This may come as a surprise to some people, but the baby boom’s retirement is already underway.

In 2004, 22% of the oldest baby boomers, then aged 48-58, were already out of the labor force. Over 2.2 million of them said they were retired. Many others are now deciding whether and when they can afford to retire. Assessing that is complex, and it is one of the most important financial decisions people will ever make.

Researchers are finding that relatively few retirees have made a good assessment of what they needed prior to leaving work. A number of people do turn to financial advisors for help, and that can be a positive and crucial step toward making a good decision. But advisors must provide clear, effective analyses that will allow clients to make informed choices.

Choosing to retire is not (usually) an irrevocable choice, but people who later discover they retired before being financially able to do so can have great difficulty getting a new job at the same compensation level as before. Therefore, it is important to get the when-to-retire decision right the first time.

How well are financial advisors now doing in advising clients about this? New research by Mathew Greenwald & Associates points to disappointing results.

The study consisted of researchers (ages 59, 62 and 65) presenting themselves to financial advisors from seven major financial firms, asking for an assessment of whether they could afford to retire. All but one advisor used a sophisticated model provided by the firm. Each understood the model and believed in it. But, as it turned out, the outputs seem to have serious flaws.

Here is how the process worked.

In the first meeting, the advisors focused on understanding their prospects’ investment risk tolerance and financial needs and goals. They also collected information on the financial assets.

At the second meeting, all but one advisor provided a projection of assets and expenditures for each year through the presumed length of retirement. Two did a Monte Carlo analysis, while the others made straight line projections of investment return, ranging from a 6% return each and every year through retirement to a 7.68% return. The two using Monte Carlo provided an estimate of likelihood of having enough funds: 75% and 90%, respectively. The others gave no assessment of the risk of outliving resources, just an assurance and a projection that the money would last.

A number of variables can affect retirement finances. In each case, the models used assumed a value certain. Example: for inflation, the assumptions ranged from under 2.25% per year to 3.5% per year, for each year over the projected 30+ years of retirement. Each advisor asked the prospect to assume retirement would last until a date certain death, ranging from ages 90 to 95. All also assumed tax rates would stay the same and that Social Security would deliver the same level of benefits it now delivers. There was almost no discussion of the possibility that these assumptions would not be borne out.

There are serious flaws in the model the advisors used, as shown in the chart.

The bottom line: the financial advisors we studied understated the level of risk that these prospects would run out of money if they retired with the projected assets and spent the projected amount.

Also, when an uncertainty is assumed to be a certainty, there is a natural tendency not to recognize the potential for financial products or strategies that protect against the risk. Therefore, the researchers found that most advisors recommended investing in mutual funds only. That’s because only mutual funds were built-in solutions in their model, so they only focused on investment risk.

Incorporating discussions of other risks as variables, not certainties, provides a better platform for developing strategies for protecting against these risks.

For example, guaranteed stream of income products help people deal with longevity risk; many people who are concerned about outliving their money would be considerably more secure if a guaranteed stream of income product was in their portfolio. (It should be noted that, in cases where the researcher expressed great concern about outliving resources, annuities with guaranteed streams of income were not recommended.)

In sum, good advice on when someone can safely afford to retire requires an assessment of all finance-related risks. Some clients will take a chance and retire early, while others may want to work longer and build up more of a cushion. All should have a good assessment of the risks and informed strategies for dealing with them. All should have advice on strategies for mitigating risk.

It seems clear that a new paradigm is needed to help people understand, evaluate and protect themselves from finance-related risk. Developing this is the biggest opportunity facing the insurance business.