Labor hiring and stock return: A model and new evidence from China

https://doi.org/10.1016/j.pacfin.2019.101256Get rights and content

Highlights

  • Hiring is negatively related to the expected stock return from the discount rate channel.

  • The negative relation becomes steeper if the firm's employment frictions are higher.

  • Positive labor supply shock leads to a flatter return-hiring relation.

Abstract

Labor input is an important factor in a firm's production and affects stock return. We use an optimization model to explore the stock return-labor hiring relation with the effects of employment frictions and labor supply. Our model illustrates that labor hiring is negatively related to the expected stock return from the discount rate channel; the negative return-hiring relation becomes steeper when the firm's employment frictions are higher; positive labor supply shock leads to a flatter return-hiring relation. Using Chinese-listed firm data, we provide evidence for the existence of the return-hiring relation and the impact of employment frictions and labor supply confirming the theoretical predictions.

Introduction

Traditional studies of stock returns mainly focus on the effects of a firm's financial variables and relevant systematic risks. These studies usually leave no role for the labor input by assuming that labor can be adjusted without costs. However, the labor market does experience frictions. For example, the labor adjustment costs paid by a firm when adjusting the labor force size reduce the income attributable to investors and thus should be reflected in its asset prices. Danthine and Donaldson (2002) suggest the impact of operating leverage on stock returns due to labor contracts. More recent studies have already looked into the effect of the labor force on financial markets and find some crucial labor features significantly affect asset prices. Donangelo (2014) suggests that labor mobility has a positive effect on a firm's stock returns since it increases the firm's operating leverage. Favilukis and Lin, 2016a, Favilukis and Lin, 2016b report that stock returns increase with wage rigidity because smoother wages make dividends riskier. Donangelo et al. (2019) find the relative size and inflexibility of labor costs result in labor leverage for a firm, and higher labor share increases the risk and equity premium. Labor hiring, as an important labor market characteristic, is closely related to the development of a firm needing competent employees. Thus, some studies investigate the impact of labor hiring. Belo et al. (2014) illustrate the U.S. firm's return-hiring relation with a neoclassical model, and Belo et al. (2017a) extend the study of this relation to the aggregate level. Belo et al. (2017b) suggest the high-skill industry in the U.S. and G7 countries has a steeper return-hiring relation than low-skill industry because workers with high-skill are more costly to be replaced.

However, previous studies on the return-hiring relation have not explored how the labor supply and firm-level employment frictions affect a firm's decision and the return-hiring relation. All else being equal, greater labor supply lowers the average wage and allows the firm to recruit workers more easily; while higher employment frictions incur more labor adjustment costs. Both of the factors may exert significant influence on the firm's hiring behavior. Therefore, this paper establishes an optimization model to study the implications of a firm's hiring for its stock returns, considering the employment friction and labor supply effects. And we use this model to empirically examine Chinese market.

Our theoretical model is constructed on the basis of an infinite horizon through the discount rate channel. We propose that a firm's labor hiring is negatively related to its expected stock returns (Proposition 1). The rationale is that a firm's optimal hiring is determined by the inter-temporal trade-off between the marginal labor adjustment cost when hiring workers at present and the discounted marginal net benefit contributed by these workers in the future. Optimal hiring decreases in accord with the discount rate (expected stock return). By maximizing a firm's current market value, we specify an objective function for a labor-hiring decision, and swiftly solve it with the Lagrange method. We further propose that the negative return-hiring relation is steeper if the firm has higher employment frictions (Proposition 2) and flatter when a positive labor supply shock occurs (Proposition 3). Higher employment frictions lead to greater employment dead-weight costs, and a positive labor supply shock increases the profits by reducing labor expenses as well as search costs. These changes affect the sensitivity of the firm's hiring to the discount rate, thus, the return-hiring relation.

Our empirical tests examine the predictability of a firm's hiring rate to its stock returns. We use China's listed firms on the A-share market of the Shanghai and Shenzhen Stock Exchanges from 2007 to 2015. The firms are sorted into 10 portfolios of equal size based on their hiring rate. The average annualized excess return of the lowest portfolio is about 4.6% higher than that of the highest portfolio, indicating a negative cross-sectional return-hiring relation. We then specify the baseline regression model, in which the lagged hiring variable is included as the explanatory variable and stock returns as the explained variable. The year- and firm-fixed effects are also controlled. The regression results show a significantly negative relation between hiring and future stock returns; specifically, a 10% increase in the firm's hiring rate is related to a 1.24% decrease in its annual stock returns. This finding confirms our first proposition.

Next, we take the proportion of staff with a bachelor degree and higher as the proxy for the degree of employment frictions. The reason is that the firm's greater reliance on high-educated workers usually incurs more costs in labor search, selection, and training. And we incorporate the interaction term between this proxy and labor hiring into the baseline return-hiring equation. We find that the interaction coefficient is significantly negative, confirming Proposition 2, that a firm's negative return-hiring relation becomes steeper if the firm has higher employment frictions.

Then, we examine the impact of labor supply. We perform regressions based on the sample split at year 2011 when China's working age population reached its peak. Intuitively, the decline of working age population after 2011 approximates a negative shock to China's labor supply, which might cause a firm to revise its labor cost policy. Thus, it is reasonable to perform the regressions for the periods before and after 2011. Our results show that the coefficients of the hiring rate in the rising phase of working age population are lower than those in the declining phase. In addition, we add an interaction term between labor hiring and labor supply growth into the baseline equation and find its coefficient is significantly positive. These results are consistent with our hypothesized relation in Proposition 3. We also employ the Fama–French (2015) five-factor model for asset pricing tests, and use alternative proxies and other means to confirm the robustness of our results.

Our study contributes to the literature on the impact of the labor market on equity returns, such as labor mobility (Donangelo, 2014), wage rigidity (Favilukis and Lin, 2016a, Favilukis and Lin, 2016b), labor leverage (Donangelo et al., 2019), labor unions (Addessi and Busato, 2009; Chen et al., 2012) and labor hiring (Belo et al., 2014, Belo et al., 2017a, Belo et al., 2017b). The vast majority of the existing literature provides empirical evidence for developed countries, mainly the U.S., while very little for emerging markets. Relating to labor hiring and stock return, this paper mainly makes two important contributions. First, we extend Belo et al., 2014, Belo et al., 2017a, Belo et al., 2017b by establishing an optimization model to explore the firm-level return-hiring relation, the effects of employment frictions and labor supply from the discount rate channel.

Belo et al., 2014, Belo et al., 2017b analyze the return-hiring relation with a neoclassical model which belongs to the type of model that conducts stock return research in business cycles (e.g., Jermann, 1998; Zhang, 2005; Papanikolaou, 2011). They explain this relation from the aspect of aggregate adjustment cost shock. One pronounced challenge of the business cycle model is to identify the sources of the aggregate shocks and accurately evaluate their impact (Romer, 2012). Greater labor supply can reduce a firm's search costs. Thus, labor supply change is an important source of the labor adjustment cost shock. In the case of a positive labor supply shock, not only the adjustment costs but also wages can be reduced. However, the model of Belo et al., 2014, Belo et al., 2017b fails to capture this shock's overall effects on the whole system. Neither its baseline wage equation nor the alternative specification reflect the reduction of wages from such a shock, underestimating the firm's final profits. Another problem is that the model of Belo et al., 2014, Belo et al., 2017b cannot provide an analytical solution with complex functions and state variables. It adopts a parameter calibration method which yields simulated data for further quantitative comparative statics. However, this simulated data generated from the structural model tends to be more regular than the real data.1 Thus, it may not be able to replicate the stock data of the emerging economies (such as China) due to their more volatile markets, which may weaken its explanatory power to these economies.

Differing from the aggregate shock approach, we use a model of optimal employment to analyze the firm's return-hiring relation and relevant effects through the discount rate channel. Our model overcomes the above problems. Belo et al. (2017a) also explain the return-hiring relation through the discount rate channel, but they proceed to discuss the connection between a firm's size, age, market beta and the return-hiring relation. Their model is also not available analytically, and their subsequent study is on the aggregate level. By comparison, we focus on the firm-level relation both theoretically and empirically. And our model achieves a precise analytical solution. This allows us to explore how the changes in employment frictions and labor supply affect the overall equilibrium outcome, which is not discussed by Belo et al., 2014, Belo et al., 2017a, Belo et al., 2017b. In addition, compared to the single labor market friction in their works, we consider a second friction, namely the labor cost stickiness. It plays an important role in constructing the labor cost equation and deriving the last proposition. and improves the generality of the model to better reflect the imperfect (frictional) labor market. Hence, our paper extends the strand of literature of exploring stock returns from the discount rate channel.

Our discussion of the effect of employment frictions extends the finding of Belo et al. (2017b). They document that the negative return-hiring relation is steeper in the high-skill industry than the low-skill industry. They explain that the high-skill industry has greater labor adjustment costs, generating higher return exposure to the aggregate adjustment cost shock. We suggest that a firm's employment frictions steepen its return-hiring relation. The hypothesis is on the firm-level, and the hiring frictions have a wide range of possible sources, such as the labor union and employment protection, although labor force structure is the main factor. Moreover, we provide a new suggestion that labor supply affects the return-hiring relation. Labor supply is a critical factor to a firm's labor expenses and search costs. Thus, it affects the firm's hiring decision in an imperfect labor market, which might further affect the return-hiring relation. This has not been considered in the previous literature.

Second, our empirical studies are conducted with the data of listed firms in China, providing new evidence from a vital emerging market to contribute to the existing literature. Belo et al., 2014, Belo et al., 2017a, Belo et al., 2017b show empirical evidence from the U.S. and G7 markets. According to our knowledge, this is the first paper that examines the return-hiring relation for an emerging market. In 2015, China had a large employed population of 774.5 million persons. And its mean weekly working hours per employed person were 45.5, and labor force participation rate was 69.9%. By comparison, the working hours per employed person in the U.S. were 38.6 per week and labor force participation rate was 62.7% in 2015. However, the ratio of China's high-quality labor force was still at a very low level. Its percentage of employed population with a bachelor's degree and higher was 8.2% in 2015, while that percentage in the U.S. was 34.13%. Another noticeable attribute of China's labor market was the decline of the working age population (those aged between 15 and 59 years) in recent years. After reaching the peak of 940.7 million persons in 2011, the working age population declined continuously, with a total reduction of 15.2 million persons in 2015. In contrast, the working age population in the U.S. continued to grow from 231.9 million persons in 2007 to 250.8 million in 2015. Thus, the Chinese labor market differed greatly from the U.S. market during our sample period. With the Chinese data, we conduct tests on the return-hiring relation and make a brief comparison with the results of the U.S. Then, we use the Chinese firm's reliance on high-educated employees to test the effect of employment frictions on stock returns. We may be also the first to use the firm-level data to measure labor adjustment costs in the existing literature. Furthermore, the declining trend of the working age population indicates the country's demographic dividend is disappearing. Our empirical tests study the effect of China's labor supply on a firm's stock returns. Therefore, we provide Chinese evidence of the return-hiring relation with a consideration of employment frictions and labor supply to enrich the existing literature.2

The rest of this paper is organized as follows. The next section presents the economic model and provides our propositions. Section 3 describes the sample data. Section 4 shows the results of the empirical studies. Section 5 examines the robustness of our findings, and Section 6 concludes.

Section snippets

Model

This model aims to clarify the argument that labor hiring is negatively related to the expected stock returns, and the relationship becomes steeper when the firm's employment frictions are greater, or flatter if a positive labor supply shock occurs. The typical firm in the model, indexed by i, is assumed to operate infinitely. Its timing is described by Fig. 1.

We intend to find how firm i chooses the optimal hiring rate in each period to maximize its market value, and inspect the expected

Data

To test the propositions, we use data from the Chinese market. The sample contains 2789 Chinese firms listed on the A-share market of the Shanghai and Shenzhen Stock Exchanges from 2007 to 2015 with 15,571 annual observations in total. Following Fama and French (2015), we match the firms' annual accounting data at the end of year t−1 to their annual stock returns from July of year t to June of year t + 1. The 6-month gap ensures that investors obtain the accounting information earlier, and the

Empirical tests

This section examines the predictability of the firm's labor hiring to its future stock returns with the data of China's listed firms, providing empirical evidence for the theoretical propositions.

Robustness check

We conduct a series of checks to test the robustness of our findings. We do not winsorize any variable, and perform the regressions of Eqs. (9), (11). The coefficient of Hiring in Eq. (9) is −0.0446 with its p-value 0.006, and the interaction coefficient in Eq. (11) is 0.0431 with the p-value 0.008. We also winsorize the interaction term of Eq. (10) at 1% of the distribution, but no other variables are so treated. The newly estimated coefficient of the interaction is −0.4358 with a p-value

Conclusion

Labor is a crucial factor of production. This paper shows that the incremental labor of a Chinese firm also has important implications for its asset prices. The economic model reveals that labor hiring is negatively related to the expected stock returns from the discount rate channel. Besides, the negative relation is steeper if the firm has higher employment frictions and flatter when a positive labor supply shock occurs. The logic is that a firm's hiring decisions reflect the trade-off

Acknowledgements

We are thankful for the sponsorship by the Chuang Xin Qiang Xiao (Innovation and Academic Thriving) Project by Department of Education of Guangdong Province in China (Grant No. 2017WQNCX115), the National Natural Science Foundation of China (Grant No. 71803066) and the Planning Project of Philosophy and Social Sciences of Guangzhou (Grant No. 2018GZGJ15). We acknowledge the helpful comments from Hedong Xu of Jinan University, Bing He and Huiying Pu of Guangdong University of Education, and

Author statement

Yuen Rong: Methodology, Software, Validation, Formal analysis, Investigation, Data Curation, Writing – Original Draft, Writing – Review & Editing, Funding acquisition.

Cunzhi Tian: Conceptualization, Formal analysis, Resources, Supervision.

Lifang Li: Methodology, Writing – Original Draft, Writing – Review & Editing, Visualization, Funding acquisition.

Xinwei Zheng: Writing – Original Draft, Writing – Review & Editing, Visualization, Project administration.

References (48)

  • J.G. Williamson

    Growth, distribution and demography: some lessons from history

    Explor. Econ. Hist.

    (1998)
  • K. Woods et al.

    Labor unions and corporate financial leverage: the bargaining device versus crowding-out hypotheses

    J. Financ. Intermed.

    (2019)
  • X. Wu et al.

    On the pricing of the persistence of earnings components in China

    Pac-Basin. Financ. J.

    (2019)
  • S. Abraham

    Can a wrongful discharge statute really benefit employers?

    Ind. Relat.

    (1998)
  • V.V. Acharya et al.

    Wrongful discharge laws and innovation

    Rev. Financ. Stud.

    (2014)
  • T. Aidt et al.

    Unions and collective bargaining: Economic effects in a global environment

  • F. Belo et al.

    Labor hiring, investment, and stock return predictability in the cross section

    J. Polit. Econ.

    (2014)
  • F. Belo et al.

    Labor Hiring and Discount Rates

    (2017)
  • F. Belo et al.

    Labor-force heterogeneity and asset prices: the importance of skilled labor

    Rev. Financ. Stud.

    (2017)
  • G.J. Borjas

    Labor Economics

    (2016)
  • A. Bozkaya et al.

    Labor regulations and European venture capital

    J. Econ. Manage. Strat.

    (2014)
  • F. Cai et al.

    Population change and resulting slowdown in potential GDP growth in China

    China World. Econ.

    (2013)
  • H. Chen et al.

    Labor unions, operating flexibility, and the cost of equity

    J. Financ. Quant. Anal.

    (2011)
  • H. Chen et al.

    Do nonfinancial stakeholders affect the pricing of risky debt? Evidence from unionized workers

    Rev. Finance.

    (2012)
  • Cited by (1)

    • The negative hiring rate premium on stock returns in the Korean stock market

      2022, Pacific Basin Finance Journal
      Citation Excerpt :

      Belo et al. (2017) further studied the effects of heterogeneous labor forces on the relation between hiring rates and future stock returns for G7 countries. Following these studies, Rong et al. (2020) showed that a negative relation between hiring and returns also holds in China's stock market. This study examines the effect of labor on stock returns in the Korean stock market.

    View full text