Questions on the 4% “safe” withdrawal rate
I have over 250 posts on the blog stretching back over the last 18 months, and you readers would only notice a comment on one of the old posts if you happened to subscribe to the “Comments” RSS feed. (It’s the orange icon over there on the right, under the “Please leave a comment!” text.) However, WordPress notifies me whenever a comment’s added to an older post, and a reader’s asked good questions on the thread “How many years does it take to become financially independent?”
Ari Lamstein of HubPages pointed out:
This is a great article. After reading this and numerous other mentions of the “4% Rule” in the early retirement community I decided to do some research on it. There is an important caveat to their research that I think you should mention: their “experiments” were only conducted over a 30 year period. And many of their sample portfolios “failed” (that is, ran out of money). I wrote an article summarizing the applicability of the 4% Rule to early retirement that you might find interesting: The “4% Rule” and Early Retirement.
Ari’s referring to the Trinity Study’s examination of withdrawal rates. One of their mainstream portfolios of 50%/50% stocks/bonds used a 4% withdrawal plan over 30 years, and during the years of 1926-95 a retiree’s portfolio would have survived that withdrawal rate for 30 years about 95% of the time. (In other words a couple tests ran out of money, hypothetically while the retiree was still alive.) The key to portfolio survival appears to be market conditions during the first 5-10 years of retirement. This was no problem for someone who retired in 1982 or the 1990s, but it’d be a significant concern to someone who retired in 2000.
The 4% “safe withdrawal rate” rule developed from the Trinity Study has been discussed to death on websites like Early-Retirement.org and Bogleheads.org. Of course it’s also mentioned in the book “The Military Guide to Financial Independence and Retirement”.
Yesterday’s post mentioned Wade Pfau’s research to extend the 4% SWR to longer retirements with higher survival rates. Two other posts examine the 4% SWR in more detail:
“Is the 4% withdrawal rate really safe?”
Details of the 4% safe withdrawal rate
While the Trinity Study looked at periods of 15-30 years, Early-Retirement.org & Bogleheads.org threads have discussed extending it out to 50 years. The issue with extending the retirement period using Trinity Study data is that there just aren’t enough rolling 50-year periods to give the historical method any validity. (The Trinity Study also ignored Social Security benefits.) For extended retirements, Monte Carlo simulations will run thousands of scenarios to cover any length of retirement. However, those simulations have their own flaws:
Problems with retirement calculators
Retirement planners and calculators
With those calculator flaws in mind, one early retiree ran his own simulation out to 55 years. Of course a financial analyst only needs to point out that future returns might be lower than those of the 20th century, so historical simulations are irrelevant.
Again, there’s not enough historical data to justify the analysis of retirements longer than 30 years. A few statisticians would claim there’s not even enough data to go that long. Monte Carlo simulation is the usual way to study retirements past 30 years. This algorithm is used by FinancialEngines.com, Fidelity Investments, USAA, and other financial firms. In general, though, its flaws tends to make it more conservative than a historical calculator like FIRECalc.com. Most early retirees run at least two different calculators and then try to decide how they want to adjust their budget, or how much longer they want to work for a paycheck.
The top three issues studied on longer retirements are (1) the trend for most retirement spending to decline with age, (2) the tendency for retirees to cut their spending during bear markets, and (3) including annuities like Social Security into retirement financial analysis. Wade Pfau’s research is blazing new trails, and his blog is worth subscribing to.
Ari went on to ask:
I’m curious, do you know of any other approaches to financing an early retirement? For example, I’ve heard some people mention trying to live off dividends, but I have not seen any academic research on this.
There are as many early retirement approaches as there are retirees, and each one is convinced that their system is the best. An early-retirement approach may have a high survival rate, but it still only works if the early retiree has faith in it and doesn’t abandon it at the first sign of a bear market. Let me mention a number of other approaches that I don’t write about very often.
Jacob Lund Fisker’s Early Retirement Extreme uses very high savings rates to build a portfolio of equities that will grow their dividends at least as fast as inflation. The idea is to build an inflation-fighting stream of income that meets a minimalist budget. The extremely high savings rates come from frugal living: smaller/cheaper housing with roommates, preparing low-cost meals, walking & bicycling, high-deductible health insurance, and solving problems without spending money on them.
A few ERs use rental property & online businesses to build up a diversified income stream. The idea is to generate as much passive income as possible with as little ongoing effort as possible. This should be built up over time– you don’t just rush out to buy an apartment building and start blogging about the experience. (Pat Flynn is one of the best at online passive income at Smart Passive Income, but he’s working crazy hours to accomplish his personal goals.) One ER turned her retail passion into buying kid’s clothes at garage sales and reselling them on eBay. It’s tough to earn a living at that, but you can certainly earn a retirement at it. I’ve learned that blogging can turn into a full-time job if you’re not careful, and landlording can also cut into your surfing time unless you spend some profits on a property-management company.
Other ERs build up a portfolio of dividend-paying stocks that are expected to keep growing their dividends, like the Dividend Growth Investor’s blog. A passive approach to this active investing would hold shares in a dividend-paying asset like the iShares Dow Jones Select Dividend fund.
John Greaney, who retired in his 30s, points out that investment expenses are just as important as living expenses. If you’re relying on high-cost actively managed funds or (even worse) paying 1% annual fees to a financial advisor, you don’t have much left over for your retirement. John’s also been following seven different investment portfolios for over a decade to show how real-life recessions affect their returns.
Finally, I’ll conclude with the “rational investing portfolio” proposed by Bob Clyatt in his book “Work Less Live More”. It uses 16 diversified funds covering nearly all asset classes, with a tilt toward small-capitalization and value-stock index funds. Where Bob’s portfolio really survives, however, is its “4%/95%” variable withdrawal system.
Bob advocates withdrawing a straight 4% of the portfolio every year, not the Trinity Study’s system of “initial 4% the first year and rising each following year with inflation”. While the Trinity Study system gives a retiree a constant spending amount (adjusted for inflation), Bob’s 4% withdrawal varies each year with portfolio performance. This is no problem when the stock market rises faster than inflation. However, the retiree starts to lose spending power when the stock market is flat. Even worse, the retiree takes a big spending cut during bear markets.
To soften the blow of bear markets, Bob introduced one of the first variable-spending retirement plans. If 4% of that year’s spending would be less than 95% of last year’s spending, then the retiree can use 95% of last year’s spending. During bear markets this raises the portfolio withdrawal rate to 5% or even 6% for a year or two, but when the stock market recovers the portfolio will too. Bob hired a financial analysis firm to backtest the portfolio against available market history.
Bob also advocates occasional part-time work during retirement. (Bob is an amazingly talented sculptor.) His thought is that early retirees will still find value in entrepreneurial self-employment, freelance writing, consulting on a project, or working a few hours a week at a retail store. (I’ve only been retired for 10 years, but I’ve had at least one job offer per year during that time.) Not only will it boost self-esteem (or at least affirm the enjoyment of retirement), but it’ll help retirees survive the emotional stress of watching a portfolio struggle through a bear market. Someone who ERs in their 40s would usually be able to consider working well into their 60s if they chose to, and that would put them within range of starting the Trinity Study’s 30-year retirement with nearly a 100% survival rate.
Ari, feel free to post follow-up questions here. Anyone else have a comment or a question?
Does this post help? Sign up for more free military retirement tips by e-mail, Facebook, or Twitter!