Brookings-Wharton Papers on Financial Services 2003 (2003) 141-193
[Access article in PDF]
Risk Measurement, Risk Management, and Capital Adequacy in Financial Conglomerates
Andrew Kuritzkes, Til Schuermann, and Scott M. Weiner
IS THERE SOMETHING SPECIAL, with respect to risk and capital, about a financial conglomerate that combines banking, insurance, and potentially other financial and nonfinancial activities? To what degree is the risk of the whole less than the sum of its parts? This paper seeks to address these questions by evaluating the risk profile of a typical banking-insurance conglomerate and highlighting some of the key analytical issues relating to risk aggregation. We provide a framework of analysis that builds up this risk profile and allows for coherent dissection of a financial conglomerate's diverse set of risks. We then use the analytical results to frame a policy debate for regulating the solvency of a multi-line financial conglomerate.
Increasing industry consolidation, financial deregulation, and globalization are fueling rapid growth in the scope of large, multi-line financial [End Page 141] conglomerates. The Joint Forum of the Bank for International Settlements (BIS) defines a financial conglomerate as "any group of companies under common control whose exclusive or predominant activities consist of providing significant services in at least two different financial sectors (banking, securities, insurance)." 1 In fact, it is fair to say that virtually all of the large, multinational financial institutions operating around the world today are, to some degree, financial conglomerates by either the strict "three of three" or the weaker "two of three" definitions.
Yet despite the increasing importance of financial conglomerates, the regulatory structure in most jurisdictions is still based largely on single business lines, on the notion that financial institutions, or their business lines, can be classified as banks or insurance companies or securities firms. 2 This is particularly true of the regulatory approach to capital adequacy, which historically has treated different regulated businesses as independent "silos" in setting capital requirements.
Analytically, the special problem of capital management in a conglomerate stems from the need to aggregate risks across diverse business lines, such as banking and insurance. Our suggested method for constructing a composite picture of risk is to follow a "building block" approach that aggregates risk at three successive levels in an organization (corresponding to the levels at which risk is typically managed):
- Level I. The first level aggregates the stand-alone risks within a single risk factor (for example, credit risk in a commercial loan portfolio).
- Level II. The second level aggregates risk across different risk factors within a single business line (for example, combining the asset, liability, and operating risks in property and casualty or life insurance).
- Level III. The third level aggregates risk across different business lines (for example, banking and insurance).
Our analysis adopts this building block approach because it isolates the incremental effects of cross-business diversification that are unique to a conglomerate.
Empirically, our findings show that diversification effects are greatest within a single risk factor (level I), decrease at the level of a business line (level II), and are smallest across business lines (level III). According to [End Page 142] our estimates, the incremental diversification benefits achievable at level III by combining a bank with an insurance company are on the order of a 5-10 percent reduction in capital requirements, depending on the business mix.
In such a building block approach, the result of adding risks at the level of the holding company can be no more accurate than the stand-alone capital measures for individual risk factors (level I) and for risks within a single business line (level II). This has important implications for capital regulation. The existing regulatory capital measures at levels I and II are inherently limited and, in particular, fail to reflect significant diversification effects at those levels.
Because of the limitations of existing capital measures, we argue that a well-constructed economic risk and capital management framework offers the best chance of overcoming the problems of the silo approach to...