Journal of Financial Planning: July 2012
Harold Evensky, CFP®, AIF®, is chairman of Evensky & Katz in Coral Gables, Florida. He is an internationally recognized speaker on investment and financial planning issues and is the author of Wealth Management and co-editor of The Investment Think Tank, Theory, Strategy, and Practice for Advisers.
As I’ve observed in prior columns, one of the major issues facing practitioners today is planning for withdrawals in an era of volatile, low-return markets. Consequently, my current review concentrates on this issue. Also, as it remains the main source for practitioner-oriented research on withdrawals, the majority of papers are from our own Journal of Financial Planning.
Sexauer, S., M. Peskin, and D. Cassidy. “Making Retirement Income Last a Lifetime.” Financial Analysts Journal (January/February). It’s always a pleasure to find a research article that has practical application, and this one fits the bill. Noting that the purchase of immediate annuities, although a theoretically optimal strategy for lifetime income, is unacceptable to most investors, the authors propose a decumulation benchmark. This benchmark is composed of a minimum risk portfolio for the first 20 years composed of laddered TIPS, designed to provide level payments in real terms. Beyond the 20th year, inflation protection is hard to achieve, so the balance of the benchmark investment is a nominal deferred life annuity. The first payment at the end of year 21 is set equal to the last payment of the TIPS portfolio. The authors suggest that this proposed benchmark does double duty: directly as an investment strategy, and as a baseline to compare other retirement income strategies.
Coopersmith, L., and A. Sumutka. 2011. “Tax-Efficient Retirement Withdrawal Planning Using a Linear Programming Model.” Journal of Financial Planning (September). Although there have been a number of excellent articles written regarding the importance of tax considerations in retirement planning, particularly by Reichenstein, Jennings, and Horan (any articles they’ve penned are worth reading), all too often the impact of taxes is not adequately addressed in planning. “Tax-Efficient Retirement” provides a number of interesting and counterintuitive examples of optimal withdrawal strategies when investors have funds in both taxable and tax-deferred accounts. The authors compare the results of the common rule for sequence withdrawals based on spending down taxable savings prior to withdrawal from tax-deferred accounts to an optimized tax-efficient strategy that provides for withdrawal from either taxable or tax-deferred accounts to generate the maximum total remaining account balance at the end of the retirement horizon. Noting that tax efficiency does not necessarily mean tax minimization, they find that a tax-efficient strategy trades off higher cumulative taxes early in retirement by depleting lower yielding tax-deferred accounts to provide greater wealth at the end of the retirement horizon along with lower taxes for heirs. They find that a tax-efficient strategy can significantly outperform the traditional strategy under conditions often encountered by practitioners, namely: taxable rate of return is greater than tax-deferred rate of return, initial taxable savings are greater than 10 percent of total retirement savings, and itemized deductions are greater than the standard deduction.
Just as the Journal of Financial Planning is the primary source for withdrawal-related research, Wade Pfau is unquestionably the major, and one of the most thoughtful, individual contributors on the subject. So it’s not too surprising that Pfau is the author/co-author of three of the papers I’ve included in this review.
Pfau, W. 2012. “Capital Market Expectations, Asset Allocation, and Safe Withdrawal Rates.” Journal of Financial Planning (January). In this first contribution, he notes that sustainable retirement withdrawal rates depend on capital market expectations, retirement duration, asset allocations, and acceptable failure probabilities. His goal in this paper is to provide a framework for practitioners on the impact these issues have in combination. This study shows that there are a wide range of asset allocations that are likely to provide equally optimal solutions, and that in many cases, a lower stock allocation may be justifiable.
One particularly handy table is titled “Sustainable Withdrawal Rate with 10% Failure Rate for 30-Year Retirement Horizon.” With an x-axis of “standard deviation of portfolio returns” and a y-axis of “portfolio real returns” and a series of plots representing sustainable withdrawal rates from 1 percent to 8 percent, the graphic offers a practitioner with specific market expectations a quick estimate of what real withdrawal his or her portfolio might support.
The next contribution by Pfau gives me an opportunity to introduce a new and valuable research source for practitioners, namely the Retirement Management Journal, a “ peer-reviewed publication serving the needs of financial advisers, practitioners, executives, and scholars who are interested in all topics relating to retirement-income planning and management.” The journal is a publication of the Retirement Industry Income Association (RIIA). Pfau was recently appointed director of curriculum for the RIIA-sponsored Retirement Management Analyst (RMA) Designation Program, the educational and ethics training program designed to prepare advisers in providing successful retirement income solutions for clients.1
Pfau, W. 2011. “Retirement Withdrawal Rates and Portfolio Success Rates: What Can the Historical Record Teach Us?” Retirement Management Journal (Fall). In this paper, Pfau makes a point that I’ve emphasized on many occasions: “Historical portfolio success rates calculated from U.S. data may present a misleading rosy picture.” His thoughtful analysis warns that during the period covered by key withdrawal rate studies, U.S. financial markets performed exceedingly well; rolling historical simulations make high stock allocations look more attractive than may be justified, and success rates based on historical projections don’t account for the reality that may be facing current retirees. He concludes with an observation that immediate annuities and TIPS may play an important role in mitigating low withdrawal rates.
Williams, D., and M. Finke. 2011. “Determining Optimal Withdrawal Rates: An Economic Approach.” Retirement Management Journal (Fall). This paper, winner of the publication’s “Academic Thought Leadership Award,” is another example of the quality articles in this new journal.2 In this study, the authors echo the academic criticism of the traditional practitioner approach to withdrawal planning of estimating safe withdrawal rates that focus on portfolio allocations that minimize shortfall risk. Theory suggests that optimal withdrawal planning would adjust consumption to reflect changing portfolio values and age; in practitioner terms, this would mean reducing withdrawals following bear markets, increasing payouts following bull markets, and reducing withdrawals as the client ages. According to theory, ignoring these adjustments and simply increasing withdrawals by the rate of inflation to hold real consumption constant results in a large opportunity cost in terms of foregone consumption.
The premise is that a client’s willingness to take portfolio risk before retirement is equivalent to a willingness to accept shortfall risk after retirement, and in order to maximize expected lifetime utility to retire he would consume more in early years and less in later years when the probability of being alive is lower. The conclusion is that optimal portfolio allocation should be based on a function of minimizing the variance of consumption as a function of the client’s unique risk tolerance rather than a fixed real withdrawal rate for all clients. Although I would look forward to debating some of these issues in the future, I believe the points raised by this paper are ones that should be considered by all practitioners.
Finke, M., W. Pfau, and D. Williams. 2012. “Spending Flexibility and Safe Withdrawal Rates.” Journal of Financial Planning (March). This is basically an extension of the prior paper. To quote the authors, “The failure to include a client’s willingness to adjust is an important shortfall of the shortfall literature. A common thread in the analysis is that all failures are counted the same, without regard to when the failure occurred or what percentage of the client’s stated aggregate spending goal was funded. Such an all-or-nothing approach to retirement simulation is inconsistent with the way trade-offs are framed in retirement.” The study concludes that the traditional 4 percent real withdrawal rate strategy may only be appropriate for risk-averse clients with moderate guaranteed income sources.
Pfau, W. 2012. “Withdrawal Rates, Savings Rates, and Valuation-Based Asset Allocation.” Journal of Financial Planning (April). The last of this medley of withdrawal-related papers is appropriately one by Pfau that incorporates another hot practitioner topic—tactical allocation. In the introduction to his paper, Pfau quotes from an interview with Bill Bengen. When Bengen was asked, “What do you see as the next step regarding research into safe withdrawal rates?” Bengen responded, “I think that was something … where you vary, now, the investment approach based on some criteria that have yet to be defined, whether it be value or something else.…” Responding to that suggestion, this paper investigates the role of valuation-based asset allocation with respect to accumulation and distribution. Although he provides a number of caveats, the author concludes, “This article provides favorable evidence based on the historical record for clients to obtain improved retirement planning outcomes (lower savings rates, higher withdrawal rates) using dynamic valuation-based asset allocation strategies.” Definitely food for serious thought.
Endnotes
- Full disclosure: I am a member of the RIIA, a member of the RMA Governance Board, and an instructor in the RMA program at Texas Tech.
- More full disclosure: Professor Finke and doctoral candidate Williams are not only my associates at Texas Tech, they’re both good friends.