It Seems Like Education is Mostly Socially Bad

Bryan Caplan’s The Case Against Education argues that education mostly serves a signaling function—it’s an easy way of proving to prospective employers that you’re a relatively smart, hard-working, mainstream member of society—and only a small part of education (maybe 20%) exists to help people learn anything or build any skills.

Excerpts from Julia Galef’s interview of Bryan Caplan:

Bryan Caplan: [… T]here’s a standard story that almost everyone tells about why education pays in the labor market, and it just says: you go to school, they pour some skills into you, you’re better at your job, and so you get paid more. What’s the problem?

And I’m happy to say, sure, that’s part of the story.

But I say there’s also a much bigger part of the story that rarely gets discussed, and that is that when you do well in school, you impress others. You get certification. You get stamped with a sign of approval saying “Grade-A Worker”. And my story is that the majority—in fact, a large majority—of the payoff from education actually comes from this.

Selfishly speaking, that doesn’t matter so much. But from a social point of view, it matters tremendously. Because if the reason why people get paid more for school is because they learn more skills, then basically it’s a way that taxpayers invest in our productive capacities and then we produce the very wealth that we are being paid for.

But on the other hand, in the signaling story, the main thing that’s going on is that you’re getting paid because you’ve impressed employers. And if everyone has a bunch of stickers on their head, this doesn’t mean everyone gets good jobs or gets paid a lot. It just means that you need a lot of stickers in order to get a job.

So the biggest sign of this, I would say, is what’s called a credential inflation, which is you just now need more education to get a job that your dad or grandfather could’ve gotten with one or two fewer degrees.

Julia Galef: And what kind of signal are you mostly pointing at? Is it the signal that someone was good enough to be accepted into a college, or the signal that someone was good enough to graduate with the grades that they did?

Bryan Caplan: Yeah, so the graduation seems like it’s a lot more important. Because if it were the first story, if it were just you get a great signal by being accepted, then people would take their admission letters and shop them around employers saying, “I got into Harvard and Stanford, so what are you going to offer me, Goldman Sachs?” And in practice, that doesn’t seem to work very well.

So I think if you’re wondering why, I would say that there’s something very odd about a person who tries to do that. They seem like they’re trying to skip out on this sacred institution of our society. So, yeah, employers are understandably nervous about someone so weird that they would get into Harvard and then try to weasel out of it.

So in terms of what is it people are signaling, I’d say it’s really a big package of different traits. Intelligence, obviously, but it’s not just that. That’s too easy to measure by itself. It’s also work ethic. And then finally, sheer conformity, which again, is very important on the job. Someone could be really smart and really hard-working, but if they’re defiant, if they don’t play as part of the team, then they’re almost useless to you. And I say really to understand a lot of what’s going on with education, we have to focus on this conformity signaling.

[… T]he whole idea of signaling is that if you come up with a really cheap way of signaling, the result isn’t that you get your signal across at a low cost, but rather that you just have to do more of it. A key idea in the signaling model is if you found a way of cutting the cost of signaling in half, the result wouldn’t be that we do half as much signaling. The result would be that we signal for twice as long. […]

My favorite example of this is suppose that someone comes up with a new way of making synthetic diamonds at 10% of the current cost. And my question is, how long would it take before people either stopped giving diamond engagement rings, or they started giving rings that were enormous?

And the key point is that since what you’re signaling with that ring is that you’re willing to go and put in a lot of money into something to indicate your devotion, if the cost per carat of diamond were to fall, it’s not that we would just keep giving the same diamonds that we’re currently giving. Instead, people would say, “Well that doesn’t really convince people very much anymore. It doesn’t say much anymore. I’d better go and either get an even bigger diamond or give something that can’t be synthesized.”

And a lot of it is really the same for education. If you were to go and have, say, free college for all, the result wouldn’t be that everybody with a college degree can get the kind of jobs that people get with it now. Instead there’d be an army of extra people going, and then you might need a Master’s degree or another advanced degree to be considered worthy of an interview.

[…M]ost specialists in both education and labor economics, they’re only looking at income. So they’re looking at the effects of education, and then there is this really circular effort to say, “Well, since there’s a big effect on the income of the person, they must’ve learned something useful,” and you say, “Yes, but the signaling model predicts the very same thing.” So that’s a big issue.

And there is an idea of, “Well of course we all know that the people are learning tons of useful stuff.” And then when you say, “Well, actually, they’re learning a ton of stuff they’re never gonna use.” And this is then where economists will often retreat to, “Oh, well, they’re learning how to learn, learning critical thinking, it doesn’t really matter what the subject is.”

And then I’ll say there’s something they really don’t know about, which is: in educational psychology, they’ve been studying this very issue for a hundred years. They want to find evidence of learning how to learn. They want to find evidence that critical thinking is being successfully taught. And yet, after a hundred years, they’re really pretty shell-shocked and say, “Look, we’re just not finding much sign of this broad, general inculcation of thinking skills that educators love to believe is actually happening.” So that’s the stuff I’d say most economists are just totally unaware of.

[… S]tudents seem so focused on getting easy As. If you were in school to acquire skills, this is pretty perverse. But if you’re in school to impress employers, then it’s pretty easy to understand why you want an easy A, because the employer doesn’t know that it was an easy A. If you find the easiest teacher of real analysis in the country, get an A+ in exchange for doing some arithmetic, people look at that and say, “Wow, he’s got an A plus in real analysis. Wow, look at that guy.” So that makes sense.

The practical implication: if, e.g., college as it exists in the real world is largely a zero-sum arms race to signal pre-existing traits (like intelligence and disposition to conform / accept instructions), rather than a positive-sum opportunity to actually learn useful or enriching material, then causing more people to go to college doesn’t improve people’s lives in aggregate.

Just the opposite, since college is expensive in time and money. If you get another 10% of people to go to college, then everyone else has to burn that many more resources to keep up in the signaling competition, but there’s still the same pool of new jobs, and people are still roughly as good at those jobs as they would have been without the education. Which means that everyone is now burning more resources just to not fall behind relative to everyone else in the ‘signal you’re a good worker’ game. Like forcing everyone in a race to run twice as fast, without doing anything to increase the reward for absolute performance at the end.

So, for example, subsidizing college education is a terrible idea that actively hurts people. A better case could be made, if anything, for taxing it as a source of net harm to society, to try to reduce how much time people spend at college etc. and thereby put more time, money, and other resources in people’s hands. Forcing poor people to get more years of education doesn’t appear to materially benefit them at all in aggregate, but handing them back money and free years of their life certainly does.

If you want to enrich people with cool ideas as an end in itself, because cool ideas are cool, then give people Internet access and free time and let them decide how to use that time. Don’t force them into camps where they have to learn classics and jump through hoops in order to be able to pay medical bills, start a family, etc. later in life.

From Caplan’s book:

Higher education is the only product where the consumer tries to get as little out of it as possible. […]

Some big blatant facts are inexplicable without the signaling model.

[1.] The best education in the world is already free. All complaints about elite colleges’ impossible admissions and insane tuition are flatly mistaken. Fact: anyone can study at Princeton for free. While tuition is over $45,000 a year, anyone can show up and start attending classes. No one will stop you. No one will challenge you. No one will make you feel unwelcome. Gorge yourself at Princeton’s all-you-can-eat buffet of the mind. Colleges do not card. I have seen this with my own eyes at schools around the country.

If you keep your learn-for-free scheme to yourself, professors will assume you’re missing from their roster owing to a bureaucratic snafu. If you ask permission to sit in, most professors will be flattered. What a rare pleasure to teach someone who wants to learn! After four years of ‘guerrilla education,’ there’s only one thing you’ll lack: a diploma. Since you’re not in the system, your performance will be invisible to employers.

[… 2.] Failing versus forgetting. You’ve studied many subjects you barely remember. You might have motivated yourself with, ‘After the final exam, I’ll never have to think about this stupid subject again.’

[… 3.] Easy As. Students struggle to win admission to elite schools. Once they arrive, however, they hunt for professors with low expectations. A professor who wants to fill a lecture hall hands out lots of As and little homework.

[… 4.] Cheating. According to human capital purists, the labor market rewards only job skills, not academic credentials. Taken literally, this implies academic cheating is futile. Sure, a failing student can raise their grade by copying an A+ exam or plagiarizing a term paper from the Internet. Unless copying and plagiarizing make people more productive for their employer, however, the human capital model implies zero financial payoff for the worker. […]

The human capital model doesn’t just imply all cheaters are wasting their time. It also implies all educators who try to prevent cheating are wasting their time. All exams might as well be take-home. No one needs to proctor tests or call time. No one needs to punish plagiarism—or Google random sentences to detect it. Learners get job skills and financial rewards. Fakers get poetic justice.

[… 5.] Teachers have a foolproof way to make their students cheer: cancel class. If human capital purists are right, such jubilation is bizarre. Since you go to school to acquire job skills, a teacher who cancels class rips you off. You learn less, you’re less employable, yet your school doesn’t refund a dime of tuition. In construction, contractors don’t jump for joy if their roofers skip shingling to go gambling. In school, however, students jump for joy if their teachers cancel class to attend a conference in Vegas.

When students celebrate the absence of education, it’s tempting to blame their myopia on immaturity. Tempting, but wrongheaded. Once they’re in college, myopic, immature students can unilaterally skip class whenever they like. Why wait for the teacher’s green light? For most students, there’s an obvious answer: When you skip class, your relative performance suffers. When you teacher cancels class, everyone learns less, leaving your relative performance unimpaired.

Human capital purists must reject this ‘obvious answer.’ Employers reward you for your skills, not your skills compared to your classmates’. Signaling, in contrast, takes the ‘obvious answer’ over the finish line. Why do students cheer when a teacher cancels class? Because they’ve escaped an hour of drudgery without hurting their GPA.

And another excerpt—the following is a relatively minor argument in a big 400-page book, but I’ve come back to it a few times, so I’ll put it here too. As a philosophy major, I get to do this without looking like I’m lording my major over others…

We can ballpark the practicality of higher education by looking at the distribution of majors. Table 2.1 breaks down all bachelor’s degrees conferred in 2008-9 by field of study—and rates their usefulness.

High usefulness: Defenders of the real-world relevance of education love to invoke engineering. Engineering students learn how to make stuff work; employers hire them to make stuff work. Engineering has well-defined subbranches, each with straightforward applications: electrical, mechanical, civil, nuclear. Before we get carried away, we should accept a key act: Engineering is a challenging, hence unpopular, major. Psychologists outnumber engineers. Artists outnumber engineers. Social scientists plus historians outnumber engineers almost two to one. […]

Medium usefulness: Majors like business, education, and public administration sound vaguely vocational and funnel students toward predictable occupations after graduation. At the same time, they teach few technical skills, and nonmajors readily compete for the same jobs. While you could dismiss these majors as Low in usefulness, let’s give them the benefit of doubt. You don’t need a business degree to work in business, but perhaps your coursework gives you an edge. You don’t need an education degree to land a teaching job, but explicitly studying education could enhance your teaching down the road. […] By this standard, about 35% of majors end up in the Medium category. […]

Low usefulness: The status of most of the majors in this bin [which contains 40% of all bachelor’s degrees]—fine arts, philosophy, women’s studies, theology, and such—should be uncontroversial. Liberal arts programs uphold the ideal of ‘knowledge for knowledge’s sake.’ Few even pretend to prepare students for the job market. You could argue I underrate the usefulness of communications and psychology. Don’t they prepare students to work in journalism and psychology? Yet this objection is almost as naive as, ‘Don’t history programs prepare students to work as historians?’ Psychology, communications, and history’s usefulness is Low because they prepare their students for fields where paying jobs are almost impossible to get. In 2008-9, over 94,000 students earned their bachelor’s in psychology, but there are only 174,000 practicing psychologists in the country. In the same year, over 83,000 students earned their bachelor’s degree in communications. Total jobs for reporters, correspondents, and broadcast news analysts number 54,000. Historians, unsurprisingly, have the bleakest prospects of all. There were over 34,000 newly minted history graduates—and only 3,500 working historians in the entire country. […]

The staunchest defenders of education reject the idea of sorting subjects and majors by ‘usefulness.’ How do you know Latin, trigonometry, or Emily Dickinson won’t serve you on the job? A man told me his French once helped him understand an airport announcement in Paris. Without high school French, he would have missed his flight. Invest years now and one day you might save hours at an airport. See, studying French pays!

These claims remind me of Hoarders, a reality show about people whose mad acquisitiveness has ruined their lives. Some hoarders collect herds of cats, others old refrigerators, others their own garbage. Why not throw away some of their useless possessions? Stock answer: ‘I might need it one day.’ They ‘might need’ a hundred empty milk cartons.

Taken literally, the hoarders are right: there is a chance they’ll need their trash. The commensense reply is that packing your house with trash is almost always a bad idea. You must weigh the storage cost against the likely benefits. […] ‘No one knows if this trash will come in handy’ is a crazy argument for hoarding trash. ‘No one knows if this knowledge will come in handy’ is a crazy argument for hoarding knowledge.

More discussion from Julia Galef’s podcast:

Julia Galef: You mentioned the case in which a teacher cancels class and the students are all happy about that and say like, “Jeez, if it was really about gaining skills that they expect to increase their productivity and value to future employers, then why would they be happy?”

You know, they already paid for tuition, and now they’re just getting less for their money. Which I do think is a suggestive and striking fact about the world.

But I felt like you didn’t quite give enough space to the alternate explanation of that—which is just, you know, people buy gym memberships because they want to lose weight or get fit, and then they find excuses not to go to the gym, or they’re happy when there’s a holiday and the gym is closed, so they don’t have to go to the gym.

It just feels like there’s this common phenomenon of a tension, of struggle between your present self’s interests and your future self’s interests, and this leads to a lot of behavior that otherwise looks irrational.

Bryan Caplan: Yeah, so I think I did have a couple sentences on that point, but you’re right, I could’ve talked more about it. […] But the main thing I say is that this myopia can explain why students don’t show up on a regular day. And yeah, typical college class in the middle of the semester, barely half the students are showing up. And that, I think you might say, “Well, it’s just myopia,” because they’re going and putting this money in, and they’re gonna get worse grades, and their life is going to be worse as a result.

But of course, there’s all the students who do show up, and why is it that those students are also happy when you cancel class? And that one seems to be that well, then I get to have this holiday without having to worry about the material that I failed to learn and that is going to lead me to get lower grades.

So yeah, I think in terms of just low attendance, you can explain it with myopia. But why people see a big difference between skipping class when everyone else is doing it, and skipping class when only half of the people are doing it, or when only you’re doing it—that’s where I think that you can detect the signaling element. It’s like, “I don’t mind missing it if everyone else misses it, but if I’m the only one missing it, then I’m dead, so no. I’ll go.” […]

Julia Galef: I just want to zoom out for a moment to note that I’ve been honing in on the parts of your argument that I find relatively less convincing—but I actually do find your argument overall pretty convincing. And if you’re correct about the standard [view] being closer to 10% signaling, I’m closer to 80% than 10%.

But yeah, I don’t know, I’ve just been thinking during our conversation about cruxes of disagreement between you and me. And I think probably one of them is I just expect that companies are less rational than you expect they are. And so I would just be less surprised if they were leaving large amounts of money on the table. Or less surprised if societal inertia or irrational biases were doing a lot of the work here. Which just changes the whole way you make sense of what’s happening.

Bryan Caplan: I mean, what’s funny is for an economist, I’d say I’m very open-minded about this stuff, and there are a bunch of cases where I’ll say, “Yeah, it looks like firms are actually not maximizing profits,” or, “They’re leaving money on the table.” But again, I think the cases that are well-documented are ones where it’s more marginal.

And there is actually a big body of literature on how firms that don’t maximize profits and have low productivity per worker have much higher attrition rates than other firms. And on the other end, the firms that have usually high productivity are just more likely to not only survive, but also to expand. It’s another thing to say, your firms are leaving money on the table for five years, but to say that it’s gone on for decades, again, this seems to go against most of what we know about selective attrition in growth of firms.

Julia Galef: Okay. Well, that’s a way bigger crux of disagreement than we can resolve in two minutes, so I’ll leave it at that. I just thought it was interesting to point out.

Bryan Caplan: Yeah, totally.

Julia Galef: And I want to make sure that I don’t forget to tell you about an ironic thing that I’ve noticed, that’s very relevant to your case, which is: Philosophy departments, in their “Why you should be a philosophy major” page on their departmental website, they always cite statistics about how there’s a high return to a philosophy major, in terms of the starting salaries you get offered. And they say, “See, this is proof that philosophy majors teach you critical thinking skills!”

Bryan Caplan: Yeah, that’s terrible.

Julia Galef: Which is especially ironic, because they’re confusing correlation and causation, which is an example of poor thinking skills in their very argument! That just struck me as a very Bryan-flavored observation.

Bryan Caplan: Yeah. Plus it’s not even true that a philosophy major is well-paid.

Julia Galef: Oh?

Bryan Caplan: It’s not at the bottom of the distribution by any means, but… actually, now that I think about it, normally the numbers that I look at actually correct for test scores. So it might be that, yeah, philosophers do come in with very high test scores. So it might be that if you just look at raw means, what you’re saying is true.

Julia Galef: Right, yeah.

Bryan Caplan: But if you go and look at how people who had the same test scores but who majored in something else do, then I think philosophy does pretty poorly. Especially if you’re not looking at people who go on to get a law degree or something like that. Those people are probably pulling up the average a lot.

Students who do unschooling seem to do totally fine; and differences between education approaches mostly don’t seem to change much. So on the face of it, mandating decades of universal formal education seems to just be burning value.

There may be subtle society-wide effects of forcing people to spend a large chunk of their life in something like a well-intentioned labor camp; but the balance of these effects might be on things like “how conformist society is in general” that I don’t think are good things to optimize for.

Some related discussion by Scott Alexander:


Odds and Ends – Mar 2021

A grab bag of interesting links and news:


The first COVID-19 human challenge trials have been approved… in February 2021, instead of February 2020.

If this is a good idea now, when the benefits are vastly lower and the risks are only slightly lower, then it was probably also a good idea a year ago. Opponents of human challenge trials should think hard about the background heuristics that caused them to get this one wrong, so we don’t have to repeat this tragic error again.

From Zvi Mowshowitz on Feb. 11:

The good news is that Johnson & Johnson has applied for emergency use authorization, and they are going to get it. There’s going to be a sprint to review the data in which every second will be used! Which started right after the official application, because you can’t review data that hasn’t been submitted in a completed application, that’s physically impossible, what are you even talking about. 

The bad news is that it’s going to take three weeks to get to the meeting, likely with additional time after the meeting before we can distribute the vaccine, but hey:

Last week I went over how we know the Johnson & Johnson vaccine is safe and effective, and there are millions of doses waiting to be distributed, and there’s no good reason we can’t start that process yesterday.

I do realize that there is a difference between, as Scott Alexander discusses, the FDA’s need to be legible and reliable, and follow proper procedures, versus my ability to apply Bayesian reasoning. 

That doesn’t mean this needs to happen three weeks after application, and attempts to justify that timeline are obvious nonsense

Mostly, it’s a call and response. You say ‘why are we letting people die for no reason?’ and they say ‘Thalidomide!’ and ‘people won’t trust it.’

So basically, one time someone had a drug that wasn’t safe. We didn’t approve that drug because our existing review process made it look unsafe, so in response to that we created a more involved and more onerous process, as opposed to noticing that the previous process actually worked in this case exactly as designed. Then we use this as a fully general excuse to freak everyone out about everything that hasn’t gone through this process, and then use that freak out (that, to the extent it exists which it mostly doesn’t, is directly the result of such warnings) as our reason to force everything through the process. Neat trick. 

Oh, and did I mention that the ‘safety data’ that requires three weeks to review is, and I quote it in its entirety, ‘nothing serious happened to anyone at all, and no one was struck by lightning.’ Either J&J has created a safe vaccine, or J&J  is committing a fraud that will be caught and get everyone involved arrested within three weeks, or they’re committing a fraud so effectively that the review won’t catch the fraud and won’t help. Those are the only possibilities. If the data isn’t fraudulent then the drug is safe, period. […]

On the actual J&J vaccine, I don’t know what more there is to say. As with Moderna and Pfizer, they’ve already done the actual approval process and confirmed that it’s going to get approved before they applied, and now we’re delaying in order to make it clear we are Very Serious People who Follow Proper Procedure and are not In Bed With Industry and Putting People At Risk or Destroying Trust in Vaccines by going ‘too fast.’ Or something like that. 

We have now done this three times. It’s one thing to have the first vaccine application point out that there’s weeks of lost time. It’s another thing to not have fixed the problem months later.


From Zvi on Mar. 4:

Meanwhile, now that we were provided a sufficiently urgent excuse that we were able to show that mRNA vaccines work, we’ve adopted them to create a vaccine for Malaria. Still very early but I consider this a favorite to end up working in some form within (regulatory burden) number of years. It’s plausible that the Covid-19 pandemic could end up net massively saving lives, and a lot of Effective Altruists (and anyone looking to actually help people) have some updating to do. It’s also worth saying that 409k people died of malaria in 2020 around the world, despite a lot of mitigation efforts, so can we please please please do some challenge trials and ramp up production in advance and otherwise give this the urgency it deserves? And speed up the approval process at least as much as we did for Covid? And fund the hell out of both testing this and doing research to create more mRNA vaccines? There’s also mRNA vaccines in the works for HIV, influenza and certain types of heart disease and cancer. These things having been around for a long time doesn’t make them not a crisis when we have the chance to fix them.


From Bryan Caplan on Feb. 16: Bioethics: Tuskegee vs. COVID. Excerpts:

[…] Why do bioethicists habitually invoke the Tuskegee experiment?  To justify current Human Subjects Review.  Which is bizarre, because Human Subjects Review applies to a vast range of obviously innocuous activities.  Under current rules, you need approval from Human Subjects merely to conduct a survey – i.e., to talk to a bunch of people and record their answers.

The rationale, presumably, is: “You should only conduct research on human beings if they give you informed consent.  And we shouldn’t let researchers decide for themselves if informed consent has been given.  Only bioethicists (and their well-trained minions) can make that call.”

On reflection, this just pushes the issue back a step.  Researchers aren’t allowed to decide if their human experiment requires informed consent.  However, they are allowed to decide if what they’re doing counts as an experiment.   No one submits a formal request to their Human Subjects Review Board before emailing other researchers questions about their work.  No professor submits a formal request to their Human Subjects Review Board before polling his students.  Why not?  Because they don’t classify such activities as “experiments.”  How is a formal survey any more “experimental” than emailing researchers or polling students?

[…] The safest answer for bioethicists, of course, is simply: “They should get our approval for those activities, too.”  The more territory bioethicists claim for themselves, however, the more you have to wonder, “How good is bioethicists’ moral judgment in the first place?”

To answer this question, let me bring up a bioethical incident thousands of times deadlier than the Tuskegee experiment.  You see, there was a deadly plague called COVID-19.  Researchers quickly came up with promising vaccines.  They could have tested the safety and efficacy of these vaccines in about one month using voluntary paid human experimentation.

[…] In the real world, researchers only did Step 1, then waited about six months to compare naturally-occurring infection rates.  During this period, ignorance of the various vaccines’ efficacy continued, almost no one received any COVID vaccine, and over a million people died.  In the end, researchers discovered that the vaccines were highly effective, so this delay really did cause mass death.

How come no country on Earth tried voluntary paid human experimentation?*  As far as I can tell, the most important factor was the formal and informal opposition of bioethicists.  In particular, bioethicists converged on absurdly (or impossibly) high standards for “truly informed consent” to deliberate infection. Here’s a prime example:

“An important principle in human challenge studies is that subjects must give their informed consent in order to take part. That means they should be provided with all the relevant information about the risk they are considering. But that is impossible for such a new disease.”

Why can’t you bluntly tell would-be subjects, “This is a very new disease, so there could be all sorts of unforeseen complications.  Do you still consent?”  Because the real point of bioethics isn’t to ensure informed consent, but to veto informed consent to whatever gives bioethicists the willies.

[…] I’ve said it before and I’ll say it again: Bioethics is to ethics as astrology is to astronomy.  If bioethicists had previously prevented a hundred Tuskegees from happening, COVID would still have turned the existence of their entire profession into a net negative for humanity.  Verily, we would be better off if their field had never existed.

If you find this hard to believe, remember: What the Tuskegee researchers did was already illegal in 1932.  Instead of creating a pile of new rules enforced by a cult of sanctimonious busybodies, the obvious response was to apply the familiar laws of contract and fiduciary duty.  These rules alone would have sent people like the Tuskegee researchers to jail where they belong.  And they would have left forthright practitioners of voluntary paid human experimentation free to do their vital life-saving work.

In a just world, future generations would hear stories of the monstrous effort to impede COVID-19 vaccine research.  Textbooks and documentaries would icily describe bioethicists’ lame rationalizations for allowing over a million people die.  If the Tuskegee experiments laid the groundwork for modern Human Subjects Review, the COVID non-experiments would lay the groundwork for the abolition of these deadly shackles on medical progress. […]


General COVID-19 thoughts, from me (someone with no relevant medical background):

1. I’ve head reports of people getting seriously ill or dying from various preventable illnesses, because they’re too scared to go to the hospital for various non-COVID-related ailments. In general, I think people are unduly scared of hospitals: going to the hospital is risky, but not catastrophically so. I’d advise people to stop going into grocery stores (if they can avoid it) long before I advised avoiding hospitals (in cases where they’re worried something might be seriously wrong).

Obviously, now is not the time to go in for routine check-ups, and video calls with doctors are a good first step in most cases, etc.

2. I’ve updated toward thinking it won’t be that hard to avoid catching COVID-19 in March/April in spite of the new strain, if you’re the kind of person who’s in a social network of very cautious people who have ~all avoided catching COVID-19 thus far. A large number of people are taking few or no precautions, and the bulk of COVID-19 exposures has been (and will continue to be) drawn from that group.

If you’re young and your whole social network has almost completely avoided anyone catching COVID-19 thus far, it’s more likely your social network is being over-cautious.

3. A lot of sources have been exaggerating the risk that you’ll be infected, or infect others, even if you’ve previously caught COVID-19 or been vaccinated. I think most people who’ve recovered from COVID-19 should mostly act as though COVID-19 doesn’t exist at all, at least for the next few months (in areas where the Brazil and South Africa strains aren’t widespread yet).

I’d say the same for people who have had two shots of the Pfizer or Moderna vaccine, as long as it’s been ~2 weeks since you had your second shot. For more detailed risk assessments than that, I recommend using the microCOVID website.

4. This is the home stretch. Universal vaccine availability is on the horizon, and our vaccines seem amazingly effective (especially for preventing deaths and hospitalizations). It goes without saying that it’s extra-unfortunate to catch COVID-19 shortly before you would have gotten vaccinated.


From Scott Alexander’s March link post:

CTRL+F “Blackrock” in this Matt Levine column for a discussion of how we accidentally stumbled into true communism for the good of all. The short version: an investing company called Blackrock owns so much of the economy that it’s in their self-interest to have all companies cooperate for the good of the economy as a whole. While they don’t usually push this too hard, the coronavirus pandemic was a big enough threat that “BlackRock is actually calling drug companies and telling them to cooperate to find a cure without worrying about credit or patents or profits”.


Also from Scott’s link post:

The class-first left’s case for why the Sanders campaign failed: he tried too hard to reinvent himself as a typical liberal to fit in, but people who wanted typical liberals had better choices, and it lost him his outsider energy (see especially the description of his “astoundingly dysfunctional” South Carolina campaign – “not only did basic tasks go unfulfilled, phone-banking and canvassing data were outright fabricated” – the article claims nobody was able to fix it because it was run by social justice activists who interpreted any criticism of them as racist/sexist. Interested to hear if anyone knows of other perspectives on this). Counterpoint: South Carolina was always going to be hostile territory for him, and maybe he didn’t reinvent himself as a typical liberal enough. I cannot find any other source confirming the South Carolina campaign allegations; interested in hearing what people think.


Marginal Revolution (also linked in Scott’s roundup) quotes Tanaya Devi and Roland Fryer’s “Policing the Police: The Impact of ‘Pattern-or-Practice’ Investigations on Crime”:

This paper provides the first empirical examination of the impact of federal and state “Pattern-or-Practice” investigations on crime and policing. For investigations that were not preceded by “viral” incidents of deadly force, investigations, on average, led to a statistically significant reduction in homicides and total crime. In stark contrast, all investigations that were preceded by “viral” incidents of deadly force have led to a large and statistically significant increase in homicides and total crime. We estimate that these investigations caused almost 900 excess homicides and almost 34,000 excess felonies. The leading hypothesis for why these investigations increase homicides and total crime is an abrupt change in the quantity of policing activity. In Chicago, the number of police-civilian interactions decreased by almost 90% in the month after the investigation was announced. In Riverside CA, interactions decreased 54%. In St. Louis, self-initiated police activities declined by 46%. Other theories we test such as changes in community trust or the aggressiveness of consent decrees associated with investigations — all contradict the data in important ways.

Neal Zupancic comments:

The authors seem to suggest it is mostly the investigations themselves causing the increase in crime, rather than any particular policy changes. The mechanism they propose is that police officers greatly reduce their quantity of policing when under federal investigation after a “viral” incident, but there is little indication that this comes about as the result of any particular policy reform – the suggestion is that police are either reducing public contact in an effort to avoid having their own actions scrutinized, or are trying to make a point (in the case of deliberate strikes and slowdowns/sickouts). There’s also a section (page 27) where the authors talk about the possible impact of increased paperwork, and estimate it might account for about 20% of the reduction in police activity in one city. I’m not sure if we’re calling this “reform” but even if we do it’s a small proposed effect.