The Role of Bank‐Affiliated Venture Capital for Parent Banks in Japan: New Evidence

• Author: Konari Uchida,Yue Sun
• DOI: http://doi.org/10.1111/ajfs.12155
• Date: 01 December 2016
• Published date: 01 December 2016

The Role of Bank-Affiliated Venture Capital

for Parent Banks in Japan: New Evidence*

Yue Sun

School of Economics and Management, Beijing University of Chemical Technology

Konari Uchida**

Faculty of Economics, Kyushu University

Received 1 April 2016; Accepted 15 October 2016

Abstract

After controlling for the effect of parent banks’ (PBs) direct ownership, we find that invest-

ment by bank-affiliated venture capitals (BVCs) is positively associated with the probability

that investee firms have loan balances with PBs in the year they issue an IPO. However,

BVCs typically sell off their holdings of investee firms within 2 years after the IPO. Further-

more, the level of BVC ownership does not have explanatory power regarding whether inves-

tee firms have loans from PBs. On the contrary, PBs’ direct shareholdings have a positive

association with the probability that investee firms have PB loans for many years after IPOs.

Keywords Bank-affiliated venture capital; Strategic objectives; Bank lending; Bank ownership;

Exit strategy

JEL Classification: G21, G24, G32

1. Introduction

This study investigates the role of bank-affiliated venture capital (BVC) for parent

banks’ lending business. BVCs are venture capital firms that are subsidiaries of a

bank. Previous studies suggest that BVCs invest in start-ups to increase their parent

banks’ (PBs’) lending opportunities (Hellmann, 2002; Wang et al., 2002). Hellmann

et al. (2008) find that the likelihood of a bank lending to a start-up increases when

the bank invests in the firm directly or indirectly through its subsidiary BVC.

*The authors would like to thank the editors (Hee-Joon Ahn and Kwangwoo Park) and two

anonymous referees. This research has been financially supported by JSPS KAKENHI Grant

Number 15H03367, JSPS Two-Country Joint Research Fund, Trust Companies Association of

Japan, and Fundamental Research Funds for Central University (No. 100102016ZY1623).

**Corresponding author: Konari Uchida, Faculty of Economics, Kyushu University, 6-19-1,

Hakozaki, Higashiku, Fukuoka 812-8581, Japan. Tel. and Fax: +81-92-642-2463, email: kuchi

da@econ.kyushu-u.ac.jp.

Asia-Pacific Journal of Financial Studies (2016) 45, 864–885 doi:10.1111/ajfs.12155

864 ©2016 Korean Securities Association

However, initial public offerings (IPOs) are commonly viewed as a profitable exit

opportunity for venture capital firms (VCs), suggesting that BVCs’ shareholdings

may be terminated at the time of the investee’s IPO. This raises the following ques-

tion: do BVCs’ investments contribute to the PBs’ lending business after investee

firms’ IPO? This research addresses the issue with particular attention to the fact

that PBs can directly hold shares of young companies. Specifically, we examine

whether BVCs’ shareholdings affect PB lending after controlling for the effect of

PBs’ direct shareholdings.

This issue is addressed by using Japanese data, which offers a number of advan-

tages. In Japan, many VC firms are subsidiaries of banks, while independent VCs

also exist. It is well known that Japanese banks maintain long-term relations with

borrowing firms.

1

Furthermore, Japanese banks have been allowed to hold up to

5% of the shares of non-financial companies together with their subsidiaries (noting

that bank groups headed by holding companies can hold up to 15% of shares of

non-financial companies).

2

It is well documented that Japanese main banks own

the shares of borrowing companies to maintain close relations. This implies that

Japanese banks choose between their direct shareholdings and indirect investments

through subsidiary BVCs. Japanese IPO data also provides information on firms’

borrowings from individual lending banks. This comprehensive data facilitates the

investigation of the effect of BVCs’ investments on PBs’ lending business after con-

trolling for the impact of direct investments.

This study finds that the existence of BVC investments increases the probability

that investee firms have a loan balance with PBs in the year they issued an IPO

after controlling for the effect of PBs’ direct shareholdings. This result suggests that

BVC investments enhance PBs’ lending business at least before the IPO of their

investee firms. However, BVCs’ ownership has much smaller marginal effects on

PBs’ lending than do PBs’ direct shareholdings, and BVCs sell off shares of investee

firms within a few years after the IPO, just as is usually done by independent ven-

ture capitalists (IVCs). This suggests that BVCs contribute little to PBs’ lending

business after the IPO of their investee firms. Furthermore, we find the BVCs’ own-

ership level does not have significant explanatory power over whether investee firms

have loan balances with PBs. On the contrary, PBs’ direct shareholdings are associ-

ated with the probability of investee firms having PB loans throughout our sample

period. In conjunction with the fact that only 47% of BVC-backed IPOs have PB

loans, these results suggest that banks choose firms for lending relationships based

mostly on creditworthiness and directly hold the shares of those companies.

1

Japanese security houses, which resemble investment banks in the United States, also launch

VCs. These VCs are likely to seek the opportunity for parent security houses to underwrite

IPOs of investee firms.

2

BVCs’ shareholdings are not included in their PBs’ shareholdings as long as investee firms

are in early stages and small in size.

The Role of Bank-affiliated Venture Capital

©2016 Korean Securities Association 865