WSJ:
Massive data-center spending has been fueling economic growth all year. Goldman Sachs recently projected that artificial-intelligence investment in the U.S. will hit a staggering 1.9% of GDP. While much of that money is spent on imported chips, the boom is also creating plenty of construction jobs in the U.S. as developers race to build server farms. The rise in construction spending has helped compensate for a decline in home-building.
Spending on imported goods surged by more than a quarter on an annualized basis, driven in part by semiconductor purchases from abroad. Overall net imports subtracted 3.3 percentage points from GDP growth.
GDP reflects the total of all spending. But since some of that spending is on imported products, rather than things made in the U.S., imports are considered a drag on GDP.
Competition On the Merits examines the incentives of state antitrust enforcers.
PROBLEM: The states bring cases against mergers whose effects fall mostly outside their borders, and ask settlement terms that have little to do with competition.
What do the states get by challenging National mergers? Attorney fees, local jobs, donations to local firms, and DEI commitments. And don't forget free PR. Many state AG's are governors-in-waiting.
SOLUTION: What can be done to realign incentives?
Eleven years old but still relevant for the trick-or-treaters this week.
Soaring cocoa costs prompted Mondelēz, the maker of Cadbury and Milka chocolate, to raise European chocolate prices by roughly 30%. Management anticipated a decline in sales. But consumers proved more price-sensitive than expected. Historically, the company had observed demand elasticities of approximately −0.4 to −0.5. Following the increases, elasticity rose in magnitude to roughly −0.7 to −0.8. On its October 2025 earnings call, CEO Dirk Van de Put noted that some chocolate bars had crossed important price thresholds, prompting consumers to reconsider their purchases. Competitors' decisions not to fully match Mondelēz's price increases also contributed to the disappointing sales response.
Marshall's Second Law of Demand says that demand tends to become more elastic as price rises. At a higher prices, more products are viable alternatives. Consumers may barely notice a price increase when chocolate is relatively inexpensive. But as prices climb, more consumers begin substituting. They may choose cheaper brands, purchase smaller packages, or forego chocolate altogether. Managers should be cautious about extrapolating historical elasticity estimates. The response to a 30% price increase is likely "out-of-sample" from the historical context.
The calculation is complicated by competitors' responses. Campbell's assumes that competing brands will not immediately match its price increases, making its products relatively more expensive. If competitors also raise prices, Campbell's could lose fewer customers than its forecast predicts. Cunfer acknowledged that the company is using different elasticity estimates across its product lines and has deliberately chosen conservative assumptions. That is, Campbells appears to be making relatively sophisticated use of elasticity concepts.
Will this have longer-term repercussions for SpaceX? When my coauthors failed to meet deadlines, I could recover from this disappointment, but I learned I could not trust them in the future. Right now, SpaceX is the only viable satellite based communications option. But there are five others either being deployed, or planning to be deployed, by 2030. Unless there are substantial switching costs, potential customers should consider these alternatives to be more trustworthy. Investors apparently recognized the value of reducing dependence on Starlink: following reports of the restrictions, shares of its European competitor Eutelsat jumped 23% in a single day.
The researchers examine Affordable Care Act insurance exchanges, where lower-income consumers receive subsidies that reduce their deductibles and copayments. These subsidies encourage greater healthcare utilization, raising costs that insurers spread across their customers through higher premiums. Unsubsidized customers consequently pay more, even though their own behavior hasn't changed. Some respond by dropping coverage, potentially raising average costs further. The surprising lesson is that moral hazard creates an externality: one customer's response to generous insurance can make coverage more expensive for everyone else.
Since national house prices increased very quickly during the pandemic - up over 40% - it seems likely that some of the usual “stickiness” will not apply. I think the most likely scenario now is nominal house prices declining 10% or more from the peak, and real house prices declining 25% or so over the next 5 to 7 years.
The knee-jerk response to high merchandise return rates is to make returns more difficult. Charge a restocking fee, shorten the return window, or require more documentation. But what if these restrictions drive away your best customers? According to a recent Wall Street Journal report, 68% of retailers now charge return fees at least some of the time, up from 43% five years ago. An increasing number of online reviews mention unclear return policies. Retailers have good reason to worry: an analysis by Riskified flagged nearly one-quarter of refund dollars in its sample as potentially abusive.
The problem is adverse selection. Customers who rarely return merchandise may value the assurance of an easy return if something goes wrong. Faced with new restrictions, they might shop elsewhere. Meanwhile, customers who frequently return merchandise may be more willing to tolerate the restrictions because they have learned how to navigate them. A recent survey by Loop found that 63% of respondents had abandoned a purchase or retailer because of its return policy. By making returns harder, a retailer could inadvertently drive away profitable customers while retaining a disproportionate share of costly ones. A potential solution may be to use purchase and return histories to wave restrictions for more reliable customers.
Although Columbus is criticized for the exploitation, enslavement, and suffering associated with European colonization, his voyages offer an instructive example of the risks of investing in exploration. In 1492, Spain's monarchs financed an expedition to find a commercially valuable route to Asia, uncertain whether Columbus would succeed—or even return. But there was another risk: would he be able to claim the rewards of a successful voyage? The Capitulations of Santa Fe promised Columbus hereditary privileges, including the offices of Viceroy and Governor of the territories he discovered. Yet the Crown subsequently removed him from office, and after decades of litigation, his heirs lost their claims to those hereditary governorships. The Treaty of Tordesillas of 1494 sought to reduce another source of uncertainty by dividing overseas claims between Spain and Portugal, although other nations and Indigenous peoples were not parties to the agreement.
Five centuries later, companies exploring the ocean floor face similar uncertainties. ABC News reports that some companies hope to recover valuable nickel, cobalt, and other minerals from the deep ocean. But collecting minerals thousands of meters underwater is technically difficult and commercial profitability remains unproven. In August, NOAA announced that it had advanced its review of an application by The Metals Company to explore and commercially recover minerals from the Pacific seabed. Even if the technology works, property rights remain uncertain. The United States has its own permitting process, but the International Seabed Authority maintains that mining in international waters requires authorization under its international framework.
Both ventures illustrate two distinct risks of exploration: the technology may fail, and even if it succeeds, (perhaps because it succeeds) the claim to the rewards may fail. The expected return on investment depends on both probabilities. Property rights are valuable only to the extent that they can be enforced.
The stated Trump administration immigration policy has been to reduce Type I errors, reduce illegal immigration. Collateral damage is an increase in Type II errors, fewer legal immigrants. The Cato Institute reports that legal immigration fell substantially more than illegal immigration. Comparing monthly entry figures through February 2026 with those before Trump took office, Cato estimates that legal entries fell by approximately 145,000, while illegal entries fell by about 53,000. The restrictions affect people seeking admission through established legal channels, including employment, family reunification, and education. A full error cost analysis would require an estimate of the cost to society of typical illegal entrants and the lost benefit of typical legal entrants. It is open to debate, but evidence suggests that immigration is a net positive and that illegal immigrants to the US are not a cost.
One frequently hears that potential immigrants should simply follow the legal process. This offers little guidance if that too is unavailable.
This resembles the familiar razor-and-blades pricing strategy: sell customers a durable product, then earn recurring revenue from the complementary products or services needed to get its full value. Oura's recent IPO filing suggests the strategy is working. Approximately 94% of ring activations convert to paid memberships, and about 85% of paying members remain subscribed after 12 months. Subscriptions account for roughly 20% of revenue but earn gross margins approaching 90%. By separating the price of the device from the price of the information it produces, Oura turns a one-time hardware purchase into a continuing stream of high-margin revenue.
The disruption has also affected oil shipments far from the Persian Gulf. Reuters reports that shipping two million barrels of U.S. crude from the Gulf of Mexico to China now costs approximately $80 million, or $40 per barrel, compared with just $8.60 per barrel before the war. The elaborate shuttle operations in the Persian Gulf have tied up tankers, reducing their availability elsewhere. Freight rates have risen so much that traders say the arbitrage opportunity for shipping U.S. oil to Asia has disappeared. Asian refiners are consequently considering alternative suppliers in the Middle East and Latin America.
The arbitrage cost is comparable in magnitude to the increase in Brent crude prices since the war began. The ability to exploit this form of arbitrage essentially put a cap on how much the price rose between between the well and the refinery.
One reason demand curves slope downward is that people tend to tire of repetition. This is what Mark Twain discovered hiking in Swiss Alps
The jodling (pronounced yodling,—emphasis on the o,) continued, and was very pleasant and inspiring to hear. Now the jodler appeared,—a shepherd boy of sixteen,—and in our gladness and gratitude we gave him a franc to jodel some more. So he jodled, and we listened. We moved on, presently, and he generously jodled us out of sight. After about fifteen minutes we came across another shepherd boy who was jodling, and gave him half a franc to keep it up. After that, we found another jodler every ten minutes; we gave the first eight cents, the second one six cents, the third one four, the fifth one a penny, contribution nothing to Nos. 5, 6, and 7, and during the remainder of the day hired the rest of the jodlers, at a franc apiece, not to jodel any more. There is somewhat too much of this jodling in the Alps.
[Mark Twain, A Tramp Abroad: Following the Equator, Other Travels, edited by Roy Blount Jr. (New York: Library of America, 2010), p. 187]
Sellers should prefer auctions to simply negotiating with a willing buyer. In bilateral bargaining, Dutch Bros has every reason to convince Salad and Go that $105 million is about as high as it will go. But the arrival of another bidder changes the negotiation. Instead of the seller having to determine how much each buyer is willing to pay, the buyers reveal information about their valuations by competing against one another. Indeed, Dutch Bros ultimately chose not to increase its offer, effectively revealing where further bidding stopped making sense for it. Buyers naturally prefer to negotiate without a competitor standing beside them. Sellers prefer exactly the opposite.
The Economist: Many surgical interventions are little better than placebo
The results are upending the field. Removing an inflamed appendix, an operation that around 5-10% of people have had at some point in their life, has turned out to be no better for most patients than a course of antibiotics. Trials comparing different surgical techniques have also revealed that cheaper and less complex operations can be more effective. Most striking, the data show that some widely used surgical procedures, such as spinal fusions and rotator-cuff repairs, have effects indistinguishable from placebo or non-surgical care such as physiotherapy.
Worse, some types of surgery may do more harm than good. A trial in Britain found that prostate-cancer surgery had no effect on patient mortality 15 years on, but worsened sexual and urological problems. A Finnish trial of a common knee procedure, meanwhile, concluded that patients who had real surgery had more problems in the affected knee ten years on than those who had had sham surgery (anaesthesia followed by a superficial incision). David Ring, an orthopaedic surgeon at the University of Texas at Austin, reckons that most operations in his field may be unnecessary. Surgery itself, it would seem, could use a major intervention.
From The Free Press:
On August 26, a wall of rock and ice broke away from Langtang Lirung in the Himalayas, resulting in a massive flood and mudflow that killed thousands of people along the border of Nepal and Tibet. Less than a month later, on September 17, an influential organization called World Weather Attribution (WWA), housed at Imperial College London, published a report on the catastrophe along with a press release touting its alarming findings.What the WWA does:
Motivated Reasoning:WWA has not been shy about the fact that it produces its studies to influence the public and to promote climate advocacy. Authors of one of the WWA’s foundational papers openly acknowledged that a primary reason for creating their methods is political: by “increasing the ‘immediacy’ of climate change, thereby increasing support for mitigation.”Detection by Correlation and Attribution by Assumption...WWA treats “climate change,” measured as a smoothed record of global temperature, as the only possible cause of any localized trend in rainfall, heat, flooding, or other types of extreme weather.People are fooledThe WWA’s advocacy campaign has worked brilliantly, of course; 50 percent of Americans now believe policies addressing climate change would reduce the number or severity of the extreme weather events that they individually, or their communities, will experience.
Who funds these idiots?What the methodology should do (IPCC, Intergovernmental Panel on Climate Change):WWA’s funders are philanthropists: the Grantham Foundation, the European Climate Foundation, and the Bezos Earth Fund, three of the largest backers of climate advocacy in the world.
The first step is detection: whether the long-term statistics associated with a particular type of extreme weather change—a century of hurricane landfalls, say, or peak river flows—can’t be explained by the normal fluctuations in weather over years and decades, called internal climate variability. [COMMENT: this is extraordinarily difficult given the long run cycles in climate data]
The second step is attribution: The IPCC starts with a detected change and assesses the possible causes underlying that change—greenhouse gases, aerosols, volcanoes, land use, modes of climate variability, etc.—using sophisticated computer models of the earth’s climate system. [COMMENT: ruling out alternative explanations from non-experimental data is also extraordinarily difficult].
IPCC Findings:If lack of knowledge were the problem, I would send WWA a copy of Chapter 17: Making Decisions with UncertaintyThe scientific reality—long understood in mainstream climate science—is that any emissions policy a government adopts today will not noticeably reduce the frequency and intensity of extreme weather, through the lifetimes of children or their children’s children.
The contrast highlights retailer actions boosting manufacturer demand. Helping a customer experience a pair of smart glasses may increase demand for Meta’s product, but Best Buy bears much of the cost of providing the service. Worse, customers can learn about a product at Best Buy and then buy it elsewhere—the classic retail-services free-rider problem. Meta’s solution is to become more deeply involved in the retail experience, helping provide dedicated specialists and an elaborate branded demonstration space. Car-audio manufacturers are responding more modestly by financing their own displays as Best Buy retreats from providing them. In both cases, manufacturers that benefit from demonstrations are taking on more of their cost, better aligning the incentives of manufacturers and retailers to provide services that help sell the product.
Prediction-market trades have turned up whose payouts nearly match college football coaches' playoff bonuses. Five line up almost exactly with Lane Kiffin's bonuses at LSU. The schools didn't place them, but third parties acting for them plausibly did.
Why would a school bet on its own success? Coaching contracts promise big bonuses for making the playoffs, advancing and winning a title. A deep playoff run can leave the athletic department owing millions it didn't budget for. A contract that pays off when the team wins covers that bill when it comes due. Like a farmer locking in a crop price, the school gives up some upside in exchange for a predictable budget.
Why go through a third party? NCAA rules forbid athletic staff from betting on games, and prediction markets count. So a school can instead pay an insurer a premium, and the insurer buys the contracts itself.
Takeaways:
Source: Josh Hendrickson, Insuring Against Success?, Economic Forces
That changes the incentive problem. If AI substitutes for routine research, analysis, or coding, producing more of those activities becomes a less useful measure of employee performance. Knowing which questions to ask, recognizing when an AI answer is wrong, exercising judgment when information is ambiguous, and finding new ways to combine AI with expertise can become more valuable. But these qualities are also difficult to measure. Counting hours, completed reports, or even AI usage is relatively easy; determining whether an employee exercised good judgment is not. EY's new program therefore allows employees at all levels to recognize colleagues, while other EY initiatives use simulations, coaching, and assessments of decision-making and collaboration to identify these less tangible skills.
Moving from objective output measures toward judgment, adaptability, and collaboration also changes the role of supervisors. Managers must have more discretion to decide whose contributions actually created value, making the quality and credibility of those evaluations more important. EY has argued that AI-era performance management should place greater weight on peer feedback, coaching, adaptability, and cross-functional impact, rather than relying simply on traditional manager ratings. It is also investing in more structured ways of assessing these skills: its new Career Residency program uses workplace simulations, coaching, and client situations to assess decision-making, collaboration, learning, and adaptation. As easily measured tasks become easier to automate, performance evaluation may consequently depend more on subjective judgment, but also on multiple evaluators and better ways of evaluating the evaluators.
Almost. The opportunity cost is what she would do with the surplus power she produces. Her utility will buyback her surplus power at $0.05-$0.08 per kwh while her retail rate is $0.45-$0.47 per kwh. Back in Texas these would be $0.07-$0.08 per kwh and $0.13-$0.19 per kwh. The opportunity cost in California is nearly 1/10 of the price while it is close to 1/2 in Texas. In Texas, I might drive a bit more if I had an EV and solar panels. In California, I would drive a lot more.
One of the most valuable decisions in retirement saving is when to start. Over the 50 years from 1976 through 2025, the S&P 500 delivered a compound annual return of about 11.82%, including reinvested dividends. Suppose a young worker had invested just $100 each month in an S&P 500 fund and earned that historical compound return until age 65. Starting at age 25 would produce about $922,000. Waiting until 30 reduces the ending balance to about $523,000, while waiting until 35 reduces it to about $295,000. Scraping together $100 when you are 25 is difficult, but if you can do it consistently, voila you will become a millionaire.
What makes the comparison striking is how little of the difference comes from the contributions themselves. Doubling your set-aside each month will merely double the ending balance. But by starting at 25 versus 30, you contribute only $6,000 more yet finish with roughly $399,000 more. Save 15% more and finish with 76% more. Compared with starting at 35, the early saver contributes just $12,000 more but finishes with about $627,000 more. The reason is compound growth: dollars contributed early earn returns, those returns earn returns, and the process repeats for decades. Of course, future stock returns may be lower than the past 50 years, and actual returns are anything but smooth. But the lesson does not depend on the return so much. The opportunity cost of waiting to save is not primarily the contributions you miss, it is the decades of compound returns those contributions could have earned.
You can use this SEC site to do your own “what if” scenarios.
