Firing Costs and Capital Structure Decisions

• Date: 01 October 2016
• DOI: http://doi.org/10.1111/jofi.12403
• Author: MATTHEW SERFLING
• Published date: 01 October 2016

THE JOURNAL OF FINANCE •VOL. LXXI, NO. 5 •OCTOBER 2016

Firing Costs and Capital Structure Decisions

MATTHEW SERFLING∗

ABSTRACT

I exploit the adoption of state-level labor protection laws as an exogenous increase

in employee firing costs to examine how the costs associated with discharging work-

ers affect capital structure decisions. I find that firms reduce debt ratios following

the adoption of these laws, with this result stronger for firms that experience larger

increases in firing costs. I also document that, following the adoption of these laws,

a firm’s degree of operating leverage rises, earnings variability increases, and em-

ployment becomes more rigid. Overall, these results are consistent with higher firing

costs crowding out financial leverage via increasing financial distress costs.

HOW DO THE COSTS ASSOCIATED with discharging workers (i.e., “firing costs”) af-

fect capital structure decisions? Prior work finds that firms with higher finan-

cial leverage ratios and those in technical default are more likely to discharge

workers (e.g., Sharpe (1994), Hanka (1998), Falato and Liang (2015)).1Yet,dis-

charging workers can involve hidden costs, such as those arising from collective

bargaining agreements, discrimination claims, and wrongful termination law-

suits. These employee firing costs create frictions that constrain a firm’s ability

to discharge workers. While it is well recognized in economics that firing costs

can have a significant effect on real outcomes such as wages, employment, and

investment, it is less clear how these costs relate to financing decisions.

From a theoretical perspective, the effect of firing costs on capital structure

decisions is ambiguous. On the one hand, an increase in firing costs could lead

∗Matthew Serfling is from the Department of Finance, University of Tennessee.I thank Michael

Roberts (Editor) and three anonymous referees for their helpful comments and guidance. I am

also grateful for the helpful suggestions from my dissertation committee members, Sandy Klasa

(chair), Kathleen Kahle, Hern´

an Ortiz-Molina, Ryan Williams, and Tiemen Woutersen. I would

also like to thank Douglas Fairhurst, Hyunseob Kim (discussant), Svetlana Orlova (discussant),

Sarah Shaikh, Shweta Srinivasan, conference participants at the 2013 Financial Management

Association (FMA) annual meeting and the 2014 American Finance Association (AFA) annual

meeting, and seminar participants at the University of Arizona, Georgia Institute of Technology,

Georgia State University, the University of Oregon, and the University of Tennessee for helpful

comments. An earlier version of this paper was circulated under the title “Labor Adjustment Costs

and Capital Structure Decisions.” The author has no relevant or material financial interests that

relate to the research described in this paper.

1Workforcereductions are an especially important cost management tool during economic down-

turns. For instance, a significant fraction of firms (67% of surveyed U.S. firms and 38% of French,

German, and Italian firms, which face substantially more rigid labor markets) cut employment dur-

ing the recent financial crisis (see “Cost Management: An Aspect of Profit & Cash Optimization,”

Accenture, January 2, 2011).

DOI: 10.1111/jofi.12403

2239

2240 The Journal of Finance R

to higher debt ratios by lowering employees’ risk of unemployment. Titman

(1984) and Agrawal and Matsa (2013) suggest that employees demand a wage

premium for bearing the increased risk of unemployment that arises from

using financial leverage. Thus, if higher firing costs reduce an employee’s risk

of dismissal, this wage premium shrinks and firms could increase debt ratios

so as to capture a larger share of the tax benefits of debt.

On the other hand, higher firing costs could lower optimal debt ratios by

increasing financial distress costs. First, because distressed firms are often

forced to discharge workers to cover cash flow shortfalls (e.g., Ofek (1993),

Kang and Shivdasani (1997)), the additional firing costs incurred as a result of

discharging workers increase the total costs of distress. Second, higher firing

costs make it more difficult to reduce employment when firms need to do so,

such as during economic downturns (e.g., Bentolila and Bertola (1990), Autor,

Donohue, and Schwab (2006), Messina and Vallanti (2007)). This effect can

make labor costs more fixed in nature, which can raise operating leverage and

increase a firm’s risk of becoming distressed (e.g., Mandelker and Rhee (1984),

Mauer and Triantis (1994), Kahl, Lunn, and Nilsson (2014)).

To test these competing predictions, one needs to overcome the endogeneity

problem associated with the fact that a firm’s firing costs are determined in

part by its financial leverage. In particular, firing costs are a function of the

expected costs from firing employees combined with the probability of firing

such employees, which increases with a firm’s debt ratio. A second concern is

that firing costs are unobservable. However, even if they were observable, it is

likely impossible to control for all factors that affect both leverage and firing

costs. In this paper, I attempt to identify the causal effect of firing costs on

capital structures by exploiting the quasi-natural experiment created by the

adoption of Wrongful Discharge Laws (WDLs) by U.S. state courts over the

period 1967 to 1995.

WDLs matured into three common law exceptions to “at-will employment”

in an effort to protect against wrongful termination.2I focus on the effect of the

adoption of one particular WDL—the good faith exception. In its broadest sense,

this exception applies when a court determines that an employer discharged

a worker out of bad faith, malice, or retaliation. In these cases, employees can

recover contractual losses and punitive damages. This exception represents

the largest deviation from at-will employment and is arguably the most far

reaching of the three (e.g., Dertouzos and Karoly (1992), Kugler and Saint-

Paul (2004)).

Importantly, WDLs increase firing costs. Jung (1997) finds that plaintiffs

won $1.29 million on average in 1996, and Boxold (2008) documents average

(maximum) awards of $0.59 million ($5.4 million) over the 2001 to 2007 period.

While these average awards are arguably small for large firms, firms can be

2At-will employment refers to a legal environment in which employers are free to terminate

any employee for good reason, bad reason, or no reason at all, with or without prior notice, and

without risk of legal liability. I discuss the three exceptions to at-will employment in more detail

in Section II.A.

Firing Costs and Capital Structure Decisions 2241

subject to several lawsuits at any point in time, which can substantially raise

their legal liability.3Also, the fear of very large settlements could alter the

behavior of risk-averse managers (Dertouzos, Holland, and Ebener (1988)),

as 46% of surveyed public firms’ managers express concerns regarding losses

arising from such lawsuits.4Consistent with these laws increasing firing costs,

prior research finds that employment levels, employment volatility, and firm

entry decrease following the adoption of the good faith exception (Dertouzos and

Karoly (1992), Autor, Kerr, and Kugler (2007)). I also find that firms experience

cumulative abnormal stock returns of −1.05% to −1.22% and a reduction in

the market value of equity of $4.3 to $5.0 million when their state adopts this

law, suggesting that the law’s adoption is not only costly but also partially

unanticipated.

To test for the effect on capital structure decisions, I use a difference-in-

differences approach in which the treatment and control groups consist of firms

headquartered in states that have and have not adopted the good faith excep-

tion, respectively. The regressions control for firm and year fixed effects and

firm characteristics known to affect capital structure decisions. I also control

for several state-level variables to help ensure that economic and political con-

ditions do not spuriously drive the results. I find that, following the adoption

of the good faith exception, book and market leverage ratios decrease by 1.5

and 1.0 percentage points, respectively. Compared to their respective sample

means, book and market leverage decrease by 6.1% and 3.6%. These results

continue to hold when I use alternative measures of leverage, such as leverage

net of cash holdings.

The key identifying assumption central to a causal interpretation of the

results is that, in the absence of the treatment, the average change in debt

ratios would have been the same for both treatment and control firms. Several

features of WDLs and results from a variety of robustness tests suggest that

this parallel trends assumption is satisfied. First, because WDLs are based in

common law, a judge’s decision to adopt the good faith exception is more likely

due to the merits of the case than political and economic factors (e.g., Walsh and

Schwarz (1996)). In line with this view, I find that out of a large set of economic

and political variables that potentially affect whether courts adopt the good

faith exception, almost none of them are correlated with the law’s adoption.

Second, due to the staggered adoption of the good faith exception, firms can be

3For example, Lawrence National Security laid off 430 employees in 2008 as the firm dealt

with the national recession and a budget deficit. However, 130 of these employees sued the firm,

claiming that the budget deficit was a “pretext to get rid of older employees who have higher

salaries, larger medical costs, and are closer to collecting their pension.” On May 10, 2013, a jury

sided with five employees who were selected to be test cases for the other 125 employees, awarding

the plaintiffs a total of $2.7 million for breach of contract and breach of the implied covenant of good

faith and fair dealing. See “Jury Awards $2.7 M against Lawrence Livermore Lab for Wrongfully

Terminating Long-Time Employees in 2008,” PRNewswire, May 13, 2013 and Andrews et al. v.

Lawrence Livermore National Security, LLC., Case No. RG09453596.

4See “U.S. Public Companies’ Perceptions of Risk, and Their Risk Mitiga-

tion Strategies,” Chubb 2012 Public Company Risk Survey, 2012. Available at

http://www.chabb.com/business/csi/chubb15930.pdf.