When planning for retirement income from your Thrift Savings Plan (TSP) account, it’s important to recognize a serious threat often overlooked by investors — but one that can unexpectedly derail or cause a serious detour in your retirement plan. That threat is a withdrawal that must be taken during a period of lower-than-expected account value.
This is not much of a threat to younger participants who don’t plan to withdraw money from their accounts for decades. In this case, the growth of the account value over time that results from contributions and investment performance tends to dwarf the fluctuations in the market value of risky investments like stocks and bonds.
But for participants approaching or already in retirement who will depend on their TSP accounts for a predictable stream of income, the situation is much different.
Each time a withdrawal is taken, it represents a percentage of the account’s balance at the time. A monthly $1,000 withdrawal represents 0.5 percent of a $200,000 balance, but this rises to 0.67 percent if the account balance has fallen to $150,000. This may not seem like much, but it’s the difference between a 6 percent annual withdrawal rate and 8 percent. Under certain circumstances, 6 percent per year may be sustainable while 8 percent is not — the difference between success and failure in the long run.
In planning for your retirement income stream, you have two options for dealing with this potential problem.
You can either avoid the risk by purchasing an inflation-adjusted fixed immediate annuity, or you can design an investment and withdrawal strategy that is robust enough to weather any bad luck that may come along. In some cases, the annuity will be the best choice, but often the income it produces is too low and the forfeiture of principal is too painful to be attractive.
Electing to continue your TSP account and take withdrawals along the way has its advantages, but relying too heavily on expected rates of return and the account values they may produce while ignoring the effects of variation in actual returns and the values they produce from year to year is a big mistake.
For the sake of simplicity, let’s say you plan to invest your entire TSP balance in the C Fund — an investment security representing the aggregate performance of the 500 stocks in the Standard & Poor’s 500 index. Let’s also say that you expect the C Fund to deliver a rate of return of 10 percent per year, on average, over the course of your retirement withdrawal period. Based on this assumption, you decide to take annual withdrawals calculated to represent 5 percent of your starting account balance, and then adjust that dollar amount each year for inflation. You’ll need these income payments to last as long as you’re alive.
The problem with this simple analysis is that it ignores the fact that it is highly unlikely that the C Fund will produce exactly 10 percent in return in any given year, much less every single year of your life. Historically, the S&P 500 has frequently produced annual returns significantly below 10 percent and has occasionally produced negative annual returns. Taking your predetermined withdrawal during one of these down periods will negatively affect the withdrawal planning equation from that point forward. And, what if you’ll need additional lump-sum withdrawals in addition to your regular income? Effectively, you have to recalculate the initial withdrawal rate that the account will support after each withdrawal is taken. In up years, this isn’t really necessary, and in fact can be dangerous, but in years that produce less return than expected, it is essential.
In order to avoid this recalculation and resulting change in retirement income, you must plan your withdrawals from the beginning while taking into account not only the expected rate of return, but also the expected variability of those returns from year to year.
This is what a simulation analysis called Monte Carlo is designed to do — expose the vulnerability of a plan to bad luck. But Monte Carlo simulation is complex and extremely sensitive to the assumptions of the financial planner using it. It is a tool to be used carefully.
Because of the complexity of the analysis required to plan for reliable and maximal retirement income from an invested balance, there is no rule of thumb that is safe for everyone.
Attempts to satisfy anxious investors with such rules are equivalent to advising you to drive from here to your destination in a straight line. While it might work, it probably won’t.
When it comes to retirement planning, sometimes the only simple advice I can offer is to explain what not to do — in this case, don’t plan as if each year’s returns will be optimal.
Written by Mike Miles
For the Federal Times
Publication May 28, 2007