A.I. may be “for real,” but it’s not going to stop a “deep” earnings downturn in the US.
That’s according to Morgan Stanley’s Mike Wilson, who dropped a well-timed tactical long in US equities around these levels back in December. Wilson, a bear, turned tactically bullish in October, a good call, but took profits towards the end of last year.
So far in 2023, Wilson’s apprehension towards equities is enduring, his skepticism unbowed. In his latest, he juxtaposed the conditions which persisted in December with the situation as it exists today.
Back then, cyclicals were leading, tech lagging and breadth was respectable. Markets were hopeful that the abandonment of “COVID zero” in China presaged a meaningful economic rebound with the potential to offset any slowdown in the West.
Fast forward five months and the differences could scarcely be any more stark. Tech is leading a rally which, on some measures, has never been more narrow and “recovery” isn’t the right word for whatever’s going on in China’s economy.
And yet, “investors are more bullish than in early December, or at least far less bearish,” Wilson wrote, citing “optimism around technology diffusion, specifically artificial intelligence.” Although I think it’s a bit of a stretch to attribute the entirety+ of this year’s gains for big-cap tech to A.I., the ChatGPT frenzy is certainly a contributing factor. At the least, it’s already a “mini bubble,” as BofA put it.
Wilson went on to suggest that recent price action is indicative of “panic buying.” On that point, he’s unequivocally right. I documented the same last week in “Stock Melt-Up Takes A Toll”+.
The seasoned veterans among you will remember last summer’s rally, which was snuffed out by Jerome Powell at Jackson Hole. Wilson worries the current grind higher could meet a similar fate.
“Valuations are not attractive, and it’s not just the top 10-20 stocks that are expensive,” he said, noting (again) that the median S&P stock’s forward multiple is above 18, in the top 15% on a three-decade lookback. Excluding tech, the S&P’s median P/E is likewise 18.
On the earnings front, Wilson is sticking by the bank’s models, even after a reporting season that came and went with no real land mines. “A very healthy reacceleration is baked into H2 consensus earnings estimates [which] flies directly in the face of our forecasts, which continue to point materially lower,” he warned, describing the model projections as “much more dire” than the collective wisdom of company analysts.
Say what you will, but the bank’s model has proven itself in the past, and it was “quite prescient” last year too, Wilson was keen to point out, on the way to suggesting consensus may be off as much as 20%.
Suffice to say Wilson isn’t convinced the market’s grenade juggling act (as I described it last week) will end safely. “While many individual stocks and sectors have traded poorly this year, the major indices are priced for simultaneous good outcomes on multiple fronts where we think risks are elevated and even increasing in several instances,” he cautioned, before suggesting liquidity risk+ from a debt ceiling resolution may ultimately pull the rug out.
“We suspect passing the debt ceiling may ironically be the catalyst that ends this bear market rally as it leads to a contraction in liquidity,” he said.
