The time has come to put Monte Carlo analysis on the shelf as a retirement planning tool alongside privatizing Social Security, 8% load funds and horoscopes. As a former advocate of stochastic modeling this is hard to admit. But when black swans become as prevalent as pigeons in Saint Mark's Square the whole notion of probability analysis as a planning tool should raise an eye brow.
Before his conversion, a certain blogger would explain Monte Carlo by piously quoting Aristotle who allegedly said when an outcome is in doubt it makes sense to do what is most probable. Who could argue with Alexander the Great's personal tutor? The logic makes sense but the question to ask is what is most probable? (This assumes there is an answer. Nassim Taleb, author of the "The Black Swan", would scoff at the notion.)
Whatever your philosophy, dramatic swings in portfolio values are not the low probability tail risk depicted by Monte Carlo analysis. In fact, an objective observer might say the white-knuckle markets we've experienced going back to 1987 seem to occur with great frequency. Accordingly, using stochastic modeling to develop an asset allocation strategy is like using the metric system to plan a cross-country road trip. Intuitively it makes sense but it's the wrong standard to use at least when traveling in the United States. This can lead to confusing outcomes that provide a false indication of where you actually are. In fairness, relying on linear return projections is equally specious. Assuming a retirement portfolio will achieve consistent positive performance year after year is like waiting for Godot - it's not happening. So how does this affect retirement planning?
The first step is to acknowledge that volatility is here to stay. This can lead to the great temptation of buying several mattresses (which doesn't mean you're diversified) and learning to sleep with one eye open. As with anything, there's risk in extremes and being fully invested in a particular asset class (or fully uninvested if there's such a thing) is imprudent for most. The key is to be honest with the amount of risk / volatility you're willing to accept. This will help determine an allocation that's unique to your particular needs and if that includes a little mattress stuffing so be it.
That's the easy part. The challenge is then sticking with it through regimented rebalancing, dollar cost averaging and periodic reassessment of your risk tolerance. There are a number of quality planning tools that can help quantify your risk tolerance (just ignore the probability analysis most of them use.) And an experienced advisor can be helpful in taking the emotion out of this process.
Retirement planning is more challenging today than it's ever been. That's why using the right tools and measurements is imperative. Reconsidering the relevance of stochastic modeling as a retirement planning tool can be a catalyst in assessing your overall strategy. Doing so may one day get you to the real Monte Carlo. Just remember they use the metric system in France.
No comments:
Post a Comment