International Regulation - Modern Banking

The international coordination of prudential regulation at global level is increasingly important. There has been a rapid growth of international banking, and financial conglomerates. A number of arguments favour global coordination of prudential regulations.

First, policy makers, bank management and regulators recognise that problems with the global institutions and markets could undermine the stability of the international financial system, and therefore the environment in which all banks operate.

Second, if a branch or subsidiary of a bank is located in another country, there is the question of which supervisory authority should have jurisdiction over the branch.

Home country regulators will want to ensure a bank’s overseas operations meet their supervisory standards because foreign operations will be difficult to monitor but affect the performance of the parent. Host country authorities are concerned with the effect the failure of a foreign bank could have on the confidence in its banking system.

They will want to see the foreign branch to be adequately supervised, but will lack information about the parent operations. For these reasons, effective international coordination will only be achieved if there is good communication between the supervisory authorities.

Third, if all multinational banks are required to meet the same global regulations, compliance costs will be similar. Hence a global approach to regulation can help to level the competitive playing field for banks with international operations.

It is worth noting that decisions taken by international regulators are increasingly being used as benchmarks for other banks. For example, the Basel Committee’s (see below) 1988 agreement on capital standards was adopted by not only the member countries, but also by governments that were not signatories to the agreement. Also, many regulators impose the Basel agreements on domestic banks.

On the other hand, it could be argued that international banking is largely wholesale, making prudential regulation less important from the standpoint of consumer protection, depending as it does on interbank and corporate business. However, the performance of a global bank will affect the confidence of depositors and investors located in the home country. Unprotected wholesale depositors are capable of starting bank runs, and the enormous size of the interbank market creates the potential for a rapid domino effect. Often, the first indication of a problem bank is when it has trouble raising interbank loans – wholesale depositors will be the first to withdraw their money.

An excellent example of this phenomenon is Continental Illinois Bank, rescued by a ‘‘lifeboat’’ in 1984. The bank was highly dependent on the interbank markets for funding, which was quickly cut off once rumours about its health began to circulate. The rapid loss in liquidity merely exacerbated the problems, prompting a rescue organised by the Federal Reserve. Furthermore, if a global bank acquires a bad reputation as a result of some international transaction, and has a retail presence in its home country, it may find itself the target of a run. Finally, global financial conglomerates, if they get into difficulties, can cause problems in more than one country.

The Basel Committee

Two major international bank failures in 1974 (Bankhaus Herstatt and Franklin National Bank) resulted in the formation of a standing committee of bank supervisory authorities, from the G-10 countries (Belgium, Canada, France, Germany, Italy, Japan, Netherlands, Sweden, the UKand USA) plus Luxembourg and Switzerland. It has a permanent secretariat (of 15) based at the Bank for International Settlements in Basel, and meets there about once every three months.

The Bank for International Settlements is owned by the central banks – it does not participate in Basel’s policy-making, provides a venue for the Committee’s secretariat and for membership meetings. Traditionally, members came from western central banks but since 1994, there are 13 member central banks from emerging markets.

The main purpose of the Basel Committee is to consider regulatory issues related to activities of international banks in member countries. Their objective is to use concordats and agreements to prevent any international banking operation from escaping effective supervision.

The 1975 Basel Concordatwas the first agreement. The home and host countries were given supervisory responsibilities as follows:

  • Country supervisor
  • Liquidity – foreign bank subsidiary Host
  • Solvency – foreign bank subsidiary Host
  • Liquidity – foreign branch Home

The Concordat stressed that consolidated data should be used to supervise the activities of a global bank, and provide an accurate picture of performance. Offshore banking centres are not party to this agreement – the Committee did not consider them to pose a major threat to international financial stability because their operations are relatively minor.

In 1983 the Committee approved a Revised Basel Concordat, when gaps in the supervision of foreign branches and subsidiaries came to light after the Banco Ambrosiano affair.

Banco Ambrosiano Bank failed in 1982, after its Chairman, Roberto Calvo, was found hanging from Blackfriars Bridge in London. Depositors panicked upon hearing the news; a lifeboat rescue was launched by the Bank of Italy ($325 million), but the bank was declaredbankrupt in late August 1982. The bank’s Italian operations were taken over by a new bank, Nouvo Banco Ambrosiano.

The Luxembourg subsidiary (BA in Milan owned 69% of Banco Ambrosiano Holdings) also suffered a run on deposits, but the Italian central bank refused to inject any cash. Nor would the Luxembourg Banking Commission. It, too, failed. As a result of this case, the Concordat was revised so that home and host supervisors now have joint responsibility for solvency problems of subsidiaries and liquidity problems from either a subsidiary or branch.

Solvency problems associated with any foreign branch are dealt with by the parent country’s central bank.

A number of issues were not addressed by either Concordat. First, no reference was made to lender of last resort (LLR) responsibilities. Recall a lender of last resort normally aids a bank in the event of a liquidity crisis. Lifeboat operations serve a similar purpose, where the central bank persuades other healthy, private banks that it is in their interest to inject liquidity into the ailing bank.

However, the Basel Committee did not feel able to offer guidelines because the LLR function is normally assumed by central banks, and the Basel Committee members do not necessarily come from the central bank.10 However, LLR intervention or lifeboat rescues have been quite frequent in most westernised countries in the post-war period.

There will be problems with achieving satisfactory international coordination if a run on foreign branches or subsidiaries occurs because the parent has run into difficulties.

Guttentag and Herring (1983) identified three types of banks that are vulnerable under the current arrangements: banks headquartered in countries with no LLR facilities (such as Luxembourg); banks headquartered in countries with non-convertible currencies or a shortage of foreign exchange reserves; and subsidiary banks with ambiguous access to the parent bank facilities.

The Basel Committee also side-stepped another issue related to financial stability – the extension of deposit insurance to all deposit liabilities. Normally deposit insurance excludes wholesale and interbank deposits, on the grounds that these depositors are better informed about the financial health of a bank and therefore do not need it. Foreign currency deposits tend to be excluded because of the concern that deposits might be shifted between the foreign bank and its parent, to the detriment of the former. However, if deposit insurance was expanded, its effects on moral hazard would have to be considered.

The 1988 Basel Accord (Basel 1)

The 1988 Basel Accord was a watershed because it established Basel’s main raison dˆetre: to focus on the effective supervision of international banking operations through greater coordination among international bank supervisors and regulators. Improved international financial stability would be a key consequence of the Committee’s actions. The 1988 Basel Accord established a single set of capital adequacy standards for international banks of participating countries from January 1993.11 With the arrival of a new revised accord (see below), the 1988 Accord will be known as Basel 1 henceforth.

Basel 1 requires all international banks12 to set aside capital based on the (Basel) risk assets ratio:

Basel risk assets ratio = capital / weighted risk assets


Capital (tier 1 & 2)
Assets(weighted by credit type)+credit risk equivalents(weighted by counterparty type)

where capital is defined as follows.

  • Tier 1 or core capital: common equity shares, disclosed reserves, non-cumulative preferred stock, other hybrid equity instruments, retained earnings, minority interests in consolidated subsidiaries, less goodwill and other deductions.
  • Tier 2 or supplementary capital: consisting of all other capital but divided into (1) upper tier 2 – capital such as cumulative perpetual preferred stock, loan loss allowances, undisclosed reserves, revaluation reserves (discounted by 55%) such as equity or property where the value changes, general loan loss reserves, hybrid debt instruments (e.g. convertible bonds, cumulative preference shares) and (2) lower tier 2 – subordinated debt (e.g. convertible bonds, cumulative preference shares).

Risk weights are assigned to assets by credit type. The more creditworthy the loan, the lower the risk weight.

  • 0%: cash, gold, bonds issued by OECD governments.
  • 20%: bonds issued by agencies of OECD governments (e.g. the UK’s Export and Credit Guarantee Agency), local (municipal) governments and insured mortgages.
  • 50%: uninsured mortgages.
  • 100%: all corporate loans and claims by non-OECD banks or government debt, equity and property.

Off-balance sheet instruments (e.g. letters of credit, futures, swaps, forex agreements) were converted into ‘‘credit risk equivalents’’, and weighted by the type of counterparty to a given claim. Again, OECD government counterparties receive a 0% weight; 20% for OECD banks and public sector agencies.

Example: SimpleBank Simple Bank plc has the following balance sheet (£ billions):

balance sheet (£ billions)

From the information given in the balance, tier 1 and tier 2 capital are:

Tier 1 = £15 + £2 = £17 billion Tier 2 = £5 + £3 = £8 billion

Assuming capital is defined as tier 1 + tier 2, total capital = £25 billion.

A simple capital assets ratio, with assets unweighted, would be capital (tier 1&2)/assets = 25/205 = 12.195%.

Assuming the Basel 1 agreement applies, the use of weightings would change the denominator of the risk assets ratio for Simple Bank:

2(0)+30(0)+20(0.2)+50(0.5)+103(1) = 4+25+103 = £132bn

The Basel 1 risk assets ratio is 25/132 = 18.9%.

This ratio is higher than the simple capital assets ratio because assets are now weighted, hence some assets (cash and OECD government bonds) go to 0 or are lower than if unweighted.

So far, off-balance sheet items have been ignored. Suppose the off-balance sheet items of Simple Bank have been computed and equal £13 billion. Then the denominator of the risk assets ratio becomes 132 + 13 = 145, and the risk assets ratio is 25/145 = 17.24%.

The Basel Accord requires banks to set aside a minimum of 8% capital; 4% for core capital. At least half the capital must be tier 1, and is set aside as a safeguard against bad credit or counterparty risk. As any July edition of The Banker shows, the average risk assets ratio for the top UK, US and other OECD banks is in fact much higher. The Banker ranks the top 1000 banks by tier 1 capital and reports other performance data, including the Basel risk assets ratio. Of the bottom 25 banks measured by the Basel ratio, it ranges from just over 8% to slightly below 4%. The vast majority of the top 50 banks (ranked by the Basel ratio) are reporting double digit risk assets ratios, some even exceed 100%. Many OECD regulators ask for higher ratios. For example, in the USA, to be labelled ‘‘well capitalised’’ banks must have a Basel ratio ≥10%.

Before the agreement was even implemented, Basel 1 was being criticised for a number of reasons. Some argued using equity as a measure of capital fails to recognise that different countries allow their banks varying degrees of access to the stock market. For example, French nationalised banks in the 1980–90s had no access and relied on government injections of capital. Though privatisation has largely resolved this issue, it does demonstrate the potential problem with using equity. On the other hand, ignoring equity would be unthinkable because it is a key source of capital for shareholder owned banks. A more serious debate is whether the book or market values should be used in the computation of tier 1 and 2 capital.

The difference between market and book values of equity is more pronounced in periods of interest rate and stock market volatility, and, indirectly, if changes in credit ratings raise or lower asset values. Regulators opted to use the book rather than the market value to compute the capital assets ratio largely because of the potential for volatility. Using market values can be the source of wild fluctuations in tier 1 capital from year to year. In the Japanese case, tier 1 would have soared in the 1980s, only to fall dramatically from late 1989 onwards, thereby adding to the pressure to find new capital (and/or reduce assets) to meet the 8% minimum.

Ambiguity about the constituents of tier 1 and 2 capital has encouraged agents to innovate to get round the regulations. Also, different standards apply in each country. Take tier 2 capital as an example. In the 1990s, Japanese banks could not issue subordinated debt but US banks did. Also, Japanese regulators allowed their banks to treat 45% of unrealized capital gains on cross-shareholdings as reserves for tier 1 capital, though regulations have since been tightened.15 Nonetheless, these points illustrate that it is difficult to obtain comparable measures of tier 1 and 2 capital. Scott and Iwahara (1994) argued differences in tax and accounting rules cause the measurement of capital to vary widely among countries, rendering different countries’ risk assets ratios incomparable.

Second, the Accord alone could never achieve the objective of a level playing field among international banks, because the degree of competition in a system is determined by other factors, such as the structure of the banking system and the degree to which a government is prepared to support its banks. Until recently, Japan’s well-known ‘‘safety net’’ meant Japanese banks could borrow capital more cheaply from wholesale markets than banks from countries where failures have been allowed. Also, they received substantial capital injections throughout the 1990s and into the early 2000s.

The use of credit risk equivalents for off-balance sheet instruments was considered far too simplistic. Effectively all on- and off-balance sheet items were treated the same, and the market or price risk associated with the growing off-balance sheet activities of many banks were largely ignored. The credit risk equivalence measure took account of the possibility of default on corporate bonds but no capital had to be set aside to allow for the possibility that the price of bonds might fall with a rise in market interest rates. Or, an OECD government bond maturing in 30 years time carries a higher interest rate risk than one maturing in a year. In Basel 1, both receive a 0% weighting in the computations, which is acceptable from the standpoint of credit risk, but the different interest rate risks (arising from differences in maturity) are ignored. Nor are liquidity, currency and operating risks accounted for. For this reason, the 1996 Market Risk Agreement replaced the use of credit risk equivalents – it applies a capital charge for the market risk associated with all traded instruments, whether on- or off-balance sheet.

It should be remembered that national bank supervisors are monitoring banks’ exposure to these risks. For example, most regulators use a liquidity ladder to estimate liquidity exposure, and monitor short and medium-term foreign exchange exposure of the banks they regulate. Also, the Basel 2 proposals (see below) deal with some of the risks largely ignored by Basel 1. The weightings used in Basel 1 are simplistic. Commercial bank loans have a 100% weighting while OECD government debt is given a 0% weight, and OECD bank claims have a 20% weight even though some corporations have a higher external credit rating than the banks they do business with. For example, a loan to Marks & Spencer or General Electric, with AAA rating, receives a 100% weight, while loans to Italian or Japanese banks are weighted at 20%, even though the long-term debt rating for the top 5 Italian banks ranges from A+ to AA−; likewise for Japanese banks – long-term debt ratings for the top 5 vary from A to A−. All corporations get the same weight, independent of whether their rating is AAA or BBB.

The weight for corporations and other counterparties is 50% for off-balance sheet items converted into credit risk equivalents, just half the risk weight assigned to corporate loans.

Basel reasoned that only the most sophisticated banks were involved in off-balance sheet activity, hence the weight could be lower.

Such anomalies can and do tempt banks to engage in regulatory capital arbitrage – using a financial instrument or transaction to reduce capital requirements without a corresponding reduction in the risk incurred. For example, a bank may agree to a 364-day credit facility on a rollover basis because no capital need be set aside for credit arrangements between banks and a customer that are rolled over within a year. If the maturity of the agreement is a year or more, it is subject to the same capital regulations as a loan that matures in 30 years. More generally, banks are tempted to keep their capital charges to a minimum by exploiting loopholes even though the overall risk profile of the bank is higher.

Basel 1 does not reward banks which reduce their systematic risk – that is, no recognition is given for risk diversification of a bank’s loan portfolio. While the Accord limits the concentration of risk among individual customers, over-exposure in a particular sector is ignored. A bank which lends ¤500 million to two sectors will set aside the same amount of capital as a bank lending ¤1 million to 500 different firms. In general, banks with a highly diversified portfolio set aside the same amount of capital as a bank with the same total value of commercial loans concentrated in just one industry. Nor is there any reward for geographical diversification.

Basel 1 is accused of being a ‘‘one size fits all approach’’ – there is little recognition that banks undertake different financial activities. A US/UK investment bank in the USA has quite different risk profiles from universal banks engaged in wholesale and retail banking activities. The balance sheets of a UK building society or German savings bank will be quite unlike the large universal (e.g. Deutsche Bank) or ‘‘restricted universal’’ banks (e.g. Barclays Bank plc). Yet all these banks are expected to conform to the same risk assets ratio requirements.

The regulations act as a benchmark, which could give some banks a false sense of security, causing them to make sub-optimal decisions. For example, since loan concentration in a specific industry is ignored by the ratio, banks may become complacent about the lack of portfolio diversification across sectors. They may also allocate too many resources to satisfy the Basel requirements (or find ways of getting round them), at the expense of other types of risk management.

In defence of Basel 1, it is worth emphasising that the Accord called for a minimum amount of capital to be set aside. As was noted earlier, many of The Banker’s top 1000 banks by tier 1 capital have ratios far in excess of 8%. Furthermore, banks are subject to additional supervision in their own countries. For example the UK’s Financial Services Authority applies a ‘‘risk to objectives’’ approach to all financial institutions, including banks. It also requires banks to satisfy other criteria. American banks are subject to scrutiny by multiple regulators, and pay different deposit insurance premia depending on the size of three different ratios. Finally, managers of publicly owned banks must answer to their shareholders. If a stock bank were to engage in excessive amounts of regulatory arbitrage which substantially increases its risk profile, it would not be long before concerns were voiced by shareholders and national regulators.

Basel Amendment (1996) – Market Risk

The Basel Committee began to address the treatment of market risks in a 1993 consultative document, and the outcome was the 1996 Amendment of Basel 1 to be implemented by international banks by 1998. It introduced a more direct treatment of off-balance sheet items rather than converting them into credit risk equivalents, as was done in the original Basel 1.

Market risk is the risk that changes in market prices will cause losses in positions both on- and off-balance sheet. The ‘‘market price’’ refers to the price of any instrument traded on an exchange. The different forms of market risk recognised in the amendment include: equity price risk (market and specific), interest rate risk associated with fixed income instruments,19 currency risk and commodities price risk. Debt securities (fixed and floating rate instruments, such as bonds, or debt derivatives), forward rate agreements, futures and options, swaps (interest rate, currency or commodity) and equity derivatives will expose a bank to market risk. Market and credit risk can be closely linked. For example, if the rating of corporate or sovereign debt is upgraded /downgraded by a respected credit rating agency, then the corporate or sovereign bonds will rise/fall in value.

In the numerator of the Basel ratio, a third type of capital, tier 3 capital, can be used by banks but only when computing the capital charge related to market risk, and subject to the approval of the national regulator. Tier 3 capital is defined as short-term subordinated debt (with a maturity of less than 2 years), which meets a number of conditions stipulated in the agreement, including a requirement that neither the interest nor principal can be repaid if it results in the bank falling below its minimum capital requirement.

Whether the Amendment raises or lowers the capital charge of a bank depends on the profile of its trading book. However, as will be shown below, banks using the ‘‘standardised’’ approach are likely to incur higher capital charges, unless positions are well hedged or debt securities are of a high investment grade. Under the Amendment, one of two approaches to market risk can be adopted, internal models or standardised.

Market risk – the internal model approach

Banks, subject to the approval of the national regulator, are allowed to use their own internal models to compute the amount of capital to be set aside for market risk, subject to a number of conditions. The market leader is JP Morgan’s Riskmetrics. This shows what Basel requires of banks if they use a VaR model. Throughout, it will be assumed they are using the Riskmetrics model, so the key equation is:

VaRx = Vx(dV/dP)_Pt ---(XIV)

Vx: the market value of portfolio x
dV/dP: the sensitivity to price movement per dollar market value
ΔPt: the adverse price movement (in interest rates, exchange rates, equity prices or commodity prices) over time t

There are several critical assumptions underlying any VaR computation. Basel has certain specific requirements to be satisfied.

  1. Bank models must compute VaR on a daily basis.
  2. The four risk factors to be monitored are interest rates (for different term structures/ maturities), exchange rates, equity prices and commodity prices.
  3. Basel specifies a one-tailed 99% confidence interval, i.e. the loss level is at 99%; the loss should occur 1 in 100 days or 2 to 3 days a year. Recall the choice of 99% is a more risk averse/conservative approach. However, there is a trade-off: a choice of 99% as opposed to 95% means not as much historical data (if it is a historical database being used – see below) are available to determine the cut-off point.
  4. The choice of holding period (t in the equation above) will depend on the objective of the exercise. Banks with liquid trading books will be concerned with daily returns and compute DEAR, daily earnings at risk. Pension and investment funds may want to use a month. The Basel Committee specifies 10 working days, reasoning that a financial institution may need up to 10 days to liquidate its holdings.
  5. Basel does not recommend which frequency distribution should be used. Recall that Riskmetrics employs a variance–covariance approach. Banks that use variance– covariance analysis normally make some allowances for non-linearities, and the Basel Amendment requires that non-linearities arising from option positions be taken into account. For either approach, Basel 2 requires the specification of a data window, that is, how far back the historical distribution will go, and there must be at least a year’s worth of data. Generally, the longer the data run the better, but often the data do not exist except for a few countries, and it is more likely that the distribution will change over the sample period.

Computation of the capital charge using the internal model

If the bank is employing its internal model once VaR is computed, the capital charge is set as follows:

[MRM(10-day market risk VaR) + SRM(10-day specific risk VaR)][trigger/8] ------(XVI)


MRM: a market risk multiplier, which is 3 or 4 depending on the regulator – the lower the multiplier, the greater the reward for the quality of the model in its treatment of systematic risk

SRM: a specific risk multiplier, which can be 4 or 5 – a lower multiplier indicates a greater reward for the way a given bank’s model deals with specific risk

trigger: the number assigned is based on the assessment of the quality of a bank’s control processes, it can vary between 8 (assigned to US and Canadian international banks) and 25 – the higher the trigger number the higher the overall capital charge

If an internal market model is used, it is estimated that a bank could reduce its capital charge by between 20% and 50%, depending on the size of the trading operations and the type of instruments traded, because the bank’s model will allow for diversification (or model for correlation between positions) whereas the standardised model does not.

The 1996 Market Risk Amendment also introduced restrictions on the total concentration of risk. If the risk being taken is greater than 10% of the bank’s total capital, the regulator must be informed, and advance permission must be obtained for any risk that exceeds 25% of the bank’s capital.

Unlike the standard approach, banks using an approved internal model can allow for the correlation between four market risk categories: interest rates (at different maturities), exchange rates, equity prices and commodity prices. Thus, banks are rewarded for portfolio diversification that reduces market risk, and so reduces the capital they must set aside.

In a theme that continues in the Basel 2 proposals (see below), the Committee is encouraging banks to have a risk management system that not only satisfies regulatory requirements, but ensures the bank has a framework to manage all the risk exposures generated by its business activities. To be approved by the regulators, in addition to a VaR model (which meets the criteria discussed above), the risk management system should:

  • Allocate the capital to various business units.
  • Use RAROC to track the performance of each business unit and the bank as a whole.
  • Record all positions in a centralised system.
  • Conduct regular stress testing and scenario analysis to ensure their risk management systems can cope with extreme market conditions, such as a sudden loss in liquidity or exchange rate crises.
  • All models should be subject to a system of continuous evaluation, and tests of the risk management system should be done independently of traders and the front office.

Basel and Related Problems with the VaR Approach

The numerous problems arising from the use of VaR, many of which derive from the assumptions underlying the model, were discussed. One, perhaps unjust, criticism is that VaR does not give a probability of bank failure. However, it was never meant to because it is designed to establish a capital requirement for market risk, one of many types of risk the bank faces. Due to the amount of attention it has received, there is a tendency to forget that it deals with just one aspect of a bank’s risk. Nonetheless, there are other problems related to the use of the Basel VaR.

Under the current Basel rules, the more sophisticated banks may employ their own advanced risk models if the country regulator approves. However, all banks will have to meet the minimum VaR standards. In a crisis, all will react the same way.

Just as Goodhart (1974) demonstrated that statistical relationships break down once employed for policy purposes, Danielsson (2000, p. 5) argues that a model breaks down once regulators use a model like VaR to contain risk.

As shown in equation (XVI) Basel requires VaR to be multiplied by 3 (sometimes 4 if there are large differences between the actual and predicted outcomes) to determine the minimum capital requirement. The larger banks have objected because the incentive to use sophisticated models is reduced. Basel justifies the requirement because of the problems with the VaR approach. In the absence of strong evidence, Shin et al. (2001) recommend a reduction in the multiplication factor, to be increased if it is found that losses are under-predicted.

Basel requires capital to be set aside for market risk based on a 10-day time horizon, Danielsson (2002) demonstrates that the production of 10 (working) day VaR forecasts is technically difficult, if not impossible. For example, suppose 1 year (250 days) is used to produce the daily VaR. To compute a 10-day VaR, 10 years’ data would be needed. To get round this problem, Basel recommends taking the daily VaR and multiplying it by the square root of 10. However, Danielsson shows the underlying assumptions with respect to distribution are violated if the square root method is used.

Most banks employ very similar VaR models, or use the standard approach. However, banks differ widely in their objectives and exposure to market risk. A small savings bank or building society in the EU is unlikely to be exposed to much market risk but must adopt the standard approach nonetheless, which is costly. Other banks may be exposed to types of market risk not well captured by VaR methods.

Market risk – the standardised approach

Banks without an approved internal model for estimating market risk exposure are required to use Basel’s standardised approach. Recall the objective: to replace the credit risk equivalents used in Basel 1 with a more sophisticated treatment of off-balance sheet items.

No VaR computation is used. Instead, the amount of capital to be set aside is determined by an additive or building bloc approach based on the four market risks, that is, changes in interest rates (at different maturities), exchange rates, equity prices and commodity prices. In every risk category, all derivatives (e.g. options, swaps, forward, futures) are converted into spot equivalents. Once the capital charge related to each of these risks is determined, it is summed up to produce an overall capital charge. The computation does not allow for any correlation between the four market risk categories. Put another way, portfolio diversification is not accepted as a reason for reducing the capital to be set aside for market risk.

Equity risk

Determining the market risk arising from equities is a two-stage process, based on a charge for specific risk (X) and one for market risk (Y). To obtain the specific risk the net (an offset of the long and short of the spot and forward position) for each stock is computed. The net exposure of each share position is multiplied by a risk sensitivity factor, which is 8% for specific and market risk, but if the national regulator judges the portfolio to be liquid and well diversified, the systematic risk factor is reduced to 4%. In the example below, it is assumed to be 4%.

Foreign exchange and gold risk

Recall that all derivatives have been converted into the equivalent spot positions. A bank’s net open position in each individual currency is obtained – all assets less liabilities, including accrued interest. The net positions are converted into US$ at the spot exchange rate. The capital charge of 8% applies to the larger of the sum (in absolute value terms) of the long or short position, plus the net gold position.

Alternatively, subject to approval by national regulators, banks can employ a simulation method. The exchange rate movements over a past period are used to revalue the bank’s present foreign exchange positions. The revaluations are, in turn, used to calculate simulated profits/losses if the positions had been fixed for a given period, and based on this, a capital charge imposed.

Interest rate risk

The capital charge applies to all debt securities, interest rate derivatives (e.g. futures, forwards, forward rate agreements, swaps) and hybrid instruments. The maturity approach involves three steps.

  1. Obtain a net overall weighted position for each of 16 time bands. Before they are summed, the net position in each time band is multiplied by a risk factor, which varies from 0 at the short end to 12.5 at the long end.
  2. 10% of each net position in each time band is disallowed to take account of the imperfect duration mismatches within each time band – known as vertical disallowance.
  3. There is another problem: the interest rates in the different time buckets may move together, which is resolved through several horizontal disallowances, which vary from between 30% and 100% (i.e. no disallowance) in recognition that the degrees of correlation will vary. The matched long and short positions between the time buckets can be offset, but:
    • a 40% disallowance applies in the first set (0–1 year);
    • a 30% disallowance applies to the other two sets of time bands, i.e. 1–4 years and over 4 years;
    • there is a 40% disallowance for adjacent time buckets, and a 100% disallowance between zones 1 and 3.

Duration is an alternative approach banks can employ to determine the capital to be set aside for interest rate risk. In each time band, the sensitivity of each position is computed by employing the duration for each instrument. The horizontal disallowance is 10% but the charge related to vertical disallowance is lower because duration allows sensitivity to be measured more accurately.

Commodities risk

This risk is associated with movements in prices of key commodities such as oil, natural gas, agricultural products (e.g. wheat, soya) and metals (e.g. silver, copper, bronze) and related risks such as basis risk, or changes in interest rates which affect the financing of a commodity. The capital charges are obtained with a methodology similar to that used for the other three categories, but it will not be discussed here.

This outline24 of the standardised approach has been kept brief for several reasons. First, depending on their activities, banks using this approach are more likely to incur substantially higher capital charges than if they opt for the internal model approach, because offsetting correlations between the four risk categories are ignored. Banks with a large trading book would be the hardest hit. Second, a bank will incur substantial costs because it still has to change its systems to comply with the standardised model. Taken together, these points provide a strong incentive for most banks to invest in a risk management system which ensures their internal model is approved.

All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd Protection Status

Modern Banking Topics