Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
Every financial plan, regardless of the way in which it was created, requires ongoing monitoring. Here is one way advisors can measure the “funded status” of a plan to determine if adjustments are necessary.
Monte Carlo models that utilize reasonable assumptions can better facilitate one-time client decisions than deterministic models because they show a range of future results and probabilities of success. On the other hand, those probabilities of success may not be as effective as metrics produced by deterministic models used in a dynamic process to facilitate ongoing planning decisions in retirement.
In this article, I discuss:
- The important difference between one-time and ongoing planning that advisors fail to adequately address with their clients when developing a plan;
- How focusing on a client’s periodically measured “funded status” makes ongoing planning easier to communicate and more effective; and
- How the plan-adjustment process can work on an ongoing basis.
One-time versus ongoing planning
Before employing a specific model to help retired or near retired clients make better financial decisions, an advisor and their client should have a “meeting of the minds” on the purpose of the planning exercise and client expectations. For example, will the client want a high probability of being able to spend $X (real) dollars per year so they can just charge ahead and spend that money even in the most extreme scenarios? Or will the client be willing to make periodic adjustments in the ongoing spending plan when necessary to either avoid spending too much or too little in retirement?
In his Kitces.com post of December 21, 2022, Dr. Derek Tharp discussed the implications between one-time and ongoing planning using Monte Carlo modeling as follows:
One key nuance to the use of Monte Carlo simulations is whether they are being used as part of a one-time plan versus an ongoing planning process. For example, a Monte Carlo simulation resulting in a 90% probability of success will mean very different things depending on whether a client will take fixed portfolio withdrawals throughout retirement based on the initial probability of success or whether they plan to run additional simulations over time and are willing to adjust their spending based on market performance. For the former client, because a 90% probability of success means that there is a 10% chance they will deplete their portfolio (though the magnitude of the failure is unknown), they might choose to aim for an even higher probability of success to decrease the likelihood that they will run out of money in retirement. But for the latter client, to suggest they have a 10% chance of depleting their portfolio is overstating the risk, as they are willing to adjust their spending in response to future simulations that show a reduced probability of success.
Tharp noted that most financial advisors anticipate (and presumably communicate to clients) an ongoing (or dynamic) planning process involving possible future plan adjustments, if necessary, rather than a one-time (or static) planning process. He wrote:
[while] few advisors are running Monte Carlo simulations intended as truly one-time projections…it appears that this understanding of the distinction between Monte Carlo in a one-time-plan context and Monte Carlo in an ongoing planning context is not well appreciated.
Instead of communicating chances of success and pushing clients to select conservatively high probabilities, he said clients who are willing to make spending changes when necessary should be using a different approach and advisors should communicate different metrics that indicate when spending adjustments may be required and the potential magnitude of such adjustments.
Focusing on a client’s “funded status” for ongoing planning
As a retired pension actuary, I agree with Tharp. The purpose of a problem instructs the best model for providing solutions. For example, at my consulting actuarial firm, I frequently used stochastic (Monte Carlo) modeling to help pension clients develop a one-time investment strategy most likely to achieve their objectives, but I generally used a deterministic model and a dynamic (annual valuation) process to calculate the client’s annual ongoing pension plan contributions and plan funded status. Client contributions would increase from year to year if the plan’s funded status deteriorated, but such contributions could be smoothed through various amortization approaches. Periodically, I would stress-test significant deterministic assumptions to assess client risk of experiencing future increased (or decreased) contribution requirements.
A simple deterministic actuarial model and dynamic process (very similar in principle to the approach used by actuaries to determine annual contribution requirements for pension-plan sponsors) can successfully be used by advisors to facilitate ongoing financial planning for retired clients who are willing to adjust their spending (or make other necessary adjustments) to reflect actual experience throughout retirement. I make such a model (the actuarial financial planner or AFP) available on my website. In addition to producing an annual spending budget, the AFP also produces a household balance sheet that compares assets and spending liabilities to generate the household’s funded status. An example of the output from the AFP (from my December 19, 2022 Advisor Perspectives article) is shown below.
The input items to generate the above output are discussed in the December 19, 2022 AP article.
The exhibit above shows Ralph’s funded status as of January 1, 2023 to be 97.29%. This result tells Ralph that if all future deterministic assumptions are realized exactly each year, Ralph spends exactly the annual spending budget (increased annually with assumed increases) each year and Social Security remains unchanged, his current assets will be sufficient to fund 97.29% of his desired future spending.
Of course, not all the deterministic assumptions made in the calculations for 2023 will be realized in the future, Ralph’s spending will not exactly equal his spending budget each year and his real Social Security benefits may be decreased or increased by other than future cost-of-living increases.
Ralph and his advisor meet annually and look at possible adjustments in Ralph’s plan to reflect actual experience and keep Ralph’s plan on track (among other items of concern to Ralph to be discussed at this annual meeting). However, instead of using a Monte Carlo model each time with a 90% calculated probability of success, Ralph’s advisor uses the AFP (or a different deterministic model that is similar.)
How the plan adjustment process can work
Ralph and his advisor agree to the following process for adjusting Ralph’s plan for deviations of actual from assumed experience, deviations of actual from budgeted spending and changes in assumptions or data used in the model.
Step 1: Adjust last year’s spending plan by:
- excluding expenses no longer expected to be incurred;
- including new expected expenses; and
- increasing adjusted plan expenses for inflation.
Step 2: Run the model this year with adjusted spending plan expenses from step 1, revised assets (like Social Security with COLA increase and updated portfolio balances), updated personal data and best-estimate assumptions.
Step 3: If the funded status developed in step 2 is less than 95%, decrease Ralph’s discretionary spending budget (or increase his assets) for the year to achieve a 95% funded status
Step 4: If the funded status is more than 120%, consider increasing Ralph’s discretionary spending budget for the year to achieve a 120% funded status.
Ralph knows that since his funded status is currently about 97%, he may have to reduce discretionary spending next year if experience is a bad as it was in 2022. Alternatively, if investment experience is favorable in 2023, his spending remains in check and price inflation subsides, he may not have to reduce his 2024 spending.
Even though he may not understand the concept of present value, Ralph understands the potential implications of a declining or increasing Funded Status and looks forward to his next meeting with his advisor.
I agree with Tharp that there are very few households (or financial advisors) who are comfortable with one-time planning approaches, but many clients (and unfortunately some financial advisors) fail to understand the real implications of selecting a 90% or higher probability of success when using Monte Carlo models. In this article, I offer a deterministic model to periodically measure a household’s funded status and a process to make previously agreed upon changes in the client’s plan when necessary.
Ken Steiner is a retired actuary with a website entitled "How Much Can I Afford to Spend in Retirement?"
More Defined Benefits Topics >