Jankowski created a fictional couple, whom we'll call the Smiths, both aged 64 and retired. Larry has an IRA with $350,000 and Joan has qualified plan assets totaling $200,000. They also have another $50,000 in savings and cash. Their mortgage is paid off, but they face expenses of $53,000. Larry receives $24,000 a year in Social Security benefits, while Joan gets $12,000 in benefits.

To keep the model simple, Jankowski assumed they would use the same static conservative asset allocation over the five years and across all four economic scenarios: 5% in cash, 30% in intermediate-term income, 25% in long-term fixed income, 15% in large growth stocks, 15% in large value stocks and 10% in small-cap stocks.

The results show that in five years the Smiths, in a Goldilocks economic environment, face a best-case scenario of $875,319. In a worst-case scenario, however, a deflationary environment, they'll see their $600,000 nest egg fall to $397,959-less than half of what they'd have in the best case. Weighting a portfolio using our assumptions for the probability of each economic scenario occurring, the portfolio will be worth $632,276 after five years.

What's troubling, however, is that most people would not project a Goldilocks outcome over the next five years. A 9.5% real return seems to be a remote possibility. In fact, the deflationary scenario is looking more realistic, as James Bullard, president of the Federal Reserve Bank of St. Louis, warned that the Fed's policies were putting the economy at risk of a Japanese style deflation. At the same time, the growing evidence of a double-dip recession has driven investors to less risky assets-toward bonds and away from stocks. 

Worse, advisors will find it hard in a deflationary recession to model financial plans. Even if they put the statistical inputs into MoneyGuide Pro, the application does not allow a negative inflation rate. The best they can do is model a zero inflation rate. Thus, accurately showing the Smiths in a financial plan how a deflationary cycle would hurt their retirement is far too complicated. And yet still necessary in this environment.

Put aside the fact that planners are often ill-equipped to judge the probability of one economic scenario over another and to attribute returns, standard deviations and correlation coefficients to differing regimes. Indeed, many planners are not nearly so analytical in their approach and are more focused on relationships, service and sales.

So the solution has to be better software. "Financial planning software applications used by CFPs all generally work the same," says Jankowski. "The assumptions built into the plan are typically based upon a single table in the software system, which uses a single set of assumptions regarding returns and standard deviations for each asset class for the entire plan period, which could be 30 or even 50 years depending upon the age of the client."

She adds that the financial planner community should challenge the way plans are currently being performed.

Though MoneyGuide's user-friendly goal-based approach set a new standard, planners will need another burst of innovation to create more sophisticated models in the future. Software that might point the way has been created for large institutional investors by money manager Mark Kritzman.

Kritzman is the president and CEO of Windham Capital Management in Boston, which manages about $35 billion and advises about 200 of the world's 250 largest investors. His firm's software development arm has built an application to conduct economic scenario investment plans for institutions. It's pricy at $12,000 a year (a $2,000 version for advisors failed to sell). But even if it's out of your price range, the underlying fundamentals of the program deserve attention, as they cut right to the heart of scenario planning.
Kritzman and his team use a statistical procedure for parti­tioning historical returns into periods of turbulence and calm in markets and apply this to portfolio construction. A set of returns is statistically unusual if one or more of the returns are much above or below average or if the returns interact in an uncharacteristic fashion.