Monte Carlo simulations are a significant improvement over forecasting methods that don't allow for variable results. But there's still a potential problem with the analyses.
Even if the simulation output includes a disastrous year like 2008, that simulated result gets averaged in with the other years' results. Clients, however, don't receive the average forecasted return on their portfolios — they have to live with portfolios that probably are down significantly and generating less income.
So, what do they do now?
Jonathan Guyton, CFP and a principal with Cornerstone Wealth Advisor in Minneapolis, has studied and written several articles on the use of simulations to determine sustainable retirement-portfolio withdrawal. (Several of the articles are available online at www.cornerstonewealthadvisors.com/pages/f_news.htm.)
He believes that you can prepare clients for years like 2008, although Monte Carlo software is not the way to do it. "
We can say to clients, suppose your portfolio went down 25 percent," Guyton says. "Let's talk about the planning we've done, the strategy we're using, and give you a sense of not only what would that be like, but how relatively well prepared you would be."
Guyton points to several problems with relying exclusively on simulation forecasts for clients' retirement plans.
The first issue is that the software doesn't help the advisor or the clients understand what went wrong in the simulations that resulted in a failure. For instance, was the portfolio depleted close to the client's life expectancy or did the money run out years earlier? While that's useful information, it's usually buried in the simulation's output.
Another problem with simulation is that it doesn't allow for mid-course corrections. In other words, the software can only run each projection to the point of failure or success: Either the money outlasts the client or the client outlives the money.
