What Drives Interregional Bank Branch Closure? The Case of Japan's Regional Banks in the Post‐Deregulation Period

• DOI: http://doi.org/10.1111/irfi.12167
• Date: 01 December 2018
• Author: Mamoru Nagano,Tatsuo Ushijima
• Published date: 01 December 2018

What Drives Interregional Bank

Branch Closure? The Case of

Japan’s Regional Banks in the Post-

Deregulation Period*

MAMORU NAGANO

†

AND TATSUO USHIJIMA

‡

†

Faculty of Economics, Seikei University, Musashino City, Tokyo, Japan and

‡

Faculty of Business and Commerce, Keio University, Tokyo, Japan

ABSTRACT

Since the elimination of branch restrictions in the Japanese banking sector,

the number of interregional branches closing has exceeded that of new

branches being established. By analyzing branch data covering 2000–2012,

we find that the probability of interregional branch closures is higher than

that of intraregional branch closures because interregional branching

worsens banks’cost efficiency. Further, we show that the geographical dis-

tance between branches does not increase the probability of intraregional

branch closures, but it does raise the probability of interregional branch clo-

sures. Moreover, banks that focus on SME markets have a higher probability

of closure in interregional markets than those that focus on household loan

markets.

JEL Codes: G21; R12; R51

Accepted: 12 November 2017

I. INTRODUCTION

Whether interregional bank branch deployment has a positive or negative

impact on bank performance is a long-term concern for many researchers. In

the United States, as reported by Dick (2006), studies first state that interre-

gional branch deployment has benefitted both lenders and borrowers signifi-

cantly since the enforcement of the Riegle-Neal Interstate Banking and

Branching Efficiency Act of 1994. However, more recent studies such as by

Keys et al. (2010) and Aguirregabiria and Clark (2016) use recent regional

data and conclude that new entrants in interregional markets increase lend-

ing contracts with borrowers with low-repayment ability. In Japan, since the

elimination of branch regulations in 1997, the number of branch closures in

interregional markets has exceeded that of newly established branches. These

* The authors thank various seminar participants for their helpful comments and suggestions.

This research is financially supported by JSPS KAKENHI, Grant Number JP15K03550.

© 2017 International Review of Finance Ltd. 2017

International Review of Finance, 18:4, 2018: pp. 595–635

DOI: 10.1111/irfi.12167

recent trends are noteworthy, since the banking sectors in both the United

States and Japan were controlled by strict branch regulations in the postwar

period until their elimination in the mid-1990s. The purpose of this study is

thus to investigate the common, general factors that determine interregional

bank branch closures and to show new evidence on what drives these

outcomes.

To determine the key driving forces behind the interregional distribution of

bank branch deployment, we focus on regional border effects in the banking

market. Here, regional border effects are defined as bank branch deployment

over regional or national borders that result in branch closure or negatively

influence a bank’s performance. Studies such as Holmes (1998) provide evi-

dence of the regional border effect in manufacturing industries, which are gen-

erally forced to incur additional transportation costs to distribute raw

materials, intermediate goods, and final products when the manufacturing

bases operate across state borders. The banking industry, which differs from

the manufacturing industry, was previously considered as not having to incur

such transportation costs and thus immune to regional border effects. How-

ever, Buch (2005) reports that the regional border effect can negatively affect

bank performance because banks handle not only hard information on bor-

rowers but also significant soft information. The study further asserts that

bank branches newly established in the interregional market are unable to cre-

ate long-term relationships with local borrowers and therefore incur more

information cost disadvantages than existing local banks that have historical

relationships with borrowers.

Our study hypothesizes that intraregional local bank–borrower relationships

strengthen under long-term interregional branch restrictions. As a result, soft

information on borrowers accumulate within local banks. Therefore, we sup-

pose that the local lending relationship and accumulation of soft information

remain the major entry barriers for outsiders, even after the elimination of

interregional branch regulations for the corporate lending market. Accordingly,

interregional bank branch deployment weakens bank performance and leads

to branch closures after the branch liberalization process as long as banks’rev-

enues depend on corporate loan interest.

This study uses Japan’s regional bank branch data, financial data, and

regional market data covering the period between 2000 and 2012. Branch data

include geographical information on existing, closing, and newly established

branches. To determine whether each branch crosses regional borders, we use

the branch’s address and geographical distance from the nearest branch. We

then investigate the influence of these variables on bank branch closures in

the interregional market. Our empirical analyses also use control variables for

regional demographic and economic performance data as well as bank size

and financial performance data. We further examine the relationship between

cost efficiency and the geographical distribution of branches.

The remainder of the paper is organized as follows. In the followingsection, we

review the extant literature in this domain as well as the historical path of bank

© 2017 International Review of Finance Ltd. 2017596

International Review of Finance

branch regulation and deregulation in Japan, and introduce our hypotheses. In

Section 3, we present our empirical model and the data used to examine our

hypotheses. Subsequently, in Sections 4 and 5, we discuss the results produced by

our empirical models. Finally, in Section 6, we offer concluding remarks.

II. EXISTING LITERATURE AND HYPOTHESES

A. Literature on regional border and Bank branch deployment

Holmes (1998) and Bartik (1985) assert that regional border effects negatively

influence the performance of the manufacturing industry. They suggest that

local taxation, local industrial policies, and local regulations vary across local

governments and that manufacturers consequently incur higher costs as they

deploy production bases across regional borders. In the banking industry,

Degryse and Ongena (2004) use the significant volume of research on survey

results to assert that both regional border effects and the lender–borrower geo-

graphical distance influence the distribution of bank branches. In their study,

they denote national and state borders as regional borders.

Degryse and Ongena (2004) define two types of economic borders: exoge-

nous and endogenous. Their definition implies that differences in local bank-

ing supervision, languages, and local political systems are exogenous economic

borders. The study also assumes that endogenous economic borders are local

market entry barriers nurtured through the formation of the local bank–

borrower relationship and accumulation of soft information on borrowers in

the local banking market.

1

Therefore, Degryse and Ongena (2004) ultimately position a series of

empirical studies concerning bank–borrower distance as a study of endogenous

economic borders. Further, Buch (2005) categorizes regional borders as regula-

tory and economic borders and verifies these impacts on bank branch deploy-

ment. In this study, an economic border is defined as a boundary demarcating

an area regionally covered by each local bank supervision authority, whose res-

idents speak common languages and to which one local administrative area

under one political district is allocated.

The enforcement of the McFadden Act of 1927 strictly restricted the estab-

lishment of interstate bank branches in the United States. Therefore, many

studies on the regulatory borders of bank branch deployment focus on the

relationship between interstate branch regulations and their regional economic

impacts. In this regard, Calomiris (2000) and Amel and Liang (1992) conclude

that regional economies were financially susceptible to external shocks such as

business cycle downturns and stock market turbulence under interstate branch

regulations in the pre-1994 period. Additionally, Jayaratne and Strahan (1998),

Kroszner and Strahan (1999), Stiroh and Strahan (2003), and Dick (2006) show

1 The definition of endogenous economic borders proposed by Degryse and Ongena (2004)

originated from empirical evidence reported by Rajan (1992) and Dell’Ariccia (2001).

© 2017 International Review of Finance Ltd. 2017 597

What Drives Interregional Bank Branch Closure?