Analyzing the Analysts: When Do Recommendations Add Value?

Abstract

We show that analysts from sell-side firms generally recommend "glamour" (i.e., positive momentum, high growth, high volume, and relatively expensive) stocks. Naïve adherence to these recommendations can be costly, because the level of the consensus recommendation adds value only among stocks with favorable quantitative characteristics (i.e., value stocks and positive momentum stocks). In fact, among stocks with unfavorable quantitative characteristics, higher consensus recommendations are associated with worse subsequent returns. In contrast, we find that the quarterly change in consensus recommendations is a robust return predictor that appears to contain information orthogonal to a large range of other predictive variables.FINANCIAL RESEARCHERS AND PRACTITIONERS have long been interested in understanding how the activities of financial analysts affect capital market efficiency. Currently in the United States, over 3,000 analysts work for more than 350 sellside investment firms. 1 These analysts produce corporate earnings forecasts, write reports on individual companies, provide industry and sector analyses, and issue stock recommendations. Most prior studies have concluded that the information they produce promotes market efficiency by helping investors to more accurately value companies. 2The Journal of Finance about the intrinsic stock values relative to their current market prices, and finally rate the investment potential of each stock. As Elton, Gruber, and Grossman (1986, p. 699) observe, their stock recommendations represent "one of the few cases in evaluating information content where the forecaster is recommending a clear and unequivocal course of action rather than producing an estimate of a number, the interpretation of which is up to the user." In short, these recommendations offer a unique opportunity to study analyst judgment and preferences across large samples of stocks.Our study investigates the source of the investment value provided by analyst stock recommendations and changes in recommendations. One possible source of this value is the ability of analysts to collect and process firm-specific information useful in identifying undervalued or overvalued stocks. Alternatively, it is possible that analyst recommendations derive their value by tilting toward stocks with particular characteristics that predict future returns. We assess the contribution of these sources of potential value.We also assess the extent to which sell-side analysts make full use of available information signals in formulating stock recommendations. We find that analysts do not fully take into account the ability of various stock characteristics to predict returns. Moreover, our evidence shows that the direction of the bias in analyst recommendations is in line with economic incentives faced by sell-side brokerage firms.We expect this research to be of interest to both financial academics and practitioners. From an academic perspective, the study contributes to a better understanding of how analysts evaluate stocks, and ...

Long-run performance of IPOs and the role of financial analysts: some French evidence

“…We found the market reaction to be similar to the changes in the recommendations than to the value of the recommendations themselves. This result is inconsistent with the evidence given by Irvine (2003) and Jegadeesh et al (2004). The characteristics of the financial analysts are not significant whatever the horizon we considered.…”

“…The rationale behind this distinction is that the market can asymmetrically react to favourable news and unfavourable news (Cooper, Day, and Lewis 2001;Jegadeesh and Kim 2006;Bradley, Jordan, and Ritter 2008a). As in previous studies, Model 3 assumes that initiation, upgrade, or downgrade conveys more information than the value of the recommendation itself (Irvine 2003;Jegadeesh et al 2004). …”

The reaction of investors to analyst recommendations of stocks listed on the WIG20 index

“…Research conducted by other authors suggests that an analysis of changes in the level of recommendation is much more informative than an analysis of a pure recommendation level (e.g., Stickel, 1995;Womack, 1996;Jegadeesh et al, 2004;. To check this assumption, recommendations are divided into nine clusters (as detailed in Table 3).…”

“…Jegadeesh and Kim (2006) confirm a dependence on the strength of the reaction and the size of the market, convincing us that a developed market reacts stronger, while show the opposite (via examples of the Austrian and German stock markets): abnormal returns on the smaller market are higher. Jegadeesh et al (2004) suggest that the research should not only involve the level of recommendation but also the change from its previous level, as it has more robust explanatory power than the level alone. The impact of recommendation is especially strong when there are the most-extensive changes, such as an upgrade to BUY from SELL or a downgrade to SELL from BUY (Stickel, 1995;.…”

How Do Analyst Recommendations Respond to Major News?

“…To date, there has been little research on the determinants of analysts' recommendation changes, and this work tends to focus on the relation between earnings announcements and recommendation changes (e.g., Bradshaw (2004) and Finger and Landsman (2003)), recommendation changes and subsequent stock returns (e.g., Green (2006) in this issue, Womack (1996), Jegadeesh, Kim, Krische, and Lee (2004)) or analysts' herding behavior (e.g., Welch (2000), Hong, Kubik, and Salomon (2000)). In terms of setting recommendation levels, findings in Hong and Kubik (2003) show that analysts are rewarded for both optimism and accuracy, which suggests that analysts trade off reputation (which is based on accuracy) and bias.…”