Current Legal Issues Affecting Central Banks, Volume V

Afterword to Chapters 8 and 9

Robert Effros
Published Date:
May 1998
  • ShareShare
Show Summary Details

Three subjects raised in Chapters 8 and 9 deserve further elaboration. First, what degree of independence should the supervisory authority have? Or, in other words, what should be the relationship between the supervisory authority and the political authority? To be effective, supervision has to be independent from political considerations for reasons similar to those that justify similar independence in monetary policy. As is well known, there is a growing consensus that the monetary authority should be independent from day-to-day political interference. An analogous, even stronger case may be made with respect to supervision, where a number of very complex issues must be addressed, and delicate judgments must be made.

Independence can be described as the ability of the supervisory authority to exercise its responsibility without political interference concerning the judgments that have to be made in the discharge of the supervision function. Needless to say, there are difficult questions to be sorted out when dealing with banks in distress. When the scope of the problem is contained—that is, limited to a few individual bank cases—there is normally little actual disagreement with the notion of independence. This is only natural because, when the magnitude of the problem at hand is small, the mileage to be gained by interference is also likely to be limited; so, why bother? But the situation changes when banking difficulties exhibit the prospect of systemic repercussions, the subject to which I now turn.

What should be the role of the government when systemic threats arise is the second topic that warrants attention. When such threats are perceived, the typical reaction of governments is to come to the protection or to the rescue of the system at large. As indicated in the introduction to Chapters 8 and 9, it is very important to allocate responsibilities properly—that is, to distribute the burden of the safeguard or the rescue equitably. Since this task typically involves public funds, it is only natural for the government to have a view on the matter. In this context, it must be pointed out that, in principle, the supervisory function—which seeks to identify those banks that are vulnerable, to determine their financial position, and to decide which banks are insolvent—needs to be distinguished from the political judgment concerning the need to rescue the banking sector—which can only be made by the government.

In practice, however, it can prove difficult to keep clear-cut, well-defined boundaries between the various functions or to reach consensus on the distribution of responsibilities among official entities. For example, as you know, views differ on whether central banks should be vested with the responsibility for bank supervision in addition to their monetary policy duties. Even where this is not the case, the fact of the matter is that to fulfill the latter duties properly, central banks do have to supervise banks.

As far as bank restructuring is concerned, however, whether or not they carry formal responsibility for supervision, it is critical to ensure that central banks do not provide the financing for such restructuring. Indeed, if by keeping supervision out of the central bank it could be assured that there would be no central bank financing for banking difficulties, this would be a strong argument for separation. But, of course, a similar out-come can be attained by keeping the supervisory authority within the central bank and making clear that no central bank resources will be made available for bank rescues.

How transparent should be the rules governing bank failures is the third subject of interest. Obviously, for individual bank problems the rules should be quite transparent. Legislation and norms on how to handle individual bank failures exist in most countries. However, when it comes to dealing with systemic threats and the exercise of the lender-of-last-resort function, arguments have been made that a measure of “constructive” ambiguity would be desirable, if not necessary. Inevitably, though, there are costs to the pursuit of such ambiguity. No matter the degree of ambiguity that may prevail ex ante, markets are likely to be aware that the central bank will support the banking system when it enters into difficulties by providing liquidity, so that discipline will be lost, and systemic threats may increase in scale and probability. The classic (and transparent) statement of the proper discharge of the lender-of-last-resort responsibility has been made by Walter Bagchot in Lombard Street (1873).1 He argued that central banks should provide support to banks that are illiquid; they should provide support without limit; and they should provide it at a penalty rate. According to him, as long as these principles are well known, confidence will be there because the system will know that the central bank would come in only to support illiquid—not insolvent—banks, and on which terms.

But the gist of the problem lies in the distinction of illiquid banks from insolvent banks. This is another instance of a concept that in theory is clear, yet in practice becomes complicated to pinpoint. In my own mind, rules should be as transparent as possible. From an economic point of view, systemic threats, if allowed to materialize, are likely to have a sizable cost for the economy at large. This is indeed the key reason why governments seek to prevent banking sectors from collapsing; such collapses would bring about severe disruptions in the real economy, with large consequent costs in terms of welfare. In effect, the protection of banking sector soundness is a public good, a public service; and like any other public service, it carries a cost. Therefore, an argument can be made that the use of taxes to protect the stability of the financial sector is proper and due. But what are the precise, unambiguous criteria by which a systemic problem is defined? Here again, there is no categorical answer. This is unfortunate, because there is a typical criticism that will always haunt governments when they act to protect the system from a threat of collapse. When they do it well and the system is kept safe, it is always possible to argue that the problem was not systemic. And yet, had the authorities not intervened, a systemic collapse could have occurred.

Another consideration is that, when in doubt, there is a tendency for governments to act before letting the systemic threat go too far—to err on the side of “well, since it looks systemic, we have to go in.” This tendency can turn out to be costly because at times authorities will have intervened when the threat was not really systemic. Yet, this is virtually inevitable because intervening late is equivalent to accepting the collapse. The difficult issue is to decide to what extent to allow market forces and discipline to be the final arbiters of the seriousness of the threat. The rationale of the New Zealand approach, examined in Chapter 8, is basically to “let the market supervise.” The logical conclusion of this approach would be that market discipline will be left to resolve any systemic problem that arises. Ignored by such an approach would be the market failures, the externalities, and similar arguments that can justify government intervention. The issue at stake here, of course, is one of defining the respective roles to be played by governments on the one hand and by market forces on the other. Needless to say, this has been one of the perennial questions in economics; to demonstrate it, let me quote John Stuart Mill, who, a century and a half ago, wrote: “One of the most disputed questions both in political science and in practical statesmanship at this particular period relates to the proper limits of the functions and agency of governments.”2

    Other Resources Citing This Publication