THE BASEL CAPITAL ACCORDS BY JOE LARSON APRIL 2011

I. Introduction Banks are a vital part of a nation’s economy. In their traditional role as financial intermediaries, banks serve to meet the demand of those who need funding. As such, banks make it possible for people to buy homes and for businesses to expand. Banks therefore facilitate spending and investment, which fuel growth in the economy. However, despite their important role in the economy, banks are nevertheless susceptible to failure. Banks, like any other business, can go bankrupt. However, unlike most other businesses, the failure of banks, especially very large ones, can have far-reaching implications. As we saw during the Great Depression and, most recently, during the global financial crisis and the ensuing recession, the health of the bank system (or lack thereof) can trigger economic calamities affecting millions of people. Consequently, it is imperative that banks operate in a safe and sound manner to avoid failure. One way to ensure this is for governments to provide diligent regulation of banks. Yet, with the advent of globalization, banking activities are no longer confined to the borders of any individual country. With cross-border banking activities rapidly increasing, the need for international cooperation in bank regulation has likewise increased. Ready to meet this need is the Basel Committee on Bank Supervision (BCBS). In its role as the international advisory authority on bank regulation, the BCBS has promulgated guidance on issues critical to ensuring health in the banking systems across the world. One such issue, and one that played an important role in the recent global financial crisis, is the regulation of bank capital. Addressing this issue has been an ongoing process for the BCBS over the past twenty
1

years, and has resulted in the promulgation of capital adequacy standards that national regulators can implement. These standards are known collectively as the Basel Accords, named after the city in Switzerland where the BCBS resides. The Basel Accords have caused disagreement at times, but they are nevertheless important to the formulation of regulatory policy relating to bank capital. In all, the BCBS has produced three such accords. Basel III, published in 2010, is the most recent Accord. Each Accord has purported to improve upon the previous one, but early indications suggest that Basel III is not flawless and so it will likely not be the last Accord. This FAQ will attempt to provide an overview of the Basel Accords and the process that has led to Basel III. In doing so, it will limit its focus to those aspects of the Accords that directly address the determination of a bank’s capital adequacy vis-à-vis the level of the credit risk (i.e. the risk of asset loss) in the bank. Part II of the FAQ will describe regulatory capital, which forms the basis of the Accords, and why it is important to the safety and soundness of the banks. Part III will provide background information to better understand the origin of the Basel Accords, including a brief description of the Basel Committee on Bank Supervision, the entity that promulgated the Accords, as well as the developments that led to its decision to address international capital regulation. Part IV will then describe Basel I and its shortcomings. Part V will first describe how Basel II addressed the faults in Basel I. Part V will also discuss the faults in Basel II itself, including those exposed by the global financial crisis. Part VI will discuss the features of Basel III and some early critiques of it. Part VII will provide some concluding remarks.

2

II. Importance of Regulating Bank Capital At the heart of all the Basel Accords is the issue of regulating bank capital or, more accurately, developing rules to ensure that banks maintain sufficient levels of capital. Thus, before one can fully understand the capital accords, one must first understand the concept of capital. As with any business, a bank has a balance sheet that is comprised of assets, liabilities, and equity. Banks fund their assets through a combination of their liabilities and equity. A bank’s liabilities represent that bank’s debt and traditionally consist of deposits of money from people who entrust the bank to hold onto their money and return it when asked to do so. On the other side of the balance sheet are a bank’s assets that, for the most part, consist of loans to its customers, from which the bank derives income in the form of interest charged to the borrowers. Obviously, there are more items that can be included in a bank’s liabilities and assets, but deposits and loans are the most common examples. If the total value of a bank’s assets exceeds that bank’s liabilities, the amount of that difference is referred to generally as the bank’s capital. Therefore, a bank’s capital represents the amount of assets not funded by debt. It follows then that if the bank’s liabilities exceed a bank’s assets, a bank is negatively capitalized. When this situation occurs, a bank owes more to its debtors than it can provide from its assets and is thus insolvent. As one can see, it is therefore important that a bank maintain a positive capital position to ensure the bank is able to repay its liabilities. Besides preventing insolvency, capital is especially important to a bank for a couple of other reasons. First, it protects against the risk of loss that is inherent in banks’ assets. Traditional banks are in the business of loaning money to people with the expectation that those people will

3

repay that amount, plus interest. However, when banks make loans, there is always the risk that the bank may not be repaid. Even the most creditworthy borrower can have an unlucky experience that prevents repayment of a loan. If such an event occurs, and the borrower defaults, the bank has lost money that it owes to its own creditors, the depositors at that bank, and must therefore rely on its capital to pay those depositors. The second reason why capital is important is to protect against the volatility in a bank’s liabilities. Banks fund much of their assets through customer deposits. However, such deposits are a risky source of funding because depositors can generally demand that the bank repay them at any time. If depositors withdraw large amounts of money within a short timeframe, the bank could find itself paying such withdrawals with its capital because most of a bank’s assets, such as loans, cannot be liquidated quickly. If the withdrawals are large enough, the bank may exhaust its capital and find itself insolvent. In the recent global financial crisis, the risk of insolvency was an important issue confronting several important banks in the United States. Because of the importance of these banks, their failure would have posed systemic risk to the financial system and were thus considered “too big to fail.” As a result, the government had to intervene and inject capital into these banks in order to prevent the collapse of the entire financial system. The role of capital in preserving a safe and sound banking system cannot be overstated. By maintaining sufficient amounts of capital, banks are able to ensure that they are capable of meeting their obligations to their creditors. Likewise, having sufficient amounts of capital will instill depositors and other bank creditors with confidence that the bank will be able to repay them, even if some of the bank’s assets default.

4

III. Basel Committee on Bank Supervision (BCBS) Up until the 1970s, bank regulation lacked, for the most part, international reach. Nations were left to decide for themselves how to best regulate the banks that did business within their borders. This all changed in 1974 when the Herstatt Bank in Germany failed. At the time Herstatt Bank failed, there were several unsettled international transactions between Herstatt Bank and American banks, where the American banks had already paid Herstatt Bank deutschmarks, but the American banks had not received the dollars owed to them in return. Before the transactions could settle and the American banks could receive the dollars owed to them, Hertstatt Bank failed, causing significant losses for its American counterparties. The Herstatt incident highlighted the significant risks that accompany international banking, and exposed the need for coherent international cooperation between nations to minimize future risks associated with international banking. In response to the Herstatt incident, the member nations of the G-10 (a group of countries with the ten largest economies in the world) established the Basel Committee on Bank Supervision (BCBS) in 1974. Initially comprised of the head of the central banks (or equivalent) of each member of the G-10, the membership of the Committee subsequently expanded to include representatives of 27 countries. The BCBS works under the auspices of the Bank for International Settlements (BIS). The BIS is an intergovernmental organization owned and operated by the central banks of numerous countries across the world and is responsible for fostering international cooperation on monetary and financial policy. The BIS also serves as a central bank to the central banks of its members. As its name suggests, the Basel Committee on Bank Supervision focuses its work on matters relating to bank supervision and regulation. The BCBS serves as a forum for its members to discuss issues and problems relating to bank regulation. In the course of these discussions, if

5

the members of the BCBS come to a common understanding or agreement on an issue, the BCBS will issue supervisory guidance and standards relating to that issue. However, the issuances of the BCBS are only advisory in nature and are not binding on the members of the BCBS or any other nation. As a matter of best practice, though, the members of the BCBS, as well as many non-members, usually adopt the recommendations of the BCBS in whole or in part. Thus, the promulgations of the BCBS form a type of “soft law.” As the work of BCBS proceeded through the 1970s and 1980s, it was becoming increasingly apparent to the Committee that one of the issues it needed to address was that of capital regulation. As the preceding section explained, the level of capital maintained by a bank is an important aspect of that bank’s safety and soundness. However, for a variety of reasons, bank regulators across the world had diverse standards relating to capital regulation, with some nations lacking enforceable capital standards. With international banking and business growing at an increasing rate, the members of the BCBS wanted to confront the issue of capital regulation at an international level to ensure that those who banked internationally would be protected by the fact that banks across the world were subject to some common degree of capital requirements. Indeed, by the mid 1980s, some of the largest and most internationally active banks took advantage of the lax capital regulation by holding extremely low levels of capital. However, not every nation at this time maintained a lax regime for regulating capital. The U.S., for example, implemented relatively strict capital requirements in the early 1980s. By doing so, though, nations like the United States were putting their banks at a competitive disadvantage vis-à-vis the banks in other countries that maintained more lax regulatory regimes.

6

A lax regulatory capital regime creates a competitive advantage for banks because lower capital requirements allow banks to lend more, which results in more interest income and, thus, higher profits. For example, compare two banks governed by two different capital standards, where Bank A is required to hold an amount of capital equal to 2% of its assets, and Bank B is required to hold an amount of capital equal to 8% of its assets. If Bank A has $100 worth of capital, Bank A can lend up to $5,000. For Bank B however, that same $100 worth of capital only allows it to lend up to $1,250. As one can see, the higher capital requirement puts Bank B at a competitive disadvantage. With the foregoing considerations in mind, the BCBS set out to promulgate standards that would harmonize the regulation of bank capital across the world, with the intent of 1) creating a safer banking environment, and 2) leveling the competitive positions of banks in different countries. The culmination of the BCBS’ efforts in this regard resulted in the promulgation of an accord entitled “International Convergence of Capital Measurement and Capital Standards,” which has since become known as the Basel I Capital Accord, or simply Basel I. IV. Basel I Basel I was finalized and approved by the BCBS in 1988. Again, because the BCBS has no binding legal authority, countries had the option to adopt Basel I’s standards. Many BCBS member countries, as well as many non-member countries, did adopt Basel I however, and thereby incorporated its features in their own domestic regulatory law. Since the BCBS’ work focused on banking at the international level, it intended that Basel I would be applied only to internationally active banks. However, many countries ultimately applied Basel I’s requirements to all banks.

7

A. Features of Basel I To determine capital adequacy, Basel I adopted the use of a capital ratio. This ratio would measure capital (the numerator) against the bank’s assets (the denominator). As will be discussed below, the value of bank assets included in this ratio was not the face value of each of the assets, but rather their value adjusted for their risk level, otherwise known as the bank’s risk-weighted assets (RWA). The capital ratio is thus expressed as capital/risk-weighted assets. To be considered sufficiently capitalized under Basel I, a bank had to maintain a capital ratio of 8%. That is, the value of the bank’s capital had to equal at least 8% of the value of the bank’s riskweighted assets. In formulating the capital adequacy ratio, the BCBS had to first address the issue of what would constitute “capital” for regulatory purposes, i.e., what would be included in the numerator. Since each member of the Committee had a different approach to defining capital, the issue was a contentious one. As mentioned previously, in the most general sense, capital is defined as the excess of a bank’s assets over its liabilities. However, there are many items that could be included in this category, with some more reliable than others in achieving the purpose of cushioning losses in the bank’s assets. Ultimately, the BCBS decided that, under Basel I, the definition of bank capital would be broken down into two components: tier 1 capital and tier 2 capital. Tier 1 capital consists of higher-quality forms of capital, in the sense that it is comprised of items that have lower priority of repayment in the event that a bank becomes insolvent, and therefore have the greatest ability to absorb asset losses. Tier 1 capital consists primarily of “core capital.” Core capital, also known as common equity, represents items arising from pure ownership in the bank and includes the paid-in value of common stock, as well as the amount of any reserves (i.e., retained earnings)

8

that the bank has accumulated and disclosed. Because these items arise from ownership in the bank, they have the lowest priority of repayment in the event of insolvency, and therefore represent the highest quality capital. As mentioned above, for a bank to be considered sufficiently capitalized under Basel I, it had to maintain a capital ratio of 8%. However, Basel I also required that half of this 8% consist of tier 1 capital (i.e., tier 1 must equal at least 4% of the bank’s risk-weighted assets). Tier 2 capital is considered less reliable and is comprised of items such as subordinated debt and reserves held for loan-losses. Subordinated debt is debt (e.g., bonds) issued by the bank that the bank need not pay back until it has paid all its other creditors. Thus, the debt that the bank owes to these creditors is “subordinate” to the debt it owes to other creditors. Therefore, a bank can use the proceeds it obtained through the issuance of subordinated debt to pay its other liabilities, including the deposits it owes to its customers. As one can see, tier 2 capital is clearly of lower quality than tier 1. Whereas tier 1 consists primarily of unencumbered equity in the bank, tier 2 is permitted to include debt held by the bank. The fact that these lower quality items were allowed to be included in the definition of capital at all reflects the fact that there were banks in the countries of some of the BCBS members that were not sufficiently capitalized with owner’s equity, but instead had to rely, at least partially, on debt. Recognizing the lower quality of tier 2 capital, Basel I limited the amount of tier 2 capital that could be included in the bank’s capital to 100% of tier 1 capital. As noted above, the purpose of bank capital is to provide a cushion against losses in the bank’s assets. Therefore, in the process of drafting Basel I, the BCBS took a risk-based approach to developing the capital adequacy guidelines. The BCBS wanted to incorporate the idea that the required level of capital should be proportionate to not only the quantity of assets held by a bank,

9

but also to the risk of loss inherent in those assets. In other words, riskier assets (i.e., those that have a higher chance of default, or loss) should be offset by a higher amount capital. This idea of risk-sensitivity was incorporated in the denominator of Basel I’s capital adequacy ratio. Instead of using a capital ratio that compared a bank’s capital to the total face value of its assets, Basel I used a ratio that would compare a bank’s capital to the value of the bank’s assets after they had been adjusted for their risk of loss or default. The resulting total was called the bank’s risk-weighted assets (RWA). To do this, Basel I established four risk categories or “buckets” (0%, 20%, 50%, and 100%) into which each of the bank’s assets would be placed. Which category an asset fell into determined how much of that asset’s value would be included in the bank’s RWA. Riskier assets were placed in higher-percentage brackets, which meant that more of that asset’s value was included in a bank’s RWA total, which, in turn, meant that a bank’s capital requirement would increase. An asset’s assignment into a risk class was predetermined by the Basel I guidelines. The BCBS established these guidelines based on the perceived risk associated with the counterparty involved in the transaction underlying the asset (e.g., the borrower of a loan). For example, holding cash poses no risk of loss to the bank. Therefore, the Basel I guidelines placed all cash in the 0% risk category, which meant that the bank would not have to include the value of its cash in its total risk-weighted assets (i.e., the denominator of the capital adequacy ratio). Likewise, according to Basel I, a loan made to a government that is a member of the Organization for Economic Cooperation and Development (OECD), such as the United States, was perceived as low-risk, and therefore would be placed in the 0% risk category. Conversely, Basel I considered assets such as commercial loans (i.e., loans made to businesses) to be high-risk and, therefore, included them in the 100% risk category. This means that 100% of the value of all commercial

10

loans would be included in the denominator of the capital adequacy ratio. Below is a chart that provides examples of the types of assets that were placed in each risk category: Risk-Weight Category 0% 20% Types of Assets Included in the Risk Category Cash; assets involving the governments of OECD countries Assets involving banks located in OECD countries; cash items in the process of collection Loans secured by mortgages on residential property Assets involving businesses; personal consumer loans; assets involving non-OECD governments (unless the transaction is denominated and funded in the same currency)

50% 100%

Basel I’s methodology to determine capital adequacy also incorporated a process to take into account the risk posed by a bank’s off-balance sheet items. As the term suggests, off-balance sheet items are those items held by the bank, but that do not appear on that bank’s balance sheet. The determination of whether an item belongs on- or off-balance sheet can sometimes be complex. However, in general, an off-balance sheet item, whether it be an asset or liability, is one that that the bank’s claim to has not materialized completely. For example, when a bank extends a home equity line of credit to a customer, any unused portion of that line of credit is considered an off-balance asset to the bank. The reason for this is that, although a line of credit is a type of loan, and therefore is like an asset, the bank cannot derive benefit from any unused portion of it because there is no balance from which the bank can earn interest. Thus, off-balance sheet assets can be thought of as contingent on-balance sheet assets that remain off-balance sheet until an event occurs (e.g., a borrower draws on a line of credit) that gives the bank the right to the benefit of that asset.

11

When an off-balance sheet asset becomes an on-balance sheet asset, it carries with it a risk of loss just like any other asset. Recognizing that off-balance sheet assets have this potential, Basel I devised a method to incorporate this risk into a bank’s capital adequacy ratio. To do so, Basel I created a two-step process. The first step involved applying a “conversion” factor to the value of the off-balance sheet asset. The application of this factor essentially converted the value of the off-balance sheet asset to take into account the probability that the off-balance sheet asset would become an on-balance sheet asset. Higher conversion factors were applied to off-balance sheet items with a higher likelihood of becoming on-balance sheet items. The application of off-balance sheet conversion factors can be illustrated by comparing a commercial letter of credit and a standby letter of credit. A letter of credit is essentially a promise by a bank to ensure payment by one its customers in a transaction with a third party. With a commercial letter of credit, if a customer, e.g., a buyer of goods, engages in a transaction with a third party, e.g., a seller of goods, a bank initially pays the seller and is repaid by its customer, once the buyer receives the goods. A standby letter of credit, on the other hand, acts as a credit enhancement and backs a financial obligation of a bank’s customer to a third party. If the customer defaults, the bank must pay the third party and is unlikely to be repaid by its customer given the fact that the customer defaulted. The commercial letter of credit supports the payment of goods in a specific transaction where neither party wants to take the risk that the other will not perform their end of the trade. A standby letter of credit supports the longer-term risk that a customer will not fulfill its end of the bargain, like repayment of a loan over time. Consequently, Basel I applied a conversion factor of 100% to the standby letter of credit because the risk is known and confined to a particular transaction. Basel I applied a 20% conversion factor to the commercial letters of credit issued by banks because the banks assume the greater risk of non-

12

performance over a longer period of time. In other words, the more likely an item will be called and the greater the risk that the bank’s customer would be unable to repay the bank, the higher the conversion factor. Once the conversion factor was applied to an off-balance sheet asset, the discounted value of the off-balance sheet asset was treated like any other on-balance sheet asset and placed in the appropriate risk category. This step resulted in the risk-adjusted value of the off-balance sheet item, which was then included in the total value of the bank’s risk-weighted assets. Once all of the bank’s on- and off-balance sheet assets were adjusted for risk, the values were added up. The resulting sum of this calculation equaled the bank’s risk-weighted assets, which represented the denominator in the bank’s capital adequacy ratio. As mentioned above, the bank must then ensure that value of its total capital levels (tier 1 + tier 2) are equal to at least 8% of the bank’s risk-weighted assets, with tier 1 capital equaling at least 4% of risk-weighted assets. B. Criticisms of Basel I As Basel I was the first coherent international attempt at regulating bank capital, it may come as no surprise that Basel I was the target of many criticisms. One of the main criticisms of Basel I pertains to its risk-weighting system. In particular, critics saw Basel I’s bucket approach to risk-weighting assets as arbitrary, overly broad, and not nearly sensitive enough to the unique risks associated with each asset held by a bank. Within each bucket, there are assets with very different levels of risk, but because they all share a common type of counterparty (e.g., businesses, governments, etc.), they are assumed to possess the same type of risk. The flaw in Basel I’s approach to risk-weighting assets can be seen when one applies it to the example of commercial loans. Under Basel I, all types of commercial loans are 100% risk-

13

weighted, thus requiring the bank to include the entire value of the loan in the total of its riskweighted assets. Yet, clearly not all commercial loan recipients have the same amount of risk. A loan to a well-established company is far less risky than a loan to a start-up company. The effect of this shortcoming in Basel I’s risk-weighting methodology is that banks have an incentive to engage in what is known as regulatory arbitrage. Essentially, regulatory arbitrage describes a situation where, if a bank is presented with two options, both of which receive the same regulatory treatment, but each of which result in differing profit-making opportunities, the bank will choose the more lucrative option. In the commercial loan example above, from a regulatory perspective, it doesn’t matter whether the bank makes the loan to the start-up company or the well-established company; in either case the bank will have to include 100% of the loan in its risk-weighted assets. However, from a profit-making perspective, the loan to the start-up company will be riskier, and therefore will demand a higher interest rate. Consequently, the bank will have an incentive to make the loan to the start-up company. The same principle holds true for potential borrowers within other risk categories, where no two borrowers will have the same risk profile, and yet all will be treated the same from a capital adequacy perspective. Given this situation, the bank will usually pursue the opportunity with higher earning potential. However, as seen with the example of the start-up company, pursuing greater profit usually means that the bank is taking on higher risk. This again leads to a situation where the level of capital required under the Basel I methodology is not sufficiently commensurate with the risk in the bank’s assets. In sum then, the common theme running through the criticisms of Basel I was that Basel I’s method to determine the proper amount of capital to be held by a bank was not sufficiently or accurately connected to the actual risks confronting the bank. As the criticisms of Basel I

14

mounted, the members of the BCBS decided that reform was in order. After several years of negotiations and consultations, the BCBS released a set of revisions to Basel I, entitled “International Convergence of Capital Measurement and Capital Standards: A Revised Framework,” also known as Basel II. V. Basel II The BCBS organized Basel II around what it called the “Three Pillar” approach. For purposes of this FAQ, however, attention will primarily be given to Pillar I, which is the Pillar that most directly addresses the issue of calculating capital adequacy, and is also the Pillar that specifically attempts to correct the deficiencies identified in Basel I. Pillars II and III, which deal with supervisory review standards and market discipline issues, respectively, while important aspects of capital regulation, do not have a direct bearing on the calculation of bank capital adequacy, and therefore are outside the scope of this FAQ. A. Features of Basel II and How they Addressed Basel I Faults Before exploring Pillar I in depth, it is worthwhile to note what portions of Basel I that Basel II did not change. Basel II still requires that a bank’s total capital equal at least 8% of the bank’s risk-weighted assets. It also still assesses banks’ capital adequacy using the same capital adequacy ratio, which is equal to a bank’s capital divided by the bank’s risk-weighted assets. Basel II did not alter Basel I’s definition of capital, i.e., the numerator of the ratio. However, as will be seen below, Basel II, or more specifically, Pillar I, does alter how a bank arrives at the denominator, i.e. the calculation for risk-weighted assets. As mentioned above, Pillar I addresses the way in which banks calculate their riskweighted assets. Pillar I was specifically designed to redress the deficiencies identified in Basel I. As such, Pillar I focuses primarily on reforming the method of measuring credit risk, i.e., the risk

15

inherent in the bank’s assets. The goal of these reforms is to ensure that the calculation of risk in a bank’s assets more accurately reflects the actual risk in those assets, which should reduce the opportunity for banks to engage in regulatory arbitrage, which was one of the major problems with Basel I. Pillar I’s approach to measuring credit risk actually consists of three approaches: the Standardized Approach, the Foundation Internal Ratings-Based Approach (FIRB), and the Advanced Internal Ratings-Based Approach (AIRB). Thus, Basel II offers a menu of options to determine the credit risk in banks. The standardized approach is the least complex of the three approaches and most similar to Basel I’s approach. The standardized approach retains the use of risk buckets to determine an asset’s risk-adjusted value. However, Basel II’s standardized approach to risk-weighting is different from Basel I’s approach in a couple of ways. First, Basel II’s standardized approach expands the number of risk buckets from four to six. In addition to the 0%, 20%, 50%, and 100% risk categories used under Basel I, the standardized approach also includes a 150% risk category, as well as a 35% risk bucket specifically reserved for residential loans secured by mortgages. The next difference between Basel I and the standardized approach is the process to determine in which bucket an asset is placed. As mentioned above, under Basel I, assets were placed in risk buckets based on the generic identity of the counterparty involved (e.g., an OECD country, business, etc.). Recognizing that no two assets bear identical risk profiles, Basel II’s standardized approach attempts to make the risk-weighting determination based on the unique risk associated with each of the bank’s assets. To achieve this, Basel II’s standardized approach utilizes credit-rating agencies, such as Standard & Poor’s and Moody’s. Accordingly, under the

16

standardized approach, assets are placed into risk buckets based on the credit rating assigned to the counterparty involved in that asset, with higher rated counterparties assigned to lower risk buckets. For example, under the standardized approach’s guidelines, if a commercial borrower receives a AAA rating from Standard & Poor’s, that loan would be placed in the 20% risk bucket. Contrast this result with Basel I, where all such commercial loans, regardless of the credit-worthiness of the borrower, were placed in the 100% risk bucket. In the event that a borrower is not rated by a credit agency, Basel II’s standardized approach automatically places that loan in the 100% risk bucket. The standardized approach makes an exception to the credit rating method for residential loans, which automatically receive a risk weight of 35%. Under the standardized approach, the risk-weighting an asset receives depends not only that asset’s credit rating, but also on whether that asset represents a claim on a sovereign national government. Reflecting the belief that government assets pose less risk, the standardized approach places government assets with a given credit rating in a lower risk category than if that asset is a claim on a private party, even if the credit rating is the same. As an example, an asset involving a AAA-rated government (using Standard & Poor’s methodology) would be riskweighted at 0%, whereas a loan involving a AAA-rated business would be risk-weighted at 20%. Below is a table that provides examples of the risk-weightings received by assets as determined by their credit ratings (using Standard & Poor’s ratings scale) and whether they represent claims on governments or private counterparties. The table excludes the 35% risk category because that category is reserved exclusively for loans secured by mortgages on residences and such assets are automatically included in that category regardless of the borrower’s credit rating.

17

Credit Rating (S&P) AAA to AAA+ to ABBB+ to BBBelow BBUnrated

Government Risk-Weighting 0% 20% 50% 150% 100%

Private Counterparty Risk-weighting 20% 50% 100% 150% 100%

The standardized approach is the simplest method in Pillar I to calculate credit risk, and is therefore more suitable for smaller banks. The two remaining approaches, the Foundation Internal Ratings-Based (FIRB) approach and the Advanced Internal Ratings-Based (AIRB) approach, are suited more for larger and more sophisticated banks. The main difference between the standardized approach and the two IRB approaches is that with the latter two approaches, banks can use their own internal methodology to determine the risk level of their assets, whereas with the standardized approach, banks must rely on the external ratings guidelines to risk-weight their assets. Because of the sophisticated nature of the IRB approaches, if banks wish to use them, they must demonstrate their technical ability to implement them and also receive approval from their regulators. To understand the methodology underlying the IRB approaches, one must first understand the concept of unexpected losses. In essence, within the IRB framework, unexpected losses theoretically approximate a bank’s credit risk, and therefore determine how much capital a bank must maintain. Unexpected losses estimate losses in a bank’s assets that are not foreseeable. Losses in a bank’s assets are a normal part of the banking business, and banks, for the most part, can anticipate and prepare for those expected future losses by looking at historical

18

loss rates. However, there are instances where a bank incurs losses greater than usual. These above-average loss levels are a bank’s unexpected losses. Since expected losses are, by definition, expected, banks often set aside reserves (called loan-loss reserves) to absorb those losses. Therefore, it is those unexpected losses that a bank’s capital levels are meant to cushion. To arrive at a bank’s estimate of unexpected losses, four inputs are used, all of which are common to both the FIRB and AIRB approaches. The first of these inputs is the probability of default. As its name implies, this factor provides an estimate of the probability, over the course of a year, that a given borrower will default on his or her loan. The next input is the loss given default (LGD). This input provides an estimate of amount the bank stands to lose if a given borrower defaults. This estimate can be thought of as the bank’s net loss, since banks are usually able to recover some amount from the borrower. The third input is the exposure at default (EAD). This input represents the additional amount that a bank could lose at the time of a borrower’s default. An example would be the unused portion of any credit line available to a defaulting borrower, where the borrower still has the ability to draw on the line, thereby creating additional assets for the bank that can also go into default. The final input is the maturity of the asset (i.e., the duration of the loan). The longer the duration of a loan, the greater the chance that something will go wrong with the borrower that causes default. Therefore, an asset with a longer maturity will, holding other factors constant, lead to a higher risk weighting for that asset. Once each of these inputs is determined for each of the bank’s assets, they are injected into complex mathematical models (the details of which are beyond the scope of this FAQ)

19

which, in turn, arrive at an estimate of the bank’s unexpected losses (i.e., its credit risk). Banks must ensure that they have capital equal to at least 8% of this amount. The primary difference between the FIRB approach and the AIRB approach is who determines the values associated with each of the inputs. Under the FIRB approach, a bank is allowed to calculate the PD for each asset, while the bank’s regulator will determine the LGD and EAD. With regard to the M input, the bank’s regulator has the discretion to assign an estimated maturity for each asset or allow the bank to use its own calculation. Under the AIRB approach, however, a bank is allowed to calculate the values for all four (PD, LGD, EAD, and M) of the inputs. If a bank is allowed to use either IRB approach, its methodology and outputs must be reviewed and verified by its regulators. As one can see, with either IRB approach, Basel II gives banks at least some degree of autonomy to devise the method to estimate their level of credit risk. Regardless of how a bank constructs its model to determine the credit risk associated with each asset in its portfolio, the output of its calculation will be used to determine the extent to which the value of that asset will be included in the bank’s risk-weighted assets. B. Critique of Basel II Just as with Basel I, Basel II experienced its own share of criticisms. First, relating to the standardized approach, many questioned the use of rating agencies to determine an asset’s risk. Since the rating agencies are paid by those they are supposed to rate, concerns arose regarding the reliability and objectivity of the ratings they provided. Maybe the clearest example of the flaws in the credit agency rating model was its failure to consider and protect against the trend of securitization that occurred before and during the global financial crisis.

20

Under Basel II’s standardized approach, all exposures (e.g., investment in securitized products) to securitization held by a bank would be assigned a risk weight. However, the riskweighting assigned to a securitized position depends on the external credit rating assigned to it. Thus, credit rating agencies play an important role in determining the amount of capital held by the banks in relation to the risks they face due to securitization. Unfortunately, as the subprime credit crisis illustrated, many credit agencies gave securitized products inaccurately high ratings, the reason for which was two-fold. First, many rating agencies relied on faulty rating methodology to assess risks associated with securitization. Second, as mentioned above, because the rating agencies were being paid by the rated parties, conflicts of interest arose which impaired the ability of rating agencies to provide objective ratings. Consequently, because the perceived risk of securitized exposures was low, as indicated by the high credit ratings, banks were required to hold less capital, which left them undercapitalized relative to the real risk level inherent in their exposures to securitized products. As a result, when the underlying assets (particularly the subprime assets) of the securitized products defaulted, banks had too little capital to absorb the losses they incurred as a result of their exposure to those securitized assets. Another concern relating to the standardized approach is the lack of a uniform rating system. Basel II does not specify which rating agency a bank must use, so banks can employ various rating agencies (e.g. S&P, Moody’s, etc.), each of which employs their own unique methodology to assign ratings, which means the consistency of credit risk assessments across banks could suffer. Critics also point out that the standardized approach does an inadequate job of differentiating risk among unrated borrowers, where it simply assigns such borrowers a 100% risk rating. Thus, in this situation, the standardized approach is a victim of the same criticisms

21

that were levied against Basel I, where it includes borrowers of varying degrees of risk in the same risk category. With regard to the IRB approaches, there appears to be significant potential to assess the real amount of risk in a bank’s assets. Using asset-specific inputs, the IRB approaches are supposed to assign risk weightings that are unique to each of the bank’s assets. As a result, instead of four risk-weightings, as with Basel I, or six risk-weightings, as with the standardized approach, the IRB models are capable of producing an indefinite number of risk weightings from 0% on up. Thus, instead of simply having broad risk-insensitive risk buckets of 0%, 20%, 50%, and 100%, the IRB models have the potential to assign risk weightings that are precisely unique to the individual assets. In addition, by allowing the banks to conduct the risk assessment, the IRB approaches are giving the people who are most familiar with the borrower the ability to determine the risk associated with that borrower, which should result in a more accurate risk rating. Therefore, in theory, the IRB approaches seem to be a significant improvement over Basel I and the standardized approach and their bucket approaches to risk-weighting. However, while the IRB approaches seem good in theory, in practice they too are not without their faults. Among other things, critics point to the fact that the assessments are produced internally. Specifically, the internal evaluation component of the IRB approaches raises concerns about consistency, since no two banks will have the same methodology to assess risk. Two banks with identical inputs, but with two different models, will produce two different risk weightings for the same asset. Thus, the regulators of banks using the IRB approaches assume greater responsibility in ensuring that the internal risk assessments are accurate. This is problematic, however, given the potential complexity of banks’ internal methodologies. Due to such complexity, regulators of such banks may find themselves deferring to the expertise of the

22

banks that devised the methodologies. If this is the case, the IRB approaches amount to nothing more than the banks regulating themselves. Another shortcoming of the IRB approaches is their tendency to promote procyclical bank behavior. In short, Basel II’s regulatory structure has the effect of inducing banks to maintain lower amounts of capital during good economic times, which forces them to take economically harmful action in bad times to maintain sufficient capital. During good economic times, borrowers are generally in a better position to repay their debt obligations. The IRB models used by banks would reflect this fact, which would result in lower estimates of risk in the bank’s assets. Accordingly, the banks’ IRB models would require the banks to hold less capital. However, during times of economic hardship, such as the crisis of 2008-2009, when credit risk was at its greatest, the IRB approaches would reflect the higher risk levels, which would drive up the bank’s RWA total, thereby raising the bank’s capital requirement. Consequently, banks would have to take action to compensate for the periods when it held low levels of capital. To do so, banks would be forced to improve their capital ratio by lending less. This phenomenon, called a credit crunch, exacerbates the severity of the economic crisis, because when banks lend less, less money is injected into the economy, which inhibits economic growth, and thus delays recovery. During the recent global financial crisis, this exact phenomenon occurred, where banks holding low-quality assets, such as sub-prime loans, were forced to reduce lending to prevent their capital ratio from becoming too small. The faults in Basel II were beginning to become apparent to the members of BCBS well before the start of the global financial crisis erupted. When the crisis began, talks on how to improve Basel II were already under way. However, the severity of the crisis made it clear that

23

Basel’s faults needed to be addressed sooner rather than later. In the fall of 2010, after much debate and negotiation, the BCBS released the latest installment of the Basel accords. VI. Basel III The BCBS promulgated Basel III in September of 2010. Formally titled, “A Global Regulatory Framework for More Resilient Banks and Banking Systems,” Basel III reflects the BCBS’ attempts to apply lessons learned from the financial crisis and apply them to the existing framework of banking regulation. Thus, Basel III does not replace Basel II, but rather augments it. The primary goal of Basel III is to improve the ability of banks to absorb asset losses without affecting the rest of the economy. In terms of capital regulation, as will be seen below, Basel III focuses mainly on the quantity and quality of capital held by banks. A. Features of Basel III Among the most important parts of Basel III is its new definition of regulatory capital, which is more restrictive and emphasizes greater quality. Basel III retains the tier 1 and tier 2 distinction, but limits their composition to higher-quality capital that is better able to absorb losses. Under Basel III, Tier 1 capital must be mostly of “core capital,’ which consists of equity stock and retained earnings. In addition, many items that were formerly included in a bank’s capital calculation under Basel II, including some forms of subordinated debt, will be excluded under Basel III. Those capital instruments that will no longer qualify as “capital” under Basel III will be phased out of a bank’s capital calculation over a ten-year period starting in 2013. This transition period will help those banks that do not currently possess the sufficient amount and types of capital comply with the new requirements. In addition to increasing the quality of capital, Basel III increases the quantity of capital that banks must hold. By the time participating countries fully implement Basel III in 2019,

24

banks are expected to maintain a total capital ratio of 10.5%, an increase from the 8% requirement under Basel II. As with Basel I and Basel II, banks under Basel III must maintain a minimum total capital ratio of at least 8% of risk-weighted assets. However, under Basel III, after a bank has calculated its 8% capital requirement, it will have to hold an additional capital conservation buffer equal to at least 2.5% of its risk-weighted assets, which brings the total capital requirement to 10.5% of risk-weighted assets. The purpose of the capital conservation buffer is to ensure that banks have sufficient capital levels to absorb asset losses, especially during periods of financial and economic stress. To improve the quality of capital held by banks, Basel III also increases the amount of tier 1 capital that banks are required to hold. As mentioned above, tier 1 capital includes higher quality capital in the sense that it is comprised of items representing ownership in the bank and unencumbered sources of funds. Under Basel III, banks will be required to maintain an amount of tier 1 capital equal to at least 6% of risk-weighted assets, a 2% increase over the current requirement of 4%. In addition, banks will also have to hold more core capital. As mentioned above, core capital is a subset of tier 1 capital that includes common equity, and thus represents the highest quality capital. Under Basel III, banks will have to hold an amount of core capital equal to at least 4.5% of risk-weighted assets, whereas in the previous Basel Accords, core capital had to represent only 2% of risk-weighted assets. The total amount of core capital that banks are required to hold increases to 7% if one includes the capital conservation buffer, which must also be comprised of core capital. To combat procyclical behavior, Basel III will require banks to maintain a countercyclical buffer. The amount of the counter-cyclical buffer will range from 0%-2.5% of riskweighted assets. The exact amount of the counter-cyclical buffer will be decided by national

25

regulatory authorities and will generally be determined by the amount of credit in an economy, with more credit leading to a higher buffer. The purpose of the counter-cyclical buffer is to ensure that banks are sufficiently capitalized during periods of excess credit growth, which usually occur when the perceived risk in assets is low. Thus, the counter-cyclical buffer can be viewed as an extension of the capital conservation buffer in the sense that it counteracts the trend of low capital levels during times of low risk. Consequently, by maintaining high capital levels during “good” economic times, banks can avoid drastic measures to conserve capital during bad economic times, and thus avoid credit crunches. Assuming a counter-cyclical buffer of 2.5%, Basel III could potentially require banks to maintain, at a minimum, a capital level equal to 13% of its total risk-weighted assets. When it issued Basel III, the BCBS also indicated that it would work with the Financial Stability Board (FSB) to implement even higher capital requirements, in addition to those mentioned above, for large and systemically important banks. The specific details of these higher capital requirements had not yet been developed at the time Basel III’s release, but the BCBS stated that they would most likely consist of a combination of capital surcharges, contingent capital, and/or bail-in debt. Like the BCBS, the FSB is another international standard-setting organization comprised of financial regulatory authorities from numerous nations. However, unlike the BCBS, the FSB advises nations on the regulation of all aspects of the financial sector, not just the banking sector. Basel III also implements a leverage ratio, which will require banks to maintain an amount of capital that is at least equal to 3% of the bank’s total assets. As opposed to the riskweighted capital ratios, which compare a bank’s capital to the bank’s risk-adjusted assets, Basel III’s leverage ratio will compare a bank’s capital level to its total assets, regardless of their risk

26

level. By requiring a leverage ratio, Basel III ensures that banks maintain at least some amount of capital at all times, and thereby limits the ability of banks to engage in practices designed to evade minimum capital requirements. Thus, the leverage ratio will serve as a capital floor to ensure that banks have at least some amount of capital to protect it against unforeseen losses. B. Critique of Basel III The Basel III has been in its finalized form only since September of 2010, so it is too early to tell whether it will be effective in practice. Nevertheless, critics have already begun to voice their opinions on Basel III. One of the obvious criticisms of Basel III surrounds the level of capital it requires banks to hold. Critics who say the amount is too high point to the impact it will have on lending. By requiring banks to have higher levels of capital, Basel III reduces the amount of money a bank can lend. For example, if a bank has $100 worth of capital, under Basel II it could lend up to $1250 of risk-weighted loans ($100 would be the 8% minimum capital level required by Basel II). However, when Basel III is fully implemented, that same $100 of capital could now represent up to 13% of the bank’s total risk-weighted assets, which means the bank can lend up to only $770. Critics point out that a reduction in lending will inhibit economic growth. Banks, and their ability to inject money in the economy through lending, are an important component in economic growth. Therefore, by imposing lending restrictions in the form of higher capital requirements, Basel III is effectively restricting banks from doing their part in sponsoring a robust and healthy economy. The true economic impact of Basel III’s requirements is, of course, debatable. Given the short amount of time that has elapsed since Basel III was approved, it is obviously too early to

27

tell with any certainty what the real impact will be. However, preliminary reports have projected what this impact might be. A report produced by the Institute of International Finance (IIF) (an organization representing banks) concluded that Basel III’s requirements would result in a 3.1% drop in a nation’s gross domestic product (GDP) for every 1% increase in a bank’s capital ratio. In contrast, a similar report produced by the Bank for International Settlements (BIS) concluded that GDP would decrease by only 0.09% for every 1% increase in the capital ratio requirement. While both reports agree that Basel III will have at least some negative impact on economic growth, the different outcomes of these reports underscore the uncertainty surrounding the extent of the negative impact of Basel III’s requirements. Related to the economic argument is the concern raised by banks that higher capital levels will hurt bank profits. Critics argue that banks will compensate for the income lost from their reduced lending ability by increasing the interest rates they will charge on loans, thus making credit more expensive to borrowers. To accomplish this, banks will take on riskier assets regardless of the concomitant higher capital requirements. Therefore, with the higher capital requirements, not only will there be less lending, but the lending that does take place will be more expensive and riskier. Alleviating the concerns that the increased capital requirements will hurt the economy and reduce profits is Basel III’s implementation timeline, which does not call for full implementation of all of Basel III’s requirements until 2019. Such a lengthy timeline should give banks plenty of time to adjust to higher capital requirements and allow for a gradual and orderly transition from the old capital rules to the new ones. Indeed, even the IIF, in the report mentioned above, admitted that the lengthy timeline for implementation resulted in a less critical appraisal of Basel III.

28

On the other hand, there are critics who hold the opposite view, that the capital requirements imposed by Basel III are too low to ensure a bank can absorb losses of the same magnitude as those experienced during the financial crisis. In support of their argument, those calling for even higher capital requirements point out that many of the banks affected by the financial crisis, especially those in the United States, already had capital levels at or above the Basel III levels. Some studies even suggest that the necessary minimum capital ratio should be closer to 15%-20%, which would effectively double Basel III’s requirements. Maybe the biggest criticism of Basel III is what it fails to do. Many of the criticisms of Basel II go unaddressed in Basel III. For example, Basel III does not address the problems associated with Basel II’s methods of assigning risk to a bank’s assets— it does nothing to change the calculation of the bank’s risk-weighted assets and leaves in place the use of external rating agencies to determine risk. Nor does Basel III do anything to harmonize the IRB approaches to prevent vastly different risk-weighting methodologies from bank to bank. Thus, while Basel III has attempted to improve the numerator of the capital ratio, it has done nothing to improve the denominator, which many would argue was what needed the most reform. VII. Conclusion

The international regime for bank regulation has evolved considerably since the BCBS was established in 1974. Yet, as the global financial crisis made clear, after the first two Basel Accords, there is still room for improvement in the regulation of bank capital. While it’s too early to tell with any certainty, Basel III seems to be a step in the right direction. Yet, if the criticisms already put forth are any indication, Basel III may be just one of many steps yet to come.

29

Sources

Bank for International Settlements, www.bis.org. Basel Committee on Bank Supervision, www.bis.org/bcbs. Basel Committee on Banking Supervision, Basel I and Market Risk Amendment, http://www.bis.org/list/bcbs/tid_21/index.htm. Basel Committee on Banking Supervision, Basel II Framework, http://www.bis.org/list/bcbs/tid_22/index.htm. Basel Committee on Banking Supervision, Basel III, http://www.bis.org/bcbs/basel3.htm. BENTON E. GUP, THE NEW BASEL ACCORD (2004). DANIEL K. TURULLO, BANKING ON BASEL (2008). DUNCAN WOOD, GOVERNING GLOBAL BANKING: THE BASEL COMMITTEE AND THE POLITICS OF FINANCIAL GLOBALISATION (2005). Felix Salmon, Basel III Arrives, REUTERS, Sept. 12, 2010, http://blogs.reuters.com/felixsalmon/2010/09/12/basel-iii-arrives/. HAL S. SCOTT, INTERNATIONAL FINANCE: TRANSACTIONS, POLICY, AND REGULATION (16TH ED. 2009). HEIDI MANDANIS SCHOONER AND MICHAEL W. TAYLOR, GLOBAL BANK REGULATION (2010). Institute of International Finance, Net Cumulative Economic Impact of Banking Sector Regulation: Some New Perspectives, http://www.iif.com/regulatory/article+831.php Jacques de Larosière, Don’t Punish The Banks That Performed Best, FINANCIAL TIMES, March 3, 2011, available at http://cachef.ft.com/cms/s/0/1b085c0e-45e1-11e0acd800144feab49a,s01=1.html#axzz1G8JzDrCS. Justin Baer and Brooke Masters, US Banks Face $100 Billion Basel III Shortfall, FIN. TIMES, November 21, 2010, available at http://www.ft.com/cms/s/0/42d42de2-f593-11df-99d600144feab49a.html#axzz1GGMA6l6B. NOUT WELLINK, CHAIRMAN, BASEL COMM. BANKING SUPERVISION, BASEL III AND BEYOND (JAN. 17, 2011). Pablo Triana, Basel III Still Contains Seeds of More Chaos, FIN. TIMES, September 19,

30

2010, available at http://www.ft.com/cms/s/0/1185b16c-c27f-11df-956e00144feab49a.html#axzz1G86VwXtw. Paolo Angelini et al., Basel III: Long-Term Impact on Economic Performance and Fluctuations (Bank for Int’l Settlements, Working Paper No. 338, 2011). Patrick Jenkins and Brooke Masters, Bank Researchers Call for Doubling Equity Safety Net, FIN. TIMES, January 27, 2010, available at http://www.ft.com/cms/s/0/1f4841ea-2a0b-11e0997c-00144feab49a.html#axzz1G86VwXtw. Standard & Poor’s, Basel III for Global Banks: Third Time’s a Charm?, March 4, 2010, http://www2.standardandpoors.com/spf/pdf/fixedincome/Basel_III.pdf Third Time’s a Charm?, THE ECONOMIST, September 13, 2010, available at http://www.economist.com/blogs/freeexchange/2010/09/basel_iii.

31