Skip to Content
MDPIMDPI

Advanced

JRFMJournal of Risk and Financial Management
PDF
  • Article
  • Open Access

26 July 2026

17 Pages

The Positivity of Earnings Conference Calls’ Tone and Cost of Equity Capital: Empirical Evidence from FTSE All-Share Companies

,
,
,
and
1
Department of Accounting, Business School, The Hashemite University, Zarqa 13133, Jordan
2
Accounting Department, King Talal School of Business Technology, Princess Sumaya University for Technology, Amman 11941, Jordan
3
College of Business Administration, University of Business and Technology, Jeddah 21448, Saudi Arabia
4
Department of Accounting, Finance and Banking, College of Business and Finance, Ahlia University, Manama 10878, Bahrain
J. Risk Financ. Manag.2026, 19(8), 557;https://doi.org/10.3390/jrfm19080557
This article belongs to the Special Issue Accounting Information and Capital Markets
Version Notes
Order Reprints

Abstract

Based on agency theory, this study examines the association between the optimistic tone of earnings conference calls and the cost of equity capital using an unbalanced panel of 342 non-financial FTSE All-Share companies (987 firm-year observations) over the period 2010–2024. Earnings conference call tone is measured using the financial sentiment dictionary and analysed using NVivo 14 software. The cost of equity capital is estimated using an implied cost of equity model. Panel specification is determined using appropriate panel-data diagnostic tests, while robustness is assessed through lagged-tone regressions, an alternative cost of equity measure, and two-stage least-squares (2SLS) estimation to address potential endogeneity. The results show a significant negative association between optimistic earnings conference call tone and the cost of equity capital (β = −7.787, p < 0.01). A statistically significant reverse association is also documented: A statistically significant reverse association is also documented: a lower cost of equity is associated with a more optimistic tone in subsequent conference calls (β = −0.001, p < 0.01). This result is interpreted as evidence of an association rather than a causal effect. Both results remain robust across alternative model specifications, lagged-tone analyses, alternative cost of equity measures, and endogeneity controls. The findings indicate that positive and transparent voluntary communication, particularly through earnings conference calls, is associated with lower information asymmetry and a lower cost of equity capital. Firms that have not yet adopted this communication channel may consider incorporating earnings conference calls into their investor-relations strategies to enhance voluntary communication with investors. This study contributes to the disclosure literature by documenting statistically significant associations between earnings conference call tone and the cost of equity capital under two model specifications in the UK market and by providing comprehensive robustness evidence supporting the stability of the reported associations.

1. Introduction

Earnings conference calls have become an increasingly important medium through which firms communicate value-relevant financial and non-financial information to investors, analysts, and other market participants. Recent research has extended the analysis of conference calls beyond transcript content by examining fixed-income calls, audio-based emotional information, differences between managerial and analyst tone, and the quality of managers’ vocal delivery (De Franco et al., 2023; Chen et al., 2023; DeLisle et al., 2024; Baik et al., 2025). Collectively, this literature indicates that both the linguistic content and the delivery of conference call communication affect information processing, market reactions, analyst behaviour, and the corporate information environment.
The cost of equity capital is a central concept in corporate finance: it represents the minimum return required by equity investors given the risk of ownership, and firms use it as a capital-budgeting threshold (Firk et al., 2020). Theoretically, voluntary disclosure is associated with a lower cost of equity through two complementary channels: (i) lower estimation risk, as investors can assess firm prospects more accurately; and (ii) lower transaction costs arising from reduced information asymmetry (Botosan, 1997; Mangena et al., 2016).
This study investigates the association between the optimistic tone of earnings conference calls and the cost of equity capital for FTSE All-Share non-financial companies over the period 2010–2024. The UK market provides an appropriate setting because earnings conference calls are widely used by large listed UK firms yet remain understudied relative to the US context. Most prior research on conference call tone has relied on US data (e.g., Bochkay et al., 2020; Fu et al., 2021); UK-focused evidence is scarce (Fyodorova et al., 2019; Athanasakou et al., 2020), and no prior study has simultaneously examined the bidirectional association between call tone and cost of equity in the UK.
The study objectives are: (i) to examine the association between optimistic conference call tone and the cost of equity capital (H1); and (ii) to examine whether the cost of equity capital is associated with the subsequent optimistic tone of conference calls (H2). H1 is grounded in agency theory, which predicts that information asymmetry between managers and shareholders increases the cost of capital and that credible voluntary disclosure can mitigate this asymmetry (Jensen & Meckling, 1976). In contrast, H2 is examined as an exploratory empirical association intended to assess whether firms’ financing and information environment is systematically associated with subsequent managerial communication, rather than to imply a reverse causal mechanism.
The study contributes to the literature in three ways. First, it extends the conference call tone literature to a UK setting, where disclosure regulation and investor-relations practices differ from the US. Second, it examines the association under two complementary model specifications—a feature that has received limited attention in prior studies—and documents statistically significant associations when each variable is alternately specified as the dependent variable. These findings are interpreted as documenting statistical associations rather than bidirectional causality. Third, it applies a comprehensive robustness framework, including lagged-tone specifications, an alternative cost of equity proxy, and 2SLS estimation to address potential endogeneity.

2. Literature Review and Hypotheses Development

Information asymmetry arises when one party to a transaction possesses superior information relative to the other, distorting decision-making by investors and other stakeholders (Goel et al., 2021). In the context of equity markets, managers typically hold private information about firm prospects that is not fully reflected in public disclosures. Voluntary disclosure mechanisms—including earnings conference calls—are intended to reduce information asymmetry by providing timely, detailed, and interactive information to market participants (Diantimala et al., 2022; Awwad et al., 2025). Empirically, prior studies have generally found such disclosure to be associated with lower levels of information asymmetry.
Tone disclosure refers to the linguistic character of corporate communications: the degree to which language is positive, negative, or neutral (Li et al., 2022). Research consistently demonstrates that tone conveys incremental information beyond the numerical content of financial reports, influencing investor perceptions, analyst forecasts, and market prices (Li et al., 2022; Price et al., 2012). In the UK, growing regulatory emphasis on transparent and meaningful corporate reporting—including narrative disclosures and social and environmental impact reporting—has heightened attention to the quality and tone of management communications.
The cost of equity capital—the minimum rate of return required by equity investors—depends critically on perceived risk, which is in part a function of information quality. Mangena et al. (2016) show that information asymmetry reduces market liquidity in a firm’s shares, forcing issuers to offer equity at a discount. When firms improve their disclosures, investors no longer need to acquire costly private information, transaction costs fall, and firms can reduce the discount at which they issue equity.
Agency theory (Jensen & Meckling, 1976) frames the manager–shareholder relationship as a principal–agent problem in which information asymmetry creates scope for managerial actions that are not in shareholders’ interests. Voluntary disclosure—such as detailed and optimistic conference call communication—is expected to mitigate this problem by enhancing the credibility of management’s representations (DeAngelo, 1981a, 1981b) and by being associated with lower investor compensation for estimation risk (Botosan, 1997). Consequently, lower information asymmetry is theoretically associated with improved share liquidity (Muller & Riedl, 2002) and a lower cost of equity capital (Jensen & Meckling, 1976).

2.1. Earnings Conference Call Tone

De Franco et al. (2023) investigate the determinants and market role of fixed-income conference calls, finding that firms with greater debt and more negative news are more likely to hold such calls, and that the calls address the distinct informational needs of debt versus equity investors. da F. Tonin and Scherer (2022) study Brazilian firms from 2010 to 2017 and find that the tone of earnings conference call transcripts, measured using the Loughran and McDonald (2011) dictionary, significantly predicts abnormal stock returns following the calls. Allee et al. (2021) examine the interplay between product market competition, disclosure framing, and casting in US earnings conference calls, finding that a more positive tone is associated with lower product market competition and higher performance uncertainty. Call et al. (2021) document that buy-side analysts participate in approximately 18% of conference calls and tend to adopt a more uncertain and less favourable tone than sell-side analysts.
DeLisle et al. (2021) show that firms with stronger ESG performance exhibit a more optimistic management tone in quarterly earnings calls, even after controlling for financial performance. Bochkay et al. (2020) find that extreme language in conference calls is associated with higher trading volume and declining stock prices, with effects concentrated in firms operating in weaker information environments. Fu et al. (2021) provide evidence that a less optimistic year-end conference call tone predicts a higher probability of a stock price crash in the following year. Fyodorova et al. (2019) analyse London Stock Exchange and Moscow Exchange firms and find that conference call tone and linguistic characteristics significantly affect abnormal stock returns.
More recent research has broadened the conference call literature by examining additional linguistic, acoustic, and contextual dimensions of managerial communication. Chen et al. (2023) employ a deep learning approach to extract emotional information from earnings-call audio and show that vocal emotions provide information relevant to analysts’ subsequent behaviour. DeLisle et al. (2024) demonstrate that environmental and social incidents are followed by a more negative managerial tone and that disagreement between managerial and analyst tone affects both short-window market reactions and post-call pricing efficiency. Baik et al. (2025) further show that managers’ vocal delivery quality deteriorates when they communicate negative or less persistent information and that investors, analysts, and the financial media respond to these vocal characteristics. Wang et al. (2025) find that more positive managerial tone during earnings communication conferences reduces corporate information asymmetry, particularly during periods of heightened uncertainty. Collectively, these recent findings suggest that conference call tone is not merely a stylistic feature but an economically relevant component of the firm’s information environment.

2.2. Cost of Equity Capital and Disclosure

Botosan (1997) establishes that higher levels of voluntary disclosure are associated with a lower cost of equity capital, particularly for firms with low analyst following. Athanasakou et al. (2020) identify a U-shaped relationship between narrative disclosure in UK annual reports and the cost of equity, suggesting that unhelpful informational clutter at high disclosure levels may increase rather than reduce the cost of equity. Firk et al. (2020) show that a more balanced distribution of information in conference calls—as guided by a balanced scorecard framework—is inversely associated with the cost of capital, with the primary mechanism being a reduction in information asymmetry. Salvi et al. (2020) find that greater intellectual capital disclosure in integrated reports is negatively related to the cost of equity. Li and Liu (2018) document an inverse relationship between CSR disclosure quality and the cost of equity for Chinese listed firms, with the effect more pronounced for state-owned enterprises and large firms. Similarly, Awwad (2022) documents that corporate environmental disclosure is positively associated with firms’ market value, further supporting the capital-market relevance of transparent non-financial communication.
Recent evidence continues to support the capital-market relevance of disclosure tone and communication quality. DeLisle et al. (2024) show that conference call tone affects the efficiency with which investors incorporate firm-specific information into prices, especially when the firm’s information environment becomes less transparent following adverse environmental or social events. Similarly, Wang et al. (2025) document that positive managerial tone reduces information asymmetry during periods of crisis. Baik et al. (2025) demonstrate that the quality of vocal communication influences real-time trading activity and the responses of information intermediaries. Taken together, these studies provide contemporary support for the argument that clearer, more credible, and more positive communication is associated with lower information-processing costs and uncertainty, which are widely recognised as important channels through which voluntary disclosure is associated with investors’ required return on equity.
Building on agency theory and the foregoing literature, a more optimistic earnings conference call tone is expected to be associated with lower information asymmetry between management and investors. Lower information asymmetry is, in turn, theoretically associated with lower estimation risk and transaction costs arising from adverse selection in equity markets. Accordingly, firms exhibiting a more optimistic conference call tone are expected to exhibit a lower cost of equity capital (Botosan, 1997; Jensen & Meckling, 1976; Mangena et al., 2016). Consistent with this reasoning, the study proposes:
H1. The optimistic tone of earnings conference calls is negatively associated with the cost of equity capital in FTSE All-Share non-financial companies.
Although managers do not directly observe a firm’s implied cost of equity at the time of the conference call, they continuously receive signals regarding the firm’s financing environment through analyst reports, investor feedback, market valuation, financing conditions, and interactions with institutional investors. The implied cost of equity can therefore be viewed as a market-based indicator of the firm’s overall information environment rather than a managerial decision variable. Firms operating in a more favourable financing and information environment, reflected in a lower implied cost of equity, may exhibit a more optimistic tone in subsequent conference calls. Accordingly, the expected relationship should be interpreted as reflecting managerial responses to market conditions rather than managers consciously reacting to the implied cost of equity itself.
Accordingly, rather than proposing a reverse causal mechanism, the second hypothesis is formulated as an exploratory empirical examination of whether firms’ financing and information environment is systematically associated with subsequent conference call tone.
H2. The cost of equity capital is negatively associated with the subsequent optimistic tone of earnings conference calls in FTSE All-Share non-financial companies.

3. Methodology

3.1. Sample and Data Collection

As shown in Table 1, the sample comprises 342 unique non-financial companies listed on the London Stock Exchange and included in the FTSE All-Share Index over the period 2010–2024. Financial firms (n = 260) are excluded because their accounting structures differ fundamentally from those of non-financial companies, making financial ratios non-comparable. The sample period begins in 2010 to avoid the confounding effects of the 2007–2009 global financial crisis and extends through 2024 to provide the most recent available evidence. It should be noted that the reported sample size (342 firms) refers to the number of unique companies rather than firm-year observations. Extending the study period from 2019 to 2024 increased the longitudinal coverage of the panel dataset rather than the number of listed firms. Consequently, the final estimation sample consists of 987 firm-year observations after excluding observations with missing earnings conference call transcripts, unavailable analyst earnings forecasts required to estimate the implied cost of equity, or missing financial data. Therefore, the study employs an unbalanced panel dataset, reflecting differences in data availability across firms and years. Earnings conference call transcripts and financial data were obtained from the Refinitiv Eikon database.
Table 1. Describes the sample composition and industry distribution.
Therefore, the study employs an unbalanced panel dataset because conference call transcripts and analyst earnings forecasts are not available for every firm in every year throughout the sample period.

3.2. Variables and Measurements

3.2.1. Cost of Equity Capital (CoE)

The cost of equity capital is estimated using the PEG (price-earnings-growth) model of Easton (2004):
r = √[(eps2 − eps1)/P0]
where P0 is the current share price, and eps1 and eps2 are analyst consensus earnings forecasts for years t + 1 and t + 2, respectively. The model requires eps2 > eps1 and both forecasts to be positive. Share prices and analyst earnings projections are obtained from the I/B/E/S database via Refinitiv Eikon. The PEG model is selected because it has less demanding data requirements than the residual-income model (Gebhardt et al., 2001) and the abnormal-earnings-growth model (Ohlson & Juettner-Nauroth, 2005), and it has been widely employed in prior disclosure and cost of equity research (Mangena et al., 2016; Athanasakou et al., 2020). As a robustness check, the Gebhardt et al. (2001) model is also applied (Section 4.3).

3.2.2. Earnings Conference Call Tone (TONE)

Tone is primarily measured as the proportion of positive words in each earnings conference call transcript using the financial-domain dictionary developed by Loughran and McDonald (2011):
TONE = Number of Positive Words/Total Number of Words in the Earnings Conference Call
The primary tone measure captures the intensity of optimistic managerial language by scaling the number of positive words by the total number of words in each transcript. This measure is appropriate because the theoretical construct examined in this study is optimistic conference call tone rather than overall sentiment polarity. Measuring positive language separately also avoids imposing the assumption that positive and negative words carry equal but opposite informational effects.
The Loughran and McDonald (2011) dictionary is preferred over general-purpose sentiment dictionaries because it is specifically developed for financial communication and reduces the misclassification of finance-specific terminology. Earnings conference call transcripts are obtained from Refinitiv Eikon and processed using NVivo to generate word frequencies. This approach is consistent with prior studies examining managerial tone, including Davis et al. (2015) and Leung et al. (2015).
To assess whether the findings are sensitive to the construction of the primary tone measure, the robustness analysis additionally employs the following alternative net-tone measure:
NET_TONE = (Number of Positive Words − Number of Negative Words)/Total Number of Words in the Earnings Conference Call
Positive and negative words are identified using the Loughran and McDonald (2011) financial sentiment dictionary. This specification incorporates both positive and negative language while scaling net sentiment by the total length of the conference call transcript.

3.2.3. Control Variables

Following prior research on disclosure and cost of equity capital (Botosan et al., 2004; Espinosa & Trombetta, 2007; Mangena et al., 2016), the following control variables are included:
LnMKT_CAPIT: natural logarithm of market capitalisation at year-end (proxy for size and visibility).
BETA: market risk (systematic risk) estimated from 36 monthly returns against the FTSE All-Share Index.
LEV: financial leverage, calculated as total liabilities scaled by total assets.
LnBTM: natural logarithm of the book-to-market ratio (inverse of growth opportunities).
ROA: return on assets (%), measuring profitability.
Loss: binary variable equal to 1 if operating income is negative, and 0 otherwise.
Industry: industry fixed effects (10 FTSE sector categories).
Year: year fixed effects (2010–2024).
All continuous variables are winsorized at the 1st and 99th percentiles to mitigate the influence of extreme values.

3.3. Model Specification and Estimation

The following models are estimated to test H1 and H2, respectively:
CoEit = β0 + β1TONEit + β2LnMKT_CAPITit + β3BETAit + β4LEVit + β5LnBTMit + β6ROAit + β7Lossit + Industry + Year + εit
TONEit = β0 + β1CoEit + β2LnMKT_CAPITit + β3BETAit + β4LEVit + β5LnBTMit + β6ROAit + β7Lossit + Industry + Year + εit
Both models are estimated using pooled OLS with industry and year fixed effects (dummy variables) and Huber–White heteroscedasticity-robust standard errors. Hausman (1978) and Breusch–Pagan Lagrange multiplier tests are reported as diagnostic panel-data tests. Although the Hausman test favours a firm fixed-effects specification, the primary analysis employs pooled OLS with industry and year fixed effects.
This approach is adopted because the panel is highly unbalanced and many firms contribute only a limited number of observations. It preserves the available sample while controlling for common industry-specific and time-specific heterogeneity. Nevertheless, persistent unobserved firm characteristics may still influence both conference call tone and the cost of equity and should therefore be considered when interpreting the reported associations.

3.4. Endogeneity Considerations

A potential concern in estimating the association between TONE and CoE is simultaneity or joint determination. Specifically, a lower cost of equity may be associated with a more optimistic tone (H2), while both variables may also reflect common determinants that are not fully captured by the regression model. To assess the robustness of the documented associations, three complementary approaches are employed: (i) industry-year median TONE (excluding the focal firm) is used as an instrument in a 2SLS framework, exploiting variation in peer-industry communication norms; (ii) one-period lagged TONE (TONEi,t−1) is used as an alternative predictor in Model (1); and (iii) the Sargan–Hansen J-statistic is reported to assess the validity of the instrumental-variable specification. The corresponding first-stage regression results are presented in the first-stage regression table in the Results section.
Although the industry-year median tone instrument satisfies conventional statistical diagnostics, the exclusion restriction cannot be verified directly. Accordingly, the instrumental-variable estimates should be interpreted as complementary robustness evidence rather than definitive identification of causal effects.

4. Results

4.1. Descriptive Statistics

Table 2 presents descriptive statistics for all study variables. The mean cost of equity (CoE) is 0.696 (approximately 6.96%), with substantial cross-sectional variation (Std. Dev. = 0.548; range: 3.9–371.8% after winsorization). The high maximum and positive skewness (2.14) confirm that the right tail is trimmed by winsorization, which is appropriate. The mean TONE is 0.019, indicating that approximately 1.9% of words in a typical conference call are classified as positive according to the Loughran and McDonald (2011) list, consistent with prior studies (Davis et al., 2015). BETA averages 1.44, suggesting the sample firms carry above-market systematic risk on average. The mean LEV of 0.741 indicates that total liabilities represent approximately 74% of total assets. ROA averages 8.59%, reflecting healthy profitability in the non-financial sample. The Loss indicator has a very low mean (0.006), confirming that loss-making firm-year observations are rare after the exclusion of the global financial crisis period.
Table 2. Descriptive statistics of study variables.

4.2. Correlation Matrix

Table 3 presents Pearson correlation coefficients; the correlation between TONE and CoE is −0.108 and is statistically significant, providing initial univariate support for H1. BETA exhibits the strongest positive correlation with CoE (r = 0.421), consistent with the fundamental risk–return relationship. LnBTM and LEV are positively correlated with CoE, while ROA and LnMKT_CAPIT are negatively correlated. All VIF values in subsequent regression models are below 5 (Mean VIF = 2.43), confirming the absence of problematic multicollinearity.
Table 3. Pearson correlation matrix.

4.3. Regression Analysis

Table 4 reports the regression results for Model (1), with CoE as the dependent variable. The coefficient on TONE is negative and statistically significant (β = −7.787, robust SE = 2.678, t = −2.91, p < 0.01), supporting H1: a more optimistic tone in earnings conference calls is associated with a lower cost of equity capital. The estimated association is economically meaningful. A one-standard-deviation difference in TONE (Δ = 0.005) corresponds to an approximately 3.9-percentage-point difference in CoE (=7.787 × 0.005), representing a meaningful proportion of the sample mean CoE of 6.96%. Among the controls, BETA enters positively and highly significantly (β = 0.340, p < 0.001), consistent with the fundamental risk–return relationship. LnMKT_CAPIT and ROA enter negatively (p < 0.01 and p < 0.001, respectively), while LnBTM is also negatively and significantly associated with CoE. LEV and Loss are statistically insignificant. The model explains 34.65% of the variance in CoE (Adjusted R2 = 32.88%), and the F-statistic of 14.33 (p < 0.001) confirms overall model significance. The Hausman test (χ2(8) = 24.17, p = 0.002) favours the fixed-effects specification, and the Breusch–Pagan LM test (χ2(1) = 312.4, p < 0.001) confirms the presence of panel effects.
Table 4. Regression results for H1 (dependent variable: CoE).
Table 5 presents results for Model (2), with TONE as the dependent variable. The coefficient on CoE is negative and highly significant (β = −0.001, robust SE = 0.0003, t = −3.33, p < 0.001), supporting H2: a higher cost of equity is associated with a less optimistic conference call tone. This finding is consistent with the interpretation that firms operating in less favourable financing and information environments are associated with a less optimistic conference call tone, rather than implying a direct causal effect of the cost of equity on managerial communication. LnMKT_CAPIT and LnBTM enter significantly and negatively, while Loss is negatively significant (p < 0.05). The model R2 is 12.10% (Adjusted R2 = 10.52%), lower than for H1 but statistically meaningful given the inherent noisiness of tone measures.
Table 5. Regression results for H2 (dependent variable: TONE).
Table 6 reports five robustness specifications. Column (1) reproduces the baseline H1 results for comparison. Column (2) replaces TONE with its one-period lagged value (TONEi,t−1) to assess the temporal robustness of the documented association. The coefficient remains negative and significant (β = −6.912, p < 0.01), indicating that the observed association persists when prior-period tone is used, although this specification does not eliminate the possibility of simultaneity. Column (3) replaces the Easton (2004) PEG cost of equity with the Gebhardt et al. (2001) residual-income model; the coefficient on TONE remains negative and significant (β = −5.341, p < 0.001), indicating that the documented association is robust to an alternative cost of equity measure. Column (4) presents the second-stage 2SLS results using industry-year median TONE as an instrument. The estimated association remains negative and statistically significant (β = −8.103, p < 0.001). The corresponding first-stage estimates are reported separately in Table 7. The industry-year median TONE instrument is positively and statistically significantly associated with firm-level TONE (β = 0.684, robust SE = 0.0996, t = 6.87, p < 0.001). The first-stage F-statistic of 47.3 exceeds the Stock and Yogo (2005) critical value of 16.38, providing evidence against weak-instrument concerns. The Sargan–Hansen J-statistic (χ2(1) = 1.12, p = 0.29) does not reject the overidentifying restrictions. Nevertheless, these diagnostics provide supporting evidence regarding instrument relevance and the robustness of the specification rather than definitive evidence of causal identification. Column (5) replaces the primary positive-tone ratio with NET_TONE, calculated as the difference between positive and negative words divided by the total number of transcript words. The coefficient on NET_TONE remains negative and statistically significant (β = −0.184, p < 0.05), indicating that the documented negative association remains robust when both positive and negative language are incorporated into the tone measure.
Table 6. Robustness checks—alternative specifications.
Table 7. First-stage regression results for the instrumental-variable estimation (dependent variable: TONE).

5. Discussion

The results provide evidence consistent with both hypotheses. The negative coefficient on TONE in Model (1) is consistent with agency theory, suggesting that a more optimistic earnings conference call tone is associated with lower information asymmetry between management and investors. Within the agency theory framework, lower information asymmetry is theoretically associated with lower estimation risk, lower transaction costs, and ultimately a lower cost of equity (Botosan, 1997; Jensen & Meckling, 1976; Mangena et al., 2016). These findings should be interpreted as documenting statistically significant associations rather than definitive causal effects. They are also consistent with prior evidence from related settings. Allee et al. (2021) show that conference call tone shapes investor perceptions, Bochkay et al. (2020) document associations between extreme managerial language, stock prices, and trading volume, and Firk et al. (2020) report an inverse association between balanced conference call content and the cost of capital. The present study extends this literature by providing evidence from a UK setting and by quantifying the economic magnitude of the documented association between conference call tone and the cost of equity.
An alternative interpretation should also be acknowledged. A more optimistic conference call tone does not necessarily imply more informative or higher-quality disclosure. Positive language may also reflect managerial communication style, impression-management incentives, or strategic framing of corporate performance. Consequently, the documented association should not be interpreted as evidence that optimistic language itself improves the firm’s information environment. Rather, the findings indicate that optimistic conference call tone is associated with a financing environment consistent with lower information asymmetry, while recognising that multiple behavioural and disclosure-related mechanisms may contribute to this relationship.
In addition, the alternative NET_TONE specification reported in Column (5) provides further evidence regarding the robustness of the findings. The coefficient on NET_TONE remains negative and statistically significant, indicating that the documented association is not driven solely by measuring optimistic language as the proportion of positive words. Instead, the results remain consistent when conference call tone is measured using an alternative net-tone specification that incorporates both positive and negative words, thereby strengthening confidence in the robustness of the reported associations across alternative sentiment measures.
The results for H2 should not be interpreted as implying that managers directly observe or respond to the firm’s implied cost of equity when participating in earnings conference calls. Rather, the implied cost of equity serves as a market-based indicator of the firm’s broader financing and information environment. Managers continuously receive information from analysts, investors, capital markets, and financing activities, which collectively reflect investor confidence and financing conditions. Accordingly, the observed association is more appropriately interpreted as managerial communication adapting to favourable market conditions that are simultaneously reflected in a lower implied cost of equity. Furthermore, the second model should be viewed as providing exploratory empirical evidence regarding this association rather than confirmation of a reverse causal mechanism.
The robustness analyses provide additional evidence regarding the stability of the documented associations across alternative model specifications. The persistence of the negative coefficient under the lagged-tone specification (Column 2) indicates that the observed association remains stable when prior-period tone is considered, although this specification does not eliminate the possibility of simultaneity. The consistency of the results across two alternative cost of equity measures (Columns 1 and 3) indicates that the documented association is not sensitive to the measurement approach. Likewise, the second-stage 2SLS estimates reported in Column (4) of Table 6 provide complementary evidence after accounting for potential endogeneity. The corresponding first-stage results reported in Table 7 show that the industry-year median TONE instrument is positively and significantly associated with firm-level TONE and that the first-stage F-statistic exceeds the relevant Stock and Yogo (2005) critical value, supporting the relevance of the proposed instrument. However, given the observational nature of the study and the identifying assumptions underlying the instrumental-variable approach, the 2SLS analysis should be interpreted as supporting the robustness of the documented association rather than establishing a definitive causal relationship. In particular, the validity of the industry-year median TONE instrument ultimately depends on the identifying assumption that peer firms’ communication affects the focal firm’s cost of equity only through its own conference call tone. Although the statistical diagnostics support the instrument, this assumption cannot be tested directly and therefore should be interpreted with appropriate caution.
Furthermore, because the primary specification does not include firm fixed effects, part of the documented association may reflect persistent firm-specific characteristics, such as disclosure policy, managerial communication style, investor-relations practices, or corporate culture.

6. Practical Implications

The findings offer several practical implications for managers, investor-relations teams, regulators, and market participants. First, managers should recognise that earnings conference calls are not merely routine communication events but value-relevant disclosure mechanisms whose tone is associated with investors’ required returns. The study documents a significant association between a more optimistic conference call tone and a lower cost of equity capital, suggesting that clear, balanced, and credible communication is associated with lower investor uncertainty and perceived information risk.
Second, companies should seek to improve the quality of their investor-relations communication. This does not imply the use of exaggerated or misleading optimism; rather, firms should provide balanced, transparent, and confidence-building narratives supported by reliable financial information. Such communication is associated with lower information asymmetry between managers and investors and may contribute to a more favourable capital-market perception.
Third, the results are relevant for boards of directors and audit committees because they highlight the importance of monitoring not only the content of earnings conference calls but also the tone and clarity of managerial communication. Boards should encourage management to adopt structured communication policies that promote consistency, transparency, and credibility in interactions with analysts and investors.
Fourth, regulators and stock exchanges may benefit from encouraging the wider adoption of earnings conference calls, particularly among firms that rely primarily on traditional financial reporting. Since the study documents associations between voluntary communication, lower information asymmetry, and a lower cost of equity capital, promoting more interactive disclosure practices may contribute to enhancing market transparency and investor confidence.
Finally, investors and analysts may consider conference call tone as an additional indicator when assessing firm risk, disclosure quality, and financing conditions. The documented associations under the two model specifications suggest that conference call tone and firms’ financing environments are closely related, making tone analysis a useful complement to traditional financial analysis.

7. Limitations and Directions for Future Research

Despite its contributions, this study is subject to several limitations that should be acknowledged.
First, the analysis is restricted to non-financial companies listed in the FTSE All-Share Index. Consequently, the findings may not be fully generalisable to financial institutions, private firms, or companies operating in other institutional and regulatory environments. Future research could examine whether the observed relationship between earnings conference call tone and the cost of equity capital differs across countries, industries, and governance systems.
Second, although the study supplements the primary positive-tone measure with an alternative net-tone specification as a robustness check, the main analysis remains focused on optimistic language. Dictionary-based textual analysis is widely used in accounting and finance research; however, such measures may not fully capture contextual meaning, negation, sarcasm, linguistic nuances, or managerial impression-management strategies. Future research may employ advanced natural language processing techniques, including transformer-based models such as FinBERT, and compare their classifications with dictionary-based and human-coded measures to provide more context-sensitive assessments of managerial tone. Future research may also distinguish between informative optimism and impression-management behaviour by combining textual tone measures with disclosure-quality proxies, analyst disagreement, or manual coding of conference call narratives.
Third, the study focuses primarily on the positive dimension of conference call tone and measures tone using the complete earnings conference call transcript without distinguishing between prepared managerial remarks and the subsequent analyst question-and-answer session. Although this approach captures the overall earnings conference call communication environment, prior research suggests that these two components may convey different informational content and may be interpreted differently by capital-market participants. Future research could separately examine scripted presentations and live Q&A discussions to investigate whether the documented associations differ across these distinct disclosure environments. It could also investigate additional linguistic attributes, such as negative tone, uncertainty, readability, emotional intensity, managerial confidence, and analyst–management interaction dynamics during question-and-answer sessions.
Fourth, while the study applies several robustness and endogeneity controls, including lagged specifications and two-stage least-squares estimation, the possibility of omitted variables and unobservable managerial characteristics cannot be entirely eliminated. Although the second model documents a statistically significant association, it should be interpreted as exploratory evidence regarding the firm’s financing and information environment rather than confirmation of a reverse causal mechanism. Accordingly, the reverse association should be understood as reflecting the firm’s broader financing and information environment, which is observable to managers through market signals such as analyst feedback, investor sentiment, and financing conditions, rather than direct managerial responses to the implied cost of equity itself. Furthermore, although the instrumental-variable approach provides complementary robustness evidence, the exclusion restriction underlying the industry-year median tone instrument cannot be directly verified. Likewise, because the primary specification does not include firm fixed effects, part of the documented association may reflect persistent firm-specific characteristics—such as disclosure policy, investor-relations practices, managerial communication style, or corporate culture—rather than within-firm changes over time. The unbalanced nature of the panel means that incorporating firm fixed effects would substantially reduce the effective sample size. Future research may therefore exploit natural experiments, regulatory changes, quasi-experimental designs, other sources of exogenous variation, or longer balanced panel datasets incorporating firm fixed-effects models to provide stronger causal evidence and further clarify the mechanisms underlying this relationship.
Fifth, the study examines the relationship between conference call tone and the cost of equity capital without explicitly considering potential moderating or forward-looking factors. Future research could investigate the moderating roles of corporate governance quality, institutional ownership, analyst coverage, ESG performance, disclosure quality, market competition, and macroeconomic uncertainty. It may also incorporate forward-looking disclosure and growth-related variables, such as management earnings guidance, future-oriented statements, analyst earnings forecast revisions, investment outlook disclosures, and future realised revenue growth, to further examine whether the documented associations remain robust after controlling for firms’ expected future performance. The inclusion of such forward-looking variables may provide additional explanatory power and may influence the magnitude and statistical significance of the documented associations by capturing information beyond general linguistic optimism.
Finally, future studies may explore whether the influence of conference call tone varies during periods of economic crisis, market turbulence, or major geopolitical events. Examining the tone–cost of equity relationship during periods such as the COVID-19 pandemic or other significant economic shocks may provide additional insights into the role of managerial communication under conditions of heightened uncertainty.

8. Conclusions

This study examines the bidirectional association between the optimistic tone of earnings conference calls and the cost of equity capital using an unbalanced panel of 342 non-financial companies listed in the FTSE All-Share Index (987 firm-year observations) over the period 2010–2024. The results document a statistically significant negative association between optimistic conference call tone and the cost of equity capital (H1: β = −7.787, p < 0.01) and a statistically significant negative association in the reverse model specification between the cost of equity capital and subsequent conference call tone (H2: β = −0.001, p < 0.001). Both associations remain robust across alternative model specifications, including lagged-tone regressions, an alternative cost of equity measure, and two-stage least-squares (2SLS) estimation using an industry-peer instrument.
The study offers three practical recommendations. First, companies should seek to provide balanced, transparent, and credible earnings conference call communications. The evidence indicates that such communication is associated with a lower cost of equity capital and lower investor uncertainty. Second, regulators may consider encouraging the wider adoption of earnings conference calls, particularly among companies that rely primarily on traditional financial reporting. Third, companies should develop investor-relations strategies that pay close attention to both the content and tone of earnings communications as part of broader efforts to strengthen investor relationships and enhance disclosure quality.
Several limitations should be acknowledged. The sample is restricted to non-financial FTSE All-Share companies listed in the UK, limiting the generalisability of the findings to other markets and the financial sector. The study relies on the Loughran and McDonald (2011) positive-word dictionary, which, although widely validated for financial texts, does not fully capture contextual meaning, irony, or sentence-level sentiment. Future research could employ transformer-based natural language processing models (e.g., FinBERT) to obtain more context-sensitive measures of managerial tone. Future studies may also utilise natural experiments, regulatory changes, or quasi-experimental research designs to provide stronger causal evidence regarding the documented associations between conference call tone and the cost of equity capital. Extending the sample period beyond 2024, including the COVID-19 period and its aftermath, would also enable researchers to examine whether periods of heightened uncertainty moderate the documented association between conference call tone and the cost of equity capital.

Author Contributions

Conceptualization, S.K. and B.S.A.; methodology, A.H.R. and B.S.A.; software, R.B.; validation, S.K., A.H.R., R.B., R.F. and B.S.A.; formal analysis, B.S.A. and A.H.R.; investigation, S.K., A.H.R. and R.F.; resources, S.K.; data curation, R.B. and B.S.A.; writing—original draft preparation, S.K., R.F. and B.S.A.; writing—review and editing, A.H.R., R.F. and B.S.A.; visualization, R.B.; supervision, B.S.A.; project administration, B.S.A.; funding acquisition, S.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Ethical review and approval were waived for this study because it relies exclusively on publicly available secondary data obtained from corporate annual reports, earnings conference call transcripts, and financial databases. The study does not involve human participants, personal data, surveys, interviews, or any sensitive information. Therefore, institutional ethical approval was not required, and the research was conducted in accordance with accepted academic and research integrity standards.

Data Availability Statement

The data supporting the findings of this study were obtained from publicly available corporate annual reports, earnings conference call transcripts, and commercial financial databases. The dataset analysed during the current study is available from the corresponding author upon reasonable request, subject to the licencing and access restrictions of the original data providers.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Allee, K. D., Do, C., & Sterin, M. (2021). Product market competition, disclosure framing, and casting in earnings conference calls. Journal of Accounting and Economics, 72(1), 101405. [Google Scholar] [CrossRef] [Scilit]
  2. Athanasakou, V., Eugster, F., Schleicher, T., & Walker, M. (2020). Annual report narratives and the cost of equity capital: UK evidence of a U-shaped relation. European Accounting Review, 29(1), 27–54. [Google Scholar] [CrossRef] [Scilit]
  3. Awwad, B. S. (2022). The role of environmental disclosure in enhancing the market value of industrial and service companies listed on the Palestinian Stock Exchange. International Journal of Entrepreneurship, 26(2), 1–12. [Google Scholar] [CrossRef] [Scilit]
  4. Awwad, B. S., Alkababji, M. W., & Razia, B. S. (2025). Artificial intelligence and the quality of accounting information in Palestinian industrial companies. Journal of Financial Reporting and Accounting, 23(4), 1402–1430. [Google Scholar]
  5. Baik, B., Kim, A. G., Kim, D. S., & Yoon, S. (2025). Vocal delivery quality in earnings conference calls. Journal of Accounting and Economics, 80, 101763. [Google Scholar] [CrossRef] [Scilit]
  6. Bochkay, K., Hales, J., & Chava, S. (2020). Hyperbole or reality? Investor response to extreme language in earnings conference calls. The Accounting Review, 95(2), 31–60. [Google Scholar] [CrossRef] [Scilit]
  7. Botosan, C. A. (1997). Disclosure level and the cost of equity capital. The Accounting Review, 72(3), 323–349. [Google Scholar] [CrossRef] [Scilit]
  8. Botosan, C. A., Plumlee, M. A., & Xie, Y. (2004). The role of information precision in determining the cost of equity capital. Review of Accounting Studies, 9(2–3), 233–259. [Google Scholar] [CrossRef] [Scilit]
  9. Call, A. C., Sharp, N. Y., & Shohfi, T. D. (2021). Which buy-side institutions participate in public earnings conference calls? Implications for capital markets and sell-side coverage. Journal of Corporate Finance, 68, 101964. [Google Scholar] [CrossRef] [Scilit]
  10. Chen, Y., Han, D., & Zhou, X. (2023). Mining the emotional information in the audio of earnings conference calls: A deep learning approach for sentiment analysis of securities analysts’ follow-up behavior. International Review of Financial Analysis, 89, 102704. [Google Scholar] [CrossRef] [Scilit]
  11. da F. Tonin, J. M., & Scherer, L. M. (2022). Market reaction to the tones of earnings conference calls. Revista de Administração de Empresas, 62(1), e2020-0301. [Google Scholar] [CrossRef] [Scilit]
  12. Davis, A. K., Ge, W., Matsumoto, D., & Zhang, J. L. (2015). The effect of manager-specific optimism on the tone of earnings conference calls. Review of Accounting Studies, 20(2), 639–673. [Google Scholar] [CrossRef] [Scilit]
  13. DeAngelo, L. E. (1981a). Auditor independence, “low balling”, and disclosure regulation. Journal of Accounting and Economics, 3(2), 113–127. [Google Scholar] [CrossRef] [Scilit]
  14. DeAngelo, L. E. (1981b). Auditor size and audit quality. Journal of Accounting and Economics, 3(3), 183–199. [Google Scholar] [CrossRef] [Scilit]
  15. De Franco, G., Shohfi, T., Xu, D., & Zhu, Z. V. (2023). Fixed income conference calls. Journal of Accounting and Economics, 75(1), 101518. [Google Scholar] [CrossRef] [Scilit]
  16. DeLisle, R. J., Grant, A., & Mao, R. (2021). Does ESG decrease information asymmetry? Evidence from earnings conference call tones and subsequent returns. Working Paper. Available online: https://ssrn.com/abstract=3832745 (accessed on 15 May 2026).
  17. DeLisle, R. J., Grant, A., & Mao, R. (2024). Does environmental and social performance affect pricing efficiency? Evidence from earnings conference call tones. Journal of Corporate Finance, 86, 102585. [Google Scholar] [CrossRef] [Scilit]
  18. Diantimala, Y., Syahnur, S., & Islahuddin, I. (2022). Recursive correlation between voluntary disclosure, cost of capital, information asymmetry, and firm value. Cogent Business & Management, 9(1), 2154489. [Google Scholar] [CrossRef] [Scilit]
  19. Easton, P. D. (2004). PE ratios, PEG ratios, and estimating the implied expected rate of return on equity capital. The Accounting Review, 79(1), 73–95. [Google Scholar] [CrossRef] [Scilit]
  20. Espinosa, M., & Trombetta, M. (2007). Disclosure interactions and the cost of equity capital: Evidence from the Spanish continuous market. Journal of Business Finance & Accounting, 34(9–10), 1371–1392. [Google Scholar] [CrossRef] [Scilit]
  21. Firk, S., Hennig, J. C., & Wolff, M. (2020). Can the balanced scorecard help in designing conference calls? The effect of balanced information composition on the cost of capital. European Accounting Review, 29(1), 115–146. [Google Scholar] [CrossRef] [Scilit]
  22. Fu, X., Wu, X., & Zhang, Z. (2021). The information role of earnings conference call tone: Evidence from stock price crash risk. Journal of Business Ethics, 173(3), 643–660. [Google Scholar] [CrossRef] [Scilit]
  23. Fyodorova, E., Sayakhov, R., Demin, I., & Afanasyev, D. (2019). The influence of conference calls’ semantic characteristics on company market performance: Text analysis. Russian Journal of Economics, 5(3), 297–320. [Google Scholar] [CrossRef] [Scilit]
  24. Gebhardt, W. R., Lee, C. M. C., & Swaminathan, B. (2001). Toward an implied cost of capital. Journal of Accounting Research, 39(1), 135–176. [Google Scholar] [CrossRef] [Scilit]
  25. Goel, A., Tripathi, V., & Agarwal, M. (2021). Information asymmetry and stock returns. Journal of Advances in Management Research, 18(1), 85–112. [Google Scholar] [CrossRef] [Scilit]
  26. Hausman, J. A. (1978). Specification tests in econometrics. Econometrica, 46(6), 1251–1271. [Google Scholar] [CrossRef] [Scilit]
  27. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behaviour, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. [Google Scholar] [CrossRef] [Scilit]
  28. Leung, S., Parker, L., & Courtis, J. (2015). Impression management through minimal narrative disclosure in annual reports. The British Accounting Review, 47(3), 275–289. [Google Scholar] [CrossRef] [Scilit]
  29. Li, S., & Liu, C. (2018). Quality of corporate social responsibility disclosure and cost of equity capital: Lessons from China. Emerging Markets Finance and Trade, 54(11), 2472–2494. [Google Scholar] [CrossRef] [Scilit]
  30. Li, S., Wang, G., & Luo, Y. (2022). Tone of language, financial disclosure, and earnings management: A textual analysis of Form 20-F. Financial Innovation, 8(1), 43. [Google Scholar] [CrossRef] [Scilit]
  31. Loughran, T., & McDonald, B. (2011). When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. The Journal of Finance, 66(1), 35–65. [Google Scholar] [CrossRef] [Scilit]
  32. Mangena, M., Li, J., & Tauringana, V. (2016). Disentangling the effects of corporate disclosure on the cost of equity capital: A study of the role of intellectual capital disclosure. Journal of Accounting, Auditing & Finance, 31(1), 3–27. [Google Scholar] [CrossRef] [Scilit]
  33. Muller, K. A., & Riedl, E. J. (2002). External monitoring of property appraisal estimates and information asymmetry. Journal of Accounting Research, 40(3), 865–881. [Google Scholar] [CrossRef] [Scilit]
  34. Ohlson, J. A., & Juettner-Nauroth, B. E. (2005). Expected EPS and EPS growth as determinants of value. Review of Accounting Studies, 10(2–3), 349–365. [Google Scholar] [CrossRef] [Scilit]
  35. Price, S. M., Doran, J. S., Peterson, D. R., & Bliss, B. A. (2012). Earnings conference calls and stock returns: The incremental informativeness of textual tone. Journal of Banking & Finance, 36(4), 992–1011. [Google Scholar] [CrossRef] [Scilit]
  36. Salvi, A., Vitolla, F., Raimo, N., Rubino, M., & Petruzzella, F. (2020). Does intellectual capital disclosure affect the cost of equity capital? An empirical analysis in the integrated reporting context. Journal of Intellectual Capital, 21(6), 985–1007. [Google Scholar] [CrossRef] [Scilit]
  37. Stock, J. H., & Yogo, M. (2005). Testing for weak instruments in linear IV regression. In D. W. K. Andrews, & J. H. Stock (Eds.), Identification and inference for econometric models (pp. 80–108). Cambridge University Press. [Google Scholar] [CrossRef] [Scilit]
  38. Wang, Y., Chen, X., & Zhang, Y. (2025). Management tone and corporate information asymmetry in times of pandemic crisis. Journal of International Accounting, Auditing and Taxation, 59, 100708. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Explaining Behavioral Intention and Actual Use of Digital Payment Systems: A Structural Equation Modeling Approach Among University Students
Previous
Nonlinear Market Coupling During COVID-19 and the Global Financial Crisis: A Convergent Cross-Mapping Analysis of US and European Equity Indices
Next