Monthly Notes, Views, & The Occasional Provocation February 2007 

This month we take issue with a recent New York Times article that suggests, provocatively, that Americans might be saving too much for retirement. Never mind that the national savings rate has arguably turned negative. The problem, according to a “small band of economists,” is that the large financial institutions have overstated the need for retirement savings in order to further their own money-management interests.

While I typically applaud efforts to highlight the conflicts of interest riddling the financial-services industry, I believe the economists may have overstated their case here. In particular, I believe they place too much faith in their research and economic models, falling prey to the trap of "precision without accuracy" that can plague any measurement problem, but particularly the complex problems inherent in long-term financial plans.

Please feel free to forward the newsletter to anyone whom you think might enjoy reading it. Also, we always love questions or comments, so please send them along.

Warm regards,

Milo Benningfield

Whether To Squander Your Youth Or Retirement
Last month The New York Times published an article with the tantalizing title “A Contrarian View: Save Less and Still Retire With Enough” that posed the equally tantalizing question, “Could it be possible that you are saving too much for retirement?” (NYT, January 27, 2007) more

Whether To Squander Your Youth Or Retirement
Last month The New York Times published an article with the tantalizing title “A Contrarian View: Save Less and Still Retire With Enough” that posed the equally tantalizing question, “Could it be possible that you are saving too much for retirement?” (NYT, January 27, 2007)

The article cites a

small band of economists from universities, research institutions and the government [who] are clearly expressing the blasphemy that many Americans could be saving less than they are being told to by the financial services industry – and spending more – while they are younger.

More specifically, the economists contend that the “ostensibly objective online calculators” of Fidelity, Vanguard, TIAA-CREF and other large financial institutions overstate the amount of money people need to save, in some cases by almost twice as much as necessary. “’There is risk in saving too much,’” Laurence J. Kotlikoff, a Boston University economics professor, said. “’You could end up squandering your youth rather than your money.’”

Point well taken. Deferring gratification now could mean deferring it forever should divorce, disability or death intervene, so there is, indeed, a risk involved in foregoing consumption now.

But there is an important counterpoint not mentioned in the article: the dollars saved in one’s youth are much more valuable than the dollars saved later, making even a few years’ delay in saving a risky proposition.

For example, consider a young engineer, age 25, who wishes to accumulate $3 million in today’s dollars by the time she is 65. If she starts saving today and earns an annualized return of 8% on her investments, she can meet her goal by setting aside less than $12,000 a year between now and retirement. But if she waits five years to begin saving, she’ll need to set aside over $17,000 a year; wait five more years and she’ll need to save over $26,000 a year, and so on, as illustrated in the chart below:

*Assumes 8% annualized investment return.

As the gentle curve on the left of the graph suggests, there is some latitude for squandering a portion of one’s youth, since the extraordinary effect of compounding investment returns can help us make up for “lost time” later. But at some point, depending on our actual investment returns, the required annual savings takes a sharp turn for the worse and can quickly soar to unattainable heights.

In addition to faulting the article for glossing over the risks involved in waiting to save, I believe Professor Kotlikoff places too much faith in his software, falling prey to the precision-without-accuracy problem that plagues many forms of measurement. As an example, your watch may allow you to read the time right down to the second – a very precise measurement – but if it’s slow ten minutes, you’ll still miss the train – an inaccurate result.

In Professor Kotlikoff’s case, he claims to have developed a better method than the online retirement calculators of estimating the target spending need during retirement. On its face, this does not seem too difficult a task, since the online calculators are designed for ease and speed of use by consumers with little or no financial training. But it is doubtful that Professor Kotlikoff’s more refined estimates will provide any more accuracy for one simple reason: the uncertainty of future investment returns.

No matter how accurate our planning assumptions are about current and future spending needs, tax rates, and a host of other variables, they will all mean very little if our assumed investment return varies even slightly from our actual experience.

For example, consider our engineer again. At age 35, she calculates that with an 8% annualized investment return she can save $3 million by age 65 if she sets aside about $26,500 each year. She has chosen $3 million as her goal, because she knows that amount will allow her to prudently withdraw $120,000 from the portfolio in her first year of retirement – 4% of its value – and then adjust upwards as necessary during retirement to keep up with inflation.

She envisions needing $120,000 in today’s dollars for core living expenses not covered by Social Security benefits and to provide her resources for travel, a sabbatical abroad, doting on her future grandkids and possibly even helping to pay for their educations, as well as reserves for those two wildcards, future tax rates and healthcare costs. She has, in short, a lot at stake in reaching her specific savings goal.

Unfortunately, even if she exhibits impeccable savings habits over the years, should her annualized investment return fall short, even a little, of 8%, she will fail to meet that goal. In fact, should the capital markets disappoint, she could be looking at having as little as $84,000 a year to spend in retirement, forcing her to make some hard choices among many competing needs.

Effect of Annualized Investment Return on Long-Term Savings
Annualized Return Savings at Retirement Annual Withdrawal Amount*
8.0% $3,000,000 $120,000
7.5% $2,700,000 $109,000
7.0% $2,500,000 $100,000
6.5% $2,300,000 $91,000
6.0% $2,100,000 $84,000

*Assumes a 4% sustainable withdrawal rate.

The lesson is not that saving to invest is worthless. Rather, given all the unknowns that await us, setting aside as much as we reasonably can as soon as we can is a compelling idea. Indeed, our ability to save and invest is one of the few variables, among the many that determine our financial success, over which we have any control. To squander one’s youth, of course, is a terrible thing. But to squander our future may be even worse.

Thank you for reading. Until next month.

back to top

Copyright 2007 - Benningfield Financial Advisors