Ratings

We base our bank credit ratings on fundamental analysis that encompasses institution-specific and system-wide factors, including external support. The system-wide factors underpin our banking industry country risk assessments (BICRAs), which establish a comparative benchmark for the baseline creditworthiness of a country's banks. The BICRA integrates several elements relating to the structure and performance of the economy, the legal and regulatory infrastructure supporting the financial system, and the competitive and operating environment of the banking system itself. The quality and effectiveness of bank regulation, and the track record of a central bank in managing turmoil in the financial sector, are key elements in the BICRA.

The BICRA excludes, however, the potential of extraordinary government intervention and rescue of failing banks, which Standard & Poor's tracks as a rating factor distinct from the BICRA.

No. We do not believe it is possible to rigorously quantify the probability that a government will rescue a failing private-sector bank and ensure timely repayment of all its unsecured debt obligations. This is particularly true with respect to healthy banks in mature market countries, where the full extent of a government rescue in the distant future is not possible to predict with precision today.

Government actions in a banking crisis, whether specific to one bank or systemic, are by nature unpredictable. For example, a government may choose to support fully some types of bank debt, such as domestic retail deposits, but to restructure other types, such as interbank borrowings. Corporate law and banking law establish the framework within which governments operate, but governments often take informal and unpredictable actions outside the established rules when the solvency of a bank or the entire banking system is threatened.

Standard & Poor's does not apply a probabilistic approach to ratings on banks, including state-owned banks. In addition to viewing the potential for intervention as not possible to quantify with rigor, we believe that a joint support approach is only valid when there is little or no correlation between the obligors. In our opinion the creditworthiness of banks and their governments are too highly correlated to apply a joint support approach to rating bank obligations. Moreover, due to the small size of the sample of defaulted banks, the correlation factor could not be set with sufficient confidence in its predictive value, in our view.

Yes, we believe that governments will continue to take various actions to prevent banks from failing. The important question is, will governments rescue banks under all circumstances and fully support all bank obligations across debt types? We believe the answer to that question is no. Bank counterparties do not rely on governments to rescue banks, including large systemically important banks, under all circumstances.

Over the past decade, a government rescue of a failing bank has been a relatively rare occurrence in most developed economies. On the contrary, when the financial position and credit quality of a bank has significantly worsened, a larger and stronger financial institution has typically purchased it, often with government influence behind the scenes. At times, private equity funds or other nonbanking investors have acquired troubled banks. Only on unusual occasions have governments taken direct control, recapitalized, nationalized, guarantee debts, or merged a troubled bank into another bank.

Moreover, we believe that future bailouts will be limited by budgetary constraints, multilateral agreements (such as those in the EU), and increased regulatory emphasis on market-oriented policies that promote market discipline.

Although private sector buyouts usually strengthen the credit of the acquired troubled bank, this is not always so. At times a rescued bank emerges from a support operation in a weakened state due to a forced sale of valuable assets or business lines, or because it is purchased by an entity, such as a private equity fund, that does not present positive rating benefits due to a lack of business synergies or the possibility of future financial aid. In addition, sometimes the support operation will lead to losses by investors, such as holders of regulatory capital instruments.

Standard & Poor's has long recognized in its credit ratings the benefits afforded to state-owned banks originating from the supporting government. We often rate state-owned banks higher than their stand-alone credit quality and stand-alone rating, due to expectations of extraordinary government support. Sometimes the rating on a state-owned bank is as high as that on the government, even in the absence of an explicit guarantee.

We designate state banks and policy financial institutions (such as export credit banks), as well as state enterprises in other economic sectors, as "government-related entities," or GREs, reflecting their closeness to the government and the strong likelihood that the government will take direct actions to support their creditworthiness. For bank GREs, this factor usually results in a final rating that is higher than the bank's stand-alone rating.

No. The governments of some countries are more oriented toward direct intervention; consequently, the potential for government rescue of failing banks is higher in these countries. At the other end of the spectrum we find countries where the potential for government action is highly uncertain. Most developed market countries, however, have policies that support banks, but favor proactive regulation over bank bailouts.

To formally integrate this factor into bank ratings, Standard & Poor's classifies governments into three categories according to the government's propensity to step in and support failing banks. The categories are:

• 1. Interventionist countries, where the government is highly likely to intervene directly and rescue failing banks.
• 2. Supportive countries, where the government relies on prudential and proactive policies, including rules and procedures to manage troubled banks, to maintain a sound banking sector.
• 3. Countries where the government may support banks but policy is unpredictable, either because of weak institutional infrastructure or due to reliance on market solutions or bank owners. We name this category "support uncertain."

Standard & Poor's will publish more information on these three categories in the coming months.

In interventionist countries, our analytical approach is to lift the ratings on systemically important private sector banks one or more notches above the stand-alone ratings on the banks. We classify several emerging-market countries as interventionist, particularly in Asia and the Middle East. Japan is also interventionist. In our opinion, the governments in these countries favor direct intervention and rescue of failing banks.

The governments of interventionist countries are very likely to direct resources to saving important banks (in size or economic role) that are crucial to the functioning of the economy. We categorize systemically important private sector banks in interventionist countries as GREs, under our methodology to classify countries by the propensity of the government to intervene and rescue failing banks. The GRE classification acknowledges the banks' closeness to the government and the strong likelihood that the government will take direct action impacting bank creditworthiness (usually positively, but sometimes negatively).

The credit ratings on private sector banks in supportive countries--the U.S., EU member countries, Australia, and Canada--receive no uplift for potential extraordinary external support above the stand-alone rating. Bank ratings in these countries do integrate important ongoing systemic elements, however, such as prudent bank regulations, central bank liquidity facilities, and supportive legal infrastructures, which are integral to the BICRA. Developed economies primarily rely on prudential supervision and private sector buyouts to prevent failures and manage troubled financial institutions.

In countries where we believe support is uncertain, such as Russia and Argentina, we do not lift the ratings on private sector banks above the stand-alone rating.

Standard & Poor's assesses the following factors to determine whether a private sector bank is systemically important:

• Market share in retail deposits.
• Overall size and attractiveness of the domestic franchise.
• Relative importance in providing credit and liquidity to the market.
• Strategic role in the economy and potential systemic implications of a default.
• Importance of niche products and services, such as payment systems and exchanges.

Systemically important banks play a major role in the national economy through classic banking intermediation or delivery of critical services; for example, payment systems or securities custody. Market share in retail deposits is an important factor because the failure of a bank with a large retail presence would have damaging consequences for confidence in the banking system. Although deposit protection systems can offset this risk, governments are typically loath to allow a problem to deteriorate to the point where deposit insurance programs kick in. We broadly consider banks with a 5% or more market share in retail deposits as systemically important.

In the coming months, Standard & Poor's will publish information on private sector banks we classify as systemically important.

Our criteria can take on an added dimension when a bank's stand-alone financial condition and creditworthiness deteriorate significantly. When the situation worsens to a point where the bank may need some type of external support, and when we have a sufficient degree of confidence that help is forthcoming, we may lift the final credit rating on a bank above the stand-alone rating. The higher our degree of confidence in external support, the stronger the source of support, and the clearer the means of that support, the higher the potential ratings uplift. In some cases, the downward movement in the credit rating on a troubled bank may be slowed or reversed by the growing prospect of third-party support, including from the government. In other words, although we do not uplift the stand-alone ratings of healthy banks in supportive countries today in expectation of future extraordinary external support, we at times uplift the credit ratings of troubled banks when we have greater visibility on specific external support actions.

For troubled banks, two opposite forces often drive the institution's creditworthiness: a deteriorating financial situation that worsens credit quality and an increasing likelihood of private sector takeover or government rescue that improves its creditworthiness. Our rating policy balances the opposite forces. We project the likelihood of takeover or government rescue and the financial condition of the entity after it emerges from the rescue operation, and set the rating accordingly.

No, we do not impose a floor that limits the decline of bank credit ratings. We do believe, however, that banking industry dynamics yield identifiable patterns that reflect the positive effect of external support actions on ratings over the past 25 years. We integrate the economic and sectoral fundamentals that underpin these patterns in our bank credit ratings.

Ratings on banks transition from 'AA' to 'A' with roughly the same frequency as those on industrial companies, but the passage from 'A' to 'BBB' and from 'BBB' to 'BB' is less frequent. Over the past 25 years, a 'BBB' rated bank in the U.S. or Europe was three or four times more likely to transition to the 'A' category over a five-year period than to the 'BB' category. Banks rated 'BBB' tend to be raised more often than lowered because, as the financial condition and creditworthiness of a bank deteriorates, external forces usually intervene to stop the slide in one of the following ways:

• The bank is bought and absorbed into a larger financial group, and is no longer rated. This is the most frequent outcome.
• The bank is bought and becomes a strategic or core member of a larger, more creditworthy group, and the rating is raised.
• Regulators take actions, such as providing liquidity or restricting a bank's activities, that stop the deterioration.
• The government intervenes financially by injecting capital, removing bad loans, or guaranteeing the bank's debts, and thereby rescues the bank.

These potential scenarios influence bank ratings at all points in the credit cycle. For example, high franchise value is an important positive factor in a bank rating and can compensate for poor financial performance. A poor financial performer with a strong franchise is more likely to be acquired than a poor performer with a mediocre or weak franchise.

The credit ratings on the major banks in Europe, North America, and Australia are in the 'AA' category, which to a significant extent reflects the regions' extensive and supportive regulatory and institutional infrastructure and low level of banking industry country risk. Access to liquidity from the central banks, together with regulation and supervision, enhance the banks' stable access to capital market and retail funding. These systemic elements, along with the banks' business profile and financial strength, are key components in the stand-alone ratings on these banks.

We believe that quantifying the probability that a government will support a healthy private sector bank and forecasting the nature of this support is a highly theoretical, and, ultimately, arbitrary exercise, however, because the scenario under which such support would be required is extremely remote. We factor extraordinary government support actions in the rating on a major bank only when the bank is in difficulty and the extent and nature of extraordinary support is sufficiently visible and predictable.

The recent upswing in Standard & Poor's bank ratings has been driven by the strong financial and commercial fundamentals of the global sector, not because of an increased expectation of future government rescue. Standard & Poor's high credit ratings on leading banks in mature economies reflect the very low probabilities of default, but that probability is not zero. Differences in the ratings on these banks reflect small but significant differences in their credit profiles.