The first half of 2022 was dominated by high inflation, central bank interest rate hikes, and fears of an economic downturn. For American investors, the most positive news was that one should not rely solely on superficial figures—such as two consecutive quarters of negative GDP—that merely suggest the onset and conclusion of a recession. While investors can depend on the official recession declaration by the NBER, should they align their portfolio decisions with NBER’s announcements?
The chart below displays the S&P 500 index alongside the recession periods as determined by the NBER since 1980. The chart clearly shows that stock market returns during recession periods were markedly negative.
So, why do we end up selling when a recession begins and fail to buy when it ends?
Of course, all macroeconomic data suffers from a time lag before being released. The NBER’s recession committee reviews various indicators and officially declares a recession only when it is convinced that a cyclical turnaround has occurred. In reality, such declarations often come more than a year after the cycle has shifted. Should investors then sell when the NBER announces that the previous cycle’s bottom has been reached, and buy when it declares that a new expansion has begun?
To explore this issue, we collected all NBER recession announcements since 1980 and calculated the average 12‑month S&P 500 return following each announcement. On average, after the NBER declared a peak and the onset of a bear market, the S&P 500 returned +15% over the next 12 months—with only one instance out of six showing a decline (namely, the bear market of 2000–2003).
Conversely, when the NBER declared that a recession had ended and a new expansion had begun, the S&P 500 fell an average of –3.5% over the following 12 months, with only two out of six cycles (in 1992 and 2003) posting positive returns.
As a market timing tool, relying on NBER announcements is clearly problematic. The lag between the actual market cycle change and the official announcement is so long that the indicator becomes countercyclical. This effect is significant—the performance difference between the NBER’s low and high points is statistically significant at around 10%.
However, these overall characteristics based on NBER announcements are merely statistical exercises. Who in their right mind would devise an investment strategy based solely on the NBER’s arbitrary announcements? Just because one can compute returns and statistical significance in an Excel spreadsheet does not mean one should base investment decisions on them. Nor do I claim that you must invest strictly according to Excel formulas.
The key to becoming an excellent investor is not the ability to crunch spreadsheet numbers, but to understand what those numbers mean. This is an essential challenge for both inexperienced young investors and quant investors alike.
Quantitative investing can easily become a case of “garbage in, garbage out” if one does not fully understand what one is doing and why certain results occur.
While the NBER recession cycle examples clearly illustrate this point, the wrecks of bankrupt hedge funds and the unknown graves of Lehman credit analysts remind us that even highly educated experts fall prey to these mistakes on a regular basis.
