Convergence to shareholder primacy corporate governance: evidence from a leximetric analysis of the evolution of corporate governance regulations in 21 countries, 1995-2014

• Pages: 849-883
• DOI: https://doi.org/10.1108/CG-07-2018-0249
• Published date: 07 October 2019
• Date: 07 October 2019
• Author: Navajyoti Samanta
• Subject Matter: Strategy,Corporate governance

Convergence to shareholder primacy

corporate governance: evidence from a

leximetric analysis of the evolution of

corporate governance regulations in 21

countries, 1995-2014

Navajyoti Samanta

Abstract

Purpose –For the past two and half decades, there has been a marked shift in the corporate

governance regulations around the world. The change is more remarkable in developing countries

where countries with little or no corporate governance regime have adopted ‘‘world class’’

standards. While there can be a debate on whether law in books actually translates into law in

action, in the meantime it might be interesting to analyse the law in books to understand how the

corporate governance regime has evolved in the past 20 years. This paper quantitativelytracks 21

countries, most of them being developing and emerging economies, over a period of 20 years. The

period covers 1995 to 2014; thus, it traverses the pre and post crisis period in 1999 and 2008.

Thus, the paper also provides a snapshot of the macrolegal changes that the countries engage in

hoping to stave off the next crisis. The paper uses over 50 parameters modelled on the OECD

Principles of Corporate Governance. The paper confirms the suspicion that corporate governance

norms around the developing economies are converging on shareholder primacy end of the

continuum. The rate of convergence was highest just before the financial crisis of 2008 and has

since then slowed down.

Design/methodology/approach –The paper uses data collected from experts. They filled up

detailed questionnaire which quizzed them on the rules relating to corporategovernance norms in

their country and asked them to retrospectively check thei r data every five years for the past

20 years. This provided an excellent overview as to how the law has evolved in the past two decades

on corporate governance. The data were then tabulated using a scoring shee t and then was put

together using item response theory(IRT) which is a Bayesian method similar to factor analysis. The

paper then follows a comparative approach using heatmaps to analyse th e evolution of corporate

governance in developingcountries.

Findings –Corporate governance norms around the developing economies are converging on

shareholderprimacy end of the continuum. The rate of convergencewas highest just before the financial

crisis of2008 and has since then slowed down.

Originality/value –This is the first time that corporate governance panel data analysis has been

carried out on top developing countries across so many parameters for such a long period.

This paper also uses Bayesian IRT modelling to analyse the evolution which is novel in its

approach especially in the corporate governance literature. The paper thus provides a clear

view on the evolution of corporate governance norms and how they are converging on a

particular ideology.

Keywords Corporate governance, Comparative leximetrics, Corporate governance evolution,

Bayesian inference, International business law

Paper type Research paper

Navajyoti Samanta is based

at Department of Law,

University of Sheffield,

Sheffield, Sheffield, UK.

Received 29 July 2018

Revised 31 January 2019

Accepted 7 February 2019

DOI 10.1108/CG-07-2018-0249 VOL. 19 NO. 5 2019, pp. 849-883, ©Emerald Publishing Limited, ISSN 1472-0701 jCORPORATE GOVERNANCE jPAGE 849

1. Introduction

In comparative law, convergence has been an oft-debated topic[1], particularly in

comparative company law and corporate governance, where one of the major areas of focus

is on the question of whether transplantation leads to convergence[2]. Convergence of

national corporate governance regulations can be functionally att ributed to prolonged

initiatives to unify commercial laws for ease of cross border trade and commerce[3],t ransfer of

‘best practices’ through investment liberalisation as a result of investor pressure[4], spread of

‘neo-liberal pro shareholder value ideologies’ (Soederberg, 2003;Underhill and Zhang, 2008),

and the harmonising role of global financial institutions[5]. There are two main ideological

branches of corporate governance –the shareholder model which can roughly be equated to

a position that companies should be run for the benefit of shareholders who provide risk

capital to companies and so have a claim to the surplus generated, a position traditionally

favoured by “free market advocates/neoclassical economists”, and the stakehold er model

which suggests that companies should be run for the benefit of all those who can affect the

company and can be affected by the company, a position generally associated with ‘left w ing/

interventionist or heterodox economists’.

The rise of modern corporate governance principles coincided with the rise to political

acceptance and apparent success of neo-liberal economic principles during the 1980s, the

fall of the Soviet Union in the early 1990s, and the relative decline of Germanand Japanese

economies in the mid-1990s seemed to provide final proof of the superiority of free market

principles. There followeda period of intense transplantation of legal and quasi-legalnorms,

and future legal historians will look back at this period and observe that, during the 20-year

period from 1995 to 2014, corporatelaw and governance around the world converged more

rapidly than during any other period in history. The only period which even comes close is

the period of imperialism and colonialism, and even then, the transplantation of law was a

relatively slow process. The drivers of this new wave of convergence were not colonial

powers but international financial organisations. One of the major corporate governance

codes available during the late 1990s was the OECD Principles of Corporate Governance

2004, which was based primarily on the shareholder value corporate governance model,

although it also offered a limited accommodation for stakeholder models. International

financial organisations recommended that individual countries should model their corporate

governance structures on OECD principles, so in effect what was being recommended to

developing countries was a shareholder value regime based on the Anglo-Saxon model.

While some scholars on the left would view these organisations as neo-imperialist, this

paper is not a denouncement of any political theory or cause. This paper is limited to

exploring whether the corporate governance regulations around the world, especially in the

emerging economies, are indeed converging on a shareholder primacy model, based on

the OECD Principles of CorporateGovernance. This is the first paper in the literature to use

Bayesian techniques to isolate the quantum and direction of shifts towards a shareholder

primacy model of corporate governance in developing economies. As such, it represents

an important and innovative methodological advance in the quantitative analysis of

corporate governance change over time. The paper also showcases how Bayesian

techniques are better able to isolate the quantum and direction of such changes by

comparing the classicaland Bayesian outcomes.

The research was undertaken in a number of steps. First, a database on the evolution of

corporate governance in 21 countries for 20 years (1995-2014) was created. Local experts

in corporate governance in thosejurisdictions were asked to fill out a detailed questionnaire

based on archival and allied qualitative research. The aim of this phase was to collect data

on fifty-two separate company and corporate governance variables based on the OECD

Principles of Corporate Governance 2004 and previous indices for 20 years (1995-2014).

The variables were scaled polynomially, i.e. the value could be zero, or one, or two which

PAGE 850 jCORPORATE GOVERNANCE jVOL. 19 NO. 5 2019

meant the survey went beyond a simple yes/no response in order to take into account

systems which use optional rules or ‘soft law’.

Second, a graded response model was used with a Kalman filter[6] to create a dynamic

corporate governance index for 21 countries over a 20-year period. This dynamic index

allowed this paper to distribute the changes identified over a period of time rather than

confining them to just one year. It is widely acknowledged that laws and regulations take

some time to show their impact, hence considering development of corporate governance

over a number of years was expected to yield morerealistic results. All previous research in

comparative corporate governance uses Classical test theory (CTT)to build an index; this is

the first time that Bayesian statistics is used for the purpose; an index utilising the Classical

test theories was also created to compare the results with that of Bayesian methods[7].

Bayesian method allows the model to improve the prediction of corporate governance

changes based on the previous year’s corporate governance score for a particular country

and also the corporate governance scores of other countries in that particular year. Thus,

the Bayesian model allows the researcher to incorporate more data in creating the index

than a frequentist system,where the calculation is isolated to one year and one country, and

does not update itself with the changes in other countries and other years. Therefore,

Bayesian modelling is able to better approximate the changes and shifts in law in real life

than frequentist methods.

Finally, to check for convergence to a shareholder primacy corporate governance regime

amongst the country studies, the dynamic corporate governance index was analysed, first

by using various quasi-experimental methods like calculating the average corporate

governance score amongst all countries and then tracking its growth, and second by

assessing the difference between the highest and lowest corporate governance index to

provide an estimate of the extent of differences in the adoption of shareholder value

corporate governance norms among the countries studied. Once the preliminary results

from the quasi experimental methods were obtained, the findings were confirmed by using

experimental methods like coefficient of determination[8], which makes it possible to track

the relative deviation within the corporate governance of the countries studied in this

research. The combination of these three methods was intended to givea robust answer as

to whether corporate governance norms around the developing world are converging on

the shareholder primacy model espoused by the OECD Principles of Corporate

Governance.

This paper is divided into four major parts, in Part II we review the literature on quantitative

comparative corporate governance, investigating the gaps especially in terms of the

methods used; in Part III we discuss the methodology used focussing on the advantagesof

Bayesian techniques over Classical test theories; in Part IV the frequentist and Bayesian

results are contrasted and analysed, showing that a Bayesian approach gives a more

reliable picture of the extent to whichconvergence is occurring.

2. Literature review

While the origins of concern about corporate governance can be traced back to Adam

Smith in the 18

th

century[9], empirical research on corporate governance began in 1932

with the publication of The Modern Corporation and Private Property. In this book, through

quantitative analysis, the authors Adolf Berle and Gardiner Means showed that due to the

wide dispersal of ownership it was possible for a small class of managers, with very little

share ownership, to effectively exercisefull control over very large companies. Though they

did not code for the systems of governance, more importantly they showed that the impact

of corporate governance can be coded from primary effects like board structure and

ownership patterns[10].

VOL. 19 NO. 5 2019 jCORPORATE GOVERNANCE jPAGE 851