To Join or Not to Join? Do Banks That Are Part of a Financial Holding Company Perform Better Than Banks That Are Not?

Article excerpt

I. INTRODUCTION

Following in the footsteps of the United States in terms of its financial regulations, in the year 2000 Taiwan started to allow banks, securities, and insurance companies to consolidate and form financial holding companies (FHCs). During the 4 yrs that followed, 14 FHCs were established and 13 banks joined 12 of the FHCs, (1) whereas 35 banks chose not to join. Under the regulatory framework for FHCs in Taiwan, an FHC is considered to be a conglomerate that combines at least two or three financial institutions, ranging from banking to securities-related business and insurance. A question that is commonly asked by policy makers, bankers, and academics is: Do banks that join an FHC (FHC bank hereafter) perform better than banks that do not (independent bank hereafter)? This phenomenon is referred to as the joining the FHC effect.

Supporters of the joining the FHC effect argue that FHC banks which engage in deposit-loan activity are able to facilitate the efficient provision of other financial services such as insurance or the underwriting of securities (Diamond 1991; Rajan 1992; Saunders and Walter 1994; Stein 2002). Similarly, securities and insurance underwriting, brokerage and mutual fund services, and other activities can produce additional information that improves loan-making decisions. Thus, a bank within a financial conglomerate enjoys the so-called 3Cs. namely, cross-selling, cost savings, and capital sharing, and benefits from economies of scale and scope that boost performance and market valuations. This diversification effect could occur only in the case of FHC banks. In addition, under the financial holding company structure, a service bank is theoretically able to provide customers with one-stop shopping for banking, insurance, and securities services.

An alternative view suggests that there is no joining the FHC effect, as independent banks may have a specialization advantage and lower systemic risk. (2) For example, some independent banks are specialized in credit card or custodian business, in which case joining the FHC has no particular advantages for them. In addition, an independent bank could have lower systemic risk because an FHC bank with diversified activities will reinforce the agency problem between insiders and outsiders. In particular, this conflict of interest tends to occur in an FHC bank when the bank makes a loan to the firm and also sells its securities (Kroszner and Rajan 1994; Puri 1996; Gande et al. 1997; Schenone 2004). Furthermore, Laeven and Levine (2007) found that the diversification discount, not the diversification premium, exists in a financial conglomerate, implying that the synergy effect is small. Stiroh and Rumble (2006) demonstrated that the diversification effect in a U.S. FHC is negatively affected by the bank's involvement in noninterest income activities. Therefore, specialization advantages, low agency costs, and diversification discount suggest that the independent bank can perform better.

There are few empirical studies comparing the performance of these two groups, probably because the direct comparison between the two groups has often been criticized, as banks with superior financial performance are more likely to be invited to join an FHC. Thus, it would not be surprising to find their continued superior performance well after they join the FHC. For example, Shen (2002) used descriptive statistics to examine CAMEL indicators of 50 U.S. FHC banks and 44 U.S. independent banks between 1997 and 1998, and found that, on average, the former outperform the latter. Hsu and Chang (2005) used ordinary least squares (OLS) to study the impact of the passage of the FHC Act on bank operational efficiency in Taiwan. They found that FHC banks outperformed independent banks both before and after the passage of the FHC Act. Thus, the above studies demonstrated that the FHC banks outperform independent banks most likely due to the endogeneity problem. …