Family Control and Expropriation at Not‐for‐Profit Organizations: Evidence from Korean Private Universities

• DOI: http://doi.org/10.1111/j.1467-8683.2012.00922.x
• Author: Woochan Kim,Seung‐Bo Kim,Kee‐Hong Bae
• Date: 01 July 2012
• Published date: 01 July 2012

Family Control and Expropriation at

Not-for-Prof‌it Organizations: Evidence

from Korean Private Universities

Kee-Hong Bae, Seung-Bo Kim, and Woochan Kim*

ABSTRACT

Manuscript Type: Empirical

Research Question/Issue: We study an agency problem in private universities – the conf‌lict between controlling families

and other stakeholders. We investigate whether universities over which controlling families have disproportionately

signif‌icant power relative to the amount of funds they contribute, that is, universities with high expropriation risk, are

associated with lower outside donations and poor quality.

Research Findings/Insights: Using a sample of Korean private universities, we f‌ind that measures of family control in

excess of monetary contributions are negatively related to the level of outside donation and measures of university quality.

We also f‌ind that universities at which the controlling family exerts disproportionate control are more likely to face disputes

between the controlling family and other stakeholders. Finally, we show that our results are not driven by reverse causality.

Theoretical/Academic Implications: While the existing literature on not-for-prof‌it organizations focuses on the conf‌lict

between professional managers and other stakeholders, we study the conf‌lict between controlling families and other

stakeholders. We investigate a situation in which the controlling family expropriates other stakeholders, a topic missing

from the existing not-for-prof‌it literature.

Practitioner/Policy Implications: This study offers insights to policymakers interested in creating private universities in an

emerging market setting. The relevance of our results is not limited to Korea. According to Altbach, family control of private

universities is prevalent in a number of countries, including Mexico, Thailand, Taiwan, Japan, Korea, the Philippines,

Argentina, India, and China.

Keywords: Corporate Governance, Non-prof‌its, Expropriation, Donations, Private University

INTRODUCTION

Family control of corporations is prevalent around the

world. La Porta, Lopez-de-Silanes, and Shleifer (1999)

show that even large corporations around the world are

typically controlled by families. Family control is not neces-

sarily a bad institutional model per se. In fact, in developed

economies, family f‌irms, particularly when founders serve

as CEO or chairman of the board, tend to be associated with

better performance and higher valuation (Anderson, Mansi,

& Reeb, 2003; Anderson & Reeb, 2003a, 2003b; Andres, 2008;

Maury, 2006; Villalonga & Amit, 2006, 2009). However, in

developing economies, family control is associated with

poor f‌irm performance. Using a sample of Korean f‌irms,

Bae, Kang, and Kim (2002), Baek, Kang, and Park (2004) and

Joh (2003) show that family-controlled Korean business

groups perform poorly. Bertrand, Mehta, and Mullainathan

(2002) f‌ind similar evidence using Indian business groups.

The main reason for the negative relationship between

family ownership and f‌irm performance in emerging

markets is that families have controlling power over f‌irms

despite holding a relatively small portion of their cash f‌low

rights through pyramidal ownership structures and cross-

holdings among f‌irms. That is, there is a signif‌icant disparity

between control rights and cash f‌low rights, which creates

strong incentives for controlling families to expropriate f‌irm

resources. Burkart, Panunzi, and Shleifer (2003) and Morck,

Wolfenzon,and Yeung (2005) develop theoretical arguments

suggesting that in a weak institutional environment, a family

can control a f‌irm without making a commensurate capital

*Address for correspondence: Woochan Kim, Korea University Business School, 5

Anam-Dong, Seongbuk-Gu,Seoul 136-701, Korea. Tel: 822-3290-2816; E-mail: wckim@

korea.ac.kr

388

Corporate Governance: An International Review, 2012, 20(4): 388–404

© 2012 Blackwell Publishing Ltd

doi:10.1111/j.1467-8683.2012.00922.x

investment, creatinglarge divergence between control rights

and cash f‌low rights. Thus, in a country with a weak insti-

tutional environment, family control is expected to be asso-

ciated with high expropriation and poor f‌irm performance,

as evidenced in the previous literature.

In this paper, we investigate whether family control

without a commensurate capital investment is associated

with poor organizational performance in non-prof‌it organi-

zations. Given the evidence of expropriation in for-prof‌it

f‌irms in emerging markets and given the weak institutional

environment in these markets, it is plausible to expect nega-

tive effects of familycontrol of non-prof‌it organizations. Spe-

cif‌ically, using a sample of private, non-prof‌it universities in

Korea, we test the hypothesis that universities over which

founding families have disproportionately large power rela-

tive to the amount of funds they contribute, that is, univer-

sities with high expropriation risk, are associated with poor

university performance as ref‌lected in donations received.1

The more controlling power the founding family has over

universities in excess of their contributions, the more incen-

tive it has to expropriate university resources to its benef‌it at

the expense of other stakeholders. Theincrease in expropria-

tion risk thus discourages future donors, leading to fewer

donations, as potential donors see thatthe risk of their dona-

tions being wasted increases with more control by founding

families.

Note that we do not measure the extent of actual expro-

priation by controlling families. Rather, we measure the

extent of expropriation risk perceived by outside donors and

examine how this expropriation risk affects outside dona-

tions. To measure expropriation risk, we create a Disparity

Index, which is intended to capture the risk of expropriation

by founding families. We f‌ind that the donation level is

negatively and signif‌icantly related to the Disparity Index.

While we use outside donation as our main university

performance variable because it can be easily measured and

its interpretation is homogeneous across different universi-

ties, we also f‌ind that universities with a higher Disparity

Index have poor quality in various dimensions. They tend to

have lower per-student funding and Scholastic Aptitude

Test (SAT) scores and engage in less faculty research, but

have a higher ratio of part-time lecturers to full-time faculty.

They are also more likely to experience a campus dispute

between the founding family and other stakeholders of the

university.

The relevance of our results is not limited to Korea.

Altbach (2005) noted that family-dominated universities are

prevalent in a number of countries, including Mexico, Thai-

land, Taiwan, Japan, South Korea, the Philippines, Argen-

tina, India, and China. Altbach (2005) also noted that family

universities are often established with the idea of making

money or wielding inf‌luence, and typically have very strong

and centralized administrative control, where administra-

tive off‌ices are in the hands of family members.

This paper proceeds as follows. The next section discusses

the related literature. The second section describes the char-

acteristics of private universities in Korea, and the third

section develops the hypotheses. Thefourth section describes

the dataand construction of our main variables.The empirical

results are presented in the f‌ifth section and the sixth section

summarizes and concludes the paper.

RELATED LITERATURE

Our paper is related to several strands in the literature:

(1) family control and organizational performance, (2) risk

of expropriation by family owners, and (3) non-prof‌it orga-

nizational governance and performance.

Family Control and Organizational Performance

This paper is closely related to the growing literature

on family f‌irms. In this review, we limit our discussion to

empirical papers that link family control to performance. For

a more extensive review of family f‌irms, see Bennedsen,

Pérez-González, and Wolfenzon (2010).

Studies of the United States and Western European coun-

tries tend to document the superior performance of family

f‌irms. For example, Anderson and Reeb (2003a) report that

family f‌irms perform better than non-family f‌irms using

S&P 500 f‌irms.2Using Western European f‌irms, Maury

(2006) f‌inds that active family control is associated with

higher prof‌itability than found in non-family f‌irms. Andres

(2008) f‌inds similar evidence using German data.

Further investigation of family f‌irms shows that not all

family f‌irms show superior performance. Using Fortune 500

f‌irms, Villalonga and Amit (2006) report that family owner-

ship creates value only when the founder serves as CEO

of the family f‌irm or as chairman with a hired CEO.3In

contrast, f‌irm value is destroyed when descendants serve

as CEOs. Pérez-González (2006) also f‌inds that f‌irms

with family succession underperform relative to f‌irms that

promote unrelated CEOs.4This f‌inding is consistent with

recent evidence on founder-director f‌irms. Li and Srinivasan

(2011) showed that pay-for-performance and turnover sen-

sitivity to performance are both higher in founder-director

f‌irms than in non-founder f‌irms. The level of corporate

transparency also appears to be important to the relation-

ship between family control and performance. Anderson,

Duru, and Reeb (2009) showed that both founder and heir

f‌irms are signif‌icantly more opaquethan diffuse shareholder

f‌irms and that the positive relationship between family

control and performance exists only in highly transparent

f‌irms.

Risk of Expropriation by Family Owners

Unlike the US evidence, in which certain subsets of family

f‌irms perform better than non-family f‌irms, studies in other

parts of the world f‌ind evidence that family f‌irms tend to

perform poorly in general. Using a sample of Korean f‌irms,

Bae et al. (2002), Baek et al. (2004), and Joh (2003) show that

family-controlled Korean business groups perform poorly.

Bertrand et al. (2002) f‌ind similar evidence using Indian

business groups.5Using Taiwanese f‌irms, Wong, Chang, and

Chen (2010) show that family control is signif‌icantly and

negatively associated with abnormal returns of corporate

venturing announcements. Studying Thai f‌irms, Bertrand,

Johnson, Samphantharak, and Schoar (2008) show that

greater involvement by sons in family f‌irms is associated

with lower f‌irm-level performance.

Why do family f‌irms perform poorly in emerging

markets? The main reason is that many family f‌irms in

FAMILY CONTROL OF NOT-FOR-PROFIT ORGANIZATIONS 389

Volume 20 Number 4 July 2012© 2012 Blackwell Publishing Ltd