In a new report, Goldman Sach’s U.S. equities chief David Kostin warned that the S&P looks alarmingly rich. Kostin writes that “equity valuations are extremely elevated on an absolute basis,” that prices appear stretched based on virtually every standard metric, metric, including price to earnings, price to book and market cap to GDP. Then Kostin notes that looks can be deceiving, and argues that U.S. large caps are still a good deal. He predicts the S&P will jump 11.5% from the all-time high of 3855 posted on January 25 to 4300 by year end.
Kostin’s rationale for reckoning that investors shouldn’t fret that all the classic yardsticks are flashing red is the familiar low-rates are the magic that will keep equities climbing argument. “After taking interest rates into account, the aggregate index actually trades below the historical valuation,” says Kostin. Many optimists take this position based on a new measure developed by the great Yale economist Robert Shiller that compares the earnings yield, the inverse of the P/E showing the cents in profits investors get for every dollar they’re investing in the S&P index, versus the inflation-adjusted return on the 10-year Treasury. When the earnings yield exceeds that real Treasury yield by a good margin, stocks are a much better buy than bonds, the theory goes.
Right now, what Shiller calls his Excess CAPE Return––the CAPE is his adjusted P/E––shows that his earnings yield of 2.83%, which looks lousy, is still 3.4% higher than the “real” yield on bonds. Hence, the Excess CAPE Return is 3.4 points. How is that possible? Because inflation running at around 1.68% exceeds the long bond rate of 1.12%. In other words, we’re experiencing an interlude when Treasuries don’t even keep you even with inflation. These are extraordinary times of “negative real rates.” A negative real rate plus a puny earnings yield that would almost always be alarming produces a pretty good Excess CAPE. But extraordinary is a problem.
Kostin’s position is similar to that of believers convinced that Shiller’s Excess CAPE beats all the traditional benchmarks, and stands alone in proving that stocks are really cheap. One problem is that the Excess CAPE isn’t nearly as favorable to equities as a few months ago, when it reached 4.88, because stock have been soaring, and bond yields rising rapidly from ultra-low levels, putting a squeeze on the bull’s prized metric. Another is that this measure doesn’t tell you much about how shares will perform in the future. The earnings yield almost always exceeds the long bond yield, simply because stocks are a lot riskier than bonds. The only exceptions are brief periods in 1998 and 2009.
Here’s what should give investors pause: the reasons behind why stocks look so much better than bonds are so unusual that they’re unlikely to last. Kostin goes on to say that we’ll see “modestly higher rates” by the close of 2021, but that a surge in earnings will more than compensate and push the S&P forward by double-digits. He’s getting at the problem with the bull case. If the economy rebounds and earnings jump, it’s logical to predict that “real” rates will rise as well. As animal spirits return, companies compete harder and harder for capital to chase promising investments, driving yields well above the rate of inflation. A world where the GDP is waxing at 5%, and businesses are thriving has never existed alongside a real interest rates at zero or below––that’s the kind of number you get in a recession!
Say the S&P reaches Kostin’s 4300 mark at the end of the year. The CBO predicts that GDP for 2021 will will be flat with 2019, and 1 point below in “real terms.” The all-time record for earning-per-share, based on trailing four quarter GAAP profits, was $139.47 in Q4 2019. If the national output’s the same size in 11 months as it was then, then at 4300 the S&P’s P/E would expand to 31 times earnings, 13% higher than the 23.2 reading at year-end 2019. That would be by far the highest number since earnings collapsed in the fall of 2009.
What about Shiller’s regular CAPE that adjusts earnings for the big swings they regularly stage by using a ten-year average of inflation-adjusted profits? The CAPE is effectively forecasting earnings per share next year at way below that $139 record. At 4300, the old-fashioned CAPE would register around 40. It hasn’t breached that level since the late stages of the 2000 bubble, and never even got close in before the crash in 1929, and collapse of 2009. Believe the yardsticks Kostin cite that point to danger and ignore the “world has changed scenario.” You never know when it will return to normal, but it always does.
