In recent weeks we have seen equity markets routinely hit record highs, buoyed by strong first-quarter earnings and a reconviction among investors in the artificial intelligence (AI) investment theme, most visibly and powerfully demonstrated in the semiconductor sector.
Yet at the same time investor demand for select stocks has seemed almost insatiable—not all signals point the same way. By one measure, equities look as unattractive relative to bonds as they did during the dot-com bubble at the turn of the millennium.
That measure (for U.S. large-cap stocks) is the equity risk premium (ERP), which is the return an investor expects from equities above what they could earn from risk-free bonds, such as U.S. Treasuries. For most of the past two decades, that premium has been meaningfully positive, rewarding investors for bearing equity risk.
Today the earnings yield on the S&P 500 index—the profit companies generate relative to stock valuations, expressed as a percentage—is roughly equal to, or in some calculations below, the yield available on U.S. Treasuries, a development last seen most strikingly at the height of the dot-com bubble.
A Yellow Flag
Before drawing conclusions, a word of caution about the metric itself. The ERP, as conventionally measured, is something of an imperfect comparison. The yield to maturity on bonds represents the discount rate which equates the market bond price with guaranteed, fixed cash flows—you know precisely what you will receive up until the bond’s maturity. Equity cash flows, by contrast, are uncertain and largely grow over time. A stock that compounds earnings at 10% annually is a fundamentally different asset than a bond yielding 4.5%, even if today's snapshot makes them look equivalent. On that basis, equities at 21–22 times earnings are elevated but far from egregious.
That said, yellow flags like this deserve attention. The last time the ERP was at these compressed levels—the late 1990s—it persisted for longer than most expected before ultimately unwinding sharply. The lesson of that episode is not that negative ERP triggers an immediate correction, but that it narrows the margin for error. When concentration is high, when a single theme dominates, and when the gap between what you pay and what you earn compresses, the portfolio becomes increasingly vulnerable to unwinds when growth disappoints relative to expectations.
We have made the case in several pieces—including AI Capex: It’s Not All or Nothing and AI: Boom? Bust? Or Both?—that the current market environment is distinct from the dot-com era, but we do concede today's market dynamics have several parallels to the late 1990s. Index concentration in a handful of mega-cap technology names remains historically elevated. The AI investment cycle—exciting, real and ongoing—has driven the bulk of equity returns and commands premium valuations. We are not dismissing it: we believe this theme has further to go and could continue to drive returns for multiple quarters. But investors who are entirely tethered to it carry a risk that diversification can mitigate without sacrificing return potential.
Broadening the Base
So what does a sensible response look like? The goal is not to reduce return potential dramatically, but to broaden the base from which returns are generated.
We would start with bonds. ERP compression has been driven by significant increases in bond yields over the past four years, not by equity valuations. Since 2022, bonds have been impaired as a reliable “hedge” to equities at key moments of duress. Much of that failure was tied to persistent inflationary pressure that made duration exposure toxic. Now, with bond yields having re-rated to more normalized levels, we believe fixed income offers attractive absolute returns and has recovered much of its diversifying utility. Bonds are a logical first port of call.
Beyond fixed income, equity breadth can be found across several areas. Our Asset Allocation Committee, for instance, favors a modest overweight to U.S. large caps—given strong earnings growth and reasonable valuations—as well as small- and mid-caps. At the same time, we believe investors are well served maintaining exposure to more attractively valued international markets. The Committee holds a modest overweight in Japan and in emerging markets equities, including China, India and Brazil. Liquid diversifiers round out the toolkit for those seeking further balance.
Discipline Over Drama
None of this, in our view, requires a dramatic repositioning. It requires discipline—the discipline to acknowledge where we are in the cycle, to respect the signal that the ERP is sending even while understanding its limitations, and to build portfolios that are genuinely diversified.
The AI supercycle may well continue. Earnings may continue to surprise. But the investors who are best positioned are those who don't need any single outcome to be right. That is what breadth means—and right now, breadth is worth seeking out.
What to Watch For
Monday 06/01:
- U.S. ISM Manufacturing Employment
- U.S. ISM Manufacturing Purchasing Managers’ Index
- U.S. ISM Manufacturing Prices
Tuesday 06/02:
- Eurozone Consumer Price Index
- U.S. JOLTS Job Openings
Wednesday 06/03:
- U.S. ADP Nonfarm Employment Change
- U.S. Services Purchasing Managers’ Index
- U.S. ISM Non-Manufacturing Purchasing Managers’ Index
- U.S. ISM Non-Manufacturing Prices
Thursday 06/04:
- U.S. Initial jobless Claims
Friday 06/05:
- U.S. Average Hourly Earnings
- U.S. Nonfarm Payrolls
- U.S. Unemployment Rate
- Eurozone GDP