- Jean L.P Brunel
With the unpleasant performance of capital markets in 2008, it should not surprise that sweeping statements should be made. While a few of these make an enormous amount of sense, others tend to reflect an imperfect understanding of history or irrational expectations. It may thus be useful to step back and meditate on three broad lessons that should both color our evaluation of 2008 and be incorporated into our future expectations: 1) fundamental analysis usually works, but not always without delays; 2) traditional investment approaches have limits that might need to be revisited; 3) strategic and tactical portfolio construction may need to become a bit more original.
Each of these, however, comes with a series of questions that point to the need for more thought and research. Though a fair amount of statistical work underpins the observations proposed below, the format of this letter requires one to limit comments to broad conclusions.
FUNDAMENTAL ANALYSIS WORKS
One of the strongly gratifying insights gained in the last year or year and a half is that there is a great deal of value in the analysis of fundamental developments. Though economics has at times rightly been called the dismal science, broad economic principles still usually hold, even in dramatic circumstances. In fact, the consistency and predictability of economic developments can usefully be contrasted with the periodically almost neurotic behavior of capital markets!
The crisis which started in the U.S. was in many ways quite predictable and the extent to which one may have missed certain aspects of it is more a function of lack of disclosure than anything else. We have—or should have—known for many years that the U.S. was sailing in the wrong direction. The imbalances that this mismanagement created were there for all to see, if they would only look. Yet, if this is true, why is it that so many people missed it until it was too late? A number of lessons must be learned:
Cultural diversification is crucial. It gives investors a better chance, as it allows them to look at the same issue from a broader range of perspectives. It would be much too easy to blame any one individual for missing that broader picture. It is a simple fact that most of us are the products of our experiences and of the cultures within which we have grown up. Thus, it should surprise no one that some form of home country bias will cloud the vision of even the most open mind. Anticipating that this should be the case is the essence of wisdom; offering a diverse cultural perspective as a possible prescription is its simple first-order solution.
A willingness to stand out from the crowd is a second requirement; behavioral finance implicitly tells us it is one of the hardest goals. It is not surprising that anyone should feel better when his or her own view appears corroborated by others. Yet, we know that “group-think” can be dangerous. There are two ways which can make it easier to stand away from a crowd and maintain some comfort level. The first is to have a willingness to “worship” data. Many errors, in this cycle and in prior instances, can be directly traced to the narrow framing which behavioral finance predicts. The second is to have a clear process through which to consider fundamental data. This process must be solidly grounded in philosophical beliefs, tested empirically, based on principles rather than rules and equipped with simple “gut” or “credibility” tests, as common sense built through experience is still a reliable indicator. It should also allow for ex post facto auditing.
Anticipate likely biases and be prepared to deal with them. Only in a theoretical world can one assume that everyone thinks in a totally objective manner. Thus, rather than looking for the utmost form of objectivity, it may be more reasonable humbly to ascertain which bias or preference one might bring to the scene. It is often better to keep looking for evidence that might disprove or at least weaken the case than to try and confirm a hunch. The investment business routinely involves a “high” probability of making the wrong decision. Though this may be truer when applied to markets than to economics, the importance of various preferences, propensities or other soft concepts exposes any forecaster to error. Being prepared to accept that risk is crucial; it naturally should lead to seeking any hint that one is proceeding down the wrong road. The three most frequent sources of error are: bias in the design of an analysis, partial perspective or incomplete data set. Testing a hypothesis for these is not a bad starting point.
Be patient. Though fundamental analysis eventually provides the right answers, it may well prove out of synch with realities or general perceptions for quite some time. This poses a challenge to the analyst who must deal with the day-to-day strictures of real life in markets. The challenge is to identify the problem and seek the catalyst that might bring the whole edifice down. Insightful scenario planning may help describe the possible paths to that eventual end point, though the timing of the reversal is accepted as hard to predict with precision. It might then make sense to adopt a different thinking process: respect markets in the short term, but avoid ever having a short-term bet that would go against a long-term expectation. In short, though it may be too early to assume that a change is coming, one should not argue that change will never come, simply because it has not occurred yet.
THE LIMITS OF TRADITIONAL PROCESSES
Many investors were unprepared for 2008 and their unpreparedness was related to certain beliefs associated with traditional finance. Among the many topics which are said to have created surprises, the following three stand out: 1) diversification does not work; 2) capital market returns are not normally distributed: the tails are much fatter than thought; and 3) alternative strategies do not achieve the goals they advertise.
Diversification still does work. This is potentially the most significant challenge to established investment thought because diversification is a crucial basis of prudent portfolio construction decisions. The logic behind the claim that diversification did not work in 2008 is that there seemed not to be much of a place to hide. Virtually anything and everything seemingly lost money! Yet, this does not prove that diversification did not work. What if it was simply a case that investors’ expectations of what diversification should have done were not met? This reflects the mistaken—and yet often promoted—belief that diversification works because certain prices go up when others go down. In reality, prices will generally move in similar directions, but with different orders of magnitude, because correlations across asset classes or strategies are more often than not positive. Thus, a first lesson from 2008 is that diversification is not nearly as powerful a shock absorber as one might think. In fact, correlations across equity markets behaved in ways that were neither really surprising nor totally unexpected. They did vary somewhat, but whatever distortion one saw was well within the range of possible outcomes.
Are returns normally distributed? Finance theory broadly assumes that most investment returns are normally distributed. Statisticians generally believe that a normal distribution should have minimal skew and kurtosis. In fact, over the long term, the data do suggest that most traditional asset classes or strategies seem to have generally normal return distributions, with the notable exception of high yield U.S. bonds, absolute return strategies and oil, which often display negative skew and excess kurtosis. Yet, except for U.S. cash, U.S. investment grade bonds, equity long/short strategies and gold, most strategies and asset classes displayed non-normal return distribution characteristics over the last three and five years. Thus, distributions which are normal in the long term can have non-normal features during shorter-term interim time periods. Recent markets were admittedly unpleasant, but their behavior is within the range of reasonable expectations, based on their long-term historical characteristics.
What about alternative strategies? It is but a truism to say that so-called alternative strategies did not meet expectations. Yet, a more careful look at their performance over both short- and long-term intervals does suggest that, with few exceptions, return patterns were not nearly as disturbing as one should have expected. It would therefore seem reasonable to conclude that a wholesale rejection of the so-called alternative strategy universe is neither called for nor sensible. Clearly, the world has not appeared normal in the recent past and this should serve to shift expectations. Yet, one would think that four general themes are worth pondering. 1) Pure arbitrage strategies seem to involve considerably more risk than initially thought, principally because of leverage and operational issues and as a result of the complexity and at times opacity of the strategies or of the instruments they use. 2) Credit and event driven strategies are doing what one would expect, although they ought to be viewed as riskier than investing in traditional bonds: they produce both higher volatility and higher return than traditional fixed income and can suffer from a dramatic lack of liquidity. 3) Equity long/short strategies should continue to be viewed as an alternative to long-only equity investments. In times of stress, they will likely suffer from a lack of liquidity and can disappoint, but there is no compelling evidence that they involve more than the risks that we know and respect. 4) Fraud is a challenging problem within the hedge fund community. Many managers do not provide the transparency required to conduct solid due diligence, and there have been massive failures among those who are tasked to provide oversight.
DEVELOPING NEW TOOLS OR APPROACHES
Our final main question relates to the potential to develop new tools or approaches to take advantage of the lessons learned. Though it does not appear that 2008 warrants a wholesale re-evaluation of the way private asset management is conducted, there still seems to be room for some improvement. This at least includes three important areas: 1) the use of goal-based strategic asset allocation; 2) the need to revisit the nature and extent of traditional capital market assumptions; and 3) the concept of “scenario-diversification” as opposed to the more traditional “statistical diversification.”
Goal-based strategic asset allocation: The last several quarters have made it abundantly clear that many individual—and quite a few institutional—investors were neither ready nor prepared for the kind of capital market dislocation they experienced. This alone should serve to remind us that the intellectual and theoretical purity of traditional finance may well not be truly appropriate in the real world. Thus, optimality as purely defined, may simply be a wonderful academic assumption, but also one that is not realistic. Individuals need to have a better understanding of why their portfolios look the way they do and why they need to be so, given the variety of needs they have and the multiple risk profiles characterizing them collectively. Goal-based allocation may, in the end and theoretically, involve some measure of sub-optimality yet it is the one approach that maximizes the chances that an individual or a family will be able to live through the most challenging times.
Different capital market assumptions and optimization approaches: Traditional mean-variance optimization only requires assumptions as to expected return, volatility and correlations. It may have the great advantage of being elegantly simple, but it shows serious limitations in the real world. These limitations fall into three distinct categories. 1) We have just seen that though returns for most asset classes and strategies can be viewed to be normally distributed over the long term, they may not be both in the shorter term for certain classes or strategies and even for others. The ability to understand, predict and illustrate the implications of non-normal return distribution events should be part and parcel of the investment advisor’s tool kit. 2) The sole use of mean-variance optimization may be doomed to failure: the presence of a “single” recommendation with single estimates for expected portfolio return and risk may project the wrong image. A statistician looking at portfolio return and risk estimates immediately transforms them into a probability distribution. Individuals may or may not visualize the same thing. Thus, the addition of wisely designed and executed scenario analyses can bring that distribution to life, particularly if advisors do not gloss over the non-central outcomes. 3) An important consideration such as liquidity, both for the portfolio as a whole and for individual elements within it, needs to be better understood and, if possible, modeled.
Statistical versus scenario diversification: Though we feel that statistical diversification really did not fail, it is not terribly helpful to keep arguing this in tough times. One element of the solution must be to improve investor education, but this has inherent limits: “tell me what time it is, not how to make a watch” is the usual investor reaction to what appears too theoretical. An alternative approach well worth investigating is to go behind the real meaning of diversification and to introduce the concept of scenario diversification. Over the long-term, it may be somewhat pointless to wonder whether there should be explicitly different scenarios under consideration. Yet, over shorter-term time intervals, the challenge of envisaging different paths is considerably less daunting. It simply involves recognizing that the uncertainty associated with current conditions and assumptions may produce a few different outcomes. Abandoning the purely institutional principle according to which portfolios should be systematically rebalanced to their long-term strategy one can begin to imagine a slightly more enlightened two-step approach to tactical portfolio management. 1) Specify the economic and corporate fundamentals scenario which one believes is the most likely outcome, and identify the likely capital market developments which should be associated with it. 2) Compile a list of the parts of the central scenario that are most likely to be wrong and identify the investment strategies that might perform best if alternative scenarios came to pass or which might be directly related to a catalyst that makes the base scenario be wrong.
That the last 12 to 18 months were unusual and unpleasant is neither hard to believe nor even really debatable. Yet, though there are a few lessons that ought to be learned and should result in modest changes in the way one approaches investment management, a somewhat dispassionate analysis of what actually happened and how does suggest that that wholesale change may be akin to the proverbial throwing of the baby out with the bath water.
We almost decided to call this Summer 2009 issue of The Journal of Wealth Management a special international issue, as it did look for a time that all articles would be sourced overseas. Though this could have been a dramatic statement, we felt it was better to be blind to the domicile of our authors in order to have as varied a menu of articles as usual and possible.
The first three articles can be thought of as focused on very broad events or issues. The first, by Dean LeBaron, Walter Deemer and Mark Ungewitter, puts recent market behavior into historical context. The second, by Taruna Shalini Ramessur, tests for the presence of altruistic bequest motives in a small liquidity-constrained economy, Mauritius, in a study that could profitably be replicated in larger countries. The third begins to offer a transition into individual investor issues as Allen Atkins and Frank Caliendo ask: Should social security benefits be initiated? The answer may not seem obvious since there is a key trade-off involved as the retiree can choose low benefits for a longer period of time, or high benefits for a shorter period of time, or something in between.
The next two articles deal with important wealth management issues. The first, by David Weinreb and Warren Litman, focuses on multi-generational wealth transfers in general and more specifically on intergenerational GRATs. The second, by Allen Hamm, focuses on the issue of long-term care planning and notes that it shares two of its goals with financial planning: asset preservation and preservation of legacy.
Our final five articles cover a wide swath of territory The first, by Mark Barnes and Vincent Lee, considers the effects of macroeconomic and firm-specific factors on shareholder wealth, focusing on their own country Australia. The second, by Dr. (Ms.) Guntur Anjana Raju and Rudresh Kunde throws light on the development of an IPO index and structuring a model IPO index for the Indian primary market. The third, by Joseph Aharony and Eli Noy, asks a very important and potentially quite topical question when dealing with strategic equity investments: What matters most in the corporate world—profit or growth? Our fourth article is highly topical as it deals with the view offered by Francois Lhabitant and Greg Gregoriou that numerous red flags ought to have led many investors to stay clear of Bernard Madoff’s Ponzi scheme. The last piece is unusual in that we have only once published a book review prior to this one, but it is a very interesting review by Vassilios Karavas of Greg Gregoriou’s Encyclopedia of Alternative Investments.
TOPICS: Wealth management, portfolio construction, financial crises and financial market history
Jean L.P. Brunel
Editor
- © 2009 Pageant Media Ltd