Central banks and financial system stability in an uncertain world
I am very pleased to be here in Antwerp this evening, and to have the opportunity of sharing with you some of the Reserve Bank of New Zealand's views on promoting financial system stability.
Everybody in this room is confronted with uncertainty in their daily interface with financial markets. For those of us who are central bankers or bank supervisors, the task is to guide the dynamic, complex and unpredictable processes in financial markets towards outcomes that meet the policy preferences of our country. For the Reserve Bank of New Zealand, our task is to deliver price stability and a sound and efficient financial system within a liberal economic policy paradigm.
The uncertain world that we operate in can be highlighted by the relatively frequent and unpredictable episodes of financial instability experienced over the past couple of decades. Since 1980, over two-thirds of IMF member countries have experienced at least one serious banking-sector difficulty. In some countries, bank losses nearly or completely exhausted the banking system's capital. In some countries, financial collapses have required significant fiscal expenditure to resolve. And, as we all know, national financial crises have been transmitted to other countries, threatening not only the economic well-being of those countries but also the stability of the international financial system as a whole.
Oceans of ink have been spilt in analysing the causes of these crises. Most detached observers seem to agree that the underlying problems can be traced to certain fundamental weaknesses, weaknesses which clearly varied from country to country, but which have included poor quality credit analysis by banks and other financial institutions, politically-directed lending by banks, the end of an asset price bubble (typically real estate), sharply increased real exchange rates in a pegged exchange rate situation, distorted incentives in financial markets, and a lack of transparency in financial markets.
The widespread incidence and significant economic cost of financial sector problems have prompted calls for concerted international action to promote the soundness of financial systems. These calls have strengthened considerably over the last couple of years. The Basel Committee has been at the forefront of this effort with the release of the Core Principles and proposals for a new Capital Accord. The IMF is making evaluation of financial supervision and regulation part of its annual country reviews, and the World Bank is emphasising the strengthening of financial infrastructure as an important part of its structural assistance programmes. While these international initiatives have many positive features, there are aspects that conflict with our approach to promoting soundness and efficiency of the New Zealand financial system and give us some concern. I will elaborate on these in a moment.
At the national level, the preservation of a stable financial system remains one of the most challenging tasks facing governments today. New Zealand is no exception.
Not all of you will be aware that New Zealand had its own banking system problems less than 10 years ago. As a result of these problems, our largest bank would almost certainly have failed had the government, as majority shareholder, not been willing on two occasions to provide a substantial capital injection. One very major financial institution (not a bank, but certainly a quasi-bank, and an institution which was in the process of applying for a banking licence) did fail, one of the largest failures in New Zealand's history. Another bank would have failed had its private sector shareholder not been in a position to inject very large amounts of additional capital, and even after that was done the institution was eventually wound up. At no point did it look likely that the whole banking system might fail, but we certainly had major problems with some of the largest participants in the system.
What lessons did we learn from that experience? There were of course a whole host of factors that caused these difficulties. One important factor was the sheer inexperience of many of our bankers: they had been accustomed to a highly protected environment, and were ill-prepared to deal with the demands of a deregulated environment. Nevertheless, I think there were four important lessons that are worth sharing with you. While every country has to make judgements and decisions in the light of its own particular circumstances - and certainly no two countries are exactly alike - some of the lessons we learnt in promoting financial system stability may have relevance to other countries as well.
I will deal with the monetary policy lessons last and somewhat succinctly, leaving most of my comments to the prudential policy lessons, where we have moved away from the more traditional `rules-based' approach to supervising banks. I have chosen to dwell mainly on the prudential policy lessons, as it is here that our experience may have more relevance for you. I suspect that I do not have much to offer the National Bank of Belgium on the monetary policy front, given your interests are likely to be more concentrated on monetary union issues with a single monetary policy for the euro area now having been in operation for over a year. (As the issue of currency union for New Zealand is currently a matter of public discussion in my country, if we have time I would welcome any comments members of the audience might want to make on your experience with monetary union so far.)
Prudential policy lessons from New Zealand
Lesson 1: encourage banks to behave prudently
First, I believe it is important that banks are given every incentive to behave prudently. This may seem a self-evident statement, but it is astonishing how frequently the importance of this principle is ignored. In New Zealand's case, we diminished this incentive to behave prudently by allowing the view to go unchallenged that banks were effectively `sovereign risk', or at least `too big to fail'. This meant that bank creditors felt little need to assess the creditworthiness of the banks with which they deposited funds - banks were, it was widely believed, effectively guaranteed by government. Bank boards and managements may have felt similarly protected against the possibility of failure, and made loans with a disregard for risk which was, in some cases, breath-taking. This so-called `moral hazard problem' may have been particularly severe in the case of two of the three institutions which got into serious difficulties in the late eighties, both of them owned wholly or in part by government. There was little or no direction of their lending by government, but the management of both institutions certainly embarked upon lending transactions in the newly liberalised environment which rapidly got them into serious difficulties.
For some of the other countries that have experienced financial instability in recent years, it is possible that the incentive for banks to behave prudently was seriously eroded not only by the impression that most large financial institutions would not be allowed to fail but also by the extent to which governments directed the lending activities of the banks themselves. After all, if governments are going to become extensively involved in directing where banks should and should not lend, it is not unreasonable if the banks and their creditors assume that governments will `see them right' if things go wrong. Bank management certainly has little incentive to carefully assess credit risk if, at the end of the day, the decision on whether or not to lend will be made by the bank's board under the influence or direction of higher political authority.
At the moment, as we all know, there is a great deal of international attention focused on how this problem of poor credit decisions in the banking sector can be dealt with. Most of the attention appears to be on how to strengthen external regulation of the banking sector, and how to make banking more independent of political influence. Certainly, freeing banks from political interference in their credit decisions is very desirable, and strengthening `rules-based' banking supervision is one possible way to reduce the risks of future problems in the banking system.
But we in New Zealand are not persuaded that strengthening `rules-based' banking supervision is the only way to proceed, or indeed is even necessarily the best way to proceed in all circumstances. When we reviewed what we were doing in banking supervision in the early nineties, we became concerned. At that time, we were conducting banking supervision along conventional Basel Committee lines. We were gathering very large amounts of confidential information from banks on a quarterly (sometimes a monthly) basis. We were laying down a large number of rules and limits designed to ensure that banks behaved prudently.
Several things prompted us to review that approach, and one of them was a worry about the risks which we were incurring on behalf of taxpayers. What would happen if, despite our banking supervision, a bank were to get into difficulties? Might depositors argue that they wanted full compensation, since while they had had no knowledge of the bank's financial condition we in the Reserve Bank were not only fully aware of that condition but were also responsible for laying down the rules and limits by which the bank had been obliged to operate?
We consoled ourselves with the thought that our banking supervision was so good that no banks would fail under our watchful eye. But then we looked abroad - at the United States, at Japan, at Scandinavia, at the United Kingdom and at Australia. We found banks going down in significant numbers, despite some extremely professional and politically-independent banking supervision. We could not be confident that traditional banking supervision would prevent bank failure, and we could be confident that, by being the sole recipient of detailed financial information on banks and the main arbiter of what constituted prudent banking behaviour, there was a major risk that we would be held liable, politically and morally if not legally, for any losses incurred by depositors.
Then we became aware of anecdotal evidence that our banking supervision was reducing the incentive for bank directors to make their own decisions about crucial aspects of their bank's operations. In other words, because the Reserve Bank was laying down maximum individual credit limits, and limits on open foreign exchange positions, and guidelines for internal controls, some bank directors were assuming that they were necessarily behaving prudently provided they were operating within those limits and guidelines. They stopped addressing the risks which their own banks were facing and simply complied with the general limits and guidelines. To the extent that that was true - and we found some evidence that it was true in some banks - we concluded that our banking supervision might actually be increasing the risk of bank failure, by reducing the incentive for bank directors and bank managers to make their own careful assessment of risk.
The outcome of our review was to substantially strengthen disciplines on the directors and managers of banks to operate their banks prudently, and to strengthen the ability of the marketplace to discipline banks. We also retain a system of supervisory discipline, which we take very seriously. But we retain only a few absolute rules within that framework, principally that all banks must at least meet the Basel capital adequacy rules. We rely mainly on a requirement that banks disclose to the public a substantial amount of financial and risk information quarterly. In addition, all bank directors must sign off these quarterly statements, at the same time attesting to the fact that the internal controls of their banks are appropriate to the nature of their banking business, and that those controls are being properly applied. We in the Reserve Bank do not attempt to tell banks what those controls should look like, but directors signing those quarterly statements without making a careful assessment of the adequacy of internal controls are exposing themselves to very considerable legal risk in the event that their bank gets into difficulty.
We have also gone out of our way on a number of occasions to make it clear to the public that neither the Reserve Bank nor the government of New Zealand is guaranteeing individual banks, and we published a booklet designed to assist the general public to interpret banks' financial information.
None of these actions is a guarantee against imprudent bank behaviour of course, but we believe that we have gone a considerable distance towards ensuring that banks face strong incentives to behave prudently. No bank operating in New Zealand is now owned by government, none is guaranteed by government, none is obliged to lend to particular sectors or companies, and our supervision is based heavily on mandatory public disclosure and director attestations. Three years ago, Alan Greenspan commented that `regulation by government unavoidably involves some element of perverse incentives. If private market participants believe that government is protecting their interests, their own efforts to do so will diminish.'1 We have tried to minimise those perverse incentives. As William McDonough, President of the Federal Reserve Bank of New York and Chairman of the Basel Committee on Banking Supervision, said in March this year, `...given the accelerating pace of change and innovation in the financial services industry, our notion of supervision must encompass discipline applied by the marketplace, as well as official supervision.'2 In New Zealand, our notion of supervision relies heavily on marketplace discipline.
Of course, to some extent our approach only works because there is a clear framework of company law which makes it clear that company directors and managers have unambiguous responsibilities. Having agreed accounting rules is important. Having risk-proofed payments systems is important. Having a vigorous media, with probing financial journalists, is also of great value, so that when a bank is forced to disclose to the public a deteriorating financial position, or a breach of one of the few rules we retain, there is at least a reasonable chance of that being picked up and sensibly analysed by the media. Not all countries are so lucky. Where countries do not have this infrastructure in place, a more hands-on involvement by supervisors in banks' activities may well be appropriate. Where this infrastructure is in place, as it is in New Zealand, we believe that a supervisory regime based largely on self-discipline and market discipline is a viable option.
I mentioned a little earlier that we have concerns about some of the international initiatives currently underway to address financial system instability. The major issue for us is that we see some of the initiatives as potentially undermining the incentives we have established for banks to behave prudently. While there is increasing recognition of the need to harness and reinforce market mechanisms in the pursuit of supervisory goals, at the same time the international supervisory community appears to be moving towards an ever more prescriptive and intrusive approach to banking supervision. An example is the supervisory validation and supervisor intervention proposals in the Basel Committee's June 1999 consultative paper on a new Capital Accord. If these proposals were implemented in New Zealand, they could considerably weaken market incentives and distort the behaviour of bank directors and managers. We see this occurring as a result of the supervisor becoming more directly involved in the management of banks, with the associated risk of the government being drawn in to underwriting banks in times of stress.
Another potential risk for us is that, because we do things differently, we might be assessed as not fully complying with the Basel Committee's Core Principles and other international standards that are being developed. The concern we have is that New Zealand banks might end up being penalised (for example, through higher borrowing costs if risk weights in the Capital Accord are linked to compliance with international standards). This could result if the process of assessing compliance with international standards becomes a `one-size-fits-all', checklist, mechanical tick-the-boxes, approach. We have stressed in our submissions to the Basel Committee that assessing compliance with any international standard needs to be focussed on looking at whether the substance, rather than the form, of the standard is being met.
I should add that, notwithstanding these concerns, New Zealand strongly supports the introduction of international standards. We consider international harmonisation and co-operation as very important in what is rapidly becoming a `borderless' global financial system. Our plea is that the implementation of such standards be done with the appropriate degree of flexibility to adequately cater for the individual circumstances of each country.
To be frank with you, it is impossible to say for sure whether our prudential policy approach is working or not. We have had no bank failures since the new policy was implemented, but then we had no bank failures in the previous 100 years either (although we certainly had some banks which would have failed, in the late eighties, had their shareholders not injected a lot more capital). Moreover, all but one of the 17 banks in New Zealand are now foreign-owned, an increase in the proportion of banks in foreign ownership over the last decade, and that also means that the New Zealand case is not necessarily a compelling demonstration of the validity of our market-based approach.
But there has been quite a lot of anecdotal evidence that bank directors are now taking their responsibilities more seriously, and there seems to have been a marked increase in the attention given to internal controls. We had a mild recession in the first half of 1998, and a related fall in the value of the New Zealand dollar of some 30 per cent against the US dollar (from peak to trough), and our banks came through this with absolutely minimal adverse effects, with low ratios of impaired assets and continuing strong profitability.
So far at least, we are well satisfied by the way in which the new system is working.
To conclude what has been a long first lesson, let me illustrate my main point with a specific event. A few months after the new system first came into operation, at the beginning of 1996, one bank was obliged to disclose to the public the fact that it had had a credit exposure to its shareholder bank which considerably exceeded the limit which we had stipulated for such exposure. The attention focused on this issue by the media, and indeed by other banks, created strong incentives for the bank never to repeat that mistake - quite probably stronger incentives than any threat of central bank sanction could have created.
Lesson 2: beware of government ownership of banks and remove barriers to foreign ownership
The second lesson from the New Zealand experience in the late eighties is that the ownership of banks is an important issue. For us, the issue was in part government ownership of banks and in part foreign ownership of banks. The government-owned financial institutions almost without exception suffered various degrees of financial difficulty - sometimes because their managers had undertaken imprudent lending, and sometimes because they had been obliged to invest in large amounts of government securities at sub-market interest rates. The large foreign-owned banks suffered to a much more limited extent from the bad debts and losses which the government-owned banks experienced.
There have been various reasons given for this difference, but the most plausible is that the large foreign-owned banks were under the watchful eye of experienced parent banks, and were therefore much less able to stray into some of the riskier propositions which tempted the government-owned institutions, especially in the years immediately after the banking sector was liberalised in the mid-eighties. (The newly-arrived foreign-owned banks, however, often did succumb to the temptation of lending on risky propositions, perhaps because, being quite small both in absolute terms and in relation to their overseas parents, they were subject to much less intensive parental scrutiny.)
More recently, New Zealand has been running a very large balance of payments deficit, probably amounting to more than 7 per cent of GDP at the present time. This balance of payments deficit has been experienced at a time when the government itself is running a fiscal surplus. In other words, it has been the private sector which has been borrowing heavily from overseas, not the public sector. And while some of this borrowing has been done by the corporate sector directly, much of it has been done by the banking sector. Comparable levels of overseas borrowing by banks in other countries have been sufficient to make foreign lenders very nervous (for example, in some Asian countries in recent years), and yet similar nervousness has not been at all evident in New Zealand. Why? I can only conclude that the foreign lenders take considerable comfort from the fact that most of the banks operating in New Zealand now are in fact wholly-owned by foreign banks, or are indeed branches of foreign banks. Those parents are seen as being financially strong, and fully able to back the operations of their New Zealand subsidiaries or branches. (It may also be relevant that, overwhelmingly, the overseas borrowing being undertaken by New Zealand banks carries no foreign exchange risk for the banks themselves.)
In some countries, there is political reluctance to allow foreign institutions unrestricted entry into local banking sectors. I would have to say that, as a country where all but one of our 17 banks are owned and controlled overseas, we have seen absolutely no disadvantages from this situation, and many advantages. We have a financially stable banking sector, with vigorously competing and highly innovative banks, all of them subject to the monetary policy influence of the central bank. I have no doubt at all that the banking sector is considerably more stable than would have been the case had all the banks been domestically-owned, whether in the private sector or in the public sector.
Monetary policy lessons from New Zealand
Lesson 3: keep prices stable
The third lesson from our experience has been the crucial importance for financial system stability of keeping prices stable. I am sure that for this audience the assignment of monetary policy to the objective of keeping prices stable is not extraordinary, as evidenced by price stability being the primary objective of the single monetary policy of the Eurosystem. But let me comment on the link between price stability and financial system stability.
By the late eighties, New Zealand had experienced nearly 20 years during which consumer price inflation had been above 10 per cent almost without a break. Interest rates after adjustment for tax and inflation were often strongly negative, and there was as a consequence a strong incentive to invest in real estate and shares, using as much borrowed money as could be obtained. This was undoubtedly an important contributor to the severe difficulties which both the corporate sector and some parts of the banking sector experienced when monetary policy was tightened in order to reduce inflation in the second half of the eighties. Interest rates went up and asset prices went down, and several banks incurred very large losses as a consequence.
In New Zealand, monetary policy has been directed at achieving and maintaining domestic price stability since the mid-eighties, with the exchange rate being allowed to float freely since March 1985. Prior to that, the objectives of monetary policy had been much more confused. In New Zealand, as indeed in many other countries, there had been a deeply engrained belief that monetary policy could be used directly to achieve faster growth in output and employment. There was a widespread belief that policy-makers could choose to tolerate a somewhat higher level of inflation in exchange for more growth and a lower level of unemployment. (After all, Bill Phillips of Phillips Curve fame was New Zealand's best known economist.) Since the mid-eighties, we have accepted that there is no such trade-off between inflation on the one hand and growth or employment on the other, and that the best thing monetary policy can do to foster sustainable growth in output and employment is to deliver predictably stable prices, or a very low rate of measured inflation. And, of course, a sound banking sector is more likely in an environment of steady economic growth.
Unfortunately, relatively low consumer price inflation is not always accompanied by low asset price inflation. There is no single explanation for asset price inflation and, as New Zealand itself discovered in the mid-nineties, with quite a strong increase in the price of both residential and rural property, it is often extraordinarily difficult to restrain asset price inflation even when inflation in consumer prices is low. Nevertheless, we are confident that monetary policy targeted on low consumer price inflation in combination with a floating exchange rate is likely to result in a weaker and less sustained asset price cycle than would otherwise be the case.
The problems caused by asset price inflation have been well illustrated in parts of Asia in recent years. It is at least possible that asset price inflation in Japan in the late eighties - the reversal of which has done so much damage to bank balance sheets in that country - was a consequence of monetary policy being kept too easy for too long, whether to appease the United States after the Plaza Accord or for some other reason I know not. Similarly, it may well be that if central banks in some other Asian countries had not been so preoccupied with trying to avoid the appreciation of their currencies against the US dollar as capital flowed into these economies in the mid-nineties, their interest rates would have been higher and asset price inflation commensurately reduced. And of course if asset price inflation had been less, the over-investment in certain kinds of real estate would presumably have been less and, with it, the subsequent fall in asset prices.
Lesson 4: avoid pegging the exchange rate
And that brings me on rather naturally to the final lesson from New Zealand's experience, and that is the danger of pegging the exchange rate unless one is prepared to go all the way to full currency union as you have done in Belgium, or at least to a currency board, as Hong Kong and some other countries have done.
In New Zealand, we had a pegged exchange rate until March 1985, as I have mentioned. Prior to that date, it was not uncommon for companies to borrow overseas, often at interest rates which were very much lower than those within the high inflation New Zealand economy. Some companies made rather spectacular losses when the New Zealand dollar was devalued from time to time, or when, even when pegged, the New Zealand dollar depreciated against the currency in which the loan was denominated. (Borrowing in Swiss francs was particularly popular, and particularly painful, for some companies.) But the losses were probably fairly moderate in comparison to the loss which the government itself incurred on behalf of taxpayers in 1984. In that year, the New Zealand dollar was devalued by 20 per cent after a foreign exchange crisis and the government, which had written very large volumes of forward exchange contracts with companies trying to protect their positions, incurred losses of many hundreds of millions of dollars.
Since March 1985, the New Zealand dollar has been freely floating, and indeed I suspect we may be the only central bank that can claim not to have intervened directly in the foreign exchange market for more than 15 years. (I say `directly' because from time to time we did adjust monetary policy when we felt that movements in the exchange rate seemed likely to threaten our single goal of low inflation.) One of the benefits of this has been that, though many companies and banks have borrowed overseas, none of this borrowing was undertaken in the belief that there was no currency risk involved. Overseas interest rates were frequently much lower than those in New Zealand but, after factoring in the exchange rate risk, the incentive to borrow offshore in foreign currency was substantially reduced.
As a consequence, when, after a period of strong New Zealand dollar appreciation between early 1993 and early 1997, the New Zealand dollar fell by some 30 per cent against the US dollar, there were very few companies unhappy about that fall - and indeed plenty of exporters who were delighted. Even fewer of our banks were caught out by the depreciation, and to the best of my knowledge none incurred losses as a consequence of the move. Because they knew that the New Zealand dollar was freely floating, they were careful to avoid taking on unhedged positions in foreign currency.
In conclusion, in a world of open capital markets, the challenges facing banks are very considerable. Central banks can not prevent all banks from getting into trouble, and nor should they try to do so. But central banks do have a responsibility in all our countries to promote the stability of the banking sector. In my own view, they can best do that by creating strong incentives for banks to behave prudently; by discouraging government ownership of banks, and removing barriers to the foreign ownership of banks; by keeping the focus of monetary policy on price stability; and by not creating the impression that borrowing in foreign currency is a riskless activity. We learnt those lessons the hard way.
1 Remarks by the Chairman of the Board of the US Federal Reserve System, Dr Alan Greenspan, at the annual conference of the Association of Private Enterprise Education, Arlington, Virginia, 12 April 1997.
2 Speech by Mr William J McDonough, President of the Federal Reserve Bank of New York and Chairman of the Basel Committee on Banking Supervision, before The Monetary Authority of Singapore, Singapore, on 24 March 2000.