Category: Investing

Is the economic/stock market boom from Artificial General Intelligence already behind us because the advent of AGI is already behind us?

by philg

A PhD physicist friend on Facebook:

The reason I’m all-in on AI investment, and there may not be a top to this “bubble”: Unlike tulips, railroads, or internet connectivity, there is no upper bound on the value of intelligence.

Me:

30-year TIPS yield is 3% real currently. So your perspective isn’t shared by the market. Investors wouldn’t lend money to the federal government at 3% (after inflation) if they thought there was no limit to NVIDIA’s value. If investors overall thought that there was a low-risk way of making super high real returns over the next 30 years they the government would have to pay 8% real, for example, if buying an AI index was expected to yield 12% real for the next 30 years.

I think AGI is already here from the perspective of most users of ChatGPT and similar. A generally intelligent human isn’t great at everything and makes mistakes… just like ChatGPT! People ask ChatGPT all kinds of questions and give the answers at least the same weight that they would give to an answer from a typical human. In that sense, ChatGPT has passed a Turing Test for general intelligence. Maybe Advanced Superintelligence is already here. The typical human is not good at mathematics. A math professor could have been described as “super intelligent” before WWII. LLMs are supposedly doing all kinds of advanced work in mathematics right now, even if they might fail at a plumbing task. The math professor in 1935 who was a failure at plumbing would still have been considered superintelligent, right?

In other words, we can’t expect hockey stick growth for the economy due to AGI/ASI arriving because the current growth is already an example of what an economy does with the gift of AGI and ASI (but maybe not robotics!).

Will the $40 trillion in federal debt be a drag on economic growth? Scott Bessent says “no”:

Let’s use a 20-year time horizon for the U.S. to potentially get out of Argentina territory (150% debt-to-GDP max; we’re at over 125% right now). For US debt to GDP to fall to the level of a high IQ society (e.g., Taiwan, which has lower-than-US tax rates and debt of about 20% of GDP), GDP growth would need to be 10%/year real for 20 years with Congress not borrowing any more money (the latter condition seems unlikely to be met, since Congress now borrows even in the most robust economies, contrary to Keynes). Investors plaintly don’t believe that this will happen because they’re willing to lend to the Feds at 2.75% real (20-year TIPS current price) and they wouldn’t do that if investing money in domestic stocks would generate a roughly 12% real annual return (real GDP growth plus 2% as a return on investment from corporate earnings).

Does this mean that we’re in an AI bubble? Not necessarily. Only that AI by itself apparently doesn’t hugely lift the overall U.S. economy (a huge part of which is government spending/welfare state!). We’ve got about 1.5% per capita real GDP growth right now. Maybe that includes the AI lift? This NBER paper by a Nobelist (sort of) predicts minimal per capita growth, but cites estimates as high as 3.4 percent per year as the boost (nowhere close to the 10% we’d need to get out debt down to Taiwan’s relative level):

(He cites McKinsey, the giant brains behind Enron!)

The only way to make $40 trillion in debt insignificant, therefore, would be to grow the U.S. population to about 1 billion humans at roughly the same level of skill/income as the current U.S. population. Until Donald Trump showed up (again!), our wise politicians were working on this, but they forgot to apply and skills test for immigrants.

(In case this blog post is going to be a source for an NPR or PBS story (example), let’s not forget that both Turing and Bessent were/are members of the 2SLGBTQQIA+ community.)

Full post, including comments

Even companies worth $1 trillion don’t want to pay Nvidia’s prices

by philg

I hope that everyone has made tons of money taking my investing advice and doing the opposite. Loyal readers may remember that I’ve been skeptical of Nvidia becoming worth more than all of the European Union combined and, therefore, never purchased NVDA (current market cap: $5+ trillion). My theory was that 50 percent profit margins for Nvidia would motivate other companies to develop chips and then the AI nerds would recompile their code (or ask AI to recompile their code). Google did this a long time ago with its Tensor Processing Unit. Intel tried and apparently failed with Gaudi (but they were very successful in DEI!). Elon Musk has his AI5 chip and, eventually, his own fab.

OpenAI is the latest entrant into this mix, apparently. “OpenAI Claims Its New Chips Can Outperform Nvidia Processors in Tests” (Bloomberg, yesterday):

The company plans to start using the new chips to support its artificial intelligence models later this year, which should significantly reduce costs for OpenAI as it rolls out more widely.

OpenAI, losing enough money every year to be worth almost $1 trillion, created Jalapeno in a partnership with Broadcom Inc., which makes custom chips for a variety of clients. The two companies, which announced their pact last year, touted the speed at which the processor was developed in June, saying it came together in record time.

Even if this thing succeeds, there will still be a huge market for Nvidia, apparently:

Jalapeno wasn’t tested against the new generation of Nvidia chips, Vera Rubin, which just began shipping. It’s also not designed for training AI models, an area where Nvidia technology excels. Jalapeno is intended for the inference phase of AI — the stage when models have already been trained and can begin responding to prompts and handling tasks.

(Why is the chip named for Vera Rubin, someone who was neither an electrical engineer nor a software developer? Google AI:

Honoring Women in STEM: The company has a specific practice of naming key platforms after pioneering women scientists. Previous examples include Grace Hopper (the Grace CPU) and Ada Lovelace (the RTX 40-series architecture)…

I guess we can’t expect the William Shockley CPU/architecture any time soon, despite the man’s Nobel for the invention of the transistor, but why not one for Jack Kilby? Kilby died in 2005, prior to the full flowering of Rainbow Science, and therefore Kilby’s gender ID is unknown (we could say the same regarding Vera Rubin, who died in 2016, some years before the trans-enhanced Rainbow Flag flew at the White House))

How much do journalists and editors need to know about electrical engineering in order to cover cutting-edge AI hardware?

Because the chip is achieving strong results at low voltage — 700 watts — Jalapeno will allow OpenAI to save money when running its data centers, where power is a key cost, he said.

Speaking of high voltage, let’s have a look at the cooling system for Google’s latest TPUs (blog entry, which Google’s massive AI brain couldn’t figure out should have a date on it; I think it is from April 2026):

It looks like they got retired Pratt PT6 engineers to work on this!

(Actually, this suggests a career path: instead of fighting it out on the crowded software front, a young person could study to become a packaging/cooling engineer.)

Vaguely related… let’s see if we can figure out where Nvidia’s passion for “Honoring Women in STEM” comes from. Perhaps nerds identifying as “women” were the principal designers of Nvidia’s hardware and software?

Prompt to Gemini (to run on Google’s TPUs so that the results aren’t biased by running on Nvidia chips):

Who were the principal engineers of Nvidia’s historically important GPUs and software architectures, including CUDA? In other words, who are the people who are primarily responsible for making Nvidia as valuable as it is today?

Answer (not in quote style):

Key Technical Leaders Behind Nvidia’s Rise

1. Curtis Priem & Chris Malachowsky (Co-Founders & Founding Hardware Architects)

  • The Role: Co-founded Nvidia in 1993 alongside Jensen Huang.
  • Their Contribution: Priem was the chief architect of Nvidia’s early graphics chips (including the NV1 and the breakthrough RIVA 128), while Malachowsky led system engineering and hardware design operations.
  • Impact: They established Nvidia’s core hardware engineering practices and designed the early pipelines that allowed Nvidia to survive the “3D accelerator wars” of the late 1990s and create the original GeForce 256—the chip Nvidia marketed as the world’s first true “GPU.”

2. Dr. David Kirk (The “Father of CUDA” & Chief Scientist 1997–2009)

  • The Role: Chief Scientist and VP of Architecture.
  • Their Contribution: David Kirk was the primary visionary behind transforming Nvidia’s fixed-function 3D graphics hardware into programmable, general-purpose parallel processors.
  • Impact: In the mid-2000s, Kirk persuaded Jensen Huang to mandate that every single Nvidia GPU include dedicated general-purpose execution units—a massive financial gamble at the time. This decision laid the groundwork for CUDA (Compute Unified Device Architecture), launching general-purpose GPU computing (GPGPU) and making GPUs usable for scientific computing and later AI.

3. Ian Buck (VP of Accelerated Computing & Original CUDA Architect)

  • The Role: Creator of the BrookGPU programming language (at Stanford) who joined Nvidia to lead the CUDA project.
  • Their Contribution: While pursuing his PhD at Stanford, Buck created Brook, an extension of C for streaming coprocessors. Jensen Huang hired Buck to lead the software and systems team that built the original CUDA software stack and runtime environment, released in 2006.
  • Impact: Buck turned the raw, complex parallel hardware engineered by Kirk’s team into an accessible software platform for C/C++ developers. Without Buck’s software stack, AI researchers in 2012 (like Alex Krizhevsky and Geoffrey Hinton) would not have been able to train AlexNet on Nvidia GPUs—the spark that triggered the modern deep learning boom.

4. John Montrym (Chief Architect, GeForce/Tesla Architecture)

  • The Role: Senior Architect and Fellow.
  • Their Contribution: Montrym was a key architect behind the NV40 (GeForce 6 series) and the landmark G80 (Tesla architecture) released in 2006.
  • Impact: The G80 was the first unified shader architecture in a PC GPU and the hardware home where CUDA was born. It abandoned separate vertex and pixel pipelines in favor of unified streaming multiprocessors (SMs), setting the blueprint for every modern GPU architecture designed since.

5. Jonah Alben (Senior VP of GPU Engineering)

  • The Role: Longtime leader of Nvidia’s hardware silicon development.
  • Their Contribution: Alben joined Nvidia in 1997 and has spent over two decades directing the physical architecture, design execution, and engineering teams behind generations of GPU microarchitectures: Fermi, Kepler, Maxwell, Pascal, Volta, Ampere, Hopper, and Blackwell.
  • Impact: Alben’s team managed the scaling and execution challenges as GPUs grew from millions to over 200 billion transistors on single silicon packages. His team integrated specialized hardware engines directly into the GPU pipeline—most notably Tensor Cores (introduced in Volta/Ampere) and Transformer Engines (in Hopper/Blackwell)—which gave Nvidia a 10x-100x efficiency advantage in AI workloads over traditional CPUs and competing chips.

6. Dr. Bill Dally (Chief Scientist & Head of Nvidia Research)

  • The Role: Former Chairman of Stanford’s CS Department, appointed Nvidia Chief Scientist in 2009 (succeeding David Kirk).
  • Their Contribution: A pioneer in computer architecture and high-speed interconnect networks, Dally leads Nvidia’s long-term research division.
  • Impact: Dally pivoted Nvidia’s research toward optical interconnects, low-precision AI math (FP8, FP4 formats), ray tracing hardware cores (RT Cores), and multi-chip network scaling (NVLink). His architectural research predicted the massive bandwidth bottlenecks of large AI models years before ChatGPT emerged.

You can see the full answer and also one from Grok.

Fun fact from Oshkosh/AirVenture 2026: I ran into a mid-level Nvidia employee. He had shown up in his new airplane, a perfectly restored P-51 Mustang ($5 million?).

Full post, including comments

New York Times says it is time to stiff the bondholders out of their $40 trillion

by philg

“An Ancient Sumerian Solution to Our $40 Trillion Deficit” (yesterday):

The U.S. federal debt has hit $40 trillion. Add the debt owed by U.S. states, corporations and consumers and the figure rises to about $77 trillion in debt, set against an annual G.D.P. of about $32 trillion. The interest on all of it is compounding constantly.

Ancient societies had another method to deal with debt. It was called an amargi — a blanket declaration of public debt cancellation. All public debts written off. Disappeared. It sounds laughable, I know. But, really, that’s just because the idea has been buried so deeply in history you’ve probably never heard of it. In the ancient world, it presented a pragmatic solution to an intractable problem. And now, faced with impossible-to-repay debts that are weighing down our economy, is the time to look at the amargi and the lessons it offers about how to think about finance.

The amargi was a response to a persistent problem that was recognized in the ancient world: Debt compounded until it destabilized society.

Understatement Department:

People who were owed money were, predictably, not always happy about the amargi.

We invented it so we can de-invent it:

The reason the practice often worked in the first place was because the ancient world understood something about our monetary system we have mostly forgotten: Money is an invented social construct. It isn’t real, not in the way a tree or a stone is real. The system of money and credit is a thing humans made up. It’s a record-keeping device for distributing resources. And since money is a human creation, we can alter it when needed.

What would happen if the U.S. were to stiff all of the bondholders, most whom are domestic (breakdown)? Would the stock market collapse because of the evaporation of money people thought that they had saved? A lot of stocks would go to zero, presumably, because a bank or insurance company could go insolvent due to the evaporation of reserves. Maybe some stocks would go up because workers wouldn’t be quite as burdened with federal taxes and, therefore, would have more spending power? That seems unlikely because our fiscal deficits are so high that ceasing interest payments on federal debt still wouldn’t result in a balanced budget (i.e., we would need higher taxes to fund the government services that we demand in a world where the federal government doesn’t/can’t borrow, even after kissing $40 trillion in debt goodbye). This is kind of remarkable. Imagine a family that is so addicted to spending that even if it is magically freed from paying a mortgage, a car loan, or any of its past credit card bills that it still can’t fund the lifestyle to which it has become accustomed and feels entitled.

If this does happen, I guess I’ll be nominally poorer (stock market collapse), but I’ll have some bragging rights because I’ve been a hater of bonds as an investment ever since the Jimmy Carter era inflation (more properly attributed to Lyndon Johnson’s Great Society program, but Carter got the blame because hyperinflation took a while to build steam).

Here’s a misleading chart, by the way. It says “total public debt” is $40 trillion, but doesn’t count debt incurred by state and local governments (also “public”). Supposedly this is only about another $4 trillion, but that doesn’t include state and local obligations to retirees, such as pensions and health insurance (impossible even to guess at what those are because we don’t know how long humans will live).

Full post, including comments

SpaceX lockup expiration today will create a lot of marriages and divorces?

by philg

How’s SpaceX stock doing today? Up 6% says the Google? Even if it had sagged to Morningstar’s number (see SpaceX is worth $63/share according to the most expert experts), there would still be a lot of newly minted rich people due to the expiration of employee lockups. It’s almost as though they won the lottery, which raises the question… Can one predict the effects of winning the lottery?

Four academics have tried:

We estimate the effects of large, positive wealth shocks on marriage and fertility among Swedish lottery players. For male winners, wealth increases marriage formation and fertility and likely reduces divorce risk. For female winners, the only clear effect is a higher short-run (but not long-run) divorce risk. The gendered treatment effects are often similar to income gradients. Our findings align with models in which wives derive greater marginal utility from consumption when single than when married, whereas the opposite holds for husbands—but only if marital property is divided unequally upon divorce. We provide descriptive evidence consistent with such unequal division.

If the academics are correct, we should expect wedding bells for some of the male employees of SpaceX and that the female employees will be filing more divorce lawsuits than they otherwise would have.

(The “if marital property is divided unequally” phrase in the above, I think, refers to prenuptial agreements in which property is kept separate. Sweden’s divorce law allows for the possibility that a lottery prize could be joint property, in which case a female partner of a male winner could cash out via initiating a divorce and also the possibility that a man’s lottery prize could belong to him alone, in which case the female partner’s only way to spend some of it would be to stay married.)

Related:

Full post, including comments

SpaceX is worth $63/share according to the most expert experts

by philg

How’s SpaceX doing on its second trading day? The most expert experts on stocks in the United States work at Morningstar. They value the company at $63/share and say that if everything went perfect, it could be worth $154. From last week:

We value SpaceX SPCX at $63 per share, a 53% discount to the upcoming IPO’s offering price. Our valuation is the result of mathematics more than skepticism. With such a wide range of possible outcomes for the company’s financial future, we created forecasts and valuations for three scenarios and probability-weighted them.

Even at $63 per share, we give SpaceX a lot of benefit of the doubt in two of the three scenarios, in which we assume the company can achieve a rapidly reusable Starship rocket enabling multiple launches per week and successfully commercialize data centers in space. Neither of these engineering problems has been solved, and we don’t expect them to be until at least 2028.

In our most optimistic “moonshot” scenario, the company would be worth $1.97 trillion, or $154 a share. That’s 14% above the offering price and a level the shares might even reach in the short term after their public launch, given widespread investor enthusiasm about SpaceX, artificial intelligence infrastructure, and the IPO. However, we assign this scenario, in which both Starship is reusable and scaled orbital data centers are highly successful, a 7% chance of happening, which is one reason our final fair value estimate of $63 is much lower than $154.

We could try to figure out how accurate Morningstar was with Tesla when Tesla was the same size ($19 billion/year in revenue) as SpaceX is now. That takes us back to 2018. Tesla stock was at $20/share vs. $400 now (adjusted for 15:1 in splits, but not for Bidenflation). ChatGPT:

Morningstar was not recommending Tesla as a buy in 2018. In ordinary buy/hold/sell language, their view was closer to sell/avoid for much of the year, or at least don’t buy at the market price. … Morningstar’s 2018 stance was “don’t buy”; in buy/hold/sell terms, it was closer to “sell” than “hold,” especially around the August 2018 $420 buyout episode.

(Keep in mind that $420 pre-splits was a $28/share offer in terms of today’s shares.)

Democrats such as Bernie Sanders who are envious that Elon Musk is a trillionaire now have a Science-based way to catch up. They say that expert advice should be followed without question, e.g., if an expert-crafted response to a virus killing Americans at a median age of 82 is to close schools for 8-year-olds for 18 months. A leveraged bet that SpaceX stock will fall to the Morningstar-estimated value could make at least $billions if not $trillions in profit.

Loosely related…

Full post, including comments

Sonos, eight years after the IPO

by philg

Readers are likely aware of my fondness for whole-house audio and, since 2005, Sonos, a company that has roots in Santa Barbara, CA and Boston, MA. The company went public at $15 per share almost exactly eight years ago. How’s the stock done? If you don’t adjust for inflation, it’s right at its original IPO price (down from the first-day bump, though). Adjusted at official CPI, though, someone who got IPO shares is down 25 percent. Adjusted for South Florida real estate costs, the investor is down about 50 percent.

Given that state governors ordered peasants to stay home for 2-3 years in a lot of high-income states, how do we account for a home audio company having done this badly? Sony has doubled, in nominal dollars, over the same period. Consumers don’t care about multi-room audio because they always have their AirPods in and the sound thus follows them?

Did Sonos fail to jump on the social justice bandwagon or ignore Is LGBTQIA the most popular social justice cause because it does not require giving money? Certainly not!

Sonos in June 2020:

In celebration of global Pride the Internet radio service introduced a limited-edition station, initiated by a group of LGBTQ+ employees.

Dmitri Siegel, VP Global Brand, extended his support becoming Pride@’s official executive sponsor. “I am so grateful to have the opportunity to be an ally in the Pride@ group,” he said. “As a straight white male, it’s common to feel intimidated or uncomfortable engaging on LGBTQ+ issues because you feel like you are going to say the wrong thing or be offensive in some way. Being a part of Pride@ has given me the opportunity to do my work—listen and learn. Getting over insecurity and discomfort unlocks all these amazing people and a whole aspect of the human experience that I would otherwise miss out on.”

The article references a corporate tweet:

The station was still up and running a year ago:

How about this month? The radio web site features state-sponsored NPR and Democratic Party-affiliated MSNBC, but I didn’t see anything about Pride.

Sonos Radio has an Instagram account. No Pride station is mentioned. (The “Full Spectrum; Celebrate Child” station is still available in the app, however.)

Sonos has an X account(!), but hasn’t posted anything for Pride 2026. Nor did they post anything regarding Pride on their (rather thin) Facebook account. Their careers page assures today’s young people that working life can be a continuous Pride celebration:

It looks as though the company has abandoned its commitment to proselytizing for Rainbow Flagism to consumers, reminding consumers that Black history is more important than other kinds of history, etc. In other words, Sonos has abandoned its principles and still can’t make money, the worst of all possible worlds for a corporation.

I still love the product, even if their lightweight powered speakers can’t overcome the laws of physics (see below regarding the latest Sonos S2 gear versus their old amps driving heavier passive (ancient) speakers). They also have great support compared to most companies. I’m wondering what they could make that would be profitable in an Apple-takes-all world.

Related:

Full post, including comments

Who has already flipped his/her/zir/their SpaceX shares?

by philg

After about 15 minutes of trading, 155 million SpaceX shares had already traded. The total size of the IPO was roughly 555 million shares (plus the underwriters have the option of buying another 83 million). I don’t think that the pre-IPO investors have the ability to sell their shares yet, a lockup period being conventional. Are people truly flipping their shares for a 20% profit within 15 minutes? (10% profit after paying short-term capital gains tax to Graham Platner.)

The total shares sold:

Speaking of Graham Platner…

Update: After one hour of trading, 272 million shares had been bought/sold on the exchange. Price is up to $168 without dramatic swings, suggesting some stabilization action by the underwriters?

Full post, including comments

I am buying a lot of SpaceX stock; what will it be worth?

by philg

As an index fund investor, I used to (indirectly) buy stocks that had been public for a while and only of companies that made a profit. The rules have apparently been tossed so now I will be a(n indirect) SpaceX shareholder. NYT:

Nasdaq, the exchange where SpaceX plans to list its stock, announced a rule change in May to allow “fast entry” into the Nasdaq-100 index by large private companies like SpaceX that go public.

Others followed. FTSE Russell recently altered its methodology, which will result in listing SpaceX in its indexes within a week of its going public.

The changes mean a large swath of index funds — which millions of Americans own in their retirement funds, pension plans and personal portfolios — are poised to hold SpaceX shares soon after the company goes public. Anthropic and OpenAI, the artificial intelligence start-ups that are planning to go public this year, would also land in index funds quickly, potentially exposing everyday investors to more financial risk whether they like it or not.

“It’s historically unprecedented,” said John Polonis, a former Wall Street lawyer who worked at J.P. Morgan and now offers financial analysis on social media. “You can try to reorient your retirement accounts to avoid funds invested in A.I. companies, but most people aren’t going to be doing that. They’re kind of left out in the dust here.”

Is there any integrity left on Wall Street?

One index provider has declined to budge. On Thursday, Standard & Poor’s said it would not change its criteria for inclusion in the S&P 500, one of the most-followed indexes, meaning SpaceX will not be eligible for inclusion until at least mid-2027. Standard & Poor’s said it had determined that exceptions to its rules “should not be granted solely based on market capitalization.”

What’s the long-term value estimate for this company? Humans cluster in metro areas, thus making fiber Internet and cell towers better than Starlink for most of us. ChatGPT: “roughly 75–85% of people worldwide who have at least a global middle-class/consumer-class standard of living live in urban, suburban, or peri-urban areas. I’d use 80% as the best single-number answer.”

SpaceX has the best rocket tech, but unless you’re Iran and want to kill infidels worldwide how many rockets do you need? Maybe the answer is military use of space rather than sending up packages that will rain down on the enemies of the righteous? But military equipment for use in space is incredibly expensive and takes forever to develop and build. Does a cheaper launch capability matter for the U.S. Department of War?

Elon says space makes sense for data centers, but I have a tough enough time maintaining the computer on my desk. If we want solar power and cheap land why isn’t Arizona, a Texas ranch, or central Florida a more sensible location?

How about the value of going to Mars? If a spaceship could accelerate continuously and indefinitely at 1g and decelerate at 1g how long would it take humans to get to Mars from Earth, all the while experiencing Earth-like gravity? Prof. Dr. ChatGPT, PhD Physics says 2-5 days. That would be an awesome tourism business. But, of course, SpaceX has no technology like that.

Perhaps the answer to the orbital-level valuation for SpaceX is that the company is run by a woman and we’re informed that female-led companies outperform their male-led peers. TIME:

Readers: What will SpaceX be worth five or ten years from now (in 2026 dollars)? What will have been seen as the main driver of value?

I’ll go first… because I believe in efficient markets, SpaceX in five years will be worth its IPO price plus 4% real return annually (about 21% over the IPO price). A narrow majority of the value will be from Starlink. (Note that this is like a probability expectation. I’m pretty sure that something dramatically good or dramatically bad will happen to SpaceX, but I can’t predict which is more likely and therefore my guess is right at the center. Analogous to the expected value of a coin flip game for $1 being $1 even though we know it will either be $0 or $2 and can’t be $1.)

A professional investor friend says that SpaceX would be a good buy at the IPO price because of the high percentage of retail purchases. “These retail investors come back in and support the price even after the inevitable post-IPO slump,” he said. “The more retail the better, contrary to previous prevailing wisdom.” If he had access to a large bock of SpaceX at the IPO price and without a lock-up, he would buy the block and sell it after a few days (i.e., right when a lot of index funds will be buying!).

“SpaceX IPO Is Said to Be More Than Four Times Oversubscribed” (Bloomberg, June 10):

SpaceX’s initial public offering has attracted demand for more than four times the available shares, according to people familiar with the matter, ahead of the Elon Musk-led rocket, satellite and artificial intelligence firm stopping taking orders.

SpaceX’s IPO is set to price June 11 and trade the following day. The company is offering 555.6 million shares at a fixed price of $135 each, which would raise about $75 billion, and value it at about $1.8 trillion.

I’m confused by the above. It says “set to price June 11” and the article is dated June 10. How was the price of $135/share already known on June 10? Separately, if the offering is oversubscribed by 4X, doesn’t that suggest the price is being set way too low? What about the duty to protect SpaceX’s existing shareholders by not giving away shares at such a low price that there are 4X as many buyers as shares? To avoid this abuse of shareholders, Google did a Dutch auction at its IPO (Google AI)

Google’s historic August 2004 Initial Public Offering (IPO) famously utilized a modified Dutch auction instead of the traditional wall street underwriting process. In this method, instead of investment banks setting a fixed price, individual and institutional investors bid directly on how many shares they wanted and the price they were willing to pay.

This is explained in “How I Did It: Google’s CEO on the Enduring Lessons of a Quirky IPO” (Harvard Business Review) by Eric Schmidt:

In mid-August the bidding began, based on our published expected IPO price range of $106 to $135 a share. … There weren’t a lot of orders, and to be frank, we wondered if we’d made a mistake in choosing an auction-based approach. The offers that did come in were at or below the low end of the range we’d anticipated. When the bidding period ended, it was clear that we weren’t going to be able to sell all the shares we had planned to sell in the price range we wanted. I met with the board to discuss whether we should delay our IPO and hope to get a higher price later. Our underwriters believed that we could close the IPO with a price around $80 to $90 a share if we reduced the number of shares for sale—a disappointing outcome. In the end we decided to close the IPO for a number of reasons, the most important being that it was time to put this chapter behind us and get back to running our business. So on August 18 we agreed to price it at $85 a share.

Perhaps this is the answer. Google expected to get more via an auction and got less. See also “Google shares took off, but the auction didn’t” (CNBC, 2014):

The rationale was simple: Take the short-term gains away from Wall Street and big money and give at least some ownership to the many consumers whose obsessive use of the search engine had allowed it to grow from a garage start-up into a multibillion-dollar phenomenon in half a decade.

But instead of pioneering a new formula for IPOs, with investment banks and big investment shops ceding control to the issuing companies and a wider universe of investors, the Google deal remains a historical anomaly.

Experts offer up two main explanations. The first is that auctions are risky. Banks get paid handsomely (7 percent of the offering amount is typical) to sell a deal to their clients, and in the process make sure prospective investors understand the business, competitive landscape and the state of the market. Through that work, they find what investors are willing to pay, so companies can be fairly confident that there’s adequate demand at the set price. An auction, meanwhile, is more like an investors’ Wild West.

The second reason is that Google’s offering wasn’t a real auction, but more of a hybrid. After all, there was clearly enough investor demand to price the stock at closer to $100, because that’s where the stock opened, but at the last minute lead underwriters Morgan Stanley and dropped it to $85. The low end of the expected range had been $108.

David Golden, a banker at JPMorgan, one of the many banks that served as an underwriter for the IPO, said the big investors decided just before the offering that without a reduction in price, they’d wait until the stock started trading and buy it on the open market rather than pay $100 a share or more in the IPO.

“Lo and behold, it was set at $85 a share, which built in a 15 to 18 percent profit that banks like to deliver to big institutional investors and that investors like to receive,” said Golden, who is now a managing partner at Revolution Ventures in San Francisco. Institutions “want to know that when they’re buying risky, illiquid securities, they’re going to have a built-in gain.”

Full post, including comments

How are bond investors doing vs. inflation?

by philg

Our government has provided us with a fresh fictitious inflation report today (the fiction is that a person can rent the same single-family home in the U.S. for 20 years and do so at a cost lower than mortgage, property tax, maintenance, insurance, etc.; “owners’ equivalent rent”).

Conventional advice for retirement saving is to buy at least some bonds rather than hold an all-stock portfolio. How can that work given the miserably low yields on inflation-protected bonds (TIPS) and the payments in nominal dollars that get ravaged by inflation any time our wise politicians feel the need to print money?

“A Yale Professor’s Investment Formula Says You Need More Stocks. See How It Works.” (Wall Street Journal, February 2026):

The formula’s central insight is that the future paychecks and retirement benefits that someone has yet to receive in their life are, when taken as a whole, like a bond because fluctuations in earnings aren’t strongly correlated with stock-market returns. For this 25-year-old, that big, bondlike chunk of future money means they could more easily weather a steep drop in stocks.

As another example, the way it treats a hypothetical middle-aged couple differs based on the amount they have saved up.

In both scenarios, the formula steers the couple toward a lower equity allocation than it did the 25-year-old, because a smaller share of their lifetime income is still to come.

But it recommends a much lower allocation, 53%, when the couple has twice as much money to invest, because in that scenario upping the equity allocation would alter the risk profile of a larger proportion of their projected lifetime resources.

“It’s more conservative when you have more money saved up,” Choi said.

For a middle-aged couple with what someone in Miami would call “no money” (either $400,000 or $800,000 saved, neither of which will pay for a kitchen renovation), the Yale genius says that two men (it’s the WSJ so both participants in the “couple” must be guys in order that they can eventually form an all-male throuple) should have as much as half their portfolio in bonds (the non-academic non-geniuses at Vanguard give these two guys only 24% bonds):

What drops out of this thinking, for people aged 70, is that they should have somewhere between 35 and 70 percent of their investments in bonds:

After living just recently through Bidenflation and having a personal memory of the Jimmy Carter-era inflation (maybe not too different, actually, if CPI were calculated in the same way), this seems intuitively wrong. The 70-year-olds could live to be 100 via GLP-1 and whatever medical miracles LLMs can come up with. The 70-year-olds might care about leaving some money for their children and grandchildren, now forced to compete with 70+ million immigrants with whom the Boomers didn’t have to.

Maybe we can simply look back. The “keep adding bonds with age” strategy is embodied in the Vanguard funds. The Vanguard 2030 fund was created almost exactly 30 years ago. It has returned about 7% per year:

2006 wasn’t a great time to buy an all-stock portfolio, since you had to make it through the Collapse of 2008. Nonetheless, ChatGPT says that SPY has enjoyed a total return (dividends reinvested) of over 11% annually since then. The difference of 4% doesn’t sound huge, but that’s 100 percent of the standard formula of how much a person can spend in retirement from his/her/zir/their assets and not go broke for 30 years. For someone who started with $100,000, the S&P 500 investment would have grown to over $830,000. The same investor in the Vanguard part-bond fund would have only $430,000. That’s the difference between a totally-pimped house in The Villages (the fun center of Elder Florida) and a used/basic house in The Villages.

Note that the above calculations don’t include federal, state, and/or local income taxes that the investor would have had to pay on the dividends received into any non-retirement account. Taxes are worse for the part-bond investor because the yield is taxed as ordinary income and not subject to the qualified dividend discount. Also, stocks deliver much of their return via appreciation rather than by paying dividends. So the corporation may pay income tax, but the investor need not. ChatGPT says the $830,000 stock portfolio would be more like $625,000.

How about all bonds for 20 years? ChatGPT says the return would have been only 3.1% year. The $100,000 would have become $184,000, a laughable $12,000 in appreciation after adjusting for official CPI (i.e., it would have shrunk in terms of the ability to buy and maintain a single-family home). This turns into a real-dollar loss if held in a taxable account, having appreciated to only $144,000 in nominal dollars. The $100,000 became $87,600 in 2006 purchasing power (official CPI).

I’ve always struggled to comprehend why investors are willing to buy bonds priced in nominal dollars, which nearly all bonds are. Someone took the other side of our absurd mortgage, issued at 3.125% just as Bidenflation was gathering a major head of steam in February 2022. I would love to meet that person and ask “What did you think was going to happen?”

Readers: Can someone please sell me on why a typical investor would have even 1% bonds in his/her/zir/their portfolio? Let’s assume we’re talking about a 65-year-old. Because this person has money, it is likely that he/she/ze/they has children (fertility vs. income shows it is Americans with zero income and those with high incomes who have kids; the working class are being bred out of existence). Our hypothetical saver wants to not run out of money even if death comes at 111, which was the age of the oldest person receiving General Motors pension and health care benefits before the company went bankrupt/got bailed out by us in 2009 because they couldn’t pay their union retirees all of the promised pension and health care benefits. Our hypothetical saver would rather leave more than less to his/her/zir/their children and grandchildren. We’ll assume that 30 percent of the saver’s portfolio is in a tax-exempt retirement account. Could the answer be “It makes sense to buy bonds when the S&P 500’s average P/E ratio exceeds a threshold and trade them for stocks when the S&P dips”?

Full post, including comments

Harvard geniuses underperform the S&P 500 by 15 percent per year

by philg

“Head of Harvard’s Endowment Tells Board He Plans to Retire” (Wall Street Journal), regarding a manager paid over $6 million per year:

N.P. “Narv” Narvekar, the head of Harvard University’s nearly $57 billion endowment, recently told the endowment’s board he plans to retire, according to people familiar with the discussions. He has served nearly a decade in the post.

In the past three years, Harvard earned an annualized return of 8.1%, a rate that topped that of Ivy League rivals Yale and Princeton and which placed it in a tie for fourth among a group of 12 top schools, according to financial technology company Markov Processes International.

The Wall Street Journal doesn’t bother to ask Edward Tufte’s question, “Compared to What?” But ChatGPT can come to the rescue:

In other words, one can get paid $6 million per year for dramatic underperformance relative to the simplest imaginable investment strategy, dumping everything into the S&P 500 (a 23% annual return vs. the 8% achieved by Mr. Narvekar and subordinates). That’s a career almost as good as “receptionist in NVIDIA branch office”!

Full post, including comments