Regulation and the connectedness of insurers to the banking sector: International evidence

• Date: 01 December 2019
• Published date: 01 December 2019
• Author: Sabine Wende,Jannes Rauch,Greg Niehaus
• DOI: http://doi.org/10.1111/rmir.12135

© 2019 The American Risk and Insurance Association

Risk Manag Insur Rev. 2019;22:393–420. wileyonlinelibrary.com/journal/rmir

|

393

Received: 25 June 2019

|

Accepted: 19 November 2019

DOI: 10.1111/rmir.12135

FEATURE ARTICLE

Regulation and the connectedness of insurers

to the banking sector: International evidence

Greg Niehaus

1

|

Jannes Rauch

2

|

Sabine Wende

2

1

Darla Moore School of Business,

University of South Carolina, Columbia,

South Carolina

2

Faculty of Management, Economics, and

Social Sciences, University of Cologne

(Albertus‐Magnus‐Platz), Cologne,

Germany

Correspondence

Greg Niehaus, Darla Moore School of

Business, University of South Carolina,

1014 Greene Street, Columbia, SC 29208.

Email: gregn@moore.sc.edu

Funding information

Jannes Rauch received a travel grant

from Deutscher Verein für

Versicherungswissenschaft e.V.

Abstract

Using variation across countries and time in thedegree to

which regulations restrict banks and insurers from

engaging in the same activities, we find that property/

liability insurers’connectedness to the banking sector

declines when regulatory restrictions increase, but life

insurers’connectedness to banks does not. The results

suggest that the connectedness between life insurers and

banks is largely due to these institutions sharing common

underlying economic and financial risk factors that exist

even when regulation restricts these institutions from

engaging in each other’sactivities.

KEYWORDS

banking, connectedness, insurance, regulation

1

|

INTRODUCTION

The financial crisis heightened concern about the connectedness of institutions in different

financial sectors, for example, the extent to which shocks to banking institutions are

transmitted to insurers, and vice versa. Consistent with this concern, Billio, Getmansky, Lo, and

Pelizzon (2012) present evidence that connectedness across financial institutions increased

during the first decade of this century and that banks are an important transmitter of shocks to

other financial sectors, including the insurance sector. Begin, Boudreault, Doljanu, and

Gauthier (2019) and Chen, Cummins, Viswanathan, and Weiss (2014) also present evidence

that banking shocks impact the insurance sector. Butler, Merrill, and Lai (2019) show that large

shocks in the stock returns of large U.S. banks and insurers tend to occur at the same time and

in the same direction during the 2006–2010 period. Baluch, Mutenga, and Parsons (2011)

attribute the increase in the correlation between stock returns of banks and insurers during the

first part of this century to insurers’increased exposure to capital markets and their adoption of

bancassurance models that combine banking and insurance under one roof.

------------------------------------------------------------------------------------

Indeed, the past few decades have seen many countries reduce or eliminate the regulatory

restrictions on banks from engaging in insurance activities and vice versa (hereafter, referred to

as regulatory restrictions). This deregulation in some countries has led to the establishment of

large financial conglomerates that combine banking and insurance businesses in the same

organization.

1

Other countries, however, continue to restrict banks from conducting insurance.

The question addressed in this paper is whether the variation in regulatory restrictions is

associated with the variation in the connectedness of insurers to the banking sector.

2

The internal capital markets literature provides one reason that insurers would be more

connected to banks in countries that allow financial institutions to do both banking and

insurance. This literature indicates that the performance of one part of an organization (say the

banking operations) can influence the management and therefore the performance of other

parts of the organization (e.g., insurance operations).

3

In Section 2 of this paper, we show that

there is also a mechanical reason why the returns on banks and insurers will be more highly

correlated when banks engage in insurance. This explanation is not based on the economic

behavior of managers, but instead, is simply due to banks doing insurance which makes their

returns more highly correlated with other insurers. Importantly, the strength of this mechanical

effect decreases as the “inherent connectedness”of insurers to banks increases.

The inherent connectedness is due to banking and insurance sharing common underlying

economic/financial risk factors that exist independent of whether banking and insurance are

conducted in the same organization. For example, GDP growth or interest rate movements could

influence outcomes in both sectors,or banks and insurers could share common counterparties, or

even contract with one another.

4

The connectedness caused by these common factors will exist

regardless of regulations that restrict banks from having insurance operations and vice versa.

Thus, if banks and insurers are connected (as we and others show they are), but the degree of

connectedness is not associated with regulatory restrictions, then connectedness is likely due to

common factors that exist independent of regulations (i.e., inherent connectedness).

An important aspect of our analysis is the distinction between life insurers and property/

liability (P/L) insurers. Specifically, we hypothesize that life insurers are more likely to share

common factors with banks than are P/L insurers. This hypothesis is based in part on the

evidence and discussion in Cummins and Weiss (2014), where they report that, relative to U.S.

P/L insurers, U.S. life insurers have (a) greater exposure to the housing market via their

investments in mortgage‐backed securities, (b) a higher percentage of assets invested in bonds

of banks and financials, and (c) products, such as annuities and guaranteed investment

contracts, that are substitutes for bank certificates of deposit. Each of these observations is

consistent with life insurers having greater inherent connectedness to banks. Moreover, these

observations suggest that an underlying mechanism generating the greater inherent

connectedness of life insurers to banks is the greater sensitivity of life insurers to interest

rates compared to P/L insurers. Support for this interpretation is provided by Carson, Elyasiani,

and Mansur (2008), who present evidence that life insurer stock returns are more sensitive to

1

The consolidation of different types of financial institutions often includes financial activities in addition to banking and insurance (e.g., securities

underwriting), which can create additional linkages among financial institutions. We focus on banking and insurance. In some European countries, the

majority of life products are sold via bancassurrance entities. For example, in Italy during 2015, 79% of gross written life premiums were sold through

bancassurance. In both France and Spain, the percentage in 2015 was 64% (see Insurance Europe, 2018; Neale, Drake, Schorno, & Semaan, 2012).

2

Increased connectedness between banks and insurers does not necessarily result in greater systemic risk (see e.g., Chen et al., 2014).

3

See for example, Lamont (1997); Scharfstein and Stein (2000); Shin and Stulz (1998); Chiang (2018).

4

We do not attempt to distinguish among these explanations; instead, we view them collectively as the reason banks and insurers are inherently connected

regardless of regulatory restrictions.

394

|

NIEHAUS ET AL.