Journal of Financial Planning: October 2011
Jonathan Guyton, CFP®, is principal of Cornerstone Wealth Advisors Inc., a holistic financial planning and wealth management firm in Edina, Minnesota. He is a researcher, mentor, author, and frequent national speaker on retirement planning and asset distribution strategies, and a former winner of the Journal of Financial Planning’s Call for Papers competition.
One of my great professional privileges is speaking to groups of financial planners around the country. Most often, the topic on which I’ve been invited concerns the generation of a sustainable income for clients in retirement.
Without exception, it’s clear that the planners who attend—whether I get to meet them beforehand or afterward—are thoughtful and serious professionals who care deeply for the well-being of their clients. They are curious, willing to challenge their beliefs, and committed to furthering their professional knowledge and skills.
Our interactions often pique my own curiosity: How do these colleagues interpret the various writings, research, and commentaries about retirement income generation? What do they believe—and on what basis? How do they communicate these beliefs to clients? How do they actually make the dozens of decisions practitioners in this area of planning are called upon to make? How do they process and implement changes in their practices when presented with differing viewpoints? What makes something a “best practice” in their eyes?
Since I only gain insights into these matters through one-on-one communication, rarely is there a sufficient sample size to draw any meaningful observations on any of them.
Rarely. But not never.
For the past several years, this Journal has published a special report devoted to retirement income planning.1 One of its regular components is a survey querying practitioners on various aspects of this subject; some questions lend themselves to ongoing tracking so that emerging trends may be more easily identified. As a committee member advising the Journal’s research staff on the makeup of this year’s survey, I can attest to their commitment to generating worthwhile insights.
Last year, 425 FPA members participated in the survey, and their collective responses were part of the report published last December. It reveals an intriguing portrait of how a pretty fair group of our fellow practitioners views this vital subject, and offers us, as professionals, a valuable reflection of ourselves. Several aspects of this portrait caught my eye and are the subject of this column. (It will be helpful to keep in mind that the survey was conducted in September 2010, only about 18 months after the darkest days of the financial crisis. The S&P 500 stood at about 1,150 and had recovered about 60 percent of its losses dating back to its 2007 high point.)
Retirement Planning Success and Withdrawal Rates
For starters, there is very good news about how these planners’ clients fared who were near or at the start of retirement: only 17 percent planned to delay their retirement date because of the recent financial and economic turmoil. By comparison, a Towers Watson survey showed that 40 percent of U.S workers planned to delay their retirement;2 similar research from Sun Life Financial showed that 52 percent expected to work at least three years longer than they had previously planned.3
In other words, these near-retirement clients were 57–67 percent less likely than the population at large to have had their retirement plan so knocked for a loop that they failed to achieve one of their primary goals on time. Impressive stuff. Hopefully, the same (or better) can be said of their colleagues nationwide!
In a separate question, these planners were asked what they “usually recommend as a sustainable withdrawal rate.” Their average response was 4.75 percent; one year earlier, it was 4.4 percent. This is either enlightening or discouraging … or both.
Research4 as well as informal and formal5 observation shows that an initial withdrawal rate sustainable for 30 or 40 years can fluctuate significantly throughout a distribution period, even when the real withdrawal dollar amount is held constant. A given year’s safe withdrawal rate is (temporarily) lower in periods of higher equity market valuations and (temporarily) higher when lower valuations prevail.
In mid-September 2009, the S&P 500 was about 1,040—about 10 percent lower than a year later. Thus, it would be discouraging to learn that our profession’s practice is to recommend higher withdrawal rates when markets are more highly valued and the opposite when they are not. It flies directly in the face of the evidence.
Unless … there is something else going on. Another possibility is that, mindful of the inverse relationship between the safe withdrawal rate and market valuation levels, planners’ recommendations are beginning to include the higher safe withdrawal rates that exist when clients are willing to adopt a flexible, policy-based distribution strategy based on research over the past decade rather than basing this advice on the maintenance of a static real withdrawal amount.6 In short, this research demonstrates that safe withdrawal rates are 0.5–1.0 percent higher when clients are willing to reduce their real gross withdrawal income by 5–8 percent following years that generate extremely low market valuation levels. Should this be the case, it would be a significant development for retirement income planning advice.
Income Distribution Strategies
The survey also asked planners to choose the income distribution strategy they most frequently employ. Ninety-two percent of respondents picked one of the following three approaches as their primary strategy: Structured Systematic Withdrawals (50 percent, up from 40 percent in 2009)—a total return, policy-based approach that draws a percentage of the portfolio periodically for income; Time-Based Segmentation (28 percent, down from 32 percent in 2009)—establishes separate pools of investments with the lowest-risk pool drawn down for income in the nearest-term time horizon with the higher-risk pools then drawn down for income in later periods; Essential-Versus-Discretionary Income (14 percent, unchanged from 2009)—low-risk investments or annuity guarantees fund essential expenses and moderate/higher-risk investments fund discretionary expenses. Several points relating to these three approaches strike me as noteworthy.
The first concerns the Essential-Versus-Discretionary approach, typically preferred because it insulates the funding of essential lifestyle expenses via very low-risk or annuity guaranteed products and applies a more growth-oriented asset allocation with the assets assigned to fund “discretionary” lifestyle components. As noted earlier, only 17 percent of planner clients in or near retirement had to significantly alter their lifestyle following the financial crisis. However, an average of 25 percent of clients whose planners primarily employ the Essential-Versus-Discretionary approach suffered a “significant” lifestyle adjustment—the highest percentage of any strategy and a correlation that proved statistically significant. How could this be?
It is likely that the assets earmarked for essential expenses in fact generated their income as anticipated without any unpleasant surprises. However, these assets require a larger initial investment to generate a given income than the other approaches. A client with $1 million seeking to fund $30,000 of inflation-adjusted essential expenses may need to put $800,000 or more in such low-risk assets, whereas even in a highly valued market, a static safe withdrawal strategy of 4.5 percent would require only $667,000, and a dynamic rate of 5.5 percent would claim just $550,000. If the remaining $200,000 were invested “more aggressively, for growth” as a counter-balance, this discretionary fund may well have declined 35 percent or more to less than $130,000. Experiencing this situation in 2008 and 2009, it is easy to imagine some retirees at least partially delaying or cancelling planned expenses that were to be funded with these assets.
Unfortunately, retirees are rarely ambivalent about their planned “discretionary” expenses. In fact, some of these may be essential to their retirement being a fulfilling time of life, even if not essential for basic food, clothing, or shelter. Income from these assets funds activities that actually help to define their quality of life—at least for that year. Even a partial delay is a “significant” adjustment. With a smaller share of assets available for such activities to begin with, the reduced discretionary fund balance early in 2010 may have been even more alarming. Sadly, an unintended consequence of this approach suggests itself: be they clients or practitioners, some of those who most desired to shelter themselves from the inevitable economic storms appear to have been the most vulnerable to their fury.
The Time-Based Segmentation (or bucket) approach, used primarily by about 30 percent of planners, sounds highly rational and feels quite secure—at least initially. After all, knowing that the first three or five years of portfolio withdrawals are safely tucked away in the most stable of fixed-income assets—with funds that will provide the income needed further down the road invested in more growth-oriented holdings—is surely comforting.
This, however, begs the question, “What happens next?” Depending on the answer, the bucket approach is then revealed as a thoughtful policy-based methodology (which places it neatly in the Structured Systematic Withdrawals category), or a rigid formula at risk of breaking in times when flexibility is required, or a well-intentioned but vacuous strategy about which Gertrude Stein might have said, “There is no there there.”
Let me explain. Various articles have explored this concept, including actual empirical research that has attempted to quantify the sustainable withdrawals it can make available to retirees. There is no result that ascribes an increased safe withdrawal rate to its use; the sum of its parts (buckets) is no greater than its whole. In other words, a series of buckets that in total puts 40 percent of assets in fixed income generates the same outcome as a balanced portfolio with a 60/40 allocation. This is by no means a criticism of the bucket approach; however, any contention that it can produce higher sustainable withdrawals because of its structure is not borne out by the evidence.
Still, practitioners using this approach face the same implementation issues as under the Structured Systematic Withdrawal method: What is the source of this year’s withdrawals? How will interest and dividend payments be handled? How do I rebalance the overall portfolio in light of current conditions? How do I handle a sustained equity bear/bull market? What, if anything, would trigger something other than an inflation-based withdrawal increase from one year to the next? Policy-based questions, all of them.
Descriptions of the bucket approach frequently treat these matters almost as though they were afterthoughts, when they often determine the very sustainability of a retiree’s portfolio income.
The Bucket and the Cushion
My concern is that placing three or five years of expenses in highly stable, short-term securities offers a false sense of security—or worse. Undoubtedly, a “cushion” of this size (or length) is not only comforting, it is also sufficient to outlast most equity bear markets—most, but not all. The reason portfolios with 60–70 percent equities consistently produce the maximum safe withdrawal rates at 99–100 percent success rates is because they contain—and need to contain—a big enough cushion to get through a far longer part of the distribution period when selling equities to fund withdrawals may be off the table. A balanced portfolio with 35 percent in short-term and U.S. government bonds can use these assets to sustain a 5 percent withdrawal rate for 8–10 years depending on its overall yield. Isolating three or five years of safe money may not be enough; thus, maintaining a safe income bucket of that size does not offer a bullet-proof sense of security absent a thoughtful integration with the other portfolio assets. If it did, portfolios with 80 percent equities would have higher safe withdrawal rates at a given confidence level than those with 60–70 percent. They do not.
A larger potential problem comes if the buckets are established under a set-it-and-forget-it approach in which they are left to be drawn down in a pre-determined sequence. What happens when an extended equity bear market ensues just as the safe bucket for the initial years is largely exhausted? Of course, a thoughtful, policy-based rebalancing method would easily mitigate this risk. However, a literal and sequential drawdown of the buckets will periodically expose clients to the very market risks and consequences they would obviously rather avoid.
Look for results of this year’s survey in the December issue of the Journal, and if next year’s survey shows up in your inbox, I urge you to take 20 minutes to share your perspective. I am grateful to colleagues—last year, this year, and in the future—who take the time to give us this revealing glimpse of how they address some of the most important and challenging issues we face in providing clients with sound retirement income planning advice. And I am equally encouraged by the advancements we are making in this area, even when flaws are revealed in our practice or knowledge. This is how we grow as a profession.
Endnotes
- FPA’s “2010 Retirement Income Planning Special Report.” Journal of Financial Planning 23 (December): 28–30.
- Towers Watson study of 9,100 employees conducted in May and June 2010: www.towerswatson.com/press/2919.
- Sun Life Financial Inc.’s Unretirement Index. October 19, 2010. www.sunlife.com/us.
- Kitces, Michael. 2008. “Resolving the Paradox—Is the Safe Withdrawal Rate Sometimes Too Safe?” The Kitces Report (May): 1–13.
- Guyton, Jonathan T. 2004. “Decision Rules and Portfolio Management for Retirees: Is the Safe Initial Withdrawal Rate Too Safe?” Journal of Financial Planning 17 (October): 56.
- Bengen, William P. 2006. “Baking a Withdrawal Plan Layer Cake for Your Retirement Clients.” Journal of Financial Planning 19 (August): pages referring to Figure
- Guyton, Jonathan T., and William J. Klinger. 2006. “Decision Rules and Maximum Initial Withdrawal Rates.” Journal of Financial Planning 19 (March): 53–57.