National Governance Bundles and Corporate Agency Costs: A Cross‐Country Analysis

• Author: Praveen Kumar,Hadiye Aslan
• Published date: 01 May 2014
• Date: 01 May 2014
• DOI: http://doi.org/10.1111/corg.12055

National Governance Bundles and Corporate

Agency Costs: A Cross-Country Analysis

Hadiye Aslan and Praveen Kumar*

Manuscript Type: Conceptual and empirical

Research Question/Issue: Comparative corporate governance performance across countries cannot be captured adequately

by ranking of single or a few national governance factors (NGFs). But it is also intractable to undertake separate comparative

analyses of the large number of NGFs that potentially impact corporate governance mechanisms. How can NGFs be

combined into national governance bundles (NGBs) that are informative of the level of agency costs at the f‌irm level and are

also convenient to use in comparative analyses?

Research Findings/Insights: We theoretically motivate and empirically identify the components of two NGBs that, at the

f‌irm level, have especially signif‌icant impact on controlling shareholder agencycosts and the agency costs of debt f‌inancing,

respectively. These NGBs (1) allow consolidation of the effects of various individual NGFs that have been considered

separately in the literature, (2) reveal the complementarity and substitutabilityeffects among NGFs and between NGFs and

f‌irm governance mechanisms, and (3) are convenient to apply through aggregate bundle scores.

Theoretical/Academic Implications: Forming NGBs by combining NGFs that have especially signif‌icant impact on specif‌ic

agency costs is useful both theoretically and empirically. The theoretical analysisguides the identif‌ication of NGFs that may

be especially inf‌luential on specif‌ic agency costs and these predictions can be verif‌ied empirically. Examining the effects of

NGFs on the relation of f‌irm-level governance mechanisms and agency costs is useful in understanding the links between

national- and f‌irm-level governance variables.

Practitioner/Policy Implications: The parsimonyand tractability afforded by the NGBs in understanding various aspects of

corporate governanceperformance should prove useful to policymakers, international business managers,and investors. By

providing readily available indicators of different agency costs at the f‌irm level, these bundles can guide policymakers in

setting the legal and regulatory framework for corporate governance at the national level.

Keywords: Corporate Governance, National Governance Bundles, Agency Costs, Ownership Structures, Cost of Debt

INTRODUCTION

The corporate governance literature emphasizes the

impact of selected national governance factors (NGFs)

on corporate governance mechanisms. For example, the lit-

erature compares common law origins with other legal tra-

ditions (La Porta, Lopez-de-Silanes, Shleifer, & Vishny,

1998), or “Anglo-American” with “continental European”

and “community-oriented” systems (Hall & Soskice, 2001;

Orrú, Biggart, & Hamilton, 1997), or “stakeholder” with

“insider-outsider” systems (Aguilera & Jackson, 2003), or

“market-oriented” with “bank-based” economies (Allen,

1993; Demirguc-Kunt & Levine, 2001). However, the inf‌lu-

ence of these NGFs on f‌irms also depends on other NGFs,

such as the strength of f‌inancial disclosure and fraud deter-

rence regulations (La Porta, Lopez-de-Silanes, & Shleifer,

2006; Leuz, Nanda, & Wysocki, 2003; Morck, Yeung, & Yu,

2000); the eff‌iciency of contract enforcement (Djankov, Hart,

McLiesh, & Shleifer, 2008); and the political environment, in

particular the presence of press freedom and democratic

accountability (Judge, Douglas, & Kutan, 2009; Roe, 2002). It

is, therefore, not surprising that there is signif‌icant cross-

national diversity in the impact of individual NGFs (such as

legal origin) on agency costs at the f‌irm level (Aguilera,

Desender, & Castro, 2011; Aguilera & Williams, 2009). For

instance, some countries may have signif‌icant agency costs

of dominant shareholder control because of weak minority

shareholder investor rights, but simultaneously have rela-

tively low agency costs of asymmetric information because

of strong f‌inancial disclosure laws, or vice versa.1

Clearly, comparative corporate governance performance

across countries cannot be captured adequately by ranking

of only a single – or even a few – NGFs. But it is also

intractable to undertake separate cross-national comparisons

*Address for correspondence: Praveen Kumar, C.T. Bauer College of Business, Uni-

versity of Houston, 334 Melcher Hall, Houston, TX 77204, USA. Tel:832-452-5625; Fax:

713-743-4789; E-mail: pkumar@uh.edu

This article was published online on 10 February 2014. An error was subsequently

identif‌ied in PanelA of Table 2. This notice is included in the online version to indicate

that this has been corrected 11 February 2014.

230

Corporate Governance: An International Review, 2014, 22(3): 230–251

© 2014 John Wiley & Sons Ltd

doi:10.1111/corg.12055

of the strengths of the large number of individual NGFs that

potentially impact corporate governance mechanisms. In

this paper, we demonstrate that individual NGFs can be

meaningfully combined into national governance bundles

(NGBs) based on their signif‌icant inf‌luence on specif‌ic

agency costs. By focusing on agency costs we are able to

theoretically motivate and empirically identify the composi-

tion of NGBs. The theoretical motivation for our approach is

that indvidualNGFs have heterogeneous effects on different

agency costs at the level of the f‌irm. For example, strong

regulations on f‌inancial disclosures and fraud deterrence are

known to be especially inf‌luential in reducing the agency

costs of external f‌inancing due to asymmetric information

(Morck et al., 2000); however, these NGFs may have rela-

tively marginal inf‌luence on governance problems stem-

ming from internal stakeholder conf‌licts between managers

and workers (Aguilera & Jackson, 2003), especially in f‌inan-

cially transparent f‌irms.

Using a structural modeling approach, we develop theo-

retically and validate empirically two NGBs that especially

impact the agency costs of dominant shareholder control

and the agency costs of debt. Dominant shareholder moral

hazard has received much attention in the literature (Aslan

& Kumar, 2012; Claessens, Djankov, Fan, & Lang, 2002; La

Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2002; Morck,

2000). Meanwhile, debt is the principal source of external

f‌inancing for f‌irms in many economies (Drucker & Puri,

2006). Agency costs of debt, which ref‌lect the higher borrow-

ing costs for f‌irms due to agency risks to lenders from stra-

tegic behavior of insiders, have been widely examined in the

literature (Jensen & Meckling, 1976; Smith & Warner, 1979).

We argue that NGFs related to the strength of creditor rights

and eff‌iciency of enforcement of debt contracts are especially

inf‌luential with respect to the agency costs of dominant

shareholder control. (We call this the “creditors’ rights and

debt enforcement” [CRDE] bundle.) Similarly, NGFs related

to the strength of accounting and f‌inancial information dis-

closure regulations can be usefully combined in a NGB for

the agency costs of debt. (We call this the “corporate infor-

mational quality” [CIQ] bundle.)

We empirically identify the effects of NGFs in the CRDE

and CIQ bundles using a cross-national sample of 22 major

European and Asian countries that cover a wide swath of

geographically and institutionally diverse national gover-

nance systems. We employ a unique hand-collected dataset

of international syndicated bank loans and dynamic owner-

ship structures that includes a large number of f‌irms and

loans. Our ownership structure data are of special interest

because, unlike earlier studies, we trace the dynamics of

control and ownership rights of f‌irms, including unlisted

companies that have direct and indirect ownership of listed

corporations.2Using a variety of estimation methodologies

that control for endogeneity and the presence of latent

common factors at the f‌irm level, we f‌ind that strong NGFs

in the CRDE and CIQ bundles have statistically and eco-

nomically signif‌icant negativeeffects on the dominant share-

holders’ ownership choice and the f‌irms’ cost of debt. But

the NGFs in the CRDE bundle appear to have greaterimpact

on dominant shareholders’ ownership structure choice com-

pared with their effects on the cost of debt,while the reverse

is true for most of the NGFs in the CIQ bundle.

To our knowledge, we provide the most comprehensive

analysis of the direct and indirect effects of NGFs on domi-

nant shareholders’ ownership structure choices and f‌irms’

cost of debt by consolidating (in the NGBs) various NGFs

that have hitherto typically been considered separately in the

literature. At the national level, countries in our sample

exhibit varying degrees of strength in the different NGFs

that comprise the NGBs, i.e., they do not tend to be uni-

formly superior or weaker in all the NGFs. We f‌ind that

CRDE and CIQ bundle scores (calculated by combining the

component NGF performance scores) allow a tractable and

meaningful comparative analysis across countries of agency

costs of controlling shareholder power and debt f‌inancing.

These results – at the national and f‌irm levels – empirically

validate the conceptual basis of our bundling approach.

Our approach also highlights the links between country-

level and f‌irm-level governance variables and, therefore,

helps bridge the so-called micro-macro gap in corporate

governance studies (Bamberger,2008; Dalton & Dalton,2011;

Minichilli, Zattoni, Nielsen, & Huse, 2012) and, in particular,

is consistent with the interest in developing further the

interface between national- and f‌irm-level governance vari-

ables (Aguilera, Filatotchev, Gospel, & Jackson, 2008; Kumar

& Zattoni, 2013; Zattoni & Judge, 2012). We show that

examining the effects of NGFs on the relation of f‌irm-level

governance mechanisms and agency costs is useful in

understanding the links between national- and f‌irm-level

governance variables.

Overall, our study helps advance both the theory and

practice of corporate governance by (1) providing scholars

with a theoretical framework to bundle NGFs and method-

ologies to empirically validate the composition of the NGBs

and (2) developing two NGBs (CIQ and CRDE) that expand

the tools available for comparative corporate governance

analyses not only for scholars but also for practitioners and

policymakers. Rather than organizing countries with respect

to single NGFs, such as legal origin or index of creditor rights

that mayprovide only a limited perspective, analysts can use

the NGB scores to evaluate cross-country performance with

respect to controlling shareholder agency costs and/or

agency costs of external f‌inancing. This type of detail is

important because our analysis shows that countries thatare

strong with respect to NGFs that moderate controlling

shareholder power are not necessarily strong in NGFs that

moderate the agency costs of external f‌inancing. Our frame-

work and empirical identif‌ication methodologies can also be

used to develop other NGBs.

Weorganize the rest of the article as follows. In the second

section, we describe the theoretical framework and specify

the empirical hypotheses. The third section describes the

data and the empirical measures, while the fourth section

lays out the empirical test methodology. The f‌ifth section

describes the results. The sixth section concludes by discuss-

ing various implications of our approach.

THEORY AND

HYPOTHESIS DEVELOPMENT

In this section, we develop a theoretical framework to moti-

vate the NGBs related to the agency costs of dominant

NATIONAL GOVERNANCE BUNDLES AND AGENCY COSTS 231

Volume 22 Number 3 May 2014© 2014 John Wiley & Sons Ltd