We love simple rules. It’s how our brains are wired. We use heuristics or mental short cuts to solve complex problems. Chances are you have heard of the ‘4% Rule’ for retirement spending. The rule states that ‘Assuming a minimum requirement of 30 years of portfolio longevity, a first-ear withdrawal of 4%, followed by inflation-adjusted withdrawals in subsequent years, should be safe.”

Rules can be very helpful, until we apply them to situations that don’t fit the rule. We get a false sense of security and possibly bad outcomes. Running out of money in retirement would be a very bad outcome.

Where Did the 4% Rule Come From?

William Begen wrote a paper called ‘Determining Withdrawal Rates Using Historical Data’. The paper was published in 1994 in the Journal of Financial Planning. The rule has been widely quoted and used by many retirement planners for years.

The analysis was very good for its day. Begen used historical returns and various equity allocations to gauge the likelihood that a 30 year retirement could be fully funded. For this, credit is due.

However, the assumptions in the analysis may not hold today. Let’s break it down. First, the analysis used historical return rates. What happens if future returns are lower? This model won’t give the correct answer. The model also assumed a 30 year retirement span. People are living longer and many hope to retire early. This assumption doesn’t necessarily hold. Finally, the measure of success used was that ‘most’ wouldn’t run out of funds in retirement.

Is There a Better Model for Estimating Retirement Income?

One of the best papers on the topic comes from Morningstar Research, ‘Optimal Withdrawal Strategy for Retirement Income Portfolios’. The authors are David Blanchett, Marciej Kowara and Peng Chen. The paper was published in 2012.

The paper described both simple rules-based withdrawal strategies as well as more complex strategies. The paper gauged the success of the rules by defining success based on what they call the ‘Withdrawal Efficiency Rate’ (‘WER’). The math behind the WER goes beyond this discussion. What is important is how ‘success is measured’. The WER defines the best strategy to be one where we spend as much as allowed such that at our death we have either zero or some pre-determined amount (for example if we want to leave assets to a charity or children). Furthermore, it builds in a penalty for running out of funds. The penalty for running out of funds is higher than for the penalty of having under spent available funds. In addition, the WER calculation can take into account both changes in investment returns and the chances that we live longer (or shorter) than expected. These are all major improvements over the original analysis used from the 4% Rule.

Which Strategy is the Best Overall Withdrawal Strategy?

Using the WER, the highest score is found with what the authors call ‘Mortality Updating Failure Percentage’. This sounds complicated, but what it really means is that the percentage withdrawal for any given year in retirement is calculated each year based on the current portfolio value, return expectations and changes to longevity given our current age. As a refresh, lifespans work in a non-intuitive way. The older we get, the higher the likelihood we will live longer.

So the idea that our withdrawal rate in any one year should be based on our age, current portfolio value and lifespan makes sense intuitively, but the number is a variable that changes year by year. This isn’t something we can easily wrap our heads around. How is it calculated? Is there any way to simplify the rule?

How is Mortality Updating Failure Percentage Calculated?

Unfortunately, the analysis requires software that includes monte carlo simulation. Most people don’t have access to this. As an RIA, we have the software and we include the numbers with annual financial plan updates.

Is There a Simple Rule That Comes Close?

We started out the post with the idea that our brains are wired for simple rules. They are much easier for us to internalize. As it turns out, the authors of the paper included the ‘RMD Rule’ as an alternative to the complete ‘Mortality Updating Failure Percentage’.

RMD Rule

The RMD Rule is almost as simple as the 4% Rule. To determine the amount we should withdraw each year for retirement spending, take the portfolio value at the start of the year and divide by the number of years remaining in our lifetime. Value / Years of Life.

Let’s say a single male age 65 has a lifespan expectation of age 92. At the start of the year, the portfolio is worth $2M. The withdrawal amount for next year would be = $2M / 27 yr = $74,074. This is equal to 3.7% of the portfolio. In this case, close to the 4% rule.

How does the RMD Rule address changes in return, inflation and lifespan? First, each years return impacts the balance. Say a balanced portfolio sees a drop of 10%? Then the next year withdrawal will go down. If we think about our retirement plan this way, we think about building a plan where the fixed spending is 70-80% of the plan and the variable spending could absorb any years with bad return. For example, in a year following bad returns cutting back on extra vacation travel or deferring new purchases such as a new car or home remodels would make sense.

Inflation is addressed by calculating the RMD from the base portfolio value. To ensure that our withdrawals are growing at the rate of inflation, we need to build a plan where the return is ‘safer’ and has a good chance of beating inflation. This suggests biasing the portfolio to include assets which have inflation exposure. Treasury Inflation Protected securities could be used. Additionally, stocks over the long run tend to exceed inflation (with higher volatility) so maintaining an equity allocation could be prudent.

How Much Equity in Retirement?

Based on Bengen, he asserts, ‘It is appropriate to advise the client to accept a stock allocation as close to 75% as possible and in no cases less than 50%’. Meanwhile, Blanchett et. al. don’t address the question directly, but their results show that the highest efficiency in most comes with an equity allocation in the 40% range.

Why such a difference? Recall that Begen was measuring success based on ‘most cases’ of covering retirement needs. In the case of Blanchett et. al. they penalized outcomes in cases where the retiree would run out of money in retirement.

What is the right answer? According to Blanchett, they note in their work that the amount of the penalty assumed for running out of funds doesn’t dramatically change their results. Given that most retirees could be considered more risk averse then risk seeking, an allocation that is closer to 40% would be more appropriate. Furthermore, even in the case of Blanchett, the results were derived using long term returns. Future risk and return assumption would also impact the final allocation.

Conclusions

Given the analysis, using software with annual updates to determine future retirement withdrawals is recommended. Be sure your advisor is updating your financial plan annually and calculating an update to date withdrawal rate.

However, for planning purposes the RMD rule with a 40% equity allocation should provide a better outcome then using the simple 4% Rule.

By way of example, using the RMD Rule estimating how much you need for retirement is as simple as taking your retirement spending goal x estimated years until death.

The 4% Rule would be replaced with the RMD Rule. To determine how much to withdraw in any year of retirement divide portfolio value by year of life remaining. Twenty years in retirement results in a 5% withdrawal rate. 30 years results in a withdrawal rate of 3.3%. 40 years results in a withdrawal rate of 2.5%. While the RMD rule results in a variable rate, the one that really matters is the one that you start with at retirement. It will give you the best estimate for planning retirement income.

 

Sources

Bengen, W.P. 1994. “Determining Withdrawal Rates Using Historical Data.” Journal of Financial Planning, vol. 7, no. 1 (October): 171–180.

Blanchett, D., M. Kowara, and P. Chen. 2012. “Optimal Withdrawal Strategy for Retirement Income Portfolios.” Retirement Management Journal, vol. 2, no. 3 (Fall): 7–20.