A new study may cast doubt on the conventional wisdom that following analyst buy or sell recommendations isn’t going to pay off for investors. This according to a recent article in CFA Institute.

The study examined 3.8 million analyst forecasts in 45 countries and regions from 1994 to 2019, comparing the top and bottom 20% of equities by consensus analyst recommendations. It found that, on an equal-weighted basis, US analysts failed to outperform.

“To be sure, this hardly qualifies as a surprise,” the article argues, adding that the US analysts tend to prefer “growth and glamour stocks.”

But the study found that “a different picture emerges as soon as the focus shifts outside the United States,” noting that in every developed market and nearly all emerging markets, following analyst recommendations did lead to meaningful outperformance over time.

The study’s findings represent a departure from earlier research, the article explains, because it covers two major bear markets (the dot-com crash of the early 2000s and the global financial crisis of 2008-2009) and was therefore able to discern whether analyst recommendations were more effective in bull or bear markets. “As we might have expected,” the article reports, “in a low sentiment phase like that of a bear market or financial crisis, analyst recommendations add more performance than in periods of bullish high sentiment.”

The article suggests two different scenarios to explain the data: It’s possible that analysts have deeper insights in the wake of a crisis and are therefore able to “sift through the rubble” more effectively than the average investor. Alternatively, investors may look to analysts for guidance and follow their recommendations more closely during a crisis.

“Whatever the answer,” the article concludes, “the study suggests that investors may want to rethink the conventional wisdom on analyst recommendations. They may add some value after all.”