Research

[“Disclosure Firmness of Corporate Language”]

Job Market Paper

Abstract: Using the MD&A section of 10-K filings, I examine the hardness/softness of corporate information as its firmness level by aggregating linguistic attributes consisting of vague, tonal, forward-looking, explanatory, numerical, and specific information, as well as novel metrics of historical and objective content based on natural language processing and machine learning. I find that firms use softer language during poorly performing years, and that this effect is stronger in less ambiguous settings. I also find that proprietary costs stemming from different types of competition differentially influence disclosure firmness. I corroborate my findings using exogenous variation in entry threat resulting from import tariff rate reductions. My results suggest that performance and competition incentives encourage managers to use an arsenal of linguistic attributes in shaping their disclosure strategy. Overall, I take a first attempt at studying the economic determinants of disclosure firmness, a concept germane to the historical debate on relevance versus reliability.

:point_right: Click here to see how I used Machine Learning in my Job Market Paper! :robot:


“Social Unrest as an Impetus for Racial Diversity: Firm Responses to the Murder of George Floyd”

with Karthik Balakrishnan, Rafael Copat, and K. Ramesh
Under revision for resubmission to Journal of Accounting Research
Presented at the 2022 JAR Conference

Abstract: George Floyd’s murder offers a unique setting to examine the valuation effects of racial diversity to shareholders. Using Black participation in boards as an observable proxy for a firm’s diversity issues, we find that firms with no Black directors experienced a share price drop of 1.5% immediately following the murder. We develop a text-based measure of a firm’s exposure to diversity issues from conference call transcripts and find that firms with diversity exposure experience a stock price drop of 0.7% around the date of the conference call. We also find that firms respond to the murder by being much more likely to appoint a Black director. The market response to the appointments of Black directors after the murder is on average positive, but the effect is muted when the additional diversity comes at the cost of increased board size. We find no evidence of a change in the skillset of incoming directors of different ethnic groups following the murder, but we find increased busyness for Black directors.

:point_right: Click here to watch the JAR Conference presentation! :tv:


“Protecting Forward Looking Statements”

with Maclean Gaulin and K. Ramesh

Abstract: We examine the increasingly prevalent managerial disclosure practice of listing specific keywords in SEC filings to identify forward-looking statements for obtaining “safe harbor” protection under the Private Securities Litigation Reform Act. We show that proxies for ex ante litigation risk, network/herding effects, disclosure supply, and economic uncertainty are strongly associated with the decision to include the keyword list and the number of keywords. Responding to transient economic circumstances, firms periodically change the number of keywords to customize their forward-looking disclosures. Using factor analysis we unravel the specific linguistic attributes implied by the keywords and show how they enable firms to tailor their disclosure of quantitative and qualitative forward-looking information. Finally, managers facing higher litigation risk find it imperative to protect disclosures that capital markets view as value relevant. Overall, the decision to include the list and the choice of the keywords it contains are neither boilerplate nor ad hoc. Together, our evidence provides an important first look at the determinants of firms’ decisions regarding a central feature of forward-looking disclosures’ “safe harbor” protection.


Works In Progress

“The Predictive Ability of Corporate Language” with Nick Guest and Mani Sethuraman
“Inferring Managerial Communication Style: A Machine Learning Approach to Extracting Linguistic Traits from Textual and Acoustic Cues” with John Gallemore
“Disclosing Strategy” with K. Ramesh