Maybe analyst recommendations DO add value

One of the things every investor knows is that the average analyst does not add value. Following buy or sell recommendations of analysts isn’t going to lead to outperformance in the long run.

Or does it?

An interesting new study by Vitor Azevedo and Sebastian Müller casts some doubt on that perceived wisdom. They looked at analyst forecasts in 45 countries from 1994 to 2019 and analysed a total of 3.8 million analyst forecasts. There are many interesting results in that paper, so I will have to come back to it one day, but let me focus on the average annual outperformance of the stocks most loved by analysts (top 20% by consensus recommendation) vs. the stocks most hated by analysts (bottom 20% by consensus recommendation). 

On an equal-weighted basis, US analysts aren’t able to provide any outperformance on average. The researchers also found that US analysts tend to prefer growth and glamour stocks which may be one reason why their recommendations don’t work in practice. But once you look at the performance of analyst recommendations in other countries, the picture changes significantly. In every developed country and almost all emerging markets, analyst recommendations create a meaningful outperformance over time. In the chart below, I have collected only the developed markets in the study, and it shows what a significant outlier the United States is.

Performance of analyst consensus recommendations

Source: Azevedo and Müller (2020).

One of the key differentiators of this study compared to the ones done in the 1990s and early 2000s that established the common perceptions of analyst consensus estimates being useless is that this study covers two major bear markets (the tech crash of the early 2000s and the financial crisis 2008). Thus, the study was able to parse if analyst recommendations work better in a bull market or a bear market. And as one might have expected, in a low sentiment phase (i.e. in a bear market or crisis) analyst recommendations add more performance than in periods of high sentiment. 

I don’t know what is cause and effect here. Is it that analysts have deeper insights than most investors and thus are better able to pick through the rubble of a crisis and select the truly good stocks? Or is it that in a crisis, investors look for guidance from analysts and tend to follow their recommendations more closely, thus turning analysts’ buy recommendations into something like a self-fulfilling prophecy.