Is there an intermediate position on immigration?
It is a common view, especially on the political right, that we should be quite open to highly skilled immigrants, and much less open to less skilled immigrants. Increasingly I am wondering whether this is a stable ideological equilibrium.
To an economist, it is easy to see the difference between skilled and less skilled migrants. Their wages are different, resulting tax revenues are different, and social outcomes are different, among other factors. Economists can take this position and hold it in their minds consistently and rather easily (to be clear, I have greater sympathies for letting in more less skilled immigrants than this argument might suggest, but for the time being that is not the point).
The fact that economists’ intuitions can sustain that distinction does not mean that public discourse can sustain that distinction. For instance, perhaps “how much sympathy do you have for foreigners?” is the main carrier of the immigration sympathies of the public. If they have more sympathies for foreigners, they will be relatively pro-immigrant for both the skilled and unskilled groups. If they have fewer sympathies for foreigners, they will be less sympathetic to immigration of all kinds. Do not forget the logic of negative contagion.
You also can run a version of this argument with “legal vs. illegal immigration” being the distinction at hand.
Increasingly, I have the fear that “general sympathies toward foreigners” is doing much of the load of the work here. This is one reason, but not the only one, why I am uncomfortable with a lot of the rhetoric against less skilled immigrants. It may also be the path toward a tougher immigration policy more generally.
I hope I am wrong about this. Right now the stakes are very high.
In the meantime, speak and write about other people nicely! Even if you think they are damaging your country in some significant respects. You want your principles here to remain quite circumscribed, and not to turn into anti-foreigner sentiment more generally.
Friday assorted links
1. Which parts of America, apart from the coasts, will benefit from AI? (NYT) Chattanooga!
2. DeepSeek on the move. Here is the report. For ease of use and interface this is very high quality. Remember when “they” told us China had no interest in doing this?
And Nathan sends me these links:
https://old.reddit.com/r/LocalLLaMA/comments/1hmxjbn/deepseek_is_better_than_4o_on_most_benchmarks_at/
https://old.reddit.com/r/LocalLLaMA/comments/1hmm8v9/psa_deepseek_v3_outperforms_sonnet_at_53x_cheaper/
https://old.reddit.com/r/LocalLLaMA/comments/1hmk1hg/deepseek_v3_chat_version_weights_has_been/
https://old.reddit.com/r/LocalLLaMA/comments/1hmn55p/deepseekv3_officially_released/
I would say the final story here is not yet in, do not leap to any particular conclusion.
4. Ruxandra on San Francisco and civilizational greatness.
5. Justin Chang reviews The Brutalist (New Yorker).
6. David Brooks picks his Sidney Awards (NYT).
Patrick Collison on classic novels
Read it here. Recommended. Excerpt:
For me the clear standouts are Middlemarch, Bleak House, Karenina, and Life and Fate. I would enthusiastically reread any of them. If I had to choose just one to go to again, I would probably select Middlemarch. There’s something memorably compelling in Eliot’s affection and empathy for almost all of her characters. If Succession is a show with no likable personalities, Middlemarch is the opposite. Bleak House is a close second. Life and Fate is quite different to the others: it’s not exactly entertaining (or even notably well-written), but it is true and profound.
And this:
Today’s scientific papers are far harder to read, and jargon-replete, than those of 1960. However, the novels of the 19th century use significantly more sophisticated construction (and vocabulary) than those of today. What should we make of the countervailing trends? To me, both seem suboptimal.
Do go and digest the whole thing.
Steve Davis, Elon Musk’s Go-To Cost-Cutter Is Working for DOGE
A Bloomberg profile of the excellent Steve Davis:
Elon Musk’s deputy Steve Davis has spent more than 20 years helping the billionaire cut costs at businesses like SpaceX, the Boring Company and Twitter ….[now] Davis is helping recruit staff at DOGE, Musk’s effort to reduce government waste, in addition to his day job as president of Musk’s tunneling startup, the Boring Company.
At Boring, Davis has a reputation for frugality, signing off on costs as low as a few hundred dollars, according to people familiar with the conversations — unusual for a company that has raised about $800 million in capital. He also drives hard bargains with suppliers of products like raw steel, sensors, or even items as small as hose fittings, said the people, who asked not to be identified discussing private information.
His favorite directive for staff doing the negotiations: “Go back and ask again.”
…Davis started working for Musk in 2003, when he joined SpaceX, at the time a new company. He had just earned a master’s degree in aerospace engineering from Stanford University, and distinguished himself at the startup by solving hard engineering problems. At one point, Musk tasked the engineer with finding a cheaper alternative to a part that cost $120,000. Davis spent weeks on the challenge and figured out how to do it for $3,900, according to a biography of Musk. (Musk emailed back one word: “Thanks.”)
…Multitasking has proved a Davis signature, dating back to his student days. While he was working on his doctorate in economics at George Mason University in Fairfax, Virginia, Davis was working full time at SpaceX and owned a frozen-yogurt shop called Mr. Yogato in Washington’s Dupont Circle. Alex Tabarrok, one of Davis’ professors, remembers him juggling the multiple roles.
“I told him, ‘Look, you’re getting a Ph.D., you can’t be having a job and running a business at the same time,” Tabarrok recalls. “Focus on getting your Ph.D.”
But Davis declined to give up any of his pursuits, at one time incorporating business trends at Mr. Yogato into an academic paper and bringing some yogurt into class for sampling. Tabarrok can’t recall Davis’ grades, but says he stood out anyway. He “had so much energy, and was so entrepreneurial,” Tabarrok says. “It’s been kind of exciting to see him become one of Elon’s most trusted right-hand men.”
Davis’s GMU training in political economy will serve him very well in Washington.
See also my previous post, an MR classic, Why We Can’t Have Nice Things–Elon Musk and the Subways.
Addendum: 2013 profile of Steve and another of his businesses, Thomas Foolery a bar in DC where you paid for drinks according to plinko. Hat tip: Kevin Lewis.
Shruti pays tribute to Zakir Hussain
An excellent appreciation, might this be the best thing written on him? Here is one excerpt:
It is hard to explain the aura of Zakir to someone who has not sat in the audience and felt it. You can try to break it down—rationalists will call it genius, the sort of brilliance that defies analysis. Others might invoke his charisma, his ability to connect with any audience, that mischief dancing in his eyes. The spiritually inclined go a step further, claiming his rhythms channel something divine, as though the tabla becomes a vessel for forces we cannot name. I have heard Zakir live at least 25-30 times—concerts scattered across cities, years, moods—and I can tell you it is all of the above. Still, I will try to describe these moments as best as I can articulate, where Zakir speaks to every audience member in their language.
Do read the whole thing, a very impressive piece.
The future of the scientist in a world with advanced AI
AI will know almost all of the academic literature, and will be better at modeling and solving most of the quantitative problems. It will be better at specifying the model and running through the actual statistical exercises. Humans likely will oversee these functions, but most of that will consist of nodding, or trying out some different prompts.
The humans will gather the data. They will do the lab work, or approach the companies (or the IRS?) to access new data. They will be the ones who pledge confidentiality and negotiate terms of data access. (Though an LLM might write the contract.) They will know someone, somewhere, using a telescope to track a particular quasar. They may (or may not) know that the AI’s suggestion to sample the blood of a different kind of gila monster is worth pursuing. They will decide whether we should be filming dolphins or whales, so that we may talk to them using LLMs, though they will ask the LLMs for cost estimates in each case.
At least in economics, this will be continuing trends that were present before current high-quality AI. The scarce input behind a quality paper is, more and more, access to some new and interesting data source. More and more people can do the requisite follow-up technical work, though quality variations have by no means been eliminated.
“Science as an employment program for scientists” will fall all the more out of favor. It remains to be seen how much that will disfavor serendipitous human discovery.
On any given day, on the quest for more data, a scientist will have to think quite creatively about what he or she should be doing.
Manmohan Singh, RIP
I am sad to hear about the passing of Manmohan Singh at age 92.
Singh was perhaps the most influential Indian policymaker in the last five decades. An Oxbridge educated trade economist, he became India’s most important technocrat in the 1980s and 1990s – occupying every top position in economic policy – finance secretary; deputy chairman of the planning commission; governor of the RBI, and chief economic adviser. And as finance minister in 1991, when he brought the Indian economy out of socialism to embrace markets and global trade. After Modi and Nehru, he is also India’s longest continuously serving prime minister over two terms from 2004-2014.
For more about his work and long career in economic policy read Changing India – a five volume collection on Singh’s work as an academic, policymaker, politician and on the family man, Strictly Personal by his daughter Daman Singh. And to learn more about India’s liberalization and economic reforms, follow the 1991 Project at the Mercatus Center led by Shruti Rajagopalan and her team.
Thursday assorted links
1. Why didn’t we get Ozempic sooner? (NYT)
2. An episode of the Aarthi and Sriram podcast.
3. David Cramer has questions for me. Lots of them.
4. China fact of the day so much of the stimulus is in fact bank recapitalization.
5. The torture of an unphilosophical life.
6. Dean Ball on the Romantic piano.
7. Will AI’s future run through the Gulf?
8. Jason Furman on a new rules-based framework for the Fed (WSJ).
Full-length documentary on the life and legacy of Rene Girard
Very well done.
The New FDA and the Regulation of Laboratory Developed Tests
The FDA under President Trump and new FDA head Martin Makary should rapidly reverse the FDA’s powergrab on laboratory developed tests. To recap, laboratory developed tests (LDTs) are the kind your doctor orders, they are a service not a product and are not sold directly to patients. Congress has never given the FDA the authority to regulate LDTs. Indeed, in 2015, Paul Clement, the former US Solicitor General under George W. Bush, and Laurence Tribe, a leading liberal constitutional lawyer, wrote an article that rejected the FDA’s claims writing that the “FDA’s assertion of authority over laboratory-developed testing services is clearly foreclosed by the FDA’s own authorizing statute” and “by the broader statutory context.”
Moreover, in addition to legal reasons there are sound public policy reasons to reject FDA regulation of LDTs. Lab developed tests have never been FDA regulated, except briefly during the pandemic when the FDA used the declaration of emergency to issue so-called “guidance documents” saying that any SARS-COV-II test had to be pre-approved by the FDA. Thus, the FDA reversed the logic of emergency. In ordinary times, pre-approval was not necessary but when speed was of the essence it became necessary to get FDA pre-approval. The FDA’s pre-approval process slowed down testing in the United States and it wasn’t until after the FDA lifted its restrictions in March that tests from the big labs became available.
In a remarkably prescient passage, Clement and Tribe (2015, p. 18) had warned of exactly this kind of delay:
The FDA approval process is protracted and not designed for the rapid clearance of tests. Many clinical laboratories track world trends regarding infectious diseases ranging from SARS to H1N1 and Avian Influenza. In these fast-moving, life-or-death situations, awaiting the development of manufactured test kits and the completion of FDA’s clearance procedures could entail potentially catastrophic delays, with disastrous consequences for patient care.
We are seeing the same kind of FDA-caused delay for tests for bird-flu.
Moreover, unlike some of the proposals associated with incoming HHS head Robert Kennedy, reversing the FDA on lab-developed tests has significant support from a wide-variety of experts. Here, for example, is the American Hospital Association:
…we strongly believe that the FDA should not apply its device regulations to hospital and health system LDTs. These tests are not devices; rather, they are diagnostic tools developed and used in the context of patient care. As such, regulating them using the device regulatory framework would have an unquestionably negative impact on patients’ access to essential testing. It would also disrupt medical innovation in a field demonstrating tremendous benefits to patients and providers.
The Trump administration has a number of options:
…the LDT Final Rule was promulgated in time to escape Congressional Review Act scrutiny; however, the executive branch and a Republican-controlled Congress have other tools to limit or vitiate FDA’s authority. These include, in no particular order:
The U.S. Department of Health and Human Services (HHS) could revoke the LDT Final Rule. The recission of a rule is treated the same as the promulgation of a new rule. If HHS revokes the final rule, the cases will likely be dismissed as moot. The timing of such action is uncertain at this time.
FDA could extend or revise its policies of enforcement discretion. LDTs are currently subject to FDA’s phaseout policy which has five stages, the last of which begins in May 2028. Specific categories of IVDs will continue under an enforcement discretion policy indefinitely as described in the preamble to the final rule. HHS could quickly issue such a revised policy or policies without prior public comment if it determines such policy meets the threshold in 21 CFR 10.115(g)(2).
Congress could act. With a Republican-controlled House and Senate to start the new Trump administration, there is a chance that efforts to legislate the regulation of LDTs could be reignited. Based on prior congressional efforts, it is likely that such legislation would place LDTs under control by CMS and CLIA, rather than require LDTs to comply with FDA requirements.
HHS could let the litigation continue. The new administration may view the U.S. District Court for the Eastern District of Texas to be sympathetic to the Plaintiffs’ arguments and therefore proceed unabridged assuming the final rule will be struck-down, if that is indeed the deregulatory objective of the new administration.
The U.S. Department of Justice (DOJ) could act concerning the litigation. DOJ options are constrained by ethics rules but DOJ could request to amend its filings, pause the case pending rule-making proceedings, or take other actions intended to stall or moot the litigation in a deregulatory fashion.
Year-end CWT episode with Jeff Holmes
Here is the audio, video, and transcript. Here is a short summary:
On this special year-in-review episode, Tyler and producer Jeff Holmes look back on the past year in the show and more, including covering the most popular and underrated episodes, fielding listener questions, reviewing Tyler’s pop culture picks from 2014, mulling over ideas for what to name CWT fans, and more.
As for an excerpt:
HOLMES: Moving on to underrated episodes, episodes that weren’t necessarily breaking download records but are still very, very good. You could think of them as a personal favorite. I’ve got my picks. Do you want to throw out a couple?
COWEN: One underrated episode was Masaaki Suzuki because most people don’t know enough about Bach to really love what he said. Plus, he had an accent; that may hurt downloads a bit. But that one, I was very fond of. Fareed Zakaria, you got to see the real Fareed. Even his son loved the episode. I don’t know how many downloads it got, but it has to be underrated. Michael Nielsen. Most are underrated. Tom Tugendhat, who did not make it to be head of the Conservative Party, but someday still might and certainly ought to be.
HOLMES: Yes, those are good picks. Masaaki Suzuki was a fan mention as well, a favorite of my wife’s, so check that one out if you haven’t. I would also throw out Stephen Kotkin, so pretty recent episode. Kotkin performed very well.
COWEN: That’s one of the best episodes of all time.
HOLMES: It clearly just established itself in the pantheon. Think about Lazarus Lake in the past, or Richard Prum, which were some of my favorites. Just as soon as you listen to it, it’s a clear favorite. If you check out the YouTube comments, many people are commenting that it’s their favorite Stephen Kotkin interview.
COWEN: And people are still listening, so that will climb in the numbers. Paula Byrne was a tremendous episode.
And this:
COWEN: We’re pleasing people too much. Is that the lesson?
Recommended. And who else would you all like to see as guests?
Is academic writing getting harder to read?
To track academic writing over time, The Economist analysed 347,000 PhD abstracts published between 1812 and 2023. The dataset was produced by the British Library and represents a majority of English-language doctoral theses awarded by British universities. We reviewed each abstract using the Flesch reading-ease test, which measures sentence and word length to gauge readability. A score of 100 roughly indicates passages can be understood by someone who has completed fourth grade in America (usually aged 9 or 10), while a score lower than 30 is considered very difficult to read. An average New York Times article scores around 50 and a CNN
article around 70. This article scores 41…We found that, in every discipline, the abstracts have become harder to read over the past 80 years. The shift is most stark in the humanities and social sciences (see chart), with average Flesch scores falling from around 37 in the 1940s to 18 in the 2020s. From the 1990s onwards, those fields went from being substantially more readable than the natural sciences—as you might expect—to as complicated. Ms Louks’s abstract had a reading-ease rating of 15, still more readable than a third of those analysed in total.
Here is more from The Economist, via the excellent Samir Varma.
Christmas assorted links
2. What Sam Enright has been reading.
3. The usefulness of AI hallucinations (NYT).
4. Michael Brendan Dougherty on his ideological migration.
5. o3 and its perception abilities.
6. Ezra Klein on stuff (NYT).
7. Richard Perry, RIP (NYT).
8. The history of FRED (NYT).
9. This Little Babe.
Merry Christmas!
Wishing all our readers a wonderful day and New Year!