The Federal Reserve Bank of New York released a seemingly innocuous study last week, really just a blog post, that contains within it a hugely important insight about the problem of inequality in the free market. In “Same Shock, Different Roads? A K‑Shaped Pattern at the Pump,” a team of Fed economists used detailed consumer expenditure data to look at how March’s spike in gas prices affected consumption. Their findings:
We find that households had very different experiences with gasoline spending: in March, high-income households increased nominal spending the most and kept real consumption essentially unchanged, while low-income households decreased real consumption of gasoline but still saw sharply increased nominal spending because of the rise in gas prices.
Thanks, Captain Obvious, I hear you say. When gas prices go up, rich people just pay more for their gas. Poor people use less gas, still end up paying more, and have to cut back elsewhere in their budgets, too. No kidding.
But pause a moment to consider the implications. Faith in market efficiency relies upon the price system allocating resources to their best (most valuable) use, which assumes that our “willingness to pay” provides a fair proxy for how much each of us values something. When it becomes more expensive to supply gasoline to the market, the price rises, encouraging more production and reducing purchases from people with a lower willingness to pay, until total demand and supply come into balance.
This is “efficient” according to economists. Those with a lower willingness to pay, suggesting they assign less value to the marginal gallon of gas, should cut back. Our limited supply goes to those who really value it. Lest you think that an oversimplification, here is Harvard professor Gregory Mankiw in his widely used textbook, Principles of Microeconomics:
Those buyers who value the good more than the price choose to buy the good; those buyers who value it less than the price do not. ... Free markets allocate the supply of goods to the buyers who value them mostly highly, as measured by their willingness to pay. ... Thus, given the quantity produced and sold in a market equilibrium, the social planner cannot increase economic well-being by changing the allocation of consumption among buyers.
And no, that’s not just the theory, after which the textbook notes the weaknesses in this framework. To the contrary, the real-world examples on the next page drive the point home: “If an economy is to allocate its scarce resources efficiently, goods must get to those consumers who value them most highly. Ticket scalping is one example of how markets reach efficient outcomes.”
What the Fed study shows is that this isn’t happening at all. Here on Planet Earth, your “willingness to pay” is often driven less by how much you value something and more by the total resources available to you. Does anyone believe that lower-income households are driving less because they value driving less, in the sense that it does less for their welfare? Come on. To the contrary, given that they already had to make tradeoffs that higher-income households did not before the spike in gas prices, it is safe to assume that the marginal gallon of gas they were consuming was already more valuable to them; when they have to cut back, it is surely more painful.
Colloquially, we can think of this as the Learjet problem. When gas prices go up, families take fewer trips to visit grandma. Corporate executives do not reduce their private jet flights to Davos. That may be “efficient,” but it obviously is not efficient in the sense of allocating resources to their best uses and maximizing welfare. Or, to take Mankiw’s example, when concert and baseball tickets get bid up to the level that real fans are priced out and the front rows are filled with lobbyists and their clients, “changing the allocation of consumption among buyers” would quite obviously “increase economic well-being.”
Maybe we don’t want to do that. For one thing, we might worry about interfering with property rights. For another, we might not have any confidence in policymakers to improve allocations in practice. For a third, we might want to preserve the incentive for people to do productive things and earn money that they can then spend as they wish, buying things that people who earn less money cannot. Those are all reasonable considerations.
What is not reasonable, however, is to pretend that—under conditions of substantial inequality—the price system is rationing resources in a way that maximizes consumer welfare (to say nothing of other forms of welfare about which we might care even more). Whether and when we can improve are important questions to ask. And we should remember that as inequality in disposable income increases, the divergence between efficiency and welfare will increase too, to the detriment of those on the lower end. — Oren
THE ART OF NO DEAL
Ahead of this week’s two-day summit in Beijing, China hawks, like us, were concerned that President Donald Trump might strike a trade and investment deal with President Xi Jinping that could reverse the trend toward decoupling he put in motion during his first term and has accelerated with his second-trade policy. What emerged, based on the initial reporting, was mercifully much less.
China has committed to or affirmed the purchase of American soybeans, beef, and Boeing aircraft. U.S. authorization for Nvidia to sell H200 chips to ten Chinese firms was the headline concession on paper, but that policy had been announced months ago and, in practice, the question remains whether China will accept the chips. A new U.S.-China Board of Trade will reportedly explore tariff cuts on “non-sensitive goods,” a potentially positive arrangement if it amounts to managed exchange in the product categories where two-way trade is genuinely benign. And that’s about it.
A nothingburger summit is, in this case, a good outcome, for the reason Oren laid out in the Financial Times on the summit’s first day: “The best plausible result is simply no deal, beyond efforts to smooth any decoupling that might proceed, giving the US time to come to its own terms with the irreconcilable differences of its great-power competitor.”
That said, the H200 chip authorization is a bad outcome on its own terms, whether or not Chinese firms ultimately take delivery. As American Compass argued last fall in Stop Selling the Rope, advanced AI chip design and manufacturing is one of the few critical sectors where the United States and its allies still hold a decisive lead. The H200 is not the most advanced chip Nvidia produces but it’s close, and exporting non-leading-edge chips still extends China’s AI capabilities by months, if not years. The so-called addiction thesis—that letting U.S. firms get a toehold in the Chinese market will lock China into dependence on U.S. technology—was wrong about China’s accession to the World Trade Organization at the turn of the century, and experience suggests it has only gotten more wrong since.
The American public agrees. New polling from American Compass and YouGov, released this week, finds that 91% of registered voters support restrictions on AI chip exports to China, with majorities in both parties favoring an outright ban. Americans are also three times more likely to view China as a rival or adversary than as an ally or potential partner, and reject by a more than two-to-one margin the idea that the United States can do business with China while steering clear of its military. Read the full report: Americans Reject Chip Exports to China.
Elsewhere in the U.S.-China economic relationship:
In Bloomberg, read about how the Pentagon’s ‘Deal Team Six’ Aims to Challenge China’s Grip on Rare Earth Power: “From an office a few blocks from the White House, a group of former Wall Streeters is at the forefront of the Pentagon’s plan to crack China’s critical-minerals stranglehold…It’s racing to put together creative deals with billions of dollars in equity stakes, long-term price floors, purchase commitments, loans and other financial tools.” And, in the same outlet, on the scale of the hole decades of failed policymaking have dug: US Needs Another Decade to Fix $1.2 Trillion Rare Earth Crisis.
Efforts to “hermetically seal” the U.S. market against the threat posed by the Chinese auto industry’s excess capacity—and the data and security risks that would come with Chinese-built cars on American roads—are gaining traction in Congress. Rep. John Moolenaar (R-MI), chairman of the House Select Committee on the CCP, and Rep. Debbie Dingell (D-MI) introduced the Connected Vehicle Security Act, legislation that would codify and expand existing restrictions by prohibiting the importation, manufacture, and sale of connected vehicles (along with the software and hardware that power them) linked to China. It’s the House companion to the bill introduced last month by Sens. Bernie Moreno (R-OH) and Elissa Slotkin (D-MI), and is backed by the United Auto Workers and General Motors. This week, Sen. Eric Schmitt (R-MO) joined as a cosponsor, and the Teamsters became the second union to issue an endorsement.
Two bonus links from the Wall Street Journal:
China’s Cars Aren’t in the U.S., But Its Auto Parts Are Everywhere
China Exports More EVs Than Traditional Cars for First Time in April
ALSO ON CAPITOL HILL
It looks increasingly likely that the Republican-controlled House will, at some point during the next couple of months, vote on the bipartisan, bicameral, labor-endorsed Faster Labor Contracts Act. The legislation would impose binding mediation and arbitration timelines to ensure that workers who vote to unionize actually secure a first contract. Under current law, roughly a third never do.
Last month, Rep. Donald Norcross (D-NJ) filed a discharge petition for the underlying bill, a measure that could force a floor vote if it receives signatures from 218 members of the House. As of this writing, the petition has 214 signatures, and is supported by four of the 17 Republicans who have cosponsored the bill, including Reps. Mike Lawler (NY), Max Miller (OH), Rob Bresnahan (PA), and Brian Fitzpatrick (PA). Four more signatures and Speaker Mike Johnson will be forced to schedule a vote. You can expect the business lobby— which has spent decades making sure those workers don’t get a first contract—to mobilize in full force, urging the other 13 Republicans not to sign the petition. For why they should, read Daniel’s article in Compact back in September: Can Republicans Help Fix Labor Law?
Over in the upper chamber: Sen. Bill Cassidy, chairman of the Senate Health, Education, Labor, and Pensions (HELP) Committee, is on the ballot this Saturday, facing off against President Trump-endorsed Rep. Julia Letlow and others in Louisiana’s Senate primary. If Cassidy loses, his replacement as the HELP Committee’s top Republican is an open question. Bloomberg reported this week on one possibility: “Though Republican Sens. Rand Paul (Ky.), Susan Collins (Maine), and Lisa Murkowski (Alaska) outrank him, they hold other premium committee chairs, so Sen. Roger Marshall (Kan.), who is close to the president and founded the Senate’s MAHA caucus, could be next in line.” Notably, Marshall is a Senate cosponsor of the Faster Labor Contracts Act…
In Family Policy news: Rep. David Valadao (R-CA) introduced the Supporting Newborn Parents Act, a new tax credit, separate from the child tax credit, that provides up to $2,000 for families welcoming a newborn child and functions as a de facto baby bonus. Read Patrick Brown’s advocacy of a baby bonus at American Compass back in 2024, and read Leah Sargeant over at the Niskanen Center with analysis of the proposed legislation.
THE GOOD, THE BAD, AND THE DYSTOPIAN
Good: U.S. Drug Overdose Deaths Fall for Third Straight Year (Wall Street Journal). “The number of people who died from drug overdoses dropped again in 2025, a promising trend as the U.S. emerges from a national fentanyl crisis that accelerated these fatalities….Drug-overdose deaths have now declined for three consecutive years, falling to levels closer to those not seen since before the Covid-19 pandemic, which intensified the drug-overdose crisis.”
Bad: ‘A’ Grades Are Suddenly Everywhere Since the Arrival of ChatGPT (Wall Street Journal): “The share of A’s in college classes heavy on writing and coding—in other words, work more prone to artificial intelligence use—has grown more significantly than in other classes since ChatGPT’s debut, according to a paper from the University of California, Berkeley, released Wednesday. Professors teaching AI-exposed classes gave out about 30% more A’s and fewer A-minus and B-plus grades.” Not what we hand in mind by technology improving student outcomes…
No, the real effect of technology on students can be found in this New York Times report, Why U.S. Test Scores Are in a ‘Generation-Long Decline’: “Almost everywhere in America, students are performing worse than their peers were 10 years ago…Compared with a decade earlier, reading scores were down last year in 83 percent of school districts where data was available. Math scores were down in 70 percent.”
Ugly: In Kalshi’s quest to monetize “any difference in opinion,” retail bettors on the prediction market platform reportedly lost more than $100 million on parlays in the first four months of this year. Defenders of prediction markets argue that the market price reveals truth more honestly than polls or punditry, because every trader has put money behind their view. Whatever the merits of that argument for binary contracts on elections or Federal Reserve rate-cut decisions, it is hard to see what truth is being discovered in a six-leg parlay with a 0.3% payout probability. A platform for prediction market parlays is a casino with a Hayek quote on the wall—not a source of information discovery. And as in any casino, the house always wins.
FINALLY, SOME GOOD READS FOR YOUR WEEKEND
The Nobel-Winning Psychologist Who Believed He Found the Secret to Happiness (New York Times): “Maximizers tend to be less satisfied with their decisions and their lives. They are typically less happy, more prone to regret and more likely to compare themselves endlessly with others. Satisficers don’t necessarily have low standards. Their standard is “good enough for me” rather than “the best out there,” and that makes it possible to feel satisfied with their choices, instead of haunted by the ones they didn’t make. This is critical today because chronic maximizing has never been easier…consumer options available to citizens of modern economies exceed those of preindustrial societies roughly by a factor of 100 million. That is an almost incomprehensible multiplication of choice, and it extends well beyond consumer goods into questions of who to be, how to live, where to work and whom to love.”
The Feed Is Fake (Vulture): That “viral” song, movie, meme, influencer, and celebrity drama was probably the product of a stealth marketing campaign.
Enjoy the weekend!
 | A guest post by
|
Price signals are working just fine for gas. What you're skipping is that the more money you have, the more choice you have, in deciding that you would rather cut spending on. Trying to claim something a simple as "the poor" is in itself disingenuous.
The poor include every thing from college students who can easily decide they would rather ride their bikes, to chuckle heads that should have stuck with public transit or bought a more fuel efficient car in the first place.
Does that include some people that made the best decisions they could and are now hurting because the temporary spike in fuel prices. Sure, but for the rest of the planet, higher fuel prices are the price signal that makes them change their behavior. For some that will mean they keep driving but decide to hold off and keep using something they own rather then purchase a new something. Which is yet another way to reduce fuel consumption.
The problem you quickly get into when trying to subsidize things like fuel usage is that for the majority of the middle and lower class people, that only encourages wasteful use of fuel.
I don't know if Oren or Daniel wrote the gas price thing, but it's about the dumbest article I've ever seen here. Normally, this place is pretty solid, but not today.
What it's describing is the disjunction between "marginal benefit" and "willingness to pay". In introductory econ textbooks, we (yes, I teach HS econ, among other things) we quickly gloss over this disjunction because it's very hard to explain without heavy math, but it's among the most well understood problems and one of the reason no one uses supply-demand graphs outside of Econ 1A.
This line of reasoning is so dangerous because the core problem here is what's called the "declining marginal utility of wealth" -- the more money you have the less you value the last $100 you earned. Once you start attacking prices on this basis though, it's very easy to end up deciding that the entire price system is hopelessly flawed and clearly there ought to be a better way. Which is essentially what Marx and Engels said. And it's a disaster.
In the case of a life-critical commodity (food, housing, energy) vouchers are a simple and very well understood way to provide relief for the poor without screwing up the price system any more than necessary.