Anywhere between 5% and 10%.

That’s how much US equities could sell off as a result of Donald Trump’s tariff broadside. According to Wall Street.

Obviously, any drawdown could be far larger or entirely negligible, depending on a whole host of factors, most of which are impossible to quantify ahead of time, let alone incorporate into any sort of modeling exercise.

But you gotta cut Wall Street some slack on this. If you’re a sell-side economist or strategist of any sort, you can’t not try to guesstimate the impact of a trade war. Clients are calling with questions, and you gotta (pretend to) have some answers.

The figure above’s just eye candy from Goldman, whose David Kostin charted the spike in Polymarket odds for the overall US tariff rate which, before “Trump 2.0,” stood at 2% on a trade-weighted average basis.

If you ask Kostin, the S&P has “near-term downside” of “roughly” 5% assuming markets price “sustained implementation of the newly-announced tariffs.” That comes from rolling up the output of an EPS model with estimates of valuation sensitivity.

Assuming (naively) the C-suite’s going to eat the higher costs associated with new taxes on imports (which, again, are paid by US companies, not by China or anyone else), profit margins will contract.

Considered as a monolith, corporate America can afford the hit. As the figure above reminds you, profit margins are (still) loitering near record highs.

That’s not how it’ll work, though. Companies aren’t going to absorb the cost of Tariff Man’s windmill-tilting. This is shareholder capitalism, after all. Shareholders come first, consumers second and everyone else later, last or not at all.

But as Kostin pointed out, passing on the costs to consumers jeopardizes volumes, so there’s really no way around lower sales, unless management can prevail upon suppliers to absorb the tariffs, but (one more time) those are US companies too.

“We estimate that every 5ppt increase in the US tariff rate would reduce S&P 500 EPS by roughly 1-2% [so] if sustained, the tariffs announced this weekend would reduce our S&P 500 EPS forecasts by roughly 2-3%, not taking into account any additional impact from major financial conditions tightening or a larger-than-expected [increase] of policy uncertainty on corporate or consumer behavior,” Kostin went on.

The figure gives you some context for what we’re looking at in terms of GPU impact from the latest tariff uncertainty. The index is now perched at levels comparable to those seen around the onset of the pandemic.

On the valuation front, Kostin reminded investors that there’s a fairly strong historical relationship between that policy uncertainty gauge and the S&P 500 ERP. The most recent spike in the GPU “should reduce the forward 12-month P/E multiple by about 3%” ceteris paribus, Kostin remarked.

Separately, the bank’s Jan Hatzius reiterated that on his math, “a sustained 25% tariff on imports from Canada and Mexico would increase the effective US tariff rate by 7ppt, implying a 0.7% increase in US core PCE prices and a 0.4% hit to GDP.” In a short note released Sunday, Hatzius said he plans to update his estimates “in light of the new tariff details.”