William Bernstein’s old advice to investors: once you’ve won the game, stop playing. Corporate pension plans just won the game.
This week’s chart shows the Milliman 100, the 100 largest U.S. corporate pension plans. They’re about 112% funded, the highest level since 2001. Milliman puts the surplus at $141 billion as of August 31. A year earlier the funded ratio was 104.9%.
Two things did the heavy lifting. A strong stock market grew the assets, and the discount rate used to value pension liabilities climbed to 6.00%, which shrinks the present value of what plans owe. Rising assets and shrinking liabilities is about as good as it gets for a pension actuary.
So what does a plan sponsor do with a surplus? Lock it in.
That could be a risk for equities, and it’s mechanical. When stocks outperform bonds, a plan’s equity weight drifts above target and gets trimmed back. Citadel Securities notes that strong funding pushes plans to “de-glide” and immunize portfolios, creating the potential for mechanical equity selling and fixed income buying, especially around quarter-end.
“De-glide” is the word I’d underline. Most corporate plans run a glide path: every time funded status crosses a trigger, they permanently move money out of stocks and into long bonds that match their liabilities. Quarterly rebalancing reverses when markets turn. Glide-path moves usually don’t.
Twenty years ago, the Milliman 100 plans held 61.7% of their assets in stocks and 28.6% in bonds. By the end of 2025 that had flipped: 24.1% in stocks and 52.7% in bonds. That trend has run one way for 20 years.
The other 23.3% is a grab bag of real estate, private equity and debt, hedge funds, commodities, and cash. That bucket has more than doubled since 2005, much of it can’t be sold quickly, and Milliman notes the companies’ SEC filings generally don’t say what’s in it. So when a plan trims risk, the selling lands on what it can actually trade, and public stocks are the easiest thing to sell.
And bonds are worth owning again. As of September 24, Treasury yields were above 5% from the 5-year out to the 30-year, with the 10-year at 5.18%. The 10-year TIPS yield was 2.85%, a real return you can lock in for a decade. For a plan that’s already won, a contractual 5% beats swinging for the fences.
Now, the case against my own argument. JPMorgan pegged June’s quarter-end rebalance at roughly $165 billion of net global equity selling. U.S. pensions were about $55 billion of it, with Japan’s and Norway’s giant funds and the Swiss National Bank supplying most of the rest. That sounds like a lot until you set it against a U.S. stock market worth something like $65 to $70 trillion. The U.S. pension piece is less than 0.1% of that, and buybacks and dip-buyers can absorb it.
And the rest of the institutional crowd hasn’t won the game. Milliman has the 100 largest public pension plans at 88.2% funded, an $816 billion hole. College endowments ($944 billion across 657 schools) keep about 86% in equities and equity-like strategies, and they spent 4.9% of assets last year. Private foundations have to pay out 5% a year. A 2.85% real yield doesn’t fund that, so don’t expect endowments and foundations to join the selling in size. They’ll rebalance at the margin, but they run no glide path.
What I expect is a missing buyer. Over the next few quarters, the most patient institutional money in the country keeps walking from stocks toward bonds, at a time when equities are priced for perfection and bonds are earning real money again. My guess is that shows up as sloppier quarter-ends, a lower ceiling on stock multiples, and perhaps steadier demand for the long end of the yield curve.
The same math applies to individuals approaching retirement. If you’ve already won the game, it’s worth asking how best to “de-glide” (and, in a taxable account, how to step back without handing a big slice of the winnings to the IRS).
Sources: Milliman Pension Funding Index, September 2026; Milliman 2026 Corporate Pension Funding Study; Milliman Public Pension Funding Index, 7/31/2026; Citadel Securities, “September Setup: The Asymmetry Has Changed” (8/31/2026); Treasury yields as of 9/24/2026 (Forbes Advisor); 10-year TIPS yield, FRED DFII10; JPMorgan quarter-end rebalancing estimate via Investing.com (June 2026); U.S. market cap context, 24/7 Wall St. (6/22/2026); FY25 NACUBO-Commonfund Study of Endowments; Mercer summary of the FY25 NACUBO-Commonfund Study.