Category Archives: links to my essays

Basic econ: costs of production

Posted on by

I am starting to work on filling in/updating some of these college economics topics. Students land on the site when they want to get help with their college econ courses. But I plan to include some occasional “improvements” to mainstream thinking. I did not much need to amend mainstream thinking in my first topic, costs of production.

Other things equal, when fixed costs are high, there will be only a few firms. When fixed costs are low, there will tend to be many firms. When the Internet reduced the fixed cost of becoming a publisher, because you no longer need a printing press, the number of providers of written content skyrocketed.

Why are polygenic scores not better?

Posted on by

Start with what I said in my review of Robert Plomin’s Blueprint.

Plomin is excited by polygenic scores, a recent development in genetic studies. Researchers use large databases of DNA-sequence individuals to identify combinations of hundreds of genes that correlate with traits.

The most predictive polygenic score so far is height, which explains 17 percent of the variance in adult height… height at birth scarcely predicts adult height. The predictive power of polygenic scores is greater than any other predictors, even the height of the individuals’ parents.

One can view this 17 percent figure either as encouraging or not. It represents progress over attempts to find one or two genes that predict height, an effort that is futile. But compared to the 80 percent heritability of height it seems weak.

Plomin is optimistic that with larger sample sizes better polygenic scores will be found, but I am skeptical.

My question, to which I do not have the answer, is this: if height is 80 percent heritable, why is the statistical correlation found between genes and height only 17 percent?

I do not know any biology. But as a statistician, here is how I would go about developing a polygenic score.

1. I would work with one gender at a time. Assume we have a sample of 100,000 adults of one gender, with measurements of height and DNA sequences. I would throw out the middle 80,000 and just work with the top and bottom deciles.

2. For every gene, sum up the total number in the top decile with that gene and the total number in the bottom decile with that gene, and see where the differences are the greatest. If 8500 in the top decile have a particular gene and 1200 in the bottom decile have the gene, that is a huge difference. 7500 and 7200 would be a small difference. Take the 100 largest differences and build a score that is a weighted average of the presence of those genes.

3. To try to improve the score, see whether adding the gene with the 101st largest difference improves predictive power. My guess is that it won’t.

4. Also to try to improve the score, see whether adding two-gene interactions helps the score. That is, does having gene 1 and gene 2 make a difference other than what you would expect from having each of those genes separately? My guess is that some of these two-gene interactions will prove significant, but not many.

It seems to me that one should be able to extract most of the heritability from the data by doing this. But perhaps this approach is not truly applicable.

Another possibility is that heritability comes from factors other than DNA. Perhaps the reliance on twin studies to try to separate environmental factors from genetic factors is flawed, and the heritability of height comes in large part from environmental factors. Or perhaps DNA is not the only biological force affecting heritability, and we need to start looking for that other force.

Another possibility is that scientists are working with much smaller sample sizes. If you have a sample of one thousand, then the top decile just has one hundred cases in it, and that is not enough to pick out the important DNA differences.

As a related possibility, the effective sample sizes might be small, because of a lot of duplication. Suppose that the top decile in your sample had mostly Scandinavians, and the bottom decile had mostly Mexicans. Your score will be good at separating Scandinavians from Mexicans, but it will be of little use in predicting heights within a group of Russians or Greeks or Kenyans or Scots.

I am just throwing out wild guesses about why polygenic scores do not work very well. I probably misunderstand the problem. I wish that someone could explain it to me.

On Mary Eberstadt’s latest book

Posted on by

In a review of Primal Screams, I wrote,

When a tribe is formed out of families, members feel secure in their status. One’s identity is established as a father, mother, sibling, uncle, aunt, or grandparent.

In contrast, when a “forced pack” is constructed out of isolated individuals, there are constant struggles to resolve the uncertainty over who belongs and where members fit in relation to one another. Eberstadt suggests that under such circumstances:

… some people, deprived of recognition in the traditional ways, will regress to a state in which their demand for recognition becomes ever more insistent and childlike. This brings us to one of the most revealing features of identity politics: its infantilized expression and vernacular.

Her thesis, about which I raise doubts in my review, is that young people turn to identity politics to try to address needs that are unmet in today’s weak family environment. I can imagine Eberstadt reading the David Brooks essay to which I referred last week and coming out with her own primal scream.

Me vs. the DISC

Posted on by

1. One of Eric Weinstein’s catch-phrases is the DISC, which I think stands for the Distributed Information Suppression Complex.

2. Recently, I was asked if I want to contribute some sections to a guide for college students of first-year economics. In looking at the guide, I was reminded of my frustrations with mainstream economics. The GDP factory. The failure to appreciate intangible factors. The failure to incorporate the business problems posed by the Internet into mainstream courses. My seemingly hopeless moonshot to overthrow neoclassical economics. My attempt with Specialization and Trade that fell with a thud. etc.

3. One idea that I extracted from Jeffrey Friedman’s turgid prose is that the economics profession probably selects for those who believe in and desire technocratic power. That seems to me what drives the DISC in economics. It leads to things like Raj Chetty’s project.

A central part of Opportunity Insights’ mission is to train the next generation of researchers and policy leaders on methods to study and improve economic opportunity and related social problems. This page provides lecture materials and videos for a course entitled “Using Big Data Solve Economic and Social Problems,” taught by Raj Chetty and Greg Bruich at Harvard University.

Gosh, if you were to just link data from tax returns, credit bureaus, and Google searches, imagine how well “seeing like a state” could work. Ugh.

4. Unfortunately, I am Bill. Let me tell you the story of Bill. In 1990, I was promoted to a low-level management position in charge of five people inside Financial Research at Freddie Mac. One of the staff I inherited was Bill. Bill was a very bright guy, the sort who is called a “computer genius” by people who are intimidated by computers, and even by some who are not intimidated. He was older, in his fifties, with the title of “economist” but doing the work of a glorified research assistant. Bill had bounced around different departments at Freddie Mac, as one supervisor would unload him for his performance issues and another would pick him up for his potential and background.

Bill was very popular with the other staff. When they had a gnarly problem in SAS or with installing new software on a PC (this was a challenge in those days), he would help. Unfortunately, he found these problems so interesting that he would gladly drop whatever assignment you gave him in order to work on the tech issues. So if he was supposed to run a report that I needed for a meeting with top management the next day, I could not count on him to do it. He was very distractable.

One day, he distractedly wandered through the tape library for Freddie Mac’s mainframe computers. I have no idea why. He pulled down a tape and, lo and behold, he found data that had been missing for years. It was data from loans that were originated in the late 1970s and early 1980s. The data was no longer needed for processing the loans, but it was priceless for research purposes. We could now correlate default rates to data from loan applications, such as the original loan-to-value ratio.

I soon hired another research assistant, Sudha. She was far from brilliant, and her computer skills were weak, but she was meticulous and organized. The other staff, who loved Bill, resented Sudha, especially because Bill always ended up doing the work for Sudha’s memos. But when I left my position, my replacement soon said to me, “Now I understand what you were doing. You needed Sudha in order to get Bill’s projects done.”

So I am Bill. I am distractable. That is who I am. That is where I live. Being distractable perhaps enables me to discover insights. But it also is a weakness. If I were like Bryan Caplan, I would spend several years delving deeply into a topic and come out with a compelling book. Maybe somebody needs to find a Sudha to pair with me.

The essay on the null hypothesis and Charles Murray

Posted on by

I am posting it below, because so many readers complained about Thinkspot. It is true that Thinkspot is not in a satisfying state as is. Please comment only the essay. I will put up a separate post on the issues with Thinkspot.

1.   If the shared environment explains little of the variance in cognitive repertoires, and
2.   If the only environmental factors that can be affected by outside interventions are part of the shared environment,
3.   Then outside interventions are inherently constrained in the effects they can have on cognitive repertoires.
–Charles Murray, Human Diversity, Chapter 13.

As an example of an outside intervention, consider reading to pre-school children.   Researchers have observed that pre-school children who have been read to a great deal by their parents subsequently perform better in school than students who have not been read to as much.
 
But this relationship is not necessarily causal.   It could be that the better school performance is due to inherited characteristics that are correlated with how much reading the parents do to their pre-school children.   In order to establish causality, one would have to conduct an experiment in which children are randomly selected into a control group that receives little reading and a treatment group that receives a lot of reading.   
 
If such an experiment were conducted, my prediction is that the effects on the treatment group would be.
 
–small to begin with.
–fade out completely within a few years, meaning that by, say, fourth grade, the treatment group and the control group show no difference.
–to the extent that the effects were non-zero and did not fade out, the results would fail to replicate in a subsequent experiment.
 
I call this prediction The Null Hypothesis, borrowing the statistical term for “no effect of the treatment.”   My reading of the literature on educational treatments is that the null hypothesis essentially always holds.   When a treatment is rigorously tested, using experimental methods, its effects are small, fade-out is complete, and/or the results fail to replicate.
 
Why does the null hypothesis hold for educational treatments (and, incidentally, for other policy treatments, such as the effect of job training programs on subsequent employment or the effect of health insurance on health outcomes)?  Consider four factors that affect human outcomes:.
 
1.   Overall cultural environment.
2.   Genetic inheritance.
3.   Gestational variation.
4.   Specific environmental interventions.
 
I believe that I have presented these in order of importance.  
 
The overall cultural environment, or “milieu” as Murray calls it, clearly matters.   If you could transport one of your children to a different historical period or to a totally different society, then you can be sure that the child’s outcomes will be affected. The Flynn Effect, in which average IQ changes across generations, is indicative of the importance of the cultural environment.   I think it only makes sense to talk about variations of the other three factors within a given environment, such as the affluent countries in the 21st century.
 
The significance of genetic inheritance is what Murray highlights.     The evidence from twin studies is persuasive in that regard.
 
Murray does not discuss gestational variation, but Kevin Mitchell’s Innate highlights its importance.   Mitchell argues that some of the variation between identical twins in cognitive repertoires is due to mutations and other accidents that occur as the fetal brain forms.  
 
In twin studies that account for variation as the sum of genetic variation and variation in the “shared environment,” the innate gestational variation tends to be misleadingly attributed to the “shared environment” component.   I believe that this leads people to be more optimistic about the potential for specific interventions than is warranted.
 
In my view, once we have accounted for the differences created by the overall cultural environment, genetic inheritance, and gestational variation, there is very little room for specific interventions to make a difference.   In The Nurture Assumption, Judith Rich Harris pointed to evidence that parental behavior makes little difference in children’s outcomes.   If the people who are most heavily involved in raising children make little difference, then what is the likelihood that, say, a particular elementary school teacher or a specific schooling method will make a difference?
 
I know that there are studies that purport to find exceptions to the Null Hypothesis.   Such studies receive wide acclaim.   But these tend to be one-off results that do not replicate.
 
I plan to write subsequently on points where my view differs from Murray’s.   But on the Null Hypothesis, my views are coherent with his.  

My critique of Case and Deaton

Posted on by

Mercatus titled it Death and Politics.

Their new book, Deaths of Despair and the Future of Capitalism, includes both an actuarial analysis of disturbing patterns of mortality in the United States and a political statement calling for government action to overhaul pharmaceutical regulation, take control of the health care system, and shift the balance of power in the economy away from capital and toward labor. It seems evident to the authors that their political statement follows from their actuarial analysis, but the connection between the two struck this reader as tenuous.

If there were a Nobel Prize for scapegoating. . .

Yuval Levin watch

Posted on by

1. My review of A Time to Build.

Levin sees today’s elites as unwilling to abide by institutional constraints. Some abuse their power within an institution. Levin terms this “insiderism”. Others only use institutional prestige to enhance their personal ambitions but eschew any obligations to bolster the institutions that support them or to conform to institutional norms. Levin calls this “outsiderism” or “platforming,” meaning using the institution as a platform from which to expand one’s personal recognition.

2. A very comprehensive interview of Yuval Levin by Richard Reinsch . Hard to excerpt, but here is a slice:

meritocracy contributes to that problem because it leaves our elites now thinking that their positions are earned, that their authority is legitimate by default because they’ve been selected into elite institutions of higher education in particular. . . an elite that doesn’t think it needs to be constrained is a very bad fit for a democratic society.

It invites the kind of resistance, frustration, and ultimately populism that we’ve seen, and I think it deserves that response. Our elites in fact don’t think enough about how to constrain themselves in ways that could make it clear to the larger society that they’re playing a legitimate and valuable role. And I think institutions have an enormous role to play in that because our elite institutions can constrain our elites in ways that put them to use for the larger society. That’s what the professions do. That’s what political and cultural institutions do when they’re functioning well.

But if we understand our institutions as performative, as just platforms for people to stand and shine on, then they don’t really function to constrain our elites. They just display our elites and increase the frustration of the larger society with them. I think part of the solution to this part of the problem our country confronts is an idea of institutionalism that requires much more constraint and formation, that requires people to understand themselves as needing to prove that they operate by some standard of integrity and public service and that would require a real cultural change in a lot of our elite institutions.

Polygenic scores

Posted on by

Charles Murray is bullish on them.

I think the application of genomic data to social science questions is roughly where aviation was in 1908. The world’s best plane, the Wright Flyer, was little more than a toy. Yet within a decade, thousands of acrobatically maneuverable aircraft were flying high and fast over the battlefields of Europe.

I will read his latest book, but I have already staked out a more skeptical position.

Plomin is optimistic that with larger sample sizes better polygenic scores will be found, but I am skeptical. Unless there are unexplored areas in the existing data sets, such as non-linearities or interaction effects, my guess is that there are diminishing returns to enlarging the sample size.

That refers to Robert Plomin and his book Blueprint, not to be confused with another recent book of the same title by a different author.

References on prestige and dominance

Posted on by

Several comments on my previous post lead me to want to include some references.

1. My views indeed are derived from books by Joseph Henrich and Kevin Laland. I see humans as particularly evolved to learn from one another, and I view prestige hierarchies in that context.

2. The distinction between prestige hierarchy and dominance hierarchy did not originate with me. I think I got it from Henrich, but I believe I also had read Kevin Simler, to whom a commenter linked. Another commenter linked to Scott Alexander’s reply to Simler.

3. I don’t think that prestige requires low-status people to suck up, as Simler implies. As a chess player (I’m not, but let’s pretend), I don’t have to suck up to Magnus Carlsen. But I recognize the reality that he is above me in the hierarchy. If you want to study chess games, you should study his rather than mine.

4. So why do humans express admiration for skilled people? As an example, think of me expressing admiration for my doctor. My motive is not to suck up to the doctor. I want my friends to take advantage of the doctor’s skills. I may think that the doctor deserves more business. I want my friends to value my judgment, and so my incentive to be honest is stronger than if I were just trying to do what Scott A calls coattail riding.

5. I go back to my contention that prestige hierarchies tend to be positive-sum. Yes, you may be jealous of somebody who has more prestige. Yes, you may waste resources trying to acquire signals of prestige. Yes, to the extent that contests to acquire prestige are set up to reward the wrong skills, the outcomes are going to be non-optimal. But that is the crux of the issue. As long as a contest rewards the right skills, then a prestige hierarchy is of great social value.

Profits in financial markets

Posted on by

Reviewing Gregory Zuckerman’s book on the trading firm Renaissance Technologies, I write

Much of the book consists of tales of the gifted mathematicians who ran the firm, along with their foibles and conflicts. These are entertaining enough, but I want to focus instead on two deeper economic issues that arise from the story.

1) Does the success of Renaissance show that financial markets are inefficient?

2) Are the social benefits of the trading conducted by firms like Renaissance commensurate with the profits that they earned?