Page 40 - Investment Advisor - November 2023

reTIremeNT PLANNING

advisors and retirement industry thought idea that the failure rate taken alone has hand, if they assume that their retire-
leaders. This includes Bryn Mawr Trust’s two big flaws. The first is that it doesn’t ment goes 36 years and that their portfo-
Jamie Hopkins. The risk-adjusted cover- speak to the timing of failure. lio in the simulation would cover just 30
age ratio is a “really fun” metric to look “Did your portfolio run out of money years, that would give them a ratio of 1.2.
into, he said in a video posted to the social super early in retirement, like in year Obviously, the more this ratio increases,
media platform X, formerly Twitter. 15, which you would find unaccept- the more cautious a client would want to
As Hopkins explained, Estrada’s paper able?” Hopkins asked. “Or did it run be about that plan and the more they may
shows how financial planners can do out of money in year 29 [of the 30-year need to consider making lifestyle adjust-
better for their clients by helping them projection period]?” ments along the way to protect themselves
to optimize and regularly update their These are two very different levels of from running out, Hopkins said. The idea
spending plan. One powerful means of failure. The other question is the magni- is that, as one navigates retirement, they
doing so is to introduce new metrics that tude of failure, which relates to the tim- will see their coverage ratio move up and
help clients to understand the “magnitude ing but is also a distinct consideration. down according to their actual spending,
of failure” concept that is often overlooked How far short did the client run at that market conditions and other factors.
in traditional Monte Carlo simulations. time? Would it be a devastating failure Alternatively, if clients assume a
Estrada is asking an important ques- or a minor inconvenience? 24-year retirement period and their
tion, Hopkins says, and is portfolio could easily sus-
pointing out that advisors “Remember, success isn’t binary tain 30 years of withdrawals,
have had too much focus on that gives them a starting
one number when it comes and neither is your retirement.” coverage ratio of 0.8, which
to deciding what retirement may signal that they could
strategy makes sense — the —Jamie Hopkins expect to spend more annu-
failure rate of a portfolio in a ally early in retirement —
traditional Monte Carlo simulation. The other key consideration is to ask especially if they don’t have big legacy
As Hopkins has explained in prior whether it is really a “successful” retire- giving goals. Estrada argues that advi-
videos and in discussion with Investment ment if clients are petrified of spending sors and clients can use this frame-
Advisor, when reporting binary Monte and end up following a very conservative work to help make ongoing adjustments
Carlo results to a client framed around plan with a 100% success projection. according to the perceived likelihood of
probability of success, anything less than This could mean they end up leaving a these different events occurring.
100% can sound scary. For example, for a large bequest — either to a spouse, chil- “Now remember, you should pick a
client with a 75% probability of success at dren or the government via estate taxes. strategy not based just off of this research
a given starting spending amount, failing “Is that a good thing? Is that even what but one that resonates with you and that
one out of every four times simply does you’re looking for?” Hopkins asked. you can understand,” Hopkins said. “You
not sound acceptable to many people. In the paper, Estrada introduces a should also not get too overly focused
It is crucial, however, to think care- new metric called the “risk-adjusted on one number. Remember, success isn’t
fully about what a 75% success result in coverage ratio,” which can help clarify binary and neither is your retirement.”
a Monte Carlo simulation actually sug- these issues. Essentially, the advisor is Estrada summarizes his findings in a
gests. While this metric does project that taking the projected number of years of similar way. “When selecting an optimal
one in four retirement scenarios will inflation-adjusted annual withdrawals retirement strategy, a retiree may aim
“fail,” the metric alone actually tells a cli- that could likely be sustained in a given to maximize the coverage ratio, a novel
ent nothing about how severe that fail- situation, and then they are dividing metric superior to the failure rate,” he
ure is. “Now here’s the thing,” Hopkins that number by the anticipated length wrote. “This article suggests focusing on
said. “Retirement is not binary. It is not of retirement for the individual client, the whole distribution of coverage ratios
success or failure. People adjust their given their health status and longevity instead, or at least on some percentiles that
spending, they adjust their lifestyles, expectations, Hopkins explained. may be of particular interest to a retiree.”
when [the] plan starts to go off course.” As a simple baseline, if clients expect Although such an approach may not be
So, as Estrada is asking, why would to make it 30 years in retirement, and as neat as making decisions based on opti-
advisors only make decisions about their withdrawal plan is projected to mizing a single variable, it does enable
what the retirement strategy should be cover all 30 years without leaving any consideration of the relevant trade-offs a
based on that outdated, binary notion? leftovers, that would provide a “1” value retiree needs to evaluate in order to find
In the paper, Estrada pushes on the for their coverage ratio. On the other an ideal retirement strategy.

38 Investment AdvIsor November 2023 | ThinkAdvisor.com