This section gives a global explanation of BIS2. It is intended to inform those who are new to BIS2 and banking. For instance a manager new to the industry, who has to be aware of BIS2 but is not interested in the details. The section is also a good start for beginners before going into the details of BIS2.
There are two articles in this section. First a short description is given of the history and function of the Bank of International Settlements (BIS). Second the principles behind regulatory capital and BIS1 is explained. Third the BIS2 framework is explained.
BIS stands for the “Bank for International Settlements” and is located in Basel. It was founded in 1930 during the aftermath of world war one. Its task was to regulate the reparation payments from Germany. This initial task of collection administration and distribution of money gave the bank its name. Besides this task the bank stimulated the development of and cooperation between central banks.
As the role of payments administrator diminished the role of coordinator grew. The BIS developed into a bank dedicated to cooperation between banks, in order to assure financial and monetary stability. The BIS accomplished this by organising meetings between representatives of central banks and regular banks. Furthermore economic research was done.
After world war two the BIS activities shifted to the main monetary events in the decade. Between 1940 en 1970 the activities mainly involved the Bretton Woods system. This was followed by the oil crisis and the debt crisis in the seventies and eighties. The debt crisis led to the desire for international banking supervision and standards. The BIS coordinated these efforts which in turn led to the first Basel Capital Accord in 1988.
The purpose of this accord was to ensure that financial institutions had enough equity to remain stable during a crisis. This would ensure that the banking system would retain its stability and therefore its credibility.
After the implementation of the first Basel Capital Accord it became apparent that the system was fairly coarse. It does not differentiate well between risks. Banks with riskier portfolios did not necessarily have to retain more equity and vice versa. Therefore discussions were started to improve the current Basel Capital Accord. This new accord is called “International Convergence of Capital Measurement and Capital Standards; A Revised Framework”, affectionately known as BIS2.
Banks hold a special role in the economy. They function as intermediates between those who have money to spare and those who need money to invest. Basically those who have money to spare open saving accounts at banks. Banks lend that money to those who need it to invest. This is an extremely simplified version of reality but close enough for our purpose.
From the accountants point of view the money deposited at a bank are loans given to the bank. In other words they are liabilities. The loans given by banks are the banks investments. In other words they are assets. Below a simplified balance sheet of a bank is given.
|Total Assets: 610.000||Total Liabilities: 610.000|
The “Cash and balances” represent the normal working capital necessary for maintaining a business. The “Trading assets” are short term investments in financial instruments like stocks, bonds, derivatives, bank-note’s, repo’s, etc. Many banks buy and sell such products for speculative profits or risk mitigation (hedging). “Loans and receivables” are the core business of a bank. They are the loans which have been given to customers (those who need money to invest). Most income is generated by the interest received on “Loans and receivables”. The “Property and equipment” are buildings etc.
“Equity” is the portion of the banks balance sheet put up by the stockholders and increased through accrual of profits. There is no direct claim on the equity. In other words there is no obligation to repay the stockholders. The “Amount due” is the counterbalance of “Loans and receivables”. This is the money put at the bank by those who have too much (for instance the savings accounts). The “Loans and receivables” and “Amount due” represent the redistribution function of a bank. The bank is obliged to repay those accounts, although they remain fairly stable through time. The “Issued Bonds” is an extra source of long term money to bridge the gap between the “Loans and receivables” and the “Amount due” and to fund the additional investments. These bonds will mature, at which point the bank will have to repay the loan and find new money to fill the gap. The “Provision” can be seen as a reservoir to absorb losses.
A general risk in banking is the fact that occasionally people go broke, and fail to repay their loan. If this happens the value of “Loans and receivables” decreases. Fortunately there is no direct link between “Amount due” and “Loans and receivables”. If there were, you could lose your savings because some company you never heard of went broke. This is unacceptable; it would cause a bank run and destroy the economy. Therefore this risk is carried by the bank. The decrease in value has to be reflected somewhere in the liabilities in order to keep the total liabilities and assets equal. There are only two items available to the bank, “Equity” and “Provisions”. Luckily the “Provisions” reservoir has been created specifically to absorb these losses. Each year the credit losses (losses due to failed loans, or the decrease in the “Loans and receivables”) are subtracted from the provision. The provision is subsequently topped up by taking a provision from the profits.
The reservoir of provisions is made for a normal (expected) amount of losses due to lenders failing on their loan. A general provision is made based on the average historical percentage of loans that fail (the general provision). Furthermore a provision is made for specific loans which are considered in trouble (the specific provision).
History has shown that the reservoir is not always deep enough. The provision is based on credit losses under normal circumstances. However at some point abnormal circumstances will occur. Furthermore losses might occur due to different risks, like bad management. Abnormal credit losses occurred, for instance during the debt crisis in the eighties. So many large loans were failing that provisions fell short. The banks had to take the decreasing value of “Loans and receivables” by decreasing the value of the "Equity". This caused a lack of "Equity", which jeapordises the stabiltiy of the banks and the stability of the monetary system.
To ensure this would no longer be possible, it was determined that banks should have a minimum buffer of equity and provisions called the regulatory capital. The buffer means that a minimum amount of equity and provisions (regulatory capital) should be present for a bank given the value (and risk) of “Loans and receivables”. Or the other way around, a maximum amount of loans may be sold given the amount of equity and provisions. The size of the regulatory capital was determined by the Basel capital accord. The essence of this buffer was to ensure that a bank could absorb the losses sustained even in the most extreme circumstances.
The regulatory capital determines the amount of capital required to cover the risk of losses by banks. For obvious reasons not all sources of capital qualify as a buffer against risk. For instance, bonds issued by a bank are burdened by a crippling obligation. The bank has to repay the loan. This means that this money cannot be used to absorb a loss, as eventually the money will have to be raised to repay the bond holders. Only capital without strings attached (or few strings attached) can be used as a buffer. Basically a bank should be able to consume the capital without failing subsequent obligations.
For this purpose the capital eligible as regulatory capital is divided into two types, according to the amount of strings attached.
Tier one capital is capital without any (or very few) strings attached.
Basically it contains the banks equity (Issued and fully paid ordinary shares/common stock and non-cumulative perpetual preferred stock, excluding cumulative preferred stock). Additionally some forms of mezzanine capital with properties like equity, can be recognised as tier one capital by your central bank.
Tier two capital is capital with either some string attached or some difficulties considering its value.
Basically it contains the banks undisclosed reserves, revaluation reserves, general provisions/general loan-loss reserves, hybrid debt capital instruments and subordinated term debt (with a minimum term to maturity of 5 years for BIS 2). There may not be more tier two capital than tier one capital. Excess tier two capital will not be excepted as an eligible buffer
Tier three capital consists of short term subordinated debt. A bank may only use for covering market risk after the central banks consent (it is a t the central banks discretion). The capital covering the market risk should consist of at least 28.5% tier one capital. The remainder may be tier two or three. Although the total tier two capital must remain equal or les than the tier one capital. Some central banks may require the sum of the tier two and three capital to be less or equal to the tier one capital. This is at the discretion of the central bank.
For BIS 1 the regulatory capital was defined as the buffer for absorbing all losses. In BIS 2 a split is made between expected losses and unexpected losses. Every year some loss can be expected and provisions are made to cover these losses. Besides the expected losses there is a probability that a certain amount of unexpected loss occurs. The regulatory capital for BIS 2 (if you use IRB) is calculated to cover the unexpected losses only. Therefore the general provisions/general loan-loss reserves are not included in the tier two capital for BIS 2. In other words, under BIS 2 the general provisions/general loan-loss reserves are subtracted from both the regulatory capital and the tier two capital. If for some reason you have more general provisions/general loan-loss reserves than your expected loss, the excess may be added to your tier two capital (up to 0.6% of risk weighed assets). If you have less, the difference must be subtracted. A subtraction must be done on both tier one and tier two capital (50% on both). The general provisions/general loan-loss reserves used for the calculation should not include general provisions/general loan-loss reserves for equity exposures
Certain balance sheet items have to be subtracted from the tier capital available. If an asset is placed on the balance sheet with a dubious high value it will most probably be reflected in the equity, as both the total assets and total liability must remain equal. However this increased equity does not represent real value on which the bank can rely. For this reason goodwill should be deducted from tier one capital. Secondly, if a bank has a banking subsidiary which has not been consolidated (and thus covers its own regulatory capital by its own tier capital), its value should be subtracted from the total capital of the bank. This is done because the extra equity created by the investment in the subsidiary is already used to cover the regulatory capital of that subsidiary. It should not be possible to also use this capital for the mother.
For BIS 2 an additional subtraction is necessary. If the expected loss for equity exposures is calculated using the PD/LGD method (see article on this site), this expected loss should be subtracted from the capital (50% from tier one and 50% of tier two).
During the development of the first capital accord it was determined that the main risk of a bank is credit risk. This is the risk that those who have to repay their loans to the bank, don’t. The maximum credit loss that a bank can suffer is equal to the value of all its outstanding loans and investments. This will never occur because it means that all the banks customers will have to go broke. Based on experience it was determined that no more than 8% of the amount of money lend to customers would fail in the most extreme circumstances. In other words 8% of regulatory capital should be present for the banks outstanding loans.
Because some loans have very secure collateral it was allowed to subtract some forms of collateral from the loan amount (outstanding amount) under BIS1. If an outstanding amount was three million and there was one million of cash collateral, the net loan over which capital should be calculated was two million. Furthermore some customers and loan types were considered less risky. For this reason these loans were risk weighted. Home mortgages are for instance rather save investments. For this reason these mortgages (under some conditions) are weighed at 50%. This means that a mortgage of two million is weighed at 50%. Thus regulatory capital must be retained for one million. This leads to a regulatory capital of 80,000 (8% of one million).
This system has a rather course risk sensitivity. Consider the following two corporate loans.
1. A corporate loan of two million to a multinational AA corporation, fully collateralized by machinery.
2. A corporate unsecured loan of two million to a paper company (no actual assets or income production capacity) in a tax haven.
In the BIS1 system an equal amount (160,000) of capital has to be retained for both deals, even though it is plain to see that the second loan is much riskier than the first.
BIS II is a redevelopment of the original capital accord BIS I. However BIS II encompasses more than just a minimum capital for credit risk. The total structure of the BIS II accord can be graphically presented as follows:
| Pillar One
Minimum capital requirements
| Pillar two
Supervisory review process
| Pillar three
The structure of BIS II consists of three main pillars.
The first pillar is the main part of BIS II. It comprises 192 pages of the total 242 pages (excluding the annexes). The pillar contains the rules for calculating the regulatory capital. The rules for capital requirements are given for three types of risk.
Credit risk is the risk of loss because a customer does not repay the full loan. Operational risk is the risk of loss due to operational issues (bad management decisions, fraud, incorrect administration, etc.). Market risk is the risk of losses due to a decrease in the value of investments (for instance stock or bond investments).
Credit risk is split into three more sections.
In the standardized approach the credit risk is calculated using standard benchmarks. This method is much like the BIS I method. The internal ratings based approach allows a bank to develop and use its own risk models to calculate the credit risk. A bank must choose which method to use. The third section is the securitisation framework. More and more banks sell the risk (and return) associated with a part of their assets (for instance the mortgages). Because in these cases the risk is (partly) not run by the bank the actual credit risk will have to be calculated.
The second pillar considers the method in which banks and central banks should review the implementation and usage of BIS II. This pillar requires banks to develop their risk management beyond the minimum requirements set out in pillar one. Additional risk types such as interest rate risk should be incorporated in this more comprehensive risk management system. Furthermore it dictates that the BIS II framework should be an integral part of the banks activities. This means for instance the banks general management should be aware of BIS II and should be involved in the decisions surrounding BIS II.
The third pillar is the inspiration for this site. It dictates that banks should publish the outcome of their risk calculations and also their methodology. The idea behind this requirement is that the market will guide the banks towards a set of well established methodologies, thereby making the capital requirements stable throughout the banking world.
The purpose of the first pillar is to set out the rules by which the regulatory capital can be calculated. The pillar is divided in three types of risk for which capital should be held.
Credit risk is the risk that those who owe you money will not pay you back. Historically credit risk is the larges risk banks run. It is at the core of its business. BIS II proposes three approaches by which a bank may calculate its required capital for credit risk.
This approach is a slight improvement on BIS I. It recognises a few more collateral types and has a few more risk weights for different asset types. Beyond that the approach is very similar to BIS I.
Internal rating based (IRB) advanced.
The internal rating based (IRB) advanced approach is the most comprehensive approach suggested by BIS II. In this approach a bank may develop their own risk models. The central bank will monitor a bank on the quality of its risk management to ensure robust calculations of the regulatory capital.
Internal rating based (IRB) foundation.
The internal rating based (IRB) foundation approach is a combination of the previous two. It is comparable to the advanced approach but with standard values given by BIS II for some of the risk parameters.
Operational risk is defined by BIS II as follows:
Operational risk is defined as the risk of loss resulting from inadequate or failed
internal processes, people and systems or from external events. This definition includes legal
risk, but excludes strategic and reputational risk.
The basic indicator approach is a very simple approach. It states that a bank should calculate required capital for operational risk by multiplying the average positive annual gross income (ignore the negative annual gross income) over the last three years by 15%.
Market risk is the risk of losses due to changes in the market price of an asset. Obviously this is only relevant if the asset is obtained for trading. If you buy a bond and intent to hold it until its maturity, the market price movements of the bond are irrelevant (as it won’t be sold). Therefore the market risk will only have to be calculated for assets in the trading book. Besides this risk the foreign exchange rate risk and commodities risk are part of the market risk. These two risk are not limited to the trading book and should be calculated for the entire bank.
Again several methods may be used:
The standardized approach dictates a set of rules for calculating the capital charge for different types of financial products available for trading.
A bank should choose which method to use for each risk type. In summery pillar one describes the following risks and measurement methods:
The second pillar considers the method in which banks and central banks should review the implementation and usage of BIS II. This pillar requires banks to develop their risk management beyond the minimum requirements set out in pillar one. Additional risk types such as interest rate risk should be incorporated in this more comprehensive risk management system. Furthermore it dictates that the BIS II framework should be an integral part of the banks activities. This means for instance the banks general management should be aware of BIS II and should be involved in the decisions surrounding BIS II. The BIS II paper itself defines the purpose of pillar two as:
The supervisory review process of the Framework is intended not only to ensure that banks have adequate capital to support all the risks in their business, but also to encourage banks to develop and use better risk management techniques in monitoring and managing their risks.”
In pillar two, four principles of supervisory review
The first principle is divided into five main features
The third principle is not divided into sub features. The idea behind this principle is to ensure that supervisors consider the pillar one, capital calculation a minimum requirement. A supervisor should expect
banks to hold capital in excess of the minimum. This can be necessary for risks which have not been incorporated in the first pillar. It can also be advisory because of bank specific issues. Or a bank may wish a larger buffer in order to position itself as a safe bank in the market. Furthermore the capital requirement will change through
time. If the available capital equals the required capital any negative change will lead to a breach of the regulatory capital requirement. No bank is expected to find such a situation acceptable.
The fourth principle is also not divided into sub features. This principle states that a supervisor should signal a move towards inadequate capital, before it is actually deemed inadequate. A supervisor has several actions which it can take if there is a serious thread that the available capital becomes less than the required capital. It can increase monitoring of a bank, Forbid (or restrict) payment of dividends, require the bank to create and implement a capital restoration plan or force a bank to attract capital from the market immediately. These actions may be used permanently or temporarily while a bank is improving the situation at hand.
The third pillar dictates that banks should publish the outcome of their risk calculations and also their methodology. The idea behind this requirement is that the market will guide the banks towards a set of well established methodologies, thereby making the capital requirements stable throughout the banking world.
The third pillar requires banks to publish the outcome of several calculations and risk estimates, but also the methods and assumptions used to make those calculations. In other words banks need to publish qualitative information and quantitative information. Investors can use both to determine whether: