Five Signs It's Time to Reduce Equity Exposure

The S&P 500's forward earnings yield — the profit return you get back relative to what you pay in — is roughly 4.2 percent. The 10-year Treasury, guaranteed by the U.S. government, yields 4.8 percent. You are getting paid less to take more risk.

That gap matters more than any headline about whether the market is in a bubble. Valuation warnings are not timing signals on their own. But when multiple independent measures point in the same direction, the question stops being "when will this correct?" and becomes "what am I actually paying for by staying in equities?"

The data right now does not tell you to sell everything and wait on the sidelines. It tells you to look at your portfolio the way a process-driven investor would — with fresh eyes, no nostalgia, and a comparison set that actually works.

The valuation stack

Here are five signals that have stacked up over the past year. Not as a checklist for panic, but as a way to read what the market is pricing.

One: The Shiller CAPE has re-entered territory seen only during past peaks. The Shiller cyclically adjusted price-to-earnings ratio smooths out earnings over a full decade to avoid being misled by a single boom year. As of late July 2026, it sat at 40.5. The long-run historical average is around 17. The only time it was higher was at the dot-com peak in 1999, when it touched 44.2 — before the S&P 500 fell 49 percent. This measure does not predict when a decline will start. It tells you that the earnings return embedded in today's prices is historically thin. If future returns follow the pattern of the past, extended periods after a CAPE above 30 have produced disappointing decade-long outcomes. The metric is backward-looking and has been wrong about timing many times, but the level itself is not in dispute.

Two: The Buffett indicator has hit a record. The total U.S. stock market sits at roughly $77 trillion — roughly 237 percent of U.S. GDP, as of early September 2026. Warren Buffett called this ratio "probably the best single measure of where valuations stand at any given moment." The 20-year average is around 130 percent. At 237 percent, the market is at its highest reading ever recorded. One legitimate critique: about 28 percent of S&P 500 revenues come from outside the U.S., so GDP is an imperfect denominator. Even adjusting for foreign revenue, the reading is well into territory that historically implies muted forward returns.

Three: The dividend yield has collapsed. The S&P 500's dividend yield today is about 0.98 percent. In earlier market cycles, a yield below 3 percent was considered a danger zone — and the old threshold was meaningful when dividends returned most of corporate profits to shareholders. Today, buybacks have absorbed the difference. The result: you can get the same cash flow through repurchases without the yield showing up on a standard metric. The practical point is not that dividends themselves matter less. It's that the total cash flow cushion investors used to rely on — the part of the market's return that came from actual corporate payouts rather than price appreciation — has shifted into a form that disappears when companies cut back on buybacks during a downturn.

Four: The equity risk premium has vanished — or flipped negative. The earnings yield on the S&P 500, at roughly 4.2 percent, sits below the 10-year Treasury yield of 4.8 percent. That means the guaranteed return on a government bond exceeds the profit return you earn by owning the broad stock market. Over the past century, the equity risk premium — the extra return stocks have paid for their additional risk — averaged 6.2 percent per year. At current levels, that premium is not just thin. It's negative. You are paying equity prices for what used to be a below-market yield. The Wall Street Journal flagged this dynamic in a headline that says it all: "The Risk Premium for Holding Stocks Over Bonds Is Vanishing."

Five: Breadth is narrowing even as the index climbs. The cumulative advance-decline line for the S&P 500 — which tracks how many stocks are participating in the rally, not just where the index sits — peaked in mid-April 2026 and has trended lower since. The index keeps reaching new highs, but fewer stocks are doing the lifting. A narrow rally is not a crash signal by itself. It is, however, a structural weakness. When leadership concentrates in a handful of mega-cap names, the average holding in a broad index fund is underperforming the index headline. And concentrated leadership tends to break.

What this does not mean

Valuation extremes do not tell you when the market will fall. The CAPE was above 30 for years during the late-1990s bull market. Japan's market stayed overvalued by historical metrics for nearly a decade starting in the 1980s. Markets can and do remain expensive for uncomfortable stretches. A high CAPE is a speed limit, not a stop sign.

The same goes for the Buffett indicator. It has stayed above 100 percent for long periods during past bull markets without triggering an immediate reversal. And the narrowing breadth signal has been present for months, not days.

None of this is a prediction that the market will crash tomorrow. It's evidence that the current prices embed an enormous amount of confidence.

What it does mean for portfolio decisions

Here is how a process-driven investor reads this stack.

The five-factor framework — valuation, growth, profitability, momentum, and earnings revisions — is designed to tell you what to own, not when to time the market. But valuation is one of those five factors, and it is currently flashing the weakest readings we've seen in a generation. That does not mean every stock is a sell. It means the default position — "I'm in the S&P 500 and I'll stay there" — deserves a harder look.

What the data supports is not capitulation. It's structure.

Reduce broad-market exposure and tilt toward factor strength. If you hold a passive index fund, the entire portfolio carries the same valuation drag. A stock-level screen identifies companies where growth and profitability are strong enough to justify premium valuations and names where the factor stack remains intact despite the elevated market. The difference between a stock that can grow into its multiple and one that cannot is the only question that matters when the overall market is expensive.

Add a barbell sleeve. When uncertainty rises, the answer is more structure, not louder predictions. Pair the growth names you own with dividend-paying businesses, REITs, or companies with durable cash flows — names that return cash to shareholders regardless of whether the broader market moves up. The S&P 500's dividend yield is near 1 percent. Individual high-quality dividend stocks and bond portfolios offer yields several times higher. That spread is not an accident. It's the market telling you where the yield lives.

Treat cash as a position, not a mistake. Berkshire Hathaway holds roughly $366 billion in cash — an unprecedented level that mirrors Buffett's behavior before both 2008 and 2020. The M2 money supply, after expanding dramatically during the pandemic, has contracted for the first time since the Great Depression. In a normal environment, cash sitting idle is a drag. When the equity risk premium is negative and multiple valuation indicators are at or near records, cash is optionality. You are not paid to be right — you are paid to be able to act.

Rebalance on a schedule, not on fear. Vanguard's latest capital markets model projects 10-year annualized US equity returns of roughly 2.8 to 4.8 percent — below what bonds are expected to return. The firm's strategic allocation leans 40/60 (stocks/bonds), a shift from the traditional 60/40 that reflects these projections. That is a quantitative conclusion, not market timing. Your own rebalancing discipline serves the same function: sell what has run, buy what has lagged, and let the process do the work.

The bottom line

Markets do not fall because indicators are high. They fall because something changes — earnings disappoint, rates rise faster, a geopolitical shock reroutes energy flows, or the AI investment cycle fails to produce the returns that justify the spending. The valuation stack I've laid out doesn't predict which trigger will fire. It tells you that the buffer between where the market sits and where it could fall is thinner than at almost any point in modern history.

The five signals here are not a prediction. They are a risk assessment. And the appropriate response to a thinning buffer is not panic — it's process. Know what you own, know why you own it, and know what would change your mind about keeping it.