Three essays on financial analysts

Li, Yaohua (2025) Three essays on financial analysts. PhD thesis, University of Glasgow.

Abstract

This thesis comprises three chapters in related to the financial analysts. The first chapter focuses on the textual information in analyst reports. The second chapter examines the impact of analyst coverage. The third chapter investigates analysts’ forecasting behavior in relation to the cultural characteristics of covered companies.

The first study in Chapter one examines the determinants and the impacts of analysts’ tendency to provide similar textual information in their reports. To implement the tests, I collect a large sample of analyst report transcripts for the S&P 500 companies from 2015 to 2020. I introduce four factors to investigate the possible determinants of analyst report similarity, which include analysts’ herding behavior, analysts’ lack of ability, analysts’ learning behavior, and the significant firm-related news. The regression results indicate that these four factors are positively associated with the analyst report similarity, which suggests that the four factors are likely to be the determinants of analyst report similarity. Next, I examine whether this textual similarity among analyst reports has implications for the market. The results indicate a negative association between the analyst report similarity and the short-term investor reactions, suggesting that market investors consider an analyst report as less informative if the report contains the textual information that is more similar to that in other prior analysts’ reports. In the additional analysis, I find that this similarity likely hinders investors’ understanding of analysts’ quantitative outputs, including the earnings forecasts, stock recommendations, and price target forecasts. In addition, I find that the observed negative investor reaction to the analyst report similarity is stronger when firm managers have more incentives to withhold relevant information, but it is alleviated if the analysts are from larger brokers.

The second study in Chapter two investigates the influence of analysts on corporate governance in the context of corporate culture. I first examine the association between the analyst coverage of a firm and the score of the firm’s culture. The baseline results show that the firm with higher level of analyst coverage is associated with a lower score of corporate culture. This is consistent with the argument that analysts can impose short-term pressure on firms, resulting in a weak corporate culture. In further tests, I find that this negative association is stronger for the long-term oriented cultural values than other cultural values. Furthermore, to deal with the potential endogeneity problems, I first employ the two-stage least squares model based on a commonly used instrumental variable in this field, the expected coverage. The results suggest that the analyst coverage has a negative impact on the corporate culture. Moreover, I implement a quasi-natural experiment based on two exogenous shocks to analyst coverage, the brokerage closures and mergers. Consistent with the above findings, the results of the Difference-in-Difference model indicate that the analyst coverage could negatively influence the covered firm’s culture. Taken together, these results are mostly consistent with the pressure effect that analysts impose short-term pressure on firms, increasing management myopia and resulting in a weak corporate culture. In additional tests, I find that analysts’ negative effect on the corporate culture is alleviated when firms are covered by more experienced analysts, when firms are more likely to reach analysts’ earnings forecasts, and when firms tend to have better corporate governance as captured by a higher competitive market.

The third study in Chapter three examines whether analysts’ forecasting behavior is affected by the covered firm’s information environment that is characterized by a strong integrity culture. I first investigate the association between analyst forecast boldness and the covered firms’ scores of integrity culture. The baseline results indicate a positive relationship between the two variables, suggesting that analysts tend to issue bolder earnings forecasts when covering the firm with a stronger integrity culture. In further analysis, the results show that analysts’ bolder forecasts for firms with a strong integrity culture are associated with decreased market reactions, indicating that analysts’ bold opinions for these firms are regarded as less informative by market investors. Moreover, I find that the firm with a higher score of integrity culture is negatively associated with the number of analysts following the firm, suggesting that the analyst’s service is less demanded for the firm with a stronger integrity culture. Additionally, other results show that analysts tend to issue less accurate earnings forecasts for firms with higher scores of integrity culture. Furthermore, I mitigate the potential endogeneity issue by conducting the two-stage least squares regression based on two instrumental variables: the average score of corporate integrity culture in an industry and the CEO age of the firm. Consistently, the regression results support the positive association between the firm’s integrity culture and analyst forecast boldness. Finally, I introduce an alternative measure of the score of corporate integrity culture based on the analyst report transcripts. Consistent with the main findings, the regression results show that the corporate integrity culture is positively associated with analysts’ forecast boldness. However, the association is not statistically significant, possibly due to the limited sample of analyst report transcripts.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: H Social Sciences > HG Finance
• Colleges/Schools: College of Social Sciences > Adam Smith Business School > Accounting and Finance
• Supervisor's Name: Aleksanyan, Professor Mark, Stasinakis, Professor Charalampos and Tsalavoutas, Professor Ioannis
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-84989
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 07 Apr 2025 08:34
• Last Modified: 07 Apr 2025 08:37
• Thesis DOI: 10.5525/gla.thesis.84989
• URI: https://theses.gla.ac.uk/id/eprint/84989