Patterns of Compliance with the German Corporate Governance Code

• DOI: http://doi.org/10.1111/j.1467-8683.2008.00696.x
• Author: Axel V. Werder,Till Talaulicar
• Date: 01 July 2008
• Published date: 01 July 2008

Patterns of Compliance with the German

Corporate Governance Code

Till Talaulicar* and Axel v. Werder

ABSTRACT

Manuscript Type: Empirical

Research Question/Issue: This study investigates whether the form of compliance with the recommendations of the

German Corporate Governance Code (GCGC) appears to be idiosyncratic to a specif‌ic company or features similarities

across f‌irms. The major aim of this researchis thus to explore the ability of a classif‌ication of compliance patterns to account

for the similarities and differences between f‌irms regarding their conformity with the GCGC.

Research Findings/Insights: Based on seven dimensions of code compliance, cluster analysis is used to identify discrete

groups of companies with similar patterns of code observance. We determine eight patterns of compliance which are

characterized by distinct forms of code conformity.

Theoretical Academic Implications: The identif‌ied cluster solution does not merely ref‌lect the number of rejected code

recommendations. Rather, companies with very similar rates of overall compliance with the GCGC are assigned to different

clusters because they feature, at the same time, different patterns of code conformity. These f‌indings imply that governance

prediction and governance performance studies have to overcome overly aggregated measures of code compliance which

only incorporate the number of rejected code recommendations.

Practitioner/Policy Implications: This study provides evidence to practitioners and policy makers that f‌irms can be

classif‌ied regarding their compliance with the code recommendations. In-depth analyses of the identif‌ied patterns of

compliance furthermore reveal that some patterns may indicate less well substantiated deviations from the code and partly

even decouplings of the declared compliance practices.

Keywords: Corporate Governance Code, Germany, Disclosure, Two-Tiered Board, Board of Directors Issues

INTRODUCTION

Many countries have issued codes of good corporate

governance. These codes contain principles and pro-

visions on the system and practices by which companies are

directed and controlled and aim to further enhance gover-

nance quality. The code precepts are not stipulated by law.

Rather, they can be characterized as a form of soft law

(Shelton, 2000) and are thus not legally binding. In order to

enhance the rate of code compliance, many codes are asso-

ciated with an enforcement mechanism that allows compa-

nies to deviate from the code norms, but at the same time

requires them to disclose these deviations in a statement

of conformity (“comply or explain”). On the one hand, the

requirement to disclose a statement of conformity facilitates

code compliance, because corporate constituencies become

aware of code deviations and can call for justif‌ications, or

even modif‌ications, of the chosen governancearrangements.

On the other hand, the comply-or-explain mechanism sug-

gests that code norms, which go beyond the law, may not

offer the single best solution under all discernable condi-

tions. Rather, the code enables companies to ref‌lect sector

and enterprise-specif‌ic requirements and to reject code

recommendations if deemed necessary. By this means,

self-regulatory codes contribute to more f‌lexibility in the

corporate constitution.

Empirical studies have tried to reveal whether higher

rates of code conformity are associated with superior f‌inan-

cial performance or lower rates with inferior f‌inancial per-

formance. However, the corresponding f‌indings are still

premature. Regarding the German Corporate Governance

Code (GCGC), for instance, Goncharov, Werner, and Zim-

mermann (2006) have shown that there is a pricing effect

associated with higher degrees of declared code observance

*Address for correspondence: Technische Universität Berlin, Department of Organi-

zation and General Management,H73, Str. Des 17. Juni 135, D-10623 Berlin, Germany.

Tel: 49-30-314-25256; Fax: 49-30-314-21609; E-mail: T.Talaulicar@ww.tu-berlin.

de

PATTERNS OF COMPLIANCE WITH THE GERMAN CORPORATE GOVERNANCE CODE 255

Volume 16 Number 4 July 2008

© 2008 TheAuthors

Journal compilation © 2008 BlackwellPublishing Ltd

doi:10.1111/j.1467-8683.2008.00696.x

and that compliance with the code is hence value-relevant

after controlling for an endogeneity bias. In contrast,

Nowak, Rott, and Mahr (2005) and Bassen, Kleinschmidt,

Prigge, and Zöllner (2006) did f‌ind signif‌icant relations

between compliance with the GCGC and different measures

of corporate f‌inancial performance. Insignif‌icant, or incon-

sistent, performance effects of overall code compliance have

also been observed concerning the governance guidelines

by the Cadbury Code in the UK (Weir and Laing, 2000;

Dedman, 2002), the Peters Committee in the Netherlands

(de Jong, DeJong, Mertens, and Wasley, 2005), the Spanish

Olivencia Code (Fernández-Rodríguez, Gómez-Ansón, and

Cuervo-García, 2004), or the Portuguese Securities Market

Commission (Alves and Mendes, 2004).

The inconsistent results on the effects of overall code

compliance on corporate f‌inancial performance come with

little surprise. Corporate performance is a non-uniformly

def‌ined, multifaceted construct, inf‌luenced by many deter-

minants that interact in a complex and dynamic manner

(e.g., Capon, Farley, and Hulbert, 1987; March and Sutton,

1997; Daily, McDougall, Covin, and Dalton, 2002). Regard-

ing the performance implications of code compliance, the

alleged performance effects will not simply depend on the

pure number of rejected code norms as previous perfor-

mance studies suggest that only incorporate the degree of

overall code compliance. Instead it will make a difference

how seriously the code is implemented and whether or not

the latitude, or discretion, left by the code norms is utilized

appropriately (cf. Samuels, Greenf‌ield, and Piper, 1996;

Burton, 2000; v. Werder and Talaulicar, 2003). Irrespective of

such still scarce governance practice studies, the comply-

or-explain mechanism moreover suggests that deviance

from specif‌ic code norms might be favorable under specif‌ic

circumstances of a particular company. Lower degrees of

overall code compliance are hence not necessarily inferior.

Finally, the degree of overall code compliance may be less

inf‌luential than the issue that specif‌ic code norms are either

applied or rejected (cf. Bebchuk, Cohen, and Ferrell, 2005;

Brown and Caylor, 2006; Bassen, Prigge, and Zöllner, 2007).

Against this background, a closer look at the declared forms

of code compliance is warranted.The same degree of overall

compliance with a specif‌ic code of corporate governance

may be accomplished by various patterns of code compli-

ance. Some patterns may differ signif‌icantly on the kind of

rejected code norms whereas the pure number of rejected

code norms is rather similar. More in-depth insights into the

concrete form of code conformity are therefore essential to

better predict why some f‌irms feature specif‌ic governance

arrangements (governance prediction studies) as well as to

analyze the consequences of these governance arrangements

(governance performance studies).1

In this paper we investigate whether the form of compli-

ance with the GCGC appears to be idiosyncratic to a specif‌ic

company or features similarities across f‌irms. The present

study thus scrutinizes whether or not homogenous patterns

of compliance with the GCGC can be identif‌ied. Previous

research has shown that overall compliance with the GCGC

tends to be high. Nonetheless, there remain code recommen-

dations that are rejected by a signif‌icant share of companies.

For reasons outlined below, we utilize these critical recom-

mendations of the GCGC to create meaningful dimensions

of code conformity. These compliance measures are sub-

sequently subjected to a cluster analysis that captures the

network of interactions among these dimensions and iden-

tif‌ies patterns of compliance with the GCGC. The eight

clusters or patterns identif‌ied are analyzed in more detail.

Our study shows that code compliance can take different

forms. We suggest that these differences, their causes, and

consequences have to be taken into consideration and offer

intriguing avenues for future research.

BACKGROUND AND RESEARCH

QUESTION

Background of the Code

In contrast to other countries, a code of corporate governance

for German f‌irms was regarded as unnecessary for a long

time, because essential governance aspects (e.g., the separa-

tion of managementand supervision) were mandatory under

German law (for details on the German governance environ-

ment and its more recent developments, see v. Werder and

Talaulicar, 2006a). However, after some private initiatives

issued drafts for a voluntary corporate governance code

that competed more than converged, the Federal Ministry of

Justice appointed a government commission to develop a

uniform code for German listed companiesin order to further

strengthen the governance quality and to make German cor-

porate governance rules transparent for both national and

international investors in a consolidated way.

The GCGC was adopted on February 26, 2002. Besides

presenting essential statutory regulations, the GCGC con-

tains standards for good and responsible governance (i.e.,

recommendations and suggestions) that are not stipulated

by law, but represent internationally and nationally recog-

nized best practice. German companies are not compelled to

comply with these standards. However, after Article 161 of

the Stock Corporation Act was amended, listed corporations

have to disclose whether they have been and are being com-

plied with the code’s recommendations or which of the rec-

ommendations are not being applied (“comply or explain”).

Three kinds of code provisions have to be distinguished

based on the level of obligatory compliance. First, the GCGC

contains precepts that f‌irms are compelled to observe under

applicable law (“must provisions”). The remaining catego-

ries (i.e., “shall recommendations” and “should or can sug-

gestions”) both consist of rules that are not required by

law. As a consequence, companies can deviate from these

norms. However, deviations from recommendations that

are marked in the text by use of the word “shall” must be

disclosed in the annual statement of conformity. These rec-

ommendations are the main subject of our study. Third, the

GCGC contains suggestions that are marked in the text

by the use of the words “should” or “can” and which can

be deviated from without disclosure. These suggestions

are intended to inspire advancements without inhibitory

requirements.

Acceptance of the Code

Regarding the acceptance of the GCGC, v. Werder and

Talaulicar (2006b) show that 1) the GCGC altogether f‌inds

256 CORPORATE GOVERNANCE

Volume 16 Number 4 July 2008 © 2008 TheAuthors

Journal compilation © 2008 BlackwellPublishing Ltd