Manager sentiment, policy uncertainty, ESG disclosure and firm performance: a large language model in corporate landscape

• Date: 23 July 2024
• Pages: 858-882
• DOI: https://doi.org/10.1108/IJAIM-08-2023-0206
• Published date: 23 July 2024
• Subject Matter: Accounting & finance,Accounting/accountancy,Accounting methods/systems
• Author: Asis Kumar Sahu,Byomakesh Debata,Saumya Ranjan Dash

Manager sentiment, policy

uncertainty, ESG disclosure and

f‌irm performance: a large language

model in corporate landscape

Asis Kumar Sahu and Byomakesh Debata

Department of Economics and Finance, Birla Institute of Technology and

Science –Pilani Campus, Pilani, India, and

Saumya Ranjan Dash

Department of Accounting and Finance, Indian Institute of Management Indore,

Indore, India

Abstract

Purpose –This study aims to examine theimpact of manager sentiment on the f‌irm performance (FP) of

Indian-listed nonf‌inancialf‌irms. Further, it endeavors to investigate the moderating role ofeconomic policy

uncertainty(EPU) and environment, social and governance (ESG) transparencyin this relationship.

Design/methodology/approach –A noble manager sentiment is introduced using FinBERT, a

bidirectional encoderrepresentation from a transformers (BERT)-type large languagemodel. Using this deep

learning-based natural language processing approach implemented through a Python-generated algorithm,

this study constructsa manager sentimentfor each f‌irm and year based on the management discussions and

analysis(MD&A) report. This research uses the system GMMto examine how manager sentiment affects FP.

Findings –The empirical resultssuggest that managers’optimistic outlook in MD&A corporatedisclosure

sections tends to present higher performance. This positive association remains consistent after several

robustness checks –usingpropensity score matching and instrumental variable approachto address further

endogeneity, using alternative proxies of manager sentiment and FP and conducting subsample analysis

based on f‌inancial constraints.Furthermore, the authors observe that the relationshipis more pronounced for

ESG-disclosedf‌irms and during the low EPU.

Practical implications –The resultsdemonstrate that the managersentiment stronglypredicts FP. Thus,

this studymay provide valuableinsight for academics,practitioners, investors,corporates and policymakers.

Originality/value –To the best of the authors’knowledge, this is the f‌irst study to predictFP by using

FinBERT-basedmanagerial sentiment, particularly in an emerging market context.

Keywords Manager sentiment, Large language model, Firm performance,

Economic policy uncertainty, ESG

Paper type Research paper

1. Introduction

The role of managers’cognitive capability and emotions in conjunction with corporate

outcomes and market reaction has garnered signif‌icant attention in the current body of

accounting and f‌inance literature. The upper echelons theory (Hambrick and Mason, 1984)

Conf‌lict of interest: The authors declared no potential conf‌licts of interest with respect to the research,

authorship and/or publication of this study.

IJAIM

32,5

858

Received18 August 2023

Revised6 January 2024

Accepted22 June 2024

InternationalJournal of

Accounting& Information

Management

Vol.32 No. 5, 2024

pp. 858-882

© Emerald Publishing Limited

1834-7649

DOI 10.1108/IJAIM-08-2023-0206

The current issue and full text archive of this journal is available on Emerald Insight at:

https://www.emerald.com/insight/1834-7649.htm

asserts that the decision-making practices of top executives are profoundly inf‌luenced by

their personal traits, cognitiveabilities and social aptitude. Kahneman’s (2003) foundational

behavioral economics theoryproposes that managers are susceptible to cognitive biases like

investors. Thus, the sentiments and emotions of individuals have implications for the

information quality of disclosureand the allocation of corporate resources (Jiang et al., 2019;

Loughran and McDonald, 2011). As a result, a growing body of empirical literature

quantif‌ies managers’tone or sentiment by using textual analysis techniques on mandated

corporate disclosures (management discussions and analysis [MD&A] reports and earning

calls) and predicts f‌irm earnings (Davis and Tama-Sweet, 2012;Li, 2010), corporate

investment (Berns et al., 2022;Durnev and Mangen, 2020), stock returns (Jiang et al.,2019)

and stock liquidity (Dang et al., 2022)and inf‌luences market reaction (Xiao Wu et al.,2021).

Taken together, extant literature addresses the issues of whether managerial tone matters

for f‌irm performance (FP) and stakeholder engagement. However, four imperative issues

warrant further empirical evidence on the association between manager sentiment and

corporate f‌inancialperformance.

Prior studies examining the implication of managerial tone on FP did not control the

implication of economic policy uncertainty (EPU). The existing literature suggests that high

policy uncertaint y affects f‌irms’business prospects, hinders corporate investment by

inducing precautionary delays, lowers diversif‌ication benef‌its and, therefore, might create

substantial risk for f‌irms (Baker et al.,2016;Gulen and Ion, 2015). In a similar vein, PÁ Stor

and Veronesi’s(2012)model argued that EPU could exogenously affect the corporate

ecosystem,contributing to investoruncertainty about f‌irm value and information asymmetry

in the f‌inancial market. This leads to the promotion of private information acquisition by

investors, potentially exacerbating information asymmetry (Verrecchia, 2001). Thus,

managers voluntarily disclose information to enhance investors’understanding of the f‌irm,

potentially substituting for private information acquisition (Baker et al.,2016). However, the

effectivenessof these mechanisms reliesupon factors like managers’self-interest, cost-benef‌it

analysis of disclosure and perceptions of the transitory nature of EPU (Nagar et al.,2019).

Hence, it isplausible to assume that policyuncertainty may impactboth manager’sbehavior,

disclosure practices and investors’uncertainty. Therefore, it is imperative to investigate the

impact ofEPU on the relationship betweenmanagement tone and FP.

Second, the rising global awareness of climate-risk hedging investment, equitable

treatment of stakeholders and growing demands for transparency in nonf‌inancial reporting,

including environmental, social and governance (ESG) aspects, indicate a s hifting landscape

for companies. For instance, corporates focus on sustainable and transparent business

strategies to attract more responsible investors and generate long-term socio-economic value

by reducing agency cost, information asymmetry, cost of capital and idiosyncratic risk

(Alareeni and Hamdan, 2020;Albitar et al., 2020;Chen and Xie, 2022;Feng et al., 2022;Wong

et al., 2021). Recently, Al-Shammari et al. (2022) introduced the dual responsibility theory,

which posits that to generate superior f‌inancial performance, corporate managers must

effectively manage both their economic and social obligations to fulf‌ill the demands of

shareholders and stakeholders, respectively. Importantly, Barney (2018) incorporates

Freeman’s stakeholder perspective with a resource-based view and emphasizes the need for

managers to acknowledge and balance the legitimate claims of stakeholders beyond

shareholders to attract prerequisite resources for economic prof‌it. According to signaling

theory, ESG reporting conveys manage rs’commitment to sustainability, transparency and

risk management, boosting the f‌irm’s reputation and attracting more investors (Chen and Xie,

2022;Khalid et al., 2022;Li and Martin, 2019). On the contrary, some other studies documented

the negative effect of ESG disclosure on FP (Branco and Rodrigues, 2008;Gillan et al., 2021;

Large language

model in

corporate

landscape

859