A little over a month ago, JPMorgan suggested US equity valuations reset structurally higher over the last couple of decades due to, among other things, increased retail investor participation and higher household equity allocations.

The gist of the bank’s analysis was that US shares will remain expensive in perpetuity. Fast forward three weeks, and US stocks were (briefly) in a bear market, and the S&P’s multiple was three turns lower, give or take.

Bad timing, I suppose, but it might’ve just been plain old misguided. If you ask BofA’s Michael Hartnett, it’s more likely that the structural shift in valuations will be lower in the new macro regime.

“We say 2025’s big picture is peak valuations in stocks and credit following the glory of the past five years,” Hartnett wrote, in his latest, noting that the elevated average multiple in the 21st century was a function of “globalization, technology, monetary indulgence” and then, post-pandemic, “US exceptionalism augmented by fiscal excess and the AI boom.”

That era’s over, though, and we may returning to something that looks and feels more like the 20th century, a period BofA described as “cursed by world wars, cold wars, depressions and stagflation.” Note that each and every one of those banes is a trending topic in 2025.

In the event we do revert to a 20th century multiple, ~14x is the destination.

As the figure reminds you, we’ve only seen valuations that cheap on a couple of occasions over the past decade: In late 2018 (the last time Donald Trump was ready to fire Jerome Powell) and in March of 2020 (at the onset of the pandemic).

Going forward, Hartnett suggested, 20x may be a ceiling for multiples “as globalization reverses, the Fed’s less independent, fiscal/monetary excess ends, the US savings rate rises” and inflation of 3-4% becomes the new normal outside of recessions.