The Problem With Living Too Long

By By Fred Reish, Pat Byrnes and Bruce Ashto
May 2011
Excepts provided below. For the complete article, please see the SOURCE URL above.

People are living longer. Undoubtedly, most of us would consider this a good thing, but from one perspective, it’s not, as it makes it much harder to save enough money to generate sufficient income in retirement. Under the defined benefit pension plans that benefitted our parents, longevity wasn’t an issue for retirees; they had a monthly income for life. But in the current 401(k) environment, most employees retire, receive lump sum distributions, roll them over to IRAs, and then have to stretch the money out over the rest of their lives. To do that, retirees need to know the time over which the money must stretch; that is, how long will they live after they retire? But, of course, no one knows the answer to that question.

Longevity is one of the most serious issues facing workers today. When a person retires, his monthly paychecks stop, but most of his monthly expenses continue. As a practical matter, this effectively forces a retiree to turn his or her 401(k) account balance into a stream of monthly payments, taking into account life expectancy, withdrawal rates, investment returns and volatility, and inflation.

The challenge is that most people do not know the probabilities for how long they will live and, in our experience, most people under-estimate the time. This means that they are under-estimating how long their funds will need to last and, thus, the lump sum amount they need and the withdrawal rate they can safely use. As explained by Craig Israelsen in an interview appearing in the July/August 2010 issue of the Journal of Indexes:

“In 1930, it was anticipated based on mortality tables that a retiree might live three to seven years in retirement. Now, for women, it could be 30 to 35 years, maybe 25 to 30 years for men….That’s a lot of money to have saved up to create an annuity stream to cover that many years of income.”

So workers today need to understand that, in setting aside funds for retirement, they should expect to live a long time after retirement. That fact affects how much they should save, but how long is a long time, and what are the chances of that happening?

In other words, if a man retires at age 65, the probability is that his retirement savings will need to last about 21 years, to age 86, with a small possibility (about 10 percent) that his money will need to last 30 years or more, to age 95. Correspondingly, if a woman retires at age 65, the median (i.e., 50 percent probability) life expectancy is 88, or 23 years. About 25 percent will live to 93 or older and almost 10 percent will survive until 98 or beyond.

Put another way, retirees who assume they will not outlive the median life expectancy will be taking a significant risk (50 percent) that they will run out of money in retirement. As a result, any calculation of retirement income over long periods should consider the impact of inflation.

Joint Life Probabilities

If the probable expectancies for single lives aren’t dramatic enough, the probabilities for the joint lives of a 65-year-old couple are stunning. At 50 percent, at least one of the spouses will be alive at age 91 -- 26 years after retirement. There is a 25 percent probability that one of the spouses will still be alive at 95 and a 10 percent probability at 99.

If a 25 percent risk is acceptable, a 65-year-old couple should plan on funding a 30-year retirement. But it is difficult to accumulate enough money to accomplish that goal, and it is equally difficult to properly invest and withdraw money during retirement in a manner such that the accumulated sums are not exhausted.

While these problems can be “solved” – at least partially – by more saving and better investing, they can also be managed by a lower standard of living in retirement, part-time work in retirement, or a delayed retirement.

© 2011 Drinker Biddle & Reath LLP. All rights reserved.