Sveriges Riksbank

Ingves: Banking reforms

Pressmeddelande   •   Nov 03, 2006 10:55 CET

Thank you for inviting me to speak at this seminar. The topic for this session reflects that the ever ongoing changing circumstances, such as globalisation and integration of markets, make old structures and approaches obsolete and call for reforms.

The title of today’s seminar is “Reform strategies for central banks and banking systems in North African and Middle East countries”. Since I do not believe in a top down approach where the authorities dictate the structure and details of a banking system, I would prefer to approach the issue by discussing how the authorities should establish a framework which promotes the growth of a modern banking system. Such a framework includes several components such as modern legislation and regulation, supervision, a financial safety net – in addition to the obvious centrepieces, namely the banks themselves and a central bank that conducts market-based monetary policy.

I will discuss a number of components which are necessary in a modern banking system to make it efficient, while at the same time flexible to accommodate developments and resilient to promote financial stability. In addition to issues about banks themselves, I will talk about the underlying preconditions for banking; the legislative and regulatory framework; the supervisory agency and its work; and also the need to supplement the banking system by other financial institutions and markets. There is no single blueprint for banking reform, and my observations come from experiences in many countries, including my own.

In my previous work as an IMF director but also as a civil servant in Sweden, I have seen numerous examples of shortcomings in the preconditions for banking, in the regulation and supervision of banking and in the conduct of banking activities. As a minimum, these shortcomings have led to inefficiencies in providing financial services to society. They have also led to higher costs and in many cases to bank problems and even systemic crises. My obvious conclusion is that banking reforms which deal with the flaws are worthwhile for all parties and countries. The cause for banking reforms is strengthened by current developments in banking, which includes a spread of cross-border institutions and activities but also new instruments and methods to handle risks in banking. Such developments generally increase efficiency and stability in banking, but need to be carefully regulated and supervised because they also contain inherent vulnerabilities and risks.

As I said at the outset – I do not believe that the authorities can create an efficient banking system “by decree”. There is a natural demand for good banking services in all countries and if you, the authorities and policy-makers, set up a reasonable framework for banking, I am convinced that you will soon see applications for new banks or for existing banks to change their structures, activities or mode of operations.

In achieving an efficient banking system, the authorities are faced by a number of questions:

What is the optimal number of banks? In modernising a banking system you would like to have a “reasonable number” of banks – not too few but not too many. In many countries the lifting of the old and restrictive regulations leads to an influx of a large number of mostly small “family-owned” banks but many of these tend to have overoptimistic plans for their activities and are closed rather soon leading to consolidation of the banking sector. Some depositors and other counterparties could suffer losses from dealing with these banks. Are such events necessary steps in the development or could they be avoided? In my experience, some countries after scrapping their old regulations became ultra-liberals and provided licenses for banking on too lenient grounds. They did not properly evaluate the prospects for the banks to conduct profitable business in a competitive environment and they set the requirement for basic minimum capital far too low. But having said that I would warn against doing the opposite – to let the authorities decide how many banks there ought to be and then distribute a fixed number of licenses. This would certainly lead to distortions. Nor would I condone the practice of requiring licenses for opening a branch in another area of the country. The banks must decide this for themselves.

The content of the license application procedure is very important. Here the authorities have an opportunity to prevent bad apples from entering the barrel. The authorities must check that both owners and managers are “fit and proper”, which means competent and not criminal. The requirement regarding skills is of course higher if they want to run complex banking activities. It is not necessary that each Board member knows every bank activity himself but between them the Board should be competent on all issues. Remember that Nick Leeson at Barings Bank easily convinced his Board that he could run a highly profitable derivatives trading operation without any risk to the bank!

The license application also includes the organisation and structure of the bank group. You should demand that this does not hinder the effective supervision of the group. Do you remember the BCCI bank which collapsed in the early 1990s? The bank’s owners had created a structure for their group with the clear intention of making full insight by the supervisors more difficult. For instance, the main activities and risk-taking did not take place where the main office and thus the main supervisor was located.

Should you prefer local owners or should you allow foreign owners freely, being open to the possibility that the foreign banks become dominant in your system? Being a free-marketeer I strongly advocate the latter. Your prime aim should be to have a banking system which provides the best services to the society, whether locally owned or foreign owned. Like many other countries your banks could also benefit from the influence of the foreign banks if they have more advanced systems or methods.

On the other hand, I would also warn against being too lenient toward foreign applicants to set up or buy into existing local banks – I have seen many countries being too trusting toward foreign owners. These should be submitted to the same scrutiny as local applicants. The scrutiny should include a test of the origin of the money used to found (or buy) the bank. Obviously only known owners should be accepted and thus no beneficial owners who are hidden behind the veils of companies established on some offshore jurisdiction. And when speaking about foreign owners: When banks operating in your country have foreign parents you must ensure that the home supervisor practices consolidated supervision which includes the entities in your country. If this is not the case, it is in your best interest to deny the application for a license.

Who should be allowed to own banks in addition to individual persons and financial institutions; should non-financial institutions be allowed and should there be holding companies? The global standard-setters are open to such ownership, but only on the condition that legislation and regulation provide the powers for effective consolidated supervision and transparency of such structures, including the parent company. This implies for instance that the bank group can be protected from the risks emanating from the non-financial owners, such as through ring-fencing. In my experience, non-financial ownerships have often caused problems for banks and the supervisors must monitor such relationships closely. This applies in particular to so called “pocket banks” which are dominated both in ownership and activities by the needs of the parent company or major owners.

Should banks have a widely spread or a concentrated ownership? I have an open mind on this issue, too. Both alternatives have their advantages and disadvantages. A broad ownership may better protect the interests of the minority shareholders, but if financial problems occur the majority owners have stronger incentives to provide capital injections. The governance of a bank is sometimes promoted by having strong owners, but there are also examples of strong owners misusing the bank for their personal purposes. All in all, you should be open to different forms of ownership but monitor them closely.

Should you allow different forms of banks, such as commercial banks, savings banks, credit unions, development banks and micro-finance institutions? In my view: Yes, of course. Competition is best served by having a range of bank ownership structures such as shareholding, mutual ownership, non-profit organisations, etcetera. But having said this I want to warn against certain practices I have experienced in some countries which I would call “compartmentalization”. This happens when the legislation prescribes unnecessary and ineffective borderlines between the allowed activities of defined banking categories. Usually, the intention is to ensure limited competition and high profits for the various categories. Such restrictions are in effect a subsidy to the banks and they are harmful to the consumers and to the overall economy. The rule-of-thumb should be that any bank should be allowed to conduct any of the generally regulated bank activities if it can prove that it possesses adequate competencies, systems and resources.

Should there be one set of prudential rules for all institutions conducting bank-like activities or should you differentiate? Could there be “light-touch supervision” for small and non-complex institutions for instance in micro-finance or local credit unions or exchange houses? There is no obvious answer to these issues. On the one hand you want to create a fair level-playing field for all bank activities which call for equal treatment. But on the other hand there is no necessity to burden small institutions with the elaborate regulations intended for large and complex banks. As I noted before, there must be a reasonable relation between the amount of regulation and its cost to the institutions. Consequently, I would accept the case for lighter supervision of small and non-complex institutions, but there must always be adequate supervision to ensure discipline also for those. Even when a small institution fails it causes disruptions and a loss of confidence in banks in general. We have recently had bad experiences in Sweden, so if you intend to introduce risk-based supervision I advise you to do it with caution.

Going one step further – how to set the boundaries between banks and other financial institutions? In my opinion, it is very important to safeguard the general public’s confidence around the concept of a bank and it must be protected by a clear definition of a bank and what it can do. In Sweden, the definition is based on the combination of a bank receiving deposits and being active in the payments system. However, a more common definition of a bank focuses on the combination of two activities, namely the receipt of deposits from the general public and the granting of credits to borrowers using these funds. In accordance with international practice, this combined activity should be a privilege of banks. Other institutions may issue deposit-like instruments in high amounts and not directed to the small savers. To highlight their different characters, such instruments should not be protected by any depositor guarantee scheme. But apart from the deposits, most bank activities might also be open to other financial institutions and vice versa. Insurance companies could sell savings products and banks could sell securities.

Should you allow “financial conglomerates” mixing banks, securities companies and insurance companies in the same groups? In line with my earlier views on a flexible financial system my opinion is that such structures should be allowed, provided of course that your laws and regulations allow for effective consolidated supervision of the whole financial conglomerate. The present development toward a blurring of the boundaries between the activities of different financial institutions makes it reasonable to allow financial conglomerates. Broad groups conducting different activities may also be better able to diversify their risks and could thus be more resilient against financial shocks.

Are there reasons for retaining state-owned banks? An argument sometimes voiced is that they promote competition and provide services for certain parts of the population which are of little commercial interest to the other banks. The Basel Committee’s core principles accept state-owned banks as long as they are run and regulated on equal terms with other banks. But my own view is that you should avoid having state-owned banks. Simply put: The government is not a good owner and manager of banks since the bank will come “too close to the politicians”. There is always a temptation for the government and parliament to use state-owned banks to provide what mistakenly looks like cheap services to the people. But there will be hidden costs for such services, not least in the form of disrupting competition in the banking sector. The government could certainly provide certain subsidized services if it so wishes, but the costs should be transparent. For instance, the government could pay the existing banking network on a commercial basis to provide such services.

Other parts of the financial system
The title of this presentation is “banking reforms”. However, an efficient and stable financial system cannot be founded only on banks. The risks of disruption to the overall economy from a break-down of the banking system are much greater if the banks are dominant than when there also exist other financial institutions and functioning markets. I would strongly encourage you to facilitate the development of other financial market participants such as insurance and pension companies, securities traders, fund management companies etcetera as efficiently functioning markets in various equities and securities. Such diversification will also increase competition and lead to better services at lower prices. Diversified domestic markets also act as a kind of insurance in bad times.
Neither should we forget the importance of the financial infrastructure, including payments and settlements systems, exchanges, custodians, etcetera. These provide the necessary “greasing” of the financial system and perform very important roles. Weaknesses in the infrastructure may lead to serious problems, so you must monitor the infrastructure as you will monitor the banks.

The development of a modern banking system is dependent on the existence of conducive external factors, often called preconditions. I will now discuss these in some detail. The preconditions include macro economic stability, an adequate legislative framework and a well functioning judicial system. There should also be adequate rules for accounting and auditing; a well developed infrastructure for the payment system and a financial safety net. Only under such an overarching set of good preconditions will the banks develop favourably.

Macro economic volatility is harmful to banks. Periods of rapid credit growth may be followed by problems for many companies. Bankruptcies could then increase leading to high credit losses in banks. High inflation will disguise the underlying profitability of a loan project and could lead to an erroneous credit decision being taken by a bank. Also, bank loans extended in foreign currencies could lead to losses if the local currency depreciates when the borrower who earns his money in the local currency runs into difficulties repaying his loan.

In order to ensure credit discipline, there must be clear legislation as well as court proceedings that are predictable and reasonably fast. If a bank cannot rely on seizing the collateral given for a non-performing loan it will be reluctant to provide loans against such collateral. Of course, the laws and the courts should also protect the depositors and borrowers from any abuse from the banks such as unfair contract terms.

In countries where the accounting and auditing rules are weak, or where the accounting and auditing firms are inadequate, various financial problems may occur. Banks will not be able to rely on the financial statements when assessing their borrowers’ creditworthiness. Nor can the depositors and other creditors to banks rely on the banks’ own financial statements. The public authorities must compensate for these weaknesses by implementing additional rules. For instance, the supervisors may need to require stricter rules for provisioning against loan losses, or to set a higher minimum level for bank capital.

An important function of the banking system is to facilitate payments between various agents in society. This is based on the fact that bank accounts provide the basis for all payments except cash payments. Most payments are executed through the payments system infrastructure, such as the systems for large value payments, cheque clearing, securities settlement systems, stock markets, other exchanges and so on. These systems must be efficient and they must also be secure. If not, there is a risk that problems in one part of the financial system will spread to other parts, including to the banks.

Many countries have neglected the need for an adequate financial safety net. The safety net includes limited but explicit depositor protection which has to be supported by the necessary legislation, institutions and procedures to conduct an orderly management, resolution or winding-up of problem banks. When bank weaknesses are identified in a country lacking a proper safety net, various problems may occur. The authorities will be reluctant to take adequate and timely remedial action since there are no clear guidelines and since they are afraid of the consequences from non-protected depositors and from other counterparts to the banks. The authorities may instead extend excessive financial support to the problem bank through the central bank for instance in the form of exceptional liquidity assistance which is sometimes abused to cover solvency problems, or through the fiscal budget. Occasionally, even failing banks’ owners have received public financial support to continue the operations of the defunct bank. The overwhelming evidence from experiences in different countries and situations is that letting problem banks continue to operate without taking adequate action simply means that the problems will increase over time and they may in the end result in a systemic and very expensive crisis.

To sum up on the preconditions, these must be taken seriously into account when you modernise your banking system. Where there are shortcomings you must deal with these in the appropriate ways, such as through legislation. While waiting for the preconditions to improve, the authorities and the banks must compensate for the shortcomings.

Let me stress this again: Your banks, as well as other financial institutions, cannot operate efficiently and soundly in an environment where the necessary preconditions are inadequate.

Banks perform special and unique services to society, such as accepting deposits and transforming them to credits, and for providing payment services. The combination of receiving short-term deposits and extending long-term loans makes banks inherently unstable. For this they may receive some protection by the public, such as the possibility of liquidity support and depositor guarantees. But the price for receiving protection is that banks must be subject to some regulation and supervision. Having acknowledged this, I am not a fan of extended bank regulation but rather regulation of a limited and transparent type, intended to protect society against major incidents and costs but allowing banks to conduct and develop their business as flexibly as possible.

Clearly, bank regulation must aim at promoting efficiency and stability. Only efficient banks will remain stable in the long run. The old-fashioned type of regulation forcing banks to do business on other than market-oriented terms must be abolished. If the government wants to subsidize certain activities, such as banking for people in remote geographical locations, mortgage lending on favourable terms, or lending to important borrowers or projects, the costs for this should be transparent and show up as a part of the fiscal budget. Such activities should not be conducted by directives from the government to the banks.

Another type of regulation which is becoming increasingly obsolete and inefficient is when the authorities try to micro-manage the behaviour of the banks by setting upper or lower limits on fees, deposit rates, lending rates, and credit expansion. The situation in the economy changes so rapidly that the authorities will never be able to catch up and their actions could then be more harmful than helpful.

The type of regulation I would like to promote is flexible. The regulation will not obstruct developments in banking while at the same time it prevents banks from behaving in a way that might harm customers or themselves, other institutions or markets, or society as a whole. Without going into detail I could mention the new Basel II framework for capital requirements on banks as an example of such flexible regulation. Under Basel II, banks may assume risks as long as they can prove to the supervisor that they have adequate capital to back up those risks. The banks must also have the necessary governance structures to ensure that they have the capability to identify, manage and control all major risks.

Please do not misunderstand me; I am not advocating that all countries should adopt the Basel II, not at all. In fact, it would even be hazardous for countries to adopt Basel II unless the underlying preconditions and regulations for basic banking are adequate. My point is only that Basel II is a type of regulation which maintains a balance between “carrots and sticks”. It lets the banks do their business as long as they manage and control it well, but if there are shortcomings in the banks’ handling then Basel II provides increased powers to the supervisor to intervene at an early stage to rectify the problems.

To sum up, Basel II mixes responsibility and flexibility for banks with strong supervision by authorities and the general public.

Whenever appropriate I would also promote a functional rather than an institutional approach to regulation. This means that a financial instrument or activity is regulated in the same way, whether it is linked to a bank or another type of financial institution. Such treatment ensures fair competition and facilitates the development of an open financial system. An institutional approach implies that there is different legislation for different categories of financial institutions, also in the cases where they perform similar services. This may sometimes be necessary, but should be avoided since it may limit market competition and development.

Another divide goes between principles-based and rules-based regulation. In the first category, general principles are provided such as “banks shall be managed in a safe and sound manner”. The authorities may then interpret this and take measures against a bank when they see fit. Rules-based regulation is more precise and says, for instance what a bank must do in order to ensure that it is managed in a safe and sound manner. I do not have a one-way view on this issue. The choice depends to some extent on your country’s legal traditions. Recent financial regulation in my own country mixes the two approaches, setting the broad principles in the legislation and delegating to the authorities to formulate the more precise and detailed rules. This approach also fits well with the fact that the financial sector is developing fast. It is quicker to change secondary regulations than the legislation itself.

When regulating financial activities, it is useful to apply an analysis of the costs and benefits to society. There is sometimes a tendency by the authorities to try to solve all problems by more regulation. It is true that we could regulate away all the risks in the banking sector – but such regulation would hamper economic development. There must be a balance. Thus for each regulation we introduce we must also conduct a fair analysis of its costs, also non-financial, as well as its benefits, also non-financial. I agree that this is not easy, partly since in most cases you will not find any clearly measurable indicators, but the process of conducting the cost/benefit analysis will in itself help you in your decision.

Good regulation is closely linked to good supervision. Supervisors must be able to build their work on a broad and relevant regulatory framework. Supervisors must also possess the necessary powers and other prerequisites to conduct efficient work. There are at least four important aspects:

First, the supervisory authority must have operational independence from industry as well as from politicians. These must not interfere in the operational decisions of the supervisors, e.g. to take remedial measures or to close banks. Such decisions must be taken on purely prudential grounds. Of course, the supervisors are still responsible for their actions and could be criticized afterwards, e.g. in Parliament hearings, but they must be able to perform their operational duties independently. Also the bank, its owners and management should be able to sue the supervisors for malpractice and they may receive compensatory payments. However, this should not stop an action started by the supervisor to deal with a presumed problem in the bank.

The supervisory management and staff must also have reasonable protection should they be sued for their bona fide decisions taken as supervisors. In some countries the supervisors are harassed by frivolous lawsuits by bank owners, managers or other parties. Even if the supervisor is in the end acquitted from any guilt, the process may take years. During this time the supervisor will be severely hampered in performing her job having to concentrate on her defence. Such lawsuits will also reduce the willingness of other supervisors to take necessary supervisory decisions, since they are themselves afraid of being sued.

A first line of defence is to take all major supervisory decisions in a collegiate fashion at the top level. Thus only the agency such can be sued as an institution. Nevertheless, should individual supervisors be sued, they must be provided assistance in the court proceedings, such as legal counsel and protection against any costs. Of course, if the supervisor is finally found guilty and have not acted in good faith, the supervisory agency could reclaim any outlays.

The second issue is that the supervisors must have a sufficiently large and skilled staff with satisfactory resources. With too few, or not adequately skilled supervisors, bank problems may not be detected early enough which may lead to major crises. With too many supervisors interventions in the banks might become excessive, thus interfering in banks’ daily business. Supervisors should never act in a way which implies that they assume responsibility for the banks’ ongoing activities.

Third, the supervisors must have a broad range of powers at their disposal to address different banking problems. This range should include limited measures such as requesting changes in the bank’s management or requesting improvements in risk management or control management. Also more far-reaching measures should be available such as stopping or restricting certain payments or activities, or the ultimate measures, to withdraw the banking license or to liquidate the bank. Preferably, some of these powers should be “pro-active” so the supervisor is allowed to act even before a problem becomes acute, for instance when the capital ratio is declining but has not yet reached the minimum level.

Fourth, the supervisors must have a well defined and documented work process. This includes a set structure of approaches to collect and analyse information and to use this as a basis for taking action. The approaches are, for instance, offsite monitoring, onsite examinations and ongoing contacts with the bank. Offsite monitoring means analysing regulatory and financial reports from banks, from their internal and external auditors, and from external media. Onsite examination is a tool to ensure that the bank actually works as reported and to gain further insights into the bank’s operations. The modern approach in this field is not to look at all transactions and documents but rather at a sample of those, including the more important ones. In addition to offsite and onsite, modern supervision includes closer, more frequent and less formal contacts between supervisors and banks. These contacts take place on different levels – on the top level of Boards and managements but also on various mid-levels and staff levels. Sometimes also the owners are contacted. The aim of the contacts is to “know your banks” meaning that the supervisors should assess how the banks operate, and if the bank managements and owners are competent and honest. Contacts, onsite and offsite supervision should be integrated so that the onsite supervisor benefits from the knowledge gained from the offsite monitoring, the contacts and vice versa. A good way to achieve this is to establish bank-specific groups of supervisors composed of offsite as well as onsite staff who are responsible for the supervision of a specific banking group.

There is a strong development towards what is known as risk-based supervision although a globally agreed definition of this concept is still lacking. Under risk-based supervision the supervisory authority uses its main resources on the major risks. With this approach, the focus is on larger banks but also on banks with more risky activities and on banks which have shown to be weaker or more vulnerable than the others. A risk-based approach also means that the supervisor focuses on collecting information from banks about those activities or operations where there might be material risks. In addition, in risk-based supervision it is important to check the corporate governance of the bank including the control mechanisms. Where there is a good balance between the Board and the managers and staff and where the control functions such as the internal audit do a good job, the supervisor can to some extent rely on their reports. In summary, “risk-based” leads to less intervention in banks as long as they are well run. Of course, this does not mean that some banks will never be supervised – all banks must provide their periodic supervisory and financial reports and they will also receive onsite visits, although some banks less frequently than others. But we should be aware that risk-based supervision also brings its own risks, namely that some minor banks may run into problems which are not detected in time – depositors and other counterparts may suffer losses.

During the last twenty years, many countries have embarked on reforms of the banking and financial services industry. Such reforms provide obvious benefits in the form of a well functioning financial sector that better supports economic growth and financial stability.

You need to retain some regulation and supervision but in more flexible and market-oriented forms than earlier. The balance between costs and benefits must be secured, as well as the balance between the objectives of financial stability and efficiency. The goals of regulation and supervision should be explicit and objective, and may include overall financial stability as well as consumer protection. The regulation should allow for a wide range of banks and other financial institutions conducting their activities on a level playing field. The degree of regulation and supervision should be commensurate to the size, nature and complexity of their activities.

I wish you luck in your endeavours to reform your banking and financial services industry. As you know, you may call on the IMF to assist you with advice or TA for capacity building or for other purposes.