Be careful, David! Donald Trump wants to keep morale up, and you’re not helping.

For the second time in a month, Goldman’s David Kostin cut his year-end 2025 S&P target, a dicey decision to the extent it runs afoul of a White House engaged in an effort to suppress any and all manifestations of dissent among prominent American institutions.

Kostin now sees the S&P ending the year at 5700, barely higher than where it started this week. He cited “higher tariffs, weaker economic growth and greater inflation than we previously assumed” in explaining the bank’s new target, which counts as the most bearish on the Street.

Goldman’s 12-month forecast sees the benchmark at 5900, which is to say Kostin doubts US equities will be perched at new records this time next year.

The figure on the left, above, shows you the bank’s new EPS forecast which, at $253, is $16 (!) below bottom-up consensus. The figure on the right’s a bit concerning too: Any material multiple compression, when applied to Kostin’s new index EPS outlook, results in sharply lower stock prices.

“With little change to consensus EPS estimates, all of the 9% selloff from the market peak in February has stemmed from valuation contraction,” Kostin wrote, adding that he expects the index multiple to fall further, reaching 19x in three months before “rising modestly to 19.5x in 12 months.”

There’s no sense ignoring the elephant in the room: If Goldman’s right about index-level (i.e., aggregate) EPS, then JPMorgan better be right to suggest valuations have reset structurally higher. Because with allowances for the contention that a “forward” multiple should apply to 2026 EPS at this point, a P/E that simply contracts to the 10-year average would translate into a bear market for the S&P. A P/E that resets to the three-decade average would result in a near 30% decline. If valuations compress to the lows seen in 2018 — and again assuming we’re applying the multiple to Goldman’s new 2025 EPS call — you’re looking at a near 40% crash.

For now, Goldman’s macro model suggests limited multiple contraction. As noted above, Kostin expects the S&P to trade on a 19.5x multiple looking out past three months. But he nevertheless sounded a cautious tone. “If the growth outlook and investor confidence deteriorate even further, valuations could decline much more than we forecast,” he wrote, noting that “during the last three major S&P 500 downturns, the P/E multiple bottomed at 15x (2022), 13x (2020) and 14x (2018).”

Over the weekend, the bank’s economists again revised their tariff assumptions. Jan Hatzius now sees the average US tariff rate rising to 18%, 15ppt above the 2024 rate. Growth will slow to 1.5% and core PCE inflation will average 3.5%, the bank reckons. Previously, those projections were 2% and 3%, respectively.

Recall that in October of 2018, Kevin Hassett said Goldman “almost at times looks like the Democratic opposition.” He was referring to Kostin, and specifically to David’s warning that slapping tariffs on everything the US imports from China could dent US corporate profits. That was during “Trump 1.0.” We’re in an entirely new environment now when it comes to what Trump’s willing to countenance in terms of criticism. And remember: He doesn’t differentiate between analysis that posits suboptimal outcomes from his policies and criticism.

Folks, I’m not joking when I say that if things continue along the current trajectory from a domestic political perspective, this sort of analysis — which is to say bearish, or even relatively bearish calls on US equities and the US economy emanating from major Wall Street banks, where research departments are the furthest thing from sacred cows — will disappear by the mid-terms.