Practical analysis for investment professionals
28 August 2018

Rethinking the 4.5% Rule

Is the standard for calculating a retirement portfolio’s maximum withdrawal rate all wrong?

In “Rethinking Retirement Rules,” Reshma Kapadia challenged the idea that a 4.5% spending rate should be applied consistently over an extended period.

The root of the problem, of course, is that we are in a protracted period of historically low interest rates.

This means that expected returns may fall below historical averages, especially in the short to intermediate term. So setting a single withdrawal rate for a 65-year-old retiree and assuming no adjustments will be needed is imprudent at best and reckless at worst.

Asset class returns, volatility, and correlations work together to create many different possible return combinations over the retirement period. We created a Monte Carlo statistical simulation to determine what spending rate best serves retirees should they encounter low returns for an extended time.

Even if returns are close to the long-term average over 30 years, we sought to home in on the question: What happens if retirees experience lower returns in the early years?

Model Assumptions

To test the utility of a 4.5% spending rate, our Monte Carlo simulation generated 2,500 iterations for each of our scenarios. We made the following assumptions in accordance with our proprietary models:

• A 65 year old invests their \$1-million portfolio in a taxable account.
• The retirement period spans 30 years based on a 95-year life expectancy.
• The dollar amount of spending increased by 2.5% inflation each year.
• The spending rate was calculated by dividing the dollar amount of the first year’s spending by the portfolio value.
• The annual expected investment return was 7%.
• Long-term returns for all asset classes remained in line with long-term historical averages.
• Annual volatility of return was 9%.
• The overall portfolio carried 90% tax efficiency.

Scenarios

We tested three base case scenarios as well as a fourth “correction” to the third scenario.

1. Average Returns, Consistently Achieved

The investor received a 7% annualized return, with 9% volatility, over all 30 years.

2. Higher Returns in the Early Years

The retiree generated an 8% annualized return, with 9% volatility, during the first 15 years, and a 6% return, with 9% volatility, over the next 15.

3. Lower Returns in the Early Years

The investor received a 6% annualized return, with 9% volatility, in the first 15 years, followed by an 8% return, with 9% volatility, for the subsequent 15.

4. Lower Returns in the Early Years, But with Spending Adjustment along the Way

Results

1. Average Returns, Consistently Achieved

The 4.5% rule succeeded 80% of the time, a result in line with Kapadia’s analysis.

2. Higher Returns in the Early Years

Spending down at a rate of 4.5% meant the retiree had an 88% chance of not outliving their nest egg.

3. Lower Returns in the Early Years

This scenario worked only 71% of the time, which is problematic since a conservative financial plan should have at least an 80% chance of success.

4. Lower Returns in the Early Years, But with Spending Adjustment along the Way

Achieving the 80% success rate required a spending reduction of 6%, from \$45,000 to \$42,300, or a 4.2% rule rather than a 4.5% rule.

Conclusions

In the base case, our results correlate with Kapadia’s hypothesis that the 4.5% rule still holds if returns stay consistent over the retirement period.

But what if volatility comes into play?

Assuming that returns over the 30 years are in line with long-term historical averages, a sustained period of low market returns early in the retirement period may warrant a spending rate adjustment for a retiree drawing down assets.

By our estimates the reduction called for could be over 5%, or the difference between a spending rate of 4.5% and 4.2%.

This suggests that, given the current interest rate environment, retirement planning should be conducted in a dynamic fashion that takes into account reasonable estimates about market volatility and asset class performance,

Advisers should watch closely for the opportunity to adjust the spending rate to ensure their clients don’t run out of savings in retirement.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

For appropriate disclaimers, please go to: https://www.fcequities.com/legal-notice

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Share On
Fred Sloan, CFA

Fred Sloan, CFA, joined FCE Group in January 2011 and serves as president and chief executive officer. He has been involved in the management of multi-strategy investment portfolios since 1990. In 2007, He founded Island Brook Partners LP, an alternative fund of hedge funds. From 2001 through 2007, he was president of Kemnay Advisory Services, which serves as investment adviser to a group of offshore clients with global hedge fund and equity portfolios. Prior to Kemnay, Sloan worked at Ivy Asset Management from 1990 through 2001. During this time, he held the positions of chief financial officer, director of research, and chief of investment management. He received a bachelor of science degree in accounting from the State University of New York at Binghamton (1978). Sloan is a CFA charterholder and serves on the Binghamton University investment committee.

Carl Friedrich, CFA

Carl Friedrich, CFA, joined FCE Group in 2014, and serves as senior managing director. He has an extensive background in portfolio management and client advisory services, having previously served as chief investment officer of Piermont Group. He helped launch the Equities Relative Value Team at Goldman Sachs, where he advised and traded for hedge fund clients from 2001 to 2009. Earlier in his career, Friedrich also worked as a derivatives trader and investment banking analyst. He holds an MBA from the Kelley School of Business at Indiana University and a bachelor of science in Economics from the Wharton School at the University of Pennsylvania. Friedrich is a CFA charterholder, Financial Risk Manager (FRM), Enrolled Agent (EA), and Certified Financial Planner (CFP).

5 thoughts on “Rethinking the 4.5% Rule”

1. Ian Cully says:

I appreciate the concise form in which this article is written. The ultimate truth is that when it comes to retirement spending, the rules of thumb that are often relied upon are not always applicable to the individual.

For example, the safe withdrawal rate plan, by design, would ultimately distribute the greatest amount of assets later in the clients life. Generally, a retiree would be spending in the exact opposite pattern. People, while still young into their retirement, would be more inclined to travel and spend on discretionary goals. This is more so than a senior, even with their increased medical costs. This spending scenario can be compared to your Lower Returns in Early Years scenario.

A goal-based financial plan through Monte Carlo is a great way to account for sequence risk and measuring the true potential success of a client’s spending goals.

2. I’m having difficulty reconciling your assumption of a 7% net expected return with an asset allocation that would be appropriate for a 65 y.o. investor contemplating retirement.

When considering the reasonableness of this figure, we need to consider that most developed capital markets have reached maturity, meaning that expected returns will almost certainly be lower than historical returns.

When we frame the issue of lower expected returns as being uniquely connected to lower interests, we ignore other important trends such as aging demographics, rising valuations, moderating volatility, and low inflation.

When all these influences are viewed synoptically, the best-fit explanation is that economies and capital markets go through phases of maturity. If this hypothesis is correct, then extrapolating expected returns from historical returns is like extrapolating developed market returns from emerging market returns: a downward adjustment is required to account for risk differentials.

Notice that under this hypothesis the source of the adjustment is not transient in nature, its structural: Expected returns are lower because they are simply not as risky as historical returns.

By extension of this logic, there is no good reason to believe that expected returns will revert to their historic averages at some point in the future.

Indeed, I believe that such logic is macro-inconsistent: For it to be valid, interest rates and volatility would have to continue trending towards zero and valuations would have to keep trending towards infinity.

Of course, my hypothesis could be proven entirely wrong, only time will tell.

My point is this: When planning for retirement, investors should behave as though this worldview was correct. If they are proven wrong, they can always make a Personal Financial Adjustment (PFA) to spend more, donate to charity or leave an estate.

Making a PFA in the other direction is far more difficult and less enjoyable.

1. Exactly right.with 50/50 bonds equities prospective returns will be 4.5 percent max pre fee pre tax so 7 is ridiculous.

2. Justin Baiocchi says:

I agree that 7% net is probably a tad high (maybe we’re all suffering from recency bias), but I’m not sure I agree with your statement:

“…we need to consider that most developed capital markets have reached maturity…”

I suspect people in 1800 or 1900 ,may well have been saying the same thing, but of course there has been significant development since both those dates. How can we believe that capital markets have reached maturity when we can’t even fathom what the world will look like in 40 years, or 80 years?

3. Mark Tapley says:

I was puzzled by the 7% figure too. The spending increased with assumed inflation, but it wasn’t clear whether the 7% was a nominal or a real return. Please clarify.
Here in the UK, pension funds enjoy a substantial ‘heritability’ advantage over other ‘wrappers’ – they can be left to the grandchildren and untaxed in their hands until drawndown, and do not count towards their own capped pension funds. So many here think of their pension as the very LAST thing to touch. (It’s called the location withdrawal problem by Sharpe).
They also think in permanent endowmnet terms. This means a) stick to real assets, b) spend no more than 3% to 3.5% of the moving average of the market value over twelve trailing quarters. This should leave the capital value at age 95 about where it was at age 65, and the grandchildren put your portrait over the mantelpiece…

﻿