Two Wrongs Make a ‘Right’? Exploring the Ethical Calculus of Earnings Management Before Large Labor Dismissals

Abstract

This paper examines whether firms strategically legitimize large labor dismissals (LLDs) by performing ex-ante downward earnings management. We further assess whether the effect is larger under stakeholder pressure and whether these practices influence the external perception of firms’ behavior. As laying off employees without an economic reason is perceived as a breach of the social contract, stakeholders pressure firms to provide economic justification for LLDs. We argue that firms strategically legitimize LLDs by artificially worsening their financial performance through downward earnings management. From a sample of European listed firms for the period 1998–2014, using a propensity score matching model to control for differences in observable characteristics between LLD and non-LLD firms, we find that firms manage their earnings downward in the year prior to LLDs. These results are more pronounced in the presence of high stakeholder pressure. As such, firms with (1) stronger trade unions, (2) higher visibility, (3) stricter employment protection legislation, and (4) more stakeholder-oriented legal environments have stronger incentives to manage earnings downward before LLDs. Finally, we document that downward earnings management prior to LLDs is effective at legitimizing LLDs, offsetting the negative external perceptions of firms in the aftermath of LLDs. This study contributes to the business ethics literature by identifying unintended consequences of stakeholder pressure around LLDs. It also brings important empirical insights into how ethically questionable decisions are strategically legitimized through further unethical practices.

Appendices

Appendix 1: Measures of Earnings Management

Our measures of earnings management (EM_NA_abs, EM_DA, and EM_DA_abs) relate to the Jones Model used in prior literature (Vladu et al. 2017). We run the following Jones (1991a) model to decompose total accruals into normal and discretionary components:

$${\text{Tacc}}_{i,t} = \, a_{0} + \, a_{1} \left( {1/{\text{Assets}}_{{i,t{-}1}} } \right) \, + \, a_{1} \Delta {\text{Sales}}_{i,t} /{\text{Assets}}_{{i,t{-}1}} + \, a_{2} {\text{PPE}}_{i,t} /{\text{Assets}}_{{i,t{-}1}} + \, \Omega_{i,t}$$

(4)

where Tacc_i,t is total accruals, defined as the difference between operating income and operating cash flow; Assets_i, t − 1 represents total assets in t − 1; PPE_i,t is the gross value of fixed assets in year t; ΔSales_i,t is growth in sales for firm i between t − 1 and t. Generally, US studies estimate Jones’ model by industry and year. In an international setting, there is a limited number of observations across numerous industries and thus industry-year regressions may produce unreliable estimates. Therefore, following He et al. (2017), we run the regressions in a given year for each country, while controlling for industry fixed-effects.

The residuals from Equation (4) measure the discretionary accruals. In this study, we focus on downward earnings management. Therefore, we examine the level of discretionary accruals (EM_DA) and negative discretionary accruals (EM_NA_abs and EM_DA_abs). We multiply the negative discretionary accruals by − 1. Greater values for EM_NA_abs and for EM_DA_abs indicate a higher level of downward earnings management for the Jones Model. Lower levels of discretionary accruals (EM_DA) can either indicate lower levels of upwards earnings management (Gao et al. 2018) or more downward earnings management (Kanagaretnam et al. 2010).

Appendix 2: Variable Definitions

Variable	Definition		Source
Earnings management variables
EM_NA_abs		Earnings Management, measured as the absolute value of the negative discretionary accruals (see Appendix 1)	EIKON
EM_DA_abs		Earnings Management, measured as the absolute value of the negative discretionary accruals and zero for positive abnormal discretionary accruals (see Appendix 1)	EIKON
EM_DA		Earnings Management, measured as the signed discretionary accruals (see Appendix 1)	EIKON
Layoff variable
Layoff		Layoff, dummy variable equals to 1 if the firm’s number of employees decreases by 10% between the periods t and t + 1, 0 otherwise	EIKON
Firm characteristic variables
Size		Firm Size, measured as the natural logarithm of total assets	EIKON
Leverage		Leverage, measured as total liabilities scaled by lagged total assets	EIKON
Large_loss		Large loss, equal to 1 if the firm’s net income is less than − 1% of assets, 0 otherwise	EIKON
Small_loss		Small loss, equal to 1 if the firm’s net income is between 0 and − 1% of assets, 0 otherwise	EIKON
PPE		Gross value of property, plant and equipment	EIKON
Sales_growth		Firm’s sales, measured as the difference in sales from year t − 1 to year t scaled by lagged total assets	EIKON
NCFO		Net cash flows from operations	EIKON
ROA		Return on Assets, measured as EBITDA scaled by lagged total assets	EIKON
Analysts		Analyst following, measured as the natural logarithm of one plus the number of Analysts following the firm	EIKON
Foreign_assets		Foreign assets, measured as foreign assets scaled by lagged total assets	EIKON
BIG4		Audit quality, equal to 1 if the auditor is one of the Big 4 audit firms, 0 otherwise	EIKON
Employment_strictness		Strictness of employment protection legislation on collective dismissals, equal to 1 if any of the four available items related to collective dismissals—“Definition of collective dismissal”, “Additional notification requirements in case of collective dismissals”, “Additional delays involved in case of collective dismissals”, and “Other special costs to employers in case of collective dismissals”—is above the sample median, 0 otherwise	OECD
Trade_unions		Trade unions strength, equal to 1 if union strength is higher than the sample median, 0 otherwise. Union strength is measured at the firm level by multiplying the union density rate (i.e., net union membership as a proportion of wage earners in employment) with the number of employees scaled by total assets	ICTWSS
Visible		Media coverage, equal to 1 if a firm’s number of press articles is higher than the median number of press articles for firms in the same decile of Size, the same country, and the same year, 0 otherwise	RavenPack
Legal_environment		Legal environment, measured as the country disclosure requirements index	LaPorta et al. (2006)
Media_sentiment		Media sentiment, equal to 1 if the media sentiment of news articles about the firm is higher than the median score of media sentiment for firms in the same decile of Size, the same country, and the same year, 0 otherwise	RavenPack
MV		Market capitalization measured as the share price multiplied by the number of shares outstanding	EIKON
CSR_report		CSR disclosure, equal to 1 if a firm issues a CSR report, 0 otherwise	EIKON
Country characteristic variables
GDP_growth		Gross Domestic Product growth of the firm’s country m from year t − 1 to year t	OECD
Corporate_tax		Corporate tax rate at the country level	OECD
Unemployment_rate		Level of unemployment at the country level in a year	OECD

Appendix 3: Legal Notice Period to Employees

This table summarizes the regulatory requirements for all sampled countries on the time lag (in days) between the announcement and the implementation of Layoffs.

Country	Notice Period to Employees (days)	Source
Austria	30	Eversheds
Belgium	60	Eversheds
Denmark	30	Eversheds
Finland	14–180	European Commission
France	30–60	Eversheds
Germany	30	Eversheds
Greece	5–160 for manual workers; 30–720 for white collar workers	European Commission
Ireland	30	Eversheds
Italy	1–120	Eversheds
Netherlands	90	Eversheds
Portugal	90	International Organization of Employers
Spain	60–150	Eversheds
Sweden	no prescribed period	Eversheds
United Kingdom	90	Eversheds