Speech The Changing Nature of Economic Crises
Talk to Australian Business Economists 13th Annual
Forecasting Conference Dinner
It is a pleasure to be here this evening to talk to the Australian Business Economists. I looked back over my files and I noted that this is only the second time I have addressed this group; the previous occasion was nearly seven years ago in February 1991. At that time, I spoke about asset price booms and busts, particularly the latter. You may think I have a one-track mind, but the same subject – only with a different location – is going to figure prominently in what I have to say tonight.
Types of Economic Crisis
I want to talk about the changing nature of economic crises, and then draw out some conclusions for Australia. At the risk of some oversimplification, I think there have been two basic types of economic crisis. The old-style economic crisis I will call Type I, and the new-style, which has come to prominence over the past decade, I will call Type II. Most of these crises occur in developing countries, but developed ones have not been immune from them.
The standard way that countries – particularly developing countries – got into trouble in the past was that their governments did the wrong thing by running bad fiscal and monetary policy. Governments ran large budget deficits and did not finance them properly – they wrote cheques on the central bank, which is colloquially known as printing money. Interest rates were held artificially low (or negative in real terms) and money supply ballooned, resulting in accelerating inflation, a loss of competitiveness, large current account deficits and a collapsing currency. This was a Type I economic crisis, and the IMF has a standard set of arrangements for dealing with this problem.
Just about every developing country ended up in this predicament at some time or other. Some of the Latin American countries did it again and again. We skirted perilously close to it in Australia, particularly in the mid seventies, but did not end up in the hands of the IMF. The United Kingdom, however, did. It suffered a classic Type I economic crisis in 1976, and needed a support package from the IMF, with the usual conditionality requirements.
The Type II problem is very different. Although it has become common in the past decade in a wide range of countries, the best examples at present are to be found in Asia. Type II economic crises are not caused primarily by bad fiscal policy or bad monetary policy. One of the reasons I have a lot of sympathy for the Asian countries at the moment is that they did not get into their present trouble through undisciplined fiscal policy, and their monetary policy in the conventional sense was quite good. This is shown by the fact that their budgetary positions are on average much better than OECD economies, and most have inflation rates that are quite low for developing countries.
A Type II crisis is really financial in nature, rather than stemming from poor macro-economic policies. Its central dynamic is the credit cycle and the main players are companies and banks, or financial institutions more generally. A Type II crisis usually traces its roots to a period when the economy was doing well, growing quickly and becoming popular with investors and lenders. Usually, but not always, an inflow of foreign capital is part of the process which bids up asset prices (and attracts more foreign capital). Domestic banks do not sit idly by – they join in to lend money for all sorts of promising projects and asset purchases. If the economy is growing strongly at the time, which it usually is, it looks as though everyone can make money. Some of the finance is invested wisely and makes a good return, but much of it is inevitably invested unwisely. A lot of it goes into property or is invested in industries which are already oversupplied.
At some point, something happens and the whole process goes into reverse. In the case of South East Asian countries, the event that triggered the reversal was the realisation that their currencies (initially the Thai baht) had become over-valued because they were tied to a strongly appreciating US dollar. The real exchange rate was also the trigger in Mexico in 1994. The mechanism by which the reversal occurs is usually capital outflow. Even without an external ‘shock’, a reversal would eventually occur, as it did in Japan's case. Domestic and foreign investors sell assets and take their funds out of the country. The early ones realise good profits in the process, the slow movers are motivated by a desire to cut their losses.
In the process, the exchange rate falls sharply, and the initial diagnosis is that the country is suffering from a ‘currency crisis’. Soon asset prices fall, with attention focusing on the most visible indicator of this – the share price index. Later it becomes apparent that businesses are going to fail, collateral values will fall, loans will go bad, and the inevitable glut of new property coming onto the market will not be able to attract buyers. If, as is usually the case, companies and banks have borrowed in foreign currency, servicing burdens become intolerable. Suspicion turns to the solvency of the banking system because its customers are failing and its capital is diminishing. It is usually not just a problem of an individual bank, but of a significant part of the banking system. Banks are reluctant to make new loans or rollover existing ones as they fight for survival.
Compared with a Type I economic crisis – which was predominantly due to government failure – the Type II crisis is predominantly a private sector and banking failure. The government is, of course, still implicated, and there are usually Type I elements involved, but the solution to the problem mainly involves reforming the banking system.
If the trigger for the crisis in the recent Asian episodes was an over-valued exchange rate, this can be counted as a macro-economic policy error by the government. This is a complicated subject because for quite a while the currency fix seemed to be contributing to the good economic growth performance. But in the end, policy placed too much emphasis on international competitiveness and the current account of the balance of payments, and neglected financial stability and the capital account. The fixed exchange rate also affected the composition of capital inflow in that it encouraged borrowing in US dollars because US interest rates were lower and there was a perception of little or no currency risk. This contributed to serious Type II problems.
The second way that governments often get involved is that they own or influence a lot of the banks and firms that get into trouble, and a lot of the physical investments that turn bad have been made at government direction.
Finally, an important qualification to the view that Type II crises are predominantly private sector in origin is the recognition that the environment in which the private sector operates is largely determined by government policies. When the insolvencies emerge in the banking sector, people will ask ‘Why did the government allow this to happen?’, ‘Why didn't the government have policies in place which prevented the build-up in this unsustainable situation?’, ‘What will the government do about resolving the crisis now that it has happened?’ The truth is that the government cannot pass the buck. The best they can say is that it was not a failure of conventional demand management policies, it was a failure of a different set of policies altogether – the policies directed at financial system stability.
The Increased Focus on Financial System Stability
There has been increased recognition in recent years that more attention should be devoted to policies directed at financial system stability. The two best examples of this are the Basle capital adequacy guidelines and, more recently, the Core Principles jointly promoted by the BIS and the IMF. One reason for this is that we now have a better understanding of the huge economic costs that are incurred whenever there is a systemic failure of the financial system. Systemic failure invariably results in a recession and can turn a probable recession into a depression. Modern scholarship now identifies a systemic banking system failure as the principal reason that the US economy experienced a depression in the 1930s, rather than a short-lived recession.
The budgetary cost of resolving a systemic failure is also extremely high. Some people may say that one way to avoid this is to let the private sector sort it out. The drawback to this suggestion is that once the problem has reached systemic proportions, the intermediation process has broken down so badly that good borrowers will be hurt as well as bad ones. The government has to step in and close down the insolvent lenders, and take the bad loans off the remaining banks so they can start with a relatively clean slate. This solution is invariably costly and may not appeal to some on long-run moral hazard grounds, but the alternatives are so much worse. In practice, this is the only practical solution to a systemic failure of the financial system. The table below shows some recent examples of the budgetary cost of systemic failures. It covers only recent episodes in OECD countries; much higher figures could have been cited if the coverage had been extended to developing countries.
|Country||Period||Cost as Per cent of GDP|
|United States (S&L)||1984–1991||3.2|
Source: Caprio and Klingebiel, World Bank, July 1996
The other reason for the recent emphasis on financial system stability is that economies appear to be more susceptible to instability as a result of the deregulated financial markets. Australia's experience in the late eighties points in this direction, but it is not an absolute rule. Just as Japan has shown that Type II episodes can occur without a lot of foreign capital flows (in fact, there were net outflows), there are many African and Latin American examples of it happening in a heavily regulated environment. The Asian countries, including Japan, which are now going through these difficulties could not be characterised as possessing highly deregulated financial sectors. The two Asian countries which most closely approximate the deregulated model are Hong Kong and, to a lesser extent, Singapore. To date, these countries have fared reasonably well.
The link between deregulation of financial markets and instability is obviously more complicated than meets the eye. It would probably be better to say that the transition phase from regulation to deregulation is a dangerous period. This will be especially so if the transition is rapid, and if it is not accompanied by increased transparency and the enhancement of other policies directed at system stability.
Problems and Remedies
In essence, therefore, a Type II economic crisis is one which is centred around a failure of the financial intermediation process. The centrepiece of it is an environment which leads to excessive competition by banks to lend, resulting in a severe reduction in credit standards. The following features are normally present:
- lending to related parties;
- excessive concentration of lending to particular borrowers or areas, e.g. property;
- excessively high loan to valuation ratios;
- inadequate covenants to restrict the activities of borrowers;
- lending based on asset values, rather than capacity to service from income;
- failure to recognise and provide for deterioration in loan quality;
- lending to firms or individuals as a result of government directive, rather than on a commercial basis.
All these developments often take place against a background of limited transparency, where investors and depositors have difficulty monitoring the capital adequacy of individual banks. There appear to be no problems during the initial phase of rapid economic growth and rising asset prices, but doubts begin to arise when asset prices fall. The recognition that some banks (and companies) have become insolvent is difficult enough in sophisticated capital markets, but is harder again where there is little necessity to revalue assets and disclose the results. Bank supervisors, as well as market participants, are thwarted by the lack of transparency. Inevitably, insolvent entities continue to operate, and there is widespread distrust of all banks because lenders and depositors cannot distinguish between creditworthy and insolvent institutions.
As a result of the presence of Type II economic crises, it is now recognised that governments require improved policies directed at financial system stability. Within this general field of policy, the one that has the most direct relevance is the prudential regulation of banks. This includes both preventative measures and crisis resolution. But prudential supervision of banks is only one policy, although an extremely important one, among the range of policies that are required to achieve reasonable financial stability. Other important policies are those that pertain to the payments system and, of course, the whole field of securities regulation, including disclosure provisions and the enforcement of commercial and criminal law. In fact, it is possible to consider the whole body of law relating to claims over property and accountancy standards as being part of the necessary infrastructure for system stability.
Relevance for Australia
It is being increasingly recognised that, in the modern world, the most likely route to a self-inflicted economic crisis or, at its extreme, an economic collapse is through a systemic failure of the banking system. Therefore, in order to get good economic outcomes in terms of sustainable economic growth, we need more than just good monetary and fiscal policies. We need good policies directed at financial system stability, including, of course, good prudential supervision of banks.
There is a close inter-relationship between monetary policy and policies that affect the stability of the financial system. While the principal aim of monetary policy should be to maintain low inflation as a pre-condition for sustainable growth, it cannot be conducted in isolation from developments in the financial system. For example, how far interest rates need to be raised to bring about a given slowing in the economy will depend partly upon how stretched asset markets are, and how exposed the banking system is to them. Similarly, the extent to which a currency depreciation can be accommodated will depend on banks' and businesses' exposure to short-term foreign currency borrowing.
Having pointed out the close relationship between monetary policy and system stability policy, I expect some of you in the audience tonight will be thinking that my next step will be to put in a plea to keep bank supervision at the Reserve Bank, rather than to move it to the yet to be formed Australian Prudential Regulatory Authority (APRA). I am sorry to disappoint you, but I am not going to do so; we did this in our submission to the Wallis Inquiry, but that chapter has ended. Although monetary policy and system stability policy are clearly related, we accept that this does not mean that bank supervision can only be carried out at the central bank. The Wallis Inquiry recommended against this, and cited a number of countries which had performed well under arrangements where monetary policy and bank supervision were carried out in separate institutions. The Government agreed with the Wallis Inquiry, as did the Opposition. We therefore accept that due process has been carried out, and that the umpire has made a ruling. We now want to get on with the job and help the Government put into place the arrangements recommended by the Inquiry.
The Wallis Inquiry was well aware of the interconnectedness of monetary policy and system stability policy, and recommended that both of these responsibilities remain with the Reserve Bank. In common with many other observers of these matters, the Inquiry saw bank supervision as a clearly identifiable and separable sub-component of system stability policy. Even though bank supervision was to be carried out in APRA, the Inquiry recommended a number of mechanisms to ensure that there was close co-ordination between bank supervision and other aspects of system stability policy. These mechanisms involve Reserve Bank membership of the APRA Board, plus a number of arrangements for information sharing.
I want to conclude my address tonight by covering two aspects of the proposed arrangements; first, I want to say something about what system stability policy involves once bank supervision is conducted elsewhere, and second I want to say something about the nature of APRA.
With bank supervision being undertaken in APRA, what is left to be done by the Reserve Bank in carrying out its system stability responsibilities? The first and most obvious task remaining with the Reserve Bank is responsibility for the payments system. In fact, the Wallis recommendations increase our responsibilities in this area so much so that we are getting a separate Board to oversee payments system issues. As you will know, we already have a lot on our plate in this area, with the RTGS project nearing the operational stage and with the next major project likely to be in the area of reducing foreign exchange settlement (including Herstatt) risk.
As well as specific responsibilities for the payments system, the Reserve Bank will also retain a capacity to contribute to the overall formulation of policies that impinge on system stability. The Reserve Bank will need to be able to evaluate the likely effects of prudential policy changes, particularly for banks but also for other financial institutions. It will have to keep abreast of the risks associated with the proliferation of new products and possibly new types of financial institutions. It will also need to form its own independent judgment of where the likely ‘pressure points’ are building up insofar as they may imperil system stability. The Reserve Bank's formal channel into the policy process will be through its membership of the APRA Board. It will also have to develop a close relationship with the Australian Corporations and Financial Services Commission (ACFSC). The new Council of Financial Regulators (CFR) will be one vehicle for this.
What we will no longer be doing is face-to-face bank supervision. We will not be dealing with individual banks when they need help with licensing, interpretation of rules, statistical returns, etc. That will all be done by APRA. Inspections will also be the responsibility of APRA, but in the interests of information sharing, the Reserve Bank will be entitled to participate from time to time if it so desires. I do not imagine this would happen very often. Overall, it is APRA which will be setting and enforcing the rules designed to minimise the chances that financial institutions will fail. It will be APRA that has to make the judgment that an institution is in danger of failing, and to take the appropriate action.
As you will see from this description of the division of responsibilities, the Reserve Bank will be relying on APRA doing a sound job. In fact, I think the whole country, and particularly the financial sector, has a big stake in ensuring that APRA is a first-rate institution. The ACFSC is also crucial to the long-run stability of the financial system, but it has the advantage that it already largely exists in the form of the ASC. APRA, on the other hand, has to be created almost from scratch. We should all support a speedy formation to make sure that the intellectual capital presently residing in its constituent parts in Sydney, Canberra and Brisbane is not lost, but is speedily brought together in one institution. It is also important that it have a high degree of budgetary independence so that it can augment its resources from private financial markets and so ensure a high degree of interaction between regulators and practitioners. I know that the Government is aware of these needs and is doing its best to bring about an early introduction for APRA. I want to finish tonight by assuring everyone that the Reserve Bank is doing everything it can to assist in the process.
Conventionally loose monetary policy in a Type I sense would involve monetisation of the budget deficit, low (or negative in real terms) interest rates, continual downward pressure on the exchange rate and rising inflation. In most Asian countries over the past decade, these conditions did not apply; in fact, the opposite was more typically the case. On the other hand, monetary policy was not tight enough to prevent the strong expansion of the balance sheets of the banks. Such a degree of tightness would not have been compatible with the maintenance of a fixed exchange rate. 
Japan in the 1990s is a classic drawn-out Type II episode, but one in which foreign capital played only a minor role.