usaa-header-banner

Back to the Trinity Study

Today’s post will be a sensuous reading experience that only a financial planner or an economist could love.

If you’re just getting started on your own finances then you probably want to skip this post (!) in favor of reading the “Armed Forces Guide to Personal Financial Planning”, or “The Military Advantage”, or another book from the Recommended Reading list. At best this post will seem boring, at worst it’ll be intimidating.

But if you’ve been reading the research papers for a few years and you’re skeptical about the 4% Safe Withdrawal Rate, then here’s more research to feast upon. Don’t worry, the 4% SWR still seems to be a good number.

First on the list is a 2010 review and update of the original 1998 Trinity Study by its original authors. Their conclusion is that a success rate of 75% is more than enough if you’re willing to watch your portfolio and make mid-course corrections. They agree that retiring into a recession is more likely to need those corrections, but if there’s not a bear market within the first few years of ER then the portfolio should survive on autopilot.

They even go so far as to suggest that 4% is the low end of an inflation-adjusted SWR, and that a fixed rate of 7% (no increases for inflation during retirement, ever) will also work. They’ve limited their study to 30 years, though, so keep in mind the effects of inflation for longer than 30 years.

Financial planners (and most military) would point out that a 75% success rate is guaranteed to be a 25% failure rate. In combat, let alone retirement planning, 25% is usually considered unacceptable. However one of my favorite financial authors, Yogi Berra William Bernstein, has pointed out that predicting the future is notoriously difficult. He claims that any portfolio success rate higher than 80% for 40 years is meaningless, and seeking a false sense of security can prolong one’s workforce misery.

One of my favorite case studies is famed early retiree Raddr’s profile of the hapless “Y2K ER”. This hypothetical fearless (and oblivious) ER (Y2K, not Raddr!) left the workforce at the end of 1999 to enjoy a 4% SWR on a portfolio flush with the outsize returns of nearly two decades of bull markets. By 2005, after five years of blissfully raising his annual withdrawals for inflation, the warning lights were beginning to flash. By 2011 his actual withdrawal rate is approaching 10%, and even the stock market’s latest recovery may not be enough to save him.

Yet nobody would persist in such reckless spending behavior. Even if Y2K ER had held out until his spending reached 7% of that year’s portfolio value, he would have started to cut expenses or found part-time work. His apparently inevitable demise is a clear example of how human behavior differs from the computer studies.

Incidentally Raddr’s website has a number of computer simulations of withdrawal rates which also conclude that 4% is challenging but not unrealistic. His studies unearthed plenty of failures but he also clearly demonstrates the difficulty of forecasting five decades of returns from history or Monte Carlo simulations. Once we humans are aware of the hazards then we’re probably sufficiently vigilant and flexible to steer around them.

One example of that vigilance and flexibility is Bob Clyatt’s “4%/95%” approach from the book “Work Less, Live More”.  Bob’s system is one of the very few “variable withdrawal rate” approaches to retirement spending.  It looks at the portfolio every year and takes a straight 4%– no adjustment for inflation because the 4% is calculated at the beginning of every year.  As you can imagine, this method is wildly popular if the stock market’s rise exceeds inflation.  It’s not so cheery if the market is flat and inflation is rising.  Even worse, spending has to be cut when a bear market inevitably rears its fanged head– which is where the “95%” kicks in.  If next year’s 4% calculation is less than 95% of this year’s 4% calculation, then next year the ER has the option to take 95% of this year’s 4%.  This is a temporarily higher withdrawal rate, but it still temporarily reduces spending to minimize damage to the portfolio.  As the bear market ends and the portfolio recovers, ER spending can rise right along with it at the next 4% calculation.

In other words, when the market drops by more than 5% then your spending only has to drop a maximum of 5%.  This allows a lot of flexibility for bear markets like the Great Recession, where the 2008 S&P500 dropped 37%.

An intriguing study takes a different approach to “failure”. Instead of agonizing over the “right” success rate, it attempts to avoid failures just before they happen.  ERs check their portfolio each year and only start to worry if it begins to steadily decline. Then they start tracking the price of an annuity that would provide enough income for the rest of their lives– either a fixed annuity that they’ll be able to live within (despite inflation) or an annuity with a cost-of-living adjustment for inflation. If their portfolio shrinks to the cost of their chosen annuities then they declare “game over” and annuitize their income for the rest of their lives. This is similar to Otar’s approach of buying a ladder of single-premium immediate annuities when a retirement portfolio seems likely to fail.

Bottom line? The academic financial research peer-review process has produced a pile of studies that hover around the 4% consensus. No one can predict the future with 100% success, but 4% seems to be what fiduciaries call a “prudent risk”. The key is your own attitude. When your portfolio is big enough to apply the 4% SWR to your budget and declare your financial independence, then you have choices.

One financially independent choice is to keep working until you’ve had enough: the “you will know when it’s time to go” approach. (Caution: anecdotal evidence indicates that when you’re financially independent, the “had enough” part happens a lot faster than you might expect.) If you aggressively save and invest during your military career, then you can achieve this within 10 years– especially with a military retirement from active duty or the Reserves/Guard.

Another choice would be to keep working until your SWR is below the dividend yield of your portfolio, so that you’re unlikely to ever have to chew into the principal. (Your heirs & beneficiaries will certainly appreciate this approach, too.) It will eventually become a trade-off between your emotional security and your tolerance for the workplace. I hope that you’ll find an avocation you love, but even the “dream job” has plenty of unrelated dissatisfiers that will truly test your commitment to showing up for work. Especially if the surf is up.

A third choice is to race for the workplace exit, intending to take an “extended sabbatical” to decide what you’d like to do with the rest of your life. If you’ve stuck with your financial and emotional retirement planning then you know how to entertain yourself for the rest of your life without spending into bankruptcy. SWR studies are very conservative and make the general simplifying assumption that you’re not going to change your spending behavior. However you know that you’re going to check your spending over the years. It’s human nature to either cut back your spending during recessions, or to use the “fire sale” opportunity to buy bargain-priced assets. Either way you’ll enhance your portfolio’s returns (and its survivability), which is a type of behavior that’s very difficult to model on a computer for a research paper.

If you’re still skeptical that there’s anything truly “safe” about predicting withdrawal rates (after all, you’re the one who has to live with the results), then you can always default to Bud Hebeler’s “Analyze Now!” negative-feedback system.  It’s more work, it’s much more conservative, and your spending limits may chafe on you from one year to the next, but Bud is an aviator who takes comfort from knowing that you’ll always have your hands on the controls.

What about you? If you’ve been retired for a few years, have you had to modify your spending or your withdrawal rate?

Related articles:
Military retirement spending: how much will I need?
Military retirement: how much can I really spend?
How many years does it take to become financially independent?
Tailor your investments to your military pay and your pension
Asset allocation considerations for a military pension

Does this post help? Sign up for more free military retirement tips via e-mail, Facebook, or Twitter!

9 Comments
  1. [...] the “0% failure” numbers. Most efforts tinker with the asset allocation to figure out the best combination of survival rate and withdrawal amount. If you’re willing to take on more risks of failure (longevity, volatility, and loss of [...]

  2. [...] The planner uses Monte Carlo: “We run 500 simulations of various investment scenarios varying the expected rates of return, inflation rates and cash flows. After running these simulations, we determine the probability of success based on the percentage of successful projections. A chance of 75% or greater is optimal for reaching your goal.” You may be a little nervous about the other 25% of the projections, but retirement research indicates that really is OK.  [...]

  3. [...] 4% withdrawal system isn’t a realistic model of retiree spending, either. Retirees do not blissfully spend 4% and raise it every year no matter what.  People tend to spend less when the economy is bad and more when the economy is good. Older [...]

  4. [...] all aspects of the Trinity Study, including this one. I’ve surveyed much of that research at this post and there’s a huge archive of further discussion at [...]

  5. [...] will I need? Military retirement: how much can I really spend? USAA: seven money rules to break Back to the Trinity Study Saving base pay and promotion raises Dealing with “retiree [...]

  6. [...] goes down, and you can do this whenever you’re not comfortable spending your yearly amount. Another option is Bob Clyatt’s variable withdrawal plan of 4%/95%.  When you reduce your withdrawals during down years, your portfolio will be able to recover that [...]

  7. [...] Related articles: Military retirement spending: how much will I need? Retirement finances: what will I spend? Military retirement: how much can I really spend? Back to the Trinity Study [...]

  8. [...] 161 Back to the Trinity Study [...]

  9. [...] throw off an average 4% per year (before inflation & taxes!) for another 30 years. She could start withdrawing 4% the first year and raise it every year for inflation.  Meanwhile if her spouse died just after the life insurance expired, and her pension/Social [...]

Comment? Question? What's on your mind?