Thursday assorted links
1. Alvin Lucier’s brain is still making music. He died in 2021.
4. America’s AI action plan. Full plan here.
5. Charles Mann on what keeps the lights on.
6. A historical political economy of Kenya.
7. How much does education benefit the global poor?
8. ACX grants.
USA fact of the day
…fraction of children living with two parents is highest it’s been since 1991 (which wouldn’t be surprising with less teen/v young pregnancy)…
That is from Kevin Bryan, here is the source material.
My excellent Conversation with Helen Castor
Here is the audio, video, and transcript. Here is part of the episode summary:
Tyler and Helen explore what English government could and couldn’t do in the 14th century, why landed nobles obeyed the king, why parliament chose to fund wars with France, whether England could have won the Hundred Years’ War, the constitutional precedents set by Henry IV’s deposition of Richard II, how Shakespeare’s Richard II scandalized Elizabethan audiences, Richard’s superb artistic taste versus Henry’s lack, why Chaucer suddenly becomes possible in this period, whether Richard II’s fatal trip to Ireland was like Captain Kirk beaming down to a hostile planet, how historians continue to discover new evidence about the period, how Shakespeare’s Henriad influences our historical understanding, Castor’s most successful work habits, what she finds fascinating about Asimov’s I, Robot, the subject of her next book, and more.
Here is an excerpt from the opening sequence:
COWEN: Richard II and Henry IV — they’re born in the same year, namely 1367. Just to frame it for our listeners, could you give us a sense — back then, what was it that the English government could do and what could it not do? What is the government like then?
CASTOR: I think people might be surprised at quite how much government could do in England at this point in history because England, at this point, was the most centralized state in Europe, and that has two reasons. One is the Conquest of 1066 where the Normans have come in and taken the whole place over. Then, the other key formative period is the late 12th century when Henry II is ruling an empire that stretches from the Scottish border all the way down to southwestern France.
He has to have a system of government and of law that can function when he’s not there. By the late 14th century, when Richard and Henry — my two kings in this book — appear on the scene, the king has two key functions which appear on the two sides of his seal. On one side, he sits in state wearing a crown, carrying an orb and scepter as a lawgiver and a judge. That is a key function of what he does for his people. He imposes law. He gives justice. He maintains order.
On the other side of the seal, he’s wearing armor on a warhorse with a sword unsheathed in his hand. That’s his function as a defender of the realm in an intensely practical way. He has to be a soldier, a warrior to repel attacks or, indeed, to launch attacks if that’s the best form of defense. To do that, he needs money.
For that, the institution of parliament has developed, which offers consent to taxation that he can demonstrate is in the national interest. It has also come to be a law-making forum. Wherever he needs to make new laws, he can make statute law in Parliament that therefore, in its very nature, has the consent of the representatives of the realm.
COWEN: What is it, back then, that government cannot do?
CASTOR: What a government doesn’t have in the medieval period is, it doesn’t have a monopoly of force. In other words, it doesn’t have a police force. It doesn’t have a professional police force, and it doesn’t have a standing army, or at least by the late Middle Ages, England does have a permanent garrison in Calais, which is its outpost on the northern coast of France, but that’s not a garrison that can be recalled to England with any ease.
So, enforcement is the government’s key problem. To enforce the king’s edicts, it therefore relies on a hierarchy of private power on the landed, the great landowners of the kingdom, who are wealthy because of their possession of land, but crucially, also have control over people, the men who live and work on their land. If you need to get an enforcement posse — this is medieval English language that we use when we talk of sheriffs and posses — the county posse, the power of the county.
If you need to get men out quickly, you need to tap into those local power structures. You don’t have modern communications. You don’t have modern transport. The whole hierarchy of the king’s theoretical authority has to tap into and work through the private hierarchy of landed power.
COWEN: Why do those landed nobles obey the king? They’re afraid of the future raising of an army? Or they’re handed out some other benefit? What keeps the incentives all working together to the extent they stay working together?
CASTOR: They have a very important pragmatic interest in obeying the king because the king is the keystone of the hierarchy within which they are powerful and wealthy. Of course, they want more power and more wealth for themselves and for their dynasty, but importantly, they don’t want to risk everything to acquire more if it means serious danger that they might lose what they already have.
They have every interest in maintaining the hierarchy as it already is, within which they can then . . . It’s like having a referee…
A very good episode, definitely recommended. I enjoyed all of Helen’s books, most notably the recent
That was then, this is now, AI edition
COWEN: They can now do Math Olympiad problems. There’s one forecast by someone working on this that, within a year, they’ll win Math Olympiads. It could be wrong, but when you say that it’s only a year away, I would take that pretty seriously.
That was from my Nate Silver CWT taped on July 22, 2024. We just taped a new episode today…
Wednesday assorted links
1. State-level arts funding is to some extent substituting for federal arts funding.
2. Norwegian sovereign wealth fund deploys AI. And more thinking is not always good for the AI.
3. Some Amish settlements have TFR above 10. And Colombia is now at 1.2. And coupling rate vs. fertility rate.
China fact of the day
China alone accounts for almost 1000 hotels that include Vienna in their name, far more than any country, including Austria itself.
This article is more generally about what you learn when you study 25,000 hotel names and look for common patterns. Via the excellent Samir Varma.
USA fact of the day
Federal Reserve Board operating expenses have *quadrupled* from 2004 to 2023, reaching ~$1 billion in 2023, according to the Annual Reports of the Board of Governors of the Federal Reserve System.
That is from Jon Hartley. It is of course correct that the other effects of the Fed far outweigh the size of these expenditures. Nonetheless, it is worth asking, given these numbers, whether the system in place is generating good decisions. That in turn said, we do not currently have an “appropriate set of askers.”
Markets in everything, bet on tariff repeal edition
Cantor Fitzgerald, a financial services company led by the sons of US Commerce Secretary Howard Lutnick, has offered to buy the right to hundreds of millions of dollars in potential refunds from companies that have paid Trump’s tariffs.
The offer means that the sons of the pro-tariff commerce secretary, Kyle and Brandon, have made a way for investors to bet that President Donald Trump’s signature tariffs will be struck down in court.
Here is the full story.
Stablecoin podcast with Garett Jones
37 minutes, fresh material in here, see also Spotify under Bluechip Dialogues.
Tuesday assorted links
1. China fact of the day: “The 10 MSAs *hardest hit* by the China Shock all had positive real wage growth since 2001.” And: “Is that just average wage growth? No! Read the post. It’s positive growth at the median. And the 10th percentile. And the 25th, 75th, 90th percentile… across the distribution.”
2. “These Restaurants, Salons and Workouts Are Free for Hot People—if They Post About Them.” (WSJ)
3. A critique of how economic history is evolving. Column version here.
4. It seems chess.com bans 100,000 cheaters every month?
5. How AI can teach general education requirements, with specifics. And Gemini has gold medal results in the IMO. And how AI can help solve cyber problems.
6. More on the potential alchemy transmutation into gold claims (FT). Seems to be real!
Gross(ery) Confusion
Zephyr Teachout’s NYTs op-ed on grocery store prices is poorly argued.
The food system in the United States is rigged in favor of big retailers and suppliers in several ways. Big retailers often flex their muscles to demand special deals; to make up the difference, suppliers then charge the smaller stores more.
Let’s be clear about what is actually going on. Costco offers its suppliers lower prices in return for bigger orders. There is nothing anti-competitive about volume discounting. Moreover, are firms dismayed or are they eager to sell to big, bad Costco? Google AI gives a good answer:
…firms are eager to sell to Costco because of the immense potential for sales and brand exposure, but they must be prepared to meet stringent requirements, negotiate competitive pricing, and be able to handle high volume and demanding logistics.
Would Americans be better off without Costco? Doubtful given that more than one-quarter of all Americans pay for a Costco membership (either individually or as a family).
Teachout’s idea that suppliers “make up the difference” by charging smaller stores more is also economically incoherent. Profit-maximizing firms already charge what the market will bear. If Costco’s volume justifies a discount, that doesn’t mean suppliers can or should charge higher prices to other buyers. Yes, there are models where costs change with volume but costs could go down with volume and, in any case, those models don’t rely on the folk theory of “making up the difference.”
That’s one of the subtler mistakes. Here’s a more glaring one:
Consider eggs. At the independent supermarket near my apartment, the price for a dozen white eggs last week was $5.99. At a major national retailer a few blocks away, it was $3.99. (For an identical box of cereal, the price difference was $3.) Any number of factors may contribute to a given price, but market power is a particularly consequential one.
Read that again: the firm allegedly abusing market power is the one charging less.
It gets stranger:
New York City has a strong price gouging law on the books, which forbids anyone — suppliers and retailers — from jacking up prices during a state of emergency unless the seller’s own costs have gone up accordingly. The city couldn’t have stopped the bird flu that devastated flocks, but maybe it can stop suppliers from cynically exploiting a crisis to justify exorbitant prices.
This makes two errors. First, she acknowledges it’s not gouging if costs rise—then cites egg prices rising due to the bird flu devastating flocks. That’s literally a textbook case of a supply shock. Maybe some firms exploited the crisis—but eggs rising in price after millions of chickens are killed is the best example you’ve got???
Second, within the span of a few paragraphs, the op-ed veers from claiming large retailers charge prices that are unfairly low to blaming them for charging prices that are too high. I’m surprised she didn’t go for the trifecta and accuse them of colluding to charge the same price.
AIs and Spontaneous Order
Tupy and Boettke in the WSJ on AI and the economy:
The belief that AI can achieve comparable results to free markets, let alone surpass them, reflects a misplaced confidence in computation and a misunderstanding of the price system. The problem for the would-be AI planners is that prices don’t exist like facts about the physical world for a computer to collect and process. They arise from competitive bidding over scarce resources and are inseparable from real market exchanges. Moreover, prices aren’t fixed inputs to be assumed in advance. They are continually being discovered and formed by entrepreneurs testing ideas about future consumer wants and resource constraints.
Economic models that treat prices as given overlook the entrepreneurial actions that create them in the first place. Ludwig von Mises made this point in 1920: Without real market exchange, central planners lack meaningful prices for capital goods. Consequently, they can’t calculate whether directing steel to railways rather than hospitals adds or destroys value.
I would another point. We are not going to have one AI to rule us all. Instead, there are going to be millions of agents who themselves will be participants in the market process. The buying and selling of the AI agents will contribute to the formation of prices but for all the Hayekian reasons that process will not be capable of being predicted.
As I said 7 years ago on Quora:
AIs will themselves be part of the economy. Firms and individuals use AIs to make decisions. Thus, any AI has to take into account the decisions of other AIs. But no AI is going to be so far advanced beyond other AIs that this will be possible. In other words, as AIs increase in power so does the complexity of the economy.
The problem of perfectly organizing an economy does not become easier with greater computing power precisely because greater computing power also makes the economy more complex.
This isn’t to say AI won’t help improve economic policy—it might, if we listen. But the future economy won’t look like a centrally planned machine. It will look like an economy of von Neumanns—autonomous agents buying, selling, and strategizing in complex interaction.
The Benefits of Scholastic Athletics
This paper uses longitudinal data to study the benefits of participation in scholastic athletics starting with high school participation and continuing with college athletics, including the benefits of intramural athletics. We study the impact of participation on a number of important life outcomes, including graduation from high school and college and wages after schooling is completed. Controlling for rich measures of cognitive and personality skills and social background, we find substantial benefits at all levels. Participation in athletics promotes social mobility for disadvantaged and minority students.
Here is the paper, by
The America vs. Europe thing, again
From my latest column at The Free Press:
I worry much more about Europe in the longer run. Let’s consider how some of the most important comparisons between America and Europe are likely to change over the next 20 years.
Two of America’s biggest problems are obesity and opioid addiction, with opioid deaths running at about 54,000 a year. Yet both of those problems are getting better. GLP-1 drugs will help us beat back obesity, and finally opioid deaths have begun to decline. If this follows the path of previous drug epidemics, the decline will continue and perhaps accelerate.
More generally, we are entering a new age of fantastic biomedical innovations. These advances likely will help Americans more than Europeans, as Europeans are more likely to be in good shape to begin with, which is why Americans are more likely to need new and better treatments. That is, of course, a ding on America, but it will matter less as time passes.
One major advantage of America is likely to increase with time, and that is one of scale. Americans do things big, think big, and have created some of the world’s largest companies, most obviously in the tech sector, where size is often rewarded. You can see this in the stock market valuations of those tech companies, including Nvidia, which at its current $4 trillion or so valuation is worth more than the entire German stock market. Europe shows few if any signs of catching up in this area, or of having a major presence in the commercial spaces for artificial intelligence. If anything, EU regulations go out of their way to prevent Europe from excelling at tech.
Is tech likely to stop growing in economic and cultural influence? Have we reached peak application for current and future AI models? You can guess at the right answers to all of those questions. They imply that America’s economic lead over Europe will widen.
The brain drain from Europe (and other regions) to the United States seems to be accelerating in the areas of tech and AI, most of all for young people. If you want to do a big, successful start-up, you probably should move to America. End of story. America has major and growing companies in these areas, full of foreigners, and Europe does not.
Of course, a lot of that talent will not pay off right away. Not all of those smart and ambitious individuals will have big commercial hits at the age of 22. But more and more of them will by the age of 40. Europe has lost an increasing number of these people, and won’t be getting most of them back. The continent feels a bit of pain now, but the talent differential will de facto increase, if only due to the mere passage of time and the rising productivity of those people. It is not just about more people leaving; rather, those who already have moved to America will make a bigger and bigger difference over the next 10 to 15 years.
And:
The more general lack of European economic dynamism also is an issue that worsens with time. One recent economic study found that “Europeans switch jobs much less frequently, and restructuring is much rarer.” That is, of course, a problem, but in the short run the associated difficulties are not so large. If your economy remains static, after a year of progress elsewhere it is only missing out on so much beneficial change. After five years it is missing out on much more, and after 10 years much more yet. The more static and less dynamic nature of European economies naturally increases in size as a problem with the passage of time.
Population aging and low birth rates are another problem that will make it harder for Europe to catch up. The U.S. total fertility rate is about 1.63, whereas in the European Union it is about 1.38. Over time, this will make it harder for Europe to afford their current system of pensions. The major European populations also will be older than the American population, and probably as a result less innovative. This difference has only started to bite, and it is likely to grow in import.
I consider some other important issues, such as immigration, at the link.
The AI culture that is Faroe
Fed up with too much planning and decision-making on holiday? The Faroe Islands tourist board says its latest initiative taps into a trend for travellers seeking “the joy of surrender” on trips “where control is intentionally let go in favour of serendipity and spontaneity”. Their needs are answered in the nation’s fleet of “self-navigating rental cars”, launched this month, which — while they are not self-driving — will direct visitors on itineraries around the archipelago devised by locals.
Each route features between four and six destinations over the course of three to six hours, with only one section of the itinerary revealed at a time to maintain an element of surprise. Along the way, the navigation system will also share local stories tied to each place.
Here is more from Tom Robbins at the FT.