The S&P 500 crossed 7,500 this week. Index funds look great. Financial media is positive. Nobody's really asking what this price actually requires to be true. Or what happens to everyone when it isn't.
I'm not predicting a crash. Prices can stay detached from fundamentals longer than any analysis expects. The people calling dot-com a bubble in 1997 were right and irrelevant for three more years. But if you have money in U.S. equities, there's a version of this story worth knowing before the cheerful headlines are the only thing you've read.
The number isn't the problem
7,500 is just a price. Every price implies a story: what earnings will look like, what growth is sustainable, how much risk is already baked in. The real question is whether the story this price tells is plausible, or whether it needs too many things to go right at the same time.
When you run the actual metrics, the answer is uncomfortable.
The valuation problem
The most honest way to measure whether a market is expensive is the Shiller CAPE ratio, a price-to-earnings multiple that smooths earnings over 10 years to strip out boom-and-bust noise. Robert Shiller built it specifically to avoid the distortions that make markets look cheap right when they're most dangerous.
Shiller CAPE Ratio — Historical Context
CAPE above 35 (dashed line) has historically preceded major drawdowns. We are at 38.
The second-highest sustained CAPE reading in recorded market history. The only period above where we are now ended with the S&P dropping 49% over two and a half years.
At this CAPE level, historical models project annual returns for U.S. equities of around 1.7% over the next decade. The long-run average is 7–10%. Vanguard's own forecast for U.S. equities sits at 4–5% nominal. Research Affiliates projects below 1% in real terms. None of that is in the headlines.
Some analysts argue the CAPE is outdated. That structural changes like higher profit margins, tech-driven earnings, and share buybacks justify a permanently higher ratio in the 25–30 range. That's a fair debate. But we're at 38. Even the most generous "new normal" argument doesn't get you there.
This is not actually a 500-company index
The S&P 500 is supposed to give you broad exposure across the U.S. economy. In practice, seven companies (Nvidia, Apple, Microsoft, Meta, Amazon, Alphabet, Tesla) make up roughly 35% of the entire index. Nvidia alone is over 7%.
The whole premise of index investing is that you're not making concentrated bets. But when seven companies drive more than a third of your returns, you are making concentrated bets. You just didn't consciously choose them.
The concentration problem
One bad earnings quarter from Nvidia. One antitrust ruling. One AI capital expenditure write-down from a major cloud provider. The index takes a real hit regardless of how the other 493 companies are performing. That's not diversification. It's concentration with extra steps.
S&P 500 Weight — Mag 7 vs Everyone Else
Nvidia alone accounts for over 7% of the index. One stock.
The AI earnings gap
The bull case for these valuations rests on earnings growth, roughly 14% for 2026, with nearly half from tech. That's what the numbers need to justify current prices. It's an aggressive assumption being treated as a base case.
Most of these companies are still in heavy capital expenditure mode. Trillions going into data centres, chips, infrastructure. AI revenue is growing but hasn't caught up to the spending yet. The earnings implied by today's prices are for work that hasn't been monetised yet.
If earnings growth lands at 7% instead of 14%, the math on current multiples breaks. Markets can handle uncertainty fine. What they're bad at is the gap between what was expected and what actually showed up.
Optimism priced in as certainty is a different thing from optimism. The distance between the two is where crashes happen.
What Warren Buffett's favourite metric says
The Buffett Indicator divides total U.S. stock market capitalisation by GDP. It's simple. It asks whether the market has grown faster than the actual productive output of the economy, and if so, by how much.
Buffett Indicator — Market Cap vs GDP
Buffett called anything above 200% "playing with fire." The current reading sits 20 points past that.
The economy hasn't grown proportionally to the market. The market got more expensive relative to what the country actually produces. One of those numbers has to move toward the other.
The part nobody talks about: what happens to everyone else
People talk about market corrections like they're a portfolio event. Number drops, you wait, it recovers. That works for moderate corrections from normal valuations. It's not how it plays out when you're unwinding from extremes.
U.S. household wealth is heavily tied to equity portfolios. When the market drops hard from this height, the sequence is predictable: consumer spending contracts, companies freeze hiring, banks tighten lending as collateral falls, pension funds already running on optimistic assumptions start showing stress.
At this scale, the S&P isn't just a financial instrument. It's a confidence mechanism for the entire economy. A serious correction from 7,500 doesn't stay in brokerage accounts. It moves into real employment, real lending, and real household spending within 6–12 months.
What history actually says about CAPE at these levels
The Shiller CAPE crossing 39 has happened three times in recorded market history.
Three times in 140 years
1929
CAPE crossed 39 in the run-up to the crash. What followed was an 86% collapse over three years.
1998–2000
CAPE crossed 39 in 1998. The market ran another 44% before peaking in March 2000. Then dropped 49% over the next two and a half years.
Now
CAPE near 38, approaching 39. The dominant narrative is that AI is a genuine structural shift that justifies the premium. In 1929 it was industrial electrification. In 1999 it was the internet. The technology was real both times.
The point isn't that a crash is coming. The 1998 comparison shows markets can run further from here before the reset. The point is you're in historically rare territory. The two prior times ended badly. That doesn't make it a certainty, but it does make it a fact you should have.
What you actually do with this
Panic-selling because of valuation concerns has historically cost investors more than the corrections they were trying to avoid. I'm not saying get out. But staying invested and not thinking about what you own are two different things.
Some questions worth actually answering for yourself: How much of your net worth is in U.S. large-cap equities, and was that a conscious choice? Do you have exposure outside the U.S. market? Non-U.S. developed market equities trade at significantly lower multiples right now. If U.S. markets went sideways for five years, what does that do to your timeline? And are you invested with money you genuinely won't need for 10+ years, or has "long term" quietly become something shorter in practice?
The actual point
Nobody can tell you when this corrects, or whether it corrects before going higher. What the data can tell you is that current valuations imply below-average returns for the next decade, that concentration in a handful of tech companies is higher than it's been in modern market history, and that the gap between market value and economic output is at its widest point ever recorded. Understanding what you own is not pessimism. It's just literacy.
The market might keep climbing. It genuinely might. But a lot of Malaysian investors with U.S. equity exposure through global ETFs, unit trusts, or brokerage accounts are operating on "it went up last year" as their risk framework. That's not a plan. It's just inertia with a good recent track record.
Ascendant Collective
Real numbers. Real mechanics. No recycled theory.
Financial concepts broken down for Malaysians who want to understand what's actually going on, not just follow the crowd. Free, every week.
Subscribe Free
Subscribe to our newsletter.
Become a subscriber receive the latest updates in your inbox.
