William Bengen’s 1994 article in the Journal of Financial Planning is widely cited as a solution for estimating a sustainable withdrawal rate for an investment portfolio during retirement. The article’s celebrated conclusion is that “about four percent” is “safe.” A reasonable first approximation, but this landmark paper is now viewed as a useful but preliminary step in modeling one of the more slippery concepts in financial planning.
The so-called 4% rule isn’t wrong per se. It’s “still a great starting point,” notes Morningstar’s head of retirement research, David Blanchett. But just as Newton’s mathematical narratives represent an early if incomplete blueprint for describing the physical universe, Bengen’s research is considered a seminal effort that launched the modern age of quantitatively analyzing spending expectations in retirement. As for delivering a genuine solution, well, that’s still a work in progress and probably always will be.
The major stumbling block is a familiar nemesis: uncertainty. Future levels of market risk and return are blurry at best. Meanwhile, every retiree’s circumstances and objectives, even for those in the upper levels of the wealth spectrum, are different.
If there’s a weakness in Bengen’s paper, it’s the suggestion that there’s one spending rate for everyone throughout retirement and it’s a number that can be confidently estimated in advance. Reality is more complicated. A truly well-heeled client may be exempt from such hitches. But a certain segment of the moneyed class—particularly those with a penchant for conspicuous consumption—may still require guidance on developing reasonable expectations for spending in retirement.
Ultimately it’s all about the details. Because of different spending goals, time horizons and other factors, the optimal withdrawal rate almost certainly varies from retiree to retiree—and from year to year. Even if running out of money is a low-risk possibility for an affluent client, modeling the future can still be a productive exercise for those planning on maximizing wealth transfers to the next generation, charitable giving and other objectives beyond living comfortably.
In other words, there’s a common set of challenges whether the nest egg is $500,000 or $50 million.
“You can’t predict how long you’ll live, what the stock market will deliver or what inflation will be,” says Jason Hull of Hull Financial Planning in Fort Worth, Texas.
In other words, trying to come up with one spending rate that’s written in stone for decades, for every portfolio, is tempting fate.
David Kitces at Pinnacle Advisor Group in Columbia, Md., says that adjusting the initial spending plan through time is reasonable and necessary. “It’s still important to set the initial number,” he says, but it needn’t remain constant. Rather, the goal should be to think of a withdrawal rate a la Bengen as a floor for spending under the worst-case scenario. “No one likes to have the floor give way,” says Kitces.
He uses Bengen’s 4% rule as an initial estimate for clients. The number is adjusted for each retiree, depending on a range of factors, he explains. If you’re willing to adjust your spending through time, for example, the flexibility allows for a slightly higher withdrawal rate. The range for Pinnacle’s clients overall is in the 3.5%-to-5% range, Kitces says.
Do Rich Clients Need A ‘Safe’ Spending Rate?
A hefty dose of wealth is the optimal solution for ensuring that the assets don’t run out in retirement. Deciding if a client is safely over the financial hump, however, isn’t always cut and dry. Does a $5 million net worth suffice? Maybe, but maybe not. Much depends on expectations and the spending rate relative to assets.
“Depending on how grand the lifestyle, spending plans have to be considered for net worth below $40 million to $50 million,” says Ed Renn, a partner in the private client department at Withers Bergman. “I have wealthy clients who go on vacation and decide to buy a $10 million house where they visited.”
To wit, it’s not how much you’re worth—it’s how much you spend. Assuming that a well-to-do client doesn’t go off the deep end, the finer points of estimating safe withdrawal rates may not be relevant. How can you tell if a client has irrevocably cleared the line? One clue: If you have to ask the question for a given client, maybe estimating a reasonable withdrawal rate is worthwhile after all … especially after factoring in taxes.
The main priorities for wealthy individuals and families tend to be intelligent planning on budgeting and tax-management issues. “Tax planning can have significant impact on the withdrawal rate” for the high-net-worth set, says Phillip Christenson at Phillip James Financial, a wealth management firm in Plymouth, Minn. Given the high tax rates that usually apply to these individuals, smart tax management can effectively be the equivalent of optimizing spending rates. An intelligent approach to minimizing taxes “can extend the life of a portfolio for years.”
Yet it’s not always easy to generalize, even for wealthy clients. Renn points out that some high-net-worth individuals hold a high degree of illiquid assets, such as real estate and private equity. Portfolios stuffed with hard-to-sell holdings can create a challenge that requires a sharper focus on modeling optimal spending rates to maintain a given lifestyle and avoid forced transactions. “For clients with way too much in residential real estate, it’s a concern,” he says.
Who Moved My ShortCut?
Regardless of the size of the asset pool, there are no easy solutions for intelligently estimating a reasonable withdrawal rate. There are, however, some obvious places to start the analytical journey:
1. Annuities
Locking in an income stream with annuities is a compelling choice to provide a foundation for retirement income, according to some advisors. “The safest way to go is with an inflation-adjusted SPIA [single premium immediate annuity],” says Hull. Paying a lump sum up front in exchange for a lifetime of guaranteed payouts that are insulated against inflation risk is a “truly safe method” for generating retirement income, he asserts.
Minds will differ, of course. The main knock against these products is the requirement that investors transfer some or all of their nest egg to an insurance company up front. There are additional risks for the high-net-worth set, some advisors warn. Robert Leahy at Leahy Wealth Management, for instance, recently wrote on his blog that “the wealthy should likely never buy tax-deferred annuities” because “sheltering too much of their investment wealth before retirement often leads to a tax bomb in retirement.”
That may be sound advice as a general rule, but annuitization as a tool for reducing or delaying taxes may still be preferable for some wealthy clients with retirement accounts that are set to trigger a taxable event. Renn points out that the recently introduced option of allowing the direct conversion of some or all of a 401(k) plan into an annuity when an employee retires can be an attractive strategy—especially if assets in the account are substantial and have a high amount of embedded capital gains. In that case, annuitization is a tempting alternative to the standard option of rolling the money from a 401(k) at retirement into, say, mutual funds—a choice that can incur a hefty tax bite.
Another twist on annuities for high-net-worth individuals as a retirement-planning tool: private placement life insurance (PPLI), a specialized form of variable universal life policies. For some clients, a customized PPLI solution can provide substantial tax advantages for managing investment-related capital gains.
The ability to select investments and wrap them in a life insurance policy while gains accrue on a tax-deferred basis is a “big advantage,” says Richard Bernstein of New York City-based Richard S. Bernstein & Associates. “The investment income is not subject to income tax as long as the policy is in force.”
The downside is that setting up a PPLI can be expensive. But the price tag may be worthwhile for wealthy clients because of their ability to customize the investment mix. A standard variable life policy offers similar tax advantages but without the control of the portfolio holdings.
Even if you’re skeptical of annuities, it’s still useful to review the numbers for perspective. Hull recommends exploring the possibilities at immediateannuities.com, a free website that provides quotes from multiple insurers. For instance, the site recently quoted a preliminary “estimate” of $5,500 in monthly income for an immediate annuity for a 65-year-old male based on a $1 million investment—the equivalent of an annual 6.6% withdrawal rate relative to the principal.
2. TIPS
Inflation-protected Treasurys (aka TIPS) are another benchmark for estimating a “safe” income stream. Their lack of credit risk plus their inflation hedge puts these securities in a class of their own.
But there are two issues to consider. One is that the longest maturity for these bonds is 30 years, which means that TIPS may not be a one-purchase solution for funding long retirements. One way around this is setting up a ladder strategy that holds bonds with a spectrum of maturity dates.
The bigger problem is that yields for TIPS are relatively paltry these days, even by the standards of recent history. The real yield on a 30-year inflation-indexed Treasury is currently in the 1% area.
As a result, TIPS aren’t a viable alternative for most retirement strategies at this point, although these bonds can be used in concert with other securities. Of course, if inflation soared in the not-so-distant future, the guaranteed real yield for TIPS would be more attractive.
In any case, the low risk aspect of these bonds make TIPS a natural benchmark, albeit an unusually low-yielding one at the moment.
3. DCDB Benchmark
Another useful yardstick is the Allianz Defined Contribution Decumulation Benchmark (DCDB). This index tracks the estimated payout of a low-risk strategy that blends a laddered TIPS portfolio out to the longest maturity with a deferred annuity that kicks in when the payout from the TIPS ladder ends. The underlying strategy is designed to offer protection against longevity risk and inflation.
Monthly updates based on current market data are available at www.dcdbbenchmark.com.
For the estimate as of June 2015, the strategy’s estimated payout in the first year is nearly 4.5%, which rises to 6.6% in the 20th year and beyond.
4. Annuitizing Payouts
The bond market and insurance products are valuable reference points for estimating a benchmark, but at some point you’ll need to crunch the numbers directly for a deeper level of perspective. There are several possibilities, including a foundational concept for projecting an annuitized stream of payments via Excel’s PMT function.
Two veterans of the institutional money-management world recently explained that “constructing a spending rule is itself an annuitization problem at heart but does not require purchasing an actual annuity.” Spending down an investment portfolio in an annuitization-based framework can be effective, too, advised M. Barton Waring (former Barclays Global Investors strategist) and Larry Siegel (advisor to Ounavarra Capital and the former director of investment research at the Ford Foundation) in this year’s January/February issue of the Financial Analysts Journal (“The Only Spending Rule Article You Will Ever Need.”)
Calculating payouts with a periodic re-annuitization method comes in two basic flavors—one for a portfolio of “riskless” assets (government bonds) vs. a risky portfolio (i.e., stocks, bonds, etc.). The key issue, the authors emphasized, is recognizing that the spending estimate must be recalculated throughout the spending period to reflect changing conditions.
5. Modeling Uncertainty
Another actuarial-based framework reverses the modeling focus by estimating the probability of ruin for a given withdrawal rate. The goal is figuring out the potential risk that a portfolio will run dry too soon. Rather than approximate the optimal withdrawal rate, this methodology identifies the price of so-called “ruin risk” for a given spending plan.
The analytical details tap into some high-level mathematics, but the heavy quantitative work has been boiled down to a single formula that’s easily executed in Excel, according to two finance professors at Canada’s York University—Moshe Milevsky (a leading authority on retirement issues) and Chris Robinson. For details, see their 2005 essay in the Financial Analysts Journal (“A Sustainable Spending Rate Without Simulation.”)
The stochastic present value (SPV) approximates the risk of ruin based on four inputs:
• Life expectancy at the time of retirement
• Expected withdrawal rate
• Expected average portfolio return
• Expected portfolio volatility (standard deviation)
The garbage-in-garbage-out caveat applies, but assuming reasonable estimates, the SPV formula offers a quantitative tool for evaluating the risk that’s embedded in a given spending plan.
Take, for example, a 65-year-old at the start of retirement with a life expectancy of 19 years (the median for that age group), a 4% withdrawal rate and an investment portfolio of stocks and bonds that are projected to earn an annual return of 7% with 20% volatility. His probability of running out of money is 12.3%, according to SPV’s estimate. Too high? Drop the rate to a 3% withdrawal rate and the probability of ruin risk falls by around half, to 6.7%. Another way to reduce the ruin risk without paring the spending rate: design a portfolio that, all else equal, exhibits less return volatility, according to SPV projections.
