“A recession is not inevitable,” Goldman’s David Kostin reiterated, in a new note. “But clients constantly ask what to expect from equities in the event of a recession.”
If Goldman’s clients are indeed “constantly” fearful of a downturn, it probably doesn’t help that both David Solomon and his predecessor made “recession” headlines this month.
“There’s a chance of recession,” Solomon told Bloomberg on Wednesday, in what sounded like a totally superfluous phone interview. He cited the bank’s econ team, which puts the odds at around 30% over the next 12 to 24 months.
For his part, Lloyd Blankfein told CBS’s “Face the Nation” that if he “were running a big company” (you know, a big company like Goldman), he’d “be very prepared” for a downturn. Consumers should be prepared for it too, he added, calling the path to a soft landing “narrow.”
Kostin walked through a hodgepodge of familiar statistics on stock performance around recessions. Across a dozen downturns since World War II, the median S&P drawdown is 24%. The average is 30%.
As the simple figure (above) shows, things can be very bad, not so bad or somewhere in-between, and the malaise can last for 30 days or 30 months.
Seen in that context, I wonder why we bother. With a sample size that small, and results that varied, statistical inference is a fool’s errand.
The two visuals (below, from Goldman) are perhaps more compelling.
In short: Dividend futures are pricing a recession. The figure on the left suggests dividends will fall next year. As Kostin wrote, referencing the figure on the right, over the past six decades “trailing four-quarter S&P 500 dividends have never fallen on a YoY basis outside of a recession.”
Ultimately, Goldman’s base case isn’t a US downturn, even if the bank is surely “very prepared” for such a scenario, as Blankfein put it, on national television last weekend.
In the event the S&P were to match the median decline shown in the first figure (above), the downside from here would be about 7%. A drop to the average post-War drawdown would find US equities falling an additional 14%.
I’m probably asking for a false level of precision here, but when say we have a downside of another 11% to reach the median 24% drawdown, it seems you are using the March high of SPX 4600, and not the year end high of about 4800. Is my math wrong, or if I’m right why not use the actual high?
It’s the 4800 figure, and the drawdown would equate to SPX 3650, but as a general rule of thumb, I use the figures from the notes I’m citing, which themselves are usually based on the prior day’s close. So, this particular GS note was published on Wednesday, but the benchmarks tumbled on Wednesday, which meant the math was different by the time it was released. I’ve changed the numbers you referenced, but I’d note that usually, it’s not as straightforward as it is in this case. Why? Well, imagine this were a 3,000-word article with eight charts, some of which referenced the “old” math. My editorializing wouldn’t match the visuals in some cases, which could lead to rampant confusion and/or make it virtually impossible for me to get much done because I’d have to constantly rerun the numbers and then try to paraphrase around the “stale” numbers from a given note. In this case, it’s much easier since the passage you’re referencing is just two standalone sentences at the end of a short piece. But generally speaking, trying to stay “current” during these kinds of whipsaw weeks is impossible.
Thanks, makes sense.
Really great graphic
Recent S&P 500 peak was on Jan 4, 2022 at just a little over 4,800. So its trough-ing process (so far) is approaching 5 months…
7% is not outside the realm of possibility for a single day, or a couple of days, given recent volatility. We could hit the trough on a Wednesday and still be up for the week by that Friday’s close.