Risk, Damned Risk, and Alpha, 2.0 (sounds like Lies, Damned Lies, and Statistics)

Over the past few days, Bloomberg columnist John Authers has been hosting an online discussion of Peter Bernstein’s Capital Ideas from 1996.  Much of the back and forth has been about using volatility (standard deviation of mostly shareprice-based returns) as the measure of risk in investing. That’s the preferred approach in Modern Portfolio Theory.  In recent years, I have made a cottage industry of bashing MPT’s definition of risk. I’ve also likened alpha (the excess or deficit of return after adjustment for risk or specific factor exposure) to Santa Claus: He may or may not exist, but it makes a lot of people happy to think that he does. And I have railed against the institutions of the modern investment industry that promote these outdated and non-business-based investment paradigms in graduate curricula and industry certification programs.

Yet, I find it tremendously ironic that for all the infrastructure of using risk-adjusted returns and seeking alpha, when push comes to shove, right now, in 2019, few investors seem to really care about either. And the indifference does not appear to be about the particular definition of risk; it seems to be about risk in any form. In my conversations with many financial advisors and even sophisticated institutions, the current goal seems to be beat the total return of the market or an index, preferably in every measurement period. Period. Risk rarely comes into the equation, or the discussion.   How much progress, one might ask in a purely rhetorical sense, have we made since 1952, when Harry Markowitz tried to bring some theoretical order to the chaos of stock market investing?

Getting Back to Business is about how the infrastructure of MPT came into being.  That infrastructure remains firmly in place—the canonical answers about how to measure and manage portfolios—but it is fair to ask how many people are following the “rules” right now in 2019.  To the extent that investors are indifferent to risk, there may be one glaringly obvious reason why: we are now ten full years into a market rally. Who cares about risk if the market keeps rising?  Yes, it paused a bit in 2011, 2015, and again in December 2018, but the 10-year chart is pretty much straight up. Annualized market returns from April 1, 2019 through March 31, 2019 are an eye-popping 15.93% for the S&P 500 Index.  And this is almost entirely share-priced gains as the market has had only a 2% or so yield throughout that period.  On March 31, 2009, the S&P 500 Index closed at 798. Ten years later, it was at 2834, for a compound annual growth rate of 13.5%. It is now hovering another 3% higher as I write, around 2900.

With that type of gross performance, perhaps it is understandable that many investors would be indifferent to risk-adjusted returns or even thinking about risk.  This has happened before.  MPT was developed in the 1950s and made somewhat practical in the mid 1960s, but got little traction in the marketplace at that time.  Why? There was no need. The markets were roaring.  The Nifty Fifty were doing just fine. It took a major market takedown in the mid 1970s for investors to turn to the risk-adjusted measurements offered by MPT.  That lasted for a while, and then the 1990s happened. You were there; you remember. No one really thought about risk.  You thought about your favorite, really cool internet or e-commerce stock.  (Mine was ASDV, Aspect Development Corp. Its meteoric provided the downpayment for my first house, for which I am very grateful. It was purchased by i2 Technologies in early 2000, at the top of the market. JDA Software bought i2 Technologies for pennies on the dollar almost a decade later in 2009. JDA succumbed to PE backed RedPrairie two years later in 2012.)

And now we are ten years into another rally. For many investors, valuations are fine; earnings are growing; the economy continues to move forward. The New Economy companies seem to be going from strength to strength. Others see warning signs. But the fact remains that ten years on, viewing the market in terms of MPT’s risk factors might seem less pressing than at other times.

That creates a second irony: In the midst of this decade-long rally, conservative, dividend-focused strategies can have very attractive long-term MPT statistics. They do so not by beating the market in absolute terms. Instead, the combination of low volatility and reasonable participation in the market’s upside generates good MPT numbers, including alpha.

You can look up the fund version of your favorite dividend manager on Morningstar.com.  The website includes MPT stats under the Ratings and Risk tab. There you will find 3-, 5-, 10-, 15-year numbers versus what Morningstar considers a best fit index, as well as the broader market index.  Compare and contrast.  The three-year view may be less positive due to a drop in market volatility in that period. It should get better with the five-year mark, and should be generally quite good at the ten-year measurement, capturing the market’s bottom in early 2009. For SMA products, you would need to check with your distribution platform. They calculate the same information and should make it available to clients.

I find the MPT verdict on conservative dividend strategies ten years into a market rally simply stunning. According to the most orthodox MPT theory, dividends should not even exist. (This is a topic taken up at length in Getting Back to Business.) And yet they actually “beat” the market in MPT terms after a decade-long rally. The tortoise beats the hare, yet again.

What about other definitions of risk?  Taleb’s extreme tail risk seems equally out of fashion at present, for the same reason that standard volatility is. Private businesses lean toward seeing risk in terms of cashflows generated by their business.  I lean in the same direction and argue in favor of a portfolio theory that incorporates cashflows rather than ever-changing asset prices.   But with the market going up and up, only those investors directly relying on dividends to meet their monthly expenses will likely define risk in those terms, at least for now. Were the rally to end, I suspect investors would be more open to measuring performance in terms of risks taken, but also to alternative definitions of risk.  In the meantime, I will wait, collect my dividends (and my alpha), and wonder whether it is not too late to look for the next Aspect Development Corp.

Please note: Nothing here should be construed as investment advice.

risk damned risk and alpha 4-18