Using Subordinated Debt as
an Instrument of Market Discipline
Banks or Bank Holding Companies?
A critical decision point for designing an SND policy is whether to apply the policy to banks or to bank holding companies. Most SND proposals seem to apply the policy to banks, and there are strong public policy reasons for doing so. Most fundamentally, insured commercial banks have direct access to the federal safety net, and thus banks are where the dangers of moral hazard and the consequent risks to the taxpayer are concentrated. Therefore, to reduce moral hazard incentives, efforts to increase market discipline should be focused on banks rather than on their parent or affiliated organizations. Also, SND at the bank level could provide increased protection for the FDIC. Finally, an SND policy applied to banks would reinforce the regulatory philosophy that the safety net and associated policies are limited to insured commercial banks. Conversely, applying the policy to bank holding companies would risk encouraging market participants to believe that the safety net extends implicitly to bank holding companies.
Although the arguments favoring a bank-oriented policy are strong, a number of counterarguments, most of which are of a practical nature, support application of an SND policy at the holding company level. One of these arguments builds on the fact that the current market for banking organization SND is overwhelmingly a market for BHC SND. Although banks issue SND, supervisory reports suggest that the vast majority of SND issued by major banks are held by the issuing bank's holding company parent and are not traded. 48 This view is supported by a recent tabulation by Board staff members that indicates that, as of March 9, 1999, only eight of the top fifty banks had SND that were rated by a major rating agency. Because a rating is virtually required for bank SND to be publicly traded, the suggestion is that few banks have traded SND. In contrast, thirty-six of the top fifty bank holding companies had rated SND. These arguments suggest that if an SND policy were applied to banks, even the very largest banks, then the existing market for bank SND would need to be increased substantially.
Although study group interviews with market participants supported the view that primarily bank holding companies currently issue traded SND, there was some indication that requiring large banks to issue traded SND might not be particularly costly for such banks. As indicated, some large banks already issue SND. Moreover, a few major banks provide augmented disclosures at the bank level. In addition, it is conventional for bank SND to trade at a lower interest rate than BHC SND, in part because banks are commonly rated one to two notches higher than their holding company parent. According to Board staff calculations, the typical bank discount is around 8 basis points, but it can rise to much higher levels in times of individual firm or systemic financial stress. Indeed, some market participants asserted that many of the bank SND currently outstanding were issued in the early 1990s, when the banking crisis made issuing SND at the bank level considerably less costly than issuing them at the BHC level. As the spreads between bank and BHC SND have declined during the 1990s, the advantages of issuing at the BHC level, such as increased flexibility in allocating funds within the total organization, have come to dominate the interest cost disadvantage.
A second reason for applying an SND policy at the BHC level is that, at least today, banks tend to dominate the holding company even at the largest banking organizations. This tendency is seen in the Bank assets/BHC assets column of table 7, which gives the estimated percentage of total BHC assets accounted for by the banking assets of each of the top fifty bank holding companies. For the top fifty bank holding companies, the estimated ratio of bank assets is greater than or equal to 95 percent at twenty-seven of the forty-one institutions for which data are available. However, there are some notable exceptions to this tendency (for example, CitiGroup at 36 percent and J.P. Morgan at 41 percent), and significantly, the exceptions are concentrated among the very largest bank holding companies. For example, among the top twenty bank holding companies, only eight have a banking ratio of 95 percent or more; and among the top ten, only one holding company (First Union) meets this criterion.
Study group interviews with market participants suggested that today the market makes little distinction between the bank and the bank holding company, beyond the basis point differences in interest rates implied by a one or two notch difference in debt ratings. In particular, market participants claimed that distinguishing differences in credit quality between the two parts of the organization was difficult. However, some speculated that the CitiGroup model, if it became more widespread, might significantly change the way analysts look at widely diversified bank holding companies. Some interviewees suggested that the CitiGroup model might facilitate issuance of SND by the bank because such a structure would force analysts to look more carefully at each major component of the holding company.
A third reason for applying an SND policy at the BHC level derives from the fact that various sources indicate that bank holding companies are often, and perhaps normally, managed on a ''product'' or a ''business line'' basis, with relatively little attention paid to any one legal entity. Recognition of this evolution is an important motivation for many current efforts at supervisory and regulatory reform, including revision of the Basel Capital Accord. Such a management approach by bank holding companies suggests that continued, much less increased, focus by bank supervisors on the legal entity of the bank may become more and more unrealistic over time and could impose significant costs on banking organizations as they were forced, in essence, to keep one set of books for their supervisor and another for their internal management and external market purposes. However, if the CitiGroup model becomes more common, application of SND or other supervisory and regulatory policies at the holding company level may be quite unattractive for several reasons, including the possibility of implying an expansion of the safety net.
A final point on the issue of whether an SND policy should be applied to the bank or the bank holding company relates to the potential impact of a bank-only policy on the likelihood that the parent holding company would continue to choose to offer SND and to the cost of reduced SND issuance at the parent level. Bank SND that were issued to third parties would, when the holding company's books were consolidated, also count as SND at the holding company level. Thus, a bank-only SND policy would probably reduce the need for the holding company to issue SND. As a consequence, while the quality of the market discipline (including the signal provided by SND prices) would be enhanced at the bank, the degree of market discipline at the bank holding company could be reduced. 49 In particular, the information content of BHC SND prices might decline. At a time when a less-intensivesupervisory regime might be in place for financial services holding companies, degradation of the quality of the price signal for the overall holding company might be costly to supervisors.
The last issue examined in this subsection is, assuming that an SND policy is applied at the bank level, how banks in a multibank holding company should be treated. Study group discussions with market participants suggest that today market participants focus on the lead (largest) bank in a multibank holding company and are aware of the cross-guarantee provisions of FIRREA. 50 Thus, limiting an SND policy to the lead bank might result in the most effective market discipline with the least disruption to current market practice. On the other hand, if an SND policy were applied to a limited number of very large and relatively complex banks, that a bank was part of a multibank holding company would appear to make little difference as to whether the policy should apply to a given bank. In any event, the easing of interstate branching restrictions is likely to reduce over time the number of multibank holding companies for which this would be an issue.