Edward Conard

Top Ten New York Times Bestselling Author

  • “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
  • “…a comprehensive explanation of the modern economy.” - Julian Robertson, Founder, Tiger Management
  • “…challenges misconceptions that distort our economic debates.” - Arthur Brooks, President of the American Enterprise Institute
  • “Unintended Consequences offers deep and well-argued analyses on almost every issue.” - The New York Times
  • “…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
  • “…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
  • “A full-throated defense of economic dynamism.” - The Wall Street Journal
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
  • “…a 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
Upside of Inequality Oxford Unintended Consequences

Macro Roundup


Thursday, July 31, 2025

AI Summary. The author estimates that AI-related capital spending contributed between 0.6% and 1.3% of a quarter's 3% U.S. GDP growth, based on information technology equipment and intellectual property investment data. Google, Amazon, Meta, and Microsoft alone spent about $69bn in the quarter, roughly half of all information technology equipment spending on an annualized basis.

Paul Kedrosky Applied Complexity
Date Posted:

Paul Kedrosky estimates that AI capex drove 20 to 40% of US GDP growth of 3% per annum in Q2, noting “for practical purposes (AI capex) ate Q2 GDP growth.”

How much of U.S. GDP growth comes from AI capital spending?

Core argument: AI capex contributed an estimated 0.6 to 1.3 points of the 3% U.S. GDP growth in the quarter, with the high end implying roughly 40% of that growth, based on BEA data on IT equipment and intellectual property products.

The key data point is IT spending (IT equipment and IP products) [while the] two categories were not entirely AI-capex-related, obviously. But a significant fraction was, certainly more than half, given their anomalous size and growth in the BEA Q2 numbers. Summing it, a reasonable estimate is that AI capex contributed perhaps 1.3% of the 3% GDP growth in the quarter, around 40%. We can back into the estimate another way, bottoms-up from known AI capex at major tech companies spending in the period. Taking just Google, Amazon, Meta, and Microsoft, and their quarterly earnings and published data, they spent around $69B in the quarter, which is $276B annualized. Total IT equipment spending in the quarter was $608B annualized, so the Big Four alone were almost half of the spending, and most of that, we know, was AI capex. Given that information processing equipment spending added 1% to GDP growth in the quarter, from the BEA's own figures, then AI capex, including both software and equipment, was at least 0.6% in that. We now have a range: AI capex's contribution to Q2 growth was somewhere between 0.6% (on the low end, undercounting smaller players) and 1.3% (on the high end).

Takeaways by Macro Roundup® AI

  1. AI capex contributed an estimated 0.6 to 1.3 points of the 3% U.S. GDP growth in the quarter, with the high end implying roughly 40% of that growth, based on BEA data on IT equipment and intellectual property products.
  2. Google, Amazon, Meta, and Microsoft together spent around $69 billion on capex in the quarter, or $276 billion annualized, nearly half of the $608 billion annualized total IT equipment spending, most of it AI capex.
  3. Information processing equipment spending added 1 point to quarterly GDP growth in the BEA's figures, from which the author derives a floor of at least 0.6 points for AI capex covering both software and equipment.

Related Articles:

  • 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 Adoption Is Accelerating In 2025 — Nvidia's data center revenue growth is used as a simple extrapolation to project the broader capital-expenditure impact of AI-specific investment on GDP. The aim is not precision but a timely sense of whether the growth contribution from AI capital spending is accelerating or potentially peaking.

AI Summary. Population aging and fertility below replacement levels are connected to a revolution in family structure in which families grow smaller, fewer people marry, and voluntary childlessness becomes common across countries. As a result, families become less able to bear the demands that may be placed on them.

Nicholas Eberstadt Fertility and Sterility
Date Posted:

Nicholas Eberstadt warns that depopulation will stress families as smaller families “become ever less able to bear weight—even as the demands that might be placed on them steadily rise.”

Does falling fertility weaken families' ability to support their members?

Core argument: Rapid aging and steep sub-replacement fertility are inextricably connected to a revolution in family structure in which families grow smaller and more atomized, fewer people marry, and voluntary childlessness takes hold country after country.

Families are society’s most basic unit and are still humanity’s most indispensable institution. Both precipitous aging and steep sub-replacement fertility are inextricably connected to the ongoing revolution in family structure. As familial units grow smaller and more atomized, fewer people get married, and high levels of voluntary childlessness take hold in country after country. As a result, families and their branches become ever less able to bear weight—even as the demands that might be placed on them steadily rise. Just how depopulating societies will cope with this broad retreat of the family is by no means obvious. Perhaps others could step in to assume roles traditionally undertaken by blood relatives. But it is hard to expect duty and sacrifice in childrearing from people who are not kin.

Takeaways by Macro Roundup® AI

  1. Rapid aging and steep sub-replacement fertility are inextricably connected to a revolution in family structure in which families grow smaller and more atomized, fewer people marry, and voluntary childlessness takes hold country after country.
  2. As families and their branches become less able to bear weight while the demands placed on them rise, how depopulating societies will cope with this retreat of the family is not obvious.
  3. Non-relatives might step into roles traditionally undertaken by blood relatives, but duty and sacrifice in childrearing is hard to expect from people who are not kin.

Related Articles:

  • The Demographic Future of Humanity: Facts and Consequences — The authors estimate the world total fertility rate is around 2.17, below the global replacement rate of roughly 2.21, meaning humanity is already below replacement, with fertility and births falling faster than anticipated. They project middle-income countries' fertility may fall below that of rich countries outside East Asia, with emigration compounding population decline.
  • World Depopulation: Prospects and Implications — The US is projected to account for a growing share of the rich world's labor force, youth, and highly educated talent, with continuing population and labor force growth and only moderate aging through 2050. Its fertility, though below replacement, exceeds that of East Asian and nearly all European countries, and sustained legal immigration is essential to this demographic advantage.
  • The Power of Demography and the Fate of East Asia — East Asia's total population has peaked and is projected to decline for at least a generation, with the pace of decline likely accelerating each decade, while US population is projected to keep growing for decades. The author argues such sustained divergence can shift the global balance of power.

AI Summary. The trade-weighted dollar against other major currencies has risen roughly 2.5% in a week, an unusually large move for such a liquid currency. The author argues the rally reflects an unwinding of extreme short positioning and scope for priced-in Federal Reserve rate cuts to be removed, and that the dollar's decline is over.

Robin Brooks Robin Brooks Substack
Date Posted:

Citing excessive market expectations of upcoming rate cuts, a backlog of short positions accumulated at the height of the trade war, and momentum from this week’s 2.5% appreciation against the rest of the G10, Brooks argues “the dollar fall is over.”

Is the dollar's decline over?

Core argument: The trade-weighted Dollar against the G10 is up roughly 2.5% this week, which the author calls a highly unusual move for such a liquid and deeply-traded currency.

The chart shows the trade-weighted Dollar against the G10. The Dollar is up around 2.5% this week, which is a highly unusual move for such a liquid and deeply-traded currency. Market pricing for the Fed is still very dovish, which means there’s still plenty of scope for rate cuts to get priced out, especially in the remainder of 2025. As trade tensions fade, markets will likely refocus on US growth outperformance, which means the US “exceptionalism” narrative may come back stronger than ever. The extreme level of Dollar shorts took many months to build, so there’s still a lot of unwinding of this positioning to be done. This short squeeze is far from over. Most importantly, the single biggest disrupter of market narratives is price action. The overriding market narrative in recent months has been that the Dollar will keep falling. This kind of appreciation episode - its speed and violence - ends this kind of story telling. The Dollar fall is over.

Takeaways by Macro Roundup® AI

  1. The trade-weighted Dollar against the G10 is up roughly 2.5% this week, which the author calls a highly unusual move for such a liquid and deeply-traded currency.
  2. The author argues market pricing for the Fed remains very dovish, leaving plenty of scope for rate cuts to be priced out, especially over the remainder of 2025.
  3. The author contends that the extreme level of Dollar shorts took many months to build, leaving substantial unwinding ahead, and that the US "exceptionalism" narrative may return stronger than ever as trade tensions fade.

Related Articles:

  • The International Role of the U.S. Dollar – 2025 Edition — The US dollar accounted for 58% of disclosed global official foreign reserves, versus 20% for the euro and 2% for China's renminbi, and roughly 50% of international payments made through the SWIFT messaging system. About 55% of international claims and 60% of liabilities are denominated in dollars, a share relatively stable since 2000.
  • Explaining the US Dollar’s Depreciation in the First Half of 2025 — The dollar depreciated by more than interest-rate differentials alone would predict, and regression analysis attributes the gap to the trade war and economic policy uncertainty, including foreign investors' concerns about proposed taxes on foreign holders of US assets and a possible currency accord. The author expects the dollar to appreciate as these uncertainties resolve.
  • Tariffs, the Dollar, and Equities: High-Frequency Evidence from the Liberation Day Announcement — Following a sweeping US tariff announcement, the dollar's effective exchange rate depreciated rather than appreciating as standard international macroeconomic models predict: all G10 currencies gained against the dollar while the most freely floating emerging market currencies gained less or fell. The authors read this as advanced economies rebalancing away from dollar assets.

AI Summary. Inflation-adjusted US house prices are 2.0% below their 2022 peak. An affordability index built from median income, a 15% down payment, a 2% allowance for property taxes, insurance and maintenance, and 30-year fixed mortgage rates shows the monthly payment on the same house rose 101% over four years, with prices up 29% and mortgage rates at 6.82%.

Bill McBride Calculated Risk
Date Posted:

Nationally, real house prices are now 2% below the 2022 peak and 10.5% above the 2006 peak. “Affordability” remains at levels last seen in the early 1980s when 30-year mortgage rates were over 14%. @calculatedrisk

How much have house prices and monthly payments changed since the peak?

Core argument: Inflation-adjusted U.S. house prices stand 2.0% below their 2022 peak, based on the Case-Shiller National Index, seasonally adjusted.

I’ve put together my own affordability index [using] median income from the Census Bureau (estimated 2024 and 2025), assumed a 15% down payment, and used a 2% estimate for property taxes, insurance and maintenance. This is probably low for high property tax states like New Jersey and Texas, and too high for lower property tax states. For house prices, I used the Case-Shiller National Index, Seasonally Adjusted (SA). Also, for the down payment - there wasn’t a significant difference between 15% and 20%. For mortgage rates, I used the Freddie Mac PMMS (30-year fixed rates). Here is what the index looks like (lower is more affordable). The monthly payment for the same house in May 2021, with house prices up 29% over four years and mortgage rates at 6.82% in May 2025, would be $3,374 - an increase of 101%! Twice the payment for the same house!

Takeaways by Macro Roundup® AI

  1. Inflation-adjusted U.S. house prices stand 2.0% below their 2022 peak, based on the Case-Shiller National Index, seasonally adjusted.
  2. The monthly payment for the same house bought in May 2021 would be $3,374, an increase of 101%, with house prices up 29% over four years and 30-year mortgage rates at 6.82% in May 2025.
  3. The affordability index uses Census Bureau median income with 2024 and 2025 estimated, a 15% down payment, a 2% estimate for property taxes, insurance and maintenance, and Freddie Mac PMMS 30-year fixed rates, with no significant difference between 15% and 20% down.

Related Articles:

  • First-Time Home Buyers Are MIA. Landlords Are the Winners — The National Association of Realtors reports 1.1m first-time home buyers last year, 380,000 fewer than the prior year and about half the two-decade average of 2.1m a year, with the sharpest sales slowdown in homes under $500,000. Unaffordable prices and high mortgage rates have slowed the flow from renting to owning.
  • Rapidly Growing Treasury Supply Crowding Out Other Types of Credit Growth — Roughly half of all fixed income issuance coming to market over the past year has been US Treasuries, meaning investors must allocate a growing share of dollars to financing the government. The argument is that lower government debt would free up more funds for consumer borrowing and business investment.
  • Why Are Young Adults in the English-Speaking World So Unhappy? — The share of young adults regularly experiencing stress and anger has risen sharply in the US, Canada, UK, Ireland, Australia and New Zealand while remaining largely stable elsewhere in the West. Real house prices since 1995 are up 85% in the US and over 200% in the other five countries.

AI Summary. Most semiconductors entering the US are not captured in trade data because they arrive embedded inside imported devices. With global chip sales of $627bn, the US is estimated to consume 25% ($156bn) while producing about $68bn and directly importing $40bn, implying roughly $50bn of net imports hidden inside other goods.

Chris Miller American Enterprise Institute
Date Posted:

“Much of the $40 billion of chips the US imports are actually made in the US, packaged abroad, and reimported,” Chris Miller argues. Thus “tariffs would senselessly penalize domestic manufacturers.”

Do trade data capture most semiconductors entering the US?

Core argument: Around $50 billion of US net semiconductor imports are "hidden" in trade data, entering embedded inside imported devices rather than recorded as chip imports.

Most of the chips that enter the US are not captured by trade data because these semiconductors are embedded inside imported devices. Global semiconductor sales in 2024 were $627 billion, of which the US is estimated to consume 25%, or $156 billion. The US produces 10% of the world’s semiconductors. On the assumption that US-made chips are roughly equivalent in value to the global average, the US thus produces $68 billion of chips. It imports only $40 billion. In other words, around $50 billion of net imports of semiconductors are “hidden” in the trade data, imported as components of other goods.

Takeaways by Macro Roundup® AI

  1. Around $50 billion of US net semiconductor imports are "hidden" in trade data, entering embedded inside imported devices rather than recorded as chip imports.
  2. Global semiconductor sales reached $627 billion in 2024, of which the US is estimated to consume 25%, or $156 billion, while producing 10% of the world's chips.
  3. Assuming US-made chips match the global average in value, US semiconductor production totals $68 billion against recorded chip imports of only $40 billion.

Related Articles:

  • TSMC To Speed Up Construction Of US Chip Plants By ‘Several Quarters’ — TSMC says it is accelerating construction of its second and third Arizona plants by several quarters to meet strong U.S. customer demand for smartphone and AI computing chips. The company has committed an additional $100bn to build five more facilities in the state and expects around 30% of its 2-nanometer and more advanced capacity to be located in Arizona.
  • TSMC’s Arizona Chip Production Yields Surpass Taiwan’s in Win for US Push — Taiwan Semiconductor Manufacturing Co. reports that the share of usable chips produced at its Phoenix, Arizona plant is about 4 percentage points higher than at comparable factories in Taiwan. Yield is a critical industry measure because it determines whether companies can cover the enormous costs of a chip plant.
  • The Chips Act Has Been Surprisingly Successful So Far — The US government has spent over half of its $39bn in Chips Act semiconductor incentives, and chip companies and supply chain partners have announced $327bn of investments over the next 10 years, according to Semiconductor Industry Association calculations. US data show a 15-fold increase in construction of computing and electronics manufacturing facilities.

Friday, August 1, 2025

AI Summary. The average effective US tariff rate facing consumers is 18.3%, the highest since 1934, and 17.3% after consumption shifts to lower-tariffed goods, the highest since 1935. The authors estimate these tariffs raise $2.7tn over 2026-35, or $2.2tn after $466bn in negative dynamic revenue effects.

Ernie Tedeschi Yale University
Date Posted:

As of August 1, the US has an overall effective tariff rate of 17.3%. Dynamically scored, Ernie Tedeschi estimates this regime will raise $2.2T over 2026–2035.

How high are tariff rates and how much revenue do they raise?

Core argument: U.S. consumers face an overall average effective tariff rate of 18.3%, the highest since 1934.

Consumers face an overall average effective tariff rate of 18.3%, the highest since 1934. After consumption shifts, the average tariff rate will be 17.3%, the highest since 1935. All tariffs to date in 2025 raise $2.7 trillion over 2026-35, with $466 billion in negative dynamic revenue effects, bringing dynamic revenues to $2.2 trillion.

Takeaways by Macro Roundup® AI

  1. U.S. consumers face an overall average effective tariff rate of 18.3%, the highest since 1934.
  2. After consumption shifts, the average effective tariff rate falls to 17.3%, the highest since 1935.
  3. All U.S. tariffs imposed in 2025 raise $2.7 trillion over 2026–35, with $466 billion in negative dynamic revenue effects bringing dynamic revenues to $2.2 trillion.

Related Articles:

  • 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.
  • 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.
  • China Emerges From Trade Chaos With Record Exports, Surplus — China posted a record half-year trade surplus of about $586bn, with exports up 5.8% year-over-year to $325bn and imports up 1.1%. Shipments to the US fell 16.1% from a year earlier, while exports to the 10 Southeast Asian nations in the Asean group rose 17%.

AI Summary. Tariff-driven inflation is now showing up in furnishings and recreational goods—two heavily China-imported categories making up 25% of "core" goods in US inflation baskets—with monthly inflation running 100–120 basis points above seasonal norms. The author argues this impulse is large enough that the Federal Reserve may not cut rates at all.

Robin Brooks Robin Brooks Substack
Date Posted:

In furnishings and recreation goods, two sectors that comprise 25% of core goods and are largely reliant on imports from China, “monthly inflation in June 2025 is running between 100 – 120 bps above where it would be normally at this time of year.”

How large is the tariff inflation impulse in imported goods categories?

Core argument: Monthly inflation in June 2025 for furnishings and recreational goods ran 100–120 bps above typical levels for that time of year in both the CPI and PCE measures.

The tariff inflation impulse took longer than expected to show up due to tariff front-running and transshipments, but it’s now hitting with a vengeance. I focus on two categories of goods - furnishings and recreational goods - which make up 25% of “core” goods in the CPI and PCE inflation baskets. Furnishings is a broad category that includes furniture, household appliances as well as tools and hardware. Recreational goods include everything from video and audio products to sporting goods and pet supplies. Both categories of goods are heavily imported from China. The top row of charts shows the path of monthly inflation for furnishings in the CPI (lhs) and PCE (rhs), while the bottom row shows the monthly path of inflation for recreational goods. In all cases, monthly inflation in June 2025 (purple line) is running between 100 - 120 bps above where it would be normally at this time of year (black line). As of today, markets price 38 bps in cuts from the Fed for the rest of 2025. Given how big the inflation impulse from tariffs is, [it is] conceivable the Fed doesn’t cut at all this year, which means the Dollar may rise a lot further.

Takeaways by Macro Roundup® AI

  1. Monthly inflation in June 2025 for furnishings and recreational goods ran 100–120 bps above typical levels for that time of year in both the CPI and PCE measures.
  2. Furnishings and recreational goods, both heavily imported from China, make up 25% of "core" goods in the CPI and PCE inflation baskets, with furnishings spanning furniture, household appliances, tools and hardware.
  3. Markets price 38 bps of Fed cuts for the rest of 2025, but the author argues the size of the tariff inflation impulse makes no cuts at all conceivable, which could push the Dollar a lot higher.

Related Articles:

  • 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.
  • Help for the Heartland? The Employment and Electoral Effects of the Trump Tariffs in the United States — The authors find that the combined effect of U.S.-China trade war import tariffs, retaliatory tariffs, and farm subsidies on employment in exposed U.S. locations was at best a wash and possibly mildly negative, while residents of tariff-protected areas became less likely to identify as Democrats and more likely to vote Republican.
  • Tariff Endgame — New tariffs would add roughly 1% of GDP to a federal tax burden that was 17% of GDP, an amount that aggregate corporate profits near $4tn and disposable personal income around $22tn could absorb at the cost of modestly slower activity. The burden would fall unevenly, concentrated in import-dependent sectors such as autos and clothing.

AI Summary. Alphabet, Microsoft, Amazon and Meta are set to spend nearly $400bn on capital expenditures in a year, largely on artificial-intelligence infrastructure, exceeding European Union defense spending. Morgan Stanley projects $2.9tn in chip, server and data-center spending over four years, contributing up to 0.5% of US GDP growth annually, with a $1.5tn financing gap.

Rolfe Winkler, Nate Rattner and Sebastian Herrera Wall Street Journal
Date Posted:

Meta and Microsoft 2025 capex is approaching 40% of revenue. MS estimates there is a $1.5T gap btw the hyperscalers’ forecasted datacenter spending of $2.9T over the next three years and their ability to self-finance the spend.

Are investors backing Big Tech's massive AI infrastructure spending?

Core argument: Alphabet, Microsoft, Amazon and Meta are set to spend nearly $400 billion on capital expenditures this year, largely on AI infrastructure, exceeding what the European Union spent on defense last year.

Alphabet, Microsoft, Amazon and Meta are set to spend nearly $400 billion this year on capital expenditures, largely to build their artificial-intelligence infrastructure. That is more than the European Union spent on defense last year. Those firms and others plan to boost outlays even more. Morgan Stanley projects $2.9 trillion in spending from 2025 to 2028 on chips, servers and data-center infrastructure. The investments, the bank says, will contribute as much as 0.5% of U.S. gross domestic product growth this year and next. The infrastructure demands are so huge, though, that companies don’t have enough cash to pay for the investments, according to Morgan Stanley, which estimates a $1.5 trillion financing gap.

Takeaways by Macro Roundup® AI

  1. Alphabet, Microsoft, Amazon and Meta are set to spend nearly $400 billion on capital expenditures this year, largely on AI infrastructure, exceeding what the European Union spent on defense last year.
  2. Morgan Stanley projects $2.9 trillion of spending on chips, servers and data-center infrastructure from 2025 to 2028, and says the investments will contribute as much as 0.5% of U.S. GDP growth this year and next.
  3. Morgan Stanley estimates a $1.5 trillion financing gap, as the companies lack enough cash to pay for their planned AI infrastructure investments.

Related Articles:

  • SPVs, Credit, and AI Datacenters — Companies funding AI data centers through jointly owned special purpose vehicles keep the borrowing off their own balance sheets while retaining shared control and use of the facilities. AI data center spending already exceeds telecom spending during the dot-com bubble, and weak returns would quickly wipe out the thin equity cushion above the vehicles' debtholders.
  • Microsoft’s Datacenter Freeze – 1.5GW Self-Build Slowdown & Lease Cancellation Misconceptions — Reported cancellations of about 2GW of Microsoft datacenter capacity cover only non-binding letters of intent, while the company holds roughly 5GW of pre-leased capacity under binding contracts slated to begin operating between 2025 and 2028. The authors estimate Microsoft accounted for more than 60% of all newly leased turnkey datacenter capacity over a recent six-quarter stretch.
  • 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.

AI Summary. Alphabet, Meta and Microsoft now hold property and equipment worth more than 60% of their equity book value, up from 20% a decade ago, and their combined capital spending with Amazon and Oracle exceeds that of all listed US industrial companies combined. Morgan Stanley estimates a $1.5tn financing gap between capital spending and cash flows over the next three years.

Economist Staff The Economist
Date Posted:

Driven by the AI buildout, Google, Meta, and Microsoft’s property and equipment constitute more than 60% of their combined book value, up 20% over the decade.

How will tech giants fund AI spending that outpaces their cash flows?

Core argument: Morgan Stanley's calculations indicate a $1.5trn "financing gap" between capital expenditure and cash flows at Alphabet, Meta and Microsoft over the next three years, with capex growing faster than cashflows.

Investors who bought shares in Alphabet, Meta and Microsoft a decade ago are sitting on eight times their money, excluding dividends. Spending on data centres means the firms possess property and equipment (accounting-speak for hard assets) worth more than 60% of their equity book value, up from 20% over the same period. Add the capital expenditure of these firms during the past year to that of Amazon and Oracle, two more tech giants, and the sum is greater than the outlay of all America’s listed industrial companies combined. Since 2023 Alphabet, Meta and Microsoft have divided $800bn of operating cashflows roughly evenly between capex and shareholder returns. Capex is growing faster than cash flows. Morgan Stanley’s calculations indicate a $1.5trn “financing gap” between the two over the next three years.

Takeaways by Macro Roundup® AI

  1. Morgan Stanley's calculations indicate a $1.5trn "financing gap" between capital expenditure and cash flows at Alphabet, Meta and Microsoft over the next three years, with capex growing faster than cashflows.
  2. Data-centre spending has lifted property and equipment at Alphabet, Meta and Microsoft to more than 60% of their equity book value, up from 20% a decade ago.
  3. Combined capital expenditure over the past year at Alphabet, Meta, Microsoft, Amazon and Oracle exceeded the outlay of all America's listed industrial companies together.

Related Articles:

  • Honey, AI Capex Keeps Eating … Everything — The author estimates that AI-related capital spending contributed between 0.6% and 1.3% of a quarter's 3% U.S. GDP growth, based on information technology equipment and intellectual property investment data. Google, Amazon, Meta, and Microsoft alone spent about $69bn in the quarter, roughly half of all information technology equipment spending on an annualized basis.
  • 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. Just over half of violent crime victims know the perpetrator and have been attacked before, and much of it occurs away from public view, while snatch theft is almost always committed in public by a stranger against a first-time victim. Shoplifting erodes the sense of public order through visible security measures and locked merchandise.

John Burn-Murdoch Financial Times
Date Posted:

While the crime rate has fallen in the US and UK, incidents of fare evasion, driving offenses, shoplifting, and dog attacks have risen since the pandemic. As public disorder has increased, the % of those who believe overall crime has increased has risen.

Does petty theft shape public perceptions of disorder more than violent crime?

Core argument: Just over half of violent crime victims know the perpetrator and have been attacked before, and such crime often occurs away from public view, making it a distant phenomenon for most people.

Violent crime is especially alarming, but most of it is not random — just over half of victims know the perpetrator and have been attacked before. It also often happens away from public view, making it a distant phenomenon for the majority. By contrast, rapidly rising “snatch theft” is almost always carried out in public by a stranger on a first-time victim, giving a much greater sense that “this could happen to me”. Shoplifting is especially damaging to the sense of public order, as people witness not only the act but the lasting consequences: the growing presence of security personnel and protective casings or locks on items from steak to cosmetics. Fare evasion in London and New York is running much higher than a decade ago, hospitalisations for dog bites have more than doubled over the past 20 years and careless driving also appears to be ticking up.

Takeaways by Macro Roundup® AI

  1. Just over half of violent crime victims know the perpetrator and have been attacked before, and such crime often occurs away from public view, making it a distant phenomenon for most people.
  2. Fare evasion in London and New York is running much higher than a decade ago, and hospitalisations for dog bites have more than doubled over the past 20 years.
  3. Shoplifting is especially damaging to the sense of public order, as people witness both the act and its lasting consequences: more security personnel and protective casings or locks on items from steak to cosmetics.

Related Articles:

  • It’s Time to Talk About America’s Disorder Problem — A meta-analysis of 56 studies found that disorder policing consistently reduces crime, but that indiscriminate "aggressive order maintenance" arrests did not significantly reduce crime while problem-oriented strategies targeting specific sources of disorder often had sizable effects without spillovers. The authors note that a large share of disorder comes from a small number of people and places.
  • The Collapse of Broken-Windows Policing in New York City, Los Angeles, and Washington, DC, 2013–22 — Arrests for minor "broken-windows" offenses—counted as misdemeanors, violations, and infractions per 100,000 residents—fell 74% in New York, 77% in Washington, DC, and 81% in Los Angeles over the period studied, declining steadily in New York and Los Angeles.
  • As America’s Marijuana Use Grows, So Do the Harms — An analysis of federal survey data estimates that about 18m US adults — nearly a third of adult cannabis users — report symptoms of cannabis use disorder, with roughly 3m considered addicted. Among those aged 18 to 25, the cannabis use disorder rate of 16.6% matches the alcohol use disorder rate of 15.1%.

AI Summary. China's daily Central Parity Rate, nominally the midpoint of the yuan's trading band, now functions in practice as a cap on daily appreciation, since the onshore yuan almost never trades stronger than the fix. On the depreciation side, the fix minus 2% still bounds how far the yuan can weaken.

Brad Setser Council on Foreign Relations
Date Posted:

Setser notes that “in the last few years, the onshore yuan almost never trades ‘strong’.” He suggests that China’s de facto FX policy is to limit appreciation pressure by treating the Central Parity Rate (the “fix”) more as an upper limit than a target.

Is China's currency fix now a cap on daily appreciation?

Core argument: The onshore yuan has almost never traded stronger than China's daily Central Parity Rate in recent years, suggesting the fix now functions as the upper limit on daily appreciation rather than the midpoint of the trading band.

The daily Central Parity Rate (“fix”) is supposed to be the middle point in the daily trading band for China’s currency. But over the last few years, the onshore yuan almost never trades “strong” (numerically, a smaller number; the yuan is stronger if it takes six yuan to buy a dollar than if it takes seven or more) relative to the fix. It looks like—de facto—China has shrunk the band and shifted the role of the fix from the midpoint of the band to the upper end of the band. That of course allows China a lot more control over its currency on a day-to-day basis. The fix defines the upper limit on the day’s appreciation, more or less. [On the other hand], when the yuan is under depreciation pressure, the fix, minus two percent, still sets the limit on how much the yuan can depreciate.

Takeaways by Macro Roundup® AI

  1. The onshore yuan has almost never traded stronger than China's daily Central Parity Rate in recent years, suggesting the fix now functions as the upper limit on daily appreciation rather than the midpoint of the trading band.
  2. China's apparent de facto narrowing of the yuan's trading band, with the fix acting as a ceiling on appreciation, affords Beijing more day-to-day control over its currency.
  3. When the yuan faces depreciation pressure, the fix minus 2 percent still sets the limit on how far the currency can weaken.

Related Articles:

  • The Case that China is Now Actively Resisting Pressure on the Yuan to Appreciate — The author argues China has been resisting pressure for the yuan to appreciate against the dollar, citing state bank foreign asset accumulation, balance of payments data, a yuan trading strong relative to its official fix, and positive foreign exchange settlement. Inflows from China's large trade surplus exceed private capital outflows.