Earn CE credits
“Perpetual distribution rates” is eligible for CE credits, see Accreditation details for more information
It might be a friend from your church, an acquaintance at a charity or a foundation, or even a concerned HNW client. Sooner or later, someone will ask you, “Can my portfolio last forever?” This continuing education course will help you answer that question.
One of the challenges of distribution planning is estimating the proper amount of withdrawals so market, longevity and inflation risks are covered for the given time horizon. In this article, we are considering a perpetual time horizon, which makes the outcome significantly more sensitive to the remaining two risk factors—market and inflation risks. Assets must be large enough to finance not only the distributions, but also provide a sufficient cushion to overcome the effects of these two risk factors in perpetuity.
If you are a retirementincome specialist, you already understand the concept of sustainable withdrawal rate (SWR). When preparing a retirement plan, say from age 65 to age 95, you might say something like “the sustainable withdrawal rate for a 30year time horizon is 3.5%.” What that means is if the client has $1 million earmarked to finance his retirement expenses, he can withdraw $35,000, indexed to CPI annually, and at age 95 he’s likely to still have income.
The perpetual withdrawal rate (PWR) is similar to SWR. The only difference between sustainable and perpetual is the SWR is for a specific time period. The PWR has no time limit attached to it; in theory it is supposed to last forever. Mathematically, forever is too long of a time horizon, so we define it as:
A portfolio is considered perpetual if, after 40 years, the asset value of the worstcase portfolio is not lower than its starting amount, and no portfolio depletes during this 40year time horizon.
We use the actual market history to calculate the perpetual withdrawal rates. We call this methodology aftcasting. It reflects the actual sequence of returns (stocks, interest rates and inflation), the actual correlation between stocks, interest rates and inflation, and actual volatility as they occurred historically. It displays the outcome of all historical asset values of all portfolios on the same chart (thin gray lines), as if a scenario starts in each of the years between 1900 and 2000. It gives a bird’seye view of all outcomes, and provides the success and failure statistics with exact historical accuracy, as opposed to manmade simulation models. The chart in Figure 1 depicts an aftcast:
Figure 1: An aftcast of perpetual portfolios
On the aftcast chart, we see “worstcase” and “median” lines. The red line indicates the worstcase portfolio performance. We use this line as our design criteria for the definition of “perpetual”. The last red point on the chart (at year 2051) must not be lower than the first red point (at year 2011). The blue line indicates the median outcome, where half of the scenarios are better and half are worse.
Keep in mind, while this is a perpetual portfolio, the aftcast shows in extreme cases, the drawdown can be as much as 50% of the initial portfolio value throughout this 40year time horizon.
Your client might have several different objectives for the capital and for the distributions—each one might be in nominal dollars, or in real dollars (indexed to CPI).
In addition, she may want to distribute part of the portfolio growth, with a natural tendency to distribute more when times are good (“ring the cash register”). In this case, the annual distributions consist of a constant base amount and a fluctuating amount. We call the percentage of growth we harvest and distribute each year the “participation rate.”
If the portfolio growth is zero or negative in the preceding calendar year, the fluctuating amount is zero and only the constant base amount is distributed that year. However, the math shows any degree of participation in the growth gives a higher total lifetime payout than having a constant distribution alone.
For example, using a 40/60 asset mix (equity/fixed income) and $10 million starting capital (not required to be indexed), we calculate a PWR of $281,000 or 2.8% with no participation in growth. The aftcast is shown in Figure 1.
When we use a 50% participation rate, however, then the PWR is 0.9% of the initial capital plus 50% of the portfolio growth during the preceding year. Note that, once you include part of the growth in the distribution, the upside of the portfolio’s growth is much more limited because of the larger amount of total distributions (see Figure 2).
Figure 2: The aftcast of PWR with 0.9% of the initial capital plus 50% participation rate
Optimum Participation Rate:
You might be tempted to use a high participation rate to harvest and distribute more of the growth. Resist that temptation and keep the participation rate between 15% and 25%.
This lower participation rate makes the distributed dollar stream more reliable while still providing a sizeable participation in growth. At your client’s next review, if you find the portfolio value is higher as a result of larger growth retention or simply a good market trend, then simply recalculate the PWR and start paying larger distributions.
If you want to smooth the floating component of the distributions over time, use a fiveyear moving average. Harvest the growth and place the money into a cash bucket. Each year, distribute 1/5th of this bucket (plus the constant portion). This allows a more even distribution over time without affecting the portfolio performance.
Optimum Asset Mix:
My earlier work (“Unveiling the Retirement Myth,” 2009) indicates that for distribution portfolios, an asset allocation of 40% equity and 60% fixed income is approximately the optimum mix. Therefore, we used that asset mix in our calculations. If you allocate a larger portion of the portfolio to equities, then you will likely be forced to reduce distributions significantly in the future.
Table 1 displays the perpetual withdrawal rates for different scenarios with different participation rates. The last column, total lifetime distributions, indicates the total amount of distribution over a 40year time horizon as a percentage of starting capital.
Table 1: Perpetual withdrawal rates (SP/TSX as equity proxy)
Scenario  Constant Portion  Floating Portion (participation rate)  Total Lifetime Distributions 

Scenario 1:

2.8%  0%  111% 
2.3%  15%  187%  
1.9%  25%  218%  
0.9%  50%  254%  
0%  71%  262%  
Scenario 2:

2.3%  0%  197% 
1.9%  15%  246%  
1.6%  25%  262%  
0.8%  50%  270%  
0%  71%  262%  
Scenario 3:

1.9%  0%  78% 
1.2%  15%  165%  
0.7%  25%  204%  
0%  38%  237%  
Scenario 4:

1.7%  0%  145% 
1.1%  15%  198%  
0.6%  25%  221%  
0%  38%  237% 
Another interesting scenario is found with certain foundations and/or charitable trusts that are annually required to distribute a minimum of 5% of the portfolio asset value of the preceding year. Figure 3 depicts the aftcast of this scenario.
Figure 3: The aftcast, minimum distribution rate 5% of the portfolio value at the preceding year end
In this case, the worstcase portfolio value (red line) declines over time. This violates our definition of perpetual, which says the end value (2051) of the worstcase asset value must not be lower than its starting value (2011). Even the median portfolio is barely flat.
To meet our definition of perpetual, we need to add cash to the asset pool over time. In this case, a minimum of 0.83% of the original capital plus all annual operating expenses must be added as new contributions to the pool of investments each year.
Table 2 shows the PWR using a constant portion and a floating portion. In this scenario neither distributions nor portfolio value are required to keep pace with CPI. The floating portion is the payout as a percentage of the portfolio value at the preceding yearend. We observe that once the floating portion exceeds 3% of the portfolio value, our definition of perpetuity is violated.
Just because this distribution strategy—withdrawing 5% of the asset value—does not meet our criteria of perpetuity, it does not mean it can’t be perpetual. When you remove a fixed percentage of the current portfolio value each year, the portfolio never runs out of money and it is perpetual. However, both the asset value and the dollar amount of distributions can shrink over time, to a level that can be significantly lower than the starting amount.
Table 2: Summary of perpetual withdrawal rates for Scenario 5 (SP/TSX as equity proxy)
Constant Portion (not indexed)  Floating Portion (% of portfolio value)  Total Distributions 

2.1%  1%  175% 
1.4%  2%  216% 
0.6%  3%  241% 
Review of Distributions:
There are two times to review distributions:

Every five years:
Recalculate the PWR anew and use this for subsequent distributions until the next review.
Since our PWR tables are based on the worstcase scenario, portfolio assets should likely be larger, triggering larger distributions.
However, we can never be sure the extremes of the future will be better than the extremes of the past. Also, we cannot know ahead of time how the fund manager will perform relative to the benchmark.
If the asset value is lower by more than 20% (excluding all cash flows) compared to the last review, a new (lower) PWR should be calculated and implemented. This is a very rare occurrence (historically, less than 1% of the time) and each of these paycuts will likely be followed by a pay increase in the subsequent review. However, it is a preemptive measure to cover our uncertainty about the degree of future market extremes; nothing more.

Interim reviews:
Undertake an interim review when your client experiences an unexpected cash flow that exceeds 10% of the current total asset value. At this point, you should calculate a new PWR. If this new PWR is lower, implement it immediately. If it’s higher, you can either implement it immediately or wait until the next periodic review.
Keep in mind market fluctuations do not trigger interim reviews; only significant cash flow events prompt these reviews. If there is a significant fluctuation of the value of investment assets, stay the course and wait until the next review to recalculate the distribution amounts.
Economic Assumptions and Forecast:
Since our analysis is based entirely on historical experience (aftcasting), you cannot impose any assumptions of future portfolio growth rates or future inflation into the calculations. Furthermore, periodic reviews mitigate the effect of any deviations from the historical experience. Our entire analysis is free of any forwardlooking economic assumptions.
Example:
XYZ Foundation has $15 million in assets, allocated as 45% stocks and 55% fixedincome. Annual administrative expenses (indexed to CPI) are $10,000. The foundation wants to distribute scholarships into perpetuity, but it is not necessary to index to CPI. It does not expect additional funding in the future. The administrator has the following questions:
1. What is the simplest distribution strategy?
The simplest strategy is to distribute a fixed amount each year, which is Scenario 1 with no growth participation. Note on asset allocation: If your equity allocation is somewhere between 30% and 55%, then use the PWR indicated in this article. If the equity allocation is outside this range, then you will likely have a smaller PWR.
2. How much can be distributed?
First, calculate the assets required to finance the administrative expenses. Since they are CPI indexed, we look at Table 1, Scenario 2. The constant portion of the distributions is 2.3%. We calculate the capital required to generate this cash flow: $10,000 / 0.023 = $434,783.
Now, we have $14,565,217 capital available to distribute, calculated as $15,000,000 less $434,783. We read on Table 1, Scenario 1 that we can distribute 2.8%. Therefore, we can distribute $407,826 each year (calculated as 2.8% of $14,565,217) until the next review.
3. Which perpetual distribution strategy would maximize total scholarship distributions over a 40year time horizon? (Do not exceed 25% growth participation rate for scenarios 1 through 4).
Use the “Total Lifetime Distribution” from each scenario:
Scenario 1 (Table 1): The total lifetime distribution without exceeding 25% participation rate is 218%. Scenario 2: (Table 1): 262%. Scenario 3: (Table 1): 204%. Scenario 4: (Table 1): 221%. Scenario 5: (Table 2): 241%
Answer: Scenario 2, paying a CPIindexed amount of $233,043 (calculated as 1.6% of the initial available portfolio asset value of $14,565,217), plus the dollar amount of 25% of the portfolio growth during each preceding calendar year, would likely pay the highest amount of distributions until the next review.
4. Five years later, at the next periodic review, assets are $19 million. The foundation hired a new parttime assistant and the annual administrative costs jumped to $26,000.
A prospective fund manager claims he can beat the benchmark index by 3%, if XYZ Foundation were to employ him to manage the investment portfolio.
How much can they expect to distribute next year, using the results from question #3?
A fund manager consistently beating the index is generally a good thing. The assets can potentially grow larger faster. However, when calculating PWR, ignore any expected or past performance figures of any portfolio manager. If the fund manager indeed outperforms the benchmark, this will be reflected in the portfolio asset value, which in turn can result in higher distributions after the next periodic review.
First, we calculate how much of the assets are needed to pay the increased administrative expenses on an ongoing basis. Since they are CPI indexed, we look at Table 1, Scenario 2. The constant portion of the distributions is 2.3%. The required capital to generate this cash flow: $26,000 / 0.023 = $1,130,435.
Now we have $17,869,565 capital available to distribute, calculated as $19,000,000 less $1,130,435.
XYZ Foundation can distribute $285,913 annually (1.6% of $17,869,565), plus the dollar amount of 25% of the portfolio growth during each preceding year until the next review.
In conclusion, the perpetual withdrawal rate is the twin of the sustainable withdrawal rate. The only difference between the two is the missing time horizon in the PWR. And because of that, the PWR is a lot lower than the SWR.
Now that you’ve finished reading, complete the exam to receive your CE credits. If your score is 85% or higher, send an email to jim@retirementoptimizer.com with your name, and proof of your score, to get a onetimeperadvisor free retirement calculator and a free pdf copy of Jim Otar’s 525page book “Unveiling the Retirement Myth.”
Jim Otar, CMT, CFP, is a financial planner, a professional engineer, a market technician and a financial writer. He is also the founder of retirementoptimizer.com. His past articles on retirement planning won the CFP Board Article Awards in 2001 and 2002. He is the author of “Unveiling the Retirement Myth – Advanced Retirement Planning based on Market History” and “High Expectation and False Dreams” Your comments are welcome: jim@retirementoptimizer.com
Next steps
 Take this testLog in and answer a multiplechoice exam about the material.
 A CE Certificate will be issued electronically through the website after you pass the exam with a score of 65% or greater.
 Accreditation details
 More CE courses