There has been a significant amount of debate recently about reforming the regulatory structure of the financial services industry - and for good reason.
In a move reminiscent of the 1930s, the governor of Rhode Island early this year declared a state banking holiday for institutions insured by a private insurance fund whose solvency was in question.
Taxpayers were told the cost to "rescue" the Bank of New England would be as much as $2.3 billion. It was the fourth large-bank rescue since 1984 estimated to cost the insurance fund over $1 billion.
FDIC reserves per dollar of insured deposits were at an all-time low prior to the recently legislated recapitalization, and the fund may be in worse condition than many perceive if existing losses at operating but insolvent or soon-to-be insolvent institution are considered.
All of this comes on top of the savings and loan debacle, which will probably cost taxpayers something approaching $200 billion.
Bank Reform Fizzles
It is typically during such catastrophic times that major legislative reforms are enacted. However, hopes for broad reform were dashed this year by the passage of a relatively narrow banking act.
The bill failed to allow broader geographic expansion or expand allowable activities to let banks offer additional insurance and securities services.
It also failed to address many of the problems associated with deposit insurance.
Given these shortcomings, there is little doubt that reform of the financial services industry will be come up again in the future. While this may not occur during a presidential election year, the problems of the industry will persist and eventually have to be addressed. Therefore, it is all-important that reform proposals continue to be evaluated.
Banking Basically Simple
In theory, the business of financial intermediation is fundamentally simple. Bankers purchase short-term funds in the marketplace and transform them into earning assets with longer maturities.
Absent deposit insurance, the providers of the working capital - the depositors - are at risk.
If the quality of assets declines then the depositors either demand a higher return on their investment commensurate with the increased risk of simply withdraw their funds - which may lead to runs.
Policymakers, however, fearing that bank runs may spread from bad to good banks, may try to head these withdrawals off. The policymaker, in effect, does not allow the disciplinary influence of deposit runs to be realized.
One way to stem the potential for these runs is to introduce deposit insurance.
In Lieu of Market Discipline
Obviously, if all deposits are insured, depositors will have no incentive to discipline banks. This necessitates oversight by regulators to substitute for the disciplining influence of depositors. However, this is costly and, as recent events show, not totally effective.
Another alternative is to limit deposit insurance to some subset of depositors, allowing the remaining ones to impose the necessary discipline.
The problem with this approach is that any subgroup of uninsured depositors large enough to discipline banks is also large enough to pose contagion problems that concern the policymakers.
This is essentially the current deposit insurance situation, in which holders of large deposits are relied upon to provide discipline.
However, contagion fears have prompted regulators to intervene and protect these depositors from loss.
This suggests that the ideal situation would be one in which discipline is imposed by the marketplace, but without any bank deposit runs.
Subordinated Debt a Solution
It is possible to devise such an environment. In fact it is quite easy. It can be achieved by using subordinated debt to cushion the deposit insurance fund and to impose discipline on financial firms.
In most nonbank firms equity alone serves the role of capital. In banking, however, subordinated debt serves the most fundamental role of capital as well as, or perhaps better than, equity. This role is as a buffer for depositors and the insurance fund against income variations.
And subordinated debt may serve this role better than equity. This is because, unlike shareholders who make a decision to accept higher risk levels as a result of mispriced deposit insurance, the debtholders do not stand to gain from increased risk.
A regulatory reform relying heavily on the use of subordinated debt to augment regulatory oversight is superior to the alternative proposals under consideration.
Underlying the proposal are the following premises:
* Market forces can serve as an effective complement to regulatory discipline.
* Stability of the banking system (not of its components) is of paramount concern.
* The current means used to price deposit insurance (risk invariant, flat rate premiums) and resolve bank failures are inadequate.
My proposal would require that a significant portion of the total capital held to satisfy the current risk-based capital requirements take the form of subordinated debt.
For example, the 8% mimimum capital requirement could be restructured to require a minimum of 4% equity and 4% subordinated debt.
The recommendation is not that banks be allowed to hold subordinated debt to meet part of their capital requirements. Rather, banks should be required to hold a significant portion of their capital in the form of subordinated debt.
This is not simply a recognition that subordinated debt can serve as a capital cushion, it is an attempt to fully utilize its superior attributes.
How It Would Operate
In order to provide bankers with appropriate incentives to control risk, the new debt would have to have certain characteristics.
It would have to contain covenants stating that it would automatically be converted into an equity stake in the bank if equity capital were exhausted.
To be conservative, the bank's dividends, growth, and deposit rates could be restricted if core (equity) capital fell below some minimum level (perhaps 2% of risk-weighted assets). And the new debt would have to carry maturities sufficiently short that the bank would be required to go to the market to roll over debt on a regular, ongoing basis.
The maturity structure is very important. There is an obvious trade-off between the increased cost resulting from having to reissue debt frequently and decreased market discipline if banks are not required to approach the market very often.
The maturity structure would have to be long enough (perhaps five years) to tie the debtholders to the firm and make the inability to run meaningful. Maturities would also have to be staggered sufficiently to enable the firm's behavior to be disciplined by the necessity to approach the market often - perhaps semiannually.
A Healthy Influence
This relatively minor adjustment in the capital structure of banks would alter the disciplinary influence to which they are subject.
Because holders of the subordinated debt would obviously be risk-sensitive, they would continuously monitor bank behavior and would demand a higher interest rate from riskier banks.
Because the size of the equity cushion would affect debt prices, banks may increase equity capital levels in their efforts to minimize the total cost of capital. That is, over some range, additional equity capital could lower the cost of subordinated debt by an amount sufficient to offset the costs of the additional equity.
Regarding this issue we do know that banks held significantly higher levels of capital before the introduction of federal deposit insurance greatly reduced depositor's incentives to monitor bank risk-taking.
The introduction of subordinated debt, by restoring market discipline, would move the industry back towar those earlier conditions. Therefore, banks may chose to hold higher than minimum levels of equity capital.
Concern About Bank Runs
The relatively minor adjustment would also alter significantly the view of regulators. Concerns about contagious deposit runs would be significantly decreased. Obviously the debtholders could not run.
Also, though contagion is commonly discussed in terms of runs from "bad" to "good" banks, the typical concern is about runs from "bad" to "better, but not-too-good" banks. With increased discipline imposed by the debtholders there would be fewer "not-too-good" banks. Thus, concerns about contagion would become much less important.
The true value of any proposal, however, lies in its success during times of stress. Under this proposal, moral hazard problems would be minimized by the continuous application of discipline by debtholders, thereby decreasing the probability of bank failure.
However, even as insolvency approaches, discipline would be applied slowly and methodically, rather than through a run on deposits. During this phase, maturing debt could not be rolled over.
Protecting the Insurance Fund
The role of depositors as a cushion to the insurance fund would decline substantially as the new debt cushion was introduced. The new cushion would protect the insurance fund (and the public) regardless of the extent of the insurance coverage.
With the debt proposal in place, as outstanding debt issues came due, refusal by the market to accept new debt would be a clear signal of a solvency problem.
Once the bank's debt capital fell below the required level, existing subordinated debtholders would be given an equity position and would have a prespecified period, such as 6 months, to recapitalize the bank or find an acquirer. Failing that, the firm would be liquidated.
It would also be necessary during this period to restrict the banks' growth, devidend payouts, and deposit rates.
Potential losses to the insurance fund would be reduced under the proposal because there would be a specified group to absorb losses once equity was eliminated. This group would have both strong incentives to avoid forbearance and its associated costs and the power to place the bank in liquidation.
Since debtholders would provide an additional layer of protection for deposits, the depositors would have no reason to withdraw their funds.
Orderly Failure Resolution
Consequenlty, and perhaps the most desirable feature of this proposal, failure resolution can proceed in an orderly manner.
There are actually several alternative events that could trigger the recapitalization process. First, and most obvious, the inability of the bank to roll over maturing debt would indicate that the market believes the equity cushion has been regulators.
Alternatively, if the regulators had superior information concerning the viability of the bank, they could require recapitalization even when the market was willing to accept the new debt.
Finally, the debtholders could be allowed to force recapitalization or closure via court petition. These triggering mechanisms are not necessarily mutually exclusive.
Compared with Other Proposals
How does the effectiveness of the subordinated debt proposal compare with that of other proposals for regulatory reform being considered in the current policy debate?
Many of the other proposals are designed to increase reliance on market forces to oversee bank behavior. The subordinated debt proposal would accomplish this as well, but would do so without the much-feared side effect of deposit runs.
Additionally, it would eliminate the inequitable treatment of uninsured depositors at too-big-to-fail banks.
Early Closure by Regulators
One proposed solution calls for earlier closure as a means to minimize (or eliminate) losses to the insurance fund.
The debt proposal provides a mechanism to achieve that and more. With early closure, you have regulatory discipline constraining the behavior of the bank as it reaches a certain capital threshold; the bank is finally closed at some positive level of net worth.
The subordinated debt proposal eliminates the arbitrary threshold and continuously applies increasing pressure to banks to manage risk as their capital position deteriorates.
Finally the debt proposal would appear to be more politically palatable than reducing deposit insurance coverages and more easily implemented than regulator determined risk-based insurance premiums.
However, it should be emphasized that subordinated debt would actually complement most of the alternative proposals.
There will no doubt be skeptics. Therefore, let me address some of the most commonly stated concerns regarding the effectiveness of the debt proposal.
First, what is to prevent the value of the firm from deteriorating quickly and "blowing through" both the equity capital and subordinated debt and impacting depositors? There is nothing in the proposal that guarantees that this could not happen.
However, evidence suggests that this is not how bank failures typically occur. Market values of assets do not change that quickly. During the 1980s, Texas land values dropped significantly, but the decline occurred over a number of years. And none of the major bank failures of the past decade was really a surprise.
Second, even if the proposal worked well for the large banks, would small banks be able to issue the new debt? Although this question remains open, there is reason to believe that they could do so relatively easily.
In fact, for small closely held banks, issuing subordinated debt would probably be preferred to issuing equity.
But even if small banks found it unfeasible to issue subordinated debt, their exclusion would not be fatal to the proposal. For example, the proposal could be limited to institutions over a certain asset size (perhaps $1 billion) and/or banks wishing to utilize expanded bank powers.
Tying it to expanded powers could eliminate the concern of those who fear that the solvency of the insurance fund would be threatened if additional product powers are allowed under the current deposit insurance structure.
Third, what is to prevent regulators "rescuing" subordinated debtholders just as they have insured depositors when a large bank encounters difficulty?
Indeed, it is imperative for the success of this proposal that debtholders react as if they are subject to losses. Otherwise, they will not exert discipline on the bank.
However, there is less reason under this proposal for regulators to intervene and keep the regulatory program from "working" than under alternative programs.
Regulators would have little incentive to avoid imposing losses on debtholders because they cannot start a run. Hence, a too-big-to-fail policy based on the fear of imposing loses on large customers, interbank accounts, and such, and the associated systemic risk concerns would no longer exist.
If a Bank Reduces Assets
Fourth, what if the bank meets its capital requirement by simply shrinking? By sufficiently reducing assets the bank could position itself to avoid having to approach the marketplace, but still meet all regulatory capital requirements.
This is particularly worrisome if, as can be expected, the problem bank accomplishes this by selling off its best assets (and retiring some liabilities).
While a bank could choose this approach, to do so would most likely be seen as a clear indication that the bank was encountering difficulty.
While holders of outstanding debt may attempt to sell their holdings, resulting in lower prices, the debt could not run from the bank.
Finally, the most serious question is whether debtholders would actually impose discipline on the bank. Financial and economic factors suggest that they would.
Given recent events, it is highly likely that regulatory reform of the financial services industry will occur relatively soon. Comparing alternative bank regulatory reform proposals, the use of subordinated debt would appear to merit placement at the top of the list.
