The trend in the Downgrade/Upgrade ratio is another indicator for the direction of the high-yield market . In a weak economic cycle this ratio will deteriorate. A reversal of this trend will occur when the economy gains on strength resulting in higher company profits. As a result, the credit quality of companies will improve and hence the Downgrade/Upgrade ratio. If this ratio is computed on the basis of the nominal amount of debt outstanding one will observe a sharp increase in 2002. The large amount of Fallen Angels is responsible for this spike.
It has to be mentioned that this ratio has a lagging character concerning the future direction of high-yield spreads because rating agencies will usually react to a change in the credit quality of companies and the whole market only with a lag of a couple of months. Nevertheless, this ratio appears to be useful for describing the current state of the credit market. Hereby the absolute number is less important than the trend, for example, in 2003 we have seen a clear deleveraging effort in the credit market even though downgrades outnumbered upgrades. The important message was that the negative trend from 2001 to 2002 was stopped.
Corporate bond ratings are not static but rather tend to move over time. The lower the rating class the more likely a rating change over a 1-year period. Corporate ratings will usually be assigned an outlook (credit watch positive, positive outlook, stable outlook, developing, negative outlook or credit watch negative). It is important to notice that the rating agencies Moody’s and S&P may change the rating of a corporate issuer with a stable outlook immediately when an event occurs which justifies the rating action. The rating migration between lower investment grade (Baa3) and noninvestment grade (Ba1) is very important for the quality of the whole credit market. In times of deteriorating quality in the credit market the number of “Fallen Angels” (companies which lost their investment grade status) will outweigh the number of “Rising Stars” (companies which were upgraded to investment grade status). In periods of improving credit quality and increased economic growth this trend will reverse and “Rising Stars” will outnumber “Fallen Angels.”
The correlation matrix shows that high yield has a low correlation with other asset classes. The highest relationship exists with small stocks like the Russell 2000 Index. The weakest correlation was measured for Treasuries. It is of interest that the correlation with High Grade shows only a reading of 0.47 which is not very strong indeed. We will show later in this chapter that those relationships measured over a long period (here: Jul. 83–Dec. 03) can vary considerably during various economic and credit cycles.
We summarize the correlation of monthly returns for the various rating classes. BB rated credits have the highest correlation with Bs even though there is also a strong relationship with BBBs. Bs have the highest correlation with C-rated paper. The relationship with BBBs is at 0.43 quite low. C-rated bonds behave almost completely independent from investment grade paper. Again we have to point out that the correlations will vary at different periods in the economic and credit cycle.
The high-yield market grows continuously and makes up around 20 percent of the high grade market. The influencing factors for its size are the volume of “Fallen Angels” (companies being downgraded to high-yield), “Rising Stars” (companies being upgraded to investment grade), redemptions and the new issue volume which is driven by demand (fund flows) and the willingness of companies to enter the high-yield market as a source of alternative financing. Particularly in Europe many companies used to depend heavily on bank loans. As banks are not willing to take more company risk a lot of companies are forced to the capital markets. High yield is an interesting alternative financing source especially for small- and medium-sized companies. High yield will also be the financing choice for many Leveraged Buy Outs (LBOs). We see high-yield market bonds as a percentage of outstanding corporate bonds by principal amount.
Generally, there are two problems with the estimation of potential losses from trading risks. Value at Risk (VaR) is an estimate of the loss that is not exceeded with a given high probability over a certain time horizon. It is derived from a bank’s own risk model. Not only do these risk models vary between banks, but they also use different parameters and sensitivities.
Hence, the same trading book positions may produce different estimates of risk. While many reported VaR measures are at the 99 percent confidence level for losses on positions held for a single day, others use different confidence levels, for example, 95 percent confidence levels and covering holding periods up to ten days. The choice of parameters feeds directly into the resulting VaR. Models will also change over time as they are updated in line with technological and business development. Consequently, even a single bank’s VaR numbers are likely to be inconsistent over time. In particular, the extent to which models account for hedging, basis risks and correlations is likely to influence the reported VaR. While there are substantial difficulties in comparing reported VaR levels, they may be helpful to give a clue as to whether a certain financial institution has a higher level of market risk than others. In particular, it may reveal whether further analysis is required. Special attention should be paid, when the losses incurred frequently are greater than those predicted by the model. This is an indicator that the level of market risk is rising or the model does not accurately reflect the risks taken.
Therefore, rating agencies tend to base their assessment of capital adequacy not only on balance sheet assets but also in relation to managed assets. In this context, managed assets additionally include assets previously originated by the bank and subsequently securitized. Typically, this relates to the securitization of credit cards and mortgages where the originator remains the servicer of the portfolio. Even though the assets are legally sold, the rating agencies frequently assume an enduring responsibility of the originator for the transaction. In the event that such a liability materializes, the financial and reputational consequences for the bank can be significant. While the above-mentioned liabilites are contingent in nature, trading risks arise from the daily business of a bank, and as such, are rather transitory. In the meantime many of the trading risks of a bank are related to the use of derivatives.
Banks now use derivatives to manage equity, FX, credit and interest rate risks and to construct products for customers. But the detail information provided on derivatives is usually rather limited, and infrequent, comparability between VaR measures low, and it is often difficult to determine exactly what constitutes trading risks and earnings. Nevertheless, it is essential to monitor these elements so that, over time, a broad picture of the risk profile evolves.
When trying to assess the risk profile of a bank, investors usually start with the information reported by the company itself. However, the aggregation undertaken by banks in their public reporting as well as the extremely heterogeneous disclosure from banks, make it a challenging task to fully perceive a bank’s risk profile. Special attention should be paid to the relationship between on- and off-balance sheet risks. This is particularly true for changing business models. In such cases, potential divergences between loans on the balance sheet, regulatory capital requirements and profitability may arise. Another point to consider are contingent risks. They are particularly hard to identify, as examples like Ahold and Parmalat have shown. Capitalia’s decision to compensate retail investors for the losses incurred by the purchase of Parmalat bonds is a typical example of a contingent liability. It highlights the extent to which reputation can influence the actions of financial institutions in certain situations.
Many large banks have experienced significant change in their business. Retail deposits are no longer sufficient to fund loan portfolio growth. Since banks increasingly use senior, secured and covered bonds to fund their business, liquidity and interest margin profiles have changed and become a critical factor in the financial performance and strategic success of a bank. Usually, a bank’s liability profile adjusts slowly to the relative cost of funds from various sources. However, if the viability of a financial institution’s business model depends on the availability, cost or pricing structure of funds the balance between funding cost and the business model can necessitate a change of the strategy. An in-depth analysis is therefore recommended if the relative importance of various funding sources changes significantly.
Although we have spent some time on the specifics of bank debt, it is worth remembering that liquidity as well as supply and demand dynamics affect bank spreads in the same way as industrial or utility spreads. Liquidity typically depends not only on the size of a bond, but also on the time since it has been issued. While investors usually demand a premium to invest in less liquid bonds, strong retail demand can hold spreads tight, even if fundamentals are deteriorating or there is new issuance. Recent experience shows that the latest issue tends to become the new benchmark bond of an issuer, enjoying strong demand and therefore often trading relatively rich compared to previously outstanding bonds of the same issuer. During the last couple of years supply in the banking sector was driven to a large part by changes of the strategy of a bank. An example is the shift of the European banking sector away from retail banking towards investment banking in the late 1990s.
M&A activity, reorganization of banking groups and recapitalization of insurance subsidiaries are further items to watch carefully.
With respect to funding costs and ratings, European banks have tended to focus more on regulatory requirements than on economic capital. If a bank wants to increase its regulatory core capital ratio, it has a variety of options. Basically it can increase its Tier 1 capital or reduce its risk-weighted assets. As stated above, issuing Tier 1 preferred or common equity are the simplest ways to increase core capital, but from a longer term perspective the enhancement of profitability and thus retained earnings also helps to improve the Tier 1 ratio. Pressure from shareholders and the lower cost of capital lead many banks to consider the issuance of Tier 1 preferred as a particularly attractive way to improve the core capital ratio. However, the upper limit for innovative hybrid Tier 1 preferred of 10 percent of a bank’s total tangible equity constrains the degrees of freedom. Overall, when a bank requires funding, raising Tier 1 or Tier 2 in the capital markets is the most compelling alternative. However, when no fresh capital is required there may be more efficient alternatives. Reducing risk-weighted assets also serves to accomplish better regulatory capital ratios. This can be done in two ways. Either through a reduction of risky assets, that is direct sales of some assets such as equity holdings or securitization of assets, or through a reduction of risk-weightings. Buying default swaps on risky assets or synthetic securitization are typical actions that help to reduce risk-weightings.
Generally, when no funding is required securitization may be the cheapest way to achieve a better regulatory Tier 1 ratio. Investors should bear that in mind when forecasting issuance activity in the banking sector.