-
Managerial FinanceAgency costs of stakeholders and capital
structure: international evidenceBing Yu
Article information:To cite this document:Bing Yu,
(2012),"Agency costs of stakeholders and capital structure:
international evidence", ManagerialFinance, Vol. 38 Iss 3 pp. 303 -
324Permanent link to this
document:http://dx.doi.org/10.1108/03074351211201433
Downloaded on: 07 October 2014, At: 17:08 (PT)References: this
document contains references to 54 other documents.To copy this
document: [email protected] fulltext of this
document has been downloaded 2275 times since 2012*
Users who downloaded this article also downloaded:Robert M.
Hull, Carmen Cotei, Joseph Farhat, Benjamin A. Abugri,
(2011),"Testing trade#off and peckingorder models of capital
structure: does legal system matter?", Managerial Finance, Vol. 37
Iss 8 pp.715-735Jean#Laurent Viviani, (2008),"Capital structure
determinants: an empirical study of French companies inthe wine
industry", International Journal of Wine Business Research, Vol. 20
Iss 2 pp. 171-194Anthony Kyereboah#Coleman, (2007),"The impact of
capital structure on the performance of microfinanceinstitutions",
The Journal of Risk Finance, Vol. 8 Iss 1 pp. 56-71
Access to this document was granted through an Emerald
subscription provided by 551360 []
For AuthorsIf you would like to write for this, or any other
Emerald publication, then please use our Emerald forAuthors service
information about how to choose which publication to write for and
submission guidelinesare available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.comEmerald is a global
publisher linking research and practice to the benefit of society.
The companymanages a portfolio of more than 290 journals and over
2,350 books and book series volumes, as well asproviding an
extensive range of online products and additional customer
resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The
organization is a partner of the Committeeon Publication Ethics
(COPE) and also works with Portico and the LOCKSS initiative for
digital archivepreservation.
*Related content and download information correct at time of
download.
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Agency costs of stakeholdersand capital structure:international
evidence
Bing YuSchool of Business, Meredith College, Raleigh, North
Carolina, USA
Abstract
Purpose This paper examines the relationship between bargaining
powers of creditors as well asemployees and financial leverage
across countries. The purpose of this paper is to explore roles
ofcreditors and employees in capital structure decisions under
different legal and political regimesacross countries.
Design/methodology/approach Using country-level creditor rights
index and labor rights indexas a proxy for bargaining powers of
creditors and employees, respectively, the author addresses
theinteraction between creditors as well as employees and
shareholders. The paper tests the impact ofemployee rights and
creditor rights on capital structure across countries.
Findings The author finds a positive relationship between
employee rights and firms use of debtand a negative relationship
between creditor rights and firm debt ratio.
Social implications The paper provides a new perspective to
interpret international variation infinancial leverage in the
world. The results obtained from this paper help us to understand
financialleverage in different countries with various corporate
governance mechanisms.
Originality/value This paper takes all stakeholders into account
when studying agency problems;it explores the role of creditors and
employees in financing decision making under various
corporategovernance patterns and political and legal systems across
countries.
Keywords Corporate governance, Capital structure, Creditors,
Employees, Agency problems,Creditor rights, Labor rights
Paper type Research paper
I. IntroductionA growing interest has been given to the impact
of non-financial stakeholders such ascreditors and employees on
corporate decisions in corporate finance literature. Thispaper
examines relationship between creditors as well as employees and
financialleverage across countries. The purpose is to explore roles
of creditors and employees incapital structure decisions under
different legal and political regimes across countries.
Shareholders, creditors, and employees have heterogeneous
utility functions incorporate context. Tirole (2001, 2006) asserts
that corporations select optimalinvestment and financing decisions
within the constraints of legal and politicalenvironments to which
they belong. Within a company, stakeholders bargain with eachother
to maximize benefits of themselves. The bargaining between
stakeholders is ruledand regulated by a countrys legal and
political regime. While legal and political regimesdiffer across
countries, the bargaining powers of stakeholders are not identical
indifferent countries. Interaction between creditors and
shareholders is mainly throughthe negotiation in debt contracting.
The bargaining power of creditors relies largely
The current issue and full text archive of this journal is
available at
www.emeraldinsight.com/0307-4358.htm
JEL classification G30, G32, G38, K3
Agency costs ofstakeholders
303
Managerial FinanceVol. 38 No. 3, 2012
pp. 303-324q Emerald Group Publishing Limited
0307-4358DOI 10.1108/03074351211201433
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
on creditor rights (CR) provided by a countrys legal system.
Employees, on the otherhand, do not have voting right nor
bargaining power unless they form unions or getprotection from
labor law. Existing literature suggests that shareholders, with
theconstraints of legal regime in a country, will seek a mechanism
within corporations toweaken creditors and employees bargaining
powers so as to maximize payoffs.Financial leverage is a tool that
shareholders can use to achieve this goal. Dronars andDeere (1991)
develop a model to describe the role of debt in limiting
employeesbargaining power when they form unions, while Matsa (2010)
finds that debt ispositively correlated with unionization rates at
firm level for firms in the USA.
This paper focuses on cross-country comparison. Using
country-level creditor rightand labor right indices as proxies for
bargaining powers of creditors and employees,I investigate the
impacts of creditor and employee rights on capital structure
acrosscountries. I argue that when employee rights are high,
employees will have strongerbargaining powers and shareholders are
more likely to be exploited by employees.If so, shareholders intend
to use more debt obligation to remove free cash flows so asto
reduce amount of revenues employees can extract. When CR are high,
creditors havemore negotiation power to obtain good terms in debt
contracting. If shareholderscannot get a favorable debt contract,
they are likely to reduce the use of debt capital.
My study extends the literature by exploring country level
factors influences and bytaking creditors and employees roles into
account when examining capital structuredecisions across countries.
This paper is directly related to the capital structureliterature
that makes cross-country comparison of financial leverage.
Empirical researchon cross-country financial leverage finds a large
variation across countries[1]. Basically,these studies merely
document differences in capital structure in different countries
orcountry groups. They identify how firm-level determinants of
capital structure such asfirm size, profitability, market-to-book
ratio, retained earnings, and growthopportunities affect capital
structure differently across countries and interpretgenerally the
empirical results based on agency problems or signaling
theories,without examining specifically the impacts of creditors
and employees on financialleverage across countries. Treating a
firm as a nexus through which shareholders andmanagers in the
productive enterprise contract with each other, law and
financeapproach represented by a series of papers by La Porta,
Lopez-deoSilanes, Shleifer, andVishny (LLSV hereafter) examines the
relationship between a countrys legal origin aswell as level of
protection for investors and finance. La Porta et al. (1997, 1998)
find thatcommon law countries provide stronger protection for
shareholders than civil lawcountries do and suggest that stronger
investor protection has positive impact offinancial market
development. Numerous studies apply this law and finance
approachand link country-level shareholder rights (SR) to corporate
finance decisions (Rajan andZingales, 1995, Claessens and Laeven,
2003, Hail and Leuz, 2006 and Pinkowitz et al.,2006). While prior
research focuses on SR, this paper extends the literature by
exploringcountry-level creditors and employees roles in capital
structure decisions acrosscountries.
Around the world, countries with different legal and political
systems providedifferent extent of supports for various
stakeholders. Some countries are in favor ofshareholders or
creditors whereas others are in favor of employees (Gourevitchand
Shinn, 2005, Roe, 2004). This variation in legal and political
institutions shapesthe characteristics of bargaining powers of
various stakeholders (Charny, 1999).
MF38,3
304
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Therefore, shareholders efforts to interact with creditors and
employees are constrainedby a countrys institutional conditions.
Since creditor and employees rights granted bylaw and regulatory
regime are exogenous, shareholders will seek reduction ofbargaining
powers of creditors and employees within a corporation. Basically,
whenshareholders use debt obligation to reduce free cash flows,
employees are less likely toobtain explicit or implicit benefits (
Jensen, 1986; Dronars and Deere, 1991). In regard tocreditors,
since stronger CR are in favor of creditors at expense of
shareholders in debtcontracting, shareholders will choose to use
less debt capital so as to mitigate thebargaining power of
creditors.
My paper is also related in general to several studies that test
the stakeholder theoryof capital structure at firm level. Klasa et
al. (2009) and Matsa (2010) analyze thestrategic use of debt
financing by firms in highly unionized industry areas in the USAand
find that those firms use more debt to remove free cash so as to
gain bargainingadvantages over employees and protect firms from
exploit of unions. Myers andSaretto (2009) find that firms increase
leverage in response to the possibility of unionstrikes when
bargaining power of unions is strong. Both Acharya et al. (2011)
and Vig(2011) find that in countries with stronger CR firms have
lower financial leverage. Theyassert that firms are reluctant to
use debt when CR are strong because financialdistress costs are too
high under such a situation.
In line with the above studies, I argue that across countries,
firms in countries withstronger employee rights will use more debt
while firms in countries with stronger CRare likely to use less
financial leverage. Shareholders will use financing
strategydifferently to mitigate bargaining powers of creditors and
employees, restricted byextents of creditor and employee rights
provided by a countrys legal regime. When afirm has less free cash
flows, employees are less likely to obtain extra benefits from
thefirm even the labor law and regulatory regime provide high
employee right in thatcountry. When shareholders intend to use
financial leverage to break employees andmanagers preference for
overexpansion and excessive risk reduction, anotherstakeholder,
creditors, will get involved. Unlike employees whose human capital
is tiedup in the firm and not well diversified, creditors can
diversify their investment well.Thus, within legal framework,
creditors can protect themselves through debtcontracting. Depending
on the creditor right provided by a countrys legal regime,creditors
can negotiate with shareholders in such terms as cost of borrowing,
limitationon dividends payment in some circumstances, and
restriction on excess borrowing inthe presence of high debt
ratio.
This study addresses the following research questions:
RQ1. What is the relationship between country-level employee
rights andcorporations financial leverage across countries?
RQ2. What is the relationship between CR and corporations
financial leverageacross countries?
While exploring the role of creditors and employees in financing
decision making undervarious corporate governance patterns and
political and legal systems across countries,this paper provides a
new perspective to interpret international variation in
financialleverage in the world. The results obtained from this
paper help us to understandfinancial leverage in different
countries with various corporate governance mechanismsand fill
significant gaps in the literature on capital structure.
Agency costs ofstakeholders
305
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
The rest of the paper is organized as follows. A conceptual
framework is discussedand testable hypotheses are developed in
Section I. Section II describes data andresearch methodologies.
Section III discusses empirical results. Section IV concludesthe
study.
II. Conceptual frameworkWhile shareholders can reduce their
investment risk via diversification, employees tietheir human
capital to a corporation. This asymmetric risk reduction
betweenshareholders and employees induces different risk aversion
levels of shareholders andemployees. The contradictory preferences
and pursuits between shareholders andemployees induce employees to
seek for protection for their interests and job securitythrough any
available channels. The most direct way employees use to protect
theirbenefits is labor contracting. However, contracting involves
negotiation and bargain.Unlike shareholders, employees have a lower
bargaining power in contracting processunless they form union to
get collective bargaining power. A union can extract no morethan
the present value of future net cash flows. Dronars and Deere
(1991) state thatfirms can use debt to limit the effect that a
union has on shareholder wealth becausedebt obligation requires
firms to repay a portion of future revenues to creditors, andhence
limit the amount of cash that employees can extract through a
unions strongbargaining power, without driving the firm into
bankruptcy.
Roe (2003) asserts that governments provide protection for
employees through theirlaw regulation in such areas as union
formation, the costs of firing employees, and thedifficulties of
firing employees. When employees obtain more benefits resulting
fromstronger employee protection provided by a countrys labor law
and regulation,shareholders suffer from the increased revenues
extracted by employees due to strongeremployee right. Therefore,
shareholders have incentives to use more debt to divertfuture cash
flows to themselves rather than to employees.
With stronger bargaining powers either through formation of
labor unions or from acountrys legal regime, employees and
creditors will bargain with shareholders topursue their best
payoff. Since employee and CR are granted by a countrys legal
regime,shareholders will choose to lessen employees and creditors
bargaining powers throughfirm-level decisions. Using financial
leverage is an effective way at firm level to mitigatebargaining
powers of creditors and employees.
Based on the above discussion, financial leverage is regarded as
a tool to limitbargaining powers of creditors and employees. The
extent of bargaining powersdepends on the level of employee
protection, SR, and CR provided by a countrys legaland regulatory
regime. Thus, I explore the association between country-level
employeeand CR and financing via testing hypotheses. Using
country-level labor right as aproxy for employee protection (Botero
et al., 2004) and creditor right as proxy forcreditor protection
(LLSV, 2006), I hypothesize:
H1. The stronger the labor right, the more debt the company will
use.
A legal and political system that provides strong employee
protection will emphasizeemployees and managers natural agenda and
demeans shareholders nature agenda.Strong employee protection makes
it hard and costly to lay off employees. Therefore,under such a
system, the pressure on the firm for low risk, unprofitable
expansionis high, and the pressure to avoid risky organizational
change is substantial.
MF38,3
306
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
However, shareholders would prefer to go slow in expanding the
firm, becauseexpansion is harder to reverse later than it would be
in a political environment thatprovides weak employee protection.
To avoid unprofitable expansion and to eliminatethe possibility of
raising employees salaries and benefits, shareholders want to
removefree cash flows from the firm via using more interest-bearing
debts. When a countryslegal regime is in favor of employees, in
order to weaken bargaining power of employees,corporations will
choose to use more debt:
H2. The stronger the creditor right, the less the debt the
company will use.
Creditor can influence a firms financing decision through debt
contracting. The strongerthe creditor right, the more negotiating
power creditors have during contracting process.High creditor right
allows creditors more likely to obtain favorable contracting.To
reduce bargaining power of creditor, shareholders are likely to use
less debt capital.
III. Data and methodologies3.1 Data sources and sample
selectionThe primary data source for the paper is Compustat Global
Vantage. All firm-levelfinancial accounting variable data are
obtained from Global Industrial file. Marketprice data are
collected from Compustat Global Issue file. Country currency
exchangerate data are from Compustat Global Currency File.
Country-level data are collectedfrom various resources.
Country-level variables are obtained from previous research ineach
aspect, respectively. I obtained SR, CR, and (LR) data from Djankov
et al. (2008),Djankov and Shleifer (2007) and Botero et al. (2004),
respectively. I collected macroeconomic data including stock market
capitalization, bond market capitalization,banking segment, GDP
growth rate, inflation rate from IMF and World Bank
annualstatistics. Government quality data are from Kaufmann et al.
(2007). Table I lists dataand variable information.
The sample period is 1990-2008. I begin sample construction by
matchingCompustat Global Industrial with Global Issue and Global
Currency files.
Rajan and Zingales (1995) point out that in any studies that
compare corporationsfinancial data across countries, the
differences in accounting practices cause samplesbias. They notice
that not every country requires firms to report consolidated
balancesheets, and corporations with unconsolidated balance sheets
appear to haveunderestimated financial leverage data than those
with consolidated financialstatements. To avoid this sample
selection bias, I select firms with fully consolidatedaccounting
statements only (consol F in Global Industrial file). Since firms
involved inmajor mergers (cstat AB in Global Industrial file) have
special capital structure(Aivazian et al., 2001), such firms are
dropped. Following literature on capital structure(Rajan and
Zingales, 1995; Aivazian et al., 2001), I exclude financial firms
(6999 . SICcode . 6000), and utility firms (4999 . SIC code .
4900). I also drop firms withnegative equity, negative sales
revenue, missing value of total assets, and negative cashflows.
I match firm-level data from Global Vantage with country-level
data from variousresources and require main three country-level
explanatory variables, SR, CR, and laborrights (LR) indices, be
available to each country included in our sample. To comply withthe
requirements of time-series cross-sectional regression, I drop the
following countrieswith less than 30 firm-year observations, Ghana,
Croatia, Jordan, Kenya, and Romania.
Agency costs ofstakeholders
307
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
After applying these filters, our sample contains 182,182
firm-year observations from21,663 unique firms over 52 countries
during the period of 1990-2008.
I use two country-level variables, CR and LR indices as proxies
for the bargainingpowers of creditors and employees, respectively.
The SR is used as a control variable.
LLSV (1998) develop a SR index. This SR index is widely used in
literature (LLSV,2000; Pinkowitz et al., 2006). Djankov et al.
(2008) update La Porta et al. (1997) SR indexto make it more
accurate. I use the updated anti-self-dealing index from Djankov et
al.(2008) as a proxy for SR.
Similar to SR index, Djankov and Shleifer (2007) use CR index to
measure forcountry-level protection for creditors. The CR index is
an accumulation of four dummyvariables that check:
. whether a country imposes restrictions, such as creditors
consent or minimumdividends to file for reorganization;
. whether secured creditors are able to gain possession of their
security once thereorganization petition has been approved (no
automatic stay);
. secured creditors are ranked first in the distribution of the
proceeds that resultfrom the disposition of the assets of a
bankrupt firm; and
Abbr. Measurement Source
Panel A: firm-level variablesVariable
Debt Debt ratio Long-term debt/total assets Global Industrial
fileMTB Market-to-book ratio (BV of total assets-BV of
equity MV of equity)/total assetsGlobal Industrial andGlobal
Issue
Profit Profitability EBITDA/total assets Global Industrial
fileCash Cash Cash balance/total assets Global Industrial fileSize
Size Log of total assets in US dollars Global Industrial fileTang
Tangibility Tangible assets/total assets Global Industrial
filePanel B: country-level variables
Proxy forSR Shareholder rights Anti-self-dealing index Djankov
et al. (2008)CR Creditor rights Creditor rights index Djankov and
Shleifer
(2007)LR Labor rights Labor union power index Botero et al.
(2004)Stock Market Stock market
developmentStock market capitalization/GDP World Bank report
GOV_QUAL Government quality Government quality index Kaufmann et
al. (2007)OWNER_CON Ownership structure Ownership concentration
index LLSV (1998)BDGDP Bond market
developmentPrivate bond market capitalization/GDP
World Bank report
GDPG Economicdevelopment
Annual GDP growth rate World Bank report
Inflation Inflation Annual inflation rate World Bank reportBKGDP
Banking
developmentDomestic bank deposits/GDP IMF Statistic report
COM Legal origin Dummy variable equals one forcommon law origin
countries and zerootherwise
LLSV (1998)Table I.Data definitions,measurements andsources
MF38,3
308
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
. whether the debtor does not retain the administration of its
property pending theresolution of the reorganization.
Roe (2004) asserts that a marginal increase in benefits of
employees would be amarginal decrease in shareholders value and
that strong LR provided by legal andpolitical systems in fact hurt
a firm value. Therefore, I use measures for LR as a proxyfor
bargaining power of employees.
There is an extensive literature on the relationship between LR
and law andregulation of labor (Besley and Burgess, 2003; Heckman
and Pages-Serra, 2000; Lazear,1990). Those studies check the law
and regulatory provisions on such aspects as thedifficulty of
firing employees, the costs of firing employees, and the easiness
of hiringemployees and explore how employees benefits are affected
due to the differences inthose provisions. With regard to employees
power to pursue maximum benefits,Botero et al. (2004) use the labor
union power index as a proxy for LR. The labor unionpower is an
average of seven dummy variables which equal one:
(1) if employees have the rights to unionize;
(2) if employees have the rights to collective bargaining;
(3) if employees have the legal duty to bargain with unions;
(4) if collective contracts are extend to third parties
bylaw;
(5) if the law allows closed shops;
(6) if workers or unions, or both, have a right to appoint
members to the Boards ofDirectors; and
(7) if workers councils are mandated by law.
3.2 Methodology and research designStudies on financial leverage
based on the trade-off theory and the pecking ordertheory use the
partial adjustment model to explore the optimal debt ratio (Harris
andRaviv, 1990; Myers, 2001) whereas studies addressing agency
problems use debt ratioto regress on firm-level determinants (Myers
and Majluf, 1984). Studies oninternational capital structure test
the different impact of firm-level factors and addcountry-level
variables as explanatory variables. Following Rajan and Zingales
(1995)and Aivazian et al., I use the following model to examine the
impact of creditor andemployee rights on financing decisions across
countries:
Debtt a1 a2MTBt a3Profitt a4CASHt a5Sizet a6Tangt a7SR a8CR a9LR
1t 1
Debtt the long-term debt ratio, computed by long-term debt
divided bytotal assets for firm i at year t (firm subscription is
suppressed inequation (1)).
MTBt the market-to-book ratio, computed by the book value of
total assetsminus the book value of equity plus the market value of
equity all dividedby the book value of total assets for firm i at
year t (firm subscription issuppressed in equation (1)).
Agency costs ofstakeholders
309
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Profit computed by earnings before interest, taxes,
depreciation, andamortization (EBITDA) divided by total assets for
firm i at year t (firmsubscription is suppressed in equation
(1)).
Sizet the log of total assets in US dollars for firm i at year t
(firm subscription issuppressed in equation (1)).
Tangt the tangibility computed by tangible assets divided by
total assets forfirm i at year t (firm subscription is suppressed
in equation (1)).
SR the SR index at country level.
CR the CR index at country level.
LR the LR index at country level.
Rajan and Zingales (1995) point out that to examine the agency
problems associatedwith debt, it is necessary to remove liabilities
like accounts payable that is used fortransactions purpose rather
than for financing purpose. Therefore, long-term debt ratiois a
more reliable measure used to address agency problems. Following
this logic, I uselong-term debt only as the dependent variable.
Frank and Goyal (2005) argue thattheoretically, the book value of
debt is a better measure of creditors liability in case
ofbankruptcy than market value of debt and that market value of
debt has measurementproblems due to the volatility of market price.
Thus, the dependent variable, Debt, iscomputed by book value of
long-term debt divided by book value of total assets foreach firm i
at year t.
As discussed in Section I, this is a research that focuses on
the impact ofcountry-specific characteristics on financing policy,
I use two groups of independentvariables in empirical tests:
firm-level variables and country-level variables. Twocountry-level
variables, the CR and LR indices are major explanatory variables
toaddress research objectives. Firm-level variables are used as
control variables. Thefirm-level variables are selected based on
capital structure theories, following up theliterature on capital
structure.
Based on the capital structure theories, empirical research
tests the impacts ofvarious variables on financial leverage and
interprets test results using one or anothermodel. Chen and Zhao
(2006) find that market-to-book ratio and profit are two
keyfirm-level determinants of capital structure in various
scenarios. Frank and Goyal (2005)examine 39 factors relating to
financial leverage and divide those factors into two tiersbased on
their reliability of relationships with leverage. The top-tier
factors include firmsize, average leverage in an industry, risk,
and market-to-book ratio. To study capitalstructure in the
international context, considering availability of data for
cross-nationalcomparison, Rajan and Zingales (1995) limit their
firm-level control variables to fourfactors: tangibility of assets,
the market-to-book ratio, firm size, and profitability.They argue
that those are factors most consistently correlated with leverage
in theliterature.
Consistent with Rajan and Zingales (1995) and Frank and Goyal
(2005), I choose touse the follow firm-level variables as control
variables: market-to-book ratio, profit,size, and tangibility. The
market-to-book ratio (MTB) is widely used in literature(Rajan and
Zingales, 1995; Aivazian et al., 2001; Chen and Zhao, 2006) to
measure forgrowth opportunities. I use the book value of total
assets minus the book value
MF38,3
310
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
of equity plus the market value of equity all divided by the
book value of total assets tocalculate the market-to-book ratio.
Market value of equity is computed by stock pricemultiplying number
of shares outstanding. Stock price information is collected
fromGlobal Issue file. All stock prices are currency exchange
rate-adjusted. Profit is definedas the ratio of EBITDA to total
assets. Profit is a proxy for internal finance capacity asthe
pecking order model suggests. Size is the log of total assets in US
dollars.Tangibility, Tang, is computed by tangible assets divided
by total assets. Both sizeand tangibility represent for
corporations operating performance. Size is also used as aproxy for
growth.
IV. Empirical results4.1 Summary statisticsI provide sample
description and summary statistics in Tables II and III,
respectively.The sample mean of debt ratio is 12.5 percent and
median is 12 percent. Norway hasthe highest average debt ratio,
23.13 percent, whereas Morocco has the lowest debtratio, 5.31
percent over the sample period. As presented in the following
section, theregression analysis on firm determinants of debt shows
that those firm-level variablesaffect debt ratios across countries
in a similar way, implying that it is
country-specificcharacteristics that cause variations in financial
leverage across countries.
Table IV presents variables that describe country
characteristics. I divide sampleinto two groups: common law and
civil countries. Consistent with literature, commonlaw countries
have better shareholder protection than civil law countries because
SRmean for common law and civil law countries is 0.736 and 0.377,
respectively. The LRmean for common law and civil law countries is
0.261 and 0.340, respectively,indicating that higher employee
rights in civil countries.
4.2 Firm-level determinants of financial leverageI start
analysis by running regression using firm-level variables only. To
address theoutliers issue, I winsorize all firm-level variables at
5 percent level[2]. I run thefixed-effect regression using panel
data as follows (firm subscription suppressed):
Debtt a1 a2MTBt a3Profitt a4CASHt a5Sizet a6Tangt 1t 2The
variables are defined the same as in Section II. To test firm
determinants of debtratio, one needs to adjust to industry effect
either by subtracting industry mean(Chui et al., 2002) or by using
industry dummy variables. Here, instead, I run theregressions using
industry segment data and pooled sample. I run regression
usingsub-samples, dividing sample groups based on industry segments
first (Frank andGoyal, 2005). Then I run the pooled sample using
industry fix effect model. Thesignificance of coefficients remains
consistent, showing that the correlation betweendebt ratio and
firm-level factors is not driven by industry difference. Table V
presentsthe regression results.
As predicted by the agency costs model and the pecking order
model, the empiricalresults are consistent with the literature on
international capital structure comparison(Rajan and Zingales,
1995; Aivazian et al., 2001). There are conflicting
theoreticalpredictions and mixed empirical findings on the effect
of size on leverage. Rajan andZingales (1995) point out that firm
size is usually regarded as a proxy both forinformation asymmetry
and for the probability of bankruptcy. These two proxies
Agency costs ofstakeholders
311
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
CountryNo. of obs No. of obs No. of obs No. of obs No. of
obsPrimary Manufacturing Advanced manufacturing Services Total
Argentina 5 99 74 37 215Australia 3,108 1,068 1,061 3,239
8,476Austria 95 219 269 201 784Belgium 110 305 270 313 998Brazil 37
433 426 257 1,153Canada 1,924 1,170 1,059 3,297 7,450Switzerland 36
437 938 634 2,045Chile 43 293 155 271 762China 241 1,679 2,387
1,790 6,097Colombia 0 44 39 33 116Czech Republic 12 24 16 40
92Germany 176 1,182 2,281 2,367 6,006Denmark 77 389 396 514
1,376Egypt 0 16 22 10 48Spain 186 387 327 424 1,324Finland 39 308
470 379 1,196France 324 1,352 1,783 2,766 6,225UK 1,583 2,855 3,347
9,078 16,863Greece 95 199 182 278 754Hong Kong 57 272 367 800
1,496Hungary 6 60 45 56 167Indonesia 145 810 459 580 1,994India 5
394 337 242 978Ireland 118 168 104 300 690Israel 6 100 93 130
329Italy 141 509 754 568 1,972Japan 2,332 6,054 11,238 14,899
34,523Korea 91 509 762 452 1,814Sri Lanka 0 9 0 31 40Morocco 0 8 19
5 32Mexico 91 229 151 316 787Malaysia 825 1,834 2,005 2,259
6,923The Netherlands 119 497 529 941 2,086Norway 191 237 347 621
1,396New Zealand 26 168 75 481 750Pakistan 20 159 85 29 293Panama 0
2 13 18 33Peru 60 30 41 28 159Philippines 212 247 152 381 992Poland
45 84 92 67 288Portugal 61 143 86 148 438Russian Federation 28 40
21 50 139Singapore 258 564 1,293 1,890 4,005Slovak Republic 13 19 7
0 39Sweden 174 408 824 1,067 2,473Thailand 165 1,034 754 1,087
3,040Turkey 11 89 155 79 334Taiwan 185 967 3,499 907 5,558USA 2,719
8,915 15,265 17,451 44,350
(continued )Table II.Sample description
MF38,3
312
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
imply two inverse effects on leverage. However, coefficients of
size are positivelysignificant. The coefficients of market-to-book
ratios are negatively significant at1 percent level in services
industry segment and pooled sample. While the results intable show
that the firm-level determinants of capital structure across
countries areconsistent, given the variations in capital structure
around the world (Aggarwal, 1990;Aivazian et al., 2001; Gaud et
al., 2007), it is necessary to explore the impact of
countrycharacteristics on capital structure across countries.
4.3 The impact of creditor and employee rights on financing
policyBased on the conceptual framework and hypotheses developed in
Section I, I turn toexplore the relationship between creditor and
employee rights and corporationsfinancing policy across countries.
The analysis is implemented by running the pooledsample ordinary
least square (OLS) regression with year and industry fixed
effects.
Robust clustering standard errors are estimated to control for
interdependenceacross firms. Based on Campbell (1996) and LLSV
(2000), I introduce seven industrygroup dummies in cross-national
regression to control for the industry effects[3]. Thereference
group is the agriculture industry group.
The H1 in Section I predicts the positive sign for LR and the
negative sign for CR.Table VI presents the regression results.
The pooled sample fixed effects regression generates positive LR
coefficients,statistically significant at 1 percent level, and
negative CR coefficients at 1 percentsignificant level. Model (1)
tests the impacts of CR and LR on debt ratio only whereasmodel (2)
adds SR as an additional independent variable. The results are
significantafter controlling for firm-level factors, firm
clustering effects, and the compoundedimpacts of SR, CR, and
employee rights[4].
To address the possible presence of heteroscedasticity and
autocorrelation, I alsoestimate the regression model with the
Newey-West standard error. The results staystatistically
significant.
To address the multicollinearity issue in OLS regression, I use
variance inflationfactor (VIF) and tolerance to diagnose
multicollinearity problem. Wooldridge (2002)defines the VIFs and
tolerance as the following:
VIF bi 1=12 R2i ; andTolerance bi 1=VIF 12 R2i
where bi is the coefficients of model and Ri2 is the unadjusted
R 2.
CountryNo. of obs No. of obs No. of obs No. of obs No. of
obsPrimary Manufacturing Advanced manufacturing Services Total
Venezuela 0 31 32 20 83South Africa 332 241 228 1,168
1,969Zimbabwe 7 7 0 18 32Total 16,534 37,297 55,334 73,017
182,182
Notes: Primary industry: SIC: 0000-1999; manufacturing industry:
SIC: 2000-2999; advancedmanufacturing industry: SIC: 3000-3999
services industry: SIC: 4000-9999 Table II.
Agency costs ofstakeholders
313
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Country MTB Profit Size Tang Debt
Argentina 5.869 0.116 6.817 0.495 0.1279Australia 1.725 0.013
4.249 0.375 0.1117Austria 1.225 0.098 5.689 0.323 0.1172Belgium
1.415 0.118 6.126 0.293 0.1465Brazil 1.128 0.125 6.919 0.458
0.1288Canada 1.696 0.079 5.552 0.465 0.1690Switzerland 1.362 0.110
6.405 0.337 0.1551Chile 1.257 0.112 5.983 0.506 0.1293China 1.420
0.073 5.619 0.391 0.0680Colombia 0.921 0.067 6.694 0.398
0.0769Czech Republic 1.101 0.137 6.695 0.568 0.0794Germany 1.417
0.105 5.731 0.256 0.0956Denmark 1.399 0.104 5.567 0.327 0.1544Egypt
1.948 0.184 6.168 0.489 0.1993Spain 1.438 0.109 6.453 0.370
0.1224Finland 1.339 0.119 6.081 0.321 0.1939France 1.404 0.111
6.005 0.200 0.1294UK 1.687 0.092 4.909 0.320 0.0984Greece 1.747
0.133 5.669 0.363 0.1127Hong Kong 1.229 0.063 5.808 0.335
0.0855Hungary 1.266 0.120 5.538 0.454 0.0869Indonesia 1.291 0.128
4.634 0.415 0.1570India 1.891 0.147 5.516 0.337 0.1895Ireland 1.720
0.083 5.117 0.348 0.1518Israel 2.059 0.095 5.801 0.240 0.1159Italy
1.194 0.087 6.588 0.265 0.1182Japan 1.215 0.060 6.201 0.301
0.1244Korea 1.068 0.099 7.456 0.408 0.1838Sri Lanka 1.043 0.101
4.634 0.447 0.0883Morocco 2.256 0.243 6.599 0.344 0.0531Mexico
1.110 0.126 7.103 0.529 0.1709Malaysia 1.420 0.085 4.712 0.374
0.0799The Netherlands 1.587 0.131 5.917 0.292 0.1237Norway 1.517
0.092 5.455 0.359 0.2313New Zealand 1.538 0.125 4.993 0.440
0.2021Pakistan 1.371 0.170 4.621 0.420 0.0949Panama 1.761 0.108
8.769 0.553 0.2031Peru 0.853 0.168 5.654 0.467 0.0992Philippines
1.107 0.072 4.914 0.407 0.1183Poland 1.447 0.130 5.405 0.419
0.0609Portugal 1.207 0.102 6.021 0.409 0.1780Russia 1.167 0.180
8.320 0.567 0.0968Singapore 1.348 0.078 4.849 0.334 0.0916Slovak
1.010 0.151 6.332 0.550 0.0899Sweden 1.498 0.078 5.726 0.269
0.1426Thailand 1.245 0.110 4.396 0.429 0.1213Turkey 1.896 0.180
6.285 0.335 0.0716Taiwan 1.557 0.092 5.929 0.348 0.1156USA 1.899
0.096 5.849 0.285 0.1641Venezuela 0.831 0.110 6.081 0.505
0.1261
(continued )
Table III.Firm-level variables foranalyses
MF38,3
314
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
It is readily seen that the higher VIF or the lower the
tolerance index, the higher thevariance of bi and the greater the
chance of finding bi insignificant, which means thatsevere
multicollinearity effects are present. Thus, these measures can be
useful inidentifying multicollinearity. Table VII presents the test
result and VIF does not showserious multicollinearity problem.
The regression results reveal a positive relationship between LR
and financialleverage level and a negative relationship between CR
and the usage of debt financing.As discussed in Section I, when
employees get strong protection from high LR, theymore easily
obtain benefits from corporations through union negotiation or
governmentintervention. Such employees benefit gain is at expense
of shareholders. Sinceprotections for employees are exogenous,
shareholders will seek a way within thecorporation to protect them
from exploiting by employees. Using higher financialleverage to
remove the free cash flow is one option shareholders can choose to
achievethis goal. When I add SR index as an additional control
variable, the coefficients of LRstay positively and increase
substantially. They increased from 0.0185 to 0.043, andfrom 0.0193
to 0.0506 in two estimations, respectively. The increased
positivecoefficients of LR in model (2) imply that in a country
where SR are higher, it is morelikely that shareholders will use
high financial leverage to mitigate agency costsof employees if
such agency costs are caused by government law and
regulatoryregimes.
The negative coefficient of CR suggests that CR affect
corporations financingdecisions differently than LR. Unlike
employees, creditors involve in debt contractingdirectly. In a
country where CR are strong, creditors have more power to negotiate
withshareholders and corporations to obtain better terms in debt
contract or can easilyapply restrictions to corporations. Such
restrictions might include the one that limitscorporation to use
excess debt. On the other side, corporations and shareholders
willchoose to use less debt since it is harder to get a favorable
debt contract if CR arestrong. This result also supports the H2,
which says the stronger the CR, the less debtthe firm will use.
4.4 Robust checkRegression analyses that use international
sample are likely to generate biased resultsdue to the sample
selection bias and the model misspecification (omitting
variable)bias. In robust tests, I address the first issue by
running the two-stage residual
Country MTB Profit Size Tang Debt
South Africa 1.516 0.146 5.701 0.376 0.0653Zimbabwe 2.210 0.244
5.069 0.334 0.1080Sample mean 1.664 0.115 5.872 0.387 0.125Sample
median 1.416 0.110 5.804 0.375 0.120
Notes: Sample period is 1990-2008; the dependent variable Debt
is the long-term debt ratio computedby long-term debt divided by
total assets; MTB is the market-to-book ratio computed by the
bookvalue of total assets minus the book value of equity plus the
market value of equity all divided by thebook value of total
assets; Profit is computed by EBITDA divided by total assets; Size
is the log of totalassets in US dollars; Tang is the tangibility
computed by tangible assets divided by total assets Table III.
Agency costs ofstakeholders
315
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
regression (Hoeffler, 2002; Chui et al., 2002) and overcome the
second bias by includingadditional control variables.
The major research objective of this paper is to examine the
impacts of country-levelCR and employee rights on financing across
countries, using firm-level variables ascontrol variables. The
pooled sample regressions have two limitations. First,
runningpooled sample regression cannot totally remove the
disturbance of firm-level variables.Second, including all countries
in the sample results in unequal weights in sample.
Country SR CR LR GOV_QUL ECO_GLB GDPG Inflation Bank
BondStock
Market
Panel B: civil law countriesArgentina 0.34 1 0.3 20.74 3.24
20.284 7.83 0.274 0.047 0.316Austria 0.21 3 0.52 1.53 5.13 1.945
1.5 1.230 0.328 0.196Belgium 0.54 2 0.6 1.32 5.5 1.945 1.58 1.172
0.449 0.570Brazil 0.27 1 0.25 0 3.44 0.87 9.33 0.577 0.087
0.310Switzerland 0.27 1 0.25 1.45 5.16 0.98 0.86 1.716 0.439
1.891Chile 0.63 2 0.12 1.41 4.63 3.779 4.1 0.546 0.159 0.865China
0.76 2 0.14 20.19 3.16 8.156 0.37 0.063 0.315Colombia 0.57 0 0.078
0.1 3.41 1.244 9.12 0.353 0.005 0.178Czech Republic 0.33 3 0.3 0.95
4.41 0.742 2.88 0.589 0.046 0.233Germany 0.28 3 0.38 1.39 4.35
1.698 0.82 1.346 0.461 0.385Denmark 0.46 3 0.8 1.81 4.42 1.618 2.13
0.962 1.099 0.486Egypt 0.2 2 0.27 20.44 3.41 2.74 3.41 0.709
0.300Spain 0.37 2 0.13 1.06 4.81 2.068 3.81 1.172 0.228
0.566Finland 0.46 1 0.84 1.7 5.15 2.424 1.52 0.714 0.284
0.902France 0.38 0 0.09 1.06 4.79 1.728 1.41 1.040 0.450
0.606Greece 0.22 1 0.354 0.79 4.65 1.451 3.45 0.738 0.023
0.389Hungary 0.18 1 0.66 1.1 4.58 1.565 8.69 0.447 0.020
0.192Indonesia 0.65 2 0.012 20.26 3.54 3.853 12.4 0.446 0.014
0.223Italy 0.42 2 0.4 0.84 3.64 1.99 2.48 0.870 0.358 0.340Japan
0.5 2 0.24 1.27 4.16 2.247 21.73 2.070 0.439 0.787Korea 0.47 3
0.138 0.7 3.64 5.763 1.94 0.712 0.465 0.477Morocco 0.56 1 20.15
3.14 1.4 0.87 0.528 0.278Mexico 0.17 0 0.4 0.43 3.55 1.335 9.7
0.314 0.074 0.282The Netherlands 0.2 3 0.28 1.65 5.57 1.726 3.42
1.339 0.416 0.946Norway 0.42 2 0.8 1.34 4.64 2.489 4.86 0.716 0.215
0.378Panama 0.16 4 0.12 0.33 4.35 1.358 0.55 0.710 0.215Peru 0.45 0
0.05 0.11 3.85 20.037 2.36 0.195 0.024 0.240Philippines 0.22 1 0.12
20.06 3.17 0.443 5.59 0.429 0.003 0.491Poland 0.29 1 0.13 0.64 3.67
3.18 3.8 0.322 0.000 0.145Portugal 0.44 1 0.35 1 4.86 2.787 3.7
1.144 0.188 0.312Russia 0.44 2 0.63 20.45 3.07 20.063 31.22 0.220
0.000 0.293Slovak Republic 0.29 2 0.5 1.08 4.22 1.063 5.55 0.565
0.000 0.074Sweden 0.33 1 0.9 1.44 5.05 1.689 1.61 0.721 0.476
0.895Turkey 0.43 2 0.12 0.21 3.75 1.429 45.38 0.289 0.002
0.189Taiwan 0.56 2 0.35 0.94 5.691 21.11 0.218 1.013Venezuela 0.09
3 0.28 21.35 3.13 21.5 26.31 0.144 0.004 0.091Civil law mean 0.377
1.722 0.340 0.667 4.150 1.986 6.159 0.745 0.215 0.455Civil law
median 0.375 2.000 0.280 0.890 4.220 1.694 3.415 0.710 0.159
0.316Sample mean 0.487 1.981 0.315 0.720 4.240 2.176 7.626 0.792
0.211 0.605Sample median 0.440 2.000 0.270 0.925 4.330 1.936 2.945
0.714 0.150 0.400
Table IV.SR, CR, and LR indicesand country-level
controlvariables
MF38,3
316
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Some countries such as the USA, Britain, and Japan have a much
larger number ofobservations than other countries do. Consequently,
the results cannot excludethe excess impact of those big countries.
To overcome such limitations, I use atwo-stage regression model to
remove the firm-level factors and to exclude the sampleselection
bias.
In the first stage, following Chui et al. (2002), I construct an
adjusted dependentvariable by following method. First, debt ratio
for firm i at year t in county j is estimatedby the following
Default (firm and county subscription suppressed):
Debtt a1 a2MTBt a3Profitt a4Casht a5Sizet a6Tangt 1t 3The
dependent and independent variables are defined the same as in
Section II. Then,I use the residual of this equation as the
adjusted debt ratio.
After building the adjusted debt ratio for each firm at each
year, in the second stage,I calculate the mean of adjusted debt
ratio for each country at each year and then usecountry mean of
adjusted debt ratio as dependent variables to run the
cross-nationalregression model:
MeanAdjDebtt bX 1 4X is the vector of country-level
variables.
Debt ratio
Primary ManufacturingAdvanced
manufacturing Services Pooled
MTB 20.0001 20.0000 20.0000 20.0001 * * * 20.0002 * * *
(0.80) (0.44) (0.24) (2.64) (5.07)Profit 20.0947 * * * 20.1774 *
* * 20.1599 * * * 20.0807 * * * 20.0588 * * *
(13.08) (23.84) (30.59) (16.14) (12.14)Size 0.0279 * * * 0.0418
* * * 0.0279 * * * 0.0305 * * * 0.0230 * * *
(26.58) (43.64) (39.35) (51.08) (53.11)Tang 0.0955 * * * 0.1009
* * * 0.1458 * * * 0.1597 * * * 0.1557 * * *
(17.42) (17.75) (28.79) (39.03) (38.16)Constant 20.0601 * * *
20.1241 * * * 20.0685 * * * 20.0739 * * * 20.0699 * *
(10.13) (20.38) (15.40) (20.28) (2.01)No. ofobs 16,472 37,286
55,333 62,090 171,181No. offirms 2,354 4,593 6,943 8,613 22,503Adj.
R 2 0.0704 0.0711 0.0602 0.0709 0.2035
Notes: Significant at: *10, * *5, * * *1 percent; absolute value
of t-statistics in parentheses; the t-statistic reported in
parentheses controls for firm clustering standard errors; this
table presents theregression results of the following Default (with
firm subscripts suppressed):
Debtt a1 a2MTBt a3Profitt a4Casht a5Sizet a6Tangt 1tsample
period is 1990-2008; the dependent variable Debt is the long-term
debt ratio computed by long-term debt divided by total assets; MTB
is the market-to-book ratio computed by the book value of
totalassets minus the book value of equity plus the market value of
equity all divided by the book value oftotal assets; Profit is
computed by EBITDA divided by total assets; Size is the log of
total assets in USdollars; Tang is the tangibility computed by
tangible assets divided by total assets
Table V.Firm and industry factors
and financial leverage
Agency costs ofstakeholders
317
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
The two-stage regression results are presented in Table VIII.
After removing firm-levelfactors totally and controlling for sample
selection bias through two-stage regression,the tests results stay
statistically significant.
To address the omitting variable issue, I run the robust tests
by adding additionalcountry-level controlling variables and re-run
the two-stage regression. Following theprior research, I add both
country-level corporate governance quality variables such
asgovernment quality index and ownership concentration index and
economic variables
Debt ratio(1) (2)
MTB 20.0002 * * * 20.0002 * * *
(6.28) (5.40)Profit 20.0620 * * * 20.0631 * * *
(12.78) (12.96)Size 0.0212 * * * 0.0221 * * *
(49.36) (50.79)Tang 0.1632 * * * 0.1609 * * *
(40.55) (40.06)SR 0.0586 * * *
(14.07)LR 0.0185 * * * 0.0473 * * *
(3.74) (8.90)CR 20.0167 * * * 20.0228 * * *
(24.01) (26.25)Constant 20.0445 20.0804 *
(1.16) (1.94)No. of obs 171,150 171,150Adj. R 2 0.2192
0.2240
Notes: Significant at: *10, * *5, * * *1 percent; robust
t-statistics in parentheses; the t-statistic reportedin parentheses
controls for firm clustering standard errors; this table presents
the regression results ofthe following Default (with firm
subscripts suppressed):
Debtta1a2MTBta3PROFITta4CASHta5SIZEt a6Tangta7SRa8CRa9LR1twhere
model (1) tests the impact of CR and LR on debt ratio and model (2)
test the compounded impact ofSR, creditor right, and labor right on
debt ratio; sample period is 1990-2008; the dependent variable
Debtis the long-term debt ratio computed by long-term debt divided
by total assets; MTB is the market-to-book ratio computed by the
book value of total assets minus the book value of equity plus the
marketvalue of equity all divided by the book value of total
assets; Profit is computed by EBITDA divided bytotal assets; Size
is the log of total assets in US dollars; Tang is the tangibility
computed by tangibleassets divided by total assets; SR and CR are
shareholder rights and creditor rights from Djankov et al.(2008)
and Djankov et al. (2007), respectively. LR is the labor rights
from Botero et al. (2004)
Table VI.Impacts of CR and LR onfinancial leverage
Variable VIF Tolerance R 2
SR 1.59 0.6306 0.3694CR 1.53 0.6515 0.3485LR 1.28 0.7826
0.2174Mean VIF 1.47
Table VII.Variance inflation factors
MF38,3
318
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
such as GDP growth rate, bond market development measure, and
banking sectiondevelopment measure. The regression results are
reported in Table IX.
The above robust tests results show that the coefficients of
major target variables:CR and LR, stay statistically significant.
These significant results support thehypotheses. Specifically, LR
have a positive relationship with debt ratio whereas CRhave a
negative relationship with debt ratio.
V. ConclusionThis paper explores the relationship between CR as
well as employee rights and capitalstructure across countries. The
results reveal the impacts of bargaining powers of creditorsand
employees on capital structure given a countrys legal and political
framework.
MeanAdjDebtCommon law countries Civil law countries Full sample
Full sample
SR 0.0421 * * * 0.0224 * * * 0.0306 * * *
(5.06) (2.83) (3.57)CR 20.0055 * * * 20.0038 * * 20.0055 * *
*
(2.89) (2.12) (2.88)LR 0.0530 * * * 0.0239 * * 0.0324 * * *
0.0321 * * *
(5.96) (2.25) (2.92) (2.92)STKGDP 20.0000 20.0001 * 20.0001
*
(0.99) (1.94) (1.94)GOV_QUAL 0.0162 * * * 0.0136 * * * 0.0126 *
* *
(3.70) (3.13) (2.87)ECO_GLB 0.0002 0.0005 0.0019
(0.06) (0.14) (0.55)Constant 20.0450 * * * 20.0293 * * 20.0474 *
* * 20.0455 * * *
(8.01) (2.27) (3.77) (3.65)No. of obs 830 814 814 814R 2 0.0987
0.1130 0.1144 0.1244
Notes: Significant at: *10, * *5, * * *1 percent; robust
t-statistics in parentheses; the t-statistic reportedin parentheses
controls for county clustering standard errors; this table presents
the regression resultsof the following model:
MeanAdjDebtt bX 1where X is a vector of country-level variables;
STKGDP, the stock market capitalization to GDP, isfrom World Bank;
GOV_QUAL is the regulation quality of government, obtained from
Kaufmann et al.(2007); ECO_GLB is the economic globalization index
from World Bank; the dependent variable,MeanAdjDebt, is the country
mean of residuals of the following model (with firm
subscriptionsuppressed):
Debtt a1 a2MTBt a3Profitt a4Casht a5Sizet a6Tangt 1twhere Debt
is the long-term debt ratio computed by long-term debt divided by
total assets; MTB is themarket-to-book ratio computed by the book
value of total assets minus the book value of equity plusthe market
value of equity all divided by the book value of total assets;
Profit is computed by EBITDAdivided by total assets; Size is the
log of total assets in US dollars; Tang is the tangibility computed
bytangible assets divided by total assets; SR and CR are
shareholder rights and creditor rightsfrom Djankov et al. (2008)
and Djankov and Shleifer (2007), respectively; LR is the labor
rights fromBotero et al. (2004); sample period is 1990-2008
Table VIII.Country-level corporategovernance factors and
debt ratio
Agency costs ofstakeholders
319
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
MeanAdjDebt(1) (2)
GDPG 0.0069 * * * 0.0076 * * *
(3.69) (3.96)Inflation 20.0002
(0.54)STKGDP 20.0098 * * * 20.0088 * *
(2.80) (2.27)Bond 0.0465 * * * 0.0505 * * *
(7.55) (6.90)Bank 0.0190 * * * 0.0234 * * *
(4.10) (5.03)SR 0.0431 * * * 0.0470 * * *
(3.55) (3.86)CR 20.0117 * * * 20.0119 * * *
(4.95) (5.24)LR 0.0375 * * * 0.0452 * * *
(4.14) (4.10)GOV_QUAL 20.0063
(1.05)ECO_GLB 0.0005
(0.14)Constant 20.0631 * * * 20.0714 * * *
(8.25) (4.77)Observations 746 746R 2 0.2504 0.2523
Notes: Significant at: *10, * *5, * * *1 percent; robust
t-statistics in parentheses; the t-statistic reportedin parentheses
controls for county clustering standard errors; this table presents
the regression resultsof the following model:
MeanAdjDebtt bX 1
where X is a vector of country-level variables; GDPG is the GDP
growth rate; Inflation is the inflationrate; Bond is the private
bond capitalization to GDP; Bank is the domestic bank deposits to
GDP;STKGDP, the stock market capitalization to GDP, is from World
Bank; GOV_QUAL is the regulationquality of government, obtained
from Kaufmann et al. (2007); ECO_GLB is the economic
globalizationindex from World Bank; the dependent variable,
MeanAdjDebt, is the country mean of residuals of thefollowing model
(with firm subscription suppressed):
Debtt a1 a2MTBt a3Profitt a4Casht a5Sizet a6Tangt 1twhere Debt
is the long-term debt ratio computed by long-term debt divided by
total assets; MTB is themarket-to-book ratio computed by the book
value of total assets minus the book value of equity plusthe market
value of equity all divided by the book value of total assets;
Profit is computed by EBITDAdivided by total assets; Size is the
log of total assets in US dollars; Tang is the tangibility computed
bytangible assets divided by total assets; SR and CR are
shareholder rights and creditor rights fromDjankov et al. (2008)
and Djankov and Shleifer (2007), respectively; LR is the labor
rights fromBotero et al. (2004); sample period is 1990-2008
Table IX.Country-level economicfactors and debt ratio
MF38,3
320
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
In corporate governance context, stakeholders such as
shareholders, creditors, andemployees have heterogeneous utility
functions. As a result, a game is played amongthose stakeholders
within a countrys legal and political framework. As firm
residualclaimants, shareholders stand on the one side of the game
whereas other stakeholdersstand on the other side. When
stakeholders other than shareholders pursue to maximizetheir
benefits and interests within corporations, their gains are at the
expense ofshareholders. This is the essential of interaction
between stakeholders.
Using country-level CR index and LR index as a proxy for
bargaining powers ofcreditors and employees, respectively, I find a
positive correlation between employeerights and firms use of debt
and a negative correlation between CR and firm debt ratio.This is
because when employee rights are high, shareholders are more likely
to beexploited by employees. If so, shareholders intend to use more
debt obligation toremove free cash flows to reduce employees
opportunities to obtain more benefits fromthe firm. When CR are
high, creditors have more negotiation power to obtain goodterms in
debt contracting, making debt less attractive to shareholders.
The empirical results are robust by controlling for sample
selection bias, test modelspecification, and a series of
country-level control variables. The results obtained fromthis
paper helps us to understand financial leverage in different
countries with variouscorporate governance mechanisms and fills
significant gaps in the literature oninternational financing
policy. These results should be of interest to managers,investors,
and policymakers.
Notes
1. Studies on international capital structure include Aggarwal
(1990), Rajan and Zingales(1995), Aivazian et al. (2001) and Gaud
et al. (2007), among others.
2. I also used 1 percent winsorized sample and original sample
to run all tests. The tests resultsdo not change qualitatively.
3. LLSV (2000) classify non-financial firms into seven broad
industrial groups: (1) agriculture;(2) mining; (3) construction;
(4) light manufacturing; (5) heavy manufacturing;
(6)transportation, communications and utilities; and (7)
services.
4. Since debt ratio is censored by zero at lower bound, we also
use Tobit model to regress debtratio on the same firm-level
independent variables, the SR, CR, and LR indices with year
andindustry fixed effect. The coefficients of CR and LR stay
significant statistically with theexpected sign. Since Tobit model
cannot generate robust standard errors, we report ourresults based
on OLS regression.
References
Acharya, V.V., Amihud, Y. and Litov, L. (2011), Creditor rights
and corporate risk-taking,NYU working paper.
Aggarwal, R. (1990), Capital structure differences among large
Asian companies, ASEANEconomic Bulletin, Vol. 7, pp. 39-53.
Aivazian, V., Booth, L., Demirguc-Kunt, A. and Maksimovic, V.
(2001), Capital structures indeveloping countries, Journal of
Finance, Vol. 56, pp. 87-130.
Besley, T. and Burgess, R. (2003), Can labor regulation hinder
economic performance? Evidencefrom India, The Quarterly Journal of
Economics, Vol. 119, pp. 91-134.
Agency costs ofstakeholders
321
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Botero, J.C., Djankov, S., La Porta, R., Lopez-de-Dilanes, F.
and Shleifer, A. (2004), The regulationof labor, The Quarterly
Journal of Economics, Vol. 114 No. 4, pp. 1339-82.
Campbell, J. (1996), Understanding risk and return, Journal of
Political Economy, Vol. 104,pp. 298-345.
Charny, D. (1999), Workers and Corporate Governance: The Role of
Political Culture, inEmployees and Corporate Governance, Brookings
Institution, Washington, DC.
Chen, L. and Zhao, X. (2006), On the relation between the
market-to-book ratio, growthopportunity, and leverage ratio,
Finance Research Letters, Vol. 3, pp. 253-66.
Chui, A.C.W., Lloyd, A.E. and Kwok, C.C.Y. (2002), The
determination of capital structure: isnational culture a missing
piece to the puzzle?, Journal of International Business
Studies,Vol. 33, pp. 99-127.
Claessens, S. and Laeven, L. (2003), Financial development
property rights, and growth,Journal of Finance, Vol. 58, pp.
2401-36.
Djankov, S. and Shleifer, A. (2007), Private credit in 129
countries, Journal of FinancialEconomics, Vol. 84, pp. 299-329.
Djankov, S., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A.
(2008), The law and economics ofself-dealing, Journal of Financial
Economics, Vol. 88, pp. 430-65.
Dronars, S.G. and Deere, D.R. (1991), The threat of
unionization, the use of debt, and thepreservation of shareholder
wealth, The Quarterly Journal of Economics, Vol. 106,pp.
231-54.
Frank, M.Z. and Goyal, V.K. (2005), Trade-off and Pecking Order
Theories of Debt, Handbook ofEmpirical Finance,
Elsevier/North-Holland, Amsterdam.
Gaud, P., Hoesli, M. and Bender, A. (2007), Debt-equity choice
in Europe, International Reviewof Financial Analysis, Vol. 16, pp.
201-22.
Gourevitch, P.A. and Shinn, J. (2005), Political Power and
Corporate Control, Princeton UniversityPress, Princeton, NJ.
Hail, L. and Leuz, C. (2006), International differences in the
cost of equity capital: do legalinstitutions and securities
regulation matter?, Journal of Accounting Research, Vol. 44,pp.
485-531.
Harris, M. and Raviv, A. (1990), Capital structure and
information role of debt, Journal ofFinance, Vol. 45, pp.
321-49.
Heckman, J. and Pages-Serra, C. (2000), The cost of job security
regulation: evidence from LatinAmerican labor markets, Economia,
Vol. 2, pp. 109-54.
Jensen, M. (1986), Agency cost of free cash flow corporate
finance, and takeovers, AmericanEconomic Review, Vol. 76, pp.
323-9.
Kaufmann, D., Kraay, A. and Mastruzzl, M. (2007), Governance
matters VI: governanceindicators for 1996-2004, World Bank Policy
Research Department working paper.
Klasa, S., Maxwell, W.F. and Ortiz-Molina, H. (2009), The
strategic use of corporate cashholdings in collective bargaining
with labor unions, Journal of Financial Economics,Vol. 92, pp.
421-42.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny,
R.W. (1997), Legal determinants ofexternal finance, Journal of
Finance, Vol. 52, pp. 1131-50.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny,
R.W. (1998), Law and finance,Journal of Political Economy, Vol.
106, pp. 1113-55.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny,
R.W. (2000), Investor protection andcorporate governance, Journal
of Financial Economics, Vol. 58, pp. 3-27.
MF38,3
322
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Lazear, E. (1990), Job security provisions and employment, The
Quarterly Journal ofEconomics, Vol. 105, pp. 699-726.
Matsa, D. (2010), Capital structure as a strategic variable:
evidence from collective bargaining,Journal of Finance, Vol. 65,
pp. 1197-232.
Myers, B.W. and Saretto, A. (2009), Union strikes and the impact
of non-financial stakeholderson capital structure, Purdue
University working paper.
Myers, S.C. (2001), Capital structure, Journal of Economic
Perspectives, Vol. 15, pp. 81-102.
Myers, S.C. and Majluf, N. (1984), Corporate finance and
investment decisions when firmshave information that investors do
not have, Journal of Financial Economics, Vol. 13,pp. 187-221.
Pinkowitz, L., Williamson, R. and Stulz, R. (2006), Does the
contribution of corporate cashholdings and dividends to firm value
depend on governance? A cross-country analysis,Journal of Finance,
Vol. 61, pp. 2725-51.
Rajan, R. and Zingales, L. (1995), What do we know about capital
structure? Some evidence frominternational data, Journal of
Finance, Vol. 50, pp. 1421-60.
Roe, M.J. (2003), Political Determinants of Corporate
Governance, Oxford University Press,Oxford.
Roe, M.J. (2004), Explaining Western securities markets, in
Grandori, A. (Ed.), CorporateGovernance and Firm Ogranization:
Microfoundations and Structure Forms, OxfordUniversity Press,
Oxford.
Tirole, J. (2001), Corporate governance, Econometrica, Vol. 69,
pp. 1-35.
Tirole, J. (2006), The Theory of Corporate Finance, Princeton
University Press, Princeton, NJ.
Vig, V. (2011), Creditor rights and corporate debt structure,
LBS working paper.
Wooldridge, J.M. (2002), Econometric Analysis of Cross-section
and Panel Data, MIT Press,Cambridge.
Further reading
Beck, T., Demirguc-Kunt, A. and Levine, R. (2001), Legal
theories of financial development,Oxford Review of Economic Policy,
Vol. 17, pp. 438-501.
Beck, T., Demirguc-Kunt, A. and Levine, R. (2003), Law,
endowments, and finance, Journal ofFinancial Economics, Vol. 70,
pp. 137-82.
Blair, M. (1999), Firm-specific Human Capital and Theories of
the Firm, in Employees andCorporate Governance, Brookings
Institution, Washington, DC.
Blair, M. and Roe, M. (1999), Employees and Corporate
Governance, Brookings Institution,Washington, DC.
Demirguc-Kunt, A. and Maksimovic, V. (1998), Law, finance, and
firm growth, Journal ofFinance, Vol. 53, pp. 2107-37.
Demirguc-Kunt, A. and Maksimovic, V. (1999), Institutions,
financial markets and firm debtmaturity, Journal of Financial
Economics, Vol. 54, pp. 295-336.
Easterbrook, F. (1984), Two agency cost explanations of
dividends, American EconomicReview, Vol. 74, pp. 605-59.
Grossman, S.J. and Hart, O. (1982), Corporate financial
structure and managerial incentives, inMcCall, J. (Ed.), The
Economics of Information and Uncertainty, University of
ChicagoPress, Chicago, IL.
Agency costs ofstakeholders
323
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
Hansmann, H. and Kraakman, R. (2004), The end of history for
corporate law, in Gordon, J. andRoe, M. (Eds), Convergence and
Persistence in Corporate Governance, CambridgeUniversity Press,
Cambridge.
Jensen, M. and Meckling, W. (1976), Theory of the firm:
managerial behavior agency costs, andownership structure, Journal
of Financial and Quantitative Analysis, Vol. 3, pp. 305-60.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny,
R.W. (1999), Corporate ownershiparound the world, Journal of
Finance, Vol. 54, pp. 471-517.
Pagano, M. and Volpin, P. (2005), The political economy of
corporate governance, AmericanEconomic Review, Vol. 95, pp.
1005-30.
Pagano, M. and Volpin, P. (2006), Shareholder protection, stock
market development, andpolitics, Journal of European Economic
Association, Vol. 4, pp. 315-41.
Rajan, R. and Zingales, L. (2001), Financial systems, industrial
structure, and growth, OxfordReview of Economic Policy, Vol. 17,
pp. 467-82.
Roberts, M. and Sufi, A. (2009), Control rights and capital
structure: an empirical investigation,Journal of Finance, Vol. 64,
pp. 1657-95.
Roe, M.J. (2005), Corporate Governance: Political and Legal
Perspectives, Oxford University Press,Oxford.
About the authorDr Bing Yu is an Assistant Professor of Finance
at the School of Business, Meredith College,Raleigh, North
Carolina, USA. Bing Yu can be contacted at: [email protected]
To purchase reprints of this article please e-mail:
[email protected] visit our web site for further
details: www.emeraldinsight.com/reprints
MF38,3
324
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)
-
This article has been cited by:
1. Ludwig Chincarini. 2013. The Impact of Quantitative Methods
on Hedge Fund Performance. EuropeanFinancial Management n/a-n/a.
[CrossRef]
2. Donald Lien, Ming-Chung Tseng, Soushan Wu. 2013. Foreign
investors in Taiwan: Their roles andgovernment perspectives.
Business Horizons 56:6, 749-756. [CrossRef]
3. Konstantinos Gavriilidis, Vasileios Kallinterakis, Mario
Pedro Leite Ferreira. 2013. Institutional industryherding:
Intentional or spurious?. Journal of International Financial
Markets, Institutions and Money 26,192-214. [CrossRef]
Dow
nloa
ded
by U
NIV
ERSI
TAS
TRIS
AK
TI, M
iss sh
ellv
ida
husn
iyah
At 1
7:08
07
Oct
ober
201
4 (P
T)