Rudi Dornbusch, the late MIT economist, coined what is today known as “Dornbusch’s Law:” Crises take longer to arrive than you can possibly imagine, but when they do come, they happen faster than you can possibly imagine.” It may well apply to what could happen to portfolios today. Recent headlines evoke the specter of the 1930s: an incipient trade war between the US, Europe and China, the scapegoating of minorities, families separated and interned, shifting alliances and uncertainty about the role of the United States in the world. To be sure, there are also major differences. The global superpowers are not currently looking to upset the prevailing world order through war, for instance.
Still, the seeds of global discontent are being sown. The G7 summit in Canada earlier this month started badly and ended in acrimony. Following this week’s EU summit, Mr. Trump will arrive in Europe for the annual NATO summit, a body he disdains. As a slap in the face to our NATO allies, the White House has confirmed he will also be meeting separately with Vladimir Putin, whose antipathy for NATO is even more vituperative. Moreover, Pres. Trump seems intent on escalating a trade war in the mold of the Smoot-Hawley Act by imposing steep tariffs on America’s trading partners. That law was passed in 1930.
This past weekend, Edward Luce, the Financial Times chief US commentator based in Washington, summarized these dynamics. “Donald Trump and the 1930s playbook: liberal democracy comes unstuck.” Luce is known for telling us what we don’t want to hear.
Are Geopolitical Developments Determinative Of Portfolio Management Outcomes?
For investors, the answer to the bolded question above is a resounding “no.” Disturbing developments are not, in the short- and intermediate-term, determinative of portfolio outcomes. An analog is, perhaps, climate change.
While it is clearly an existential threat, the timing, magnitude, and resulting consequences are not immediately known and provide only limited actionable information. In practice, geopolitical headlines often pull our eye off the ball—real market risk. Take a cursory look at the above chart. Notice the movement of the upward sloping white curve, the S&P 500, since Donald Trump assumed the Oval Office in January 2017. Given the geopolitical gyrations since, the ever-rising market certainly appears impervious to what is happening around the world. In fact, harkening back to our last post, Black-Scholes, Volatility, & Risky Tales, implied volatility of a one-year OTM put contract (red curve), the “unprovable assumption about the future,” remains quite low. Since 2013 the trading of that contract has continued to signal “all’s clear.” The shorter-term VIX fear gauge has been trending lower overall for the last five years. The 2018 spike was a symptom of reckless speculating within a narrow sector of the market and had little or nothing to do with escalating geopolitical tensions. That begs the next question:
If Not Geopolitical Developments, What Forces Are Determinative Of Portfolio Outcomes?
The answer is the ever-fluctuating price investors are willing to pay for estimated future per-share corporate cash flows. Inferring from the rising S&P 500 index, investors are bullish on those growth prospects. Whether it’s the Tax Cuts and Jobs Act of 2017, in which corporate taxes were slashed from 35% to 21%, the dismantling of business and finance regulations, or the presumptively growth-friendly trade tariffs newly imposed, pragmatists in the investment community seem to be embracing Trump’s supposedly pro-business agenda.
It’s our contention that investors are overpaying, perhaps in the extreme, for earnings prospects that are anything but certain. Historically, much of what we observe today tends to occur in the terminal phases of an expansionary cycle. The current uptrend in corporate profits, given the comparatively anemic growth in GDP, was no more widely anticipated than the utter collapse of 2008-2009. Reported earnings are highly variability in the short-term. That’s precisely why Robert Shiller derived the 10-year moving average Cyclically Adjusted P/E Ratio (CAPE). It smooths out the short-term earnings volatility so as to capture the overarching trend in valuations.
The practical goal is to improve investment decision-making by using more consistent valuation techniques. This avoids being lulled into complacency when earnings are on an eventually unsustainable tear or, equally important, to avoid being paralyzed by fear when reported earnings are freefalling.
Using logarithmic scale, the above analysis compares real reported annual earnings with a 10-year moving average of them. Two forces are anecdotally apparent. First, as you examine this comparison going back to 1880 it seems that reported earnings are, in their own unpredictable but nonetheless inevitable way, mean reverting. Reported earnings are the tail on the economic dog. If the dog’s inflation-adjusted pace through time was 1.78%, the constant tail wagging, whether extreme or moderate, had little effect on the dog’s speed. Secondly, Newton’s third law of motion, “For every action, there is an equal and opposite reaction,” seems to apply. The farther (or perhaps longer) earnings rise above their long-term average, the greater is a likelihood that they will largely compensate for the excesses in the next downturn.
As is obvious, when annual earnings are rising rapidly, the 10-year moving average of earnings will lag. Trailing 12-month (TTM) reported earnings are currently $112, well above the $84.64 in 10-year average earnings. Thus, the CAPE will be higher than the TTM P/E; i.e., 32.9 against 24.9, respectively, based on mid-June S&P index prices. Using either methodology, the average P/E for the S&P 500 going back to 1870 is about 16.5.
The TTM P/E was higher than today’s level on only two occasions: in the final four years of the Dot-com bubble and, interestingly, at the low point in the earnings bust during the financial crisis when it skyrocketed off the charts to 116 times in March 2009 (see below). The tortoise-like CAPE was 13.32. The CAPE has proven to be a reliable valuation tool at both market troughs and market peaks.[1]
The first chart in this post includes white dashed horizontal lines that represent the S&P index at 1600 and 1000 respectively. Holding Shiller’s 10-year average real earnings constant at today’s $84.64, those levels would see the S&P trading at a CAPE of 18.9 and 11.8, respectively. At S&P 1600, the CAPE would not yet have mean reverted. Even at 1000, it would still be well above the major historical secular lows of the mid-single digits.
Based on our extensive research, the single most important determinant of dismal if not disastrous long-term total returns from equities is the beginning Shiller CAPE ratio. The CAPE is more important than reported earnings growth and fluctuations in interest rates. While the geopolitical variables mentioned above are important in many respects, they are not especially so in terms of portfolio outcomes.
A Return to Volatility
It is hard to keep the eye fixed on valuation. While investor bullishness has not been distracted by the geopolitical events, premonitions of the 1930s notwithstanding, it has certainly failed to heed the warnings sounded by Shiller’s data. The price of options contracts belies the misplaced confidence.
That is unsurprising, though. Valuation does not figure in to the calculation of option prices. It is volatility that plays the staring role in that equation. As we wrote in our last post, “In his 2008 letter to the shareholders of Berkshire Hathaway, Warren Buffett wrote: ‘I believe the Black–Scholes formula, even though it is the standard for establishing the dollar liability for options, produces strange results when the long-term variety are being valued… The Black–Scholes formula has approached the status of holy writ in finance … If the formula is applied to extended time periods, however, it can produce absurd results.’”
Today the prices of long-term options seem nearly as absurd as the current geopolitical realignment, a phenomenon almost unimaginable a decade ago. For portfolio outcomes, however, it is a much more common signal that should garner our greatest attention—valuation.
[1] Earnings charts reflect Shiller data.