Dan Brown, author of the bestseller “The DaVinci Code,” states that “History is always written by the winners. When two cultures clash, the loser is obliterated and the winner writes the history books — books that glorify their own cause and disparage the conquered foe. As Napoleon once said, ‘What is history, but a fable agreed upon?’”
Famed wit and former British Prime Minister Winston Churchill may have agreed. He spoke of the three kinds of lies: “There are lies, there are damned lies and then there are statistics.” The implication was that statistics (and the artistic way some present them) become the biggest lies of all. They are fables made for illusion more than illumination.
With all due respect to Churchill, let’s agree that statistics don’t lie. To the contrary, they simply reflect what is measured. Trouble happens when the wrong things are measured or the right things are measured incorrectly. And this bit of alchemy is the charmer’s playground from which all manner of “statistical truths” is possible, and tales of doom or divinity may be spun according to the fortune teller’s design.
The statistics investors care about most are portfolio returns. Those who deliver the finest will see the world beat a path to their door. With the benefit of hindsight, these performance numbers are incontrovertible. When the investment timeline, initial investment cost, periodic cash flows and closing balance are all known, there’s no way to misrepresent the investment return. At that point, it’s all quite explicit and each dollar (invested and earned) is accounted for. A 12-year-old with a cellphone app can do that math. All is known and in the rearview mirror. No mystery there.
But how should savvy investors interpret forecasted returns? Can they be reasonably trusted? Consider that a fixed-income investor might forecast a specific total return (nominal yield + price change) assuming a future market yield (price) for a given investment period and be quite close to achieving the forecasted total return if the “guestimated” future yield (price) proves valid. However, when this premise is dissected, a corollary emerges. There appears to be a proportional relationship that suggests the accuracy of this forecasting hinges on the prevalence of a bond’s explicit, constant and known properties. Thus, the more obvious and measurable the features of a bond, the more likely the forecast will reflect reality. And by default, the more inexplicit the features of a bond, the more likely the forecast is incapable of producing predicted outcomes.
A subtle indicator that forecasting is likely to fail is the modeling of specific future yield (price) predictions on securities with abundant complexities, which are not uniformly understood by the marketplace. These bonds tend to price according to the migrating passions of frenzied bond traders and without a well-articulated market.
Home equity conversion mortgages (reverse mortgages, a.k.a. HECMs) are an excellent example of this phenomenon. They are promoted as though they are well understood and properly digested by the larger market. Often, a portfolio manager will be further convinced to employ a complementary strategy of purchasing other securities with properties designed to inversely balance the HECMs. Yet HECMs are thinly traded and rarely possess known prepayment characteristics or trading patterns. Thus, they are likely to surprise (and not the good kind). Accordingly, when the primary strategy falls apart, so does the complementary strategy. At this point the colorful presentations that illustrated complementary benefits appear to be frothy props to market high profit securities that have very few market-makers. Despite the artistic and attractive wrapping paper their champions employ when presenting them, this is not a formula for highly predictable returns, but it may be a formula for sincere disappointment. A computer programmer might describe this dilemma as “garbage in, garbage out.” Failure looms when intricate components and complex properties are illustrated to produce outcomes that are not necessarily accepted by the broader market.
If forecasted returns are to be considered reasonable, the securities must behave in a manner that is credibly illustrated for a given data set. High quality, non-callable bonds are the best example of this, and uncertainty accrues with every additional feature (call provisions, prepayments, fluctuating coupons, credit perception, market breadth, tax status, etc.). Uncertainty and predictability are opposing forces. The greater the former, the less likely forecasted returns should be considered reliable.
When securities fall out of favor there’s no analysis that can accurately forecast pricing, thus total return predictions are doubtful. Consider TruPS, CDOs, Alt-A and private label mortgages of the past decade. Anguished investors were sold a good story and promised exciting returns. Were those debacles accurately predicted?
A good story is a lot of sizzle. A good meal is a lot of steak. Smart investors prefer a fine steak and avoid being sold inferior meats with a sizzle. What is often forecasted with complex securities may actually be achievable with explicit, simple securities like bullets, plain vanilla MBS and high-quality municipal bonds. These predictably satisfying entrées should be staples in your portfolio diet.
Chris Thompson is executive vice president in the Capital Markets Group at Country Club Bank in Kansas City.