Boom Loops

When the market only ever goes up, something's going down.

In the summer of 1976, a man named Jack Bogle launched a mutual fund through his new firm, Vanguard. The idea was so damning that Wall Street took it as a personal insult: you cannot beat the market, so you should stop paying the people who swear they can. Not that you shouldn’t try—that you can’t, not reliably, especially not once the fees for trying are subtracted from whatever you’d have made otherwise. So his fund wouldn’t try. It would buy all five hundred companies in the S&P index, each in proportion to its size, and then do nothing, charging almost nothing, forever. The underwriters had hoped to raise a hundred and fifty million dollars. They only raised eleven. The industry called it “Bogle’s Folly.” The chairman of Fidelity reportedly scoffed that no investor would ever settle for average.

Then Bogle started beating his competitors, for a reason closer to arithmetic than skill. Everyone invested in the market, added together, is the market. Since the actively managed funds charged high management fees, the average fund earned the market returns, minus the fees. Bogle’s fund also earned the market returns, because he just bought the whole market. But since he charged lower fees, he beat the average fund automatically.

The average active dollar doesn’t trail the index because the manager is dim. It trails because the manager is expensive. Bogle built the company to drive that cost toward zero. By the time Bogle died in 2019, Warren Buffett was writing that if a statue were ever erected to the person who had done the most for American investors, “the hands-down choice should be Jack Bogle.” The folly had become the orthodoxy, and the orthodoxy was correct.

It is still the most responsible, boring thing you can do with a paycheck. Don’t pick stocks. Don’t try to beat the market. Don’t pay fees to the clever broker¹ in the good suit. Buy everything at once, payroll-deducted, preferably in your 401(k), for forty years, and never look at it until you retire. More than half of American households now have money in the market, much of it invested in roughly this way, most of them having never heard Bogle’s name. They followed the advice. They bought diversified index funds.

What almost nobody noticed, including, for most of his life, the man who built it, was that the advice worked because the people taking it were a minority. An index fund has no opinion about whether anything is cheap or dear, and that indifference is the entire source of its discipline and its microscopic fee. It free-rides on everyone else’s pricing: it lets the stock-pickers do the sweating and the arguing over what Apple is worth, and then simply buys their answer at whatever the answer turned out to be. That free ride is a marvelous deal when the rider is a small passenger on a very large market. It becomes something stranger when the passenger grows into the vehicle. Bogle himself, in the last years of his life, was the one who said it out loud: in a 2018 op-ed he warned that a handful of index managers were on track to own “30% or more of the U.S. stock market—effective control,” and that he did not believe “such concentration would serve the national interest.” The patron saint of the index spent one of his last public breaths worried about over-indexing.

He was right about that too, and you can see it in your own account. The fund still carries the same Vanguard name, the same trivial fee, the same reassuring pie chart. But it no longer holds what the pie chart says it holds. What was sold as the broadest, safest bet in finance has quietly become a bet on a handful of names—and concentration is only the first of the surprises. The bigger issue is what it’s done to prices.

The Thermostat

A price is information, and the cleanest place to watch it do its work is a market too dull to be worth rigging: the dairy aisle. Say milk gets expensive. That isn’t the start of a problem; it’s the start of the cure. A high price is a message—make more milk—sent to exactly the people who can. So they do. Dairies breed more cows, and a season later there’s more milk than the market wants, and the price drifts back down. The signal that announced the scarcity is the thing that ends it. Push the system and it pushes back. Economists call this negative feedback. You can think of it as a thermostat: the reason a market doesn’t overheat, quietly at work behind every honest price you have ever paid.

And there are two cures in that, not one. The high price pushes the dairies to make more milk—that’s supply. It also pushes some shoppers to buy less, to reach for the soy milk or the orange juice or to skip it entirely—that’s demand. More coming in, less going out, and the price has nowhere to go but back down. Two brakes, working opposite ends of the same number, and both of them just the market responding to a real fact: milk is scarce, do something about it.

A share of stock is built to ride those same two brakes. When a company’s stock gets expensive, it can do exactly what the dairy does when milk gets expensive: make more. It issues new shares at the rich price and plows the cash into the actual business—more stores, more trucks, more R&D, more output. More cows, in other words, and more milk, until the price eases back toward what the now-bigger company is worth. That’s the supply brake. The demand brake is a person: the investor who looks at a stock priced at a hundred and thirty times its earnings, decides that is insane, and won’t pay it. His money goes somewhere cheaper instead, and every buyer who walks away is a bid the price loses—the same downward tug as a shopper reaching for the soy milk. Two brakes, the same as the dairy, and the market is built to ride them both.

But just as Bogle feared, today the market is riding neither. One brake has been wired to run in reverse. The other has been overrun—the thin line of investors who still care about price, swamped by an army of passive index money that doesn’t. What’s left is a price that no longer answers to what anything is worth. The one signal it still reflects is the sheer volume of money pouring in.

Culling the Herd

Start with the supply brake—the one running backward.

To see how that happens, forget stocks for a moment and picture a rancher. Imagine he isn’t paid a salary; he’s paid in cattle, and his bonus rises and falls with the price of a cow. When cattle get expensive, the textbook says he should breed more of them and sell into the high price—but he won’t, because a barn full of new calves is exactly what would drive the price back down and gut his own bonus. So he does the opposite: he thins the herd. A smaller herd means scarcer beef, a higher price per head, and a fatter check for him. The response that’s supposed to cool an overheated market has quietly reversed into one that stokes it, because the man at the lever is paid by how high the price climbs, rather than how much value is produced.

This is almost exactly how a modern public company is run. Its executives are paid in stock—options, grants, packages that vest on the share price—so the share price isn’t one of their concerns. It is the concern. And the fastest way to lift a share price is not to build a better business, which is slow and uncertain and may not pay off until long after the options have vested or the executive has moved on. The fastest way is to shrink the number of shares. So when the stock is expensive, instead of doing the textbook thing—issuing new shares, raising cheap capital, plowing it into something real—the company does the reverse. It spends its cash buying its own shares back, hundreds of billions of dollars a year, retiring them, thinning the herd until each surviving share is a claim on a slightly scarcer thing.

None of this is a crime, or even necessarily foolish. Handing cash back to shareholders is respectable enough, and a buyback doesn’t change what the business underneath is worth—it’s the same company, just sliced fewer ways. But the executives running it have no incentive to breed more calves. An expensive stock is supposed to summon something real into existence: new capacity, new products, more of the actual thing the company makes, more milk. A buyback summons none of it. It produces no new milk at all—more earnings per investor, but not a penny more overall. The supply brake didn’t fail quietly. It was rebuilt into a valuation accelerator, by people whose pay depends on the pedal staying down.

The Bid That Doesn’t Read Prices

The demand brake is stranger, because it hasn’t been removed at all. The skeptic who’d refuse to overpay is still right there refusing. He’s simply been swamped—by a buyer who shares none of his hesitation because it hasn’t got an opinion to hesitate with. A passive index fund does not judge whether anything is cheap or overpriced. It can’t; that indifference is the whole source of its discipline and its microscopic fee. Money comes in, and it buys every company in proportion to what that company already costs, at whatever today’s quote happens to be, on the fifteenth and the thirtieth, or whenever your withholding hits your account, whether the market is on a tear or on sale or on fire. It is the purest price-insensitive buyer ever assembled at scale: a bid that never once asks what it’s paying. And it is no longer a niche. Index funds overtook the active stock-pickers in total assets around 2024 and now manage more than nineteen trillion dollars between them—grown from the eleven million Bogle’s fund scraped together in 1976.

On its own, that’s harmless. A market-cap fund is a proportion-keeping machine, not a concentration machine. If a company is seven percent of the market, your dollar puts seven cents into it—you’re not piling into the winner, you’re just holding your slice. Left alone, a price-insensitive fund would sit there in perfect proportion forever and distort nothing.

But it is not left alone. It is fed—every two weeks, payroll-deducted, automatic, forever. And a buyer who pays any price and never stops is almost impossible for the other side to discipline. Nothing matches it on the sell side—no automatic seller dumping shares every payday, no matter the price. Only people, and people run out. The skeptic who thinks the giant is wildly overvalued can sell his shares. Once. Then he’s out of stock and out of ammunition, and the bid is still there on the thirtieth, and the fifteenth after that. To actually walk the price down, you wouldn’t need one skeptic, or ten. You’d need essentially everyone who still cares about value to sell at the same moment, in concert, into a bid that keeps buying through all of it—and they never coordinate. Each one’s sale is swallowed by a nation’s worth of 401(k) money buying that same morning. The price stops being something the market negotiates its way to. It becomes whatever the bid is, and the bid isn’t looking at the price.

A finance professor might object that index funds are a majority of the market’s assets but only a sliver of its daily trading, and that prices are set by whoever’s trading. True, narrowly, and beside the point. Most of that trading volume is active managers selling to each other, one fund handing Nvidia to another, churn that nets to nothing and discovers no new price. What moves the level isn’t the churn; it’s the net direction of the flow, and the net direction is a one-way street. Volume is the noise of a thousand people swapping the same shares around. The flow is the quiet, relentless, price-blind flood underneath it, driving everything higher.

The man whose job is to lean against this—the skeptic, the brake—doesn’t just get outnumbered. He gets destroyed for being right. He looks at the giant at thirty-five times earnings, judges it overpriced, and positions against it, and then he gets run over by twelve straight quarters of payroll-funded buying that does not care that he is right. The active managers watch the bears die on the side of the road, and decide—of course—they’d better move with traffic. The brake doesn’t just fail. It gets fed into the engine. The market used to carry the memory of every prior mania in the form of someone willing to bet against the next one. Price-blind money is amnesiac.

Drown the price-sensitive seller—not remove him, just drown him—and the index level stops measuring anything but the size of the bid. How big is the payroll deduction, the 401(k) contribution bid this month? There’s still a number getting printed. It just doesn’t have much to do with the performance or value of any given investment in the market. Which leaves the one question the whole machine depends on: how does a bid like that never run out of money to spend?

The Bottomless Bid

A bid that never runs out of money needs a source that never runs out of money. There is one, and it isn’t a figure of speech. The government spends far more than it takes in, and it covers the gap by selling bonds; when there aren’t enough willing lenders to absorb them all at a tolerable rate, the central bank buys the rest, with money it never had and never collected. A bond goes in, and dollars that didn’t exist that morning come out—printing, with a couple of respectable steps in between so that nobody respectable has to say so. Dollars made that way pile up faster than real production to spend them on, which grows at maybe two or three percent a year, and the surplus has to land somewhere. It can’t sit in milk and gasoline, where a high price just summons more supply and sinks back down. It pools instead in the things no one can make more of on command—shares, land, the scarce and the already-built. It goes to assets.

There is supposed to be a limit on this, and the limit is the interest rate. Make borrowing expensive enough and borrowers retreat—you don’t buy the bigger house when the mortgage rate doubles, the company shelves the new plant, and the flow of fresh money slows at its source. That works on every borrower who flinches at a price, which is every borrower but one. When the rate on the government’s debt doubles, it doesn’t borrow less; it borrows more—enough to cover the spending it was already going to do, plus enough to cover the extra interest. Today, the interest bill alone is already past nine hundred billion dollars a year, more than the entire military, the largest on the planet, and every uptick gets rolled into next year’s deficit and financed with even more debt. So the one tool built to slow borrowing down speeds this borrower up: it is the single actor in the economy that answers a higher price by buying more. You can’t raise the rate high enough to shut it down, because the rate is the thing that turns it up.

And the delivery is automatic. A retirement system built over those same decades routes a slice of every paycheck into the index by default—payroll-deducted, target-dated, rebalanced by software, never once pausing to ask what anything costs or what it’s worth. The money that has to go somewhere meets a machine designed to buy regardless of price, on schedule, forever. That is the bottomless bid: not a verdict the market reaches, but the default setting on a hundred million paychecks doing the “responsible” thing.

Thirty Years of Profit

Of course, the responsible thing isn’t to buy a company at any price. Sophisticated investors value companies in proportion to their earnings. A price-to-earnings multiple is just a number of years: it’s how long the company would have to hand you its entire profit, every dollar of it, before you’d have gotten your money back. At ten times earnings, you’re buying ten years of the company’s whole output to break even. That’s a deal that pencils. At thirty times, it’s thirty years—you’ll have your capital returned around the time you retire, assuming nothing goes wrong for three decades. At sixty, you’re not really hoping to get your money back, so much as you’re hoping your grandchildren do. But today, nobody flinches at stocks that trade at a P/E of 60, even though anyone would be suspicious of the equivalent statement: you are pre-paying sixty years of the company’s entire profit.

To be fair, investors don’t just buy a company for its current earnings; they buy it for its growth, and if profits compound fast enough, a thirty-year sticker pays back far sooner. For a single exceptional company that can be right—Tesla has spent years priced at several hundred times its earnings, and sits north of three hundred even now, a number that only resolves if you assume it conquers industries it hasn’t entered yet. Maybe it will. But the index can’t make that argument, and the index is what the price-blind money buys. One company can grow into a heroic multiple; the whole market cannot, because aggregate earnings are roughly chained to the size of the economy and margins can’t climb past a hundred percent. So when the aggregate market trades at thirty-two times current earnings—more than thirty years of payback, past forty once you smooth a full business cycle—no story about disruptive growth rescues it. A high multiple doesn’t hand you the growth; it means you’ve already paid for it, and your return now rides on reality beating an expectation that’s already priced optimistically.

History keeps the receipts on this, and it keeps them in a graveyard of names that everyone once thought of as invincible. In March of 2000, Cisco Systems was the most valuable company on earth, worth more than half a trillion dollars, trading north of two hundred times its earnings. A person who bought at that price waited more than twenty-five years just to see the stock back where he’d started—not adjusted for inflation, just nominally, a quarter of a century to break even. And Cisco won. It’s still here, still profitable, still selling the plumbing of the internet. The company was never the problem. The price was. Microsoft is the strongest case the optimist has, the firm that genuinely grew into its promise and became one of the great businesses of the age, and even Microsoft handed its end-of-1999 buyer the better part of seventeen years of dead money before the stock reclaimed its old high. That is the reward for being completely right about the best franchise of the era. You simply paid too much for it, and “too much” took a decade and a half to forgive.

So paying a few hundred times earnings isn’t optimism. It’s a bet that this company sits in the thinnest sliver of corporate history—the handful that compounded at extraordinary rates across an entire human lifetime—and a bet priced as though the sliver were a certainty. The price-blind money makes similar bets every two weeks, for huge multiples on an index highly concentrated in a few names, on your behalf, without anyone in the chain ever asking whether the odds are any good.

The Diversification That Isn’t

The antidote to Cisco or Tesla or any other company failing to live up to irrational exuberance is simple. Diversify.

And diversification is the other selling point in the indexes’ favor. In principle, you bought the fund so that you wouldn’t have all your eggs in a few baskets. Except, today, the ten largest companies are nearly two-fifths of the entire index—more top-heavy than at any point in half a century, longer than most investors have even been investing, and more concentrated than the market was at the peak of the dot-com bubble. Weight every company equally instead of by size and the index multiple drops by roughly a quarter, the widest that gap has run in the fifteen years the data covers. The headline valuation isn’t a story about five hundred companies. It’s a story about ten. The fund that sold you safety-in-numbers has quietly become a concentrated bet on a handful of names.

Cap-weighting invests your money in proportion to price, not earnings. The more the market falls in love with a company, the higher its multiple climbs and the heavier it sits in the index—so the fund buys the most of exactly the companies whose price already assumes years of perfection, the ones with the most that can go wrong. That might suit a desk running high-variance bets with analysts watching every quarter. It’s a shaky foundation under a nation’s retirement savings.

The problem with high growth expectations isn’t the growth, or even really the expectations. It’s the results. Eventually, and inevitably, companies fail to deliver on the hype. A quarter will come that will disappoint. There are some number of GPUs, or some number of Teslas, or some number of data centers that will simply be enough. The market will be saturated. In a low-multiple business, the price corrects down proportionally, and a well-diversified portfolio muddles on. But in high-multiple businesses, a disappointment doesn’t simply reprice this quarter, it reprices expectations over the next several decades. And in highly concentrated, and highly correlated, indexes the disappointment tends to be contagious.

That’s the downside of being only a little wrong. The crueler half is the upside of being mostly right: a company priced for perfection that delivers perfection has merely performed to expectations already priced in, and the multiple drifts back to earth anyway. Disappoint, and the concentrated, correlated names fall together. Deliver, and the gain was prepaid. Heads you lose; tails you don’t really win.

The Boom Loop

Market crashes always earn a sort of infamy among investors. Black Monday, the Dot-Com Crash, the GFC. We remember the sinking feeling when suddenly everything is repriced down, dramatically, all at once. The memories are visceral, emotional, almost physical.

But we always forget the crash up. The boom loop that preceded each of those notorious days. It runs like this: a price-blind bid pushes the market up, and the rising market makes indexing look like the only safe choice, which pulls in more money, which widens the bid and pushes the market up again. The higher the number climbs, the better the advertisement—every record high brings in the next dollar, and the next dollar lifts the price again. And nothing in the machine leans the other way: a government that borrows more the more it costs to borrow keeps the cash coming, the cash floods into assets because there’s nowhere productive left for it to go, and the index funnels it into fewer and fewer names at higher and higher multiples. Every part feeds the next, and the whole thing runs one direction. Up.

At the end of the day, there’s exactly one surefire way to lose your investment. Buy high, sell low.

So the next time the index closes at a record and the headline calls it a resilient economy that just won’t quit, don’t ask whether it’s a bubble—that’s the old question, and it’s asking for a date nobody can give you. Ask the harder one: what does the price signify, and what is the company worth? You’ve been buying a smaller and smaller share of the same economy, at a longer and longer payback, through a machine built so that nobody in it ever has to decide whether it’s worth it. The number goes up. But that part was never in question.

1. The old joke: You know why they call him a broker? Cause he’s broker than you.