Withdrawal Strategies (without numbers)

Jim Williams

Decision architecture on withdrawal strategies for funding retirement from investment portfolios; how much, how to decide is written about extensively.  Most of this writing is rich in calculations, charts, graphs and other numerical presentation.  I think many readers find that kind of source material off-putting and lose sight of the forest by being surrounded by trees.  This essay will seek to explain a variety of withdrawal frameworks with a minimum of recourse to numbers.  It will outline the basics of a few primary withdrawal strategies and address the costs, risks and benefits of each strategy.

Constant Dollars adjusted for inflation - Hell or High Water

A seminal study of withdrawal rates was published in the early 1990's by Bill Bengen, a pioneer financial planner.  Bill used prior periods of market data to illustrate what he termed "a safe withdrawal rate" for a 30-year period.  The gist of the study was to start withdrawals at a certain dollar amount (expressed as a percentage of the starting portfolio) and then inflate that withdrawal dollar amount based on historical inflation for the same period and see what (starting) withdrawal rate gave the most money with a reasonable success rate (not running out of money).   The results of the study gave rise to the now-widely-used 4% withdrawal rate benchmark. (Editor's note: it is now nigh impossible to discuss portfolio withdrawal rates without using the words "four percent".)  This was groundbreaking work, and it formed the benchmark for much of the withdrawal framework study going forward: it took into account a variety of market conditions and it explicitly took inflation into account.  Moreover, the study put in place the very notion that one ought to have a framework and methodology for determining withdrawals from an investment portfolio in retirement.

As instrumental as the Bengen study was and despite the fact that we occasionally cite the study in our conversations with clients, I think it had one major flaw: people don’t live that way.   I think of the Bengen study as the "Hell or High Water" withdrawal method.  This means that the investor/retiree starts with a dollar number, inflates that number each following year for cost of living, and withdraws that number, come what may.  No reference to changes in lifestyle, no reference to balance in the portfolio, no reference to changes in life expectancy.  Just take the annually inflated amount, spend it, and go on to the next year.  I think it is, and would be, suboptimal to base one's withdrawal in 2020 only on information that was available in 2000 and simply ignore the intervening years and the information that might be brought to bear on the withdrawal decision for 2020.  Again, people just don't live that way.

Withdraw a constant percent of the portfolio

Rather than a fixed dollar amount increasing by inflation each year, this method extracts a fixed percentage of the portfolio each year.  Given that one can expect the portfolio to increase over time, this method can offer an increasing amount of money over time with virtually no risk of exhausting the portfolio.  The trick in this method is to set the withdrawal rate at a level that allows the portfolio to continue to grow.  That means when considering expected return, minus fees and other expenses, and minus withdrawal rate leaves something on the table for growth.  You can also chose a higher withdrawal rate that will put the portfolio into decline.  But as long as the percentage method is followed, the portfolio will never be exhausted.  Further, the amount of withdrawal will vary from year to year and will certainly decline in those years when the portfolio declines.

Withdraw an increasing percent of the portfolio based on age/life expectancy.

This is the essence of the Required Minimum Distribution (RMD) Method imposed on qualified retirement plans: IRAs, 401(k)s, etc.  The purpose of this method is to increase the percent of the portfolio withdrawn each year without exhausting the portfolio over a lifetime.  Think about it this way:  each year, withdraw the amount that if withdrawn equally over the owner's remaining life expectancy in years, will entirely liquidate the portfolio.  As the owner gets older and life expectancy declines (a decline of less than one year for each year lived, by the way), the number of remaining years declines, and the factor applied to the remaining portfolio goes up accordingly.  This is a quite sensible method of decumulation for a retirement portfolio.  The amount of the withdrawal will vary from year to year and grow in early years and decline in later years.  The decline takes over when the withdrawal rate exceeds the net expected return of the portfolio.

Withdraw only income; never touch principal

This approach to portfolio withdrawal is suitable in only very limited circumstances.  It can be well applied to money that is explicitly intergenerational in its purpose, or money that is explicitly permanent, such as an endowment.  Even in those cases, a unitrust concept (like the constant percent of the portfolio approach described above) might be more effective and more efficient.  As applied to a retirement portfolio, the reference to and focus on "income" will lead the investor to make suboptimal investment choices and position the portfolio in a poor risk/return profile; unnecessarily sacrificing growth for cash income or sacrificing cash income for more suitable growth.

Keep in mind:

• Although all spending is withdrawal, not all withdrawal need be spent.  Consider IRA Required Distributions; you have to take the money out (and pay tax on the withdrawal), but you need not spend the money.  
• It makes sense to apply different withdrawal methodologies to different components of the portfolio.  The only discretion on a retirement account is to withdraw the RMD or to withdraw more.  Other (non-tax-deferred) portions of retirement assets may be withdrawn under different schemes, including fixed dollar, fixed percentage, and ad-hoc.
• Not infrequently, account holders wish to augment their regular withdrawal with ad-hoc withdrawals for special purposes such as special travel, home renovation/improvement, gifting, charity, etc.   It is prudent in these cases to check in on the overall and cumulative withdrawal rate in the context of the overall financial plan.
• Crossover point - When the rate of withdrawal exceeds the rate of expected net total return in the portfolio, the portfolio will necessarily decline in value.  This decline may be intentional or unintentional.

 Reminder to get your quarterly Credit Report from:      TransUnion

The table below shows the returns through March 31, 2022, for selected investment asset classes.   In most cases, the results below are appropriate benchmarks for the related mutual funds in your investment portfolio.