Earnings Signal Remains Bullish for Stocks

What is the right multiple to pay for S&P 500 earnings? Historically since 1935, the average P/E, using four-quarter-trailing reported earnings, was 17.1. It was 23.7 during Q1-2018. That’s a high reading. However, since 1990—a period of relatively low inflation and interest rates—the average P/E was 25.5.

I am not a fan of P/Es based on trailing earnings, as I explain in my new book, Predicting the Markets. I prefer to focus on the forward P/E of the S&P 500, which is based on the time-weighted average of analysts’ consensus expectations for operating earnings during the current year and coming year. It has averaged 13.9 since the start of the monthly data series during September 1978. It recently rose to a 16-year high of 18.5 during January and fell to 16.5 during May.

The forward P/E based on estimated earnings has been lower than the multiple based on four-quarter-trailing earnings because analysts are looking forward, not backward. There’s one major flaw with the forward valuation approach: Analysts’ earnings estimates tend to be unrealistically inflated in advance of recessions because analysts collectively never see recessions coming until it is too late. Once they do, they slash their earnings estimates at the same time as reported earnings take a dive, and share prices take an unanticipated hit. Conversely, the flaw with P/Es based on trailing earnings, which produce the higher multiples of the two approaches, is that they tend to turn bearish much too early in a bull market. The higher the multiple, the more likely it is to be deemed unduly heady as a bull market continues.

Since I don’t see a recession in the foreseeable future, I continue to focus on the forward P/E, which isn’t alarmingly high, in my opinion. The further out that a recession is perceived as likely to happen, the more sustainable are above-average P/Es. That’s because long expansions give investors the time to see earnings grow, as predicted by industry analysts.

I currently don’t expect a recession over the rest of this year or in 2019. What about 2020? Ask me again in 2019. As long as inflation remains subdued, as I expect, odds are that the expansion will go on and on—until further notice.

My Blue Angels analysis compares the S&P 500 stock price index to its implied value using weekly forward earnings multiplied by forward P/Es of 10.0 to 19.0 in increments of 1.0. Just for fun, let’s compare the index to its implied value using a multiple of 15.0, which seems to be widely viewed as a fair-value multiple for the S&P 500 both by forward-looking and backward-looking investment strategists. The monthly version of this analysis starts in September 1978, while the weekly version is available going back to March 1994.

As long as the economy is growing, I like to think of the implied S&P 500, derived by multiplying forward earnings by 15.0, as the underlying signal that determines the direction of the stock market. The actual S&P 500 is driven by the signal and buffeted around that signal by “noise.”

Currently, the signal is very strong thanks to Trump’s tax cuts. Also contributing to the strong signal is solid global economic growth, which has been bullish for the earnings of commodity companies, especially in the energy sector. The noise recently has been mostly about Trump’s protectionist threats, which already seem to be dissipating. There is also some noise about a pickup in inflation, which remains subdued. Fed officials have provided a relatively steady signal about their intention to normalize monetary policy in a gradual fashion, as confirmed by the latest minutes of the FOMC. (Chapter 13 in my book is all about predicting corporate earnings. Chapter 14 is all about predicting valuation.)

Widely Followed Strategist and Economist Ed Yardeni Shares His Unconvent...

What Are They Smoking?

I would like to try some of whatever industry analysts are smoking. You can compare my earnings forecasts to their consensus estimates on a weekly basis in YRI S&P 500 Earnings Forecast on our website. I say “tomato.” They say “tomahto.”

My earnings-per-share estimate for 2018 is $155.00 (up 17.4% y/y). The analysts continue to up the ante and are currently at $160.40 (up 21.5%). My estimate for 2019 is $166.00 (up 7.1%). Theirs is $175.72 (up 9.6%). Perhaps the analysts are just high on life.

They could be right about 2018, especially since they just boosted their Q1-2018 sights based on results reported so far during the earnings season. At the same time, they have been conservative on the remaining quarters of this year. Their growth estimate for next year seems too high to me since I expect 2019 earnings growth to settle back down to the historical trend of 7%.

You can drive a truck between my earnings estimates and theirs. However, both suggest that the stock market is likely to be at new record highs by the end of this year. As the year progresses, the earnings estimate for this year will be less relevant, while the estimate for 2019 will be more so. By the end of the year, the market will be discounting analysts’ consensus earnings estimate for 2019, not my estimate.

However, analysts tend to be too optimistic and often lower their estimates as earnings seasons approach. So let’s split the difference between my estimate and their current estimate for 2019. That would put consensus earnings for 2019 at roughly $170 per share by the end of this year. Now let’s apply forward P/Es of 14, 16, 18, and 20 to estimate where the S&P 500 might be at year-end:

2380 (down 12.3% from Friday’s close)
2720 (down 0.3%)
3060 (up 12.8%)
3400 (up 25.3%)

I pick the third scenario as most likely. My year-end target for the S&P 500 remains 3100. In my scenario, inflation remains subdued around 2.0% for the foreseeable future. Real GDP grows between 2.5%-3.0% this year. The Fed raises the federal funds rate to 2.25%-2.50% by the end of the year. The 10-year Treasury bond yield trades between 3.00% and 3.50% over the rest of the year. Investors conclude that interest rates aren’t likely to move much higher in 2019 and increasingly believe that the economic expansion might last beyond July 2019, when it will be the longest one on record. Also fears of a trade war are likely to subside as the year progresses. (For more on “Predicting Corporate Earnings,” see Chapter 13 of my new book, Predicting the Markets: A Professional Autobiography.)

S&P 500 Real Earnings Yield Says Market Isn’t Too Pricey

The May 7 issue of Barron’s included an interview with my good friend John Apruzzese, the chief investment officer of Evercore Wealth Management. In my new book, Predicting the Markets, Chapter 14 is titled “Predicting Valuation.” I explore various models for assessing whether the stock market is undervalued, fairly valued, or overvalued. I discuss a model that John and I both favor as follows:

“The earnings yield of the S&P 500, which is simply the reciprocal of the P/E based on reported earnings, is highly correlated with the CPI inflation rate on a year-over-year basis. The real earnings yield (REY) of the S&P 500 is the difference between the nominal yield and the inflation rate. The result is a mean-reversion valuation model that logically includes inflation. The average of the real yield since 1952 is 3.3%. The model tends to anticipate bear markets when the yield falls close to zero. John Apruzzese ... examined this model in a November 2017 paper titled A Reality Check for Stock Valuations. Based on the REY model, he found that ‘stocks appear more reasonably priced than the conventional P/E ratio suggests during periods of low inflation and rising markets, and more expensive during periods of high inflation and falling markets when they otherwise might seem cheap.’”

As I observe in my book, there are lots of valuation models. None are infallible. None are right all the time. I like the REY model because it does reflect the impact of inflation on valuation. As John observes, “Inflation is absolutely crucial for long-term investors. It’s the most important macro factor. Oddly, the market is stuck on the P/E ratio.” I agree and devote Chapter 4 of my book to “Predicting Inflation.”

The average value of the REY since 1952 has been 3.3%. Presumably, the market is fairly valued around this level, undervalued above it, and overvalued below it. The REY was 2.6% during the first quarter of this year. Since the late 1960s, it worked relatively well as a bear market leading indicator when it fell closer to zero. It also turned out to be a relatively good bull market indicator when it rose back above zero and exceeded its historical average.

In his paper, John concludes, “As of September 30, REY is 3.0%, near its 60-year average, based on a nominal earnings yield of 4.7% and a 1.7% core inflation rate. That indicates that the stock market is fairly valued. By no means does it look like one of the most overvalued markets in history, as measures such as the trailing P/E ratio and the Shiller P/E ratio suggest.” I agree.

High Noise-to-Signal Ratio In Stock Market

The 10-year US Treasury bond yield edged up above 3.00% yesterday. This widely feared level didn’t faze the S&P 500, which rose 1.0% on Wednesday to 2697.79. Rising oil prices, in response to Trump’s no-deal with Iran, helped to lift the S&P 500 Energy sector. But the Tech and Financial sectors also had a good day yesterday.

While the 10-year yield suggests that credit conditions are tightening, the yield spread between US high-yield corporate bonds and the 10-year Treasury bond continues to fluctuate in a tight range at a level that matches previous cyclical lows. All of the volatility in the S&P 500, and the downside pressure on this index, so far this year have been attributable to the forward P/E of the index, which has dropped from a high of 18.6 on January 23 to a low of 15.9 on May 3.

The forward P/E has been very noisy so far this year on fears of higher inflation, tighter Fed policy, and trade protectionism. It has masked the underlying bullish trend of the S&P 500 forward earnings. Thanks to Trump’s tax cuts at the end of last year, this weekly measure of industry analysts’ consensus estimates for earnings over the coming 52 weeks has been soaring since the start of the year.

My Boom-Bust Barometer (BBB), which is the ratio of the CRB raw industrials spot price index to initial unemployment claims, continues to be highly correlated with S&P 500 forward earnings. The BBB rose to a record high at the end of April. The BBB is also highly correlated with the S&P 500 stock price index, but less so than with forward earnings. That’s because there is more noise than signal in the S&P 500 than in forward earnings as a result of the short-term volatility of the P/E.

I have managed to tune out most of the short-term noise and focus on the underlying signal coming from earnings since the beginning of the current bull market. I hope I will continue to do so. In my opinion, the signal remains clearly bullish for now despite the noise.

The Dividend Yield Scare

Contributing to the stock market’s agita so far this year has been the prospect that the 10-year US Treasury bond yield may be on the verge of rising above 3.00%, a level that for some reason is perceived as particularly dangerous for stocks. I suppose that’s mostly because a few widely respected market gurus have been warning that the risks of a bear market in stocks increase above this totally subjective threshold level. Perversely, at the same time, there has been some consternation over the fact that the yield curve has been flattening. That implies that the bond yield hasn’t increased fast enough relative to the federal funds rate and relative to the two-year Treasury note yield! So what are we supposed to be rooting for?

Complicating matters some more are that as the Fed has hiked the federal funds rate, short-term Treasury bill and note yields have risen to match or even exceed the S&P 500’s dividend yield. A few market commentators deem this development as bear-market provoking. So we have nothing to fear but that interest rates will continue to rise above the dividend yield and that short-term rates will rise faster than long-term rates. Consider the following:

(1) S&P 500 yields. I like to look at the S&P 500 dividend yield along with the S&P 500 earnings yield. The latter is derived from the former. On average over time, half of earnings tends to be paid out as dividends. During Q1-2018, the dividend yield was 1.89%, while the earnings yield during Q4-2017 was 4.78% (for the latter, Q1 data aren’t yet available).

(2) Treasury bill rates. The one-year US Treasury note yield rose to 2.06% during March and above the S&P 500 dividend yield for the first time since June 2008. I am hard pressed to see a predictable pattern showing that the spread between the one-year and the dividend yield can be useful in calling bear markets. They tend to occur when interest rates rise high enough to cause a recession. Simply crossing above the dividend yield isn’t a sure-fire signal of an impending recession and bear market.

(3) Treasury bond yields. Comparing the 10-year Treasury bond yield to the dividend yield makes even less sense as a bear market indicator. This Treasury yield is usually compared to the earnings yield, since the total return of stocks tends to be driven by overall earnings. Whether it makes sense for investors to compare just the dividend yield to the bond yield is a debatable issue.