Sharpe. He suggests investing in a series of zero-coupon bonds for a 4.46% withdrawal rate. Or why not give clients a choice: risk-free distributions with bonds or a variable rate based on a market portfolio?
The 4% rule has become near-gospel in retirement planning circles, thanks to
a groundbreaking study by William T. Bengen in 1994, which showed that an
initial 4% withdrawal rate—increased by inflation each year—would last at least
30 years as long as 50% to 75% of the portfolio is invested in stocks and the
rest in bonds.
The beauty of the 4% rule is its simplicity. It is easy for clients to
understand, easy for advisors to implement, and it gives savers a concrete goal
to work toward. Flip the numbers around and you can instantly give a client
their ” number,” which is 25 times their first-year spending goal in retirement.
A client who wants to spend $40,000 in the first year of retirement will need 25
times that, or $1 million on the day he clocks out for good.
The 4% rule has also given math geeks plenty of fodder for number crunching.
What if you change the allocations? What if you change the sequence
of investment returns? What if you take taxes into account? And is 4% really
the ideal withdrawal rate? Computers have been spitting out all kinds of
variations on the formula to find the holy grail of retirement income planning:
the optimal sustainable withdrawal rate.
Sharpe’s opinion
One of the latest papers to address this issue is called “The 4% Rule—At
What Price?,” by William F. Sharpe and his colleagues at Financial Engines
in Palo Alto. Sharpe argues that the 4% rule attempts to “finance a constant,
non-volatile spending plan using a risky, volatile investment strategy.” As a
result, clients accumulate unspent surpluses when markets outperform and face
spending shortfalls when markets underperform. These inefficiencies raise the
cost of the strategy compared with either (1) investing in risk-free bonds and
being 100% sure that the money will never run out, or (2) varying the withdrawal
rate in accordance with portfolio volatility.
Sharpe and colleagues determined that a retiree can guarantee a set dollar
amount of inflation-adjusted income every year for the next 30 years by buying a
series of risk-free, zero-coupon bonds. The inflation-adjusted income from such
a strategy would equal 4.46% of the initial principal amount. (This is not
dependent on current rates; see the paper for the math.)
The risk-free withdrawal rate is therefore deemed to be 4.46% and is the
baseline against which other withdrawal strategies should be measured. The paper
notes that this risk-free strategy never has a surplus, never has a shortfall,
and is the cheapest way to receive a constant, guaranteed payout every year.
Adding risk, by investing in a portfolio of 60% stocks and 40% bonds (called
a market portfolio), can increase the withdrawal rate over the risk-free rate.
But it necessarily introduces the possibility of failure—even if withdrawals are
less than or equal to the 4.46% guaranteed rate.
So one has to wonder: Why is a market portfolio used to give retirees
withdrawals of only 4%? The client gets less income than the guaranteed rate,
and it assumes more risk than the risk-free portfolio. You could say it’s the
worst of both worlds. Or, in the words of Sharpe and his colleagues, an investor
withdrawing at the guaranteed rate (4.46%) and investing in the market portfolio
(12% volatility) has allocated 13.5% of his portfolio to surpluses and only
86.5% to actual spending. The surplus could be used to increase every retirement
payout by nearly 16% (=13.5%/86.5%).
What’s the solution? Give clients a choice: either a risk-free portfolio with
guaranteed inflation-adjusted withdrawals of 4.46% or a market portfolio with
variable withdrawals. Clients might intuitively be happier with this anyway.
Consider a retiree who is planning on spending $40,000 a year for the next 30
years. If his portfolio suffers a 20% loss in the first year, he might willingly
cut back on his withdrawals in order to spread the pain of the loss across all
his remaining years. This would go against strict adherence to the 4% rule,
which mandates spending an inflation-adjusted $40,000 a year until the portfolio
is exhausted, but it might make the client happier. Or, as economists would say,
it would increase the client’s utility.
Client realities
Financial engineering makes a valuable contribution to the literature and
informs many of the choices you offer your clients. Computer models can quantify
probabilities that human beings have difficulty measuring on their own, and they
tell a kind of truth that eludes people with whole brains.
But the problem with any withdrawal model is that it doesn’t take into
account the messy factors of real life. What if the client has a health crisis?
What if she lives longer than 30 years?
What if hyperinflation raises living costs far higher than expected? What if
stocks and bonds fail to deliver historical returns? What will future tax rates
be? What if the client receives an inheritance or other windfall that changes
the retirement equation in a good way? What if he doesn’t retire in the
traditional sense but transitions toward retirement over a period of 10 or 20
years? The 4% rule—or any preset withdrawal strategy—isn’t designed to
accommodate such variations in spending needs over the retirement
lifespan.
Models that state the exhaustion of assets as probabilities don’t work for
clients because any number larger than zero is unacceptable. And because the
models do not account for unexpected events, their probabilities are probably
flawed anyway.
Clients need a plan that allows them to cover fixed expenses and have assets
in reserve for inflation, possible health crises or other unexpected events, and
vacations and other discretionary expenses. The client may not be fully
optimizing all the possible income from the portfolio based on the models, but
the peace of mind that comes from knowing that possible future contingencies are
covered cannot be measured in dollars.
One solution is to make sure essential current expenses are covered by
annuitized income. Sources of guaranteed annuitized income might include Social
Security, corporate pension, and an immediate annuity in an amount determined by
the shortfall and purchased with available assets.
For example, let’s say a 65-year-old married couple has essential annual
expenses of $60,000. If Social Security is $30,000, the remaining $30,000 can be
funded with an immediate annuity costing roughly $400,000 depending on terms.
Then, if the client had started retirement with $1 million, the remaining
$600,000 could be allocated to an investment portfolio according to the client’s
risk tolerance and personal outlook for life expectancy, discretionary spending
plans, and how well prepared he or she wants to be for possible emergencies.
Withdrawals could be taken as needed for extraordinary expenses.
What clients are likely to love about this plan is that they are assured of
being able to pay their bills and have a rainy-day fund to cover unexpected
future expenses. The bulk of the client’s longevity risk is borne by the Social
Security system and the annuity carrier, both of which utilize risk pooling to
keep the program actuarially sound. If your client is one of the unfortunate
ones who dies early, his heirs will miss out on the funds used to purchase the
annuity—that’s the price of risk transfer—but they will receive the balance of
the investment fund.
On the other hand, if the client enjoys a very long life, fewer personal
assets will need to be consumed thanks to the annuitized income, and heirs will
receive more than if the client had relied on personal assets alone for all of
her income other than Social Security.
Let clients decide
The big question is whether or not the $600,000 will be enough to handle
inflation, health care crises, and other emergencies. Unfortunately, you can’t
answer that. By definition, this fund is earmarked for unexpected events.
Clients who want to draw it down early do so at their peril.
Conversely, refusing to touch it results in a lower standard of living than a
less conservative client would enjoy. But it’s the client’s choice. All you can
do is provide some guidelines on expected inflation rates, average health care
costs, typical discretionary expenditures by retirees, and so on.
You can’t—nor should you be expected to—make assumptions about a client’s
individual health care needs or other reasons why it may be necessary to tap the
rainy-day fund. As much as we’d all like to make distribution planning a
risk-free, fail-safe process, life takes too many twists and turns to carry out
a withdrawal strategy with ultimate certainty.
Above all, whatever shape a retirement income plan takes, it should be
revisited at least annually so adjustments can be made as necessary. A
“set-and-forget” withdrawal plan, whether 4% or some other number, does not
allow for the dynamics of clients’ changing lives.
The 4% rule—or, more accurately, the 25-times-first-year-income rule—might be
a good rule of thumb during the accumulation phase, because it gives clients
something to shoot for. But its major flaw as a distribution strategy is that it
fails to adjust for changes in spending needs and can’t give a clear “yes”
answer to the client’s number-one question: Will my money last?
