Close Close
ThinkAdvisor
Wade Pfau

Retirement Planning > Spending in Retirement > Income Planning

Why 1966 Was the Worst Year to Retire (and Why It Matters in 2023)

By John Manganaro

X
Your article was successfully shared with the contacts you provided.

What You Need to Know

  • The difficulty experienced by retirees between 1966 and 1995 is the basis of the 4% withdrawal rule.
  • Retirement simulations are useful, researcher Wade Pfau says, but they are limited in profound ways.
  • He suggests rerunning simulations as circumstances change and using flexible spending approaches.

Most financial planning professionals are able to articulate the basic premise of the 4% safe withdrawal rule, but that doesn’t mean they fully appreciate either the real power of the retirement spending framework or its significant real-world limitations.

They also might be unaware of where the 4% figure came from. As retirement income researcher Wade Pfau recently pointed out, the popular guideline for how much money is safe to spend annually in retirement was calculated based on a retirement beginning in 1966.

“In the original analysis, this was basically the toughest 30-year period on record for a new retiree,” he said on a recent episode of the Economics Matters podcast.

In general, financial planners struggle to fully understand and accurately contextualize Monte Carlo simulations — of which the 4% withdrawal rule is perhaps the most famous and widely cited example, Pfau said.

As Pfau told podcast host and Boston University-based economist Laurence Kotlikoff, the topic of poorly contextualized Monte Carlo simulations and the shortcoming of the 4% withdrawal rule might sound like overly academic or esoteric matters, but they are actually of paramount practical importance to financial planners serving investors focused on retirement.

“Don’t get me wrong, the 4% rule does have a lot of practical use,” Pfau says. “It is, to put it simply, a research guideline that can allow for the start of a solid conversation about income planning.”

What is critical to understand, however, is that this type of modeling is highly sensitive to the inputs and assumptions being used, Pfau warns. Monte Carlo simulations, with their focus on generating binary success-failure probabilities, can mask a lot of nuance in middle-ground cases where success and failure are harder to define, “such that we have to view all retirement simulations with a significant degree of caution.”

According to Pfau and others, an overreliance on probability-focused Monte Carlo simulations is one key problem for the planning industry to address, and another is figuring out how to more clearly and effectively communicate with clients about the interplay of complicated sources of risk.

Ultimately, Pfau argues, now is a great time for advisors to learn and leverage some of the key planning concepts being put forward by academics, and he says studying the history of the 4% withdrawal rule is a decent place to start.

Where the 4% Rule Really Comes From

“You might not expect it, but we can actually still learn a lot by going back and looking at the study that first brought about the 4% withdrawal rule,” Pfau says, citing the work of Bill Bengen, the researcher and retired advisor credited with inventing the spending framework.

“For example, it is really interesting to look back and see that the 4% ‘safe’ withdrawal figure itself comes from what would have been safe to spend during the 30 years from 1966 to 1995,” Pfau explains.

As Pfau notes, the period in the late 1960s and early 1970s was a tough time to retire. Inflation ran rampant, and the S&P 500 scored several significantly negative years in that period. Returns were particularly poor in 1966, 1969, 1973 and 1974.

“Notably, after 1982, or about halfway through the 30-year retirement that started in 1966, the markets actually did really well,” Pfau observes. “The key takeaway here is that, even though the average return to a portfolio was decent between 1966 and 1995, the sequence of returns was really difficult for retirees to deal with.”

In other words, by the time a retiree hit 1982, their portfolio had essentially been decimated from the need to sell assets to generate income while prices were significantly depressed. Only by limiting their spending to 4% per year from the start of the retirement period could the 1966 retiree reliably avoid running short of funds.

“Conversely, 1982 was actually an amazing year to retire,” Pfau points out. “You could spend something close to 10% and it would be safe in the simulation. That is the kind of nuanced information that is commonly missed in a Monte Carlo simulation.”

The Dangers of Monte Carlo Simulations

According to Pfau, if there is one lesson financial planners should take away from his discussion with Kotlikoff, it is that Monte Carlo simulations are only as good as the inputs and assumptions fed into them — and the ability of the advisor and client to soberly interpret the results.

“This makes sense if you keep in mind that the Monte Carlo simulation, while a sophisticated technique, is really just generating a sequence of returns that helps you to see a probability of success and failure for a given income strategy and average return expectation,” Pfau observes.

As he notes, depending on the level of optimism in the assumptions being used, the exact same retirement income strategy can be shown to have anything from a 95% probability of success to something more in the range of 60% or even 50%.

“When you start to play around with these simulations in earnest, one of the first things you realize is that the outcome can vary so much without making a big change in the inputs,” Pfau explains. “In practice, this is really important. We actually have to be humble with our projections, because we know in real life the input variables of market performance and spending levels can vary so much from our initial expectations.”

Pfau says the projected volatility input is “always one that is really hard to grapple with.”

“This just comes from the fact that forecasting volatility is really hard, and so we tend to rely on historical volatility averages,” Pfau says. “But this has its limitations, as we have seen, because average volatility over a given time period can vary so much from the volatility experienced in a given year.”

Takeaways for Financial Planners

To be clear, Pfau does not argue that advisors should simply lay aside this potentially powerful analytical tool, but they do need to tread with caution.

“The real lesson here is that retirement simulations are useful, but they are limited in some pretty profound ways,” Pfau says.

This means it is sensible to constantly rerun and revisit simulations and see how the success probability may be changing over time, given what has actually happened so far in a person’s savings journey or retirement period.

“Also, this where the importance of embracing flexible spending approaches comes into play,” Pfau argues. “The ability to modulate spending is a major safety value, if you will. It helps you address the real sting of sequence of returns risk, which a standard Monte Carlo simulation can mask. What we see in practice is that, if you can modestly cut your spending during market downturns, that has a big impact on outcomes.”

Pictured: Wade Pfau 

John Manganaro

NOT FOR REPRINT

© 2024 ALM Global, LLC, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to [email protected]. For more information visit Asset & Logo Licensing.