Monday, April 16, 2007

How much longer will you really live, redux?

Re: "Training to Be Old," NYT Business 4/10/2007

The number of remaining years of life is a critical parameter for insuring a happy and secure retirement, and yet it seems to be a difficult number to know, with much popular confusion. The workers approaching retirement in "Training to Be Old," NYT Business, 4/10/2007, at least seem to be erring on the side of caution in expecting a much longer retirement than their parents and grandparents had, which is wise.

But they seem to be receiving advice regarding their remaining years that is of varying quality. One statistic in the article, that there is a 40 percent change that at least one member of a married couple at age 65 will survive to 90, is quite accurate, based on Social Security cohort life tables. The first statistic cited, that retirees will live 11-13 more years than their parents or grandparents, is just wrong and probably based on the wrong statistics. Done correctly, the comparisons show that a male approaching retirement, say age 57, is likely to live to age 81, or 8 more years than his father probably did and 11 more years than his grandfather. Women are likely to live to about 84, only 5 more years than their mothers and 8 more years than their grandmothers. Why the discrepancy? Women's still large lead in life expectancy has been diminishing for some time now.

Rules of thumb are extremely important for planning. But it would be nice to use correct ones.

Monday, April 9, 2007

How much longer will you really live?

An article in the New York Times on Sunday, April 8, quoted a male "soon to be 59 years old, and life expectancy is 76" for him.

In fact, remaining life expectancy for a male of his age today is more like 22 years according to Social Security cohort life tables, or around 30% higher than the 17 years implied in the man's statement. Another way of stating this is that the man's expected life span as of today is 81 years rather than 76.

There are really two things going on here:

  • One is that remaining life expectancy plus your age is always higher than the "life expectancy" you may read about in the paper, which is always life expectancy at birth.

    So even though a male infant born today might face an average life expectancy of about 80 years, a male in that cohort who survives to 59 should expect about another 25 years, for a total expected life span as of reaching age 59 of 84.

  • The other point is that an individual's length of life is based on what we call cohort life tables, which are actual observed mortality for those who have already died plus forecasts of future mortality for those who are not.

    In practice, cohort life expectancies will not deviate from period life expectancies much at advanced ages, where the trends in mortality over time are more muted.

Although it is difficult to find unless you know what you are looking for, these statistics are compiled, forecast, and reported by the Social Security Administration and are available in Actuarial Study No. 120 by Bell and Miller.

You could instead examine remaining life expectancy in the SSA's period life tables, which gives you an underestimate of your average remaining years of life because mortality rates will continue to fall. Based on that period life table, a 59 year old man
has about 21 years remaining.

So what is the point of all this?
    Re: "No Longer an Outsider Looking In," NYT Real Estate 4/9/2007

    If he is as healthy as he sounds in the article, Bob Trenta is likely to enjoy about 22 years or more in his new fourth floor walk-up on the Upper West Side, or about 30 percent longer than he reportedly expected. Remaining life expectancy is a critical parameter for an enjoyable and financially secure retirement, but it is unclear how many have accurate information. To be sure, one cannot save too little by underestimating it, and speaking for younger generations, I can vouch for the benefit of inheritances. But unless he is a smoker or has a preexisting health condition, Mr. Trenta should definitely keep taking the subway rather than taxicabs.

If you underestimate your remaining life span, you risk saving too little. Check published life tables when planning for retirement.