Category: Economics

Time Constraints for Tenure, Promotion, and Funding Decisions Lead Academic Biologists to Over-Study Already-Studied Genes

George Stigler argued that when most economists were self-funded business practitioners economics was more applied and empirical, while after most economists were academics funded by endowments or the government economics became less applied and more formal. [In a quick search I failed to identify the article where Stigler says this–sorry.] A similar point was made to science more broadly by Terence Kealey in his thought-provoking The Economic Laws of Scientific Research. The article quoted below argues persuasively that research on human genes is aligned with the career survival goals of academics, rather than with either the faster advance of science or the quicker cure of diseases like cancer. The alignment could be improved if more of research funding came from a variety of private sources.

(p. D3) In a study published Tuesday [Sept. 18, 2018] in PLOS Biology, researchers at Northwestern University reported that of our 20,000 protein-coding genes, about 5,400 have never been the subject of a single dedicated paper.

Most of our other genes have been almost as badly neglected, the subjects of minor investigation at best. A tiny fraction — 2,000 of them — have hogged most of the attention, the focus of 90 percent of the scientific studies published in recent years.

A number of factors are largely responsible for this wild imbalance, and they say a lot about how scientists approach science.

. . .

It was possible, . . ., that scientists were rationally focusing attention only on the genes that matter most. Perhaps they only studied the genes involved in cancer and other diseases.

That was not the case, it turned out. “There are lots of genes that are important for cancer, but only a small subset of them are being studied,” said Dr. Amaral.

. . .

A long history helps, . . . . The genes that are intensively studied now tend to be the ones that were discovered long ago.

Some 16 percent of all human genes were identified by 1991. Those genes were the subjects of about half of all genetic research published in 2015.

One reason is that the longer scientists study a gene, the easier it gets, noted Thomas Stoeger, a post-doctoral researcher at Northwestern and a co-author of the new report.

“People who study these genes have a head start over scientists who have to make tools to study other genes,” he said.

That head start may make all the difference in the scramble to publish research and land a job. Graduate students who investigated the least studied genes were much less likely to become a principal investigators later in their careers, the new study found.

“All the rewards are set up for you to study what has been well-studied,” Dr. Amaral said.

“With the Human Genome Project, we thought everything was going to change,” he added. “And what our analysis shows is pretty much nothing changed.”

If these trends continue as they have for decades, the human genome will remain a terra incognito for a long time. At this rate, it would take a century or longer for scientists to publish at least one paper on every one of our 20,000 genes.

That slow pace of discovery may well stymie advances in medicine, Dr. Amaral said. “We keep looking at the same genes as targets for our drugs. We are ignoring the vast majority of the genome,” he said.

Scientists won’t change their ways without a major shift in how science gets done, he added. “I can’t believe the system can move in that direction by itself,” he said.

Dr. Stoeger argued that the scientific community should recognize that a researcher who studies the least known genes may need extra time to get results.

“People who do something new need some protection,” he said.

For the full commentary see:

Carl Zimmer. “Matter; The Problem With DNA Research.” The New York Times (Tuesday, September 25, 2018 [sic]): D3.

(Note: ellipses, and bracketed date, added.)

(Note: the online version of the commentary has the date Sept. 18, 2018 [sic], and has the title “Matter; Why Your DNA Is Still Uncharted Territory.” Where there are differences in wording between the versions, the passages quoted above follow the online version.)

The paper in PLOS Biology co-authored by Thomas Stoeger and mentioned above is:

Stoeger, Thomas, Martin Gerlach, Richard I. Morimoto, and Luís A. Nunes Amaral. “Large-Scale Investigation of the Reasons Why Potentially Important Genes Are Ignored.” PLOS Biology 16, no. 9 (2018): e2006643.

Kealey’s book, praised above, is:

Kealey, Terence. The Economic Laws of Scientific Research. New York: St. Martin’s Press, 1996.

Ozempic Profits Poured into Massive Supercomputer Meant to Power AI for Future Drug Development

I think AI is currently being oversold. But I am very ignorant and could be wrong, so I favor a diversity of privately-funded bets on what will work to bring us future breakthrough innovations.

(p. B2) Two of the world’s most important companies are now in a partnership born from the success of their most revolutionary products. The supercomputer was built with technology from Nvidia—and money from the Novo Nordisk Foundation. The charitable organization has become supremely wealthy as the largest shareholder in Novo Nordisk, which means this project was made possible by the breakthrough drugs that have sent the Danish company’s stock price soaring.

To put it another way, it’s the first AI supercomputer funded by Ozempic.

It was named Gefion after the goddess of Norse mythology who turned her sons into oxen so they could plow the land that would become Denmark’s largest island.

. . .

Whatever you call it, Gefion is a beast. It is bigger than a basketball court. It weighs more than 30 tons. It took six months to manufacture and install. It also required an investment of $100 million.

. . .

When it’s fully operational, the AI supercomputer will be available to entrepreneurs, academics and scientists inside companies like Novo Nordisk, which stands to benefit from its help with drug discovery, protein design and digital biology.

For the full commentary see:

Ben Cohen. “It’s a Giant New Supercomputer That Might Transform an Entire Country.” The Wall Street Journal (Saturday, Nov. 2, 2024): B2.

(Note: ellipses added.)

(Note: the online version of the commentary has the date November 1, 2024, and has the title “Science of Success; The Giant Supercomputer Built to Transform an Entire Country—and Paid For by Ozempic.”)

“Most Published Research Findings Are False”

(p. 10) How much of biomedical research is actually wrong? John Ioannidis, an epidemiologist and health-policy researcher at Stanford, was among the first to sound the alarm with a 2005 article in the journal PLOS Medicine. He showed that small sample sizes and bias in study design were chronic problems in the field and served to grossly overestimate positive results. His dramatic bottom line was that “most published research findings are false.”

The problem is especially acute in laboratory studies with animals, in which scientists often use just a few animals and fail to select them randomly. Such errors inevitably introduce bias. Large-scale human studies, of the sort used in drug testing, are less likely to be compromised in this way, but they have their own failings: It’s tempting for scientists (like everyone else) (p. C2) to see what they want to see in their findings, and data may be cherry-picked or massaged to arrive at a desired conclusion.

A paper published in February [2017] in the journal PLOS One by Estelle Dumas-Mallet and colleagues at the University of Bordeaux tracked 156 biomedical studies that had been the subject of stories in major English-language newspapers. Follow-up studies, they showed, overturned half of those initial positive results (though such disconfirmation rarely got follow-up news coverage). The studies dealt with a wide range of issues, including the biology of attention-deficit hyperactivity disorder, new breast-cancer susceptibility genes, a reported link between pesticide exposure and Parkinson’s disease, and the role of a virus in autism.

Reviews by pharmaceutical companies have delivered equally grim numbers. In 2011, scientists at Bayer published a paper in the journal Nature Reviews Drug Discovery showing that they could replicate only 25% of the findings of various studies. The following year, C. Glenn Begley, the head of cancer research at Amgen, reported in the journal Nature that he and his colleagues could reproduce only six of 53 seemingly promising studies, even after enlisting help from some of the original scientists.

With millions of dollars on the line, industry scientists overseeing clinical trials with human subjects have a stronger incentive to follow high standards. Such studies are often designed in cooperation with the U.S. Food and Drug Administration, which ultimately reviews the findings. Still, most clinical trials produce disappointing results, often because the lab studies on which they are based were themselves flawed.

For the full essay see:

Harris, Richard. “Dismal Science In the Search for Cures.” The Wall Street Journal (Saturday, April 8, 2017 [sic]): C1-C2.

(Note: bracketed year added.)

(Note: the online version of the essay was updated April 7, 2017 [sic], and has the title “The Breakdown in Biomedical Research.”)

The essay quoted above is adapted from Mr. Harris’s book:

Harris, Richard. Rigor Mortis: How Sloppy Science Creates Worthless Cures, Crushes Hope, and Wastes Billions. New York: Basic Books, 2017.

The 2005 paper by Ioannidis mentioned above is:

Ioannidis, John P. A. “Why Most Published Research Findings Are False.” PLoS Medicine 2, no. 8 (2005): 696-701.

“I’m Sick of It, I’m Leaving” Are First Words of Children in Primitive Village Routinely Eating Grubs and Starch Tasting Like “Gummy Mucous”

(p. C9) As she tended soldiers during the Crimean War, a British nurse found herself appalled by the wretched, vermin-infested conditions at the army’s hospital in Istanbul. She began collecting figures showing the devastating effects of the filth and the dramatic benefits of the sanitary improvements she implemented. Her presentation on the need for cleaner care facilities, published in 1858, led to reforms that ultimately saved millions of lives and increased life expectancy in the U.K. Florence Nightingale, it turns out, was a pioneering data scientist.

Data, when used to reveal the value of hospital hygiene or the harm of tobacco smoke, can be a vital force for good, as Tim Harford reminds us in “The Data Detective.”

. . .

Imprecise and inconsistent definitions are one source of confusion.  . . .  . . . “infant mortality,” a key data point for public health, varies depending on the specific time in fetal development when the line is drawn between a miscarriage and a tragically premature birth.

. . .

To learn from data, it’s essential to present it well. For her analysis after the Crimean War, Florence Nightingale created one of the first infographics, using shrewdly designed diagrams to tell a memorable story. From the outset, she regarded visually compelling data displays as indispensable to making her arguments.

. . .

An authentically open mind can make a difference, Mr. Harford says, noting that the top forecasters tend to be not experts but earnest learners who constantly take in new data while challenging and refining their hypotheses. Data, Mr. Harford concludes, can illuminate and inform as well as distract and deceive. It’s often maddeningly hard to know the difference, but it would be unforgivable not to try.

For the full review see:

Wade Davis. “To Hear a Dying Tongue.” The Wall Street Journal (Saturday, Aug. 10, 2019 [sic]): C9.

(Note: ellipses added.)

(Note: the online version of the review has the date Aug. 9, 2019 [sic], and has the title “‘A Death in the Rainforest’ Review: To Hear a Dying Tongue.”)

The book under review is:

Kulick, Don. A Death in the Rainforest: How a Language and a Way of Life Came to an End in Papua New Guinea. Chapel Hill, NC: Algonquin Books, 2019.

Rationality-Defender Stigler Saw Voting as Irrational, but Did It Anyway

Nobel Prize winner George Stigler contributed to the Public Choice literature and was a staunch defender of rationality. One example would be his paper with Gary Becker, “De Gustibus Non Est Disputandum.” One popular, much discussed conclusion of some public choice theorists is that it is irrational to vote. The argument goes that the marginal effect of one vote is almost always miniscule, so the expected benefit to the voter is equally miniscule. On the other hand, the time and effort it takes to vote are always more than miniscule. So the expected costs of voting exceed the expected benefits. Ergo it is irrational to vote. When I was a graduate student, taking courses in philosophy and economics, and for a couple of years as a post-doctoral fellow, I frequently stopped by the office of the Journal of Political Economy where Stigler was an editor. I believe it was there that I heard Stigler, definitely on an election day, say “Here I go to do something irrational.”

Stigler is well-known for his humorous biting comments. These could be tough on others. But this story shows that they also could be directed at himself.

I do not know if anyone has fully solved the paradox of the irrationality of voting. I guess you would have to say something about how the effects of all good people ceasing to vote would be far from marginal and far from good.

I once mentioned to distinguished Public Choice theorist Dwight Lee that a positive result of the personal benefits of voting being miniscule to a voter, is that the voter was freed from voting their personal narrow self-interest, and could vote their conscience about what served the general good. (Maybe something like what Rawls hoped for behind his “veil of ignorance” in A Theory of Justice.) I believe that Dwight told me that he already published a paper that expressed this positive result, but I never took the time to look for that paper.

Milton Friedman Bubbled with Energy as He Grabbed His Sunday New York Times

During my first year in graduate school at the University of Chicago, I lived in a dorm for graduate students that had been built with money from John D. Rockefeller. It was next to a several story apartment tower that I had heard was built by Milton Friedman who owned and lived in the top apartment. On Sunday mornings, on more than one occasion, I remember Friedman used to bounce down the hallway of International House and go up to the mail counter, which always had a pile of The Sunday New York Times for sale. He would buy one, and bounce back down the hallway. Friedman was curious, energetic, optimistic, and engaged in the broad world of policy. A libertarian who wants to move the intellectual consensus, benefits from reading The New York Times.

“Mass Deportation” Is Not in Trump’s Heart, but Is a Warning to Future Illegal Aliens

I am stressed by the image of the “mass deportation” of those who entered the U.S. illegally, but otherwise have been decent hard-working people. My plausible hope is that deep in his heart, Trump does not really mean it or plan it. Why “plausible”? Read the passage quoted below describing Trump’s visit with The Wall Street Journal editorial board.

At this year’s Republican National Convention, Mr. Trump vowed to undertake “the largest deportation operation in the history of our country.” Editorial board member Kyle Peterson asks how large—does Mr. Trump intend to deport aliens who are law-abiding except for their illegal presence in the country, even if they have American spouses and children? Maybe not, Mr. Trump says: “We have a lot of good people in this country, and we have to do something about it, and I’d like to see if we can do it.”

Pressed for specifics, he demurs: “Well, I don’t want to go too much into clarification, because the nicer I become, the more people that come over illegally.” When he was president, “I said, ‘We’re going to separate your family.’ . . . It doesn’t sound nice, but when a family hears they’re going to be separated, you know what they do? They stay where they are, because we couldn’t handle it. . . . But the interest from the heart, yeah, something’s going to be done. . . . I mean, there’s some human questions that get in the way of being perfect, and we have to have the heart, too. OK?”

The implication is that the optimal immigration policy is a happy medium between restriction and openness. That’s certainly true and perhaps a truism. Mr. Trump suggests that he, the bully with a heart of gold, is just the man to strike the balance.

For the full commentary/interview see:

James Taranto. “The Weekend Interview; Trump Tangles With the Journal’s Editors.” The Wall Street Journal (Saturday, October 18, 2024): A13.

(Note: ellipses in original.)

Europeans Tire of Costly and Ineffective Climate Transition Policies

(p. A15) The 2015 Paris Agreement aspired to “reduce the risks and impacts of climate change” by eliminating greenhouse-gas emissions in the latter half of this century. The centerpiece of the strategy was a global transition to low-emission energy systems.

. . .

U.S. and European governments are trying to induce an energy transition by building or expanding organizations and programs favoring particular “clean” technologies, including wind and solar generation, carbon capture, hydrogen production and vehicle electrification. Promoting technological innovation is a worthy endeavor, but such efforts face serious challenges as costs and disruptions grow without tangible progress in reducing local, let alone global, emissions. Retreats from aggressive goals are already under way in Europe, with clear signs of mandate fatigue. The climbdown will be slower in the U.S., where subsidies create constituencies that make it more difficult to reverse course.

. . . It means that today’s ineffective, inefficient, and ill-considered climate-mitigation strategies will be abandoned, making room for a more thoughtful and informed approach to responsibly providing for the world’s energy needs.

For the full commentary see:

Steven E. Koonin. “The ‘Climate Crisis’ Fades Out.” The Wall Street Journal (Tuesday, June 11, 2024): A15.

(Note: ellipses added.)

(Note: the online version of the commentary has the date June 10, 2024, and has the same title as the print version.)

Koonin’s commentary, quoted above, is related to his book:

Koonin, Steven E. Unsettled: What Climate Science Tells Us, What It Doesn’t, and Why It Matters. Dallas, TX: BenBella Books, 2021.

In Walt Disney’s Disneyland Youth Could “Savor the Challenge and Promise of the Future”

(p. 12) President Dwight D. Eisenhower once praised Walt Disney for his “genius as a creator of folklore.” When Disney died in 1966, the line made it into his obituary, evidence of its accuracy. Folklore, defined broadly, is an oral tradition that stretches across generations. It tells people who they are, how they got here and how they should live in the future. The company Disney created appointed itself keeper of these traditions for Americans, spinning up fresh tales and (more often) deftly repackaging old ones to appeal to a new century.

It started with Mickey Mouse, but as his company turns 100, Disney’s legacy — advanced in hundreds of films and shorts and shows, mass-produced tie-in merchandise, marvelous technical advancements, gargantuan theme parks around the world — was the production of a modern shared language, a set of reference points instantly recognizable to almost everyone, and an encouragement to dream out loud about a utopian future. Walt Disney was a man who gazed backward and forward: speaking at the opening of Disneyland in 1955, he proclaimed: “Here age relives fond memories of the past, and here youth may savor the challenge and promise of the future.”  . . .

Disney told stories of folk heroes (Davy Crockett, Paul Bunyan), princes and princesses, and even, occasionally, a mouse, all while leading the pack on ever-shifting technologies. (He was, among other things, the first major movie producer to make a TV show.) A sense of optimism ruled Disney’s ethos, built on homemade mythologies. The lessons of his stories were simple, uplifting and distinctly American: believe in yourself, believe in your dreams, don’t let anyone make you feel bad for being you, be your own hero and, most of all, don’t be afraid to wish upon a star. Fairy tales and legends are often disquieting, but once cast in a Disney light they became soft and sweet, their darker and less comforting lessons re-engineered to fit the Disney ideal. It was a distinctly postwar vision of the world.

And we ate it up, and we exported it, and we wanted to be part of it, too.

For the full commentary see:

Alissa Wilkinson. “The Wonderful World of Disney?” The New York Times, Arts&Leisure Section (Sunday, December 17, 2023): 12 & 14.

(Note: ellipsis added.)

(Note: the online version of the commentary was updated Dec. 18, 2023, and has the title “Disney Is a Language. Do We Still Speak It?”)

AI Algorithms Lack Intelligence Since They Are “Just Predicting the Next Word in a Text”

(p. B5) Yann LeCun helped give birth to today’s artificial-intelligence boom. But he thinks many experts are exaggerating its power and peril, and he wants people to know it.

. . .

On social media, in speeches and at debates, the college professor and Meta Platforms AI guru has sparred with the boosters and Cassandras who talk up generative AI’s superhuman potential, from Elon Musk to two of LeCun’s fellow pioneers, who share with him the unofficial title of “godfather” of the field. They include Geoffrey Hinton, a friend of nearly 40 years who on Tuesday was awarded a Nobel Prize in physics, and who has warned repeatedly about AI’s existential threats.

. . .

LeCun thinks AI is a powerful tool.

. . .

At the same time, he is convinced that today’s AIs aren’t, in any meaningful sense, intelligent—and that many others in the field, especially at AI startups, are ready to extrapolate its recent development in ways that he finds ridiculous.

If LeCun’s views are right, it spells trouble for some of today’s hottest startups, not to mention the tech giants pouring tens of billions of dollars into AI. Many of them are banking on the idea that today’s large language model-based AIs, like those from OpenAI, are on the near-term path to creating so-called “artificial general intelligence,” or AGI, that broadly exceeds human-level intelligence.

OpenAI’s Sam Altman last month said we could have AGI within “a few thousand days.” Elon Musk has said it could happen by 2026.

LeCun says such talk is likely premature. When a departing OpenAI researcher in May talked up the need to learn how to control ultra-intelligent AI, LeCun pounced. “It seems to me that before ‘urgently figuring out how to control AI systems much smarter than us’ we need to have the beginning of a hint of a design for a system smarter than a house cat,” he replied on X.

He likes the cat metaphor. Felines, after all, have a mental model of the physical world, persistent memory, some reasoning ability and a capacity for planning, he says. None of these qualities are present in today’s “frontier” AIs, including those made by Meta itself.

Léon Bottou, who has known LeCun since 1986, says LeCun is “stubborn in a good way”—that is, willing to listen to others’ views, but single-minded in his pursuit of what he believes is the right approach to building artificial intelligence.

Alexander Rives, a former Ph.D. student of LeCun’s who has since founded an AI startup, says his provocations are well thought out. “He has a history of really being able to see gaps in how the field is thinking about a problem, and pointing that out,” Rives says.

. . .

The large language models, or LLMs, used for ChatGPT and other bots might someday have only a small role in systems with common sense and humanlike abilities, built using an array of other techniques and algorithms.

Today’s models are really just predicting the next word in a text, he says. But they’re so good at this that they fool us. And because of their enormous memory capacity, they can seem to be reasoning, when in fact they’re merely regurgitating information they’ve already been trained on.

“We are used to the idea that people or entities that can express themselves, or manipulate language, are smart—but that’s not true,” says LeCun. “You can manipulate language and not be smart, and that’s basically what LLMs are demonstrating.”

For the full commentary see:

Christopher Mims. “Keywords: This AI Pioneer Thinks AI Is Dumber Than a Pet Cat.” The Wall Street Journal (Saturday, Oct. 12, 2024): B5.

(Note: ellipses added.)

(Note: the online version of the commentary was updated Oct. 11, 2024, and has the title “Keywords: This AI Pioneer Thinks AI Is Dumber Than a Cat.” The sentence starting with “Léon Bottou” appears in the online, but not the print, version. Where there are small differences between the versions, the passages quoted above follow the online version.)