Retirement planners and calculators (part 2 of 2)
This is the final post on the choices of retirement planners and their strengths/weaknesses. If you’re new to this blog (or if you’ve discovered this post through a search engine) then you might want to read about the problems with retirement calculators. They all have advantages & drawbacks (no matter how well they work or how user-friendly they are) but the first step in using these tools is understanding their limits. Then you should read about the most popular mainstream retirement planners.
The “problems” of FIRECalc and FinancialEngines are the same as all historical and Monte Carlo retirement planners: they don’t completely & accurately model investment history or persistence. But there are several other types of retirement planners that attempt to plug these holes or completely avoid the issues. These retirement planners aren’t as well-known but they cover areas that the mainstream planners don’t handle.
The first planner uses a variable withdrawal system instead of a fixed one. It’s not a stand-alone planner, although it can be included in FIRECalc. However it was extensively modeled by a team of financial analysts and an author who eats his own cooking. It’s Bob Clyatt’s “4%/95%” variable withdrawal plan from his book “Work Less, Live More”. It starts with the Trinity Study 4% rule, but it adds a variation for bear markets.
Most retirement planners assume a fixed withdrawal system (perhaps adjusted for inflation). For example, the Trinity Study withdrawal scheme starts at 4% and raises that amount every year by the rate of inflation. It’s constant (in “real” dollars) and it’s a lot easier for researchers to model on a computer. However it’s not the way that human beings operate– we tend to reduce our spending during bear markets, no matter what withdrawal the software claims is “safe”.
Bob’s 4%/95% system starts with a 4% withdrawal every year and isn’t adjusted for inflation. Every year you withdraw 4% from the portfolio. However if the market had a bad year then your portfolio might have lost quite a bit of value, and a 4% withdrawal might be a much smaller dollar figure than the year before. This could put an unacceptable crimp into a spending plan. (In 2008 the S&P500 dropped 37%. Nobody wants to cut their retirement spending by over a third.) The “95%” part of Bob’s system is a compromise– it reduces spending a small amount during down years by “allowing” the withdrawal to be as much as 95% of the previous year. That way if next year’s 4% withdrawal is less than 95% of last year’s withdrawal, you can take the 95% amount. It’s more than a 4% withdrawal, but it’s still less than the Trinity Study withdrawal. Because the 4%/95% system varies its withdrawals, even the 95% option gives a bear-chewed portfolio some recovery time before withdrawals begin to rise again.
Bob had a financial research firm analyze the 4%/95% system in more historical detail than the Trinity Study. They showed that it works with diversified portfolios of 50% stocks. Better yet, Bob’s recommended portfolios still grew well enough to keep up with inflation. Although retirees only withdraw 4% each year (no boost for inflation), the portfolio keeps up with inflation so the 4% withdrawal preserves buying power. Best of all, Bob’s variable withdrawal system had a 100% success rate out to 40 years (a third longer than the Trinity Study).
Bob’s analysis is so popular that the FIRECalc planner includes a 4%/95% variable-spending feature on the “Spending Models” tab.
Next, Bud Hebeler’s “Analyze Now!” website offers another variable withdrawal system. Bud has the same credibility as FIRECalc’s designers and Bob Clyatt: he eats his own cooking. He’s a retired Boeing finance exec who was not happy with the advice he received when he retired. He put together his own tools and these days, happily retired, he gives most of them away.
Bud’s an aerospace engineer, too, so he based his analysis on the “negative feedback” controls used in aircraft autopilot systems. His retirement planner is not only conservative, but each year of spending is based on the previous year’s expenses and the portfolio’s returns. Instead of depending on the history of the entire stock market (or a series of randomly generated returns), his planner relies on your performance history and asset allocation. It’s another variable system, but you can do the calculations each year to make sure that your portfolio survives a bear market. In addition, Bud offers a number of decision tools to determine when to begin drawing Social Security (hint: the later the better) and when to withdraw from an IRA or convert it to a Roth IRA.
Third, ESPlanner also uses a variable spending model, but it focuses on “consumption smoothing”. The idea is to take your retirement spending from the right types of accounts at the right time to minimize expenses & taxes. It also helps you plan out “lumpy” expenses like a child’s college degree, a fantasy vacation, or a new roof. You might recognize ESPlanner’s creator, Professor Laurence Kotlikoff, from his economics research and his books.
The program is available in a number of different configurations, from “free” (but very limited) to $250. It’s a program that only a retirement-planner geek could love, but the data entry is detailed and extensive. (It takes several hours. To us diehard retirement planners, that’s very reassuring.) You can tweak just about every retirement scenario: moving to a new state, downsizing your home, starting or delaying Social Security, converting a conventional IRA to a Roth IRA, and even running “What if?” scenarios on state/federal taxes or projected inflation. The program’s price is its biggest drawback.
Finally, but definitely not in last place, is Jim Otar’s retirement planner. His book, “Unveiling the Retirement Myth”, debunked a number of financial-planning philosophies and showed why investor psychology keeps many retirees from following through on their spending plans. Instead of trying to “fix” the retirees, Otar fixes the planning. Retirement-planning assumptions aren’t necessarily “wrong”, but they’re not correct 100% of the time. Many retirees are willing to risk their future on planners that offer “95% success” or “<1% failure”, but all of us want to know exactly how to avoid those unpleasant results. Otar focuses on the dark, scary corners of retirement planning where things can fall apart.
Otar’s retirement calculator has only been out for a couple of years, and it’s not as aggressively marketed as ESPlanner, but it may already be more popular. It costs $100 but it offers nearly as much data entry and flexibility in a spreadsheet format. It’s a historical calculator, so it tests portfolios against 20th-century markets like FIRECalc. However its most powerful feature is its analysis of retirement failures and ways to avoid them. Its most innovative feature (from his book) is an annuity calculator that recommends whether to insure some or all of your retirement income. It includes a “stop loss” feature that suggests buying annuities if the portfolio is in danger. It even lets users analyze various combinations of strategies like downsizing, cutting spending in bad years, or limiting spending to a certain percentage.
Alert readers may have noticed that I’ve skipped over some of the most widely used retirement planners on the Internet: Fidelity, Vanguard, and T. Rowe Price. These planners aren’t “bad”, but they use the same “approximations” as the mainstream historical or Monte-Carlo planning tools. They don’t adequately document their assumptions or thoroughly explain what happens when you change your inputs. You also have to put up with some oversimplifications and even some marketing.
The main advantage of these planners (including USAA’s planner) is that they’re easy to use. They’re very convenient, too– you can tap into your portfolio holdings and even consolidate assets from other brokerages. However their asset-allocation recommendations may be biased toward their own products. Regardless of how easily you can use them, you shouldn’t rely on these calculators as your only source of retirement planning. They’re all great basic tools that should be the foundation for using a more detailed planner.
How does your brokerage’s retirement planner stack up against the competition? If you’re a Vanguard customer then should you open an account at Fidelity or USAA just to be able to use their tools? Personally I think that FIRECalc or Bob Clyatt or Bud Hebeler offer enough information and reassurance to make a good decision. Some retirees (me among them) want to worry constructively by spending more time & effort on their planning. In that case Otar’s planner, FinancialEngines, or ESPlanner can help convince even the most skeptical & paranoid retirees. The “good” news about the detailed planners is that you will know you haven’t missed anything.
If you’re still curious how the various planners compare with each other, you can scan a detailed summary at the Bogleheads Wiki on retirement calculators.
Run as many different planners as it takes to make you confident that you’ve thought of everything. You won’t work any longer than you “need” to, you’ll know that you’ve planned for every contingency, you’ll worry less during your retirement, and you’ll enjoy yourself more!
Does this post help? Sign up for more free military retirement tips via e-mail, Facebook, or Twitter!