4% rule is no longer an ideal retirement strategy, Milevsky says

Retirement income planning is not a rigid, inflexible and one-dimensional process. This makes the popular 4% rule problematic, according to Moshe Milevsky, finance professor at York University’s Schulich School of Business.

The 4% rule originated in a 1994 article by William Bengen, a retired American financial advisor, published in the Journal of Financial Planning.

“Assuming a minimum requirement of 30 years of portfolio longevity, a first-year withdrawal of 4%, followed by inflation-adjusted withdrawals in subsequent years, should be safe,” he wrote.

Milevsky laid out his concerns about Bengen’s 4% rule at a Financial Planning Association of Canada (FPAC) webinar on Monday.

What if things, other than inflation, change? What if the stock market declines 15% or 30% next year? What should the client do then? According to Milevsky, the response with the 4% rule is: “No, stick to the plan.”

“When you explain the 4% rule this way, you’re telling [people], ‘We’re going to tell you, today, exactly what you should do for the next 30 years, because historically, this would have worked,’” he said. “You kind of wonder, ‘Really, I’m pre-committing to this for the next 30 years? No matter what happens, I don’t adjust?’”

Milevsky noted at the FPAC webinar how “optimal decumulation” is a time-consuming process that requires quant skills. He equated decumulation expertise to tax and legal expertise — areas that may be out of the realm of a financial advisor’s knowledge base and that require special advising.

Decumulation is “an intrinsically mathematical problem” that requires mathematical expertise, in which software is not a sufficient solution, he said.

“Decumulation plans have to be multi-dimensional. A good decumulation plan tells me what to do not just based on inflation (i.e. adjust up, adjust down). [It] tells me what to do in many states of nature, like a matrix,” he said.

“If the market is up, this is how much you adjust. If the market is down, this is how much you adjust,” he added. “Telling someone to withdraw the exact same amount of money, in inflation-adjusted terms, no matter what, on a going-forward basis, doesn’t make a lot of sense.”

Milevsky provided a hypothetical example of two clients seeking decumulation advice.

Jack is 75 and receives $77,000 per year from a defined benefit pension (a real index-linked, inflation-adjusted, defined benefit pension plan). He has an additional $100,000 in investible assets.

Jill is 65 years old and receives $20,000 per year from a real defined benefit pension. She has investible assets worth $1 million.

Who should be spending more, in percentage terms?

“They are both equally wealthy on a family balance sheet perspective because the [present value] of the pension, plus investible wealth, is exactly $1.5 million,” Milevsky said. “Should they both be told to spend the same percentage of the portfolio? What about next year, and the year after?”

It varies depending on the client and what is going on with their respective balance sheets, he noted.

Milevsky provided another hypothetical example examining a withdrawal rate and spending strategy based on a sequence of investment returns. In the scenario, an investor has $100 in their nest egg and four years to live — giving them $25 per year if the money earns them $0.

However, what if the sequence of investment returns is: +25% (in year one), -15% (in year two), +10% (in year three) and -14.4% (in year four)? The average return is 0%. The maximum the investor can spend yearly is actually more than $25 ($27.80 per year, to be exact).

Milevsky noted that it’s better to have the good returns in the earlier years of retirement, and the not-so-good returns in the later stages. This is especially the case when the longevity of a retiree, as well as when exactly they’ll retire, are both unknown.

“Retirement success, and retirement failure, is being positioned as a binary variable. That binary variable is creating mathematical problems,” Milevsky said. “Probability-based [retirement planning] I don’t think is the way to go.