The Only Withdrawal Plan You Will Ever Need

Transcription

1 The Only Withdrawal Plan You Will Ever Need November 14, 2016 by Ken Steiner Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives. It seems like every other week we read here at Advisor Perspectives or elsewhere in the retirement planning media about the latest and greatest strategic withdrawal plan (SWP) to use to tap one s savings in retirement. The new and improved strategy may be fixed, variable or a hybrid of the two. It may use a safe withdrawal rate. It may have guardrails, floors or ceilings. It may be a variation of the IRS required minimum distribution (RMD) rules. It may involve using the Excel payment (PMT) function. It may calculate a rate that retirees should feel free withdrawing, or it may be one of the many approaches that adjusts the 4% rule in some manner to supposedly make it better. I call these approaches rule of thumb (RoT) approaches. All of these RoT approaches miss an important point. Your clients want to know approximately how much they can afford to spend each year and meet their financial objectives, not how much they can withdraw from their investment portfolio. There is only one SWP that financial advisors (FAs) should be using for their clients: If the client chooses to spend his or her spending budget for the year, the amount to be withdrawn from savings for the year should be equal to: 1. The calculated sustainable spending budget (SSB) amount for the year, minus 2. The total amount of income from other sources, expected to be received during the year (IFoS). If IFoS is expected to be relatively constant throughout retirement in real dollar terms, it is possible that the sum of IFoS and the withdrawals based on one of the RoT approaches may produce a reasonable SSB for a retiree. An example of this situation could include a retiree whose only other source of retirement income is from Social Security (and the retiree has not chosen to defer commencement of such a benefit). In situations where the retiree has sources of income that may be front-loaded, deferred or temporary (i.e., not expected to be relatively constant in real dollars over the entire retirement planning period (RPP)), the sum of the withdrawals based on one of the RoT approaches plus IFoS for a specific year may not result in a reasonable SSB. An example is provided below. Page 1, 2018 Advisor Perspectives, Inc. All rights reserved.

2 It is my understanding that most FAs already use the SSB minus IFoS approach to determine possible withdrawals for a year. This was a revelation to me. Of course this begs the question of why we see so much literature describing the next and best RoT approach, unless these approaches are primarily either for research purposes (where relatively constant real dollar IFoS is assumed) or for "media purposes." In order to avoid misleading those FAs and DIY retirees who may believe that the best way to develop a reasonable SSB is to add IFoS to the withdrawals determined under one of these RoT approaches, we must make it clear that they have assumed relatively constant real dollar IFoS, and their approach may not produce a reasonable spending budget if IFoS is not expected to be relatively constant in real dollars over the retiree s RPP. Developing a reasonable and sustainable spending plan and an annual SSB As noted above, the key to determining how much should be withdrawn from accumulated savings each year lies in developing a reasonable and sustainable spending plan that provides a reasonable SSB each year. In order to meet the client s financial goals, the tool and process used to develop a reasonable and sustainable spending plan should have the following characteristics: It should consider all the retiree s: assets, including the future income expected to be received from various sources, and liabilities (future expected expenses), including future non-recurring expenses, such as unexpected expenses, long-term care expenses and bequest motives, as well as future expected recurring expenses. It should permit the FA or client to make different assumptions with respect to various types of future expenses. For example, it may be reasonable to assume that future medical expenses will increase at a faster rate in the future than discretionary expenses. It should permit the FA or client to make different assumptions about the future, with respect to investment returns, inflation and RPP. If all assumptions about the future are exactly realized and the client spends exactly her SSB each year: Future expected SSBs should increase at the desired rate selected by the client The process should show the expected decumulation of accumulated savings over the client s RPP, and The client s projected accumulated savings at the end of the selected RPP should equal the amount selected by the client for a bequest motive. If all assumptions about the future are not realized and actual future spending does not exactly match the SSBs (by far the more normal situation), future SSBs should be adjusted to keep the retiree on track to meet financial goals. This may involve some smoothing of SSBs (and/or smoothing of actual spending) from year to year. To help formulate investment strategies and assess risk tolerance, it should facilitate modeling the impact on future SSBs of actual investment and/or spending that differs from assumptions made. Page 2, 2018 Advisor Perspectives, Inc. All rights reserved.

3 In my September 7, 2015 Advisor Perspective article, Think Like an Actuary to Become a Better Advisor, I encouraged FAs to check out our website, which contains spreadsheets that can be used as a tool to develop a reasonable spending plan and SSB. Our spreadsheets are reasonably transparent and involve the use of basic math to match one s assets (including the PV of income from all sources) with the PV of future expected non-recurring and recurring expenses, in a manner similar to that used by pension actuaries to determine a defined benefit pension plan sponsor s annual contribution requirements. As noted in the previous article, the spreadsheets are available for free, and we receive no compensation from visits to the website. As is generally accepted practice with pension actuaries, the matching of a retiree s assets and liabilities in our spreadsheets is accomplished by employing deterministic assumptions about the future. This practice appears to be an anathema to FAs, many of whom believe the use of deterministic assumptions to be an inferior approach, and who also believe it is important to be able to tell their clients that based on the 10,000 (or more) scenarios examined in their Monte Carlo analysis, the client has a 93.7% probability of meeting spending objectives with a given investment and spending strategy. Suffice to say that I am not as impressed with the supposed benefits of Monte Carlo modeling as most FAs when it comes to retirement planning (perhaps a topic for a later article). Even if you are a FA with access to sophisticated planning tools that satisfy all the requirements outlined above, you may still wish to try out the actuarial budget calculator (ABC) contained in our website. At a minimum, it will either confirm the results of your tools, or it will provide you an additional data point you can use to help your clients. The ABC spreadsheet has been recently modified to enable FAs and their clients who are not yet retired to develop spending/savings budgets using the same basic actuarial principles. Development of Joe s 2017 spending budget and related withdrawal from savings This section provides a simple example of how the ABC can be used to develop a SSB and expected withdrawals from accumulated savings for a hypothetical retiree whose IFoS is not expected to remain constant in real dollar terms throughout his expected RPP. Let s assume Joe is age 65 at the beginning of 2017 and is unmarried. He has current accumulated savings of $250,000 and is receiving an annual Social Security benefit of $20,000 and an annual fixed dollar life annuity pension of $12,000. He is also working in part-time employment earning $15,000 per year. He owns a home, but to simplify the example, we are going to assume that the PV of the equity he can pull out of his home in the future is equal to the PV of his future expected long-term care expenses plus other unexpected expenses. With the assistance of his FA, Joe has selected the following assumptions about the future: Joe s assets will earn 4% per annum and this rate will be used to discount his future spending liabilities, Inflation will be 2% per annum, and Joe s RPP at his current age (65) is 30 years. Page 3, 2018 Advisor Perspectives, Inc. All rights reserved.

4 Joe and his FA have analyzed Joe s expected recurring expenses (essential health, essential nonhealth and non-essential expenses) and have concluded that in total they will increase in the future by inflation minus 0.5% (1.5%) per annum. Joe has also expressed a desire to work only five more years in part-time employment, and he expects his part-time employment income to remain at $15,000 in the future. Finally, he wants to have $200,000 (in nominal dollars) remaining at the end of his RPP. Under the selected assumptions and Joe s data, the present value (PV) as of the beginning of 2017 of Joe s assets (ignoring home equity) is $994,437. This amount is the sum of: his current assets ($250,000), the PV of his Social Security benefits ($459,184), the PV of his life annuity pension benefits ($215,805) and the PV of his expected part-time employment income ($69,448). From this total PV of Joe s assets ($994,437), Joe s FA subtracts the PV of Joe s bequest motive ($61,664), to get the PV of Joe s future SSB s ($932,773). Joe s 2017 spending budget ($43,280) is calculated by dividing the PV of Joe s future SSB s ($932,773) by the PV of a period-certain annuity of $1 per annum payable for 30 years and increasing at a rate of 1.5% per annum ($ ). These calculations are all shown in the Present Value Calcs tab of the ABC spreadsheet. If Joe decides to spend $43,280 during 2017, the amount he will withdraw from his accumulated savings during the year will be ($3,720). Yes, this is a negative amount that is determined by subtracting his expected IFoS of $47,000 (consisting of $20,000 of annual Social Security benefit, $12,000 of annual fixed dollar life annuity pension and $15,000 from part-time employment) from his spending budget of $43,280. Thus, to stay on his sustainable spending plan, he will need to save $3,720 of the $47,000 of income he expects to receive from sources other than savings during By comparison, had he used the 4% Rule to determine that he could withdraw $10,000 (4% of $250,000) during 2017, and added that amount to the $47,000 of expected IFoS, he would have developed a spending budget of $57,000 for 2017, an amount nearly 32% greater than the sustainable spending budget developed above. If all the assumptions made by Joe and his FA are exactly realized in the future and Joe spends exactly his spending budget each year, his 2017 spending budget of $43,280 is expected to increase by 1.5% for each future year, and his projected accumulated savings at his death will be $200,000. Of course, not all the assumptions about the future will be exactly realized, and Joe s actual spending is unlikely to be exactly equal to his annual spending budget. For this reason, Joe s financial plan also includes meeting with his FA to revisit his spending budget at the beginning of each year for the rest of his life to keep his spending on track, monitor results and make appropriate adjustments when necessary. From time to time, Joe and his FA will also model the impact of significant deviations from assumed experience and consider alternative courses of action (including modifying Joe s investment strategy or the expected period of his part-time employment) if such deviations were to occur. Ken Steiner is a retired actuary with a website entitled, "How Much Can I Afford to Spend in Page 4, 2018 Advisor Perspectives, Inc. All rights reserved.

5 Retirement?" Page 5, 2018 Advisor Perspectives, Inc. All rights reserved.