I’ve made this argument before. But it bears repeating because of a recent bipartisan proposal in Congress to fund the Social Security trust fund with the profits the U.S. Treasury could earn by borrowing $1.5 trillion and investing it in the stock market.
The proposal is garnering increasing support, not just on Capitol Hill but also from Wall Street.
Read: Social Security’s COLA for 2024 is 3.2%, vs. 2023’s historic 8.7% inflation-fueled adjustment
It’s easy to understand why many are attracted to it. The proposal claims to overcome Social Security’s multitrillion-dollar deficit without raising taxes or cutting benefits. What’s not to like about that?
If only it were that easy.
The fundamental problem with the proposal is that it’s extremely risky: More than half the time since the founding of the U.S. in the late 1700s, bonds have outperformed stocks. So if the future is like the past, there’s a better-than-even chance that the Treasury’s foray into the stock market will cause the U.S. government to lose money. And Social Security will be in worse shape than it is already.
Read: Inflation is already racing past next year’s Social Security COLA
Proponents of the proposal calculate that the odds of success are much better than this. But they are guilty of focusing only on that portion of U.S. history in which stocks greatly outperformed bonds. Unless there is a good theoretical reason to ignore the more than half of U.S. history in which bonds outperformed stocks, the proponents’ arguments represent a triumph of hope over experience.
The accompanying chart plots the cumulative performance of stocks and bonds since 1793, courtesy of Edward McQuarrie, a professor emeritus at the Leavey School of Business at Santa Clara University who has spent years constructing a database of U.S. stock and bond market history. Notice that, as late as 1933, stocks’ cumulative performance since 1793 was below that of bonds. That 140-year period of bond superiority constitutes more than half the 230 years of U.S. market history.
I’m not accusing the proponents of the Congressional proposal of consciously excluding these 140 years, which would be shameful. I doubt they are even aware of that history. The Center for Research in Security Prices (CRSP) at the University of Chicago, which is the gold standard for historical market data, starts its database in December 1925. Almost all other stock and bond yearbooks follow suit.
Ignorance is not a good defense, however, especially when trillions of dollars are at stake. And there is no theoretical justification for the December 1925 start date, McQuarrie said in an email. “My understanding is that the University of Chicago team that collected the original data for CRSP in the early 1960s ran out of time and / or money as they got back that far.”
As a result of this unfortunate historical accident, however, the period prior to the mid-1920s is invisible to most students of the markets.
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]