A lot of retirement experts get pretty excited about what is commonly referred to in discussions about spending in retirement as “sequence of returns risk.” This risk is generally the negative impact on retirement portfolios associated with spending during periods of poor investment performance. Frequently the experts will focus on the period just after retirement when they claim this risk is highest. For example, in his recent Forbes articles “Navigating One of the Greatest Risks of Retirement Income Planning” and Evaluating the Impact of Sequence Risk on Retirement Income”, Dr. Wade Pfau states, “Even with the same average returns over a long period of time, retiring at the start of a bear market is very dangerous; wealth can be rapidly depleted as withdrawals are made from a diminishing portfolio, leaving little money to benefit from a subsequent market recovery”; “if negative returns are experienced when you start spending from your portfolio, you will face an insurmountable hurdle that cannot be overcome even if the market offers higher returns later in retirement” and “In the withdrawal phase of retirement, the specific sequence of market returns matters a great deal.”
I agree that the sequence of returns matters. I’m not so sure that it matters more when you have just retired vs. when you have been retired for quite a while, but the point of this post is to show you that sequence of return risk can be mitigated by decreasing spending during unfavorable return periods. Since the Actuarial Approach recommended in this website encourages retirees to annually determine a spending budget that matches their assets to their liabilities on a “mark to market” basis, it automatically adjusts spending budgets to mitigate or reduce sequence of returns risk.
Let’s look at a simple example to illustrate this point. Throughout this example, I will be using the five-year projection tab of the Actuarial Budget Calculator spreadsheet from this website. This tab allows the user to vary future investment returns and actual spending to see the effect on retiree assets and future spending budgets. We are going to look at a 65-year old retiree with an annual Social Security benefit of $20,000 and $500,000 of accumulated savings, no bequest motive and a desire to have future spending budgets increase with inflation. For simplicity sake, we are going to look at our example retiree’s spending budget as a whole and not separate it into different budgets by expense type as we would normally recommend.
If we enter the sample retiree’s data and recommended assumptions into the “input” tab of the spreadsheet, we develop a first year spending budget under the Actuarial Approach of $41,751 ($20,000 from Social Security and $21,751 from accumulated savings). The “run-out” and “inflation-adjusted run-out” tabs of the spreadsheet are based on the assumption that the 4.5% investment return that we entered in the input tab will be exactly realized each future year. As discussed in prior posts, this is clearly not a realistic assumption about the future, but we include these tabs simply to show that the math works. These tabs show that if the retiree earns 4.5% each year on his assets and all the other assumptions are realized, his real dollar spending budget will remain at $41,751 each year and his accumulated savings at after five years at age 70 is expected to be $492,651 in nominal dollars ($435,432 in inflation-adjusted dollars).
With this example retiree as background, we are going to shift to the 5-year projection tab and model a different sequence of returns to see how real spending amounts and assets remaining after the five-year projection period are affected. Instead of assuming constant 4.5% investment returns, we are going to assume the following sequence of returns: -25%, -5%, 10%, 20% and 32.5%. The geometric average of this sequence of returns is approximately 4.5% per annum. So, if you had invested one dollar at the beginning of the 5-year period and didn’t make any withdrawals, you would have approximately the same amount of money ($1.246) at the end of 5 years under either the constant 4.5% return sequence or this alternative sequence.
Now we will look at the impact on spending budgets and assets at the end of the five-year projection period of using different spending approaches. The four different spending approaches considered are: the Actuarial Approach without Smoothing, the Actuarial Approach with Smoothing, the 4% Rule and the Guyton Decision Rules starting with a 5.5% withdrawal rate. The Actuarial Approach without smoothing is just application of the approach recommended in this website without any smoothing of the budget or spending amount. The Actuarial Approach with Smoothing applies the following smoothing algorithm: Increase the prior year’s budget amount by inflation but the result must fall within 10% of this year’s calculated budget using the Actuarial Approach. The 4% Rule takes 4% of the initial year’s accumulated savings and adds that year’s Social Security benefit. Each year thereafter, the initial year’s amount withdrawn from savings is increased by inflation irrespective of investment return for that year. Under the Guyton Decision Rules (discussed more in our post of April 26, 2015), the amount withdrawn from savings is increased by inflation each year, but is reduced by 10% for a year in which the current year’s withdrawal rate is more than 20% higher than the initial withdrawal rate and increased by 10% for a year in which the current year’s withdrawal rate is less than 20% below the initial withdrawal rate. In this example, the initial withdrawal rate used was 5.5%.
 |
| click to enlarge |
The graph shows that the 4% Rule produces the smoothest pattern of real dollar spending budgets for this particular example retiree. The Actuarial Approach without smoothing produces the most year to year variation in real dollar spending. The two smoothing approaches provide more smoothing of real dollar spending than the Actuarial Approach without Smoothing but less than the 4% Rule. At the end of fifth year, the Actuarial Approach without Smoothing has assets of $492,644. This amount is basically the same (off by rounding) as the assets expected ($492,651) under the 4.5% per year investment return run-out in nominal dollars. By comparison, the end of the fifth year assets under the Actuarial Approach with smoothing is $475,758; it is $463,848 under the 4% Rule and $445,657 under the Guyton Decision Rule starting at a 5.5% withdrawal rate. Thus, approaches that involve high initial withdrawal rates, like the Guyton Decision Rule in this example, or that don’t adjust spending, like the 4% Rule, are more sensitive to sequence of return risk. While the Actuarial Approach without smoothing produces the most year to year variation of real dollar spending budgets, its use completely mitigates the sequence of risk problem in this particular example.
As I have indicated in previous posts, you can either smooth your spending budget or you can smooth your actual spending. You can also spend more now and thereby increase the risk of having to spend less later. Smoothing your spending budget or front-loading your spending can increase your sequence of return risk. Reducing your spending after experiencing poor investment returns, as recommended under the Actuarial Approach with or without smoothing can reduce this risk. If you use a smoothing approach to develop your spending budget, I encourage you to annually monitor the spending called for under the approach you use with that called for under the Actuarial Approach to see how far off track you may have wandered.
This month, the Bipartisan Policy Center’s Commission on Retirement Security and Personal Savings issued a report entitled, “Securing Our Financial Future: Report of the Commission on Retirement Security and Personal Savings.” This report includes many policy recommendations designed to strengthen the retirement system in the U.S., and with one exception that I will discuss below, I believe the recommendations are very well thought-out. If you are interested in possible changes in Social Security and retirement related law in the U.S., I encourage you to read this report. The report’s recommendations are organized into six major areas:
I. Improve Access to Workplace Retirement Savings Plans
II. Promote Personal Savings for Short-Term Needs and Preserve Retirement Savings for Older Age
III. Facilitate Lifetime-Income Options to Reduce the Risk of Outliving Savings
IV. Facilitate the Use of Home Equity for Retirement Consumption
V. Improve Financial Capability Among All Americans
VI. Strengthen Social Security’s Finances and Modernize the Program
By far the most controversial recommendations to strengthen our retirement system are the ones regarding Social Security. According to the report, long-term solvency is achieved under their proposal by increasing system revenues by about 53% and by decreasing scheduled net benefits (there are some proposed increases) by 47%. The Commission notes that their proposed package of changes would not only solve the 75-year actuarial deficit, but it would also result in “sustainable solvency” as that term is defined by the Social Security actuary.
The only real bone I have to pick with the Commission’s conclusions with respect to Social Security’s long-term solvency is that these conclusions are valid only if the 2015 Trustees assumptions are exactly realized (or are more favorable) and not changed over the next 75 years. For example, the Commission proudly announces that, “the commission’s package of recommendations would extend Social Security’s ability to pay benefits without abrupt reductions through the end of the 75-year projection period” and “if adopted, the commission’s recommendations would secure the program’s trust funds for 75 years and beyond…” The commission neglects to point out that these statements are conditioned on the accuracy of the assumptions made for the 75-year period. And since very few people can accurately predict the future (not even actuaries), it would seem imprudent to assume that assumptions made in 2015 will be accurate for 75 years or longer. After all, even though the assumptions made in 1983 weren’t horribly inaccurate, they were still wrong to some degree, and we now find ourselves facing the very significant changes recommended by the Commission earlier than predicted in 1983. No sound “actuarial” process involves making assumptions for 75 years without anticipating making periodic adjustments in future years. The current approach just isn’t sustainable.
As I said in my post of May 17, 2016, “There exists no process in current law to automatically adjust the System’s tax rates to maintain a balance between system assets and system liabilities. Imbalances (in the form of deficits in the annually calculated 75-year actuarial balance) may occur as a result of the previously unrecognized deficits…, or because of changes in assumptions, experience losses or gains, or from other sources. Unfortunately, the Commission does not address this problem in their recommendations, so instead of achieving their goal of providing predictable Social Security benefits that workers can plan on, workers relying on Social Security could once again in the near future find themselves in a position similar to Charlie Brown trying to kick a football being held by Lucy van Pelt.
As I have said in previous posts on Social Security financing, we just need to look at what Canada did with their Canada Pension Plan for an example of how to use sound actuarial principles to provide “Self-Sustaining Sustainable Solvency.”
The Society of Actuaries has published an article authored by me in its May/June issue of the Pension Section News. The article is entitled, “Using Sound Actuarial Principles to Better Manage Retirement Finances.” In this article I make a case for why I believe the actuarial profession should take a more active role in helping retirees and near retirees develop reasonable spending budgets.
The problem of determining how much to spend in retirement is a basic actuarial problem that requires an actuarial solution. A November, 2014 Survey of financial advisors by Russell Investments concluded that not enough financial advisors were using “math and science to develop spending budgets for their clients and should be periodically comparing the client’s assets with the client’s liability (the present value of the future withdrawals from the accumulated assets) similar to how actuaries measure the funded status of pension plans.” This was a clear shout-out to the actuarial profession to step up its game and become part of the solution.
The public voice of the actuarial profession is the American Academy of Actuaries. The stated mission of this profession body is to serve the public and the United States actuarial profession. “Through its public policy work, [the Academy] seeks to address pressing issues that require or would benefit by the sound application of actuarial principles.” I suggest in the article that helping retirees and near retirees develop reasonable spending budgets is indeed a pressing issue for our country that would benefit significantly by the sound application of actuarial principles.
This post is a follow-up to my post of May 9, where I described all the adjustments to the 4% Rule recommended by Charles Schwab to make it work. And Schwab is not alone in this pursuit. There is no shortage of experts out there proposing adjustments to the 4% Rule to come up with what they believe is a better safe withdrawal rate approach. As indicated in my previous post, the Motley Fool suggested that perhaps instead of using the 4% Rule, you might want to use 3% or perhaps you may want to withdraw “more” than the spending called for under the 4% Rule after a good investment year and “less” after a poor investment year. In my June 24, 2015 post, I looked at Michael Kitces’ proposal to “ratchet up” spending under the 4% Rule by 10% whenever the retiree’s account balance exceeds more than 150% of the initial account balance. He further proposed that if the account balance continues to remain high thereafter, the retiree can continue to apply further increases every three years. He indicated that these spending increases can be “ratcheted” up without much concern about subsequent declines.
As readers of this blog know, I’m not a big fan of safe withdrawal rate (SWR) approaches. My most recent list of the disadvantages of SWR approaches is contained in my post of April 25, 2016. There are, however, two significant potential advantages of using a SWR when compared with a more dynamic approach such as the Actuarial Approach: simplicity and stability of withdrawals. Of course, if you implement all these recommended adjustments, these potential advantages quickly fade away.
Thanks once again to Martin from Maine for providing me with more grist for my blog mill. This time, he alerted me to yet another individual who believes that the 4% Rule needs to be adjusted to work properly. Rob Bennett has developed a “new school” of safe withdrawal rate analysis that adjusts the safe withdrawal rate to reflect market valuations at time of retirement. His website, PassionSaving.com, includes a safe withdrawal rate calculator that requires four input items: “(1) the [Shiller] P/E10 (valuation) level that applies at the start of the retirement; (2) the real return being paid on Treasury Inflation-Protected Securities (the non-stock investment class examined by the calculator); (3) the stock-allocation percentage; and (4) the percentage balance that the retiree desires at the end of 30 years.” For 80% stock allocation percentages, the safe withdrawal rates developed by Mr. Bennett’s calculator vary dramatically depending on the inputted P/E10 level. Using the default assumptions, for example, the safe withdrawal rate (95% probability of not-running out of assets over 30 years and no desired legacy assets) varies from 9.13% for an initial P/E10 ratio of 8 to 2.02% for an initial P/E10 ratio of 44.
Mr. Bennett’s “new school” differs from the “old school” in that current market expectations are expected to affect future investment experience rather than old school techniques that project future experience based on historical results without regard for current market conditions. Mr. Bennett’s default results for Scenario 3 (P/E10 of 26, which is approximately the current level) are not much different from results obtained by Blanchett, Finke and Pfau in their article, Retiring in a Low-Return Environment, which used similar concepts to account for current market conditions (Mr. Bennett’s calculator produces somewhat higher safe withdrawal rates).
If I had to use a SWR approach, I might consider the results of Mr. Bennett’s calculator as another data point to use in my selection process. Fortunately, I am not required to use a SWR approach, and I will stick with the Actuarial Approach, which I believe to be a much sounder approach. And, as I have indicated many times in prior posts, the Actuarial Approach produces a spending budget for the year; it does not tell you how much you must spend. If you are bothered by the potential volatility associated with the Actuarial Approach, you can always smooth the results from year to year (or smooth your actual spending or reduce the expected volatility of your investments). If you insist on using a SWR approach, you can still use the Actuarial Approach to see how far off track you may have strayed. But, of course if you do this, you are adding yet another layer of complication to all those adjustments the experts want you to make to the “simple” 4% Rule.
Frequently we see advice from “experts” regarding how much of your nest egg you should withdraw each year. For example, a recent Motley Fool article, “Forget the 4% Rule: Here’s a Better Way to Approach Your Savings” tells us that the 4% Rule may not actually be all that bad as a rough guide but you need to be flexible in its application. Like most rules of thumb (Rot) approaches, the 4% Rule (or even a more flexible version of the 4% Rule) is a methodology used to “tap” one’s retirement savings and is generally not part of a strategic plan to meet a retiree’s financial objectives in retirement. Since the typical Rot approach doesn’t even consider a retiree’s specific circumstances or specific financial goals, it is unlikely to have a high probability of successfully achieving such goals.
In his most recent blogposts on May 17th and May 27th, Dirk Cotton encourages retirees and their financial advisors to use strategic planning processes similar to those used in business to develop retirement plans for individual retiree households. As part of his recommended process, Dirk advocates that the retiree household adopt a mission statement, whose purpose is “to identify their strategic objectives [or goals], or those things that, at retirement's end, they would need to have achieved in order to consider their retirement to have been successful.” According to Dirk, “The challenge of retirement planning is to find a strategy (and there may be several) that meets the desires of our mission statement but also falls within the limits imposed on us by the economy and our household’s resources.” I encourage you to read Dirk’s recent posts.
The Actuarial Approach advocated in this website can help financial advisors and retirees refine their strategic goals by indicating which goals are financially “possible” [fall within the limits imposed by household resources]. For example, a household may desire to leave a significant legacy to heirs but they may have insufficient assets at this time to fund expected future essential expenses. The “Budget by Expense” tab of the Actuarial Budget Calculator can help the household plan how they will spend their assets (if all current assumptions about the future are realized) enabling them to determine which strategic goals are most important to them. The calculator can also help the household make decisions about how much risk to assume for certain types of future expenses, how their home equity should be spent, etc. The Actuarial Approach also provides a reasonable measure of whether the household is progressing satisfactorily toward the achievement of the goals selected (or to revise goals in the future). Unless your goal is to simply not run out of your savings, adoption of a Rot approach is not likely to be as successful at achieving your strategic goals.
