Tuesday, January 3rd, 2012
the haircut

At the start of a new year, there is a tendency to think grand thoughts.  So, the first few postings of 2012 will focus on some big ideas about investment decision making.

They are all simple.  Not simple to implement, but simple in concept.  Yet they are practical, not theoretical.

The first I’ll call “the haircut.”  I speak not of a haircut on the stated value of a security for margin purposes, capital requirements, or the like, but rather of a haircut of expectations that should be a part of investment planning (and almost never is).

While it’s a bit of a fool’s game to predict asset class returns, it is necessary to have some baseline values as a framework for planning — and in some situations (as with pension plans) there is a requirement to put a stake in the ground with a specific expected return.  It’s most common to look to historical returns (like “the famous nine percent”the research puzzle | This piece is from that famous fall of 2008.), when it makes more sense to take a probabilistic approach that incorporates the fundamental building blocks of future return, most importantly the current valuation of each asset class.  With either approach, when you come up with a number, that’s when the haircut needs to be applied, and it shouldn’t be just a modest trim.

Planning should build in a margin of safety — a big margin at that.  Not a few tenths of a percent off of the estimate of future returns, but several percentage points.  Yet as I look around, I don’t see that being done.

Take pension plans.  Corporate and government plans consistently have used expected return projections that are too high.  In most cases, I’d argue that they have been too high even if the goal is to predict the most likely return profile, but that shouldn’t be the goal.  The goal should be to minimize the likelihood of substantial shortfalls, which arrive at just the moment when they are hardest to remedy.  The same goes for foundations and endowments.

And individuals.  A financial plan that is based upon return assumption that are too aggressive is inherently risky; the normal functioning of financial markets means that the assets in the plan could be under pressure just as the vagaries of life make themselves known.  Without a sizable haircut, historical returns clearly qualify as “too aggressive” when making plans.

So, why do organizations and individuals keep adopting these ill-advised forecasts?  Because it hurts to do the right thing.  Lower projected pension plan returns mean benefit cuts for workers, more contributions from them, or higher taxes for government plans and lower earnings for the companies that have plans.  Charitable organizations would have to scale back their operations to adopt a more conservative approach.  Individuals would have to save more or spend less to meet their goals.

The other reason for the status quo is that there are enablers who bless the projections.  Pension consultants and other experts christen some level of return as reasonable, based upon modern financial market history (which itself is too limited in time to provide an accurate framework for estimation), and don’t adjust it to reflect the skewed risks from a poor forecast.  Financial planners may do a Monte Carlo simulation when preparing their advice, but most plans still use return projections that are much too optimistic.

Conversely, what’s the worst that can happen if modest assumptions are used?  In situations where shortfalls develop, by definition they would always be less onerous than those under the standard way of operating.  And greater-than-expected results would deliver a flexibility that would allow future goals to be met even more easily.  (That is, if the decision makers resist the age-old temptation to increase return forecasts just because recent results have been good.)

It’s important to note that this is not a prediction about anything.  I have written about how the realities of “bond math”the research puzzle | From a series I wrote on investment styles. in the current environment make many fixed income return projections wildly out of line, but it could very well be that other assets could pick up the slack.  This is, instead, a finger pointed at a practice in the industry that has done great damage.  It’s time to call it a mess and take the scissors to it.