As recently as 10 short years ago, asset allocation for wealthy individuals and their families was almost totally reliant on the framework that had been successfully applied to the institutional world. Families frequently did not really like the actual outcomes, but we, as advisors, did not have anything much better to offer. Led by a few original thinkers, the industry looked for more satisfactory alternatives. An initial step was to recognize that the mean–variance framework relied on a few assumptions that might or might not be appropriate for wealthy individuals and their families. A few of these are highly technical, and minor modifications to the framework have traditionally been applied, with no visible adverse consequence—we discussed a few of them in our Winter 2010 letter. Others are more complex and require significant modifications, but, as shown here three months ago, still eventually rely on the same generic framework. We concluded then, and still believe today, that the generic mean–variance framework may be the worst alternative, except all others!
Our point today is thus not to revisit that conclusion. Rather, it is about suggesting a revised process to get to the point where the quantitative work must take over. It is about looking at each of three challenging assumptions and viewing them as bifurcations in a strategic integrated wealth management decision tree, the purpose of which is to arrive at a finer matching of a family’s dreams, nightmares, and circumstances to the deployment of its assets.
Let us turn to these three assumptions. First, the traditional framework assumes that individuals and their families have but one goal and one risk profile. Second, it assumes that they also have but one time horizon. More subtly, but equally importantly, it assumes that they have but one view of the world. Although these assumptions may be eminently reasonable when one deals with a board made up of professional investors, we intuitively and empirically know that they are questionable when it comes to wealthy individuals and their families. The lack of a well-defined and easily quantifiable liability stream makes it doubly difficult to abide by what may appear as too “cold” a process.
Behavioral finance helped us appreciate that families have more than one goal and that each goal is often associated with a different risk profile. This naturally led us to goal-based allocation or—as Das, Markowitz, Scheid, and Statman have called it—mental accounting–based allocation. Jaeger helped us think of the issue in slightly different terms but in a way that is still quite consistent with behavioral finance: Individuals and their families may have more than one time horizon! Thus whether one focuses on goals, which may have different time horizons, or on time horizon, which may be associated with a different view of one’s goals, the issue essentially boils down to the need to be able to break the hitherto unassailable tenet that only a solution that considered the whole portfolio as one was acceptable.
The next step in this logic, one which we began to discuss here a few quarters ago, relates to the third level of complexity: What if the future was not continuous? What if one could not realistically describe the future capital market environment with a single set of numbers—a return and risk assumption for each asset class or investment and a correlation matrix uniting the whole? What if the future could take so many realistic and different shapes that this single-scenario forecast essentially made little or no sense? One is reminded of a variant of the famous quip by an old statistics professor: “If your head is in the oven and your feet are in the refrigerator, does it mean that, on average, you are comfortable?” This question may well be particularly relevant as we write these words.
Imagine that one worries that a systemic change may be coming. History offers precious few examples from which to learn. Furthermore, these examples are so few and far between as to make them virtually useless as potential guides to the future. Consider the last two: the loss of the role of the pound sterling as a global currency in the late 1960s–early 70s and the end of the Japanese economic growth miracle in 1989. The former occurred at a time when global capital markets were considerably less integrated, when another currency—the U.S. dollar—had already taken on a major reserve role, and when global capital flows were not nearly as free as today. The latter example occurred in a country that, as a net exporter with a somewhat controlled currency, could experience an internal meltdown without much impact on the world. By contrast, as a major net importer, the U.S. would likely export at least some of the lower demand associated with a major case of domestic weakness to the world’s exporting community. Yet, irrespective of how few insights one can really glean about the future, one thing should be clear: how can one blissfully ignore such a potentially crucial development when planning for a family’s financial future?
One of the most important issues with which we have learned to deal is the risk that any strategic decision will at one time or another become so uncomfortable that it will be put into question at potentially the worst possible instant. We have dubbed this “decision risk” and have defined it as the risk of changing horses in the middle of the race. We have thus postulated that any strategic asset allocation or investment policy exercise should be mindful of and take into account the need for the outcome to be sustainable through time. Recent events go one step further; they suggest that a strategic asset allocation also should comprise decision rules or processes that allow it to evolve in a somewhat predictable manner should circumstances change.
Our focus is thus on the challenge associated with the twin goals of mapping a family’s circumstances and financial or real assets as closely as one can and of doing so in a way that is sustainable through good and, more importantly, bad times.
The seminal research and associated article by Das, Markowitz, Scheid, and Statman, which we discussed here three months ago, provides an interesting framework to address the crucial issue. To reiterate for readers who might have missed these comments last quarter, the authors compared what they called mental accounting (MA) (or goal-based allocation) and mean–variance theory (MVT). They concluded that there is a practical equivalence. Integrating MVT and MA requires one to consider that individuals need a multiple sub-portfolio structure, to define risk as the failure to reach a certain threshold outcome for each sub-portfolio, and to define the optimization process as maximizing expected returns subject to some minimal probability of failing to achieve some desired outcome. This equivalence has three important consequences according to the authors: 1) the optimal mental accounting sub-portfolios are on the efficient frontier; 2) the mental accounting constraints can be mapped to some implied risk aversion; and 3) a combination of several optimal mental accounting sub-portfolios may very closely approximate a single optimal portfolio. In short, one can simply conclude that an individual’s multiple goals and individual risk tolerances for each goal can be viewed as a depiction of his or her utility as defined in traditional finance. Practically, in the context of this discussion, one can view these multiple sub-portfolios as an initial set of bifurcations in a decision tree.
The recent debate as to whether the current environment may, or may not, involve one of those few epoch-marking structural changes raises the next important dimension. To what extent is it possible that this could be viewed as another bifurcation in our decision tree? It clearly behooves each advisor or wealth advisory firm, in the end, to form an opinion as to whether these complexities are manageable and should or should not be a part of their solution set.
Let us consider, however, a potential and somewhat generic decision tree that comprises three consecutive forks—which we will in due course suggest could be viewed as strategic “dials”—each addressing an important issue and each subject to certain variations over time. Our tree is thus defined around our three important issues: multiple goals, multiple time horizons, and multiple capital market scenarios. Note that we can view the original mean–variance theory design as being one that postulates one goal, one time horizon, and one capital market scenario. It is thus a perfectly valid approach, but it may turn out to be a special form of a more general solution.
This conceptual tree might thus start with what is often the first key question asked by a family: Can I divide my wealth between the assets needed to support my lifestyle and those that meet other goals? This question simply recognizes that the worst nightmare that families often formulate is the need for them to be forced to change their lifestyles, especially in a way that outsiders might notice.
Dividing the portfolio into two sub-components would allow one to look at two broad buckets. The first is designed to defease lifestyle needs for a period long enough that one could minimize the risk that any short-term mark-to-market loss in a risky asset is not turned into a permanent loss of capital as one has to liquidate risky assets to meet lifestyle needs. Note that this lifestyle portfolio would be defined to include all assets that are needed to meet one’s lifestyle, both the financial assets required to fund spending and the real assets that provide any personal utility, such as shelter.
The second is designed to bring together all other needs, excepting any need that involves ongoing spending, which is incorporated in the first bucket. Note that this may be the point where a family might need to take a major step back and begins to identify separate “entities,” each of which ought to have its own decision tree. These could include irrevocably gifted assets—whether in trusts or through a foundation—that the family still views as a part of itself but that realistically should be handled separately. Alternatively, one might need to view individual family members through distinct trees if their needs are so potentially radically different from one another as to make the “averaging” process senseless.
The next important issue relates to the question of time horizon. At one level, although there is a clear time horizon issue with the non-lifestyle bucket, one could argue that the time horizon issue is principally concerned with the lifestyle bucket. Indeed, the non-lifestyle bucket really deals with needs that have quite a long fuse by definition. Thus, breaking that horizon into long-term and really long-term sub-horizons might amount to splitting hairs. The same cannot be said of the lifestyle defeasing bucket.
History would suggest that the lifestyle bucket might be designed to cover at least the next 15 years if not more, simply because this is the minimum time frame over which equity markets have typically always produced average compounded positive returns. Yet, the experience of Japan since 1990 and the last 10 years in the U.S. might require one to be prepared to lengthen that minimum time horizon. Given the fact that the required time is thus quite long, one would have to build some risk exposure into one’s investments to ensure that real purchasing power is preserved. So, with the portfolio volatility associated with such inflation hedges, it is conceivable that many families or individuals would want to consider at least two different time horizons: 1) some shorter-term period over which investment risks would be minimal in terms of return volatility but might be material in terms of loss of real purchasing power and 2) another where the risk associated with the inflation hedge is acceptable.
Turning to the non-lifestyle bucket, an important initial issue relates to the management of the risk. It requires the family and the advisor to address an initial question that is often sidestepped. Most advisors would simply argue that any non-lifestyle investment should assume risk. Yet, why should the family take risk? Here, the issue revolves around identifying the reasons that justify risk taking, rather than simply assuming that “risk is good!” There are many such potential reasons, but five immediately come to mind:
1. the need to deal with unexpected inflation;
2. the need to be able to deal with unexpected lifestyle changes;
3. the need to deal with the potential wealth fragmentation that accompanies the generational process;
4. the need to deal with “mission” investing, which is defined here as an area of interaction between investment and philanthropy;
5. the need to keep score, defined here as the effort to minimize the risk that a sustained deviation between actual returns and headline returns (the returns which one reads in the press) might lead to a measure of discomfort and thus to the risk of unsustainability in one’s policy.
There are probably other reasons that one family or another might identify. Advisors would be well-inspired not to create their own exhaustive list but rather to keep listening to the way their clients frame the issue.
The final question at this point relates to the potential for there to be multiple capital market scenarios. This involves an initial dichotomy between a “steady state” capital market environment and another, which would postulate several possible scenarios. Whether these are defined in terms of “macro-economic” nightmares, in terms of “major environmental or political concerns,” or simply in terms of a structural change in the world economic order is really up to each family to decide, with or without the help of their advisors. It is indeed almost a truism to postulate that one could draw up a virtually endless list of what might potentially go wrong and still end up with nothing terribly relevant or actionable. On the other hand, if a family has a number of real concerns, providing for a means of dealing with them at the strategic level is not only important but, in fact, crucial if one views the management of decision risk to be an integral part of the process.
Once this initial apportioning between steady state and multiple scenarios is made and the multiple scenarios are defined, one can imagine a process leading to the formulation of sub-portfolio parameters that would guide the composition of each of these individual sub-portfolios.
Such a process might initially appear quite complex, and let’s be honest, it probably is. Yet, it possesses two important characteristics that many families might find quite desirable. The first is that it allows each family to visualize a much more direct relationship between who they are and how their assets are enabling them to proceed toward their goals and dreams and away from their nightmares. It is also possible that a family can view the initial branching in the tree as some apportioning of beneficial interests. The lifestyle bucket is really meant to comprise and bring together all the assets that the family needs to live; it can be seen as the assets of which the family, imagined here as a couple, is the sole beneficiary. The other bucket can be viewed as bringing together all the other interests of this couple—such as dynastic or philanthropic—plus some residual interest they may have in those assets, should the need arise.
The second important characteristic of this process is that it contains a measure of dynamism; a change in one’s initial insight or opinion with respect to any one of our three overriding dimensions (goals, time horizons, and capital market scenarios) can be processed and can directly lead to a policy change without upheaval.
This is what earlier led us to offer the notion that these various “forks” in the tree should in fact be viewed as “strategic dials.” Although tactical investment discussions might address these concerns, they typically ought to be shaded by the relationship one perceives between the reality of the issue and the extent to which it may already be discounted by market participants. Strategically, however, the family’s concerns with a particular issue might evolve. It is thus important to allow for periodic re-evaluation of the initial sizing. which the family went through for each “end bucket,” with a view to ensuring that it is still appropriate. A “dial” in this case can be viewed as an opportunity to vary the importance of—and thus the funds allocated to—certain goals or certain issues. A concern might indeed become more pressing over time, while another might dissipate to some extent. The way this is translated into policy is important, as it serves to preserve the “intimate relationship” that the family has thus created between itself, its goals, and its concerns and the assets they wish to have or need to deploy to deal with them.
Ostensibly, an important additional requirement would have to be to define the critical reporting requirements that such a design might encompass. These would be needed for the family to be able to answer a simple question: How am I doing? However potentially challenging, intellectually and operationally, such requirements would seem absolutely necessary to allow our hypothetical family to deal with its ongoing financial wealth management and to maintain comfort with respect to both the adequacy of its current activities and their congruency with the balance of their wealth management efforts.
The Spring 2011 issue of The Journal of Wealth Management is as “diversified” as ever. Our first three articles delve into important governance and personal family issues, which are clearly becoming increasingly important in the agenda of wealthy individuals and their families. First, Alex Wang aims to understand younger generations’ investing behaviors in mutual funds in order to help wealth advisors understand how better to work with younger generations; he concludes that gender emerges as the most important factor that differentiates younger generations’ investing behaviors in mutual funds. Then, Fredda Herz Brown focuses on responsibility and accountability to one’s own family and proposes one way to teach confidentiality to the next generation in a social media society. Our final article in this section is by Fredda Herz Brown and Dennis Jaffe, and it addresses a crucial issue in wealthy families: overcoming entitlement. The authors suggest that in order to assist younger family members in dealing with this sense of entitlement, there needs to be a thoughtful and systematic process, and they offer a model for thinking about the process and specific suggestions to deal with it.
The next five articles are related to various aspects of asset management. The first, by Philip Maymin and Gregg Fisher, focuses on preventing emotional investing and suggests that an important service provided by investment advisors, and apparently desired by individual investors, is the barrier the advisor provides to prevent the individual from aggressively trading and thereby losing money. Next, Robert Kim, Edward Dougherty, and Miriam Klein revisit an issue that we have not extensively covered in the recent past and which is yet absolutely crucial: tax efficiency. They conclude that looming higher tax rates, combined with a greater emphasis by hedge funds on client service and retention, are likely to make the hedge fund industry more tax-conscious, arguing that tools and techniques for expressing investment ideas in a more tax-efficient manner are available and that the task for managers will be to employ them as appropriate, while keeping investment strategy firmly in the driver’s seat. Next, Peter Mladina develops a methodology to estimate forward-looking long-term active and passive investment returns for major publicly traded asset classes from the perspective of a taxable investor who consumes triple net returns—after all expenses, taxes, and inflation. This section’s fourth article, by James Chong and Michael Phillips, revisits the old issue of “dartboard” investing and conclude that it remains vital for a financial planner to know a client’s tradeoff between the possibilities of higher return from individual stocks, with commensurate chances of loss, or more predictable but lower returns from portfolios of mutual funds. The fifth piece is by Jor Molchan and Fabrice Rouah and is more highly technical: It focuses on the characteristics of sophisticated options and illustrates how delta and gamma hedges break down, explaining one alternative static hedging technique.
Our final article is by Carol Boyer and looks into a topic that we cover only rarely! Boyer examines the relationship between the market for art and the economy. She finds a strong relationship between the market for art and the economy. In general, her results show that as the economy expands, the market for art improves. There is some evidence of a negative relationship between the stock market and art market in the near term and a positive relationship when the stock market is lagged two months.
Although our readers will get this issue after the beginning of the New Year, we still want to offer our most sincere wishes for a happy and prosperous year on behalf of the whole JWM team and to thank you for your continued confidence.
TOPIC: Wealth management
Jean L.P. Brunel
Editor
- © 2011 Pageant Media Ltd