Bad News Withholding and Stock Price Crash Risk of Banks

• Author: Taejin Jung,Young Jun Kim,Hyun Jong Na,Natalie Kyung Won Kim
• Published date: 01 December 2019
• Date: 01 December 2019
• DOI: http://doi.org/10.1111/ajfs.12279

Bad News Withholding and Stock Price

Crash Risk of Banks*

Taejin Jung

College of Business Administration, Seoul National University, Republic of Korea

Natalie Kyung Won Kim

College of Business Administration, Seoul National University, Republic of Korea

Young Jun Kim

College of Business, Hankuk University of Foreign Studies, Republic of Korea

Hyun Jong Na**

School of Business, The George Washington University, United States

Received 5 November 2018; Accepted 6 September 2019

Abstract

Using US banks’ quarterly data from 1995 to 2014, this study examines the mechanism by

which delayed expected loss recognition (DELR) affects the stock price crash risk of banks.

We first show that greater DELR is positively associated with a subsequent crash in stock

price. We then find that this association is only present when bank managers have more dis-

cretion in concealing bad news, which is proxied by the high proportion of heterogeneous

loans. These findings provide policy implications for bank regulators regarding the impor-

tance of specific loan types and time horizons when monitoring the accounting treatment of

banks.

Keywords Bad news withholding; Banks; Delayed expected loss recognition; Loan loss provi-

sion; Stock price crash risk

JEL Classification: G12, G21, M41

*This work was supported by the Hankuk University of Foreign Studies Research Fund of

2019. We thank Xuan Tian (editor) and two anonymous reviewers for their comments and

suggestions. We are also grateful to the workshop participants at Seoul National University,

the 2018 International Finance and Accounting Conference, and the 2018 Conference on

Asia-Pacific Financial Markets for their helpful comments and suggestions.

**Corresponding author: The George Washington University School of Business, Duques

Hall, 2201 G Street NW, Washington, D.C., 20052, United States. Tel: +1-202-710-4750,

Fax: +1-202-994-5164, email: hna@gwu.edu.

Asia-Pacific Journal of Financial Studies (2019) 48, 777–807 doi:10.1111/ajfs.12279

©2019 Korean Securities Association 777

1. Introduction

Large declines in bank stock prices (i.e., crash risk) greatly affect investors, regula-

tors, and the financial market. Banks represent over 20% of the total public equity

market capitalization in the United States (Fields et al., 2004), and troubled banks

are known to aggravate systematic crises in the financial market. Prior research

shows that the banks’ accounting decisions exhibit real implications on their down-

side risks. For example, banks’ crash risk (or similar extreme tail risks) is asso ciated

with their credit cycles, earnings management (Cohen et al., 2014), opacity (Bush-

man and Williams, 2015), and conservatism in accounting treatments (Andreou

et al., 2017).

However, empirical evidence on the exact mechanism by which banks’ account-

ing decisions affect their downside risk is, at best, weak and mixed. There are two

possible mechanisms by which banks’ loan loss accounting treatments can result in

tail risk. On the one hand, delayed expected loan loss recognition (DELR) of banks

could occur because managers withhold bad news from investors (Jin and Myers,

2006; Bushman and Williams, 2015; Zhu, 2016). Managers at banks with high

DELR can hoard bad news until a tipping point where the accumulated bad news is

released simultaneously, thereby causing a sudden stock price crash. On the other

hand, DELR may represent severe performance deterioration and may be associated

with a high default risk. Given that firms with a high default risk are more likely to

fail than those with a low default risk, banks with high DELR are more likely to

experience a stock price crash or a positive price jump

1

than those with low DELR.

The two mechanisms significantly differ from each other given that DELR does not

automatically trigger a stock price crash in the latter mechanism. Although high

(low) default risk is commonly associated with high (low) DELR, high default risk

confounds the link between managers’ hoarding of bad news and banks’ stock price

crash. High default risk may simultaneously cause an increase in DELR and an

increase in the likelihood of a sudden stock price crash. Therefore, we perform vari-

ous falsification tests to verify the relationship between DELR and stock price crash

risk.

This study investigates the bad news withholding mechanism by using loan type

categorization to exploit variations in the ability of bank managers to conceal bad

news. We follow prior research (Liu and Ryan, 1995; Ryan and Keeley, 2013) and

categorize loans into homogeneous loans vis-

a-vis heterogeneous loans based on the

Federal Reserve Form Y-9C filings data. Ryan and Keeley (2013) argue that homo-

geneous loans (e.g., consumer loans and residential mortgages) are small, similar,

and rarely renegotiated. In addition, their loss reserves are primarily determined

through a statistical analysis of historical loss rates. By contrast, heterogeneous loans

1

A negative price jump would indicate a stock price crash, but the term “stock price crash” is

more common than “negative price jump” in prior literature. Therefore, we follow the tradi-

tion and use the term “stock price crash” in contrast to a “positive price jump.”

T. Jung et al.

778 ©2019 Korean Securities Association