Date of Award
Spring 2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Management
First Advisor
Barberis, Nicholas
Abstract
This dissertation consists of three essays in financial economics. The first essay is titled “Financial News Production.” I establish that financial news production can be strongly influenced by factors unrelated to the arrival of, and demand for, information. Fluctuations in real economic activity, such as advertising, generate cash-flow shocks to the media sector, which reacts by changing news quantity and quality. Such endogenous dynamics in news production then shift the levels of uncertainty and information asymmetry about firms, affecting real and financial outcomes. Implementing a within-firm estimator on a comprehensive data set of media advertising revenue, news, and job postings, I compare news production about the same firm by different news media whose advertising revenues are differentially exposed to industry-level advertisement shocks. Financial news production is procyclical at the aggregate level and serves as a channel for economic shock transmission and amplification.The second essay, joint with Song Ma, is titled “Persuading Investors: A Video-Based Study.” Persuasive communication functions through not only content but also delivery, i.e., facial expression, tone of voice, and diction. This paper examines the persuasiveness of delivery in startup pitches. Using machine learning algorithms to process full pitch videos, we quantify persuasion in visual, vocal, and verbal dimensions. We find that positive (i.e., passionate, warm) pitches increase funding probability. Yet conditional on funding, startups with higher levels of pitch positivity underperform. Women are more heavily judged on delivery when evaluated in single-gender teams, but they are neglected when co-pitching in mixed-gender teams. Using an experiment, we show that persuasion delivery works mainly through leading investors to form inaccurate beliefs. The third essay is titled “Self-Selected or Designated: Which SIC Code Is True?” I investigate the difference between firms' self-selected industry codes on the public information platform (i.e., EDGAR) and industry codes designated in the financial database (i.e., Compustat). The discrepancy is pervasive—36% of the codes disagree at the four-digit level, and 17% disagree at the two-digit level. Firms are more similar to peers in their designated industries in terms of a higher overlap in products and a higher correlation in production processes. Firms tend to self-select into industries with higher valuations. After self-selecting new industry codes, firms yield positive abnormal returns (4.5% on average) and double the return comovement with their self-selected industry peers. Meanwhile, firms are more likely to restate their financial reports and invest more in the future.
Recommended Citation
Hu, An, "Essays in Financial Economics" (2024). Yale Graduate School of Arts and Sciences Dissertations. 1440.
https://elischolar.library.yale.edu/gsas_dissertations/1440
