Determinants of Underwriting Fees by New Entrant Banks: Evidence from the Japanese IPO Underwriting Market

• Published date: 01 June 2018
• DOI: http://doi.org/10.1111/fima.12191
• Author: Keiichiro Koda,Kazuo Yamada
• Date: 01 June 2018

Determinants of Underwriting Fees by

New Entrant Banks: Evidence from the

Japanese IPO Underwriting Market

Keiichiro Koda and Kazuo Yamada∗

This researchinvestigates how banks expand market share after entering the underwriting market

by examining the relation between commercial bank equity investments and underwriting fees.

First,we f ind that not only bank underwriterswith private information about issuers but also those

without private information discount their fees, especially forsmaller and riskier f irms. This result

is robust when using multiple firm-bank relationship measures or when changing the investing

stage. This is consistent with the strategic discount view that predicts that bank underwriters

discount fees to expand bank market shares in underwriting markets.

This article analyses how commercial banks expand market shares after entering the under-

writing market, focusing on the fee structure. Although traditional studies base their arguments

on information asymmetry, we examine strategicbehavior by bank entrants into the underwriting

market. This argument is supported by evidence from the period during which commercial banks

were allowed to enter the underwriting market in Japan. The findings reveal that investment

banks that have no previous business experience in underwriting and no reputation discount fees

regardless of the degree of private information about issuers.

Two hypotheses are proposed. First, the information advantage hypothesisstates that in prepar-

ing an issuance, underwriters playan impor tant role in mitigating asymmetric information between

firms and investors and certifying the quality of issuers for investors (Booth and Smith, 1986;

Carter and Manaster, 1990; Chemmanur and Fulghieri, 1994). If a bank has private information

about the quality of an issuer, it can avoid the costs of acquiring private information, resulting in

discounted fees, whereas an underwriter normally experiences increased costs for acquiring such

private information (Fang, 2005; Kutsuna, Smith, and Smith, 2007). This cost increases as the

degree of asymmetric information between insiders and outsiders increases (Puri, 1996; Fang,

2005). Thus, a long-term relationship with an issuer enables a bank to obtain private informa-

tion about a firm (Diamond, 1984; Hoshi, Kashyap, and Scharfstein, 1990; Petersen and Rajan,

1994). In contrast, bank underwriters without private information must pay obtain private infor-

mation, which increases fees. Based on this argument, only underwriters with privateinformation

Weare gratefulto Utpal Bhattacharya (Editor) and an anonymous referee for helpful comments and suggestions. Wewould

like to thank Sofia Johan, Kenji Kutsuna, Michelle Lowry(Discussant at ENTFIN), Yusuke Matsuki, Daisuke Miyakawa

(Discussant at JFA),Hideaki Miyajima, Hiroyuki Okamuro,Masayo Shikimi, Hidenori Takahashi, and Konari Uchida for

their helpful comments. Wealso acknowledge other seminar participants at Nagoya University, Oita University,Waseda

University, KyushuUniversity, 2014 IFABS,2014 World FinanceConference, 2015 Japan Finance Association, and 2016

ENTFIN. Wethank Hidenori Takahashi who kindly provided us his underwriting feedata set. This research is supported

by JSPS Grant Nos. 25780256, 16H05952, 16H02027, and 17H02525 and the Ritsumeikan University KokusaikaSuishin

Grant.

∗Keiichiro Koda is an Assistant Professor on the Faculty of Economics at Hiroshima University of Economics, Japan.

Kazuo Yamada is an Associate Professoron the Faculty of Economics at Nagasaki University, Japan.

Financial Management •Summer 2018 •pages 285 – 307

286 Financial Management rSummer 2018

about issuers can discount their fees, and the discount is more pronounced for firms with higher

information asymmetry.

Second, the strategic discounting hypothesis predicts that banks will offer lower underwriting

fees for all customers. This argument is based on the managerial economics and marketing

literature (Tellis, 1986; Raman and Chatterjee, 1995), whichargues that in some settings, entrant

firms discount fees to expand their market share. Because commercial banks are entrants in the

underwriting business and have no reputation in the market, they have an incentive to offer lower

fees to attract clients and to compete with other investment banks that have long histories in the

market. In this situation, underwriting fees from entrant bank underwriters are lower for all firms

regardless of the degree of asymmetric information between the bank and issuer.

Our empirical findings support the strategic discounting hypothesis. Results show that banks

discount underwriting fees not only when they haveprivate information about the issuers but also

when they do not havethis infor mation. Twomeasures are used to def ine firm-bank relationships.

The first is equity investment history by banks or subsidiary venture capital; if the banks invested

before the initial public offering (IPO), they can acquire privateinformation about the issuer. The

second is main bank information, which refers to the bank that has the longest transaction history

with and most private information about an issuer. Based on both measures, we find that banks

discount fees for both high and low degree of asymmetric information, which is measured by firm

size. The discount for small firms is pronounced. For example, we find that a decrease of one

standard deviation of firm size leads to a 0.25-standard-deviation discount in the underwriting

fee. Importantly, the discount for small firms is observed by bank underwriters both with and

without private information implying that the existence of private information does not matter

for the determination of underwriting fee.

A subsample analysis is also conducted by dividing the sample into two periods: 2002–2005

and 2006–2011. If banks use private information, as the first hypothesis predicts, fees should be

discounted throughout the sample periods. However, if all banks discount fees to expand market

shares, more discounts will be offered in the second versus the first sample period because banks

began underwriting in 1999. The findings show that discounts offered by bank underwriters occur

only during the early sample period, which is consistent with the second hypothesis.1

So, why do banks enter the underwriting market even if they discount the fee? One rea-

son is that they can obtain subsequent business with issuers, as pointed out by Hellmann,

Lindsey, and Puri (2008).2The main bank is usually the primary lender for firms in Japan.

In addition to lending, the bank can obtain additional business such as underwriting subse-

quent equity offerings (Krigman, Shaw, and Womack, 2001) and bond issuance (Yasuda, 2005).

Therefore, even if the IPO underwriting is not profitable, the bank can obtain additional profit

by accessing subsequent business with the issuer. Along this line, Fang, Ivashina, and Lerner

(2013) analyze the behavior of banks in private equity markets and find that banks invest-

ing as private equity entities can obtain subsequent lending business and underwrite security

issuance.

This study uses data from the Japanese IPO market, where commercial banks enter the un-

derwriting business by creating investment bank subsidiaries, as allowed by deregulation en-

acted in 1999. These policy changes allow banks to own investment bank subsidiaries and offer

1Weuse the ter m “bank underwriter” to represent the bank’s affiliated security company.

2Another reason we do not consider in this article is risk diversification. By adding investmentbank business, commercial

banks can diversify their source of profit and idiosyncratic risk of the entire business.