DeLong's Grasping Reality: Economy in the 2000s & Before: Macro Outlook

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

##contango-backwardation-and-the-current-freaking-out-of-the-oil-market
##macro-outlook
#backwardation
#contango
#oil-markets
#futures-curves
#commodity-prices
#convenience-yield
#cost-of-carry
#real-interest-rates
#intertemporal-prices
#oil-shortage
#financial-jargon
#city-of-london
#term-structure

ORANGE MAN, BAD MACRO: Trump’s War on Powell & Economic-Policy Reality

Brad DeLong — Mon, 12 Jan 2026 15:10:30 GMT

Bessent, Lutnick, Hassett, Yared, & company should all have resigned last night, if they want anybody ever to regard any of them as a man. Just saying…

The most important thing for you to focus on right now is: ORANGE MAN VERY BAD.

The Trump-Noem ICE is populated by murderous fascist thugs, people whose brutality arises because, as Noah Smith puts it, of “unprofessionalism and low recruitment standards” plus that they have been brainwashed. As he goes on to say:

Noah Smith: <https://twitter.com/Noahpinion/status/2010177162960367693>: ‘ICE is recruiting people using the Great Replacement ideology, so it's getting agents who think they're in an EXISTENTIAL RACE WAR…

And they are starting to wage that war against one-third of America: the foreign-born, with or without valid H1-B and other visas and green cards; citizen children of the foreign-born; other U.S. citizens who look funny to them; and liberals.

But, shift gears. Also we have:

Right now the ORANGE MAN—President Donald Trump—is trying to put Federal Reserve Chair Jay Powell in jail. Who is Jay Powell? He is the man who Donald Trump chose eight years ago as the best human in the entire world to do the job—the very best out of all eight billion of us. Remember: Trump could have nominated anyone, and the Republican senators would have confirmed his nomination. Trump could have nominated his favorite horse Incitatus. In fact, he could not have done that. But the blockage is not because the Republican senators would have nixed such a nomination nomination. It is because Trump is such a total misanthropic narcissistic alexithymic that he is incapable of having a favorite horse.

He chose Powell. Now he has regrets. And he wants to put Powell in jail:

Jay Powell: Statement by Federal Reserve Chair Jerome H. Powell <https://www.youtube.com/watch?v=KckGHaBLSn4>: ‘Good evening. On Friday, the Department of Justice served the Federal Reserve with grand jury subpoenas threatening a criminal indictment related to my testimony before the Senate Banking Committee last June. That testimony, concerned, in part, a multi-year project to renovate historic Federal Reserve office buildings.
“I have deep respect for the rule of law and for accountability in our democracy. No one, certainly not the chair of the Federal Reserve is above the law. But this unprecedented action should be seen in the broader context of the administration’s threats and ongoing pressure. This new threat is not about my testimony last June or about the renovation of the Federal Reserve buildings. It is not about Congress’s oversight role. The Fed, through testimony and other public disclosures, made every effort to keep Congress informed about the renovation project. Those are pretexts.
The threat of criminal charges is a consequence of the Federal Reserve setting interest rates, based on our best assessment of what will serve the public, rather than following the preferences of the President. This is about whether the Fed will be able to continue to set interest rates based on evidence and economic conditions, or whether instead, monetary policy will be directed by political pressure or intimidation.” ⁠
I have served at the Federal Reserve under four administrations, Republicans and Democrats alike. In every case, I have carried out my duties without political fear or favor, focused solely on our mandate of price stability and maximum employment. Public service sometimes requires standing firm in the face of threats. I will continue to do the job the Senate confirmed me to do with integrity and a commitment to serving the American people.” ⁠
Thank you.

<https://www.project-syndicate.org/commentary/trump-tariffs-threats-incentive-for-canada-mexico-and-everyone-else-to-break-free-by-j-bradford-delong-2025-03>

Donald Trump says right now that he does not like Jay Powell’s Republican-grandee relatively hard-money orientation toward Fed policy?

This is the bed he made.

He had two very well-qualified Fed officials who did not share that orientation to hand when he chose Powell. He had two who were better equipped, in my mind at least, to do the job: then-Fed Chair Janet Yellen and then-Fed Governor Lael Brainard. He rejected both of them. For Powell.

And now he is trying to put Powell in jail.

Why?

For doing his job as he sees it. And, under any cool and rational assessment, doing a very good job as well.

The monetary policy of Powell and his committee has done a very large number of very good things, successfully:

kept the economy growing smoothly and efficiently through the majority of Donald Trump’s first term, until it was knocked off the rails by the sudden emergence of the COVID-19 plague.
kept the flow of spending through the economy during the plague, so that the collapse of employment did not produce a deep and unneeded depression, but that the recession elements of 2020 were those imposed on us by nature and by our desire not to repeat worldwide the public-health disasters of Lombardy in March and New York in April of 2020.
put the pedal to the metal after the creation of the lifesaving mRNA vaccines to get the U.S. and the global economy very quickly back to full employment, in striking contrast to the extraordinarily long lost-half-decade slog of delayed recovery from the GFC recession of 2007-2009 presided over by Bernanke and Obama. (Yes, the Obama people say none of that was Obama’s fault; I disagree. Powell-Biden did substantially better given the table set for them. Obama and his people were not the major bad actors—those were the Republican austerians, the counterparts in the U.S. of the Cameron, Osborne, and Clegg who committed the first half of the relative wrecking of the British economy that has taken place in the past fifteen years.)
and thus prudently guarded against a return to the painful global macroeconomic environment of the safe-asset shortage that pushed short-term safe interest rates down to near-zero worldwide, and made anything close to full-employment unattainable without giving lots of free money away to businesses undertaking stupid investments.
created enough flexibility in relative prices during the post-plague recovery that the economy wheeled into a near-optimal configuration in which people did not just get back into jobs but got back into highly productive jobs—many fewer in-store salespeople, many more delivery drivers, more goods-producers—given the societal learning about commerce in the information age forced by the plague year.
managed to keep an economy afflicted by disastrous random chaos-monkey Trumpist policies from either falling into recession or seeing a renewed jump-up in inflation during a very difficult economic-policy management year indeed.
wound up blessing an extra jump-up in the price level (relative to the 2.5%/year desired CPI-basis trend that is the Federal Reserve’s target) of 11%. (But, in my view, at least half of that 11% is the doing of Vladimir Putin; not of Jay Powell and his committee failing to find a utopian path in which all good things come together.

Yes, people are very unhappy about (7). But the way the cards were dealt, the non-Putin part of (7) was a necessary concomitant of achieving (5), (4), and (3).

And people would be far more unhappy if we had avoided 5%/points of post-2020 cumulative inflation at the cost of having had another lost-half decade with incomplete recovery and interest rates still at the zero lower bound. You leave rubber on the road when you rejoin the highway at speed in order to avoid being rear-ended.

The rational thing to do is to accept that, rather than to carry on about how the driver should have figured out a way to rejoin the highway at speed without leaving any rubber on the road.

And, of course, I have written about all of this before:

Rescuing America’s Economy from Trump

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

##orange-man-bad-macro
##orange-man-very-bad
##trumps-war-on-powell-economic-policy-reality
##macro-outlook
##neofascism
#jay-powell
#fed-independence
#soft-landing
#inflation
#price-level-step-up
#safe-asset-shortage
#kayfabe-politics
#full-employment-recovery
#powell-biden-record
#janet-yellen
#lael-brainard
#fed-credibility

Lessons for Debt Control from Clinton's Success in the 1990s

Brad DeLong — Thu, 01 Jan 2026 19:52:43 GMT

Time to fly my left-neoliberal freak flag! For a failure to get the history right may well lead us to inaccurate conclusions about what our government-debt outlook really is, & mistake how resulting economic & political-economic problems should be dealt with…

Credible fiscal anchors crowd in private investment when the central bank leans against demand shortfalls and technology makes capital cheaper. Clinton’s OBRA 93 deficit-reduction Reconciliation package accelerated the economy, because macro reality beats tribal signaling. Fiscal credibility, pro-work redistribution, and a supportive Fed plus falling ICT prices crowded in investment, the 1994 yield-curve shift move reflected growth strength and MBS mechanics, not fears of “austerity gone wrong.” The Clinton 1990s really were a fabulous decade that delivered rising employment, low inflation, and real wage gains, with the EITC expansion the single biggest pro–working-poor social-insurance expansion. Labeling this “austerity” misses how structure + demand + technology produced more capital, higher productivity, and a richer America.

And taking claims that Clinton’s OBRA 93 would tank the economy as real fears by credible macroeconomic analysts is to mistake political bullshit for real analytical judgements and warranted fears.

I write because I think the very sharp Marcus Nunes gets this one wrong here.

He is reviewing the the fiscal adjustment that was the Clinton 1993 Reconciliation deficit-reduction bill—OBRA 93 <https://en.wikipedia.org/wiki/Omnibus_Budget_Reconciliation_Act_of_1993>

It crowded-in a truly extraordinary boost to investment in America, which was further amplified by the secular fall in the relative price of information-communications capital goods that was the internet boom. And American economic growth was, thereafter, stronger by perhaps 0.5%-points per year. Figure that America today is 15% richer because of Bill Clinton and those of us who worked for and supported him.

OBRA 93 was viewed by us left-neoliberals—us Rubin Democrats—who pushed this in 1993 as very much a second installment of what had been OBRA 90: the George H.W. Bush-Mitchell-Foley deficit-reduction package of three years before. OBRA 93 did have more tax increases and fewer spending cuts in the mix, but by a narrow margin. OBRA 93 was more progressive than OBRA 90, but again by a narrow margin. But both combined revenue increases with spending restraint rather than relying on one side of the ledger alone, both raised top‑bracket income tax liabilities and closed loopholes/preferences to broaden the base and increase progressivity, both tightened discretionary spending caps and enforced them with sequestration/PayGo‑style budget rules to deter backsliding, both protected core social insurance pillars while trimming growth rates in selected programs rather than cutting benefits outright, both sought business‑confidence effects by committing to predictable multi‑year deficit paths to “crowd‑in” private investment, and both were sold as responsible long‑run policy over short‑run optics.

Indeed, the theses of us left-neoliberals—us Rubin Democrats—who pushed this, and convinced Clinton to throw his weight and his administration 100% behind it, was that:

it would indeed crowd-in investment and boost economic growth,
we had a commitment from Alan Greenspan at the Federal Reserve that he would do his damnedest to adjust monetary policy so that recovery from the 1990-1991 recession was not interrupted by any shortfall in aggregate demand,
neglecting the urgent need for deficit reduction might well wind us with much higher interest-rate risk premiums that would disrupt economic recovery,
because there were signs in financial- and exchange-market reactions to news that the U.S. debt was rising high enough to endanger the dollar’s safe-haven status,
we would get votes from sensible deficit-hawk Republicans and so it would become a “bring us together” bipartisan initiative,
success at boosting the pace of American economic growth would mean that, after a 1990s of the first rapid real wage increases in a generation, those who had become Reagan Democrats in the 1980s would be much more willing to support equity policies as they would no longer feel under as much family financial stress.

(6) was 100% wrong.

(5) was 200% wrong—it turned out that there were ZERO sensible deficit-hawk Republicans: none were willing to accept balanced deficit reduction hitting both spending cuts and tax increases, even though they would talk a good game in the abstract.

(4) was something I was 100% certain of at the time—and maybe it was true then that the safe-haven exorbitant-privilege debt capacity of the U.S. was then not that much more than 70% of annual GDP. But, if so, debt capacity rapidly and extraordinarily expanded once the 2007-8 GFC hit the world economy:

(3), thus, has to be put in doubt. It was an argument I very strongly believed back then. I recall a cold December 1992 night I spent carrying (a) simulation runs projecting interest rates implicit in the then-current yield curve under fiscal business-as-usual and (b) assessments of how the Bush 41 administration had begun to see “bad news” about the deficit not strengthen but weaken the dollar over to Bob Reich’s house, so he could carry them down to DC for Transition-Planning meetings. I still believe it was a risk. But I cannot believe it was an overwhelming risk.

However, (1) and (2) still look very very good indeed. And those by themselves are enough to place us, all of us who worked on and supported OBRA 93 public benefactors, among those whose names are written brilliantly and boldly in the Book of Life—especially as Gene Sperling managed to get and keep in OBRA 93 the EITC expansion that was then and is now the biggest pro-working poor expansion of the American social insurance system ever.

But we got zero Republican votes for OBRA 93. OBRA 90 had passed the House 227-203 (Democrats 217-40, Republicans 10-163, but with an unknown number of Republican “yeas'‘ in reserve if needed) and the Senate 54-46 (Democrats 44-10, Republicans 10-35). OBRA 93 passed the House 218-217, and the Senate 51-50, all Democrats both times.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

So here is Marcus:

Marcus Nunes: The Transition to Fiscal Dominance <https://marcusnunes.substack.com/p/the-transition-to-fiscal-dominance>: ‘In the 1990s… a rising debt ratio since 1980 was reduced… by Bill Clinton and the congressional Democrats]... through significant “austerity”, with both goverment spending falling and government revenues rising…. Monetary policy was appropriate, managing to keep NGDP on a stable level growth path, while inflation was low and stable. Throughout the adjustment unemployment was falling and reached 3.9% by the time Clinton left office…. However, there was significant economic debate—and widespread professional concern—that Clinton’s fiscal consolidation would damage growth. Many prominent economists predicted the 1993 deficit reduction package would cause recession or stall the recovery from the 90/91 recession. Just to give a few examples of ‘big names’ that were skeptical:
1. Republican Economists’ Consensus View: The Republican economic establishment was nearly unanimous: this would damage growth. Herbert Stein (Nixon/Ford CEA Chairman): Warned the tax increases would slow recovery. Martin Feldstein (Reagan CEA Chairman): Argued higher taxes would reduce investment and employment. Predicted the package would “significantly reduce economic growth.” Robert Barro (Harvard): Concerned about growth effects of higher marginal tax rates on labor supply and investment. Michael Boskin (Bush CEA Chairman): Predicted negative growth effects, particularly from top rate increases affecting entrepreneurs and small business.
2. Wall Street Consensus: Major investment banks and forecasters predicted slower growth: Goldman Sachs economists initially forecast the deficit reduction would subtract ~0.5 percentage points from GDP growth. Many Wall Street economists worried the fiscal tightening would abort the fragile early-1990s recovery. Bond market initially sold off on fears that fiscal consolidation would hurt growth, reducing tax revenues and making deficit reduction self-defeating.
3. Political Predictions: Every single Republican in Congress voted against the package, many citing economic harm: - Senate: 0 Republican votes (50-50, VP Gore broke tie). - House: 0 Republican votes (218-216). Newt Gingrich predicted: “The tax increase will kill the recovery... This is the Democrat machine’s recession, and each one of them will be held personally accountable.” Dick Armey (House Majority Leader): “The impact on job creation is going to be devastating.” Phil Gramm (Senator): “I believe hundreds of thousands of people are going to lose their jobs... I believe Bill Clinton will be one of those people.” (In 1996, Clinton was reelected!)…

(Parenthetically, not mentioned by Marcus: John Kasich of Ohio, back on July 28, 1993: “This plan will not work. If it was to work, then I’d have to become a Democrat and believe that more taxes and bigger government is the answer…” <https://crywolfproject.org/quotes/quote-%E2%80%93-rep-john-kasich-r-oh-cnn-1>. John Kasich lied. He never became a Democrat.)

Now let me pick my bones:

First, damned if I know why Marcus writes “Bond market initially sold off on fears that fiscal consolidation would hurt growth, reducing tax revenues and making deficit reduction self-defeating.” Look at the 10-Year Treasury:

Damned if I can see any bond-market selloff as George H.W. Bush went from clear favorite to win reëlection to loser, and as Clinton shifted to his left-neoliberal deficit-reducer incarnation, as OBRA 93 moved through the congress with only Democratic votes and with only one-vote victories in either chamber, and as it then began to take effect.

Zooming in, adding in the dates that Alan Greenspan raised the Federal Funds rate:

The bond-market selloff came starting only half a year after the passage of OBRA 93, and was a reaction to two things: (A) The first was the extraordinary strength of the U.S. economy, as high-tech emerged as a leading sector and as investment in information and communications technology roared ahead greatly in excess of our expectations. (B) The second was the endogenous duration of mortgage-backed securities, which were then a new thing: as interest rates rose, people stopped refinancing mortgages, holders of MBS found themselves holding assets of much longer duration than they had counted on, and so they dumped long-duration Treasuries into the market; the consequences was that instead of a 1-to-4 gearing of 10-Year to 3-Month interest rate increases, we were surprised by a 3-to-4 gearing.

Both (A) and (B) struck us by surprise, and they made my life working for the Treasury in 1994 extremely interesting—very stressful—and never dull.

But the bond-market selloff did NOT, REPEAT NOT, reflect any fear that “fiscal consolidation would hurt growth, reducing tax revenues and making deficit reduction self-defeating”. The deficit was, by then, falling much faster than according to the benchmarks our initial 1993 forecasts had set.

And the Wall Street economist fears—well, Greenspan was 100% on board with deficit reduction and OBRA 93, and so we interpreted those worries (and the Goldman-Sachs forecasts) as coming from people who were not so much making forecasts of the consequences of a combination of fiscal austerity and monetary ease, but rather people making noises to put themselves into ideological alignment with their largely-Republican client base, by parroting what the Republican politicians were saying.

And the Republican politicians? The Doles had been powerful drivers and advocates of the Bush 41 deficit-reduction package, OBRA 90, three years before OBRA 93. Had OBRA 93 been proposed under a second-term Bush 41 presidency, they would have been strong advocates as well. It was, in their view, good policy. But because the person at the head of the government was not Republican George H.W. Bush 41 but Democrat Bill Clinton 42, root-and-branch opposition to it was good politics. As for the Gramms, the Gingriches, and the Armeys, they were not making forecasts but rather one-way bets: if the economy went into recession for any reason, their predictions that OBRA 93 would be useful; if the economy did not, they knew that the supine press corps would never hold them to account. And it did not.

And now we come to the Republican economists. Were Stein, Feldstein, Barro, and Boskin serious in their fears that OBRA 93 would damage economic growth? Barro, yes. But Stein and Feldstein had been big OBRA 90 boosters, and Boskin had been an OBRA 90 designers. By far the most significant difference between OBRA 90 and OBRA 93 was the partisan identity of the President who would sign it into law. Now I suppose it is possible that Stein, Feldstein, Boskin thought that OBRA 90 was bad policy, and only went along with it because they were team players—professional Republicans. It could be. I never asked any of them. But my hunch is that it is overwhelmingly more likely that the polarity is reversed: that it was their opposition to OBRA 93 rather than their support of OBRA 90 that was subservience to their political masters.

Thus I think Marcus has it more-or-less completely wrong when he says that “there was significant economic debate—and widespread professional concern—that Clinton’s fiscal consolidation would damage growth. Many prominent economists predicted the 1993 deficit reduction package would cause recession or stall the recovery from the 90/91 recession…” The “professional” concern was—Barro aside (and Barro, recall, is the person unhinged enough to claim that the Trump-Ryan-McConnell tax cut of 2017 would raise investment in America by as much as it increased from 1993 to 2000 and increase America’s steady-state capital-output ratio by 40%) a professional Republican, not a professional economist concern.

There had been, recall, no significant economic debate over OBRA 90.

But the rest of what Marcus has to say about fears that we may be undergoing a transition to “fiscal dominance” is good!

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

##left-neoliberal-freak-flag
#macro-outlook
#clinton-obra93
#crowding-in
#eitc-expansion
#yield-curve-1994
#growth-engine
#policy-mix
#rubin-democrats
#business-confidence
#paygo-caps
#historical-context

HOISTED FROM THE ARCHIVES: I Don't Fully Buy Stiglitz's Argument That Our Macro Problems Have Deep Structural Roots. But I Do See Its Coherence

Brad DeLong — Tue, 09 Dec 2025 22:21:33 GMT

From 2011-12-16: there is demand for this, so here it is back again... <https://web.archive.org/web/20111219144609/ttps://delong.typepad.com/sdj/2011/12/i-dont-fully-buy-stiglitzs-argument-that-our-macro-problems-have-deep-structural-roots-but-i-do-see-its-coherence.html>

I Don’t Fully Buy Stiglitz’s Argument That Our Macro Problems Have Deep Structural Roots. But I Do See Its Coherence

Truth is, I don’t understand all the hating on the Vanity Fair Stiglitz piece, “A Banking System is Supposed to Serve Society, Not the Other Way Around”.

Perhaps I am giving Joe too charitable a reading. But I really do not see a serious analytical problem. Not that I buy the argument--I don’t think that our macro problems today have deep important structural roots in the decline of manufacturing due to its rapid productivity growth, and I don’t think that a massive governmental borrow-and-spend program is the only way out (but Stiglitz may be right in his claim that it would be the best way out).

However, even though I do not fully buy it I do think I understand the argument. And I do not think it is incoherent.

As I understand the Greenwald-Stiglitz hypothesis--about the Great Depression as applied to agriculture and about today as applied to manufacturing--it goes like this:

Rapid technological progress in a very large economic sector (agriculture then, manufacturing now) leads to oversupply and steep declines in the sector’s prices. Poorer producers have less income. They come under pressure to cut back their spending. Others--consumers--are now richer because they are paying less for their food (or their manufactures), but their propensity to spend is lower than that of the stressed farmers or ex-manufacturing workers.
Moreover, the oversupply of agricultural commodities (or manufactured goods) means that only an idiot would invest at their normal pace in those sectors. To the shortfall in consumption spending is added a shortfall in investment spending as well.
Thus we have systematic pressures pushing spending down below economy-wide income. These aren’t going to go away until the declining sector (agriculture then, manufacturing now) is no longer large enough to be macroeconomically significant.
Macroeconomic balance requires that the economy generate offsetting pressures pushing spending up. What might they be?
For a while, those receiving the income that farmers (or ex-manufacturing workers) have lost and those who use to invest in the declining sectors can lend it to the farmers (or ex-manufacturing workers) so that they can keep up with the Joneses. But lending more and more to poorer and poorer debtors is, like lawn darts, only all fun-and-games until somebody loses an eye.
An alternative possibility is to switch investment away from the farm value-chain complex (or the manufacturing value-chain complex) to something else. But what? Nobody really knows. The future is uncertain. Other investments are clearly riskier then funneling money into the old channels of boosting the capital of the farm value-chain complex (or the manufacturing value-chain complex) had been. Given the extra risks, this pressure can only manifest itself if the cost of capital falls. But here we hit the zero lower bound on interest rates. And we are off to the secular liquidity-trap races. This won’t work either.

Now at this point I disagree with Greenwald-Stiglitz. I see three plausible ways to fix the unemployment-generating aggregate demand shortfall:

(a) This could be fixed by expectations of inflation that push you sustainable real cost of capital down below zero far enough that savings no longer exceeds planned investment at full employment. (But normal monetary ease it does not produce expectations of secular inflation won’t do anything useful.)
(b) This could be fixed by government loan- or bank-guarantee programs that transfer the risk of new and untried investments away from entrepreneurs and investors onto taxpayers, so that even without expected inflation planned investment at full employment no longer falls short of saving when the cost of capital is at the zero nominal lower bound. so that you don’t need a cost of capital less then be expected rate of deflation. It can be fixed by the government running up debt and buying stuff.
(c) This could be fixed by having the government borrow and spend on a large scale.

Stiglitz, however, doesn’t see either the “expected inflation” or the “loan- and bank-guarantee” roads as possible. Stiglitz’s conclusions:

Two conclusions…. The first is that the economy will not bounce back on its own, at least not in a time frame that matters to ordinary people. Yes, all those foreclosed homes will eventually find someone to live in them, or be torn down. Prices will at some point stabilize and even start to rise. Americans will also adjust to a lower standard of living—not just living within their means but living beneath their means as they struggle to pay off a mountain of debt. But the damage will be enormous. America’s conception of itself as a land of opportunity is already badly eroded. Unemployed young people are alienated. It will be harder and harder to get some large proportion of them onto a productive track. They will be scarred for life by what is happening today. Drive through the industrial river valleys of the Midwest or the small towns of the Plains or the factory hubs of the South, and you will see a picture of irreversible decay.
Monetary policy is not going to help us out of this mess…. [A]nyone who believes that monetary policy is going to resuscitate the economy will be sorely disappointed. That idea is a distraction, and a dangerous one.
What we need to do instead is embark on a massive investment program—as we did, virtually by accident, 80 years ago—that will increase our productivity for years to come, and will also increase employment now. This public investment, and the resultant restoration in G.D.P., increases the returns to private investment. Public investments could be directed at improving the quality of life and real productivity—unlike the private-sector investments in financial innovations, which turned out to be more akin to financial weapons of mass destruction.

I think that if I asked him Stiglitz would say that (a)--zero interest rates and expected inflation--would help if you could get there, but that you cannot get there through monetary policy. Only if people expect full employment will there be enough inflation to sustain a full employment equilibrium. And since they don’t expect full employment there isn’t enough expected inflation no matter how easy monetary policy is. I think he would say that (b) might work in the sense of restoring full employment in the short run, but it would be an unfair upward redistribution of wealth from taxpayers to financiers, and moreover would not help resolve the underlying structural problems that created the shortfall between savings and planned investment at full employment in the first place.

Stiglitz does say that (c) is best:

The private sector by itself won’t, and can’t, undertake structural transformation of the magnitude needed—even if the Fed were to keep interest rates at zero for years to come. The only way it will happen is through a government stimulus designed not to preserve the old economy but to focus instead on creating a new one… out of manufacturing and into services that people want—into productive activities that increase living standards, not those that increase risk and inequality…. Education…. [B]asic research. Government investment in earlier decades—for instance, to develop the Internet and biotechnology—helped fuel economic growth…. Meanwhile, the states could certainly use federal help in closing budget shortfalls…. [C]leaner and more efficient energy production…. [O]ur decaying infrastructure, from roads and railroads to levees and power plants, is a prime target for profitable investment.

And once we have both restored full employment and accelerated the structural transformation needed then we can return to normality. But in order to get there:

we must fix the financial system. As noted, the implosion of the financial sector may not have been the underlying cause of our current crisis—but it has made it worse, and it’s an obstacle to long-term recovery…. What’s needed is to get banks out of the dangerous business of speculating and back into the boring business of lending. But we have not fixed the financial system. Rather, we have poured money into the banks…. We have, in a phrase, confused ends with means. A banking system is supposed to serve society, not the other way around…

I’m not sure that I buy Stiglitz’s argument that massive government borrow-and-spend is the only, or even the best, way out of our current mess. And I agree that Stiglitz’s piece could have used a couple more paragraphs about life at the zero nominal interest rate lower bound explaining why Stiglitz thinks the belief “that monetary policy is going to resuscitate the economy will be sorely disappointed. That idea is a distraction, and a dangerous one”.

Perhaps I have seen so much really bad macroeconomics over the past four years that I now suffer from the soft bigotry of low expectations. But this does not seem to me to be, as Nick Rowe calls it:

really bad macroeconomics. God it’s depressing. If you want to talk about a deficiency of aggregate demand, and why Say’s Law sometimes fails, you really do need to talk about monetary exchange economies and an excess demand for money at the aggregate level. You can’t just do partial equilibrium analysis and cobble it all together.

So then why does Nick have such an adverse reaction to Stiglitz?

I think it is because Stiglitz is at bottom a Wicksellian and Rowe is a Fisherian. A Wicksellian is a believer that the key equation in macro is the flow-of-funds equation S = I + (G-T), savings S equals planned investment I plus government borrowing (G-T), and that the money market exists to feed the flow-of-funds an interest rate that has a (limited) influence on planned investment I. A Fisherian is a believer that the key equation in macro is the money market’s quantity theory equation PY = MV(i), and that the flow-of-funds exists to feed the quantity theory an interest rate that has a (limited) influence on velocity V.

Thus they have a hard time communicating. From the Fisherian viewpoint, the Wicksellians are talking nonsense because they spend their time on things that have a minor impact on velocity while ignoring the obvious shortage of money. From the Wicksellian viewpoint, the Fisherians are talking nonsense by ignoring the obvious fact that movements in money induce offsetting effects in velocity unless they somehow alter the savings-investment balance.

And it is we Hicksians, of course, synthesize both positions into a single unified and coherent whole…

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#hoisted-from-the-archives
#macro-outlook
#macro
#great-depression
#john-hicks
#john-maynard-keynes
#I-dont-fully-buy-stiglitzs-argument-that-our-macro-problems-have-deep-structural-roots-but-i-do-see-its-coherence

CROSSPOST: NOAH SMITH: The AI Bust Scenario That No One Is Talking About

Brad DeLong — Tue, 09 Dec 2025 15:01:47 GMT

Think of airlines: very useful, very high-tech, immense user surplus, next to no profits. Even if AI works, and even if it gets adopted very fast, it might not make much in the way of profits for anyone. The part of Noah Smith’s post above his paywall...

Very much worth reading, because, I think, highly likely to be correct.

Noah sees three ways that the current AI-bubble and MAMLM-capex buildout could spectacularly crash:

(1) The Virtual Reality Scenario is if “AI” as we know it is just a useful enough technology to justify the capital-expenditure build-out, given that it is not write-once run-everywhere for-all-time, but rather train and infer, train and infer, train and infer again and again and again without mammoth economies of scale.
(2) The Railroad Scenario is if the financing seizes up as the profits do not appear in time before those who have been shorting the build-out and those who own the leveraged debt decide it is time to take their profits and cash-in their chips, respectively. The historical model is the Panic of 1873 that took down Jay Cooke & Co. In this case the sector and the technology are fine, but only after a very disruptive five-year final AI-Winter.
(3) The Airline Scenario is one of a very useful high-tech technology deployed at scale, but because markets are contestable and increasing returns largely absent, there are next to no profits as prices keep getting pushed down to short-run marginal cost.

This last is, I think, the most likely. And Noah thinks so too: The technology works. It diffuses. It becomes infrastructural. Yet the rents flow not to the operators, and in the long run not that many rents flow to the toolmakers and the suppliers. Instead, what rents there are flow to the downstream complementors. And the overwhelming bulk of value flows to customers in the form of user surplus. It is glorious (provided our AI can preserve our attention from being hacked to our detriment by Zuckerberg’s AI)! But it is not profitable for investors! And it is profitable for lucky entrepreneurs who keep their wits about them and understand that their trees will not grow to the sky. And it is profitable for only those VCs who understand that they are basically rerunning their crypto grift, and prioritize exit while the getting is still good and the crash has not yet come

There is one very important feature of the situation that, I think, Noah misses. It is a very powerful, and I think likely decisive, factor that pushes us toward the (3) Airline Scenario.

It is this:

Google and Facebook and Amazon do not want anybody—anybody—to take their search and social-media profits away from them by providing front-end natural-language interfaces to exploring the internet and to connecting with friends and entertainment and to shopping. Apple does not want anybody—anybody—to take its iPhone hardware sales profits away from it by providing a better cloud-based natural-language software interface to cheap Android smartphones than the iPhone’s software suite provides. Microsoft does not want anybody—anybody—to take its office-work and cloud-enterprise profits away from it by providing better natural-language interfaces to smithing words and numbers and bureaucratic paperflow.

All of these will do what Microsoft did to Netscape before they will let any of that happen.

If it ever looks like OpenAI or anyone else is attaining platform-aggregator scale by selling AI-based services, Google or Facebook or Amazon or Apple or Microsoft—or probably all five at once—will stop that from happening by giving good-enough substitutes away for free.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

The AI bust scenario that no one is talking about

Even if AI works, and even if it gets adopted very fast, it might not make profit.

Noah Smith

Dec 08, 2025∙ Paid

<https://www.noahpinion.blog/p/the-ai-bust-scenario-that-no-one>

Noahpinion

The AI bust scenario that no one is talking about

4 months ago · 230 likes · 33 comments · Noah Smith

I’ve actually already written a number of posts about the possibility of an AI bubble and bust. Back in August, I wondered if the financing of data centers with private credit could cause a financial crisis if there was a bust. I followed that up with a post about profitability, and suggested that the AI industry might be a lot more competitive than people expect. In October, I wrote about how AI is propping up the U.S. economy.

But I feel like I need to write another post, because almost all of the discussion I see about an AI bubble seems to leave out one crucial scenario.

Since I wrote those posts, popular belief that there’s an AI bubble and impending bust has only grown. A lot of prominent people in the industry are talking about it:

“Some parts of AI are probably in a bubble,” Google DeepMind CEO Demis Hassabis told Axios’ Mike Allen at the [Axios] AI+ Summit on Dec. 4. But, he added, “It’s not a binary.”…“I, more than anyone, believe that AI is the most transformative technology ever, so I think in the fullness of time, this is all going to be more than justified,” Hassabis said…“I think it would be a mistake to dismiss [AI] as snake oil,” OpenAI Chairman and Sierra co-founder Bret Taylor said at the AI+ Summit…Taylor acknowledged that there “probably is a bubble,” but said businesses, ideas and technologies endure even after bubbles pop. “There’s going to be a handful of companies that are truly generational,” Taylor said.

And:

Every company would be affected if the AI bubble were to burst, the head of Google’s parent firm Alphabet has told the BBC…Speaking exclusively to BBC News, Sundar Pichai said while the growth of artificial intelligence (AI) investment had been an “extraordinary moment”, there was some “irrationality” in the current AI boom.

The market is also starting to get skeptical. Here’s a chart from Bloomberg:

Source: Bloomberg

Almost everyone I read is basically talking about two scenarios for an AI bust. I call these the Virtual Reality Scenario and the Railroad Scenario. I’ll go over these, and then talk about the third scenario

The Virtual Reality Scenario

What I call the Virtual Reality Scenario is if AI, in its current form, turns out to just not be a very useful technology at all — or at least, not nearly useful enough to justify the amount of capital expenditures. This might happen because AI hallucinates too much, or because progress in AI comes to a halt. Bloomberg reports:

The data center spending spree is overshadowed by persistent skepticism about the payoff from AI technology…[R]esearchers at the Massachusetts Institute of Technology found that 95% of organizations saw zero return on their investment in AI initiatives…More recently, researchers at Harvard and Stanford [found that e]mployees are using AI to create “workslop,” which the researchers define as “AI generated work content that masquerades as good work, but lacks the substance to meaningfully advance a given task.”…
AI developers have also been confronting a different challenge. OpenAI, Anthropic and others have for years bet on the so-called scaling laws — the idea that more computing power, data and larger models will inevitably pave the way for greater leaps in the power of AI…Over the past year, however, these developers have experienced diminishing returns from their costly efforts to build more advanced AI.

This is basically what happened to VR technology. Meta spent $77 billion on developing the virtual reality “Metaverse”, but outside of gaming and some niche entertainment applications, nobody really wanted VR for anything, no matter how good the headsets got. Meta is now throwing in the towel and pivoting away from the Metaverse.¹

But I just don’t think this is going to happen to AI. Very few people use VR, even a decade after it started getting hyped. But AI is being adopted more rapidly than any technology in history. As far back as a year ago, 40% of people were already using AI at work:

Source: Bick et al. (2024)

Household adoption is similarly rapid.

Humans just know when a technology works. If AI weren’t useful, we’d see people trying it for a while and then setting it aside. But we don’t see that. Despite hallucinations and the other limitations of the technology, most people are finding some reason to keep using AI after they try it.

Sorry, haters. This tech is for real.

As for progress hitting a wall, I just don’t think that’s an important question anymore. The consensus in the industry ² seems to be that scaling up in terms of training AI models on more data has hit a wall, but that our ability to improve AI’s capabilities via inference scaling (basically, having it “think harder” before answering) is still going, and improvements in reinforcement learning and other algorithmic techniques are still coming.

But I don’t think this is the right question to be asking, because “better chatbots” are only one of many ways that AI can create value. The world of AI applications, including “agents” (AI that does stuff on its own) is still in its infancy. Andrej Karpathy is good at talking about this; I recommend his interview with Dwarkesh Patel.

Essentially, we haven’t even begun to build AI yet. Anthropic sort of has — it’s focused on AI business applications, and it’s making some money doing this. But most of the actual technology we’re going to create with AI is still in the future. And there’s a chance that AI as it exists today will be able to provide the foundation for a whole bunch of incredibly useful applications, even without any continued improvement in capabilities. The typical pattern is for most of the useful (and lucrative) applications to manifest decades after a new general-purpose technology is introduced.

The Railroad Scenario

Which brings us to the second scenario: the Railroad Scenario. Railroads were even more economically useful and lucrative than their biggest boosters in the 1860s imagined. But there was still a huge bust in 1873, because those economic benefits didn’t show up before railroad companies’ debt came due.

Here’s what I wrote about that back in October:

The railroad buildout in the 1800s was, in percentage terms, the greatest single feat of capital expenditure in U.S. history, dwarfing even what the AI industry is spending on data centers right now. In 1873, a bunch of railroad-related loans went bust, causing a banking crisis that threw the economy into a decade-long depression.
And yet despite the carnage in the railroad industry, the number of miles of railroad in the U.S. never stopped increasing — it just slowed down briefly and picked back up again!…In other words, the great railroad bust did not happen because America built too many railroads. America didn’t build too many railroads! What happened was that America financed its railroads faster than they could capture value…
There’s basically no technological limit on how many…loans [the financial system] can disburse in a short period of time…And yet there is a limit on how fast businesses can create real value. In order for a railroad to pay off, you need to build it, and then you need to find people to pay you to ship things on it. That takes time, especially because the full economic value of the railroad doesn’t manifest until new cities, new industries, and new supply chains that are enabled by the railroads get created.
Just to give one example, the famous Sears Catalog — which allowed people all across America to order products and have them delivered by railroad — eventually revolutionized American retail. But it didn’t even start to do that until 1888 — fifteen years after the big railroad crash of 1873…
Why is this important for AI? Because even if AI creates all the value its biggest boosters say it’ll do — supercharges growth, enables the automation of most kinds of production, and so on — it might not do it fast enough for the data center “hyperscalers” to pay back everything they borrowed. In that case, there will be a wave of defaults on bonds and loans.

We don’t know how likely this scenario is, because we don’t know how fast AI value creation will increase. But we can get some basic idea of the risk of this scenario by looking at the financing side.

If the companies building and operating the data centers (the main cost of AI) are spending less than they make, then everything is basically safe. Suppose you have a company that makes $50 billion in profit every year, and you spend $40 billion every year on data centers. Even if AI suffers a catastrophic crash and all your money is wasted, you’re safe; you just take a hit to your profit margins for a year, your stock price goes down, and then you just move on. This is true even if you borrowed the money to build the data centers; if things go bad you can afford to pay off the loans.

If you’re spending $70 billion a year things get dicey; you might have to take a couple of years of losses to pay back the loans if there’s a bust. And there’s some level of borrowing and spending where you’re actually in danger of bankruptcy. There’s no hard and fast rule for when a certain amount of spending becomes dangerous; it’s just sort of a sliding scale of worry.

Right now, much of the AI buildout is being done by big tech companies like Google, Microsoft, Amazon and Meta that make lots and lots of profit.³ Until recently, these “hyperscalers” have been making enough cash to cover their AI spending. But spending is rising, so that may not be true for much longer. If spending keeps increasing, some companies, like Amazon, may start to have to borrow against future cash flow soon. And Meta, which doesn’t have its own cloud business, and thus has to pay other companies to do its AI stuff, may be in greater danger.

Meanwhile there are a bunch of other companies that are investing a ton of money in AI that don’t make enough profit to fund it out of their own pockets, but which have borrowed a bunch of money to invest in AI. These include the big model-making companies — OpenAI, Anthropic, and xAI. They also include some cloud providers like CoreWeave that aren’t attached to a big profitable business. And they include various construction companies and service providers.

If AI takes 10 or more years to generate enough value to pay back all these debts, many of these companies could go bankrupt. Whatever financial institutions they’ve borrowed money from — private credit firms, banks, etc. — may fail or have to pull back significantly when their loans suddenly go bad. And that could touch off a financial crisis, even if AI continues to advance and companies continue to build data centers. That would be like what happened in 1873 with the railroads. AI itself would be fine, but the economy, the financial system, and a bunch of specific companies could be hurt.

This scenario seems at least fairly likely, given that this is basically what happened with both the railroads and the telecoms in past industrial booms. Currently, many observers are highly skeptical that AI companies will be able to earn enough revenue to pay back their debts by 2030, even under optimistic assumptions. As I said, the typical pattern is for the economy to take a long time to figure out how to use each new general-purpose technology, and the financial system may not be able to wait around.

But there’s also a third scenario, which relatively few people seem to be paying attention to. Even if AI works and manages to create value very quickly, that value may not be captured by the AI companies themselves. AI itself may turn out to be a commoditized, low-margin business, more like solar power…or airlines.

The third scenario: The Airline Scenario

The third scenario for an AI bust is that AI succeeds as a technology, but that the companies that make AI models — OpenAI, xAI, and so on — don’t manage to capture much of the value that AI creates…

1 You could also argue that something similar has partially happened to crypto. Crypto has certainly found plenty of application in the worlds of crime and corruption — ransomware, drug purchases, evading capital controls, and enabling illicit payments to and from corrupt politicians. But that’s all just regulatory arbitrage; in the world of above-board commerce, crypto hasn’t found any widespread application that couldn’t be handled by an Excel spreadsheet.

2 And at the risk of doing some real actual journalism, let me note that I do talk to a lot of people in the industry, and they agree with this.

3 Note that what people often talk about is not profit, but free cash flow. These tend to be pretty correlated.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#noah-smith
#ai-bubble
#mamlms
#the-ai-bust-scenario-that-no-one-is-talking-about
#macro-outlook
#subturingbradbot

The AI-Bubble's Most Likely Endgame Looks to Be Not Apocalypse, But an Awful Lot of Useful Compost

Brad DeLong — Fri, 05 Dec 2025 17:18:17 GMT

All the digging will not uncover a Golden ASI Pony. But the digging will spread an awful lot of very useful fertilizer around, in which very useful and valuable things will grow...

Proprietary LLMs are likely to prove lousy businesses—high opex, thin moats, fast commoditization. But their capex won’t vanish. And the value from it will diffuse broadly and equitably : better models for everyone; upgraded grids; repurposed GPU farms; millions of open repos. No Deep Thought, but better weather forecasting, antibiotics, copilots, and the replacement of bureaucratic cookie-cutter with bespoke algorithmic classification isn’t failure.

In short: AI is unlikely to mint new platform monopolies. It is likely to manure the next generation’s digital commons. The bubble finances infrastructure and code that, post‑panic, underpins broad gains.

That will be the case, unless, monkey’s paw-like, freely provided AI-service flows turn out to be very expensive indeed. We can build our own AI-tools to protect our attention from being harvested by the malignity of the Zuckerbergs. But if we do not, their AI tools will harvest our attention to our detriment.

This now looks to me like the most likely AI-Bubble endgame scenario:

FT Alphaville: What if OpenAI is worth more dead than alive? <https://ftav.substack.com/p/what-if-openai-is-worth-more-dead>: ‘Here’s the theory. Proprietary large-language models[’] core technologies still have value, but their developers’ individual business models marry persistently high operating costs with shallow commercial moats and their innovations quickly become commoditised…. The result is better machine learning for all. Companies will die, but the spoils of more than $1tn in capital expenditures can trickle down equitably to a nation’s underfunded labs, studios, factories and faculties. We won’t get Artificial General Intelligence, but we might get improved weather forecasting and some new antibiotics…. YOLO indeed. A common fear about AI is that the bubble’s bursting will metastasize into a financial crisis demanding of government interventions and value-destructive shotgun mergers — but what if it doesn’t? What if all it leaves behind are worthless share options, defaulted bonds, written-off GPU clusters, some slightly upgraded power grids, and several million open-source repositories full of fairly useful code? LLM commoditisation won’t give us Deep Thought, but can be a force for economic good in myriad small ways. Whereas China’s open-model approach seems to already accept this possibility, the Western version can still succeed if a handful of companies stay irrational for longer than they can stay solvent…

Share DeLong's Grasping Reality: Economy in the 2000s & Before

We also get very good on-device natural language interfaces to structured and unstructured databases, which are themselves very valuable. And the GPU clusters can do lots of very useful very high-dimension flexible-form big-data classification analyses, which will change and enrich the world in ways we do not yet know—provided we can keep the Zuckerbergs of the world from hacking our brains to distract and commodify our attention in ways that make us sick.

The most important thing to grasp here is that this is the most likely trajectory because Google, Facebook, and Amazon are not spending money like water on the AI-buildout to make money, but rather to defend their existing platform-monopoly profit flows.

There are lots of others, overlapping sets of them, in the businesses: Microsoft, Open-AI, Anthropic, ByteDance, Baidu, Alibaba, TenCent, Perplexity, MidJourney, StabilityAI, and a host of others—Character.ai, Runway, ElevenLabs, Grammarly, Notion, Adobe Firefly, Atlassian, Palantir, Snowflake, Cohere, come immediately to mind—are hoping to become the natural-language and copilot platform monopolist; Microsoft, Oracle, Google, Amazon, and Salesforce are also hoping to make and are making lots of money selling digital pick-and-shovel services to the Bubbleteers; NVIDIA, BroadCom, and TSMC are making lots of money by making the digital picks-and-shovels; Tesla is off high on ketamine, meth, and booze, staggering around raving at clouds; Apple is hoping to have good-enough on-device models that, with their privacy commitment, they can keep the Android smartphone wolf at the door without having to lose a fortune buying NVIDIA chips; and that is the lineup.

But all of these are maneuvering in the environment in which Google, Facebook, and Amazon are going to spend whatever it takes from anybody else’s being able to make enough money via copilots and natural-language interfaces to then take a run at taking the search-aggregation monopoly, the social-network-aggregation monopoly, or the shopping-aggregation monopoly. And, behind them, Apple stands ready to flick the switch if it decides that privacy plus on-device is not enough of a moat to protect its smartphone sales profits and that it too needs to burn money like water in data centers.

There is no more a path to Open-AI or Anthropic or ByteDance becoming another highly profitable platform monopolist than there was a path in the late 1990s for Netscape to take Microsoft’s Windows and Office profits away from it by turning the browser into the operating system and turning Windows into a collection of poorly debugged device drivers. Indeed, the browser became half of the operating system. But that half was Internet Explorer, provided for free and stapled to the continuing Windows enterprise profit flow.

That is the way things are going right now.

And yet, at the moment, the U.S. economy’s escape from recession relies on the YOLO continuing, as the flow of money willing to be burned to train and infer drives the data-center expansions that fund the massive profits of current data-center operators and, behind them, NVIDIA and the rest. But money burned to train and infer so that consumer and enterprise AI-services providers can then make money by charging customers for things that Google Gemini, FaceBook’s MetaAI, and Amazon’s Alexa-Q will provide them for essentially free.

Recognition that this is the case percolates—and it is percolating: it is Dario Amodei who says: “On the economic side, I have my concerns. Even if the technology fulfils all its promises… some players… are YOLOing… [because they] constitutionally want… to YOLO… or just like… big numbers… [and] may turn the dial too far…” This recognition is likely to start what Minsky and Kindleberger called “profit-taking”.

Then come panic, and crash.

But as long as the AI-buildout is overwhelmingly financed by equity and venture rather than by debt, panic and crash will indeed rhyme with 2000. It will indeed result in “better machine learning for all… [as] the spoils of more than $1tn in capital expenditures… trickle down equitably to a nation’s underfunded labs, studios, factories and faculties…”

Could I be wrong? Yes. But I really do not see how I could be far wrong.

There are, I think, things to watch for as this develops:

How Has U.S. Manufacturing Been Going Over the Past Generation?

Brad DeLong — Wed, 03 Dec 2025 23:35:41 GMT

Measure what users value in what they actually buy, rather than the numbers of boxes that leave the factory. Fix the deflators and follow things through the input-output tables summarizing the production networks and value chains, and the sector’s post‑1990 productivity story looks much brighter—and especially so in computers and electronics. But an even wider gap between those high-tech sectors and the rest of manufacturing emerges, and raises increasing puzzlement in my mind at least. Still, the slowdown is real, but it comes after the GFC, and the level attained is higher: claim of absolute “stagnation” appear to be mismeasurement…

We have a very nice take on trends in manufacturing sector producrtiviy from a very strong team here:

Enghin Atalay & al.: Why Is Manufacturing Productivity Growth So Low? <https://enghinatalay.github.io/manufacturing.pdf>: ‘Nearly all measured TFP growth since 1987—and its post-2000s decline—comes from a few computer-related industries. We argue conventional measures understate manufacturing productivity growth by failing to fully capture quality improvements… [with] mismeasurement in standard industry deflators. Using an input-output framework, we estimate that TFP growth is understated by 1.7 percentage points in durable manufacturing and 0.4 percentage points in nondurable manufacturing, with no mismeasurement in nonmanufacturing industries…
…

We have long known that nearly all measured manufacturing TFP growth since 1987 comes from a few computer-related industries. But are these measurements credible? Televisions illustrate the issue: from 1997 to 2023, consumer-facing PCE prices fell at a rate of 15.4% per year, vis-à-vis 3.0% in the producer-facing indices.

Atalay, Hortaçsu, Kimmel, and Syverson calculate that, since producer-facing deflators miss quality improvements, measurements are wrong. They calculate user-facing hedonic quality adjustments to the standard deflators, and then use their input-output framework to consistently calculate estimates of TFP consistent with the different hedonic-adjusted intermediate-good quantities generated. Concretely, they replace producer‑facing deflators (PPI/import indices used in gross output) with consumer‑facing deflators from the PCE/CPI, where those show larger quality‑adjusted price declines,notably in durables, and especially computers and electronics. They then push those price-index corrections through the input–output structure to adjust both output and intermediate input prices before recomputing TFP via the dual price identity.

Thus their procedures are onceptually broader than “hedonic adjustment” alone. They undertake a system-wide revaluation of deflators consistent with production networks. The input–output structure helps separate commodity prices from margins, and allocate corrected deflators to the right NAICS commodities.

Their calculated manufacturing sector-wide TFP diverges sharply upward from official measurements after 1997, reaching about 141 by 2023, vis-à-vis 117 reported by the government. Their calculations show TFP growth understated by 1.7 percentage points in durable goods and 0.4 points in nondurables (with effectively no mismeasurement outside manufacturing). Mismeasurement is largest in computers and electronics at 5.7%-points per year, with durable goods as a whole seeing productivity growth understated by 1.6%-points per year, and nondurables by 0.5%-points per year.

However, overall manufacturing still shows a substantial slowdown after the 2007-2009 GFC, but at higher levels: growth fell from 2.2% (1997–2009) to 0.6% (2009–2023). We still have recent stagnation in manufacturing excluding computers and electronics. The difference is that the stagnation begins after the 2007-2009 GFC rather than after 1990.

I think that this is an area that researchers should crowd into. I would very much like to see people chase down:

Why do computers and electronics dominate the TFP story?: Nearly all sector TFP since 1987—and the post-GFC decline—comes from the few computer-related industries. But why? There are lots of ways in which you can push manufacturing productivity forward other than figuring out how to make smaller devices in which you push electrons through doped silicon crystals. And all the technologists and engineers working in non-electronic manufacturing sectors did not vanish after 1990.
The ICT subindustries’ deceleration: Within NAICS 334, computers (3341) and semiconductors (3344) show the biggest post‑2009 slowdown: 15%-points per year vis-à-vis pre‑GFC for computers; and 11%-points per year vis-à-vis pre‑GFC for semiconductors. How much of this is simply Intel’s missteps? And where did the rest of the deceleration come from?
How robust are these calculations to different views of margins?: TFP in increasing-returns industries where firms have substantial market power is a very tricky beast to even think about. I would love to see more people think about it—and derive not a one- but rather a multi-dimensional index of technological capability and its interactions with both scale and with market power-utilizing profit harvesting.
The role of globalization in all this: The paper performs some robustness checks vis-à-vis different views of the value of globalization, the effects of transfer prices, and so on. But I would like to see more.
Policy and narrative implications: If indeed quality growth is thus undercounted, and the “manufacturing stagnation” narrative this overstated, then it is highly likely that investment and innovation spillovers are much larger than the numbers most of us have been keeping in our heads. Manufacturing then matters even more than ever, and is worth more policy attention. be larger than official data suggest.“Correcting for the under-counting of quality improvements implies that manufacturing productivity is still growing… Our results reshape… the desirability of public policies targeting the manufacturing sector.”
Source: Conclusion, manufacturing.pdf

And great kudos to the people who started this off and have pushed it forward before this. I am thinking of Byrne & Corrado (2015a) <https://www.federalreserve.gov/econres/feds/prices-for-communications-equipment-rewriting-the-record.htm>, Byrne & Corrado (2015b) <https://www.federalreserve.gov/econresdata/notes/feds-notes/2015/recent-trends-in-communications-equipment-pricing-20150929.htm>, and Byrne, Oliner & Sichel (2018) <http://www.roiw.org/2018/n3/8.pdf>, among others.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#how-has-us-manufacturing-been-going-over-the-past-generation
#total-factor-productivity
#hedonic-price-adjustments
#input-output-tables
#value-chains
#production-networks
#computers-and-electronics
#high-tech
#economic-growth
#macro-outlook

On the Faux-Puzzlement of Dan Davies

Brad DeLong — Thu, 30 Oct 2025 18:22:08 GMT

The Global Financial Crisis scarred global economies for a decade; COVID did not. The difference was cash, certainty, and a lender of last resort. Bagehot wrote the manual in 1873; Keynes added the addendum in 1936. Economists misplaced both. As Don Kohn said at the time, in dealing with a financial crisis: “Go fast, go hard, go soon”—and fix moral hazard later. Policy knew this once; it should again…

The Great Financial Crisis was not a destructive hurricane that laid waste to the fundamentals of wealth. It was, rather, a head injury to the economy’s coordination system. The muscles that drove the engine of wealth were still there, but useless for a long time. For when uncertainty about interbank obligations spiked, the system froze—and only a deus ex machina could restore certainty. That deus had a name—lender of last resort—but the profession second-guessed it. And still has not reinternalized it, as Dan Davies writes today:

Dan Davies: are we having fear yet? <https://backofmind.substack.com/p/are-we-having-fear-yet>: ‘The Great Financial Crisis [of 2008]…. The overall economic system… [afterwards] was no longer able to perform a vital function of balancing present consumption and investment for the future…. Economics ought to regard this as more of a puzzle than it does [that] the financial crisis, which caused practically zero physical destruction of productive capability, left scarring and after-effects for more than a decade, while the COVID-19 pandemic, which killed millions and left lots of buildings unusable, was all reversed within a couple of years, with a fairly trivial inflation by past standards as the worst economic consequence…

For, of course, he then gives the answer:

Dan Davies: are we having fear yet? <https://backofmind.substack.com/p/are-we-having-fear-yet>: ‘The actual reason… the Global[1] Financial Crisis had such bad long-term effects and the pandemic didn’t is that one… was accompanied by a deluge of public sector money and private sector loan forbearance… precisely because the previous experience had been so bad; flooding the zone with cash made sure that it wasn’t overwhelmed with the flashing red lights of “BORROWER CAN’T MEET CASH CALL”. It would be nice to know that this is now the standard operating practice for similar crises. But I don’t think we do know that, and this is worrying…

But the thing is: This was not a lesson that should have had to have been re-learned after 2008.

Why not? Because it was a lesson that had already been well-learned long before, by 1873, when Walter Bagehot published his Lombard Street. The message was settled and clear. In a financial crisis, to deal with it, you need a central back to:

lend freely,
at a penalty rate,
on collateral that is good in normal times.

So why did economics as a profession not know this 150 year-old wisdom? And why has it still not internalized it properly? That is a hard question to which I—still—do not have a very good answer. But let me endorse Dan Davies’s Call to Action. And let me set out some disorganized and largely repetitive musings.

Let me start by quoting Bagehot:

Walter Bagehot: Lombard Street <https://archive.org/details/lombardstreet00bage>; ‘These advances… should be made to obtain the object… to stay the panic…. These loans should only be made at a very high rate of interest… as a heavy fine…. [And] if it is known that the Bank of England is freely advancing on what in ordinary times is reckoned a good security… the alarm of the solvent merchants and bankers will be stayed. But if securities, really good and usually convertible, are refused by the Bank, the alarm will not abate… and the panic will become worse and worse…

Share DeLong's Grasping Reality: Economy in the 2000s & Before

To my mind, the only significant addition to the playbook since 1873 was John Maynard Keynes’s 1936 observation that should things get really bad, perhaps—probably—more will be required than just financial-monetary policy. Bagehot had recognized the possibility:

Walter Bagehot: Lombard Street <https://archive.org/details/lombardstreet00bage>: ‘It may be said that the reserve in the Banking Department will not be enough for all such loans. If that be so, the Banking Department must fail. But lending is, nevertheless, its best expedient. This is the method of making its money go the farthest, and of enabling it to get through the panic if anything will so enable it. Making no loans, as we have seen, will ruin it ; making large loans and stopping, as we have also seen, will ruin it. The only safe plan for the Bank is the brave plan…

And Keynes provided the answer:

John Maynard Keynes: The General Theory of Employment, Interest & Money <https://archive.org/details/in.ernet.dli.2015.50092/>: ‘It seems unlikely that the influence of banking policy on the rate of interest will be sufficient by itself to [always] determine an optimum rate of investment. I conceive, therefore, that a somewhat comprehensive socialisation of investment will prove the only means of securing an approximation to full employment; ‘though this need not exclude all manner of compromises and of devices by which public authority will , co-operate with private initiative…

And without full employment, the market system cannot rebalance the economy into any tolerable approximation of a new, taught, productive configuration. Price signals are there to guide people into making decisions about what activities are truly valuable for the community. Its guidance is always very imperfect: wealth inequality and externalities are outside of its vision. But its guidance is much butter than nothing. But when unemployment is not low and aggregate demand is slack, the market flashes a big red “NOT VALUABLE!” sign in front of nearly every possible expansion of economic activity. And so the rebalancing cannot take place, and definitely will not take place rapidly.

All this has been known for a century and a half, settled doctrine for nearly a century, and brought down from the Mount Sinai of economic theory into a narrative form nearly everyone can comprehend for half a century, ever since my teacher Charlie Kindleberger published his Manias, Panics, & Crashes <https://archive.org/details/maniaspanicscras0000kind/page/n6/mode/1up>.

So why was this Bagehot playbook not SOP during and after the GFC of 2008? And why does Dan Davies feel that he has to translate the financial-crisis playbook of Walter Bagehot into a cognitive-neuroscience register of financial crisis as concussion or stroke with this?:

Dan Davies: are we having fear yet? <https://backofmind.substack.com/p/are-we-having-fear-yet>: ‘The Great Financial Crisis [of 2008]…. a cognitive shock—a kind of head injury to the overall economic system… so that it was no longer able to perform a vital function of balancing present consumption and investment for the future…. I wrote about this…. “The ‘Lehman Moment’ was entirely a control failure… the information processing system was hit with something it couldn’t handle…. Snce a lot of the economic system was built in such a way as to require the particular mix of control, information and accounting co-ordination that is provided by the financial system… the real economic damage was wildly out of proportion…. The system couldn’t handle was a very rapid increase in the uncertainty attached to the value of financial assets… [because] the system was built on the assumption that certain kinds of interbank obligations could be treated as completely risk free…. If you relaxed the assumption of zero risk even a little bit, the bandwidth needed to manage the system increased hugely. What’s needed in that sort of situation is for some deus ex machina to step in and restore certainty… [That] was very difficult, because the system wasn’t set up for this kind of centralised action…

But the system was so set up. Or, at least, with central banks set up to follow the Bagehot SOP it was.

Part of the problem is that a side-effect of restoring certainty was that as a result the wicked would flourish like the green bay tree. But as Donald Kohn said shortly afterwards:

Don Kohn: <https://elischolar.library.yale.edu/cgi/viewcontent.cgi?article=1118&context=journal-of-financial-crises>: ‘I guess it must have been at Jackson Hole 2007; the focus of the discussion was moral hazard. At that time, in retrospect, we’d hardly done anything. We’d made the discount window cheaper. We’d encouraged people to borrow. Maybe we’d eased policy a little bit. Then, there was all this discussion that, “Oh my God, you’re creating moral hazard. There’s a Bernanke put, et cetera.” Ignore that, I guess would be the advice. Just go fast. Go hard. Go soon. Correct the moral hazard issues later with regulation like in Dodd-Frank. Don’t worry about it in the middle of the crisis…

The rest of the problem was what Paul Krugman has long described as the coming of an intellectual Dark Age—one of willful obscurantism, as people who ought to have known better stopped trying to understand the economy but instead turned to solving irrelevant models with spiffy new tools they had overinvested in learning—to a large chunk of economics <https://www.princeton.edu/~pkrugman/keynes_and_the_moderns.pdf>.

Things were, perhaps, at their nadir in 2009. I remember being at an INET conference in 2011 <https://larrysummers.com/news-item/brenton-woods-speech/> at which Larry Summers was being taunted and tormented by Martin Wolf with respect to the uselessness of academic economics in dealing with the GFC:

Martin Wolf & Larry Summers: A Conversation on New Economic Thinking <https://larrysummers.com/news-item/brenton-woods-speech/>:
Martin Wolf: ‘There are some very big questions, if not some pretty obvious symptoms, of a profound failure in modern economics…. Larry… how far [do] you share that perspective[?] How far do you feel that what has happened in the last few years… just simply suggests that economists didn’t understand what was going on?
Larry Summers: ‘There are things economists didn’t know… things economists were wrong about…. things where some economists were right…. I got a lot of papers…. attempted to read all the ones that used the words ‘leverage,’ ‘liquidity,’ ‘deflation’ or ‘depression.’ And I attempted to read none of the ones that used the words ‘neoclassical,’ ‘choice theoretic,’ ‘real business cycle,’ or ‘optimizing model of.’ There were more in the second category than there were in the first. But there were a reasonable number in the first, and they told you a lot. There is a lot in Bagehot that is about the crisis we just went through. There’s more in Minksy and perhaps more still in Kindleberger. There are enormous amounts that are essentially distracting, confusing, and problem denying in the stuff that is the substance of the first year courses in most PhD programs…. Economics knows a fair amount… has forgotten a fair amount…. And it has been distracted by an enormous amount…. People who were practical understood… liquidity finding its way into… asset price inflation and being problematic…. But… those concepts… were at the very edge of, and in many cases not even at the edge, of contemporary macroeconomics to the great detriment of contemporary macroeconomics.
On the other hand… everyone who doesn’t like economics has piled on at this moment to regard this crisis as a repudiation of economics, and I don’t think that’s right. I think the wisdom that’s in the Bagehot, Minsky, Kindleberger, Eichengreen, Akerlof, Shiller…

As I noted at the time to Barry Eichengreen, Larry had just called him the wisest living economist with respect to dealing with financial crises.

This story of the great loss of public-policy and economic-theory intellectual capital was one I watched happen over the course of the first two-thirds of my career so far. But I really do not understand it, not even now.

The person who comes closest to understanding it is, I think, Paul Krugman. This is his best shot.

Paul Krugman:“Mr. Keynes & the Moderns” <https://www.princeton.edu/~pkrugman/keynes_and_the_moderns.pdf>. ‘The… moderate economic policy regime… that by and large lets markets work, but in which the government is ready both to rein in excesses and fight slumps… can last for a generation or so, but not much longer… [due to] Minsky-type financial instability… [but also] equally crucial are the regime’s intellectual and political instability.
Let me start with the intellectual instability…. Paul Samuelson back in 1948… combines the grand tradition of microeconomics, with its emphasis on how the invisible hand leads to generally desirable outcomes, with Keynesian macroeconomics, which emphasizes the way the economy can develop what Keynes called “magneto trouble”, requiring policy intervention…. It’s a deeply reasonable approach—but it’s also intellectually unstable… [because in] micro, you assume rational individuals and rapidly clearing markets… [and in] macro, frictions and ad hoc behavioral assumptions are essential. So what?… The map is not the territory, and it’s OK to use different kinds of maps depending on what you’re trying to accomplish….
Given human propensities, plus the law of diminishing disciples, it was probably inevitable that a substantial part of the economics profession would simply assume away the realities of the business cycle, because they didn’t fit the models. The result was what I’ve called the Dark Age of macroeconomics, in which large numbers of economists literally knew nothing… and, of course, went into spasms of rage when their ignorance was pointed out.
To this intellectual instability, add political instability. It’s possible to be both a conservative and a Keynesian… [look at] Keynes…. But in practice, conservatives have always tended to view the assertion that government has any useful role in the economy as the thin edge of a socialist wedge. When William Buckley wrote God & Man at Yale, one of his key complaints was that the Yale faculty taught—horrors!—Keynesian economics…. Monetarism.. [was] an attempt to assuage conservative political prejudices without denying macroeconomic realities… [with Milton] Friedman was saying was, in effect, yes, we need policy to stabilize the economy—but we can make that policy technical and largely mechanical…. When monetarism failed… it was replaced by the cult of the independent central bank…. And this worked for a while….
It worked in part because the political insulation of central banks also gave them more than a bit of intellectual insulation, too. If we’re living in a Dark Age of macroeconomics, central banks have been its monasteries, hoarding and studying the ancient texts lost to the rest of the world. Even as the real business cycle people took over the professional journals, to the point where it became very hard to publish models in which monetary policy, let alone fiscal policy, matters, the research departments of the Fed system continued to study counter-cyclical policy in a relatively realistic way. But this, too, was unstable.
For one thing, there was bound to be a shock, sooner or later, too big for the central bankers to handle without help from broader fiscal policy. Also, sooner or later the barbarians were going to go after the monasteries too; and as the… [ca. 2011] furor over quantitative easing shows, the invading hordes have arrived.
Last but not least, there is financial instability… The very success of central-bank-led stabilization, combined with financial deregulation—tself a by-product of the revival of freemarket fundamentalism—set the stage for a crisis too big for the central bankers to handle. This is Minskyism: the long period of relative stability led to greater risk-taking, greater leverage, and, finally, a huge deleveraging shock…

But I am afraid I do not find this completely satisfactory. One note: Milton Friedman did not say that government intervention to stabilize the economy can be made technical and largely mechanical: he said that whatever policy stabilizes the economy—which he long thought was a stable growth rate for the “M2” monetary aggregate—was by definition neutral and non-interventionist. A neat rhetorical trick, at the time, in allowing the maintenance of the “government failure when it intervenes in the economy is always bigger than the market failure it purports to correct” bright line for True Believers. But intellectually profoundly unhelpful.

Bagehot, Walter. 1873. Lombard Street: A Description of the Money Market. London: Henry S. King & Co. <https://archive.org/details/lombardstreet00bage>.
Davies, Dan. 2025. “are we having fear yet?” Back of Mind. October 30. <https://backofmind.substack.com/p/are-we-having-fear-yet>.
Haggerty, Maryann. 2020. “Lessons Learned: Donald Kohn”. Journal of Financial Crises. 2:3. <https://elischolar.library.yale.edu/cgi/viewcontent.cgi?article=1118&context=journal-of-financial-crises>.
Keynes, John Maynard. 1936. The General Theory of Employment, Interest, & Money. London: MacMillan. <https://archive.org/details/in.ernet.dli.2015.50092>.
Kindleberger, Charles P. 1978. Manias, Panics, & Crashes: A History of Financial Crises. New York: Basic Books. <https://archive.org/details/maniaspanicscras0000kind/page/n6/mode/1up>
Krugman, Paul. 2011. “Mr. Keynes & the Moderns”. Princeton. <https://www.princeton.edu/~pkrugman/keynes_and_the_moderns.pdf>.
Summers, Lawrence & Martin Wolf. 2011. “A Conversation on New Economic Thinking”. INET. <https://larrysummers.com/news-item/brenton-woods-speech/>.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#on-the-faux-puzzlement-of-dan-davies

Yes: AI Is a Bubble. But It Is a Bubble-Plus. & That Makes a Substantial Difference

Brad DeLong — Thu, 30 Oct 2025 12:37:50 GMT

Bubble-double-plus: how the AI boom props up the economy with debt, data centers, faith, and valuations without rational expectations of durable and monetizable productivity growth…

The number of people uneasy about the flywheel between the macroeconomy, the stock market, and the ongoing AI bubble is growing. Here is Tracy Alloway:

Tracy Alloway: Here’s what Tracy is thinking about <https://www.bloomberg.com/news/newsletters/2025-10-29/some-thoughts-on-today-s-fed-decision>: ‘Just how central is the stock market to the overall economy? If wealthy people are driving spending and the assets at the heart of that wealth keep hitting new highs, then it would make sense that equity markets become an increasingly critical support for jobs and economic growth…. As stocks have notched new records, we’re seeing new products and strategies designed to help the wealthy tap into those gains…. People using box spread strategies to create a sort of synthetic loan at a lower rate than they might get from a bank… effectively borrowing against their portfolios…. Stelrix… pitches itself as the first credit card that enables users to automatically borrow against their stock portfolio…. The target market for this card — which apparently is “crafted in proprietary glass by the same minds behind the Black Amex Centurion” — seems pretty obvious…. The money’s not just getting spent as stocks go up, it’s getting multiplied…

Yes: the AI-bubble is pushing up investment spending directly via construction of data centers, and indirectly via valuation effects on consumption. And then the multipliers kick in. If not for sky-high optimism about AI, the economy would now be in recession with high probability. This sky-high optimism has pushed stocks up to remarkably high valuation ratios rivaling those of the very end of the 1990s:

And this in spite of long-term bond rates that are quite low in historical perspective—inconsistent with any sort of strong expected future productivity boom that might generate income to validate such high valuation ratios. Low bond rates even when there is a furious AI bubble-driven data-center capital investment boom—they are certainly registering the presence of a global savings glut, if not of full-blown secular stagnation:

As best as I can see, this sky-high optimism about stocks is not warranted.

And here is the very sharp Gillian Tett is right now sounding the alarm about the shift in AI data-center funding to debt and to circular vendor financing. When the people who are giving you the money do not trust that your speculative enterprise, that is a bad sign. And when you can only get customers by lending them the money to buy your products—that is a very bad sign too. Gillian Tett:

Gillian Tett: AI has a cargo cult problem <https://www.ft.com/content/f2025ac7-a71f-464f-a3a6-1e39c98612c7>: ‘Spending vast sums and inflating an investment bubble is no guarantee of unleashing technological magic: Is it just a Ponzi scheme? That is the question that currently haunts American tech — and wider markets — as the valuation of artificial intelligence-linked groups soars to evermore eye-popping heights…. Ten lossmaking AI start-ups — such as OpenAI, Anthropic and Elon Musk’s xAI — now command a collective valuation of close to $1tn, while venture capital has poured $161bn into AI overall this year…. Fw of these entities expect to turn a profit anytime soon — and these valuations are being boosted by variants of cross-cutting vendor financing, like recent deals between OpenAI, Nvidia, Oracle, AMD and Broadcom… [in] a pattern of circular flows that echo some of the hairball of interconnections that emerged between banks and insurance companies via credit derivatives before 2008… [which] resulted in unseen concentrations of risk — and subsequent contagion when the bubble burst…. And… even tech luminaries, like Jeff Bezos, admit there is excessive exuberance…

Share DeLong's Grasping Reality: Economy in the 2000s & Before

But, right now, it is a bad sign only for those invested in the technology’s long-term superprofitability and in making money by building and running data centers. For while it is to some degree a Ponzi scheme, it is not “just” a Ponzi scheme. And that makes a big difference: that means that it will probably go on for quite a while, and that when there is the ultimate shakeout the tech world will look very different and very weird because of all the money poured into data centers, software development, and business-model experimentation.

Back up. These days I like to say that the AI bubble is eight things. It is:

one part: reasonable expectation of providing and financially capturing true end-user value,
two parts: millennarian religious hype on the part of those hoping for the Rapture of the Nerds,
three parts perhaps-reasonable expectation of improved advertising targeting for user good and user ill,
four parts: grifters seeking easy investor marks moving over from crypto,
five parts: platform near-monopolists fearing the loss of their profit flows to a Christensenian disruption from the Next New Big Thing,
six parts:
- speculative possible enormous increases in utilizer surplus from the ability to add natural-language interfaces to every interaction with structured and unstructured databases,
- or decreases in human flourishing as natural-language interfaces serve as a stalking-horse for yet another round of hacking users’ attention, and not for their benefit,
- but in either case not a likely source of Google-Facebook-Apple-style platform monopoly profits,
seven parts: the downstream consequences for human society and culture flowing from enormous increases in human data-analysis capabilities driven by the unbelievably large scope of huge data, enormous dimension, highly flexible-function classification tools coming online now and soon, and last:
eight parts: the downstream consequences for human society and culture flowing from MAMLM-mediated—Modern Advanced Machine-Learning Model-mediated—human interaction with the infosphere.

Of these, only (1) and (3) are likely sources of superprofit for investors.

(6), (7), and (8) are transformative, potentially, for human society and culture, but not likely sources of superprofits from investors in providing AI-services or AI-support right now. (5) is a defensive move: not an attempt to boost the profits of platform monopolists, but to spend a share of those profits—a large and growing share—defending them against Clayton Christensenian disruption. (2) and (4) are culturally important, and are driving much of investor interest on the belief that with so much excitement about this pile of manure there must be a pony in there somewhere. But focusing on them is not likely to lead to good investment decisions.

Yes, the grifters moving over from crypto are definitely trying to run a Ponzi scheme: if the person trying to sell you something has just spent a decade selling BitCoin, DogeCoin, and Web3 use cases coming real soon now, you should probably block them and add them to your spam list.

And anyone saying either of these two things: (i) soon everyone will be under threat from a malevolent digital god that will soon control all of our minds through flattery, misdirection, threats, and sexual seduction, the negative millennarians; or (ii) they are on the cusp of building a benevolent digital god. Shake your head and walk away. And if they then turn to “give me money! lots of money!”—well, then run.

oth groups are mostly conning themselves. But do not believe any forecasts or accept any unsecured debt obligations from anyone talking about ASI—Artificial Super-Intelligence—coming within a decade.

And if whatever project is pitched to you requires that the enterprise make a lot of money by using some form of natural-language interface to eat away at and grab a share of Google’s, Facebook’s, Amazon’s, or Apple’s platform-monopoly profits, run away: Google, Facebook, Amazon, and Apple will in the end put five times as many programmers on-task as whatever you have invested in, and give away their product for free to keep you from being successful. Think: Microsoft vs. Netscape in the 1990s. Oracle, Microsoft, and OpenAI may make money at a reasonable pace by selling data-center and chatbot-interface services; firms like CoreWeave, Nebius, Lambda, Equinix, Aligned, Digital Realty may make substantial amounts of money too; as will AMD, BroadCom, TSMC, ASML and most of all NVIDIA (but those are all very richly valued by now even for the likely future profit flows).

But the market disagrees. So we do read, day by day, about things like:

Ruchir Sharma: America is now one big bet on AI <https://www.ft.com/content/6cc87bd9-cb2f-4f82-99c5-c38748986a2e>: ‘Large companies and investors… are increasingly confident that artificial intelligence is such a big force, it can counter all the challenges. Lately, this optimism has become a self-fulfilling prophecy. The hundreds of billions of dollars companies are investing in AI now account for an astonishing 40 per cent share of US GDP growth this year…. AI companies have accounted for 80 per cent of the gains in US stocks so far in 2025. That is helping to fund and drive US growth, as the AI-driven stock market draws in money from all over the world, and feeds a boom in consumer spending by the rich….
Without all the excitement around AI, the US economy might be stalling out, given the multiple threats.… [The] immigration boom-bust cycle… alone will reduce America’s growth potential by more than a fifth…. Government deficits and debt are increasing faster in the US than in other developed markets…. AI is regarded as a magic fix… [because] it is expected to deliver a significant boost to productivity growth…. The possibility of a productivity miracle to come has cemented the faith of investors…. Foreigners poured a record $290bn into US stocks in the second quarter and now own about 30 per cent of the market…. AI [had] better deliver for the US, or its economy and markets will lose the one leg they are now standing on…

If anything, this is understated:

It is not that the U.S. economy “might be stalling out”.
It is that the U.S. economy would be stalling out, or worse.
We will have a truly ungodly huge amount of datacenters and GPUs and (largely natural gas) power plants when this build-out is done. For example, only the first of many is StarGate 1 in Abilene, Texas: 500,000 square feet and 200 MW per building; 8 buildings planned; a campus roughly 1 square mile.
These concentrations of computer power will be useful for something.

Why are we here? Well, I gave eight reasons above, of which only one, and of that the smallest, was the idea that those undertaking the programming and data-center building-out are doing so as a result of their reasonable expectation of providing and financially capturing true end-user value.

But there is also a ninth reason. The ninth reason is that, right now, the bear case is simply not being made by anyone outside the extended AI-knowledgeable community.

I think the extent to which that case is not being made is best grasped by looking at Matt Yglesias here:

Matt Yglesias: The AI boom is propping up the whole economy: It’s a fragile basis for growth, whether or not it’s a “bubble.”… When we were pitching the idea that became Vox… the people who were interested in the pitch were interested in Ezra Klein and Melissa Bell and Matt Yglesias. And that was not a crazy calculation…. And the other thing… is that, at the time, investors were enthusiastic about ad-supported digital media… [on the] thesis (wrong, as it turned out) that you could make a lot of money…. That’s where it gets hard to judge these things. Investing in start-ups… [as a] business… depends less on having a very high batting average than on the idea that home runs will be extremely lucrative…. The thesis lives or dies primarily on the question of how big the hits will be. And if I could tell you that, I wouldn’t be writing a subscription newsletter for a living…. The leading AI labs are genuinely seeing rapid revenue growth. I don’t think “the trajectory of model capabilities improvement and associated revenue will continue on its current path indefinitely” is the most sophisticated investment thesis in the world. But it’s also not unmoored from reality. It’s a plausible but possibly wrong guess about the future state of the world — i.e., a pretty normal investment hypothesis…. The nice thing about financial markets is that the people investing in them have very strong incentives to at least try to avoid wishful thinking and motivated reasoning and instead focus on cold hard cash…

And here Matt has, I think, served his readership badly.

He should say: The question of how big the hits will be and whether you or anyone should invest hinges on the interaction of (i) the technology and its future development, (ii) the industrial market structure in that it has to be conducive to both the commodification of all of the pieces of the value chain complementary to your enterprise and durable moat-construction for your piece, and (iii) the demand which needs to be potentially sky-high.

He should also say: If someone—or the market as a whole—brings you what they claim is such an opportunity, remember that there is a famous passage in the American musical “Guys & Dolls”:

Sky: Nathan, let me tell you a story.
Nathan: Have we got a bet?
Sky: Nathan, let me tell you a story. On the day I left home to make my way in the world, my daddy took me to one side.
“Son,” my daddy says to me, “I am sorry I am not able to bankroll you to a large start, but not having the necessary lettuce to get you rolling, instead, I’m going to stake you to some very valuable advice.
“One of these days, a guy is going to show you a brand-new deck of cards on which the seal is not yet broken. Then this guy is going to offer to bet you
that he can make the jack of spades jump out of this brand-new deck of cards and squirt cider in your ear.
“But, son, you do not accept this bet because, as sure as you stand there, you’re going to wind up with an ear full of cider…”

He should then also say: I do not believe I can judge the chances of a big hit—the only people who can are the people who know enough about industrial structure and user demand to get by but know a huge amount about the technology, the people who know enough about the technology and about potential user demand to get by but know a huge amount about industrial structure, and the people who know enough about the technology and industrial structure to get by and know a huge amount about potential user demand.

And he should penultimately say: If the person offering you this deal is not in one of those three groups—and the market as a whole right now definitely isn’t—what the hell are you doing listening to their representations that this is one of those cases in which you should invest anyway? And if the person offering you this deal is in one of those three groups—and so knows whether this is actually a favorable-odds bet or not—what is their motive for offering you a piece of the action rather than keeping it all for themselves and their friends?

And so the ultimate punchline should be quite different than what Matt says. Matt says that it may well not be a bubble—that investments in AI programming and the AI build-out right now may well make sense—because people in financial markets “have very strong incentives to at least try to avoid wishful thinking and motivated reasoning”, and hence Democrats should think very hard about “how to win in 2028 if unemployment stays low and growth remains robust”.

But look at the eight things driving this—reasonable profit expectations, another ride of the crypto-grifters, millennarian religious hype, advertising targeting, platform-monopoly defense, huge but nearly impossible to turn into profit shifts in human behavior, the wild card of enormous new analytical capabilities, and the downstream consequences of a large shift in the modality of human communication.

Look at all that, and the obvious conclusion is that the fact that this is a bubble does not mean any sort of economic bill in terms of high unemployment and falling production is going to come due.

Trees don’t grow to the sky. But bubbles that are far more complicated than simple Ponzi schemes do not collapse on any schedule. Especially if the Ponzi-scheme element is only a very small part of the total econo-techno-cultural-socio story.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#yes-ai-is-a-bubble-but

On the Surface, the U.S. Economy Looks Better than Its Fundamentals

Brad DeLong — Thu, 23 Oct 2025 02:31:43 GMT

The stock market says “all good.” Consumer sentiment says “2009.” Both can be true—if the scoreboard is rigged, and sends the wrong signals in the context of AI-overexuberance and a two-class system that suppresses labor while juicing capital…

Unemployment is low. Unemployment counts joblessness. It does not count fear. It does not count the worker who won’t complain, won’t organize, won’t quit—because deportation or lost health insurance is worse. Sentiment captures that reality first. The live question, therefore, for capital: do domestic gains from rent extraction enabled by labor suppression beat global losses from deglobalization and trade chaos? I think probably not. But the stock-market tape, also juiced by the AI bubble says yes. Maybe we have a genuine two-speed economy. Or maybe we are genuinely flirting with recession, with AI-hype euphoria papering over a broader stall. Thus if you want a historical rhyme, don’t fear 2009. Fear the late 1920s: ticker-tape joy atop sectoral distress and fragile distribution. When you watch the wrong scoreboard, you understate the risks of crashes. Perhaps the workers and consumers with strongly negative confidence are putting more accurate numbers up on the scoreboard.

Paul Krugman writes:

Paul Krugman: The U.S. Economy is in Worse Shape Than it Looks <https://paulkrugman.substack.com/p/the-us-economy-is-in-worse-shape>: ‘No recession so far, but the no-hiring economy is hurting workers…. Consumer sentiment is much weaker than it was pre-Covid, in fact comparable to its level at the depths of the 2008-2009 financial crisis…. There are some objective, measurable reasons to say that the US economy, which appears OK by the most commonly used measures, is definitely not OK…. AI is booming, but the rest of the economy isn’t…. While there have been no mass layoffs so far, people who have lost their jobs or are just entering the work force are finding it very hard to get new jobs…. [And] there are clear signs that middle-to-low income consumers are struggling… [while] the top 10% of the income distribution now accounts for nearly half of all consumer spending…. [Perhaps] Trump’s wildly erratic policies are creating huge uncertainty which is deterring many companies – essentially those that are not in the AI sector or a sector catering to the affluent – from making investments. And those forgone investments include hiring new workers.… The number of long-term unemployed — would-be workers who have been jobless for more than 6 months — had soared as of August, and has probably continued to rise since then:
Workers believe, rightly, that if they should happen to lose their current job they will have a hard time finding another. This obviously means that workers have much less bargaining power…

My take: The disconnect between official metrics and what we seem to see in felt experience runs deeper than a mismatch between aggregate statistics and individual anxieties. There has been a vibe shift to a new régime of labor suppression, one that operates through immigration enforcement, regulatory abandonment, and the systematic dismantling of worker protections. The stock market’s buoyancy masks what may be either a genuine two-speed economy or—more troubling—a weak economy-wide recession papered over by AI-driven irrational exuberance.

Start with what most economic commentary ignores: the precarity that now defines the status of all non-citizens in America. Many people I talk to do not grasp how thoroughly the ground has shifted. This is not merely about immigrants without papers, though their vulnerability has intensified. Even if you hold a green card—that putative guarantee of permanent residence:

Kristi Noem can ask Marco Rubio to revoke it.
He will do so.
And then you are gone from these shores.

The administrative machinery exists; the political will to deploy it has been demonstrated; the checks that might once have constrained such actions have evaporated.

Or consider the H-1B visa holder. You face not just the risk of attracting political attention from Noem and Rubio, but the employer’s power to destroy your legal status by firing you and dropping a dime to ICE. What is your ability to negotiate wages, resist exploitation, or assert workplace rights collapses under this Sword of Damocles? The visa ties you to your employer in a bond of indenture. Your company knows it; you know it; the entire labor market knows it. This knowledge reshapes bargaining power across the board.

These are more than vibe shifts. They are, I think, already having powerful consequences:

Thus we now have a large workforce segment that cannot risk asserting rights, cannot job-hop for better wages, cannot organize, cannot complain about wage theft or unsafe conditions. The spillover effects extend well beyond the directly affected workers. When a substantial portion of your labor market operates under these constraints, it exerts downward pressure on wages and working conditions for everyone competing in adjacent labor markets.

Now layer onto this foundation the systematic abandonment of other worker protections. No checks on union-busting remain. The regulatory apparatus that might investigate, penalize, or prevent the suppression of organizing efforts has been dismantled or deliberately starved of enforcement capacity. Wage theft—the practice of simply not paying workers what they are owed—proceeds without meaningful consequence. The agencies that might intervene have been neutered.

Extend the logic further: no checks on corporate consumer financial fraud either. The regulatory framework that might protect consumers from predatory lending, deceptive practices, or financial manipulation has been hollowed out. The Consumer Financial Protection Bureau, once a source of genuine constraint on corporate behavior, has been transformed into a decorative shell.

The message to capital is clear: the rules, such as they remain, will not be enforced against you.

Add in the latest Republican tax cut for the rich. This is not merely a fiscal policy choice; it is part of a comprehensive architecture for transferring income from labor to capital. The immigration enforcement regime suppresses wages. The abandonment of worker protections prevents labor from organizing to capture productivity gains. The gutting of consumer protections allows capital to extract rents through financial manipulation. The tax cuts then formalize and accelerate the transfer by directly moving resources from public goods and income support toward the already-wealthy.

These mechanisms are not independent policy choices but interlocking elements of a system. The result is considerable pressure transferring income from labor to capital—and this transfer is not accidental or incidental but the explicit goal of the policy architecture.

And so the stock market climbs. Add in irrational exuberance from the AI bubble, and it is no surprise that equity values soar. The market anticipates productivity gains from AI that may or may not materialize, extrapolates them into the indefinite future, capitalizes them into current valuations at discount rates that assume risk has disappeared, and adds on a rocket boost from labor suppression.

And so here we encounter an old problem: the American press’s strong tendency to view the stock market as the scoreboard for the economy. This habit always serves us badly. It serves us especially badly now. The stock market rises on the back of labor suppression and speculative exuberance. And all the media are treating it as evidence of economic health, thus producing dangerous confusion.

The question before capital is this: Do the benefits to capital from labor suppression outweigh the costs to capital from the rest of the world slowly deglobalizing its production value chains from America? I do not think they do. The Trump administration’s chaos-monkey approach to trade policy, its contempt for alliance relationships, and its systematic destruction of the institutional frameworks that enabled global supply chains—these are smashing many of American capital’s wealth gains from globalization.

But the stock-market disagrees. And I recognize that I might be wrong. This might actually be good for capital even in the long term. It might be that we genuinely have a two-speed economy: gains from domestic rent extraction exceed the losses from international retrenchment. Then the economy looks strange because it is strange—one sector genuinely thriving while another suffers.

There is an alternative interpretation: The wealth losses to capital from deglobalization are real and substantial. But these losses get papered over by AI-driven over-exuberance in equity markets and by systematic underestimation of how many of American capital’s wealth gains from globalization are being destroyed by Trump’s policies. In this scenario, we are not in a two-speed economy but stagnant, flirting with a weak recession that we are failing to recognize because we are reading the wrong indicators.

In this case, the historical parallel that should haunt you is not 2008-2009, despite similar consumer sentiment levels. Perhaps the parallel to fear is the late 1920s, when stock market exuberance and agricultural depression coexisted, when the maldistribution of income reached levels that made the economy fragile, when observers focused on the stock ticker rather than on underlying structural weaknesses.

The crash, when it came, surprised people who had been watching the wrong scoreboard.

Share DeLong's Grasping Reality: Economy in the 2000s & Before

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#on-the-surface-the-us-economy-looks-better-than-its-fundamentals

For a Normal U.S. Government to Shoulder "Credit Risk" Can Be Fine. But This Is Not a Normal U.S. Government, Is It?

Brad DeLong — Wed, 08 Oct 2025 12:36:51 GMT

Chaos monkeys promising unconditional U.S. financial support for unsustainable Argentinian economic policies is a royal road to disaster. Failing to make that transparently clear at the start of the piece is doing a disservice to your readers. So why are you doing this, Matt Klein?…

Et tu Matthew? Normalizing chaos monkeys is not a good look:

Matthew Klein: What Should the U.S. Try to Achieve in Argentina? <https://theovershoot.co/p/what-should-the-us-try-to-achieve>: ‘Argentina’s longstanding vulnerabilities to capital flight and inflation have reasserted themselves with a vengeance…. The U.S. Treasury is preparing to commit tens of billions of dollars to Argentina…. The IMF is apparently “coordinating support”, which could involve loans from the ~$175 billion of Special Drawing Rights (SDRs) held by the Treasury. According to Treasury Secretary Scott Bessent, the goal is to give Argentina “the tools to defeat speculators, including those who seek to destabilize Argentina’s markets for political objectives.” At one level, this is incredibly stupid, because the Argentine government will almost certainly fail to repay the Treasury without doing something rash. At another level, however, I think that there is a germ of a good idea in Bessent’s proposal. The U.S. government should be more willing to take credit risk for the sake of mitigating balance of payments crises. The question is whether Argentina is the best case for doing this…

The next to last sentence is a genuine point. Indeed: the U.S. government should be more willing than it has been to take credit risk for the sake of mitigating balance of payments crises. The sentence before it and the sentence after it, however, are embarrassing. Because it is Trumpist Bessent’s proposal, it is not a good idea at all—not even a “germ”. And because it is the Trump chaos-monkey administration doing it, Argentina is not the best case.

As I said a week ago:

Dan Drezner said last week that one good thing that could be said about Scott Bessent was that he was 10% less crazy than the other Trumpist chaos monkeys. Scott Bessent’s response?: “Hold my beer!” Stating that the U.S. will pay the piper but that Javier Milei will call the tune and that U.S. financial support for Argentina is “unconditional” may well be the most unprofessional thing I have ever heard a Treasury Secretary say… <https://braddelong.substack.com/p/no-there-are-no-even-half-competent>

Lender-of-last resort operations involve (a) lending freely (b) at a penalty rate (c) on collateral good in normal times. In the international context:

“on collateral good in normal times” means that repayment financial flows need to be earmarked.
"at a penalty rate” means that one does not say that the support is “unconditional”: rather, one says that support is conditional on policy changes—and that those are already in train, and unless they happen the support will not go forward.
absent those two, claims that one is lending freely are simply not credible.

This is smelling like the British Tories’ defense of the pound back in 1992, the thing that made George Soros’s career as he took ten figures of pounds sterling away from the British government. And since that story has been told and retold everywhere in the Soros affinity all the time since, Bessent knows it very well.

So what does Bessent think he is doing? And what does Klein—smart, honorable, well-intentioned—think he is doing in providing even half-hearted praise for the “germ of a good idea”, and stating that it is just a question of whether here and now?

Maybe Klein thinks, rhetorically, that if he starts his piece sympathetically he will bring more people along to his ultimately negative conclusion. But that is not how this works: that may be what he wants Trumpist and Trumpist-curious readers to hear. That is not what they will hear.

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#for-a-normal-us-government

HOISTED FROM OTHER PEOPLE'S ARCHIVES: Unprofessionals Spreading Misinformation Edition

Brad DeLong — Wed, 10 Sep 2025 16:02:13 GMT

In which John Cochrane claims that Walter Bagehot's "Lombard Street" & all subsequent analyses of lenders-of-last-resort are not economics...

People are pushing back on me—nobody believing that John Cochrane in March 2009 was as moronic as he was.

Few people believing that in the late winter of 2009 one of the most important things Sebastian Mallaby thought he should do from his perch at the CFR was to run a conference in which people who understood anything at all about a financial crisis were as scarce as hen’s teeth.

But at least some of the lurkers in email will credit that, back in Those Days, the usually reliable Chrystia Freeland offered no pushback at all to a panel she was running that was, largely, insane.

However, it now can be found only at the WayBack Machine. CFR has trashed the record of its conference.

So here is what John Cochrane was saying back then:

Nobody in the Great Depression… was fooling enough to think that toxic assets were the problem or that the government, by somehow stirring up the liquidity of $10 trillion of toxic assets, was going to make the banks look all right again. This is an amazing fairy tale we've been telling ourselves for eight months now. Think of every step of the chain. Stirring the pot is going to make everything more valuable. Making everything more valuable is going to make the banks look solvent again. Making the banks look solvent again means people are going to start buying bank stocks. Banks are just dying to go and lend if only they had more equity -- it's not just going to happen. Well, that one we're inventing on our own.
My job is a finance professor so let me mention a few things about securities. Fundamentally what happed in the '20s to '30s and in the recent time to now has very little to do with the Fed. The risk premium was very low in the boom and is very high now. There's almost no economics that describes how the Federal Reserve, by monkeying around with three months of Treasury bills and reserves, can lower long-term interest rates, but absolutely no economics that says how the Federal Reserve is in charge of the risk premium, and that's what was going on. People were willing to hold mortgages, stocks, risky bonds at amazingly low premiums on the way up and now they want very high premiums on the way down. That's not something the Federal Reserve is in charge of.
Let me say something nice also about Glass-Steagall. It had many faults but its essential wisdom is that if there's something systemic that you're going to bail out, you draw a circle around that and then you say, this is not systemic and will be allowed to go under. That's what we're going to have to do, and our current Treasury plans still don't envision that. They envision we're going to let these monster businesses keep going and somehow perch a little fairy of regulation on everybody's shoulders to make sure they do the right thing. That just is not going to work.
General -- well, a last couple of general comments. Many things are depressingly the same. Policy is chaotic. Who would invest in this climate? It's not about toxic assets; it's about who wants to go in on a deal with Darth Vadar, who can change his mind at any moment? That's the uncertainty that's keeping things from getting going and that's what's slowing the rebuilding of financial markets. We're facing growth-destroying marginal tax rates, an excuse for the government takeover of large and completely unrelated sectors, class warfare, vindictive ex post taxations. This is the chance for a credit crunch -- which normally resolves itself fairly quickly -- to turn into a Great Depression. And perhaps most of all there is the danger of learning the wrong lessons; that our grandchildren will have to come back to the next meeting to say, what were the lessons -- the lessons mis-learned of the last time around?
My great hope is that the bounce-back will be quick before the quack medicine can be said to have worked. (Chuckles.) Just as we sort of -- as people think that this insane idea of fiscal stimulus -- which I'll go on with later if I get a chance -- came from Roosevelt's experience with no reason why it should work, there is a danger of thinking all of the crazy stuff they're doing now will have caused the bounce-back, if that happens, in five years, but my only hope is that it happens quickly and doesn't leave us with another Great Depression.

It’s not as if Cochrane thinks the entire subfield of financial-crisis-&-lender-of-last-resort economics from Walter Bagehot’s Lombard Street through Charlie Kindleberger’s Manias, Panics, & Crashes to, say, Andrew Hauser’s (2021) “From lender of last resort to market maker of last resort via the dash for cash” is wrong.

Cochrane is unaware that that subfield exists.

Which is a stunning and unprofessional degree of ignorance that still boggles my mind.

<fred.stlouisfed.org/ser…>

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#hoisted-from-other-peoples-archives
#unprofessional-people-calling-themselves-economists
#financial-crises
#lenders-of-last-resort
#walter-bagehot
#charles-kindleberger

Employment Growth Stalled: America's Labor Market in Tariff-Uncertainty Twilight

Brad DeLong — Fri, 05 Sep 2025 13:08:12 GMT

The Federal Reserve is boxed in—with neither a reduction in rates to fight possible recession nor an increase to fight possible inflation prudent—with sputtering payroll as continued random tariff threats cloud economic policy and Federal Reserve decision-making…

Let me extend my note from Friday September 5, 2025 dawn:

: ‘“Tim Cook would be in pretty good shape,” Trump said Thursday of the Apple chief executive officer…. “I’ve discussed it with the people here, chips and semiconductors, and we’ll be putting tariffs on companies that aren’t coming in,” Trump said. “We’ll be putting a tariff very shortly. You probably are hearing we’ll be putting a fairly substantial tariff, or not that high, but fairly substantial tariff”…\n\nSo the right and prudent policy for the Fed at its September meeting is, still, to stand pat.\n\n","body_json":{"type":"doc","attrs":{"schemaVersion":"v1"},"content":[{"type":"paragraph","content":[{"type":"text","text":"Well, well, well: seasonally-adjusted, payroll employment in July was no higher than we had thought yesterday it had been in June. Our estimate of the June level has been revised down by 21,000—yes, the last report that maddened Trump was more optimistic than it looks now like the situation warranted. And the seasonally-adjusted gain from June to July was only 22,000:"}]},{"type":"blockquote","content":[{"type":"paragraph","content":[{"type":"text","marks":[{"type":"bold"}],"text":"BLS"},{"type":"text","text":": Employment Situation Summary: ‘Total nonfarm payroll employment changed little in August (+22,000) and has shown little change since April, the U.S. Bureau of Labor Statistics (BLS) reported today. The unemployment rate, at 4.3 percent, also changed little in August. A job gain in health care was partially offset by losses in federal government and in mining, quarrying, and oil and gas extraction…. The change in total nonfarm payroll employment for June was revised down by 27,000, from +14,000 to -13,000, and the change for July was revised up by 6,000, from +73,000 to +79,000. With these revisions, employment in June and July combined is 21,000 lower than previously reported…"}]}]},{"type":"paragraph","content":[{"type":"text","text":"Total gain over the past three months: 88,000, which 29,000 per month: genuine stall speed. "}]},{"type":"paragraph","content":[{"type":"text","text":"If the tariff situation were at all settled, this would put me on the side of looking through what are (probably) still one-off price-level increases and lead me, were I on the FOMC, to prefer an 0.50% policy interest rate cut."}]},{"type":"paragraph","content":[{"type":"text","text":"But the tariff situation is not at all settled, is it?:"}]},{"type":"blockquote","content":[{"type":"paragraph","content":[{"type":"text","marks":[{"type":"bold"}],"text":"Josh Wingrove"},{"type":"text","text":": Trump Says ‘Fairly Substantial’ Chips Tariffs Coming ‘Shortly’ <"},{"type":"text","marks":[{"type":"link","attrs":{"href":"https://www.bloomberg.com/news/articles/2025-09-05/trump-says-fairly-substantial-chips-tariffs-coming-shortly?srnd=homepage-americas","target":"_blank","rel":"nofollow ugc noopener","class":"note-link"}}],"text":"https://www.bloomberg.com/news/articles/2025-09-05/trump-says-fairly-substantial-chips-tariffs-coming-shortly?srnd=homepage-americas"},{"type":"text","text":">: ‘“Tim Cook would be in pretty good shape,” Trump said Thursday of the Apple chief executive officer…. “I’ve discussed it with the people here, chips and semiconductors, and we’ll be putting tariffs on companies that aren’t coming in,” Trump said. “We’ll be putting a tariff very shortly. You probably are hearing we’ll be putting a fairly substantial tariff, or not that high, but fairly substantial tariff”…"}]}]},{"type":"paragraph","content":[{"type":"text","text":"So the right and prudent policy for the Fed at its September meeting is, still, to stand pat."}]},{"type":"paragraph","content":[{"type":"text","text":"<"},{"type":"text","marks":[{"type":"link","attrs":{"href":"https://fred.stlouisfed.org/series/PAYEMS#","target":"_blank","rel":"nofollow ugc noopener","class":"note-link"}}],"text":"https://fred.stlouisfed.org/series/PAYEMS#"},{"type":"text","text":">"}]}]},"restacks":0,"reaction_count":0,"attachments":[{"id":"7b4feb74-0d49-4d40-95e8-f398a2836134","type":"image","imageUrl":"https://substack-post-media.s3.amazonaws.com/public/images/5c2c7451-5718-4529-b058-b22d43c106c9_1968x698.png","imageWidth":1968,"imageHeight":698,"explicit":false},{"id":"1e5b1183-924b-41bf-9b21-87639152d80d","type":"link","linkMetadata":{"url":"https://fred.stlouisfed.org/series/PAYEMS#","host":"fred.stlouisfed.org","title":"All Employees, Total Nonfarm","description":"All Employees, Total Nonfarm","image":"https://substack-post-media.s3.amazonaws.com/public/images/3b313e31-9ca1-4ca5-a44e-a91e3793128c_880x463.png","original_image":"https://fred.stlouisfed.org/graph/fredgraph.png?width=880&height=440&id=PAYEMS"},"explicit":false}],"name":"Brad DeLong","user_id":16879,"photo_url":"https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2Fea5ae644-9822-4ca5-ac6b-e18c017d8fbc_1189x1208.png","user_bestseller_tier":1000}}" data-component-name="CommentPlaceholder">

<https://substack.com/@delongonsubstack/note/c-152668421>

Well, well, well: seasonally-adjusted, payroll employment in July was no higher than we had thought yesterday it had been in June. Our estimate of the June level has been revised down by 21,000—yes, the last report that maddened Trump was more optimistic than it looks now like the situation warranted. And the seasonally-adjusted gain from June to July was only 22,000:

BLS: Employment Situation Summary: ‘Total nonfarm payroll employment changed little in August (+22,000) and has shown little change since April, the U.S. Bureau of Labor Statistics (BLS) reported today. The unemployment rate, at 4.3 percent, also changed little in August. A job gain in health care was partially offset by losses in federal government and in mining, quarrying, and oil and gas extraction…. The change in total nonfarm payroll employment for June was revised down by 27,000, from +14,000 to -13,000, and the change for July was revised up by 6,000, from +73,000 to +79,000. With these revisions, employment in June and July combined is 21,000 lower than previously reported…

Total gain over the past three months: 88,000, which 29,000 per month: genuine stall speed.

If the tariff situation were at all settled, this would put me on the side of looking through what are (probably) still one-off price-level increases and lead me, were I on the FOMC, to prefer an 0.50% policy interest rate cut.

But the tariff situation is not at all settled, is it?:

Josh Wingrove: Trump Says ‘Fairly Substantial’ Chips Tariffs Coming ‘Shortly’ <bloomberg.com/news/arti…>: ‘“Tim Cook would be in pretty good shape,” Trump said Thursday of the Apple chief executive officer…. “I’ve discussed it with the people here, chips and semiconductors, and we’ll be putting tariffs on companies that aren’t coming in,” Trump said. “We’ll be putting a tariff very shortly. You probably are hearing we’ll be putting a fairly substantial tariff, or not that high, but fairly substantial tariff”…

So the right and prudent policy for the Fed at its September meeting is, still, to stand pat. It is in a box, and there is no good path out of it until we get some policy stability.

<https://fred.stlouisfed.org/series/ADPWNUSNERSA#>

Given the corruption of the Trump Administration, it is probably time to shift my attention to the ADP employment series. It has problems, and is different from the BLS establishment survey in its biases in ways I do not understand. But my knowledge base from an adult lifetime of learning the biases and quirks of the BLS establishment survey is about to be burned in a fire, isn’t it? And looking forward big disagreements between BLS and ADP have to be presumed to be something going very wrong at the BLS.

Here’s ADP payroll, monthly changes:

<https://fred.stlouisfed.org/series/ADPWINDMANNERSA#>

And here is the ADP monthly manufacturing change:

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#employment-growth-stalled
#macro-outlook
#macro
#chaos-monkey
#tariffs

A Note on the Unusual Procedures Surrounding the Expiration of Federal Reserve Gubernatorial Appointments...

Brad DeLong — Tue, 26 Aug 2025 20:58:21 GMT

...that I repost here, for reasons...

The Governors of the Federal Reserve have—uniquely—14-year terms putting them as far beyond any “unitary executive” control by the presidency that fantasists like C. Harvey Manfield and his epigones who wish the founders had sought to establish not a chief magistrate by a king, a tsar, a supreme sole autocrat might imagine in their drug-addled minds.

More interesting, right now, is that the congressional founders of the Fed envisioned that their terms might be considerably longer: §10 of the Federal Reserve Act says that: “Upon the expiration of their terms of office, members of the Board shall continue to serve until their successors are appointed and have qualified…” (12 U.S.C. §242).

<https://substack.com/profile/16879-brad-delong/note/c-149419205>

Were I General Counsel of the Federal Reserve today, I would issue an opinion. It would state that:

Donald Trump’s purported act to cause the premature expiration of Lisa Cook’s term “for cause” triggers this provision,
thus removal requires not just that Trump find “cause” but that he nominate and the Senate confirm a successor,
hence there is no need for any action to enjoin Trump;
any case is not yet ripe until the Senate confirms a successor,
any successor must then sue to require that Lisa Cook move out of her office,
failing Senatorial action, Lisa Cook remains a Governor of the Federal Reserve as long as she wishes, even unto the Last Trump, the Last Day, and the Resurrection,
perhaps she remains a Governor beyond: we know little of monetary-policy institutions in the New Jerusalem, after all.

And I would further state that there is thus no “case or controversy” which any federal court could use to elbow itself into the situation. There is no facially-qualified successor. Hence there is no one with standing to sue to evict Lisa Cook from her office.

“This is stupid!” you say.

I might agree, with the proviso that it is a lot less stupid than John Roberts’s rationale for his Medicaid decision in NFIB vs. Sibelius was.

And it is infinitely stupid a lot less stupid than the Four Fascists and the Two Cowards claim on their Shadow Docket that although “the Constitution vests the executive power in the President, see Art. II, §1, cl. 1, [so] he may remove without cause executive officers who exercise that power on his behalf, subject to narrow exceptions…” one of which is “the Federal Reserve [because it] is a uniquely structured, quasi-private entity that follows in the distinct historical tradition of the First and Second Banks of the United States…”

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#lisa-cook
#federal-reserve
#neofascism
#a-note-on-the-unusual-procedures-surrounding-the-expiration-of-federal-reserve-gubernatorial-appointments

Chaos Monkeys Trying to Fire Fed Governor Lisa Cook to Lower Interest Rates Succeed in Raising Them

Brad DeLong — Tue, 26 Aug 2025 11:49:10 GMT

Chaos monkeys chaosing as they caravan down Constitution Avenue from the White House to the Eccles Building. The attack on Fed independence spooked investors, widened yield spreads, & raised the risk premium on Treasuries. & in aiming at Lisa Cook they guarantee that the intellectual center of gravity of the FOMC will be less receptive to the technocratic arguments for lower interest rates…

Another Trumpist chaos-monkey own goal:

Ruth Carson, David Finnerty, & Alice Gledhill: Long US Bonds Fall as Threat to Fed’s Cook Stirs Inflation Worry <https://www.bloomberg.com/news/articles/2025-08-26/dollar-falls-with-treasuries-as-trump-seeks-to-oust-fed-s-cook?sref=rvrmfDby>: ‘Donald Trump intensified efforts to oust Federal Reserve Governor Lisa Cook, exacerbating investor concerns his attacks on the central bank’s independence and lobbying for lower interest rates will end up fanning inflation. The yield on 30-year bonds rose, reflecting the market’s expectation that price pressures could heat up if Trump were to succeed in replacing Cook with a policymaker more inclined to lower borrowing costs. The President has repeatedly complained that Fed Chair Jerome Powell and his colleagues have been too slow to cut rates. Trump’s move is a reminder that the US administration “remains unconventional and unpredictable,” said Andrew Ticehurst…. The push to dismiss Cook “is a good reminder that no institution in Washington can insulate itself from Trump’s bullying,” said Sarah Binder, political science professor at George Washington University. “Nor are claims of ‘independence’ sufficient to protect the Fed from Trump’s ambitions…”

And:

Jonnelle Marte & Myles Miller: Threat to Fed’s Cook Stirs Inflation Worry <https://www.bloomberg.com/news/articles/2025-08-26/dollar-falls-with-treasuries-as-trump-seeks-to-oust-fed-s-cook?sref=rvrmfDby>: ‘“Even if you don’t believe anything is going to come of this — that there won’t be any changes at the Fed — in the near term what it implies is, the risk premium for holding long-term Treasuries needs to go even higher,” Kathy Jones, chief fixed income strategist at Charles Schwab & Co, told Bloomberg TV…

Register this: The interest rates that matter for boosting spending are not the short but the long-term interest rates. Donald Trump wants spending boosted in the economy by lower interest rates, so he said a thing on Truth Social and sent a letter to Fed Governor Lisa Cook purporting to dismiss her that caused long-term interest rates to go up.

Truly a genius level move here. Congratulations, Bessent, Lutnick, Hassett, Miran, and company: Here we have proof that you really are as incompetent as the right-wing Washington and New York whisper machines have been saying.

There are currently twelve regional Fed Bank Presidents and, after Adriana Kugler’s resignation, six sitting Governors of the Federal Reserve—Powell, Jefferson, Bowman, Barr, Waller, and Cook. Of these, there is one who feels in her bones the strength of the case for a low-interest rate policy to generate labor-force upgrading in a high-pressure economy: Cook. If there is a case to be made for lowering interest rates right now, she is the most likely Governor to make it, and to make it persuasively. And if she is not making that case strongly, it is because it is not now a strong case. Why isn’t it? Because for the case to be strong the chaos monkeys need to be giving the Fed fiscal and supply-side policies that make reducing interest rates right now a prudent thing to do, and they are not.

If we are to rank current sitting Fed Governors in terms of their expertise and ability to make convincing arguments these days with respect to monetary policy, the ranking goes like this, starting from the bottom:

Bowman—out of her depth.
Waller—he definitely does have an affinity who say he is highly competent in his understanding of monetary policy, and I am sure that in private he is; but in public this year the arguments he has made have been very low-quality ones.
Barr—smart and expert in his specialties, but not a monetary-policy guy.
Powell—a superb manager and consensus-builder, a truly worthy Republican Worthy.
Jefferson—very sharp, but chosen for labor-market knowledge depth rather than deep exertise on monetary affairs.
Cook—yep, at the top; someone who has not just studied finance and monetary economics since her teens, focusing on the interplay between macroeconomic stability, financial markets, and long-run economic growth, with a central theme of hers being the integration of financial stability into mainstream macroeconomic analysis; and also someone who has lived through a hyperinflation, Russia between 1992 and 1994 as the ruble lost value at rates approaching 2000%/year.

I arrived at Berkeley as a professor just as Lisa Cook was finishing up her Ph.D. She was clearly smart as a whip, hard-working, had a broad range of knowledge and expertise casting a very wide intellectual net over the literature and the world, and able to actually get things done quickly and professionally.

Should Trump and the Trumpists succeed in moving her out and moving their own white guy in, they will have to draw from a pool of Trumpist sycophants and whatever Republican worthies remain who have no shame at all. That is probably a big pool. But from that big pool, there are very few who feel the strength of the case for a low-interest rate policy to generate labor-force upgrading in a high-pressure economy in their bones. Maybe zero? And even if there are some, I cannot believe that they will find even one who knows enough about the technocratic issues of monetary policy. At least, I cannot think of any. And should someone come forward with a Trumpist sycophant or shameless Republican worthy Fed-Governor candidate with (a) technocratic experience who (b) feels the high-pressure economy case in their bones, I would (c) give 9-1 odds that they are dumb as a post, and unable to make arguments in the policy-discernment scrum that is the FOMC that would convince anyone.

Plus (d) they will be rotten and tainted goods—everyone will know from the get-go that their arguments are pure bulls**t.

By contrast, Lisa Cook makes genuine and earnest points she believes in, intelligently formulated and thought through, based on a deep knowledge of monetary and financial affairs, from a perspective that feels the importance of maintaining a high-pressure economy in her bones.

If Trump were rational, if Bessent were rational, if Miran were rational, if Hassett were rational, one of them would have called a halt to this by now—would have said: “Wait a minute! This Governor is the one who is making our case within the FOMC discussions in a much better way than anyone we could replace her could! This one is our ally!”

And if the response was: “But she really isn’t making that case effectively enough, is she?” The rational counter would be: “That is because the case is not there to be made, and if it is to be made we need to first give her and the rest of the FOMC fiscal and supply-side policies that make reducing interest rates now prudent”.

To repeat myself: Lisa Cook is, by Trumpist lights, a better member of the Board of Governors than they could find to replace her, and a much better Governor than they deserve.

But if there is a rule about Trump’s chaos-monkey crew, it is that they are chaos monkeys.

And so they want interest rates to go down, and in an attempt to do so they try to push a better advocate for that case than they could appoint off the Board of Governors, and so the interest rates that matter go up.

Surprise, surprise.

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#chaos-moneys-trying-to-fire-fed-governor-lisa-cook-to-lower-interest-rates-succeed-in-raising-them
#lisa-cook
#federal-reserve
#chaos-monkeys
#neofascism

Trump’s Fed Chair Candidates: Sub-Mediocrity of Moral Character, at Best. & the Worst Plumb the Challenger Deep Along Several Axes

Brad DeLong — Thu, 14 Aug 2025 19:04:23 GMT

Why the next Federal Reserve Chair appointment may well be both the most consequential & the most badly compromised in advance ever. Intelligence, experience, and character: All in short supply among likely live possible picks…

Donald Trump once championed Jay Powell as the paragon of central banking. Now, with Powell out of favor and a slate of dubious potential successors on deck, the Senate’s role as gatekeeper would be vital, if there were a snowball’s chance in hell of there being enough sane Republicans for it to be able to actually, you know, do its job.

Of all the eight billion people in the world, eight years ago Donald Trump thought—strongly—that Jay Powell was the best person in the world to be Fed Chair. Hold on to that.

Today Donald Trump loathes Powell for, as best as I can see, doing his job. Trump has been applying pressure on him to imprudently and prematurely, given uncertainties and risks, reduce the policy interest rates the Fed controls. Trump has been “flooding the zone” with pressure along all axes, up to and including threatening to fire Powell for cause. What cause? The renovation of the Fed’s HQ has come in over budget, in substantial part because years ago Powell acceded to demands from Trump henchmen that the renovation utilize more marble and less concrete.

Now Donald Trump is about to nominate somebody else to be Fed Chair.

Looking at his candidates, the right thing for the U.S. Senate to do is to reject any of the likely candidates we now see. All of the candidates lack either the intelligence and analytical skills, the central-banking knowledge and experience, or the moral character to be a good candidate for the job. Appointments are made by the president with the advice and consent of the Senate. The Senate should—if it were to do its job—follow the lead of what its left-wing Democratic members will do when the nomination comes to the floor, which is to advise Donald Trump that they do not consent, and ask him to rethink. And given Donald Trump’s past inability, by his own lights, to choose a good Fed Chair, surely he should delegate the job of choosing one to somebody less inept at the task.

Robin Wigglesworth in the Financial Times cogently and concisely summarizes the same market and economist views that I am hearing:

Robin Wigglesworth: Rating & slating all the Fed chair candidates <https://www.ft.com/content/07e92a4b-f4ef-40b5-8172-eebc983f532f>: ‘Chris Waller…. The pretty obviously cynical jump from advocating for “more caution” on rate cuts in the autumn of 2024 to suddenly being the FOMC’s leading dove might have dinged his credibility, but Waller… is easily the favoured candidate of most investors…. And despite his overt keenness on the job, Waller would probably be welcomed by most of the Fed’s staffers as well….
Kevin Hassett… once mostly famous for co-writing the… “most spectacularly wrong investing book ever”… now primarily known as Trump’s premier economist cheerleader…. The increasing extremity of his Trump sycophancy has unnerved a lot of investors and analysts…. His modelling skills need work….
Kevin Warsh… briefly worked as an investment banker… served on the National Economic Council… George W Bush… nominated him for a Fed governorship in 2006 despite just being 35…. Investors and economists who Alphaville has spoken to are unimpressed… a lightweight, instinctive hawk who has desperately tried to reinvent himself as a dovish Trump devotee to win a job that on pure merit he probably shouldn’t be anywhere near. But he’s married to the daughter of Trump megadonor and fellow Greenland fan Ronald Lauder….
Steven Miran… might not be able to convince his FOMC colleagues to slash rates immediately, but his longer-term vision of a more ~~supine~~ accountable Fed is more than a little unnerving….
Miki Bowman… has also been making noticeably dovish coos lately… [to] be in the mix. Or at least [she] is very keen to be in it… joined Waller in voting for an interest rate cut at the July FOMC meeting…

When I look at Bowman, I see someone whose experience is too thin to be a credible candidate. And I see a definite lack of moral character. Her blaming house-price inflation in the United States on demand from the flood of immigrants was a definite low point. Another low point was her, last month, joining Waller in dissenting from the Federal Open Market Committee consensus.

The conventional norm for the past two generations, at least, is that while Fed Governors can quarrel with Chair decisions about interest rates, in the end they vote with the Chair. This norm arises because the voting members of the FOMC consist of seven president-appointed-senate-confirmed officers of the United States plus five of the twelve regional central-bank presidents who have an ambiguous private-public status. Things might become politically fraught, should the deciding vote on what is a government decision of monetary policy come from a person of ambiguous status. Hence the seven governors almost always form a united voting block on the FOMC. Mark Olson broke from this norm back in 2005. Since then, IIRC, no governor had—until Miki Bowman, last September, dissented. That dissent was because she thought Jay Powell was making too-accommodative monetary policy. Now she (and Waller) dissent because they claim they think that Jay Powell is making too-restrictive monetary policy. That is just not credible.

Strong moral character is important here. FOMC members should remember that, according to those I have talked to, 1970s Fed Chair Arthur Burns’s late life was shadowed by his constant memory of how he had not had the strength to resist pressures from the Nixon administration during the Great Inflation of the 1970s. And he did not, as a result of succumbing, even get the revenues from an appointment as Justiciar for Wales.

My opinion of Hassett is reasonably well-known. I think he makes the trifecta here—current Trump sycophancy showing negative moral character, double-counting of retained earnings as both investments and payouts showing negative intelligence and modeling skills as well, and clearly insufficient central-banking knowledge and experience. Miran is largely in the same boat—although there are people who claim to me that his modeling and analytical skills are head-and-shoulders above Hassett’s.

Warsh at least has the central-banking knowledge and experience. But Wigglesworth’s channeling of “investors[’] and economists[’]” opinons as “unimpressed… a lightweight, instinctive hawk… who… on pure merit… probably shouldn’t be anywhere near” the job rings very true to everything I know. A network-nepotist promoted well beyond his Peter-Principle level by George W. Bush’s posse who is the son-in-law of Trump affiliate Ronald Lauder. And the lack of moral character shown by his “desperately tr[ying] to reinvent himself as a dovish Trump devotee to win a job” would, in a well-working world, lead him to get zero senate-confirmation votes.

And so we are down to Waller as the best of a rather scurvy lot.

Waller’s affinity presses me that what Wigglesworth characterizes as his “pretty obviously cynical jump from advocating for ‘more caution’ on rate cuts in the autumn of 2024 to suddenly being the FOMC’s leading dove…” is the price he has to pay to stay in the mix, and that on central-banking knowledge, central-banking experience, intelligence, analytical chops, and even moral character—as his current bending the knee to Trump is his only flaw, and it is a flaw shared by very many now—he play in a much better league than any of the others. Point. Several points, in fact. His affinity is not wrong.

But in a good world, on a sane planet, the senate would tell Trump: You have a demonstrated a stunning inability to pick a good Fed Chair even by your lights, let alone anyone else’s. Eight years ago you thought very strongly that Jay Powell was the best person in the world to be Fed Chair. Remember that. So delegate this task, that you did such a lousy job on last time. Let the members of the Senate Banking Committee hold an open-hearings process, and then pick a truly excellent meritocratic candidate for your approval.

However, we are not on a sane planet, are we?

I hope for Waller.

I do greatly fear.

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#trump’s-fed-chair-candidates-sub-mediocrity-of-moral-character-at-best-and-the-worst-plumb-the-challenger-deep-along-several-axes

Macroeconomy Now Below "Stall Speed"?

Brad DeLong — Fri, 01 Aug 2025 17:25:32 GMT

Job creation slows, dementia-addled chaos-monkey tariff risks continue, data-center construction booms with little ability to rapidly move resources into that part of the construction-investment sector, and so the likelihood of stagflation in the near-term future rises. The U.S. economy is growing, but no longer fast enough to clearly outrun the shadow of rising unemployment, and yet inflation risks driven by supply and narrow sector-bottleneck shocks rise as well…

Job creation in the U.S. has slowed to a crawl, with recent months averaging just 35,000 new payrolls—a figure well below what’s needed to keep unemployment from rising. Yet inflation risks are not falling but growing, stubbornly present, amplified by dementi-addled chaos-monkey tariff-shock and other supply-chain uncertainties, plus sectoral-investment surges that may well run into inflation-producing bottlenecks. With Federal Reserve policymakers now publicly divided and the macroeconomy below stall speed, the risks of policy missteps and unexpected shocks loom large. The macro stakes are high.

This AM we have the U.S. Bureau of Labor Statistics:

BLS: Employment Situation Summary <https://www.bls.gov/news.release/empsit.nr0.htm>: ‘Transmission of material in this news release is embargoed until 8:30 a.m. (ET) Friday, August 1, 2025…. JULY 2025: Total nonfarm payroll employment changed little in July (+73,000) and has shown little change since April…. The unemployment rate, at 4.2 percent, also changed little in July. Employment continued to trend up in health care and in social assistance.… The unemployment rate has remained in a narrow range of 4.0 percent to 4.2 percent since May 2024…. The number of long-term unemployed (those jobless for 27 weeks or more) increased by 179,000 to 1.8 million… 24.9 percent of all unemployed people….
Revisions for May and June were larger than normal. The change in total nonfarm payroll employment for May was revised down by 125,000, from +144,000 to +19,000, and the change for June was revised down by 133,000, from +147,000 to +14,000. With these revisions, employment in May and June combined is 258,000 lower than previously reported…

With:

Over the past three month—only 106,000 new payroll jobs, seasonally adjusted; that is only 35,000 per month.
Since the end of last year—only 597,000 new payroll jobs, seasonally adjusted; that is only 85,000 per month.

(2) may, actually, not be a slow-enough rate of growth of payroll unemployment to put upward pressure on the unemployment raite and thus on the difficulty of finding a job. Jared Bernstein quotes Jed Kolko on this point:

Jed Kolko: 25-5 As US Population Growth Slows, We Need to Reset Expectations for Economic Data <https://www.piie.com/sites/default/files/2025-07/pb25-5.pdf>: ‘US population growth has slowed sharply in the past 18 months, as the immigration surge of the early 2020s has ended and the population continues to age. Fewer jobs are needed to keep up with the growth of the labor force…I estimate the breakeven rate of monthly payroll growth in the jobs report needed to keep up with the labor force has fallen from 166,000 jobs in early 2024 to 86,000 jobs in June 2025…

Smack-equal to what we have seen so far in 2025. An economy not at stall speed for the year as a whole so far, but worryingly close, and slowing down significantly over the past three months.

And Jared Bernstein—very sharp, but who, since he had worked for Biden since 2009, is one of those really ought back in 2022 to have been telling me to start pushing hard for a new candidate for president, and did not—summing up the sitch:

Jared Bernstein: Job Creation Slows To A Crawl: Monthly Avg: 35,000 over Past 3 Months <https://econjared.substack.com/p/job-creation-slows-to-a-crawl-monthly>: ‘The big news is the payroll revisions… “employment [growth] in May and June combined is 258,000 lower than previously reported."... [Plus} manufacturing employment is down three-months in row…. US manufacturing depends on imported inputs… [so] Trump’s escalating trade war exacerbates this…. Kicking up the effective tariff rate to somewhere around 18%… will hurt, not help, our manufacturers….
A few brighter spots: —Wage growth remains solid, at 3.9%, handily beating inflation. —Layoffs have been slowly trending up, but we’ve yet to see a spike that’s commensurate with severe labor market deterioration.
Chris Waller was one of the dissenters on the Fed’s decision earlier this week to hold interest rates where they are. One of Waller’s key motivators for cutting rates is his view that the labor market is weakening… “while the labor market looks fine on the surface… downside risks to the labor market have increased. With inflation near target and the upside risks to inflation limited, we should not wait until the labor market deteriorates before we cut the policy rate….” Chair Powell… took the other side… repeatedly describing the labor market as solid…. Today’s report obviously leans Waller’s way…. But… slower labor supply… deportations and considerably slower immigration….
Hard data—growth, jobs, inflation—are [now] starting to show what happens when the economy is relentlessly subjected to destructive economic policy…
Jared’s Substack
Job Creation Slows To A Crawl: Monthly Avg: 35,000 over Past 3 Months
I well know that today is yet another Liberation Day (I was talking about it on Morning Joe at 6am!), but it’s also jobs day, so let’s dig into the jobs data, recognizing that the two are related, as I stress below…
Read more
9 months ago · 33 likes · 7 comments · Jared Bernstein

A word about Waller: I think it unprofessional right now to assert, unqualified, that “upside risks to inflation are limited.” The recent history of U.S. inflation should make everyone humble: we have seen how quickly how what we judge small transitory shocks—whether from supply chains, energy prices, or geopolitical disruptions—can morph into much larger cost pressures self-sustaining for a time, and threatening to exit the transitory mode for the persistent one. Plus a professional central banker must always keep a weather eye on all potential sources of risk, and never unduly minimize any of them.

To declare, as Waller did, that such risks are “limited” is to attempt to lull markets and policymakers into complacency—precisely the opposite of what a prudent monetary authority should do. At the very least the standard of professionalism would require a nuanced and conditional framing: “While recent data suggest upside risks to inflation are moderating, those risks could return—particularly if supply constraints return or if aggregate demand surprises to the upside.”

I would have liked to see something like: If the enormous uncertainty about future tariff increases were to be removed, then upside risks to inflation would be significantly reduced. The message that the Fed should be sending to Donald Trump right now is this: if you are serious about lower interest rates, give us policies that make a lower interest rate path prudent.

Fed Governors, even tempted to go the extra mile in currying favor with Donald Trump, perhaps out of hope that their fealty will land them in the big chair at the Federal Reserve—really should behave better. Remember Arthur Burns in the early 1970s, whose willingness to accommodate the Nixon administration’s political imperatives did lasting damage to the Fed’s credibility and amply watered and fertilized the sprouting Great Inflation of the 1970s. The central bank’s power is mostly a power of narrative and expectation management. Once the perception takes hold that policy is being tailored to the whims of the incumbent or the ambitions of a would-be Fed Chair, little can be done, and what can be done quickly becomes very costly. Fed Governors must always remember who their true constituency is: not the president du jour, but rather the long-term economic and financial heualth of the republic. Now matter what a corrupt Supreme Court may claim, Fed Governors have their 14-year terms precisely because they are not instrumentalities to carry out the whims of an underbriefed chaos-monkey executive-branch president with dementia, but rather independent technocratic powers making decisions about coining money and regulating the value thereof as directed by the congress.

All that is to say: A false claim that upside inflation risks are limited is not helpful.

Today, the sources of potential upside risk to inflation are not low: chaos-monkey dementia-fueled tariff craziness continued, renewed energy shock, a reversal in globalization, an unexpected fiscal expansion, or, as we are already observing, the AI-driven data-center construction boom that is channeling a torrent of investment into the real economy. Central bankers need to never dismiss these risks out of hand, but to take the prudent course, which is to acknowledge uncertainty and to keep all options on the table, rather than to lull the polity into a false sense of security that may soon prove costly.

And yet, that said: Waller has, I think, a sound desire to become Fed Chair. rooted both in personal ambition and in a genuine conviction that, among the names likely to appear on a hypothetical Trump short list, he would be the most responsible and competent steward of U.S. monetary policy. In my view, he is correct. I would breathe somewhat easier, at a disaster (probably) averted, if Waller were to become Fed Chair next year. Waller is not like those whom one might politely call “mercurial” or “ideologically-driven” who might catch Trump’s chaos-monkey dementia-addled eye. He has the analytic and procedural norms that keep monetary policy on an even keel in his bones. The prospect of someone with his background and temperament at the helm is, in this fraught moment, a source of some comfort—if, admittedly, of only some and rather cold comfort.

A note on one of the things creating upside inflation risks right now:

Consider the sheer magnitude of the data-center construction boom fueled by the current AI bubble.

This is a phenomenon that, I think, deserves far more attention from both policymakers and market-watchers. The scale is staggering. Over the four years 2024-2027 we now expect to see more than $1.8 trillion spent building not bricks-and-mortar but rather generator, wire, silicon, plywood, concrete, pipe, and steel data centers worldwide. The hyperscalers and others are each pouring tens of billions of dollars into new facilities globally, in a frantic bid to secure the computational horsepower and energy infrastructure needed to train and deploy ever-larger natural-language models. This is not your garden-variety tech cycle; it is a capital expenditure arms race, and the urgency is driven less by exuberant optimism for future profits than by a defensive fear—fear that failing to provide customers with the lowest-latency inference and the highest-quality training will mean irrelevance in the next platform transition.

The result is a wave of demand for construction labor, steel, concrete, specialized semiconductors, and, most crucially, power—already putting upward pressure on input prices in regions like northern Virginia, Texas, and the Pacific Northwest. Historically, such sectoral investment booms have had the power to push up wages and prices even in the absence of broad-based consumer-demand surges. The AI data-center surge is not a sideshow: it is a macroeconomic force that could well become a significant contributor to inflation dynamics in the quarters ahead, and one that the Fed ignores at its peril.

The investment triggered by other bubbles died off as businesses saw themselves as unlikely to make profits from further investments, even with the extraordinary investment-financing terms that the bubble’s irrational exuberance offered them…

Not this time: these AI data-center investments are defensive insurance, not attempts to massively grow their profits. That is, unlike the dot-com or fiber-optic booms of the late 1990s, or the shale oil bonanza of the 2010s, the current wave of capital expenditure is less about chasing speculative upside than about preempting existential risk. The buidlers are not betting on blue-sky demand projections or the hope that “if we build it, they will come.” Instead, they are acting out of fear that, should they fail to keep pace in providing the lowest-latency inference and the highest-quality model training, they will be left behind in the next technological paradigm shift. This is primarily a race to avoid irrelevance and the loss of their current tech oligopoly-platform profits, not to seize new territory.

The result is a remarkable stickiness to investment, even as financing costs have risen and macro uncertainty abounds. The willingness to sign multi-year power purchase agreements, lock in supply chains for specialized semiconductors, and pre-commit to massive construction contracts all reflect this insurance logic. It is a dynamic that, I think, makes the potential inflationary potential of the current boom more persistent and less susceptible to the usual self-correcting mechanisms that have tamed previous bubbles.

No. Rather than believe that inflation risks are no longer worth considering, much better to recognize that there is more than a whiff of stagflation in the air:

Jared Bernstein: Inflation ticks up; real consumer spending still flat for the year <https://econjared.substack.com/p/inflation-ticks-up-real-consumer>: ‘Core PCE inflation… rose 0.3% in June and 2.8% over the past year…. Headline… up 2.6%…. Given that the tariffs are now clearly taking hold, I’m concerned about their direction of travel.… Core is stuck… overall… is ticking back up…. These are far from alarming rates… but… we are now seeing two quite challenging developments…. Trump’s trade war is escalating, with tariffs spewing out of the White House like toxic smoke from a dumpster fire, and two, we’re now seeing consumer price-passthrough at both anecdotal and statistical levels…. There’s a whiff of stagflation…. Growth for the first half of the year, at 1.2%, has been subpar…. Triffs, not to mention deportations that are particularly punishing to certain sectors… are taking a toll on prices and production…. These impacts are… gradual…

And the magnitude of this bubble-driven boom has a consequence. Normally I would react to a jobs report like this by stating that there is now at least a 50-50 chance of a near-term recession. But not this time. I cannot make any sort of likely recession call—not with this going on.

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#macro
#macro-friday
#federal-reserve
#macroeconomy-now-below-stall-speed
#on-the-federal-reserve
#on-the-ai-bubble

NOTE TO SELF: On Katie Martin's "Magnificent Seven" Today

Brad DeLong — Thu, 24 Jul 2025 07:05:57 GMT

Engines of growth reshaping the world, but then increasingly becoming rent collectors. The fortunes of the “Magnificent Seven”. What happens in the New Reality of Big Tech? And what are the answers to the natural questions about competition, equity, and the future of innovation?…

We have:

Aiden Reiter & Hakyung Kim: The Magnificent 7 growth slowdown <https://www.ft.com/content/3769fe15-f122-4fb5-874f-5ee48bca2bed>: ‘he market’s three-month recovery has been driven by Big Tech: From a market cap perspective, the Mag 7 is now 31 per cent of the S&P 500, also near an all-time high…. Nvidia[’s]… earnings growth in 2023 and early 2024 is too off the chart…. Tesla[’s]… earnings are all over the place and a clear outlier…. Investors have often punished Mag 7 stocks for earnings results that were great, rather than astounding…. Bank of America predicts the Magnificent 7 is only expected to keep its earnings growth edge over the rest of the S&P 500 for another 18 months: That deceleration is concerning…. Slowing earnings growth, and the high potential for bad guidance on tariffs and capex, will be a challenge…

The “Magnificent Seven”, in the Katie Martin-coined phrase, are: Nvidia (22% of the Mag 7), Microsoft (20%), Apple (17%), Amazon (13%), Google (12%), Facebook (10%), and Tesla (6%).

They are 31% of the S&P 500 with nearly $20 trillion in market capitalization.

Pause and let that sink in.

That is staggering. Their dominance is a testament to the extraordinary value creation (or is it value capture?) as we leave the globalized-value-chain era and enter the era of the attention info-bio tech economy. Scale, platform-construction, network-effect, and technological-moat effects have propelled a bare handful of firms to heights that would have seemed fantastical just a decade ago and are fantastical now. hat a mere seven firms now account for nearly a third of its value is both a marvel of innovation (but what kind of innovation?) and a striking concentration risk.

Over the past five years, they have delivered an astonishing average real annual return of 26%, in stark contrast to the more modest 7% average real annual return posted by the remaining 493 companies in the S&P 500. This yawning performance gap is the central fact of contemporary equity markets, driven as it has been by technological advance, technological disruption, market power, and—especially in the case of Nvidia—a near-monopoly on the infrastructure of the AI revolution. Meanwhile, the broader index has trudged along at a pace more consistent with historical norms. The 26% real annual return for the Magnificent 7 is not simply the result of clever management or product innovation—it is the mathematical outcome of a system in which the spoils of growth accrue disproportionately to those who control the key levers of technology and data.

The sources of value for today’s tech titans—especially the Magnificent 7—can be classified into four interwoven categories:

(i) genuine productivity gains, as these firms deploy cutting-edge technology to automate, optimize, and scale operations;
(ii) rent collection from their platform ecosystems, whereby network effects and user lock-in allow them to extract ongoing economic value from expanding value created by platform participants;
(iii) rent collection by “tightening the screws,” or leveraging their dominant market positions to squeeze suppliers, partners, or customers locked into their platforms—often through pricing power, restrictive contracts, or data control;
and (iv) the “gonzo meme stock” phenomenon, in which valuations are propelled less by underlying fundamentals and more by narrative, speculation, and the collective enthusiasm (or mania) of retail investors.

All four of these elements are present, in varying proportions, across the Magnificent 7, shaping both their current profitability and their future prospects.

To begin, the first source—genuine productivity gains—remains foundational, particularly for firms like Nvidia and Microsoft, whose technological breakthroughs in AI chips and cloud infrastructure have enabled entire new industries and made existing processes vastly more efficient. These productivity gains are quantum leaps in computational power, data processing, and software capabilities, which in turn drive broader economic growth.

However, even as these firms innovate, they also benefit from the second source: platform-based rent collection. Apple’s App Store, for example, extracts a significant share of revenue from every transaction on its platform, while Amazon’s marketplace fees and Google’s advertising auction model similarly skim value from vast digital ecosystems. This rent-collecting dynamic is amplified by network effects: the more users and developers an ecosystem attracts, the harder it becomes for rivals to compete, and the more entrenched the incumbent’s position.

The third source, “tightening the screws,” sees dominant firms use their market power to impose terms that maximize their own profits, sometimes at the expense of suppliers, partners, or even consumers. Nvidia’s recent pricing strategies for AI chips, for example, reflect a recognition that it faces little meaningful competition in the short term; Apple’s relentless negotiation with suppliers and its push to grow “services revenue” are similarly designed to extract every available dollar.

Finally, the fourth source—the gonzo meme stock phenomenon—is a product of the current market zeitgeist. Tesla’s valuation, for instance, has at times seemed detached from traditional metrics, buoyed instead by the cult of personality around Elon Musk and the willingness of retail investors to buy into a narrative of limitless disruption. This speculative fervor can create extraordinary volatility and, at times, disconnect the market price from underlying economic value.

These four sources of value interact and reinforce. And the result is a tech landscape in which profitability and market capitalization are driven as much by structural dynamics—platform lock-in, regulatory arbitrage, and speculative enthusiasm—as by traditional measures of productive contribution.

Nvidia & Apple are primarily a mesh of (i) & (iii): a blend of genuine productivity gains and aggressive rent-collection strategies. For Nvidia, its dominance in the GPU market—particularly as the indispensable supplier of chips for machine learning and large language model (MAMLM) applications—has allowed it to not only drive technological progress but also to extract economic rents from the entire AI ecosystem. Firms across industries now find themselves with little choice but to pay Nvidia’s premium prices if they wish to participate in the AI revolution, cementing Nvidia’s position as both innovator and gatekeeper. Apple, meanwhile, continues to improve the technical sophistication of its hardware, but the company’s corporate strategy has shifted: rather than focusing solely on product excellence, Apple now leverages its market power to squeeze suppliers and maximize the extraction of “services revenue”—from App Store fees to subscription bundles—transforming its ecosystem into a formidable rent-generating machine.

Together, Nvidia and Apple illustrate the duality at the heart of the current tech landscape: genuine innovation and productivity gains are inseparable from the structures of rent extraction and market power. Both companies have built platforms that are indispensable to their respective domains, but the rewards they reap are as much a function of their strategic positioning and ability to control chokepoints as of their technical prowess. For investors, this means that future returns may depend less on the next breakthrough device or algorithm and more on the continued ability of these firms to maintain their bottlenecks and enforce their tolls. For the broader economy, it raises questions about competition, innovation, and the long-term sustainability of such concentrated power.

Microsoft’s principal source of value in the current tech landscape is primarily (ii): its prowess as a creator and provider of the enterprise platform that makes IT easy for businesses, and then rents, particularly through its cloud-compute services, as demands for IT expand with technological progress and general economic growth. The company’s Azure cloud platform has become an essential utility for enterprises and startups alike, offering scalable computing power, storage, and AI tools without the need for firms to build and maintain their own infrastructure. By lowering the barriers to entry for digital transformation and enabling rapid deployment of machine learning and data analytics, Microsoft positions itself as an indispensable intermediary—one whose services are deeply embedded in the workflows of its clients. This strategic positioning allows Microsoft to extract ongoing economic rents from the growing value created by the platform community it hosts, as customers become increasingly locked into its ecosystem of software, cloud, and AI offerings. The rise of Azure is the core here. Microsoft’s cloud-compute services are not just a convenience for customers; they are the linchpin of a new platform-centric mode of value capture in the digital age.

Amazon, Google, and Facebook derive their remarkable profitability primarily from (ii) and (iii): (ii) rent collection from their platform ecosystems and (iii) rent collection by “tightening the screws” on those dependent on their platforms. Each has constructed a digital infrastructure that is not only central to the daily experience of billions but is also designed to capture a portion of the value created by every participant—be they advertisers, merchants, app developers, or content creators. Their platforms are not going anywhere, but also are not changing in terms of what people do on them that can be a source of rent extraction. Thus overall growth continues to enlarge the scale. Plus there is the tightening of the screws—what Cory Doctorow has taught us to call the “enshittification” of their platforms as we have to wade through more and more SEO- and now AI-slop to do what we were initially looking to do. Is anyone having more fun wading through Google or Amazon search results or Facebook ad-ridden feeds these days than trying to find something actually useful on Apple’s IOS app store?

Amazon’s e-commerce platform, for instance, charges third-party sellers a variety of fees for access to its marketplace, fulfillment services, and advertising tools, effectively taxing the commercial activity it enables. Google, through its dominance in search and digital advertising, has built an auction-based system where advertisers must outbid one another for visibility, with Google extracting a margin from every transaction. Meta, as the custodian of the world’s largest social networks, monetizes user attention and data, leveraging its platform’s reach to command ever-higher prices from advertisers while simultaneously altering algorithms and policies in ways that reinforce its own power and profitability.

And Tesla.

I should stop there.

Tesla is now selling 15% fewer vehicles than it was a year ago, and Elon Musk claims that future profits will not come from Tesla as an auto-producer but as an “AI” company—that the future of Tesla is robotaxis, and then, beyond that, humanoid robots. But how much of the profits from Elon Musk’s ventures into humanoid robots, robotaxis, and, indeed, natural-language interfaces and actually working self-driving for cars will go to the 60% Musk-owned xAI rather than the 13% Musk-owned Tesla? Back in The Day, the Big Four of the publicly-held Central Pacific Railroad Company of California—Leland Stanford, Collis Huntington, Mark Hopkins, and Charles Crocker—were also the four owners of the privately-held Charles Crocker & Co. construction firm. What company did the Central Pacific pay to actually build the railroad? Crocker& Co. Who decided on the price? The Big Four. How much of the value committed by the investors in the Central Pacific was tunneled out to the Big Four by their setting a high price for construction services? How much do you think?

Maybe Elon Musk will be generous to the outside shareholders of Tesla. Maybe it will be Tesla and not xAI that is the corporate form that profits most from Musk’s humanoid robots and robotaxis. If there are profits. Maybe.

But at least as I see it, the story of Tesla is of a meme-stock company. And the meme-stock aura surrounding Tesla is a double-edged sword. While it has enabled the company to raise capital cheaply, attract top engineering talent, and command a level of media attention unmatched by peers, it also introduces risks. The disconnect between valuation and fundamentals makes the stock susceptible to sharp corrections. Moreover, the focus on narrative over execution can obscure operational challenges.

Now, do not get me wrong about the Magnificent 7: the creation of their respective technology platforms—Apple’s iOS ecosystem, Microsoft’s Windows and Azure cloud, Google’s search and advertising infrastructure, Amazon’s e-commerce and AWS, Facebook’s social networks, Nvidia’s AI hardware stack, and, ahem, Tesla—are truly extraordinary leaps in wealth creation. They have delivered enormous societal benefit. These platforms fundamentally reshaped the digital landscape, enabling new industries, business models, and forms of communication, while driving productivity gains and consumer surplus on a scale rarely seen in economic history. The rise of the smartphone, the democratization of cloud computing, the global reach of social media, and the explosion of AI capabilities can all be traced back to the foundational work done by these companies in building and scaling their platforms.

But over the past five years, and looking forward:

The increased profits and soaring valuations of the Magnificent 7 are less a direct reflection of their own productive innovation and more an indicator of two intertwined forces: (a) the prosperity generated by real economic growth elsewhere in the economy, and (b) their enhanced ability to collect, scale, and extract economic rents from their dominant positions. As the broader economy has expanded—driven by consumer demand, capital investment, and global trade—these platform firms have positioned themselves at critical chokepoints, enabling them to capture a disproportionate share of the value created by others. Their business models increasingly rely on leveraging network effects, data control, and platform lock-in to extract recurring fees, commissions, and advertising rents, rather than on delivering ever-greater leaps in genuine productivity or consumer surplus. This shift is evident in the way Apple’s services revenue has outpaced hardware growth, Google’s ad auctions have become ever more lucrative, Amazon’s marketplace fees have multiplied, and Microsoft’s cloud subscriptions have locked in enterprise customers.

Plus, at present, the Magnificent 7 are deeply engaged in the build-out of massive artificial intelligence and machine learning model (MAMLM) infrastructure, a technological transformation that promises to generate enormous user surplus for those individuals and organizations who can effectively harness the capabilities of advanced GPT LLMs and similar AI tools. The productive potential here is vast: AI models can automate complex tasks, generate creative content, accelerate scientific discovery, and unlock new business models across industries. However, the economic gains from this transformation are likely to be distributed unevenly. While users who master these tools may reap significant benefits, the lion’s share of profits is poised to accrue to infrastructure providers—most notably Nvidia, whose GPUs and specialized chips are the indispensable hardware backbone of modern AI. The costs of building, training, and operating these models—especially the massive electricity bills for data centers—will further erode margins for most players, meaning that, for investors, the return on capital in the broader AI ecosystem may be modest, with only, cough, Nvidia, able to capture increasing profits at scale.

Moreover, what is good for the Magnificent 7 is increasingly no longer good for the broader economy. Once heralded as engines of innovation and productivity, these firms are now more frequently encountered by other businesses and sectors as collectors of costly economic rents rather than as providers of ever-more-valuable services. As their dominance has grown, the Mag 7’s interests have diverged from those of the wider marketplace: their business models rely on extracting value from their privileged positions atop digital platforms, supply chains, and data networks, often at the expense of smaller competitors, suppliers, and even consumers. This shift is evident in rising platform fees, more restrictive contractual terms, and the growing use of proprietary ecosystems to lock in customers and partners, all of which serve to transfer wealth from the rest of the economy to the tech giants’ balance sheets.

The Magnificent 7—once the undisputed engines of market returns and the very symbol of American technological dynamism—have become a study in contrasts. Their performance and narratives are now sharply differentiated, shaped by a mix of sector-specific challenges, competitive pressures, and the rapid pace of technological change. Some, like Nvidia and Microsoft, are riding the wave of artificial intelligence and cloud computing, while others, such as Apple and Tesla, are grappling with slowing growth, regulatory headwinds, or shifting consumer demand.

This divergence signals the end of an era when these companies could be treated as a single, unified trade or story. Investors and policymakers must now move beyond simplistic narratives. Instead, they need a more nuanced, case-by-case analysis that carefully weighs the true sources of value—distinguishing between genuine innovation and the extraction of economic rents. It is also crucial to assess the risks that come with market concentration and to evaluate the prospects for continued productivity growth in a landscape where the easy gains of platform dominance are fading.

References:

Barrett, Claer, & Katie Martin. 2024. “Transcript: The Five Minute Investor — Magnificent Seven.” Financial Times, April 23. <https://www.ft.com/content/6c9a99d6-b1ae-4d60-a682-438985986637>.
Reiter, Aiden & Hakyung Kim. 2025. “The Magnificent 7 Growth Slowdown.” Financial Times, July 22. <https://www.ft.com/content/3769fe15-f122-4fb5-874f-5ee48bca2bed>.
Rosen, Phil. 2025. “Magnificent 7 Outlook: Wall Street’s Favorite Big Tech Trade Is Splintering.” Opening Bell Daily, July 23. <https://www.openingbelldailynews.com/p/earnings-outlook-magnificent-big-tech-investors-trump-nvidia-microsoft-tesla-apple-goog>.

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#on-katie-martins-magificent-seven-today

Policy Uncertainty, Not "AI"-Automation, Is Almost Surely Behind the Bulk of Recent Graduates' Job Discontent

Brad DeLong — Wed, 23 Jul 2025 18:53:40 GMT

Why are new college graduates finding it harder than usual to get a foot in the door, even as unemployment rates stay low? Do not (yet) focus on “AI”…

Whatever is making new college graduates these days think that their life is unusually difficult for their labor market segment, it is almost surely not “AI”. Amanda Mull:

Amanda Mull: What the Tough Job Market for New College Grads Says About the Economy <https://www.bloomberg.com/news/articles/2025-07-17/tough-job-market-for-new-college-grads-is-worrying-for-us-economy>: ‘Ernie Tedeschi… [says] hiring rates for new grads are… in line with the latter half of the 2010s…. “We still see low unemployment, and we do see pretty solid job gains,” says Allison Shrivastava, an economist at… Indeed…. But it… isn’t as lush as it was in the recent past…. [And] graduates in computer science, computer engineering and graphic design all have unemployment rates at 7% or greater…. Nathan Goldschlag… [says] the new-grad unemployment rate… “is low for things like accounting and business analytics, which… when you’re coming from the AI space… are ripe for automation”….
Stochastic uncertainty…. Companies are waiting as long as they can… in hopes of catching a glimpse of what tariffs and AI and inflation and anti-immigration policies will do to their business, which means many are delaying hiring. “Everything is just kind of stalled out and frozen,” Shrivastava says…

And Paul Krugman:

Paul Krugman: Bad Times for College Graduates <https://paulkrugman.substack.com/p/bad-times-for-college-graduates>: ‘What we’re looking at now isn’t the worst job market college graduates have ever seen. It is, however, the worst such market compared with workers in general that we’ve ever seen, by a large margin…. Young workers always have higher unemployment than workers as a group. College graduates always have lower-than-average unemployment. But normally education trumps age: Even recent college graduates have relatively low unemployment. But not now…

Although “bad” is relative, relative to other groups, to other countries, and tro previous booms only. Plus things do indeed still look very good in the health-care sector. The aging of the American population, combined with rising demand for mental health and other medical services, has kept job growth in health care robust, even now. But health care is only 1/6 of the economy.

Derek Thompson <https://www.theatlantic.com/economy/archive/2025/04/job-market-youth/682641/?gift=o6MjJQpusU9ebnFuymVdsJ1qwI70CnAkjDXBfrYqvHw> reports that the gap between total and recent-grad unemployment went from 3%-points in the depressed economy of 2012 down to 1%-point in the mid-2010s, which was markedly lower than the 1.75%-points typical of the 1990s. The gap went negative even before the plague. From 2014 to 2024, with zigs and zags, it fell from +1%-point to -1% point—a fall of 0.2%-points per year—before rapidly falling to nearly -2%-points today.

In tech, it actually may well be “AI”, or at least have an “AI” component,.

Tech has the most “AI” true believers by far. There are tech bosses saying, but we do not really knwo how representative they are, that workers should see this year whether ChatGPT instantiations can be their interns. Why pay for a junior analyst to draft reports or summarize documents when a large language model can do it in seconds? But similar fears have accompanied every wave of technological change, from the spreadsheet’s arrival in the 1980s (which, some predicted, would eliminate the need for accountants) to the earlier automation of switchboard operators and typists in the mid-20th century. The reality is that while some tasks are indeed automated away, new roles and new forms of work tend to emerge, though not always at the same pace or for the same people.

More important, probably, is that money that would go to new hires is instead going to buying NVIDIA chips. In the current tech boom, companies are pouring vast sums into the hardware that powers artificial intelligence—most notably, the high-performance graphics processing units (GPUs) produced by NVIDIA. These chips are the backbone of machine learning and generative AI, and demand has been so intense that NVIDIA briefly became the world’s most valuable company. For firms, the calculus is straightforward: Investing in AI infrastructure is seen as a ticket to future competitiveness, while hiring junior staff is a cost that can be postponed. The opportunity cost, however, is that young people seeking a first job may find doors closed—not because their skills are obsolete, but because capital is being allocated elsewhere. For a college freshman, this is a reminder that macroeconomic trends and corporate priorities—often far removed from undergraduate coursework—can shape the contours of the job market in unpredictable ways.

Thus there is still now hard and not even a semi-convincing soft narrative that “AI is to blame” for entry-level job scarcity. It seems to be almost surely, still, merely a convenient scapegoat. It is tempting, especially for the media and for anxious graduates, to pin the blame for a tough job market on the rise of artificial intelligence. After all, stories about robots taking jobs are both dramatic and easy to understand.

But hiring slowdowns are driven by broader forces: economic uncertainty, shifts in business investment, and the cyclical ebb and flow of demand. Blaming AI allows both policymakers and business leaders to avoid grappling with deeper, structural issues—such as the mismatch between what colleges teach and what employers need, or the long-term stagnation in productivity growth that has made firms more cautious about expanding payrolls, or short run policy uncertainty.

It is here that you want to place your bets on what is going on: Policy uncertainty—over trade, immigration, inflation, and technology—is driving the short-run piece of the jobs cycle. It has paralyzed business planning, reinforcing a cycle of hiring freezes and risk aversion that hits new entrants hardest. When companies face an unpredictable policy environment—say, uncertain tariffs on imported goods, shifting rules about work visas, or volatile inflation—they tend to delay major decisions, including hiring. A tech firm unsure whether it will be able to recruit international talent may postpone expanding its junior staff. Similarly, a manufacturing company facing the threat of new tariffs might hold off on bringing in new graduates until the dust settles. This risk aversion is particularly damaging for those at the start of their careers, who rely on a steady flow of entry-level openings to get a foot in the door.

The deeper sclerosis is this: risk aversion among both employers and workers, leading to fewer opportunities for mobility and advancement. In a healthy labor market, people change jobs frequently, chasing better pay or more interesting work, and employers are willing to take risks on new hires. Today, by contrast, both sides are playing it safe. Workers are staying put, afraid that if they quit, they won’t find something better—a sentiment reflected in historically low quit rates. Employers, meanwhile, are reluctant to hire or promote, preferring to “wait and see” rather than invest in untested talent. This mutual caution creates a feedback loop: With fewer people moving, fewer positions open up, and the market becomes stagnant. Job-change rates are abnormally low on both accessions and separations. “Just get your foot in the door” is much less effective when doors are not opening as quickly or as widely.

For the longer-run, the rise in the college wage premium is over, and a decline has (probably) begun. For decades, earning a college degree was a reliable ticket to higher earnings, as the labor market rewarded those with advanced skills and credentials. But in recent years, this “college wage premium”—the extra money earned by graduates compared to non-graduates—has plateaued and may even be falling. The reasons are complex: The supply of degree holders has grown faster than the demand for high-skill jobs, and technological change has not produced a corresponding surge in new, well-paying roles for graduates. The value of a degree is still strongly positive. But it is no longer increasing.

Plus: Today, the dominant corporate ethos is flexibility: Firms prefer to hire for specific needs, often on a temporary or contract basis, and are less willing to invest in employee development. This shift is driven by competitive pressures, shareholder demands, and the rise of management philosophies that prioritize efficiency over loyalty. For students entering the workforce, this means that career ladders have become career lattices—less predictable, more fragmented, and requiring more self-direction and networking to climb.

Bottom line: Unemployment is low overall. New college graduate unemployment is not high. But, compared to others, America’s newest graduates are finding it relatively harder to find jobs than they ever have before. Blame corporate reluctance to invest in talent, the start of what may be a long-term shift away from steady rising wage premiums for college degree-holders, and—likely most important—unprecedented policy volatility, and thus unusual risk.

That means weaknesses in young workers’ educational pedigrees are likely to shadow them throughout their careers. Those who start out behind—whether because of a less prestigious college, a lower GPA, or fewer internships—will find it especially difficult to catch up or leapfrog, even as their actual skills improve over time.

So pay attention in junior high school algebra, kids!

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#policy-uncertainty-not-ai-automation
#paul-krugman
#amanda-mull
#mamlms
#labor-market

Saturday Macro: Fed Independence as Your Screentime-Limit Phone App

Brad DeLong — Sat, 19 Jul 2025 12:57:17 GMT

I laugh at a metaphor from Joe Weisenthal. Legal independence is no shield against political winds—but so far this is one of the very few things Trump is doing that is still benign...

Joe Weisenthal quotes the wise Carola Binder:

Carola Binder (2020): De Facto & De Jure Central Bank Independence” <https://www.suerf.org/wp-content/uploads/2024/01/doc_f2217062e9a397a1dca429e7d70bc6ca_11095_suerf.pdf>: ‘Legal CBI… is virtually orthogonal to frequency of political pressure. And there are plenty of examples of central banks with strong legal CBI that face political pressure…. I is political pressure on central banks, rather than legal independence, that seems to have more explanatory power for inflation. I find that even when the central bank reportedly resists the political pressure, inflation still tends to rise following an episode of pressure. Perhaps even if the central bank does not change its monetary policy in response to political pressure, the pressure damages central bank credibility. The possibility that it might respond to pressure leads to higher inflation expectations and in turn to higher inflation…

And Joe writes:

Joe Weisenthal: <https://www.bloomberg.com/news/newsletters/2025-07-15/central-bank-independence-is-a-spectrum>: ‘Kevin Warsh, a longtime hawk who is now seen as a possible successor to Jay Powell, is suddenly a fan of rate cuts. FHFA Bill Pulte has been promulgating unfounded rumors that Powell might resign, while also criticizing him for not having cut rates sooner. OMB Director Russ Vought has slammed Powell for the Fed’s expenditure on office renovation. And of course Trump himself has criticized Powell’s rate policy repeatedly…. The Fed functions,,, as a somewhat independent body in Washington DC. But its status is still something like that screentime app, where the political will or desire to actually abide by these self-imposed institutional guardrails is starting to decay…. This could eventually have undesirable macro-economic effects…

<https://fred.stlouisfed.org/series/PCEPILFE#>

The question with respect to Kevin Warsh is, with respect to whether he actually becomes Federal Reserve Chair, who will he then double-cross? Will he double-cross his affinity, to whom he is well-known as a reassuring hard-money guy. Or will he double-cross Donald Trump as he attempts to fill the shoes of Alan Greenspan?

And I would say that the political pressure has already been effective. Under Biden, there was a very solid FOMC majority that thought it was worth running substantial risks to try to ensure that the inflation rate got and stayed on a path to kiss the 2%/year PCE-index inflation target. Since Trump’s election the average for the core PCE inflation rate has been 2.7%, and there seems to be no enthusiasm on the committee to push it any lower. They see themselves as now having much bigger fish to fry than 2% vs. 2.5% vs 3.0%.

Mind you, I think this is a good thing. I have long thought that the 2%/year target was too low, and that the “credibility” you gained by sticking to it was a reputation for being unwilling to do the clearly right thing out of some combination of stubbornness and pride, and that was not the kind of reputation that you want to purchase. The danger I see is that the Federal Reserve will not hold the target line below 5%/year, or that markets will not believe that the Trump-pressured Fed will hold the target line below 5%/year.

This is an especially good thing because I think we have to look forward to a substantial downward shift in our real GDP numbers. Back before last November we had confidence that the trend real GDP growth number going forward would have a “3” as its first-digit handle. Now we will be lucky if it does not have a “1”: the “crackdown” on immigrants, the damage done to American science, and the breaking of production networks as the U.S. is forced into substantial decoupling from the globalized value-chain economy are all going to be substantial headwinds. Debt amortization burdens wreak havoc when nominal GDP growth is low. And now that real growth is likely to fall off, we need to stay above 2% per year inflation more than ever.

References:

Binder, Carola. 2020. “De Facto & De Jure Central Bank Independence”. In Ernest Gnan & Donato Masciandaro, eds. Populism, Economic Policies, & Central Banking. Vienna: SUERF, the European Money and Finance Forum. <https://www.suerf.org/wp-content/uploads/2024/01/doc_f2217062e9a397a1dca429e7d70bc6ca_11095_suerf.pdf>.
Weisenthal, Joe & Tracy Alloway. 2025. “Central Bank Independence Is a Spectrum.” Bloomberg Odd Lots Newsletter, July 15. <https://www.bloomberg.com/news/newsletters/2025-07-15/central-bank-independence-is-a-spectrum>