The fundamentals are sound!
How many times have I said that recently amid a US equity rally which, built on towering profits or not, feels dangerously detached from nebulous notions of “reality”?
A lot. I’ve said it a lot. And for good reason. Unlike many of the other factors we allude to when we claim soaring stocks are “oblivious,” profits are the “reality” that matters most, followed by the discount rate, the cost of capital and all the other boring inputs into the B-school valuation models most of us forget the minute we’re no longer compelled to remember them.
Setting aside clever quips you and I can both make about the irony of calling non-GAAP accounting “reality,” and similar jokes about the inherent peril in treating EPS projections as though they’re realized earnings (i.e., treating forecasts as something other than guesstimates), profitability has at least as much claim on being stocks’ “reality” as geostrategy, sociopolitical friction, sundry gauges of “speculative froth” and all the other things we habitually cite while deriding equities’ for their purported careless indifference.
With that in mind, have a look at the figure on the left, below, from Goldman, which shows you the trailing four-quarter ROE for the index hitting a record 22% in Q1 of this year, one of the strongest quarters for profit growth in decades excluding 2021.
Over the past four decades, index ROE expanded by 800bps. As Goldman’s Ben Snider remarked, 150bps of that came during the past four quarters alone.
Not surprisingly, ROE maps well onto multiples. If you plot the ROE series with the CAPE ratio, there’s a pretty tight fit from 1990 onward, as shown on the right.
“Record profitability has been one of the factors supporting high valuations,” Snider wrote, adding that while multiples are obviously high versus history, the index P/E’s “roughly in line” with model-implied fair value.
So what’s the problem? Well, the figures below give you a hint.
Mega-cap tech’s doing the heavy lifting. Indeed, ROE for the median stock’s drifted steadily lower, and there’s now a yawning disparity between the index and the median constituent.
“ROE for the aggregate S&P 500 has benefited from the growing weight of mega-caps,” Snider wrote. That’s a potential issue given concerns around hyper-scaler capex and doubts about the sustainability of the associated fillip for semi profitability.
“While ROE for the mega-cap tech stocks is extremely elevated, the AI capex boom is likely to weigh in coming years,” Snider went on, adding that although insatiable demand for chips has “pushed semiconductor profitability to extremely high levels, investors remain skeptical about the durability of those earnings.”
According to company analysts, ROE will decline in 2027 from 2026 for all of the stocks Goldman counted as “mega-cap tech” with the exception of Broadcom, where profitability should be flat YoY.
