Most research regarding withdrawal rates focuses on asset allocation, longer distribution periods and increasing withdrawals based on inflation. However, at the start of retirement, no one knows how long our lifespan may actually be.
Consequently, Larry Frank, John Mitchell and David Blanchett have set out to uncover the reality that none of us know how long we will live, and, therefore, we do not know how much we will need. They share their thoughts in “Transition through Old Age in a Dynamic Retirement Distribution Model” from a 2012 article in Journal of Financial Planning.
As financial advisors, we often see two types of individuals: the spenders and the savers. Spenders will be forced to adjust their spending habits usually some time in their 70s as the realization becomes apparent that there is a high risk of outliving one’s retirement. The savers will not outlive their money, because they have an internal monitor to guard against excessive spending.
But the reality remains that we don’t know how long we will live and, therefore, cannot predict how much money we will need for retirement. Frank, Mitchell and Blanchett state, “Essentially, the retirement pot of money can only support so much. To some extent, the cash flow during early retirement years can provide a sustained cash flow into later retirement years, should the retiree continue to survive. This transition between the two life stages generally occurs during the retiree’s mid-70s. The retirement pool is much like a candle, which can only give out so much light based on the amount of wax and wick. Burn it brightly early and you’ll get less light later.”
The authors indicate that three levers exist: portfolio volatility, distribution period and sequence risk. Portfolio volatility refers to how much a portfolio increases or decreases given changes in the market. Distribution period refers to the numbers of years one lives in retirement; and sequence risk is the risk of a sudden decline in the market that would cause withdrawals to be lower. The authors also suggest a fourth lever: superannuation, which indicates that the longer a person lives, the longer a portfolio is needed, but with increasingly shorter time periods.
To contend with all four levers, the authors suggest two remedies: to annually use longevity percentile tables, and to adjust withdrawal rates by subtracting a joint (for couples) life expectancy of 1/n, where n is the number of years of life expectancy.
The article describes a process that helps smooth out withdrawal rates as individuals transition from an early age with high withdrawal periods to an older age with substantially lower withdrawal periods.
The biggest finding is that withdrawal rates need to vary based on the individual’s age and that the older an individual is, the more money he will require in the future.
The authors present a question that is worth further research. On one hand, they indicate that the portfolio allocation has the least impact on changing withdrawal rates. But sequence risk, the impact of a sudden decline in the market, has the greatest impact on withdrawal rates.
Certainly, the rate of returns from one year to the next is unknowable. However, in the 1998 study, Valuation Ratios and the Long Run Stock Market Outlook, John Campbell and Robert Shiller indicate that sequence risk increases in probability as valuations grow higher. Consequently, one could change the portfolio allocation to mitigate the impact of sequence risk. The idea of changing portfolio allocation based on valuation levels to reduce sequence risk is worth a second glance.
Overall, the authors raise an important question: should we vary withdrawal rates based on annual life expectancy, rather than fixed periods of time like 20, 30, or even 40 years out?
For a financial planner, alternatives exist to provide a process for taking withdrawals and providing an annuity that may cause either loss of control of the portfolio or an additional cost, but may provide security in a specific glide path of withdrawals that need little oversight.
The benefit of the research provided by Frank, Mitchell and Blanchett is that advisors can provide clients with options, such as tying up money in an immediate annuity, working to gain benefit from a variable annuity, or having more flexibility with a process in order to vary withdrawal rates based on annual life expectancy.