Roger Aliaga-Diaz: We’re underweight in stocks in our time-varying portfolios. For example, we’re more conservative than the traditional 60 percent stocks / 40 percent bonds risk profile. We’re at 40/60.

Why? It’s not pessimism about AI or the economy. It’s about risk from a stock market correction.

U.S. equity market valuations are stretched. Sure, stocks have gone gangbusters for much of 2025 and, who knows, the momentum and hype may continue for a while. But our analysis of fundamental drivers points to greater odds that longer-term returns will be subdued, below historical averages.

Meanwhile, bonds remain attractive in the high-interest-rate environment. Our models project that, over the next decade, a 40/60 portfolio can achieve similar returns to a 60/40, but with less risk. In other words, markets may not appropriately reward the additional risk of stocks.

We also balance risk within asset classes. Corporate debt valuations are stretched, so we’re cautious about credit and high yield, but we’re overweight Treasuries.

Within equities, our cautious views stem from growth stocks. Their prices already reflect high expectations for AI, so they have little room for expansion.

Our view on value stocks is more constructive. It may seem counterintuitive, but if AI and technology deliver on their promise, value stocks are the ultimate beneficiaries.

Finally, non-U.S. stocks have enjoyed a big runup in 2025, so their valuation gap with U.S. stocks isn’t as great as it once was. But their return prospects remain attractive, and they play a valuable portfolio diversification role.