How much can we spend in retirement? We have debated the money side of retirement for a long time, oscillating weekly between the need of triple-checking everything in our massive Excel simulation, and the desire to just say “Let’s just do it and play it by ear”. Looking back, I think we have spent too much time thinking about this. At the end of the day, the plan is simple: we stop working, and use our savings to fund our retirement. Yeah, what could possibly go wrong?
Just went over my bank account and figured out I can live comfortably without working for the rest of my life as long as I die on Thursday
There is, of course, no way to predict the future. So the best thing to do is to turn to researchers, and see whether someone can give us an answer, or at least a pointer. Of course, the most famous person to turn to for answers is Bill Bengen.
Bill Bengen and the 4%
I just wrote this title and thought: if I ever learn how to play an instrument, I’ll put together a Rock’n’Roll band made of retirees and will call it “Bill Bengen and the 4%”.
Beside making for a great Rock’n’Roll band name, Bill Bengen and 4% are forever linked in the mind of early retirees. In 1994, this (then) unknown american accountant crunched lots of historical data, and determined that the ‘safe’ withdrawal rate over 30 years is 4%.
In his own words:
Assuming a minimum requirement of 30 years of portfolio longevity, a first- year withdrawal of 4 percent, followed by inflation-adjusted withdrawals in subsequent years, should be safe.
Bengen’s also checked other rates, before settling on 4%:
- 3% (or even 3.5%) was found to be an ‘absolutely safe‘ rate, as every retiree would not run out of money for at least 50 years
- 4% would make money last for at least 33 years
- 5% would work in many scenarios, but not all. For example, people beginning their retirement in the late 1960s and early 1970s would have only preserved their money for 20 years due to market crashes and other economic events.
- At 6%, a good 60% of retirees would run out of money before 30 years.
There are a few notes and caveats in Bengen’s paper, most notably:
- Bengen’s goal was to avoid for the retiree to run out of money within 30 years, so ending with a portfolio worth just $1 on year 31 was considered a success. This could be risky for early retirees who may want for their money to last for 40 or 50 years (or more)
- Tax is not considered in this study, so the income is gross and tax payments may reduce it.
The Trinity Study
In 1998, three professors from Trinity University, Texas, published a study (“Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable”) which was very similar to Bengen’s, although it looked at the probability of how long savings would last at a specific withdrawal rate.
Did this study confirm Bengen’s findings? Yes, more or less. According to the Trinity study, a 4% withdrawal rate has a 95% chance of not running out of money in 30 years (which increases to 98% if the portfolio is 75% stocks).
Why 95% instead of 100%, as Bengen found? The difference is probably explained by these two factors:
- The trinity study used longer historical data (from 1926 to 1995)
- The type of assets used is slightly different: Bengen used Treasury Bonds; The Trinity study used Corporate bonds.
Trinity Study gets updated
In 2009, a researcher named Wade Pfau replicated the Trinity study, using even longer historical data: 1926 to 2009. He wrote:
I think people are sometimes curious if [Trinity’s] result may have changed if the study were repeated today, especially in light of the recent financial crisis. The simple answer is: no.
According to Pfau, the extended data places the success rate of the 4% withdrawal rate at 96%, 1% more than the Trinity study.
In 2011, the Trinity team updated their own research. They confirmed Pfau’s findings (96% success rate for a 4% withdrawal rate), and also increased this to 100% for portfolios which have 75% shares.
Bengen updates Bengen
Bengen himself, updating his own research in 2001, introduced a twist to his own 4% rule. Instead of a fixed 4% withdrawal, he suggested using a variable amount based on the previous year’s market returns. This approach, which he called “Floor and Ceiling” system, would work like this:
- You make your initial withdrawal. Say, $40,000.
- In a bull market, the next annual withdrawal can be 25% higher ($50,000)
- In a bear market, the withdrawals will be 10% lower ($36,000).
- In both cases, the increases (or decreases) are calculated on the initial withdrawal adjusted for inflation.
This approach, according to Bengen, increases the safe withdrawal rate to 4.58% for 30 years. In a recent and very interesting Reddit AMA (Ask Me Anything) he commented:
The average safe withdrawal rate […] is, believe it or not, 7%! However, if you experience a major bear market early in retirement, as in 1937 or 2000, that drops to 5.25%. Add in heavy inflation, as occurred in the 1970’s, and it takes you down to 4.5%.
Bengen also added:
As your “time horizon” increases beyond 30 years, as you might expect, the safe withdrawal rate decreases. For example for 35 years, I calculated 4.3%; for 40 years, 4.2%; and for 45 years, 4.1%. […] If you plan to live forever, 4% should do it.
The “4% rule” is extremely popular, and for a good reason: it is easy to understand and implement. It has been tested successfully on historical data, but is it really safe for us who – hopefully! – will need to keep withdrawing for 50-60 years?
Unfortunately there isn’t enough research covering this extended period of time. However, a 3% withdrawal rate is usually considered to be ultra-safe in any scenario.
The Family’s take on the 4% rule
Well…we totally buy the reasoning and the results, but as as usual, one thing is the theory, one thing is the practice! We want to be able to sleep at night, so we consider Bengen’s 4% a sort of upper limit, something we never want to go above. For our retirement we are targeting a 1 to 1.5% rate, so there is room to double that and still and be at 3%, which is still considered ultra-safe.
All of this is very academic, though. There is only one way to know whether it works: it’s to try it!