Friday, December 28, 2018

The 20th Century as a Boorish Uncle

I wrote an appreciation of negative role-models here. These are people who set such a bad example that nobody close to them will be tempted to emulate them. They're an evil fun-house mirror showing you how bad it can get, like the boorish uncle who berates his wife in full view of the extended family.

I think of the 20th century as the ultimate negative role model. So many horrible things were tried. This is true also of previous centuries, but this time we had modern technology to record it and preserve this bad example for posterity's sake. Fascism. Eugenics, leading to whole-sale industrial genocide. Communism. Socialism. Hyper-inflation. So many terrible ideas were tried, and what's more they were tried to the hilt. We saw pure, undiluted examples of these phenomena. Eugenics is so horrifying because it's morally revolting to see it in actual practice, even if some pre-Nazi eugenicists had insufficient imagination to see the horror in theory. Same with communism. Even if centralized planning of an entire society had been intellectually respectable in theory, history has repudiated it.

My darkest thought is that history had to try these things to see just how bad they are. It would be easier to indulge empty ruminations if such-and-such a people didn't exist if it weren't for the horrific examples of Nazi genocide. It seems obvious now to us right-thinking folk, but I don't think it would be quite so obvious without a stark negative role-model in recent historical memory. To take another example, full-fledged communism is still respectable in some circles, but far less so than it would otherwise be. Coffee house intellectuals and college sophomores like to pontificate about how awesome society might be if someone were in control. History's repudiation of communism hasn't been quite as fully learned as the repudiation of eugenics and genocide. But the lesson is still there for the taking. Someone more knowledgeable than a coffee house intellectual can point to horrific counter-examples supplied by the 20th century. This would be a lot harder if communism hadn't been tried (and tried and tried and tried...).

I actually have a darker thought still. These lessons will fade from living memory, and we'll have to re-learn them with another round of horrific counter-examples. We'll begin to regard Hitler and Stalin as benignly as we now regard, say, Attila the Hun or Genghis Khan. Or worse, we'll start to think, "You know, we have the data an the computing power now to do central planning. And our understanding of social science is a lot more mature than it was in the 20th century. Let's give it another go." (Salvador Allende thought he could centrally plan the economy of Chile with the computing power available to him in the early 1970s, see here. I can imagine someone with better computers having the hubris to think they can naturally do better.) I don't know if we're really headed for that. I tend to believe the optimistic Stephen Pinker/Matt Ridley/Julian Simon narrative that life is getting better, cleaner, less violent, and more enlightened. But I'm not fully sure of that, and it terrifies me.

Thursday, December 27, 2018

There Isn't Enough Moral Philosophy in Public Health

I know philosophy gets a bad rap these days. Many people think it's just a bunch of useless navel-gazing. Saying there isn't enough philosophy in something practical like public health is like saying there isn't enough interpretive dance in civil engineering, or there isn't enough basket weaving in meteorology. But this is a mistake. Without some kind of philosophical grounding, it's impossible to say which decisions are correct. It's impossible to figure out how to trade off one cost versus another, and it's impossible to justify curtailing people's freedoms, as public health institutions often do.

I've criticized the term public health in multiple previous posts. As far as I can tell, the term doesn't have a coherent definition, at least not one that justifies the existence of government "public health" institutions in any serious way (the FDA, the CDC). Public Health is like a magic incantation that lets the government interfere with our decisions without actually justifying itself. It's as if the goal were to expand human lifespans at any cost, or reduce this or that morbidity rate regardless of people's actual preference for risk. In fact public health officials often explicitly use this kind of language.

If you want examples, read anything by Jacob Sullum on the FDA's misguided efforts to regulate vaping. (Here and here for example.) Sullum repeatedly points out that any restrictions placed on vaping will cause some number of adults to decline to switch from cigarettes to vaping. Vaping is almost certainly safer than smoking cigarettes (though I think there are legitimate concerns that there might be unseen long-term effects), so any missed opportunity to make the switch is bad news even from a strict public health perspective. Sullum even quotes Gottlieb, the FDA chief:

"It's implausible for me to say that there aren't kids out there who are using e-cigarettes instead of combustible tobacco and probably, if they never had this opportunity, would have used combustible tobacco." But he added that it's hard for the FDA to consider that as "a public health justification" when "our mandate is that no child should be using a tobacco product." That suggests the FDA's mission to reduce underage vaping may conflict with the public health goal of minimizing morbidity and mortality.
Even if "public health" means something narrow like "minimizing overall mortality and morbidity across the population," then the FDA is apparently bad at it. They have to follow their mandate off a cliff, without even considering other public health considerations. This doesn't even get into any of the philosophical questions about how to weigh the costs and benefits of restricting people's freedoms. Suppose people really like vaping, even though it turns out vaping removes, say, a year from the life of the average vaper. Vapers decide that the cost is tolerable. Does "public health" demand that we restrict vaping above and beyond the way people voluntarily restrict their own vaping habits? Or here's another philosophical consideration. The FDA seems so concerned with protecting children ("our mandate is that no child should be using a tobacco product") that it's willing to throw a large number of adults under the bus. (An infinite number, as Gottlieb's recitation of his mandate would seem to imply? I hope not, but I don't see what would constrain them.) But wait a minute. Is it so terrible if a "child" vapes? (Does "child" mean "anyone under 18"? Does anyone remember being this age and resenting their lack of autonomy? Does this get any weight at all in the social calculation?) Most of the health consequences of smoking, and presumably vaping, come much later when the child has become an adult and this adult has continued their bad habit for a lifetime. If smoking were acutely toxic, killing some random users instantly, there would be a much stronger case for restricting youth use. Admittedly, it's probably easier to prevent a child from ever starting the habit than it is for an adult to end their long-standing habit. But the framing of "We're protecting children" seems misguided. "We're protecting adults, who would otherwise have a harder time controlling their behavior" doesn't sound quite as compelling as "We're protecting the children." But it would be more honest.

Consider the so-called "opioid epidemic." From a public health perspective, people only see that upward-trending line showing the increase in opioid-related deaths. The only valid goal from a public health perspective is to mechanically bring that trend line back down. The CDC's initial reaction was to do something like this. They issued misguided prescribing guidelines, which completely ignored the variability in actual pain patients' responses to the medicine. These "voluntary" guidelines ended up being used as justification for law enforcement to harass doctors and patients who were prescribing/prescribed "too much." This was a foolish overreaction by the CDC to something that wasn't even a real "public health" crisis in any meaningful sense. Some patients were hurt by these restrictions, and by "hurt" I mean "cast into crippling agony." (I argue here, it was no more a "crisis" than increased driving leading to more driving deaths is a crisis.) And what is the public health benefit of these restrictions? That people who want to use opioids to get high have a harder time finding them? Is that even a benefit? Again, public health obliterates free choice with an all-trumping mandate bring the trend line down. Pleasure counts. It damn well ought to, at any rate.

In his book Good Calories, Bad Calories, Gary Taubes describes a miserably austere diet that would have a small impact on heart disease and obesity rates. You'd basically have to starve yourself for your adult lifetime in order to gain a few months of life. Some doctors and public health officials advocated for the diet anyway, apparently taking an "improve society's health metrics at all costs" perspective. This kind of "no man left behind" thinking is just indefensible.

I'm thinking of other examples that are not necessarily from our public health institutions. Here are several. Sometimes people at my jiujitsu school get injuries, and they ask their doctors how to deal with them. Sometimes they get the clueless response, "Well, quit doing the thing that's getting you injured!" (Some doctors are better than others. One of my injured colleagues found a sports doctor who had worked with the American Olympic weightlifting team, so he understood the notion of training though injuries and training to overcome injuries.) To this doctor, this hobby was all cost. If you can't understand the love of the sport, you won't understand why someone would risk injury to continue it. Russ Roberts has mentioned on Econtalk (I can't find the episode right now) a friend who likes to ride his motorcycle, but who broke his arm in an accident once. His doctor told him to stop riding or he'd stop treating him. Russ' friend, bless him, said, "What the hell do you mean? I love riding my motorcycle." To Russ' friend, cycling was an enriching hobby, and the risk was a price worth paying. I spoke to one older gentleman who was on testosterone treatment. On a high dose, he could get away with less sleep, but on a lower dose, the one that his doctor ultimately prescribed to him, he had to sleep for 11 or 12 hours a day. Suppose we get more total hours of waking life under one treatment regime, even if technically we die at a younger age. Does "public health" have anything at all to say about this trade-off? If it does, and if it conflicts with an individual's preference for "more total waking hours, but a shorter overall life", does public health trump individual choice? (I can't say that any official public health policy led to any of the above examples, but the "better health at all costs" mindset is pervasive anyway.)

I think we really need to realign the mission of our public health institutions so that they are consistent with a free society. (First step: realign them so they are actually coherent.) Doing this might require some squishy moral philosophy. All that public health data is useful, and props to the CDC for collecting it and making it publicly available. But what society ultimately does with that information is not their call.

Do Immigrants Bring Corrupt Institutions With Them?

I did three previous posts about John Carreyrou's book Bad Blood. The book discusses one anecdote that I found incredibly jarring.

Sunny Balwani, president and COO of Theranos, was entertaining one of his buddies, and Indian software consultant. The consultant had struck another employee's car in the parking lot and declined to tell anyone. John Fanzio, a supply-chain manager at the company, noticed the damage and noticed a same-sized gash on the Indian consultant's vehicle. (He actually used a tape measure to check.) John confronted the consultant about it when the man went outside for a smoke-break. The consultant denied it and complained to Sunny, who promptly fired Fanzio. (Supposedly screaming at him "Oh, really, you want to be a cop? Go be a cop!") 

I'm trying to imagine this happening anywhere else, and my imagination is coming up empty. It just seems so beyond-the-pale corrupt that a high-rolling consultant would lie about a minor traffic accident, and that the president of a company would actually cover for him! Maybe I'm sheltered and someone will comment to tell me that this kind of shit happens all the time. Apologies if I'm way off base here. (Maybe Sunny and his friend are just enormous assholes. Or, hell, maybe Sunny's friend was actually innocent.) This seems like the kind of corruption that just isn't done, at least not here in the U.S.

I can't help but think: Is this an example of someone importing corrupt practices from their home-country? I'm an open-borders person. I believe we should have a regime of vastly liberalized immigration, compared to what we currently have. But I think we need to be on guard about stuff like this. Americans have different social norms and political institutions than other parts of the world. Read The Beautiful Tree, which describes in detail the corrupt practices in Indian political institutions. Paying bribes is seen as normal, just another cost of doing business. (The Beautiful Tree is a book about private schools in the third world. Indian private schools often don't meet "regulations", which are often designed with the express intent of making private schools unfeasible. They get around these regulations (e.g. requiring an impractically large recreational area in the middle a crowded city like Delhi)  by bribing inspectors. This kind of tit-for-tat, two-wrongs-make-a-right dynamic definitely breeds corruption.) Or if you like, read Games Primates Play, which is ostensibly a book about human behavior but describes in detail the corruption of Italian institutions. Or read Thomas Sowell's Migrations and Cultures, or any of the other books in his ...and Cultures trilogy for that matter. In her book Infidel, Ayaan Hirsi Ali describes her attempt to socialize Somali immigrants who came to the Netherlands. She had to explain to mothers of small children that it's not okay to hit back. Violence isn't okay just because someone else nominally "started it." (My last book plug for this paragraph: read The Law of the Somalis for a good discussion of Somali legal institutions, and how tit-for-tat violence can mostly keep people in check in that particular society. Ali has a great discussion of why these institutions lead to civil war in Somalia, but the Dutch expectation that "violence is totally unacceptable" leads to social harmony in the Netherlands.) Culture matters, and people do tend to bring their culture with them. Sometimes one person's cultural expectations can clash with another's. You might even say that some cultures are more consistent with a liberal order, with functioning markets and pluralistic social norms, than other cultures. 

Conservatives and skeptics of open immigration worry about this kind of thing. They warn that we'll be overrun by backwards people, that we'll import their bad cultural norms, that they'll pollute our political institutions with citizens who will vote for bad policy. They warn that immigrants with bad political ideas will "kill the goose that laid the golden egg", and we'll no longer have the liberal society and corruption-free institutions that we currently enjoy. I think this is overblown and doesn't justify an outright refusal to admit an immigrant to the country, but I don't want to pretend that this isn't even a thing. I don't know what the answer is. The Dutch supposedly make immigrants watch videos showing men kissing and women sunbathing topless. Supposedly this is directed at Muslim immigrants from Morocco and is an effort to drive home the point "you're going to see stuff like this here, so get comfortable with it or don't come here." A part of me finds this kind of social engineering creepy. But on the other hand, if someone comes from a culture that shames female sexuality and treats homosexuality as a capital offense, we're under some sort of obligation to inform them that those cultural attitudes are totally unacceptable and beyond the pale in the society they are entering. ("Bro, just between you and me. The following behaviors are totally normal and you are likely to encounter them frequently... Got that? Okay, good. Responding with violence to said behaviors is totally uncool. Just so you know." I don't know if this has to be delivered with a waving finger or if it can be delivered as a quiet aside without the finger-wagging "your culture is backwards and inferior" overtones.) I don't know if this kind of "immigrant orientation" should be part of official government policy or be left to private actors who set those boundaries. I like the argument that it's more merciful to conditionally admit an immigrant than to unconditionally say "No" to them. Even when those conditions seem unfair (unfair compared to what? Languishing in third-world poverty?). So if Immigration Services wants to identify cultural problems that are causing trouble and try to nix them with an "orientation" video or some kind of written exam, that's okay if it leads to more overall immigration. It also might be a way to sell liberalized immigration policy to skeptics, as in "We have a program that already addresses this problem."

This all feels very creepy, and just writing it down makes me feel very uncomfortable. But I don't see a way around it. It seems like someone has to point out to a new immigrant that "Some of the cultural norms and expectations you're used to will not fly here. We tolerate alternative lifestyles. Sexism is very much frowned upon, even a fireable or possibly arrestable offense. We report traffic accidents immediately. Unlike in your home country, tax compliance here is very high. Bribes simply are not done, and you will probably be arrested if you try it." It's not necessarily finger-wagging cultural imperialism. It's a downright merciful warning, similar to a "Bridge Out" or "Clearance 9' " sign. We might even say to German and Swiss immigrants, with a culture of overcompliance with rules, "Psst. People flout traffic laws and 'Walk' signals all the time. It's okay. You're kind of weird if you actually wait for the 'Walk' signal when there's no traffic, and it's odd if you don't speed by 5-10 over." If we ever get truly open borders and immigration increases dramatically, we should expect a need for more of this kind of orientation. Contra conservatives, I think this is a fixable problem. And contra some open borders advocates, I think it is a problem with a non-zero magnitude.

Bryan Caplan and Garett Jones have a debate/discussion of the long term effects of open immigration here. I'm reading Garett Jones as saying that "bad culture" effects are real and harmful, and I'm reading Caplan as saying they're not a big deal and not likely to be as harmful as Jones is claiming. I'm more on the Caplan side of this debate than the Jones side. But I think we should admit that this is potentially a problem that needs to be solved to make open immigration more feasible and palatable.

Wednesday, December 26, 2018

Unbroken Brain's Interesting Take on Privilege

I'm listening to Unbroken Brain on audiobook. I've read it twice before on my Kindle, an I wrote a review of it here. There was one sticking point that I was reminded of on this third time around.

Maia Szalavizt recounts here interaction with the criminal justice system. She had been a cocaine and heroin addict and she sold drugs to support her habit. Somehow she got on the cops' radar and was arrested for dealing. She never did any actual prison time, though she was forced into various treatment programs. She describes her white-ness as helping her get privileges that a black defendant couldn't necessarily count on, which ultimately kept her out of prison. "It's impossible to know how much the privileges of race and class were what enabled me to get decent addiction treatment, return to college, and help my lawyer demonstrate to the court that I could soon be a respectable tax-payer. They certainly mattered, however."

Fair enough. But then, she also mentions that her father was a Hungarian Holocaust survivor. He was a successful chemist when he came to the states. (This particular demographic was, famously, incredibly successful. Interesting post about that at Slate Star Codex.) I don't know exactly what "privilege" means, but I think if the official government policy is that you should be exterminated, you don't have it. There isn't necessarily a contradiction here. Maybe a Hungarian Jew is maximally "un-privileged" in Nazi-occupied Hungary, but suddenly privileged when s/he moves to the United States? I believe there are versions of the "white privilege" story whereby discrimination leaves deep and lasting scars on the psyches of the discriminated-against races, even getting passed on to the next generation. The success of Jews who fled the Holocaust would seem to discredit this particular story, at the very least implying that "deep psychic scars" aren't inevitable or don't inevitably lead to higher rates of social problems.

Szalavitz describes her father as suffering from a deep depression. She points out that this is a pattern among Holocaust survivors. She even has a long exposition about epigenetics in Unbroken Brain, about how stress in a parent's or grand-parent's environment can have an impact on the child/grandchild's gene expression. There's a "deep scars" story to be told here, for sure. But to weave it into a "privilege" narrative, someone has to square it with the economic and social success of these deeply scarred peoples.

None of this is to deny that privilege exists. Rather, I'd like to see more rigor and discipline in nailing down the phenomenon, explaining the causal mechanisms involved, addressing counterexamples seriously, etc. It's hard work, but it wins you more converts when you do it right.

The Theranos Story Contains a Deep Lesson About How the World Works

Rupert Murdoch owns the Wall Street Journal, where John Carreyrou first published his take-down of Theranos. Murdoch had a substantial investment in Theranos, about $125 million according to this account. People at Theranos knew that a big, damaging news story was coming. CEO Elizabeth Holmes tried to get Murdoch to bury the story. She met with him four times to plead with him to kill the story. Murdoch repeatedly told her that he trusted the integrity and professionalism of his newspaper's editorial board with regard to running the story. He would not interfere in their decision. Props to Murdoch for his journalistic integrity on this count.

What's more interesting to me than one old man showing journalistic integrity is the broader lesson about how the world works. I don't think this is just an isolated anecdote about one decision made by one man. It repudiates a widely believed worldview that goes something like: The rich dominate the world. There are different versions of this narrative with different degrees of plausibility, and they differ in the degree of paranoia. The most extreme goes something like: a cabal of billionaires totally dominate politics and markets and basically get their way all the time. A less extreme version is given by Martin Gilens in his book Affluence and Influence. This book presents evidence that when the upper class (defined in terms of income quantiles) disagrees with the lower classes about policy, the rich have a slight tendency to get their way. Even this moderate version is probably wrong, but it's widely believed that something like this happens. Presumably there are gradations of this narrative between these two extremes and most people fall somewhere on this continuum between "cabal of billionaires utterly calling the shots" and "gentry class slightly getting their way in policy decisions." In fact, I think most people couldn't even articulate what version of this story they believe, they just vaguely think "the rich call the shots" because it sounds right. (I think most people are sleep-walking with respect to their political opinions, in the sense that they don't give deep or serious thought to what they actually believe.)

By what mechanism might someone in this cabal of wealthy masters exert their influence? Presumably making editorial and staffing decisions at a newspaper that you yourself own is a start. Murdoch could have probably killed the story and fired Carreyrou with little more than a phone call. Or he could have demanded access to all of Carreyrou's information and thus had millions of dollars (maybe hundreds of millions) worth of insider information. There are many ways he could have leveraged his position or private information to unload his worthless investment on someone else. But he declined to do any of these things. It seems like if the "rich cabal dominates the world" narrative is correct, we'd see more instances of this happening.

There are various ways of dismissing this anecdote. "It is just an anecdote, and actual examples of underhanded influence are happening all the time." Maybe, but this just seemed like the prime opportunity for someone to use their influence. Instead this arch-conservative media mogul and business man, kind of a boogie-man of the left, actually showed journalistic integrity. "It was a huge tax write-off for him, so Murdoch didn't care about the loss." This is a fair point, and Carreyrou actually makes it in his book. Still, 65% of a hundred million dollar loss is a big loss. It's nice to offset your gains with your losses, but it's even nicer to not have those losses in the first place. His tax liability would have been higher if Theranos had been a gain instead of a loss, but on net he would have made more money. There are also ways of dismissing the anecdote that actually support my point. "He would have committed insider trading and been criminally liable." If this is the case, then the constraints that apply to the rest of us must also apply to this supposed secret ruling class. "Theranos would have eventually been found out, and Murdoch's huge investment loss would simply have come later." Again, this dismissal implicitly accepts that a media mogul actually doesn't have all that much influence, even with regard to information, even with regard to a story in which his own newspaper is the head of the spear.

I don't want to overstate my case. At the end of the day it really is just one story about one guy making one decision. But combine it with the fact that the richest men in the world are trying to out-do each other in how effectively they can give away their money to charity. Combine it with the fact that politicians get utterly destroyed by even minor instances of campaign finance malfeasance or other untoward behavior. Combine that with the very high rates of taxation on capital (capital gains, the corporate tax (which mercifully was reduced very recently), inheritance taxes, property taxes); either rich people like being taxed or, contra the "rich domination" story, populists actually get their way in public policy.  Some version of the "rich people tend to get their way" story might be true, but the notion of utter domination by a moneyed interests is constantly being contradicted. Maybe it can be saved with a "that's what they want you to think" conspiricism, whereby contrary evidence is "really" proof that they have their hooks in everything and can control the narrative. But that's getting paranoid in the extreme.

I want to preempt a "You would argue something like that, you libertarian you!" kind of response. The argument I'm making above is actually kind of inconvenient for libertarians. Some libertarians like to argue that government is just utterly corrupt to the core, and that's why we should slash the state. I think this is mistaken. For sure there is graft and special interests profit from public funds, but overall out political institutions function with little corruption. I see more of a "Baptist and Bootlegger" story, where some popular cause, perhaps a morally self-righteous one, initially has public support for well-intentioned reasons. But moneyed interests actually do the hard work of writing the laws and allocating the funds and lobbying for funding or regulatory action. Suppose a populist president promises tariffs and other trade restrictions, but then sympathetic interest groups (auto workers, farmers) that are harmed by the tariffs get special dispensations. (Crazy, right?) Those sympathetic interest groups will fight very hard for their special treatment, and perhaps carve out special exemptions from the tariffs that specifically benefit themselves. But the tariffs wouldn't happen in the first place without popular support, nor would the special dispensations be possible without the interest groups being sympathetic (from the point of view of the voting public, as a whole). Rule by secret oligarchs sounds awful, but so does rule by unconstrained populism. I think American politics is dominated more by the latter than the former.

Tuesday, December 25, 2018

Lying? Or Mutually Agreed-Upon Hallucination?

This will be another post on the Theranos story, which is described in detail in the book Bad Blood. I want to give a contrarian take on the story. For a good start, it's worth reading this piece in Tech Crunch called In Defense of Theranos. It was published before bad press about Therano's misdeeds completely destroyed the company, but I think some of it holds up well and generalizes beyond this particular story.

Theranos founder Elizabeth Holmes' vision was to have blood tests done by a finger-stick rather than having to draw a full vial of blood from a vein. Theranos often fibbed to its investors and partner companies about how far along their technology was. Walgreens, for example, ran many of its blood tests through Theranos. They didn't realize that Theranos was actually using standard industry machines (for example, from the German firm Siemens) to do its blood testing, not Theranos' proprietary machine (a device called "The Edison", which never quite worked). At least Theranos wasn't misleading patients with unreliable blood testing technology, but then again it wasn't exactly being forthright with its investors.

But hold on a second. People were raising questions and concerns about the technology, and these warnings were being ignored. Here's a typical example from Bad Blood. A major investor was George Schultz, an elder statesman who had worked in the Nixon and the Reagan administrations. Schultz comes off as foolish and self-delusional in this story. His grandson Tyler briefly worked for Thernaos. Tyler noticed terrible problems with Theranos' lab practices and became a major whistle-blower. Tyler tried to warn his grandfather, who basically blew him off and even tried to strong-arm his own grandson into signing a non-disclosure statement in front of two Theranos lawyers. One way of telling this story is: Theranos led a concerted, deliberate effort to deceive its investors, and they succeeded in fooling most of them. Another way of telling this story is: Theranos and its investors both mutually consented to uphold a fiction that the technology was working well, or that imminent improvements would make it so. Theranos was a private company, after all. If private investors wanted to put their money toward a moonshot, they have the right to do so. One could make an "everybody knows tech start-ups exaggerate" kind of argument. See the Tech Cruch piece, where it talks about hype outpacing reality. Maybe these investors each had some money in each of 100 different moonshots, expecting most of them to fail. Its even possible that a necessary condition for success is for the CEO, senior leadership, and the company's board to have irrational optimism about the prospects of success. It's hard to bring in hundreds of millions of dollars and top talent if you're telling everyone, "Meh, we're probably one of the 99 failures, but we'll give it a go anyway! (raises fist weakly)." It seems like Theranos did this to an excessive degree and probably crossed the line into outright lying, but to some degree this was mutually agreed-upon hallucination.

Walgreens also had many chances to figure out the truth. Walgreens' own lab scientists were skeptical of Theranos' tech and called for audits and clinical trials. Theranos pushed back, and Walgreens management ultimately sided with Theranos against their own scientists. Frustrated senior scientists at Walgreens complained about Theranos' tech not working, but Walgreens senior management basically told them, "If this works, and if our competitors are in on this but we aren't, we'll lose everything." Classic moonshot thinking. I don't know if this is irrational optimism or calculated risk-taking at work.

Carreyrou interviewed a number of scientists in the field of blood testing while researching for his Wall Street Journal piece and his book Bad Blood. A number of them said some pretty damning things. What Theranos was trying to do, to replace a blood test normally done on a full vial of blood with a single drop from a finger stick, wasn't technically feasible. Doing it would require solving many medical and engineering problems that weren't even close to being solved. A finger stick results in capillary blood, which is different from the blood you'd get from a vein ("venous" blood). Basically, the capillary blood has a lot of extra crap in it that can pollute a blood test. A scientist told Carreyrou that solving this problem would be a major engineering accomplishment, and that it was less plausible than finding out Theranos had invented a working time machine. Maybe that's an overstatement of the difficulty, but plainly some very knowledgeable people were saying Theranos' tech wasn't possible. It seems like it shouldn't have been that hard for investors to discover contrarian information. At the very least, they might have revised their earnings estimates or their expectations of pay-back times to reflect the immature status of the technology. To some extent, the investors were snowed. But in a deeper sense, they allowed themselves to be snowed.

It's also worth thinking about this from the following perspective. Suppose the investors were truly joining Theranos in a mutual hallucination. They knew it was unlikely to work, but thought it was worth trying anyway. How do they react when a big story hits the news, telling the world about this mutual self-deception? An investigative journalist tells the world that this young college dropout (Elizabeth Holmes) fooled them all. These are elder statesmen, like George Schultz and Henry Kissinger, and business moguls like Rupert Murdoch. Surely this damaging news story results not just in declining market value, but also hurt feelings and bruised egos, even damaged reputations. Such an investor is likely to feel aggrieved and embrace the "They lied to us" narrative over the "We mutually agreed to indulge an optimistic fiction" narrative. Carreyrou doesn't exactly dodge the fact that some of the investors were delusional, but if I were to write a book like his I'd have a longer discussion of how investors were in part responsible for letting Theranos get away with fibbing for so long. They had a hand in generating those multi-million dollar losses to their own portfolios.  Don't feel too sorry for them.

The Tech Crunch piece linked to above makes this point well: Who exactly is the aggrieved party here? Not the investors, who took a calculated risk. Not the customers, who ultimately got blood tests that were run on standard industry equipment. (Carreyrou finds a few cases of patients who made medical decisions based on bad Theranos tests, or were needlessly worried by positive test results. But it wasn't a wholesale fraud with regard to actual patients. It's telling that Carreyrou could only find a few victims.) Really, the most aggrieved party is the media. So many reporters wrote puff pieces about Theranos and ended up looking foolish. Naturally, they're going to think it's a huge deal when someone points out Thernos snowed them, even if all parties with a concrete interest in the matter are okay with the state of things. It's an overstatement to say that the media is the only aggrieved party. Theranos shamelessly and ruthlessly harassed whistle-blowers, sicking high-powered lawyers and private investigators on them. (David Boies, possibly the most respected lawyer in America, comes off looking like quite the goon in this story.) They even threatened doctors who had complained about Theranos' inaccurate blood tests, which had mislead and needlessly terrified some of their patients. I hope this post doesn't come off as defending these behaviors, which I regard as sleazy and probably illegal. Seriously, good on Carreyrou for demonstrating to the world that this kind of corporate malfeasance still exists. But with respect to the investors, I think it's less useful to think of this as a "con men separate fools from their money" story and more useful to think of it as a "mutual hallucination" story. I think something like this dynamic is at play with every tech start-up. Maybe half the time it sort of works, and the other half the time you lose all or most of your money. At the top 99.99th percentile, you get a Google or an Apple. At the bottom 0.01th percentile, you get a Theranos or an Enron. There's probably some value in doing a postmortem narrative on the failures, but keep in mind we all know ahead of time that most of these things fail.

Sunday, December 23, 2018

The Theranos Debacle Shows Free Markets Working Well

I recently read John Carreyrou's book Bad Blood, which outlines the rise and fall of the company Theranos. A brief summary for people who aren't familiar with the book: Theranos lied repeatedly about the capabilities of its blood testing technology. It fooled investors and partner companies. It then used an army of aggressive, well-paid lawyers to sue any attempted whistle-blowers. (It had legal standing to do this because it made all its employees sign strict non-disclosure forms.) Carreyrou's investigative reporting broke the story open and made it impossible for Theranos to keep fibbing about its technology.

Some people who are familiar with the story might take issue with my title. "Free markets working well? Didn't a private company fool its investors? Didn't it use private contracts to silence whistle-blowers?" Also, a major piece of evidence in Carreyrou's story was a government inspection report of a Theranos lab, which gave the company poor marks. So, didn't the regulatory state have a hand in bringing down Theranos? On the surface, this looks like a story about market actors behaving badly and government regulation reigning them in.

I think it's useful to think about what might have happened if Theranos faced no market discipline whatsoever. A short summary of the story might be: "Theranos lied to its investors." What's interesting about this is that Theranos felt some kind of obligation to impress investors. They had to find some way of getting other people to voluntarily give them money. In order to do that, they had to convince those investors of Theranos' vision of transforming medical testing. They had to make some kind of credible case that those investments would pay off. Theranos had to show their investors earnings reports and revenue projections. In the case of Theranos these reports were wildly optimistic, probably impossible, and certainly not done in good faith. Still, it's worth pausing for a moment to reflect on why Theranos saw any need to make such projections.

Imagine a world where a bad actor like Theranos doesn't even have to bother with these formalities. Imagine a medical firm can make wildly optimistic claims about the social value of its future products, unmoored by investor discipline of any kind. They get their allocation of funds, equipment, top scientists, and talented engineers paid top-level salaries year after year. It would be a tremendous waste of resources if this firm never had to show anything in return. It would be like the worst, most useless parts of academia, but now spread across the entire economy. Without investor discipline, profligate spending on useless projects goes on forever. Sure, maybe a free-market think-tank does a study pointing out the negative social return on such investments, or maybe a few investigative reporters write stories about labs that never produce anything of real value. But these labs simply fire back with vapid slogans about how "we must invest in our future." A public with a short attention span ignores the criticisms, not having the time to cost-benefit analyze every government program.

The Theranos debacle shows markets operating at their worst. A company lied to its investors and got away with it for a while. But does this impugn markets? Does corporate malfeasance at the 99.9th percentile impugn markets in general? Not really. What matters is what happened next. Theranos got its comeuppance, long before the government stepped in to fix the problem. Do we have proof that our system is a failure every time someone steals something? Every time some guy murders another guy? No, that would be silly. Problems happen in every society and under any system of government. What matters is how well problems are dealt with when they happen. What matters is how institutional arrangements create the right incentives, and how well these incentives avoid problems in the first place. The Theranos story is so compelling and news-worthy because it's so unusual.

Here's an analogy. I had a mouse in my house the other day. At first blush, my two cats failed. I jokingly sent a text to my wife saying "The cats are fired." On second thought, though, it was very useful to have cats around. They managed to chase the mouse under a cabinet and keep it trapped there. I couldn't find any evidence that the mouse got into our pantry, which means the mouse didn't have free reign to cause trouble. (We had mice in the house several months ago, so I know what that looks like.) Did the cats fail, because we had a mouse in the house? Or did they succeed, because as soon as a mouse got into the house they managed to trap it? I don't think I should judge my cats by a "mice never happen" standard, but rather by a "How quickly does a mouse scenario resolve itself?" standard. I read the Theranos story in the same light.

Friday, December 21, 2018

Does Ayn Rand Taking Social Security Make Her a Hypocrite?

This silly claim gets brought up now and then. I heard it again recently, so I thought it was worth commenting on.

Presumably, Rand was compelled to pay into Social Security. She was forced to buy a financial product she didn't necessarily want. She didn't have the option of saying "No thanks" and keeping her contributions. So is it problematic from a libertarian point of view if she then accepts payments from Social Security? Not really. I'd consider it problematic if she applied for welfare. It's much harder to simply "opt in" to welfare, but it's almost a default that people opt in to Social Security. If she went out of her way to make herself the beneficiary of government largesse, that would impugn her character. Or if Social Security is a really good deal. Suppose it earns a return on contributions much, much higher than the market rate. Those inflated earnings are being subsidized by young workers, who won't earn as much when it's their turn. Claiming benefits from such a program might be un-libertarian and hypocritical, even if most people do it.

An analogy: I'm forced to buy more health insurance than I otherwise would because of bad federal policy (mainly the ACA, but also mandates that predate the ACA). But given that I've bought this financial product, I might as well use it when I have some kind of medical expense. It's not hypocritical for me to make a claim against my absurdly over-priced health insurance. Someone compelled me to buy it. I complied with that law. That's where we are today. I have more coverage than a rational person would buy in a libertarian market for health insurance, but given that I paid for it I'm going to make use of it.

A better example of a prominent libertarian flouting libertarian principles is Robert Nozick suing his landlord. His apartment was supposed to be rent controlled, but his landlord raised the rent on him for several years in a row. Nozick figured this out and sued, despite his landlord confronting him directly with his libertarian philosophy (supposedly shoving a copy of Anarchy, State, and Utopia in Nozick's face).

But does any of this matter? So what if some prominent libertarians fail a purity test or can't resist the treasures that the state dangles in front of it? Nozick could have totally nailed the case for libertarianism, but then had a failure of personal character. The arguments stand on their own. It's pure ad hominem to say that libertariansim is compromised as a system of values because Rand and Nozick couldn't live up to their principles. This is just like people claiming that Al Gore's personal energy consumption makes him a hypocrite on environmental issues. Of course it doesn't. You are supposed to evaluate the argument, not the person.

The following shouldn't need to be said, but unfortunately it does. Even if someone like Rand or Nozick were so blatantly hypocritical that we could infer they didn't believe their own arguments, there are other serious people making similar arguments or perhaps building on their work. That work isn't impugned by someone else's hypocrisy or weak character, either. 

Wednesday, December 19, 2018

Cost-Benefit Analyzing My Efforts At Policy Analysis and Advocacy

I want to do a back-of-the-envelope cost-benefit evaluation of my own efforts at policy analysis/evangelizing, at least to put some kind of bounds on my contribution. The policy I’ve put the most thought and effort into is drug policy, more specifically the “opioid epidemic.”

Heroin/fentanyl overdose deaths (accidental) rose from 2,124 in 1999 to 33,534 in 2017, according to data published by the CDC. I consider this rise to be entirely the fault of bad drug policy. In fact, I even think the 2,124 figure from 1999 is a high estimate of what overdoses would be in a legal market, with safe injection facilities, pharmaceutical-grade drug supplies, and assistance services that are likely to exist in a free market for drugs. (When something is highly in demand but potentially lethal, products and services typically emerge to make it safer. And people tend to purchase these kinds of safety features.) But that’s not too important. Let’s just suppose that the number of unnecessary heroin overdose deaths is about 2,000 under total legalization and that bad drug policy raises it to 33,500. (Yes, I’m assuming the conclusion that I like for the sake of this argument, having supplied that supporting argument elsewhere.) A total reversal of bad drug policy would save 31,500 lives a year, about 86 lives a day. If my efforts shift public opinion and the opinions of important policy makers ever so slightly, and if I bring about some kind of policy reform one day earlier than what would have otherwise been, it will save about 86 lives. That’s a big deal to me. I could make this calculation a lot more conservative, making it contrary to my preferred conclusion, and it’s still easily worth the time and effort I’ve put into it.

What’s my actual contribution? Hard to say. Maybe I’ve had no effect whatsoever, and all my words are mere dust in the wind. I think I’ve had some impact on elite opinion, and in some cases I’ve been the point of the spear on this topic. I’ve had e-mail correspondence with some of the important public intellectuals who are particularly outspoken on this topic. Some of them have told me that I influenced their opinions, and one asked me to write a paper with him. Three years ago when I really started researching, I could find very little published material that was skeptical of the standard narrative. And much of what I found was incomplete or not very satisfying. I think I’ve made some unique observations and brought them to attention at least a day sooner than they otherwise would have been. What impact this has on any actual policy change, I don’t know. Maybe the zeitgeist changes slowly and deterministically, regardless of any individual person's evangelism. Maybe I’m just a cause and not an effect of shifting opinions on drug policy. Whatever. The point is there is a massive problem to be solved, and I can be happy if I’ve had even a small impact on shifting policy in the right direction.

Maybe I’m 180 degrees wrong and moving policy in the wrong direction. Or maybe I’m right but arguing my point so poorly that the blow-back will tilt policy in the wrong direction. In either of those cases, my contribution would be negative in a big way. So it’s really important to do this right. Policy analysis is hard. It requires serious thought and careful communication.


I kind of knew this would happen. I wanted to write this post, but I thought “Geez, this is going to sound extremely self-congratulatory!” I want to warn against self-delusion. Anyone who thinks they are single-handedly saving dozens of lives probably needs a reality check. Scott Alexander’s piece Stop Adding Zeroes makes this point really well: Don’t just come up with a huge estimate of the problem, multiply it by something small, and say “Even if we only do this much it’s worth it!” I hope it doesn't sound like that's what I'm doing in this post. That said, it is very important to get policy right, because public policy applies to hundreds of millions of people. A bad policy with a small effect can kill thousand of people if it's applied to a large enough population. Ronald Coase once said something to the effect (I can't find the exact quote at this moment): If a public servant can delay for two weeks a bad policy that costs a billion dollars per year, he's earned his salary for a lifetime. Maybe this is slightly question-begging because it assumes that public servants can easily identify good and bad policies. (Should anyone feel confident about this ever, given that this is what we're all arguing about all the time?) Still, I think there's something salvageable here: It's really important to get policy right. 

Sunday, December 2, 2018

Drug Poisonings Update for 2017

The CDC has released the mortality data for 2017. This post will be my annual update. In this post I’ll point out some of my pet peeves with respect to news stories on the “opioid epidemic.” I’ll try to give my readers the tools needed to interpret the numbers and warn you against sloppy reporting.

In 2017, there were 70,320 total drug poisoning deaths, up from 63,536 in 2016. So far, this has nothing to do with the opioid epidemic. There are many kinds of drugs and people are poisoned by many combinations of substances. If an “opioid epidemic” story anchors you to the 70,320 figure, be skeptical. They are starting with an inflated number. It’s true that deaths from heroin, synthetic narcotics, and prescription opioids constitute a large fraction of these overdoses, something like 2/3 of the total in recent years. But a large fraction of these deaths involving cocaine, methamphetamine and prescription ADHD medicine (collectively called “psychostimulants with abuse potential”), benzodiazepines, and antidepressants, often in combination with alcohol. (Pure alcohol poisoning deaths are not counted in the 70,320 figure above; alcohol has its own set of ICD10 cause of death codes. But many drug poisonings involve alcohol, as alcohol interacts with many kinds of drugs.)
I think the “two-thirds of seventy-thousand” framing is a bit of a trick. German Lopez at Vox has done this multiple times. I criticize his piece here and I see a more recent piece where he uses the same trick for the 2017 data. I don’t get it. 
“Dude, imagine there’s this guy. And he’s nine feet tall!”
“Whoa, that’s pretty tall.”
“Right? Okay, the dude I’m going to actually tell you about now. He’s 2/3 of that!”
“Um, okay. So he’s six feet tall?”
“Yeah. So this guy…”
“Wait. Why did you start me off at nine feet? You’re only talking about a six-foot guy, right? Why did you start me off by imagining this giant?”
“Dude, the three feet of space above his head exists. It’s just that I’m going to focus on the six feet of guy.”
 “O…kay. Fair enough.”
Before we start splitting these deaths up by substance, let’s separate out accidental overdoses and suicides. Of the 70,320, a majority (61,487) were accidental deaths. The rest were suicides (4,977), “undetermined intent” (3,714), and murder (142). When people talk about the crazy mortality of the “opioid epidemic”, I think they are really only talking about the accidental overdoses. If someone makes the decision to end their own life and they happen to use an overdose of opioids as the means, I don’t think we can blame that death on loose prescribing practices by pain doctors. We can’t really say, “But for the opioids, that person would still be with us.” On the contrary, but for the opioids, that person would likely have chosen another means of ending their life. Someone could make the case for adding in a fraction of the “undetermined intent” deaths, because some fraction of these are suicides and some fraction of these are accidental. But then again, the growth of the “undetermined intent” category is small (growing from 2,476 in 1999 to 3,714 in 2017) compared to the growth in accidental deaths (which grew from 11,006 to 61,487 over the same period). When people talk about the explosive growth in drug poisonings, they are mostly talking about the growth in accidental deaths. So my post will focus on those deaths.

(Including suicides and undetermined intent makes the overall numbers higher, but makes the slope of the trend line over the past 20 years less intense. I’m happy to discuss how any conclusions change if you include or exclude these in the count totals. Also, someone might make the argument for including opioid-related suicides in the counts. "Loose prescribing practices for painkillers have created a lot of miserable, hopeless addicts. They become depressed with their lives and end them." But this kind of story-telling speculates wildly, beyond what the raw numbers imply and contrary to other evidence about opioid addiction rates. Also, this could easily cut the other way; it's likely that the lack of readily available opioids leads to a lot of suicides from desperate chronic pain patients.)

Here is a table of accidental drug poisonings by substance since 1999. (Apologies if these tables and graphs look small on your screen or overlay the stuff in the right pane of this blog. You can click directly on these figures to enlarge them for a clearer look.)

Each of these columns corresponds to an ICD10 code. Heroin is T40.1. "Other Opioids" is code T40.2, and it corresponds to Oxycontin, Hydrocodone, and some other common prescription opioids. Methadone is T40.3. "Synthetic narcotics" is T40.4, which corresponds to fentanyl and other super-powerful synthetic opioids. Cocaine is T40.5, benzodiazepines is T42.4, "Psychostimulants with abuse potential" is T43.6, and antidepressants is T43.0 and T43.2.

If you're a more visual person, here are the same numbers as a chart.

Naively, everything is going up.
"What's the trend in accidental poisonings from 'other opioids'?"
"It's going up!"
"What about benzoidazepines?"
"That's the  orange line. It's going up, too!" 
You have to be really careful here. Most drug poisoning deaths involve multiple substances. I've written about that here and here, and in many other previous posts. What that means is that if overdose deaths from one substance are skyrocketing, it can drive up deaths from other substances. For example, even if deaths from prescription opioids excluding heroin are declining and deaths from benzodiazepines excluding heroin are flat, the overlap with heroin-related deaths can mean all deaths involving prescription opioids and benzodiazepines are increasing. Here is a table showing how different substances relate to each other in 2017. The top table shows raw numbers. (For example, 3,851 deaths involved both heroin and cocaine; 2,552 deaths involved both heroin and benzodiazepines.) The bottom table shows how many deaths involving the substance named in the row label also involved the substance in the column label. (For example, 26% of heroin deaths also involved cocaine; 46% of benzodiazepine deaths also involved synthetic narcotics such as fentanyl or carfentanyl. Basically, divide all the numbers in each row by the diagonal element for that row, and that gets you the bottom table.) This should give you a rough idea of which substances are likely to be found in combination when there is an overdose death.

Given this, let's take a second look at the trends by substance. I'm going to do some grouping. I'm defining "prescription opioids" to mean other opioids or methadone (so it involves ICD10 codes T40.2 or T40.3). In practice, heroin overdoses are sometimes labeled with T40.1 and (the ICD10 code for heroin) and sometimes with T40.4 (the ICD10 code for synthetic opioids). So much black market heroin is tainted with fentanyl and other super-potent synthetics, and toxicology screenings often do not distinguish between them. So it makes sense to count deaths involving heroin or synthetic opioids as being in the same category (generically, we could call these "heroin overdoses"). In the table below, "Inclusive" means it's counting all accidental poisonings involving that substance. "Exclusive" means it's excluding heroin and synthetic narcotics (shortening to just "heroin" for the discussion below).

This tells a slightly different story than you would see by looking at the raw figures. The "Prescription Opioids Inclusive" column shows a steady increase since 1999 (with a slight dip in 2012 and 2013). The "Prescription Opioids Exclusive" column, which once again is filtering out heroin deaths, has actually been declining since 2011. Benzodiazepine-related deaths look like they are steadily increasing, but excluding heroin, deaths involving benzos have been basically flat or even declining since 2010. Cocaine-related deaths hit a high in 2006, then started coming back down, then rose again. This is still true when filtering out those cocaine-related deaths also involving heroin, but the rise is not quite as dramatic. Psychostimulants are rising dramatically, but not quite as dramatically if you filter out the heroin-related deaths. The take-away here: heroin is so deadly and has become so prevalent it is driving up deaths in all the other major drug categories.

Keep this in mind when someone reports on the CDC numbers by saying "deaths from prescription opioids are at an all-time high." Such a report is really a commentary on the deadliness of heroin, not of prescription opioids, and the reporter is failing to separate these into different topics. You could just as truthfully say "Deaths from prescription opioids have been declining since 2011" or "Deaths from prescription opioids are at their lowest level since 2007." In accidental drug poisonings involving both street heroin and prescription pills, it seems likely that heroin would be the more dangerous substance and more likely to be responsible for the death. Given that, I think the "prescription painkiller deaths are declining" framing is more accurate.

If you're a visual person, here's the same information in graph form:

Some context for this. Some recreational opioid users use benzodiazepines to "potentiate" the effects of whatever else they're taking. This is incredibly dangerous and apparently accounts for a large number of accidental deaths, but some people apparently take this risk willingly.

Some context for the cocaine figures. Cocaine and heroin is a common combination, known as speed-ball. It was involved in the deaths of John Belushi, Chris Farley, Mitch Hedberg, and Philip Seymoure Hoffman. What's likely happening here is that the "heroin" used to create this concoction is increasingly tainted with fentanyl, sufentanyl, carfentanyl, etc.. What might look to a naive observer like an increasingly severe "cocaine problem" is really a "black-market heroin" problem. (Note the automatic axis scaling on this one; the bottom of the y-axis is not at 0. See the table above for the numbers.)

Policy Implications

Facts and number by themselves have no policy implications whatsoever. You need a theory or narrative to interpret them. This hasn't stopped some commentators from drawing unwarranted, even silly, conclusions from recent evens. Some have claimed that recent events have caused them to start "questioning my drug-libertarianism." (As if drug libertarianism has been tried as a matter of actual policy. It hasn't.) Supposedly the increase in prescription painkillers and the subsequent rise in opioid-related drug poisonings are a black eye for drug-libertarianism. That might have been a semi-defensible position up until 2010, when people started noticing the rise in opioid-related deaths. But the 2010 to present period represent a huge black eye for drug prohibition. In recent years, legal restrictions, informal guidelines, and attitudes about prescribing practices have made it harder to obtain opioid painkillers legally. This led to an increase in the use of street heroin by recreational drug users. Drug dealers have "innovated" by adding fentanyl and even more potent synthetic opioids to the "heroin" they sell (in some instances it seems to be a complete substitution of the synthetic for the heroin rather than simply adulterating).

The 2010 to present increase seen in the figures above is enormous, even overshadowing the 1999-2010 increase. (What was happening prior to 1999 is less clear, because the then-used ICD9 system did not separate out deaths by type of substance, at least not as well as ICD10. Still, it looks like overdoses were rising prior to 1999.) The libertarians had this one right all along: you can't control drugs, and you can't stop drug use. There are too many work-arounds. When dealers can collapse a one-ton shipment of heroin down to something that fits into a briefcase, it becomes impossible to stop drug trafficking. In his process, there is no guarantee that the final product has been properly mixed or diluted. The end users have no idea what dosage they are taking, which results in a large number of accidental fatal overdoses (as is evident in the dramatic rise in synthetic opioid deaths).

The 1999-present period did not actually see an increase in prescription painkiller abuse. Some commentators will attempt to blame today's heroin epidemic on prior loose prescribing practices, as in "Today's heroin users are yesterday's legitimate opioid patients; loose prescribing practices turned these people into addicts." But the flat-line in the "abuse" statistics is a huge break in that chain of causation. Apparently you can triple the amount of opioid prescriptions (roughly what happened in the 1999-2010 period) and not actually get any more addicts. Even putting that aside, it's not necessarily problematic if prescriptions for a drug triple and deaths associated with that drug also triple. If an enlightened decision is made to increase the use of a particular treatment tool, which I would argue is an accurate description of the decision to use opioids to treat more chronic pain, we should expect the side effects of that treatment to triple. It's not even clear that all of these deaths that are labeled "drug poisonings" are actually that. A standard textbook on drug-related pathology warns repeatedly against assuming something was a drug overdose based only on toxicology screenings, circumstantial evidence, or convenience. The repeated warnings, and statements made by the author elsewhere, imply that this mistake is common.


Here's a back-of-the-envelope calculation regarding mis-labeled causes of death. About 3% of the US population are on high-dose opioids (so about 9.7 million people). These people have a higher-than-average mortality, about 1.6 times the typical all-cause mortality (because they're older and sicker, not primarily because they are dying of overdoses). The population average mortality in the US is ~850 per 100k. So about 130,000 people on high-dose opioids die in a given year from all causes (9.7 million * 850/100k * 1.6). If 1% of these are mis-classified as drug poisonings (and that must be a tempting diagnosis if someone in this population dies suddenly and mysteriously, say from a heart arrhythmia), that's 1,300 deaths mis-labeled as drug overdoses. If 5% are mis-classified, there are 6,500 mis-classified drug poisonings related to prescription opioids (out of a total of 8,691 in 2017, again counting accidental only and excluding heroin-related deaths). Even if the mis-classification rates for cause of death are low, mis-labeled "opioid poisonings" could represent a large fraction of the total and could over-state the trend. Seen in this light, the rise in "opioid deaths" in the CDC's official numbers may not be alarming at all. (The discussion in this paragraph is strictly about prescription opioids, not heroin. I believe the recent rise in heroin-related deaths is very real. The "at-risk" population for heroin is a lot smaller, so it's not likely that a low mis-classification rate could be responsible for such a large increase in the official figures.)

Sunday, November 25, 2018

This Is Why We Can’t Have Nice Things: Healthcare Edition

Suppose I were to snap my fingers and transport us to my preferred health policy regime. Everyone pays out of pocket for most of their medical expenses, and most people carry catastrophic health insurance policies for truly bank-breaking medical episodes. I think there would be a very strong tendency in this world to push back in the direction of "insurance pays for everything." If we ever got there in the first place, preserving the "first-party pays for most medical expenses" regime requires vigilance.

The problem is that there are always marginal claims, which according to the insurance policy language might or might not be covered. Insurance companies try to nail down policy language so as to minimize this confusion, but there are always ambiguous claims. There is even a tendency for insurers to cover claims that they are clearly not on the hook for, because they like to show good faith to their customers. (I once overheard someone who was on a committee that decided how these marginal claims get paid out. She said something like, "How are we going to stay in business if we keep paying these claims that people don't have coverage for?" Presumably some insurers specialize in going the other direction, saying "No" to as many ambiguously covered claims as they can get away with.)

Everyone wants these marginal claims to be adjudicated in their favor. But at the same time, everyone wants their insurance premiums to stay affordable. It requires some discipline on the part of the insurer to say "No" to these claims and resist mission creep. And it probably requires a more mature understanding on the part of the insurance customers: with low premiums, diligent claims defense is part of the deal. The insured with the marginal claim needs to be able to say, "Low insurance premiums entail a lot of self-insurance for these kinds of claims. This is a cost I agreed to pay, and now I'm paying it."  This would probably require a massive cultural shift on the part of the American public.

I think this kind of "cover everything" mission creep in health insurance can happen even in the absence of wrong-headed legislation or foolish judicial decisions, declaring that this or that treatment must be covered by insurance.

So that's one problem. People have this, "I have health insurance, so it should pay for everything health-related" mentality, which has lead to spiraling health costs in the U.S. and elsewhere.

The other problem is that there are very high fixed costs in medicine (that's not the problem), and that they engage in price discrimination to extract more money from rich, price-inelastic customers (that's still not the problem). The people who pay high prices bitch and moan about how somebody else got the same treatment for a lower price (that is the problem). Some commentators on health policy, even conservative and free-market commentators, have called for total price transparency for all medicine. As in, the hospital lists all their prices somewhere (on a menu, on their website, whatever). Trying to charge more than the list price is fraud, in the eyes of these reformers. I'm sorry, but I don't see how you get away from price discrimination. I explain the concept here (specifically with respect to medicine) and here (more generally). By charging lower prices to price-elastic customers, hospitals and clinics can attract customers who still earn them a profit, but who are below the average price they would need to charge to keep themselves in business. Offer everyone the low price, and there isn't enough to pay support staff, keep the lights on, amortize the cost of the building, and so on. Offer everyone the high price, and the hospital/clinic misses out on these price-inelastic patients, who are still profitable. This situation is common when you have low marginal costs and high fixed costs. The price-inelastic customers, the ones who are sore about paying the higher costs, actually benefit from price discrimination, because the clinic has more customers to spread their fixed costs over (even if they don't load quite as much of the fixed costs on these low-price customers). These reformers probably imagine that under pure price transparency, everyone obtains the same amount of care but nobody gets price-gouged. This is wrong, because some people won't seek treatment except at the lower price. (I hesitate slightly to make this point, because some places have opted for total price transparency. The Surgery Center of Oklahoma and clinics offering laser eye surgery are good examples.)

As I write this, I can't help but be reminded of someone I know who complained about a surgery that cost him $20,000, in his telling for a trivial amount of time and effort on the part of the surgeon. This was someone in a very well-paid management job. I'm thinking, You're the rich patron who's keeping the lights on and subsidizing lower-income patients. Be happy to play that role. The surgeon could probably do the same surgery for 1/10th the cost for a price-elastic patient. But he couldn't stay in business and wouldn't have bothered becoming a surgeon in the first place if he had to offer that price to everyone.

I suspect a functioning market in medicine has to embrace a little bit of price discrimination, which probably means tolerating a lack of total transparency in prices. This means a culture of not bitching and moaning that someone else got a better deal than you.

Wednesday, November 21, 2018

Tabarrok and Cowen on the Economics of the Drug War

This is from their Modern Principles: Microeconomics text. Ellipses for omitted text, most of which is a description of a supply and demand chart.
Why the War on Drugs Is Hard To Win 
It's hard to defeat an enemy that grows stronger the more you strike against him or her. The war on drugs is like that. We illustrate with a simple model.
The U.S. government spends over $33 billion a year arresting over 1.5 million people and deterring the supply of illegal drugs with police, prisons, and border patrols. This in turn increases the cost of smuggling and dealing drugs.
(Description of a chart accompanying the discussion follows)
The most important that the demand curve is inelastic. It's hard to get good data on how the quantity of drugs demanded varies with price, but most studies suggest that the demand for illegal drugs is quite inelastic, approximately 0.5. Inelastic demand is also plausible from what we know intuitively about how much people are willing to pay for drugs even when the price rises. Economists have much better data for the elasticity of demand for cigarettes, which one can think of as the elasticity of demand for the drug nicotine and it too is about 0.5.
What happens to seller revenues when the demand curve is inelastic and the price rises?... When the demand curve is inelastic, an increase in price increases the seller revenues....Prohibition increases the cost of selling drugs, which raise the price, but at a higher price, revenues from drug selling are greater even if the quantity sold is somewhat smaller.
The more effective prohibition is at raising costs, the greater are drug industry revenues. So, more effective prohibition means that drug sellers have more money to buy guns, pay bribes, fund the dealers, and even research and develop new technologies in drug delivery (like crack cocaine). It's hard to beat an enemy that gets stronger the mroe times you strike against him or her.
Then follows a paragraph discussing the use of a tax to deter drug use, as opposed to outright drug prohibition. They reference a paper by Gary Becker (and Kevin Murphy and Paul Grossman) titled The Economic Theory of Illegal Goods: The Case of Drugs. I did a write-up of this excellent paper a couple of years ago. The conclusion: even if we decide it's a good idea to deter drugs (which is a dubious goal), it's better to do so with a tax than an outright ban.

This discussion in the Modern Principles text nails the economic case against drug prohibition. So I don't understand how Cowen can be such a teetotaler. Maybe Tabarrok wrote that part of the textbook, and Cowen doesn't really know the standard economic analysis of drug prohibition? Or maybe Cowen does understand the standard economic treatment in the Becker paper, but perhaps he thinks that something's missing from this analysis? (But has never, to my knowledge, explained what is missing.) Since I wrote that blog post, Cowen posted this piece with excepts from a New York Times piece. Tyler adds his own commentary, saying "I increasingly believe that decriminalization will prove a more stable solution than outright legalization." Nothing in Cowen's post or the original NYT piece supports this belief, so I guess he's just sharing his stray thoughts with us. "Decriminalization" means no penalties for users but prohibition on the supply side. But the economics of illegal goods implies that the black market grows stronger (in the "greater revenues" sense) the more you crack down on it. So I don't get it. Decriminalization keeps the high revenues for black markets intact (with all the associated black-market violence and corruption) while not discouraging the demand side (except through higher prices driven by interdiction, which price-inelastic users willingly pay). To be fair, Cowen never outright defends drug prohibition, but he's always lukewarm on the benefits of legalization and (in my reading of him) exaggerates the harms of drug use.

In a recent assorted links post, Tyler links to Scott Alexander's post on marijuana prohibition, in which Scott references some (in my opinion, poorly done) recent studies claiming that traffic accidents rise after states legalize marijuana. (I remember reading Scott's post and thinking, "Tyler is going to post a link to this, because this drug war meta-contrarianism stuff is like catnip for him." I even e-mailed a couple of people at Cato saying that this nonsense is going to make the rounds in the news cycle again, so we might want to get ahead of it.)  My reservations about these kinds of studies are outlined here. Not to fault Scott or Tyler for referencing respectable-looking, superficially convincing research on this question. What else are they supposed to do? But there is contrary research showing the opposite conclusion. I worry that it's becoming easy to claim "studies show legalization increases traffic accidents" when in fact that's still an open question (and in my reading of the evidence to date, probably not true). Scott is careful to hedge his conclusions in conditional terms and say why the studies he cites might not be right, but the thrust of his post is about how his previous cost-benefit assessment of marijuana legalization was wrong and that this new evidence might flip the conclusion. Be careful, guys! You both have enormous influence on elite opinion. 

Temp Agencies and Employer Learning

Paging Bryan Caplan.

In this episode of Econtalk featuring Noah Smith, Russ Roberts brings up the topic of temp agencies. Why, Roberts asks, do temp agencies command such a high premium? They take a large fraction of their workers' hourly wages. Why should temporary workers be willing to pay such a premium to someone to find them work? (This is particularly baffling when you consider that these are people with already-low wages.) Both Noah and Russ seem confused and don't have an easy answer, but some commenters present plausible explanations.

From a listener called MF:
Towards the end of the conversation, Russ and Noah talk about the role of temp agencies in employment.

My company (A small manufacturer, just under $10M in rev and ~30 employees) uses the temp agency as a long-term screening tool.  Basically, we have found that people are generally good at getting through an interview and a month or two of employment while “holding it together” but people tend to be who they are by say 90 days of being part of the team.
 When you hire someone as a regular full-time employee and then after 90 days or so they start acting out, slacking off, being overly confrontational, etc it is really hard to get rid of them and even if you properly document all the behavior/performance issues they often still end up on unemployment benefits which drives up the premium for the companies unemployment insurance.
 For us, (I can’t say why other companies use temp labor) it is well worth the premium paid to the temp agency to help manage this process.  Especially when considering that bad behavior or animosity towards co-workers can bring about horrible results on a production line, in a busy warehouse, etc…basically anywhere that heavy equipment is running.
 If a temp employee does start acting out, behaving dangerously on the line, etc there is NO paperwork and all we need to do is call the temp agency and say “person X is no longer needed” and that’s it.  The agency may send a replacement for that person the next day if needed.
 However, it is much better value in the long run for the company to hire them, and indeed at my company, those who are eager to learn, want to work as part of the team, be safe, etc are most often offered full employment and benefits with the company in the long term.
I think this fits well with Bryan Caplan's discussion of "employer learning" in his book The Case Against Education, which I wrote about previously. Employers who use temp agencies are in effect laundering the firing process. You're not "fired." You're simply "no longer needed."

I think this is very sad. It means our labor markets are inflexible, and that inflexibility is particularly harmful to low-skilled laborers. I wish the labor market was more fluid. I wish these low-wage laborers could float in and out of jobs more easily. It would be nice if they could work directly for the employers at the normal wage, rather than having to fork over huge sums to the temp agencies. Yes, that might mean some of them get fired more often. However, knowing that they can fire at will, employers would be more willing to actually take a chance on these low-skilled workers. That would be a much more forgiving labor market. It would create opportunities for low-skilled workers, who perhaps have had behavioral issues and a spotty work history.

Why is our labor market so inflexible? It's hard to escape the notion that labor regulations make the problem worse than it needs to be. Mandatory benefits make wages inflexible, because if I have to give you $X worth of benefits, I have to reduce your wages by $X (in an on average and all else equal sense). If I have to pay you a huge severance when I fire  you, it means firing you will be costly, which means I'm less willing to take a chance by hiring you in the first place. There might be good reasons for offering severance packages that have nothing to do with labor regulations. Preserving morale of the existing workers, convincing them the process is fair, etc. But this might make less sense for low-wage workers, who earn their employers pretty slim margins. Common "wisdom" says that labor regulations protect workers from exploitation. I think the shape of the labor market, particularly for low-skilled laborers, implies that there is a huge cost to these regulations. 

Some of the commenters on the Econtalk episode give plausible explanations for the temp agency phenomenon, and some of this is coming from employers who actually make hiring and firing decisions. They are explaining their actual reasons for being inflexible in hiring/firing decisions. If you're usually dismissive of arguments about regulations having counter-productive effects, I implore you to not dismiss so easily. This excellent essay in Regulation Magazine by Warren Meyer seems appropriate to this discussion.


By the way, Noah Smith makes the absurd argument that big companies can keep wages down because they wield "market power." That would imply that big companies earn enormous returns on their low-wage workers. Given the temp agency phenomenon discussed above, given that low-wage workers are the first to lose their jobs during an economic down-turn, and given that these workers experience the highest rates of unemployment, I find these arguments kind of ridiculous. Maybe I'm being dense, and the high unemployment is actually part of the "monopsony/ market power" story: "Large employers employ fewer people than they would if they were forced to offer the market wage. That's how they keep wages down, by keeping people unemployed. Fewer workers means you're lower on the supply curve for labor, which means you can pay lower wages. D'uh!" But that doesn't make sense to me. Someone who is turned down by Wal-Mart can go work for McDonalds or KFC or Walgreens. At least one of these companies will hire this worker, whom they could supposedly exploit for a huge return on investment. The monopsony story implies more than one big employer wielding market power. It implies that all big employers are conspiring together to keep labor costs down, and they are all resolutely resisting the urge to earn huge profits by "cheating" on the deal. This is a "secret cabal of evil businessmen" level conspiracy theory.

A second aside. Caplan has a long section in his book about "IQ laundering." As in, explicit IQ testing is illegal if it leads to racial bias (which it often does), so educational credentials launder the IQ test. Universities effectively do the IQ testing for the employer, via years of coursework and student selection. Employers can then merely impose a degree requirement for a job and escape the threat of lawsuits they would incur if they used explicit IQ tests. Caplan does some back-of-the-envelope math to show that the effective penalty for IQ testing is a pittance compared to overall labor costs. If it were beneficial to do so, most employers would gladly flout the law and do IQ testing.

If someone did a similar back-of-the-envelope calculation for "wrongful termination" lawsuits, I'm curious how that would turn out. Maybe employer fear of wrongful termination lawsuits is vastly exaggerated, but they still are averse to firing. But if this were the case it would undercut Caplan's story about IQ laundering, because it would demonstrate that employers sometimes have irrational fear of small costs. Caplan's estimate of the "IQ testing penalty" is based on legal defense costs and settlement amounts, but maybe the threat of bad press about being a racist employer is more salient in the minds of HR professionals than the explicit dollar costs.

Tuesday, November 20, 2018

You Need a Reason to Refuse an Immigrant Entry

I'm an advocate for open borders. I think it is the only moral immigration policy. It's the only policy that actually respects everyone's fundamental rights. Immigration restrictions violate the rights of citizens in the country that imposes restrictions. If I can't rent out an apartment to whomever I choose or hire whomever I choose to work at my business, my country has violated my property rights and my right to freedom of association. The would-be immigrant's rights are also violated, and to add injury to insult they are relegated to dire poverty (or horrifying civil war or corrupt institutions or political persecution).

But the term "open borders" freaks people out. I think some advocates brandish the term in a way that's deliberately provocative, like "Watch me bite this bullet!" It's like calling yourself an "atheist" just because the term sounds harsher than "agnostic", even when the second term gets the point across. If I were trying to make the case for open borders to someone who is very skeptical to the idea, I would frame it differently. I would try to frame it as: You need a reason to say "No" to this person. "Immigrants" aren't an undifferentiated mass of humanity. This is a person you're talking about. A migrant worker who wants to earn first-world wages and send remittances back to his family at home, or a mother who wants a better life for her children. If they want to come here, and someone who already lives here wants to host them and hire them (perhaps marry them, or be their colleague), then you need a reason to say "No" to this specific person. If you can't look that person in the eye and explain to them why your personal uneasiness with immigrants trumps their right to move here, you don't really have the right to tell them "No."

Folks in the anti-immigration crowd often point to crimes committed by immigrants as if that's a legitimate reason to restrict immigration (even though, by all accounts, immigrants commit fewer crimes than natives). Okay, if that's your reason, that's fine. Would you tell someone, to their face, "I'm sorry, but the social statistics imply that the probability that you will commit a crime is X%. I'm not willing to take that risk. Please return home to live in dire poverty." That's the policy decision we're making with immigration restrictions, so let's just be upfront and honest about it. We're effectively doing this millions of times over, but without the moral courage to actually say what we're doing.

Some anti-immigration folks are worried about, how should I say this, the "cultural makeup" of America (or whatever their home-country is). I don't particularly like it when people fling around loose accusations of racism, but in this case I can't help it. At least some people who are anti-immigration are worried that America will become "less Anglo" and become more Hispanic, Chinese, and Indian. (Scott Sumner has written insightfully on this point; restrictionists aren't worried about being swarmed by Canadians and Britons and Aussies, or even by Europeans. If you comment to tell me that you're in favor of immigration restrictions but you're not racist and the racial makeup thing doesn't bother you at all, that's fine, I believe you. I'm just pointing out that these people exist. I've literally heard them voice their concerns.) I still think you need a reason to tell someone "No." When faced with someone who would like to immigrate here, to work for a willing employer and rent from a willing land lord, you need to be able to say, "Your presence here pollutes a cultural ideal that I would like to preserve. I'm not willing to endure that cost."

Too often, immigration debates deal in aggregates. We need to do a better job of reminding people what the proposition is at the individual level. (Many open borders folks have actually been great on this point.) The crime and "cultural makeup" objections to immigration seem to be weighing insignificant costs against tremendous benefits. I don't see how these could yield a "No" answer for a typical immigrant in a fair process. (Saying "No" to a known criminal, on the other hand, might make sense, depending on the severity of the crime and recidivism risk.)

If "You need a reason to say 'No'" yields "We almost never actually say 'No'", and if "We never actually say 'No'" implies "Basically open borders", so be it. We followed a fair process to it's logical conclusion, even if the resulting policy is surprising to many people. Moral inquiry often leads to surprising conclusions. But I wouldn't start the conversation at "Open borders, or else you're an awful person."

With all that said, there are often more sensible solutions than just saying "No" and outright denying someone entry. "You can come here if you put up a bond, which you forfeit if you commit a crime." Or "You can come here if you secure employment and housing first." Or "You can come here if you forfeit your right to certain provisions of the welfare state." All of these conditions seem a little unfair and demeaning, but they are far better than an outright and unconditional "No!"

This Econtalk interview with Bryan Caplan made me an instant convert to open borders, without my having previously given the topic much thought. Caplan rests much of his argument on the unfairness of immigration restrictions to the individual.