Edward Conard

Top Ten New York Times Bestselling Author

  • “…reminds us that inequality sends a signal of what society lacks most, in America’s case, entrepreneurship and risk taking.” - Lawrence Lindsey, CEO, The Lindsey Group, former Director of the National Economic Council
  • “…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
  • “…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
  • “…challenges misconceptions that distort our economic debates.” - Arthur Brooks, President of the American Enterprise Institute
  • “…a very valuable contribution.” - Larry Summers, former Secretary of the Treasury and director of the National Economic Council, president emeritus, Harvard University
  • “…reminds us that inequality sends a signal of what society lacks most, in America’s case, entrepreneurship and risk taking.” - Lawrence Lindsey, CEO, The Lindsey Group, former Director of the National Economic Council
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “Unintended Consequences is far smarter and more thought-provoking than most economics written for the general public” - Greg Mankiw, Harvard University, Former Chairman of the Council of Economic Advisors
  • “…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…a very valuable contribution.” - Larry Summers, former Secretary of the Treasury and director of the National Economic Council, president emeritus, Harvard University
  • “…a comprehensive explanation of the modern economy.” - Julian Robertson, Founder, Tiger Management
Upside of Inequality Oxford Unintended Consequences

Macro Roundup


Tuesday, September 2, 2025

AI Summary. U.S. sectors with greater exposure to large language models are estimated to have seen statistically significant gains in total wages paid, employment, and hourly wages, with larger hourly wage gains for younger and more educated workers in exposed occupations. Sectors intensive in human–AI collaboration tasks gained employment and wages, while those with more directly automatable tasks saw declines.

Andrew Johnston and Christos Makridis Social Science Research Network
Date Posted:
Is Database:
Database
Is Important:
Important

Over 2021 – 2024, sectors that were more exposed to LLMs saw significant overall increases in wages and employment following LLMs’ release, mainly for young, educated workers, reflecting cases where AI augments, rather than substitutes for, human effort.

How does language model exposure affect sector employment and wages?

Core argument: U.S. state–industry pairs with higher generative AI task exposure show statistically significant gains in the wage bill, employment, and hourly wages in the years surrounding the public release of large language models, alongside more insignificant increases in establishments.

This paper provides early empirical evidence on the labor market and output effects of generative AI exposure across U.S. [state and industry pairs] in the years surrounding the public release of large language models. Using task-level LLM exposure indices and variation in pre-existing occupational structure pre-2020, we find that sectors with higher exposure experienced statistically significant gains in the wage bill, employment, and hourly wages alongside more insignificant increases in establishments. These results are consistent with individual-level data from the Census Bureau: hourly wages rise more in exposed occupations for younger and more educated workers, which is consistent with early adoption, faster skill acquisition, or task reallocation toward workers better positioned to integrate AI into workflows. We also find that decomposing exposure matters. Sectors intensive in tasks that require human–AI collaboration show gains in both employment and wages, while sectors with higher shares of directly automatable tasks experience declines in wage bills and headcount. Aggregated exposure measures can conceal offsetting movements within industries between complementary and substitutable roles.

Takeaways by Macro Roundup® AI

  1. U.S. state–industry pairs with higher generative AI task exposure show statistically significant gains in the wage bill, employment, and hourly wages in the years surrounding the public release of large language models, alongside more insignificant increases in establishments.
  2. Sectors intensive in tasks requiring human–AI collaboration show gains in both employment and wages, while sectors with higher shares of directly automatable tasks experience declines in wage bills and headcount.
  3. Census Bureau individual-level data show hourly wages rising more in exposed occupations for younger and more educated workers, which is consistent with early adoption, faster skill acquisition, or task reallocation.

Related Articles:

  • Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence — Using payroll records covering millions of US workers, the authors find employment for workers aged 22–25 in occupations most exposed to artificial intelligence, such as software developers and customer service representatives, declined 6%, while employment for older workers in those occupations rose 6–9%. Employment in less-exposed occupations grew comparably across age groups.
  • AI and Jobs: The Final Word (Until the Next One) — Unemployment has risen by 0.30 percentage points for the fifth of workers most exposed to artificial intelligence, compared with 0.94 percentage points for the least exposed fifth, leaving no visible AI effect on jobs. The share of workers leaving the labor force within a year is lowest among the most AI-exposed workers and roughly flat.
  • Expertise — The authors' model of automation predicts that occupations with rising expertise requirements see higher wages but falling employment, while those with declining requirements see lower wages alongside rising employment. Analyzing employment and earnings data by occupation across four decades, they find changes in occupational expertise strongly predict changes in occupational wages.

AI Summary. Chinese factories install about 280,000 industrial robots a year, half the global total, with roughly half made by domestic producers that undercut foreign rivals on price, lifting China's robot-to-worker density above Germany's. Trade data show China's global export share in labour-intensive goods rose, including small manufactured goods up 9 percentage points to 52.3%.

Ryan McMorrow, Haohsiang Ko and William Langley Financial Times
Date Posted:
Is Database:
Database

Btw 2019 and 2023, automation helped China increase its export share in labor-intensive sectors. During that period, Chinese manufacturers added ~280,000 industrial robots every year, 50% of the global total.

Are low-cost domestic robots expanding China's export share in labor-intensive goods?

Core argument: Chinese factories install roughly 280,000 industrial robots annually, half the global total, lifting China's robot-to-worker density above Germany's and closing in on leader South Korea.

China’s homegrown robot makers are driving a wave of low-cost automation that is helping local factories churn out more goods at lower prices, allowing the country to increase its share of exports, even in labour-intensive products. Chinese factories are installing about 280,000 industrial robots every year, or half the global total, bringing the country’s robot-to-worker density ahead of Germany and closing in on leader South Korea. Data from Chinese research group MIR Databank shows that about half of those robots are made by domestic groups such as Chengdu CRP Robot Technology, which has won local customers by undercutting global rivals on price. Trade data compiled by Harvard’s Growth Lab shows China grew its global export share in a swath of labour-intensive industries from 2019 to 2023. The country’s global export share of small manufactured goods, such as brooms, mops and pens, rose 9 percentage points to 52.3% over the four-year period. Furniture exports gained about 1.5 percentage points of market share, while China’s proportion of the world’s toy exports rose from 54.3% to 56.9%.

Takeaways by Macro Roundup® AI

  1. Chinese factories install roughly 280,000 industrial robots annually, half the global total, lifting China's robot-to-worker density above Germany's and closing in on leader South Korea.
  2. China's global export share of small manufactured goods such as brooms, mops and pens rose 9 percentage points to 52.3% between 2019 and 2023, according to trade data compiled by Harvard's Growth Lab.
  3. About half of the robots installed in China are made by domestic groups such as Chengdu CRP Robot Technology, which has won local customers by undercutting global rivals on price, according to MIR Databank.

Related Articles:

  • America Is Missing The New Labor Economy – Robotics Part 1 — Building an identical robotic arm in the US costs about 2.2x more than in China, and US-labeled components still depend heavily on Chinese parts and materials with no scalable alternative. The author argues China is the only country positioned to capture full-scale intelligent robotic automation, which would let it outcompete the US across production.
  • Global Robot Density in Factories Doubled in Seven Years — Global factory robot density reached a record 162 robots per 10,000 manufacturing employees, more than double the level of seven years earlier. The European Union averages 219 units per 10,000 employees, North America 197, and Asia 182, with Korea, Singapore, mainland China and Japan among the ten most automated economies.
  • China is Suffering its Own ‘China Shock’ — Academic analysis of 12 labour-intensive Chinese manufacturing industries estimates average employment fell about 14%, or nearly 4mn jobs, between 2011 and 2019, with textile roles down 40%; a further analysis of the same sectors found 3.4mn more jobs lost through 2023. Researchers argue industrial upgrading and automation are cutting demand for lower-skilled workers and depressing their wages.

AI Summary. The U.S. Court of Appeals for the Federal Circuit ruled 7-4 that the International Emergency Economic Powers Act — a statute that primarily authorizes sanctions and does not mention tariffs — was unlikely to have been intended by Congress to grant the president nearly unlimited tariff authority. The White House is preparing a Supreme Court appeal.

Tony Romm and Ana Swanson New York Times
Date Posted:

The administration will appeal to the Supreme Court an appellate court ruling that the International Emergency Economic Powers Act does not allow the President to impose broad import taxes.

Does emergency powers law give the president unlimited tariff authority?

Core argument: The U.S. Court of Appeals for the Federal Circuit ruled 7–4 against President Trump's use of the International Emergency Economic Powers Act, finding it unlikely Congress intended to grant the president nearly unlimited authority to impose tariffs under that law.

[Last] Friday, the U.S. Court of Appeals for the Federal Circuit ruled 7 to 4 against Mr. Trump and his use of the International Emergency Economic Powers Act, finding that it was unlikely that Congress had intended to grant the president nearly unlimited authority to impose tariffs as part of that law. The statute, which primarily allows the president to impose sanctions, does not specifically mention tariffs. But Mr. Trump has invoked it repeatedly to issue withering global duties, most recently in an executive order imposing taxes on imports from about 90 countries in early August. President Trump is preparing to ask the Supreme Court to overturn [the] ruling that found many of his punishing tariffs to be illegal, claiming that any erosion in his power to wage a global trade war would inflict severe financial damage on the United States. An administration official, who spoke on condition of anonymity, confirmed the White House’s legal plans on Tuesday.

Takeaways by Macro Roundup® AI

  1. The U.S. Court of Appeals for the Federal Circuit ruled 7–4 against President Trump's use of the International Emergency Economic Powers Act, finding it unlikely Congress intended to grant the president nearly unlimited authority to impose tariffs under that law.
  2. The International Emergency Economic Powers Act primarily allows the president to impose sanctions and does not specifically mention tariffs, though Trump invoked it repeatedly, most recently in an early-August executive order taxing imports from about 90 countries.
  3. Trump is preparing to ask the Supreme Court to overturn the ruling, claiming that any erosion in his power to wage a global trade war would inflict severe financial damage on the United States, according to an administration official.

Related Articles:

  • Tariffs Are Generating Meaningful New Revenue — New U.S. tariffs are estimated to generate $1.3tn in net new revenue through the end of the current presidential term and $2.8tn through 2034 before economic effects, about $600bn more than previously enacted tariffs. If a court ruling striking down some tariffs is upheld, the remainder could raise as little as $800bn through 2034.
  • Trump’s Tariffs Are Making Money. That May Make Them Hard to Quit — Tariff revenue may make future administrations of either party reluctant to roll back tariffs, since repeal would add to the federal debt and replacing the revenue with other taxes would require a politically risky vote in Congress. Democrats may also be tempted to use tariff revenue to fund new social programs.
  • Tariffs are Dumb Enough to (Almost) Work — A universal 15% tariff on imported goods could raise roughly $300–$400bn annually, a fraction of what a value-added tax would generate but real revenue given U.S. imports of about $4tn. The author argues tariffs are a poorly designed form of consumption tax that could build political groundwork for replacing them with a more efficient value-added tax.

AI Summary. A Wall Street Journal-NORC poll found that 25% of Americans say they have a good chance of improving their standard of living, a record low in surveys dating to 1987, while nearly 70% said the idea that hard work leads to getting ahead no longer holds true or never did.

Lindsay Ellis and Aaron Zitner Wall Street Journal
Date Posted:
Is Database:
Database

A WSJ-NORC poll finds the % of Americans who say they have “a good chance of improving their standard of living” fell to 25%, a record low since 1987. The survey results are more negative than predicted by macro conditions, suggesting unmeasured fears.

Do Americans still believe hard work leads to getting ahead?

Core argument: The share of Americans saying they have a good chance of improving their standard of living fell to 25% in the WSJ-NORC poll, a record low in surveys dating to 1987.

A new Wall Street Journal-NORC poll finds that the share of people who say they have a good chance of improving their standard of living fell to 25%, a record low in surveys dating to 1987. More than three-quarters said they lack confidence that life for the next generation will be better than their own, the poll found. Nearly 70% of people said they believe the American dream—that if you work hard, you will get ahead—no longer holds true or never did, the highest level in nearly 15 years of surveys. Republicans in the survey were less pessimistic than Democrats, reflecting the longstanding trend that the party holding the White House has a rosier view of the economy. An index that combined six poll questions found that 55% of Republicans, as well as 90% of Democrats, held a negative view of prospects for themselves and their children.

Takeaways by Macro Roundup® AI

  1. The share of Americans saying they have a good chance of improving their standard of living fell to 25% in the WSJ-NORC poll, a record low in surveys dating to 1987.
  2. Nearly 70% of respondents said the American dream—that hard work leads to getting ahead—no longer holds true or never did, the highest level in nearly 15 years of surveys, and more than three-quarters lacked confidence the next generation's life will be better.
  3. An index combining six poll questions found 90% of Democrats and 55% of Republicans held a negative view of prospects for themselves and their children, reflecting the longstanding trend that the party holding the White House views the economy more favorably.

Related Articles:

  • The Cost of Money is Part of the Cost of Living: New Evidence on the Consumer Sentiment Anomaly — Variation in the University of Michigan consumer sentiment index that inflation and unemployment cannot explain is strongly correlated with growth in consumer borrowing costs, such as mortgage and car loan rates. The authors estimate that including interest payments and homeownership costs in the consumer price index would raise measured inflation from 3% to 9%.
  • Zero-Sum Thinking and the Roots of U.S. Political Divides — Zero-sum thinking — viewing gains for one group as losses for another — is associated with greater support for liberal economic policies, stronger alignment with the Democratic Party, and weaker alignment with the Republican Party. The authors also find it linked to believing the QAnon conspiracy theory holds some truth and to sympathy for the attack on the U.S. Capitol.
  • Asymmetric Amplification and the Consumer Sentiment Gap — Consumer sentiment surveys show Republican respondents' views swing about ±15 index points relative to predicted levels depending on which party holds the presidency, while Democrats' views swing only about ±6 points. The authors estimate that adjusting for this asymmetric amplification shrinks the gap between observed and predicted consumer sentiment by 30%.

AI Summary. Gold reached a record above $3,500 a troy ounce amid a weak dollar and expectations of a US interest rate cut. Prices have nearly doubled since the start of 2023 as central banks add holdings and investors seek an inflation hedge, with gold now 20% of global official reserves.

William Sandlund Financial Times
Date Posted:

Gold prices hit a record high of $3,508.70/troy ounce; prices have doubled since the beginning of 2023. Last year, gold became the world’s second-most common central bank reserve asset after the dollar.

Are a weak dollar and rate-cut expectations driving gold to records?

Core argument: Spot gold rose 0.9% to a record $3,508.70 a troy ounce in early Asian trading before paring gains to $3,497, surpassing the previous high set in April after President Donald Trump's "liberation day" tariffs.

Gold prices jumped to an all-time high on Tuesday, surpassing a previous record in April after President Donald Trump’s “liberation day” tariffs, amid a weak dollar and expectations of a US interest rate cut this month. The spot price of gold rose 0.9 per cent to $3,508.70 a troy ounce during early trading in Asia, before paring gains to trade at $3,497. Gold prices have nearly doubled since the start of 2023 as central banks increase holdings and investors look for a hedge against inflation. Last year, the metal overtook the euro to become the world’s second-most common central bank reserve asset after the dollar, accounting for 20% of global official reserves.

Takeaways by Macro Roundup® AI

  1. Spot gold rose 0.9% to a record $3,508.70 a troy ounce in early Asian trading before paring gains to $3,497, surpassing the previous high set in April after President Donald Trump's "liberation day" tariffs.
  2. Gold prices have nearly doubled since the start of 2023 as central banks increase holdings and investors seek a hedge against inflation, with a weak dollar and expectations of a US interest rate cut supporting the move.
  3. Gold overtook the euro to become the world's second-most common central bank reserve asset after the dollar, accounting for 20% of global official reserves.

Related Articles:

  • Gold Overtakes Euro as Global Reserve Asset, ECB Says — Gold has overtaken the euro as a global reserve asset, with central bank gold holdings reaching 36,000 tons, near the mid-1960s peak of 38,000 tons, helped by a 30% rise in the gold price. The European Central Bank finds countries geopolitically close to China and Russia accumulated gold more than others.
  • How Gold Became the World’s Refuge from Uncertainty — Gold prices reached a record intraday high in real terms, surpassing the prior 1980 peak, with roughly a 30% gain over the year. Central banks have bought more than 1,000 tonnes net annually for three years — record levels — led by emerging markets including China, India and Turkey as they scale back dollar exposure.
  • Taking Stock: Dollar Assets, Gold, and Official Foreign Exchange Reserves — Gold accounts for about 10% of aggregate official foreign exchange reserves, roughly its level in the early 2000s. International Monetary Fund country data indicate most added official gold holdings came from a few central banks, with over half of reported accumulation from China and Russia and about a quarter from several emerging markets.

AI Summary. Regrouping US college majors to match Chinese disciplines yields a 2:1 ratio of science and engineering to humanities and social sciences, the same ratio as in China. The author argues America's building slowdown reflects structural economic shifts—urbanization reached 70% by 1960 and 85% by 2020, shifting needs toward maintenance—rather than a lawyerly culture.

Jonathon Sine Cogitations
Date Posted:
Is Database:
Database
Is Important:
Important

Jonathon Sine argues China “is moving beyond its breakneck industrial prime, facing similar dilemmas to those America confronted in the 1960s and 70s.” The ratio of science/engineering to humanities undergraduate majors is 2:1 in both the PRC and US.

Does America's building slowdown reflect structural shifts rather than lawyerly culture?

Core argument: Regrouping US college majors according to Chinese disciplines yields a 2:1 ratio of science/engineering to humanities/social sciences, the same ratio as in China when management is grouped with science/engineering in both.

Dan Wang’s “big idea” [is] “China is an engineering state, building big at breakneck speed, in contrast to the United States’ lawyerly society, blocking everything it can, good and bad.” I re-group US college majors according to Chinese disciplines to allow for rough comparison. Surprisingly, the ratio of science/engineering to humanities/social sciences is 2:1, the same as in China (if one groups management with science/engineering, as I also do for China). As with China today, America’s breakneck building phase was decidedly winding down by the 1960s. Urbanization went from 40% in 1900 to 70% by 1960, and grew much more incrementally over the next 60 years to 85% by 2020. The country simply did not need to continue building dams, expressways, and energy production facilities at breakneck pace. It became much more a matter of maintaining and upgrading (which has not gone well, at least according to the American Society of Civil Engineers’ report card). The American [building/investment slowdown that started after the 1970s] may be more about structural economic shifts than lawyers.

Takeaways by Macro Roundup® AI

  1. Regrouping US college majors according to Chinese disciplines yields a 2:1 ratio of science/engineering to humanities/social sciences, the same ratio as in China when management is grouped with science/engineering in both.
  2. US urbanization rose from 40% in 1900 to 70% by 1960, then grew more incrementally to 85% by 2020, with America's breakneck building phase winding down by the 1960s.
  3. The US building and investment slowdown that began after the 1970s may owe more to structural economic shifts than to lawyers, qualifying Dan Wang's contrast between China's engineering state and America's lawyerly society.

Related Articles:

  • Industrial Colossus: China vs 1950s America — China's growth in its share of global manufacturing has slowed from roughly 1.5 percentage points per year in the early 2010s to just over 0.5 percentage points annually, about one-third the earlier pace. The author argues China's share of global manufacturing is likely near its peak.
  • The Real China Model — China generates more electricity annually than the United States and European Union combined, having added capacity equivalent to the United Kingdom's total supply each year. Electricity makes up nearly 30% of China's energy use, versus 22% in the US and 21% globally, and its share is growing about 6% a year.
  • America’s Housing Affordability Crisis and the Decline of Housing Supply — Housing production intensity has fallen substantially, especially in fast-growing Sunbelt markets once known for building, converging toward the low rates long seen on the coasts and in declining industrial centers. The authors argue incumbent-controlled regulation explains the weakening link between high prices, high demand areas, and new construction.

Wednesday, September 3, 2025

AI Summary. Among the largest US industrial firms, those founded between 1880 and 1920 remain the most represented, while top retailers and wholesalers have averaged 60–70 years old and large services firms cluster around founding dates of 1960–1980. The authors model such a "special generation" as arising from adopting a scalable technology plus learning-by-doing.

Yueran Ma, Benjamin Pugsley, Haomin Qin and Kaspar Zimmermann National Bureau of Economic Research
Date Posted:
Is Database:
Database
Is Important:
Important

An outsized share of the top manufacturing firms in the Fortune 1000 as of 2018 originated circa 1900, apparently the persistent echo of an “industrial revolution,” while the less technology-based retail and wholesale sectors lack a “special generation.”

Why do the largest industrial firms share a common founding era?

Core argument: Among 2018 Fortune 1,000 industrials, firms founded between 1880 and 1920 are the most represented cohort, and top manufacturing firms in the 1910s, 1950s, and 2010s alike predominantly date back to around 1900.

In manufacturing, top firms in the 1910s, 1950s, and 2010s predominantly date back to around 1900, [though] even as this special generation persists, turnover among [these] top firms has been substantial. In Figure 1, we plot the number of 2018 Fortune 1,000 companies that originated in each decade, based on the firm’s industry in 2018. For industrials, we observe that the cohort from 1880 to 1920 is most represented among today’s largest industrials. In contrast the largest retailers and wholesalers have always stayed 60 to 70 years old on average, without a special generation. Today, birth years of services companies cluster around 1960 to 1980; few large ones existed before, and few services companies would qualify for the largest businesses in the economy until recently. In a model of firm dynamics a special generation can arise from an industrial revolution, through the adoption of a scalable technology and learning-by-doing; lasting nonstationarity among the dynamics of top firms can result from the long echoes of technological change. [For] retailers and wholesalers, learning appears absent, so a special generation would be harder to sustain.

Takeaways by Macro Roundup® AI

  1. Among 2018 Fortune 1,000 industrials, firms founded between 1880 and 1920 are the most represented cohort, and top manufacturing firms in the 1910s, 1950s, and 2010s alike predominantly date back to around 1900.
  2. The largest retailers and wholesalers have always averaged 60 to 70 years old with no special founding generation, while large services companies cluster in birth years from 1960 to 1980.
  3. In a model of firm dynamics, a special generation can arise from an industrial revolution through adoption of a scalable technology and learning-by-doing, with learning appearing absent among retailers and wholesalers.

Related Articles:

  • Which U.S. Stocks Generated the Highest Long-Term Returns? — Of the 29,078 U.S. common stocks in a database spanning nearly a century, 51.6% had negative cumulative returns, while 17 stocks returned more than 5m% — led by Altria at 265m%. Those top performers averaged annualized returns of just 13.47%, underscoring the value of time invested.
  • Long-Term Shareholder Returns: Evidence From 64,000 Global Stocks — A study of about 64,000 global stocks estimates net global stock market wealth creation of $75.7tn, with returns highly concentrated: the top 159 firms (0.25%) account for half of net wealth creation and the top 1,526 (2.39%) account for all of it. Fewer than half of stocks beat one-month US Treasury bills, 42.6% of non-US and 44.8% of US stocks.
  • The Long Goodbye? Incumbent Competitiveness in the 21st Century — Value creation among large American companies has become concentrated in a newer cohort of firms: businesses founded after World War II accounted for 8% of the top 50 companies' total market value in 1990 but 77% by 2024. The leading 20th-century incumbents are fading as this newer group takes over.

AI Summary. Weaker-than-expected US job openings data prompted bets on more aggressive Federal Reserve rate cuts, driving a global bond rally that halted a sell-off which had lifted borrowing costs in several major economies to multi-year highs. The 30-year US Treasury yield fell 0.07 percentage points to 4.9%.

Ian Smith, Leo Lewis and Arjun Neil Alim Financial Times
Date Posted:
Is Database:
Database

Amid a global sell-off of long bonds, the 30-year Treasury yield hit 5% until disappointing data on US job openings caused it to retreat to 4.9%, and Japan’s 30-year yield briefly hit a record high.

Does weak US labor data spur a global bond rally?

Core argument: The 30-year US Treasury yield fell 0.07 percentage points to 4.9%, after climbing to 5% for the first time since July, as disappointing US job openings data spurred bets on more aggressive Federal Reserve rate cuts.

Bond markets rallied on Wednesday, recovering from a global sell-off, as disappointing data on US job openings prompted bets that the Federal Reserve would cut rates more aggressively to support the world’s biggest economy. The rally halted a bond market slide that had pushed borrowing costs in some big economies to their highest levels in years. The yield on the 30-year US Treasury fell 0.07 percentage points to 4.9%, recovering after climbing to 5% for the first time since July. Bond yields move inversely to prices. The moves powered a recovery in sentiment towards sovereign bonds around the world, after Japan’s 30-year yield earlier hit a record high and the UK’s 30-year gilt yield reached a fresh post-1998 high at 5.75%. The 30-year gilt yield fell back to 5.62%.

Takeaways by Macro Roundup® AI

  1. The 30-year US Treasury yield fell 0.07 percentage points to 4.9%, after climbing to 5% for the first time since July, as disappointing US job openings data spurred bets on more aggressive Federal Reserve rate cuts.
  2. The UK's 30-year gilt yield retreated to 5.62% after reaching a fresh post-1998 high of 5.75%, while Japan's 30-year yield had earlier hit a record high.
  3. The bond rally halted a global slide that had pushed borrowing costs in some big economies to their highest levels in years.

Related Articles:

  • Why Is The Yield Curve Steepening? Because Of Coming Fed Cuts And Fiscal Challenges — The US yield curve is steepening across both the 2-to-10-year and 10-to-30-year segments. The article attributes this to Federal Reserve rate cuts, the risk that markets read those cuts as politically motivated and raise inflation expectations and long-term rates, and growing Treasury debt issuance pushing long rates up.
  • The Race Between Asset Supply and Asset Demand — The authors estimate that demand for assets has outpaced supply by roughly 400% of GDP versus 100%, making rising asset demand the leading contributor to falling interest rates. They project that debt sustainability requires fiscal consolidation exceeding 10% of GDP, though debt could reach 250% of GDP without raising interest rates.
  • Data Update 4 for 2025: Interest Rates, Inflation and Central Banks! — Interest rates are ultimately driven by macroeconomic fundamentals, with an "intrinsic risk-free rate" defined as the sum of the inflation rate and real GDP growth tracking the observed 10-year Treasury bond yield closely. On this view, central banks' influence comes from preferential access to data on those fundamentals, their capacity to alter them, and their credibility.

AI Summary. The authors estimate US tariffs raised $146bn in net customs duty revenues over the first eight months of 2025, with roughly $88bn of that attributable to new 2025 tariffs. August's new-tariff revenue of about $23bn ran roughly a fifth below the $29bn per month projected from statutory rates.

Ernie Tedeschi Yale University
Date Posted:

Ernie Tedeschi estimates tariffs have raised ~$88B in new revenue since January, with the new tariffs raising ~$23B in revenue in August alone. The actual average effective tariff was likely ~11.5% in August relative to 2.4% at the start of the year.

How much revenue have new tariffs actually collected so far?

Core argument: The US collected an estimated $146 billion in total net customs duty revenues in calendar 2025 through August, of which roughly $88 billion came from the new 2025 tariffs.

[We] estimate that the US has raised $146 billion in total net customs duty revenues in calendar year 2025 so far through August. This total stems from all US tariffs, not just those adjusted in 2025. Before the 2025 tariff announcements, the average effective US tariff rate was 2.4% which raised about $7 billion a month in net customs duty revenues. The new 2025 tariffs have raised roughly $88 billion in revenues so far, with about $23 billion of that new tariff revenue posted in August alone when the average effective tariff rate was likely 11-12%, up from 10% in June and July. This comes to roughly 0.8% of GDP each month, a meaningful amount of deficit reduction. The new tariffs are raising a bit less revenue than would be expected given where policy has been set. One way of illustrating is comparing actual to expected revenues. TBL projects that new 2025 tariff policy as of August 6 would be expected to raise an additional $29 billion per month in net tariff revenues, but as mentioned above, actual revenues in August will likely be $23 billion higher, about a fifth less than we would project given current statutory tariff rates.

Takeaways by Macro Roundup® AI

  1. The US collected an estimated $146 billion in total net customs duty revenues in calendar 2025 through August, of which roughly $88 billion came from the new 2025 tariffs.
  2. About $23 billion of new tariff revenue posted in August alone, when the average effective US tariff rate was likely 11–12%, up from 10% in June and July and 2.4% before the 2025 tariff announcements.
  3. Actual August tariff revenue of roughly $23 billion came in about a fifth below the $29 billion per month that TBL projects from statutory tariff rates as of August 6.

Related Articles:

  • Tariffs Are Generating Meaningful New Revenue — New U.S. tariffs are estimated to generate $1.3tn in net new revenue through the end of the current presidential term and $2.8tn through 2034 before economic effects, about $600bn more than previously enacted tariffs. If a court ruling striking down some tariffs is upheld, the remainder could raise as little as $800bn through 2034.
  • Trump Signals Imminent Supreme Court Appeal to Protect Tariffs — The U.S. Court of Appeals for the Federal Circuit ruled 7-4 that the International Emergency Economic Powers Act — a statute that primarily authorizes sanctions and does not mention tariffs — was unlikely to have been intended by Congress to grant the president nearly unlimited tariff authority. The White House is preparing a Supreme Court appeal.
  • Are Businesses Absorbing the Tariffs or Passing Them on to Their Customers? — About three-quarters of manufacturing and service firms facing tariff-driven cost increases reported passing at least some of those costs to customers through higher prices, while roughly a quarter of each said they absorbed the full increase without raising prices. Nearly a third of manufacturers and about 45% of service firms reported full pass-through.

AI Summary. Technology-sector capital spending has accounted for 35%–45% of US GDP growth over three quarters, implying that without it growth would be near 1% given weak consumption. Among S&P 500 companies, capital spending rose 24% and research and development 10%, but the median stock grew both by only 3%–4%.

Michael Cembalest J.P. Morgan
Date Posted:
Is Database:
Database

Capital spending by the tech sector drove 35–45% of US GDP growth over the past three quarters. Without this capex, Mike Cembalest notes, “US growth might already have converged to ~1% due to weakness in consumption.”

Is US growth dependent on tech capital spending?

Core argument: Tech sector capital spending has contributed 35%–45% of overall US GDP growth over the last three quarters, and without it US growth might already have converged to ~1% given weakness in consumption.

Tech sector capital spending has been contributing 35%-45% of overall US GDP growth over the last three quarters. This is remarkable; in other words, without tech capital spending, US growth might already have converged to ~1% due to weakness in consumption. Some GDP growth impacts are exaggerated since tech equipment is partially imported from Taiwan, but the general trend is the same: tech capital spending is an increasingly important contributor to both growth and equity valuations. Investment spending growth for the S&P 500 looked good in Q2 at 24% for capex and another 10% for R&D; but for the median stock, both grew by just 3%-4%.

Takeaways by Macro Roundup® AI

  1. Tech sector capital spending has contributed 35%–45% of overall US GDP growth over the last three quarters, and without it US growth might already have converged to ~1% given weakness in consumption.
  2. S&P 500 investment spending grew 24% for capex and 10% for R&D in the second quarter, but for the median stock both grew by just 3%–4%.
  3. Some GDP growth impacts of tech capital spending are exaggerated because tech equipment is partially imported from Taiwan, though tech capital spending remains an increasingly important contributor to both growth and equity valuations.

Related Articles:

  • How America’s AI Boom Is Squeezing The Rest Of The Economy — About a sixth of America's 2% real GDP growth over the past year came from investment in computer and communications equipment and data centres, and one estimate puts the artificial intelligence boom's total contribution to growth at 40%. Average electricity bills rose 7%, attributed in part to data-centre demand, while housebuilding and non-AI business investment slumped and real consumption flatlined.
  • Back to Our Regularly Scheduled Programming — The four largest US hyperscalers spent $440bn on capital expenditure and research in 2024, with $596bn projected for 2025, a 35% increase, leaving large gaps between AI spending and AI revenues. These firms account for more than half of all capital expenditure growth in the US equity market.
  • Honey, AI Capex is Eating the Economy — Capital spending on AI data centers is around 20% of railroads' peak spending as a share of GDP and is rising quickly, having already surpassed the dot-com era peak in telecom investment. Unlike long-lived rail infrastructure, data centers are short-lived assets requiring frequent hardware replacement, and the capital is being reallocated from other sectors.

AI Summary. Employment for workers aged 22-25 and 26-30 has declined relative to its pre-ChatGPT trend and relative to older workers in the same occupation, but the age groups shown are not equally sized. The authors argue that combining all age groups shows software developer jobs have continued to grow overall.

David Deming Forked Lightning
Date Posted:
Is Database:
Database

Deming argues that the drop in headcount of software developers aged 22–25 may merely be a reflection of the experience of a particular cohort constituting only 4 years of observations, not the “canary in the coal mine” suggested by Brynjofsson et al.

Are software developer jobs still growing despite declines among young workers?

Core argument: Employment for workers aged 22–25 and 26–30 has fallen relative to its pre-ChatGPT trend and relative to older workers in the same occupation, according to the headline finding the authors cite.

The authors emphasize the headline finding, which is that employment for young workers (age 22-25 in blue, and age 26-30 in orange) has decreased relative to its pre-ChatGPT trend and also relative to older workers in the same occupation. While the figure above shows that young workers are suffering more, it’s a little misleading. Those lines each represent a group of workers, but they are NOT the same size. Your eyes are drawn to the blue line, but it only has 4 years in it. The top line (in purple) has 9 years in it (ages 41-49). If age were uniformly distributed, those two lines would roughly cancel out when weighted by size. If you smashed all the lines together, you’d see that software developer jobs have continued to grow overall through the release of ChatGPT.

Takeaways by Macro Roundup® AI

  1. Employment for workers aged 22–25 and 26–30 has fallen relative to its pre-ChatGPT trend and relative to older workers in the same occupation, according to the headline finding the authors cite.
  2. The authors argue the age-group chart misleads because the bands differ in size, with the 22–25 group spanning 4 years and the 41–49 group spanning 9, so under a uniform age distribution the two lines would roughly cancel when weighted by size.
  3. Aggregating all age groups, software developer employment has continued to grow overall through the release of ChatGPT.

Related Articles:

  • Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence — Using payroll records covering millions of US workers, the authors find employment for workers aged 22–25 in occupations most exposed to artificial intelligence, such as software developers and customer service representatives, declined 6%, while employment for older workers in those occupations rose 6–9%. Employment in less-exposed occupations grew comparably across age groups.
  • AI and Jobs: The Final Word (Until the Next One) — Unemployment has risen by 0.30 percentage points for the fifth of workers most exposed to artificial intelligence, compared with 0.94 percentage points for the least exposed fifth, leaving no visible AI effect on jobs. The share of workers leaving the labor force within a year is lowest among the most AI-exposed workers and roughly flat.
  • Technological Disruption in the US Labor Market — The authors present four signs that US labor market change has accelerated, possibly from technology: polarization has ended as low- and middle-paid employment declined while highly paid work grew, low-paid service job growth stalled, the STEM employment share rose over 50%, and retail sales employment fell 25%. They argue artificial intelligence's effects will unfold over decades.

AI Summary. General Social Survey data show the share of US adults ages 18–64 reporting weekly sex fell from 55% in 1990 to 37% by 2024. The decline extends to married adults across age groups, whose weekly rate fell from 59% to 49%.

Grant Bailey and Brad Wilcox Institute for Family Studies
Date Posted:
Is Database:
Database

According to the General Social Survey, the share of Americans (18–64) reporting they had sex weekly dropped from 55% in 1990 to 37% in 2024. 24% of young adults 18–29 report not having sex within the last year.

Is the share of Americans having weekly sex declining?

Core argument: The share of U.S. adults ages 18–64 reporting weekly sex fell from 55% in 1990 to under 50% by 2010 and 37% in 2024, according to the General Social Survey.

Americans are having a record low amount of sex. We find that in 1990, 55% of adults ages 18–64 reported having sex weekly, according to the General Social Survey (GSS). But around the turn of the millennium, that number began to dip: by 2010, less than half reported having sex weekly, and by 2024, of the more than 1,000 men and women queried on this topic by the GSS, that number had fallen to just 37%. Notably, the sex recession is making inroads among married couples. Between 1996 and 2008, 59% of married adults, ages 18-64, reported having sex once a week or more. That number fell to 49% for the period of 2010 to 2024. Married couples are seeing declines in sexual frequency across age groups.

Takeaways by Macro Roundup® AI

  1. The share of U.S. adults ages 18–64 reporting weekly sex fell from 55% in 1990 to under 50% by 2010 and 37% in 2024, according to the General Social Survey.
  2. Among married adults ages 18–64, 59% reported sex at least weekly over 1996–2008, falling to 49% over 2010–2024, with declines in sexual frequency across age groups.
  3. The General Social Survey's 2024 figure of 37% reporting weekly sex draws on more than 1,000 men and women queried on the topic.

Related Articles:

  • 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.
  • Disconnected: The Growing Class Divide in American Civic Life — 17% of Americans report having no close friends, and survey data show a wide educational gap: 24% of those with a high school education or less have no close friends versus 10% of college graduates. Three decades ago, Americans with less formal education reported social circles at least as large.
  • Declining Life Satisfaction and Happiness Among Young Adults in Six English-speaking Countries — Eleven studies find declines in life satisfaction and happiness among young adults in the US, Australia, Canada, Ireland, New Zealand, and the UK. In these countries life satisfaction now rises with age across the lifespan, as young people's life satisfaction dropped sharply while that of people 45 and over stayed flat.