Employee firing costs and accounting conservatism: Evidence from wrongful discharge laws

• Published date: 01 October 2023
• Author: Huasheng Gao,Jun Huang,Tianshu Zhang
• Date: 01 October 2023
• DOI: http://doi.org/10.1111/jbfa.12672

Received: 17 June 2021 Revised: 10 November2022 Accepted: 14 November 2022

DOI: 10.1111/jbfa.12672

ARTICLE

Employee firing costs and accounting

conservatism: Evidence from wrongful discharge

laws

Huasheng Gao1Jun Huang2Tianshu Zhang3

1Fanhai International School of Finance, Fudan

University, Shanghai, China

2Institute of Accounting and Finance, Shanghai

University of Finance and Economics,

Shanghai, China

3School of Accounting, Shanghai University of

International Business and Economics,

Shanghai, China

Correspondence

Huasheng Gao, Fanhai International School of

Finance, FudanUniversity, No. 220 Handan

Road, Shanghai 200433, China.

Email: huashenggao@fudan.edu.cn

Fundinginformation

MOE Project of KeyResearch Institutes of

Humanities and Social Science in Universities,

Grant/AwardNumber: 18JJD790010;

National NaturalScience Foundation of China,

Grant/AwardNumbers: 71632006,

71973029, 72072107, 72172081; Program

for Professor of Special Appointment (Eastern

Scholar) at Shanghai Institutions of Higher

Learning,Grant/Award Number: TP2018001;

Dawn Programof the Shanghai Education

Commission, Grant/AwardNumber: 19SG50;

the 111 Project, Grant/AwardNumber:

B18033

Abstract

This paper identifies the causal effect of a firm’s employee

firing costs on its conditional conservatism, using the stag-

gered adoption of US state wrongful discharge laws (WDLs)

that increase a firm’s cost of firing employees. We find

that the adoption of WDLs leads to a significant increase

in conditional conservatism. This result is greater for firms

that are more labor-intensive, have higher propensities to

fire employees, make more firm-specific investments and

have greater risk. Overall, our findings support the view

that higher firing costs lead to greater demand for condi-

tional conservatism to decrease investment inefficiencies

because higher firing costs make inefficient investments

(including overinvestment in negative-net present value

(NPV) projects and delays in disinvesting poorly performing

projects) costlier for the firm.

KEYWORDS

conditional conservatism, firing costs, investment inefficiency,

wrongful discharge laws

JEL CLASSIFICATION

G30, J63, M41

1612 © 2022 John Wiley & Sons Ltd. J Bus Fin Acc. 2023;50:1612–1632.wileyonlinelibrary.com/journal/jbfa

GAO ETAL.1613

1INTRODUCTION

Understanding the determinants of firms’ accounting choices is an important goal in accounting research. Although

labor is one of the key inputs of a firm’s operation, how labor marketfrictions affect firms’ accounting choices is rela-

tively under-examined. In this article, we shed light on this strand of research byinvestigating the effect of employee

firing costs on conditional conservatism.

Our study is based on the staggered adoption of state-level wrongful discharge laws (WDLs), which significantly

increase firms’ employee firing costs. Weexpect the adoption of WDLs in a state to increase conditional conservatism

for firms located in that state for the following reasons. WDL makes inefficient investments (i.e., overinvestment in

ex-ante negative-NPV projects and delays in disinvesting ex-post poorly performing investment projects) costlier for

the firm because when the bad outcomes finally accumulate and losses need to be recognized, these losses will also

include the firing costs, and WDLs will enhance such costs. Therefore, considering that conditional conservatism is

effective in decreasing investment inefficiencies (Ball & Shivakumar,2005; Bushman et al., 2011; Balakrishnan et al.,

2016; García Lara et al., 2016), the demand for conditional conservatism is expected to increase following the state’s

adoption of WDL.

The staggered adoption of WDLs provides a good opportunity to examinethe causal effect of employee firing costs

on conditional conservatism for two reasons. First, US states pass WDLs in order to protect the interests of employees

by encroaching on the employment-at-will doctrines. As the adoption of WDLs is not related to accounting conser-

vatism, it is likely an exogenousshock. Second, since several states adopted WDLs in different years, we can provide

clearer identification than the studies based on a single shock.

Based on a sample of 95,599 firm-year observations from 1969 to 2003, we find that the adoption of WDLs leads

to an increase in firms’ conditional conservatism. The results are consistent with the view that higher firing costs lead

to greater demand for conditional conservatism after the adoption of WDLs because conservative accounting policies

can help reduce firms’ investment inefficiencies.

The essential assumption for our difference-in-differences model is the parallel trend assumption: Treatmentand

controlfirms have similar trends in conditional conservatism before the adoption of WDLs. Our parallel trends analysis

indicates that the two groups indeed have similar trends in conditional conservatism prior to the passage of WDLs.

Further,the influence of WDLs on firms’ conditional conservatism appears after the law adoption, which supports our

causal interpretation.

One possible concern is that the adoption of WDLs is correlated with local business conditions, which influence

firms’ accounting choices. Toalleviate this problem, we focus on a subsample of treatment firms and control firms near

the state borders. Considering that companies in such proximity share similar economic conditions, if our results are

driven by local business conditions (rather than WDLs), we should expect no effect in this subsample. However, we

still find a significantly positive effect of WDLs on conditional conservatism for treated firms on one side of the state

border relative to their neighboring control firms on the other side of the state border.

We further investigate the cross-sectional variation of the treatment effect. We find that the positive effect of

WDLs on conditional conservatism is more pronounced for firms that are more labor-intensive, havea higher propen-

sity to fire workers, makemore firm-specific investments and have higher risk. These results provide further evidence

that the effect of WDLs on conditional conservatism is indeed related to firms’ hiring costs.

Finally,we perform a number of robustness checks on our main findings: using other conservatism measures, con-

trolling for the confounding effect of cost stickiness, implementing alternative difference-in-differences methods,

addressing the changes in accounting standards during our sample period and conducting a placebo test. The positive

effect of WDLs on conditional conservatism remains.

Our paper makes at least three contributions to the literature. First, it contributes to the literature on the rela-

tion between accounting conservatism and investment decisions. Bushman et al. (2011) show that the sensitivity of

investment to declining investment opportunities is increasing in country-level accounting conservatism, supporting