Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Excess employment, media coverage and corporate valuation: Labour Contract Law in China

Excess employment, media coverage and corporate valuation: Labour Contract Law in China China Journal of aCC ounting StudieS , 2016 Vol . 4, no . 1, 34–52 http://dx.doi.org/10.1080/21697213.2016.1144972 Excess employment, media coverage and corporate valuation: Labour Contract Law in China* a b Deming Yang and Can Zhao a b Management School, Jinan university, guangzhou 510632, China; Management School, o cean university of China, Qingdao 266100, China KEYWORDS ABSTRACT excess employment; firm Previous research has found that firms’ excess employment will reduce value; labour contract law; corporate valuation as well as lead to a series of negative economic media coverage consequences. However, these studies cannot explain why the excess employment would lead to a series of negative economic consequences. Using the exogenous event of the promulgation of the new Labour Contract Law (‘the Act’) enacted in 2007 in China, this paper finds that an increase in excess employment will reduce the valuation of firms. With the promulgation and implementation of the Labour Contract Law, the negative correlation between excess employment and corporate valuation is significantly strengthened, which shows that excess employment has a more serious negative economic impact after the law is implemented. It has been shown elsewhere that labour costs have increased since the implementation of the Act. This paper reveals that the mechanism relating excess employment to negative economic consequences is that excess employment leads to increasing labour costs rather than reducing enterprise productivity. Furthermore, we present evidence that media coverage significantly reduces the negative correlation between excess employment and the valuation of the firm, which extends the information intermediary theory of media. 1. Introduction Low-cost labour is regarded as an important source of economic growth in China (Cai, 2010). However, there are negative effects owing to low-cost labour. For example, low-cost labour may misappropriate labour rights, and it may also increase the gap between the rich and the poor (Cai, 2010). To overcome these negative effects and improve the labour contract sys- tem, the Labour Contract Law (‘the Act’) was revised and issued in 2007. The promulgation and implementation of the Act have had a profound impact on the Chinese economy. We expect that the Act will have had a major impact on enterprise employment in micro-level enterprises. The theoretical and empirical evidence shows the Act would greatly increase CONTACT d eming Yang yangdeming2001@sina.com Paper accepted by d onghua Chen. t he a ct was passed on 29 June 2007 by the g eneral assembly of the t enth national People’s Congress Standing Committee of the 28th Meeting and was carried out on 1 January 2008. © 2016 a ccounting Society of China CHInA JoUrnAL oF ACCoUnTIng STUDIeS 35 labour costs (Ding, 2010; Liu & Liu, 2014). This increase in labour costs would affect enterprise employment after the Act. The promulgation and implementation of the new version of the Labour Contract Law also provide a good opportunity for researching excess employment, as experienced in China, also referred to as the over-staffing problem. The issues of excess employment may be the result of either firms reluctantly hiring employees or firms being unable to dismiss employees (Zeng, 2006). For instance, the Chinese government bears the responsibility for increasing employment, so the government might affect firm hiring and firing through a series of administrative actions (Lin, Liu, & Zhang, 2004). In this environment, firms naturally cannot dismiss or hire employees according to necessity. Some firms, especially state-owned firms, have to bear huge responsibilities to help the government address employment, which leads to a number of excess employees (Zeng, 2006; Zeng & Chen, 2006). Previous Chinese literature draws the conclusion that firms’ excess employment brings a series of negative consequences, including reducing corporate valuation, increasing labour costs, and weakening corporate governance (Zeng, 2006; Zeng and Chen, 2006; Zhang, Wang, & Liu, 2013). However, the literature does not address the mechanism that explains why excess employment can trigger these negative economic consequences. This mechanism may either be a reduction in production efficiency or an increase in labour costs, both of which are caused by excess employment. The Chinese Labour Contract Law enacted in 2007 increases firms’ labour costs (Ding, 2010; Liu & Liu, 2014), although there is no evidence that shows the Act can reduce employees’ production efficiency. If the first mechanism is true, the negative correlation between excess employment and corporate valuation should be strengthened with the promulgation and implementation of the Act because there is a negative correlation between labour costs and corporate valuation (Zeng, 2006). on the contrary, if the mechanism of reduced production efficiency is true, the negative correlation between excess employment and corporate valuation should be weakened or remain unchanged as a result of the act’s promulgation and implementation because there is a positive correlation between employees’ production efficiency and corporate valuation (Zeng, 2006). The previous literature concerns the relationship between excess employment and the valuation of firms, but it never addresses the issues from the perspective of media coverage or media monitoring. The issue of excess employment is closely related to pressing social issues; for example, social justice, social stability, and corporate governance (Lin et al., 2004). So, the media are concerned about these hot issues, and because of this, according to the information intermediary theory of media (Bushee, Core, guay, & Wee, 2010), media cover - age can lower information asymmetry, systematically affecting investors, analysts, creditors and other market participants. Based on this theory, according to Liu and McConnell (2013), with decreasing information asymmetry, media coverage can weaken the impact of adverse factors on corporate valuation. If corporate managers take some action that is not benefi- cial to shareholders, such as hiring a great number of excess employees, it should have a negative effect on Tobin Q. However, in a situation of information symmetry, the negative impact should be small (Liu & McConnell, 2013). Because firms’ excess employment brings a series of negative consequences (Zeng, 2006; Zeng & Chen, 2006; Zhang et al., 2013), that is, excess employment is harmful to firms, we expect that media coverage will reduce the relationship between excess employment and the valuation of firms. a ccording to d yck et al. (2008), media coverage is a proxy variable for media monitoring. 36 D. YAng AnD C. ZHAo Using the exogenous event of the proclamation of the Chinese Labour Contract Law in 2007 and its implementation in 2008, this paper studies the relationship between excess employment and the valuation of firms using data for Chinese A-share listed companies. We also study how media coverage affects the relation between excess employment and the valuation of firms. o ur research finds that a firm’s excess employment can significantly lower the firm’s value, and this negative correlation has been significantly strengthened with the implementation of the Labour Contract Law. The Act has raised the cost of excess employment. This implies that the mechanism is increasing labour costs rather than reduc- ing production efficiency. Further evidence shows that media coverage will significantly weaken the relation between excess employees and the valuation of firms, which supports the information intermediary theory of media. The contributions of this paper are as follows. First, some studies (Zeng, 2006; Zeng & Chen, 2006; Zhang et al., 2013) find that excess employment produces a series of negative economic consequences. However, these stud- ies do not systematically analyse the mechanism by which excess employment produces these negative economic consequences. o ur paper tests the mechanism. The result shows that the mechanism is increasing labour costs rather than reducing enterprise productivity. Second, this paper approaches the information intermediary theory of media from a new perspective, which directly contributes new evidence to the literature. Liu and McConnell (2013) find that a value-reducing acquisition has a significantly negative impact on the valuation of the firm; however, media coverage can significantly lower the negative effect on the valuation of the firm. Consistent with their study, our paper presents evidence that media coverage significantly reduces excess employment’s negative effect on the valuation of the firm, which extends the information intermediary theory of media. The remainder of this paper proceeds as follows. Section 2 discusses the institutional background. Section 3 is a review of the relevant literature. Section 4 discusses theoretical assumptions. Section 5 presents the research design. Section 6 presents the results of our empirical research. Section 7 presents robustness tests, and we conclude in Section 8 with a brief discussion. 2. Institutional background Before 1980, under the planned economy system, Chinese enterprises, especially state- owned enterprises, carried out a kind of highly centralised labour employment system (Zeng, 2006; Zeng & Chen, 2006). In this system, the government uniformly assigned jobs and job placement in an administrative way. The basic characteristics of the system were that the government controlled the number of employees and workers’ wages and employers and workers had no choice (Shen, Liang, & Chen, 2013). enterprises could hire employees but had no right to fire workers. To maintain social stability, the government strictly limited the ability of enterprises to fire workers for any economic reasons (Zeng & Chen, 2006). The employment problems of state-owned enterprises at this time were entirely the result of government behaviour (Zeng, 2006). Since 1980, China has carried out a policy of reform and opening up. With the implemen- tation of the policy, the Chinese economy has gradually begun to transform from a planned economy to a market economy. In this process, the Chinese labour-employment system has also adopted gradual reform (Zeng, 2006). In the reform of the labour-employment system, CHInA JoUrnAL oF ACCoUnTIng STUDIeS 37 the state council issued a rule named ‘The provisional regulations on the labour contract system in state-owned enterprises’ in 1986 (Shen et al., 2013; Zeng, 2006). The formulation allowed Soes to dismiss workers, which increased the risk of unemployment. However, Soes were still not able to dismiss workers in full accordance with the economic goals of enter- prises in the 1980s (Zeng, 2006). At that time, there were several reasons why the scale of Soes’ employees was still subject to government intervention. First, the existing political and economic system could not be substantially changed in a short time. If a large number of employees had been laid off, these unemployed employees would have led to social instability. Second, firing workers was also subject to a series of policy constraints. Although enterprises had some right to hire or fire workers, it was very difficult to fire workers for economic reasons (Zeng, 2006). In such a climate, Chinese enterprises, especially state-owned enterprises had to bear many policy burdens, which had a direct impact on the generally poor performance of Soes (Lin et al., 2004). To solve these policy burdens and promote the reform of Soes, the labour department proposed the establishment of the re-employment project in 1993 (Zeng, 2006). As part of the project, the government allowed a certain level of unemployment; thus, Soes began to dismiss a large number of workers. However, the number of laid-off workers was still tightly restricted by the government and administrative system (Zeng, 2006). For example, the labour law that was promulgated in 1994 stipulated that the department that either needed to dismiss workers for poor performance or be faced with bankruptcy could dismiss workers according to legal procedure. Meanwhile, the government, especially local government, gradually gained some free right over Soes dismissing workers because the government had the right to influence Soes’ performance and corporate government by administrative intervention (Zeng, 2006). Due to unemployment affecting social stability, the government, especially local government, pays special attention to the issue and limits firms’ ability to dismiss workers. Thus, government pressure to bolster employment is always an important factors that leads to firms’ excess employment (Zeng & Chen, 2006). For the capital market, Chinese stock exchanges (one in Shanghai and another in Shenzhen) were created as an experiment in mixing a market economy with central plan- ning (Chen & Yuan, 2004). g overnmental intervention has been a constant throughout, which has affected listed companies, especially state-owned listed companies (Chen & Yuan, 2004). Listed Soes can spin off their social burdens, strengthen corporate governance, and improve information transparency. In the capital market, the costs of government interven- tion should be reduced with high information transparency. However, it is an important goal for the Chinese government to control unemployment and promote employment. If the local government faces employment pressure, they let the local Soes share the pressure or restrict the local Soes’ ability to dismiss workers (Xue & Bai, 2008). Thus, pressure from the government to solve employment is one of the most important factors that always influence state-owned listed companies (Xue & Bai, 2008; Zeng & Chen, 2006). Then, we discuss the question of the methods by which the government achieves its motive to affect employment in state-owned listed companies. The government can influ- ence Soes through the performance evaluation system. In China, the State-owned Assets Supervision and Administration Commission (SASAC), as the representative of the gov- ernment, is responsible for maintaining and increasing the value of state-owned assets (Ma, Wang, & Shen 2013). To maintain and increase the value of state-owned assets, SASAC has taken a series of measures that include the appointment of senior executives and the 38 D. YAng AnD C. ZHAo establishment of a performance evaluation system. Because SASAC is responsible for the appointment of senior executives, SASAC can affect state-owned listed companies’ corpo - rate governance through the executives who are appointed by them; thus, the government actually affects the behaviour of state-owned listed companies’ employees and, similarly, can affect behaviour through the performance evaluation system (Ma et al., 2013; Zeng & Chen, 2006). For these reasons, there are a number of excess employed in state-owned listed com- panies because of the government’s administrative intervention (Zeng & Chen, 2006). The government can achieve its objectives through SASAC and other methods. We now discuss and analyse whether there is a certain amount of excess employees in non-state-owned listed companies. Because non-state-owned enterprises have developed in the market environment, these enterprises use production elements according to market forces of supply and demand (Zeng, 2006). Accordingly, there should be less excess employment in these enterprises. However, if non-state-owned enterprises want to issue shares, these enterprises have to face a great deal of administrative intervention (Liu, Sun, & Liu, 2003). In the Chinese capital market, it was important to help Soes solve the problem of poor performance in 1990s because Soes generally faced operating difficulties. In this situation, the companies in the capital market mainly consisted of Soes at the stock market initial stage (Liu et al., 2003). According to the annual statistics of the China Securities r egulatory Commission (CSr C), with the development of mergers and acquisitions, the creation of a small and medium enterprises board (SMe board), and the creation of a growth enterprise market (geM), the number of non-state-owned listed companies has risen continuously. either IPo from the SMe board or geM or going public through mergers and acquisi- tions (M&A), non-state-owned listed  companies have to face strict supervision and strict administrative intervention (Liu et al., 2003). The government can also affect non-state- owned listed companies by taxes, or government subsidies. Compared with state-owned listed companies, intervention in non-state-owned listed companies is weaker because the government has no direct role in the affairs of non-state-owned listed companies. So, there are a number of excess employees in non-state-owned listed companies, but the number in these companies is significantly less than the number in state-owned listed companies (Zeng and Chen, 2006). We consider that the issue of excess employment has changed with time and with impor- tant reforms of the system. For instance, the Labour Contract Law in 2008 might have an ee ff ct (Liu & Liu, 2014 ). In general, the Act enforces a greater degree of employee protection. The employee protections include the following. (1) The Act demands that firms ensure a minimum wage. (2) The Act demands that firms offer rest and vacation, labour safety, health regulations and insurance benefits, which indirectly increase the labour cost of firms. (3) The Act increases the cost of signing, carrying out, modifying and terminating the labour contract, which directly increases the labour cost of firms. (4) The greatest impact on the labour cost of listed companies should be the labour contract that does not have a fixed term. The following is the content of the Act relating to such a contract. Article 14: A non-fixed-term labour contract shall refer to a labour contract for which the employer and the worker have agreed that the date of termination is not fixed. Upon negotiation and consensus between the employer and the worker, a non-fixed-term labour contract may be concluded. Under any of the following circumstances, a worker may propose CHInA JoUrnAL oF ACCoUnTIng STUDIeS 39 or agree to renewal or conclusion of labour contract; except where a worker proposes for the conclusion of a fixed-term labour contract, a non-fixed-term labour contract shall be concluded: (1) the worker has worked for the employer for a period of 10 years consecutively. (2) when the employer first implemented the labour contract system or when a new labour contract is concluded upon restructuring of a state-owned enterprise, the worker has worked for the employer for a period of 10 years consecutively and will attain his/her statutory retirement age in less than 10 years. (3) w here a fixed-term labour contract has been concluded twice consecutively and the worker who does not fall under any of the categories stipulated in Article 39 and item (1) and item (2) of Article 40 renews his/her labour contract. Where an employer fails to conclude a written contract with a worker after one year has lapsed since the date of employment, the employer and the worker shall be deemed to have concluded a non-fixed-term labour contract. Article 14 means that provided an employee has worked for ten years for an employer, the employee has the right to sign a fixed-term labour contract, and the employer has no right to dismiss the employee. The impact on listed companies may be reflected in two respects: first, employers of listed companies have no right to dismiss seasoned employees who have worked for a company for ten years or who are eligible under the fourteenth term of the Act. Second, due to the difficulty of dismissing seasoned employees, listed companies will reduce their intention to hire new employees. In short, the Act has increased the minimum wage standard and labour costs (Ding, 2010; Liu & Liu, 2014), which would affect excess employment in listed companies. 3. Literature review The literature of excess employment consists of two branches. The first studies the factors that affect excess employment. The second addresses the economic consequences of excess employment. Some of the theoretical literature considers excess employment to be closely related to political benefits (e.g., Shleifer & Vishny, 1994). empirical studies support the view of Shleifer and Vishny (1994) in providing evidence on the factors that affect excess employment. For example, Lin et al. (2004) indicate that there are widespread ‘policy burdens’ for Chinese companies in transition economies and that China is a labour-surplus country. To solve the labour surplus issue, the government would require companies, especially state-owned companies, to hire a large number of employees, even though these companies are reluc- tant to hire too many employees; this policy creates social policy burdens on companies. Similarly, Zeng (2006), Zeng and Chen (2006) also believe that listed companies have to bear the burden of helping the government solve the unemployment issue, so there are generally excess employees in Chinese listed companies. Because Soes face greater administrative intervention, the number of excess employees in state-owned companies is significantly higher than that in private listed companies (e.g. Zeng & Chen, 2006). Xue and Bai (2008) provide evidence that, compared with those in low unemployment areas, companies in high unemployment areas have to take on more responsibility to help the government bolster employment. These companies also have to bear more of the local governments’ pressure and administrative intervention. o ther findings include Ma et al. (2013) who discover that Chinese Communist Party organisations of state-owned enterprises (Soes) can participate in corporate governance, 40 D. YAng AnD C. ZHAo which increases excess employment significantly. Their empirical evidence also shows that administrative factors play a very important role in the generation of excess employment. The literature regarding the economic consequences of excess employment offers a consistent conclusion; that is, excess employment brings a series of adverse consequences. For example, excess employment reduces the valuation of firms significantly (Zeng, 2006), as well as increasing labour costs significantly (Zeng & Chen, 2006). Analogously, Xue and Bai (2008) find that excess employment reduces the performance of firms significantly and that firms in high unemployment areas bear more excess employment, as a consequence of which these firms receive more fiscal subsidies than other firms. The adverse consequences also include the fact that excess employ - ment may reduce the sensitivity between executive compensation and corporate performance; in other words, excess employment might weaken corporate governance (Zhang et al., 2013). o verall, excess employment has a negative impact on corporate performance to some extent. o ur research is related to other research fields, specifically with regard to the corporate gov - ernance role of the media. In recent years, more studies have begun to pay close attention to this field of research. Dyck, Volchkova and Zingales (2008) explain why the media have corporate governance functions. The role of the media is to collect, select, certify, and repackage informa- tion (Dyck et al., 2008). By doing so, they increase the number of people who learn about the behaviour of others, thereby increasing the effect of reputation. As top executives, managers care more about their companies’ reputation. To avoid the loss of reputation, managers have the incentive to improve company performance or improve corporate governance. Thus, the media can play an important role in corporate governance by increasing reputation cost. The media can play an important role in corporate governance in another way, such as monitoring (Miller, 2006). Miller (2006) draws the conclusion that the media can affect corporate governance by performing the function of oversight, i.e. the role of ‘watchdog’. Some Chinese studies (e.g. Li & Shen, 2010; Luo, Liu, & Zhang, 2013; Yang, Lu, & Luo, 2013; Yang & Zhao, 2012) determine that the media play an important role in corporate governance by administrative interference. Why should this be the case? Chinese enterprise executives usually do not care about their companies’ reputation but do care about administrative regulation. Thus, the media can usually affect corporate governance by having an influence on administrative supervision instead of executives’ reputation. These studies in practice support Miller (2006). While many studies suggest that the media have an impact on public perception in some manner, one group of studies indicates media coverage lacks in-depth research and tends toward sensationalism. For example, Jensen (1979) argues that the media have become a form of entertainment and that articles are written to appeal to the lowest common denominator, so the media cannot play an important role in the macro and micro economy. Consistent with this, Core, guay, and Larcker (2008) study media coverage of compensa- tion and conclude that the media sensationalises their coverage by focusing on large ex post stock gains rather than compensation expenses to the company. They conclude that media coverage has no impact on compensation behaviour. Using a clinical study approach, DeAngelo, DeAngelo, and gilson (1996) suggest that simplistic media coverage of junk bonds may have skewed the economic behaviour of customers and regulators. As a group, these studies question the validity of the media as an important information intermediary in the economy and even suggest it may play a negative role. Starting with f ama (1980), the finance literature has recognised the importance reputation plays in disciplining corporate managers. t he early literature, f ama (1980) and f ama and Jensen (1983), emphasises that manager reputations would affect potential employers, who determine future jobs and wages. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 41 In contrast, other studies suggest the media form an important component of the infor- mation environment in society. Some studies use cross-sectional variation in the national characteristics of the media to investigate the role of the media in corporate governance and economic development (Dyck et al., 2008). This literature generally concludes that com- petitive media can have a positive impact on the political and economic development of a country. This impact is presumed to be driven by the oversight function of the media. o ther research also supports that the media have a supervision function. For example, Miller (2006) investigates the role of the press as a monitor or ‘watchdog’ for accounting fraud. The research finds that the press fulfils this monitoring role by rebroadcasting information from other information intermediaries (analysts, auditors, and lawsuits) and by undertaking original investigations and analysis. Liu and McConnell (2013) propose that the media are key conduits through which managers’ reputation capital is built and, potentially, destroyed. Their research supports the theory of reputation put forward by Dyck et al. (2008). o verall, although these conclusions are controversial, it has gradually become the consen- sus that the media play an important supervisory function. In China, although the media do not have total freedom, they do play an important supervisory function (e.g. Li & Shen, 2010). From another perspective, there is a lack of literature that discusses how media coverage influences the relation between excess employment and the valuation of firms. 4. Hypothesis development The issues of excess employment may be the result of firms reluctantly hiring a large number of employees or being unable to dismiss employees. The number of excess employees in listed companies indicates the extent to which these employees are not able to produce enough value and their wages or labour costs are too high. In brief, the value creation does not match the wages of these employees, which causes a decline in the value or performance of listed companies (Zeng & Chen, 2006). If the capital market is efficient, it can draw a reasonable valuation for this kind of phenomenon. So, we expect that excess employment should reduce the valuation of firms. In addition, excess employment can also be seen as a proxy variable for government intervention or administrative intervention, which might lead to the weakening of corporate governance as well as declining performance (Xia & Fang, 2005; Zeng, 2006). Based on the above discussion, we state our first hypothesis as follows. H1: There is a negative relationship between excess employment and corporate valuation. Previous literature draws the conclusion that firms’ excess employment creates a series of negative consequences (Zeng, 2006; Zeng & Chen, 2006; Zhang et al., 2013). However, these studies do not explain why excess employment can trigger these negative economic conse- quences. This mechanism may either be reducing production efficiency or increasing labour costs, both of which are due to excess employment. The Labour Contract Law increases labour costs, as has been shown in some studies (Ding, 2010; Liu & Liu, 2014). For example, Article 14 in the Act shows that an employee has the right to sign a fixed-term labour con- tract if the employee has worked for ten years for an employer, which indicates that listed companies have to hire and pay wages to an employee with a fixed-term labour contract even though listed companies do not need the employee, which also indicates an increase in the labour costs in listed companies. Because it is difficult to dismiss an employee with a 42 D. YAng AnD C. ZHAo fixed-term labour contract, listed companies will refrain from hiring new employees, which means an increase in the labour costs in listed companies because seasoned  employees usually have higher wages and better benefits than new employees. Through the mecha- nism of increased labour costs, excess employment can cause a deeper decline in the value or performance of listed companies, and we state our second hypothesis as follows: H2: The negative relationship between excess employment and corporate valuation should be strengthened with the promulgation and implementation of the new version of the Labour Contract Law. According to the information intermediary theory of the media (Bushee et al., 2010), media coverage can lower information asymmetry and systematically affect investors, analysts, creditors and other market participants. Corporate valuation ( Tobin’s Q) is the shareholders’ evaluation of corporate valuation, which is affected by shareholders and retail investors’ transaction behaviour. If corporate managers take some action that is not beneficial to shareholders, such as hiring a great number of excess employees, it should have a negative effect on Tobin’s Q. However, in a situation of information symmetry, the negative impact should be small. For example, the majority of shareholders and retail investors understand the rationality of managers’ behaviour; consequently this negative impact should be small. Based on theory, according to Liu et al. (2013), with lowering information asymmetry, media coverage can weaken the impact of adverse factors on corporate valuation. Because excess employment is likely to be harmful to firms, we expect that media coverage will reduce the relationship between excess employment and the valuation of firms. So, we state our third hypothesis as follows. H3: The negative relationship between excess employment and corporate valuation will be weakened with the improvement of media coverage. 5. Sample, descriptive statistics and research design 5.1. Sample We draw our sample from A-share listed companies in the period 2006–2012, during which the number of non-financial listed companies provided by the WInD database were 1,420, 1,521, 1,575, 1,721, 2,072, 2,301 and 2,422, respectively. We delete financial listed compa- nies because these companies’ financial data are very different from those of other listed companies. on this basis, after deleting those missing the main variable, our final sample, which comprises 1,2761 firms, meets the data requirements for our hypotheses testing. The observations by year are 1,371, 1,478, 1,549, 1,672, 2,000, 2,272 and 2,419, respectively. The data in this study consist of media coverage details that were manually collected from the WInD database. The WInD database includes everyday news reports on Chinese listed firms from more than 100 important newspapers. According to Yang and Zhao (2012), the data on media monitoring or media coverage in the WInD database mainly cover the financial media reports of Chinese listed firms. o ther financial data are also provided by the WInD database. In this paper, all of the continuous variables are winsorised at the 1st and 99th percentiles. t he issues of excess employment may be the result of either firms reluctantly hiring employees or firms being unable to dismiss employees (Zeng, 2006). t hat is, there are a number of redundant employees in a firm with excess employment. t hus, the increasing labour costs of employees can increase the cost of excess employment. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 43 Table 1. d efinitions of the main variables. Variable Definition Number t he number of the staff in a firm Emp (10,000× Number)/total assets Exemp1 t he residual of regressing equation (1) by using the total sample Exemp2 t he residual of regressing equation (1) industry by industry and without year by year TobinQ t he ratio of firms’ market value to replacement cost News t he logarithm of one plus the number of articles in media YearN YearN equals 1 if the sample distribution in 2009, 2010, 2011, and 2012, and 0 if the sample distribution in 2006, 2007 and 2008 Lever Lever equals total assets divided by total liabilities at the end of the accounting period. Size t he logarithm of a firm’s total assets at the end of the period State State equals 1 if the list is a state-owned listed company, and 0 otherwise Roa t he ratio of earnings divided by the total assets at the end of the accounting period Growth Sales growth in an accounting period Ppe t he ratio of fixed assets to total assets at the end of the accounting period 5.2. Research design Zeng and Chen (2006) indicate that there are some factors, including corporate size, capital intensity, company growth and industry characteristics, that can affect the size of the work - force. Furthermore, corporate performance and financial situation can also affect the number of the employees (Liu, Zhang, Wang, & Wu, 2010). In according with Zeng and Chen (2006) and Liu et al. (2010), this paper uses the following formula to estimate excess employment: Emp =  +  Size +  Lever +  Roa +  Growth +  Ppe + (1) 1 2 3 4 5 The definition of variables in formula (1) is shown in Table 1. In formula (1), Emp is a depend- ent variable applied to measure the number of employees. Size, Lever, Roa, Growth and Ppe are independent variables. We show the definition of these variables in Table 1. According to Zeng and Chen (2006), Liu et al. (2010), the residuals of formula (1) can represent the abnormal number of employees, which is excess employment. Based on this principle, Liu et al. (2010) regress formula (1) according to year by year and industry by indus- try, and the regression’s residuals are regarded as the proxy variable of excess employment. Although there is no clear economic implication of the regression’s residual, the relative value of the residual is significant; that is, the firm having the bigger residual has more excess employees than the firm with the smaller residual. Because our paper compares firms’ excess employment in different years, it is unsuita- ble for regressing formula (1) year by year. In our paper, we get a variable named Exemp1, which is the residual of regressing formula (1) using the total sample, and we get the other variable named Exemp2, which is the residual of regressing formula (1) industry by industry and not year by year. After calculating the proxy variables of the excess of employment, namely, Exemp1 and Exemp2, we use equations (2) to (4) to test Hypotheses 1, 2 and 3, respectively. TobinQ =  +  Exemp + ControlVarible + (2) i,t 1 i,t i,t TobinQ =  +  Exemp +  (Exemp × YearN )+  YearN + ControlVarible + (3) i,t 1 i,t 2 i,t i,t 3 i,t i,t TobinQ =  +  Exemp +  (Exemp × News )+  News + ControlVarible + (4) i,t 1 i,t 2 i,t i,t 3 i,t i,t 44 D. YAng AnD C. ZHAo In these formulae, subscript i denotes the sample firm, and subscript t denotes the year in the sample period. TobinQ is a dependent variable reflecting corporate valuation. According to guo (2012), Lei, Li, and Wang (2009), Wu (2009) and Xia and Fang (2005), TobinQ is the ratio of firm mar - ket value to replacement cost. Exemp is an independent variable applied to measure excess employment and comprises Exemp1 and Exemp2. The other variable is YearN, which is a dummy variable. YearN equals 1 if the sample dis- tribution in 2009, 2010, 2011 and 2012, and 0 if the sample distribution in 2006, 2007 and 2008. Because the Chinese Labour Contract Law enacted in 2007 was implemented on 1 January 2008, given the legal lag effect, the Act may have had no effect in 2008. Based on this consideration, that YearN equals 1 shows the Act has already affected listed companies. on the contrary, that YearN equals 0 shows the Act has not affected listed companies. In the robustness test, we make YearN equal 1 if the sample distribution in 2008, 2009, 2010, 2011 and 2012, and 0, if the sample distribution in 2006 and 2007, which can guarantee the robustness of this paper’s result. With equation (2), we can test H1. If H1 is supported, the coefficient of β in equation (2) should be negative. (Exemp  × YearN ) is the interaction between Exemp and News. (Exemp  × YearN ) consists i,t i,t i,t i,t of two variables, namely,(Exemp1  × YearN )and (Exemp2  × YearN ). (Exemp1  × YearN ) i,t i,t i,t i,t i,t i,t represents the interaction between Exemp1 and YearN, and (Exemp2  × YearN ) represents i,t i,t the interaction between Exemp2 and YearN. With formula (3), we can test Hypothesis 2. If Hypothesis 2 is supported, the coefficient of β in formula (3) should be significantly negative. news is an independent variable reflecting media monitoring following Dyck et al. (2008) and Core et al. (2008). (Exemp   ×  News ) is the interaction between Exemp i,t i,t and News. (Exemp   ×  News ) consists of two variables, namely, (Exemp1   ×  News ) and i,t i,t i,t i,t (Exemp2   ×  News ). (Exemp1   ×  News ) represents the interaction between Exemp1 and i,t i,t i,t i,t News, and (Exemp2  × News ) represents the interaction between Exemp2 and News. With i,t i,t formula (4), we can test Hypothesis 3. If Hypothesis 3 is supported, the coefficient of β in equation (4) should be significantly positive. We choose control variables according to Wu (2009) and Xia and Fang (2005), and we also control the fixed effects of industry and year. The specific definitions of the control variables are shown in Table 1. 5.3. Descriptive statistics In Table 2, we group the sample according to the YearN variable and provide the descriptive statistics of the main variables. In Figures 1, 2 and 3, we provide the annual average number of three variables, which includes Number, Exemp1 and Exemp1, respectively. From Figures 1–3 and Table 2, we can find certain characteristics. First, from 2006 to 2012, there was a growth trend in listed companies’ average number of employees, but the growth trend changed in 2010. on the contrary, listed companies’ average number of employees decreased in 2010. The result may be related to the Chinese Labour Contract Law enacted in 2007. Because labour costs increased after the Act, the growth trend in listed companies’ average number of employees was reduced. o f course, this is an inference from Figure 1. From Table 2, listed companies’ average number of employees in the sample after 2008 is significantly higher than those in the sample before 2008 (including 2008), which shows the T statistic is significant, but the T statistic is not significant. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 45 Table 2. d escriptive statistics. Variable Mean Median Standarddeviation obs. T-statistics Z -statistics Number 4,892.920 1,651.000 18,808.180 12,761 –2.477** –1.555 Exemp1 0 –0.002 0.009 12,761 8.221*** 4.105*** Exemp2 0 –0.002 0.009 12,761 10.891*** 8.796*** TobinQ 2.255 1.728 1.726 12,761 –6.781*** –12.149*** News 3.684 3.584 0.878 12,670 –21.236*** –19.728 Roa 6.807 6.190 7.723 12,761 –4.129*** –3.797*** Growth 18.635 13.780 40.864 12,761 3.575*** 2.215** Ppe 0.250 0.214 0.180 12,761 16.755*** 17.136*** *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. refer to t able 1 for the definition of the variables. t he t -statistic is the mean test between the sample of YearN = 0 and YearN = 1; the Z-statistic is the Wilcoxon test between the sample of YearN = 0 and YearN = 1. 2006 2007 2008 2009 2010 2011 2012 Figure 1. t he listed companies’ annual average number of employees. -.0005 -.001 2006 2007 2008 2009 2010 2011 2012 Figure 2. listed companies’ annual average number of excess employees. ( t he y-axis represents exemp1.) Second, Table 2 shows that the listed companies in the sample after 2008 have signifi- cantly better performance (Roa) and better growth (Growth) than the listed companies in the sample before 2008 (including 2008), in addition to lower Ppe and Lever. 46 D. YAng AnD C. ZHAo -.0005 -.001 2006 2007 2008 2009 2010 2011 2012 Figure 3. listed companies’ annual average number of excess employees. ( t he y-axis represents exemp2.) Table 3. regression results for equation (2). Variables (1) (2) Constant 13.824*** 13.871*** (0.000) (0.000) Exemp1 –4.489*** (0.001) Exemp2 –2.282* (0.097) Lever –0.292*** –0.296*** (0.000) (0.000) Size –0.574*** –0.576*** (0.000) (0.000) State –0.075*** –0.076*** (0.006) (0.006) Year effect Controlled Controlled Industry effect Controlled Controlled observations 12761 12761 adj. r-sq 0.354 0.354 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. 6. Empirical results Table 3 reports the regression results of equation (2). After controlling for other factors, the results show that Exemp1 has a coefficient of –4.489 (significant at the 1% level with a p-statistic of 0.001), and Exemp2 has a coefficient of –2.282 (significant at the 10% level with a p-statistic of 0.097). These negative and significant coefficients support H1. The results show that excess employment can lead to certain economic consequences, such as reduced corporate valuation, but these results do not show the mechanism that explains why excess employment can trigger these negative economic consequences. The mechanism may either be reducing production efficiency or increasing labour costs, both of which are caused by excess employment. next, we will research the mechanism by testing H2. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 47 Table 4. regression results for formula (3) and formula (4). Variables (1) (2) (3) (4) Constant 13.869*** 13.875*** 16.659*** 16.708*** (0.000) (0.000) (0.000) (0.000) Exemp1 –1.380* –15.070*** (0.079) (0.005) Exemp2 –0.653* –14.344** (0.050) (0.011) YearN 1.483*** 1.486*** (0.000) (0.000) Exemp1× YearN –0.793** (0.032) Exemp2× YearN –1.876** (0.035) News 0.506*** 0.505*** (0.000) (0.000) Exemp1× News 2.763* (0.064) Exemp2× News 3.144** (0.046) Lever –0.292*** –0.294*** –0.395*** –0.400*** (0.000) (0.000) (0.000) (0.000) Size –0.576*** –0.577*** –0.789*** –0.791*** (0.000) (0.000) (0.000) (0.000) State –0.076*** –0.076*** –0.086*** –0.086*** (0.006) (0.005) (0.001) (0.001) Year Controlled Controlled Controlled Controlled Industry Controlled Controlled Controlled Controlled observations 12761 12761 12670 12670 adj. R-sq 0.354 0.354 0.394 0.393 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. Table 5. Sub-sample regression results for equation (1) (grouped by year). Sub-sample: before 2008 (including Sub-sample: after 2008 2008) Variables (1) (2) (3) (4) Constant 17.369*** 17.398*** 10.444*** 10.424*** (0.000) (0.000) (0.000) (0.000) Exemp1 -2.277** -0.793* (0.012) (0.095) Exemp2 -2.036** -1.776* (0.022) (0.039) Lever -0.034 -0.664*** -0.530*** -0.528*** (0.591) (0.000) (0.000) (0.000) Size -0.663*** -0.082** -0.415*** -0.414*** (0.000) (0.016) (0.000) (0.000) State -0.082** -0.287 -0.084* -0.084* (0.017) (0.877) (0.067) (0.067) Year Controlled Controlled Controlled Controlled Industry Controlled Controlled Controlled Controlled observations 8363 8363 4398 4398 adj. R-sq 0.368 0.368 0.346 0.346 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. 48 D. YAng AnD C. ZHAo Table 6. Sub-sample regression results for equation (1) (grouped by media coverage). Variables High media coverage Low media coverage (1) (2) (3) (4) Constant 14.255*** 14.281*** 20.278*** 20.341*** (0.000) (0.000) (0.000) (0.000) Exemp1 –1.937 –5.650*** (0.334) (0.001) Exemp2 0.303 –4.072** (0.882) (0.021) Lever –0.836*** –0.836*** –0.073 –0.079 (0.000) (0.000) (0.258) (0.219) Size –0.556*** –0.558*** –0.902*** –0.905*** (0.000) (0.000) (0.000) (0.000) State –0.181*** –0.180*** 0.028 0.026 (0.000) (0.000) (0.473) (0.492) Year Controlled Controlled Controlled Controlled Industry Controlled Controlled Controlled Controlled observations 6370 6370 6391 6391 adj. R-sq 0.382 0.382 0.400 0.400 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. In Table 3, the control variables in formula (2) including Lever, Size and State have a neg- ative coefficient in all columns. The coefficients of control variables are consistent with the previous research (e.g. Wu, 2009; Xia & Fang, 2005). In Table 4, Exemp1 and Exemp2 have negative coefficients in all columns, which also supports H1. In column (1), when it is included along with YearN and the interaction term (Exemp1× YearN), the interaction term has a negative coefficient (significant at the 5% level with a p-statistic of 0.032). In column (2), when it is included along with YearN and the inter- action term (Exemp2× YearN), the interaction term has a negative coefficient (significant at the 5% level with a p-statistic of 0.035). The results indicate that the negative correlation between excess employment and corporate valuation is signic fi antly strengthened with the promulgation and implementation of the Labour Contract Law after 2008, which supports H2. Because the Act has increased labour costs, it has not reduced production efficiency. The results also imply that the mechanism that causes excess employment to reduce firms’ value is increasing labour costs rather than reducing production efficiency. In Table 4, columns (3) and (4) show that the news coefficients are positive, both at the 1% significance level. These results are quite reasonable. Because media coverage can lower information asymmetry, the market would give companies with higher media coverage a higher valuation. In column (3), the interaction term of the (Exemp1× News) coefficient is pos - itive at the 10% significance level. In column (4), the interaction term of the (Exemp2×  News) coefficient is positive at the 5% significance level. These results support H3, which indicates that the negative relationship between excess employment and corporate valuation should weaken with the improvement in media coverage. These results also support and extend the information intermediary theory of the media (Bushee et al., 2010). In Table 5, we provide regression results in sub-samples that are grouped by year. The group after 2008 represents the time in which the Chinese Labour Contract Law enacted in 2007 may have an effect on listed companies. The other group represents the time in which the Act may have no effect on listed companies. From Table 5, comparing the results between these CHInA JoUrnAL oF ACCoUnTIng STUDIeS 49 Table 7. instrumental variable regression results. Variable (1) (2) (3) (4) Exemp1 –0.219* –28.179*** (0.051) (0.001) Exemp2 –1.087** –27.525*** (0.043) (0.003) Lever 0.082 0.083 0.045 0.045 (0.176) (0.174) (0.452) (0.448) Size –0.677*** –0.677*** –0.894*** –0.893*** (0.000) (0.000) (0.000) (0.000) State –0.006 –0.006 0.024 0.024 (0.857) (0.847) (0.462) (0.451) News 0.480*** 0.480*** (0.000) (0.000) Exemp1× News 7.568*** (0.001) Exemp2× News 7.800*** (0.002) Constant 16.789*** 16.791*** 19.552*** 19.561*** (0.000) (0.000) (0.000) (0.000) Industry Controlled Controlled Controlled Controlled observations 10337 10337 10275 10275 adj. R-sq 0.248 0.248 0.285 0.286 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control the year and industry fixed effects. We take Exemp1_lag and Exemp2_lag as the instrument variables of Exemp1 and Exemp2, respectively. We also take (Exemp1× News)_lag and (Exemp2× News)_lag as the instruments variables of (Exemp1× News) and (Exemp2× News), respectively. sub-samples (column (1) vs column (3), column (2) vs column (4)), we find that the negative correlation between excess employment and corporate valuation is significantly strength- ened after 2008. For example, the Exemp1 coefficient is –2.277 at the 0.012 significance level in column (1), while the Exemp1 coefficient is –0.793 at the 0.095 significance level in column (3). For another example, the Exemp2 coefficient is –2.036 at the 0.022 significance level in column (2), while the Exemp2 coefficient is –1.776 at the 0.039 significance level in column (4). In Table 6, we provide regression results in sub-samples that are grouped by media cover- age. The high media coverage sub-sample means media coverage is greater than the median of media coverage, and the other sub-sample means media coverage is less than the median of media coverage. From Table 6, comparing the results between these sub-samples (column (1) vs column (3), column (2) vs column (4)), we find that the negative correlation between excess employment and corporate valuation is significantly weakened with the increase in media coverage. For example, the Exemp1 coefficient is –1.937 at the 0.334 significance level in column (1), while the Exemp1 coefficient is –5.65 at the 0.001 significance level in column (3). As  another  example, the Exemp2 coefficient is 0.303 at the 0.882 significance level in column (2), while the Exemp2 coefficient is –4.072 at the 0.021 significance level in column (4). These results can also support H3 from another angle. 7. Robustness test The following methods are used to carry out the robustness test. First, we adopt different methods to define some variables. For example, we make YearN equal 1 if the sample distribution in 2008, 2009, 2010, 2011, and 2012, and 0 if the sample 50 D. YAng AnD C. ZHAo distribution in 2006 and 2007. r egardless of the methods we adopt, the results are not substantially changed. Second, we group the sample through ownership. We retest H1, H2 and H3, controlling for each sample fixed effects. These three hypotheses are supported both in the state-owned sub-sample and no-state-owned sub-sample. Third, an immediate concern with our analysis is the potential of endogenous relation- ships between excess employment and corporate valuation because these two factors would influence each other. We use instrumental variables to solve the endogenous problem. As instruments for Exemp1 and Exemp2, we use Exemp1_lag and Exemp2_lag, both of which represent lagged variables of Exemp1 and Exemp2, respectively. In the field of accounting and finance, it is sometimes the case that researchers take lagged variables as instrumental vari- ables. For instance, Caramanis and Lennox (2008) study the relationship between the effort of audits and earnings management. There may be endogenous relationships between the effort of audits and earnings management because these two factors would influence each other. To solve the endogenous problem, Caramanis and Lennox (2008) take lagged variables of auditor working hours as instrument variables. Similarly, we take (Exemp1× News)_lag and (Exemp2× News)_lag as the instrumental variables of (Exemp1× News) and (Exemp2× News). In Table 7, we provide instrumental variable regression results. In this table, F statistics of Wu-Hausman are all significant (all significant at the 10% level), which can reject the hypoth - esis that Exemp1, Exemp2, (Exemp1×  News) and (Exemp2×  News) are exogenous variables. Meanwhile, these variables are highly correlated to their instrumental variables, which means there are no weak instruments. The results in Table 7 show that after controlling for other factors and the endogeneity problem, the results show that Exemp1 has a coefficient of –0.219 (significant at the 10% level with a p-statistic of 0.051), and Exemp2 has a coefficient of –1.087 (significant at the 5% level with a p-statistic of 0.043). These results also support H1. In columns (3) and (4), the interaction terms of the (Exemp1×  News) and (Exemp2×  News) coefficients are all positive at the 1% significance level, which supports H3 after controlling for other factors and the endogeneity problem. 8. Conclusion Using the data for A-share listed companies, we find that increasing excess employment will reduce the valuation of firms. Furthermore, with the promulgation and implementation of the Labour Contract Law in 2007 and its enforcement in 2008, the negative correlation between excess employment and corporate valuation is significantly strengthened, which shows that excess employment has a more serious negative economic impact after the prom- ulgation and implementation of the Act. It has been shown in prior research that labour costs increased after the Act. This paper reveals that the mechanism relating excess employment to negative economic consequences is that excess employment leads to increasing labour costs rather than reducing enterprise productivity. Furthermore, this paper presents evidence that media coverage signic fi antly reduces the negative correlation between excess employment and the valuation of firms, which extends the information intermediary theory of the media. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 51 Acknowledgements We thank two anonymous reviewers, Professor Liansheng Wu, Professor Zhengfei Lu and Professor Dong Lu for their helpful comments and suggestions. Disclosure statement no potential conflict of interest was reported by the authors. Funding The authors acknowledge financial support from the national natural Science Foundation of China (Media report, media bias and the governance of financial scandals, no. 71372168). References Bushee, B. J., Core, J. e., guay, W., & Wee, J. (2010). The role of the business press as an information intermediary. Journal of Accounting Research, 48, 1–19. Cai, F. (2010). Demographic transition, demographic dividend, and Lewis turning point in China. Economic Research Journal, (4), 4–13 (In Chinese). Caramanis, C., & Lennox, C. (2008). Audit effort and earnings management. Journal of Accounting and Economics, 45, 116–138. Chen, K. C. W., & Yuan, H. (2004). earnings management and capital resource allocation: evidence from China’s accounting-based regulation of rights issues. The Accounting Review, 79, 645–665. Core, J. e., guay, W., & Larcker, D. F. (2008). The power of the pen and executive compensation. Journal of Financial Economics, 88, 1–25. Ding, S. H. (2010). An analysis of minimum wage effects on the labor market: effect of interaction with the law of the Pr C on employment contracts. Social Sciences in China, (1), 85–102 (In Chinese). DeAngelo, H., DeAngelo, L., & gilson, S. (1996). Perceptions and the politics of finance: junk bonds and the regulatory seizure of first capital life. Journal of Financial Economics, 44, 475–511. Dyck, A., Volchkova, n., & Zingales, L. (2008). The corporate governance role of the media: e vidence from russia. Journal of Finance, 63, 1093–1135. Fama, e. F. (1980). Agency problems and the theory of the firm. The Journal of Political Economy, 88, 288–307. Fama, e. F., & Jensen, M. C. (1983). Separation of ownership and control. Journal of Law and Economics, 26, 301–325. guo, F. (2012). Foreign exchange risk hedging and firm value: e vidence from the MnCs in China. Economic Research Journal, (9), 18–31 (In Chinese). Jensen, M.C. (1979). Toward a theory of the press. Working paper, SSrn. (ssrn.94038) Lei, g. Y., Li, S. F., & Wang, X. J. (2009). Political association, auditors select and corporate value. Management World, (9), 145–155 (In Chinese). Li, P. g., & Shen, Y. F. (2010). The corporate governance role of media: empirical evidence from China. Economic Research Journal, (4), 14–27 (In Chinese). Lin, Y. F., Liu, M. X., & Zhang, Q. (2004). Policy burden and the SBC enterprises: an empirical study from China. Management World, (8), 14–27 (In Chinese). Liu, B., & McConnell, J. (2013). The role of the media in corporate governance: Do the media influence managers’ capital allocation decisions? Journal of Financial Economics, 78, 1–17. Liu, H. L., Zhang, M., Wang, Y. P., & Wu, L. S. (2010). Political connections, compensation incentive, and employee allocation efficiency. Economic Research Journal, (9), 109–121 (In Chinese). Liu, S. J., Sun, P., & Liu, n. Q. (2003). The ultimate ownership and its shareholding structures: Does it matter for corporate performance? Economic Research Journal, (4), 51–62 (In Chinese). Liu, Y. Y., & Liu, B. (2014). employment protection legislation, cost stickiness, and corporate response. Economic Research Journal, (5), 63–76 (In Chinese). 52 D. YAng AnD C. ZHAo Luo, H., Liu, B., & Zhang, W. (2013). The monitoring role of media on executive compensation. China Journal of Accounting Studies, 2, 138–156. Ma, L. F., Wang, Y. F., & Shen, X. X. (2013). Soe party governance, redundant employees and executive compensation contract. Management World, (5), 100–115 (In Chinese). Miller, g. S. (2006). The press as a watchdog for accounting fraud. Journal of Accounting Research, 44, 1001–1033. Shen, Y., Liang, S., & Chen, D. (2013). Wage and accounting conservatism: evidence from China. China Journal of Accounting Studies, 1, 32–46. Shleifer, A., & Vishny, r. (1994). Politicians and firms. The Quarterly Journal of Economics, 109, 995–1125. Wu, L. S. (2009). State ownership, preferential tax and corporate tax burdens. Economic Research Journal, (10), 109–120 (In Chinese). Xia, L. J., & Fang, Y. Q. (2005). g overnment control, institutional environment and firm value: evidence from the Chinese securities market. Economic Research Journal, (5), 40–51 (In Chinese). Xue, Y. K., & Bai, Y. X. (2008). national ownership, redundant employee and company performance. Management World, (10), 96–105 (In Chinese). Yang, D., Lu, Z. F., & Luo, D. L. (2013). Political connections, media monitoring and long-term loans. China Journal of Accounting Research, 7, 165–177. Yang, D. M., & Zhao, C. (2012). Media monitoring, media governance and managers’ compensation. Economic Research Journal, (6), 116–126 (In Chinese). Zeng, Q.S. (2006). State block-holder, over-employment and corporate value. Shanghai University of Finance and economics, Doctoral Thesis (in Chinese). Zeng, Q. S., & Chen, X. Y. (2006). State stockholder, excessive employment and labor cost. Economic Research Journal, (5), 74–86 (In Chinese). Zhang, M., Wang, C. F., & Liu, H. L. (2013). r edundancy burden and Soes’ executive incentive. Journal of Financial Research, (5), 140–151 (In Chinese). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png China Journal of Accounting Studies Taylor & Francis

Excess employment, media coverage and corporate valuation: Labour Contract Law in China

Yang, Deming; Zhao, Can
China Journal of Accounting Studies , Volume 4 (1): 19 – Jan 2, 2016
Download PDF
19 pages

Loading next page...
 
/lp/taylor-francis/excess-employment-media-coverage-and-corporate-valuation-labour-Ok5BrXYSIn

References (36)

Publisher
Taylor & Francis
Copyright
© 2016 Accounting Society of China
ISSN
2169-7221
eISSN
2169-7213
DOI
10.1080/21697213.2016.1144972
Publisher site
See Article on Publisher Site

Abstract

China Journal of aCC ounting StudieS , 2016 Vol . 4, no . 1, 34–52 http://dx.doi.org/10.1080/21697213.2016.1144972 Excess employment, media coverage and corporate valuation: Labour Contract Law in China* a b Deming Yang and Can Zhao a b Management School, Jinan university, guangzhou 510632, China; Management School, o cean university of China, Qingdao 266100, China KEYWORDS ABSTRACT excess employment; firm Previous research has found that firms’ excess employment will reduce value; labour contract law; corporate valuation as well as lead to a series of negative economic media coverage consequences. However, these studies cannot explain why the excess employment would lead to a series of negative economic consequences. Using the exogenous event of the promulgation of the new Labour Contract Law (‘the Act’) enacted in 2007 in China, this paper finds that an increase in excess employment will reduce the valuation of firms. With the promulgation and implementation of the Labour Contract Law, the negative correlation between excess employment and corporate valuation is significantly strengthened, which shows that excess employment has a more serious negative economic impact after the law is implemented. It has been shown elsewhere that labour costs have increased since the implementation of the Act. This paper reveals that the mechanism relating excess employment to negative economic consequences is that excess employment leads to increasing labour costs rather than reducing enterprise productivity. Furthermore, we present evidence that media coverage significantly reduces the negative correlation between excess employment and the valuation of the firm, which extends the information intermediary theory of media. 1. Introduction Low-cost labour is regarded as an important source of economic growth in China (Cai, 2010). However, there are negative effects owing to low-cost labour. For example, low-cost labour may misappropriate labour rights, and it may also increase the gap between the rich and the poor (Cai, 2010). To overcome these negative effects and improve the labour contract sys- tem, the Labour Contract Law (‘the Act’) was revised and issued in 2007. The promulgation and implementation of the Act have had a profound impact on the Chinese economy. We expect that the Act will have had a major impact on enterprise employment in micro-level enterprises. The theoretical and empirical evidence shows the Act would greatly increase CONTACT d eming Yang yangdeming2001@sina.com Paper accepted by d onghua Chen. t he a ct was passed on 29 June 2007 by the g eneral assembly of the t enth national People’s Congress Standing Committee of the 28th Meeting and was carried out on 1 January 2008. © 2016 a ccounting Society of China CHInA JoUrnAL oF ACCoUnTIng STUDIeS 35 labour costs (Ding, 2010; Liu & Liu, 2014). This increase in labour costs would affect enterprise employment after the Act. The promulgation and implementation of the new version of the Labour Contract Law also provide a good opportunity for researching excess employment, as experienced in China, also referred to as the over-staffing problem. The issues of excess employment may be the result of either firms reluctantly hiring employees or firms being unable to dismiss employees (Zeng, 2006). For instance, the Chinese government bears the responsibility for increasing employment, so the government might affect firm hiring and firing through a series of administrative actions (Lin, Liu, & Zhang, 2004). In this environment, firms naturally cannot dismiss or hire employees according to necessity. Some firms, especially state-owned firms, have to bear huge responsibilities to help the government address employment, which leads to a number of excess employees (Zeng, 2006; Zeng & Chen, 2006). Previous Chinese literature draws the conclusion that firms’ excess employment brings a series of negative consequences, including reducing corporate valuation, increasing labour costs, and weakening corporate governance (Zeng, 2006; Zeng and Chen, 2006; Zhang, Wang, & Liu, 2013). However, the literature does not address the mechanism that explains why excess employment can trigger these negative economic consequences. This mechanism may either be a reduction in production efficiency or an increase in labour costs, both of which are caused by excess employment. The Chinese Labour Contract Law enacted in 2007 increases firms’ labour costs (Ding, 2010; Liu & Liu, 2014), although there is no evidence that shows the Act can reduce employees’ production efficiency. If the first mechanism is true, the negative correlation between excess employment and corporate valuation should be strengthened with the promulgation and implementation of the Act because there is a negative correlation between labour costs and corporate valuation (Zeng, 2006). on the contrary, if the mechanism of reduced production efficiency is true, the negative correlation between excess employment and corporate valuation should be weakened or remain unchanged as a result of the act’s promulgation and implementation because there is a positive correlation between employees’ production efficiency and corporate valuation (Zeng, 2006). The previous literature concerns the relationship between excess employment and the valuation of firms, but it never addresses the issues from the perspective of media coverage or media monitoring. The issue of excess employment is closely related to pressing social issues; for example, social justice, social stability, and corporate governance (Lin et al., 2004). So, the media are concerned about these hot issues, and because of this, according to the information intermediary theory of media (Bushee, Core, guay, & Wee, 2010), media cover - age can lower information asymmetry, systematically affecting investors, analysts, creditors and other market participants. Based on this theory, according to Liu and McConnell (2013), with decreasing information asymmetry, media coverage can weaken the impact of adverse factors on corporate valuation. If corporate managers take some action that is not benefi- cial to shareholders, such as hiring a great number of excess employees, it should have a negative effect on Tobin Q. However, in a situation of information symmetry, the negative impact should be small (Liu & McConnell, 2013). Because firms’ excess employment brings a series of negative consequences (Zeng, 2006; Zeng & Chen, 2006; Zhang et al., 2013), that is, excess employment is harmful to firms, we expect that media coverage will reduce the relationship between excess employment and the valuation of firms. a ccording to d yck et al. (2008), media coverage is a proxy variable for media monitoring. 36 D. YAng AnD C. ZHAo Using the exogenous event of the proclamation of the Chinese Labour Contract Law in 2007 and its implementation in 2008, this paper studies the relationship between excess employment and the valuation of firms using data for Chinese A-share listed companies. We also study how media coverage affects the relation between excess employment and the valuation of firms. o ur research finds that a firm’s excess employment can significantly lower the firm’s value, and this negative correlation has been significantly strengthened with the implementation of the Labour Contract Law. The Act has raised the cost of excess employment. This implies that the mechanism is increasing labour costs rather than reduc- ing production efficiency. Further evidence shows that media coverage will significantly weaken the relation between excess employees and the valuation of firms, which supports the information intermediary theory of media. The contributions of this paper are as follows. First, some studies (Zeng, 2006; Zeng & Chen, 2006; Zhang et al., 2013) find that excess employment produces a series of negative economic consequences. However, these stud- ies do not systematically analyse the mechanism by which excess employment produces these negative economic consequences. o ur paper tests the mechanism. The result shows that the mechanism is increasing labour costs rather than reducing enterprise productivity. Second, this paper approaches the information intermediary theory of media from a new perspective, which directly contributes new evidence to the literature. Liu and McConnell (2013) find that a value-reducing acquisition has a significantly negative impact on the valuation of the firm; however, media coverage can significantly lower the negative effect on the valuation of the firm. Consistent with their study, our paper presents evidence that media coverage significantly reduces excess employment’s negative effect on the valuation of the firm, which extends the information intermediary theory of media. The remainder of this paper proceeds as follows. Section 2 discusses the institutional background. Section 3 is a review of the relevant literature. Section 4 discusses theoretical assumptions. Section 5 presents the research design. Section 6 presents the results of our empirical research. Section 7 presents robustness tests, and we conclude in Section 8 with a brief discussion. 2. Institutional background Before 1980, under the planned economy system, Chinese enterprises, especially state- owned enterprises, carried out a kind of highly centralised labour employment system (Zeng, 2006; Zeng & Chen, 2006). In this system, the government uniformly assigned jobs and job placement in an administrative way. The basic characteristics of the system were that the government controlled the number of employees and workers’ wages and employers and workers had no choice (Shen, Liang, & Chen, 2013). enterprises could hire employees but had no right to fire workers. To maintain social stability, the government strictly limited the ability of enterprises to fire workers for any economic reasons (Zeng & Chen, 2006). The employment problems of state-owned enterprises at this time were entirely the result of government behaviour (Zeng, 2006). Since 1980, China has carried out a policy of reform and opening up. With the implemen- tation of the policy, the Chinese economy has gradually begun to transform from a planned economy to a market economy. In this process, the Chinese labour-employment system has also adopted gradual reform (Zeng, 2006). In the reform of the labour-employment system, CHInA JoUrnAL oF ACCoUnTIng STUDIeS 37 the state council issued a rule named ‘The provisional regulations on the labour contract system in state-owned enterprises’ in 1986 (Shen et al., 2013; Zeng, 2006). The formulation allowed Soes to dismiss workers, which increased the risk of unemployment. However, Soes were still not able to dismiss workers in full accordance with the economic goals of enter- prises in the 1980s (Zeng, 2006). At that time, there were several reasons why the scale of Soes’ employees was still subject to government intervention. First, the existing political and economic system could not be substantially changed in a short time. If a large number of employees had been laid off, these unemployed employees would have led to social instability. Second, firing workers was also subject to a series of policy constraints. Although enterprises had some right to hire or fire workers, it was very difficult to fire workers for economic reasons (Zeng, 2006). In such a climate, Chinese enterprises, especially state-owned enterprises had to bear many policy burdens, which had a direct impact on the generally poor performance of Soes (Lin et al., 2004). To solve these policy burdens and promote the reform of Soes, the labour department proposed the establishment of the re-employment project in 1993 (Zeng, 2006). As part of the project, the government allowed a certain level of unemployment; thus, Soes began to dismiss a large number of workers. However, the number of laid-off workers was still tightly restricted by the government and administrative system (Zeng, 2006). For example, the labour law that was promulgated in 1994 stipulated that the department that either needed to dismiss workers for poor performance or be faced with bankruptcy could dismiss workers according to legal procedure. Meanwhile, the government, especially local government, gradually gained some free right over Soes dismissing workers because the government had the right to influence Soes’ performance and corporate government by administrative intervention (Zeng, 2006). Due to unemployment affecting social stability, the government, especially local government, pays special attention to the issue and limits firms’ ability to dismiss workers. Thus, government pressure to bolster employment is always an important factors that leads to firms’ excess employment (Zeng & Chen, 2006). For the capital market, Chinese stock exchanges (one in Shanghai and another in Shenzhen) were created as an experiment in mixing a market economy with central plan- ning (Chen & Yuan, 2004). g overnmental intervention has been a constant throughout, which has affected listed companies, especially state-owned listed companies (Chen & Yuan, 2004). Listed Soes can spin off their social burdens, strengthen corporate governance, and improve information transparency. In the capital market, the costs of government interven- tion should be reduced with high information transparency. However, it is an important goal for the Chinese government to control unemployment and promote employment. If the local government faces employment pressure, they let the local Soes share the pressure or restrict the local Soes’ ability to dismiss workers (Xue & Bai, 2008). Thus, pressure from the government to solve employment is one of the most important factors that always influence state-owned listed companies (Xue & Bai, 2008; Zeng & Chen, 2006). Then, we discuss the question of the methods by which the government achieves its motive to affect employment in state-owned listed companies. The government can influ- ence Soes through the performance evaluation system. In China, the State-owned Assets Supervision and Administration Commission (SASAC), as the representative of the gov- ernment, is responsible for maintaining and increasing the value of state-owned assets (Ma, Wang, & Shen 2013). To maintain and increase the value of state-owned assets, SASAC has taken a series of measures that include the appointment of senior executives and the 38 D. YAng AnD C. ZHAo establishment of a performance evaluation system. Because SASAC is responsible for the appointment of senior executives, SASAC can affect state-owned listed companies’ corpo - rate governance through the executives who are appointed by them; thus, the government actually affects the behaviour of state-owned listed companies’ employees and, similarly, can affect behaviour through the performance evaluation system (Ma et al., 2013; Zeng & Chen, 2006). For these reasons, there are a number of excess employed in state-owned listed com- panies because of the government’s administrative intervention (Zeng & Chen, 2006). The government can achieve its objectives through SASAC and other methods. We now discuss and analyse whether there is a certain amount of excess employees in non-state-owned listed companies. Because non-state-owned enterprises have developed in the market environment, these enterprises use production elements according to market forces of supply and demand (Zeng, 2006). Accordingly, there should be less excess employment in these enterprises. However, if non-state-owned enterprises want to issue shares, these enterprises have to face a great deal of administrative intervention (Liu, Sun, & Liu, 2003). In the Chinese capital market, it was important to help Soes solve the problem of poor performance in 1990s because Soes generally faced operating difficulties. In this situation, the companies in the capital market mainly consisted of Soes at the stock market initial stage (Liu et al., 2003). According to the annual statistics of the China Securities r egulatory Commission (CSr C), with the development of mergers and acquisitions, the creation of a small and medium enterprises board (SMe board), and the creation of a growth enterprise market (geM), the number of non-state-owned listed companies has risen continuously. either IPo from the SMe board or geM or going public through mergers and acquisi- tions (M&A), non-state-owned listed  companies have to face strict supervision and strict administrative intervention (Liu et al., 2003). The government can also affect non-state- owned listed companies by taxes, or government subsidies. Compared with state-owned listed companies, intervention in non-state-owned listed companies is weaker because the government has no direct role in the affairs of non-state-owned listed companies. So, there are a number of excess employees in non-state-owned listed companies, but the number in these companies is significantly less than the number in state-owned listed companies (Zeng and Chen, 2006). We consider that the issue of excess employment has changed with time and with impor- tant reforms of the system. For instance, the Labour Contract Law in 2008 might have an ee ff ct (Liu & Liu, 2014 ). In general, the Act enforces a greater degree of employee protection. The employee protections include the following. (1) The Act demands that firms ensure a minimum wage. (2) The Act demands that firms offer rest and vacation, labour safety, health regulations and insurance benefits, which indirectly increase the labour cost of firms. (3) The Act increases the cost of signing, carrying out, modifying and terminating the labour contract, which directly increases the labour cost of firms. (4) The greatest impact on the labour cost of listed companies should be the labour contract that does not have a fixed term. The following is the content of the Act relating to such a contract. Article 14: A non-fixed-term labour contract shall refer to a labour contract for which the employer and the worker have agreed that the date of termination is not fixed. Upon negotiation and consensus between the employer and the worker, a non-fixed-term labour contract may be concluded. Under any of the following circumstances, a worker may propose CHInA JoUrnAL oF ACCoUnTIng STUDIeS 39 or agree to renewal or conclusion of labour contract; except where a worker proposes for the conclusion of a fixed-term labour contract, a non-fixed-term labour contract shall be concluded: (1) the worker has worked for the employer for a period of 10 years consecutively. (2) when the employer first implemented the labour contract system or when a new labour contract is concluded upon restructuring of a state-owned enterprise, the worker has worked for the employer for a period of 10 years consecutively and will attain his/her statutory retirement age in less than 10 years. (3) w here a fixed-term labour contract has been concluded twice consecutively and the worker who does not fall under any of the categories stipulated in Article 39 and item (1) and item (2) of Article 40 renews his/her labour contract. Where an employer fails to conclude a written contract with a worker after one year has lapsed since the date of employment, the employer and the worker shall be deemed to have concluded a non-fixed-term labour contract. Article 14 means that provided an employee has worked for ten years for an employer, the employee has the right to sign a fixed-term labour contract, and the employer has no right to dismiss the employee. The impact on listed companies may be reflected in two respects: first, employers of listed companies have no right to dismiss seasoned employees who have worked for a company for ten years or who are eligible under the fourteenth term of the Act. Second, due to the difficulty of dismissing seasoned employees, listed companies will reduce their intention to hire new employees. In short, the Act has increased the minimum wage standard and labour costs (Ding, 2010; Liu & Liu, 2014), which would affect excess employment in listed companies. 3. Literature review The literature of excess employment consists of two branches. The first studies the factors that affect excess employment. The second addresses the economic consequences of excess employment. Some of the theoretical literature considers excess employment to be closely related to political benefits (e.g., Shleifer & Vishny, 1994). empirical studies support the view of Shleifer and Vishny (1994) in providing evidence on the factors that affect excess employment. For example, Lin et al. (2004) indicate that there are widespread ‘policy burdens’ for Chinese companies in transition economies and that China is a labour-surplus country. To solve the labour surplus issue, the government would require companies, especially state-owned companies, to hire a large number of employees, even though these companies are reluc- tant to hire too many employees; this policy creates social policy burdens on companies. Similarly, Zeng (2006), Zeng and Chen (2006) also believe that listed companies have to bear the burden of helping the government solve the unemployment issue, so there are generally excess employees in Chinese listed companies. Because Soes face greater administrative intervention, the number of excess employees in state-owned companies is significantly higher than that in private listed companies (e.g. Zeng & Chen, 2006). Xue and Bai (2008) provide evidence that, compared with those in low unemployment areas, companies in high unemployment areas have to take on more responsibility to help the government bolster employment. These companies also have to bear more of the local governments’ pressure and administrative intervention. o ther findings include Ma et al. (2013) who discover that Chinese Communist Party organisations of state-owned enterprises (Soes) can participate in corporate governance, 40 D. YAng AnD C. ZHAo which increases excess employment significantly. Their empirical evidence also shows that administrative factors play a very important role in the generation of excess employment. The literature regarding the economic consequences of excess employment offers a consistent conclusion; that is, excess employment brings a series of adverse consequences. For example, excess employment reduces the valuation of firms significantly (Zeng, 2006), as well as increasing labour costs significantly (Zeng & Chen, 2006). Analogously, Xue and Bai (2008) find that excess employment reduces the performance of firms significantly and that firms in high unemployment areas bear more excess employment, as a consequence of which these firms receive more fiscal subsidies than other firms. The adverse consequences also include the fact that excess employ - ment may reduce the sensitivity between executive compensation and corporate performance; in other words, excess employment might weaken corporate governance (Zhang et al., 2013). o verall, excess employment has a negative impact on corporate performance to some extent. o ur research is related to other research fields, specifically with regard to the corporate gov - ernance role of the media. In recent years, more studies have begun to pay close attention to this field of research. Dyck, Volchkova and Zingales (2008) explain why the media have corporate governance functions. The role of the media is to collect, select, certify, and repackage informa- tion (Dyck et al., 2008). By doing so, they increase the number of people who learn about the behaviour of others, thereby increasing the effect of reputation. As top executives, managers care more about their companies’ reputation. To avoid the loss of reputation, managers have the incentive to improve company performance or improve corporate governance. Thus, the media can play an important role in corporate governance by increasing reputation cost. The media can play an important role in corporate governance in another way, such as monitoring (Miller, 2006). Miller (2006) draws the conclusion that the media can affect corporate governance by performing the function of oversight, i.e. the role of ‘watchdog’. Some Chinese studies (e.g. Li & Shen, 2010; Luo, Liu, & Zhang, 2013; Yang, Lu, & Luo, 2013; Yang & Zhao, 2012) determine that the media play an important role in corporate governance by administrative interference. Why should this be the case? Chinese enterprise executives usually do not care about their companies’ reputation but do care about administrative regulation. Thus, the media can usually affect corporate governance by having an influence on administrative supervision instead of executives’ reputation. These studies in practice support Miller (2006). While many studies suggest that the media have an impact on public perception in some manner, one group of studies indicates media coverage lacks in-depth research and tends toward sensationalism. For example, Jensen (1979) argues that the media have become a form of entertainment and that articles are written to appeal to the lowest common denominator, so the media cannot play an important role in the macro and micro economy. Consistent with this, Core, guay, and Larcker (2008) study media coverage of compensa- tion and conclude that the media sensationalises their coverage by focusing on large ex post stock gains rather than compensation expenses to the company. They conclude that media coverage has no impact on compensation behaviour. Using a clinical study approach, DeAngelo, DeAngelo, and gilson (1996) suggest that simplistic media coverage of junk bonds may have skewed the economic behaviour of customers and regulators. As a group, these studies question the validity of the media as an important information intermediary in the economy and even suggest it may play a negative role. Starting with f ama (1980), the finance literature has recognised the importance reputation plays in disciplining corporate managers. t he early literature, f ama (1980) and f ama and Jensen (1983), emphasises that manager reputations would affect potential employers, who determine future jobs and wages. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 41 In contrast, other studies suggest the media form an important component of the infor- mation environment in society. Some studies use cross-sectional variation in the national characteristics of the media to investigate the role of the media in corporate governance and economic development (Dyck et al., 2008). This literature generally concludes that com- petitive media can have a positive impact on the political and economic development of a country. This impact is presumed to be driven by the oversight function of the media. o ther research also supports that the media have a supervision function. For example, Miller (2006) investigates the role of the press as a monitor or ‘watchdog’ for accounting fraud. The research finds that the press fulfils this monitoring role by rebroadcasting information from other information intermediaries (analysts, auditors, and lawsuits) and by undertaking original investigations and analysis. Liu and McConnell (2013) propose that the media are key conduits through which managers’ reputation capital is built and, potentially, destroyed. Their research supports the theory of reputation put forward by Dyck et al. (2008). o verall, although these conclusions are controversial, it has gradually become the consen- sus that the media play an important supervisory function. In China, although the media do not have total freedom, they do play an important supervisory function (e.g. Li & Shen, 2010). From another perspective, there is a lack of literature that discusses how media coverage influences the relation between excess employment and the valuation of firms. 4. Hypothesis development The issues of excess employment may be the result of firms reluctantly hiring a large number of employees or being unable to dismiss employees. The number of excess employees in listed companies indicates the extent to which these employees are not able to produce enough value and their wages or labour costs are too high. In brief, the value creation does not match the wages of these employees, which causes a decline in the value or performance of listed companies (Zeng & Chen, 2006). If the capital market is efficient, it can draw a reasonable valuation for this kind of phenomenon. So, we expect that excess employment should reduce the valuation of firms. In addition, excess employment can also be seen as a proxy variable for government intervention or administrative intervention, which might lead to the weakening of corporate governance as well as declining performance (Xia & Fang, 2005; Zeng, 2006). Based on the above discussion, we state our first hypothesis as follows. H1: There is a negative relationship between excess employment and corporate valuation. Previous literature draws the conclusion that firms’ excess employment creates a series of negative consequences (Zeng, 2006; Zeng & Chen, 2006; Zhang et al., 2013). However, these studies do not explain why excess employment can trigger these negative economic conse- quences. This mechanism may either be reducing production efficiency or increasing labour costs, both of which are due to excess employment. The Labour Contract Law increases labour costs, as has been shown in some studies (Ding, 2010; Liu & Liu, 2014). For example, Article 14 in the Act shows that an employee has the right to sign a fixed-term labour con- tract if the employee has worked for ten years for an employer, which indicates that listed companies have to hire and pay wages to an employee with a fixed-term labour contract even though listed companies do not need the employee, which also indicates an increase in the labour costs in listed companies. Because it is difficult to dismiss an employee with a 42 D. YAng AnD C. ZHAo fixed-term labour contract, listed companies will refrain from hiring new employees, which means an increase in the labour costs in listed companies because seasoned  employees usually have higher wages and better benefits than new employees. Through the mecha- nism of increased labour costs, excess employment can cause a deeper decline in the value or performance of listed companies, and we state our second hypothesis as follows: H2: The negative relationship between excess employment and corporate valuation should be strengthened with the promulgation and implementation of the new version of the Labour Contract Law. According to the information intermediary theory of the media (Bushee et al., 2010), media coverage can lower information asymmetry and systematically affect investors, analysts, creditors and other market participants. Corporate valuation ( Tobin’s Q) is the shareholders’ evaluation of corporate valuation, which is affected by shareholders and retail investors’ transaction behaviour. If corporate managers take some action that is not beneficial to shareholders, such as hiring a great number of excess employees, it should have a negative effect on Tobin’s Q. However, in a situation of information symmetry, the negative impact should be small. For example, the majority of shareholders and retail investors understand the rationality of managers’ behaviour; consequently this negative impact should be small. Based on theory, according to Liu et al. (2013), with lowering information asymmetry, media coverage can weaken the impact of adverse factors on corporate valuation. Because excess employment is likely to be harmful to firms, we expect that media coverage will reduce the relationship between excess employment and the valuation of firms. So, we state our third hypothesis as follows. H3: The negative relationship between excess employment and corporate valuation will be weakened with the improvement of media coverage. 5. Sample, descriptive statistics and research design 5.1. Sample We draw our sample from A-share listed companies in the period 2006–2012, during which the number of non-financial listed companies provided by the WInD database were 1,420, 1,521, 1,575, 1,721, 2,072, 2,301 and 2,422, respectively. We delete financial listed compa- nies because these companies’ financial data are very different from those of other listed companies. on this basis, after deleting those missing the main variable, our final sample, which comprises 1,2761 firms, meets the data requirements for our hypotheses testing. The observations by year are 1,371, 1,478, 1,549, 1,672, 2,000, 2,272 and 2,419, respectively. The data in this study consist of media coverage details that were manually collected from the WInD database. The WInD database includes everyday news reports on Chinese listed firms from more than 100 important newspapers. According to Yang and Zhao (2012), the data on media monitoring or media coverage in the WInD database mainly cover the financial media reports of Chinese listed firms. o ther financial data are also provided by the WInD database. In this paper, all of the continuous variables are winsorised at the 1st and 99th percentiles. t he issues of excess employment may be the result of either firms reluctantly hiring employees or firms being unable to dismiss employees (Zeng, 2006). t hat is, there are a number of redundant employees in a firm with excess employment. t hus, the increasing labour costs of employees can increase the cost of excess employment. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 43 Table 1. d efinitions of the main variables. Variable Definition Number t he number of the staff in a firm Emp (10,000× Number)/total assets Exemp1 t he residual of regressing equation (1) by using the total sample Exemp2 t he residual of regressing equation (1) industry by industry and without year by year TobinQ t he ratio of firms’ market value to replacement cost News t he logarithm of one plus the number of articles in media YearN YearN equals 1 if the sample distribution in 2009, 2010, 2011, and 2012, and 0 if the sample distribution in 2006, 2007 and 2008 Lever Lever equals total assets divided by total liabilities at the end of the accounting period. Size t he logarithm of a firm’s total assets at the end of the period State State equals 1 if the list is a state-owned listed company, and 0 otherwise Roa t he ratio of earnings divided by the total assets at the end of the accounting period Growth Sales growth in an accounting period Ppe t he ratio of fixed assets to total assets at the end of the accounting period 5.2. Research design Zeng and Chen (2006) indicate that there are some factors, including corporate size, capital intensity, company growth and industry characteristics, that can affect the size of the work - force. Furthermore, corporate performance and financial situation can also affect the number of the employees (Liu, Zhang, Wang, & Wu, 2010). In according with Zeng and Chen (2006) and Liu et al. (2010), this paper uses the following formula to estimate excess employment: Emp =  +  Size +  Lever +  Roa +  Growth +  Ppe + (1) 1 2 3 4 5 The definition of variables in formula (1) is shown in Table 1. In formula (1), Emp is a depend- ent variable applied to measure the number of employees. Size, Lever, Roa, Growth and Ppe are independent variables. We show the definition of these variables in Table 1. According to Zeng and Chen (2006), Liu et al. (2010), the residuals of formula (1) can represent the abnormal number of employees, which is excess employment. Based on this principle, Liu et al. (2010) regress formula (1) according to year by year and industry by indus- try, and the regression’s residuals are regarded as the proxy variable of excess employment. Although there is no clear economic implication of the regression’s residual, the relative value of the residual is significant; that is, the firm having the bigger residual has more excess employees than the firm with the smaller residual. Because our paper compares firms’ excess employment in different years, it is unsuita- ble for regressing formula (1) year by year. In our paper, we get a variable named Exemp1, which is the residual of regressing formula (1) using the total sample, and we get the other variable named Exemp2, which is the residual of regressing formula (1) industry by industry and not year by year. After calculating the proxy variables of the excess of employment, namely, Exemp1 and Exemp2, we use equations (2) to (4) to test Hypotheses 1, 2 and 3, respectively. TobinQ =  +  Exemp + ControlVarible + (2) i,t 1 i,t i,t TobinQ =  +  Exemp +  (Exemp × YearN )+  YearN + ControlVarible + (3) i,t 1 i,t 2 i,t i,t 3 i,t i,t TobinQ =  +  Exemp +  (Exemp × News )+  News + ControlVarible + (4) i,t 1 i,t 2 i,t i,t 3 i,t i,t 44 D. YAng AnD C. ZHAo In these formulae, subscript i denotes the sample firm, and subscript t denotes the year in the sample period. TobinQ is a dependent variable reflecting corporate valuation. According to guo (2012), Lei, Li, and Wang (2009), Wu (2009) and Xia and Fang (2005), TobinQ is the ratio of firm mar - ket value to replacement cost. Exemp is an independent variable applied to measure excess employment and comprises Exemp1 and Exemp2. The other variable is YearN, which is a dummy variable. YearN equals 1 if the sample dis- tribution in 2009, 2010, 2011 and 2012, and 0 if the sample distribution in 2006, 2007 and 2008. Because the Chinese Labour Contract Law enacted in 2007 was implemented on 1 January 2008, given the legal lag effect, the Act may have had no effect in 2008. Based on this consideration, that YearN equals 1 shows the Act has already affected listed companies. on the contrary, that YearN equals 0 shows the Act has not affected listed companies. In the robustness test, we make YearN equal 1 if the sample distribution in 2008, 2009, 2010, 2011 and 2012, and 0, if the sample distribution in 2006 and 2007, which can guarantee the robustness of this paper’s result. With equation (2), we can test H1. If H1 is supported, the coefficient of β in equation (2) should be negative. (Exemp  × YearN ) is the interaction between Exemp and News. (Exemp  × YearN ) consists i,t i,t i,t i,t of two variables, namely,(Exemp1  × YearN )and (Exemp2  × YearN ). (Exemp1  × YearN ) i,t i,t i,t i,t i,t i,t represents the interaction between Exemp1 and YearN, and (Exemp2  × YearN ) represents i,t i,t the interaction between Exemp2 and YearN. With formula (3), we can test Hypothesis 2. If Hypothesis 2 is supported, the coefficient of β in formula (3) should be significantly negative. news is an independent variable reflecting media monitoring following Dyck et al. (2008) and Core et al. (2008). (Exemp   ×  News ) is the interaction between Exemp i,t i,t and News. (Exemp   ×  News ) consists of two variables, namely, (Exemp1   ×  News ) and i,t i,t i,t i,t (Exemp2   ×  News ). (Exemp1   ×  News ) represents the interaction between Exemp1 and i,t i,t i,t i,t News, and (Exemp2  × News ) represents the interaction between Exemp2 and News. With i,t i,t formula (4), we can test Hypothesis 3. If Hypothesis 3 is supported, the coefficient of β in equation (4) should be significantly positive. We choose control variables according to Wu (2009) and Xia and Fang (2005), and we also control the fixed effects of industry and year. The specific definitions of the control variables are shown in Table 1. 5.3. Descriptive statistics In Table 2, we group the sample according to the YearN variable and provide the descriptive statistics of the main variables. In Figures 1, 2 and 3, we provide the annual average number of three variables, which includes Number, Exemp1 and Exemp1, respectively. From Figures 1–3 and Table 2, we can find certain characteristics. First, from 2006 to 2012, there was a growth trend in listed companies’ average number of employees, but the growth trend changed in 2010. on the contrary, listed companies’ average number of employees decreased in 2010. The result may be related to the Chinese Labour Contract Law enacted in 2007. Because labour costs increased after the Act, the growth trend in listed companies’ average number of employees was reduced. o f course, this is an inference from Figure 1. From Table 2, listed companies’ average number of employees in the sample after 2008 is significantly higher than those in the sample before 2008 (including 2008), which shows the T statistic is significant, but the T statistic is not significant. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 45 Table 2. d escriptive statistics. Variable Mean Median Standarddeviation obs. T-statistics Z -statistics Number 4,892.920 1,651.000 18,808.180 12,761 –2.477** –1.555 Exemp1 0 –0.002 0.009 12,761 8.221*** 4.105*** Exemp2 0 –0.002 0.009 12,761 10.891*** 8.796*** TobinQ 2.255 1.728 1.726 12,761 –6.781*** –12.149*** News 3.684 3.584 0.878 12,670 –21.236*** –19.728 Roa 6.807 6.190 7.723 12,761 –4.129*** –3.797*** Growth 18.635 13.780 40.864 12,761 3.575*** 2.215** Ppe 0.250 0.214 0.180 12,761 16.755*** 17.136*** *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. refer to t able 1 for the definition of the variables. t he t -statistic is the mean test between the sample of YearN = 0 and YearN = 1; the Z-statistic is the Wilcoxon test between the sample of YearN = 0 and YearN = 1. 2006 2007 2008 2009 2010 2011 2012 Figure 1. t he listed companies’ annual average number of employees. -.0005 -.001 2006 2007 2008 2009 2010 2011 2012 Figure 2. listed companies’ annual average number of excess employees. ( t he y-axis represents exemp1.) Second, Table 2 shows that the listed companies in the sample after 2008 have signifi- cantly better performance (Roa) and better growth (Growth) than the listed companies in the sample before 2008 (including 2008), in addition to lower Ppe and Lever. 46 D. YAng AnD C. ZHAo -.0005 -.001 2006 2007 2008 2009 2010 2011 2012 Figure 3. listed companies’ annual average number of excess employees. ( t he y-axis represents exemp2.) Table 3. regression results for equation (2). Variables (1) (2) Constant 13.824*** 13.871*** (0.000) (0.000) Exemp1 –4.489*** (0.001) Exemp2 –2.282* (0.097) Lever –0.292*** –0.296*** (0.000) (0.000) Size –0.574*** –0.576*** (0.000) (0.000) State –0.075*** –0.076*** (0.006) (0.006) Year effect Controlled Controlled Industry effect Controlled Controlled observations 12761 12761 adj. r-sq 0.354 0.354 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. 6. Empirical results Table 3 reports the regression results of equation (2). After controlling for other factors, the results show that Exemp1 has a coefficient of –4.489 (significant at the 1% level with a p-statistic of 0.001), and Exemp2 has a coefficient of –2.282 (significant at the 10% level with a p-statistic of 0.097). These negative and significant coefficients support H1. The results show that excess employment can lead to certain economic consequences, such as reduced corporate valuation, but these results do not show the mechanism that explains why excess employment can trigger these negative economic consequences. The mechanism may either be reducing production efficiency or increasing labour costs, both of which are caused by excess employment. next, we will research the mechanism by testing H2. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 47 Table 4. regression results for formula (3) and formula (4). Variables (1) (2) (3) (4) Constant 13.869*** 13.875*** 16.659*** 16.708*** (0.000) (0.000) (0.000) (0.000) Exemp1 –1.380* –15.070*** (0.079) (0.005) Exemp2 –0.653* –14.344** (0.050) (0.011) YearN 1.483*** 1.486*** (0.000) (0.000) Exemp1× YearN –0.793** (0.032) Exemp2× YearN –1.876** (0.035) News 0.506*** 0.505*** (0.000) (0.000) Exemp1× News 2.763* (0.064) Exemp2× News 3.144** (0.046) Lever –0.292*** –0.294*** –0.395*** –0.400*** (0.000) (0.000) (0.000) (0.000) Size –0.576*** –0.577*** –0.789*** –0.791*** (0.000) (0.000) (0.000) (0.000) State –0.076*** –0.076*** –0.086*** –0.086*** (0.006) (0.005) (0.001) (0.001) Year Controlled Controlled Controlled Controlled Industry Controlled Controlled Controlled Controlled observations 12761 12761 12670 12670 adj. R-sq 0.354 0.354 0.394 0.393 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. Table 5. Sub-sample regression results for equation (1) (grouped by year). Sub-sample: before 2008 (including Sub-sample: after 2008 2008) Variables (1) (2) (3) (4) Constant 17.369*** 17.398*** 10.444*** 10.424*** (0.000) (0.000) (0.000) (0.000) Exemp1 -2.277** -0.793* (0.012) (0.095) Exemp2 -2.036** -1.776* (0.022) (0.039) Lever -0.034 -0.664*** -0.530*** -0.528*** (0.591) (0.000) (0.000) (0.000) Size -0.663*** -0.082** -0.415*** -0.414*** (0.000) (0.016) (0.000) (0.000) State -0.082** -0.287 -0.084* -0.084* (0.017) (0.877) (0.067) (0.067) Year Controlled Controlled Controlled Controlled Industry Controlled Controlled Controlled Controlled observations 8363 8363 4398 4398 adj. R-sq 0.368 0.368 0.346 0.346 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. 48 D. YAng AnD C. ZHAo Table 6. Sub-sample regression results for equation (1) (grouped by media coverage). Variables High media coverage Low media coverage (1) (2) (3) (4) Constant 14.255*** 14.281*** 20.278*** 20.341*** (0.000) (0.000) (0.000) (0.000) Exemp1 –1.937 –5.650*** (0.334) (0.001) Exemp2 0.303 –4.072** (0.882) (0.021) Lever –0.836*** –0.836*** –0.073 –0.079 (0.000) (0.000) (0.258) (0.219) Size –0.556*** –0.558*** –0.902*** –0.905*** (0.000) (0.000) (0.000) (0.000) State –0.181*** –0.180*** 0.028 0.026 (0.000) (0.000) (0.473) (0.492) Year Controlled Controlled Controlled Controlled Industry Controlled Controlled Controlled Controlled observations 6370 6370 6391 6391 adj. R-sq 0.382 0.382 0.400 0.400 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control for the year and industry fixed effects. In Table 3, the control variables in formula (2) including Lever, Size and State have a neg- ative coefficient in all columns. The coefficients of control variables are consistent with the previous research (e.g. Wu, 2009; Xia & Fang, 2005). In Table 4, Exemp1 and Exemp2 have negative coefficients in all columns, which also supports H1. In column (1), when it is included along with YearN and the interaction term (Exemp1× YearN), the interaction term has a negative coefficient (significant at the 5% level with a p-statistic of 0.032). In column (2), when it is included along with YearN and the inter- action term (Exemp2× YearN), the interaction term has a negative coefficient (significant at the 5% level with a p-statistic of 0.035). The results indicate that the negative correlation between excess employment and corporate valuation is signic fi antly strengthened with the promulgation and implementation of the Labour Contract Law after 2008, which supports H2. Because the Act has increased labour costs, it has not reduced production efficiency. The results also imply that the mechanism that causes excess employment to reduce firms’ value is increasing labour costs rather than reducing production efficiency. In Table 4, columns (3) and (4) show that the news coefficients are positive, both at the 1% significance level. These results are quite reasonable. Because media coverage can lower information asymmetry, the market would give companies with higher media coverage a higher valuation. In column (3), the interaction term of the (Exemp1× News) coefficient is pos - itive at the 10% significance level. In column (4), the interaction term of the (Exemp2×  News) coefficient is positive at the 5% significance level. These results support H3, which indicates that the negative relationship between excess employment and corporate valuation should weaken with the improvement in media coverage. These results also support and extend the information intermediary theory of the media (Bushee et al., 2010). In Table 5, we provide regression results in sub-samples that are grouped by year. The group after 2008 represents the time in which the Chinese Labour Contract Law enacted in 2007 may have an effect on listed companies. The other group represents the time in which the Act may have no effect on listed companies. From Table 5, comparing the results between these CHInA JoUrnAL oF ACCoUnTIng STUDIeS 49 Table 7. instrumental variable regression results. Variable (1) (2) (3) (4) Exemp1 –0.219* –28.179*** (0.051) (0.001) Exemp2 –1.087** –27.525*** (0.043) (0.003) Lever 0.082 0.083 0.045 0.045 (0.176) (0.174) (0.452) (0.448) Size –0.677*** –0.677*** –0.894*** –0.893*** (0.000) (0.000) (0.000) (0.000) State –0.006 –0.006 0.024 0.024 (0.857) (0.847) (0.462) (0.451) News 0.480*** 0.480*** (0.000) (0.000) Exemp1× News 7.568*** (0.001) Exemp2× News 7.800*** (0.002) Constant 16.789*** 16.791*** 19.552*** 19.561*** (0.000) (0.000) (0.000) (0.000) Industry Controlled Controlled Controlled Controlled observations 10337 10337 10275 10275 adj. R-sq 0.248 0.248 0.285 0.286 *** significant at the level of 1%; ** significant at the level of 5%; * significant at the level of 10%. P-values are reported in parentheses. t he dependent variable is TobinQ. refer to t able 1 for the definition of other variables. We control the year and industry fixed effects. We take Exemp1_lag and Exemp2_lag as the instrument variables of Exemp1 and Exemp2, respectively. We also take (Exemp1× News)_lag and (Exemp2× News)_lag as the instruments variables of (Exemp1× News) and (Exemp2× News), respectively. sub-samples (column (1) vs column (3), column (2) vs column (4)), we find that the negative correlation between excess employment and corporate valuation is significantly strength- ened after 2008. For example, the Exemp1 coefficient is –2.277 at the 0.012 significance level in column (1), while the Exemp1 coefficient is –0.793 at the 0.095 significance level in column (3). For another example, the Exemp2 coefficient is –2.036 at the 0.022 significance level in column (2), while the Exemp2 coefficient is –1.776 at the 0.039 significance level in column (4). In Table 6, we provide regression results in sub-samples that are grouped by media cover- age. The high media coverage sub-sample means media coverage is greater than the median of media coverage, and the other sub-sample means media coverage is less than the median of media coverage. From Table 6, comparing the results between these sub-samples (column (1) vs column (3), column (2) vs column (4)), we find that the negative correlation between excess employment and corporate valuation is significantly weakened with the increase in media coverage. For example, the Exemp1 coefficient is –1.937 at the 0.334 significance level in column (1), while the Exemp1 coefficient is –5.65 at the 0.001 significance level in column (3). As  another  example, the Exemp2 coefficient is 0.303 at the 0.882 significance level in column (2), while the Exemp2 coefficient is –4.072 at the 0.021 significance level in column (4). These results can also support H3 from another angle. 7. Robustness test The following methods are used to carry out the robustness test. First, we adopt different methods to define some variables. For example, we make YearN equal 1 if the sample distribution in 2008, 2009, 2010, 2011, and 2012, and 0 if the sample 50 D. YAng AnD C. ZHAo distribution in 2006 and 2007. r egardless of the methods we adopt, the results are not substantially changed. Second, we group the sample through ownership. We retest H1, H2 and H3, controlling for each sample fixed effects. These three hypotheses are supported both in the state-owned sub-sample and no-state-owned sub-sample. Third, an immediate concern with our analysis is the potential of endogenous relation- ships between excess employment and corporate valuation because these two factors would influence each other. We use instrumental variables to solve the endogenous problem. As instruments for Exemp1 and Exemp2, we use Exemp1_lag and Exemp2_lag, both of which represent lagged variables of Exemp1 and Exemp2, respectively. In the field of accounting and finance, it is sometimes the case that researchers take lagged variables as instrumental vari- ables. For instance, Caramanis and Lennox (2008) study the relationship between the effort of audits and earnings management. There may be endogenous relationships between the effort of audits and earnings management because these two factors would influence each other. To solve the endogenous problem, Caramanis and Lennox (2008) take lagged variables of auditor working hours as instrument variables. Similarly, we take (Exemp1× News)_lag and (Exemp2× News)_lag as the instrumental variables of (Exemp1× News) and (Exemp2× News). In Table 7, we provide instrumental variable regression results. In this table, F statistics of Wu-Hausman are all significant (all significant at the 10% level), which can reject the hypoth - esis that Exemp1, Exemp2, (Exemp1×  News) and (Exemp2×  News) are exogenous variables. Meanwhile, these variables are highly correlated to their instrumental variables, which means there are no weak instruments. The results in Table 7 show that after controlling for other factors and the endogeneity problem, the results show that Exemp1 has a coefficient of –0.219 (significant at the 10% level with a p-statistic of 0.051), and Exemp2 has a coefficient of –1.087 (significant at the 5% level with a p-statistic of 0.043). These results also support H1. In columns (3) and (4), the interaction terms of the (Exemp1×  News) and (Exemp2×  News) coefficients are all positive at the 1% significance level, which supports H3 after controlling for other factors and the endogeneity problem. 8. Conclusion Using the data for A-share listed companies, we find that increasing excess employment will reduce the valuation of firms. Furthermore, with the promulgation and implementation of the Labour Contract Law in 2007 and its enforcement in 2008, the negative correlation between excess employment and corporate valuation is significantly strengthened, which shows that excess employment has a more serious negative economic impact after the prom- ulgation and implementation of the Act. It has been shown in prior research that labour costs increased after the Act. This paper reveals that the mechanism relating excess employment to negative economic consequences is that excess employment leads to increasing labour costs rather than reducing enterprise productivity. Furthermore, this paper presents evidence that media coverage signic fi antly reduces the negative correlation between excess employment and the valuation of firms, which extends the information intermediary theory of the media. CHInA JoUrnAL oF ACCoUnTIng STUDIeS 51 Acknowledgements We thank two anonymous reviewers, Professor Liansheng Wu, Professor Zhengfei Lu and Professor Dong Lu for their helpful comments and suggestions. Disclosure statement no potential conflict of interest was reported by the authors. Funding The authors acknowledge financial support from the national natural Science Foundation of China (Media report, media bias and the governance of financial scandals, no. 71372168). References Bushee, B. J., Core, J. e., guay, W., & Wee, J. (2010). The role of the business press as an information intermediary. Journal of Accounting Research, 48, 1–19. Cai, F. (2010). Demographic transition, demographic dividend, and Lewis turning point in China. Economic Research Journal, (4), 4–13 (In Chinese). Caramanis, C., & Lennox, C. (2008). Audit effort and earnings management. Journal of Accounting and Economics, 45, 116–138. Chen, K. C. W., & Yuan, H. (2004). earnings management and capital resource allocation: evidence from China’s accounting-based regulation of rights issues. The Accounting Review, 79, 645–665. Core, J. e., guay, W., & Larcker, D. F. (2008). The power of the pen and executive compensation. Journal of Financial Economics, 88, 1–25. Ding, S. H. (2010). An analysis of minimum wage effects on the labor market: effect of interaction with the law of the Pr C on employment contracts. Social Sciences in China, (1), 85–102 (In Chinese). DeAngelo, H., DeAngelo, L., & gilson, S. (1996). Perceptions and the politics of finance: junk bonds and the regulatory seizure of first capital life. Journal of Financial Economics, 44, 475–511. Dyck, A., Volchkova, n., & Zingales, L. (2008). The corporate governance role of the media: e vidence from russia. Journal of Finance, 63, 1093–1135. Fama, e. F. (1980). Agency problems and the theory of the firm. The Journal of Political Economy, 88, 288–307. Fama, e. F., & Jensen, M. C. (1983). Separation of ownership and control. Journal of Law and Economics, 26, 301–325. guo, F. (2012). Foreign exchange risk hedging and firm value: e vidence from the MnCs in China. Economic Research Journal, (9), 18–31 (In Chinese). Jensen, M.C. (1979). Toward a theory of the press. Working paper, SSrn. (ssrn.94038) Lei, g. Y., Li, S. F., & Wang, X. J. (2009). Political association, auditors select and corporate value. Management World, (9), 145–155 (In Chinese). Li, P. g., & Shen, Y. F. (2010). The corporate governance role of media: empirical evidence from China. Economic Research Journal, (4), 14–27 (In Chinese). Lin, Y. F., Liu, M. X., & Zhang, Q. (2004). Policy burden and the SBC enterprises: an empirical study from China. Management World, (8), 14–27 (In Chinese). Liu, B., & McConnell, J. (2013). The role of the media in corporate governance: Do the media influence managers’ capital allocation decisions? Journal of Financial Economics, 78, 1–17. Liu, H. L., Zhang, M., Wang, Y. P., & Wu, L. S. (2010). Political connections, compensation incentive, and employee allocation efficiency. Economic Research Journal, (9), 109–121 (In Chinese). Liu, S. J., Sun, P., & Liu, n. Q. (2003). The ultimate ownership and its shareholding structures: Does it matter for corporate performance? Economic Research Journal, (4), 51–62 (In Chinese). Liu, Y. Y., & Liu, B. (2014). employment protection legislation, cost stickiness, and corporate response. Economic Research Journal, (5), 63–76 (In Chinese). 52 D. YAng AnD C. ZHAo Luo, H., Liu, B., & Zhang, W. (2013). The monitoring role of media on executive compensation. China Journal of Accounting Studies, 2, 138–156. Ma, L. F., Wang, Y. F., & Shen, X. X. (2013). Soe party governance, redundant employees and executive compensation contract. Management World, (5), 100–115 (In Chinese). Miller, g. S. (2006). The press as a watchdog for accounting fraud. Journal of Accounting Research, 44, 1001–1033. Shen, Y., Liang, S., & Chen, D. (2013). Wage and accounting conservatism: evidence from China. China Journal of Accounting Studies, 1, 32–46. Shleifer, A., & Vishny, r. (1994). Politicians and firms. The Quarterly Journal of Economics, 109, 995–1125. Wu, L. S. (2009). State ownership, preferential tax and corporate tax burdens. Economic Research Journal, (10), 109–120 (In Chinese). Xia, L. J., & Fang, Y. Q. (2005). g overnment control, institutional environment and firm value: evidence from the Chinese securities market. Economic Research Journal, (5), 40–51 (In Chinese). Xue, Y. K., & Bai, Y. X. (2008). national ownership, redundant employee and company performance. Management World, (10), 96–105 (In Chinese). Yang, D., Lu, Z. F., & Luo, D. L. (2013). Political connections, media monitoring and long-term loans. China Journal of Accounting Research, 7, 165–177. Yang, D. M., & Zhao, C. (2012). Media monitoring, media governance and managers’ compensation. Economic Research Journal, (6), 116–126 (In Chinese). Zeng, Q.S. (2006). State block-holder, over-employment and corporate value. Shanghai University of Finance and economics, Doctoral Thesis (in Chinese). Zeng, Q. S., & Chen, X. Y. (2006). State stockholder, excessive employment and labor cost. Economic Research Journal, (5), 74–86 (In Chinese). Zhang, M., Wang, C. F., & Liu, H. L. (2013). r edundancy burden and Soes’ executive incentive. Journal of Financial Research, (5), 140–151 (In Chinese).

Journal

China Journal of Accounting Studies – Taylor & Francis

Published: Jan 2, 2016

Keywords: Excess employment; firm value; labour contract law; media coverage

There are no references for this article.