Last week, The Wall Street Journal published an article by Eleanor Laise titled “How to Cash out in Retirement”. The sub-title was “A look at four strategies that could help make a retiree’s savings last a lifetime.”
How many of those strategies for making “savings last a lifetime” do you think involved the use of mortality credits or any annuity products? Hint: The answer is the same number of times I have been on a beach and mistaken for Matthew McConaughey.
In Ms. Laise’s 1,800 word article, she reviewed:
- the 4% withdrawal rule;
- making retirement spending a function of performance; and
- bond ladders.
The fourth “strategy” is a reminder to “remember tax efficiency”. Good advice, though curiously the tax efficient aspects of immediate annuities were not discussed. Her only reference to annuities was a parenthetical at the end of the 6th paragraph; “[a]nnuities, of course, may still be a good retirement-income solution for some people.”
Before I illustrate that very point, allow me to make an observation:
Jay ascends soapbox
Mortality credits are a valuable, under-utilized asset class, particularly for retirees who do not have huge retirement portfolios or other sources of guaranteed income, like pensions. If The Wall Street Journal (!) does an article on strategies for helping retirees make their “savings last a lifetime” and mortality-contingent products are not included in the article, one has to ask if the industry has done an adequate job educating the investing public (including advisors and reporters) on the virtues of morality credits.
Jay descends soapbox
The “4% rule”
I will limit this post to addressing the “4% rule”, since it is so frequently cited. As generally applied, the retiree begins withdrawing 4% of his or her portfolio and increases the withdrawals each year to keep up with inflation. The scenario she describes is a retiree with a $1 million nest egg (60%/40% stocks/bonds), and an income need of $40,000 per year. The author put this through the T. Rowe Price Retirement Income Calculator and determined the investor had a 10% change of running out of money at age 97 (when I entered the author’s assumptions, my results repeatedly showed an 11% chance of running out of money by age 95, I assumed retirement at 65, perhaps she used a later age). Disclosure: My wife has an IRA at T. Rowe Price.
A roughly 10% chance of running out of money by age 95 doesn’t sound so bad, unless your 95 and have run out of money. For a couple retiring at 65 today, there is about a 50/50 chance that at least one of them will live to age 95. That’s a lot of broke old people. To be fair, Ms. Laise, does not depict the “4% rule” as the “4% solution”. Indeed, she quotes economist Laurence Kotlikoff that “[t]his (rule) is a prescription for getting people into serious trouble.”
The “4% rule” vs. immediate, life-contingent annuities
Suppose the same 65 year-old (I like giving my hypothetical people a name, let’s call him “Bruce”), instead purchased a life-contingent immediate annuity at age 65 to cover his $40,000 in estimated expenses at retirement. According to www.immediateannuities.com (the source for all annuity payment assumptions relied on in this post), it would cost Bruce $525,312. Let’s assume Bruce is very risk averse and invests the balance in 90 day T-Bills. Here’s Bruce’s strategy
- For years 1-4 after retirement, Bruce draws down on that nest egg as his expenses increase beyond $40,000 due to inflation.
- In year 5 after retirement, Bruce purchases another life-contingent annuity in an amount equal to his expected expenses that his current annuity payments do not cover.
- Bruce repeats steps 1 and 2 until he goes he rambles on to the Thunder Road in the sky.
How does Bruce fair?
The answer of course depends on Bruce’s real, risk-free return – how much does Bruce earn (if anything) in his T-Bill account after accounting for inflation.
Considering the different inflation scenarios highlights a critical limitation in the T. Rowe Price model used in the “4% rule”: it assumes a flat 3% inflation annually.
I looked at the period 1928 – 2010 and took inflation numbers from the Bureau of Labor Statistics and T-Bill returns from the St. Louis Fed, to come out with real, risk-free returns. From there I ran 4 inflation/risk-free return scenarios:
- Average real, risk free return from 1928 – 2010
- Actual real, risk free return numbers from the last 35 years
- The average real, risk free returns for the worst 35-year period (from 1934 to 1968, the average annual real, risk-free rate of return: -1.24%)
- The actual real, risk free returns each year for the worst 35-year period (1934 – 1968).
Here are the results:
|Scenario||Assets at 101st birthday|
|Last 35 years||$130,000|
|Average of worst 35 years (1934 – 1968)||$85,000|
|Actual time series for worst 35 years||$0 (Bruce went broke between age 97 and 98.|
For Bruce to run out of money in his 100th year, his 35 year average annual real, risk-free rate of return would have to be about -2%; substantially worse than the average for the worst 35-year time period. You will note that Bruce would go broke in Scenario 4 at age 97. That is because for the actual time series, the bad returns were frontloaded in the early years; for the 15-year period from 1934 to 1948, the average real, risk-free return the first 15 years was about -4% per year.
The point isn’t that an immediate annuity ladder is the solution for all, or even most retirees. It isn’t. However illustrations such as this one need to make their way into the general business press; not as a specific strategy recommendation, but as a way to help investors understand the value of mortality credits and how they might be employed, to some extent, in their retirement portfolios.