Longer Retirements Don?t Cost Much More

Beyond Buy-and-Hold #88

I?ve seen a number of articles in recent years arguing that middle-class people need to continue working in retirement. The argument that is often made is that we live longer than we did in the days when age 65 became the conventional retirement age. If we are going to live longer, we cannot expect our retirements to survive until death unless we are willing to work through retirement, the thinking goes.

No. The logic here does not follow.

It IS true that adding some income from work to your retirement plan in the early years makes a huge difference. So adopting a plan to work in the early years of retirement makes a lot of sense for a lot of people. If you don?t have enough saved by the time you want to retire, the best way to cover the shortfall is to agree to work during the first ten years of retirement or so.

Why saving more doesn?t necessarily make retirement assets last longer

The point that I am objecting to is the idea that you need to add a lot of cash to your retirement plan if you expect it to last not until age 75 but to age 85 or 90 or 95.

There?s a counter-intuitive dynamic at work here. Because of the compounding effect, some stages of retirement matter a great deal more than others. You don?t want to suffer losses in the early years. That?s why it is so important to take valuations into account when putting together a retirement plan. Price crashes always come at times of high prices and a price crash experienced in the first ten years has a devastating effect.

Conversely, reducing the amount you need to withdraw during the first ten years by taking on a part-time job has a very big positive effect.

The first 10 years of retirement are the key

However, stretching the retirement out another 10 or 15 years costs a lot less than you would expect. Once you get past that 10 year hurdle during which you need to avoid crashes, you are pretty much on your way.

I?ll enter some numbers into the Retirement Risk Evaluator and the Investment Strategy Tester to try to give you a sense of how these two factors influence your chances of enjoying a successful retirement.

The P/E10 level today is 23. Say that you were turning 65 next week, had a portfolio value of $1 million, a stock allocation of 60 percent, were confident that you could earn 2 percent real in assets not invested in stocks and wanted to preserve at least 50 percent of your portfolio value at death to pass it on to heirs or charities. Say also that you need $50,000 to live on, a 5 percent withdrawal. The Risk Evaluator tells us your odds of meeting your goals.

This retirement plan is doomed!

The Risk Evaluator tells us that a withdrawal rate of 4.7 percent has only a 20 percent chance of surviving 30 years (which would bring you to age 95). This plan will not fly. Back to the drawing board!

A post-retirement part-time job to the rescue

Now let?s retain all the old assumptions except for the annual withdrawal, assuming that you will work at a part-time job paying $800 per month ($10,000 per year) for the first 10 years of retirement. The Risk Evaluator does not permit me to make changes in the withdrawal assumption during the course of the retirement, so I plugged these numbers into the Strategy Tester.

It tells me that the most likely portfolio value at the end of 30 years for the plan calling for a $50,000 withdrawal in each year is $542,673. That means that there is almost a 50 percent chance that the goal of having the portfolio retain 50 percent of its real value at death will not be met.

But change the withdrawal number to $40,000 only for the first ten years (leaving it at $50,000 for the remaining years) and the most likely 30-year value jumps to $872,904. In this scenario, there is only a 20 percent chance that the Year 30 value of the portfolio will be less than $631,157. Success!

The long-term effect of less in withdrawals in the first 10 years

Please note that by adding $100,000 in contributions to the plan ($10,000 per year for each of the first 10 years of retirement), you increased the most likely Year 30 value by more than $300,000. The first ten years of retirement are the years that matter most!

Say that you were to die at age 85 rather than at age 95? Would that make much of a difference?

It would not. The most likely Year 30 value for the portfolio in which you withdrawal $50,000 in every year is $614,741 at the end of 20 years, only about $70,000 higher than the number that applies at the end of 30 years. The most likely Year 30 value for the portfolio in which you withdrawal $40,000 for three first 10 years and $50,000 for the remaining years is $798,355, only about $75,000 more than the number that applies in Year 30.

Planning to work during retirement to lower the size of the withdrawals you make in the early years makes sense. Planning to work during retirement to cover the possibility that you will live longer than you expect does not have as much of an impact as you expect.

The first ten years of retirement are the critical years. Stretching out a retirement that does well for the first ten years is not too difficult.

Rob Bennett has written about what really matters most to the success of a money management strategy. His bio is here. For background on the Big Fail of Buy-and-Hold and on the need to move to Valuation-Informed Indexing, please check out the ?About? page at the ?A Rich Life? blog.

( Photo from Flickr by Helico )

2 Responses to Longer Retirements Don?t Cost Much More

  1. Sure that makes sense, but it seems when you are older and need more health care, the expense will shoot way up. So the last few years of retirement matters a lot too if you have chronic conditions. Right?

  2. Your point makes perfect sense, Joe.

    I’m afraid it’s a little above my pay grade to figure out what those health expenses are going to be.


Leave a reply