In recent weeks, as the S&P 500 has climbed to yet another record high, I have been getting more requests to discuss valuation.
Chapter 14 of my book, “Predicting the Markets” (2018), examines the various models that are widely followed by investors to gauge valuation in equities.
I concluded in the book that valuation, like beauty, is in the eye of the beholder. In other words, it is more art than science. It is more subjective than objective. Nevertheless, I continue to monitor all the major valuation models, especially those that include the influence of inflation and interest rates. History shows that when both are high, valuation multiples tend to be low, and vice versa.
The conundrum is that in the current environment, we have historically low inflation and interest rates, but they are attributable to subpar domestic and global economic growth. Should investors be willing to pay high price-to-earnings (P/E) ratios when inflation and interest rates are low, but growth is weak? That doesn’t seem like a good deal. Then again, the combination of low inflation and interest rates with slow growth may result in a longer-than-usual economic expansion. The current one became the longest one on record just this month. If it keeps going, even at a slow pace, maybe it makes sense to pay relatively high P/Es.
On balance, I believe that the current mix of inflation and interest rates and economic growth merits relatively high, above-average valuations, especially for stocks of companies that can generate consistently above-average earnings growth. Let’s have a look at some of the relevant valuation models:
1. Misery index. The simplest model incorporating inflation is the one showing the strong inverse correlation between the S&P 500 forward P/E and the misery index, which is the sum of the unemployment rate and the CPI inflation rate (figure 1). The misery index was very low at just 5.3% during June, justifying the month’s forward P/E of 16.8, which is above the historical average reading of roughly 15.
2. Fed’s stock valuation model. There was a relatively good fit between the S&P 500 forward earnings yield and the 10-year U.S. Treasury bond yield TMUBMUSD30Y, -1.18% from 1979 to 2001 (figure 2). Since then, they have diverged, suggesting either that stocks are a screaming buy relative to bonds or that bonds are grossly overvalued. At 2%, the bond yield matches the dividend yield of the S&P 500 SPX, +0.00% making dividend-yielding stocks with growing dividends very attractive, in our opinion (figure 3).
Read: These dividend stocks haven’t been scooped up by investors yet
3. Real earnings yield model. A valuation model that explicitly incorporates inflation is the one tracking the spread between the S&P 500 reported earnings yield and the CPI inflation rate (figure 4). Since 1952, the quarterly spread has averaged 3.3%. Bear markets were preceded by declines in the real earnings yield toward zero. During the first quarter, it rose back to the average, registering 3.5%.
4. Buffett ratio. If you are looking for overvaluation, you’ll find it in the so-called Buffett ratio — named for investor Warren Buffett — which is the market capitalization of all U.S. equities (excluding foreign issues) divided by nominal GNP (figure 5). In the first quarter, it stood at 1.85, only a bit below its record high at the end of the bull market of the 1990s.
However, even Warren Buffett has cautioned that this valuation metric doesn’t reflect that both inflation and interest rates are at record lows. So he is ignoring it. In a May 6 CNBC interview, Buffett said stocks are a huge bargain if interest rates remain at their low levels. “I think stocks are ridiculously cheap if you believe ... that 3% on the 30-year bonds makes sense,” Buffett said.
Ed Yardeni is president of Yardeni Research Inc., a provider of global-investment-strategy and asset-allocation analyses and recommendations. He is the author of “Predicting the Markets: A Professional Autobiography.” Follow him on LinkedIn, Twitter and his blog. Institutional investors may sign-up for a free four-week trial to his research service.