Perspective

Your stock portfolio, the core of your wealth and retirement security, is being reduced every year by a secret 2% tax, one you didn’t vote for and won’t find disclosed in any prospectus. And it will o... - 81% of the 2400 stocks in the MSCI's global stock index are already suffering losses due to extreme heat, costing $3.9 trillion every year. There are no safe havens from climate change, not even in a stock portfolio Gift link to my @opinion.bloomberg.com column

Based on charts like the one at right, Paul Krugman and many others are arguing that: Massive spending on data centers is crowding out investment in everything else in the economy. Not so fast. Can you see one immediate problem? How is $51b in data center construction spending crowding out $120b in everything else! Indeed, crowding out in the standard loanable funds model is generically less than 1 to 1 –that is, $51b of additional spending will crowd out less than $51b of other investment spending. The standard analysis is that increased investment demand/borrowing increases interest rates–the higher interest rate reduces investment demand elsewhere (that’s the crowding out) but the higher interest rate also increases savings so investment elsewhere falls by less than the increase in tech investment. Here’s the classic graph from our textbook, Modern Principles.

Your stock portfolio, the core of your wealth and retirement security, is being reduced every year by a secret 2% tax, one you didn’t vote for and won’t find disclosed in any prospectus. And it will only get bigger as the planet gets hotter. - Extreme heat saps productivity by $3.9 trillion a year, translating into lower revenue for companies and hurting investors

Not a week goes without another piece of datacenter FUD. We’ve already debunked in depth why moratoriums aren’t as impactful as one may think. Despite that, many still misunderstand the pace of the global AI datacenter buildout. Just the other week, Brad Gerstner from Altimeter showed our data on the All-In Podcast and stated that “SemiAnalysis’ forecast of +43GW in 2027 is too aggressive” and that he thought only about 25GW would get built, close to half of our number, based on his “hunch”. But let’s not operate on hunches. It’s very tempting to cede to the FUD given all the constraints on the datacenter buildout: power, labor, supply chains, capital, and many more. But at SemiAnalysis, we care about the truth, not about narratives. And here’s the truth for you: one picture, >24GW of datacenters in advanced stage of construction, most of it to be delivered in 2027.

Ramez Naam is, in my opinion, one of the best futurists alive. But although he has a Substack (which you should definitely follow), he doesn’t write nearly enough. So I’ve been encouraging him to write more by cross-posting his posts on my blog. Somewhat to my surprise, Ramez has turned out to be a lot more of an AI skeptic than I would have though. He’s certainly not dismissive of the technology, but he disagrees with many of the more expansive and dramatic claims coming out of the big labs (especially Anthropic). In his first post in this series, he expressed skepticism that recursive self-improvement (RSI) would lead to a fast takeoff of AI capabilities in the near future: In this post, he expresses skepticism about the potential for agent swarms — large numbers of AI agents working in parallel — to offer dramatic increases in AI capabilities.

From the October WSJ survey: Figure 1: Ten year Treasury yields (black), October WSJ survey mean (blue square), July WSJ survey mean (light green triangles), all in %. Source: Treasury via FRED, WSJ. A mix of improving growth, new inflationary pressures and concerns about the U.S. fiscal trajectory have pushed bond yields higher since the start of the summer. Many economists said the cocktail has forced them to re-evaluate their assumptions about investors’ demand for Treasurys, whose yields rise when their prices fall. The low prediction for December 2027 is 3.95%, high is 5.63% (the indomitable James Smith, and Allen Sinai, respectively).

As I documented in my last primer, AI-related capital expenditure is now dramatically crowding out investment in almost every other sector of economy – especially housing construction and manufacturing. Furthermore, data centers are incredibly capital-intensive: that is, they deploy vast amounts of capital while employing very few people. A given amount of capital employs far more workers when used to construct a factory, an office building or even a distribution center. So when investment is diverted away from other sectors to AI, the overall demand for labor falls relative to what it would have been before the shift. Hence, the AI boom will depress the demand for labor in the economy. Despite the recent panicked claims that AI will wipe out millions of jobs, the shift towards AI probably won’t lead to mass unemployment. However, it will lead to slower growth in wages.

The abstract of a new paper coauthored with Laurent Ferrara, “Missing Recessions: Revisiting the Term Spread’s Predictive Power”: The failure of a recession to materialize in 2024 has forced a reappraisal of the term spread as the most reliable recession predictor for the US. In this paper, we re-evaluate the predictive power of the term spread, and explain the false positive signal in terms of sample specificity, simultaneity, and omitted variables, focusing on the United States, but including other advanced economies for which the term spread has proven to be a useful predictor in the past. We confirm that prior to the 2020 recession, the yield curve was an accurate predictor of recessions in the US, Canada, and Germany, but proved less accurate thereafter.

and Monthly GDP for August, manufacturing production. First, NBER BCDC indicators: Figure 1: NFP employment (bold blue), civilian employment with smoothed population controls (bold orange), industrial production (red), personal income excluding current transfers in Ch.2017$ (bold light green), manufacturing and trade sales in Ch.2017$ (black), and monthly GDP in Ch.2017$ (pink), GDP (blue bars), all log normalized to 2025M01=0. Source: BLS via FRED, BLS, Federal Reserve, BEA 2026Q2 3rd release, S&P Global Market Insights (nee Macroeconomic Advisers, IHS Markit) (9/2/ 2026 release), and author’s calculations.

In a world of competing currency blocs, the euro is the dollar’s most credible rival. But for it to become a genuine counterweight to the greenback, the European Union must strengthen its financial foundations and expand its reach, starting with the member states that have not adopted it despite being obliged to do so.

In 2012, when Japan was dealing with chronic deflation, policymakers were right to embrace monetary easing, including lower interest rates. Today, however, Japan faces the opposite problem: a weak yen, combined with higher interest rates in the US, calls for a much tighter monetary policy.

Yesterday, in Part 1: Current State of the Housing Market; Overview for mid-October 2026 I reviewed home inventory and sales. In Part 2, I will look at house prices, mortgage rates, rents and more. House Prices Here is an interesting graph from Realtor.com on median list house prices. According to Realtor.com, list prices peaked in 2022 (when interest rates moved sharply higher) and have moved mostly sideways for four years. These are median prices and can be distorted by the mix, but median list prices have been down year-over-year for eleven consecutive months and are the lowest for September since 2021. Also, June median prices are typically the peak for the year - so this measure of prices will continue to decline through January. The Case-Shiller National Index increased 1.9% YoY in July and will likely be up about the same in the August report (based on other data).

In 1973, the great MIT economic historian Charles P. Kindleberger published The World in Depression 1929–39. In its wide-ranging coverage, it still stands as the definitive history of the Great Depression. In 2022, during the recovery from COVID, Perry Mehrling (1959-2026) published an intellectual biography of Charles P. Kindleberger, which I had the honor of reviewing in the New Statesman. As Mehrling makes brilliantly clear, Kindleberger, like Minsky, held fast to the obvious fact that money mattered. And it mattered not just nationally. Through the channels of global capitalism, it mattered globally. What Mehrling helps us to understand is that Kindleberger was not just an analyst of inter-national finance, of balance of payments and currency, of the gold standard and Bretton Woods. Kindleberger was something even more interesting.

The automobile changed our world and symbolizes prosperity and personal freedom for billions of people. Making automobiles, meanwhile, is a miserable, grinding affair. - China produces more than two cars for every one sold there. Its excess capacity is everyone's problem, but there's a way out

Contrary to tech optimists' rosy predictions, AI may prove to be a drag on growth. By concentrating wealth and economic opportunities in a few cities while destroying the low-cost jobs that have historically fueled global development, the technology could end up making much of the world poorer and more divided.

Andrea Kendall-Taylor and Richard Fontaine, former members of the National Security Council, argue that the members of what they call the “axis of upheaval”—Russia, China, Iran, and North Korea—are deepening their cooperation. Meanwhile, Washington is creating friction with many of its own partnerships. If these developments continue unabated, the United States will squander the strength of its network just as its adversaries learn to make the most of theirs.

Summary Following today’s PPI release, we are updating our nowcast by +2bps to +34bps m/m and by −1bp to 3.09% for the y/y September core PCE reading (a correction to the 3.12% y/y figure in our earlier email). We maintain our base case that the FOMC will hold rates steady at the October meeting and hike at the December meeting. Today's upside stemmed from the airfares, while the rest of the PPI inputs into PCE were more benign. Today's release came with sizable downward revisions to the August airfares and hospitals data. We now anticipate the m/m August core PCE to be revised down by roughly 2bps to +23bps. PCE Inflation Overshoots by PPI Component Trimmed Mean & Other Summary Gauges With the exception of the Dallas Fed's gauge, trimmed mean measures are running below core PCE (2.65–2.76% vs 3.09%), but still above the Fed's 2% target.

This 2-part overview for mid-October provides a snapshot of the current housing market. Part 1 will review inventory and sales, and Part 2 will review house prices, rents, mortgage rates, and more. The key story for existing homes are: Sales in September were down 2.2% YoY and historically low. Sales SAAR (seasonally adjusted annual rate) have been very low for 4 years. Months-of-supply is at the highest level in over a decade. Inventory growth might have picked up again year-over-year. The median price is up 1.3% YoY, but has peaked for the year. There are also significant regional differences. And 2026 has been a difficult year for homebuilders. There are still a large number of completed homes for sales, a larger than normal number of unsold homes under construction and builders have been reducing prices to clear inventory.

The first time in the memory of living man; before—even in the 1970s—a short-term rate cut was seen as a sign the long-term bond should read and follow, as the Fed probably knew something...

One of the great ironies of American politics is that Donald Trump won the 2024 election basically because a significant number of credulous voters believed his promises to bring prices down beginning on “day one.” What actually happened was that America, along with the rest of the world, experienced a temporary surge in inflation in 2021-22, mainly because of the lingering effects of Covid disruptions. And voters still blamed Joe Biden for that surge. But by 2024 the war against inflation had been largely won, at startlingly low cost — the recession many analysts claimed would be necessary to get inflation down never happened, but inflation came way down anyway. And then Trump snatched defeat from the jaws of victory. There is still some controversy among economists about how much responsibility Biden administration policies bear for the 2021-22 inflation surge. My view is, very little.

As an OpenAI whistleblower recently explained, AI executives have knowingly fostered a culture that “guarantees periodic failures,” the scale of which is “growing as systems get more capable.” This approach hinges on an expectation of impunity, which is why AI firms must be held legally accountable for the harm their agents cause.

Seasonal postings are running ahead of last year — and ahead of 2019 — while some of the biggest seasonal employers are changing how they’re hiring. The post Seasonal Hiring Is Off to a Fast Start, With an Eye Toward 2027 appeared first on Indeed Hiring Lab.

From the Great Recession until 2019, U.S. labor productivity growth disappointed. Nonfarm-business output per hour grew about 1 percent a year from 2010 to 2019, down from 3 percent in the late 1990s. This slowdown was especially pronounced in manufacturing, and it coincided with a slowdown in the growth of revenue per worker among public firms. However, since 2019, growth has picked up to twice the 2010s rate, around 2 percent. In this post, I use data from firms’ 10-K filings to unpack this recent acceleration. I first show that the acceleration appears in public firms’ revenue per worker, then document how the gains went into profit margins, and finally, what firm characteristics predicted significant profit margin increases. The increase in profit margins is consistent with ongoing declines in the labor share of output.

The Fed’s odds of threading the needle and getting inflation back down to target while avoiding a recession get much lower; & Fed Chair Warsh’s oaths of fealty to Trump are not helpful at all...

Today I’m going to follow up on affordability, sparked by a remarkable Wall Street Journal poll in which large majorities, left and right, support populist economic policies. “The rent is too damn high.” So, people respond, pass rent control, subsidize rents, force builders to construct and rent out so-called “affordable units,” or have the government build and provide housing. They give the same answers for childcare, health care, prescription drugs, and energy. Why are people so enamored of policies that have failed time and again, over centuries? Why are simple cause-and effect truths seemingly so hard to understand? “Don’t transfer income by distorting prices” is a bedrock economic principle.

Far-right presidential candidate Flávio Bolsonaro holds up Argentine President Javier Milei’s economic program as a model for Brazil. But Milei's economic experiment has been a failure, leaving millions of Argentinians worse off and effectively turning their heavily indebted country into a client of the United States.

The AI boom is showing up in different corners of the U.S. economy, including international trade. Imports of semiconductors, servers, and computing equipment have climbed sharply yet the U.S. current account deficit has been largely unchanged. To better understand how the AI boom has not substantially affected the trade balance, consider that the current account is dictated by the gap between domestic saving and investment spending. It turns out that aggregate saving and investment spending have not changed much during the AI boom. Aggregate investment has not risen due to weakness outside the AI sector and business saving has remained robust.

From Steve Kamin: Until early in President Donald Trump’s second term, the sensitivity of the dollar to the difference between the slope in the U.S. yield curve and foreign yield curves was mainly positive, as expected. Increases in U.S. 10-year yields relative to those abroad boost the dollar. But immediately after “Liberation Day” in April 2025, when Trump shocked the world with his announcement of “reciprocal” tariffs, that sensitivity turned negative and started fluctuating widely around a lower baseline. The change in the dollar’s sensitivity is even more apparent when we compare its average value over the pre-Liberation Day period with that during the post-Liberation Day period.

Scott Singer and Pavlo Zvenyhorodsky, researchers at Carnegie’s China AI Initiative, argue that China is betting that AI dominance will come from combining hardware and software, or AI-powered robots. By focusing only on software such as AI models, they argue, Washington and Silicon Valley risk falling behind.

As fall starts to show itself in cooler weather, I want to discuss a topic that usually raises temperatures: inflation. Like a budget-conscious consumer, the Federal Reserve is highly attuned to changes in the price of goods and services; minimizing such swings is one of the Fed’s primary duties, after all. But how does the Fed’s view of prices differ from that of a consumer? And how does the Fed go about stabilizing prices writ large when it lacks the ability to set prices directly? In this post, I explore that apparent paradox from the perspective of a central banker (though I remain a budget-conscious consumer when off duty).

Robert Zoellick, an expert on the global economy who served as U.S. trade representative and World Bank president, makes the authoritative case that President Donald Trump’s antagonism toward Canada and Mexico is destroying an economic and security partnership that that enables the United States to project power globally.

AI safety is on fire. On 12 September 2026, Dario Amodei published an essay arguing that frontier labs must deliberately slow the pace at which they improve model capabilities. China was central to his geopolitical argument: export controls and tighter security could preserve America’s lead, creating room to slow down and leverage for a future agreement with Beijing. Within a day, Sam Altman and Elon Musk supported Dario’s call. However, President Trump scoffed. On 14 September 2026, he wrote that fears of AI destroying humanity are “a HOAX” and that “the only one that is happy about it is China.” Beijing scoffed too, with the state-run Global Times calling Dario’s essay a “Cold War playbook.” The debate then moved onto the diplomatic calendar. On 20 September 2026, Treasury Secretary Scott Bessent and Vice Premier He Lifeng held the first US-China AI dialogue.

The “Honey Badger don’t care” stock market: why and how twenty “AI” stocks and only twenty “AI” stocks shrug off the big shock of rising interest rates in our weird two-speed stock market: The longest-duration assets in the market should be the most sensitive to interest rates. Share Instead they are the least. Why? Either the marginal AI investor does not care about discounting cash flows but only about being blessed at the Ω-point by the coming Digital God, or it is the AI-driven datacenter buildout’s capital requirements and pretty much only the AI-driven datacenter buildout’s capital requirements that is driving rates up right now: Give a gift subscription Katie Martin & Rob Armstrong: There Is Only One Trade: PODCAST OF THE DAY > > Katie Martin: Stocks right now are not really what they seem.

Here is the ICE October Mortgage Monitor report (pdf). Press Release: ARM Demand Hits Nearly Four-Year High on Rising Mortgage Rates Intercontinental Exchange, Inc. … today released the October 2026 ICE Mortgage Monitor Report. The analysis finds that adjustable-rate mortgages (ARMs) are drawing renewed borrower interest as mortgage rates climb. ARMs accounted for more than 11% of rate locks, their largest share in nearly four years, as ICE’s Conforming 30-year Fixed Rate Index reached 7.2% on Sept. 24. “ARMs are becoming more attractive to borrowers looking for relief from today’s higher fixed rates, but the overall market exposure to adjustable payments remains relatively limited,” said Andy Walden, head of Mortgage and Housing Market Research at ICE.

Until now, discussions of the potential economic problems created by the AI boom have mostly focused on the future. Will it destroy millions of jobs? Is there a bubble and will it cause a severe recession if it bursts? Will it destroy humanity? While these are doubtless very important questions, attempts to answer them are at best educated guesses. We don’t have to guess, however, about one major effect of the AI boom, because it’s happening right now: Massive spending on data centers is crowding out investment in everything else in the economy. “Crowding out” is a familiar term in economics, usually considered a consequence of government spending. When a government engages in deficit spending and the economy is at or near full employment, it drives up interest rates.