The household used to be the factory

Happy Sunday.

$65 dollars.

That is what a 22-year-old in Florida paid for three weeks of an injectable peptide stack he ordered online. Shipped to his door. No prescription, no diagnosis, none of it approved for what he is using it for. He is not sick. He wants to feel better by Thursday, and he has at least forty years of runway to keep paying for it.

Millions of people are buying some version of that box right now. Nobody knows how many. Nobody is required to find out. I spent most of this week trying to size it and could not find anybody who had accurately.

Now the other number…. last year the American birth rate hit 53.1 births per thousand women aged 15 to 44. The lowest ever recorded here, down 23% from the 2007 peak. Not the lowest since the financial crisis. The lowest since 1909, which is as far back as the numbers go.

We know that one to the decimal point. Statisticians count it, demographers fight about it, and Congress builds sixty-year budgets on top of it. We can even price the cause. Seventy-one percent of American adults say kids are not affordable for most people, and it takes $109,152 of household income to qualify for a mortgage on the median American home, before you have found a dollar of the down payment.

You could draw a line between those two numbers. I am not going to, because I cannot prove it, and anybody who tells you they can is selling something.

The point is simpler than that. One of these markets gets counted, argued about, and eventually corrected. The other does not, and it is not small.

That is today's letter.

What actually got sold

The old healthcare transaction had four gates.

You had to be sick. A doctor had to write it. An insurer had to pay for it. And the FDA had to have proven it worked on somebody else, years earlier.

The new one keeps a doctor on staff and removes the other three.

You do not have to be sick, because the product treats being young instead of being ill. You do not need insurance, because it is ninety-two dollars a month on a card. And you do not need an approval, because compounded product never required one.

So what is left is a subscription to a better version of yourself, billed monthly, that never ends, because you never get well. You just get closer. The best revenue in the world is recurring, price-insensitive, and attached to an outcome the customer can never actually reach. Somebody built exactly that and aimed it at the kid in Florida.

It works. Hims & Hers $HIMS did more than $753 million in revenue last quarter with roughly 2.9 million people paying that ninety-two dollars a month. That is a phone bill. And it is one company's average, sitting on top of whatever else that person already buys.

Now the part almost nobody wrote up. When Hims shifted off compounded product toward branded, gross margin dropped to 64 percent from 76. Twelve points gone.

That was not a strategy call. The FDA declared the semaglutide and tirzepatide shortages over, ended the grace periods for compounding, sent warning letters to thirty telehealth companies in a single day this February, and in April proposed stripping semaglutide, tirzepatide and liraglutide off the list that let outsourcing facilities make them in bulk. The agency said plainly that being cheaper does not count as a medical reason.

So the twelve points were the regulatory gap. The regulator noticed and took them back.

Here is the part that should worry you more than the margin. Closing the legal lane does not close the market. Patient-specific compounding at your corner pharmacy is still legal. And the mail-order research-chemical business, the one that shipped that $65 box to Florida, was never inside the FDA's reach to begin with. Squeeze the half-visible lane and demand walks over to the invisible one. It does not evaporate. It just gets harder to see.

Why your employer is buying

One correction to what I just said. The payer did not disappear everywhere. It disappeared from the cash lane. In the branded lane it is spending like it never has.

Bank of America spends more than $250 million a year on GLP-1s for its people. Roughly 13 percent of the entire health budget for 211,000 employees, up from zero in about four years. Brian Moynihan said some of those employees will leave before the company ever sees a benefit, and that he considers it a good investment anyway.

Every writeup I read treated that as a benevolence story. I think it is a labor story.

Moynihan is not underwriting a 2050 mortality curve. Nobody underwrites a 2050 anything on a four-year CEO clock. He is underwriting turnover, recruiting and sick days this fiscal year, in a category his people will switch jobs to get. The long-run health benefit is the line in the press release. The line that clears the capital committee is that a benefit your competitor does not offer is a retention product.

So the fastest-growing item in American corporate health spending is justified on one time horizon and paid for on another. Worth knowing before you assume that money keeps flowing through a bad quarter.

Hold onto the difference between those two lanes. It is the whole rest of the letter.

Today’s Sponsor

125,000+ podcasts, transcribed and searchable in minutes.

Particle transcribes 125,000+ podcasts where executives, officials, and analysts talk candidly and publicly, transcribed and queryable within minutes of airing.

Watch a company, a person, or a theme with the Particle API and get alerts whenever they're mentioned.

Particle is built by former Twitter and Tesla engineers, using an AI-native transcription pipeline that delivers high accuracy and extensive data enrichment.

Start with $10 in free credit.

The market selects for what you can feel

This is the section I would keep if I had to cut everything else.

The CDC has logged 2,777 confirmed measles cases this year, a 35-year high. More than 93 percent of the people infected were unvaccinated or nobody knows. Same country, same year, millions of people are paying cash for injections nobody has approved.

Everybody reads that as hypocrisy. It is not. Look at what each product asks the customer to notice.

A vaccine's benefit is an event that does not happen to you. You will never feel it. Not once. Its entire value is statistical, which means believing in it requires trusting somebody else's arithmetic over your own senses.

A peptide's benefit is a feeling you have on Thursday. Your shoulder aches less. Your skin looks better in the mirror. You slept well. Whether the compound caused any of it is unknowable to you and completely beside the point, because the feedback showed up in seventy-two hours with your own name on it.

KFF's polling found that hard anti-vaccine belief is rare. Fewer than one in ten Americans called each common myth definitely true. The common state is uncertainty, with at least half landing somewhere in the gray. Uncertain people do not follow evidence. They follow whichever signal arrives first and closest.

So here is the rule I think runs this entire market.

An unsupervised health market does not select for what works. It selects for what the customer can feel working.

Follow that down and it gets bleak, because the treatments with the best evidence in all of medicine are the ones with the weakest felt feedback. Vaccination. Statins. Blood pressure control. Not smoking. Every one of them pays off as an absence, decades out, invisible. Every one of them loses to a compound that makes you feel something by Thursday.

The placebo response is not a defect in this category. It is the product spec. It costs nothing to manufacture and no regulation touches it.

The evidence bar did not fall. Authority moved from the institution to the anecdote, and it moved for a structural reason. The anecdote pays out faster.

Four circuits, all cut

Fine, you say. Bad products die. A lawyer files, a short seller publishes, the stock reprices. The market cleans itself.

It cleans itself through four circuits. This category has quietly cut all four.

Reporting. Compounding pharmacies are not required to report serious adverse events to the FDA. The agency has still collected more than 455 reports tied to compounded semaglutide and more than 320 tied to compounded tirzepatide, and every one of those arrived voluntarily, from people nobody obligated to send anything. That is not a count. That is a floor, and nobody knows how far the floor sits above zero.

Causation. For several of these substances there is no universally accepted chemical definition. You cannot prove a compound hurt somebody when you cannot establish what the compound was.

Liability. Gray-market product ships labeled for research use only, not for human consumption. That is not a warning to the customer. It is a wall between the seller and a courtroom.

Payer. Cash removes the one party in American healthcare with a financial reason to ask whether a treatment was necessary. Insurers are unpleasant for a reason, and that reason is the cheapest quality control in the system. Bank of America asks that question every renewal. Nobody asks it about the $65 box.

Every mechanism that normally turns a bad health product into a repricing event has been disconnected, and not one of those disconnections was an accident.

Look at what that does to the shape of the risk. Risk you can measure arrives slowly. Small studies, small settlements, a gradual de-rating everybody has time to react to. Risk nobody is allowed to measure does not do that. It arrives all at once or it never arrives at all. There is no middle of the distribution, because the middle is made of data and the data is not being collected.

That is not a growth stock. That is a short volatility position that pays you in subscribers.

If you want to know what the world does with a risk it cannot price, look at the other letter in your mailbox. Homeowners insurers did not leave those states because they proved the fires were coming. They left because they could no longer price them. Absence of data reads as safety to a consumer. In an underwriting room, absence of data is the definition of uninsurable.

Today’s Sponsor

Forget Elon's Gadget. Buy the Companies Behind Its Tech.

Every breakthrough device runs on chips, parts, and materials from other companies — most of them public and overlooked. Our analyst named 3 positioned to profit from Elon's July 22 launch, plus the most undervalued name in the supply chain.

Get the free report.

The rigor pays for the promise

Not all of this is froth, and the good version deserves its paragraph.

In July the FDA approved Merck's Lipfendra, the first pill that does the job of an injected cholesterol biologic. It is a macrocyclic peptide, same broad family of chemistry as the stuff in the $65 box, and it cut LDL by about 56 percent against placebo across roughly 2,900 patients. Here is the detail I love. Even with the approval in hand, there is still a trial running to find out whether lowering LDL this way actually prevents heart attacks. They got the win and they are still checking.

Then in August, Merck and Moderna reported that their individualized mRNA cancer therapy hit its primary endpoint in a Phase 3 melanoma trial. 1,137 patients, built one tumor at a time. First positive Phase 3 for anything in that category, ever. They did not publish how big the benefit was, which tells you something about how carefully these people talk.

Both got a fraction of the attention the gray market gets. And here is the ugly part.

The two economies feed each other. Every real breakthrough in personalized medicine raises the plausibility of every unproven personalized product, because the public does not distinguish between a molecule with a Phase 3 endpoint and a molecule with a testimonial. The headline is the same headline. Personalized. Precision. Built for you.

Moderna paid for the trial. The gray market keeps the credibility.

The rigor pays for the promise and the promise never pays it back. That is why the unproven side compounds faster. It runs on borrowed trust with no interest expense.

The honest ugly

Every letter here gets one. This one needs four.

The first 80 percent is free and mostly ignored. Persistent smokers lose about ten years of life against people who never smoked. Seven thousand steps a day is associated with far lower mortality than two thousand. The biggest observed benefit from strength training showed up at 30 to 60 minutes a week. Per week. These are observational findings, not guarantees, but the size of the gap is not close. They work, they cost nothing, and they feel like nothing while you do them, which is exactly why an unsupervised market prices them at zero and sells you the thing you can feel.

It grows because nobody argues about it. David Kelly opened a piece on the federal debt this week with a story about buying Godiva squares in years when he and his wife could not afford the house or the car. Nobody goes broke on chocolate squares. It was never big enough to make it into the budget fight. The wellness spend is the chocolate square. Multiply it across a subscriber base and it is a multibillion-dollar company. Multiply it across a workforce and it is the fastest-growing line in a corporate health plan. It grows because it is too small to fight about anywhere it actually gets spent, which is the same psychology Kelly says keeps us from touching the deficit.

The harm is real and nobody can see it. Clinicians report that GLP-1s have set off or worsened disordered eating in some patients. Pediatric endocrinologists have raised alarms about growth hormone peptides in teenagers. Because those four circuits are cut, whatever the true harm rate turns out to be, we will learn it late and by anecdote. (If you or somebody close to you is struggling with eating or body image, the National Alliance for Eating Disorders helpline is 866-662-1235.)

And the case against me. Maybe this generation is right and I am the old guy at the window. They lift more than my cohort did. They sleep with data. They take fitness seriously in their twenties instead of their fifties, and those are the interventions with the best evidence in the whole field. If the injectables are froth sitting on top of a real shift toward the boring stuff that works, the outcome is good and my worry is misplaced. I would take that trade happily. I would just like somebody to be counting.

So what

Usual reminder. I am not a financial advisor and none of this is advice.

Three questions I would ask before putting money anywhere near this.

Who is collecting the data, and what happens to this position if nobody ever does? In the regulated lane there is an answer. In the compounded lane the honest answer is nobody, by design. That does not make the business bad. It makes the range of outcomes barbelled, and a barbell should never be priced like a growth curve.

Am I long the evidence or long the enthusiasm? Those are two different businesses trading at similar multiples. One survives a bad headline. The other is the bad headline.

And what does the regulator do next? We just watched twelve points of margin vanish when the FDA closed one door. There are more doors. But every door it closes pushes volume toward the lane it cannot reach at all, which is worse for the customer and invisible on the income statement.

Back to the two numbers I started with.

$109,152 is what it costs to buy into the life our parents treated as the default. $65 is what it costs to buy three weeks of a “better version of yourself.” I do not think the second one caused the first. I think they are both prices in the same catalog, and one of them is a lot easier to say yes to on a Tuesday night.

So the country is running the same bet twice.

Once in the desert, in public, at eleven trillion dollars, on audited financial statements, with a capex line and analysts screaming about it every ninety days.

Once in bathrooms, on autopay, for a rounding fraction of that money, with no reporting requirement, no chemical definition, no liability and no payer asking questions.

The sizes are not close and I am not pretending they are. That was never the point. The point is that both are wagers on staying healthy and productive without doing the boring, expensive, slow thing.

Both may pay off. But only one of them was built in a way that can ever tell us it was wrong.

And that is the one your kids are already living inside.

Stay curious 😎

- John