The ‘eighties: Liberalization and crisis
Until the mid 1980s, New Zealand had just four banks – one fully owned by the government, one owned by Lloyds Bank in the UK, and two owned by large Australian banks. In the mid 1980s, as the New Zealand economy experienced a far reaching liberalization across virtually every area of policy, non-bank financial institutions were allowed to convert to full banking status, and foreign banks were allowed to establish branches or subsidiaries. By the time I became Governor of the Reserve Bank of New Zealand in September 1988, there were a total of 15 registered banks, including the four “original” banks.
The Reserve Bank had responsibility for issuing licences to the new banks – the original four were “deemed” banks by legislation, and so did not need to get a licence – and for supervising all of them. In keeping with the policy environment in New Zealand at that time, supervision was light-handed, and regulation even more so. We didn’t dream of on-site inspection and were relaxed about whether foreign-owned banks operated as subsidiaries incorporated in New Zealand or as branches of the foreign parent. When I first arrived at the Bank, there was no limit on risk concentration.
I was particularly surprised at the absence of any limit on risk concentration, having just come from a commercial banking environment, and soon after my arrival we instituted a limit of 35% of bank equity for any single counterparty.
But then in the late 1980s New Zealand was hit by recession. In part, this was a result of an aggressive attack on inflation by the central bank and in part a result of a dramatic fall in share and property prices following the October 1987 crash on Wall St. (The fall in share prices was much more abrupt in New Zealand than in the US.) The government-owned bank, the Bank of New Zealand, suffered severe loan losses and would almost certainly have failed had the government not stepped in and injected additional capital. One of the new banks, owned by a foreign-owned insurance company, would also have failed had it not been for massive support from its British parent. And a near-bank, DFC, did fail, the largest failure of any financial institution in New Zealand history. The failure of DFC was particularly traumatic for financial markets because until a year or two earlier it had been wholly owned by government, and its “privatization” was seen by many of its creditors (which were mainly US and Japanese banks and other institutions) as still leaving the government primarily responsible for the institution (80% of the shares were purchased by the National Provident Fund, a government agency responsible for providing pensions to many thousands of public servants).
Supervisors vs Economists: Towards a new regime
The more or less automatic reaction of those of us in the Reserve Bank was that we needed to intensify our supervisory activities, and get a great deal more information from the banks than we had been receiving previously. Where we had been getting information from banks quarterly, we moved to get it monthly, and in some cases even more frequently. We adopted the Basel I rules for minimum capital, both Tier 1 and Tier 2.
But some of the senior economists at the Bank began asking questions about the rationale for what we were doing. In particular, they asked whether the additional information we were seeking from banks was intended to be an alternative to more intensive regulation or as an addition.
And thus began an internal debate which raged for several years. The economists tended to argue that supervision and regulation were largely futile and could even be counter-productive. They pointed out that in the almost five decades since World War II the only major banks which had failed in the developed world had failed for one of two reasons. First, they may have failed because of fraud – think BCCI (or a little later, Barings) – and bank supervisors almost never detect fraud, especially if there is collusion between bank management and bank customers. Second, they more often failed because of a dramatic fall in asset prices, usually property prices – and bank supervisors seem no better able to anticipate those than the management of banks. Therefore, claiming to be able to substantially reduce the probability of bank failure risks damaging central bank credibility when the inevitable bank failures eventually come. The economists argued that compelling banks to make more disclosure to the public could be a substitute for tighter regulation.
Moreover, the economists were worried that gathering substantial amounts of confidential information from banks significantly increased the problem of moral hazard. As supervisors, we – but not bank customers or the wider public – would know about the strength of the banks and would feel some responsibility if a problem began to emerge. We would be “fixed with knowledge” and face all the wrong incentives.
On the other side of the debate, our bank supervisors were absolutely sure that regulation and supervision of banks was essential and cited the usual arguments of asymmetric information and systemic risk. They were convinced that our regulation and supervision were essential to keep the banks sound, and that without our oversight the banks would quickly get into trouble – as indeed, they had done in the late eighties.
An economist by background myself, I was sympathetic to the argument of the economists, and I was strengthened in that view when one of my predecessors as Governor told me he thought that the Bank was on a hiding to nothing by carrying responsibility for bank supervision. When things are going well, the Bank will get no credit; but when things turn sour, the Bank will be blamed!
In September 1992, a chance conversation strengthened my conviction that the economists were right. I was a guest at one of the innumerable dinners which mark the annual meetings of the IMF and World Bank in Washington and found myself sitting beside a man who had had a very distinguished career in the UK Treasury. He had just been appointed a director of one of the largest British banks. I asked him how he found being a bank director after a lifetime of working in the Treasury. Funny you should ask that, he replied. Banking is all about measuring and pricing risk, and in the Treasury I haven’t had experience of that. So I was greatly relieved to discover that all I had to worry about as a director of the bank was whether we were complying with the Bank of England’s rule (this was before the creation of the FSA of course).
In the end, we announced a new regime to take effect from the beginning of 1996. We retained the minimum capital rules from Basel I because we weren’t quite brave enough to totally reject the international orthodoxy, and we retained rules on related-party exposures. But we scrapped any limit on credit concentration; we had no minimum requirements for liquidity; we had no rules on foreign exchange exposure; we had no rules about risk control systems; we did not reserve the right to approve bank directors; we still had no on-site inspections.
Instead, we had two things. First, we required a high level of financial disclosure not just to the central bank but also – and primarily – to the market, on a quarterly basis. So, for example, banks had to disclose the extent of their risk concentration – how many exposures they had to individual counterparties which exceeded 10% of equity, how many which exceeded 20% of equity, and so on. Similarly, banks had to disclose the extent of any open foreign exchange position, and the maturity structure of their assets and liabilities. We reasoned that the obligation to disclose this information would in itself impose a strong discipline on the banks without the Reserve Bank having to specify some arbitrary rules.
Secondly, we required all bank directors – executive directors and non-executive directors – to sign off these quarterly disclosure statements, attesting to their belief both that the information therein was accurate and that the bank’s risk control systems were appropriate to the nature of the bank’s business.
These director attestations prompted a reaction. Shortly before they were implemented, the managing director of the Australian parent of one of New Zealand’s largest banks flew to Wellington to protest that it was quite wrong of us to require these attestations. Most bank directors, he claimed, know absolutely nothing about banking, so it is totally unfair to expect them to sign these attestations.
I had to accept his statement that many bank directors did not know a lot about banking – and that was partly because the bank supervisor had been willing to make key decisions about capital adequacy, risk control systems, and all the rest for them. Too often banks had appointed people as directors because of their strong connections in the corporate world, or their reputation for probity and integrity, and while those things are clearly advantageous, it was also, in our view, important that bank directors took primary responsibility for bank governance.
Thirdly, we required all banks to have an independent chairman and at least two independent directors – independent from any parent bank that is. We wanted to ensure that, if the bank group of which the New Zealand operation was a part got into difficulty, there were at least some strong independent voices at the board table to look out for the interests of the New Zealand operation.
Disclosure and governance
Has this approach to bank regulation worked? It’s impossible to be dogmatic because of course nobody knows what the counterfactual would have been.
But we do know that banks themselves keep a close eye on the disclosure statements of other banks, and this has to be constructive. Shortly after the new regime was introduced, the New Zealand subsidiary of an American bank was forced to disclose in its quarterly disclosure statement that it had lent more money to its parent than it had equity in New Zealand. Another bank protested vigorously to that subsidiary because it had had very substantial funds on deposit with the subsidiary at a time when the latter effectively had a negative net equity position in New Zealand. To the best of my knowledge, the American bank never repeated that behaviour.
There is no doubt in my mind that the requirement for bank directors to make quarterly disclosure statements has improved the quality of bank governance in New Zealand. Early on, it was established that signing the attestations does not require bank directors to become internal auditors, but it does require them to put in place policies and procedures designed to give them a high degree of confidence that what they are being asked to sign is in fact accurate and reliable. And I have no doubt that that has strengthened bank governance. Some years after I resigned as Governor of the Reserve Bank in 2002, I became a director of the largest commercial bank in New Zealand – and I can personally confirm that bank directors take signing these attestations very seriously indeed!
As an aside, the disclosure statements come in two forms – a so-called Key Information Summary of no more than three or four pages, which must be available on demand in every branch of the bank; and a much longer and more comprehensive statement, which must be provided within five days of any request being made. This reduces the number of forests that must be felled in producing the more comprehensive statement in a situation where the number of people who will actually read it carefully can almost certainly be counted on the fingers of two hands (though maybe a few toes may be needed also). On reflection, I think the Bank (and I take full responsibility for the decision) required an excessive level of detail in the comprehensive statements: the sheer volume of that detail probably obscures more than it reveals to all but the most diligent and determined of observers.
But can it be claimed that the New Zealand approach to banking regulation and supervision has actually worked? If an improvement in bank governance is the criterion for making that assessment, I would argue that it has done so, as noted above. And there haven’t been any bank failures since the regime was introduced in the mid 1990s.
But New Zealand is not a good laboratory to test the regime. Until the last few years, almost every bank in the country has been a subsidiary or a branch of a major international bank, and the four largest banks, which overwhelmingly dominate the New Zealand banking sector, are all wholly-owned by financially strong Australian banks. Those Australian “parents” are supervised by the Australian Prudential Regulation Authority, which operates an entirely orthodox supervisory regime. The fact that there is a widespread assumption that the Australian parent banks would support their New Zealand subsidiaries means that very few people, even very few financial analysts, pay any attention at all to the quarterly disclosure statements.
Moreover, since I left the Reserve Bank in 2002 the New Zealand regime has drifted gradually towards a more orthodox regime also. The Bank still does not conduct on-site inspections but it does, for example, now stipulate that a minimum percentage of every bank’s funding should come from retail or long-term wholesale funding to reduce reliance on short-term (and therefore potentially fickle) wholesale funding. And all directors and senior management of banks must be approved by the central bank. I have a good deal of sympathy for putting in place rules around the maturity structure of bank funding, given the vulnerability that banks which are heavily dependent on short-term wholesale funding have. I am much less convinced of the merit of requiring a central bank tick for all bank directors and senior management: if something should go wrong with the bank, depositors could with some justification come to the central bank demanding compensation.
Dealing with foreign control
The Bank’s thinking has evolved in other areas also, during my own time as Governor and subsequently. When I first became Governor, we were entirely relaxed whether a foreign bank operating in New Zealand chose to incorporate locally or simply operate as a branch, without local incorporation. We could see considerable merit in having foreign-owned banks operating as a branch because this enabled the local operation to derive the full benefits of the parent’s balance sheet.
This indifference began to change after the failure of BCCI in 1991. We saw central banks and other official organs put their hands around whatever assets of BCCI happened to lie within their jurisdiction, regardless of legal form, but that appeared to be somewhat easier to do without legal challenge when the operations were incorporated locally.
Then we became focused on the implications of the Australian Banking Act: that legislation provided that, in the event of an Australian bank being wound up, a priority would be given to Australian depositors of the bank at least insofar as the Australian assets were concerned. But what constituted the “Australian assets” of the bank? Could a court find that the “Australian assets” included assets owned through a branch structure in New Zealand? To make matters worse, US legislation was changed during the nineties providing the same kind of priority to American depositors in the event of a US bank being wound up. What would happen to New Zealand depositors with the local branch of a US bank which got into trouble?
While the Reserve Bank had (and still has) no legal obligation to protect bank depositors (our responsibility was to protect the stability of the banking system), we decided that the political implications of a situation where an Australian or American bank in liquidation was able to pay out its own national depositors in full but able to pay nothing or very little to New Zealand depositors would be intolerable. In the end, we decided that all foreign banks which engage in substantial retail banking business in New Zealand and which are based in countries which give their own national depositors a priority in the event of the bank’s being liquidated must be incorporated locally. We decided that banks which are only engaged in commercial or corporate banking can choose to incorporate or not as they please, on the assumption that commercial and corporate customers are mature enough to make an informed judgement about the risks themselves. (Of course, banks which come from countries where all depositors, whether local or foreign, are treated pari passu have no obligation to incorporate in New Zealand, whatever the nature of their banking business.)
As a consequence of this policy, one large Australian bank which had operated in New Zealand for more than a century as a branch was obliged to incorporate locally, while an American bank which had a small retail banking operation in New Zealand and which was not incorporated locally withdrew from retail banking operations but continued with commercial and corporate banking business.
Too big to fail and operational separability
One of the major issues confronting any banking supervisor is how to deal with a bank which gets into serious trouble. Like all central bankers with responsibility for supervising banks, I steadfastly maintained the public stance that no bank was too big to fail. But when we actually did some analysis of what would happen if one of our four big banks did fail, it was rapidly apparent that closing the bank would be out of the question – not just because of the effect on those with deposits in the bank, or even those dependent on loans from the bank, but because of the disastrous consequences for the payments system if one of the four largest players were to close. So we started trying to work out how a systemically important bank might be allowed to fail without actually being closed – or at least without being closed for more than, say, 24 hours.
I understand the Reserve Bank has continued to work on this issue since I left in 2002. It’s entirely consistent with moves in some other countries to require banks to provide “living wills” to their supervisors, so that in the event that a bank gets into trouble the supervisor is able to step in and either wind up the bank in a sensible manner or alternatively keep it going, perhaps by imposing some form of haircut on one or more categories of bank creditors. (We referred to this as BCR, for Bank Creditor Recapitalization.)
But of course for it to be feasible to continue the operations of a failed bank it would be essential that whoever steps in to run the bank has the practical ability to do so – in other words, has the ability to access the bank’s loan processing facilities, its computer systems, and so on. If these are inside the international headquarters of the parent bank, that could be very difficult in practice, particularly if, as is likely, the foreign parent is itself in difficulty and could even be under the direct control of a foreign supervisor. For this reason, the Reserve Bank has required foreign-owned banks in New Zealand to maintain control over their own back-office processing facilities. They do not need to be in New Zealand, but there must be robust contractual arrangements which provide a high degree of confidence that, in the event of a crisis, they would be under the effective control of whoever has stepped in to take control of the bank in New Zealand.
Having said that, I acknowledge that all those responsible for the regulation and supervision of banks, whether in New Zealand or elsewhere, are only too aware that what to do with a systemically-important bank in serious difficulty remains a major conundrum. Nobody wants to acknowledge that a major bank is too big to fail, but everybody recognises the serious dilemmas involved in letting a major bank close its doors. While Bank Creditor Recapitalization might conceivably work for a bank-specific crisis, where everybody could see that the problems facing one bank were unlikely to be facing other banks in the system, it seems unlikely to be a sufficient solution to a crisis which extended to the banking system as a whole – the sort of crisis which has affected the banking systems of the US and Europe over the last few years.
The limits of supervision
I think there is a role for bank supervision[1], but I am more than ever convinced in the light of the bank failures internationally over the last few years that bank regulation and bank supervision are only part of the answer. Neither supervision nor regulation can prevent all bank collapses: after all, despite the public myth that banks were entirely free of regulation and supervision over the last couple of decades, large US banks at least were subject to intensive supervision, with supervisors on site permanently. Not only did that not prevent some major banks getting into serious trouble, that level of supervision may even have made matters worse, by encouraging both bank management and bank directors to assume that, as long as the supervisors didn’t blow the whistle, things must be just fine. Nothing could be further from the truth.
[1] See also “Banking Regulation after the Global Financial Crisis”, D T Brash, speech to the Alamos Alliance conference on 19 February 2010 (http://donbrash.com/banking-regulation-after-the-global-financial-crisis/).
Copyright © 2024 Don Brash.