Minimum wages and non-listed firm’s earnings management: Evidence from a geographic discontinuity design☆
Introduction
This study evaluates the causal effect of minimum wages on the information production of non-listed firms. Although minimum wage is a crucial policy affecting labor costs, related studies mostly focus on whether such a provision affects employment, household income, inequality, and poverty. Surprisingly, little is known about how and to what extent the minimum wage policy shapes the information production of firms. Given that wage-related costs typically represent more than 70% of economy-wide value added, an evaluation of the causal effects of minimum wages on firm-level decisions should be of particular interest to economists and regulators who are concerned with labor policies and information efficiency.1
In this study, we focus on one of the most tangible signs of distorted information (Massa et al., 2015), that is, earnings management. Earnings management can be used to “mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers.”2
We expect that minimum wage hikes could affect firm’s costs and capital allocation and therefore, their earnings and profits. This assumption implies that a firm may exhibit incentives to manage its earnings due to outside pressure (e.g., raising external capital or avoiding debt covenant violations). The literature well supports the idea that a minimum wage policy substantially increases the labor costs of firms (Grossman, 1983; Gustafson and Kotter, 2021), and the minimum wage policy in China has been considered a major force driving increases in wages and putting labor cost pressure on firms (Geng et al., in press).
However, the impact direction of minimum wage on firm’s earnings management is ambiguous. On the one hand, minimum wage hikes may provide the incentive for a firm to manage earnings upwardly because the increasing labor cost could crowd out capital investment. Anecdotal evidence from media reports argues that increases in minimum wages squeeze small firms.3 Empirical evidence (Dechow et al., 2011, Linck et al., 2013, Dou et al., 2016) shows that the earnings transfer is useful information to banks or other market participants, and firms which want to borrow from external channels upwardly manage earnings to signal a positive outlook. Linck et al. (2013) find that financially constrained firms with valuable projects can use discretionary accruals to credibly signal positive prospects and in turn help the firm obtain bank loans. Thus, to raise external capital to offset the rising costs, firms exhibit incentives to manage earnings upward and deliver positive information after an increase in minimum wage.
On the other hand, minimum wage hikes may lead a firm to manage earnings downward. Labor investment is more flexible and reversible than capital investment (Hall, 2016, Beladi et al., 2020). Therefore, firms affected by wage hikes might cut labor-related expenditures or do not proportionally increase all employees’ wages in response to increasing labor costs. Accordingly, the firm needs a defense to cost-cutting, and managers have incentives to strategically withhold good news and promote bad news to preserve their bargaining power against labor (Bova, 2013, Chung et al., 2016).
Given the lack of a unique prediction about the effect of minimum wage on earnings management, the issue is essentially an empirical one. While previous literature treats minimum wage hikes as plausible exogenous shocks (Aaronson et al., 2012, Aaronson, 2001), we still attempt to construct identification design to address the potential confounding factors in minimum wage endogenously with regional and market conditions.4
To overcome this difficulty, we introduce an improved geographic information system (GIS) technique and use policy discontinuities at county borders to identify the variation in minimum wages within each of these county pairs located on opposite sides of a county border (Dube et al., 2010). Given that unobserved heterogeneity in economic activity is likely to be reduced when we consider a contiguous county-pair or a metropolitan area, we come closer to “quasi-experimental” conditions than extant studies using counties’ variation. We thus consider that firms located on opposite sides of county borders may feature variation in minimum wage and constitute counterfactual pairs. Firms in approximate geographical areas demonstrate a sufficiently high level of similarity in labor supply, geographic resource endowment (e.g., geographic and geomorphic conditions), public infrastructure (e.g., airport and railway), market condition, local culture, and other unobservable factors. Therefore, we can observe and examine variations within each of the contiguous county pairs with a minimum wage gradient. We enhance the cross-sectional variation by relying on the large geographical and inter-temporal variations of minimum wage policies in China, where the minimum wage varies across more than 2,800 counties. During our sample period (2004–2013), China implemented more than 19,406 local minimum wage changes, more than 53% of which were greater than 10%.
China, as the study setting, provides an ideal laboratory to examine the effects of minimum wage on a firm’s earnings management. First, China promulgated the minimum wage regulations on January 20, 2004, which expanded the scope of application of the minimum wage in more than 2850 counties in 31 provinces or autonomous regions. Moreover, the governments require adjustment in minimum wage frequency. Given the exogeneity, large intertemporal and geographic variations are present in China’s minimum wage policy, providing an ideal quasi-experimental scenario to address the endogenous issue.
Second, our sample is based on a large dataset of the China Industrial Enterprises Database (CIED), which is a census dataset that comprises the entire universe of Chinese manufacturing firms. Data from CIED allow sufficient observations and geographical information to conduct our identification strategy. Private firms are predominant in nearly all economies, and more than 99% of limited liability companies are not publicly tradable.5 Bank financing is usually a major source of finance in private firms (Zou and Adams, 2008), which accordingly induces the earnings management of private firms. By using the unique dataset, we overcome the obstacle of obtaining a large sample of financial data at the firm level.
To conduct the empirical study, following Dube et al., 2010, Geng et al., in press, we introduce GIS techniques to pinpoint the precise location of each firm according to their longitude and latitude. The GIS technique allows us to exploit the discontinuities of minimum wage laws at the same county borders. Firms in approximate geographical areas demonstrate a sufficiently high level of similarity in labor supply, geographic resource endowment (e.g., geographic and geomorphic conditions), public infrastructure (e.g., airport and railway), market condition, local culture, and other unobservable factors.
We construct contiguous county pairs that contain firms located within a short distance (i.e., 25, 50, and 75 km) of the shared border. Our baseline results accounts for the 14.84% increase in earnings management relative to its mean. To check the robustness of our results, we introduce three alternative measures of earnings management, and highly consistent results are found. Our finding is consistent with the hypotheses that minimum wages causally increase the incentives of firms to distort the earnings information for obtaining advantages in external financing. Then, we conduct a number of cross-sectional tests on the bases of the level of average wage, labor capital and IT technology. We find that the incentive of earnings management is stronger in low levels of average wage, labor capital and IT technology firms when they react to minimum wage hikes.
Our paper makes two contributions to the literature. First, we present the causal effects of labor cost on the distortion of firm information, deepening our understanding of the determinants of a firm’s information production and resource misallocation at the firm level. Goldstein and Yang (2019) point out that, following the recent global financial crisis, one main conclusion in the literature is that firms should be required to provide better disclosure and more precise information. Given that non-listed firms constitute the majority of the modern economy and the absence of stock price in aggregating dispersed information, studying information production behaviors of non-listed firms is relevant. Second, we contribute to the emerging literature of labor and finance by evaluating the economic consequences of a prevalent labor policy worldwide (i.e., minimum wage policy). Prior studies mostly investigate whether such provisions affect employment, income, inequality, and poverty. Little is known about how and to what extent minimum wages shapes firm’s earnings information production.
In addition, we also believe that this topic offers clear policy implications to regulators because labor cost, the most important production input, typically represents roughly two-thirds of economy-wide value added (Gan et al., 2016) and information production at the firm level eventually affects resource allocation of the entire economy.
Section snippets
Minimum wage policy in China
In 1994, China started to implement a minimum wage policy in various cities across the country (Gan et al., 2016, Lin and Yun, 2016). On January 20, 2004, the Provisions on Minimum Wages of the Ministry of Labor and Social Security of the People’ s Republic of China (MLSS), that is, the Order No. 21 of MLSS, was promulgated to expand the applicable scope of the (monthly) minimum wage to the entire country, including 31 provinces, autonomous regions, and more than 2850 counties. Article 2 of the
Identification strategy
During our sample period (2004–2013), minimum wage policies in China varied across approximately 2850 counties and more than 300 prefecture-level cities. However, owing to the severe empirical challenges in identifying causality, the literature on how minimum wage affects the earnings information production of firms remains sparse.
Given that the determinants of minimum wage might not be orthogonal to economic fundamentals or firm characteristics, identifying the causal effects involves an
Preliminary tests
In this subsection, two preliminary tests are conducted to examine the effects of minimum wage policy on the labor costs of firms. To assess how minimum wage affects labor costs and employment behavior of manufacturing firms, we use the following model:where i, c, p, s, and t denote firm, county, contiguous county pair, province, and time, respectively. The outcome of interest, denoted yit, is either the firm’s total wage
Conclusions
This study investigates the effect of labor cost on firms’ earnings management at the firm level. Our identification strategy is based on contiguous county pairs and policy discontinuities at county borders and considers that firms located on opposite sides of county borders may feature variations in minimum wage and constitute counterfactual pairs. We show that minimum wage hikes significantly increase firms’ earnings management. Firms adjust earnings management upward for easy access to
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (59)
- et al.
The spending and debt response to minimum wage hikes
Am. Econ. Rev.
(2012) Price pass-through and the minimum wage
Rev. Econ. Stat.
(2001)- et al.
A model of growth through creative destruction
Econometrica
(1992) - et al.
China’s financial system: growth and risk
Found. Trends Finance
(2014) - et al.
Earnings quality in U. K. private firms
J. Account. Econ.
(2005) The timing of asset sales and earnings manipulation
Account. Rev.
(1993)- et al.
Unemployment governance, labour cost and earnings management: evidence from China
World Econ
(2020) - et al.
The financial reporting environment: Review of the recent literature
J. Account. Econ.
(2010) Labor unions and management’s incentive to signal a negative outlook
Contemp. Account. Res.
(2013)- et al.
WTO accession and performance of Chinese manufacturing firms
Am. Econ. Rev.
(2017)
Creative accounting or creative destruction? Firm level productivity growth in Chinese manufacturing
J. Dev. Econ.
The importance of reporting incentives: Earnings management in European private and Public firms
Account. Rev.
Firing the wrong workers: financing constraints and labor misallocation
J. Financ. Econ.
Employee quality and financial reporting outcomes
J. Account. Econ.
Time-series minimum-wage studies: a meta-analysis
Am. Econ. Rev.
Self-selection of auditors and audit pricing in private firms
Account. Rev.
Chinese culture, materialism and corporate supply of trade credit
China Finance Rev. Int.
Earnings management and capital resource allocation: evidence from China’ s accounting-based regulation of rights issues
Account. Rev.
The effect of rural-urban migration on corporate innovation: evidence from a natural experiment in China
Financ. Manage.
Do managers withhold good news from labor unions?
Manage. Sci.
Real and accrual-based earnings management in the pre-and post-sarbanes-oxley periods
Account. Rev.
Detecting earnings management
The Accounting Review
Predicting material accounting misstatement
Contemp. Account. Res.
Labor unemployment insurance and earnings management
J. Account. Econ.
Minimum wages and firm profitability
Am. Econ. J. Appl. Econ.
minimum wage effects across state borders: Estimates using contiguous counties
Rev. Econ. Stat.
Earnings management, stock issues, and shareholder lawsuits
J. Financ. Econ.
The higher costs of doing business in china: minimum wages and firms' export behavior
J. Int. Econ.
Cited by (10)
Minimum wage and corporate investment efficiency: Evidence from China
2024, Finance Research LettersStock market liberalization and earnings management: Evidence from the China–Hong Kong Stock Connects
2023, Finance Research LettersGovernments' fiscal stress and firm decentralization
2023, International Review of Financial AnalysisManagerial ability and earnings management: The moderating role of career concerns<sup>✰</sup>
2023, Finance Research LettersHow does minimum wage affect firm pollution discharges: Evidence from China
2023, Journal of Cleaner ProductionLabor cost and corporate tax avoidance
2023, Journal of Economic Behavior and OrganizationCitation Excerpt :We also drop all cross-province county pairs that straddle two provinces because other source factors in different provinces might produce endogeneity (Note that this conduction only affect a small number of observations and our results are robust to keeping all bordering pairs, even if they straddle provincial lines). We construct contiguous county pairs that contain firms located within a short distance (i.e., 25, 50, and 75 km) of the shared border, which is consistent with the literature Xiang et al., 2022. We further keep each of the two counties in a county pair that contains at least five firms in each year.
- ☆
We appreciate the insightful comments and suggestions of Marco Trombetta (Editor-in-Chief), the specialized editor, two anonymous referees, Donghua Chen, Guohua Jiang, Qinglu Jin, Dan Li, Zengquan Li, Shasha Liu, Danglun Luo, Wei Luo, Xi Wu, Lijun Xia, Deren Xie, Qingquan Xin, Yu Xin, Nianhang Xu, Kangtao Ye, Hongqi Yuan, Guojian Zheng, Jigao Zhu, Kai Zhu, Jian Zhang, and seminar participants at Peking University, Shanghai Jiaotong University, Shanghai University of Finance and Economics, Wuhan University, and Zhongnan University of Economics and Law. We appreciate Yun Dai (Sun Yat-Sen University) for sharing minimum wage data at the county level. We acknowledge the financial support from the Major Project of National Social Science Foundation of China (Grants: 21ZDA010) and National Natural Science Foundation of China (Grants: 72102248; 71991473; 71772178). All authors contribute equally to this paper. All errors and omissions are our own.