Do Corporate Ethics Enhance Financial Analysts’ Behavior and Performance?

Abstract

This study investigates the relationship between corporate ethics and the information intermediation element of public companies’ information environment. Drawing on the well-established virtue, deontological, and consequential ethical theories, we predict that higher corporate ethics standards have a positive effect on financial analysts’ behavior and earnings forecasts. Using a sample of 5276 firm-year observations from 780 publicly listed US companies, multivariate regression analyses document a significant positive association between company’s level of ethical commitment and analyst coverage and forecast accuracy. Furthermore, the results show that firms with fewer incidents of ethical misconduct are associated with higher analyst consensus. These findings hold across a battery of robustness tests and indicate that a firm’s ethical commitment enhances its corporate information environment and allows financial analysts to play a more effective intermediary role in capital markets.

1. Introduction

Capital market and corporate governance research in accounting and finance has extensively studied the dynamics of companies’ information environment, which is also known as the corporate information ecosystem (von Koch and Willesson 2020). This is not a surprise, bearing in mind that the modern economic model has developed around the idea of a separation between ownership and management, and the ensuing concept of agency risk. Agency risk, the risk of managers failing to act in the best interest of the company’s shareholders, is rooted in two main causes: information asymmetry and a conflict of interest between agents and principals (Jensen and Meckling 1976). To reduce agency risk to an acceptably low level and allow for financial markets to subsist and flourish, several governance mechanisms have been deployed over the years, particularly corporate reporting. Corporate reporting is a mechanism that tackles the information asymmetry problem and consists in the statutory and discretionary financial and non-financial information disclosed by companies using diverse vehicles such as annual and quarterly financial statements, and CSR, ESG, and sustainability reports (Bushman et al. 2004). However, repeated high-profile financial scandals have called into question the reliability of public companies’ disclosures and the transparency of their information ecosystems (Soppe et al. 2012). According to Chih et al. (2008), managers seem determined to deceive external stakeholders through an opaque information environment. This lack of transparency exacerbates information asymmetry between listed firms and the public, which limits investors’ ability to adequately assess the parameters that underlie share price formation (Firth et al. 2015).

Different definitions of the corporate information environment (CIE) have been proposed in the literature. For instance, Lang et al. (2003) stipulate that the CIE is a synonym for corporate transparency, which encompasses “corporate reporting, private information acquisition, and information dissemination” (Bushman et al. 2004). Later, in a seminal paper, Beyer et al. (2010) distinguish three dimensions of the CIE, including mandatory reporting, voluntary disclosures, and financial intermediation. Despite its critical importance, the third dimension has received less, yet growing, attention from scholars and regulators compared to the two other dimensions. This could be explained by the complex relationships that third-party intermediaries have with companies, the lack of research data, and regulators’ focus on firms’ accountability (Imam and Spence 2016; Brauer and Wiersema 2018). The financial intermediation dimension of the CIE includes all parties external to the company and who participate in the processing, communication, and interpretation of the company’s information. Third-party intermediaries such as auditors, financial analysts, brokers, and the media, play a critical role in shaping the company’s information ecosystem (Armour et al. 2016).

Financial analysts are crucial information intermediaries in capital markets (Liang et al. 2022; Hussain and Su 2024). They are viewed as the most knowledgeable and sophisticated users of corporate information as they have the expertise required to collect, analyze, and interpret information from various public and private sources (Lang et al. 2012; Gu and Hackbarth 2013; Armstrong et al. 2014; Chen et al. 2014; Upadhyay 2014). As a result, their earnings forecasts play a prominent role in the successful functioning of capital markets, shaping investors’ perception of a firm’s performance and prospects and considerably affecting their investment decision-making process (Zadeh et al. 2021). In prior research, three attributes of analysts’ behavior have been examined, namely analysts’ coverage, prediction accuracy, and forecast dispersion. Along these lines, an expanding body of research has documented the significant impact of financial analysts’ output on a wide range of corporate policies (Hussain et al. 2023; Bradley et al. 2021; Qian et al. 2019; Adhikari 2016; He and Tian 2013; Yu 2008). Nonetheless, there is a lack of research about the antecedents of the information intermediaries’ dimension of the CIE, including the firm’s factors that significantly affect analysts’ predictions (Bouteska and Mili 2022). Previous studies in this area have focused on firms’ financials and governance, such as size (Bonini et al. 2010), risk (Kerl 2011), bankruptcy (Clarke et al. 2006), ownership structure, board characteristics (Bouteska and Mili 2022), and ESG performance (Luo and Wu 2022)1. Surprisingly, to the best of our knowledge, no previous studies have examined the effect of corporate ethics on analysts’ behavior and performance.

Since the beginning of the century, the issue of corporate ethics has gained prominence following well-publicized accounting failures, involving large companies such as Enron, WorldCom, and Xerox, and, more recently, following the credit crisis in the United States (Kaptein 2010). Several authors (e.g., Toffler and Reingold 2003; Palazzo 2007; Chua and Rahman 2011; Biggerstaff et al. 2015) argue that the unethical behavior of these companies and their leaders is the root cause of these debacles and the undermining of investors’ trust in public companies and their managers. This is now even more the case, given the major money-laundering scandals of the past few years (e.g., Lloyds Bank, Danske Bank). The topic of corporate ethics is therefore gaining unprecedented attention, and the business community is under pressure to improve its ethical practices (Chua and Rahman 2011; Biggerstaff et al. 2015, Parris et al. 2016). Prior research has evidenced the significant positive impact of corporate ethics on several business outcomes, including improved reputation, client fidelity and trust, staff productivity and engagement (Gino and Pierce 2009; McShane and Von Glinow 2021; Sardžoska and Tang 2015), and sustainable performance (Cheung and Lai 2023; Aftab et al. 2022).

This study aims to contribute to the growing literature about the antecedents of financial analysts’ behavior and the more established research on the consequences of corporate ethics, by examining whether a higher corporate ethical commitment is associated with more favorable analyst outputs. Using a sample of 5276 firm-year observations from 780 publicly listed US companies over the 2010–2016 period, the empirical analyses provide evidence that companies that are known to behave ethically attract more coverage from financial analysts. In addition, the results document that analysts issue more precise and less dispersed earnings forecasts for ethically committed companies. These findings, which hold across a battery of additional tests and robustness checks, suggest that corporate ethics commitments enhance the information intermediation dimension of the company’s information environment, which is expected to greatly benefit the firm’s stakeholders and capital markets.

The reminder of this paper is organized as follows. Section 2 presents the study’s theoretical background and hypotheses. Section 3 describes the research method. Section 4 provides the results. Section 5 discusses the results and concludes.

2. Literature Review and Hypotheses Development

2.1. Theories of Corporate Ethics

Business ethics in general and corporate ethics in particular have drawn the attention of scholars, practitioners, and regulators for decades due to their importance in inhibiting unethical conduct and cultivating a trusting environment among business partners and capital market participants. Overall, the corporate ethics literature provides three main theoretical approaches to explain the role of firms’ ethical commitment in enhancing their corporate transparency: the virtue theory, the deontology approach, and the consequential perspective. Following Thomas Aquinas’ moral philosophy, das Neves and Vaccaro (2013) state that transparency is necessary to achieve the virtues of honesty, justice, and prudence. They define transparency as being honest and ensuring the truth by giving an exact and well-intentioned image of an object. Therefore, the mere disclosure of information, although imperative, is not sufficient to accomplish transparency. For this reason, the authors propose to measure the distance between the truth and the information disclosed by two attributes, relevance and completeness, and claim that only complete and relevant information guarantees the truth. While the disclosure of misleading information is always ethically unacceptable and contradicts the virtue of honesty, the authors argue that restraint in the disclosure of information can be ethically desirable under certain circumstances to achieve the virtues of prudence and justice. Sometimes these virtues require the careful control of information by keeping information confidential when its disclosure could have negative or unfair consequences.

Vaccaro and Madsen (2009) adhere to the principle of justice of John Rawls and suggest that an individual’s right to know the potential personal effects of a company’s activities is an example of a fundamental freedom. Thus, the decision to disclose information is governed by the principle of stakeholders’ “right to information.” According to this deontological perspective, firms have a moral obligation to be transparent and to provide exact and timely information to their stakeholders. Economic actors, individuals, and companies must be transparent vis-à-vis stakeholders and accountable for the impacts of their decisions and activities (Vaccaro and Madsen 2009). However, similar to the virtue theory, economic agents may decide to withhold information due to other ethical considerations, including privacy and the poor reliability of the information, as well as socioeconomic forces such as pressure from competitors, investors, and government institutions (Fung et al. 2007; Hess 2007).

Drawing on the consequential approach, Tapscott and Ticoll (2003) argue that, since the financial scandals of the early 2000s, when public trust in the business world fell apart, companies that demonstrate the ethical values of openness and frankness retain a competitive advantage over their peers. Previous studies show that the more transparent the company, the more the stakeholders are convinced of its ethical commitment and the more they trust its managers (e.g., Cramer 2003; Williams 2005; Vaccaro and Madsen 2007, 2009). In fact, sound transparency practices promote stakeholder confidence (Vaccaro and Madsen 2009) and strengthen corporate reputation (Linthicum et al. 2010). Pr. Johan Graafland’s most recent textbook, entitled Ethics and Economics, is entirely devoted to the explanation of the critical role played by the virtue, deontology, and consequential perspectives in the inception, growth, and success of the modern Western economic model. In particular, the economic and regulatory foundations and subsequent evolutions of the free American financial markets, which host two of the largest stock exchanges in the world, the New York Stock Exchange (NYSE) and the NASDAQ, are primarily based on and guided by these theories (Graafland 2022).

Modern financial markets have been built around the idea of a separation between capital providers (or principals) and their agents (or managers), who are entrusted with the running of the company on behalf of the shareholders. Due to the information asymmetry about managers’ behavior, investors are compelled to depend on the transparency of the company’s information ecosystem, especially the information disclosed by public companies under the scrutiny and supervision of the securities commissions. The academic and professional literature, as well as the financial press, show that the effective functioning of the US economy and its capital markets are inextricably dependent on investors’ confidence in the integrity and faithfulness of the information disclosed by listed firms. Each time the lack of transparency in these companies is exposed through major accounting scandals, trust between investors and large corporations’ wanes, and financial markets decline (Peasnell et al. 2011).

2.2. Corporate Ethics and Transparency

In their seminal paper, Beyer et al. (2010) structure the concept of the CIE around three distinct dimensions, including statutory reporting, discretionary disclosures, and financial intermediation. A large and constantly expanding body of research (e.g., Labelle et al. 2010; Choi and Pae 2011; Elayan et al. 2016; Cheung and Lai 2023) has studied the impact of corporate ethics on the first two dimensions of the CIE (Bushman et al. 2004). Most of the early empirical research that has examined the link between corporate ethics and the quality of financial reporting has focused on earnings management (Chih et al. 2008; Labelle et al. 2010; Kim et al. 2012; Elayan et al. 2016). Elayan et al. (2016) finds a positive relation between changes in the ethical performance of companies, measured by the Covalence Ethical Quote, and the quality of financial information proxied by earnings management. Later studies have investigated the quality of auditing (Houqe et al. 2015) and other proxies of the quality of financial disclosures. For example, Felo (2000) shows that, according to analysts’ assessments, financial disclosures made by companies with an ethics program supervised by the board of directors are more credible than those who are not supervised by the board. Afterwards, using the results of the Standard & Poor’s Transparency and Disclosure Survey, Felo (2007) found that companies with an ethics program overseen by the board of directors disclose more general information, more financial information, and more information on management and the board of directors. Chih et al. (2008) provide evidence that companies included in the FTSE4Good index are less likely to smooth their results to avoid the disclosure of losses or decreases in profit.

Using the KLD indexes of social responsibility, Hong and Andersen (2011) and Kim et al. (2012) indicate that companies that are more socially responsible are less likely to manipulate their earnings. More recently, Mao et al. (2024) document a negative relationship between ESG performance and earnings management. In addition, this association is positively moderated by ESG rating divergence. Wu and Zhou (2022) find that companies that adopt integrated reporting are less (more) inclined to engage in accrual-based (real) earnings management. Finally, Wu et al. (2024) provide evidence of a significant negative relationship between ESG ratings and real earnings management. Furthermore, this negative association is accentuated by internal control quality, external supervision, analyst attention, and corporate governance2.

Taken together, these studies suggest that corporate ethics commitments improve the quality of financial reporting. In fact, companies have a certain degree of flexibility when applying accounting standards to prepare their financial statements. However, the excessive manipulation of accounting information while remaining within lawful limits (Choi and Pae 2011) is considered unethical if it could mislead current and potential investors and deters the public’s confidence in managers and financial markets (Graafland 2022). Therefore, a firm’s degree of ethical commitment can be decisive regarding the faithfulness of its accounting choices, since it influences the way the company applies responsible and reasonable professional judgment to interpret accounting regulations (Jones 1991; Loe et al. 2000).

In the extant corporate ethics literature, previous studies have focused on one dimension of the CIE, namely, the quality of mandatory disclosures, with limited or no research on the dimensions of voluntary reporting and third-party intermediaries. This study aims at addressing this research gap by investigating the relationship between corporate ethics and the information intermediation element of the CIE. The Western capital markets, especially in the US, involve many third-party intermediaries such as auditors, brokers, and the financial press (Huang et al. 2022). Given the critical role played by financial analysts in enriching companies’ information ecosystems, this paper concentrates on the effect of corporate ethics on financial analysts’ behavior and performance.

2.3. Corporate Ethics and Financial Analysts’ Behavior

Financial analysts are considered among the most important third-party intermediaries in the American capital markets (Bradshaw et al. 2021). They are commonly viewed as one of the most experienced and sophisticated users of corporate reporting. By issuing forecasts about firm’s future performance, producing research reports, and assessing the value creation potential of the securities issued by public companies (Jegadeesh et al. 2004; Bradshaw et al. 2021; Bouteska and Mili 2022), they play a pivotal role in improving the quality, relevance, faithfulness, and accessibility of the information available to investors, thereby enhancing market trust, efficiency, and liquidity (Cao et al. 2022). For instance, Dyck et al. (2010) indicate that compared to auditors and regulators, financial analysts are more likely to detect corporate fraud.

Prior research on financial analysts has studied the consequences of their behavior, including its impact on investor decisions (e.g., Bhat et al. 2006; Ramnath et al. 2008; Dhaliwal et al. 2012; Lang and Lundholm 1996; Roulstone 2003; Lang et al. 2012; Gu and Hackbarth 2013; Hui and Matsunaga 2015; Balakrishnan et al. 2019). Brauer and Wiersema (2018) argue that analysts can even influence the media and investor response to corporate misconduct by highlighting the implications of these unlawful actions on a firm’s performance. While several scholars have studied the various consequences of analysts’ behavior, Brauer and Wiersema (2018) note the sparsity of the research about its antecedents. In this regard, Luo et al. (2015), Alazzani et al. (2021), and Luo and Wu (2022) show that analysts do consider and assess a firm’s social performance in their forecasts. In the same vein, Bouteska and Mili (2022) and Yu and Wang (2018) document the positive impact of the quality of the company’s corporate governance system on analysts’ performance. In addition, a burgeoning literature has examined the effects of analysts’ individual (e.g., Jiang et al. 2016) and organizational (e.g., Huang et al. 2022) attributes on their behavior. In their paper, Brauer and Wiersema (2018) call for future work on the effects of corporate ethics (among other factors) on analysts’ forecasts. To the best of our knowledge, this is the first study investigating the impact of corporate ethics on financial analysts’ behavior and performance.

The focus on companies’ ethical commitment distinguishes our research from prior studies that investigate the impact of CSR and ESG on the financial intermediation dimension of the CIE. As described above, the concept of ethics is grounded in moral philosophy, involving the notions of right/wrong, justice, fairness, integrity, etc. (Carroll 2000; Fukukawa et al. 2007). Approaches to corporate ethics revolve around ethical principles and values that guide the behaviors and actions of a company and its main constituents such as board members, executives, and employees, helping them determine the appropriate, or ethical, course of action in various business contexts and circumstances, ranging from routine operations to strategic decisions (Schermerhorn 2002; Fischer 2004). Therefore, the scope of corporate ethics is extremely broad. It involves each and every employee, activity, and hierarchical level of a company, and influences and guides all its internal and external decisions and actions. However, the concepts of ESG and CSR are rooted in the principles of sustainability, social equity, and responsible governance. More specifically, ESG is explicitly defined around three main pillars (environmental, social, governance), and CSR revolves around community and societal welfare (Nugroho et al. 2024). Most importantly, these principles include practical considerations about specific societal and environmental issues (Furlotti and Mazza 2022). In fact, ESG and CSR focus on the external effects of the company’s operations and activities on its different stakeholders (Furlotti and Mazza 2022). For instance, the European Commission Green Paper (European Commission 2001) defines CSR as “a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis.” (p. 6). By their very nature, ESG and CSR are, hence, more explicit, tangible, and visible than corporate ethics, and their scopes are narrower; they emphasize actions over intentions. For this reason, they are easier to measure and link to performance indicators. Although the concepts of business ethics and ESG/CSR are very distinct conceptually, they are related because ESG and CSR can be influenced by ethical principles and values. This is not a surprise, since corporate ethics commitments represent the moral backbone of the company and influence all its activities and operations.

Prior research has studied two main aspects of financial analysts’ behavior and performance: analyst following (or coverage) and the characteristics of their earnings forecasts, that is, accuracy (or precision) and consensus (or dispersion). To properly assess the financial performance of a firm, financial analysts need reliable and timely information that reflects its economic reality (Krishnan and Parsons 2008). Previous studies show that analysts are more likely to follow companies that are more transparent and honest (Bhushan 1989; Lang and Lundholm 1996; Healy et al. 1999; Bushman et al. 2004). On the other hand, Felo (2007) provides evidence that ethical companies, which have ethics programs overseen by the board of directors, are more compliant with the principle of stakeholders’ right to information and disclose more financial and non-financial information. Therefore, consistent with the consequential approach, we expect that companies that are more ethically committed will be followed by a higher number of analysts. In addition, a firm’s involvement in ethical controversies increases its riskiness and reduces its appeal to investors due to the ensuing depletion of public trust (Yu et al. 2022; Wong and Zhang 2022). As a result, financial analysts will be less inclined towards following the company because of the diminished investors’ interest and the analysts’ inability to provide accurate forecasts due to the lack of reliable information about the firm (Zadeh et al. 2021). Accordingly, our first hypothesis is formulated as follows:

Hypothesis 1:

Corporate ethics are positively associated with analysts’ coverage.

Financial analysts make their earnings predictions based on the corporate information ecosystem, which includes mandatory and voluntary information disclosed by the company, but also private information collected by the analysts from other diverse sources such as news articles, industry reports, economic indicators and data, third-party research and databases, and social media platforms and online forums. When the company’s information ecosystem is richer, analysts will have more relevant and reliable information at their disposal. Hence, they will be able to develop more accurate earnings forecasts (Lang and Maffett 2011). Lang and Lundholm (1996) provide evidence that the analysts’ earnings forecasts are more consistent with the company’s actual earnings when the firm’s disclosures are more informative about its future performance. Similarly, Chiang and Chia (2005) argue that corporate transparency yields more favorable and accurate forecasts from financial analysts. As documented by Felo (2007) and Labelle et al. (2010), firms that are more ethical are more honest and tend to disclose more financial information of a higher quality, in line with the deontological perspective. As a result, we propose that ethical companies gain substantial competitive advantages as stipulated by the consequential approach and will be associated with more accurate analyst forecasts. Furthermore, firms involved in ethical incidents are riskier and more volatile. As a result, it will be more difficult for financial analysts to issue highly precise forecasts about the future performance of those firms. Therefore, our second hypothesis is stated as follows:

Hypothesis 2:

Corporate ethics are positively associated with analysts’ forecasts accuracy.

Analysts’ decision-making processes are overly complex and involve numerous stages and inputs (Beccalli et al. 2015). According to Huang et al. (2022), individual analysts do not have access to the same technologies and sources of information within their organizations and networks. Moreover, they have different types and levels of knowledge, experience, sophistication, and professionalism, and different personality traits and backgrounds (Cao et al. 2022; Fang and Yasuda 2014). Hence, analysts following the same firm will naturally have dissimilar inputs to their individual decision-making processes, will follow disparate heuristics, and will probably produce discrepant forecasts (Herrmann and Thomas 2005). Lang and Lundholm (1996) show that financial analysts’ forecasts are less dispersed when the firm is more transparent, which is the case for ethical entities (Felo 2007; Labelle et al. 2010). This is explained by the fact that the honest companies that respect investors’ right to information provide analysts with the same sufficient quantity and appropriate quality of information to develop their forecasts, and analysts will, hence, have little or no need to seek extensive additional input data (Gul et al. 2013). In addition, consistent with the consequential approach, the richer information landscape of ethically committed companies will reduce the complexity and uncertainty of analysts’ forecasting models, thereby reducing divergence between the analysts’ model outputs. On another note, companies that have ethical issues will be inherently riskier and more challenging to evaluate. Even analysts who have access to similar information about the firm may have divergent opinions about the potential detrimental effects of an ethical incident on the entity’s performance and sustainability. To summarize, because ethical companies are more transparent, less risky, and easier to assess, we expect that they will be associated with a higher consensus among analysts. Our third hypothesis is, therefore, stipulated as follows:

Hypothesis 3:

Corporate ethics are positively associated with analysts’ forecasts consensus.

3. Method

3.1. Empirical Model

To test our research hypotheses, we use the following regression model:

[Trans]_i = α * Ethic + ⅀β * Control variables + ε

where [Trans]_i (i = 1, 2, 3) represents the three dependent variables that capture the information intermediation dimension of the CIE, that is, financial analyst following, the accuracy of analysts’ forecasts, and analysts’ consensus. The variable of interest, Ethic, corresponds to corporate ethics as proxied by the ethical controversies score developed by Sustainalytics.

3.2. Variables

3.2.1. Dependent Variables

In the accounting and corporate finance literature, the information intermediation dimension of a company’s information ecosystem is commonly captured using three variables, including the extent of the firm’s coverage by financial analysts (or analyst following), the accuracy (or precision) of analysts’ earnings forecasts, and the consensus (or dispersion) of analysts’ earnings projections (e.g., Bushman et al. 2004; Lang et al. 2012; Gu and Hackbarth 2013; Armstrong et al. 2014). Following Byard et al. (2006) and Sengupta and Zhang (2015), we measure financial analyst following by the average number of analysts who follow the company during the last month preceding the end of the fiscal year. Similar to Lang and Lundholm (1996), Lang et al. (2012), and Chen et al. (2014), we measure the accuracy of analysts’ forecasts by the opposite (minus) of the absolute value of the difference between the company’s current annual earnings per share and the average of analysts’ forecasts of earnings per share disclosed during the period prior to the end of the fiscal year. Finally, like Armstrong et al. (2014), we measure analysts’ consensus by the opposite (minus) of the standard deviation of average analysts’ forecasts divided by the average consensus forecast announced during the period before the end of the fiscal year.

3.2.2. Independent Variable

Corporate ethics commitments are measured using the business ethics incidents score developed by Sustainalytics for a sample of 6000 companies from 50 countries over the 2009–2017 period (now called “legacy ESG ratings”). This unique score takes into account actual incidents related to financial accounting misconduct, fraudulent reporting, tax avoidance and evasion, bribery and corruption, money laundering, and breaches of intellectual property rights. For food companies, the indicator also considers animal welfare incidents. The business ethics incidents score is determined based on a thorough controversy analysis conducted by Sustainalytics experts, which considers both the number and the gravity of the firm’s ethical controversies, including potential impacts on stakeholders and firm reputation. To ensure the validity of their score, Sustainalytics experts leverage artificial intelligence tools to collect data about potential ethical issues and risks from various sources such as regulatory filings, public documents, news and media coverage, NGOs, and industry associations. The Sustainalytics ethics score ranges from a minimum value of 0 to a maximum value of 100. The higher the score, the less the company engages in ethical misconduct.

This exclusive score from Sustainalytics has the merit of overcoming the limits of the internal measures of corporate ethics commonly used in the extant literature, such as the code of ethics and the ethics programs adopted by companies. It is less prone to manipulation and “ethics washing” by managers since it is based on genuine facts and actual ethical incidents, not perceptions or impressions. Unfortunately, this score was discontinued in 2017. In the subsequent Sustainalytics “ESG Risk Ratings” database launched in 2018, the concept of “corporate ethics” per se is not really measured. In fact, the measurement of the new “business ethics” score (MEI.4) follows a completely different methodology, as it provides a rating of “company’s management of risks related to business ethics” (Sustainalytics 2024) and is initially measured at the industry level. Therefore, compared to our corporate ethics proxy, which focuses on the actual ethical incidents in which the company was involved, the new score is less specific, adopts a risk management lens, and involves a significant amount of judgement.

3.2.3. Control Variables

Prior research on corporate transparency identifies several factors that could influence transparency, particularly the following firm characteristics, size, financial performance, MTB, leverage, governance, and loss. Firm size is measured by the logarithm of total assets. Financial performance is measured by the return on assets. MTB is measured by the year-end market price per share divided by the book value per share ratio. Leverage is measured by total debt scaled by total assets. Corporate governance is measured by the Sustainalytics governance score, which considers various indicators of good governance practices such as board independence, board diversity, and the separation between the chief executive officer and the chairman. Loss is a dichotomous variable that takes the value of one if the company discloses a loss in the current year, and zero otherwise. Finally, consistent with prior research on financial analysts, we control for the forecast horizon, which is defined as the number of days that elapses between the announcement of the earnings forecast by the analyst and the disclosure of the actual earnings by the company.

4. Results

4.1. Sample and Data

The initial sample is composed of 1472 US companies from the Sustainalytics ethics database over the 2010–2016 period. Our study period starts in 2010, at a time when the American economy and markets started to recover from the 2007–2008 global financial crisis. This crisis has reignited the debate about the importance of transparency and ethical conduct and practices in the US markets. In addition, 2010 was marked by the introduction of an important piece of legislation in the US, the Dodd–Frank Wall Street Reform and Consumer Protection Act, which had a significant impact on the US capital markets. Although the focus of the new legislation was primarily on the financial industry, many of its provisions aimed at enforcing more ethical behavior through increased accountability and transparency.

Financial information about the companies was collected from the Compustat database. Information about financial analyst following and forecasts were obtained from the I/B/E/S database. Due to missing information, the merging of the data obtained from the sources mentioned above yielded a final sample that consisted of 5276 firm-year observations from about 780 companies over the period from 2010 to 2016. The companies in our final sample are indexed on the two major US stock exchanges, the NYSE (75% of the firms) and NASDAQ (25%).

4.2. Descriptive Statistics

4.2.1. Descriptive Statistics of the Sample

Table 1. Descriptive statistics of the sample.

Panel A. Frequency of observations per year
Year	Number	Percentage
2010	688	13.04
2011	743	14.08
2012	780	14.78
2013	780	14.78
2014	780	14.78
2015	767	14.54
2016	738	13.99
Total	5276	100
Panel B. Frequency of observations by industry
Industry	Frequency	Percentage
Agriculture and Forestry (01–09)	14	0.27
Mining (10–14)	321	6.08
Construction (15–17)	83	1.57
Manufacturing (20–39)	1880	35.63
Telecommunication (48)	121	2.29
Wholesale (50–51)	145	2.75
Retail (52–59)	403	7.64
Financial3 (60–67)	1089	20.64
Services (70–88)	702	13.31
Others	518	9.82
Total	5276	100.00

provides descriptive statistics about the study’s final sample by year (panel A) and by industry (panel B). It is worthy to note that the number of companies is evenly distributed over the years. Regarding industry classification (Hui and Matsunaga 2015), the table shows that the manufacturing (35.6%) and financial (20.7%) sectors dominate the sample composition.

4.2.2. Descriptive Statistics of Variables

Table 2. Descriptive statistics of the variables.

Variable	Mean	SD	Min	Median	Max
PRECISION	−0.082	0.225	−1.714	−0.02	0
CONSENSUS	−0.049	0.128	−1	−0.014	0
ANAL_FOL	15.475	8.204	1	15	38
ETHIC	95.694	11.734	0	100	100
TA (Total Assets)	39,214.72	163,000	227.792	7752.614	2,573,126
SIZE (log (TA))	9.155	1.402	6.503	8.956	13.502
ROA	0.053	0.065	−0.195	0.047	0.241
MTB	3.348	5.107	−18.505	2.439	31.961
LEV	0.619	0.217	0.126	0.612	1.296
FHORIZON	50.042	11.69	28	48	75
GOVERNANCE	64.356	9.048	37	65	95

presents the descriptive statistics of the variables. On average, companies are followed by 15 analysts. The average (median) accuracy of the analyst forecasts is −0.082 (−0.02), and the average (median) analyst consensus is −0.049 (−0.014). The independent variable corporate ethics, i.e., the firm’s ethical commitment, measured by the Sustainalytics business ethics incidents score, shows an average value of 95.69 with a standard deviation of 11.72.

Regarding control variables, the average size of the companies (total assets) in the sample is USD 39 trillion. To mitigate concerns with the measurement scale, the natural logarithm of total assets is used as a measure of firm size; the average size is therefore 9.15. The average return on assets is 5%. The MTB average value is 3.34. The average leverage level is 62%. The average score of the governance variable is 64.33. The average forecast horizon for analysts is 50 days.

4.3. Correlation Matrix

Table 3. Correlation matrix.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
(1) PRECISION	1.000
(2) CONSENSUS	0.636	1.000
(3) ANAL_FOL	0.182	0.115	1.000
(4) ETHIC	0.007	0.018	−0.208	1.000
(5) SIZE	0.043	0.029	0.342	−0.345	1.000
(6) ROA	0.297	0.335	0.149	0.026	−0.189	1.000
(7) LEV	−0.049	−0.051	−0.066	−0.079	0.359	−0.274	1.000
(8) MTB	0.080	0.070	0.129	0.005	−0.092	0.202	−0.044	1.000
(9) FHORIZON	−0.163	−0.151	−0.294	0.114	−0.299	−0.161	−0.072	−0.085	1.000
(10) GOVERN.	0.065	0.066	0.053	0.156	−0.097	0.079	−0.040	0.068	0.092	1.000
(11) LOSS	−0.329	−0.388	−0.044	−0.011	−0.095	−0.578	0.073	−0.043	0.173	0.012	1.000

presents the correlation matrix for all our variables. In addition, the variance inflation factor (VIF) test provides low coefficients for the independent variables. Thus, there are no multicollinearity issues in our empirical models.

4.4. Multivariate Analyses

Table 4. Regression of the characteristics of analysts’ forecasts on corporate ethics.

	(1)	(2)	(2)	(4)
	ANAL_FOL 1	PRECISION	CONSENSUS	TRANSIND
ETHIC	0.0009	0.0008	0.0004	0.0026
	(0.060) *2	(0.006) ***	(0.026) **	(0.011) **
SIZE	0.1855	0.0087	0.0055	0.0593
	(0.000) ***	(0.008) ***	(0.001) ***	(0.000) ***
GOVERNANCE	−0.0006	0.0002	0.0004	0.0021
	(0.517)	(0.658)	(0.074) *	(0.139)
MTB	0.0077	0.0007	0.0000	0.0024
	(0.000) ***	(0.069) *	(0.954)	(0.070) *
LEV	−0.2767	−0.0159	−0.0069	−0.1044
	(0.000) ***	(0.386)	(0.497)	(0.092) *
FHORIZON	−0.0072	−0.0015	−0.0005	−0.0047
	(0.000) ***	(0.000) ***	(0.010) **	(0.000) ***
LOSS	0.1419	−0.1758	−0.1172	−0.7485
	(0.000) ***	(0.000) ***	(0.000) ***	(0.000) ***
ROA	1.2020	0.4128	0.2816	1.8606
	(0.000) ***	(0.000) ***	(0.000) ***	(0.000) ***
_cons	1.4129	−0.1152	−0.1126	−0.4801
	(0.000) ***	(0.000) ***	(0.027) **	(0.007) ***
Year fixed effects	Yes	Yes	Yes	Yes
Industry fixed effects	Yes	Yes	Yes	Yes
P	0.000 ***	0.000 ***	0.000 ***	0.000 ***
R2		0.230	0.190	0.261
N	5246	5155	5243	5153
VIF	1.669	1.669	1.669	1.669

¹ The analyst following variable is discrete and its distribution follows a quasi-Poisson model. A binomial negative regression is therefore used for Model 1. ² p-values in parentheses * p < 0.10, ** p < 0.05, *** p < 0.01.

presents the regression results for each of the four models in our study. The dependent variables for the first three models (Models 1–3) are ANAL_FOL, PRECISION, and CONSENSUS, respectively. In Model 4, we use a blended transparency index (TRANSIND) as our dependent variable. This index is constructed using a factor analysis of the three dependent variables from Models 1–3, that is, analyst following, accuracy of analyst forecasts, and analyst consensus. In all our models, we winsorize the continuous variables at the 1st and 99th percentiles to mitigate the impact of outliers, and we control for year and industry fixed effects.

The Model 1 results document a positive and significant relationship between the number of analysts following a public company, ANAL_FOL, and its level of ethical commitment, ETHIC (p = 0.06). As predicted by our first hypothesis and consistent with the consequential approach for corporate ethics, this finding indicates that financial analysts are more motivated to cover ethically committed firms due to the higher investor interest in these companies and analysts’ greater ability to produce accurate earnings forecasts for more transparent and honest firms who value stakeholders’ right to information. Similarly, Model 2 indicates that the accuracy of analysts’ forecasts, PRECISION, is positively and significantly associated with corporate ethics commitments (p = 0.006). This result confirms our second hypothesis and reveals that companies that are more ethical, that is, less affected by ethical controversies, receive more accurate earnings forecasts from financial analysts. This finding could be attributed to the honest company’s richer and more reliable information environment, as explained in our literature review and hypotheses section. In Model 3, we find that our CONSENSUS variable is positively and significantly associated with ETHIC (p = 0.026), which supports our third hypothesis. This result suggests that the more ethical the company, the less dispersed the analysts’ forecasts. In line with the consequential and deontological ethical theories, analysts’ consensus could be viewed as one of the major competitive advantages that ethically committed firms benefit from. Collectively, Models 1–3 provide concurrent evidence that corporate ethics commitments significantly enhance financial analysts’ behavior and performance. Finally, the regression results from our Model 4 document a positive and significant association between our transparency index (TRANSIND) and ETHIC. Therefore, our overall conclusion from Models 1–3 above still holds when we combine the three independent variables, analyst coverage, accuracy, and consensus, into a single variable that measures analysts’ behavior and performance. It is worth noting that the results for our control variables are in accordance with previous studies in the field.

4.5. Additional Tests

To validate the robustness of our results, we run a battery of additional tests. Specifically, we use two approaches to address potential endogeneity issues between our analysts’ variables, corporate ethics, and control variables. First, we re-estimated our Models 1–4 while lagging all our variables by one year. The findings are illustrated in

Table 5. Main models with lagged independent variables.

	(1)	(3)	(2)	(4)
	ANAL_FOL	PRECISION	CONSENSUS	TRANSIND
L.ETHIC	0.0010	0.0012	0.0004	0.0037
	(0.035) **	(0.003) ***	(0.045) **	(0.005) ***
L.SIZE	0.1857	0.0083	0.0044	0.0554
	(0.000) ***	(0.032) **	(0.023) **	(0.000) ***
L.GOVERNANCE	−0.0011	0.0005	0.0004	0.0023
	(0.227)	(0.292)	(0.102)	(0.141)
L.ROA	1.2069	0.3379	0.2145	1.4886
	(0.000) ***	(0.000) ***	(0.000) ***	(0.000) ***
L.MTB	0.0087	0.0010	0.0000	0.0029
	(0.000) ***	(0.035) **	(0.940)	(0.073) *
L.LEV	−0.2849	−0.0039	0.0037	−0.0450
	(0.000) ***	(0.847)	(0.734)	(0.513)
L.FHORIZON	−0.0066	−0.0020	−0.0010	−0.0077
	(0.000) ***	(0.000) ***	(0.000) ***	(0.000) ***
L.LOSS	0.0975	−0.1248	−0.0810	−0.5242
	(0.003) ***	(0.000) ***	(0.000) ***	(0.000) ***
_cons	1.5919	−0.1209	−0.0816	−0.3686
	(0.000) ***	(0.048) **	(0.007) ***	(0.064) *
Fixed year effects	yes	Yes	yes	yes
Fixed industry effects	yes	Yes	yes	yes
P	0.000 ***	0.000 ***	0.000 ***	0.000 ***
R2		0.159	0.180	0.210
N	4475	4469	4411	4405
VIF	1.65	1.65	1.65	1.65

*: significant at 10%; **: significant at 5%; ***: significant at 1%.

and are consistent with our main results presented in the previous section.

Next, we use the two-stage least squares (2SLS) approach. In the first 2SLS equation, we regress ETHIC on our control variables (size, performance, MTB, leverage, and governance) as well as an instrumental variable, namely, the difference between the variable ETHIC and its industry average, as suggested by Zona et al. (2018). We find that the instrumental variable is significantly related to the ETHIC variable. In the second 2SLS equation, we regress the financial analyst-related variables (precision, consensus, and transparency index) on the control variables, while instrumenting the exogenous variable. As illustrated in

Table 6. 2 SLS regression.

	(1)	(2)	(3)	(4)
	ANAL_FOL	CONSENSUS	PRECISION	TRANSIND
ETHICres	0.158
	(0.447)
ETHIC	0.015	0.0004	0.0008	0.0026
	(0.026) *	(0.024) **	(0.005) ***	(0.010) ***
GOVERNANCE	0.042	0.0004	0.0002	0.0021
	(0.001) ***	(0.072) *	(0.657)	(0.136)
SIZE	1.982	0.0055	0.0087	0.0593
	(0.000) ***	(0.001) ***	(0.008) ***	(0.000) ***
ROA	−2.12	0.2686	0.4128	1.8607
	(0.063) *	(0.000) ***	(0.000) ***	(0.000) ***
MTB	0.029	0.0000	0.0007	0.0024
	(0.014) ***	(0.954)	(0.066) *	(0.068) *
LEV	−1.229	−0.0069	−0.0159	−0.1044
	(0.057) ***	(0.493)	(0.382)	(0.089) *
FHORIZON	−0.073	−0.0005	−0.0015	−0.0047
	(0.000) ***	(0.009) ***	(0.000) ***	(0.000) ***
LOSS	0.1420	−0.1172	−0.1758	−0.7485
	(0.000) ***	(0.000) ***	(0.000) ***	(0.000) ***
_cons	1.4088	−0.1151	−0.1126	−0.4793
	(0.000) ***	(0.000) ***	(0.025) **	(0.006) ***
P	0.000 ***	0.000 ***	0.000 ***	0.000 ***
R2	0.03	0.230	0.189	0.261
N	5245	5155	5243	5153

*: significant at 10%; **: significant at 5%; ***: significant at 1%.

, the findings show that the ETHIC variable remains positively and significantly correlated with analyst coverage, forecast accuracy, and consensus. Furthermore, we run the Sargan–Hansen test to verify that our instrument does not violate the exogeneity assumption. The results of the test are not significant for all our models, which indicates that the test’s null hypothesis cannot be rejected and suggests that our instrumental variable is valid, i.e., exogenous and does not correlate with the error term.

5. Discussion and Conclusions

Repeated high-profile accounting scandals in the past decades indicate that many companies tend to operate with business standards and practices that encourage financial opacity and deter transparency. As a result, public trust in public companies and the capital market has eroded to unparalleled levels. Many scholars have called attention to the pressing need to promote ethical awareness and conduct by firms and their managers, with the aim of enhancing companies’ information ecosystem and its transparency and rebuilding public confidence in corporations (Parris et al. 2016). This study contributes to this movement by investigating the effects of a firm’s ethical commitment on the financial intermediation dimension of its information environment, particularly financial analysts’ behavior and performance.

Using a sample of 5276 firm-year observations from 780 publicly listed US companies over the 2010–2016 period, our empirical analyses show that ethical companies attract more of a following by financial analysts. This result is consistent with virtue and deontological ethical theories and is explained by the fact that ethical companies are genuinely honest with their stakeholders and respect their right to reliable information. Moreover, as suggested by the consequential approach, their high transparency is a competitive advantage that appeals not only to investors but also financial analysts because it makes their projection processes more effective and efficient. Conversely, companies with ethical incidents are riskier, more challenging to evaluate, and less attractive to both investors and analysts (Yu et al. 2022; Wong and Zhang 2022).

Along the same lines, we document a significant and positive relationship between corporate ethics and the accuracy of analysts’ earnings forecasts. This finding is in accordance with the virtue and deontological perspectives. In fact, analysts following ethically committed companies greatly benefit from these firms’ honesty (das Neves and Vaccaro 2013) and the importance they attach to the principle of investors’ “right to information” (Vaccaro and Madsen 2009). Due to the abundance and faithfulness of the information that ethical companies share with the public, including financial analysts, the latter will be able to provide precise earnings projections for those firms, which is considered an important competitive advantage for publicly listed companies and increases the liquidity and stability of their stocks. Our results are, therefore, consistent with previous studies that showed a positive relationship between the quality of corporate governance and analyst forecast accuracy (e.g., Bouteska and Mili 2022; Yu et al. 2022; Byard et al. 2006). On the contrary, firms involved in ethical controversies are more opaque, riskier, more volatile, and hence more difficult to assess, which increases the risk that analysts fail in providing accurate predictions of their earnings.

Furthermore, our regression analyses reveal a significant positive association between corporate ethics and analyst consensus. This outcome is in line with the virtue and deontological ethical approaches as well. As explained earlier, ethical companies are more inclined to honestly adhere to capital providers’ rights for sufficient and appropriate information in order to make informed investment decisions. Therefore, analysts covering ethically committed firms will have access to most (and even all) of the information they need to form their opinions through companies’ mandatory and voluntary disclosures and will not need to collect considerable additional private information. In addition, this will reduce the complexity and uncertainty of the analysts’ forecasting models. With similar inputs and less complex estimations, analysts’ outputs will probably be quite similar and less dispersed, which results in the increased consensus. Overall, our results hold for a battery of additional tests and provide strong evidence that corporate ethics commitments enhance the third-party intermediaries dimension of the company’s information environment, which could contribute to the recovery of investors’ trust in corporations and capital markets.

This study makes a valuable contribution to the field of research concerned with the dynamics of a company’s information ecosystem, especially the topics of corporate transparency and third-party intermediaries. To the best of our knowledge, this is the first study that features corporate ethics as an important driver of financial analysts’ behavior and performance. The focus on companies’ ethical commitment distinguishes our research from prior studies that examined the effects of CSR and ESG on the financial intermediation dimension of the CIE. This study also adds to the literature about the determinants of financial analysts’ effectiveness (e.g., Bradshaw et al. 2021; Luo and Wu 2022; Bouteska and Mili 2022; Huang et al. 2022) in response to Brauer and Wiersema (2018)’s call for more work in this area. Additionally, this research contributes to the large literature about the positive consequences of business and corporate ethics (e.g., Duong et al. 2022; Elayan et al. 2016; Labelle et al. 2010) and uses a unique proxy for the concept of corporate ethics. Furthermore, the study has significant implications for managers, practitioners, investors, and regulators in accounting, finance, and business ethics. By documenting the significant benefits that could result from a firm’s superior ethical commitment, this study raises managers’ and practitioners’ awareness about the importance of compliance with the principle of stakeholders’ right to information and the virtue of honesty, as well as the considerable competitive advantages ensuing from increased analyst following, accuracy, and consensus. Capital providers need to increase their pressure on public companies to adhere to the highest ethical standards and practices. Having more ethical and transparent firms will facilitate analysts’ work, which eventually benefits investors by helping them achieve more effective and efficient capital allocation decisions. For this to happen, regulators need to develop new policies that encourage ethical conduct and transparency, while deterring deception, opacity, and financial fraud. In other words, adhering to business and corporate ethics is essential for enhancing the efficiency of financial markets and reducing the occurrence of financial scandals.

Our study offers several worthy research avenues. First, it would be interesting to replicate this study in different economic contexts to determine if the positive associations between corporate ethics and financial analysts’ behavior and performance remain significant in countries other than the United States. Given that the American legal and financial environment is stricter and more stakeholder-oriented, it would be desirable to examine the role of legal regimes and the degree of their stakeholder orientation on the relationships documented in the paper. Another possible avenue for future research is to study managers’ earnings forecasts and the corporate ethics nexus, especially since managers of ethical and transparent companies are likely to provide more accurate and more frequent earnings projections. Finally, future research can build on this study to identify the causal mechanisms by which corporate ethics commitments influence financial analysts’ work. This would certainly help us better understand the dynamics of this association, including the identification of potential moderating and/or mediating channels between the two variables.