Thanks to Mark Chamberlain, Co-Founder of The Open Architecture 2020 Group, for pointing us to an interesting Retirement Management Journal paper by James B. Sandidge entitled, “Odds Are Retirees Don’t Care about the Odds.” Mr. Sandidge’s well-expressed reservations about Monte Carlo Analyses typically used by Financial Advisors struck a chord with us as we have expressed our own misgivings in many of our prior posts. In fact, in our most recent post of January 10, 2021, we said,
“As discussed in our post of May 19, 2020, current Monte Carlo modeling generally by financial advisors falls short of the Actuarial Approach in many functional areas. The fact that many financial advisors may use more aggressive real investment return assumptions (arguably without adequately reflecting increased risk) to inflate initial spending rates does not, in our opinion, make this approach superior. We believe that when returns are properly risk-adjusted, spending budgets that anticipate higher returns on risky assets will converge to the Actuarial Budget Benchmark.”
We encourage you to read Mr. Sandidge’s paper. It includes much more discussion (and research) of behavioral and psychological considerations than we, as actuaries, would normally consider, but we found his arguments very compelling (probably because we agree with his conclusions).
In this post, we will include some of Mr. Sandidge’s quotes regarding Monte Carlo models typically used by financial advisors, and, as suggested by Mr. Sandidge, we will include results of a “worse-case” investment environment projection under our Recommended Actuarial Financial Planning Approach for Hank and Marie, the hypothetical couple from our previous post.
Selected Quotes from “Odds Are Retirees Don’t Care about the Odds”
Here are a few quotes from Mr. Sandidge’s paper that resonated with us.
“Although ubiquitous within the financial services industry, Monte Carlo analysis is likely an ineffective tool that wastes resources and distracts most investors from the essence of the problem.”
“Monte Carlo is wildly inaccurate in its predictions of how long a retiree’s savings are likely to last and employs a methodology that is the opposite of what retirees want.”
“Advisors should articulate how their retirement income strategy would have performed in a recent worst-case environment, emphasizing short- and long-term principal erosion.”
“Lifestyle risk is the much more likely worst-case outcome for most people (instead of the probability of running out of money).”
“Being prepared to manage risk and cash flow every year creates the multitude of solutions needed for truly personalized retirement income plans.”
“The flawed inputs of Monte Carlo are not approximately accurate and its output is nowhere close to vaguely right, but the wealth of data it generates makes it precisely wrong.”
“Using flawed inputs and a systematic approach creates an illusion of control and suboptimal risk analysis that leads to flawed forecasts of portfolio longevity in a practical setting.”
“Balancing [retiree] goals requires planning for the worst-case and actively managing risk and cash flow to adapt to the environment. Working backwards from these goals does not lead to Monte Carlo analysis…
As suggested above by Mr. Sandidge, the next section will illustrate how performance of our recommended planning process might be “articulated” in a “worst-case” environment.
Stress testing Hank and Marie’s Retirement Plan under Our Recommended Actuarial Financial Planning Process
In our previous post, we estimated that Hank and Marie would need an initial Floor Portfolio of assets of about $1,471,586 to fund their future essential expenses, leaving about $625,000 to invest in equities in their Upside Portfolio. If they used the same assumptions as used to price their essential expense liabilities (except with a 1% per annum annual expected future increase), they estimated that this would cover their $20,000 per annum discretionary expenses and leave $200,000 (in 2021 dollars) upon the last death within the couple if all assumptions were realized. To fund their Floor Portfolio with guaranteed income, they decided to invest $375,000 of their accumulated savings of $1,000,000 in lifetime income annuities ($200,000 for a single life immediate life annuity for Marie and $175,000 for a single life immediate annuity for Hank). These purchases resulted in annual benefits of $10,428 for Hank and $10,740 for Marie. This investment left them with $625,000 which they decided to invest 100% in equities. This investment in annuities actually increased the present value of their total assets by $47,392 from $2,095,422 to $2,142,814 under the default assumptions, and increased the present value of their discretionary expenses by the same amount.
To stress test Hank and Marie’s plan, we assumed investment returns on equities equal to those proposed for stress testing in our post of April 23, 2018 with approximately the same experience as 2008/2009 for the first two years and a total of five years to get back to where Hank and Marie started.
Results of the five-year projection are shown below. All income and expense amounts are assumed to be paid as of the beginning of the year. Social Security benefits are projected to increase by 2% per year, Hank and Marie’s essential non-health expenses are projected to increase by 2% per year and their health expenses are projected to increase by 3% per year. We ran five years of ABCs for Hank and Marie with updated ages, benefits and expenses to produce this summary. The default assumptions were assumed to remain unchanged during the projection period.
 |
| (click to enlarge) |
As shown in this exhibit, under Hank and Marie’s plan, they will not be required to reduce essential spending during this “worse-case” scenario, but they may be required to reduce discretionary spending. If they use the default assumptions for investment return and lifetime planning periods and they do not “dip into” their legacy goal, they will need to reduce discretionary spending from $20,000 to about $10,500 in 2022 and $8,000 in 2023.
Hank and Marie can use this type of analysis to determine if they are comfortable with the risk inherent in their retirement plan. And while their accumulated savings are 100% invested in equities under their plan, the last column shows the percentage of their total assets that are subject to equity investment risk is relatively low. However, if they are uncomfortable with this level of risk, they can always invest a portion of their investible assets in less risky investments. This exercise also highlights the importance of determining which expenses are considered “essential” and which may be considered “discretionary.”
Conclusion
We agree with Mr. Sandidge’s concerns regarding Monte Carlo models typically used by financial advisors. We believe our recommended planning approach, utilizing basic actuarial principles and processes (including periodic stress testing) and Liability Driven Investing principles can be a more effective retirement planning tool than Monte Carlo models and can lead to better financial decisions. We encourage DIY households to try our free actuarial models and processes, and we encourage financial advisors to incorporate these models and processes into their financial consulting toolkits.
