Confident that you’ll be able to afford your standard of living during retirement?
You might want to cultivate a bit more humility. A recent study of near-retirees’ behavior found widespread evidence of overconfidence: Many of those who believed their financial knowledge was above average actually were among those with greater-than-average ignorance of crucial aspects of their retirement portfolios.
Humility is a virtue, in other words, especially to those who don’t have it. We would do well to remember what Socrates reportedly said: “I am the wisest man alive, for I know one thing, and that is that I know nothing.”
The study, “Financial knowledge overconfidence and early withdrawals from retirement accounts,” was conducted by Sunwoo Tessa Lee and Sherman Hanna, both of Ohio State University. The researchers found that many investors are taking early withdrawals from a 401(k) or IRA “without understanding [the] possible consequences” to their “retirement financial security.” Previous research had documented that such withdrawals represent a significant loss to retirement savings and thus have a marked impact on investors’ retirement standard of living.
Up until recently there were significant restrictions on investors’ ability to take these early withdrawals. But the Coronavirus Aid, Relief, and Economic Security (CARES) Act waved the 10% penalty for those withdrawals. This meant that the remaining major barrier to taking an ill-advised early withdrawal is investor understanding of what is at stake.
To test whether lack of understanding plays a role in the decision to take such a withdrawal, the researchers tested investors’ ability to answer a famous financial literacy test. (I devoted a Retirement Weekly column last August to this test; its questions and answers appear in that column.) They found that only 5% of the investors who scored the highest on this test took an early withdrawal from their 401(k) or IRA, compared with 24% of the investors who scored the lowest.
Even more telling was the frequency of early withdrawals among those investors who, despite scoring low on this test, rated themselves as having above-average understanding of financial issues. Among this group of overconfident investors, 37% took an early withdrawal.
Note carefully that not all early withdrawals are a bad idea. In some cases they are a rational response to one’s situation. But you’d otherwise expect that the frequency of such withdrawals would, on average, be the same for the various groups in the researchers’ study. But they weren’t, even after the researchers controlled for myriad other variables that otherwise might account for more early withdrawals—such as “age, gender, respondent’s education, marital status, race/ethnicity, employment status, household income, homeownership, financial education, financial hardship, risk tolerance, retirement planning experience, and income drop.”
That these other variables couldn’t explain the researchers’ results suggests that overconfidence is playing a big role.
Another illustration of the dangers of overconfidence
This new study would be important for retirees and near-retirees even if taking early withdrawals was the only occasion for dangerous overconfidence. But it’s not, and it’s helpful to review other examples so that you can be on your guard against it.
One of the biggest sources of overconfidence is the way in which investment strategies’ returns are reported to investors. Average returns are presented without acknowledging wide year-to-year variability, unwittingly leading investors to downplay the role of luck and sheer random variation. (Confession: I have been guilty of this on occasion.)
Consider the “Sell In May and Go Away” seasonal pattern that is based on the historical tendency for the stock market to perform better between Halloween and May Day (the “winter” months) than in the “summer” months (May Day to Halloween). Since 1928, the S&P 500
has produced an average gain of 5.3% during the winter, versus 2.1% in the summer—a difference that is significant at the 95% confidence level that statisticians often use when determining whether a pattern is real.
That usually is all that gets reported when introducing the “Sell In May and Go Away” seasonal pattern to investors. And it certainly seems impressive and significant enough to justify being taken into account when managing your portfolio.
But consider the range of returns across individual summers and individual winters, as outlined in the table below.
|Average||Best||Worst||% of time rising|
The data in the table certainly paint a different picture, don’t they? Winters may have a better average return, but notice that the worst six-month return occurred during the winter, not the summer. Notice also that that odds of the market rising in the summer, while modestly less than during the winter (66% versus 71%), are still much better than 50%.
The data in the table show us why we shouldn’t be confident that, in any given year, the stock market in the summer would do worse than in the winter. Even on the assumption that the future is like the past, the only statistically sound way of betting on the “Sell In May and Go Away” pattern is to do so consistently over many years—a couple of dozen at a minimum.
Absent that consistency and discipline over the very long term, this seasonal pattern is not worth following—despite having one of the stronger statistical track records of any of the patterns that capture Wall Street’s attention.
The bottom line: Luck and sheer randomness play a large role in our performance, especially over the short term but even over longer terms as well. We forget that at our peril.
The investment implication is that we should build a large margin of error into all aspects of our retirement planning. False precision is just another manifestation of overconfidence.
I’m reminded of the joke about how we know if an economist has a sense of humor: He uses decimal points!
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at [email protected]