Category: Science
DEI vs. the Chips Act
The Hill has a good op-ed by Matt Cole and Chris Nicholson on how labor and DEI regulations are strangling the CHIPS act. It’s somewhat over the top, failure is overdetermined, but this is an important op-ed and directionally correct.
The Biden administration recently promised it will finally loosen the purse strings on $39 billion of CHIPS Act grants to encourage semiconductor fabrication in the U.S. But less than a week later, Intel announced that it’s putting the brakes on its Columbus factory. The Taiwan Semiconductor Manufacturing Company (TSMC) has pushed back production at its second Arizona foundry. The remaining major chipmaker, Samsung, just delayed its first Texas fab.
This is not the way companies typically respond to multi-billion-dollar subsidies. So what explains chipmakers’ apparent ingratitude? In large part, frustration with DEI requirements embedded in the CHIPS Act.
…The law contains 19 sections aimed at helping minority groups, including one creating a Chief Diversity Officer at the National Science Foundation, and several prioritizing scientific cooperation with what it calls “minority-serving institutions.” A section called “Opportunity and Inclusion” instructs the Department of Commerce to work with minority-owned businesses and make sure chipmakers “increase the participation of economically disadvantaged individuals in the semiconductor workforce.”
…Handouts abound. There’s plenty for the left—requirements that chipmakers submit detailed plans to educate, employ, and train lots of women and people of color, as well as “justice-involved individuals,” more commonly known as ex-cons. There’s plenty for the right—veterans and members of rural communities find their way into the typical DEI definition of minorities. There’s even plenty for the planet: Arizona Democrats just bragged they’ve won $15 million in CHIPS funding for an ASU project fighting climate change.
…tired of delays at its first fab, [TSMC]flew in 500 employees from Taiwan. This angered local workers, since the implication was that they weren’t skilled enough. With CHIPS grants at risk, TSMC caved in December, agreeing to rely on those workers and invest more in training them. A month later, it postponed its second Arizona fab.
Now TSMC has revealed plans to build a second fab in Japan. Its first, which broke ground in 2021, is about to begin production. TSMC has learned that when the Japanese promise money, they actually give it, and they allow it to use competent workers. TSMC is also sampling Germany’s chip subsidies, as is Intel.
Intel is also building fabs in Poland and Israel, which means it would rather risk Russian aggression and Hamas rockets over dealing with America’s DEI regime. Samsung is pivoting toward making its South Korean homeland the semiconductor superpower after Taiwan falls.
…The CHIPS Act’s current identity as a jobs program for favored minorities means companies are forced to recruit heavily from every population except white and Asian men already trained in the field.
Remember that there is an Extreme Shortage of High-IQ Workers. The United States is big and rich and can afford to handicap itself in many ways but less so when it comes to high-end semiconductor manufacturing, the most difficult and complicated manufacturing process ever attempted by human beings. For that we want the Dream Team, the very best, chosen on merit alone.
Addendum: Scott Lincicome has a good overview of the problem.
*How Life Works*
The author is Philip Ball, and the subtitle is A User’s Guide to the New Biology. I thought this book was wonderful, one of the best popular science books I’ve read in a long time. I’m sure its contents are familiar to many MR readers, but for me it was a very good introduction to debunking Richard Dawkins-like “primacy of the gene” stories, rather seeing genes as part of a broader, fairly flexible biological ecosystem.
It is also a very good book for explaining just how much computation goes on in biological systems.
I learned the word “gastrulation.”
Have you ever wondered how the salamander grows its tail back in exactly the right way? It turns out we are not sure why:
These creatures maintain a reserve of pluripotent stem cells for such repair jobs. But making the missing part seems to entail an ability of the regenerating cells to “read” the overall body plan: to take a peek at the whole, ask what’s missing, and adapt accordingly to preserve morphological integrity. Levin believes that this information is delivered to the growing cells via bioelectric signaling. But there are other possibilities. To account for the ability of the zebrafish to regrow a truncated tail to exactly the shape it had oringlaly — stripe markings and all — cell biologist Stefano Di Talia believes that a memory of the target shape is somehow encoded within the cells throughout the tail. In effect, he suggests, the different cell growth rates needed to recapitatulate the missing part are recorded along the edge of the wound.
And I learned about “xenobots“, a new kind of living creature, sort of:
Levin and colleagues discovered xenobots from a “what if” experiment: they wondered what might happen if embryonic frog cells were “liberated” from the constraints imposed by making an embryonic frog body. “If we give them the opportunity to re-envision multicellularity,” he asked, “what is it they will build.”
I found much of interest in this book, definitely recommended. Here is one good review of the book.
It’s happening
Is DNA all you need?
In new work, we report Evo, a genomic foundation model that learns across the fundamental languages of biology: DNA, RNA, and proteins. Evo is capable of both prediction tasks and generative design, from molecular to whole genome scale. pic.twitter.com/BPo9ggHhmp
— Patrick Hsu (@pdhsu) February 27, 2024
Dwarkesh Patel with Patrick Collison
Here is my episode with @patrickc
We discuss:
– what it takes to process $1 trillion/year
– how to build multi-decade APIs, companies, and relationships
– what's next for Stripe (increasing the GDP of the internet is an open ended task, and the Collison brothers are just… pic.twitter.com/Wx52cJI9Ve— Dwarkesh Patel (@dwarkesh_sp) February 21, 2024
Silk vaccines?
The invention of the hypodermic needle in 1844 brought major benefits to the practice of medicine, but ran headlong into an unexpected quirk of human nature. It turns out that millions of people feel an instinctive horror at the thought of receiving an injection – at least ten percent of the US adult population and 25 percent of children, according to one estimate. This common phobia partly explains the widespread reluctance to receive vaccinations against Covid-19, a reluctance which has led to tens of thousands of unnecessary deaths.
But a company in Cambridge, Massachusetts, called Vaxess Technologies plans to sidestep this common fear by abandoning stainless steel needles and switching to silk.
Vaxess is testing a skin patch covered in dozens of microneedles made of silk protein and infused with influenza vaccine. Each needle is barely visible to the naked eye and just long enough to pierce the outer layer of skin. A user sticks the patch on his arm, waits five minutes, then throws it away. Left behind are the silk microneedles, which painlessly dissolve over the next two weeks, releasing the vaccine all the while.
The silk protein acts as a preservative, so there’s no need to keep it on ice at a doctor’s office.
Here is more from Hiawatha Bray, at the new issue of Works in Progress.
Scientific Talent Leaks Out of Funding Gaps
Lack of speed kills:
We study how delays in NIH grant funding affect the career outcomes of research personnel. Using comprehensive earnings and tax records linked to university transaction data along with a difference-in-differences design, we find that a funding interruption of more than 30 days has a substantial effect on job placements for personnel who work in labs with a single NIH R01 research grant, including a 3 percentage point (40%) increase in the probability of not working in the US. Incorporating information from the full 2020 Decennial Census and data on publications, we find that about half of those induced into nonemployment appear to permanently leave the US and are 90% less likely to publish in a given year, with even larger impacts for trainees (postdocs and graduate students). Among personnel who continue to work in the US, we find that interrupted personnel earn 20% less than their continuously-funded peers, with the largest declines concentrated among trainees and other non-faculty personnel (such as staff and undergraduates). Overall, funding delays account for about 5% of US nonemployment in our data, indicating that they have a meaningful effect on the scientific labor force at the national level.
That is the abstract of a new paper by Wei Yang Tham, Joseph Staudt, Elisabeth Ruth Perlman, and Stephanie D. Cheng. Here is my earlier piece, with Collison and Hsu, and what we learned doing Fast Grants.
What should I ask Michael Nielsen?
I will be doing a Conversation with him. No description of Michael quite does him justice, but here is Wikipedia:
Michael Aaron Nielsen (born January 4, 1974) is a quantum physicist, science writer, and computer programming researcher living in San Francisco.
In 1998, Nielsen received his PhD in physics from the University of New Mexico. In 2004, he was recognized as Australia’s “youngest academic” and was awarded a Federation Fellowship at the University of Queensland. During this fellowship, he worked at the Los Alamos National Laboratory, Caltech, and at the Perimeter Institute for Theoretical Physics.
Alongside Isaac Chuang, Nielsen co-authored a popular textbook on quantum computing, which has been cited more than 52,000 times as of July 2023.
In 2007, Nielsen shifted his focus from quantum information and computation to “the development of new tools for scientific collaboration and publication”, including the Polymath project with Timothy Gowers, which aims to facilitate “massively collaborative mathematics.” Besides writing books and essays, he has also given talks about open science. He was a member of the Working Group on Open Data in Science at the Open Knowledge Foundation.
Nielsen is a strong advocate for open science and has written extensively on the subject, including in his book Reinventing Discovery, which was favorably reviewed in Nature and named one of the Financial Times’ best books of 2011.
In 2015 Nielsen published the online textbook Neural Networks and Deep Learning, and joined the Recurse Center as a Research Fellow. He has also been a Research Fellow at Y Combinator Research since 2017.
In 2019, Nielsen collaborated with Andy Matuschak to develop Quantum Computing for the Very Curious, a series of interactive essays explaining quantum computing and quantum mechanics. With Patrick Collison, he researched whether scientific progress is slowing down.
Here is Michael’s Notebook, well worth a browse and also a deeper read. Here is Michael on Twitter. So what should I ask him? (I’m going to ask him about Olaf Stapledon in any case, so no need to mention that.)
My Conversation with the excellent Ami Vitale
Here is the audio, visual, and transcript. Here is the episode summary:
Ami Vitale is a renowned National Geographic photographer and documentarian with a deep commitment to wildlife conservation and environmental education. Her work, spanning over a hundred countries, includes spending a decade as a conflict photographer in places like Kosovo, Gaza, and Kashmir.
She joined Tyler to discuss why we should stay scary to pandas, whether we should bring back extinct species, the success of Kenyan wildlife management, the mental cost of a decade photographing war, what she thinks of the transition from film to digital, the ethical issues raised by Afghan Girl, the future of National Geographic, the heuristic guiding of where she’ll travel next, what she looks for in a young photographer, her next project, and more.
Here is one excerpt:
COWEN: As you probably know, there’s a long-standing and recurring set of debates between animal welfare advocates and environmentalists. The animal welfare advocates typically have less sympathy for the predators because they, in turn, kill other animals. The environmentalists are more likely to think we should, in some way, leave nature alone as much as possible. Where do you stand on that debate?
VITALE: It depends. It’s hard to make a general sweeping statement on this because in some cases, I think that we do have to get involved. Also, the fact is, it’s humans in most cases who have really impacted the environment, and we do need to get engaged and work to restore that balance. I really fall on both sides of this. I will say, I do think that is, in some cases, what differentiates us because, as human beings, we have to kill to survive. Maybe that is where this — I feel like every story I work on has a different answer. Really, I don’t know. It depends what the situation is. Should we bring animals back to landscapes where they have not existed for millions of years? I fall in the line of no. Maybe I’m taking this in a totally different direction, but it’s really complicated, and there’s not one easy answer.
And:
COWEN: As you know, there are now social networks everywhere, for quite a while. Images everywhere, even before Midjourney. There are so many images that people are looking at. How does that change how you compose or think about photos?
VITALE: Well, it doesn’t at all. My job is to tell stories with images, and not just with images. My job as a storyteller — that has not changed. Nothing has changed in the sense of, we need more great storytellers, visual storytellers. With all of those social media, I think people are bored with just beautiful images. Or sometimes it feels like advertising, and it doesn’t captivate me.
I look for a story and image, and I am just going to continue doing what I do because I think people are hungry for it. They want to know who is really going deep on stories and who they can trust. I think that that has never gone away, and it will never go away.
I am very happy to have guests who do things that not everyone else’s guests do.
Vaccine Induced Social Amnesia
NYT: In 2022, there were 941 reported cases of measles in the World Health Organization’s European region. Over just the first 10 months of last year, according to an alarming bulletin the W.H.O. issued in mid-December, there were more than 30,000.
This is the kind of spike — a 3,000 percent increase — that looks implausible in headlines….But as the year drew to a close, the European measles outbreak kept growing. Through December, case numbers in the region eventually reached over 42,000, and although the largest outbreaks were in countries most Americans regard as pretty remote (Kazakhstan, Kyrgyzstan and Russia), there is also a vicious surge in Britain, which may look plausibly to us as the canary in a coal mine. There, in just one of England’s nine regions, the West Midlands, 260 cases have been confirmed and dozens more suspected, in a country which, as a whole, recorded just two cases as recently as 2021.
As David Wallace-Wells explains, vaccination rates are still 90%+ and down only slightly but measles is so infectious that even amid an otherwise well-vaccinated population, it can sometimes find pockets of low vaccination populations and spread like wildfire.
Measles is an especially nasty infection because it can induce “immunological amnesia“, thereby making individuals more susceptible to pathogens that they previously were able to resist.
Ironically, just as measles can induce immunological amnesia, vaccines can induce social amnesia about the severity of diseases, thereby making society more susceptible to pathogens that they previously were able to resist.
My new podcast with Dwarkesh Patel
We discussed how the insights of Hayek, Keynes, Smith, and other great economists help us make sense of AI, growth, risk, human nature, anarchy, central planning, and much more.
Dwarkesh is one of the very best interviewers around, here are the links. If Twitter is blocked to you, here is the transcript, here is Spotify, among others. Here is the most salacious part of the exchange, highly atypical of course:
Dwarkesh Patel 00:17:16
If Keynes were alive today, what are the odds that he’s in a polycule in Berkeley, writing the best written LessWrong post you’ve ever seen?
Tyler Cowen 00:17:24
I’m not sure what the counterfactual means. Keynes is so British. Maybe he’s an effective altruist at Cambridge. Given how he seemed to have run his sex life, I don’t think he needed a polycule. A polycule is almost a Williamsonian device to economize on transaction costs. But Keynes, according to his own notes, seems to have done things on a very casual basis.
And on another topic:
Dwarkesh Patel 00:36:44
We’re talking, I guess, about like GPT five level models. When you think in your mind about like, okay, this is GPT five. What happens with GPT six, GPT seven. Do you see it? Do you still think in the frame of having a bunch of RAs, or does it seem like a different sort of thing at some point?
Tyler Cowen 00:36:59
I’m not sure what those numbers going up mean, what a GPT seven would look like, or how much smarter it could get. I think people make too many assumptions there. It could be the real advantages are integrating it into workflows by things that are not better GPTs at all. And once you get to GPT, say, 5.5, I’m not sure you can just turn up the dial on smarts and have it, like, integrate general relativity and quantum mechanics.
Dwarkesh Patel 00:37:26
Why not?
Tyler Cowen 00:37:27
I don’t think that’s how intelligence works. And this is a Hayekian point. And some of these problems, there just may be no answer. Like, maybe the universe isn’t that legible, and if it’s not that legible, the GPT eleven doesn’t really make sense as a creature or whatever.
Dwarkesh Patel 00:37:44
Isn’t there a Hayekian argument to be made that, listen, you can have billions of copies of these things? Imagine the sort of decentralized order that could result, the amount of decentralized tacit knowledge that billions of copies talking to each other could have. That in and of itself, is an argument to be made about the whole thing as an emergent order will be much more powerful than we were anticipating.
Tyler Cowen 00:38:04
Well, I think it will be highly productive. What “tacit knowledge” means with AIs, I don’t think we understand yet. Is it by definition all non-tacit? Or does the fact that how GPT-4 works is not legible to us or even its creators so much? Does that mean it’s possessing of tacit knowledge, or is it not knowledge? None of those categories are well thought out, in my opinion. So we need to restructure our whole discourse about tacit knowledge in some new, different way. But I agree, these networks of AIs, even before, like, GPT-11, they’re going to be super productive, but they’re still going to face bottlenecks, right? And I don’t know how good they’ll be at, say, overcoming the behavioral bottlenecks of actual human beings, the bottlenecks of the law and regulation. And we’re going to have more regulation as we have more AIs.
You will note I corrected the AI transcriber on some minor matters. In any case, self-recommending, and here is the YouTube embed:
An important new paper on the costs of climate change
Forthcoming in ReStud, I haven’t had the chance to read it yet:
To analyze climate change mitigation strategies, economists rely on simplified climate models — so-called climate emulators — that provide a realistic quantitative link between CO2 emissions and global warming at low computational costs. In this paper, we propose a generic and transparent calibration and evaluation strategy for these climate emulators that is based on freely and easily accessible state-of-the-art benchmark data from climate sciences. We demonstrate that the appropriate choice of the free model parameters can be of key relevance for the predicted social cost of carbon. The key idea we put forward is to calibrate the simplified climate models to benchmark data from comprehensive global climate models that took part in the Coupled Model Intercomparison Project, Phase 5 (CMIP5). In particular, we propose to use four different test cases that are considered pivotal in the climate science literature: two highly idealized tests to separately calibrate and evaluate the carbon cycle and temperature response, an idealized test to quantify the transient climate response, and a final test to evaluate the performance for scenarios close to those arising from economic models, and that include exogenous forcing. As a concrete example, we re-calibrate the climate part of the widely used DICE-2016, fathoming the CMIP5 uncertainty range of model responses: the multi-model mean as well as extreme, but still permissible climate sensitivities and carbon cycle responses. We demonstrate that the functional form of the climate emulator of the DICE-2016 model is fit for purpose, despite its simplicity, but its carbon cycle and temperature equations are miscalibrated, leading to the conclusion that one may want to be skeptical about predictions derived from DICE-2016. We examine the importance of the calibration for the social cost of carbon in the context of a partial equilibrium setting where interest rates are exogenous, as well as the simple general equilibrium setting from DICE-2016. We find that the model uncertainty from different consistent calibrations of the climate system can change the social cost of carbon by a factor of four if one assumes a quadratic damage function. When calibrated to the multi-model mean, our model predicts similar values for the social cost of carbon as the original DICE-2016, but with a strongly reduced sensitivity to the discount rate and about one degree less long-term warming.
The social cost of carbon in DICE-2016 is oversensitive to the discount rate, leading to extreme comparative statics responses to changes in preferences.
That is the abstract from Doris Folini, Aleksandra Friedl, Felix Kübler, and Simon Scheidegger,
What happened in 17th century England (a lot)
East India Company founded — 1600
Shakespeare – Hamlet published 1603
England starting to settle America – 1607 in Virginia, assorted, you could add Harvard here as well
King James Bible – 1611
The beginnings of steady economic growth – 1620 (Greg Clark, JPE)
Rule of law ideas, common law ideas, Sir Edward Coke – 1628-1648, Institutes of the Laws of England, four volumes
Beginnings of libertarian thought – Levellers 1640s
Printing becomes much cheaper, and the rise of pamphlet culture
John Milton, Aeropagitica, defense of free speech, 1644
King Charles I executed – 1649 (leads to a period of “Britain without a King,” ending 1660)
Birth of economic reasoning – second half of 17th century
Royal African Company and a larger slave trade – 1660
General growth of the joint stock corporation
Final subjugation of Ireland, beginnings of British colonialism and empire (throughout, mostly second half of the century)
Discovery of the calculus, Isaac Newton 1665-1666
Great Plague of London, 1665-1666, killed ¼ of city?
Great Fire of London, 1666
John Milton, Paradise Lost, 1667
Social contract theories – John Locke 1689
Bill of Rights (rights of Parliament) — 1689
Birth of modern physics – Newton’s Principia 1687
Bank of England — 1694
Scientific Revolution – throughout the 17th century, places empiricism and measurement at the core of science
The establishment of Protestantism as the religion of Britain, both formal and otherwise, throughout the century, culminating in the Glorious Revolution of 1688.
London – becomes the largest city in Europe by 1700 at around 585,000 people.
England moves from being a weak nation to perhaps the strongest in Europe and with the strongest navy.
Addendum: Adam Ozimek adds:
…first bank to print banknotes in Europe, 1661
Discovery of the telescope 1608
First patent for a modern steam engine 1602
Economic ornithology
But a hundred years ago, birds were seen as the best remedy for the weeds and insect pests that threatened the country’s food supply and cost farmers hundreds of millions of dollars every year. And in order to identify the precise impact that birds had on agriculture, a field called economic ornithology was born. According to one of its leading practitioners, economic ornithology was “the study of birds from the standpoint of dollars and cents … in short, it is the practical application of the knowledge of birds to the affairs of everyday life.”1 And from the 1880s to the 1930s, birds were widely seen as economic agents, working alongside farmers in the fight against the insect hordes.
By the 1940s, economic ornithology had become discredited and obsolete. Effective and affordable pesticides had entirely replaced the birds’ bug-killing role, while economic ornithologists could never prove that their methods actually increased the number of helpful birds. But before their role in agriculture was dismissed, there was a time when we believed that we depended on birds for our food, and for our very survival.
And here was the method of economic ornithology:
In 1916, Gilbert Trafton summarized the primary approach used by economic ornithologists: “The practical value of birds to man, whether helpful or harmful, depends chiefly on their food habits,” and by examining what they eat, “the exact economic status of a bird is determined.” Sometimes this was done by observing the behaviors of birds in the field, but it usually meant dissecting birds and seeing what they had in their stomachs.
Here is the full Substack, by Robert Francis. Via Philip Wallach.
Department of Uh-Oh, economic research edition
We assess statistical power and excess statistical significance among 31 leading economics general interest and field journals using 22,281 parameter estimates from 368 distinct areas of economics research. Median statistical power in leading economics journals is very low (only 7%), and excess statistical significance is quite high (19%). Power this low and excess significance this high raise serious doubts about the credibility of economics research. We find that 26% of all reported results have undergone some process of selection for statistical significance and 56% of statistically significant results were selected to be statistically significant. Selection bias is greater at the top five journals, where 66% of statistically significant results were selected to be statistically significant. A large majority of empirical evidence reported in leading economics journals is potentially misleading. Results reported to be statistically significant are about as likely to be misleading as not (falsely positive) and statistically nonsignificant results are much more likely to be misleading (falsely negative). We also compare observational to experimental research and find that the quality of experimental economic evidence is notably higher.
That is from a new paper by Zohid Askarov, Anthony Doucouliagos, Hristos Doucouliagos, and T. D. Stanley.
Via my colleague Jonathan Schulz.
How women are perceiving the economics profession
Fewer women reported being satisfied with the climate in the economics profession in 2023 compared to five years ago, despite efforts during that time to improve conditions for women in the field, according to a new survey.
About 17% of women in economics said they strongly agreed or agreed with a statement about being satisfied in the profession, down from 20% in 2018, according to the topline results of a survey conducted in the fall. The preliminary findings were presented by University of Chicago Booth School of Business economist Marianne Bertrand Friday at the American Economic Association’s annual meeting in San Antonio.
The gap between women and men’s experience in economics widened slightly over the past five years, with 39% of men saying they were satisfied with the profession’s climate, compared to 40% in 2018.
Women made up just 17.8% of full economics professors in 2022. While representation is higher among students and associate professors, the share of new economics doctoral degree recipients that were women fell in 2023, Bertrand said Friday.