Should you save enough to live to 100?

common-retirement-mistakesFirst, you were supposed to die at 85. Then 90. Now 95 and even 100 are common defaults when financial planners tell people how much to save for retirement.

Except that’s nuts.

In the U.S., the typical man at age 65 is expected to live another 18 years. The typical woman, about 20. Yet many financial planners contend we should save as if we’re all going to be centenarians.

In my new column for the Associated Press, why we need to save for the retirement we’re most likely to have.

Related Posts

  • Q&A: Keeping pace with retirement saving Dear Liz: My wife is distressed by your recent column about how many multiples of salary are needed to retire. She interpreted the column […]
  • Retirees’ top money regrets In a previous column, I detailed retirees' biggest lifestyle regrets, such as not traveling more before their health gave out and not […]
  • Can’t afford to save for retirement? Dear Liz: Is it reasonable for a 50-year-old single man helping with support of a teenage child and earning a steady $35,000 a year to […]
  • Tuesday’s need-to-know money news Today's top story: How to create a retirement 'paycheck'. Also in the news: Handy money rules of thumb for a quick financial checkup, how […]


  1. Robert Rosenwald says

    Your “Should you save enough to live to 100” column appeared in Sunday’s “Savannah Morning News” In it you cited a 35-year old wanting to replace 60% of her current $60,000 salary at age 65 would need about $1.2m at retirement age if he expects to live to age 85. You noted an assumed inflation rate of 3% and return-on-investment of 7%. If the idea was to keep this example simple, you missed the mark. I had to do several fairly involved calculations, involving compounding inflation and investment return to find out that at age 84 your example would still be above water but at 85 the example individual would be some $33K in the hole. And I’m not sure I got it “right” but only note that portion of your column did little to educate or enlighten what I take to be your target audience, a relatively unsophisticated investor.

    • Liz Weston says

      I’m not sure the average reader would want to have to go through the calculation, but I can walk you through it if you have a financial calculator. This is what they teach in the CFP training courses for calculating a lump sum that generates income for a future need.

      1. Establish the replacement rate. (Someone earning $60,000 a year wants to replace 60 percent of her income in retirement, or $36,000.)

      2. Inflate the sum to the amount needed the first year of retirement. (If she’s 35 and wants to retire at 65, inflate $36,000 by 3 percent for 30 years, or $87,381.)

      3. Figure out the lump sum needed at retirement to generate that payment for the number of years in retirement.

      a. Assume an interest rate that adjusts expected return for inflation. If your return is 7% and inflation is 3%, the real return figure is 3.8835% (1.07/1.03 minus 1 times 100).

      b. The calculation on an HP12C goes like this for a 20-year retirement: 20 N (number of years) 3.8835 I (real return) -87381 PMT (annual payment needed) 0 FV (for zero future value, because you’re assuming you’ll exhaust the fund), BEG (for amount needed at the beginning of the period), solve for PV (present value) = $1,246,479

      c. The same calculation for a 35 year retirement = $1,721,398.

      4. The difference between the two sums is $474,919, so the sum required for a 35 year retirement is 38% greater than what’s needed for a 20 year retirement.

      In the real world, planners would want to do a Monte Carlo analysis to account for the fact that returns won’t always be 7% and inflation won’t always be 3%. And they’d shoot for a high probability of success to improve the odds you’ll still have money when you die. But this calculation is a simple way to show how much more you’ll need as the length of retirement increases.