Journal of Financial Planning: January 2012
Harold Evensky, CFP®, AIF®, is chairman of Evensky & Katz in Coral Gables, Florida. He is the author of The New Wealth Management and co-editor of Retirement Income Redesigned.
As the saying goes, variety is the spice of life, so for this contribution I decided to do something different. Normally I cover a broad spectrum of publications, but this time I thought I would focus solely on contributions to the Journal of Financial Planning. Over the past 12 months there were so many interesting articles, my problem was how to limit my selection to keep my comments within my editorial word count. After reviewing recent papers, I found a particularly rich vein of work addressing retirement income, a major planning concern for most practitioners, so I resolved my problem by concentrating my review on five papers addressing this subject. As you’ll see, each provides important insights on this critical issue, and I hope after reading my review you’ll take the time to read each one in its entirety.
“Portfolio Success Rates: Where to Draw the Line,” by Philip L. Cooley, Carl M. Hubbard, and Daniel T. Walz, Journal of Financial Planning, April 2011. If these authors sound familiar, they should—they, along with Bill Bengen,1 contributed the early seminal papers on the subject. In this update to their earlier work2 extending their study from January 1926 through December 2009, they reconfirm their earlier work and that of Bengen, concluding that “clients who plan to make annual inflation adjustments to withdrawals should plan lower3 withdrawal rates in the 4 percent to 5 percent range.”4 Although reconfirmation of past studies alone is useful, the paper also includes a series of tables framing portfolio “success rates” by withdrawal rate, portfolio composition, and payout period. The purpose of the tables is to assist practitioners in developing “adaptive” withdrawal plans, utilizing the information in the tables to revise withdrawal strategies as a client’s positions and market returns vary over time. Finally, one of the most valuable contributions of their work is an excellent literature review appendix that describes in some detail a dozen of the major prior studies of retirement income distribution.
“The Impact of Aging on Retirement Income Decision Making,” by Gregory W. Kasten and Michael Kasten, Journal of Financial Planning, July 2011. Although this paper, unlike the others in my review, does not address the mathematics of retirement income distribution, it introduces an important concept practitioners need to consider in addition to risk tolerance and risk capacity. The authors argue that we need to consider our client’s cognitive ability to “grasp and integrate the implications of the retirement income solution.” Their position is based on recent research that concludes “cognitive function declines dramatically over an investor’s life span, starting at age 20.”5 Noting that age-driven mental decline can be somewhat offset by experience, the authors suggest that the sweet spot for decision making is age 50. Subsequent to that age, analytic cognitive ability falls at a pace that trumps experience. This diminution does not imply that most investors beyond 50 will have clinical dementia or Alzheimer’s; however, they will progressively face cognitive impairment, especially as related to financial decisions. Given this conclusion the authors make two points.
First, recent brain scan research suggests that the brain “sets aside rationality when it gets the benefit of supposed ‘expert’ opinion,” and retirees are likely to be particularly susceptible to this uncritical trust of experts. As a consequence the authors conclude that these clients are best served by a fiduciary relationship.
Second, even for investors with limited cognitive declines, a critical aspect of retirement income management is effective communication. Toward this end the paper introduces an interesting graphic ranking various solution alternatives relating various forms of risk (inflation, market volatility, longevity, and standard of living) to a risk scale. They also describe a 12-topic retirement income report that might well serve as a practitioner’s model for a firm’s retirement planning policy.
“Safe Savings Rates: A New Approach to Retirement Planning over the Life Cycle,” by Wade D. Pfau, Journal of Financial Planning, May 2011. If the author of this paper and the next two seem similar, there’s a good reason. All three papers are by Wade Pfau. I am including all three as I believe Professor Pfau’s series introduces a unique approach to retirement income planning that deserves our attention. The basic premise of his work, introduced in this paper, is that “focusing on a ‘safe withdrawal rate’ and then deriving a ‘wealth accumulation target’ to achieve by the retirement date may not be the best way to approach retirement planning.” Although I do not necessarily agree that this is a totally accurate description of how many practitioners approach retirement planning, I do believe his conclusion that “the focus of retirement planning should be on the savings rate rather than the withdrawal rate” deserves serious consideration.
The important concepts highlighted in this paper include the sobering fact that not only is the minimum necessary savings rate (MSR) for investors extremely volatile (significantly dependent on the specific period), the required rate is surprisingly high. Based on the modest goal of replacing 50 percent of final salary, Pfau’s study suggests that the necessary fixed savings rate exceeded 20 percent almost 40 percent of the time.6 A second intuitive but often overlooked consideration addressed in the paper is that long periods of subpar performance are usually followed by long periods of significant performance, and significant performance is followed by subpar performance. He provides two powerful examples to highlight this interaction. Both assume an investor saves at a fixed 16.62 percent rate over a 30-year working career.
An unfortunate investor retiring in 1921 would have accumulated only 5.52 times his final salary, whereas a lucky investor retiring in 2000 would have accumulated 19.07 times his final salary.7 However, that’s only half the story. Although the unfortunate 1921 retiree, because of the relatively small accumulated nest egg, would need to withdraw at a 9.06 percent rate to maintain his targeted living standard, subsequent market performance was such that the minimum sustainable withdrawal rate (MWR)8 was 9.78 percent. Unfortunately for the 2000 retiree, Pfau suggests that she may be facing the opposite result—although she has accumulated significant assets, subsequent market returns may result in a sustainable withdrawal rate of less than the 4 percent to which practitioners tend to anchor. The point Pfau makes is that when investors achieve “traditional wealth targets earlier than expected, which may have caused people to either cut back on their savings or even retire early, while unbeknownst to them, post-retirement market conditions could have resulted in a lower-than-expected sustainable withdrawal rate.” Significant food for thought.
“Can We Predict the Sustainable Withdrawal Rate for New Retirees?” by Wade D. Pfau, Journal of Financial Planning, August 2011. This second paper follows the theme of the first, bringing into focus the importance of incorporating contemporaneous market metrics, concluding (persuasively) that “market valuations and yield measures at retirement may provide a tool to help predict how much retirees can sustainably withdraw from their portfolio.” Based on this conclusion Pfau warns that the estimated “MWR for the 2000 retiree is 2.7 percent, but further declines continue until 2008 when the estimated MWR reaches 1.5 percent. For 2010 retirees it rises back to only 1.8 percent.”
“Getting on Track for a Sustainable Retirement: A Reality Check on Savings and Work,” by Wade D. Pfau, Journal of Financial Planning, October 2011. The last of Professor Pfau’s trilogy focuses on the mid-career investor. Noting that because of the compounding of returns most of a portfolio’s growth tends to occur shortly before retirement, it is difficult to estimate accumulation success until just before actual retirement.9 Even for periods as short as five years, the performance of a 60/40 portfolio varies significantly. For example, from 1870–2010, the average compounded annual five-year rolling real return was 5.26 percent; however, specific periods ranged from –8.5 percent to +19 percent, and returns during 21 of the 135 periods were negative. Included in the paper is a fascinating table that quantifies the degree of predictability for retirement date wealth provided by varied asset allocations and years before retirement.
I hope I’ve stirred your curiosity enough to get you to read the full articles. I believe you’ll be well rewarded.
Endnotes
- Bengen, W. P. 1994. “Determining Withdrawal Rates Using Historical Data.” Journal of Financial Planning 7, 4 (October): 171–180.
- Cooley, P. L., C. M. Hubbard, and D. T. Walz. 1998. “Retirement Spending: Choosing a Suitable Withdrawal Rate.” Journal of the American Association of Individual Investors 20, 2 (February): 16–21.
- “Lower” as compared to their conclusion that at a 74 percent success level a portfolio of at least 50 percent large-company common stocks can sustain a 7 percent nominal withdrawal.
- For portfolios with at least a 50 percent equity allocation.
- Agarwal, S., et al. 2009. “The Age of Reason: Financial Decisions over the Life Cycle with Implications for Regulators.” Brookings Papers on Economic Activity (October 19).
- Based on a 30-year working career and based on 110 30-year periods between 1870 and 2010.
- Assumes a 60 percent equity/40 percent bond allocation rebalanced annually. Taxes and transaction cost are not considered.
- Bengen’s SAFEMAX rate.
- He notes that Basu and Drew identify this phenomenon as “the portfolio size effect” in their 2009 “Portfolio Size Effect in Retirement Accounts: What Does It Imply for Lifecycle Asset Allocation Funds?” Journal of Portfolio Management 35, 3 (Spring): 61–72.