What will U.S. equities do in 2026?

With the start of 2026, investors inevitably ask the same question: What will the S&P 500 do this year? It’s a fair question—and one that invites no shortage of confident forecasts. Unfortunately, it is also a question no one can answer with any reliability.

At Frontier, we do not model or publish one-year return forecasts. That is not an oversight; it is a deliberate choice. Short-term market outcomes are driven by variables that are inherently unpredictable: policy decisions, geopolitical events, earnings surprises, sentiment shifts, unforeseen external shocks, etc. Over a 12-month horizon, even valuations provide limited guidance.

Valuations matter—eventually

By most historical measures, U.S. equities today are considered “expensive”. Forward price-to-earnings ratios sit well above long-term averages, and history suggests that elevated starting valuations tend to be associated with more muted long-term returns. When we examine the relationship between valuation levels and subsequent 10-year returns, the pattern is clear and persistent: higher valuations have typically led to lower average returns over the following decade.

The exhibit below from Apollo shows the forward Price/Earnings Ratio and subsequent 10-year annualized returns. The horizontal axis represents the forward Price-to-Earnings Ratio or how much one pays for unit ($1) of earnings. Moving to the right means investors are paying a higher valuation multiple for earnings. The vertical axis shows the annualized average return over the NEXT 10 years. The clustering of points suggests a clear relationship: lower starting P/E ratios have generally been associated with higher 10-year forward returns, illustrating the classic “buy low/sell high” principle. Note the red dot suggesting that the next 10 years could well be muted given the relatively high current valuations.

However, this is a statement about long-term averages—not next year.

High valuations today say surprisingly little about what the market will do over the next 12 months. Over short horizons, forecasting the return is often close to a coin flip. Especially trying to get specificity within a single percentage point. A range? Maybe. The time frame is simply too brief. Such short time periods for valuations to exert consistent predictive power are pretty limited.

Below is a similar view from Charles Schwab, but instead of the next 10 years, it shows the next 12 months. In contrast with the previous graph, the axes for this chart are reversed: the horizontal axis represents the equity return, and the vertical axis represents the forward P/E ratio. Here, the equity return is for the next 12 months, rather than the next 10 years as in the prior chart. The takeaway is that the data points for the returns are seemingly scattered, suggesting there is no consistent relationship between valuation levels and returns over the following year. Close to a coin flip.

In short, valuations matter a great deal in the long run. In the short run, they often do not.

A long-term framework

Frontier’s modeling and portfolio construction are intentionally long-term in nature. Our most recent capital market forecasts used in our optimization process reflect an expected return of approximately 6.1% for U.S. Large Cap equities. This figure represents a long-term, average return expectation—not a forecast for 2026, or any specific calendar year. And equally important is the relationship relative to the other 15 asset classes that we model.

That distinction is critical. Long-term expected returns are useful for asset allocation, risk budgeting, and strategic planning. They are not market timing tools.

Interestingly, history offers a complementary perspective. When the S&P 500’s trailing three-year annualized return has exceeded 23%, subsequent three-year returns have averaged roughly 6% annualized. Since 1970, this has occurred 49 times. Today, the trailing three-year return is just over 23%. While this observation is not part of our formal modeling process, it is broadly consistent with our long-term expectations—and again, says nothing definitive about 2026 itself.

A Final Thought

Earlier in my career, we used to hold an informal contest on guessing the S&P 500’s return for the coming year. Despite the participation of seasoned investment professionals, the winners were more often than not the executive assistants or colleagues from marketing. The “propeller heads” in research rarely had the best answer.

There is a lesson in that.

Precision over short horizons is more often than not an illusion. Discipline, diversification, and a long-term perspective remain far more reliable guides than annual predictions.