How do broker-employed stock analysts operate? In their March 2013 paper entitled “Inside the ‘Black Box’ of Sell-Side Financial Analysts”, Lawrence Brown, Andrew Call, Michael Clement and Nathan Sharp summarize the results of a survey and follow-up interviews designed to discover key inputs and incentives affecting sell-side equity analyst outputs, including earnings forecasts and stock recommendations. They conducted the 23-question survey via email during January 9 through February 6 of 2013, with a promise of strict confidentiality. Using information from 365 responses from sell-side analysts (10.9% of 3,341 targeted) and 18 detailed follow-up interviews (17 by telephone and one in person), they find that:
- Regarding inputs to earnings forecasts and stock ratings:
- Industry knowledge is most important.
- Private telephone calls with company management are also extremely important (more useful than earnings conference calls), even in the post-Regulation FD environment. More than half of respondents report five or more direct contacts a year with the CEOs or CFOs of companies they follow. Maintaining strong relationships with these executives is therefore key to career success.
- Regarding determinants of compensation:
- Industry knowledge is most important.
- Client satisfaction, as measured by trading commission allocations to their brokerage firms, is second most important.
- Accurate earnings forecasts and stock ratings are relatively unimportant.
- Their own managers have pressured 17% (24%) of survey respondents to raise (lower) earnings forecasts and 24% (15%) to raise (lower) stock ratings.
- Respondents rely more on simple valuation metrics (price-earnings ratio and price-to-earnings growth ratio) than on sophisticated models (such as residual income) to support stock ratings.
- Issuing earnings forecasts or stock ratings that are well below consensus tends to enhance credibility with clients, and is unlikely to affect compensation or promotion opportunities.
In summary, evidence from surveys indicates that equity analysts rely heavily on industry expertise and simple valuation metrics in formulating earnings estimates and stock ratings.
As noted in the paper, “surveys have limitations, such as potential response bias, potentially small sample sizes, social desirability biases, and possible construct validity issues.” This caution applies especially to the small set of follow-up interviews.
