earlier post reacting to the book, and I stand by it. But I took some time to re-read and annotate so I could deliver a more thoughtful review. 

When I wrote the earlier post, I was under the misapprehension that Koppelman thought of himself as a libertarian. The whole book (title included) has this air of "You're doing it wrong, here's the true path." Which would presumably be coming from the position of someone who was following the true path. He actually never identifies as a libertarian. He is some kind of pro-market progressive (my description, not his). The book occasionally provides a flourish of rhetoric in favor of markets. But when it comes to particular regulations or government programs, Koppelman is almost always pro-state. 

Too many on the left fail to grasp that the original libertarian strategy has been massively vindicated. The capacity of markets to alleviate poverty has been so overwhelmingly demonstrated in recent decades that it is silly to keep denying it. Too many on the right fail to grasp that unregulated markets cannot deliver a livable world. Moderate libertarianism can bridge some of the bitterest divisions of contemporary American politics.

Amen to the first two sentences. This book could be a useful corrective to the excesses of the left if Koppelman expounded on this a little.  Here is a passage that seems right out of Steven Landsburg's Price Theory text:

Should a farmer grow beans or zucchini? It depends on what consumers want. How can he know what they want? And how can he know if consumer preferences change—if, for example, there is a surge of interest in fried zucchini flowers? Only prices in a market can convey the necessary information quickly and accurately. If the price of zucchini rises, farmers will plant more of it without needing to know anything about culinary fashions, or any of the incalculably many other factors that affect the demand for vegetables. The irreplaceable virtue of free markets, Hayek thought, is their capacity to manage that flood of information. Mises had argued, against socialism, that without a market, one can’t rationally calculate how to allocate capital. Hayek argued, more radically, that the necessary knowledge could not even be collected. Innovation often emerges from individuals’ tacit knowledge, inexplicit even to themselves. Its value is not reflected in existing prices, because the innovation does not yet exist. The problem is not the complexity of calculation, but the lack of information. No central authority can know all the competing demands for resources.

He seems to have a mature understanding of how markets allocate resources (an understanding that most leftists lack). But, again, he could have provided a few more examples of bad government interference in markets. On my second time through the book, I was specifically listening for any such instances. At one point (apropos of nothing) he says, "There is no excuse for farm subsidies." It wasn't connected to a discussion or cost-benefit analysis of farm subsidies, it was just given as an individual instance of a bad government program. Articulating the reason why farm subsidies are bad might have highlighted a higher principle or helped the reader to identify other bad government programs. But I suspect Koppelman is trying hard to avoid discovering any such principle. I suspect he thinks there are wise technocrats who can very thinly "slice the salami", identifying and rejecting the bad policy proposals and implementing the good ones. The libertarian perspective cautions that this is a fraught path. Once empowered, governments do a poor job of discerning good from bad. Politicians even self-consciously implement bad policies because they are popular. Even "wise technocrats" have a tendency to get the calculus wrong, or to be captured by populism or by trendy "science", or to optimize the wrong thing, or to minimize/maximize along a single dimension with a total disregard for other considerations. (Think of the public health establishment trying to maximize life-years, with no consideration whatsoever given to the enjoyment people get from their vices. Or think of the response to Covid, which seemed to be shooting for minimization of Covid deaths disregarding all other costs, even deaths from other causes.)

Political Philosophy

Koppelman presents a cursory review of some of the major thinkers of libertarianism. He covers the philosophy and writings of Ayn Rand, Murray Rothbard, and Robert Nozick, using them as foils to be contrasted with wise libertarians like Hayek. He also includes a long discussion of John Rawls, whose philosophy he sees as delivering a devastating rebuke to libertarianism. 

I understand what Koppeleman is trying to do here. Most libertarians would probably consider themselves somewhat grounded in philosophy. Many would cite one of the above authors as being a major influence on their political philosophy. (See "It usually begins with Ayn Rand", for instance.) So it makes sense from this perspective to poke holes in their arguments or point out their weaknesses. At the same time, I think he's overestimating the reliance of libertarians on the arguments left behind by dead men and women. No doubt Rothbard and Rand have some fanatical devotees, but they are a small (though perhaps loud) minority of the movement. I have read a few books by Rothbard (his book on the Great Depression, For a New Liberty, and parts of his Man, Economy, and State). Also Nozick's Anarchy, State, and Utopia. (I haven't read anything by Rand, and I don't feel like any part of my own libertarianism hinges on her philosophy.) I feel like I derived some value from these books, but I also found myself dismissing or disregarding the pieces that were poorly argued. It's not like Koppelman can dismantle Rothbard's moral premises and the entire structure of libertarian thought falls apart without that scaffolding. I see these books as early works of libertarianism, which have been built upon (or, as is often appropriate, disregarded) by subsequent scholars. Call them first drafts of a movement. Individual libertarians can take them or leave them. One can arrive at the libertarian view on a whole slate of policy issues without knowing any of these authors' names. A smart person armed with a little training in economics and some understanding of cost-benefit analysis may end up supporting privatization of schools, repealing minimum wage laws and other labor market regulations, liberalizing insurance markets, ending tariffs, etc., without ever needing to become steeped in moral or political philosophy. This theme of the book simply did not land for me, and I suspect most libertarian readers will likewise disregard Koppelman's arguments.

I'm imagining a libertarian author attempting the same exercise, writing a book about progressivism that takes aim at three "influential" progressive thinkers. Most self-identified progressives will not have heard of any of these authors (some may be familiar with the names but unfamiliar with said authors' arguments) and will simply shrug at this hypothetical libertarian author's devastating rebuke. Perhaps they'd say, "I wasn't really relying on any of those premises or arguments, so...who cares?" Koppelman alleges that libertarian arguments have infiltrated the conservative movement as well. Imagine this libertarian author waving his book, his devastating takedown of progressivism, in the faces of adjacent movements (socialists and left-leaning classical liberals, maybe?). Imagine him insisting that he's undercut the basis of their support for the welfare state, or labor market regulation, or public education. The reaction would be confusion, and a rightful dismissal. 

Koppelman says near the end of the book, 

This book engages with the philosophical arguments for libertarianism: consequentialist claims like those of Hayek, Epstein, and Mises, and rights-based arguments like those of Rothbard, Nozick, and Rand. But political philosophers sometimes need to get over themselves. Most people hold their political views for reasons that have nothing to do with philosophy.

Emphasis mine. I was thinking this throughout his entire discussion of Rothbard, Nozick, Rand, and Rawls. Libertarians mostly are not staking themselves to a rigid moral philosophy and following it off a cliff. (Contra the quote, the book barely engages with consequentialist arguments. That would require dueling cost-benefit analyses of various policies, and the book barely even touches on that.) 

The book's discussion of Rawls would probably be interesting to someone who has never heard of the Rawlsian veil of ignorance. The basic idea is that if we're designing the politics of our world, we shouldn't be able to stack the deck in our own favor. We should craft a world that would work out well if we didn't know who we're going to be. So I might personally favor some policy that makes actuary and physician salaries artificially high because that would directly benefit me, but I should really favor or disfavor that policy with regard to its effect on society as a whole. If I was going to be born into a random body (not one that I already know the conditions of), I'd choose a different slate of policies that are less slanted in my favor. I might opt for a generous social welfare system, not knowing if I'm going to be rich or poor. 

Koppelman gives some details of Rawls's life that I didn't know. He had younger siblings who died of disease, which they probably contracted from him. You could imagine (and Koppelman does imagine) that this experience affected his philosophy. It was mere bad luck that determined which siblings in Rawls's family would survive to adulthood. Plausibly, this experience of being the "lucky" sibling had a profound impact on Rawls' philosophy. This is all fine, but I think Rawls' fans overestimate the political implications of his musings on luck and misfortune. For instance, they vastly overstate the role of luck in determining personal income. (BTW, this whole discussion of luck struck me as odd. There is no way to "redistribute" luck or health such that his siblings would have survived, so the implications of these events for the welfare state are far from obvious.) His philosophy is basically building up to that punchline: incomes are unequal, so we need to redistribute.  This is a non sequitur, though, because people's incomes are largely determined by the choices they make. People choose a profession (or no profession), which has a profound impact on their lifetime earnings. Anyone with at least moderate intelligence and a willingness to hunker down and study can join the professional class: doctors, lawyers, actuaries, business executives, etc. Rich and poor people aren't stamped with an "R" or a "P" on their forehead at birth. There are exceptions, of course. Being born in a third world country dramatically limits your options. So does being born with a serious disability, or suffering a debilitating injury. These conditions can be called "bad luck." Most libertarians don't categorically oppose government aid to people with serious disabilities. And libertarians tend to favor more liberalized immigration (not all libertarians do), such that the person born into a poor country can move to better opportunities. But it's bizarre to frame the differences in income as this mysterious, fortuitous phenomenon known as "income inequality." If you are comparing the annual incomes of two people of modest intelligence who went to the same high school, the difference will be utterly explicable in a large majority of cases. Some of those differences will be quite large, though stemming from similar beginnings. The future auto-mechanic may be sitting next to a future doctor in the same class at the same school. These are people with nearly equivalent opportunities. The difference is often in dedication to study, the decision of which career to pursue, the avoidance of vices, etc., all of which are volitional. (I often observe the same differences in annual incomes between siblings, people who are literally from the same nuclear family. These differences aren't all based on choice, nor are they all based on luck. My point here is that Rawlsians, and people who implicitly rely on Rawls's arguments, vastly overstate the role of luck. They generally argue based on raw income statistics, which affords no correction of any kind for justifiable, explicable differences that one should not wish to correct for with a redistribution scheme.) Koppelman's attempt to make Rawls the hero of political philosophy just didn't land for me.


Amazingly, Koppelman tries to blame opposition to Obamacare on libertarians. He points to a few libertarian thinkers (e.g. Randy Barnett) who opined on Obamacare in order to make his point. But this had almost nothing to do with Republican opposition, which was almost entirely an expression of crass partisan politics. Republican congressmen and their base of Republican voters had a negative emotional valence toward Obama. They weren't deriving their deeply reasoned opposition from the writings of libertarian legal theorists (much less from Rothbard or Rand). By the way, the inverse is also true. The support for Obamacare was largely based on a positive emotional valence from Obama's base, not some deeply held philosophy on the nature and responsibility of the state. 

He also dismisses too cavalierly the constitutional arguments against Obamacare. There is a very simple one, which I think he would probably dismiss with contempt, but which I insist is still in play. Congress does not have an enumerated power to force individuals to purchase health insurance. Article I, Section 8 of the constitution explicitly states which powers the US government has, and there is no provision for anything resembling Obamacare. The federal government has been enacting laws that is has no constitutional authority to enact for at least the past hundred years. Government lawyers have been inventing legal fictions to justify this state of affairs, like playing "Six degrees of interstate commerce." Since the federal government has the explicit authority to regulate interstate commerce (this power was initially granted to prevent states from slapping tariffs on each other, not to give the federal government overarching control of all commerce), they simply need to present an argument, no matter how vacuous, that the activity being regulated is "interstate commerce". (Is growing wheat on your own land for your own consumption interstate commerce? How about growing cannabis that you will personally use for a medical condition? How about declining to purchase healthcare? Plainly not, but the federal government argued that all of these things are "interstate commerce" because it wanted to seize the power to regulate these things.) 

The formal process for doing something like Obamacare involves amending the constitution to grant the federal government the authority it seeks. Critics of this "enumerated powers" argument note that this makes it very hard for the federal government to do a lot of things. Requiring a constitutional amendment means it's easy to create blocking coalitions. What these critics don't appreciate is that this is a feature, not a bug. It's a good thing that big, sweeping, expensive government programs require a high degree of unanimity to pass. Legislation such as Obamacare that fails to pass at the federal level can instead be tested at the state level. But I am just an ignorant pleb, not of the legal tribe. I'm sure legal scholars like Koppelman can provide all kinds of clever arguments as to why a straightforward reading of the Constitution doesn't apply. Actually, these clever arguments are usually wrong. The plain meaning, legible generally to all citizens, is the law of the land, even if the government has been getting away with some other arrangement. For "the law" to be meaningful as a concept, it has to be comprehensible and predictable. It can't be something that shifts with trendy new legal theories. We can't empower government lawyers to rewrite the constitution by pulling and poking at the clear boundaries it sets, inventing clever edge cases and asserting that everything is some kind of exigency. 

(Another example of ignoring/obliterating constitutional process is Congress's abdication of it's duty to formally declare war. It hasn't done so since World War II. Instead, it passes "authorizations of use of military force", which unconstitutionally transfers the power to make war to the executive branch. Modern politicians and legal scholars can sniff and sneer at us "purists" who insist on following process. But it's not an obsession with process for the sake of process. The process is an important check on government power.) 

We don't really need to get into the constitutional arguments, though, because Obamacare has almost nothing going for it from a cost-benefit perspective. This was always my challenge to Obamacare proponents: demonstrate some measure of aggregate public health that has improved substantially (and improved beyond what the pre-existing trendline would have predicted) since passage and enactment of the legislation. I'm not talking about the intermediate metric of how many people have "coverage." If that coverage is actually leading to better care, if the legislation is serving its stated purpose of providing needed medical care to those who would otherwise lack it, we should see an improvement in aggregate health outcomes. Infant mortality. All-cause mortality. Morbidity, particularly for treatable diseases. We should expect to see the benefits concentrated in low-income populations (assuming the data is collected in a way that allows us to analyze such a result). Maybe the impact of Obamacare is too subtle to see in aggregate statistics? (In which case, how could anyone have any confidence that it's "working"?) The narrative about recent trends in public health (past 20 years or so, but certainly incorporating the post-Obamacare era) is that so-called "deaths of despair" have reduced life expectancy for the first time since we've been keeping records, at least in some age demographics. The financial security supposedly provided by Obamacare has not alleviated this despair, apparently. Someone could try to argue that the opioid epidemic is some kind of external shock. But for that shock to the system, Obamacare would have brought life expectancy up. But absent entertaining fantastic counterfactuals, it's hard to find evidence of Obamacare's efficacy. 

We shouldn't be surprised by Obamacare's null effect on health outcomes. The Rand health insurance experiment, in which thousands of families were randomly assigned to groups that got more or less free coverage, saw no effect of additional medical spending on health outcomes. Oh, more coverage correlated with greater use of medicine. The additional medicine was often deemed "essential" by the doctors who prescribed it (despite the fact that the control group, who would forego such additional medicine, faired no worse). Two other randomized controlled trials have arisen since Obamacare's passage, both finding the same null effect. The Oregon Medicaid health experiment used a lottery to assign some applicants to be eligible for Medicaid. A control group was non-eligible (there was a waitlist due to a limited resources, which allowed for a randomization of treatment). From Wikipedia (thought this is consistent with commentary I've seen elsewhere, and this is frankly a plain interpretation of the study's results): Approximately two years after the lottery, researchers found that Medicaid had no statistically significant impact on physical health measures, but "it did increase use of health care services, raise rates of diabetes detection and management, lower rates of depression, and reduce financial strain." A third, and much larger, RCT in the same vein comes from the Karnataka Hospital Insurance Experiment. Here's Robin Hanson on that study: So a new randomized experiment on ordinary health residents of India had 6.8x as many subjects as the RAND experiment, and also found no net effect on health. It only looked at the effects of hospital treatment, but to many that is the crown jewel of medicine. This appears to hold up pretty well. We've run the experiment of giving people tons of free medicine. If "lack of resources" or "lack of access to medicine" were a lethal threat to a large number of people, you should see some effect on aggregate health outcomes when you give a lot of people free medicine.* This result is seriously embarrassing to the left's worldview. If we take the result seriously, it dramatically undercuts any agenda for nationalizing healthcare and regulating medicine to provide "more coverage." This notion of Dickensian deprivation leading to illness and death from lack of medicine is mostly a myth (or, at any rate, adding the supposedly missing medicine does not improve outcomes in any observable way).

Koppelman appeals to an absurd estimate of lives that might have been saved by Obamacare if states had implemented it, citing this study:

Had every state adopted the Medicaid expansion, about fifteen thousand deaths would have been prevented during the program’s first four years.

I can't speak to the quality of this particular study. There are a lot of garbage studies in this space that basically compare the all-cause mortality of people who have insurance to those who don't, then attribute the entire difference to coverage. (Even though people who have health insurance are different in other relevant ways, like having higher incomes and higher levels of education and more likely to be married.) There are various attempts to control for known differences, but all of this is trying to use statistics to simulate the answer you would have gotten from a randomized controlled trial. (Meaning randomization and statistical models are both trying to make the treated and untreated groups equal and tease out just the effect of the treatment.) We have three randomized controlled trials, and they all give us a null result. (A lot of regression studies come to the same conclusion, btw.) Koppelman can cherry-pick studies that make Obamacare opponents look like mass murderers, but he's not on solid ground. 

I think Koppelman doesn't appreciate how messed up healthcare markets are due to government interference and third-party payments. He repeats the trope that healthcare is too expensive, but he states it as thought this is merely a feature of the universe. It's not, it's the result of bad policy. We over-insure ourselves because the government creates bad incentives to do so. "Health insurance" behaves like no other form of insurance, in that it steps in to pay for routine medicine (regular checkups, medications such as antibiotics for mild illnesses, the occasional office visit for illness, etc.). That causes us to consume way more medicine than we otherwise would without regard to the price tag. We're not the payer, so we're not the customer. So providers get away with insanely high bills that get charged to insurance or the government. They suggest more tests and treatments than are truly necessary. And we don't have any particular reason to say "No" to excessive treatment, since we wouldn't personally pocket the savings from foregoing it. Obamacare more or less doubled down on all these bad features of the existing system. In the few enclaves of medicine where there are truly private markets and first party payments, prices have been reasonable, even falling, in contrast to the system that is mostly funded by third party payments. See the Surgery Center of Oklahoma, for instance. Or see Lasik surgery, which has also seen falling costs. This is a working model of what the rest of the healthcare system could be. If we had a 90% private market for health care, I wouldn't object on libertarian grounds to having some subsidies for the truly indigent. If Obamacare were merely a safety net to catch those who couldn't otherwise afford care, it would probably be fine. But that's not what's being proposed. Obamacare proponents accept none of the blame for the bad features of our system that have caused costs to explode. They should, though. 

"Tough Luck"

Koppelman apparently thinks libertarians are unique in having to say "tough luck" to people in dire straights. He uses the phrase repeatedly throughout the book, and it was even in the title of his earlier book on Obamacare, The Tough Luck Constitution and the Assault on Health Care Reform. A couple of examples:

If you get sick and you can’t afford to pay for a doctor, that’s your tough luck.


The only arguments of principle available (and that is what you need in court) rested on a categorical rejection of redistribution—what I have elsewhere called Tough Luck Libertarianism. If you are too poor to buy medical treatment or health insurance, tough luck: you can’t get something in a market unless you pay for it.

This framing is nonsense, because every system imaginable has to say "tough luck" some times. What if you ignored the mandate to acquire coverage, and you get sick or injured? Tough luck. What if, as was the case in Oregon (and the premise for running a randomized experiment), there aren't enough slots when you sign up for Medicaid? Tough luck. What if bad policy has made healthcare needlessly expensive, and you get stuck with a gigantic bill? Tough luck. What if you're struck with a terminal illness and there's no cure at any price? Tough luck. With the "tough luck" taunt directed at us, we're being invited to imagine there's an imminent problem with a solution that will work with absolute certainty. The problem is lack of access to affordable medicine, the solution is some government transfer program. But the government "solution" can fail for any number of reasons. The voting public underfunds the social safety net. (Tough luck!) Or the government declines to pay for certain treatments that would benefit the patient, because it fails some cost-benefit test (rightly or wrongly, tough luck!). Or the government nationalizes the hospitals, but the system gets captured by public sector unions that are more interested in feathering their nests than providing quality healthcare. (Tough luck!) We have to resist the notion of government as a lever of control over social problems, where the only obstacle to solving them is the lack of political will to pull that lever. There is simply no mechanism for wresting control of social problems. 

(Bryan Caplan made this point more eloquently about ten years ago in a post titled "Tough Luck." The late Tibor Machan made a similar point in this video, after some intellectual springs a lame "gotcha" question on him.)

None of this is to concede Koppelman's hypothetical. He's trying to get us to imagine that if you can't afford medicine in a free market system, you simply go without and you die (or otherwise suffer some horrible fate). That generally isn't true. If you require emergency care, hospitals have to provide it for you whether you can pay or not. They may take you to collection to recover some of their costs. But if you collapse from a blockage in your heart, you will get an angioplasty without anyone checking your assets or creditworthiness. That is, you'll get better medical care today if you're penniless than the President of the United States got when a heart attack struck him down (Eisenhower) in 1955. Hospitals are required by law to help, but that's not the only reason the indigent would get free or affordable care. The book Ensuring America's Health by Christy Ford Chapin describes how medicine was practiced in the early and middle parts of the 20th century, prior to the era of widespread health insurance. Doctors had sliding fees that took into account the patient's ability to pay. They would often give out free care, possibly out of a sense of professional duty and possibly as a tax write-off. (There can be both selfish and unselfish motives at play.) They would make up for their losses by charging higher rates to their high income patients (who were probably easily identifiable as such, possibly they were country club buddies with the doctors in the community). This was an era of free market social welfare and redistribution in medicine. I should also bring up the various ways people raise funds when they come upon a sudden illness. Often their friends in the community will organize some kind of collection. I've seen numerous fliers saying, "Let's help out Ben!" or something like it. Sometimes the sick person appeals directly to GoFundMe. That's not to claim this is a totally reliable solution, but of course neither is depending on the government. If a society is full of selfish, uncharitable people, then that society's welfare state will tend to be underfunded. 

I feel like I should say something about the source of libertarian objections to the welfare state. It's almost never about the money. Very few libertarians are fretting about the tiny portion of their tax bill that is extracted to pay for government services for the poor. (Where I have heard resentment about this expense, it's usually from lower middle-class laborers. They resent having to work for a living while other people live as well as they do without putting in the labor. Even here, it's more about the unfairness than the expense.) Almost none of the policy analysis I have seen comes from a place of "I don't want to pay for poor people's medicine." The more common libertarian objection to the welfare state is summed up nicely by a recent post by John Cochrane: generous welfare payments that steeply drop off as you earn more creates very high marginal tax rates on the poor. Sometimes the implicit marginal tax rate is 100% or higher, meaning your final take home income (post-tax and post-transfer) is unchanged or even decreases if you earn an extra dollar. This can disrupt a person's natural career trajectory. "Why work more if I don't earn more?" But this is short-termism. If this person worked diligently at some profession, even starting at an entry level job, they could eventually climb into the upper income quantiles. I don't oppose these programs because they're a bad deal for me. I think they're a bad deal for the recipients. 

Likewise, most libertarians who have strong opinions about socialized medicine aren't simply saying, "Give me my money back" or "I don't want to pay for poor people's medicine." Usually they'll argue that the current system makes things worse even for poor people. We've regulated and subsidized medicine to the point where it's unaffordable, but that's a self-inflicted wound. It's not an immutable feature of the world. A libertarian purist might insist that it's always wrong to take money by force to provide for someone else's welfare. But if we're listing the things the government does in rank order of how bad they are, this outlay set aside for the indigent would be pretty far down on the list. (The welfare state is vastly more expensive than it needs to be because it doesn't merely redistribute to the poor. We mostly redistribute from the young to the old, without account of need. If we stuck to providing for the needy, the welfare state would be a rounding error in the budget, and we'd barely need to have this argument.)

Libertarians could be wrong in any of the particulars. Maybe we're wrong about how a free market in medicine would play out, and maybe we're overestimating the disincentive to work produced by the welfare state. My point is that Koppelman's book is not really engaging with libertarianism! He's swatting away silly arguments by a few high-profile libertarian authors who subscribe to extreme deontology. To be fair, those authors have some dedicated fans. There are some bullet-biting followers who swallow these libertarian gurus' dogma. They can engage with Koppelman's argument if they like. But that conversation would be a minor and unimportant side-show. Most libertarians aren't lashing themselves to the mast of unquestionable moral premises then following them off a cliff. Most are happy to talk about why government programs are dysfunctional from a practical standpoint, fail to pass cost-benefit tests, and have unintended (though predictable!) consequences. Contra Koppelman, "tough luck" isn't even in most libertarians' vocabulary. This was another theme that utterly failed to land. 

Freedoms and Rights

Koppelman uses a kind of argument that I find extremely distasteful. It sneaks in government paternalism and regulation using the language of "rights". That is, if I can artfully, surgically constrain your option set, I can actually make you freer rather than less free. After a discussion of auto safety regulations requiring back-up cameras, he writes the following:

Freedom isn’t the absence of regulation. It is the capacity to construct one’s life as one likes. That means the capacity to focus on what matters to us, in reasonable security that we will not be blindsided by all the matters we inevitably neglect. Some paternalistic intervention is merely a collective delegation of certain decisions to the state, in the way that one delegates responsibilities to any professional. I’m not competent to decide what toxins are unsafe to inhale at the workplace, but I’m sure that I don’t want to leave that decision to my boss. Interference with my liberty of contract makes me freer.

Emphasis mine. This kind of thinking proves too much. It always has some nasty implications that the speaker does not intend. It recasts "freedom" as some kind of optimization problem where wise technocrats make you better off by fiddling with your option set (by banning and taxing things), rather than the more obvious definition whereby "freedom" simply means leaving people unmolested. If I were a religious fanatic, I could argue that government regulation of religion expands your freedom. If you are "free" to follow the wrong religion, you will damn your soul for all of eternity. A trivial amount of force for a finite amount of time in the physical world would preserve an eternity of freedom for all those would-be lost souls. Likewise, couldn't the government make you freer by restricting the kinds of ideas you are exposed to? Shouldn't a government that's concerned with optimizing your freedom take a keen interest in the books you're allowed to read or the arguments you're able to hear? You won't be very free if you're exposed to Covid misinformation and end up dead because of it. Of course, these are contemptible arguments. But we only recognize them as such because we've internalized ideas like freedom of religion and freedom of the press. For some reason there is a double standard when it comes to anything involving market transactions. You are free to damn your soul for all of eternity and speak contemptible speech. Our devotion to rights is strong enough that we're willing to let each other make those mistakes. But god forbid you make a mistake on your employment contract or consumer purchase. Our freedom to interact with our fellow creatures is basically untrammeled with respect to speech, religion, and consensual romantic relationships, even though we can and do make terrible decisions in these domains. There is no reason to nullify this basic understanding of freedom when money changes hands, as if such human interactions were less worthy of our respect. 

Implicit in this double-standard is the notion that "bigness" necessarily implies power. Employers and the sellers of consumer products are large and well-financed and can run roughshod over their employees and customers, the thinking goes. Koppelman makes this explicit (emphasis mine):

 Consider arbitration provisions in contracts, which typically require the economically weak contracting party—the consumer or employee—to waive access to the courts for disputes that arise out of the contract. Instead, such disputes are remitted to a forum designed to the specifications of the stronger party, which drafted the provision. Should such waivers of legal rights be permitted?

This notion of "bigness implies power" basically has no merit, not if you give even a moment's thought to how consumers and employees make decisions. I think the best dispelling of this idea was given in Private Governance by Edward Peter Stringham. He entertains the idea of Amazon (or a similar vendor) slanting the rules, say by declining to fulfill some random orders and then adjudicating all disputes in its own favor. (Koppelman is strongly implying that "bigness" grants large entities this power. He basically comes out and says as much in the quote above.) Here's an excerpt from Private Governance:

Even if one party is larger and offers the terms of adjudication on a take-it-or-leave-it basis, the requirement for a compromissum creates incentives to maximize joint benefit and minimize potential problems such as bias or waste. Suppose, for example, that buyers on eBay accurately predict a 10 percent chance of the seller not delivering and that eBay adjudication always found in favor of these delinquent sellers (and suppose for the moment this 90 to 10 percent ratio is stable). Here a $ 100 tie would be sold only for $ 90 (or less if we add risk aversion), meaning the cost of any proseller bias would be borne by the seller. Whether the sellers are big corporations and the buyers are little guys is irrelevant. Rules that harm buyers negatively affect their willingness to pay for a transaction.

Whether we are talking about potential disputes between stockbrokers and clients, banks and customers, cell phone carriers and users, or landlords and tenants, attempting to stack the deck against one side does not pay when that party can adjust his willingness to enter an agreement. The cost of biased rules or procedures is borne by the party that the rules were intended to bias.

When you consider that the smaller (likely poorer and less diversified) party is likely risk-averse, the large party is probably paying more than $10 when it tries to statistically skim an expected $10 off each order. I think the worldview Koppelman subscribes to obliterates the agency of typical consumers and employers, who are actually quite savvy. Resource-poor consumers shop around for the best price and spurn companies with bad customer service or products that break unexpectedly. Workers are generally well aware of their employment options. They know about poor working conditions related to safety or discomfort, usually well before they join a firm. If anything (again because of risk aversion) they will tend to overweight rare but costly mishaps that are stacked in the employer's favor. Suppose one out of every 1,000 employees suffers a $100,000 injury. It seems like Koppelman is insisting that employers can make the unfortunate employee simply eat the cost (thus effectively skimming $100 off each employee, in expected value terms). In reality, such a scofflaw employer is likely to attract bad press, and its reputation of mistreating workers will certainly get out to potential employees. Being less attractive as an employer, it will need to pay its employees at least a $100 premium to join (relative to similar firms that don't cheat their workers), erasing any "gains" from cheating random workers out of a proper settlement. (Again, the cheating employer likely needs to pay more than a $100 premium because the individual workers are undiversified and thus risk averse.)

Again, we don't really need to get into the talk of "rights" or "freedom". The regulations requiring backup cameras have some fairly obvious unintended consequences, which could completely erase any value added. Such regulations mandating safety features (or fuel efficiency) make automobiles more expensive, which means car owners are likely to hold on to their older, less-safe (or less efficient) models for longer.  The regulation he's championing doesn't mandate retro-fitting older models in the existing fleet to add back-up cameras. This is the folly of optimizing in one dimension. Even without any government regulation, newer model vehicles tend to get safer and more fuel efficient. The mandate could well lead to more car-years of service of less safe automobiles (even if the newer models are slightly safer than they otherwise would be...and models with backup cameras would be available for people who want them at any rate). Koppelman keeps inviting us to think of libertarians as following simple moral principles off a cliff. No, a cold, rational technocrat with a basic understanding of economic reasoning will tend to reach libertarian conclusions, at least in the examples presented in Burning Down the House.**

CO2 and Climate Change

Koppelman is obviously interested in discussing pollution in this book, because it is the kind of problem that markets don't always handle well. For obvious kinds of pollution, such as dumping buckets of toxic chemicals onto your neighbor's property, there actually isn't much of a problem. Even in an anarchocapitalist society (or a society in which the legislature fails to take up the issue of environmental pollution), you have redress against your neighbor for such dumping. You can sue him in whatever judicial system exists in your society. More interesting is the topic if extremely diffuse pollution that doesn't have an obvious victim. Instead of dumping highly concentrated chemicals on someone's lawn, thus directly poisoning my neighbor's well water and killing him, say I instead evaporate the pollution and send it into the atmosphere. My pollution is now mildly affecting a much larger number of people (rather than obviously poisoning a single person). Maybe I'm still killing one person per unit of pollution under this diffuse method, it's just very hard to identify who that is because it shows up in, say, higher cancer rates. This is admittedly a hard case for free markets. Various free market solutions have been floated, but the obvious solution is to have some kind of society-governing body (the legislature, or maybe a judicial ruling) regulate the pollution. 

Granting all of that, I think he says some absurd things about CO2. Essentially nobody doubts that higher levels of CO2 have caused some warming over the past century and a half. But there are valid differences in opinion over how much of that warming is due to CO2, how harmful that warming is likely to be, how sensitive the climate is to CO2 levels, and how effective any proposed mitigation measure is likely to be. The accusation that someone is a "denier" is almost always an empty smear, directed at someone who takes issue with alarmism on any of the topics in the prior sentence. The boogeyman climate denier, who disputes the basic science of CO2 as a greenhouse gas, is basically not a thing. The book veers toward the extreme alarmist side of the spectrum when it comes to CO2 and climate change.

Koppelman says the following:

There are about $6 trillion in fossil fuel reserves. If all of those are used, the earth’s temperature will rise by 16 degrees Fahrenheit. The tolerable upper bound is 3.6 degrees. The largest increase that has happened since life appeared on earth was a 6-degree rise during the end-Permian extinction, 251 million years ago, which wiped out 95 percent of the species on the planet. The 16-degree rise won’t happen, because everyone would die first and the burning of fossil fuels would then cease.

This is absurd. Fossil fuels are the remains of plants that died hundreds of millions of years ago. They took carbon out of the atmosphere and turned it into their bodies. Then they died and were buried deep underground. Pressure and time turned these dead plants into coal, oil, and natural gas. All of that carbon was once part of the atmosphere. Koppelman is claiming that returning to this state is inconsistent with life, even though obviously that's the state of the atmosphere when life on Earth was thriving. (Is he talking about temperature deltas rather than actual temperature levels?) No doubt, an extremely rapid temperature rise would be very disruptive, and a large enough rise could end civilization. Certainly the species alive today aren't optimized for a very high CO2 atmosphere, as it was hundreds of millions of years ago. But almost nobody is actually projecting such a catastrophic change. 

CO2 levels rose from about 280 parts per million pre-industrial revolution to around 420 ppm today. The temperature response to carbon is logarithmic! That is, if doubling the amount of carbon dioxide gets you X degrees of warming, we must double that again to get 2X degrees of warming. CO2 levels would have to be growing at an exponential rate over time to get a steady (linear) increase in global temperatures. (That relationship is basic physics and is not in dispute, as far as I can tell.)

Anyway, I'll chalk this up to Koppelman being not very knowledgeable about climate science. That's no sin, it's not his specialty. He's likely taking his cues from the alarmist climate press rather than reading directly from the literature. But given that, he probably shouldn't be holding this up as an awesome example of market failure. He certainly shouldn't have any confidence that he's identified the solutions to such a market failure. 

Books That Supplement With Useful Context

I was fortunate to have read Alex Epstein's new book Fossil Future immediately after reading Burning Down the House. It's a useful inoculation against lazy climate doom saying, the authors of which expect their audience to uncritically accept their alarmist projections. In this vein, I will also recommend The Myth of American Inequality by John Early, Phil Gramm, and Robert Ekelund. Burning Down the House repeats a lot of stale bromides about how American income inequality is increasing, and how this implies our market system is inherently unfair. This is basically not true at all if you count all of the resources at a household's disposal rather than just counting their market income. That is, the lower quantiles have been getting more "generous" transfer payments, and the higher quantiles pay very high taxes. This results in a sever flattening of household incomes. To the people who gripe about income inequality, why would you not give the American system credit for what it's actually doing about the "problem"? Inequality-griping is always a lead-in to an argument for more  income redistribution. Why would you not take account for the current levels of redistribution? The climate and inequality stuff in Koppelman's book is a lazy repetition of leftist talking-points. Again, maybe I'm wrong about some of the particulars. The point is that Koppelman's book is attempting to engage with libertarianism. It could have dealt with more sophisticated versions of the climate and inequality discussions. The basic leftist talking points have been answered (at least responded to). Rather than respond to the responses, Koppelman simply repeats the step-1 talking points.***  

I should note that there's nothing particularly libertarian about one's opinion on the state of climate science or the proper way to measure inequality. These are technical topics, matters of fact, that stand on their own and should not be polluted by political philosophy. But libertarians end up focusing on these topics because statists use them as intellectual ammunition when building their case for various government programs. If the statists are routinely misstating facts about the world, if they are overstating various problems and generally catastrophizing, then they're going to attract a chorus of libertarians contradicting them, speaking as if in the same voice. 


I was very curious to read a serious critique of libertarianism. But this book fell flat. It told me some things I already knew, that there have been some nutty libertarian theorists who make terrible arguments. I know Rand and Rothbard made some silly arguments, because other libertarians are constantly critiquing them! Koppelman vastly overstates the influence of these authors on modern politics by finding occasional similarities in rhetoric or noting that some politician claims Rand or Hayek as an intellectual influence. It was a surprise to me that libertarians have all this political power, considering the vast and unceasing expansion in government over the past hundred years. 

If I'm trying to be constructive, I'd recommend a few changes for a second edition, or perhaps a sequel. Maybe even an entirely new book by a different author. For one thing, make it a little bit clearer what mistakes leftists are making. Koppelman is far too vague about the pro-market part of his ideology. He could identify a slate of bad government interventions that leftists love. He doesn't hesitate to wag his finger at bad libertarian arguments, but he seems shy about scolding the left for its mistakes. I suspect nothing he does here will satisfy me, because he is far too accepting of government regulation even when it's poorly justified. He basically swallows the left's "bigness is power" framing, even in cases where there is not an externality (e.g. the stuff about employers and companies slanting contracts in their favor). But he could at least disavow the most destructive or the most obviously bad market interventions: import quotas and tariffs, excessive zoning, rent control, occupational licensure, the Jones Act, etc. He could scold the left-aligned constituencies that favor these programs, like labor unions and local NIMBY homeowners. He could also give some examples of policies that the left likes, but which marginally fail a cost-benefit test (he gives some examples that go the other way). 

On the inequality and climate change stuff, again I feel like we should be engaging at Step 3 of the following interchange:

Step 1: Leftist claim/argument about (policy topic)

Step 2: Libertarian counter-argument about same policy topic.

Step 3: Leftist counter-argument to the libertarian counter-argument.

His book is mostly stuck on step 1. He does spend a lot of time engaging with libertarian arguments, to be fair. But again I feel like he's not engaging with the most current state of the response-counterresponse discourse. 

In my opinion, the intellectual center of the current libertarian movement isn't Rand, Rothbard, and Nozick. Richard Epstein and Randy Barnett are interesting figures, but they don't jump out at me as influencing the movement in a significant way. David Friedman, who gets a single mention in Koppelman's book (in a rebuke to Rothbard's environmentalism), should get center-stage in a proper treatment. See his debate with George Smith on rights-based vs. economic arguments for libertarianism. (Again, we libertarians don't need Koppelman to point out the silliness of extreme deontology. There is a lively debate about this within the movement.) Or see his debate with Bob Murphy on Austrian versus Chicago school economics. 

The econ-libertarian blogosphere and the George Mason University economists barely get mentioned. A second book could correct this omission by grappling with some of these titans of libertarian thought. When I think of my libertarian influences, near the top of the list are these excellent scholars. Bryan Caplan on immigration, education, and voter rationality; Tyler Cowen on anything (and he often comes to non-libertarian conclusions); Alex Tabarrok on pharmaceutical regulation and policing; Garrett Jones on IQ, governance, and macroeconomics; Don Boudreaux on applying basic economics to policy analysis; Peter Leeson on pirates, anarchy, and private/cultural systems of governance; Robin Hanson on medicine, prediction markets, and VERY BIG QUESTIONS! There are also non-GMU economists who are adjacent. Steven Landsburg or David Henderson or Arnold Kling on foundational economics applied to policy. Scott Sumner on macroeconomics. Mike Munger on public choice theory. Jeff Miron on drug policy. Jonathan Meer on minimum wages. Russ Roberts (formerly of GMU) of Phil Magness on anything. Then there is the Cato crowd. Alex Nowrasteh on immigration policy. Jeffrey Singer on drug policy and medicine. Michael Cannon on healthcare. 

Moving a step back, we could include the Chicago school in this pantheon. Koppelman does include several of them in his book. Milton Friedman (mentioned), Gary Becker (mentioned and quoted), and George Stigler (not mentioned). There is also the generation of scholars who were students of these giants, people like Deirdre McCloskey and Thomas Sowell, who have interesting and distinct arguments. Read the works of these scholars and get a feel for the texture of their arguments. They don't start off with "Since we all know government is bad..." or "Applying the non-aggression principle...". The pro-market punchlines of their arguments are conclusions, not premises. Nor is "undercut the case for government" an animating motive for these intellectuals. (Sowell and McCloskey both admit to starting their intellectual lives as extreme leftist, Sowell having once been a Marxist. And Milton Friedman was a Keynesian who became a free market advocate as he matured.) 

This is where Koppelman's entire project fails. I get the sense that he was hoping he could knock out a few simple foundational arguments and the entire edifice of libertarianism would come crumbling down. But modern libertarianism is more eclectic. It would be almost impossible to write a single book "correcting" libertarianism by responding to arguments of all the thinkers mentioned above. There are individual policy critiques that stand on their own, with no obvious through-line or libertarian principle uniting them. Such analyses don't start off assuming some weird Rothbardian or Randian moral premise. They often still end up reaching the libertarian conclusion. 


*Here is Robin Hanson's classic Cato Unbound post titled Cut Medicine In Half. He gives a good rundown of the literature at the time (it was written in 2007). He has three hostile intellectuals responding to him with essays of their own in this Cato forum. In my opinion, Hanson gets the better of them, and none bother to seriously engage with his point. 

**Here is Koppelman early in the book, expressing his discomfort with "rights":

The standard defense of gay rights was simple: people should have a right to do whatever they want as long as they don’t violate anyone else’s rights. I thought it was too simple. The principle had two embarrassing implications. First, it meant that the law could impose no restrictions on self-destructive conduct: corner stores would have to be allowed to sell heroin and methamphetamine. Second, it undercut all of antidiscrimination law, which forces people into transactions that they would prefer to avoid. The world was just more complicated than that. So I was led to defend gay rights on an entirely different basis. (I argued, and still think, that discrimination against gay people is a form of sex discrimination.) 

Casting gay rights as a basic form of human freedoms, as in the freedom to transact and interact with your fellow creatures any way you like so long as it is mutually agreed upon, is a more obvious framing than putting it under the more modern umbrella of anti-discrimination law. Koppelman's objections to this way of thinking are notable. He appears to be objecting to legalizing drugs, which would put him on the side of one of America's worst, most destructive domestic policies. He makes it clear much later in the book (and in several podcasts) that he's no drug warrior. (He later states: "America’s war on drugs has been so destructive that there is a serious case to be made for complete legalization.") It was simply jarring to me that this is what snapped him out of embracing rights language? Libertarians have diverging opinions on antidiscrimination law. Some of us are maximalists when it comes to the right to associate, which necessarily implies the right to not associate. Even if it's for a dumb reason, like disliking someone's race or wanting to create an exclusive boys' club. At the same time, even those people will usually say antidiscrimination law is not the hill they're going to die on. It's way down on the list of "bad things the government does," and repealing such laws is not a priority. Other libertarians positively embrace antidiscrimination law. And some have argued there is pre-existing common law (so preceding and not dependent on anti-discrimination legislation) that compels businesses to serve all comers. As in, putting out your shingle and advertising your services to the public is a kind of implicit contract, which you can't simply rescind because you don't like someone's face. I find this common-law argument persuasive.

***I am confused about Koppelman's position on inequality. At one point he writes the following:

The fundamental error of the anticapitalist left is focusing on inequality when its most pressing concern should be poverty. Effectively addressing poverty will require acceptance of substantial, undeserved inequality. Attacks on capitalism foolishly hurt the very people they are intended to help.

Amen, brother! But he then repeats all the standard talking points about how the middle income quantiles haven't seen their incomes increase in fifty years, how the top quantiles have been taking all the gains, etc. There isn't necessarily a hard contradiction here, but there is a tension. 

Wednesday, December 28, 2022

Drug Poisoning Update for 2021 Data

I haven't done one of these in almost three years, but I wanted to do an update. Here is my most recent one with data through 2018, also with links to prior years' analyses. I think not much has changed about the overall narrative, so I'll direct readers to the previous years' posts for those details.

The CDC's large mortality file gets posted here, in case anyone wants to download and do their own data mining. 

There were 107,027 drug-related deaths in 2021. This includes 98,722 accidents, 4,239 suicides, 3,852 incidents of "undetermined intent", and 214 murders. (Not included in the 107,027 figure is 7,017 "other" deaths, where drugs were mentioned on the death certificate but weren't listed as the immediate cause of death. Some of these really are caused by drug use, so presumably they should count for anyone who's concerned about drug-related public health issues.) In the past I've made a big deal about the difference between the total and accidental deaths. Sensationalist journalists reach for the largest number they can find, so they cite the total. This will often be in the context of an "opioid epidemic" story. But it doesn't make much sense to blame suicides on the opioid epidemic. A suicide is a death that would most likely have happened anyway, just by another means. At any rate, as the years go on the "accident" category is becoming such an overwhelming majority that it feels a little pedantic to make this distinction. Still, bonus points for honesty to any journalists who cite the 98k figure rather than the 107k figure. Here's a chart of the proportion of deaths by the various categories over time, where you can see "accidental" swallowing up a greater and greater share of the total:

I do suspect a lot of these "accidental" overdose deaths have a question mark over them. A lot of these people are taking such tremendous risks (like shooting up alone with a powder of unknown origin and potency) that it's worth asking whether it should count as an attempted suicide. But the best I can do is analyze the data we have and caveat with my personal suspicions. 

Above is a chart showing the deaths by major substance categories. The pink line that dominates in the most recent years is "synthetic narcotics", which mostly refers to fentanyl being sold as a street drug. Fentanyl has largely replaced heroin as the major opioid of abuse. This view can be slightly misleading in the following sense: most drug poisonings are multi-drug interactions. So if one category of drugs is skyrocketing, it can drive up the apparent deadliness of other drugs (even if the risk or abuse rates of, say, cocaine or benzodiazepines are basically unchanged). This is actually my value-added. Because this CDC file is at the individual record level, I can count, say, cocaine deaths excluding fentanyl and heroin. I think what looks like a general trend of increasing drug deadliness is in reality almost entirely a fentanyl story. 

Below I have created a similar time series, but this time I'm splitting each substance into two lines. The first line is unconditional count, and it should match the chart above. The second (necessarily lower) line excludes deaths involving heroin or fentanyl. The normal green line is cocaine deaths, while the pukish-green line is cocaine deaths excluding those that also involved heroin or fentanyl. The first looks like it's taking off, the second looks like it's relatively flat. This is clearly a case of fentanyl driving up cocaine-related deaths. Presumably in most of these deaths, the cocaine was either not a factor or not sufficient to kill without the added effect of a potent opioid. 

For clarity (the above chart is too busy to read clearly), here are prescription opioids inclusive and exclusive of heroin/fentanyl. The dishonest way to report on this is to look at the raw number and say "prescription opioid deaths are still rising." A more honest assessment is that prescription opioid deaths are falling, but the extremely high mortality of illicit fentanyl users is spuriously driving up the number of death certificates that mention prescription opioids. There are fentanyl users who happen to have prescription pills on them when they fatally overdose, and there are people who abstain from fentanyl but who fatally overdose on prescription pills. The non-overlap in these categories explains the huge gap between the two lines.

Here's benzodiazepines. Almost all benzodiazepine-related deaths involve some other substance, typically an opioid but sometimes alcohol or some other prescription medication. This is once again a story of fentanyl driving up the total.

Here are "psychostimulants with abuse potential", which mostly means methamphetamine. I think I have heard some commentators suggest that meth is on the rise again, which possibly exonerates opioids. The rising rates of meth use, this narrative goes, are simply a continuation of a longstanding trend of increasing overdose rates, which extends back to the 1970s (possibly earlier). Meth just happens to be in vogue again. This appears not to be the case. Undoubtedly the red line is increasing dramatically, but in the most recent year the blue line is almost three times as high as the red line. Pure meth-related mortality is several times higher than it was twenty years ago, but at the same time the vast majority of meth-related drug poisonings also involved fentanyl or heroin. 

By the way, heroin deaths seem to have reached a peak in 2017, at 14,870. By 2021 they were down to 9,064. So overwhelmingly the most fatal street drug is fentanyl, which appears to have almost totally taken over the heroin market. 

If you want more information about which drugs in combination are killing people, here is a cross table. (To read the lower table, take the second column of the first row. Read this as "22.48% of heroin deaths also involved cocaine.") 

The elephant in the room here is that I've added the Covid years to the analysis. My previous update went through 2018, where drug-related deaths appeared to be leveling off. Well, 2019 looks like a clone of 2017, if not slightly worse. 2020 and 2021 both saw large increases over the prior year. I don't know if the "deaths of despair" narrative makes sense here, or if we're just seeing a continuation of that longstanding trend reaching back to the 1970s. Shouldn't despair have been worse in 2020, when lockdowns were harsher, society was more closed, and drug users would have been more likely to shoot up alone (thus not having anyone who could call for help if they overdosed)? Or was there some kind of accumulation of despair from 2020 to 2021? That is, even though 2021 was milder in terms of social isolation, that isolation was still wearing people down and driving them to risky drug use. 

I have limited patience for the notion that drug-related mortality can exponentially increase for multiple decades. Obviously this can't go on forever. I haven't done a population dynamics calculations to check this, but at some point you'd start running out of drug users. A little Googling suggests there are around a million heroin users in the United States. (Maybe double or triple that if you think most heroin users are homeless and out of reach of household surveys of drug use). If these are the people being killed by fentanyl, then with the numbers we're seeing it would only take a few years for them to all die off. Even with new addicts refreshing their ranks, the mortality is high enough that some depletion should be happening. And these very high mortality rates should be scaring off a lot of potential new users. (Presumably that's happening, but some countervailing force is attracting them?) We should probably take seriously the notion that a large number of deaths are misclassified. Does anyone have a story for why methamphetamine-related deaths, excluding those involving heroin/fentanyl, have increased 30-fold since 1999? Were we perhaps undercounting back then and/or are we overcounting now? I don't want to selectively dismiss or down-weight government data, but at the same time this kind of dramatic increase makes me really suspicious. My attempt to find some pattern in all-cause mortality or a redistribution from one cause-of-death category to another was pretty inconclusive, but that's not to say a deeper dive wouldn't turn up something interesting. 

How to Means Test Social Security and Medicare

Social Security and Medicare make little sense as anti-poverty programs. They pay out based on age, not need. A retiree who earned a respectable income over their lifetime shouldn't need Social Security to provide for their retirement. They had the means to save back enough to cover their retirement years. Such a person could easily have a million or so dollars saved up.

The obvious way to means test observe your means. Look at the retiree's assets, then apply some kind of threshold at which point benefits are subtracted. (More likely a sliding scale, such that holding an extra dollar doesn't magically cause thousands of dollars worth of annual payments to disappear.) This may seem superficially fair, because such means testing would avoid the bad optics of making government welfare payments to millionaires and billionaires. But there's an obvious moral hazard problem. If saving less results in a higher Social Security income, people will be encouraged to save too little for retirement. Some people save back nothing for retirement, leaning on the lazy assumption that Social Security will be adequate to cover living expenses. (It's not for most retirees.) Means testing based on current assets would exacerbate this problem. Some fools would deliberately aim for zero assets to literally maximize Social Security. That's an extreme case, but even normal, rational people would be nudged in the direction of inadequate savings.

The smart way to means test is to count total lifetime income, not current assets at the time of retirement. Two people who have had identical incomes over their lifetime have had roughly equal opportunities to save. If one saves frugally and the other saves back nothing, we shouldn't feel compelled to help out the latter. Both people had the same lifetime means, but they made different choices for what to do with their earnings. There still may be some frugal individuals with modest incomes who retire millionaires. They may get Social Security payments because their lifetime income was low, even though they don't technically need it. And there will be people with high lifetime earnings but inadequate savings who get nothing. Hopefully, the government makes a credible commitment to not bail out these people. The best scenario is that they get the message early in their working lives that nobody is coming to save them, and that they should set aside what they'll need for their own retirement. 

This scheme mostly avoids the moral hazard problem. There might be some people who deliberately earn less over their lifetimes to game the system, but a smartly designed payment scheme would avoid any such gaming. There should be no "cliffs", whereby earning one extra dollar causes thousands of dollars of Social Security payments to disappear. And there should be no point on the sliding scale where earning a dollar in income loses you more than a dollar in Social Security payments. Social Security payments should be such a gentle slope with respect to lifetime income that there is no temptation to earn less in the marketplace. This should be easy enough to do if we think of it as a program to avoid indigence in old age, rather than a program to fund the retirements of middle-income earners. A minimalist income floor can be provided without distorting the incentives to earn and save over a lifetime. 

Medicare is a little more complicated, because it's not just an entitlement program. It vastly distorts the market for medicine by being one of the largest single payers. We would need to have pro-market reforms, whereby people get comfortable paying out of pocket for most of their medical expenses rather than having third parties pay for almost everything. Patient price shopping and transparent pricing would bring costs down and eliminate a lot of wasteful spending. But given that, the same arguments would apply. 

I think there are a couple of major barriers to this kind of reform, one ideological and one selfish (both are cynical). The selfish one is that current and near-future retirees don't want to see anything happen to a government program that they're planning to draw an income from. I think the kind of means testing I describe here wouldn't impoverish many retirees, because people with modest or high incomes typically save for retirement (beyond what they expect to collect from Social Security). Still, it would be a little unfair to change the deal on people who are in or near retirement. So a more palatable (you could even say fairer) plan would be to start the means testing on people who will be retiring more than, say ten years in the future. Means testing could be on a gradient that scales up over time, so people retiring in ten years could get essentially the same benefits as current retirees while people retiring 25 or 30 years in the future could get the full effect of means-testing. This solution is so obvious that the "It's unfair to current retirees" argument carries no water at all. We should be asking, "What's the optimal state of this welfare program? What does the transition look like if we have decades or a century to ease into it?" The specter of current retirees being left in the cold, having their incomes revoked without warning, is nothing but a scare tactic. 

The ideological opposition comes from the left's knee-jerk love of entitlement programs. They prefer to have a broad base of recipients for these programs so it becomes difficult to reform them, even if that means throwing government money at people who don't need it. They want everyone to have "buy-in" so the whole population becomes part of the program's constituency. When politicians start talking about fiscal responsibility and reigning in the unsustainable spending of Social Security and Medicare, they want boomer-cons and senior Trump voters to fight back saying, "Get your government hands off my Social Security." This dynamic serves up some delicious hypocrisy, but it also makes government transfer programs politically untouchable. It's a dishonest way to sell something to the public. If transfer programs that specifically target the needy are politically unpalatable, if the American public wouldn't vote for them or would underfund them because of a lack of buy-in, then we shouldn't have them in their current incarnation. But it looks like we're stuck with them for the long term. 

Sunday, November 27, 2022

Covid-19 Deaths In 2020 In the United States

In 2020 there were 385,293 deaths that mentioned Covid-19 on the certificate. Of those, 351,530 listed Covid-19 as the underlying cause of death. (In other words, 33,763 deaths, about 9%, listed Covid-19 as incidental but not necessarily the ultimate cause.) I wanted to take a closer look, because this seemed lopsided to me. Quick overview, the CDC's mortality data list an ultimate cause of death, the single thing that is fingered as the reason the person died. It also lists up to 20 (though typically only three to five) contributing causes of death, things that are relevant but not deemed the ultimate cause. So if someone has a chronic lung condition and dies after contracting covid, most likely the chronic condition will be listed as a contributing cause with "Covid-19" listed as the ultimate cause. 

If you've followed my blog for a while, you've seen my analyses of the CDC's statistics on drug-related mortality. One pattern I've noticed is that if a drug is mentioned, it's overwhelmingly (>90% of the time) listed as the ultimate cause of death. At the same time, you see a lot of chronic illnesses getting listed on these records as contributing causes of death. Is there some bias whereby if drugs are anywhere in the picture they get fingered as the ultimate cause? Or are medical examiners assiduously avoiding any mention of drugs if they don't think they're relevant to the causal story? (While at the same time promiscuously reciting details about the decedents medical history in drug cases?) 

I had a similar thought about the 2020 covid deaths, so I took a closer look. What does a "covid caused" death look like compared to a "covid mention" death (in which presumably covid was incidental)? Early in the pandemic, I suspected that there was some kind of misclassification happening, where covid was blamed even if it was incidental. Such a bias in reporting would inflate the death toll. The excess death statistics quickly ruled that out as a plausible explanation, because those are a measure of total overall mortality and it's implausible that those numbers could be inaccurate. But it could still be true that there is a significant amount of misclassification happening. Excess deaths don't exactly overlap with covid deaths, so there's room for error here. 

After looking into the data, my takeaway is that the deaths really do look like covid deaths. I don't have the actual death certificates, but I do have the large mortality file from 2020 (the Multiple Mortality Cause Files in this link). With this I can see which conditions are listed on a given record. I see a lot of conditions such as: pneumonia, chronic obstructive pulmonary disease, hypertensive heart disease without congestive heart failure, adult respiratory distress syndrome, cardiac arrest, respiratory failure unspecified, etc.  A typical record will list multiple conditions. The average age of decedent is 76. Taken together, this corroborates the story that this disease mostly struck down the sick and the elderly. 

I actually didn't see a huge difference between the records that were marked as covid being the underlying cause versus the ones where some other cause was marked as primary. They are listing the same set of conditions, and the age distributions are similar. If you look at the underlying cause of death for the "covid mentions" records, you see a lot of chronic conditions. Malignant neoplasms (cancer) of the prostate, Parkinson's disease, Alcoholic hepatitis, Other Obesity, Heart failure, Unspecified dementia, etc. You see a lot of these listed as contributing causes of death on the records for which covid was selected as the ultimate cause.  I don't know exactly by what rubric a death is ruled to be "caused" by covid rather than having it listed as a contributing condition. Presumably someone is making that ruling, but these things can be a little arcane. 

I think there's more work to do here, particularly when the 2021 file comes out (which should happen near the end of 2022). There should be a more thorough analysis comparing the official counts of covid deaths to the excess mortality numbers to see if they match up or not. If they don't match fairly closely, it could imply that there is excess mortality (or less than expected mortality) due to some of the lockdown measures. Presumably the disruption of medical services had an effect on some people getting necessary healthcare, and it seems that suicides and "deaths of despair" increased during the past two and a half years. Excess deaths for younger demographics far exceed the official covid counts. It's not plausible that we're under-reporting covid deaths in these demographics by such an extreme factor. The more likely story is that deaths of despair really are taking a bite. Given my review of the CDC data and the close match between the official covid counts and "excess mortality" estimates for 2020, I would guess that there isn't a serious misclassification problem for that year. But there is more to do when the 2021 data becomes available. It's possible that excess deaths due to deaths of despair and disrupted healthcare take a bigger bite later in the pandemic, but these could be concurrent with a reporting bias that over-attributes deaths to covid. Someone could look at covid deaths versus "excess mortality" by month of the year or age or race to see if there are serious mismatches, which could suggest a reporting bias. (See the link above, excess mortality in the younger demographics far exceeds the official covid count, so obviously it's not straightforward to say "excess mortality can all be attributed to covid" as some careless commentators have done.) Overall I'm fairly confident that the official counts aren't wildly off. There have probably been around 1 million covid-related deaths in the US since the start of the pandemic. But it's worth checking. Official narratives have been wrong before. 


I think we have to start getting really sophisticated about what "excess deaths" means in the current environment. Has anyone projected out what's expected to happen in coming years? Didn't covid kill off a lot of vulnerable people, who would have died anyway within the next few years? Won't this cause overall mortality rates to fall in future years? Are "long covid" or disrupted healthcare or deaths of despair taking enough of a bite that it's overwhelming this expected decline in mortality? I think someone would have to build a reasonably sophisticated statistical model to do this, accounting for mortality at various ages by various conditions, aging the population to project out to future years. "Excess mortality" requires having an expected baseline to measure the actual deaths against. It's probably valid to compare mortality in 2019 to 2020 or 2021 and call the difference "excess mortality", but the further you go into the future the more you have to adjust the baseline and the more assumptions you need to build in. 

Wednesday, November 23, 2022

Burning Down the House

If you're a libertarian who wants to read something by a progressive that attempts to convert you to progressivism, you should read Burning Down the House by Andrew Koppelman. Alternatively, if you're a progressive who wants to feel an unearned sense of intellectual and moral superiority to libertarians, you should also read Burning Down the House. The book's subtitle is How Libertarian Philosophy Was Corrupted by Delusion and Greed. This is provocative, because it teases the idea of a political philosophy that once made sense in the good ole days, but whose current incarnation is a betrayal of its roots. Modern libertarians are doing it wrong apparently and are foolishly misinterpreting the philosophy's founding documents. While the title and intro hint at this, it never really delivers.

Koppelman frames himself as a pro-market liberal. He's a nearly-extinct species of libertarian who is aware of the flaws of a market economy and the need of government intervention. It was a frustrating book to read, because it all comes off as standard progressivism. The only clue that Koppelman's philosophy is favorable to free markets is his own say-so. He offers few if any examples of favoring pro-market reforms. I heard him concede in a podcast, though not in the book, that the elimination of the Civil Aeronautics Board and deregulation of interstate trucking in the 70s were good. It would have been helpful for him to expound upon this in his book so we know where he positions himself on the spectrum of being "pro-market". His discussion of income inequality is just standard leftism. I was expecting a mature discussion of the reasons for income inequality, or perhaps a neutral, reasonable explanation of why inequality increased in recent decades. No such luck. He offers some statistics that demonstrate income inequality has increased in recent years, but with no argument whatsoever that the increase was unreasonable or that any identifiable group was being undervalued by the market. He spends a lot of time discussing John Rawls and and the Rawlsian "veil of ignorance". (See this piece highlighting some of Rawls's basic logical errors. Despite his popularity, his philosophy seems a bit confused.) Unfortunately, there's little connection between the Rawlsian philosophy and the modern income statistics. (Like, is any increase in inequality always bad? How do I know that the recent increase is too extreme? Or how can I determine what the right level of inequality is so I'll know when society has achieve justice? Koppelman doesn't really tell us.) That is the general theme of the book. He presents a theoretical argument for government regulation, but fails to justify government in actual practice. (He asserts that certain regulations are cost-benefit justified without really making his case, in my opinion. Show you're work if you want to be taken seriously by critics.) It might have been useful for him to contrast himself with standard progressives by giving some examples of markets being positive, examples which progressives would actually disagree with. Or he might have pointed to some period in history when inequality increased but did so for perfectly justified and understandable reasons. (Something like, "Such-and-such era also saw an increase in inequality, but in this case it was because some people chose to work longer hours while others adopted more leisure. Also, this was an age of genuinely beneficial innovations, and it was socially just to reward the innovators..." This would have been a nice contrast to the discussion in his book, which seems to have an implicit premise that any time inequality increases, it's unjustified.) 

The book seethes with an element of  Look at all these libertarian policies doing so much harm in the world! As a libertarian, it's pretty irksome to be catching blame for bad policy outcomes when we don't wield any power. Contra Koppelman, we haven't exactly been getting our way. Our critics grossly overstate our influence on policy. They somehow miss the fact that government has been growing in size and scope (contrary to the policy prescriptions of any prominent libertarian). It reminds me of the drug policy commentators who apparently think we've been living in a libertarian regime for the past several decades, and therefore recent trends in drug overdoses can be blamed on libertarian-influenced permissive drug policy. As someone who knows what the recent drug regime looks like and what the various libertarian policy prescriptions are, I can assure you we haven't recently lived through a libertarian drug policy utopia. This notion that we libertarians are getting our way with policy,and thus society's ills can be blamed on us, is pretty absurd. 

The book's title is a reference to an incident in which a fire department allowed a house to burn down because the owner failed to pay a $75 annual fee. Koppelman's argument bizarrely tries to pin this on the rise of libertarian ideology. In his interview with Koppelman, Reason's Nick Gillespie politely eviscerates this narrative of the event. Gillespie points out that the fire department was not a private service, but instead was a neighboring city's department which contracts with nearby municipalities that don't have their own departments. This is hardly an example of private markets going wild. It sounds more like an extension of the DMV's model of "customer service" applied to government-run emergency services. Gillespie points out that in similar such instances, the private fire department typically puts out the fire and charges the homeowner some reasonable multiple of the original fee. (Some will be tempted to cry foul on this kind of surcharge. But if there weren't any surcharge for non-payment, then everyone would simply decline to pay until their house is on fire, so there would be no funding and thus no fire department.) Private businesses that provide emergency services are very concerned about public perception and bad PR, but a municipal agency pretty much has carte blanche to let someone's house burn down and say, "I don't really give a shit. Not my jurisdiction." I think Koppelman is basically 180 degrees wrong to pin this on libertarianism. This story is an example of the callous indifference of a government bureaucracy, and Gillespie has the right model of how private companies work in similar situations. In Koppelman's defense, he might not be claiming that the fire chief was a devoted Rothbardian/Randian/Reaganite libertarian, applying market fundamentalism to his profession. There was probably no causal connection between libertarian philosophy and the actual event, which would make Koppelman's fixation on it quite strange. He makes the connection more obliquely. Koppelman points to a few prominent libertarian-adjacent commentators (including Glen Beck) who argued in defense of the decision to not fight the fire. This is a little strange, to say the least. The titular event of the book was not caused by libertarian ideology, but because some vaguely libertarian-leaning commentators discussed it after the fact, he can blame us for it? 

I'm a fairly well-read libertarian. I don't think of Glen Beck as being a particularly representative example of this clan. He almost never comes up in the libertarian blogs or books I read or the podcasts I listen to. He may self-describe as libertarian, but comes off as more conservative. Koppelman also lists Kevin Williamson and Jonah Goldberg as having commented on the event, both writing in National Review Online. It almost seems like Koppelman couldn't find any actual libertarians making the argument he wanted to tag us with, so he instead turned to conservative outlets. This is a little strange, as the book makes it seem like the libertarian blogosphere was buzzing with unanimous approval of the decision to let a house burn down. Here is Glen Beck's discussion; it's the transcript of a radio show, not a carefully penned article, so I'd cut him some slack for sloppiness in terms of language or logical structure. Here is Kevin Williamson's (incredibly short) piece. Also see Jonah Goldberg's two-paragraph commentary. Both Beck's radio clip and Williamson's short piece point out a crucial detail, that the South Fulton Fire department (the one that declined to put out a homeowner's fire in another municipality) until recently had declined to put out any fires outside the city limits. The $75 opt-in subscription to their services needs to be contrasted with the alternative of not having a the services of a fire department at all, not at any price. I think Koppelman and other people of his ideological stripe are morally obtuse in failing to comprehend (or declining to entertain) this important truth. He's apparently insisting on a "principle" that once you are capable of providing a service, you then become obligated to provide that service to anyone who needs it. It's worth noting (as Beck, Williamson, and Goldberg all do) that if you don't have some consequences failing to pay up, almost nobody will pay, and the fire department will be unfunded and thus serve no one. 

See this roundup at The Atlantic of pieces that commented on the event at the time. Note that it includes a post by Ron Beasley, a leftist who says he agrees with Williamson. He makes an argument of the flavor: the $75 fee was like a tax to fund a government service, and if you don't pay your taxes you shouldn't get those services. A search on Reason's website turns up one single link, which doesn't express an opinion on the justice of the fire department's inaction. Here's an old post at Volok Conspiracy, to name another libertarian-inflected outlet (Volok Conspiracy has since moved to Reason). The Volok piece appears to be presenting this as an interesting question rather than a justification of the fire department's decision. It even says "one obvious solution is to make the service mandatory: Require homeowners to pay the $75 as a mandatory tax". This is weird. It feels like Koppelman searched for the libertarian take on this question, couldn't find any sources that told his narrative, so he instead reached for any sources making the argument he was looking for. He ended up with a handful of conservatives and apparently zero libertarians, unless you want to reclassify Glen Beck and Jonah Goldberg. 

If you want the authentic libertarian take on this, Gillespie nails it in the Reason interview. The fire department probably should have showed up to put out the fire, especially because some of the neighbors were paying subscribers, and their homes were threatened by the fire. (The fire department took measures to prevent the spread of the fire to those properties, though apparently not by putting it out at the source, which would have been the more humane and direct way of dealing with the problem.) They should have put out the fire and perhaps taken them to collection for some multiple of the $75 fee, or just acknowledged that the homeowner had previously been a dues-paying subscriber who simply forgot to pay one year (as appears to be the case). 

The book has a lot not to like about it. I've only just scratched the surface in this short piece. Basically this post is aiming at the book's title and grand theme. To do a more thorough and targeted critique of the arguments in the book, I would like to re-read and dissect Koppelman's exact language with greater care. I was eager to read it after hearing Nick Gillespie describe it positively in a podcast. I have to say it was a let down. But it's a useful foil in that it's a book-length criticism of libertarianism that basically misses the mark.