How cross‐shareholding influences financial reporting: Evidence from Japan

• Published date: 01 September 2020
• DOI: http://doi.org/10.1111/corg.12333
• Date: 01 September 2020
• Author: Tomomi Takada,Katsuhiko Muramiya

ORIGINAL ARTICLE

How cross-shareholding influences financial reporting:

Evidence from Japan

Katsuhiko Muramiya

1

| Tomomi Takada

2

1

Graduate School of Economics and Center for

Mathematical Modeling and Data Science,

Osaka University, Toyonaka, Japan

2

Graduate School of Business Administration,

Kobe University, Kobe, Japan

Correspondence

Tomomi Takada, Graduate School of Business

Administration, Kobe University, 2-1,

Rokkodai, Nada, Kobe 657-8501, Japan.

Email: takada@pearl.kobe-u.ac.jp

Funding information

JSPS, Grant/Award Numbers: JP17H04783,

JP21730369, JP25780283

Abstract

Research Question/Issue: This study investigates the effect of cross-shareholding on

financial reporting and the information environment in the Japanese market and

hypothesizes that higher cross-shareholding is related to lower contract-efficient

earnings recognition and higher information asymmetry because cross-shareholding

entrenches managers. Furthermore, we investigate whether managers try to maxi-

mize their self-interest or to enjoy a quiet life when they are isolated from market

pressure and entrenched owing to cross-shareholding.

Research Findings/Insights: Using a unique dataset on cross-shareholding among

Japanese listed firms, we find that higher cross-shareholding is associated with (1) less

timely loss recognition and (2) higher probability of information-based trading. Fur-

ther investigation clarifies that higher cross-shareholdings are related to (3) lower

absolute forecast error and (4) lower R&D expenditures. The first two findings indi-

cate that cross-shareholding entrenches managers, and the last two findings show

that entrenched managers enjoy a quiet life.

Theoretical/Academic Implications: This study contributes to the corporate gover-

nance literature by providing new evidence that cross-shareholding influences finan-

cial reporting and the information environment in the market by inducing managers

to become entrenched and to enjoy a quiet life.

Practitioner/Policy Implications: This study provides implications for policymakers in

markets in which cross-shareholding is a common ownership format. To enhance

contract-efficient financial reporting and/or reduce information asymmetry in the

market, it may be necessary to reduce cross-shareholding.

KEYWORDS

corporate governance, archival data, business groups, financial disclosure, Japan

1|INTRODUCTION

Corporate governance has become increasingly prominent since

issues surrounding the separation of security ownership and control

were first addressed in the literature (Grossman & Hart, 1980;

Hart, 1995). Research has thoroughly investigated the governance

effects of ownership concentration on corporate behavior

(e.g., Claessens, Djankov, Fan, & Lang, 2002; Shleifer & Vishny, 1997).

One format of concentrated ownership is cross-shareholding, which is

defined as a mutual shareholding relationship between at least two

firms and is widely observed in several European and East Asian coun-

tries (Claessens, Djankov, & Lang, 2000; Khanna & Yafeh, 2007;

Masulis, Pham, & Zein, 2011). Despite the common usage of cross-

shareholding, there is a shortage of empirical evidence on this form of

concentrated ownership. Several studies document the unfavorable

impact of cross-shareholding on corporate value (Chang, 2003;

Received: 12 February 2019 Revised: 20 May 2020 Accepted: 21 May 2020

DOI: 10.1111/corg.12333

Corp Govern Int Rev. 2020;28:309–326. wileyonlinelibrary.com/journal/corg © 2020 John Wiley & Sons Ltd 309

Claessens et al., 2002; Hiraki, Inoue, Ito, Kuroki, & Masuda, 2003), but

they generally overlook its consequences for financial reporting,

despite its critical role in maintaining sound market systems. Further-

more, the impact of cross-shareholding on the securities market is

underinvestigated despite its relevance to financial statement users

and authorities. To contribute to the corporate governance literature,

we examine the financial reporting and stock market consequences of

cross-shareholding using Japanese data.

We investigate Japanese firms because they provide an ideal

research opportunity for examining the effect of cross-shareholding.

First, Japanese firms exhibit great variation in ownership structure

(Jackson & Miyajima, 2007; Nitta, 2008), but cross-shareholding is still

one of the important ownership structures characterizing the market.

Second, other types of unique ownership formats, such as pyramid

structures and share issuance with differential voting rights, are less

common in Japan than in other non-US countries, which enables us to

detect the independent effect of cross-shareholding. Third, a well-

coordinated commercial database on cross-shareholding for Japanese

listed companies is available.

Cross-shareholding is viewed as a problem because it prevents a

firm's efficient usage of capital and effective monitoring of company

managers. Because cross-shareholding counterparts are generally sup-

pliers, business partners, or allied firms within a group (e.g., keiretsu in

Japan), they have less incentive to monitor management to maximize

corporate value (Forum of Investors Japan, 2015).

1

Instead, the coun-

terparts aim to smooth business transactions, share private informa-

tion, and ensure mutual protection in the event of hostile takeovers

within the cross-shareholding structure. Accordingly, managers with

higher cross-shareholding are isolated from market pressure and scru-

tiny, and thus, they are entrenched.

Figure 1 demonstrates that entrenched managers do not maxi-

mize but, rather, jeopardize corporate value. Specifically, Figure 1

shows the relationship between the level of cross-shareholding and

Tobin's Q for 1989–2012 and indicates that the level of cross-

shareholding is negatively related to Tobin's Q. Figure 1a reports the

relationship for the entire period, and Figure 1b reports the relation-

ship in subperiods (up to 1994, 1995–1999, 2000–2004, 2005–2009,

and from 2010). Interestingly, the gap between theTobin's Q between

the firms with no cross-shareholdings and those with cross-

shareholdings (C0 vs. C1 to C6) is remarkable in Figure 1a, which

implies that the impact of cross-shareholding on management

entrenchment is material even if the percentage of the shareholding is

very small. Because the relationship shown in Figure 1 suggests that

cross-shareholding weakens corporate governance and induces man-

agement entrenchment, our primary research question is how

entrenched managers owing to cross-shareholding influence a firm's

financial reporting and the information environment in the market.

To analyze the impact of cross-shareholding on financial

reporting, we utilize timely loss recognition as an outcome measure

(Basu, 1997), because it is widely used in the literature on corporate

governance (e.g., Ahmed & Duellman, 2007; Lafond &

Roychowdhury, 2008; Ramalingegowda & Yu, 2012). Timely loss

recognition, which is also called conditional conservatism, is defined

as accountants' tendency to require a higher degree of verification

for recognizing good news than bad news in financial statements

(Basu, 1997, p. 4). The literature has documented that timely loss

recognition enhances the usefulness of financial statements for con-

tract purposes because it emits warning signals regarding invest-

ments with a negative net present value to stakeholders in a timely

manner, thereby allowing the parties to take rapid corrective actions

(Ball, 2001; Kothari, Ramanna, & Skinner, 2010; Watts, 2003a). As

such, timely loss recognition is generally demanded and

implemented in financial accounting (Basu, 1997; Lafond &

Roychowdhury, 2008). Given this ceteris paribus condition of timely

loss recognition, we hypothesize that cross-shareholding induces

lower levels of timely loss recognition because it weakens managers'

discipline. Furthermore, if corporate stakeholders develop amicable

relationships through cross-shareholding, they make fewer demands

for public disclosures because they have alternative paths through

which to obtain private information, such as through sitting on the

board, which also induces lower levels of timely loss recognition.

FIGURE 1 The relationship between cross-shareholding and

Tobin's Q. (a) Cross-shareholding and Tobin's Q for Japanese firms in

1989–2012. (b) Cross-shareholding and Tobin's Q in subperiods.

Panel (a) reports the averageTobin's Q (= (Total Assets −Book Value

of Equity + Market Value of Equity)/Total Assets) according to the

levels of cross-shareholding from 1989 to 2012. Here, C0 (C6)

represents firms with no (the most) cross-shareholding. We categorize

observations into six groups according to the level of cross-

shareholding as follows: C0,CROSS = 0%; C1,0%< CROSS < 5%; C2,

5% ≤CROSS < 10%; C3,10%≤CROSS < 15%; C4,15%≤CROSS < 20%;

C5,20%≤CROSS < 25%; and C6,25%≤CROSS.Theboxplotsof

Panel (b) represent the upper adjacent value, 75th percentile, median,

25th percentile, and lower adjacentvalue of Tobin's Q in subperiods

310 MURAMIYA AND TAKADA