Amish health care spending

Scott Alexander writes,

Amish people don’t have health insurance, and pay much less than you do for health care. But their health is fine. What can we learn?

He explores various hypotheses, but he does not address my main hypothesis.

In 2006 when I wrote Crisis of Abundance, I asked why health care cost more in 2006 than in 1970. I determined that much of the answer was what I called “premium medicine,” which uses expensive equipment and medical specialists. You can think of getting an MRI when you hurt your back or getting a routine coloscopy screening as recommended at age 65.

Premium medicine contributes a lot to health care spending but relatively little to aggregate health outcomes. Those outcomes are affected more by lifestyle considerations. If in a society you increase substance abuse at the same time as you increase overall use of premium medicine, the net impact on longevity will not be particularly positive.

The Amish philosophy on technology in general is to be late adopters. That is, they do adopt new inventions, but the process is very gradual. My guess is that they are a lot less inclined to use premium medicine.

In my book, I pointed out that in 2006 we could easily afford health insurance that covered only those procedures that were prevalent in 1970. Perhaps the Amish are following that approach.

Eric Weinstein and Tyler Cowen, partially annotated

The podcast is here.

For the first hour, it reminds me of conversations I occasionally had at home as a teenager, in which I would climb onto an intellectual ledge and my father would try to gently talk me down. Here, Tyler plays the role of my father.

At one hour and six minutes, Tyler himself climbs onto a ledge. In talking about the stagnation that began in 1973, he speaks of the “feminization” of our society.

One possibility is that he was thinking, “Our society reduced its rate of risk-taking and novelty-seeking. Women tend to like risk-taking and novelty-seeking less than men. Therefore, it is fair to speak of feminization.”

The first two statements are controversial, but suppose that they are true. The conclusion still does not follow. We need some link that connects the premises to the conclusion at the level of society. Did women, starting around 1973, acquire significant power to direct corporate investment and/or the regulatory apparatus? Or did men in these positions lose their, er, manhood around this time? I don’t think that’s a ledge you want to stand on.

Another interpretation of “feminization” works much better but is somewhat less interesting. That is, we can say that because of the decline of manufacturing and the rise of the New Commanding Heights industries of education and health care, employment opportunities for women improved while those for men worsened. At the same time, widening the door for women to enroll in universities was like opening up professional sports to African-Americans in that the formerly-excluded were able to compete effectively and some of the formerly-protected were left worse off.

This latter interpretation allows “feminization” to explain why earnings of the median working-age male have shown at best disappointing growth over the past 35 years. But it does not work as a broad sociological explanation for other phenomena, such as a slowdown in scientific discovery or an apparent decline in the productivity of civil engineering.

About 15 minutes later in the podcast, Eric tries to interest Tyler in a comparison of mainstream and heterodox thinkers. Tyler will have none of it. He says that we are the last generation that will understand the distinction. His view appears to be that institutional brand names, such as “New York Times” or “Harvard Economics Department” will not impress the Internet generation.

So where does that lead? What becomes of what Eric would call society’s “sense-making apparatus”? One scary scenario is that it doesn’t get any better than it is today, so that the loss of the information Leviathans with which we grew up will lead to a sort of Hobbesian “war of all against all.” A more optimistic scenario would be a “cream rises” outcome in which to attain broad credibility you have to rise to a very high level of intellectual rigor. Think of Scott Alexander as an example.

On Tom Wolfe, Ken Kesey, and LSD

Scott Alexander writes,

The best I can do in making sense of this story is to think of Kesey as having unique innate talents that made him a potential cult leader, combined with the sudden rise in status from being a famous author and the first person in his social scene with access to LSD. Despite the connotations of “cult leader”, Kesey was overall a good person, genuinely wanted to help people’s spiritual development, and genuinely thought LSD could do this. LSD formed the content of his cult, the same way Messianic Judaism formed the content of Jesus’ cult. It also made his life easier because of the drug’s natural tendency to make people think they are having important insights. When he, attempting to genuinely discover a spiritual path, decided to change the content and go beyond LSD, he lost that crutch, his people betrayed him, he became less confident in himself, and eventually he gave up.

My interpretation of Kesey is a bit different. He could only depict the power struggle between Nurse Ratched and Randle McMurphy because he understood the strengths of both. (Note that more than one pundit has seen an echo in the contest between Hillary Clinton and Donald Trump.) So he had a natural sense of personal power, but I think in the end he did not want it. Maybe it bored him. Maybe having a cult following even turned him off after a while. These days, you can watch Jordan Peterson and suspect that he sometimes longs to retire to a life of isolation and anonymity, which is what Kesey did.

When I was in high school, I read and re-read Jack Kerouac’s On the Road, Kesey’s One Flew Over the Cuckoo’s Nest, and Tom Wolfe’s The Electric Kool-Aid Acid Test. Alexander gives a good plot summary of the latter. Of the three, I would say that Cuckoo’s Nest holds up the best, followed by Acid Test.

I recommend Alexander’s whole post. The key paragraph:

One of two things must be true. Either psychedelics are a unique gateway to insight and happiness, maybe the most powerful ever discovered. Or they have a unique ability to convince people that they are, faking insight as effectively as heroin fakes happiness. Either one would be fascinating: the first for obvious reasons, the second because it convinces some pretty smart people. If the insight of LSD were fake, its very convincingness could tell us a lot about the mind and about how rationality works.

My money is on the latter.

Cultural evolution vs. memetic evolution

Scott Alexander writes,

Cultural evolution may be moving along as lazily as always, but memetic evolution gets faster and faster. Clickbait news sites increase the intensity of selection to tropical-rainforest-like levels. What survives turns out to be conspiracy-laden nationalism and conspiracy-laden socialism. The rise of Trump was really bad, and I don’t think it could have happened just ten or twenty years ago. Some sort of culturally-evolved immune system (“basic decency”) would have prevented it. Now the power of convincing-sounding ideas to spread through and energize the populace has overwhelmed what that kind of immunity can deal with.

Think of gender roles. For many generations, they evolved very slowly. The pace of change in the twentieth century, which seemed rapid at the time, seems glacial by today’s standards. Back then, women steadily increased their participation in the work force. Over a period of decades, sexual taboos came to be relaxed, notably concerning divorce and pre-marital sex. Next came gay liberation, which took place roughly from 1970 to 2000.

But in the last five years, the memetic evolution has sped up enormously. It seems like we’ve had a new cool gender-identity flavor every month, and even “ordinary” gays are feeling as threatened as old-fashioned straights.

We have no idea whether these trendy gender fluidity memes represent progress. I certainly have my doubts. But it feels to me as if our culture is a passenger in a car with no brakes.

I agree that a Trump presidency would not have been possible a dozen years ago. To the Claremont folks, his victory is our way of trying to stop the runaway car. But I think it is more plausibly explained by Martin Gurri’s idea of the revolt of the public, made possible by the new media environment. The car is still going ahead full speed, just without the support of the Secretary of Education–for now.

Until very recently, the party elites and the mainstream media were powerful enough to prevent an outsider rebel like Mr. Trump from gaining a major party nomination, if they wanted to do so. Goldwater and McGovern made it past the party establishments, but each of them claimed to be aligned with the establishment in a more pure form, which made the establishment unwilling to wholeheartedly resist. Another reason that the establishment put up weak resistance to their insurgencies was that in both cases they expected to lose the general election, anyway.

Mr. Trump’s approach to politics is more personal than ideological. The establishment resistance to him was more highly motivated than the establishment resistance to Goldwater or McGovern, but it was utterly ineffectual.

Questioning the Baumol-effect story

Scott Alexander writes,

Factory workers are not getting paid more. That makes it hard for me to understand how rising wages for factory workers are forcing up salaries for violinists, teachers, and doctors.

. . .College really does seem to be getting less affordable. So do health care, primary education, and all the other areas affected by cost disease. Baumol effects shouldn’t be able to do this, unless I am really confused about them.

Suppose that the economy consists of apples and string quartets, and productivity doubles in apple picking. The Baumol-effect story is that we are now richer, and we can afford to spend more on both apples and string quartets. The increased spending on apples is more than offset by the higher productivity, so apple prices fall. But the productivity of violinists stays constant, so the increased spending on string quartets causes their prices to rise.

As I see it, Alexander is asking: if this is the scenario, then why does it seem as though the apple pickers have not gotten richer?

In a straightforward Baumol-effect story, when the productivity windfall hits the apple industry, some workers should be released from the apple-picking sector to work as violinists, so that we now have more string quartets as well as more apples. Everyone is richer.

Instead, Alexander’s data and anecdotes seem to indicate that we have had a big redistribution of income away from apple pickers and toward violinists. How do you get that? A combination of very inelastic demand for apples and little ability to shift from apple picking to string-quartet playing? That would seem necessary, but it may not be sufficient.

Note that in our national economic data, concepts like “real wages” may be calculated using price indexes that are constructed in a way that treats the demand for apples as totally inelastic, regardless of whether this is actually true. So perhaps the absence of real wage growth in the data is a mere statistical artifact, which opens up a different kettle of worms entirely.

So here is the issue: if Tabarrok and Helland are correct that the Baumol effect explains rising prices in health care and education, then it seems that we should have observed broad-based increases in real incomes. Instead, what we seem to have experienced is a significant redistribution of incomes toward the providers of services in health care and education. If so, then the Baumol-effect story may not suffice, and we need another explanation.

“Subsidize demand, restrict supply” comes to mind.

Tradition or momentary reason?

This post is inspired by a lot of recent reading, too much to reference here. Some of it pertains to Sohrab Ahmari David French. But most of it pertains to Scott Alexander’s recent posts inspired by Joseph Henrich’s work. (Note that I also praised the Henrich book myself.)

In the latter post, Scott writes,

We are the heirs to a five-hundred-year-old tradition of questioning traditions and demanding rational justifications for things. Armed with this tradition, western civilization has conquered the world and landed on the moon. If there were ever any tradition that has received cultural evolution’s stamp of approval, it would be this one.

Sometimes, there is a conflict between the approach that you arrive at using your reasoning of the moment and the existing tradition. For example, Bryan Caplan argues that a reasoning libertarian should oppose immigration restrictions.

Under such circumstances, which should prevail: your momentary reason or tradition?

Conservatives argue for paying considerable respect to tradition. Your individual, momentary reason is not sufficient to overwhelm generations of experience. Henrich’s anthropology supports that (although Henrich does not define himself as a conservative). Always going with momentary reason would mean depriving ourselves of cultural intelligence.

But obviously, if you always go with tradition, you never evolve in a better direction. So you want some experimentation.

The Whig history is that our current society reflects retention of successful experiments. The dour conservative point of view is that it has all been downhill since. . .the radical Social Justice turn of the last five years. . .or the 1960s. . .or Rousseau. . .or John Locke. Take your pick.

A few hundred years ago, a lot of cultural transmission depended on the elderly. Old people knew more than young people, so it was hard for young people to question tradition.

Today, old people don’t know how to use smart phones as well as young people do. So why should young people think old people aren’t equally antiquated on issues of race relations, gender, or free speech?

I wish that old people and traditions had somewhat higher status than they do with young progressives, and I wish that momentary reason had somewhat lower status.

UPDATE: After I wrote this post but before it was scheduled to appear, Scott Alexander elaborated further. I will have another post on this tomorrow soon.

The genes that did not matter

For predicting depression. The authors of this study report

The implication of our study, therefore, is that previous positive main effect or interaction effect findings for these 18 candidate genes with respect to depression were false positives. Our results mirror those of well-powered investigations of candidate gene hypotheses for other complex traits, including those of schizophrenia and white matter microstructure.

Read Scott Alexander’s narrative about their findings.

As I understand it, a bunch of old studies looked at one gene at a time in moderate samples and found significant effects. This study looks at many genes at the same time in very large samples and finds that no one gene has significant effects.

The results are not reported in a way that I can clearly see what is happening, so the following is speculative:

1. It is possible that the prior reports of a significant association of a particular gene with greater incidence of depression are due to specification searches (trying out different “control” variables until you find a set that produces “significant” results).

2. It is possible that publication bias meant that although many attempts by other researchers to find “significant” results failed, those efforts were not reported.

3. These authors use a different, larger data sample, and perhaps in that sample the incidence of depression could be measured with greater error than in the smaller samples used by previous investigators. Having a larger data sample increases your chance of finding “significant” results, but measurement error reduces your chances of finding “significant” results. The authors are aware of the measurement-error issue and they conduct an exercise intended to show that this could not be the main source of their failure to replicate other studies.

4. If I understand it correctly, previous studies each tended to focus on a small number of genes, perhaps just one. This study includes many genes at once. If my understanding is correct, then in this new study the authors are now controlling for many more factors.

Think of it this way. Suppose you do a study of cancer incidence, and you find that growing up in a poor neighborhood is associated with a higher cancer death rate. Then somebody comes along and does a study that includes all of the factors that could affect cancer incidence. This study finds that growing up in a poor neighborhood has no effect. A reason that this could happen is that once you control for, say, propensity to smoke, the neighborhood effect disappears.

In the case of depression, suppose that the true causal process is for 100 genes to influence depression together. A polygenic score explains, say, 20 percent of the variation in the incidence of depression across a population. Now you go back to an old study that just looks at one gene that happens to be relatively highly correlated with the polygenic score.

In finance, we say that a stock whose movements are highly correlated with those of the overall market is a high-beta stock. The fact that XYZ corporation’s share price is highly correlated with the S&P 500 does not mean that XYZ’s shares are what is causing the S&P to move. Similarly, a “high-beta” gene for depression would not signify causality, if instead a broad index of genes is what contributes to the underlying causal process.

Further comments:

(1) and (2) are fairly standard explanations for a failure to replicate. But Alexander points out that in this case it is not just one or two studies that fail to replicate, but hundreds. That would make this a very, very sobering example.

If (3) is the explanation (i.e., more measurement error in the new study), then the older studies may have merit. It is the new study that is misleading.

If (4) is the explanation, then the “true” model of genes and depression is closer to a polygenic model. The single-gene results reflect correlation with other genes that influence the incidence of depression rather than direct causal effects.

If (4) is correct, then the “new” approach to genetic research, using large samples and looking at many genes at once, should be able to yield better predictions of the incidence of depression than the “old” single-gene, small-sample approach. But neither approach will yield useful information for treatment. The old approach gets you correlation without causation. The new approach results in a causal model that is too complex to be useful for treatment, because too many genes are involved and no one gene suggests any target for intervention.

I thank Russ Roberts for a discussion last week over lunch, without implicating him in any errors in my analysis.

The null hypothesis for policy

Scott Alexander writes,

the same argument that disproves the importance of photolithography disproves the importance of anything else.

His post gives a number of examples where progress follows a straight line. This is sometimes used as an argument that no individual policy (or invention, as in the case of photolithography) matters. Alexander wonders whether we are deceiving ourselves into believing the null hypothesis for policy.

I think that in the case of inventions it can be difficult to discern an effect at the point in time when the invention occurs. The process of developing complementary inventions, adapting to the new technology, and achieving widespread adoption takes time. See the work of economic historian Paul David. As a result, even in a world of discrete innovations, the overall path of progress is smooth.

In the case of policy, I think that one must also allow for time lags. For example, changes in labor market incentives may not have large effects in the short run, but over time the culture can be affected.

But in general, I think that if one fails to see any historical break point in an outcome following the adoption of a policy, that justifies a presumption that the policy did not better. I would suggest more careful analysis if that is possible. A clever researcher may be able to find a “natural experiment” that has more power against the null hypothesis. For example, Tyler Cowen posted about a study that found that a carbon tax had little effect on carbon dioxide emissions by comparing across regions. In principle, that study provides more persuasive evidence that the null hypothesis holds for the carbon tax.

Wages and productivity

Scott Alexander writes,

Median wages tracked productivity until 1973, then stopped. Productivity kept growing, but wages remained stagnant.

This is called “wage decoupling”. Sometimes people talk about wages decoupling from GDP, or from GDP per capita, but it all works out pretty much the same way. Increasing growth no longer produces increasing wages for ordinary workers.

Is this true? If so, why?

He makes a valiant effort to summarize and assess the economic literature. But this is where orthodox economics is hopeless.

Productivity by definition is output divided by the amount of labor input. Let me make three points:

1. You can’t measure the numerator very well.

2. You can’t measure the denominator very well.

3. The U.S. is not just one big GDP factory. Both the numerator and the denominator are affected by shifts in the composition of the economy, even if actual productivity and wages were not changing at all.

The numerator is output. How many people work in businesses with measurable output? Scott Alexander doesn’t. I never have. Most of my readers never have. There are entire industries, like health care, education, and finance, where we do not have any idea how to measure output. And even within an industry that has quantifiable output, we still have the issue that, as Garett Jones pointed out many years ago, most workers are not engaged in actual production; they are building organizational capabilities. Even if the factory managers can count widget production, they cannot measure the productivity of the tax accountants or of the team developing a new marketing initiative.

The denominator is labor input. But most of labor input consists of human capital. To measure labor input, you need to be able to measure quality, not just quantity. What is the incremental value of X years of schooling and Y years of experience? We do not have a reliable way to do that. One approach uses wage rates as an indicator of quality, but that amounts to assuming that productivity and wage rates are tightly coupled, but that amounts to assuming away the question that Alexander is raising.

We are not in a GDP factory. As the share of GDP devoted to health care and education goes up and the share devoted to manufacturing goes down, we are giving more weight to a sector where real output and the quality of labor input are extremely difficult to measure.

I think that for economists to say anything useful about productivity and wages, they should try to study individual units of individual firms. My guess is if you were to undertake such a study, you would be overwhelmed by doubts about the precision of your measurements and the difficulty of obtaining a decent signal-to-noise ratio. It’s perverse that you would instead look at the aggregate statistics cranked out by the Commerce Department and the Labor Department and pretend that it’s 100 percent signal.

Government is a branch of culture

Scott Alexander proposes that we think of culture as a branch of government.

Each branch of government enforces rules in its own way. The legislature passes laws. The executive makes executive orders. The judiciary rules on cases. And the culture sets norms. In our hypothetical world, true libertarians are people who want less of all of these. There are people who want less of the first three branches but want to keep strong cultural norms about what is or isn’t acceptable . . .The real libertarians also believe that cultural norms enforced by shame and ostracism are impositions on freedom, and fight to make these as circumscribed as possible.

Sounding like one of Alexander’s “real libertarians,” one of my commenters complains about,

. . . A tiny subset of the population, media-ready and always on, always practicing PR, permanently in performance mode, not expressing themselves except in precisely those expressions that can be guaranteed to win the approval of the bigots and authoritarians who appointed themselves the police of society. Enforcers of conformity. Stamping out creativity. Stomping on self-expression.

I think it is best to take Scott Alexander’s view and turn it around: government is a branch of culture. I suggest defining culture as socially communicated practices and beliefs. We may think of government as the subset of practices and beliefs that are defined formally and enforced coercively.

Take property rights. We can think of them as culturally defined, even in the absence of government. But property rights take on more significance when the government establishes and enforces them. De Soto in The Mystery of Capital argues that without formal property rights an economy cannot develop properly.

Just as an economy has both a formal sector and an informal sector, culture has both a formal and an informal sector. The formal sector is where norms are enforced by government.

As a metaphor, think of footpaths. The paths where people walk are culture. Those paths that are not paved are the informal sector. Those paths that are paved are the formal sector.

Alexander argues that the proper libertarian position is to oppose enforcement of social norms, either formally or informally. But you cannot have a society without social norms, and you cannot have meaningful social norms without enforcement.

I think that a more viable libertarian position is that where social norms are contested, contests should be resolved peacefully. You don’t want the Protestants and Catholics burning heretics and fighting civil wars. But if Protestants want to engage in nonviolent attempts to set standards of behavior for Protestants and to convert Catholics to those standards, then that is ok.

A hard case for libertarians is when Google fires James Damore on religious grounds. Whose religious freedom should concern us most, Google’s or Damore’s? And once we choose sides, do we want the formal cultural institution, namely government, to enforce our point of view?

Libertarians often seek black-and-white answers, but I don’t think they are always easy to find.