“…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
“…challenges misconceptions that distort our economic debates.” - Arthur Brooks, President of the American Enterprise Institute
“…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
“…a very valuable contribution.” - Larry Summers, former Secretary of the Treasury and director of the National Economic Council, president emeritus, Harvard University
“A full-throated defense of economic dynamism.” - The Wall Street Journal
“…a very valuable contribution.” - Larry Summers, former Secretary of the Treasury and director of the National Economic Council, president emeritus, Harvard University
“Unintended Consequences offers deep and well-argued analyses on almost every issue.” - The New York Times
“…a comprehensive explanation of the modern economy.” - Julian Robertson, Founder, Tiger Management
“Unintended Consequences provides a provocative interpretation of the causes of the global financial crisis and the policies needed to return to rapid growth. Whether you agree or not, this analysis is well worth reading.” - Nouriel Roubini, New York University; Chairman, Roubini Global Economics
“There are an amazing number of good ideas and interesting points made in Unintended Consequences. The thinking underlying it, and the obvious depth of understanding of the author, are very impressive.” - Steven Levitt, coauthor of Freakonomics; 2004 John Bates Clark Medal
“…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
“…a comprehensive explanation of the modern economy.” - Julian Robertson, Founder, Tiger Management
AI Summary.U.S. survey data on high school students show the share of 9th graders identifying as lesbian, gay, or bisexual fell from 17% to 11% over two years, compared with a decline from 18% to 16% among 12th graders. The drop was concentrated among girls, from 27% to 22%, versus 9% to 7% among boys.
Jean TwengeGeneration Tech
Date Posted:
Is Database:
Database
Among US high-school students, the LGB share fell btw 2023 to 2025 by 5pp for girls (27% to 22%), and 2pp for boys (9% to 7%). The preceding rise was likewise larger for girls: +128% from 2015 to 2021, versus +99% for boys to 2023.
Are younger teens less likely to identify as lesbian, gay, or bisexual?
Core argument: The share of U.S. 9th graders identifying as lesbian, gay, or bisexual fell from 17% in 2023 to 11% in 2025 in the CDC’s Youth Risk Behavior Survey, while 12th graders declined from 18% to 16%.
[New data from the Youth Risk Survey of U.S. 9th to 12th graders from the CDC shows] for 12th graders, 17 and 18 years old, identifying as LGB declines only a few percentage points between 2023 and 2025 (from 18% to 16%). But the headline result is for 9th graders (14 and 15 years old): Identifying as LGB absolutely plummets between 2023 and 2025, from 17% to 11%. That really does look like a free fall. The decline appears to be driven almost entirely by girls. Identifying as LGB dropped from 27% in 2023 to 22% in 2025 among girls, and from 9% to 7% among boys. [In terms of the share of high school students reporting engaging in same-sex activity] most high school students are not sexually active, so this is only a subsample of the larger group. For boys, there’s been a fairly steady rise in same-sex partners, not too different from the pattern of identifying as LGB among boys. The story is different for girls, where having non-heterosexual sex peaked in 2021 and then fell steeply.
Takeaways by Macro Roundup® AI
The share of U.S. 9th graders identifying as lesbian, gay, or bisexual fell from 17% in 2023 to 11% in 2025 in the CDC’s Youth Risk Behavior Survey, while 12th graders declined from 18% to 16%.
The decline appears driven almost entirely by girls, whose LGB identification fell from 27% in 2023 to 22% in 2025, compared with a drop from 9% to 7% among boys.
Among sexually active teens, the share of boys reporting at least one same-sex partner has risen fairly steadily, while for girls it peaked in 2021 and then fell steeply.
Related Articles:
Is There Really a Generational Difference in Identifying as Lesbian, Gay, or Bisexual?— General Social Survey data show that increases among young American adults appear in same-sex sexual behavior as well as in identification, with 20% reporting a same-sex partner versus 18.5% identifying as lesbian, gay, or bisexual. The increases are larger among 27- to 41-year-olds.
Teens Used To Be Happier. What Changed?— The share of US high school seniors describing themselves as "very happy" peaked at 24.8% and remained around 24.4% before falling to 11.4% at its low, recovering only to 12.7%, well below the 17.9% recorded before the pandemic. The decline coincided with smartphones and social media becoming pervasive.
US College Students’ Well-Being— A study of US college students finds moderate-to-severe depression reported by 26% of students at one elite institution, 22% reporting moderate-to-severe anxiety and 10% contemplating suicide. Poorer mental health is associated with being female, little exercise, low grades, and financial aid; financial stress is associated with higher suicidal thoughts.
AI Summary.Racial disparities in US crime, arrest, and prison admission rates have declined across offence types, driven by reductions among Black Americans. Black-White disparity in drug-offence prison admissions fell from 12x to 2.9x, attributable to a 75% drop in the Black drug admission rate, while declines for violent and property offences were smaller.
Jason Robey and Michael LightNature Human Behaviour
Date Posted:
Is Database:
Database
Is Important:
Important
From their respective recent peaks in 2000 and 2006 to 2019, prison admissions for violent crime fell 20% for Whites and 43% for Blacks. Race-specific NIBRS estimates suggest actual violent crime fell 18% and 47%, respectively.
Why are racial gaps in US arrests and imprisonment narrowing?
Core argument: Black–White disparity in drug-offence prison admission rates fell from 12 to 1 in 2000 to 2.9 to 1 in 2019, attributed almost entirely to a 75% drop in the Black prison admission rate for drug offences.
From their respective peaks [2000 for whites, 2006 for blacks], prison admission rates for violent crime fell 20% for Whites and 43% for Blacks. [Because UCR crime totals do not identify offender race, the authors use the National Incident-Based Reporting System (NIBRS), which records offender race as reported by victims, witnesses or police, to allocate national violent-crime totals by race. NIBRS contains about three times as many violent offenders as arrestees. As a check on crimes not reported to police, they use National Crime Victimization Survey victim reports; results are similar.] [Recorded violent crime fell 18% for Whites and 47% for Blacks See Table 2]. For Whites, the decline in prison admissions is primarily a function of a decline in crime and arrests, with case processing dampening the decline. For Blacks, the decline in violent prison admissions is exclusively attributed to a 47% reduction in violent crime, as policing and case processing trends worked against an otherwise larger decline. [Arrests per crime and prison admissions per arrest both rose slightly; See Table 2].
Takeaways by Macro Roundup® AI
Black–White disparity in drug-offence prison admission rates fell from 12 to 1 in 2000 to 2.9 to 1 in 2019, attributed almost entirely to a 75% drop in the Black prison admission rate for drug offences.
Racial disparities in crime, arrest, and prison admission rates have declined across all offence types since about 2000, largely attributable to reductions in crime, arrests, and imprisonment among Blacks.
Declining drug admissions accounted for 53% of the overall drop in Black prison admission rates, while for Whites 62% of the decline came from property offences, with violent (20%) and drug (19%) admissions contributing roughly equally.
Related Articles:
America’s Incarceration Rate Is About To Fall Off A Cliff— The U.S. imprisonment rate stayed below 120 per 100,000 people from the end of World War II until the mid-1970s, then doubled by 1985 and grew about 8% a year, passing 1m prisoners in 1994 and peaking in 2009 — 18 years after violent crime peaked.
Number of Young People Accused of Serious Crimes Surges in New York City— New York City Police Department figures show 4,858 major crimes with a minor accused or arrested in the most recent year, up 37% from 3,543 seven years earlier, with robberies and felony assaults continuing to rise. Police officials attribute the increase to a state law diverting most 16- and 17-year-old defendants from adult courts.
How Many Are Criminals— In the United States, between one in four and one in three Black men eventually serve prison time, compared with one in seven Hispanic men and roughly one in fifteen non-Hispanic white men. In Denmark, about 9% of men of Danish origin and 27% of men of non-Western origin have a criminal conviction by age 24.
AI Summary.An analysis of authors at a leading AI research conference finds China is the largest originating source of elite AI talent, with its share of researchers holding Chinese undergraduate degrees rising 11 percentage points to 57%, while the US share fell to 13%. The US remains a net importer of AI researchers, gaining 2,145 while China lost 1,729.
Damien Ma and Binyi YangCarnegie Endowment for International Peace
Date Posted:
Is Database:
Database
Surveying education/employment histories of authors at the Neural Information Processing Systems conference, Ma and Yang infer China educated 57% of elite AI talent as of 2025, versus 13% for the US. Yet the US gained 2,145 researchers y/y while China lost 1,729.
Does China produce the most elite AI talent while America attracts it?
Core argument: Damien Ma and Binyi Yang find China’s share of elite AI talent, defined by undergraduate degrees earned in China, rose 11 percentage points to 57% in 2025 versus 2022, while the U.S. share fell to 13%.
[Using a methodology based on paper authors from the Neural Information Processing Systems conference (NeurIPS) Damien Ma and Binyi Yang find] China has established a commanding lead as the largest originating source of elite AI talent globally. Compared to 2022, the share of Chinese-origin AI talent, defined as those who received undergraduate degrees in China, increased 11 percentage points to 57% in 2025, while the U.S. share fell to 13%. The United States remains a magnet for global AI talent, particularly Chinese talent. Despite U.S.-China tensions, among AI researchers working in the United States, the 2025 share of those holding undergraduate degrees from China rose by 4 percentage points. When it comes to brain gain vs. brain drain, the United States saw a net gain of +2,145 researchers in 2025 while China registered a net loss of -1,729 researchers.
Takeaways by Macro Roundup® AI
Damien Ma and Binyi Yang find China’s share of elite AI talent, defined by undergraduate degrees earned in China, rose 11 percentage points to 57% in 2025 versus 2022, while the U.S. share fell to 13%.
Among AI researchers working in the United States, the share holding Chinese undergraduate degrees rose 4 percentage points in 2025, despite U.S.-China tensions.
The United States recorded a net gain of 2,145 AI researchers in 2025, while China registered a net loss of 1,729, under Ma and Yang’s methodology based on NeurIPS paper authors.
Related Articles:
China Lures Home Its Top AI Talent From Silicon Valley— Elite artificial intelligence researchers have left US firms including Google DeepMind and OpenAI for Chinese companies such as ByteDance, Tencent, and Alibaba. Three China- and San Francisco-based headhunters say they relocated more than 30 US-based researchers to China over 12 months, up from low single digits the prior year.
China and the Future Of Science— China graduates about 5x as many medical and biomedical students as the US each year, 7x as many engineers, and nearly double the number of science and engineering PhDs, while one-sixth to one-fifth of US science and engineering graduates are themselves Chinese. One study found returning Chinese scientists lead-author 2.5x more papers than peers who stay in the US.
Why Europe Produces Nobel Prize Winners but Not Elon Musks— Europe trains more engineering graduates per capita than China and more STEM PhDs than any region outside the United States, yet the only EU-born companies to reach $100bn market value in the past fifty years, Spotify and BioNTech, listed on US exchanges. Europe captures 6% of global AI startup funding versus 61% for the United States.
AI Summary.The authors estimate that more than half of the rise in the top 1% income share since 1985 — 5.8 of 10.5 percentage points — flowed through pass-through businesses, counting roughly 3m wealthy private business owners with average net worth near $25m running law firms, dealerships, and medical practices.
Torsten SløkApollo
Date Posted:
Is Database:
Database
Is Important:
Important
More than half the increase in the top 1% income share since 1985 flowed through pass-through businesses to ~ 3mm private business owners with a mean net worth of ~$25mm.
Did most of the top 1% income gain come from private businesses?
Core argument: Owen Zidar and Eric Zwick find that 5.8 of the 10.5 percentage-point rise in the U.S. top 1% income share since 1985 flowed through pass-through businesses, more than half of the increase.
The income that lifted America's top 1% did not come mainly from tech or Wall Street, according to Owen Zidar and Eric Zwick. More than half of the rise in the top 1% income share since 1985, 5.8 of 10.5 percentage points, flowed through pass-through businesses. Zidar and Zwick, counted roughly 3 million wealthy private business owners, with average net worth near $25 million, running law firms, car dealerships, medical practices, commercial contractors and regional restaurant chains, the kind of businesses that make their owners wealthy and well known in their own communities without ever making them nationally famous.
Takeaways by Macro Roundup® AI
Owen Zidar and Eric Zwick find that 5.8 of the 10.5 percentage-point rise in the U.S. top 1% income share since 1985 flowed through pass-through businesses, more than half of the increase.
Zidar and Zwick count roughly 3 million wealthy private business owners with average net worth near $25 million, running law firms, car dealerships, medical practices, commercial contractors and regional restaurant chains.
The income that lifted America’s top 1% did not come mainly from tech or Wall Street, according to Zidar and Zwick, but from businesses whose owners are well known locally without becoming nationally famous.
Related Articles:
What’s Missing in the Fed’s Data about Ultrarich Portfolios— The Federal Reserve's Distribution of Financial Accounts defines "private business" narrowly as proprietor's equity in noncorporate, nonfinancial businesses, excluding S-corporations, other private corporations, and private equity and hedge fund partnerships, which are lumped into "corporate equities." The Survey of Consumer Finances instead shows private business is roughly 45–50% of top 0.1% wealth.
They’re Rich but Not Famous—and They’re Suddenly Everywhere— An analysis of Federal Reserve data estimates about 430,000 U.S. households are worth $30m or more, including roughly 74,000 worth at least $100m, with growth in the number of very rich households outpacing overall population growth over recent decades. Nearly 72% of the top 0.1%'s wealth consists of corporate equities, mutual fund shares, and private businesses.
Capitalists in the Twenty-First Century— Most income at the top of the US distribution is non-wage income, primarily profits from closely-held, mid-market firms in skill-intensive industries owned by working-age individuals. The authors find private business profit falls by three-quarters after an owner retires or dies prematurely, indicating most top earners are "working rich" deriving income from human capital.
AI Summary.Census data show the number of U.S. families with inflation-adjusted incomes above $150,000 grew from under 3m in 1967 to over 31m in 2025, while families earning under $75,000 fell by almost 3m in absolute terms. The share of families with multiple earners has not increased much over that period.
Jeremy HorpedahlEconomist Writing Every Day
Date Posted:
Is Database:
Database
Is Important:
Important
In 1967, just 5.5% of American families had an inflation adjusted income of $150,000 or more. In 2025, 35.7% did. During that timeframe, the share of families with a real income less than $75,000 declined from 63.4% to 32.8%.
Are more American families reaching higher inflation-adjusted income levels?
Core argument: The number of U.S. families with incomes above $150,000 in 2025 inflation-adjusted dollars rose from fewer than 3 million in 1967 to over 31 million in 2025, according to Census data.
In 1967 there were fewer than 3 million families with incomes over $150,000 (in 2025 inflation-adjusted dollars). By 2025, that had grown to over 31 million families. If we look at the highest income threshold in this Census data — over $200,000 — the number of families has grown from just 1 million in 1967 to over 20 million in 2025. And there are fewer lower-income families too: the number of families under $75,000 shrunk, not just as a proportion of the total but even in absolute terms by almost 3 million families from 1967 to 2025. Of course, some of these families do have more earners than in 1967, though we shouldn’t overstate that too much. Using other data from Census, we can see that the share of families with multiple earners hasn’t increased much since 1967, and especially hasn’t since the mid-1990s.
Takeaways by Macro Roundup® AI
The number of U.S. families with incomes above $150,000 in 2025 inflation-adjusted dollars rose from fewer than 3 million in 1967 to over 31 million in 2025, according to Census data.
Families earning over $200,000, the highest income threshold in the Census data, grew from about 1 million in 1967 to over 20 million in 2025.
Families with incomes under $75,000 declined by almost 3 million in absolute terms from 1967 to 2025, while Census data show the share of families with multiple earners has not increased much since 1967, and especially not since the mid-1990s.
Related Articles:
Income in the United States: 2025— Median U.S. household income rose 2.6% to $87,460, the highest level on record in data going back to 1967. Median income increased 3% for White households, 2.9% for non-Hispanic White households, and 4.8% for Black households, while changes for Asian and Hispanic households were not statistically significant.
The Middle Class Is Shrinking Because of a Booming Upper-Middle Class— Analyses showing a shrinking middle class typically use relative income thresholds that rise with median income, so they can register decline even when living standards rise across the distribution. Using an absolute definition, the authors find the core middle class shrank because the upper-middle class grew from 10% of families to 31%, doubling its income share.
Who Sees Themselves as Working Class?— Survey data indicate 60% of U.S. adults say the term "working class" describes them well, including half of those with a bachelor's degree and half of upper-income Americans. Identification is highest among blue-collar workers at 77%, versus 61% in other occupations, and ranges from 52% of Asian adults to 62% of White adults.
AI Summary.The official U.S. poverty rate fell 0.5 percentage points to 10.2%, a third consecutive annual decline and among the lowest on record, with 34.5m people in poverty. Poverty rates hit historic lows for Hispanic individuals, down 1.2 points to 13.9%, and children under 18, down 1.0 point to 13.4%.
Christina Bijou and Emily ShriderU.S. Census Bureau
Date Posted:
Is Database:
Database
According to the Census Bureau, the poverty rate in the US fell .5pp in 2025 to 10.2%, one of the lowest rates on record. The poverty rate for children fell to a historical low of 13.4%. Census’s income measure does not include taxes or non-cash government transfers.
Is the U.S. poverty rate falling to record lows?
Core argument: The official U.S. poverty rate fell 0.5 percentage points to 10.2% in 2025, the third consecutive annual decline and one of the lowest rates on record, with 34.5 million people in poverty.
In 2025, the official poverty rate fell 0.5 percentage points to 10.2%, the third consecutive annual decline and one of the lowest rates on record. There were 34.5 million people in poverty in 2025. Poverty rates for Hispanic individuals and children under age 18 both fell to historic lows in 2025. Between 2024 and 2025, the poverty rate for Hispanic individuals fell 1.2 percentage points to 13.9%. The child poverty rate fell 1 percentage point to 13.4%.
Takeaways by Macro Roundup® AI
The official U.S. poverty rate fell 0.5 percentage points to 10.2% in 2025, the third consecutive annual decline and one of the lowest rates on record, with 34.5 million people in poverty.
The poverty rate for Hispanic individuals fell 1.2 percentage points to 13.9% between 2024 and 2025, a historic low.
The poverty rate for children under age 18 fell 1.0 percentage point to 13.4% in 2025, a historic low.
Related Articles:
Poverty and Dependency in the United States, 1939–2023— Extending a consistent poverty measure back to 1939, the authors estimate US absolute poverty fell from 48.5% to 19.5% between 1939 and 1963, driven by growth in market income rather than government transfers net of taxes. Poverty fell no faster in the 24 years after the War on Poverty than before.
Evaluating the Success of the War on Poverty since 1963 Using an Absolute Full-Income Poverty Measure— The authors estimate that an absolute full-income poverty measure, which counts a fuller set of income sources and adjusts thresholds only for inflation, fell from 19.5% to 1.6%, versus an official decline to 10.5%. They also find relative poverty reductions modest and government dependence more than doubled among working-age adults.
The Great “Transfer”-mation— Government transfers rose from 8.2% of Americans' total personal income in 1970 to 17.6% in 2022, making them the third largest income source after work and investments. The share of U.S. counties drawing a quarter or more of personal income from transfers rose from under 1% in 1970 to 53% in 2022.
AI Summary.The authors estimate the heritability of labor market outcomes by comparing the excess similarity of same-sex over opposite-sex twins with the corresponding difference among ordinary siblings. They caution that social interaction between twins, assortative mating, genetic dominance, population stratification, or indirect parental effects can bias such estimates upward or make them reflect non-genetic influences.
David Card, Jesse Rothstein and Moises YiNational Bureau of Economic Research
Date Posted:
Is Database:
Database
Is Important:
Important
Analysis of 44,500 twin pairs suggests ~one-third of the variation in log earnings is due to genetic differences – likely an upper bound, because same-sex twins may influence one another and share more similar environments than ordinary same-sex siblings.
How reliable are twin-based estimates of heritability in earnings?
Core argument: The authors identify heritability of U.S. labor market outcomes from a difference-in-differences comparing same-sex versus opposite-sex twins against the corresponding difference among ordinary siblings, adjusting for the estimated share of same-sex twins who are identical.
Our model identifies heritability from the difference-in-differences of same-sex vs. opposite-sex twins vs. siblings. [Specifically], the model attributes to genetic differences the excess similarity of same-sex over opposite-sex twins over and above the corresponding difference among ordinary siblings, adjusting for the estimated share of same-sex twins who are identical. The ordinary-sibling comparison is intended to remove family influences that differ between sons and daughters. Social interactions may lead to an over-estimate of the genetic component in earnings. The estimated heritability of outcomes may be biased in the presence of assortative mating, genetic dominance, population stratification, or violations of the assumption that the home environment is shared to the same degree for MZ as for DZ twins. Even when estimated without bias, heritability need not indicate direct genetic effects – it also may reflect indirect or cultural effects (e.g., effects of the parents’ genes on a child’s outcomes, not operating through the child’s own genes. Twins may [also] influence each other directly – for example in their choice of jobs -- inflating the correlation in their outcomes.
Takeaways by Macro Roundup® AI
The authors identify heritability of U.S. labor market outcomes from a difference-in-differences comparing same-sex versus opposite-sex twins against the corresponding difference among ordinary siblings, adjusting for the estimated share of same-sex twins who are identical.
Social interactions between twins, such as influencing each other’s job choices, may inflate the correlation in their outcomes and lead to an over-estimate of the genetic component in earnings.
Estimated heritability may be biased by assortative mating, genetic dominance, population stratification, or unequal sharing of the home environment between identical and fraternal twins, and even unbiased estimates may reflect indirect or cultural effects such as parents’ genes.
The Inheritance Of Social Status: England, 1600 to 2022— Using a genealogical database of 422,374 English people with rarer surnames, the authors find social status persists strongly even among distant relatives, with status correlations declining by a constant factor of 0.79 per step in the lineage across all status measures and eras studied. The authors argue this constancy suggests social interventions have modest effects.
A Chip Off the Old Block? Genetics and the Intergenerational Transmission of Socioeconomic Status— Linking Dutch genetic data to tax records, the authors estimate that moving a person 10 percentiles higher in a genetic index raises their own income by 0.9 percentiles and their child's by 0.7 percentiles. They attribute about 50% of this to transmitted genes and the rest to environmental pathways.
AI Summary.Median U.S. household income rose 2.6% to $87,460, the highest level on record in data going back to 1967. Median income increased 3% for White households, 2.9% for non-Hispanic White households, and 4.8% for Black households, while changes for Asian and Hispanic households were not statistically significant.
Melissa Kollar and Zach SchererU.S. Census Bureau
Date Posted:
Is Database:
Database
Is Important:
Important
Real US median household income increased 2.6% year over year in 2025 to a record high of $87,460.
Has median U.S. household income reached its highest level on record?
Core argument: Median U.S. household income reached $87,460 in 2025, up 2.6% from $85,210 in 2024 and the highest on record in data dating back to 1967.
In 2025, median household income was $87,460, an increase of 2.6% from the 2024 estimate of $85,210. Household income in 2025 was the highest on record dating back to 1967. For all demographic characteristics of the householder the median household income estimates in 2025 were either higher than the 2024 estimates or were not statistically different from 2024. Median household income increased by 3% for White households, 2.9% for non-Hispanic White households, and 4.8% for Black households. Median income did not change significantly for Asian or Hispanic households.
Takeaways by Macro Roundup® AI
Median U.S. household income reached $87,460 in 2025, up 2.6% from $85,210 in 2024 and the highest on record in data dating back to 1967.
Median household income rose 4.8% for Black households, 3% for White households, and 2.9% for non-Hispanic White households in 2025, while Asian and Hispanic households saw no statistically significant change.
Across all demographic characteristics of the householder, 2025 median household income estimates were either higher than or not statistically different from 2024 estimates.
Related Articles:
Income in the United States: 2024— U.S. median household income was $83,730, not statistically different from the prior year's $82,690, according to Census Bureau estimates. Across most demographic subgroups, median household income was higher or not statistically different from the prior year, with the exception of Black households, which saw a decline.
US Focus: The Effect Of Soaring Profits— US nominal pre-tax corporate profits in the national accounts grew 23% over the past year, a pace not seen outside post-recession spikes since the mid-2000s. Wider margins were the key driver, as domestic profit growth far exceeded the 8% rise in corporate value added, pushing margins near all-time highs while the labor share hit new lows.
The College Wage Premium in the Generative AI Era— The U.S. college wage premium fell from 0.626 to 0.575 after four decades of expansion, implying the first sustained negative growth in relative demand for college-educated labor. The authors estimate that weaker wage growth in occupations highly exposed to artificial intelligence accounts for roughly 28% of that decline, though the patterns are not causal.
