277: Global Equity Markets: Can the Stock Market Rally Continue Beyond AI?

Carrie King · BlackRock

S&P 500 gains since ChatGPT are mostly earnings growth, not valuation expansion.

Since ChatGPT launched in November 2022 the S&P 500 has roughly doubled. S&P 500 earnings rose about 75% over that period, while the price-earnings multiple moved from just under 18x to about 19x, roughly an 8% revaluation. Cumulative dividends added about another 10%. Earnings growth was dominated by tech, moving from semiconductors, hardware and memory into data centers, power, electrical equipment and cooling. The market's rise is therefore grounded in actual earnings, not sentiment.

“Most of that stock market return is driven by actual earnings growth rather than any kind of revaluation of the companies.”at 6:30

Alex Brazier · BlackRock Investment Institute

Market concentration reflects real earnings concentration, not valuation hype, so it is not a problem per se.

Earnings growth has been concentrated in a few companies, and that is now reflected in concentrated market capitalization. This is not driven by valuations or hope about the future. It reflects what is happening on the ground now in AI-linked earnings. So concentration is not a problem per se, although it leads investors to examine how much AI build-out exposure they hold.

“the earnings growth that we've seen has been pretty concentrated, and that's now reflected in a concentration of market capitalization, which, to emphasize what you both said, isn't to do with valuations and hope about the future. It's actually reflecting what's happening on the ground now.”at 10:38

Alex Brazier · BlackRock Investment Institute

The AI trade will bring heavy volatility because it is a long-duration bet on which small daily views matter.

Stocks tied to the AI build-out are up about 40% and the market has relied on them, yet only about a third of US stocks trade above their 50-day moving average. Semiconductors, the heart of the AI trade, moved up or down more than 5% on more days this summer than in any summer in 24 years. Because the AI trade is a long-duration play, small day-to-day differences in view can change the outcome a lot. Investors with long horizons can ride the volatility, while shorter-term investors who care about it are stepping back.

“The AI trade, even if you're a great believer in the long run returns associated with it, you know it's gonna come with a lot of volatility.”at 4:32

Carrie King · BlackRock

Hyperscaler cloud revenue growth accelerating toward 50% shows AI capex is generating returns.

The consensus expects about $7 trillion of cumulative AI build-out spending over five years, and the key question is when these companies earn a return on it. In the latest quarter the top three hyperscalers' cloud revenue, which she uses as a measure of AI usage, accelerated from 15% growth a year earlier to nearly 50%. Growth is accelerating on a base that has doubled, and more acceleration is expected in Q3. This shows the spending is enabling increased AI usage and supports the long-term upside case for AI.

“The cloud revenue, again, a measure of AI usage from the top three hyperscalers accelerated from 15% growth in the second quarter last year to nearly 50% in the second quarter this year.”at 9:09

Alex Brazier · BlackRock Investment Institute

Real US yields near 3%, highest since 2008, are driven by AI borrowing and squeeze non-AI equities.

Inflation-adjusted US government bond yields have reached nearly 3%, the highest since 2008. This reflects heavy borrowing to fund the AI build-out, which is creating intense competition for capital, so all borrowers pay more. Equities must now generate higher returns to compete. Companies that drive the AI build-out can earn those returns, but companies that bear the consequences without being the cause are under pressure. The US equal-weight index and the European index, which are less concentrated in AI, are down about 5% since the start of summer.

“rates on US government debt adjusted for inflation reach nearly 3%. That's the highest since 2008. And it's really a reflection of all this investment that's going on. People have to pay more to borrow”at 12:57

Carrie King · BlackRock

Healthcare is an attractive non-AI diversifier, with the best upside earnings revisions outside AI.

Healthcare has the strongest upside earnings revisions of any sector outside AI. Policy headwinds are starting to dissipate. Demographics are a tailwind: about 20% of the US population will be 65 by 2030, and that cohort spends three times more on healthcare than the rest of the population. Healthcare also has AI optionality, with a 50% year-on-year increase in the number of healthcare companies reporting tangible economic benefits from AI. When semiconductor stocks fall, healthcare stocks have tended to rise, which makes it a useful diversifier.

“Health care as a sector has the strongest upside earnings revision of any sector outside of AI.”at 16:05

Carrie King · BlackRock

US cyclical recovery is underway, with logistics, transport and airline earnings turning positive.

US manufacturing and services PMIs have shown six straight months of expansion. Earnings in areas such as logistics, transportation and airlines are swinging from negative to positive growth, moving from a recessionary to a growth scenario. This recovery is separate from the AI tailwind, and she sees interesting valuation opportunities in these areas.

“PMIs indicate six straight months of expansion in manufacturing and services in The US, and we're seeing earnings swing from negative to positive growth.”at 17:18

Alex Brazier · BlackRock Investment Institute

Brazil exposure is attractive as a low-correlation AI diversifier with limited downside after the election rally.

After the first round of Brazil's election, the currency appreciated and the local stock market rose. Investors are buying Brazil because it is not well correlated with the AI build-out trade and has limited downside risk or return drag. They are not seeking outsized returns; the value lies in avoiding downside while offering low correlation to AI, which now dominates many markets.

“for anything that's not well correlated with the AI build out and where you don't have big downside risks or a large return drag. They're not asking for big outsized returns. The value in these exposures is just in not having downside, but having low correlation to the AI build out trade.”at 20:00