The Stumble of Bull Run

There’s a fierce debate going on about whether U.S. stocks have become too expensive after a historic eight-year bull run. I argued recently that the disagreement is actually about corporate earnings.

Consider how the market’s valuation changes depending on assumptions about earnings. Based on the S&P 500’s closing price of 2,439.07 on Friday and analysts’ expected earnings of $145.46 for 2018, the index’s price-to-earnings ratio is a modest 16.8. That’s roughly in line with the market’s average one-year trailing P/E ratio of 15.6 since 1871.

However, if you think that last year’s earnings of $108.54 are a better proxy for future earnings, then the P/E ratio jumps to a more worrisome 22.5. And if you peg earnings using the 10-year trailing average of $90.62, then the P/E ratio jumps even higher to 26.9.

Investors’ level of anxiety about the market, therefore, is likely to depend on their confidence in corporate earnings.

Until recently, that confidence was in short supply. The S&P 500’s year-over-year earnings declined for seven consecutive quarters from the fourth quarter of 2014 through the second quarter of 2016. It seemed as if the great bull run was finally coming to a close. The market’s response was appropriately muted — the index returned an annualized 3.6 percent over that period.

Then the S&P 500’s year-over-year earnings staged a sharp comeback. They grew 1.7 percent in the third quarter of 2016, 8.5 percent in the fourth quarter, and an astonishing 17.7 percent in the first quarter of this year. The bulls were back, and the index responded with a return of 16.1 percent from July 2016 through Monday.

There are some inconvenient details in the earnings numbers, however. As Bloomberg News reported this morning, a growing number of companies in the S&P 500 are losing money. By my count, 3.4 percent of companies in the S&P 500 were in the red in fiscal year 2014. That number swelled to 8.3 percent in 2015. It grew again to 10.1 percent in 2016.

It’s tricky to compile comparable numbers for historical periods, primarily because the composition of the S&P 500 changes over time. Roughly one-half of the index’s constituents either weren’t in the index three decades ago or their earnings numbers for those prior periods aren’t available. But an easy way to gauge the effect of unprofitable companies on the S&P 500’s aggregate earnings is by looking at the ratio of positive earnings — or earnings that exclude those unprofitable companies — to aggregate earnings. Let’s call this the Negative Earnings Ratio.

For example, the S&P 500’s positive earnings were $113.63 in 2016 and aggregate earnings were $108.54, which translates into a Negative Earnings Ratio of 1.05. The higher the ratio, the larger the share of negative earnings in aggregate earnings. That ratio was 1.01 in 2014 and 1.03 in 2015, which confirms that negative earnings are on the rise.

The reason investors ought to pay attention is that a rise in negative earnings has historically been associated with bigger problems. The correlation between the Negative Earnings Ratio and real GDP growth has been a negative 0.5 since 1990 — the earliest year for which earnings numbers are available. Similarly, the ratio’s correlation with earnings growth has been a negative 0.6.

That makes intuitive sense. When the economy stumbles, more companies struggle to make money, which eventually drags down aggregate earnings. And as the market has reminded investors in recent years, as earnings go so go stock prices.

The percentage of unprofitable companies in the S&P 500 hasn’t reached the levels seen during the previous two downturns. That number climbed to 19.4 percent in 2002 and 17.1 percent in 2009.

Still, investors shouldn’t be lulled by recent earnings growth. The devil is in the negative earnings.

Bloomberg Gadfly,