Retiree withdrawal rates alone are misleading. Here’s why

Recently released research from Morningstar suggests that those who begin retirement today may want to consider withdrawal rates as low as 3.3% — lower than the commonly quoted, but flawed, 4% rate.

This sounds like bad news for retirees, but focusing on any withdrawal rate can be misleading. In reality, retirees spend dollars, not percentages.

Americans who retired in November 1965 could have withdrawn roughly 4% of a 60-40 stock-bond portfolio, adjusted that withdrawal for inflation over time and successfully funded a 30-year retirement. All other historical periods would have allowed for higher initial withdrawal rates.

Does that mean that the mid-1960s were the worst time to have retired? Not quite! Depending on pre-retirement investment returns, individuals who invest through their working years have quite different balances at retirement, even with the same savings plan. Differences in nest eggs can lead to divergent standards of living in retirement.

Let’s assume simplistically (and now, anachronistically) that workers followed the same routine for the last 150 years: saved money, invested it and then spent down those investments during retirement.

Now let’s set a different worker off from each monthly point since 1871. Each will save $1,000 per month (in today’s dollars) and invest in a 60-40, stock-bond portfolio.

With this savings behavior, our 1965 retiree with the 4% withdrawal rate could have had more than $86,000/year in income, which is at the 45th percentile of historically available withdrawals. Viewed in this way, this is not the worst time to have retired — it’s more like the average.

Retirement income suffered most in the decades leading up to the 1929 market crash and in the “lost decade” of the 1970s. The worst retirement date in history by this measure was August 1918, with annual sustainable income of $52,000. In the 1970s, retirement bottomed out in February 1974 at $62,000/year. Surprisingly, these two troughs had distinct, but historically middling withdrawal rates, neither of which was even close to 4%.

In other words, withdrawal rates alone tell us very little about possible standards of living in retirement. This means that a 3.3% withdrawal rate doesn’t necessarily signal hard times for retirees.

To explain these variations in historically possible dollar withdrawals, we need to look at portfolio balances at retirement, which reflect the pre-retirement investment experience of savers. Historically, retirement balances and possible withdrawal rates would have traveled in opposite directions: when balances were high, withdrawal rates were low and vice versa.

Neither the size of the nest egg nor the withdrawal rate does much on its own to predict historically possible standards of living in retirement. To know whether a period in history was a good time to retire we need to know both the pre- and post-retirement experience. After all, a high balance (which reflects a good pre-retirement sequence of returns) with a low withdrawal rate (which reflects a poor post-retirement sequence of returns) can yield the same income as a low balance with a high withdrawal rate.

Same income via different paths
Examples of this are not hard to find. Our hypothetical February 1921 retiree could have produced about $55,000/year via a 11.4% withdrawal from a $580,000 nest egg. A May 1916 retiree could have had a nearly identical income, but with a 4.8% withdrawal from a portfolio of almost $1.4 million. Retirees from November 1985 and March 1961 could have achieved nearly twice this income, but again with different mixes of balance and withdrawal rate.

All else being equal, we would probably prefer to retire at a time when balances are high, as in 1916 and 1961, even though prudent withdrawal rates might therefore be lower. Having just lived through good market returns, we would likely have had high confidence and been able to live with a more conservative withdrawal rate. If Morningstar is correct, today’s retirees may find themselves in exactly this position.

If we had found ourselves in the opposite position, with a comparatively small nest egg, it may have felt risky to adopt a higher withdrawal rate initially, even though historically this has often been possible. Instead, we might have begun retirement with a more restricted budget.

But beginning retirement cautiously doesn’t necessarily mean a permanently reduced standard of living. As a 1921 or 1985 retirement played out, it would have become clear that high post-retirement returns were allowing for more income than originally planned. A well-informed retiree could have then increased their income, and even their withdrawal rate, accordingly and captured a much higher standard of living than expected — closer to the levels shown above.

The takeaway? Even simple historical scenarios demonstrate that retirement planning is complex and doesn’t lend itself to simple rules of thumb regarding withdrawal rates. Historical patterns suggest that when formulating retirement plans advisors and retirees should keep in mind the cyclical nature of retirement nest eggs and sustainable withdrawal rates and, more importantly, be ready to adjust as retirement plays out.