Cross‐Industry Product Diversification and Contagion in Risk and Return: The case of Bank‐Insurance and Insurance‐Bank Takeovers

• Author: Elyas Elyasiani,Sotiris K. Staikouras,Panagiotis Dontis‐Charitos
• Date: 01 September 2016
• DOI: http://doi.org/10.1111/jori.12066
• Published date: 01 September 2016

CROSS‐INDUSTRY PRODUCT DIVERSIFICATION AND

CONTAGION IN RISK AND RETURN:THE CASE OF

BANK‐INSURANCE AND INSURANCE‐BANK TAKEOVERS

Elyas Elyasiani

Sotiris K. Staikouras

Panagiotis Dontis-Charitos

ABSTRACT

We investigate the impact of domestic/international bancassurance deals on

the risk-return profiles of announcing and nonannouncing banks and

insurers within a GARCH model. Bank-insurance deals produce intra- and

interindustry contagion in both risk and return, with larger deals producing

Elyas Elyasianiis at the Department of Finance, Temple University, Philadelphia,Pennsylvania.

Elyasiani can be contacted via e-mail: elyas@temple.edu. Sotiris K. Staikouras is at the

Department of Finance, Cass Business School, City University, London EC1Y 8TZ, UK.

Staikouras can be contacted via e-mail: sks@city.ac.uk. Panagiotis Dontis-Charitos is at the

Department of Accounting,Finance and Governance, Westminster Business School, University

of Westminster, London NW1 5LS, UK. Dontis-Charitos can be contacted via e-mail: p.dontis-

charitos@westminster.ac.uk. The first author is a visiting professor and Dean’s Fellow at

Hebrew Universityof Jerusalem, Israel and a fellow at the Wharton Financial Institution Center,

Philadelphia, PA. This article was revised when he was on a sabbatical leave from Temple

University at the JerusalemSchool of Business, Hebrew University. He would like to thank both

Temple and Hebrewuniversities for support. He would also like to thank Temple University for

Summer Research Grant in 2010. Earlier versionsof this paper were presented at the Financial

Management Association Annual Meeting 2010; the FMA European Annual Conference 2010;

The INFINITI Conference on International Finance 2010; the British Accounting Association

Annual Conference 2010; the Department of Risk Management, Insurance and Health Care

Management at Temple University, Philadelphia, PA in 2010; the Centre for Banking, Finance

and SustainableDevelopment at the School of Management, University of Southampton,U.K. in

2011; the Jerusalem School of Business at Hebrew University, Israel, 2012; the FMA European

Annual Conference 2012; the FMA Annual Meeting 2012; the British Accounting and Finance

Association Conference 2012; and the Financial Engineering and Banking Society Conference

2012. The authors would like to thank the participants, in particular, Hong-Jeng Abraham,

JuhaJoenvaara,and Sebastia

´n Nieto Parra who servedas discussants, as well as DoronAvramov,

Taufiq Choudhry, Keith Cuthbertson, Ranadeva Jayasekera, Gerhard Kling, Ian McManus,

Mary Weiss, Richard Werner, Zvi Wiener,Simon Wolfe, YishayYafeh, Ling Zhang, Gang Zhao,

and Nan Zhu who provided useful comments.We are also grateful to Elena Kalotychou, Harris

Schlesinger(the co-editor), and an anonymous referee fortheir very useful comments. The usual

disclaimer applies.

© 2015 The Journal of Risk and Insurance. Vol. 83, No. 3, 681–718 (2016).

DOI: 10.1111/jori.12066

681

greater contagion. Bidder banks and peers experience positive abnormal

returns, with the effects on insurer peers being stronger than those on bank

peers. Insurance-bank deals produce insignificant excess returns for bidder

and peer insurers and positive valuations for peer banks. Following the deal,

the bank bidders’ idiosyncratic (systematic) risk falls (increases), while

insurance bidders exhibit a lower systematic risk and maintain their

idiosyncratic risk.

INTRODUCTION

A widespread mode of corporate restructuring within the financial intermediation

industry is the bank-insurance interface, or bancassurance. Arguments in favor of this

structure include diversification benefits, scale and scope economies, efficiency gains,

strength to withstand competition, and managerial discipline through takeover

threats. Skeptics argue, however, that the bancassurance structure is vulnerable due

to conflicts of interest/culture, allocative distortions, regulatory arbitrage, subsidiza-

tion of nonbank affiliates via the bank safety net (e.g., bailouts), affiliation risk (bank

runs due to nonbank affiliate problems), and the creation of financial and political

super powers (Herring and Santomero, 1990; Flannery, 1999; Santomero and Eckles,

2000). A main concern among regulators is whether the risk inherent in these

conglomerates has a greater potential to spill over to other financial firms and,

ultimately, to the real economy (Parsons and Mutenga, 2009), thus causing a systemic

failure.

These concerns resurfaced during the 2007–2009 crisis when a number of financial

intermediaries (FIs) failed or were bailed out at the expense of the taxpayers.

Contrary to the banking sector, the insurance industry came through the crisis

relatively unscathed, mainly due to the ability of insurers to generate new capital

through external funds and dividend cuts (Berry-Stolze, Nini, and Wende, 2014). In

the aftermath of the crisis, the issue of effective regulation and/or supervision of

systemically important FIs (SIFIs), and the need to minimize the moral hazard

effects associated with bailouts became the center of the regulatory debate.

Moreover, the issue of “narrow banking,” that is, whether the range of permissible

activities of FIs should be restricted in order to limit systemic risk, was revived as

a major regulatory concern (Boot and Thakor, 2009; Morrison, 2009) and also

gained ground in the academic literature (De Jonghe, 2010; Wagner, 2010;

Ibragimov, Jaffee, and Walden, 2011; Chen et al., 2014; Cummins and Weiss,

2014). As far as insurers are concerned, the regulatory debate is likely to focus on

strengthening the mechanisms for insurer supervision (Cummins and Weiss, 2014)

and preventing the extension of the government guarantees beyond the banking

sector (Harrington, 2009).

Regardless of the regulatory process, in practice, banks and insurers have witnessed a

considerable level of product convergence. The overlap in the two sectors is especially

apparent in markets where products offered by banks, such as credit-default swaps,

closely resemble a casualty insurance policy, albeit without either an insurable-

interest requirement or a role for an insurance adjuster (Saunders and Cornett, 2008).

These similarities make the bank-insurer interface a natural process, and in a sense, a

“fait accompli” in spite of the regulatory concerns.

682 THE JOURNAL OF RISK AND INSURANCE

A thorny issue concerning policymakers is whether product diversification by FIs

affects their risk-return profiles and/or those of peers operating in the same sector.

Despite the importance of this subject, the academic literature has failed to shed light

on it.

1

Thus, it is important to examine how domestic and international bank-

insurance mergers and acquisitions (M&As) will affect the risk and returns of the

acquiring firms and their peers and what factors determine the magnitudes of the

effects. These matters will be investigated in the current study.

The article contributes to the li terature on five fronts. First, we investigate the effects

of bancassurance deals on the acquirers’ risk-return profile. We employ all publicly

available domestic and cross -border acquisitions satisfying our sele ction criteria

over an extended time period. De al size and the timing of deals before and after the

financial crisis are also take n into account to distinguish any differentia l impact on

acquirers and their peers. Se cond, we broaden the analysis to investigate th e

respective spillover effec ts on both bank and insurer peers, in order to deter mine

whether the wealth effects of su ch M&As are limited to the firms directly invo lved, or

they spill over to peer firms a s well. Third, we examine the changes in total risk

and its systematic and unsystema tic components for the acquiring firms, as well as

associated spillovers to t heir peers, using a risk decomposition approach. Fourth,

we estimate the cross-sectio nal determinants of bidder abnormal retu rns around

the announcement of bank-insuran ce partnerships, in order to identify the factors

contributing to their abnor mal performance. Fifth, we examine the wealth/risk

effects of bancassurance cor porate restructurings within a generalize d auto-

regressive conditionally h eteroskedastic (GARCH) model. This speci fication

accounts for volatility clus tering, nests the traditional asset pricing mo dels and

measures the shock persiste nce on return volatility. The above issues hav e not

been addressed by studies foc using on wealth or risk effects of bank-ins urance

ventures.

Our analysis is motivated by an economic as well as an academic perspective.

Bancassurance ventures may increase or destroy value, with the net effect being

observable in the returns and risks of the involved firms. Similarly, peer firms

may be positively (negatively) affected if a deal favorably (adversely) alters their

competitive position and/or the standing of their industry. This can translate

into positive (negative) returns and/or a reduction (increase) in their risk.

Moreover, the direction of the effect on peers and its magnitude can be seen as

an important indicator of the interdependencies in the industry and is of interest

to regulators seeking to minimize systemic risk. The effects of M&As on peer

firms have been the focus of some earlier studies (Eckbo, 1985; Carow, 2001a). Yet,

these studies look into the reaction of peers to a single event, with the exception

of Eckbo (1985), while no study looks into risk spillover effects on peers. The

article proceeds as follows: The “Relevant Literature” section reviews the literature,

the “Data and Methodology” section describes the methods, the “Empirical

Findings” section reports the findings, and the “Summary and Conclusions” section

concludes.

1

See, the “Relevant Literature” section for a pertinent discussion of the existing literature.

CROSS‐INDUSTRY PRODUCT DIVERSIFICATION AND CONTAGION 683