Credit Risk Management and the Basel Accords Explained

Introduction to Banking and Credit Risk

Banks are an important part of the economy of any nation. Traditionally, banks operate as financial intermediaries serving to satisfy the demand of people in need of various forms of financing. Through this function, banks enable individuals to purchase homes and businesses to expand. These financial institutions therefore facilitate investment and spending that fuel economic growth. In spite of their vital role in the economy, banks are nevertheless prone to failure, and just like other types of businesses, they can go bankrupt. Unfortunately, bank failures carry far more significant implications than failures in most other industries. As witnessed during the Great Depression, and more recently following the global economic crisis and recession, the stability — or lack of it — in the banking system can trigger economic epidemics that impact millions of people.

With respect to this vulnerability, it is important for banks to operate in a sound and safe manner. One means of achieving this is ensuring that governments have put in place practical and strict regulations for banks. At the same time, with the spread of globalization, the activities of banks are no longer restricted within the borders of individual countries. As a result, there is an increasing need for international cooperation in the regulation of the banking system (Larson, 2011).

There appears to be some light at the end of the tunnel, as the Basel Committee on Banking Supervision (BCBS) is positioned to meet this need. As an international advisory authority on bank regulation, the BCBS has issued guidance on matters crucial to ensuring the healthy operations of banks across the globe. One such pressing issue is the regulation of bank capital. The process of addressing this issue has been ongoing for over two decades and has led to the promulgation of capital adequacy standards that can be implemented by regulators of individual countries. Collectively, these standards are referred to as the Basel Accords. At times these Accords have resulted in disagreements, yet they remain critical to the formulation of regulatory policy associated with bank capital. The BCBS has so far produced three such Accords; Basel III, published in 2010, is the most recent. Each Accord purports to improve upon the specifics of the previous one. However, there are indications that the most recent Accord is not without flaws, and it may not be the last (Larson, 2011).

Rather than focusing narrowly on the stability of individual banks or the Basel framework in isolation, it is worth considering whether the Basel Committee has addressed even the smallest details of the new Accord correctly — a task that could greatly impact the bottom lines of banks. It is therefore more productive to focus on some general and fundamental questions. For instance: To what extent has Basel I succeeded in accomplishing its stated goals? How successful is Basel II expected to be in achieving its goals? Are those stated goals themselves desirable? And perhaps most fundamentally: is the Basel Accord — specifically, the international harmonization of bank capital standards — necessary or desirable for the stability of the financial system? (Rodriguez, 2003).

After several years of deliberations following the Latin American sovereign defaults of 1982, the BCBS finally completed the Basel Capital Accord in 1988. It was established with two main objectives: to strengthen the stability and soundness of the international banking system, and to achieve a high level of consistency in its application across separate nations — thereby reducing a source of competitive inequality among international banks. To this end, the Basel Accord requires banks to satisfy a predetermined minimum capital ratio mandatorily equivalent to at least eight percent of total risk-weighted assets (Rodriguez, 2003).

The Banking Sector: History and Financial Intermediaries

The BCBS focused on capital standards for two major reasons: first, because Congress directed banking regulators to work with regulators from other nations to ensure that banks maintained sufficient capital bases (Kapstein, 1991; Oatley & Nabors, 1998); and second, because capital acts as a buffer protecting bank deposits in the event of asset losses (Rodriguez, 2003).

For a long time, banks and other financial institutions have been subject to greater government oversight than most other sectors of the economy. Bank regulations have historically taken the form of entry restrictions, limitations on activities, reserve requirements, geographical limitations, and capital requirements (Benston, 1998, pp. 27–85; Kroszner, 1998, p. 421; Kane, 1997; Goodhart et al., 1998, Chapter 9). Today, most regulation rests on the rationale of safety, soundness, or consumer protection.

According to Kroszner, the primary historical reason governments regulated banks and financial institutions was to finance wars (1998, p. 419). However, a long-standing tradition among economists, dating back to Adam Smith, has held that the banking system differs from other industries because of the nature of the activities in which banks engage — and that this difference justifies a form of regulation and supervision.

Smith alluded to the inherent instability of banks operating under a fractional reserve system, which, if true, justifies the need for regulation (Smith [1776] 1937, pp. 285, 308). Banks act as financial intermediaries: they make profits by taking deposits, extending loans, and investing in marketable securities and other financial assets. In the aggregate, this process produces a multiple expansion of the money supply. The liabilities of banks are normally fixed in value and payable on demand, while the value of banks' assets is variable and not immediately collectible. For these two reasons, banks are generally considered more prone to failure and bank runs — particularly in the event of a sudden, large-scale withdrawal of funds by depositors who have lost confidence. This could adversely affect solvent institutions through a contagion effect, in turn negatively impacting the entire financial system. This largely explains why banking regulation is vital (Rodriguez, 2003).

Banks operating under the fractional reserve system are inherently fragile and at risk of runs when depositors have limited information about a bank's activities and financial health (Diamond et al., 1983; Dowd, 2001). Moreover, a run on a bank can theoretically destabilize otherwise sound institutions.

Historically, the private sector was adept at managing this fragility before government-sponsored deposit insurance began addressing the issue. Initially, banks would make their capital levels known to depositors and investors to boost confidence. As Benston (1998, p. 39) notes, banks regularly advertised the size of their capital and surplus prominently in newspapers and within their branches. It is worth noting that capital and surplus levels at that time were considerably higher than they are today (Kaufman, 1988).

Additionally, depositors and investors continuously monitored bank activities and demanded higher rates of return if they suspected their bank was taking on excessive risk. Before government-sponsored deposit insurance, banks also formed private clubs and clearinghouses to assist one another. Timberlake explains that a bank seeking membership in such associations had to meet specified requirements regarding capital levels, permissible activities, and risk profiles (1993, Chapter 14). Furthermore, option clauses in contracts permitted banks to defer payments for a set period in exchange for paying depositors a higher interest rate on the debt. These clauses were used extensively during the Scottish free banking period of the 18th century and effectively eliminated panic runs while giving banks breathing room to reorganize assets without engaging in fire sales. Finally, debt holders signed covenants with banks restricting the activities and investments banks could pursue.

Market discipline by shareholders and depositors worked well in preventing runs and, when runs occurred, in containing their spread. In the period between the end of the Civil War and the end of World War I, bank failures in the United States were low relative to non-financial firms. Moreover, banks that failed were typically insolvent before the run, not as a result of it. In this regard, runs on insolvent banks actually served the useful economic purpose of eliminating firms within the financial system that were motivated to engage in risky lending to recover their solvency — a strategy that would have harmed other market participants. During the Great Depression, depositors were able to distinguish between banks suffering from liquidity problems (though solvent) and those that were genuinely insolvent (Calomiris & Mason, 1997).

The large number of bank failures during 1929–1933 resulted in the restructuring of the banking industry along product lines and led to the creation of the Federal Deposit Insurance Corporation (FDIC) following the passage of the Banking Act of 1933 (Friedman & Schwartz, 1963). The FDIC was established to achieve three goals: to restore public trust in the banking system, to safeguard the payments system, and to protect existing branching limitations (White, 1997). Its creation had three major effects: it deprived shareholders and depositors of incentives to monitor bank activities; bank runs became rare, even though they were not catastrophic to financial system stability when they did occur; and by charging a flat fee, the FDIC created a classic moral hazard by effectively subsidizing risk-taking by banks (Kaufman, 2002).

Despite a broad reform program covering capital markets — including clearinghouses, shadow banking, and over-the-counter (OTC) derivative markets — banking remains central to the financial system. Attention must be given to several implications for the credit system, beginning with banks' capital structure and micro-regulatory regimes, followed by new macroprudential policies and their impact on credit conditions (Tucker, 2013).

Banking reforms at the micro-level have two significant components. The first is a step change in regulatory requirements on leverage, liquidity, and capital, aimed at reducing the likelihood of bank failure. The second entails recognizing that failure cannot and should not be entirely eliminated, and establishing effective and credible resolution regimes. Individually and together, these components will change how risks in bank portfolios are distributed across bondholders, shareholders, taxpayers, and depositors (Tucker, 2013).

The most significant impacts are expected from making orderly resolution credible for the largest and most complex institutions. Funding costs for any bank where a bailout was still widely anticipated would remain heavily subsidized, as in earlier periods. Where uncertainty was high, there would be a tendency toward more or less subsidized funding alongside a bias toward short-term financing — both signs of distress. Such arrangements, however, are not sustainable. The policy rationale is clear: taxpayers should not be required to provide solvency support to banking companies (Tucker, 2013).

Rather, losses exceeding a bank's equity base should fall on bondholders and other uninsured creditors, consistent with the creditor hierarchy applicable in bankruptcy. Exposing bondholders to risk will tend to raise banks' general cost of financing during normal times compared to the past. Investing in longer-term debt while bearing risk — despite a fixed return — will induce market discipline through both price and rationing. This will most likely incentivize banks to become better capitalized through some combination of less risky business activities and less leveraged balance sheets. For society, with credible resolution plans in place and a reduced likelihood of failure, public finances should become more resilient and government bond yields lower than they would otherwise be (Tucker, 2013).

When fully implemented and when equity surcharges of up to 2.5 percentage points for systemically significant institutions are included, Basel III raises equity capital requirements by nearly an order of magnitude. This means that in the future, banks will be capable of absorbing larger losses while remaining a going concern. In other words, more of the risk in bank portfolios is being shifted onto stockholders, leaving less with creditors. Less clear, however, is the effect on banks' overall funding costs (Tucker, 2013).

Bank Capital Structure and Regulatory Reform

More than fifty years ago, Modigliani and Miller (1958) famously showed that, under certain conditions, a company's overall cost of funds is independent of how it is financed. Equity is more expensive than debt finance because it absorbs losses first. Raising the equity-funded share of a firm's balance sheet increases the proportion of the more costly form of financing. However, it also reduces the risk to debt holders, lowering debt financing costs; and the cost of each additional unit of equity falls as the balance sheet becomes less leveraged. The Modigliani-Miller proposition is that these two effects precisely offset each other, leaving the overall cost of financing unchanged. The argument is straightforward: provided that the returns on a firm's asset portfolio do not change as its funding structure changes, the cumulative cost of its financing is constant regardless of how those risks and returns are distributed between debt holders and shareholders (Tucker, 2013).

If this were fully accurate, higher equity requirements would impose no cost on banks or the real economy. In reality, several features of the real world depart from the Modigliani-Miller framework. Like other firms, banks can deduct interest paid on debt from corporation tax, reducing the after-tax cost of debt relative to equity. All else equal, the average funding cost can therefore be reduced by issuing debt. That cost advantage should largely be passed on to clients (Tucker, 2013).

Banks are also unique in that they finance themselves with retail deposits that provide monetary services — the nature of their liabilities is central to their business, not merely a feature of how the business is funded. Most such deposits are protected by guarantee schemes, making the interest rates on transactional deposits relatively insensitive to the risks a bank is running, and thus making deposit-based funding cheaper for banks than standard debt finance. Unlike ordinary firms, part of the value of a banking institution derives from its capital structure itself (Tucker, 2013).

There is also a broader point concerning the economy-wide benefits of the maturity-transformation services that banks provide by funding longer-term loans with shorter-term liabilities. In the context of fractional-reserve banking — combining committed lines of credit with demand deposits — banks provide liquidity insurance to their clients. This allows firms and households to economize on liquid savings, directing more of the economy's savings toward higher-risk projects that drive longer-term growth. There are genuine social benefits from the deposit-financing model of banking (Tucker, 2013).

However, not everything about banks argues for more leverage. There is a material asymmetry: for banks, unlike ordinary firms, the costs of approaching failure are severe even before the point of actual insolvency. As a bank nears failure, short-term funding will tend to run, and asset values will be cut down by forced sales required to raise liquidity. Beyond the point of failure, bankrupt banks find it harder to enforce contracts and collect debts, imposing significant costs on the failed institution's creditors and on society more broadly (Tucker, 2013).

Together, these departures from the Modigliani-Miller framework imply that a bank's overall funding costs are indeed dependent on its capital structure, but in a non-linear way. As long as a bank's capital buffer is sufficient to make bankruptcy seem remote, the bank will tend to economize on equity, partly for tax reasons. Conversely, if capital levels are dangerously thin, bankruptcy becomes a real possibility, and economizing on equity capital may prove counterproductive as debt holders — factoring in liquidation or resolution costs — demand progressively higher interest rates (Tucker, 2013).

Whether it is resolution or liquidation that looms, an effective resolution regime can materially reduce both social and private bankruptcy costs. Relative to a world in which liquidation is a plausible threat, having a credible resolution regime will tend to reduce bond financing costs and increase the share of bonds in the capital structure. However, relative to a world in which government bailouts are confidently expected, an effective resolution regime will tend to raise bond costs and reduce their share. Effective resolution regimes are therefore necessary to give bank managers and investors appropriate incentives to maintain a sound capital structure (Tucker, 2013).

Even so, banks' private capital structure choices will not produce outcomes that are socially optimal, given the negative externalities and spillovers from failure. Resolution may help reduce those spillovers, but banks should be required to maintain a capital structure that makes organized resolution practically feasible (Tucker, 2013).

This discussion has been largely a matter of comparative statics: would one prefer to finance a new bank mainly with equity or mainly with debt? It has glossed over what happens when a firm considers changing its existing capital structure. There are three relevant issues (Tucker, 2013).

The first arises from asymmetric information. Equity issuance may be costly if investors fear that a new issue signals that management believes the share price is too high — i.e., that earnings and assets are worth less than the market has assumed. This may occur even if the bank is not at risk, but simply less well capitalized than it should be. One remedy is for the prudential regulator to require the firm to raise the required equity when a deficit is identified. Another is for banks to issue high-trigger contingent capital instruments (CoCos) — bonds that convert to equity if the bank's capital ratio falls below a prescribed but realistically high level. In a stable state, for a bank with a minimum equity ratio of ten percent, such a trigger might be set at eight percent: low enough that the insurance provided by these CoCos is not prohibitively costly, but high enough that the bonds would convert to equity while the bank was still able to fund itself in the market (Tucker, 2013).

The second, and in some respects larger, issue relates to the debt-overhang problem. Suppose that, following a significant deterioration in the macroeconomic environment, a bank's equity base — even after high-trigger CoCo conversion — proves too thin to cover business risks. Debt spreads widen and the value of bonds in issue falls. A new equity injection would raise the value of the firm by reducing the probability of bankruptcy — but because bondholders ultimately bear the costs of bankruptcy, they and other creditors would be the primary beneficiaries of recapitalization. Shareholders in a badly capitalized firm therefore have an incentive to wait and hope conditions improve (Tucker, 2013).

A specific problem could arise if losses were large enough to deplete high-trigger CoCos, leaving the bank undercapitalized but not yet at the point where solvency is in doubt. By deleveraging and continuing to operate, the bank's supply of credit to the economy would be impaired. An alternative in such circumstances might be low-trigger CoCos to force recapitalization. Nevertheless, the bank might suffer a run, requiring resolution. Therefore, bonds must convert into equity and absorb first losses upon entry into resolution. This is the most coherent way to think about the Point-Of-Non-Viability (PONV) instruments that count toward "full capital" in the existing Bank Capital Accord (Tucker, 2013).

In short, the system cannot be relied upon to recapitalize itself as the probability of failure rises. There is therefore a premium on getting micro-prudential rules for minimum capital broadly right rather than, as occurred in the past, seriously wrong. If the equity cushion is too thin, the system risks collapse; if banks are forced to hold too much equity, the genuinely valuable liquidity services they provide to the economy may be stifled (Tucker, 2013).

This analysis informs the general shape of a richer regulatory Capital Accord for the future — one that carefully distinguishes between the different stages of a bank's life cycle. Regulatory intervention is clearly needed to set a minimum equity level providing sufficient going-concern loss absorbency. That is the primary function of the existing Basel III Accord. Nonetheless, it is insufficient on its own. Regulation is also needed to specify a minimum level of term bonded debt to provide gone-concern loss absorbency (Tucker, 2013).

In response to private cost savings arising from the tax regime, governments should authorize minimum issuance of term bonds from prescribed parts of banking groups so as to make resolution executable. Rather than short-term debt — which, while disciplining through its capacity to run, cannot absorb losses — longer-term bonds can recapitalize a firm in resolution and impose market discipline through both price and rationing (Tucker, 2013).

A richer Accord might go further, authorizing high-trigger CoCos to encode recovery measures into a bank's capital structure, and even low-trigger instruments to facilitate resurrection if a bank's equity is severely impaired but it is not yet on the verge of bankruptcy. Together, such instruments constitute a robust Accord covering: normal times (equity); resolution (PONV instruments or low-trigger CoCos); and recovery (high-trigger CoCos) (Tucker, 2013).

Reforms to capital requirements and resolution regimes — even a more comprehensive Capital Accord — may not alone be sufficient to maintain systemic stability. This is for two reasons. First, however rich, any reasonable regulatory capital regime will eventually prove inadequate, either because the economic environment proves riskier than anticipated at calibration, or because of regulatory arbitrage. Second, behavior within the monetary system can generate elevated risks through myopia — a tendency to overlook risks during upbeat conditions. Risk eventually crystallizes, prompting recognition of excessive leverage and asset overvaluation. Credit then tightens, exacerbating the macroeconomic downturn. Addressing the too-big-to-fail (TBTF) problem is necessary but not sufficient to eliminate boom-bust cycles; banking systems composed of many small banks have historically driven themselves over the cliff as well (Tucker, 2013).

For regulation to respond dynamically to changing circumstances so as to preserve systemic stability, macroprudential policy is required. In the UK, the newly established Bank of England Financial Policy Committee (FPC) has as its primary objective the enhancement and protection of the resilience of the UK financial system, with a secondary objective of supporting employment and growth. Parliament has granted the FPC authority to direct changes in capital requirements for banks, and to make recommendations to UK prudential supervisors and securities regulators on anything pertinent to stability (Tucker, 2013).

In several respects, the effects of a temporary macroprudential change in capital requirements resemble those of a permanent change to the capital regime. Any action that raises a bank's equity base by, for example, ten percent will raise its loss-absorbing capacity by ten percent. This makes the system better equipped to withstand losses if they crystallize. Regardless of whether such action dampens the boom phase of the credit cycle, it will reduce the severity of the bust. Fewer banks are likely to fail; a credit supply tightening is therefore less likely to become a full-blown credit crunch; the resulting slowdown in output growth should be less severe; and default rates on bank loan portfolios should be lower. The primary objective of macroprudential intervention — improving the resilience of the financial system — may therefore be achieved even if the credit boom itself was not substantially dampened (Tucker, 2013).

Whether macroprudential measures can quell a credit boom depends significantly on their effect on the cost of finance. Much analysis of this question is oversimplified, based on the assumption that a rise in capital requirements will always and everywhere tighten credit conditions and slow the boom. An expected substitution toward more costly equity finance will tend to push banks' funding costs upward, but the overall effect on funding costs will depend on whether, and to what degree, debt financing costs fall as a result of a lower probability of insolvency (Tucker, 2013).

There is also a crucial distinction between banks being required to hold more equity in a stable state versus a macroprudential authority intervening temporarily to raise capital requirements. The policy action of a macroprudential authority conveys information about policymakers' views on current risks to financial or economic stability and about the authority's reaction function — matters directly relevant to how banks and markets interpret the signal (Tucker, 2013).

Conclusion and Criticism

Changes to capital requirement rules were proposed by Basel II. The BCBS, in its efforts to reform the Basel framework, sought to incorporate linearly greater complexity into the system in order to utilize the portfolio invariance properties of the Basel model. This complexity has made it harder to align capital rules with the actual structure of financial institutions (Blundell-Wignall et al., 2013).