Broadly speaking, the financial services industry has been divided into two competing paradigms since roughly 1950. One, articulated by Harry Markowitz in his famous paper in a 1952 issue of the Journal of Finance, and strongly supplemented by the Brinson research, suggests that financial/investment advisors add value primarily by creating diversified portfolios, preferably optimized along the efficient frontier. A number of studies have shown that (despite the recent unpleasantness in the global investment markets) broad diversification produces a smoother investment ride and greater terminal wealth than concentrated portfolios.
Three years prior to this first articulation of modern portfolio theory, Benjamin Graham published The Intelligent Investor, suggesting that an investment advisor's primary role is to maintain a constant and vigilant evaluation of prices that the market is offering for individual securities against his/her personal definition of 'fair value.'
Generally speaking, the former approach had a scientific look and feel to it, while the latter could be described as art. For more than 50 years, art and science blended about as well as oil and water. The situation that faces the investment world today is not unlike what today's physicists are grappling with as they try to reconcile the equations defining quantum mechanics with relativity: it's clear that the reigning paradigms have served us well. But it's also clear that both are incomplete, and that they belong together in a consolidated formulation.
The questions that advisors will need to answer for themselves is: what will that next paradigm look like, and how can I begin to apply it now?
There is evidence that this is not far from their minds. In a recent poll of the readers of the Inside Information newsletter service, advisors were asked to project the future real returns for the S&P 500 and intermediate bonds, and also the inflation rate, and to explain why they selected the numbers they did and what use they were putting them to in their client-facing activities.
Before the 2008-9 market storm, this audience of thoughtful advisors would overwhelmingly have cited the historical averages. This time was interestingly different: we received more than 400 pages of explanations on different ways to evaluate the opportunity set in the investment marketplace. Reading through the responses, it appears that the New Paradigm, if it is to emerge, will have several characteristics.
First, it will apply Ben Graham-espoused principles, not to individual securities, but to asset classes. More than 100 of the responses talked about how tricky it is to determine the actual, intrinsic value of stocks, bonds and other components of the investment opportunity set. But they seem to be taming the problem. For stocks, many advisors now appear to be coalescing around some form of the trailing 10-year P/E number favored by Robert Shiller, which has fewer problems than a valuation that relies on notoriously unreliable projected earnings. For bonds, advisors said that they start by looking at the current vs. historical yields on Treasuries as a baseline measure, and then evaluate the spreads between different Treasury maturities, and spreads among those and various corporate/municipal paper of different grades and maturities.
Unfortunately, these valuation measurements cannot be made in a vacuum; a low interest rate in a period of low inflation, or a high P/E at a time of low interest rates, are very different things from low interest rates during persistently high inflation or expensive stocks when interest rates are in the double digits. Many of the responses talked about making macroeconomic evaluations, tracking M1 and M2, the economic cycle and various leading indicators, and (currently) things like the prospect of a double-dip recession. You might call this first characteristic of the New Paradigm: evaluating each asset class's price in a broader economic context.
This implies the second characteristic of the New Paradigm: that advisors, in tending client portfolios, will incorporate some kind of active response to the shifting valuations in the context of shifting economic winds. This has variously (depending on the degree to which the observer approves of this activity) been referred to as "market timing" or "dynamic asset allocation."
To see how the first and second element might work together, consider an interesting hybrid proposal that was put forth by David Jacobs, an advisor who practices in Honolulu, HI. Jacobs starts by looking at the current valuation of an asset class--let's say the P/E10 of the S&P 500--and then compares that to the lowest historical valuation of that asset class. For example, if the S&P 500 overall is trading at a P/E10 of 20, and the Black Swan lows from the 1974 oil shock, the Korean War or World War II brought the P/E10 down to 8, then the current S&P 500 allocation's estimated downside risk, purely in terms of the pricing of the security, is a 60% drop (1-8/20). Jacobs then adds in an estimate of yearly earnings growth and dividend payments (current dividend payments plus, perhaps, the average growth of the economy), and calculates a real downside risk that can be applied to any time period.
If clients can't tolerate that much risk, Jacobs proposes to buy Treasury Income Protected Securities (TIPS) with a duration equal to the client's time horizon to offset some portion of his calculated downsize risk, and produce a portfolio that will, under the worst case scenario, provide approximately a nominal loss equal to the client's expressed risk tolerance over a particular time frame.
This is obviously not a static process. Jacobs notes that in October 2007, the P/E10 ratio for the S&P 500 had risen to 27, while the dividend yield was 1.8% and the 20-year TIP was yielding a real 2.26%. Since the equity downside risk was now 70% in equities, each client's individual risk aversion would argue for a reduced exposure to equities and higher allocation to TIPS. In March 2009, when the P/E10 had fallen to 13 and the S&P 500's dividend yield had climbed over 3.5%, the downside risk exposure was dramatically lower, arguing for higher across-the-board allocations to the stock market. One might call this entering the market timing world through a side door.
A third characteristic of the New Paradigm is a search for what might be termed "expanded diversification," reflecting, more colloquially, an intense dissatisfaction with the diversification results provided by stocks (of various flavors and nationalities), bonds, real estate and commodities.
The search appears to be focused on asset classes which are not traded publicly--and, therefore, are not subject to the mass psychology forces which drive correlations close to 1.0 during periods of market panic or euphoria. Larger advisory firms talked about pooling their due diligence efforts to invest in portfolios of real estate properties, producing oil wells and timberland. Indeed, one study group--called 20-20--has made pooled due diligence a centerpiece of its discussions and value to each other. Similarly, advisors appear to have shown greater interest in creating LLC structures which will aggregate client assets, so that many retail clients can participate as a single investor in high-minimum venture capital, oil and gas or producing timber operations. (More recently, this aggregation has become easier and more efficient using unified management account (UMA) platforms.)
Meanwhile, a fourth characteristic of the New Paradigm is modifying the mathematics of MPT itself, to more closely conform to the actual data set of historical returns. More than a few responses to the Inside Information poll noted that the return dispersions of different categories of equities are clearly not fully explained by what is variously called the "normal," "gaussian" or bell curve distribution, defined by a mean and standard deviation. Some advisors now change their optimizer settings to distribution curves that incorporate leptokusis (where the peak is not at the exact center of the curve), Pareto-Levy distributions (where the peak is higher and tails are fatter than the bell curve) or Cauchy-Lorenz curves (potentially infinite tails). Smart Portfolios, Inc. in Seattle draws a customized distribution curve from an equation that is defined by the historical data itself.
The challenge here is to select the appropriate data set. The PIMCO organization has pointed out that since 1916, the U.S. equity markets have experienced 58 trading days where the losses exceeded 17 standard deviations, and six days where the losses would have been 61 standard deviations out on the tail of a traditional bell curve. James notes that the Dow's 7.18% drop on October 27, 1997, under the mathematics of MPT, could reasonably be expected to happen roughly once in 50 billion trading days, or every 130 million years or so. A sequence of three market drops over a seven-day period in July 2002 should happen, at most, once in the entire 13.7 billion duration of the universe.
Yet yearly return data has not produced outliers nearly this dramatic. The 2008-9 bear market produced a decline roughly 2.2 to 2.4 standard deviations from the mean--meaning that it was an event which should, according to the mathematics, happen roughly every 60-120 years.
Finally, the fifth characteristic of the New Paradigm is an effort to constrain returns using a variety of market instruments. The simplest approach is to use variable annuities with various guarantees, or to employ a simple algorithm where the advisor would notice when the client has accumulated a portfolio large enough to sustain a comfortable retirement--and, before the market can take back some of this accumulated wealth, use the assets to buy an immediate annuity. In the survey, one advisor offered a detailed study which suggests that this approach has a higher probability of success than the traditional approach.
A more recent option is the deferred annuity (sometimes called longevity insurance), where the client is encouraged to buy an annuity today with a single premium, and receive monthly payments from the insurance company at some specified age (80? 85? 90?) in the future, thus turning the retirement savings process into a period certain effort.
However, a growing number of advisors are exploring more complex options, involving futures and options contracts traded on the CBOE or other exchanges, or structured products offered by investment banking firms. In either case, the principle is essentially the same: the advisor buys and sells contracts to create a position where, for instance, clients receive two times the return of an index if it rises, up to a ceiling return, and also receive a downside cushion. Thus, if the market were to rise 10% over the next 13 months, clients would receive a 20% return. If the market dropped 10%, they wouldn't experience any decline at all; if the market fell 20%, they would only experience a 10% decline.
This trend appears to have been accelerated by the acceptance of ETFs in client portfolios, since ETFs, being publicly-traded, are more amenable to options and futures strategies than mutual funds.
This obviously is not a crystal clear picture of the New Paradigm, and it leaves out a number of other characteristics, such as rebalancing monthly and tax management, which are also becoming increasingly workable due to improvements in software and built-in tools on the investment platforms themselves. But it leads to some interesting conclusions. First, the creation and management of client portfolios is likely to become far more complicated and time-consuming than it has been in the recent past. If advisors are using tools that rely on mathematical formulations drawn from generalized auto-regressive conditional heteroscedasicity, Cupola dependency models, evaluating downside risk factors and reading 60 or more data sources on a regular basis, this could eventually reduce their important client-facing time.
Another is that the risk of getting this new paradigm wrong will be increased dramatically. Any new set of tools carries the risk of misuse, and the more powerful they are, the greater the likelihood of injury. If a lot of drivers from the 1950s were put behind the wheel of modern Formula 1 race cars, or carpenters accustomed to using a hand saw were handed chain saws, we would not be surprised if there were a statistical rise in injuries.
Finally, the evolution of a New Paradigm may lead to a new bifurcation in the advisory profession. Many advisors might choose a different paradigm altogether: they would outsource the creation of client portfolios to persons who are more adept at or more interested in using these complex tools and models. In the final analysis, the New Paradigm may be adopted broadly, but applied by few.
Bob Veres publishes Inside Information and Media Reviews (www.bobveres.com), a newsletter service for financial planning professionals, and now offers a Client Articles service of custom-written articles that advisors can send to their clients on a variety of topics.
Read more articles by Bob Veres
