Notes, Views, and the Occasional Provocation July / August 2014 

It's fire season in California, now through late October, which happens to coincide with some of the more unpleasant investment months historically. But how much light can history shed on what will happen next in the markets? This is the topic of this month's newsletter.

Mark Twain undoubtedly put it best: "October: This is one of the peculiarly dangerous months to speculate in stocks. The other are July, January, September, April, November, May, March, June, December, August, and February." - Pudd'nhead Wilson's Calendar

Best Regards,

Milo Benningfield




What Historical Snapshots Can't Tell Us
A picture may be worth a thousand words, but it takes far more than that to pin down a complex social system like the stock market. more



What Historical Snapshots Can't Tell Us
The photographer's shop in the village of Point Reyes Station was small, but didn't feel that way because of all the expansive outdoor vistas in the framed black and white photographs on the walls. A large panorama of Point Reyes National Seashore looked down from the top of Mount Vision across dense pine forests and a series of placid, shining estuaries that linked to the sea. In the Pacific to the left were the Farallon Islands; to the right, the faint markings of a squall driving rain toward shore. Magnificent clouds and shadows on ocean ripples completed the picture of creation.

The photographer Marty Knapp stood beside me. "That view doesn't exist any more," he said. He'd taken the photo in 1995 a few months before the devastating Mount Vision fire. In the trees that would have been behind us, four teenagers had failed to extinguish their illegal campfire. The forest had exploded and burned to the ground, along with forty-five houses in the village of Inverness. A huge expanse of the national seashore was reduced to black, smoldering moonscape. "You could call this a historical document," Marty said, as we stared at the peaceful vista.

Much of what passes for stock-market analysis is likewise comprised of historical snapshots that appear to present things as they are, but, at best, show a world that no longer exists. Often, they distort even the historical view.

A few years ago, after the financial crisis, the dominant caption read, "Lost Decade for Stocks." Investor money flowed out of a broad swath of stock funds too quickly to warn people of how closely the picture had been cropped.

Yes, the ten years ending in 2009 had produced a negative one percent return, but primarily for large U.S. growth stocks, as measured by the S&P 500 Index. Anyone holding more diversified allocations that included foreign and other types of U.S. stocks had, depending on the mix, quite positive returns.

Today's captions are reversed and underscore how well the S&P 500 Index is doing compared to foreign stocks. For example, a recent Wall Street Journal article discussed how a portfolio of S&P 500 Index stocks beat a globally diversified stock allocation over the past twenty years. Earlier articles asked whether "the story for emerging-market stocks is over."

Far from being actionable information, these types of return snapshots are so highly dependent on the data's start dates and endpoints as to constitute statistical noise. Extending the timeframe can help, but can itself hide important information.

For example, taking a multi-decade view from 1970 (when the MSCI EAFE Index began) through 2013, we see large foreign and U.S. stocks ending neck to neck: the S&P 500 Index returned 10.4 percent per year versus 10.0 percent for the MSCI EAFE. Who needs foreign stocks, right?

However, the S&P 500 Index eked out its slight advantage in just the last four years from 2010 through 2013 when it returned 15.9 percent per year versus 8.6 for the MSCI EAFE, suggesting that U.S. stocks alone are a riskier bet than more diversified allocations.

If past returns are an easily distorted basis on which to make investment decisions, are there more reliable measures? One of the more popular stock-valuation measures right now is the "cyclically adjusted price-earnings ratio" popularized by professor and Nobel laureate Robert Shiller at Yale University. Also known as the Shiller PE ratio, it seeks to provide information useful for long-term asset allocations by looking at corporate earnings over ten years rather than the traditional one-year look-back and comparing long-term earnings to current market prices.

At a current value of about 26, well above its long-term average of about 16, the ratio suggests that stocks are richly overvalued today, making a lot of investors nervous. As the chart below shows, today's high value was seen only three times during the twentieth century, in the years 1929, 1999, and 2007, just before major market declines.

However, asking the Shiller PE to tell us if the market is about to crash is akin to asking a photograph to tell us whether a forest fire is imminent. Neither device can possibly convey the richness of all the necessary information – e.g., how hot the temperature is or how fast the wind is blowing in the case of the photo; how myriad economic variables combine and interact in the case of the valuation metric.

Critics of the Shiller PE have long contended that too much has changed since the railroad days of Cornelius Vanderbilt for the ratio's long-term average to be much use today. Different corporate accounting practices, different dividend payout practices, and better management of interest rates by the Federal Reserve have all potentially contributed to permanently higher stock valuations. (An admittedly dangerous argument, as Professor Shiller notes, given how it was famously made just before the 1929 market crash.)

Professor Shiller himself recently acknowledged that the ratio has been saying the stock market is overvalued for almost two decades, which begs two questions that he has tried, but been unable to answer:

  • Have stocks really been overpriced over the past two decades compared to the long-term average of the Shiller PE ratio?
  • Or has the ratio itself, and thus market values, drifted permanently higher because of systemic factors?1

He suggests, though cannot prove that the mystery of stock valuations lies somewhere in the murky realm of investor psychology and behavior, which, if true, likely means that no quantitative measure will ever be able to serve as a crystal ball to guide our investment decisions.

Rather, prudent investments will continue to be based more on an assessment of our own financial circumstances and preferences for risk, rather than on guesses about where the markets are headed. Market history and valuation metrics can still be useful tools, but should serve as something akin to the Fire Danger alerts at the entrances of national parks – helpful in terms of whether to take extra precautions given the climate, but nothing more.

After the Mount Vision fire in 1995, it was difficult to imagine the area would ever be as beautiful again. But native plants that had evolved to survive forest fires soon thrived better than before, since invasive non-native competitors had been wiped out. Within a few months, wildflowers were blooming, and within a few years, the pine forests reappeared.

Most of the Inverness residents rebuilt their homes, showing devotion to the land as well as faith that the benefits of living in such a beautiful area would outweigh the costs. Ultimately, stock investors rely on a similar faith in progress and human ingenuity despite – or more accurately, because of – the fires that will inevitably occur.

Thank you for reading. Please look for our next newsletter in October.

Best regards,

Milo Benningfield
back to top




1 "The Mystery of Lofty Stock Market Elevations," The New York Times, August 16, 2014

Copyright 2014 - Benningfield Financial Advisors