TOPIC • 18 CLAIMS
Bubbles
What separates a bubble from a bull market, and whether the distinction is usable in real time.
HOWARD MARKS/
He splits bubbles into two kinds, which is the memo’s main contribution: inflection bubbles, built on technology that ultimately delivers, versus mean-reversion bubbles, financial fads that leave nothing behind.
IS IT A BUBBLE? →HOWARD MARKS/
On the central valuation question he argues the historical comparison cuts the other way — today’s leaders trade cheaper than the 1999 leaders did.
- P/E ratios for Microsoft, Cisco and Oracle in 1998-2000 were much higher than current ratios for Nvidia, Microsoft, Alphabet, Amazon and Meta
HOWARD MARKS/
His concern is located in the financing of the buildout rather than in the equity multiples.
- roughly $5 trillion of data center buildout
- against about $350 billion held collectively by Microsoft, Alphabet, Amazon, Meta and Oracle
HOWARD MARKS/
He cites private-market pricing as the clearer symptom of excess.
- Thinking Machines raised $2 billion at a $10 billion valuation with no released product, later valued at $50 billion
HOWARD MARKS/
He notes how narrow the market’s dependence on one theme has become.
- AI stocks accounted for 75% of S&P 500 gains, 80% of profits, 90% of capex
HOWARD MARKS/
He grounds the downside in what actually happened to earlier inflection bubbles, where the technology survived and the shareholders did not.
- RCA lost 97% after 1929
- aviation stocks fell 96% by 1932
HOWARD MARKS/
He lays out the paradox at the center of performance assessment: the actuarial return is the only thing that matters in the long run and tells you nothing in any single year, so the short-run yardstick has to be a relative one — yet matching peers in a frothy market may only prove you were as imprudent as they were, and the policy portfolio lets whoever set it off the hook.
- a 6¼% actuarial assumption
HOWARD MARKS/
On bubbles his selection pairs the psychology with the excuse: the dot-com boom rested on excessive belief in the new new thing, and the phrase by which soaring prices often get justified is that this time it is different — which he tested against a list of such theories rather than dismissing outright.
- nine such theories examined in This Time It’s Different
HOWARD MARKS/
Value pulls on price like a magnet, but only over the long run: prices can go to crazy extremes in either direction, and that capacity is exactly what produces bubbles and crashes. He treats betting on convergence as intellectually sound and betting on it happening soon as dangerous.
THE CALCULUS OF VALUE →HOWARD MARKS/
He stacks up the behavioral indicators alongside the multiple, including the credit market, where the compensation for giving up Treasury safety has thinned to near its historical minimum.
- the S&P 500 valued at more than 3.3 times sales, an all-time high (Financial Times, July 25, citing Bloomberg)
- a Barclays equity-euphoria indicator at twice its normal level, in territory associated with asset bubbles
- the ratio of US market capitalization to GDP at an all-time high, and understated because companies stay private longer
- yield spreads approaching all-time lows
HOWARD MARKS/
He states the bull case rather than caricaturing it — that today’s index is increasingly made of companies that grow faster, are less cyclical, need less capital to grow and hold stronger moats, and therefore deserve above-average multiples — and grants that the explanation makes complete sense. What he will not do is adjudicate: he says he cannot tell whether the greater error today is falling for "it’s different this time" or failing to notice when it is, and warns that in most new new things far too many companies, often the wrong ones, get treated as winners.
- Templeton put it at 20 percent of the time that things really are different; Marks would bet on more than 20 percent today
HOWARD MARKS/
A bubble is a state of mind rather than a calculation, he argues, and the surest tell is hearing “there’s no price too high.”
ON BUBBLE WATCH →HOWARD MARKS/
Concentration is the fact he leads with, and it now exceeds the peak of the bubble he wrote about 25 years earlier.
- the top seven were 32-33% of S&P 500 capitalization at the end of October
- roughly double the leaders’ share five years earlier
- against a prior 28-year high of roughly 22% in 2000
- US stocks over 70% of the MSCI World Index, the highest since 1970
HOWARD MARKS/
Newness lets prices escape gravity — no history, nothing to tether them — and a mania usually has a grain of truth taken too far.
- a 200x payoff makes a 1% chance of success mathematically worth funding
- stocks listed down 99% in the aftermath of the TMT bubble
HOWARD MARKS/
His own baptism supplies the evidence: the best companies in America, bought at any price, lost most of their value in five years.
- well over 90% lost by buying in September 1969 and holding five years
- p/e ratios falling from the range of 60 to 90 to the range of 6 to 9
- the market as a whole down about half in 1973-74
HOWARD MARKS/
Starting valuation and the next decade’s returns are strongly related, with minor variations but no serious exceptions.
- J.P. Morgan Asset Management: just short of 324 monthly observations from 1988 through late 2014
- J.P. Morgan Asset Management: buying at a multiple in line with today’s 22 always produced ten-year returns between plus 2% and minus 2%
- today’s p/e well into the top decile of those observations
HOWARD MARKS/
Two straight 20% years have now happened five times, and three of the earlier four saw declines after — a small sample, he notes.
- the S&P 500 up 26% in 2023 and 25% in 2024, its best two years since 1997-98
- Bitcoin up 465% in the last two years
HOWARD MARKS/
He then argues against himself: the multiple is high but not insane, and he hears no one saying “there’s no price too high.”
ON BUBBLE WATCH →
Claims are restated; figures are the authors' own. Each links to the document it came from, which in turn links to the original in full.