اتحادیه های کارگری، قدرت چانه زنی و گسترش بازدهی اوراق قرضه شرکتی: چشم اندازهای مدل اعتبار سازه

This study investigates labor union effects on bond yield spreads from perspectives of structural credit models by employing American bond observations from 2001 to 2007. This research finds that union strength significantly and positively relates to bond yield spreads (this effect is roughly equal to that of issuer rating for one standard deviation change when controlling for well-known variables). The empirical results also show that the positive effects become weaker when management has higher bargaining power. Additionally, union strength volatility significantly and negatively relates to bond yield spreads and capital structure (leverage). The above results are robust when controlling for credit ratings, collinearity concerns, industry effect and tax effect.

Among structural credit model literature, Merton’s (1974) model is the forerunner and has many extensions. Merton’s (1974) model assumes that firm value follows a diffusion process and default occurs when firm value falls below debt face value (i.e., the default threshold) at the debt maturity date. The model estimates a firm’s asset value from its equity market value using the option-based theory, so that the determinants of an option value, such as asset volatility, firm leverage and others, influence corporate bond values.1 The extended structural credit models add other risk variables into Merton’s (1974) framework, such as interest rate (Longstaff and Schwartz, 1995), optimal capital structure (Leland and Toft, 1996), time-varying default threshold (Collin-Dufresne and Goldstein, 2001), incomplete accounting information (Duffie and Lando, 2001), and default event risk (Driessen, 2005).2 It reveals that both the systematic risk and a firm’s idiosyncratic risk affect corporate credit risk. Recent studies document that a firm’s idiosyncratic risk also plays a significant role in explaining yield spreads (Campbell and Taksler, 2003) and results from a firm’s characteristics or real activities.3 However, few existing studies explore the relations between a firm’s idiosyncratic risk and its real activities. The purpose of this study therefore is to fill part of the important gap by investigating the idiosyncratic risk effects resulting from labor unions on corporate credit risk. Most labor union related studies emphasize on exploring how unions affect the wealth of equity holders rather than that of debt holders and are from the perspectives of profitability and strikes (Lewis, 1986 and Becker and Olson, 1986). According to Merton’s (1974) framework, changes in a firm’s profitability and firm value variations caused by unions may also affect the value of debt holders’ claims, and therefore also the firm’s credit risk. However, few existing studies consider the effects of unions on debt holders’ wealth and incorporate them into corporate credit model settings, or investigate their effects on bond yield spreads from perspectives of structural credit models.4 To address this issue, the current study employs American bond market data to examine the effects of unions on bond yield spreads from structural credit model perspectives. This investigation also contributes to the line of literature of exploring the determinants of corporate bond yield spreads, which is an important topic in credit risk management. This study explores the potential effects of unions on a firm’s credit risk from four perspectives: asset returns, asset volatilities, default thresholds and a firm’s cash flows. The first three are the major elements of Merton’s (1974) structural credit risk framework (also called stock-based credit risk) while the last one is the key element of flow-based credit risk (Chen et al., 2011). Lewis (1986) states, from the viewpoint of asset returns, that unions are commonly linked to wage raises, which generates additional costs on employers so that strong unions have negative impacts on a firm’s productivity and profitability (Clark, 1984 and Vender and Gallaway, 2002). In addition, union work rules which aim to protect union members’ jobs may slow down the production process. Card (1996) shows that union workers earn 17% more than non-union workers. Ruback and Zimmerman, 1984 and Bronars and Deere, 1994 find that union-organizing activities significantly lower a firm’s market value (a proxy for a firm’s future profitability) compared to its industry peer’s. Higher wages demanded by union members lead to higher production costs and lower earnings. Addison and Hirsch (1989) also document a negative impact of unions on a firm’s profitability. Becker and Olson (1986) show that firms lose sales and profits during strikes, which cause a decrease in shareholder wealth.5 As a result, the wage and strike effects suggest that firms facing stronger unions would have lower operating cash flows and asset returns.6 Therefore, from the asset returns perspective, unions increase a firm’s credit risk and bond yield spreads. The effects of unions on asset volatility are twofold: generating uncertainty in operating performance and reducing R&D and risky investments. Becker and Olson (1986) show that firms with unions might face the threat of strike if their wage proposals in the collective bargaining cannot meet unions’ demands. As a result, the threat of strike casts uncertainty on firm operating performance and increases volatility of a firm’s asset value distribution.7 On the other hand, beginning with Connolly et al. (1986), many studies support that union firms have lower capital investments and R&D activities. Chen et al. (forthcoming-a) show that firms in unionized industries adopt a less risky investment policy and this action is beneficial for bondholders’ wealth. The R&D intensity effects of a firm on bondholder returns are not definite in the literature. Hence, from the asset volatility perspective, the effects of unions on bondholders’ wealth are indefinite. Existing studies discuss the effects of unions on default threshold from two angles: capital structure and additional priority claims to a firm’s assets. Regarding capital structure, because a higher level of financial liquidity encourages unions to demand higher wages, unionized firms tend to have a higher debt level which reduces a firm’s liquidity and mitigates union’s wage demand. Bronars and Deere (1991) find a positive relationship between industry unionization rate and industry leverage ratio. Matsa (2010) obtains similar results. In addition, Chapter 7 (liquidation) of bankruptcy rules explicitly regulates that labor claims have a higher priority than secured and unsecured creditors’ claims to bankruptcy liquidation proceeds.8 A firm with higher union strength has a higher average wage which results in higher obligatory payments prior to bond holders’ claims and decreases bond holders’ recovery rate (or almost equivalent in effects to an increase in the level of default threshold). The above two viewpoints suggest a positive relation between union strength and a firm’s credit risk (or bond yield spreads). Hilary (2006) finds that management has an incentive to maintain information asymmetry over outsiders when facing organized labor.9 Information asymmetry contributes to an imprecise knowledge of firm value for outside investors. Firms also have an incentive to engage in earnings management when facing collective bargaining. DeAngelo and DeAngelo (1991) find that US major steel companies suppress earnings before contract negotiation to extract union concessions in 1980s. Since debt holders face greater uncertainty on firm’s asset value distribution (including both asset returns and volatility) under a higher degree of information asymmetry or earnings management, a unionized firm’s credit risk will be higher (Duffie and Lando, 2001, Yu, 2005 and Lu et al., 2010). Summarizing the union effects from the above three perspectives of structural credit models, this study preliminarily proposes that unions decrease a firm’s asset returns and increase its default threshold. Union effects on a firm’s asset volatility are indefinite. Within Merton’s (1974) structural credit model framework, a firm’s asset returns negatively relate to corporate credit risk while the firm’s asset volatility and default threshold positively relate to it. Therefore, this study hypothesizes that union strength positively relates to credit risks from the perspectives of asset returns and default threshold, while the relation is uncertain from asset volatility perspective. Besides the above three perspectives of structural (or also called stock-based) credit models, this study also considers the labor union effects from the perspective of flow-based credit risk (the risk of insufficient liquidity for repaying obligations due). Klasa et al. (2009) suggest that firms hold less cash to shelter corporate income and gain bargaining advantages against unions. They also provide empirical evidences of a negative relation between corporate cash holdings and industrial unionization rate. Holding less cash decreases a firm’s internal liquidity and leads to an increase in the firm’s credit risk. As a result, unions also increase a firm’s credit risk from the perspective of flow-based credit risk. Therefore, both perspectives of structural (or stock-based) and flow-based credit risks suggest that unions increase a firm’s credit risk and its bond yield spreads except for the asset volatility viewpoint. Accordingly, this study hypothesizes that the stronger the union, the more pronounced its effects on firms’ credit risk. This study empirically investigates whether or not union strength significantly influences corporate bond yield spreads when controlling for well-known yield spread determinant variables, such as leverage, equity volatility, maturity, coupon, issuance amount, information uncertainty, information asymmetry, and cash flow volatility, by employing a sample consisting of 1607 firm-year data, which accounts for 5446 annual bond observations from 2001 through 2007. Following Hilary (2006), this study uses the product of labor intensity (later denoted as LINT) and industrial unionization rate (later denoted as R) as the proxy for union strength (later denoted as LSTR). The intuition is that unions’ impacts will be greater for firms in a highly unionized industry and with high labor to capital ratio. Empirical results of this study show that union strength significantly and positively relates to bond yield spreads. The yield spreads increase 21.43 bps per standard deviation increase in union strength. When controlling for credit ratings, leverage ratio and other well-known variables, the influence of union strength on yield spreads is roughly equal to that of issuer (firm) credit rating and approximately a quarter of that of leverage ratio (the yield spreads increase 11.40 bps, 13.21 bps and 45.97 bps per standard deviation increase in union strength, firm credit rating, and leverage ratio, respectively). The results of this research also show that the positive effects of union strength on a firm’s credit risk become weaker when a firm has higher bargaining power (as measured by leverage ratio, cash flow volatility, cash holdings and degree of information asymmetry). Furthermore, union effects are significant when additionally controlling for well-known idiosyncratic risk variables related to incomplete information stated in Duffie and Lando (2001) and others. This study also illustrates that union strength volatility (LSTR_Vol) significantly and negatively relates to bond yield spreads and capital structure (leverage ratio). The results are robust when controlling for credit ratings, collinearity concerns, industry effect and tax effect. Therefore, this study supports that the idiosyncratic risk effects resulting from unions have a substantial impact on corporate credit risks. The remainder of this paper is organized as follows. Section 2 presents the theories and hypotheses. Section 3 summarizes major variables used in the empirical examinations. Section 4 presents and analyzes empirical results. Finally, Section 5 provides concluding remarks.

Few existing studies investigate the labor union effects on bond yield spreads from perspectives of structural credit models. The current study fills part of the important gap by employing 5446 American annual bond observations from 2001 through 2007. Empirical results of this study show that union strength significantly and positively relates to bond yield spreads when controlling for spread determinant variables well known in the literature, such as leverage, equity volatility, maturity, coupon, issuance amount, information uncertainty, information asymmetry and cash flow volatility. This is consistent with the predictions based on the perspectives of structural (or stock-based) and flow-based credit models except that of asset volatility. The results of this research also demonstrate that the positive effects of union strength on bond yield spreads become weaker when a firm has higher bargaining power. A firm’s idiosyncratic risk effects of unions on bond yield spreads also exist in the presence of incomplete information variables mentioned in Duffie and Lando (2001) and others. This study also finds that union strength volatility significantly and negatively relates to bond yield spreads, leverage ratio and current union strength. Finally, union strength helps traditional structural credit risk models explain the bond yield spreads from the perspective of idiosyncratic risk effects. The results are robust when controlling for credit ratings, collinearity concerns, industry effect and tax effect.