Edward Conard

Top Ten New York Times Bestselling Author

  • “Unintended Consequences offers deep and well-argued analyses on almost every issue.” - The New York Times
  • “…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
  • “…challenges misconceptions that distort our economic debates.” - Arthur Brooks, President of the American Enterprise Institute
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…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
  • “…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
  • “…a comprehensive explanation of the modern economy.” - Julian Robertson, Founder, Tiger Management
  • “…a comprehensive explanation of the modern economy.” - Julian Robertson, Founder, Tiger Management
Upside of Inequality Oxford Unintended Consequences

Macro Roundup


Wednesday, July 22, 2026

AI Summary. The number of reported software vulnerabilities has been rising, with 1.5%–2.0% exploited each year, and the average time from a vulnerability's disclosure to its first confirmed exploitation fell below 100 days, with median times reaching zero. The same AI tools that find and exploit flaws can also propose code fixes.

Michael Cembalest J.P. Morgan
Date Posted:

The median time to exploitation of a vulnerability has fallen from ~ one year in 2021 to one day in 2026, and is projected to decline to one minute by 2027 as AI tools defuse. Cembalest notes, “The same tools that are used to detect and exploit vulnerabilities can also be used to propose code fixes and remediate them.”

Can AI fix software vulnerabilities as fast as it exploits them?

Core argument: Average time between a vulnerability's initial publication and its first confirmed exploitation fell below 100 days in 2025, with median times to exploitation already falling to zero days, and some attacks occur before patches become available.

Even before Mythos and GPT 5.5 came along in 2026, companies and governments had to react with increasing speed once a vulnerability is disclosed. The number of reported vulnerabilities has been rising, of which 1.5% -2.0% get exploited each year [and] the time between initial publication of a given vulnerability and its first confirmed exploitation (TTE) declined to less than 100 days on average in 2025, with median times to exploitation already falling to zero days. In fact, some attacks are occurring even before patches become available. The most important epiphany is the following: the same tools that are used to detect and exploit vulnerabilities can also be used to propose code fixes and remediate them. Claude Security, GPT 5.5-Cyber and OpenAI’s Codex Security Plugin are notable examples.

Takeaways by Macro Roundup® AI

  1. Average time between a vulnerability's initial publication and its first confirmed exploitation fell below 100 days in 2025, with median times to exploitation already falling to zero days, and some attacks occur before patches become available.
  2. Reported vulnerabilities have been rising, of which 1.5%–2.0% are exploited each year, requiring companies and governments to react with increasing speed once a vulnerability is disclosed.
  3. The same tools used to detect and exploit vulnerabilities can also propose code fixes and remediate them, with Claude Security, GPT 5.5-Cyber, and OpenAI's Codex Security Plugin as notable examples.

Related Articles:

  • Misanthropic: On Mythos, Bad Human Behaviors and Systems Vulnerabilities — Anthropic reports that its Mythos model has detected thousands of high-severity cyber vulnerabilities, including ones found by chaining multiple obscure software weaknesses, and that these hacking skills are emergent rather than a designed capability. Mythos completed a corporate network attack simulation estimated to take an expert over 10 hours, which no other frontier model had solved.
  • Emerging Details of Chinese Hack Leave U.S. Officials Increasingly Concerned — U.S. intelligence and national security officials have told major telecommunications companies to find a permanent way to expel Chinese hackers from the deepest parts of the nation's communications networks. Critical parts of the system date to the late 1970s or early 1980s and are too old to upgrade with modern cybersecurity protections.
  • 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.

AI Summary. The US 30-year Treasury yield has traded above 5% for its longest stretch since 2007, with roughly 19% of trading sessions above that level, reflecting investor concerns about a growing debt pile and sticky inflation.

Michael MacKenzie and Ye Xie Bloomberg
Date Posted:
Is Database:
Database
Is Important:
Important

So far this year, the US 30-year has traded beyond 5% for 27 days ~19% of trading days, the most since 2007.

Are rising long-term bond yields signaling worries about debt and inflation?

Core argument: The US 30-year Treasury yield has traded above 5% for 27 days this year, about 19% of all sessions, the most since 2007, according to data compiled by Bloomberg.

The US 30-year bond yield is trading above 5% for the longest stretch since the dawn of the financial crisis, echoing investor concerns about a growing debt pile and sticky inflation. So far this year, the 30-year has traded beyond 5% for 27 days — or about 19% of all sessions, the most since 2007, according to data compiled by Bloomberg. It traded above that level for 50 days that year.

Takeaways by Macro Roundup® AI

  1. The US 30-year Treasury yield has traded above 5% for 27 days this year, about 19% of all sessions, the most since 2007, according to data compiled by Bloomberg.
  2. The 30-year yield traded above 5% for 50 days in 2007, according to data compiled by Bloomberg.
  3. The 30-year yield's longest run above 5% since the dawn of the financial crisis echoes investor concerns about a growing US debt pile and sticky inflation.

Related Articles:

  • How Sky-High Deficits Threaten the Bond Market — The gap between 30-year and 10-year US Treasury yields has widened to 0.5 percentage points from 0.2, which suggests investors are more wary of future deficits, inflation, or random shocks. Dealers now buy only 10–15% of Treasurys at auction, down from 40–50% previously, according to JPMorgan Chase data.
  • How Might Fiscal Policy Respond to the Rise of Artificial Intelligence? — The Congressional Budget Office estimates that a permanent 0.5 percentage point increase in annual productivity growth would lower publicly held federal debt after 30 years by 39% of GDP, offsetting roughly half the projected rise from 101% of GDP to 175%. Higher revenue and slower spending growth outweigh higher interest costs.
  • The Budget and Economic Outlook: 2026 to 2036 — The Congressional Budget Office projects the federal deficit will grow from $1.9tn (5.8% of GDP) to $3.1tn (6.7%) over a decade, above the 3.8% averaged over the past 50 years, with rising net interest costs driving much of the increase. Federal debt held by the public is projected to rise from 101% to 120% of GDP.

AI Summary. US after-tax non-financial corporate profit margins remain elevated by historical standards, estimated at 7.6% versus a post-1949 high of 8.2%, while employee pay as a share of non-financial corporate gross value added has fallen from 66% to 56%. One argument holds that subdued demographic growth leads companies to prioritize maximizing margins over expanding capacity.

Tan Kai Xian Gavekal
Date Posted:
Is Database:
Database
Is Important:
Important

In Q1 2026, US after-tax non-financial margins were estimated at 7.6%, just short of the post-1949 high of 8.2% in Q2 of 2021. Tan Kai Xian argues US corporate profits may not mean revert due to increased automation as the workforce shrinks relative to the economy.

Does slower population growth push companies to maximize profit margins?

Core argument: US after-tax non-financial corporate profit margins were estimated at 7.6% in the first quarter of 2026, just below the post-1949 high of 8.2% set in the second quarter of 2021.

US corporate profit margins remain elevated by historical standards. In the first quarter of 2026, US after-tax non-financial margins were estimated at 7.6%, only just below the post-1949 high of 8.2% reached in the second quarter of 2021. The second-quarter earnings season has also begun strongly, according to FactSet data. Employee remuneration as a share of gross value added for US non-financial corporate businesses declined from 66% in the fourth quarter of 2001 to 56% in the first quarter of 2026. [Will Denyer argued] "When strong demographic growth powers rapid demand growth, companies concentrate on expanding capacity and sales—on growing along with the growing market. That means they tend to focus less on their margins—less on getting the biggest profit they can out of every dollar of revenues. In contrast, when demographic growth is subdued, and hence there is less potential demand growth, companies are not so bent on expanding capacity as fast as they can, and are far more interested in squeezing the maximum possible margin from every dollar of sales."

Takeaways by Macro Roundup® AI

  1. US after-tax non-financial corporate profit margins were estimated at 7.6% in the first quarter of 2026, just below the post-1949 high of 8.2% set in the second quarter of 2021.
  2. Employee remuneration as a share of gross value added for US non-financial corporate businesses fell from 66% in the fourth quarter of 2001 to 56% in the first quarter of 2026.
  3. Will Denyer argued that strong demographic growth leads companies to prioritize expanding capacity and sales, while subdued demographic growth pushes them to squeeze the maximum possible margin from every dollar of sales.

Related Articles:

  • The Record Divide Between Corporate Profits and Worker Pay — Labor's share of gross domestic income fell to 51%, the lowest since records began in 1947, while profits' share rose to 12.1%, the highest since 1950. Inflation-adjusted hourly wages are up 3% since the end of 2019, compared with a 50% rise in profits.
  • The Post‑COVID Decline in the Labor Share — The share of US national income paid to workers fell 1.6 percentage points below its pre-pandemic level, reaching a post-war low. The authors find this decline stems mainly from changes within industries rather than shifts in activity across sectors, following cyclical patterns seen in earlier recessions.
  • End of an Era: The Coming Long-Run Slowdown in Corporate Profit Growth and Stock Returns — The authors estimate that declining interest and corporate tax rates accounted for over 40% of real corporate profit growth from 1989 to 2019, while falling risk-free rates explain the entire expansion in price-to-earnings multiples. Real earnings before interest and taxes grew slightly more slowly than in 1962–1989, at 2.2% versus 2.4% per year.

AI Summary. China's manufacturing export surplus is around 2% of world GDP, roughly twice the largest surplus Japan ever ran, while domestic demand has stalled. Unexplained negative net income of about $125bn on its $4tn in net foreign assets, versus an expected $100bn surplus, supports the argument that the renminbi is about 30% undervalued.

Martin Wolf Financial Times
Date Posted:
Is Database:
Database

Martin Wolf estimates that under reasonable assumptions, China should be earning ~$100B annually on her $4T of foreign assets. Yet, China shows an unexplained negative net income of around $125bn on its net foreign asset position.

Is China's currency undervalued as its export surplus hits record levels?

Core argument: China's manufacturing export surplus stands at roughly 2% of world GDP, described as roughly twice the largest surplus Japan ever ran, while exports surge and domestic demand stalls.

China’s macroeconomics are very different from those obtained before the global financial crisis and, even more so, before the end of its property boom. Exports are surging, but domestic demand has stalled. China's export surplus in manufacturing is now around 2% of world GDP, which is “roughly twice the largest surplus Japan ever ran.” Most importantly, China shows an unexplained negative net income of around $125bn on its net foreign investments. But it should, on reasonable assumptions about the income it earns on its $4tn in net foreign assets, have a surplus of some $100bn. As a result, Brad Setser argues, the renminbi is by now some 30% undervalued. China’s policies look unavoidably threatening to much of the world. If a major trading partner runs huge and persistent trade and current account surpluses, everyone else has to run offsetting deficits. This leads to the shrinkage of sectors specialising in producing tradable goods and services, as well as huge domestic financial deficits.

Takeaways by Macro Roundup® AI

  1. China's manufacturing export surplus stands at roughly 2% of world GDP, described as roughly twice the largest surplus Japan ever ran, while exports surge and domestic demand stalls.
  2. China reports an unexplained negative net income of about $125bn on its net foreign investments, though reasonable assumptions about returns on its $4tn in net foreign assets imply a surplus near $100bn.
  3. Brad Setser argues the renminbi is some 30% undervalued; persistent Chinese trade and current account surpluses require offsetting deficits elsewhere, shrinking sectors producing tradable goods and services and producing large domestic financial deficits.

Related Articles:

  • The Dangerous Triumph Of Neo-Mercantilism — Neo-mercantilism is thriving in both China, which pairs infant-industry promotion with large trade surpluses, and in a United States focused on external deficits and protecting domestic markets. The author argues this will likely fracture the world economy, and that a new multilateral treaty could include a mechanism to combat large structural trade surpluses.
  • Sticking with What Works: Raising GDP Forecast on China’s Manufacturing Push — Stronger Chinese growth driven by export-oriented manufacturing may not produce net positive spillovers: for every 1pp export-driven boost to China's GDP, other economies may see a 0.1–0.3pp drag, with high-tech producers such as Europe and Japan facing greater pressure. Cheaper Chinese goods continue to exert a disinflationary impulse.
  • China’s Economy: Rightsizing 2025, Looking Ahead to 2026 — The authors argue China's official growth figures are inconsistent with ten consecutive quarters of deflation, estimating that real GDP is around 11% below official levels, with most of the adjustment in investment. That would imply a drop of roughly 9 percentage points in China's share of global investment, helping explain its export surge.

AI Summary. US voters most likely to identify as "very liberal" hold postgraduate degrees but household incomes of $30,000–$60,000 a year, an unusual combination given that education and income are typically positively correlated. In the Democratic Socialists of America's most recent member survey, 80% of members aged 25 or older had bachelor's degrees while 45% reported household incomes below $60,000.

Nate Silver Silver Bulletin
Date Posted:
Is Database:
Database
Is Important:
Important

Silver finds that the most liberal voters are highly educated and lower-income. Only 19% of Americans with a college degree had a household income of $60,000 or less, but responders to a DSA survey were ~ twice as likely to fall into that category.

Do highly educated, lower-income voters identify as very liberal?

Core argument: In the Democratic Socialists of America's 2021 member survey, 80% of members aged 25 or older held bachelor's degrees while 45% reported household incomes below $60,000 per year.

The US voters most likely to identify as “very liberal” are those with postgraduate degrees but lower-to-middle household incomes of $30K to $60K per year. This is very much also the sweet spot for the DSA. In the DSA’s most recent member survey in 2021, 80% of members aged 25 or older had bachelor’s degrees, but 45% had household incomes below $60,000 per year. This is unusual because education and income are usually substantially positively correlated. In the composite CES data, only 19% of Americans with bachelor’s degrees or higher had household incomes of $60K or below, while respondents to the most recent DSA survey [were twice as likely to fall into this category].

Takeaways by Macro Roundup® AI

  1. In the Democratic Socialists of America's 2021 member survey, 80% of members aged 25 or older held bachelor's degrees while 45% reported household incomes below $60,000 per year.
  2. Composite CES data show only 19% of Americans with bachelor's degrees or higher had household incomes of $60,000 or below, while DSA survey respondents were twice as likely to fall into that category.
  3. US voters most likely to identify as "very liberal" are those with postgraduate degrees and household incomes of $30,000 to $60,000 per year, an unusual pairing since education and income are usually substantially positively correlated.

Related Articles:

  • America’s Support for Capitalism Has Declined Over Last Decade — Just under half of Americans say capitalism is working very or somewhat well, down from 60% about a decade ago, according to a Wall Street Journal-NORC survey. Only 12% say democracy is working very or extremely well, and 16% say average citizens have considerable influence on politics.
  • Zero-Sum Thinking and the Roots of US Political Differences — The authors construct an index of "zero-sum thinking" — the belief that one group's economic gains come at another's expense across ethnic groups, citizens versus non-citizens, and countries in international trade. Among both moderate and strong Democrats, those in the top quartile of zero-sum thinking voted for Donald Trump at a disproportionately high rate.
  • Political Representation Gaps and Populism — A study comparing survey responses of citizens and parliamentarians across 27 European countries finds that policymaking reflects voters' economic attitudes well, but the average parliamentarian is about one standard deviation more culturally liberal than the national mean voter. The author finds this cultural gap spans nearly all issues, countries, and established parties, and that right-wing populists fill it.

Thursday, July 23, 2026

AI Summary. A long-run model in which health spending, the federal budget, and capital investment are jointly determined by supply and demand projects U.S. federal debt reaching 151% of GDP in 2036 and 309% in 2056, versus the Congressional Budget Office's 120% and 175%. Rising real interest rates compound deficits and debt.

Mark Warshawsky American Enterprise Institute
Date Posted:
Is Database:
Database

Warshawsky projects US debt-to-GDP will be 151% in 2036, well above CBO’s 120% projection, as higher real rates and a dynamic Baumol cost-disease effect will increase healthcare spending relative to GDP in an aging economy.

Does modeling health spending and interest rates jointly raise debt projections?

Core argument: The model projects U.S. federal debt-to-GDP at 151% in 2036, 218% in 2046, and 309% in 2056, above CBO's 120% in 2036 and 175% in 2056.

The Congressional Budget Office makes certain simplifying assumptions on health care spending, and is based on current law. By contrast, in the model presented in this paper, these variables are simultaneously determined by supply and demand, based on logical functional forms and parameter estimates from the literature or empirical analysis, and the model is based on current policy. This approach better reflects real economic relationships—between health care spending, the federal budget, and investment in capital—and changing underlying conditions, especially demographics. Within the next twenty or so years, the model predicts that federal government debt will grow significantly beyond historical experience, to be judged unsustainable because the real interest rate exceeds real economic growth. Debt-to-GDP will be 151% in 2036, 218% in 2046, and 309% in 2056, compared to CBO’s 120% in 2036 and 175% in 2056. Real interest rates rise for several decades, ratcheting interest payments, deficits, and debt in a vicious cycle.

Takeaways by Macro Roundup® AI

  1. The model projects U.S. federal debt-to-GDP at 151% in 2036, 218% in 2046, and 309% in 2056, above CBO's 120% in 2036 and 175% in 2056.
  2. The model judges the projected debt path unsustainable because the real interest rate exceeds real economic growth, with real rates rising for several decades and ratcheting up interest payments, deficits, and debt.
  3. Unlike CBO's current-law projections with simplifying assumptions on health care spending, the model determines these variables simultaneously through supply and demand under current policy, using functional forms and parameter estimates from the literature or empirical analysis.

Related Articles:

  • Can AI Avert the Impending Federal Budget Crisis? — U.S. GDP per capita, about $87,000, is projected to reach $102,000 by 2035 with 1.8% annual productivity growth, or $110,000 with 2.7% growth. Because the healthcare sector has been slow to adopt AI, health spending is projected to rise from 18.3% of GDP to 21% under slower growth and 21.7% under faster growth.
  • How Might Fiscal Policy Respond to the Rise of Artificial Intelligence? — The Congressional Budget Office estimates that a permanent 0.5 percentage point increase in annual productivity growth would lower publicly held federal debt after 30 years by 39% of GDP, offsetting roughly half the projected rise from 101% of GDP to 175%. Higher revenue and slower spending growth outweigh higher interest costs.
  • The Budget and Economic Outlook: 2026 to 2036 — The Congressional Budget Office projects the federal deficit will grow from $1.9tn (5.8% of GDP) to $3.1tn (6.7%) over a decade, above the 3.8% averaged over the past 50 years, with rising net interest costs driving much of the increase. Federal debt held by the public is projected to rise from 101% to 120% of GDP.

AI Summary. The author estimates "shadow" 30-year government bond yields for Germany, Spain, Italy, Greece and Japan — what long-term yields would be if central banks withdrew support such as Japanese government bond purchases and the European Central Bank's backstop facilities. These estimates assume no outright central bank selling, so they represent a lower bound on how high yields could go.

Robin Brooks Robin Brooks Substack
Date Posted:

Robin Brooks estimates that, given its high government debt to GDP ratio, Japan’s 30-year yield would, in the absence of BoJ intervention, be at least 300bps higher – a 7% yield as opposed to the observed 4%.

How much higher would long-term bond yields be without central bank support?

Core argument: The author estimates "shadow" 30-year government bond yields for Germany, Spain, Italy, Greece and Japan, comparing actual market yields as of July 21st with where yields would sit absent central bank yield caps.

The scatter plot shows my estimates for what 30-year shadow yields are for Germany (DE), Spain (ES), Italy (IT), Greece (GR) and Japan (JP). The solid red dots are actual yields in markets as of [July 21st], while the hollow dots are where I think yields would be if central banks pulled back from yield caps. In the case of Japan, this would be an end to gross purchases of JGBs. For the ECB, it means cancelling TPI and going back to the OMT as the only bailout facility. It would obviously also preclude the kind of comments President Lagarde is fond of making on French yields. I don’t anticipate actual selling of government bonds from anyone, so my shadow yields are a lower bound for how high yields might go if central banks really exited bond markets.

Takeaways by Macro Roundup® AI

  1. The author estimates "shadow" 30-year government bond yields for Germany, Spain, Italy, Greece and Japan, comparing actual market yields as of July 21st with where yields would sit absent central bank yield caps.
  2. For Japan, the author's shadow yield scenario assumes an end to gross purchases of Japanese government bonds, while for the ECB it assumes cancelling the Transmission Protection Instrument and reverting to Outright Monetary Transactions as the only bailout facility.
  3. The author treats the estimated shadow yields as a lower bound for how high yields might go if central banks exited bond markets, anticipating no actual selling of government bonds by any central bank.

Related Articles:

  • Global Debt Report 2026 — Gross central government borrowing across OECD countries rose 7% to a record $17tn, about 23% of GDP, and is projected to reach roughly $18tn, or 24% of GDP. The United States and Japan account for close to 80% of the area's refinancing requirements, with the US share rising from 57% to 70%.
  • Are Government Bonds Safe in Times of War and Pandemic? — Government bonds severely underperform in wartime, in absolute terms and relative to equities, housing, and economic growth, unlike in financial crises when bonds tend to outperform stocks. The U.S. and U.K. reduced war-driven debt-to-GDP ratios without explicit default, using surprise inflation and financial repression that shifted fiscal burdens onto bondholders.
  • Across The Rich World, Fiscal Crises Loom — An analysis estimates most rich countries could keep debt stable as a share of their economies with only small primary surpluses—Britain's required surplus is 0.3% of GDP—even if all debt were refinanced immediately at current rates, but the tightening needed to stabilize debt ratios exceeds 2% of GDP in Britain and America and 3% in France.

AI Summary. Japan's return to headline inflation of 1.5%, modest by global and historical standards, feels to many households like a severe cost-of-living crisis because real wage gains had until recently failed to keep pace with more expensive imports. Japanese households spent 30.75% of income on food, a 25-year peak and among the highest in the developed world.

Leo Lewis Financial Times
Date Posted:

Reflation in Japan has left real wages below their 2020 level, though headline inflation is only running at a modest 1.5%.

Why does modest inflation feel severe to Japanese households?

Core argument: Japan's Engel coefficient, the share of household income spent on food, hit a 25-year peak of 30.75% in December, one of the highest levels in the developed world.

[Japan’s] return to what is, in global and historical terms, modest headline inflation of 1.5% feels, for many households, like a severe cost of living crisis, as real wage increases have failed, until recently, to match the costs of more expensive imports. Japan’s Engel coefficient, a “pain index” that measures the proportion of household income spent on food, stood at a 25-year peak of 30.75% in December — one of the highest in the developed world.

Takeaways by Macro Roundup® AI

  1. Japan's Engel coefficient, the share of household income spent on food, hit a 25-year peak of 30.75% in December, one of the highest levels in the developed world.
  2. Japan's headline inflation of 1.5%, modest in global and historical terms, registers for many Japanese households as a severe cost of living crisis.
  3. Real wage increases in Japan failed, until recently, to keep pace with the cost of more expensive imports.

Related Articles:

  • 2026 Japan Economic Outlook: Steady Fundamentals, Policy Risks Ahead — Japan's government debt-to-GDP ratio has continued to decline despite an ongoing primary deficit, because the government's interest payment rate (0.8%) is far below nominal GDP growth (3.9%), a gap the authors call the widest since 1980 and worth roughly -7 percentage points a year.
  • Yen Hits Four-Decade Low in Historic Slide That’s Rattled Japan — The yen weakened past 161.95 per dollar to its lowest level since 1986, putting traders on alert for Japanese intervention to support the currency. The weak yen is lifting exporters' profits and Japanese stocks to record highs while raising import costs for oil and gas and fueling inflation that hurts consumers.
  • Japan’s Debt Puzzle: Sovereign Wealth Fund from Borrowed Money — Japan's public sector operates as a highly leveraged sovereign wealth fund, funding risky assets by borrowing from banks and depositors at below-market rates, with returns sustaining its debt-to-GDP ratio above 200% without primary surpluses. The authors argue the resulting duration and currency mismatches expose Japan to mark-to-market losses that could trigger a fiscal crisis or a run on the yen.

AI Summary. A warning sign in a boom cycle is when companies closest to final demand weaken while capital spending beneficiaries keep thriving, as when telecom operators' shares rolled over while communications equipment stocks kept rising. Semiconductor stock prices and free cash flow are soaring while large cloud providers' shares stagnate and their free cash flow plummets.

Michael Cembalest J.P. Morgan
Date Posted:

Cembalest notes the rising stock prices and expanding free cash flow of firms supporting the AI buildout, even as the hyperscalers, who are closest to final demand, see their stock prices stagnating and their free cash flow plummeting.

Does weakening demand-side performance amid thriving capital spending signal a boom's end?

Core argument: In late 1999 and early 2000, communications services companies including Verizon, AT&T, WorldCom, Bell South, GTE and Sprint rolled over even as communications equipment stocks such as Cisco, Nokia, Nortel, Qualcomm and Motorola kept rising.

The worrisome part of a boom cycle is when companies closest to final demand roll over, even as capital spending beneficiaries of the cycle keep on thriving. In late 1999 and early 2000, communications services companies rolled over (Verizon, AT&T, WorldCom, Bell South, GTE, Sprint) even as communications equipment company stocks (Cisco, Nokia, Nortel, Qualcomm, Motorola) kept on rising. The obvious parallel to today is soaring stock prices and free cash flow for semiconductor companies (the caboose) even as the front end of the train (hyperscalers) is seeing stock prices stagnate and free cash flow plummet. Tech sector earnings typically track reasonably well with sectors that rely heavily on technology (finance, manufacturing, media, transport, and healthcare) and pay for all that equipment and software. The recent divergence begs the question: who is going to pay for the AI boom if the tech sector’s primary customers are not making more money as well?

Takeaways by Macro Roundup® AI

  1. In late 1999 and early 2000, communications services companies including Verizon, AT&T, WorldCom, Bell South, GTE and Sprint rolled over even as communications equipment stocks such as Cisco, Nokia, Nortel, Qualcomm and Motorola kept rising.
  2. Semiconductor companies show soaring stock prices and free cash flow while hyperscalers, at the front end of the chain, see stock prices stagnate and free cash flow plummet.
  3. Tech sector earnings typically track reasonably well with sectors that rely heavily on technology — finance, manufacturing, media, transport and healthcare — raising the question of who pays for the AI boom if those customers are not making more money.

Related Articles:

  • AI: The ROI Runway Could Be Long Outside the Tech Sector — AI company valuations rest on the assumption that profit margins across non-tech firms will eventually rise, but outside a handful of sectors adoption is slow, creating a divergence between front-loaded valuations and slower cash flows. If the per-use costs of AI models converge toward zero, revenue may be insufficient for all major cloud providers.
  • 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.
  • Cover Ratios for Hyperscaler Bonds Declining — Investor demand for hyperscaler bond deals has weakened, with the cover ratio — dollars of investor orders per dollar of bonds issued — falling from nearly 5x to below 2x. This suggests investors may require wider spreads to absorb additional hyperscaler bond supply.

AI Summary. Alphabet raised its projected annual capital expenditures to $195bn–$205bn from a prior forecast of $190bn, and its shares fell amid concern over artificial intelligence costs. The Google parent reported negative free cash flow of $5.9bn in the quarter, its first negative quarterly cash flow since going public more than two decades ago.

Davey Alba and Julia Love Bloomberg
Date Posted:

Alphabet’s free cash flow was negative (-$5.9B) in Q2 for the first time since going public over 20 years ago, as the firm raised its forecast of 2026 capex to between $195–205B, up from $190B.

Are rising AI spending plans pressuring big tech cash flow?

Core argument: Alphabet raised its projected capital expenditures for the year to $195 billion–$205 billion, up from a prior forecast of $190 billion, and its shares fell amid concern over AI spending costs.

The Google parent company boosted its projected capital expenditures for the year to a range of $195 billion to $205 billion, up from a previous forecast of $190 billion. The company’s quarterly cash flow was negative for the first time since going public more than two decades ago, according to data compiled by Bloomberg. Negative free cash flow came in at $5.9 billion in the period.

Takeaways by Macro Roundup® AI

  1. Alphabet raised its projected capital expenditures for the year to $195 billion–$205 billion, up from a prior forecast of $190 billion, and its shares fell amid concern over AI spending costs.
  2. Alphabet posted negative free cash flow of $5.9 billion in the quarter, its first negative quarterly cash flow since going public more than two decades ago, according to data compiled by Bloomberg.

Related Articles:

  • More Money, More Problems — Public filings indicate AI compute backlogs—contracts for future supply generating no current revenue—exceed $2.2tn, but under reasonable assumptions the net present value of those rental payments after required capital spending equals only about 15% of the backlog. Slowing AI revenue or lost capital-markets access could raise risk premiums on these obligations.
  • Hyperscaler CapEx Sits on Thin Implied Equity — The present value of future rental payments from the roughly $2.2tn artificial intelligence compute backlog, after deducting the capital spending needed to deliver that compute, equals only about 15% of the backlog. That thin implied-equity layer means the contracts hold value only if customers stay creditworthy, capital remains available, and the infrastructure earns competitive returns for years.
  • 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.