With evidently more passion to learn than they may ever have shown as students, several hundred financial planners packed a double-wide hotel meeting room last week to hear William Reichenstein talk about tax-efficient withdrawal strategies at the Financial Planning Association’s annual conference in Boston.
Reichenstein spoke rapidly at the podium; he has given this slide show before. But the audiences are always fresh, and this one was rapt. Affluent retirees hate taxes, of course (perhaps because taxes can be their single biggest expense). And lowering clients’ taxes is a good way for advisors to earn their asset-based fees.
A professor at Baylor University, Reichenstein is well known in retirement circles. Many of the advisors seated in chairs or at tables or parked along the walls of the room in the Boston Convention and Exhibition Center knew his 2011 book, “Social Security Strategies” (with William Meyer), and some had no doubt read his recent article in the Financial Analysts Journal (with Meyer and Kirsten Cook) on which this day’s presentation was based.
His presentation compared five different withdrawal strategies: i.e., five sequence patterns in which to draw a client’s income in retirement from a combination taxable accounts, tax-deferred accounts (e.g. rollover IRAs) and tax-exempt accounts (e.g., Roth IRAs).
The punch line: The fifth strategy can increase the life of a hypothetical portfolio to 36.17 years, or about three years longer than the “conventional” strategy of drawing income from taxable accounts first, then traditional IRAs/401ks, then Roth IRAs, and 20% longer than withdrawing Roth IRA first, then traditional IRAs, then taxable money.
The fifth strategy is a bit complicated, so pay close attention (or read the academic paper). First, assume a 65-year-old woman who needs $81,400 in after-tax income each year, has an $11,500 standard deduction/personal exemption, and who can withdraw $47,750 from her tax-deferred accounts each year without breaching the ceiling of the 15% tax bracket. Also assume a 50/50 stock/bond portfolio with an average bond return of 4% and stocks returning a 4% geometric average. She withdraws once a year.
At the beginning of each year, the retiree makes two separate Roth conversions of $47,750, which takes her income to top of 15% bracket. She also withdraws funds from her taxable account until exhausted and then from her tax-exempt Roth IRA.
One Roth will contain short-term bonds and the other US stocks (with repeating returns sequence of – 12.6%, 22.6%, and 5%). At the end of the year, she re-characterizes the Roth (back to a traditional IRA) with the lower value and retains the other. In late retirement years, she withdraws funds from tax-deferred account and tax-exempt account.
“A key to a tax-efficient withdrawal strategy is to withdraw funds from [tax-deferred accounts] such that the investor minimizes the average of the marginal tax rates on these withdrawals,” write Cook, Meyer and Reichenstein in their 2015 FAJ article. “Contrary to the conventional wisdom, the TEA is not more tax advantaged than the TDA.”
Americans Funds as decumulation tools
Purveyors of index funds no longer need to prove that their investment methodology has value, as they once did. Today, sellers of actively managed funds bear that burden. At the FPA convention, Steve Deschenes of the Capital Group (formerly of Sun Life and, before that, MassMutual) made the case for active management in a talk about his firm’s American Funds.
Certain American Funds perform especially well, relative to their benchmarks, during the withdrawal stage (i.e., retirement), Deschenes said. His calculations showed that, all else being equal, a portfolio of A-share American funds (half “moderate allocation” and half “world allocation”) delivered hypothetical ending balances after 20 years of annual withdrawals that were two to almost three times larger than a portfolio consisting of the comparable index fund.
The results included the annual expenses of the American funds but not the initial sales charge, which can reach as high as 5.75%. The case study assumed a starting balance of $500,000 and either 4%, 5% or 6% withdrawal rate (plus annual inflation increases of 3% the previous year’s withdrawal) over a 20-year period ending in December 31, 2014.
Three key fund attributes drove the outperformance of selected American funds and others with the same characteristics, Deschenes said. The success factors were high rates of manager ownership in the fund, relatively low expense ratios, and low “downside capture” ratios. (In a down market where an index registered a loss, a comparable active fund that registered a much smaller loss would have a low downside capture ratio.)
The “true cost” of retirement
Fee-based (who earn AUM fee and commissions) and fee-only (AUM fees) financial planners tend not to use annuities to mitigate their clients’ risk of running out of money in retirement. Traditionally, they rely on rules of thumb about income replacement ratios and safe withdrawal rates.
But what if those rules of thumb are based on unfounded assumptions? David Blanchett, Morningstar’s retirement guru and one of the FPA conference headliners, gave a presentation in which he argued that each retiree household has its own specific income and spending needs that vary over the course of retirement. As a result, the rules of thumb don’t necessarily apply.
The “true cost” of retirement for most people will probably be lower, as much as 20% lower, Blanchett argues. For instance, he believes that most people won’t need income in retirement that’s 70% to 80% of their pretax income. In many cases, that number will be closer to 50%.
Why? Because most people won’t live until age 95. In the old days, advisors thought it was rational for retirees to plan on reaching the average life expectancy. Then they were urged to plan for a very long life. Now, Blanchett seems to be saying, it’s OK to plan on average life expectancy.
Blanchett offers a similar insight about the safe withdrawal rate, long assumed to be 4% of the original balance per year, increased slightly each year to account for inflation. For most people—i.e., those who live only 15 or 20 years in retirement—a 5% payout rate will offer enough safety. Using the higher payment rate means either of two things: You’ll be able to spend 20% more each year or you can save 20% less.
The moral of Blanchett’s story: “The true cost of retirement is highly personalized based on each household’s unique facts and circumstances, and is likely to be lower than amounts determined using more traditional models.”
© 2015 RIJ Publishing LLC. All rights reserved.