Since the late 1980s national authorities from the G-10 and some other countries have engaged in increasingly complex national and international regulatory reforms in the banking sector (Simmons 1999; Lütz 2000). The aim of regulatory activity, a large part of which concentrates on bank capital-asset ratios (CAR), has been to mitigate bank solvency problems that may destabilize national and international financial systems. The motivation behind the proliferation of regulatory activity is found in the presumption that the banks, if left alone, would select to remain undercapitalized relative to the socially optimal level.
In this paper we attempt to explicitly evaluate the validity of this popular presumption. We first construct a model that explains how banks select their capital-asset ratios. The model explains why and how competition among banks can support high capital-asset ratios. It suggests that "market discipline" (competitive forces) can not only assist the implementation of capital adequacy regulations but may even substitute for such regulations.
In the absence of explicit or implicit, catholic –that is, covering all types of bank liabilities- guarantees by the government, banks may have an incentive to satisfy "implicit capital adequacy requirements" in order to lower their riskiness and hence create liabilities at more favorable terms. We show that whether the banking sector is under- or overcapitalized relative to the social optimum depends on the relative importance of idiosyncratic bank risk (the risk profile of individual banks) and systemic risk (the "healthiness" of the entire banking sector) for banks' borrowing costs. Undercapitalization - and hence a need for regulation - is more likely to occur when systemic bank risk dominates. We also argue that the presence of other government interventions such as government deposit insurance and government-led bank bailouts makes under capitalization even more likely.
We then examine whether market forces have acted as a substitute for capital adequacy requirements by linking differences in borrowing costs across banks to differences in banks' own and industry-wide capital-asset ratios. The results of pooled cross-section times series regressions with fixed effects for more than 15,000 U.S. banks in the 1990s (approximately 130,000 bank-years) show that better capitalized banks experienced lower borrowing costs. The analysis also demonstrates, however, that competition cannot substitute for capital adequacy regulation because systemic effects dominate. Although these findings suggest that banks are under capitalized relative to the social optimum, banks may be overcapitalized relative to the official regulatory requirements when the latter fall short of the socially optimal levels. This possibility cannot be ruled out given that the existing capital adequacy requirements seem rather ad hoc. That is, these requirements have not been derived from explicit social objective functions.
We draw two conclusions from these findings. First that capital adequacy regulation may indeed be socially desirable. And second, regulation aiming at creating and sustaining competition among banks, notably through increased transparency, can play an important role in mitigating bank solvency problems. Consequently, ongoing reform efforts at national and international levels should focus strongly on increasing transparency and strengthening competition among banks as an efficient way of promoting higher capital-asset ratios and greater bank solvency.
