Journal of Financial Planning: September 2011
“There are two kinds of forecasters: those who don’t know, and those who don’t know they don’t know.” —John Kenneth Galbraith
Alice Lowenstein, CFP®, is a partner and director of managed portfolios at Litman/Gregory Asset Management LLC, a California-based independent investment adviser and publisher of the AdvisorIntelligence research service for financial professionals.
We face a high degree of uncertainty as to how the aftermath of the 2008–2009 financial crisis will play out over the next few years, and the range of potential outcomes is particularly wide. This high level of uncertainty reflects the fact that in dealing with the extremely challenging debt/deficit situation in the United States and globally, we (and central bankers and policy makers) are in largely uncharted waters. Yet against this backdrop, advisers must continue to prudently manage client assets to meet their clients’ financial objectives.
Today’s environment may represent an extreme, but uncertainty is always an element in what we do. Investing requires us to assess and act on an inherently unknowable future. So to effectively navigate the current landscape and be successful over time, it is essential to have an approach that acknowledges and addresses this reality.
What Don’t You Know?
As advisers, we need to have strong convictions and use them to make good decisions. But this can only happen if the strength of those convictions is deserved. Some things can be measured precisely, such as the interest rate yields of two sectors of the bond market. Others are so difficult to assess with confidence that it doesn’t make sense to base investment decisions on them, such as what the political will for deficit cutting might be following the next election. To tie this back to Galbraith’s comment is first to acknowledge that we don’t know what the future will hold, and second, to recognize this reality but build a sound, long-term process for making forward-looking decisions.
My firm operates with a few such practices that can be described in simple terms (though they are, of course, far more complex and difficult in execution).
Ranges and Variations
As a core component of our investment decision making, we think in terms of ranges of outcomes. To do this requires answering a number of questions: What outcomes are possible? How likely are they over the time span I’m assessing? What is the magnitude of their impact if they do play out? One way to get answers is through a well-thought-out process of scenario analysis.
To assess the absolute and relative return potential of investment asset classes, my firm starts by analyzing the broad economic environment and underlying conditions and evaluating the similarities to and differences from prior periods. Our goal is to determine the range of macro-economic scenarios we think have a reasonable likelihood of playing out over the next five years.
These probable or possible outcomes take the form of scenarios, and in each we make assumptions for key underlying variables such as interest rates, inflation, earnings, the dollar, and investor risk appetites. The goal is not to predict specifically any of these data points but rather to use them in helping us sketch out scenarios that represent a range of possibilities from optimistic to pessimistic. An example of a scenario we consider is one in which we see high inflation but sluggish growth, or so-called stagflation. Although this wouldn’t be the most likely scenario among our current set, we can use the assumptions within this and our other scenarios to determine probable return ranges for each asset class we analyze.
For example, our valuation model for equities focuses on earnings growth and valuation. In the stagflation scenario, we use lower earnings and valuation assumptions than in more positive economic environments where inflation is contained.
Example: Recent Scenarios
Currently, in our most optimistic scenario we assume that we have a more normal recovery (as opposed to the tepid recovery we’ve seen thus far). Our more negative scenarios include stagflation as well as a deflationary climate. These scenarios arise from our over-arching assessment of the big picture, which factors in the realities of financially stressed households, stretched government resources, and the risk of unintended consequences and outright mistakes as policy makers try to manage through this highly challenging environment. We also consider positive factors such as the likely long-term impact on earnings from emerging markets’ demand. Our analysis is also informed by what happened in other cycles with a similar debt/asset bust.
We view the range of potential scenarios over the next five years as skewed toward subpar return outcomes for equities. Moreover, (1) we do not have a high level of conviction in any particular scenario playing out over the next several years, (2) in most scenarios our analysis suggests the returns to traditional stock and bond indices will be subpar, and (3) we see some potentially very negative scenarios that could play out as a result of the excessive global debt problem. Therefore, our balanced portfolios currently have a conservative bias. They are constructed with the aim to generate satisfactory returns in the scenarios we think are most likely, while mitigating (but not avoiding) losses in the more severe scenarios that we think are lower probability but cannot be ruled out.
Of course, we can never be 100 percent certain, so diversification is always prudent. It’s also the case that no matter how good your decision-making process, portfolio results are still vulnerable to the fact that markets can be irrational for uncomfortably long periods. So we limit and hedge our bets as protection.
Where Could I Be Wrong?
In an economic world where the ground can quickly shift underneath us and where what we are experiencing today is outside virtually everyone’s frame of reference, it is hugely important to think carefully about what could make you wrong. This practice shouldn’t be limited to the current period. It makes sense to regularly question your own assumptions and judgments in order to make sure you are peeling back all the layers of the onion to identify and understand all the important variables that could affect any investment being considered.
To use an extreme example, in 2008 our commitment to intellectual honesty allowed us to recognize that the financial crisis had become extremely dangerous and that reducing equity exposure was a prudent course of action. For the prior 20 years, selling stocks after a 20 percent decline was unthinkable and counter to the beliefs of most in our industry, including us. But new information made us realize we were facing an environment outside of our frame of reference. As a result, we reduced equity exposure in mid-September 2008 (though we wish we had done so more aggressively), and were criticized by some followers of our work at the time. In the months that followed, this decision helped somewhat reduce losses and gave us “dry powder” to deploy in capturing the types of opportunities that often emerge in times of extreme investor emotion. High-yield bonds were one such example.
One way we test our views is by drawing on the perspective of more than one person. For example, each asset class has a lead analyst who maintains the models that incorporate assumptions and forecasts for each key metric in various economic scenarios. The most senior members of our team also follow all the asset classes and help vet the analysis, and everyone looks to identify possible errors in the assumptions or analysis. In fact, in many cases, our research process includes assigning a dedicated “devil’s advocate” whose sole assignment is to poke holes in the arguments being made.
It may not be realistic to take this particular approach if you work alone or don’t have peers with the skills needed to challenge your views. So another way to meet similar objectives is to regularly seek out ideas and information from external sources, including published research, conferences, high-quality blogs, study groups, and so on. It can be of particular use to seek out views that strongly differ from your own. Assuming this contradictory opinion is credible, it may uncover things you haven’t considered or haven’t weighted as important. The next step is to assess this information and incorporate it where necessary. Certainly this iterative process requires additional time, but it can result in important fine-tuning adjustments and even occasional opinion changes.