In the world of corporate bonds, the yield spread over the sovereign benchmark reflects the liquidity premium and, more importantly, investors’ compensation for default risk. Accurate measurement of default risk is critical in determining the required risk premium on bonds. But the official default rate, which is measured and published by rating agencies, can be subject to distortions because certain capital losses, corporate buybacks, tenders and exchanges may not be accurately represented in the default rate calculations. In a macroeconomic environment likely to be characterized by greater performance differentiation among credit markets, sectors and issuers, paying closer attention to these potential distortions will prove invaluable to the credit selection process.
Conventional Default Rate Calculations May Omit or Discount Important Factors Typically, annual default rates are calculated by dividing the total value of euro-denominated defaults by the average value of all outstanding euro-denominated bonds within a particular category of bonds in any given year. The annual default rates are then added and averaged over a longer time horizon to provide an estimate of the average annual default rate within a given rating category. When gauging the inherent default risk of a specific bond, investors tend to consider the credit quality of the issuer along with a range of other factors including macroeconomic growth prospects and lending standards.
In the event of a default – when the issuer is not able to pay the interest on the bond or pay back the principal at maturity – the bonds are likely to drop to their recovery value, which is usually above zero. (The recovery value, or recovery rate when stated as a percentage of par value, is the price of the debt at the end of the month in which the default occurs.) Over the longer term, the average default rate for high yield issues (those rated below BBB) has typically been between 4%–5%, while the average recovery rate has been around 40% of the notional amount. The European high yield default rate is still lower than the U.S. default rate, though it has started to rise in the last few years. According to Moody’s Default Report (January 2010), the issuer-weighted default rate in European high yield rose from 1.1% in 2007 to 2.0% in 2008 and again in 2009 to 10.3%. We think it is likely that the default rate has peaked in this cycle and should decline to levels somewhere around 4%–6% in 2010, a rate closer to the long-term average.
Under their existing measurement framework, the rating agencies may not always take into account certain capital losses when calculating the official default rate. For example, the official default rate does not include defaults of unrated bonds, nor defaults in the loan market (the latter due to their private nature). Other potential risks for bondholders that are only partially factored into standard default rate analysis are bond buybacks, tenders and exchanges. If investors do suffer a loss as a result of these actions, this is usually compared to a (partial) default on the company’s obligations – it is therefore worth questioning the prevailing methodology behind “official” default rates. Are current default rates underestimating the actual losses bondholders incur? Unfortunately, there is no standard answer to this question, not least because there are different categories of activities that may increase default risk or otherwise complicate the calculations.
Buybacks and Beyond: Categories of Potential Default-Related Risks Companies can quietly buy back their own bonds in the open market. They usually do this through investment banks so the seller is less likely to know if the bond is being repurchased by the issuer, and in any case, the sale is voluntary. For the issuer, these buybacks are opportunistic and usually don’t exceed a small portion of an existing bond issue.
In order to buy back a more meaningful portion or even the entire bond issue, companies will typically launch a formal tender, where they extend an official offer to buy back their own bonds. Usually, the company will either identify a price or price range plus a maximum notional amount of bonds it is interested in repurchasing. In the case of a price range, the final tender price is determined by a (reverse) Dutch auction – the lowest price at which the desired quantity can be repurchased.
We can say both of the above activities fall within the “par tender” category because they are instances where companies tender for their bonds at around par value. Examples include cash-rich companies for which debt reduction is a very efficient way to use their cash. In this situation, investors generally have the option of keeping their bonds or tendering them in exchange for a cash payment. Such companies, even if they have a non–investment grade rating, do not usually suffer from any material operational issues at the time of the tender. Investors are likely to make their decisions based on factors such as tender price, bond liquidity and the future prospects of the company and its industry. There have been recent examples in the European high yield market where healthy companies have reduced their debt by 15%–25% via a par tender offer.
However, even within the par tender category, there are times when bondholders do not have the option to keep their bonds until maturity and are forced to tender their bonds, thereby awarding the company the opportunity to repurchase the entire bond issue. One very efficient way for companies to secure this opportunity is to combine the tender process with a covenant waiver: In this case the bonds will be stripped of their covenants if a certain threshold of tendered bonds is reached, usually greater than 50% of the issue. Investors will typically tender their entire holdings to avoid ending up with uncovenanted bonds.
An exchange is another means for a company to repurchase an entire issue, and it also puts pressure on investors. Instead of a cash payment, the investor is offered new bonds with longer maturities. This is a relatively low-risk way for the company to extend its debt maturity profile. While investors have the option to sell their securities in the market if they are not interested in the longer-term debt, they would have to accept secondary market prices. Investors will typically base their decisions on the longer-term outlook of the company as well as their individual needs and potential maturity constraints. Again, holding onto the bond until maturity is not an option. The investor may see a capital loss if the new exchanged bond carries a lower coupon than what the company would normally have to pay. Another way in which investors could potentially lose capital is when they are offered the new longer-dated security for 97% (for example) of the par value of the existing bond plus a cash payment for the remaining 3%. However, this cash payment will only be made if the bonds are tendered before a deadline; the cash payment is lost if this deadline is missed. This potential loss is not likely to be captured by prevalent default rate calculation tools.
The second category comprises companies tendering for their bonds far below par. Companies have tendered for their 100-par bonds in the 60s and 50s and even down in the 20s. Again, every case is different, but there are many instances where investors have no choice but to accept the tender and take the loss.
In some of these cases the companies are acting opportunistically, especially those looking to exploit advantages arising out of market dislocation in times of distress. In the recent financial crisis, some companies in the high yield sector launched tender offers for subordinated debt. But tender offers have been used to an even larger extent by banks and other financial institutions, which have tendered for an aggregated notional debt amount to the tune of approximately €110 billion between March 2009 and January 2010. Their objective is to reduce the debt burden and interest rate costs. For financial institutions, the additional motivation for below-par tendering is to book the difference between the tender price and par value as a profit and strengthen core equity for regulatory purposes.
Investors in bonds that are subject to a below-par tender offer usually have the option to hold onto their bonds. However, they are typically enticed by a tender price that is slightly above secondary market levels and by the fact that they are offered liquidity in an otherwise relatively illiquid instrument.
Better the Devil You Know Investors in companies facing operational and financial problems usually have fewer options. As the companies struggle to service their debt, default becomes a likely outcome. However, if there is still some cash on the balance sheet and debt covenants are loose or non-existent (e.g., covenant-lite deals), it could be months or even years before a default occurs. In the meantime, the operational environment could improve. In such cases the companies’ owners, often private equity firms, try to retain ownership. But in order to create some optionality with their investment they need to substantially reduce debt levels, so it is not unusual for these companies to resort to even stronger measures to push investors into distressed exchanges. For example, investors are sometimes offered a lower notional value of new bonds, and in order to encourage them to accept the exchange, the new bonds tend to be higher up in the capital structure, ensuring a higher recovery rate in the event of a default. However, unless the owners inject new cash (equity) into the company, a restructuring via a bond exchange would solely be at the cost of bondholders. A reduction of the notional value (a “haircut”) therefore locks in a permanent loss for the bondholder. In other situations, bondholders may be offered a percentage of equity in the company in exchange for their bonds, transforming the investors from lenders with a fixed claim into profit-sharing co-owners. The details of the exchange or tender largely depend on the level of distress the company is under.
Investors are likely to accept these distressed-scenario tenders and exchanges if they think the alternative is far worse. In most of these situations, the alternative is bankruptcy, where the recovery value for the bondholders will hinge largely on the decision of a judge. Many bondholders would prefer to avoid such uncertainty.
The most effective way for investors to fend off coercive and disadvantageous exchanges is to unite and form a bondholder group. This allows them to combine their votes and reach the necessary majority to block an offer. However, even this method can prove unsuccessful if the company has bought bonds in the open market before officially announcing a tender. This will serve to give the issuer the majority needed to dictate the outcome of the tender.
Even if a company applies a lot of pressure to force bondholders into an exchange to reduce the debt burden, a number of profitable investment opportunities could still arise depending on the situation. However, there is of course the threat that the company could file for bankruptcy if a certain minimum percentage of exchanged bonds are not reached. But this threshold, set by the company, has to be comfortably below 100% of the total bond amount. For example, the fact that some bondholders might not be reachable should be factored in. This can then allow some bondholders to base their decision on game theory. By not exchanging their bonds and avoiding a loss, they hope that a sufficiently high number of holders will tender their bonds and therefore avoid bankruptcy. But in reaching this decision, there needs to be some understanding of the way majority bondholders think and the conviction that the company will continue to operate as a going concern even if the threshold is barely met.
A Less-than-Perfect Measure of Default and Recovery RateIt is critical for high yield investors to know the expected loss rate (determined via default and recovery rates). At present, investors typically glean this knowledge from the actual and projected default rates published by rating agencies and various other sources. In addition, spread levels are used to measure implied default rates, i.e., the probability of default that is currently “priced into” the bond.
However, default rates are now being impacted by an increasing number of tenders and buybacks. With only some exchanges and buybacks counting as defaults, the concern is that current statistics are not reflective of wider market trends. Opportunistic buybacks and exchanges from strong credits are excluded from the standard default rate calculation. But beyond that, the criteria for exclusion becomes less obvious due to the large number of subtle differences in bond exchanges. As Moody’s states in a March 2009 report on distressed exchanges, “…[T]he determination of whether an exchange constitutes a default event is inherently a judgement call.” The rating agencies have developed and published a catalogue of criteria for counting an exchange as a default, but it still leaves room for discretion. One common criterion to assess if a bond exchange should be counted in the default rate is through the bond’s rating, but this again varies by agency. For example, Standard & Poor’s treats a bond exchange as a default if the bond rating is B− or lower. At Moody’s the threshold rating is Caa1. If a rating is slightly higher, a substantial degree of discretion is applied by the agencies. These differences, among others, explains why the European default rate at the end of 2009 was 7.5% at S&P and 10.3% at Moody’s.
Exchanges that are counted as defaults bring the default count forward in time, since the exchange usually takes place earlier then the actual default. On the other hand, not all losses incurred by bondholders are shown in the default rate if certain exchanges are not counted. Moreover, some investors prefer to calculate the default rate on the basis of the defaulted notional bond amount instead of the issuer count. If an exchange or tender is counted as a default, only the bonds which actually get exchanged or tendered enter the “default rate”. This could be significant if the company has multiple bonds outstanding and only some of them will be tendered or exchanged. Therefore, the default rate would be higher if the company was to enter into bankruptcy, since all of the outstanding bonds would be included in the default statistic. The recovery rate, however, could be lower in an exchange than in a default, depending on the bondholder’s crystallized losses.
Such distortions make the projection of default rates even more difficult. Investors should be mindful that actual default rates may not cover all capital losses in a given period and therefore take this into account when relying on the default rate. What’s more, capital losses could be even higher than published numbers indicate during periods of crisis and distress.
This material contains the current opinions of the author but not necessarily those of PIMCO and such opinions are subject to change without notice. This material has been distributed for informational purposes only and should not be considered as investment advice or a recommendation of any particular security, strategy or investment product. Information contained herein has been obtained from sources believed to be reliable, but not guaranteed. No part of this material may be reproduced in any form, or referred to in any other publication, without express written permission.
No part of this material may be reproduced in any form, or referred to in any other publication, without express written permission. Pacific Investment Management Company LLC, 840 Newport Center Drive, Newport Beach, CA 92660, 800-387-4626. ©2013, PIMCO.
Are you sure you would like to leave?
You are currently running an old version of IE, please upgrade for better performance.