Published Research Articles

CEOs’ capital gains tax liabilities and accounting conservatism

Journal of Business Finance & Accounting, 2024

Abstract: Recent studies show that the tax-induced lock-in effect discourages CEOs from unwinding their unrestricted equity and subsequently exacerbates their risk-aversion. I investigate how CEOs’ unrealized capital gains tax liabilities (tax burdens) influence financial reporting conservatism. I find that the demand for accounting conservatism decreases with CEO tax burdens. Further analyses show that the negative relation between CEO tax burden and conservatism is stronger when the firm has high leverage and high default risk and when the CEO's incentives are more aligned with equity holders. This highlights the shareholder–creditor agency conflicts mitigation role of CEO tax burdens in reducing creditors’ demand for conservatism. I exploit the Federal Taxpayer Reform Act of 1997 and staggered state-level tax cuts that significantly decreased personal capital gains tax rates as identification strategies.

CEO Past Distress Experience and Risk-Taking: Evidence from Property–Liability Insurance Firms

with Sangyong Han,  Chia-Ling Ho, and Gene Lai, Risk Management and Insurance Review, 2024

Abstract: We investigate the relationship between CEOs’ past distress experience and risk-taking in U.S. property-liability insurance companies. Our evidence shows that CEOs’ past distress experience is negatively associated with insurers’ risk-taking behavior, suggesting that CEOs with distress experiences tend to take lower levels of risk in making financial decisions for their firms. The results are robust to using alternative measures of risk-taking, including value at risk, expected shortfall, volatility of stock return, idiosyncratic volatility, systematic volatility, underwriting risk, and investment risk. Additionally, our results pass a placebo test, and we mitigate endogeneity concerns with the propensity score matching method.

Industry tournament incentives and corporate hedging policies

with Ahmet Nart and Ahmet Tuncez, Financial Management, 2022

Abstract: This paper examines how a tournament among CEOs to progress within the CEO labor market influences their corporate hedging policies. We employ a textual analysis of 10-Ks to generate corporate hedging proxies, finding that the likelihood and intensity of hedging grow as the CEO labor market tournament prizes increase. We also explore the mitigating impact of corporate hedging on the adverse effects of risk-inducing industry tournament incentives (ITIs) on the cost of debt and stock price crash risk, noting that these could be possible reasons behind the relation. Additionally, we observe that the relationship between ITIs and corporate hedging is less pronounced for firms that demonstrate more financial distress and for firms whose CEOs are the founders of the company or are of retirement age. We identify a causal relation between ITIs and corporate hedging using an instrumental variable approach and an exogenous shock sourced from changes in the enforceability of noncompetition agreements across states.

Industry tournament incentives and corporate innovation strategies

with Lingfei Kong and Ahmet Nart, Journal of Financial Research, 2022

Abstract: This paper examines how the tournament-like progression in the CEO labor market influences corporate innovation strategies. By exploiting a text-based proxy for product innovation based on product descriptions from 10-Ks, we find a significant positive relation between industry tournament incentives (ITIs) and product innovation. We then explore the trade-off effects of ITIs on product innovation created through long-term patenting technologies and short-term “routine” product development. We discover that ITIs strengthen routine product development activities but decrease patent-based innovation. Further analyses show that the effect of ITIs on product innovation is stronger when the product market is more competitive and when CEO characteristics indicate a higher probability of winning the tournament prize.

edgar: An R Package for the U.S. SEC EDGAR Retrieval and Parsing of Corporate Filings

with Bharat Patil and Nilesh Raut, SoftwareX, 2021

Abstract: This paper introduce edgar package in R statistical language which facilitates a robust framework for bulk download and parsing of EDGAR filings in a single query. It provides the following functionalities: downloading filings, searching for filing information and keyword list, extracting filing header information, extracting business descriptions and MD&A section, retrieving Form 8-K events, generating sentiment measures of filings.