Government vs. Private Control, Political Loans, and the Privatization of
Jaewook Ana, Sang Kun Baeb, Ronald A. Rattic,∗∗
aDepartment of Economics, Kyung Hee University, Seoul, Korea 130-701
bKorea Economic Research Institute, Seoul, Korea
cDepartment of Economics, University of Missouri, Columbia, MO 65211,USA
This paper compares the performance of banks with and without effective government control
in appointment of chief operating officers in Korea using panel data. A privatization program
succeeded in spreading ownership of banks widely among the public. Government retention
of an ownership stake in an institution meant de facto control by government, decision-
making subject to political objectives, and lack of government respect for investors’ property
rights following privatization. A model is presented in which political loans are a constraint
on banks subject to government control. It is found that banks controlled by government,
despite charging lower loan rates, experienced disproportionately bad loan performance, and
were inefficient compared to privately controlled banks.
JEL classification: G 21; G 28
Keywords: Government control, bank privatization, political loans
∗ The authors wish to thank to J. Huston McCulloch, Paul Evans and Hyun Park for helpful comments.
An earlier version of the paper has been presented at the Korea Law and Economics Association
Monthly Meeting, the Korea Economics Association Annual Meeting in Seoul on February 15, 2001,
and the Korea Econometric Society Annual Meeting in Seoul on June 15, 2001. We thank participants
in these seminars for thoughtful comment.
∗∗ Corresponding author. Tel.: 573-882-6474; fax: 573-882-2697. E-mail address: email@example.com
Government vs. Private Control, Political Loans, and the Privatization of
There has been a spread of privatization programs, whereby state owned enterprises
are sold by government to agents in the private sector, in recent years. In their review of the
growing literature on this issue, Megginson and Netter (2001) find that privatization has for
the most part been successful. Kikeri et al. (1992), Galal et al. (1994), and Megginson et al.
(1994) report that many newly privatized firms become more efficient. These and other
authors carefully document the circumstances under which the sale of assets to private
economic agents will result in firms becoming more efficient and profitable.1
The objective in this paper is to present evidence on the performance in Korea of
banks privatized, but controlled de facto by government, and on the performance of privately
owned and managed banks using panel data from 1987 to 1997.2 Prior to 1980 commercial
banks had been state owned in Korea, and bank lending was an instrument of industrial policy
by directing scare funds to priority projects considered highly productive.3 Proponents of state
ownership of banks could also argue that this helped overcome market failures due to
Starting in the early 1980s a privatization program intended to bring market orientated
1 Frydman et al. (1999) and Jones and Mygind (1999) show that privatizations have limited success (in
transition economies) when there has been an overhang of worker and managerial ownership.
2 There has been recent interest on the issue of the relative efficiency of privately controlled and
government controlled banks. La Porta et al. (2000) finds that government ownership of banks retards
financial and economic development in an analysis of large banks in 92 countries. Barth et al. (2000)
report that greater state ownership of banks across 60 countries tends to be associated with less
developed banks and poorly functioning financial systems.
3 In Korea’s experience, the industrial policies of the 1960s that favored chemical and heavy industry
seriously misallocated real and financial resources. These policies created overcapacity in those industries
and denied resources to medium and small businesses.
4 Arguments about market failure are not firmly established. Rolnick and Weber (1984) found no
evidence of any contagion effects from bank runs during the "free-banking" era (1837-1860) in the
U.S. In a review of the national banking era (1863-1913), Kaufman (1988) demonstrated that evidence
of contagions in the panics of 1878, 1893, and 1908 is weak except for 1893. Wicker (1980) found that
the contagion effects had been limited regionally during the "Great Contraction" of 1930-1933, and that
prior to 1932 runs were confined for the most part to banks suffering from pre-run insolvency or to
banks affiliated with insolvent firms.
decision making to the financial sector succeeded in spreading ownership of banks widely among
members of the public. The intent of the privatization program was undermined, however, by a
restriction on ownership meant to prevent domination of banks by chaebol business groups, since
those institutions in which government retained an ownership stake (with no major shareholders
from the private sector) continued to be effectively controlled by government, in that government
determined the appointment (and decisions regarding retention) of chief operating officers at
these banks.5 This situation of overwhelming (but widely dispersed) private ownership and
strong government influence raises potentially important issues about efficient decision making,
since top managers at these institutions owe their positions and prospects for further
advancement to the government. Government influence over decisions on loans and bank
resources could result in outcomes adverse to the interests of bank shareholders, and
ultimately the general public through implicit guarantee of security of deposits. Those
institutions free of a government stake or with foreign investors as major shareholders
(interestingly, ownership restrictions did not apply to foreigners) were controlled by the private
The simultaneous existence of privately controlled banks and banks subject to
government control of the appointment of chief operating officers allows us to directly compare
their performance. It is argued that banks effectively controlled by government are more likely
to make loans for political reasons or in line with broad public policy issues, and that this will
raise costs.6 A model is presented in which political loans are a constraint on banks effectively
controlled by government. The data suggest that despite higher loan rates at privately
5 Clark and Cull (2001) argue that the means whereby privatization is achieved is important in
determining the outcome of privatization. They present a model of the incentives faced by government
and private investors in setting the parameters for privatization.
6 Altunas et al. (2001) finds that state-owned banks are somewhat more efficient than private banks in
Germany. Our results by type of control apply with regard to privatization in Korea and the way in
which government operates in Korea. Laeven (1999) does not directly compare the efficiency of
privately owned versus government owned banks, but shows that family owned and company-owned
banks in East Asia took on more risk than foreign-owned banks. For banks in the U.S., Spong et al.
(1995) and Sullivan and Spong (1998) find active involvement by major shareholders to be
characteristic of efficient banks. The absence of major shareholders in some of the banks in Korea may
have contributed to their relative inefficiency.
controlled banks, and the increased potential for greater adverse selection in the awarding of
loans, the ratio of bad loans to total loans is much smaller at privately controlled banks than at
government controlled banks. In empirical work we find that banks subject to government
control in the appointment of chief operating officer have statistically significantly higher
costs and lower profits than privately controlled banks.
The work in this paper is related to the literature on ownership issues, in that
management at banks controlled by government is not subject to capital market discipline.
This is not because of exclusive public ownership, but because following privatization
government retained control of career paths of top management and performance was
evaluated on criteria besides efficiency.7 Political interference forced banks subject
government control to make loans on bases other than profit and market orientated criteria,
and may well have contributed to the severity of the Korean financial crisis at the end of
1997. The experience of Korea indicates that failure on the part of government to respect
investors’ property rights following privatization was a major reason for the poor
performance of banks subject to government control. Following the crisis of late 1997, a
massive infusion of public money has been spent on restructuring the banking sector and
effective recreation of government banks. The episode of privatization during the pre-crisis
period provides lessons on what to avoid in setting up an efficient market orientated banking
The paper is organized as follows. Section I provides an overview of Korea banking
system. Section II presents a simple bank model for economic performance that introduces
political loans that require real resources as a constraint on management behavior at banks
7 Public choice theory suggests that the public sector performs less efficiently than the private sector
(see Buchannan and Tullock (1962), Niskanen (1975), and Shapiro (1975)). In contrast, Liebenstein
(1966) argues that monopolies are likely to be X-inefficient, regardless of ownership. De Alessi(1983)
criticizes Leibenstin’s X-inefficiency argument. Boardman and Vining (1989) provide comprehensive
discussion of these issues and a summary of empirical evidence on relative efficiency of publicly
owned and private owned firms. There has also been a wider debate on the public choice model and
ownership issues (including contributions by Jensen and Meckling (1976), Fama (1980), and Fama and
subject to government control. Section III discusses the estimation methodology, the panel
data, and presents the empirical results. Conclusions appear in Section IV.
I. Overview of the Korean Banking System
A few large banks with extensive branch networks dominate the commercial banking
sector in Korea. There were only five commercial banks with nation wide operations at the
end of 1970s: Commercial Bank, Chohung Bank, Hanil Bank, First Bank, and Seoul Bank.
After financial liberalization and restructuring starting in 1980, eleven new ‘commercial’
banks entered the sector over time and there were sixteen commercial banks with nation wide
operations at the end of 1997. These banks are listed in Table I. The eleven new banks entered
the sector at the following dates: Shinhan Bank in 1982; Koram Bank in 1983; Dongwha
Bank, Dongnam Bank, Daedong Bank, and Korea Exchange Bank in 1989; Hana Bank and
Boram Bank in 1991; Peace Bank in 1992; Kookmin Bank in 1995; and Housing Bank in
1997.8 The total number of domestic branches of the nationwide commercial banks amounted
to over 4,100 and held approximately 85 percent of the total assets of commercial banks in
1997 (small regional commercial banks held the balance of assets).
Prior to 1980, government owned the large nationwide commercial banks and
directed them to provide priority to strategic sectors and industries with subsidized credit.
Government appointed top bank managers, supervised the bank budget, and organized the
business activities of banks. In an attempt to move toward private sector-oriented
management of economy, financial institutions, including commercial banks, were privatized
over the period from 1981 to 1983 by selling equity to the general public. Partial privatization,
the laws governing ownership, and the gradual granting of commercial bank charters to
institutions with very different histories resulted, as we will see, in a dual system in the
8 In the last few years the number of banks has been reduced from sixteen to ten in order to create
larger and hence more viable economic units. Boram bank has merged with Hana Bank, Commercial
bank and Hanil Bank have been merged into Hanvit Bank, and Dongwha Bank, Daedong Bank, and
Dongnam Bank have been acquired by Shinhan Bank, Kookmin Bank, and Housing Bank, respectively.