U.S. Stocks: Great Expectations, Even Greater Valuations

For many months, U.S. equities have appeared to be heading in one direction – up. However we’ve been crunching key financial metrics and the gap between what investors are paying for U.S. stocks compared to internationally has reached a worrying gap that history tells us may lead to slimmer future returns.

Add in a record concentration in a handful of mega-cap stocks, such as Nvidia, Microsoft and Tesla, and the U.S. stock market increasingly looks priced to impossible perfection. Something has to give.

Five Numbers That Matter

1) ~23× vs ~15× – The Valuation Gap

If we take the 12-month forward P/E, the U.S. market is trading at around 23×, as per the MSCI USA Index, while international markets are coming in more like 15×. That’s a roughly 51% difference.

What this means in basic terms is that investors are paying $23 today for $1 of next year’s earnings in the U.S., while abroad, they’re paying a cool $15.

When you pay that much for anticipated growth, expectations are essentially sky-high with limited room for error. Strong results can keep prices steady, but any disappointments could lead to sharp corrections.

2) ~65% – U.S. Share of “Global” Equity Indexes

The U.S. makes up about 65% of the MSCI All Country World Index, so even most “global” portfolios are heavily tilted to America by default.

This means if investors put $1 in a “world” fund, about 65 cents of that will actually end up in U.S. stocks.

Lots of investors understand that the U.S. absorbs a significant portion in many funds, but not all truly appreciate the size of that skew, and what it means for even their supposedly “globally diversified” funds.

Ultimately, many retail investors will be taking a much bigger single-country bet than they may realise, and even diversified holdings won’t be insulated from waning enthusiasm.

3)~26% vs ~15% – America’s Slice of the World Economy

The U.S. produces roughly 26% of global GDP (nominal) and more like 15% after adjusting for living costs, according to data from the IMF. This is far below its approximately 61% share of world equities.

This shows another huge gap – America’s stock market is far bigger than its share of the world economy, with investors paying a premium while being heavily exposed to how the U.S. performs.

Now it’s fine for substantial gaps between market capitalization and economies to persist, but they do raise the bar: profits must keep outrunning the rest of the world for the premium to make sense.

4) ~66% U.S. Weight in a “Total World” ETF (VT), With ~4% in Nvidia

Vanguard’s Total World Stock ETF (VT) still invests about 66% in U.S. stocks, while roughly 4% sits in a single name, Nvidia. That’s one stock making up around 1/20th of the entire fund.

Even a fund that is supposed to “own everything” is two-thirds in the U.S., with a single stock taking up a decent chunk. What happens if, or should we say when, Nvidia falters…

A single hiccup in a top holding equity could dent portfolios far more than individual investors may expect.

5) ~50% – AI’s Contribution to 2025 S&P 500 Gains

This year, about half of the S&P 500’s returns have come from companies linked to AI (chips, platforms, infrastructure), a statistic echoed by Bloomberg.

That is either a tremendous achievement for this potentially revolutionary technology, a significant worry, or most likely – both.

This means if the AI narrative wobbles, and some headlines are hinting at that, then the index could start to look shaky too.

Thus far the shift in coverage across trusted financial outlets, heading more towards “Maybe AI firms are actually overvalued” doesn’t seem to be trickling into the share price of major firms, with Nvidia alone enjoying an over 7% bump in the last 5 days.

However, it often takes some time for the message to echo across retail investor circles, even in today’s world of instant technology.

Based on SPDR’s 30 Sept, 2025 data, with each S&P 500 sector’s index weight multiplied by its year-to-date returns to estimate sector contribution to the index. 

Why This Data Spells Fragility Even If It’s Not Quite “Crash”

If we rewind the clock, we can see that expensive starting points can start to cap future returns. High forward P/Es aren’t timing tools, but they do tend to soften multi-year returns because so much good news is already priced in.

Similar types of valuations showed up in 2001–02 (dot-com) and 2020–21 (pandemic tech surge). In fact, since the end of 2001, U.S. stocks have delivered about 9% Compound Annual Growth Rate (CAGR). But in the decade after 2001 (when stocks were still very expensive by conventional measures), the market went nearly a decade without any nominal gains.

Now with roughly half of 2025’s gains tied to AI names, market performance primarily depends on a small group of winners. One domino wobbles, another starts to look shaky, and before you know it the train could be leaving the station and heading straight off the cliff.

Where Cracks Might Show First

Assuming the U.S. stock market is overvalued (it is) and that it won’t go on like this (it can’t), then it’s prudent to consider which sectors and equities could see a shift in their prospects first:

1. Semiconductors/AI platforms: Nvidia (NVDA) is top of our list. If AI demand or spending pauses, a lot of the future success already priced in could unwind quickly. So far it’s managing to make its financials look rosy, in part through circular investments and loans from other AI firms. But it’s got to really deliver and meet almost impossibly high expectations,
2. EVs/speculative growth: Tesla (TSLA) had dodged the naysayers for too long. Its valuation still leans heavily on its “hyper-growth” label, yet the automobile market is facing margin pressure and rising competition from pretty much every part of the market. Tesla generates approximately 75% of its revenue from autos, and 12-13% each from energy and services/other. Auto companies commonly trade at 0.6x-0.8x revenue. Taking the top-end of that range, that means ~95% of Tesla’s value is coming from highly capital-intensive moonshots projects that it’s promised, but shown no evidence of producing. The company is currently 2-3% of the S&P 500 and thus a pillar in many global portfolios.
3. Crypto-exposed finance: Coinbase (COIN) is waving the crypto banner from the front. However, its revenues are heavily dependent on trading volumes, so a drop in activity or tighter regulations could hit earnings fast. It might feel unlikely with the existing U.S. administration and some institutional interest, but who knows what the mood will be around the corner.

Thinking Ahead

The current state of U.S. equities may well make some investors consider diversifying across more regions and elements.

With the U.S. at around 61% of world indexes, deliberately seeking out more exposure beyond the U.S. , i.e. developed and emerging markets, might reduce the single-country risk without abandoning the U.S.’ historical growth engine.

Equally, investors could think more about how to separate great technology from great entry prices. AI can be both transformative and over-discounted in prices, so expect some potentially rocky payoffs across both winners and losers.

This analysis reflects personal opinions and does not necessarily represent DayTrading.com. It is for information/education only and is not investment advice, nor an offer or solicitation. Readers should do their own research and consider professional advice. All investments carry risk, including loss of capital; past performance is not a reliable indicator. Any forward-looking statements are uncertain and may change without notice. Data and figures are believed reliable but not guaranteed.