Bank–client cross‐ownership of bank stocks: A network analysis

• Published date: 01 June 2022
• Author: James R. Barth,Sunghoon Joo,Kang‐Bok Lee
• Date: 01 June 2022
• DOI: http://doi.org/10.1111/jfir.12275

Received: 20 October 2019

|

Accepted: 5 April 2022

DOI: 10.1111/jfir.12275

ORIGINAL ARTICLE

Bank–client cross‐ownership of bank stocks: A

network analysis

James R. Barth

1

|Sunghoon Joo

2

|Kang‐Bok Lee

3

1

Department of Finance, Auburn University,

Auburn, Alabama, USA

2

Department of Accounting, Finance, and

Economics, California State University

Dominguez Hills, Carson, California, USA

3

Department of Systems and Technology,

Auburn University, Auburn, Alabama, USA

Correspondence

Sunghoon Joo, Department of Accounting,

Finance, and Economics, California State

University, Dominguez Hills, 1000 E. Victoria

Street, Carson, CA 90747, USA.

Email: sjoo@csudh.edu

Abstract

We examine the effects of banks' client stock ownership

structure on their governance mechanism, risk taking, and

systemic risk in the financial system. We apply a dyadic

level of analysis to provide new insights into the relevance

of such cross‐ownership as an effective monitoring

mechanism and as a source of interconnectedness between

and among financial institutions. Our empirical results

indicate that bank–client cross‐ownership of bank stocks

is negatively associated with the riskiness of bank holding

companies and positively associated with systemic risk.

Moreover, the effects of such cross‐ownership on systemic

risk are stronger in times of a financial crisis.

JEL CLASSIFICATION

G23, G24, G28

1|INTRODUCTION

The global financial crisis has drawn considerable attention to the effects of the interconnectedness among financial

institutions on the financial system. As noted in Allen and Babus (2009), networks—broadly understood as a

collection of nodes and links between nodes—can be a useful representation of financial systems. Because the

ownership structure of economic entities is often characterized by a complex network of interdependent owners

(Dietzenbacher & Temurshoev, 2008), cross‐ownership among bank holding companies (BHCs) can be a potential

source that contributes to the complexity and interdependency in the financial system. In the banking sector, the

literature linking banks' ownership structure and bank risk has mainly focused on the effects of ownership

J Financ Res. 2022;45:280–312.280

|

wileyonlinelibrary.com/journal/JFIR

This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which

permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no

modifications or adaptations are made.

© 2022 The Authors. Journal of Financial Research published by Wiley Periodicals LLC on behalf of The Southern Finance

Association and the Southwestern Finance Association.

concentration and the nature of owners

1

on either banks' governance or risk taking (Boyd & Hakenes, 2014;

Iannotta et al., 2007,2013; Laeven & Levine, 2009; Saunders et al., 1990). These studies show that the ownership

structure of individual banks alters owners' risk‐taking behavior through either the changing standard risk‐shifting

incentives or the interactions with bank regulatory policies.

Yet, we have been unable to find research on the effects of banks' cross‐ownership structure on their

governance and systemic risk in the financial system. We aim to fill this gap in the literature by providing new

insights into the relevance of the bank ownership ties by bank clients in reducing banks' risk‐taking behavior. We

also examine whether such ties are a source of herding behavior of those clients that may contribute to systemic

risk in the banking sector. More specifically, we first examine bank–client ownership ties as a type of ownership

structure that may serve as an effective mutual monitoring mechanism using a sample of 537 US BHCs from 2008

to 2015. We then examine how such ownership interconnectedness among financial institutions by bank clients is

related to the stability of the financial system. To the best of our knowledge, no research examines such a

connection between the ownership structure of bank stocks by bank clients and individual bank's risk‐taking

behavior, as well as systemic risk in the banking system.

Table 1presents a matrix of cross‐ownership among the major US BHCs as of the end of 2015. This cross‐

ownership matrix shows a directionality that goes from the rows to the columns. For example, JPMorgan Chase

(first row) holds 1.70% of common shares of Bank of America for its clients as of the end of 2015, and Bank of

America (second row) holds 1.57% of common shares of JPMorgan Chase for its clients as of the end of 2015.

The percentage of cross‐ownership in Table 1accounts for only a small fraction of a BHC's shares outstanding.

Nevertheless, such seemingly small percentages can translate into significant dollar amounts. Table 2presents the

dollar amount of the 5 largest BHCs' holdings for their client's common shares outstanding of the other 536 BHCs

in the sample from 2007 to 2015. JPMorgan Chase, Bank of America, Wells Fargo, Citigroup, and Goldman Sachs

are the 5 largest BHCs, and each holds $13, $13, $6, $5, and $14 billion, respectively, in equity stakes in the other

536 BHCs for their clients as of the end of 2007. We compare these equity stakes with total equity capital in Panel

A and tangible equity capital in Panel B. Tangible equity capital includes core capital elements

2

and therefore plays a

critical role in assessing the viability of financial institutions. In the third column of each BHC, we present how much

a BHC's ownership stakes held for its clients account for total equity capital in Panel A and for tangible equity

capital in Panel B, in percentage. As can be seen, these percentages are not negligible and the difference in the

percentages between Panels A and B is notable, especially in 2007, for the 5 largest BHCs. Overall, Table 2

illustrates that total ownership stakes held for the banks' clients among BHCs are not negligible in dollar amount.

In our article, a high degree of interrelatedness originates from the direct ownership connections between

BHCs through their clients' bank stock holdings. Figure 1shows a client stock ownership network among the major

BHCs in our sample and illustrates that the major BHCs are closely interconnected through clients holding shares of

one another. Given the sheer size of BHCs in the overall financial system, such linkages through bank–client cross‐

ownership of shares across BHCs in financial networks creates a complex web of ownership network that may

contribute to systemic risk.

Given that a BHC's ownership stakes on behalf of its clients can account for a large fraction of its own tangible

equity capital, cross‐ownership among BHCs may be an overlooked factor in explaining systemic interconnected-

ness in the financial system. Specifically, if the clients of the banks have information not available to other owners

of bank stocks and they decide to sell their shares all at once, sending the price of the stocks plunging, this may

create a panic with regard to the banks' solvency. This, in turn, can lead to uninsured deposit runs and the inability

of banks to continue their funding with repurchase agreements.

The main contribution of our article is that we relate two topics examined in isolation in the literature. The first

topic is the effects of banks' client stock ownership structure on their governance mechanism and their risk‐taking

1

The nature of owners is mainly classified into three groups: privately owned stock banks, mutual banks, and government‐owned banks.

2

See Rules and Regulations (2013) for the components of tangible equity capital.

CROSS‐OWNERSHIP OF BANK STOCKS

|

281

behavior, and the second topic is the effects of such ownership ties in the banking sector on systemic risk in the

financial system. We apply a dyadic level of analysis (pair of entities as a unit of analysis). This enables us to provide

new insights into the relevance of such cross‐ownership as an effective mutual monitoring channel and as a

possible source of interconnectedness between and among financial institutions.

2|THEORETICAL BACKGROUND AND HYPOTHESIS DEVELOPMENT

The primary source of profits for most banking firms comes from the collection and investment of depositors' funds

in such a way that banks earn the interest margin by charging lower interest on deposits and higher interest on

loans (Walter, 1991). The major assets in which banks invest depositors' funds are loans. Banks can improve asset

quality through the enhanced monitoring system of their loan portfolio (Goetz et al., 2016). In our context, cross‐

shareholdings among BHCs by bank clients may provide a more direct form of mutual monitoring and potentially

align the incentives of one BHC with those of the other BHCs.

There are several reasons to suggest that bank–client ownership ties as a type of ownership structure serve as

an effective mutual monitoring mechanism. First, banks are holders of equity and thus principals in other banks, and

they are also agents of their clients who invest in their funds and thus have a fiduciary duty to monitor contractual

relations with their clients. This creates a dual agency relationship (Arthurs et al., 2008) wherein shareholders serve

as principals, discharging duties concerning managerial agents, and as agents, who are themselves charged with the

duty of investing toward the objectives of ultimate shareholders in a particular fund (Connelly et al., 2010). Studies

show that the dual identity of these agents indeed makes them more effective monitors (Arthurs et al., 2008; Pratt

& Foreman, 2000).

Second, different types of institutional investors (i.e., pension funds, banks and trusts, insurance companies, and

mutual funds) are subject to different legal restrictions and expectations. Accordingly, institutional investors' level

of monitoring or their degree of influence on management can differ by type. For example, mutual funds are legally

TABLE 1 Matrix of cross‐ownership between the major US bank holding companies (BHCs) as of the end

of 2015

JPMorgan

Chase

Bank of

America

Wells

Fargo CitigGroup

Goldman

Sachs Prudential

State

Street

BNY

Mellon

JPMorgan

Chase

. 1.70% 1.97% 1.55% 1.22% 1.58% 0.66% 0.44%

Bank of

America

1.57% . 1.03% 1.04% 0.97% 0.78% 0.69% 0.50%

Wells Fargo 0.65% 0.61% . 0.38% 0.94% 0.47% 0.21% 0.28%

Citigroup 0.10% 1.20% 0.00% . 2.49% 0.12% 0.00% 0.00%

Goldman

Sachs

0.57% 0.63% 0.30% 0.52% . 1.37% 0.20% 0.47%

Prudential 0.42% 0.42% 0.30% 0.64% 0.60% . 0.18% 0.15%

State Street 4.18% 3.98% 3.69% 4.07% 4.76% 4.03% . 4.28%

BNY Mellon 1.28% 1.06% 0.96% 1.31% 0.85% 1.66% 1.03% .

Note: This table shows the matrix of cross‐ownership between the major US BHCs, in percentages, as of the end of 2015.

This matrix has a directionality that goes from the rows to the columns. For example, JPMorgan Chase (first row) owns

1.70% of common shares of Bank of America as of the end of 2015, and Bank of America (second row) owns 1.57% of

common shares of JPMorgan Chase as of the end of 2015.

282

|

JOURNAL OF FINANCIAL RESEARCH