Employee treatment and its implications for bondholders

• Published date: 01 September 2019
• Date: 01 September 2019
• Author: Tsung‐Kang Chen,Hsiao‐Lin Yang,Yan‐Shing Chen
• DOI: http://doi.org/10.1111/eufm.12196

DOI: 10.1111/eufm.12196

ORIGINAL ARTICLE

Employee treatment and its implications

for bondholders

Tsung-Kang Chen

1

|

Yan-Shing Chen

2

|

Hsiao-Lin Yang

3

1

Department of Management Science,

College of Management, National Chiao

Tung University, Hsinchu, Taiwan

Email: vocterchen@nctu.edu.tw

2

Department of Finance, College of

Management, National Taiwan

University, Taipei, Taiwan

Email: yanshing@ntu.edu.tw

3

Department of Finance, College of

Commerce, National Chengchi

University, Taipei, Taiwan

Email: 101357501@nccu.edu.tw

Funding information

Excellent Research Projects of National

Taiwan University and National Science

Council of Taiwan (NSC 101-2410-H-224-

012-)

Abstract

We examine the various channels through which the quality

of a firm's employee relations can affect the welfare of

bondholders. Our evidence suggests that better employee

treatment benefits bondholders and leads to a lower bond

spread by enhancing a firm's productivity, and by reducing

the likelihood of product failures, labor strife, and employee

turnover. However, a higher level of satisfaction is costlier

for bondholders in firms facing more severe financial

constraints or agency problems.

KEYWORDS

bond yield spreads, cost of debt, employee treatment

JEL CLASSIFICATION

G12, J53

1

|

INTRODUCTION

As quality and innovation become more crucial than quantity, human capital has become a key

competitive asset for modern corporations. Research suggests that firms with good human resources

management practices have higher productivity, profitability, and shareholder value (Edmans, 2011;

Filbeck & Preece, 2003; Huselid, 1995; Ichniowski & Shaw, 1999; Ichniowski, Shaw, & Prennushi,

1997; Kempf & Osthoff, 2007). Huselid (1995) finds firms which adopt the High Performance Work

The authors gratefully acknowledge the constructive suggestions and comments from John Doukas (the Editor), two

anonymous referees, Chuan-Yang Hwang, and seminar participants at the 2015 International Conference of Taiwan Finance

Association. Yan-Shing Chen acknowledges the financial support from Excellent Research Projects of National Taiwan

University and National Science Council of Taiwan (NSC 101-2410-H-224-012-). All remaining errors are those of the

authors.

Eur Financ Manag. 2018;1–33. wileyonlinelibrary.com/journal/eufm © 2018 John Wiley & Sons, Ltd.

|

1

1047Eur Financ Manag. 2019;25:1047–1079. wileyonlinelibrary.com/journal/eufm © 2018 John Wiley & Sons, Ltd.

Practices systems show lower turnover rates and higher productivity and profitability. Ichniowski et al.

(1997) find that the adoption of new work practices, including work teams, flexible job assignments,

employment security, training in multiple jobs, and extensive reliance on incentive pay, produces

substantially higher levels of productivity. Ichniowski and Shaw (1999) find that the Japanese

production lines, which exhibit superior human resource management practices, are more productive

than US lines. In particular, Edmans (2011) shows that a portfolio consisting of ‘100 Best Companies to

Work for in America’ earns an abnormal annual return of 3.5%. Bondholders and rating agencies also

pay attention to human resources and employee relations management issues because of their influence

on the borrower's credit risk. For example, the labor strife in the mid-1990s in a Bridgestone/Firestone

plant was responsible for the production of defective tires (Krueger & Mas, 2004). The company was

forced to recall an estimated 6.5 million tires in 2000 and was subsequently downgraded by rating

agencies. This case shows that good employee relations may benefit bondholders by lowering the risks

of costly product flaws and labor strife. However, the expenses associated with better employee

benefits may also constrain a firm's investments and weaken its ability to meet its debt obligations.

Thus, whether bondholders appreciate better employee treatment is an empirical question.

This study provides evidence on the channels through which better employee treatment may

positively or negatively affect bondholders’ welfare. The clearest beneficial effect of better employee

treatment for bondholders, as for shareholders, is that it induces greater worker effort (Akerlof, 1982;

Akerlof & Yellen, 1986) and thus enhances a firm's productivity and operating cash flows.Akerlof

(1982) suggests workers increase effort in response to higher compensation as a gift from employers.

The efficient wage theory proposed by Akerlof and Yellen (1986) suggests workers receiving excess

compensation tend to increase effort to avoid being fired from such jobs. Chen, Liao, and Chen (2014)

show that better production efficiency results in lower yield spreads. In addition, better employee

treatment may also benefit bondholders by enhancing a firm's production efficiency and reducing cash-

flow volatility through the following three channels. First, employee satisfaction reduces the likelihood

of product failure. Economic models of fairness (e.g., Fehr & Gachter, 2000; Rabin, 1993) indicate

that, when being treated well and in good faith, workers are more cooperative and less prone to commit

sabotage. Conversely, employees who feel their expectations are ignored show less attention to product

quality. Tense employee relations may increase product failure rates (Krueger & Mas, 2004; Turnley &

Feldman, 1999).

Second, better employee treatment may also alleviate the risks of labor strife or costly employee

litigation. Firms with improper working environments or infringement on labor rights are more likely

to face employee litigation and incur significant damage, legal, and reputational costs (Bradford, 2005;

Karpoff & Lott, 1999).During 1990–2006, employment-related complaints accounted for roughly 50%

of all civil rights complaints in the United States. Third, better employee treatment facilitates the

retention and the recruitment of skilled employees in a knowledge-based industry. Maslow (1943) and

Hertzberg (1959) argue that satisfaction, which cannot be purchased with cash and can only be

provided by firms, is an efficient form of compensation. Mobley, Griffeth, Hand, and Meglino (1979)

and Muchinsky and Turtle (1979) show that dissatisfied employees are more likely to quit their jobs

than are their satisfied colleagues. Thus, better employee treatment reduces the risk of losing key

employees to other firms and is aligned with debtholders’ interests.

However, bondholders may not appreciate better employee treatment in some circumstances since

good relations with employees are not free. Higher salaries, better working environment, the provision

of profit-sharing plans, pension plans, and health and benefit plans all add to the firm's financial

burdens and may weaken a firm's ability to meet its debt obligations (Carroll & Njehaus, 1998; Lee,

2008; Maher, 1987).When a firm goes bankrupt, US bankruptcy laws give labor claims a higher

priority than creditor claims.

2

|

CHEN

ET AL.

e

t

e

a

p

c

u

a

a

B

p

a

y

t

r

(

i

g

e

a

b

d

p

s

p

s

s

e

m

T

f

e

o

c

w

e

a

i

c

b

t

t

C

C

2

|

CHEN

ET AL.

1048

.

,

s

e

e

o

o

e

e

s

g

s

e

.

y

e

;

f

.

s

)

e

-

d

e

t

t

&

e

y

;

%

e

d

e

)

s

y

e

n

l

,

r

L

.

To investigate the various effects of employee treatment on the welfare of bondholders, this paper

employs a corporate bond sample in the United States between 2003 and 2014. Using information from

the MSCI ESG STATS database (formerly known as KLD STATS), we develop three measures of

employee treatment: (i) satisfaction index (hereafter SAT); (ii) dissatisfaction index (hereafter DISAT);

and (iii) net satisfaction index (hereafter NSAT), which equals SAT minus DISAT.STATS provides

positive (negative) indicators for identifying a firm's strengths (concerns) in employee relations, which

covers a subset of publicly traded companies in the United States since 1991. STATS ratings are widely

used in previous studies to construct indexes of employee well-being or satisfaction. See, e.g., Turban

and Greening (1997), Landier, Nair, and Wulf (2009), Verwijmeren and Derwall (2010), Bae, Kang,

and Wang (2011), Ghaly, Dang, and Stathopoulos (2015), Guo, Huang, Zhang, and Zhou (2016), and

Ben-Nasr and Ghouma (2018). We find that yield spreads are negatively related to NSAT and SAT, and

positively related to DISAT when controlling for a number of bond and firm characteristics, and firm-

and year-fixed effects. In particular, a one standard deviation increase in NSAT leads to a decrease in

yield spreads by 10.99 basis points (bps), which is equivalent to a reduction of 5.48% when evaluated at

the median of yield spreads.

We continue to find a negative (positive) effect of NSAT (DISAT) on yield spreads in the following

robustness tests: (i) using alternative specifications to control for the industry effects more explicitly;

(ii) using alternative methods to aggregate the information from STATS and construct the three

indexes; and (iii) applying two-stage least squares (2SLS) regressions in which the average of

geographical neighboring firms’ employee treatment is used as the instrumental variable. However, the

effect of SAT on spreads becomes statistically insignificant in some of these alternative settings.

We provide further evidence on the effects of productivity enhancement and risk reduction

associated with better employee treatment. First, we find a more pronounced negative relationship

between NSAT and yield spreads in labor-intensive firms, in which labor plays an important role in the

determination of product quality, and firms operating only in a single product market, in which a

product failure is more likely to affect the borrower's ability to meet debt obligations. These results

suggest that the spread-reduction effect of better employee treatment is partly due to the reduction in

product failures. Second, we find a more pronounced positive relationship between DISAT and yield

spreads in firms primarily operating in industries with higher unionization rates and firms experiencing

strikes in recent years, suggesting that the reduction in costly labor disputes and employee litigation can

explain the beneficial effect of better employee treatment.Powerful unions tend to use strikes to express

members’ dissatisfaction and use litigation to strengthen their bargaining position (Garden, 2013;

Tracy, 1986; Vroman, 1989). Third, we find that SAT is more negatively related to yield spreads for

firms in high-tech industries and for firms with more patent citations, suggesting that the beneficial

effect of better employee treatment is related to the better retention of skilled employees.

More directly, we show that firms with a higher NSAT have higher productivity, a lower likelihood

of experiencing product recalls, lost working days due to labor disputes, and employee-related

controversies, and a lower employee turnover after controlling for firm characteristics. Interestingly,

while a lower DISAT contributes to all these beneficial effects, a higher SAT is only related to a lower

employee turnover. This is consistent with our previous findings that the spread-reduction effect

associated with a decrease in employee dissatisfaction is more evident than that associated with an

increase in satisfaction.

We further show several cases in which an improvement in employee satisfaction may not be

consistent with bondholders’ interests. In particular, we find that the spread-reduction effect of SAT

becomes weaker: (i) in financially constrained firms, in which the employee-related expenses are likely

to constrain a firm's investment in profitable projects, and thus weaken the borrower's ability to repay

their debt (e.g., Campbell, Dhaliwal, & Schwartz, 2011; Rauh, 2006); and (ii) in firms with weaker

CHEN ET AL.

|

3CHEN ET AL.

|

3

L

.1049