Should Your Withdrawals Mirror Your RMDs?

While retirees may gripe about the required minimum distributions that they must take from their tax-sheltered accounts once they hit age 70 1/2, one complaint that most can safely set aside is that taking and spending their RMDs will cause them to prematurely deplete their assets.

That's because the Uniform Lifetime Table, which most retirees use to guide their RMDs, rests on some pretty conservative assumptions. Specifically, the distribution period that determines the RMD amount is calculated using the account owner's life expectancy plus an additional 10 years to ensure that assets aren't depleted during the lifetime of a spousal beneficiary. The net effect is that unless a retiree has a strong bequest motive, a very long life expectancy, and/or a substantially younger spousal beneficiary, basing total portfolio withdrawals on RMDs won't lead to premature capital depletion.

The fact that the RMD system that most retirees use is pretty conservative raises another question: Could RMDs be employed as a general guide to withdrawals, much as some retirees and advisors use the 4% guideline or other withdrawal systems? After all, the Uniform Lifetime Table results in conservative withdrawals, and many people would rather err on the side of safety than risk running out later in life. Taking withdrawals also has simplicity on its side. Moreover, RMD-based withdrawal amounts are automatically updated to account for portfolio value and life expectancy. Thus, the RMD method syncs up with much of the current thinking on retirement portfolio withdrawals--namely, that dynamic withdrawal systems are better than static, dollar-based ones.

The Shortcomings of a Static Approach Withdrawal-rate strategies have received extra scrutiny from researchers in recent years, largely because the global financial crisis got everyone thinking about worst-case scenarios and the problems that arise from withdrawing too much in down markets.

The 4% guideline for retirement portfolio withdrawals, the most familiar withdrawal strategy, is static; withdrawals don’t adjust to account for changes in portfolio value, such as when the market craters, or changes in life expectancy. Under the 4% guideline, a retiree withdraws 4% of his or her balance in year 1 of retirement, then inflation-adjusts that dollar amount each year thereafter. A $40,000 first-year withdrawal from a $1 million portfolio would become $41,200 in year 2 (assuming a 3% inflation rate) and so on.

Such a static withdrawal system might seem appealing because it ensures a stable standard of living in retirement, but it's not ideal for a few reasons. As Morningstar Investment Management's head of retirement research, David Blanchett, has written, it doesn't take into account the fluctuations in portfolio value that will inevitably accompany a long-term portfolio with volatile assets. If withdrawals are too rich at a time when the portfolio's value has declined, that leaves fewer assets in place to recover when the market does, potentially impairing the portfolio's sustainability over a 25- to 30-year time horizon.

To illustrate that point, let's go back to the earlier example of a retiree with a $1 million portfolio who withdrew $40,000 in year 1 of retirement. If the portfolio drops 30% in value in year 2, the year 2 inflation-adjusted withdrawal would amount to more than 6% of the depressed year 2 balance (the $41,200 year 2 withdrawal divided by the $672,000 remaining in the portfolio after the year 1 withdrawal and market decline). If those bad years happen to coincide with the first years of retirement, and too much comes out of the portfolio early on, that leaves fewer assets in place to heal and eventually grow when the market recovers, imperiling the portfolio's long-run sustainability. That's what retirement researchers call sequence risk.

Not only that, but the static approach also doesn’t take into account that, as a retiree ages, he can take a higher percentage of the portfolio than at the beginning of retirement, simply because the withdrawal period has declined. Because RMDs rely on actuarial data to determine life expectancy and, in turn, distribution period, they also take into account that the longer you live, the longer you’re likely to live.

The RMD approach corrects for these issues--changes in portfolio value and changes in life expectancy--and does so in a relatively simple fashion. A retiree calculates her withdrawal based on the account’s value at the end of the previous year divided by a life expectancy factor, or distribution period.

What's Not to Like? Yet even as it's admirably dynamic, the RMD method isn't perfect.

In contrast with the static withdrawals that are central to the 4% guideline, RMD-type withdrawals can result in extremely lumpy cash flows, which can wreak havoc with a retiree’s budget. On the Uniform Lifetime Table, for example, a 75-year-old retiree with a $500,000 IRA would arrive at her RMD by dividing her portfolio value at the end of the previous year by 22.9, resulting in a withdrawal of $21,834. If her portfolio dropped by 25% the following year, her withdrawal would amount to $16,301 (her $500,000 prior-year balance, less her $21,834 withdrawal and the 25% depreciation). That’s a $5,500 swing in income. Big swings like that are apt to be particularly painful for people with more-modest portfolios than very large ones.

Moreover, the success of the RMD method depends on how those RMDs are calculated. As noted above, the Uniform Lifetime Table that most people use to calculate RMDs builds in a healthy cushion--an individual's life expectancy plus 10 years. That will tend to make that RMD system quite conservative, especially for single folks and similarly aged married people. Many of us would rather be safe than sorry, but foregone consumption--and potentially a big remaining balance at death--isn't ideal, either, especially for people who don't have a strong bequest motive. Indeed, a study of RMD-based withdrawals using the Uniform Lifetime Table concluded that strategies based on the Uniform Lifetime Table led to a high level of wealth "unconsumed" relative to other withdrawal strategies. At the opposite end of the spectrum, a withdrawal system based on the Uniform Lifetime Table may well be too aggressive for very wealthy individuals and couples; such retirees are more likely to have a strong bequest motive.

How about customizing your own life expectancy, as Morningstar contributor Mark Miller discusses here, and using that to guide your own RMD-based system? David Blanchett took a closer look at withdrawal strategies based on RMDs/life expectancy in this paper. His conclusion was that RMD-based withdrawals using simple life expectancy were effective, as measured by consumption maximization and not running out of money, for periods of less than 15 years. However, he found them to be less useful for longer time periods.

In the end, some of this research, plus a healthy dose of common sense, can help retirees find their way to a withdrawal system that makes sense for them. Retirees with longevity on their sides and a strong bequest motive should favor a more conservative withdrawal system. That's easier to accomplish, of course, with portfolios that are large at the outset. On the other hand, retirees who aren't concerned about leaving a bequest and are willing to live with the possibility that they might have to reduce consumption in retirement may want to (or have no choice but to) use a more aggressive withdrawal rate. And no matter the baseline approach, being willing to rein in spending if a market correction materializes is a best practice for ensuring a portfolio's sustainability, as outlined here.