One of the most common buzzwords in Financial Planning is “withdrawal rate.” The question of “If I retire today, what rate of withdrawal can I count on in retirement?” is one of the most common I hear from clients. The heart of this question revolves around the idea that a retiree can assume a certain withdrawal rate from their portfolio throughout retirement to guarantee the longevity of their assets. This idea is rooted in findings from the mid 1990s by William Bengen, along with follow up analysis from the Trinity study, that concluded given historical return and inflation numbers, a 4% withdrawal rate is a standard “safe withdrawal rate.” Over the past few years, though, a wealth of analysis has been done to change our perception of what has become known as the 4% rule.
A leader in this research has been Wade Pfau, Ph.D, CFA, who is a professor of retirement income at The American College. Through several studies, Professor Pfau has come to find that the idea of a ‘safe’ withdrawal rate of 4% isn’t holding up, and more importantly, the question of a ‘safe withdrawal rate’ itself is much more complex. He notes in the October 2013 Journal of Financial Planning “In terms of all the contradictory findings within safe withdrawal rates and systematic withdrawals, it’s important for clients to be thinking more broadly so they don’t get too hung up on the idea that there is a safe withdrawal rate from a volatile portfolio,” and he goes on to note about the 4% rule specifically, “It introduced success rates and failure rates and established in the public’s mind the idea that 4% has a 95% chance for success over 30 years with a 50% (equity)/50% (fixed income) portfolio. It works 95% of the time in history, but that doesn’t mean it has a 95% chance of working for someone who retires today.” In looking at sustainable withdrawal rates of other development economies outside of the U.S. and also incorporating the volatile returns of the past decade, Pfau concludes that if a sustainable withdrawal rate does exist it is most likely less than 4%. This idea of a sustainable withdrawal rate itself, though, seems to be illogical when addressed outside of a classroom.
The major problem with trying to find a ‘safe’ withdrawal rate for an individual is rooted in the unpredictability of the market and the individual. In regards to the market, factors such as high inflation, low bond yields, or poor stock returns at the onset of retirement can have a huge effect on a portfolio. As retirement progresses, the interaction of these market factors can also shape one’s retirement. As Pfau notes, having a manager to understand current conditions can help set acceptable withdrawals; “What recent research starts to show is that it matters what the current conditions are. If the stock market is overvalued, then that implies lower future stock market returns, which in turn implies a lower sustainable withdrawal rate. If bond yields are low, then that implies lower future bond returns, which in turn implies a lower future withdrawal rate. We see that there are links, and where you are at today can help determine what the sustainable withdrawal rate is going to be.”
One response to the current market conditions problem is to invest in a fixed annuity, which provides a set amount of money over the life of a client. Outside of losing upside potential and possible fee drawbacks, the idea of locking in a return can leave a retiree with few options if their retirement situation changes, irrespective of the market. This is the other reason thinking that there is a singular safe withdrawal rate can be foolhardy—the assumption that a situation won’t change. Health problems, tax changes, legacy factors, relationship difficulties, etc., can all impact a person’s retirement situation, and in turn affect someone’s ‘safe’ withdrawal rate.
Recent discussion has regarded withdrawal rates as a more dynamic process rather than a set systematic or ‘static’ approach. As noted, retirement variables are far from static, so the idea that withdrawals should be static as well seems fairly backward. David Blanchett, CFP®, CFA, head of retirement research at Morningstar Investment Management, recently produced some valuable research regarding this idea in the September 2013 Journal of Financial Planning. According to Blanchett, “A static approach assumes that a retiree selects a withdrawal rate at retirement and subsequently increases the portfolio level of consumption, regardless of portfolio performance, expected mortality, or the retiree’s changing needs.” In other words, this approach entails selecting a set percentage withdrawal from the portfolio, and then increasing this withdrawal dollar figure annually by the rate of inflation until the portfolio is exhausted.
In contrast to this static approach, Blanchett favors “a dynamic technique because the portfolio withdrawal amount adapts to ongoing expectations and actual experiences during retirement.” Therefore, a favorable outcome is seen when a retiree allows withdrawal rates to change given market trends and personal life changes. Models for creating the ‘ideal’ dynamic withdrawal vary, and each model comes with different benefits and drawbacks. The ‘RMD’ approach, which uses the RMD actuarial withdrawal table associated with IRA required minimum distributions and applies it to all retirement assets, is a nice approach due to its simplicity, but as Blanchett suggests, “the more dynamic and adaptive a withdrawal strategy is, the better off it makes the retiree.” Therefore finding an approach that is both encompassing and feasible for a retiree can best be found through extended discussion with a financial planner.
A dynamic approach to retirement is not limited to withdrawal rates. In explaining dynamic withdrawal rates, Blanchett notes, “The optimal behavior for a retiree in the face of longevity risk is to consume some amount of income based on survival probabilities and current likelihood of being able to achieve some level of income, versus blindly withdrawing constant income for life.” For most retirees, goals are focused around mitigating sustainability risk or, in other words, maximizing retirement income while limiting the risk of running out of money (or dipping into funds originally left for inheritance). Like withdrawals, both retirement date and survival risk should be looked at from a dynamic perspective. Consequently, by continually revisiting and revaluating all these factors, a retiree may have the best opportunity to mitigate sustainability risk.
Longevity risk is a factor that many retirees and planners alike may not factor into retirement spending enough. Retirement projections most people encounter provide a picture of retirement given a portfolio amount, return rate, inflation rate, and withdrawal rate. These projections are then shown until the retiree reaches 95 or 100. The fact that may not be discussed enough is that very few people actually live to age 100. Recent research from Grant Gardner, Ph.D, and Sam Pittman, Ph.D, in the December Journal for Financial Planning took a deeper look into this trend. Their main finding was that if “two sources of uncertainty determine sustainability risk: (1) uncertainty regarding the investor’s lifespan and (2) investment return uncertainty, then the prevalent approach captures investment return uncertainty, but fails to represent lifespan uncertainty.” It makes sense that if an advisor is consistently reviewing retirement conditions such as market returns and withdrawal habits, perhaps more emphasis should be placed on longevity analysis. As Gardner and Puttman mention as an example, “If a client has only a 5% chance of living to age 99 and a 70 percent chance of having wealth if he lives to age 99, his odds of dying penniless are much less than 30 percent. Nonetheless, the client may decide to reduce his spending, forgoing a better lifestyle in retirement because of faulty risk assessment.” In a retirement landscape much less reliant upon lifelong payment vehicles like pensions, it is important that a retiree look at all risk factors, including long-term health, when reviewing his or her current portfolio.
The need to look at retirement from a more dynamic perspective is not limited to those in retirement. The question of “When can I retire?” is also changing as we move from Defined Benefit plans to more Defined Contribution plans. In Defined Contribution plans, individuals add a piece of every pay check over the 40-plus years they are working. Each account is assigned to the individual; therefore each individual is responsible for the account’s returns. As explored by Michael Kitces, CFP ®, CLU ®, ChFC®, RHU, REBC, in the December issue of the Journal of Financial Planning, this saving structure can create an issue when considering the volatility of the market place; “After the first decade or two contributions (to a savings plan) have less and less of an impact, as the results are increasingly dominated- for better or worse- by the portfolio returns. In fact, by the time the retirement date approaches, the outcome becomes so dependent upon returns that a poor final decade of performance can drastically derail the entire retirement plan.” When worked out mathematically, we see that a “25 year old who accumulates $1 million by age 65 with regular savings and compounded investment returns is just barely crossing the $500,000 line around age 55.”
Therefore we see that the classic 401(k) retirement model relies heavily on the last decade of retirement. If this decade were 2000–2010, a period when the market was nearly flat, you can see how an individual may have to rethink his or her retirement. In most cases the result will be a need to put off retirement and thus delay the need to withdraw from the retirement portfolio. Furthermore, as we saw with last year’s market, long periods of a down market are usually followed by an upswing if given enough time. Once this upturn has been taken advantage of, a retiree may be able to decrease portfolio risk and look more toward creating portfolio retirement income. Until reaching this point of being able to lower portfolio risk and still achieve retirement goals, though, a retiree may need to continue working. This may seem obvious, but many people who might be accustomed to a pension system with set retirement dates can be in for a harsh realization. As Kitces puts it, “Following a plan with a high reliance on key returns in a small number of years leading up to retirement leads to a concentrated exposure to risk. In some cases, prospective retirees may still wish to take such risks, but perhaps it’s time to acknowledge the ramifications of such risks not in standard deviations of returns and wealth, but in real-world outcomes of potential retirement delays and retirement date risk.” As we have seen with other retirement factors, retirement date may need to be looked at more as a dynamic, moving target given the defined contribution landscape.
Retirement, like economics and finance, is not an exact science—as the past 15 years can attest. As much as we try to develop rules to dictate the retirement process, the guidelines seem to never stop changing and evolving. In retirement, whether it is because of changing market conditions, changing desires, changing opinions, or changing circumstances, the presence of human and market inputs means that at some point any retirement rule can go out the window. Given this fact, working with a trusted advisor to make sense of varying inputs and risks may be the most productive way to handle a dynamic retirement. In the new retirement landscape, making tough decisions is inevitable, but these should be educated, fully informed decisions. By having a clear picture of all variables, tough decisions can be deliberate decisions. I know this response may not be as nice as knowing there are a few rules that can ensure retirement happiness, but I can at least say this is a response that should stand the test of time.
© 2017 Howe and Rusling, Inc.