“…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
“…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
“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
“…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
“Unintended Consequences represents the most cogent and persuasive analysis of the Financial Crisis to date.” - Andrei Shleifer, 1999 John Bates Clark Medal Winner
“…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
“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
“Unintended Consequences is full of substance, it is one of the must-read books of the year, and once I finish it I will be giving it a second read through right away.” - Tyler Cowen, Professor, George Mason University
“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
“Unintended Consequences offers deep and well-argued analyses on almost every issue.” - The New York Times
“…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
Research on business investment, capital formation, equipment, structures, software, and the conditions that turn spending into higher productivity. Coverage connects financing, expected returns, policy, technology, and long-run growth.
AI Summary.A survey of 396 enterprises finds only 11% forecast their AI spending within 10% of actual, down from 15% a year earlier, while 62% say an unexpected AI cost altered a business decision and one in four delayed or canceled an AI initiative. 81% report moderate-to-high gross margin impact from AI costs.
Torsten SlokApollo
Date Posted:
Slok flags a survey of 396 enterprises that found AI adaptation hasn’t improved gross margins in aggregate, with 81% of firms noting a “moderate-to-high gross margin impact” from their AI spending, while 63% saw a 6 to 15pp decrease.
Are companies losing control of their AI spending forecasts?
Core argument: Only 11% of the 396 enterprises surveyed by Mavvrik and Benchmarkit forecast their AI spending within 10% of actual, down from 15% a year earlier.
A new survey by Mavvrik and Benchmarkit of 396 enterprises finds that companies are adopting AI faster than they are building the financial controls to keep its costs in check. Only 11% of firms forecast their AI spending within 10% of actual, down from 15% last year. The survey also shows that those misses are changing decisions, with 62% of firms saying an unexpected AI cost altered a business decision and one in four delaying or canceling an AI initiative. The impact is showing up in profitability too, with 81% of companies reporting moderate-to-high gross margin impact from AI costs, including 63% who see a 6 to 15 percentage point decrease.
Takeaways by Macro Roundup® AI
Only 11% of the 396 enterprises surveyed by Mavvrik and Benchmarkit forecast their AI spending within 10% of actual, down from 15% a year earlier.
In the Mavvrik and Benchmarkit survey, 62% of firms said an unexpected AI cost altered a business decision, and one in four delayed or canceled an AI initiative.
The survey reports that 81% of companies see moderate-to-high gross margin impact from AI costs, including 63% citing a 6 to 15 percentage point decrease.
Related Articles:
Tech’s Trillion-Dollar Internal Inconsistency— Wall Street analysts covering technology project the sector's operating cash flow will more than double to roughly $2.4tn by 2028, a gain of over $1.2tn, while analysts covering the rest of the S&P 500 — technology's customers — project those firms will add far less cash flow. These sector-by-sector forecasts are internally inconsistent.
How Much Revenue Does The AI Buildout Need In 2030?— The authors estimate AI capital spending of $560bn in 2026 and, in a high-investment case, $3.5tn of net AI capital in service by 2030, requiring roughly $1.3tn in annual infrastructure revenue once depreciation, operating costs and a 15% return are covered. That implies 7.4x growth from current annualized revenues, with downstream revenues of $1.9–2.6tn.
Writing Code vs. Shipping Code: Productivity Effects Across Generations of AI Coding Tools— The authors estimate that AI coding tools raise developers' task-level productivity by about 40% with autocomplete, 140% including synchronous agents, and 180% with asynchronous agents, with larger effects for less active developers. Gains attenuate up the production chain: synchronous agents raise lines of code 741% and pull requests 65% but releases only 20%.
AI Summary.Across 12 middle-income countries, call-center employment grew in line with other white-collar office work before the spread of AI chatbots but has since stagnated or declined even as other white-collar jobs held up. The authors argue this divergence makes artificial intelligence's role hard to dismiss.
Caelan Wilkie-RogersRevelio Labs
Date Posted:
A Revelio Labs analysis finds global call center employment is now 5.3% under its December 2023 peak. Call center employment started diverging from other white collar work in middle income countries around November 2022, when ChatGPT was released.
Is AI reducing call center employment in middle-income countries?
Core argument: Across 12 middle-income countries, call-center employment showed sluggish growth or outright declines after ChatGPT’s release, while other white-collar office and administrative roles remained fairly robust.
Across the 12 middle-income countries, the pattern is recognizable: in the pre-ChatGPT era, call-center headcount generally kept pace with other office and administrative, "white-collar" roles. In the post-ChatGPT era, we see sluggish growth or outright declines in call centers—even as the rest of white-collar work looks fairly robust. So, although we cannot observe which firms replace people with chatbots, this divergence in headcounts between call centers and the rest of their labor markets makes AI’s potential role in this trend hard to wave off. Moreover, call centers’ high-volume, scripted work is exactly the profile the International Labour Organization flags as most exposed in these countries. The industry is shrinking because hiring slowed, not because separations picked up.
Takeaways by Macro Roundup® AI
Across 12 middle-income countries, call-center employment showed sluggish growth or outright declines after ChatGPT’s release, while other white-collar office and administrative roles remained fairly robust.
Before ChatGPT, call-center headcount in these 12 middle-income countries generally kept pace with other office and administrative employment.
The ILO flags high-volume, scripted work—the profile of call-center jobs—as most exposed in these middle-income countries, though which firms replace workers with chatbots cannot be observed.
Related Articles:
AI’s Most Exposed Industry Keeps Hiring— Call center employment in the Philippines has reached nearly 2m workers and has grown every year since 2016, including through the artificial intelligence boom. As AI lowers the cost per customer interaction, companies are buying more of these services rather than fewer — an example of Jevons paradox.
Looking for the Ladder— The article argues that the hiring downturn in occupations most exposed to generative AI reflects interest-rate sensitivity rather than AI itself, since roughly 38% of workers in the most AI-exposed quintile work in information, finance and insurance, and professional and technical services, versus under 2% in the least-exposed quintile.
The Early Impacts of AI on Employment among Recent College Graduates— An expanded unemployment measure for young college graduates — including those outside the labor force who report wanting a job — averaged 10.4% in the most recent summer, the highest of the five summers compared and 0.3 percentage points above the prior high. Estimated associations with occupational AI exposure are positive but not statistically significant.
AI Summary.The United States generates almost twice as much economic output per dollar invested as China, the opposite of what would be expected from a mature developed economy versus a fast-growing economy catching up. The gap points to overexuberance in Chinese investment.
Robin Rivaton and Ben SouthwoodWorks In Progress
Date Posted:
The US generates ~2x as much output per dollar of capital as China. Rivaton and Southwood argue that relentless competition among Chinese subnational governments, who account for ~85% of public expenditure, drives industrial growth at the cost of reduced capital efficiency.
Does China get less economic output per dollar invested than America?
Core argument: The United States obtains nearly twice as much GDP from every dollar invested as China does, measured by the capital efficiency ratio of output to invested capital stock.
Overexuberance is visible in economic data. Interest rates that banks pay to depositors are capped, giving banks artificially cheap capital. Entrepreneurs may be backed to attempt extremely unlikely, borderline schemes. The capital efficiency ratio measures how much output a country gets out of the capital stock it has invested – how many dollars of GDP a country can buy with a dollar of investment. The United States gets almost twice as much economic output out of every dollar it invests as China does, precisely the opposite of what we would expect from a developed country with a mature economy versus a rapidly growing country playing catch up.
Takeaways by Macro Roundup® AI
The United States obtains nearly twice as much GDP from every dollar invested as China does, measured by the capital efficiency ratio of output to invested capital stock.
China’s capital efficiency inverts the expected pattern, in which a rapidly growing catch-up economy would extract more output per dollar of investment than a mature developed economy.
Overexuberance is visible in Chinese macroeconomic data, with China’s investment generating less economic output per dollar than that of the United States.
Related Articles:
The Year of the Trojan Fire Horse: China’s Imbalanced Economy and Unrelenting Mercantilism— About 30% of Chinese industrial companies are unprofitable "zombies" kept alive by subsidies far larger than those elsewhere, rather than exiting amid falling industrial producer prices. Countries with large increases in manufactured imports from China also saw slowdowns in domestic manufacturing, including declining manufacturing shares of output across most of Southeast Asia.
Xi’s Superpower Ambitions Fuel a Dangerous Debt Spiral in China— China's credit to the nonfinancial sector—corporate, government and household borrowing—topped 300% of gross domestic product, 60 percentage points above its end-2019 level and larger than in the US, according to the Bank for International Settlements. The International Monetary Fund estimates broad Chinese government debt, including local financing vehicles, reached 127% of GDP, up from 98% in 2020.
Is China’s High-Quality Investment Output Economically Viable?— China's technological progress and infrastructure do not refute claims of overinvestment, since overinvestment often produces real physical assets and technological gains yet still counts as overinvestment when it generates less real value than the resources, subsidies, and transfers it consumes. When such losses are capitalized, assets are carried above their real economic value.
AI Summary.Goldman Sachs estimates that artificial intelligence–related stocks account for 57% of Chinese equity trading turnover and that nearly $5tn in AI-related market value has been created, including $2tn from new listings. China also holds dominant shares of the physical supply chain, including 97% of humanoid robot shipments, 75% of electric vehicle production, and 90% of rare-earth refining.
Paul KedroskyApplied Complexity
Date Posted:
China accounts for 77% of global token usage, but only 22% of token spending, according to a GS analysis. Kedrosky notes, “This is what widespread AI adoption at low prices looks like. And it is the channel through which Chinese competition will pressure prices elsewhere.”
How large is AI's footprint in Chinese markets and supply chains?
Core argument: AI-related stocks account for 57% of trading turnover in Chinese equities, and Goldman estimates almost $5 trillion in AI-related market value has been created since the DeepSeek moment, including $2 trillion from IPOs.
AI-related stocks now represent 57% of trading turnover in Chinese equities. Since the DeepSeek moment, Goldman estimates almost $5 trillion in AI-related market value has been created, including $2 trillion from IPOs. AI dominates Chinese equity markets even more than it does U.S. markets. It's a one-factor world. China shipped 97% of humanoid robots in the first half of 2026. Alongside China’s 75% share of EV production and 80% of battery production, China basically owns the physical side of AI and AI-adjacent technologies. The industrial base behind the AI buildout is already concentrated in China: 60% of global transformer production capacity, 60% of PCB and optical-module production, 77% of new foundry capacity, and 90% of rare-earth refining. There are more than 880 AI-related companies focused mainly on China’s domestic market, with $3.2 trillion in market value. To put it in context, this is nearly half the total market of the Nasdaq at the peak of the dot-com frenzy.
Takeaways by Macro Roundup® AI
AI-related stocks account for 57% of trading turnover in Chinese equities, and Goldman estimates almost $5 trillion in AI-related market value has been created since the DeepSeek moment, including $2 trillion from IPOs.
China shipped 97% of humanoid robots in the first half of 2026, alongside a 75% share of EV production and 80% of battery production.
China holds 60% of global transformer production capacity, 60% of PCB and optical-module production, 77% of new foundry capacity, and 90% of rare-earth refining, concentrating the industrial base behind the AI buildout.
Related Articles:
Who’s Afraid of Chinese Models?— Demand for frontier artificial intelligence models exceeds supply because computing capacity is scarce, allowing chip suppliers and their customers to resell compute at high margins. The author argues alarm over Chinese models is overblown, expecting the inference market to grow faster than training costs so labs can thrive at lower prices.
Semiquincententacles— Leading artificial-intelligence labs are moving to usage-based token pricing on a cost-plus basis that includes total ownership costs, pushing token prices higher and giving some enterprises reason to migrate to cheaper open models. Frontier models remain necessary for cybersecurity, distillation training, agents, and scientific discovery, where maximum performance is required.
Why .400 Hitters Disappeared — and What It Means for AI— Artificial intelligence models' capability gains are shrinking in percentage terms, falling from about 15% relative to the baseline early on to roughly 10% for the same absolute increment, narrowing the gap between frontier and good-enough models. The author argues this points to a commoditizing market where price, not minor model differences, becomes the main differentiator, pressuring frontier firms' margins.
AI Summary.Foreign investors made roughly $1.5tn in net purchases of US equities from the start of 2024 through July, with chipmaking economies like South Korea and Taiwan helping fund American tech firms' AI buildout. Bloomberg Economics estimates about 20% of foreign US equity holdings were AI-related in 2025, versus roughly 10% in 2019.
Shawn Donnan, Estelle Ou and Jorgelina Do RosarioBloomberg
Date Posted:
The profits of South Korean and Taiwanese firms are flowing into AI-related US equities. A Bloomberg analysis found ~20% of all foreign holdings of US equities were in AI-related sectors, up from ~10% in 2019.
Are foreign investors funding America's AI buildout?
Core argument: Foreign investors held roughly $4.9 trillion of US-listed shares in AI-centered industries at the end of June, a sum roughly equivalent to Germany’s annual GDP, according to Bloomberg Economics’ analysis of Treasury data.
The AI boom is creating massive demand for high-end chips and new pools of savings in places that make them like South Korea and Taiwan which are flowing to fund American tech companies and their $9 trillion AI build out. Foreign investors made roughly $1.5 trillion in net purchases of US equities alone since the beginning of 2024 through the end of July. A Bloomberg Economics analysis of Treasury data shows that in 2025 roughly 20% of all foreign holdings of US equities were in AI-related sectors, up from roughly 10% in 2019. That number has grown with soaring valuations since then. But even assuming that 20% level, foreign investors had about $4.9 trillion of exposure to US-listed shares of companies in industries at the center of the AI boom at the end of June. That’s roughly equivalent to Germany’s annual GDP.
Takeaways by Macro Roundup® AI
Foreign investors held roughly $4.9 trillion of US-listed shares in AI-centered industries at the end of June, a sum roughly equivalent to Germany’s annual GDP, according to Bloomberg Economics’ analysis of Treasury data.
AI-related sectors accounted for roughly 20% of all foreign holdings of US equities in 2025, up from roughly 10% in 2019, per Bloomberg Economics’ analysis of Treasury data.
Foreign investors made roughly $1.5 trillion in net purchases of US equities from the start of 2024 through the end of July, with savings from chipmaking economies such as South Korea and Taiwan flowing to American tech companies and their $9 trillion AI build-out.
Related Articles:
Califoreboding— A basket of 42 AI stocks accounts for 60%-80% of all S&P 500 earnings growth, price returns and capital spending, while 60% of S&P 500 stocks sit 20% or more below their highs and only 30% outperform the index. AI-related construction spending has risen 38%, the only positive trend as manufacturing, commercial real estate and residential construction decline.
Financing the AI Buildout— The authors estimate roughly 57 gigawatts of operating US data center capacity against a planned development pipeline of about 509 gigawatts, which is not a completion forecast since projects may be delayed, scaled back, or abandoned. Their central scenario has 182.8 gigawatts of added capacity becoming operational through 2032, 117.2 gigawatts completed later, and 226.9 gigawatts never completed.
One Hundred Years in the U.S. Stock Markets— Over a century, U.S. publicly traded stocks generated roughly $91tn in shareholder wealth enhancement, with returns strongly positively skewed and dominated by a relatively small number of high-performing firms. Because most possible outcomes fall below the mean in a skewed distribution, the authors argue financial planning should focus on median outcomes or the full distribution.
AI Summary.An artificial intelligence investment boom that is insensitive to interest rates is drawing capital, electricity and labor away from rate-sensitive sectors such as housing and autos, which the author argues will force the Federal Reserve to hold rates higher for longer. The durable fix is expanding power, chip and infrastructure supply.
Torsten SløkApollo
Date Posted:
Sløk warns the rate-insensitive AI buildout has triggered unbalanced growth in the US economy by crowding out capital, electricity, and labor from vulnerable sectors like housing and autos.
Does a rate-insensitive AI boom force the Fed to keep rates higher?
Core argument: The AI boom’s insensitivity to interest rates is draining capital, power and labor from rate-sensitive sectors such as housing and autos, producing a “Dutch disease” dynamic in the US economy.
AI's rate-insensitive boom is draining capital, power and labor from vulnerable sectors like housing and autos. Trapped between inflation driven by the AI boom and weakness in rate-sensitive industries, the Fed will have to keep interest rates higher for longer. Because monetary policy cannot resolve this sectoral imbalance, the only durable fix is expanding the supply of power, chips and infrastructure.
Takeaways by Macro Roundup® AI
The AI boom’s insensitivity to interest rates is draining capital, power and labor from rate-sensitive sectors such as housing and autos, producing a “Dutch disease” dynamic in the US economy.
Caught between AI-driven inflation and weakness in rate-sensitive industries, the Federal Reserve will have to hold interest rates higher for longer.
Monetary policy cannot resolve the imbalance between AI and rate-sensitive sectors, leaving expansion of the supply of power, chips and infrastructure as the only durable fix.
Related Articles:
Financing the AI Buildout— The authors estimate roughly 57 gigawatts of operating US data center capacity against a planned development pipeline of about 509 gigawatts, which is not a completion forecast since projects may be delayed, scaled back, or abandoned. Their central scenario has 182.8 gigawatts of added capacity becoming operational through 2032, 117.2 gigawatts completed later, and 226.9 gigawatts never completed.
AI Set to be Largest CapEx Cycle Ever … and Soon Majority Externally Financed— Hyperscaler and data-center investment is tracking toward $1.5tn annually by 2030, which in real terms would exceed the 2005-06 housing boom, the 1990s fiber peak, and the 1882 railroad peak combined. External debt and equity are projected to fund close to 90% of the buildout, making it dependent on credit markets.
The Macro Implications of the AI Capex Boom— Artificial intelligence capital spending is estimated to add roughly 1.4 percentage points to US growth in one year and about 1.5 points the next, comparable to business investment's contribution during the tech bubble. A material part of that estimate reflects profits at Nvidia and a few other chipmakers that are unlikely to be recycled back into the economy.
AI Summary.A basket of 42 AI stocks accounts for 60%-80% of all S&P 500 earnings growth, price returns and capital spending, while 60% of S&P 500 stocks sit 20% or more below their highs and only 30% outperform the index. AI-related construction spending has risen 38%, the only positive trend as manufacturing, commercial real estate and residential construction decline.
Michael CembalestJ.P. Morgan
Date Posted:
Since January 2024, JPM’s 42 AI-related stocks drove 67% of S&P 500 returns, 58% of earnings growth, and 81% of capital spending. Cembalest notes, “the technology share of GDP growth has risen from almost zero in 2022 to ~33%,” and only 20% of states are outpacing national GDP growth, the lowest level since 1979.
How concentrated are US market and economic gains in AI?
Core argument: A basket of 42 AI stocks accounts for 60%–80% of all S&P earnings growth, price returns and capital spending since January 2024, with AI primarily benefitting AI supply chain companies rather than the companies adopting it.
A basket of 42 AI stocks accounts for 60%-80% of all S&P earnings growth, price returns and capital spending since Jan 2024. At the same time, 60% of S&P 500 stocks are down 20% or more from their all-time highs despite the market being close to all-time highs. Only 30% of stocks are outperforming the S&P 500, one of the lowest figures since 1991. AI is primarily benefitting AI supply chain companies rather than the companies adopting it. The technology share of GDP growth has risen from almost zero to 2022 to ~33%. Only 20% of US states are growing faster than the national average, almost the lowest figure since 1979. AI accounts for the only positive trend in US construction spending, having risen by 38% since January 2024. Unfortunately the other three main categories are down since then (manufacturing -28%, commercial real estate -13% and residential -6%). That’s why the decline in planned data center projects is something to watch.
Takeaways by Macro Roundup® AI
A basket of 42 AI stocks accounts for 60%–80% of all S&P earnings growth, price returns and capital spending since January 2024, with AI primarily benefitting AI supply chain companies rather than the companies adopting it.
60% of S&P 500 stocks are down 20% or more from their all-time highs even as the index sits near record levels, and only 30% of stocks are outperforming the S&P 500, one of the lowest figures since 1991.
AI accounts for the only positive trend in US construction spending, rising 38% since January 2024, while manufacturing is down 28%, commercial real estate 13% and residential 6%.
Related Articles:
Financing the AI Buildout— The authors estimate roughly 57 gigawatts of operating US data center capacity against a planned development pipeline of about 509 gigawatts, which is not a completion forecast since projects may be delayed, scaled back, or abandoned. Their central scenario has 182.8 gigawatts of added capacity becoming operational through 2032, 117.2 gigawatts completed later, and 226.9 gigawatts never completed.
How Much Revenue Does The AI Buildout Need In 2030?— The authors estimate AI capital spending of $560bn in 2026 and, in a high-investment case, $3.5tn of net AI capital in service by 2030, requiring roughly $1.3tn in annual infrastructure revenue once depreciation, operating costs and a 15% return are covered. That implies 7.4x growth from current annualized revenues, with downstream revenues of $1.9–2.6tn.
One Hundred Years in the U.S. Stock Markets— Over a century, U.S. publicly traded stocks generated roughly $91tn in shareholder wealth enhancement, with returns strongly positively skewed and dominated by a relatively small number of high-performing firms. Because most possible outcomes fall below the mean in a skewed distribution, the authors argue financial planning should focus on median outcomes or the full distribution.
AI Summary.Large US cloud and artificial-intelligence companies have raised $48bn in European-currency bonds, more than triple the prior year's total, spread across €27bn, £13bn and CHF7.5bn. The issuance adds highly rated US borrowers to European credit indices and deepens the thin market for longer-dated maturities, with few signs so far of crowding out other issuers.
Huw van SteenisApollo
Date Posted:
Hyperscalers have raised $48B in European-currency bonds this year, more than 3x all of 2025. They now account for 3% of euro IG issuance, 10% of sterling and 22% of Swiss franc, versus 8% in the US.
Are European bond markets funding the US AI buildout?
Core argument: Hyperscalers have raised $48 billion in bonds denominated in European currencies this year — €27 billion, £13 billion and CHF7.5 billion — more than triple the full-year 2025 total.
Hyperscalers have raised $48 billion in bonds in European currencies this year, already more than triple the entire amount raised last year. This breaks down as €27 billion, £13 billion, and CHF 7.5 billion. There are few signs of crowding out: non-hyperscaler IG issuance remains strong and little changed YoY, and corporate spreads remain tight. Rather, hyperscalers are adding highly rated US borrowers (AA- or higher) to European credit indices that have traditionally had more A-to-BBB exposure. Hyperscalers account for 7% of euro issuance with maturities of 10 years or more, versus 3% across all maturities, and 3% of the 10-year-plus euro IG index versus 1% of the overall index. This adds depth at longer-dated maturities—thin outside sovereign issuance—at a time when some governments, including the UK, have been shortening duration.
Takeaways by Macro Roundup® AI
Hyperscalers have raised $48 billion in bonds denominated in European currencies this year — €27 billion, £13 billion and CHF7.5 billion — more than triple the full-year 2025 total.
Hyperscaler issuance shows few signs so far of crowding out European borrowers, with non-hyperscaler investment-grade issuance little changed year-over-year and corporate spreads still tight.
Hyperscalers account for 7% of euro issuance with maturities of 10 years or more versus 3% across all maturities, adding borrowers rated AA- or higher to indices traditionally weighted toward A to BBB.
Related Articles:
Financing the AI Buildout— The authors estimate roughly 57 gigawatts of operating US data center capacity against a planned development pipeline of about 509 gigawatts, which is not a completion forecast since projects may be delayed, scaled back, or abandoned. Their central scenario has 182.8 gigawatts of added capacity becoming operational through 2032, 117.2 gigawatts completed later, and 226.9 gigawatts never completed.
AI Set to be Largest CapEx Cycle Ever … and Soon Majority Externally Financed— Hyperscaler and data-center investment is tracking toward $1.5tn annually by 2030, which in real terms would exceed the 2005-06 housing boom, the 1990s fiber peak, and the 1882 railroad peak combined. External debt and equity are projected to fund close to 90% of the buildout, making it dependent on credit markets.
AI Is Driving Up Treasury Yields: ‘It Just Touches Everything’— Investment-grade companies have sold nearly $1.5tn of bonds, up 36% from a year earlier, and some investors have sold US Treasuries to buy higher-yielding corporate debt. Bank of America economists estimate the surge in corporate-debt sales, along with more mortgage-backed securities issuance, pushed 10-year rates up by about 0.3 percentage point.
