Cashing in on Monte Carlo

"... Suppose you have a $500,000 portfolio when you retire. How much can you take out each year?

"... People should consider a full range of possible scenarios for investment returns. Don't rely on a potentially misleading long-term average.

"... What is the last thing you do before you climb on a ladder? You shake it," says Sam Savage, senior research associate at Stanford University. And that is Monte Carlo simulation."

"... Monte Carlo tools can play a powerful role in making retirement planning more realistic and accurate."

— ELLEN ROSEMAN, The Toronto Star

Full Review

Wow. That was some scary week on the stock markets.

Toronto's S&P/TSX composite index had its worst week in two years, falling 4 per cent by the closing bell on Friday. Stocks rallied early in the day yesterday before falling back to a flat finish.

Ups and downs are normal when you invest in securities. You can't avoid them. You have to plan for them.

But you wouldn't know it by reading the stuff churned out by the financial-services industry.

A frightening descent? A roller-coaster ride? Forget it. You're more likely to encounter terms like "volatility" and "standard deviation."

I'm glad to see one mutual fund manager, Fidelity Investments Canada Ltd., taking a different tack, urging financial advisers to stop relying on "the flaw of averages" in retirement planning.

In a research report, Fidelity points a finger at one of the most common and potentially disastrous mistakes advisers make: basing a plan on historical average returns and then projecting those returns out in a linear manner for 20 years.

It's somewhat like deciding to wade across a river based on the average depth.

"The average may be four feet, but that won't help you when you're in the middle of a section that's 12 feet to the bottom," Fidelity says.

So, what's the solution?

You know that markets don't move in a straight line. You know that investment returns occur in a random fashion over time.

The Monte Carlo simulation is a mathematical model for computing the odds or probability of an outcome by testing thousands upon thousands of possible results.

As a forecasting tool, Monte Carlo analysis has been used for years in economics, chemistry, nuclear physics, city planning and the programming of slot machines.

But the tool can also help you figure out how long your retirement nest egg will last, given the inevitability of stock-market ups and downs.

Here's an example from the Fidelity report.

Suppose you have a $500,000 portfolio when you retire. How much can you take out each year?

Your financial adviser may tell you that a conservative portfolio such as this one has earned a hypothetical return of 5.9 per cent a year in the past. You'll be safe withdrawing 5 per cent a year on an inflation-adjusted basis. Your portfolio will deliver a reliable income stream over 30 years or more and you'll be close to your 100th birthday when the money runs out.

That's terrific, you think. You feel secure.

Now, let's look at what happens when you run the same portfolio with the same withdrawal rate through a Monte Carlo simulation model.

The simulation shows a very wide range of possible outcomes.

Instead of a simple straight-line graph, you see an eye-popping array of squiggly lines.

In real market conditions, this portfolio, which looked rock solid based on projecting past averages into the future, actually has only a 73 per cent chance of delivering income for 25 years through age 90; and only a 57 per cent chance of delivering income to age 95.

Perhaps more troubling, the portfolio also has a significant chance of running out in less than 20 years. That's a possibility that an income projection based on averages completely masks.

A Canadian software program called RetireWare has a $79.99 Monte Carlo edition you can download at the Internet website, http://www.retireware.com. The program is updated once a year by Toronto-based Apeiron Software Ltd. [Our emphasis added]

Monte Carlo should not be viewed as a certainty test. It is a probability test, the software maker says.

"Ultimately, there will be only one outcome, but knowing not to take more risk than necessary and finding a safe spending level is invaluable information."

I like the line from a BusinessWeek article posted at the RetireWare website.

"What is the last thing you do before you climb on a ladder? You shake it," says Sam Savage, senior research associate at Stanford University.

"And that is Monte Carlo simulation."

Fidelity plans to create a seminar to help financial advisers talk to clients about retirement-income planning.

People should consider a full range of possible scenarios for investment returns. Don't rely on a potentially misleading long-term average.

Monte Carlo tools can play a powerful role in making retirement planning more realistic and accurate.

So, ask your financial advisers if they can do such an analysis for you.

Ellen Roseman's column appears Wednesday, Saturday and Sunday.

© Toronto Star Newspapers Limited. All rights reserved.

Press