A New Defense of Portfolio Diversification

‘Perhaps Markowitz had it right all along.’

(Illustration via the Library of Congress)

(Illustration via the Library of Congress)

In 1952, Nobel Prize laureate Harry Markowitz famously said “diversification is the only free lunch in investing.” In recent years, however, investors have lost their appetite for the economist’s counsel, instead embracing maxims like “diversification fails when needed most” and “all correlations go to one in a crisis.”

An academic paper published last month threw its weight behind Markowitz, the architect of modern portfolio theory, arguing that “diversification remains as relevant to investors today as ever before,” adding that it has never been easier to construct a globally diversified portfolio.

“Although steep losses have been an integral part of the story, the analysis here shows markets over the past 75 years have shown remarkable resiliency and a globally diversified portfolio even more so,” writes Rodney Sullivan, CFA, CAIA, Executive Director, University of Virginia Darden School of Business, and Editor, Journal of Alternative Investments. “Perhaps Markowitz had it right all along.”

The paper examined the long-term record of market drawdowns and recoveries, focusing on long-only portfolios.


[Like this article? Subscribe to RIA Intel’s’ twice-weekly newsletter.]

To tabulate returns on 60/40 portfolios, Sullivan allocated 60% to large-cap U.S. stocks and 40% to intermediate U.S. government bonds, rebalancing monthly.

Bonds have provided material support during steep selloffs. “Our 60/40 portfolio experienced an average maximum peak-to-trough drawdown of -15.9 percent over the full period, and - 11.7 percent since WWII, versus -26.3 percent and -20.6 percent declines for large cap stocks, respectively.”

This contradicts the notion that diversification “fails when it’s needed most,” says Sullivan. Moreover, he adds that “the recovery for the 60/40 portfolio is also at least as rapid (or more so) as that of the stock-only portfolio.”

Sullivan ran the numbers on a further diversified long-only globally diversified portfolio that included additional asset classes to complement U.S. stocks and bonds.

He earmarked 60% – comprising “more risky assets” – to global equities (including developed and emerging markets), global bonds (developed, emerging markets, and global high yield), U.S. REITs, and gold. The “less risky” 40% was comprised of the U.S. bond aggregate and global developed market bond aggregate.

Lacking complete data going back to 1926, Sullivan examined data involving drawdowns around recessions since 1994, the common starting date for all assets examined here.

“The global diversified portfolio, while not eliminating declines altogether, nonetheless has historically even further mitigated drawdowns during turbulent markets versus the all-equity and 60/40 portfolios.”

From January 1994 to June 2020, the average annual return for the S&P 500 was 10.3%, followed by the global diversified portfolio at 7.5%, and the 60/40 portfolio at 7%.

Sullivan adds that the in addition to falling the least during periods near recessions, the global diversified portfolio recovered to its pre-crisis levels before the all stock and 60/40 portfolios.

“The key to effective diversification is, no surprise, using more assets (again, that are lowly correlated and have positive expected risk premiums).”

Given the fierce headwinds that bonds are facing with rock-bottom interest rates, the 60/40 portfolio is increasingly finding less support for the 40% traditionally allocated to bonds. A shift from bonds to other assets makes particular sense in this low-rate environment, furthering the argument for a more diverse global diversified portfolio.

Greg Bartalos (@gregorianchance) is editor of New York City-based RIA Intel.

Subscribe to RIA Intel’s twice-weekly newsletter and follow the publication on Twitter and LinkedIn.

Related