What year was the worst possible time in U.S. history to retire, from an investment point of view? To put this question another way: What year’s U.S. retirees had the greatest difficulty sustaining their retirement standard of living, relative to any other year’s retirees of the last two centuries?
If you’re like almost everyone else, your first guess is the summer of 1929, just prior to that year’s stock market crash. Close runners up include early 2000, just prior to the bursting of the Internet bubble, 1987 prior to that year’s stock market crash, or October 2007 before the Great Financial Crisis.
Each of these guesses is wrong, however.
The correct answer, according to just-completed research from Edward McQuarrie, an emeritus professor at Santa Clara University’s Leavey School of Business, is 1965. That’s because, beginning in that year, both the stock and bond markets embarked on a sustained period of negative inflation-adjusted returns. In the years after 1929, 1987, 2000, and 2007, in contrast, a balanced stock-bond portfolio performed much better, either because interest rates declined and bonds performed well, or the stock market quickly recovered from its losses, or both.
McQuarrie arrived at this 1965 worst-retirement-start date when measuring the probabilities that a retiree would outlive his or her money. He analyzed a much larger data set than most previous studies, which have largely focused on just the U.S. markets since the 1920s. McQuarrie, in contrast, analyzed U.S. data back to 1793 and global markets back to 1864.
He came up with both good and bad news. The bad: The odds of running out of money in retirement are significantly greater than prior studies had concluded. That’s because those prior studies based their return assumptions on U.S. history since the mid-1920s. Other countries’ markets, as well as the U.S. markets prior to the mid-1920s, frequently performed less well.
The good news: Retirees and near-retirees have several portfolio fixes available to them that dramatically reduce the odds that a stock-bond portfolio will run out of money for they pass. Here’s a summary of these fixes that McQuarrie identified:
Invest in a total bond market index fund instead of long-maturity bonds
This seems like a minor fix, but actually has a big impact, according to McQuarrie. By investing in a total bond market index fund, you significantly reduce the average duration of the bonds you own. A balanced portfolio containing such a fund and equities did a better job dealing with “the diverse challenges a retiree might face” than a portfolio of equities and long-term bonds.
To illustrate the reduction in duration that comes with shifting from a long-term bond fund to a total bond market index fund, consider that the Vanguard Long Term Bond ETF
has an average duration currently of 16.5 years, while the Vanguard Total Bond Market ETF
has an average duration of 6.9 years, less than half as long.
It’s helpful to remember why retirees should want to mix equities with bonds. It is not because bonds outperform stocks, but because an equity-bond portfolio is less vulnerable to an equity bear market—especially if that bear market were to begin right at the onset of retirement. McQuarrie’s argument is that intermediate-term bonds hardly ever do a poorer job than long bonds of reducing the risk of a retiree’s portfolio, and sometimes do far better.
This was certainly the case for those who retired in 1965, for example. In the stagflation era that ensued for the subsequent 15 years, long bonds were crushed. Intermediate-term bonds were “considerably more successful in coping with stagflation” than long bonds.
Note that McQuarrie’s recommendation to favor a total bond index fund over long bonds is not motivated by a market timing judgment that inflation and interest rates will rise over coming years. Given the recent spike in inflation to the highest level in 40 years, many are betting that long bonds are entering into a long-term bear market. They may very well be right, of course. But it’s not out of the question that the global economy could suffer a deflationary collapse; we saw a hint of that at the beginning of the COVID-19 pandemic, you may recall. A total bond index fund would serve a retiree well in both the inflation and deflation scenarios, while long bonds would be appropriate only in the second.
Allocate a portion of the fixed-income allocation to TIPS
McQuarrie found that in some limited circumstances a small allocation to inflation-protected bonds was a good idea. Those circumstances are when everything goes wrong at the worst possible time, in both the stock and nominal bond markets.
McQuarrie adds that, but for these circumstances, TIPS rarely help a retiree’s portfolio last longer. And he stresses that the circumstances in which they do help have been very limited historically. He says that a modest TIPS allocation (perhaps 20%, taken from the fixed income allocation) is therefore most appropriate for the extremely risk-averse retiree.
McQuarrie next investigated whether global diversification increased the longevity of a retiree’s portfolio. He found that it did in the case of equities, for the familiar reason that it reduces portfolio volatility. However, at least for U.S.-based investors, he found mixed results when global diversifying the bond portion of a retiree’s portfolio. That’s because, as McQuarrie put it, “there’s too much tail risk [for the U.S. investor] in foreign government bonds.”
He notes, however, that global diversification of the bond portion of a portfolio may be a better idea for “residents of a small nation whose governing structures have historically not been stable.” They “might not regard bonds from their own government as in any way superior to foreign bonds.”
Lastly, McQuarrie found some evidence that the addition of residential real estate to a portfolio increases the odds that that the retiree will not run out of money. He stresses that this conclusion is tentative, because of limited historical data on the performance of residential real estate. That he was even able to reach a tentative conclusion about real estate is due to the monumental project of several academic researchers to construct a long-term database that includes “the rate of return on everything.” In any case, McQuarrie found that “the addition of real estate to [a retiree’s portfolio]… was generally successful,” and for investors outside the U.S. “sometimes remarkably so.”
McQuarrie, in an interview, acknowledged the difficulty putting this finding into practice, since there is no available investment that is benchmarked to the performance of residential real estate generally. At a minimum, however, he said that his findings suggest retirees shouldn’t automatically exclude their real estate holdings when constructing their financial plans for dealing with all possible contingencies. The key will be whether their residential real estate holdings perform at least as well as the national average.
The bottom line from McQuarrie is good news: “Two centuries of global market history indicate that exhaustion of tax-sheltered retirement savings before the age of 100 is unlikely to occur for retirees” who hold a balanced portfolios.
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@example.com.