Default and Recovery Rates of Canadian Corporate Bond Issuers,

Transcription

1 Special Comment April 2006 Contact Phone New York Sharon Ou David T. Hamilton Toronto Andrew Kriegler Summary Default and Recovery Rates of Canadian Corporate Bond Issuers, This report is Moody s fourth annual review of the default, recovery and credit loss experience of Canadian corporate bond issuers. In this study, we examine the results for 2005 as well as the historical period since The key findings include: There were no Moody's-rated corporate bond defaults in Two unrated bond issuers defaulted on a total of C$86 million of bonds: SR Telecom Inc. (C$75 million) and Unisphere Waste Conversion, Ltd. (C$11 million). The default rate for all rated Canadian issuers dropped to 0% in 2005, down from 0.6% in The default rate for speculative-grade rated Canadian issuers also fell to 0% in 2005 from 1.8% in Credit quality trends were more favorable for Canadian corporate bond issuers in 2005 compared to The upgrade per downgrade ratio rose from 0.26 in 2004 to 0.95 in Five out of 128 issuers were downgraded from investment grade to speculative grade in 2005, pushing the fallen angel rate up to 3.9%, nearly double the historical average rate of 2.1%. Since 1989, 67 rated and unrated Canadian bond issuers have defaulted on C$32.6 billion of bonds. Thirty-four of the 67 defaulting issuers were rated by Moody s with a total of C$28.3 billion of bonds affected. The average default rate for Canadian issuers has been 1.7% over a one-year horizon and 7.1% over a five-year horizon. Multi-year average default rates by rating category for Canadian bond issuers between 1989 and 2005 have been very similar to the U.S. issuers over the same time period. Among the 34 Moody s-rated defaults, credit quality deterioration has historically been reflected in credit rating downgrades well in advance of default. In particular, the median rating of Canadian defaults was B1 at three years prior to default, falling to Caa1 in the year just prior to default. Defaulted bond recovery rates by lien position have been roughly the same for Canadian and U.S.-based defaults. For example, senior unsecured bond recovery rate for Canadian issuers has averaged 36.5% of par, compared to 37.5% in the U.S. Default Counts Annual Canadian Corporate Bond Default Counts and Volume Unrated Counts Rated Counts Default Volume (Bil C$) Default Volume (Bil C$)

2 Table of Contents Page Introduction... 3 The Year 2005 in Review... 3 Default Counts and Volume lowest since Default rates drop to zero... 4 Aggregate credit quality improves, but fallen angels rise Outlook... 6 Historical Experience on Credit Losses... 6 Multi-year default rates... 6 Moody's ratings as predictors of default... 7 Recovery rates... 8 Credit loss rates... 9 References... 9 Appendix A Moody's definition of default Estimated senior ratings Default rate calculations Recovery rates Appendix B Related Research Moody s Special Comment

3 Introduction The year 2005 turned out to be a strong year for corporate credit quality in the Canadian bond market. Default activity slowed considerably. There were no Moody's-rated defaults in 2005, and only two small, unrated issuers defaulted on a total of C$86 million of bonds. In terms of total default volume (for both rated and unrated bonds), 2005 marked the lowest default volume since From a broader credit quality prospective, Canadian bond issuer credit quality improved in 2005: the upgrade per downgrade ratio rose from 0.26 in 2004 to 0.95 in Looking into 2006, however, we expect default rates of Canadian issuers to gradually pick up along with the global trend. In the first section of this report we review the defaults and credit trends in Section two presents Moody's default outlook for 2006 and beyond. The final section documents the historical default and recovery experience of Moody's-rated Canadian-domiciled corporate bond issuers over the time period. 1 Throughout the report, we compare and contrast the credit rating performance of Canadian corporate bond issuers with U.S.-domiciled issuers. The U.S. is a natural benchmark against which to measure the performance of Moody's Canadian ratings. Although Moody s has been rating Canadian corporate bond issuers since 1901, our analysis begins in 1989, when Canadian corporate debt issuance and the number of issuers obtaining credit ratings began to grow at an accelerated rate. For the purposes of this study, a corporate issuer is considered to be domiciled in Canada if it is legally incorporated or has the majority of its operations and assets in Canada. Debt may have been placed in multiple markets and in currencies other than Canadian dollars. In addition to bankruptcy under the Companies' Creditors Arrangement Act and the Bankruptcy and Insolvency Act, events of default may have occurred under foreign legal frameworks (such as U.S. Chapter 11). Appendix A contains additional detail on the data sources and methodologies we used in writing this report. The Year 2005 in Review DEFAULT COUNTS AND VOLUME LOWEST SINCE 1997 In 2005, only two corporate issuers defaulted on a total of C$86 million of bonds, neither of which was rated by Moody s 2. The larger defaulter is SR Telecom, Inc., which missed the principal and interest payment on C$75 million debenture matured in April. The smaller one is Unisphere Waste Conversion, Ltd. that failed to make the interest payment on C$11 million bonds in January. While the total number of rated and unrated defaulters remained unchanged from 2004 (2 issuers), default volume declined sharply for the third consecutive year, falling from C$400 million in 2004 to C$86 million in s total default volume (for rated and unrated bonds) marks the lowest level since Exhibit 1 shows the annual default counts and volume since Exhibit 1 Annual Canadian Default Counts and Volume, Counts Volume (millions C$) Year Rated Unrated Total Rated Unrated Total , , , , , , , , , , Total ,309 4,310 32, An analysis of historical rating migration trends for Canadian bond issuers can be found in Hamilton and Ou (2005) and Kryzanowski and Menard (2001). 2. Two Canadian affiliates of U.S.-based firms also defaulted in 2005: Calpine Canada Energy Finance ULC and Calpine Canada Energy Finance II ULC. They are two pure financial arms of Calpine Corporation, therefore are not considered independent corporate bond issuers in this study. Moody s Special Comment 3

4 As shown in Exhibit 1, 67 Canadian corporate issuers defaulted on a total of C$ 32.6 billion of bonds between 1989 and Thirty-four of the 67 defaults were rated by Moody s, representing C$ 28.3 billion of bonds. Similar to the global trend, default activity peeked during the period between 2000 and AT&T Canada, which filed for bankruptcy in 2002 with approximately C$4.6 billion of bonds, was the largest default during the period. A detailed list of public bond defaults is included in Exhibit 11 in Appendix B. Among the 67 total defaulters, 42% were by issuers in the industrial sector. Within this sector, defaults were concentrated in traditionally important Canadian industries: metals and mining experienced 11 defaults, and forest products/paper experienced 7 defaults. Measured by default volume, however, the telecommunications and technology sector topped the others by contributing nearly half of the default total. The industry sector distribution of default counts and volumes by broad industry sector is presented in Exhibit 12 in Appendix B. DEFAULT RATES DROP TO ZERO As there were no Moody s-rated defaults in 2005, the default rates for all Canadian corporate bond issuers dropped to 0% last year. The last recorded Moody's-rated default was by Hollinger, Inc. in March In 2004, with 1 rated default, the default rate for all Canadian corporate bond issuers stood at 0.6%. Among speculative-grade rated issuers, the default rate also fell to 0% in 2005, down from 1.8% in The last time Canadian default rates dropped to 0% was in Exhibit 2 presents annual default rates for Moody's-rated Canadian corporate bond issuers. As shown in the chart, annual default rates for Canadian speculative-grade issuers have declined in recent years after peaking at 12.9% in the year As for investment-grade issuers, the Canadian default rates were 0% for the entire period except for the years 2000 and The highest Canadian investment-grade default rate occurred in 2002 when AT&T Canada and Teleglobe defaulted. These two companies, together with Laidlaw 4, are the only three Canadian investment grade defaults since Exhibit 2 Annual Default Rates for Canadian Corporate Bond Issuers, % Speculative-Grade Investment-Grade 40% 30% 20% 10% 0% AGGREGATE CREDIT QUALITY IMPROVES, BUT FALLEN ANGELS RISE Along with low default counts and volume, credit quality among Canadian-based bond issuers also showed strengthening as measured by the ratio of rating upgrades to rating downgrades. Exhibit 3 shows the annual upgrade-downgrade ratio for Canadian corporate bond issuers between 1989 and Our estimate of rating upgrade-downgrade ratio is simply the number of issuers downgraded divided by issuers upgraded in a year 5. A ratio greater than 1 indicates that rating upgrades outnumbered rating downgrades. In 2005, the upgrade-downgrade ratio arrived at 0.95, which means 3. The unusually high speculative-grade default rates in 1989 and 1990 are the result of the very small number of rated issuers outstanding in those years. 4. Laidlaw defaulted in May The upgrade-downgrade ratio measures the direction but not the magnitude of changes in aggregate credit quality. 4 Moody s Special Comment

5 the number of rating upgrades is matched nearly one for one by rating downgrades. The year 2005 s upgrade-downgrade ratio is up noticeably from 0.26 in 2004 and just surpasses the historical average of 0.92 from 1989 to Exhibit 3 Annual Upgrade-Downgrade Ratio for Canadian Corporate Bond Issuers, Annual Ratio Average Ratio ( ) Although aggregate credit quality - as measured by the upgrade-downgrade ratio - showed an improvement, the number of fallen angels increased in Exhibit 4 presents the number of fallen angels in each year from 1989 to The chart shows that fallen angels are relatively rare in the Canadian market, with three or less per year. In 2005, however, 5 out of 128 Canadian issuers lost their investment grade ratings, marking the second highest number of fallen angels after Moody s research has shown that fallen angels generally have greater default risk during the two years after falling to speculative grade compared to firms that had never been investment grade but held the same ratings as the fallen angels on the dates that they fell 6. Exhibit 4 Number of Canadian Fallen Angels, See Mann (2003). Moody s Special Comment 5

6 Exhibit 5 lists the 5 Canadian bond issuers who lost their investment grade ratings in All of these five companies have experienced multi-notch downgrades when their ratings were cut from investment grade. Additionally, three of them received further downgrades after entering the speculative-grade zone. Among these five fallen angels, General Motors Acceptance Corp of Canada Ltd. and General Motors Nova Scotia Finance Company are affiliates of General Motor Corporation, the world s largest auto maker. The Canadian unit of Ford Motor Company - Ford Credit Canada Limited - became a fallen angel in January 2006 when its rating was lowed from Baa3 to Ba2. Exhibit 5 List of Fallen Angels in 2005 Company Downgrade date From To As of 3/31/06 Domtar Inc. 6/20/2005 Baa3 Ba2 B1 General Motors Acceptance Corp. of Canada Ltd 8/24/2005 Baa2 Ba1 Ba1 Sears Canada Inc. 11/16/2005 Baa2 Ba2 Ba2 Quebecor World Capital Corporation 8/26/2005 Baa3 Ba2 Ba3 General Motors Nova Scotia Finance Company 8/24/2005 Baa3 Ba2 B Outlook Steady economic growth, relatively low interest rates, and healthy credit quality set the stage for lower default rates in Canada in Looking ahead, however, Moody's expects that the default rate for Canadian issuers may pick up gradually in tandem with the global rate. Moody's global speculative-grade default rate forecasting model predicts a 3.1% default rate by the end of February Although Moody's does not forecast Canadian default rates separately, we expect that many of the factors driving our prediction of higher global default rates including rising interest rates, shrinking liquidity, and lower average credit quality among speculative-grade issuers will also impact the Canadian default rate over In addition, as discussed in the previous section, the rising number of fallen angels also implies potential default risk for certain Canadian issuers. Historical Experience on Credit Losses MULTI-YEAR DEFAULT RATES While annual default rates are helpful for measuring trends in aggregate credit quality, multi-year default rates are more relevant for investors with longer investment horizons. Exhibit 6 presents multi-year, weighted average cumulative default rates for both Canada and the U.S. by whole letter rating category. Each row of the table shows the cumulative default rate for a holding period of up to five years. For example, a portfolio of Canadian bond issuers rated Ba held for five years has experienced a default rate of 8.2%. Exhibit 6 - Average Cumulative Default Rates for Canadian and U.S. Corporate Bond Issuers, * Canadian Issuers United States Issuers Year 1 Year 2 Year 3 Year 4 Year 5 Year 1 Year 2 Year 3 Year 4 Year 5 Aaa 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Aa 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% A 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.2% 0.3% 0.4% Baa 0.5% 1.3% 2.2% 2.6% 3.1% 0.2% 0.6% 1.0% 1.5% 2.0% Ba 0.6% 1.6% 3.5% 6.0% 8.2% 1.3% 3.6% 5.8% 7.8% 9.6% B 3.2% 8.3% 12.5% 16.6% 20.9% 5.7% 12.6% 18.8% 24.2% 29.2% Caa-C 28.4% 45.2% 59.3% 71.4% 76.5% 19.7% 31.8% 41.3% 49.5% 56.4% Investment-Grade 0.2% 0.4% 0.7% 0.8% 1.0% 0.1% 0.3% 0.4% 0.7% 0.9% Speculative-Grade 5.0% 9.8% 14.2% 18.6% 22.0% 5.5% 11.0% 15.9% 20.0% 23.5% All Corporates 1.7% 3.3% 4.8% 6.1% 7.1% 2.3% 4.5% 6.4% 7.9% 9.2% * Calculated using monthly cohorts The chart shows that the aggregate cumulative default rates for Canadian corporate issuers are generally similar to those of the U.S. over all investment horizons. For example, the average default rate for Canadian speculative-grade issuers increases from 5.0% at a one-year holding period to 22.0% at a five-year holding period. In the U.S., the one- 6 Moody s Special Comment

7 and five-year default rates are 5.5% and 23.5%, respectively. A visible difference is the relatively high default rates for Canadian Caa-C rated issuers. Although the relatively high default rate for Canadian issuers is no doubt due to attrition in the denominator (which removes defaults from prior years and rating withdrawals, see Appendix A), it also reflects the fact that Canadian issuers who subsequently defaulted were identified as high credit risks and were assigned correspondingly low credit ratings. As for investment grade credits, no issuers rated single A or above have ever defaulted within a year of holding those ratings in Canada. Although the Baa default rates for Canadian issuers across all time horizons are somewhat higher than those for the U.S, it should be noted that the small sample size for Canadian Baa-rated issuers amplifies the default rate estimates. MOODY'S RATINGS AS PREDICTORS OF DEFAULT The default rates shown above reveal a strong correlation between Moody's ratings and the probability of default over multiple investment horizons, even given the relatively short time period and small data set of this study. For the average defaulter, credit quality deterioration is reflected in ratings downgrades that occur well in advance of the time of default. Exhibit 7 plots the median and average senior unsecured rating for defaulted Canadian and US issuers up to three years before default. Exhibit 7 Median and Average Ratings Prior to Default, Ba3 B1 B2 B3 Caa1 Caa2 US Average Canada Average US Median ` Canada Median Caa Months prior to default The chart reveals that corporate issuers that subsequently defaulted held low speculative-grade ratings three years before default occurred. The median and average rating has been near B1 for both Canadian and the U.S. defaulters three years prior to default. As the default date approaches, Canadian defaulters were downgraded earlier, so that one year prior to default, the median rating for Canadian defaulters was Caa1 compared to B2 for US defaulters. This pattern is largely an artifact of the time period available for this study, during which aggregate credit quality deteriorated sharply, and the distribution of industry sectors rather than a stricter rating standard being applied to Canadian issuers. In addition to identifying issuers that ultimately defaulted with low credit ratings, Moody's rating system has also demonstrated its ability to distinguish issuers that do not default with high credit ratings. Exhibit 8 shows the Cumulative Accuracy Profile (CAP) for a one year time horizon. The CAP curve in Exhibit 8 sorts the Moody's-rated pool of issuers into risk percentile (the horizontal axis). The percentage of defaulters captured in each risk percentile is shown on the vertical axis. This CAP curve can be described as a "power curve," since it shows how effective a risk scoring system (rating system) is at detecting defaults from the population. The further the curve bows toward the northwest corner, the greater the fraction of all defaults that carry low ratings and the more efficient ratings are at sorting defaults from non-defaults. In Exhibit 8, we see that both the Canadian curve and the U.S. curve bow significantly toward the northwest corner, indicating that Moody's ratings are powerful signals of default risk. The chart further shows that Canadian ratings are slightly more powerful in default prediction as 75% of Canadian defaulters were captured in the 10% riskiest population; while for U.S. defaulters, roughly 60% of defaults were identified in the same decile. Moody s Special Comment 7

8 Exhibit 8 One Year Cumulative Accuracy Plots (CAP), Cumulative Proportion of Defaults 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% C B3 Caa B2 Caa2 Caa3 Ca B1 Ba Ba2 Ba1 Baa3 Baa2 Baa1 A3 A2 A1 Aa3 Aa Aa1 Aaa 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% US Canada Cumulative Proportion of Rated Issuers RECOVERY RATES Moody's credit ratings are statements of expected credit loss, which is a function of default probability and recovery in the event of default. Therefore, empirical estimates of recovery values for bonds in default are just as important as the calculation of default rates. Exhibit 9 presents descriptive statistics for both issuer- and issue-weighted defaulted bond recovery rates for Canadian and U.S. defaulted bonds during 1989 to Moody's methodology for computing issuer-weighted averages is to take the mean of recovery rates for a security class for each issuer before calculating the average across all issuers. The issue-weighted average calculation simply averages across every defaulted bond of all firms by security class. Recovery prices in Exhibit 9 are measured by trading prices in the secondary market one month after default 7. The table shows that recovery rates for Canadian defaulted bonds are correlated with their priority in the capital structure. Bondholders with a higher priority of claim enjoy a higher average rate of recovery in default. Exhibit 9 further shows that average recovery rates for Canadian defaulted bonds have been roughly the same as those in the U.S. For example, the issuer-weighted mean recovery rates of Canadian senior unsecured bonds are 36.5% compared to 37.5% for U.S. issuers. On an issue-weighted basis, however, recovery rates of Canadian bond issuers appear to be slightly lower than those for those of U.S.-based issuers. Exhibit 9 Average Recovery Rates for Canadian and U.S. Defaulted Bonds, * Issuer-Weighted Issue-Weighted Canada United States Canada United States Priority All Bonds Exclude Telecom All Bonds Exclude Telecom All Bonds Exclude Telecom All Bonds Exclude Telecom Sr. Secured $54.0 $51.3 $ $54.0 $51.3 $ Sr. Unsecured $36.5 $39.2 $ $31.7 $39.6 $ Sr. Sub. $29.5 $29.5 $ $29.5 $29.5 $ Sub. $13.6 $13.6 $ $12.4 $12.4 $ Jr. Sub. n.a. n.a. $ n.a. n.a. $ *Excludes discount bonds A final note about recovery rates is that the significant proportion of telecommunications defaults in the past couple of years has strongly influenced the average recovery rates. In Exhibit 9, we find that the average issuer-weighted recovery rate for senior unsecured bonds in Canada rises to 39.2% if telecommunications defaults are removed. For U.S. unsecured bonds, the average issuer-weighted recovery rate rises to 40.0% excluding telecommunications defaults. 7. See more discussion on recovery rates measurement in Appendix B 8 Moody s Special Comment

9 CREDIT LOSS RATES Based on the default rates and recovery rates that were discussed above, we can calculate credit loss rates which give the loss of total return of a bond portfolio due to defaults. Mathematically, credit loss rates can be defined as the product of default frequency and loss severity (i.e., 1-recovery rate). Exhibit 10 presents the average one- year senior unsecured credit loss rates for Canadian and U.S. issuers. The chart demonstrates that for both Canadian and U.S. issuers, Moody s ratings are negatively correlated to credit loss rates. As there have been no Canadian defaults in the Aaa to A rating categories, the one- year loss rates for Canadian issuers holding these ratings are naturally zero. One-year credit loss rates for Baa-and Caa-C-rated issuers have been higher in Canada than in the U.S., while the opposite is true for Ba-and B-rated issuers. Despite these differences, credit loss rates are, for the most part, consistent between Canada and the U.S. Indeed, given the high variance in the underlying default and recovery rates, we can conclude that average senior unsecured credit loss rates are not statistically different between Canada and the U.S. Exhibit 10 Average One-Year Credit Loss Rates for Canadian and U.S. Issuers, % Canada US 18.1% 15.0% 12.3% 10.0% 5.0% 0.0% 3.6% 2.0% 0.0% 0.0% 0.0% 0.3% 0.1% 0.4% 0.8% Aaa Aa A Baa Ba B Caa-C References Cutler and Ederer (1958), Maximum Utilization of the Life Table Method in Analyzing Survival, Biostatistics, December. Hamilton and Ou (2005), Default and Recovery Rates of Canadian Corporate Bond Issuers, Moody s Global Credit Research, April. Hamilton (2005), Moody s Senior Ratings Algorithm & Estimated Senior Ratings, July. Kryzanowski and Menard (2001), Migration Behavior of Long-Term Bond Ratings of Canadian Corporate Bond Issuers, Canadian Investment Review, Fall Mann (2003), What Happens To Fallen Angels? A Statistical Review , Moody s Glogal Credit Research, July. Moody s Special Comment 9

10 Appendix A MOODY'S DEFINITION OF DEFAULT Moody s definition of default includes three types of credit events: A missed or delayed disbursement of interest and/or principal, including delayed payments made within a grace period; Bankruptcy, administration, legal receivership, or other legal blocks (perhaps by regulators) to the timely payment of interest and/or principal; or A distressed exchange occurs where: (i) the issuer offers bondholders a new security or package of securities that amount to a diminished financial obligation (such as preferred or common stock, or debt with a lower coupon or par amount, lower seniority, or longer maturity); or (ii) the exchange had the apparent purpose of helping the borrower avoid default. The definition of a default is intended to capture events that change the relationship between the bondholder and bond issuer from the relationship which was originally contracted, and which subjects the bondholder to an economic loss. Technical defaults (covenant violations, etc.) are not included in Moody s definition of default. ESTIMATED SENIOR RATINGS The credit ratings used in Moody's annual default studies do not refer to the ratings of any specific debt obligations. Rather, they are notional, issuer-level ratings derived from each issuer's outstanding rated bonds. The resulting ratings are called estimated senior unsecured ratings or, more concisely, estimated senior ratings. Estimated senior ratings are derived by a statistical algorithm: first, the rating on each company's senior unsecured bonds is used; if an issuer does not have any rated senior unsecured bonds, the estimated senior rating is derived by statistically implying such a rating on the basis of rated subordinated or secured bonds, or senior implied ratings. 8 Deriving estimated senior ratings is necessary because Moody's credit ratings for corporate bonds are opinions of expected credit loss, which is comprised of (1) the expected probability of default, as well as (2) the expected severity (loss of principal and interest) in the event of default. Therefore, in order to calculate historical default rates by rating category, which are estimates of the expected default probability component, one must hold severity considerations constant. In most cases, estimated senior ratings will yield an assessment of default risk that is relatively unaffected by special considerations of collateral or of position within the capital structure. DEFAULT RATE CALCULATIONS Moody s employs the standard method used by statisticians to analyze cumulative survival (and hence default) probabilities using discrete time data. 9 In calculating default rates for individual cohorts, Moody s adjusts the denominator of the default rate downward for issuers whose ratings are withdrawn to account for survivorship bias. Once an issuer s rating is withdrawn, the observation of its future default experience is treated as a censored observation. 10 By way of example, if while measuring a 10-year CDR an issuer's rating was withdrawn in year 5, certainty regarding default history can only be gained for the period of time between the time the cohort was formed and year 5. The issuer would therefore be considered "exposed" to default for 5 years, not 10 years. Cumulative default rates (CDRs) are calculated from marginal default rates (MDRs), which represent the probability that an issuer that has survived in the sample through a particular date will default over a short interval following that date. By definition, the denominators used to estimate MDRs must adjust for withdrawals and defaults prior to that date. We use the following formal for calculating MDRs, which also makes a very minor adjustment (and inconsequential) adjustment for the withdrawals that occur during the short interval in question. The marginal default rate, d(t), is the ratio of the number of defaulting issuers in period t divided by the number of issuers exposed to the risk of default in period t: d t = n t x xt 1 [ ] 2 wt wt 1 t See Hamilton (2005) for details. 9. The classic reference is Cutler and Ederer (1958). This approach is the discrete time analogue of the continuous hazard rate model which is familiar to most economists. 10. Moody s does attempt to follow these companies subsequent experience and occasionally becomes aware of instances that issuers with withdrawn ratings default. It is not possible, however, to use this information in an unbiased way because we do not know what happens to all withdrawn issuers. Moreover, issuers that have their ratings withdrawn because they no longer have any debt outstanding and are liquidated cannot be used to gain insight to the question: what would happen if I have credit exposure to an issuer beyond that date of liquidation? 10 Moody s Special Comment

11 Cumulative default rates, D(T), for horizons of length T, for specific cohorts are built up from the MDRs as follows: T (1 d1) d 2 + (1 d1)(1 d 2 ) d (1 ) t 1 D T d ( ) = = For example, the two year cumulative default rate is the sum of the marginal default rate in year one (d 1 ) and the marginal default rate in year two (d 2 ) for issuers that survived past year one, (1 d 1 ). When calculating average CDRs over many cohorts, the above formula is also used, but the MDRs are constructing by averaging across cohorts. In this manner, when estimating long-term CDRs, information from recent cohorts need not be ignored entirely. Rather, near-term cohorts are used to obtain more precise estimates of the near-term marginal default rates used in deriving long-term CDRs. Our default rate calculation method has two important advantages over the more simplistic so-called static pool method used by some researchers: If you expect to have a ten-year exposure to an issuer, you want an estimate of the risk of ten years of exposure. Moody's method produces an unbiased estimate of this risk, whereas the static pool method produces downwardly biased default rates estimates since it implicitly assumes that issuers that leave the sample would never have defaulted. Moody's method uses all the available information about default/survival in the dataset to estimate long-term CDRs, including the information from recent cohorts, not just cohorts that have history that span the entire horizon of the CDR calculation. Moody's method uses such information in estimating short-term MDRs that enter into the long-term CDR calculations, whereas the static approach simply ignores that data. 11 RECOVERY RATES Moody's generally measures recovery rates in this report using bid prices on defaulted debt obligations observed roughly 30 days after the date of default, specifically measured as a ratio of prices relative to par. Using a post-default price to measure loss severity parallels common practice in the credit default swaps market. Moreover, recovery rates measured in this way are most relevant for the many cash bond investors who liquidate their holdings shortly after default as required by their portfolio governance rules or their own investment objectives. For investors holding defaulted securities until ultimate resolution, prices observed shortly after default are generally accepted as the market's discounted expected ultimate recovery rates. The alternative approach measuring ultimate recoveries realized in bankruptcy resolutions presents a number of estimation challenges including estimation of the discount rate to apply to each cash-flow, assumptions about the value of certain payments made during the process, and estimates of values of securities used to pay creditor claims, which may include illiquid new securities (e.g., equity and derivative instruments) as well as physical assets. The table below defines the various ways Moody's sometimes calculates and presents its recovery rate estimates. For the purposes of measuring expected loss rates, we rely on issuer-weighted mean recovery rates. d t d T Term Issuer-Weighted Mean Recovery Rates Value-Weighted Mean Recovery Rates Definition Issuer-weighted mean recovery rates are derived by estimating mean recovery rates for each issuer, then averaging them across issuers. They are useful for predicting recovery rates for portfolios that are well diversified across issuers. Value-weighted recovery rates represent the average of recovery rates on all defaulted issuers, weighted by the face value of those issues. These estimates are useful for predicting recovery rates on the market portfolio. 11. For example, Moody's 10-year cumulative default rate is not calculated only using issuers with 10 years of history available. Issuers with one year of history contribute to the first year marginal default rate; issuers with two years of history contribute (up to) the second year marginal default rate, etc. Moody s Special Comment 11

12 Appendix B Defaulted Company Exhibit 11 Chronological List of Canadian Bond Defaults (Rated and Unrated) Default Year Default Event Default Amount Rated? * 360networks, Inc Missed interest payment C$ 2,204 Yes Ainsworth Lumber Company Limited 2001 Grace period default C$ 286 Yes Air Canada 2003 Bankruptcy C$ 1,872 Yes Algoma Steel, Inc Bankruptcy C$ 533 Yes American Eagle Petroleum Ltd Missed interest payment C$ 13 No American Eco Corporation 2000 Missed interest payment C$ 175 Yes Anvil Range Mining Corporation 1998 Bankruptcy C$ 30 No AT&T Canada, Inc Missed interest payment C$ 4,663 Yes BCE Development Corporation 1990 Missed interest payment C$ 126 No Beatrice Foods Inc Missed interest payment C$ 206 Yes Bramalea Limited 1992 Missed interest payment C$ 511 No Call-Net Enterprises, Inc Distressed exchange C$ 3,434 Yes Canadian Airlines Corporation 2000 Missed interest payment C$ 408 Yes Canlan Ice Sports Corporation 2000 Missed interest payment C$ 20 No Confederation Life Insurance Company 1994 Seized by regulators C$ 235 No Confederation Treasury Services Ltd Seized by regulators C$ 287 No Consumers International, Inc Missed interest payment C$ 216 Yes Consumers Packaging, Inc Missed interest payment C$ 95 Yes Consumers Packaging, Inc Missed interest payment C$ 63 No Crown Packaging Enterprises Ltd Bankruptcy C$ 117 Yes Crown Packaging Holdings Limited 2000 Bankruptcy C$ 175 Yes Crown Packaging Limited 2000 Missed principal payment C$ 114 Yes Cuddy International Corporation 2000 Missed interest payment C$ 112 Yes Doman Industries Limited 2002 Missed interest payment C$ 1,070 Yes Financial Trustco Capital Ltd Missed interest payment C$ 130 Yes Galactic Resources, Limited 1992 Missed interest payment C$ 46 No General Trustco of Canada 1993 Missed principal and interest payments $56.20 No Greenstone Resources Ltd Suspension of payments C$ 100 No GT Group Telecom Inc Bankruptcy C$ 1,336 Yes Hollinger, Inc Grace period default C$ 161 Yes HQ Office Supplies Warehouse Inc Missed interest payment C$ 13 No Hurricane Hydrocarbons Ltd Missed interest payment C$ 263 Yes International Utility Structures Inc Missed interest payment C$ 84 No Laidlaw, Inc Missed interest payment C$ 2,944 Yes Lavalin Industries 1991 Missed interest payment C$ 45 No Livent Inc Chapter 11 C$ 180 Yes Loewen Group International, Inc Bankruptcy C$ 1,341 yes Loewen Group, Inc Bankruptcy C$ 200 no Maxwell Communications Finance Canada Ltd Chapter 11 C$ 102 yes McWatters Mining, Inc Bankruptcy C$ 24 no Microcell Telecommunications Inc Missed interest payment C$ 1,497 yes Nevada Bob's Golf, Inc Bankruptcy C$ 7 no Ntex, Inc Grace period default C$ 114 yes OCS Technologies Corp Missed interest payment C$ 18 no PCI Chemicals Canada, Inc Missed interest payment C$ 266 yes Pegasus Gold Inc Chapter 11 C$ 165 no * 12 months prior to default 12 Moody s Special Comment

13 Exhibit 11 Chronological List of Canadian Bond Defaults (Rated and Unrated) (Continued) Defaulted Company Default Year Default Event Default Amount Peoples Jewellers Limited 1992 Bankruptcy C$ 23 no Philip Services Corp Chapter 11 C$ 393 no Philip Services Corp Suspension of payments C$ 40 yes PWA Corporation 1992 Missed interest payment C$ 96 no Rea Gold Corp Bankruptcy C$ 13 no Redekop Properties, Inc Missed interest payment C$ 10 no Royal Oak Mines Inc Distressed exchange C$ 252 yes Saskatchewan Wheat Pool 2003 Missed interest payment C$ 300 no Semi-Tech Corporation 1999 Chapter 11 C$ 975 yes Southland Canada Inc Chapter 11 C$ 50 yes SR Telecom Inc Missed principal and interest payments C$ 75 no Stelco, Inc Bankruptcy C$ 240 no Tee-Comm Electronics, Inc Receivership C$ 100 no Teleglobe, Inc 2002 Bankruptcy C$ 1,913 yes Telesystem International Wireless, Inc Distressed exchange C$ 856 yes Trizec Corporation Ltd 1993 Missed principal and interest payments C$ 744 no Uniforet, Inc Missed interest payment C$ 210 yes Unisphere Waste Conversion, Ltd Missed interest payment C$ 11 no Vicwest Corporation 2002 Missed interest payment C$ 131 no William Resources Inc Missed interest payment C$ 98 no Woodward's Ltd Bankruptcy C$ 33 no Rated? Exhibit 12 - Distribution of Canadian Bond Defaults by Broad Industry Category, Industry Counts Count % Volume* Volume % Banking 2 3.0% % Consumer Products 4 6.0% 1, % Energy 2 3.0% % Financial (Non-Bank) % 2, % Hotel, Gaming, & Leisure 2 3.0% % Industrial % 5, % Media 1 1.5% % Miscellaneous 2 3.0% 1, % Retail 4 6.0% % Telecom & Technology % 16, % Transportation 4 6.0% 5, % Total % 32, % * In millions of Canadian dollars. Moody s Special Comment 13

14 Canada Exhibit 13 - Annual Default Rates for Canadian and U.S. Corporate Bond Issuers, Year Aaa Aa A Baa Ba B Caa-C Investment- Grade Speculative- Grade All Corporates % 0.0% 0.0% 0.0% 0.0% 40.0% n.a. 0.0% 33.3% 2.2% % 0.0% 0.0% 0.0% 0.0% 100.0% n.a. 0.0% 50.0% 2.0% % 0.0% 0.0% 0.0% 0.0% n.a. n.a. 0.0% 0.0% 0.0% % 0.0% 0.0% 0.0% 0.0% 0.0% n.a. 0.0% 0.0% 0.0% % 0.0% 0.0% 0.0% 0.0% n.a. n.a. 0.0% 0.0% 0.0% % 0.0% 0.0% 0.0% 0.0% 0.0% n.a. 0.0% 0.0% 0.0% % 0.0% 0.0% 0.0% 12.5% 0.0% 0.0% 0.0% 3.7% 1.2% % 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% % 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% % 0.0% 0.0% 0.0% 0.0% 5.9% 0.0% 0.0% 3.4% 1.3% % 0.0% 0.0% 0.0% 4.7% 0.0% 17.4% 0.0% 4.5% 1.8% % 0.0% 0.0% 2.4% 0.0% 3.0% 53.3% 0.9% 12.9% 5.7% % 0.0% 0.0% 0.0% 0.0% 6.9% 44.4% 0.0% 10.0% 3.5% % 0.0% 0.0% 4.1% 0.0% 0.0% 53.3% 1.6% 7.9% 3.4% % 0.0% 0.0% 0.0% 0.0% 4.4% 0.0% 0.0% 1.8% 0.6% % 0.0% 0.0% 0.0% 0.0% 0.0% 20.0% 0.0% 1.8% 0.6% % 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% United States Year Aaa Aa A Baa Ba B Caa-C Investment- Grade Speculative- Grade Exhibit Letter Rating Migration Rates for Canadian Issuers All Corporates % 0.0% 0.0% 0.6% 2.8% 8.5% 25.0% 0.2% 5.5% 2.4% % 0.0% 0.0% 0.0% 3.2% 16.1% 58.8% 0.0% 10.0% 4.2% % 0.0% 0.0% 0.3% 5.4% 14.5% 36.8% 0.1% 10.4% 4.0% % 0.0% 0.0% 0.0% 0.3% 9.3% 26.7% 0.0% 5.1% 1.7% % 0.0% 0.0% 0.0% 0.6% 5.9% 28.6% 0.0% 3.8% 1.3% % 0.0% 0.0% 0.0% 0.3% 4.1% 2.7% 0.0% 2.1% 0.7% % 0.0% 0.0% 0.0% 0.6% 5.1% 12.8% 0.0% 3.6% 1.4% % 0.0% 0.0% 0.0% 0.0% 1.5% 14.8% 0.0% 1.9% 0.7% % 0.0% 0.0% 0.0% 0.2% 2.4% 12.0% 0.0% 2.2% 0.9% % 0.0% 0.0% 0.0% 0.5% 4.5% 16.7% 0.0% 3.9% 1.6% % 0.0% 0.0% 0.2% 1.0% 6.1% 20.5% 0.1% 5.9% 2.6% % 0.0% 0.0% 0.4% 1.1% 6.6% 19.4% 0.2% 7.0% 3.2% % 0.0% 0.3% 0.3% 1.6% 8.9% 33.7% 0.2% 11.0% 4.8% % 0.0% 0.3% 1.1% 1.2% 4.9% 23.5% 0.6% 7.2% 3.2% % 0.0% 0.0% 0.0% 1.2% 2.3% 20.3% 0.0% 5.4% 2.2% % 0.0% 0.0% 0.0% 0.3% 0.8% 13.3% 0.0% 2.9% 1.2% % 0.0% 0.0% 0.3% 0.0% 1.4% 8.1% 0.1% 2.3% 1.0% From \ To Aaa Aa A Baa Ba B Caa-C D WR Aaa 100.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Aa 0.0% 92.6% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 7.4% A 0.0% 2.3% 86.0% 4.7% 0.0% 0.0% 0.0% 0.0% 7.0% Baa 0.0% 0.0% 3.7% 79.6% 3.7% 3.7% 0.0% 0.0% 9.3% Ba 0.0% 0.0% 0.0% 0.0% 73.3% 26.7% 0.0% 0.0% 0.0% B 0.0% 0.0% 0.0% 0.0% 7.4% 74.1% 3.7% 0.0% 14.8% Caa-C 0.0% 0.0% 0.0% 0.0% 0.0% 60.0% 20.0% 0.0% 20.0% 14 Moody s Special Comment

15 Related Research Special Comments: Default and Recovery Rates of Canadian Corporate Bond Issuers, , May 2005 (92050) Default and Recovery Rates of Corporate Bond Issuers, , February 2006 (96546) The Performance of Moody's Corporate Bond Ratings: December 2005 Quarterly Update, January 2006 (96330) To access any of these reports, click on the entry above. Note that these references are current as of the date of publication of this report and that more recent reports may be available. All research may not be available to all clients. Moody s Special Comment 15

16 To order reprints of this report (100 copies minimum), please call Report Number: Author Sharon Ou David Hamilton Senior Production Associate Mark A. Lee Copyright 2006, Moody s Investors Service, Inc. and/or its licensors and affiliates including Moody s Assurance Company, Inc. (together, "MOODY S"). All rights reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODY S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY S from sources believed by it to be accurate and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided as is without warranty of any kind and MOODY S, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness for any particular purpose of any such information. Under no circumstances shall MOODY S have any liability to any person or entity for (a) any loss or damage in whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of MOODY S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without limitation, lost profits), even if MOODY S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information. The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider purchasing, holding or selling. MOODY S hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and preferred stock rated by MOODY S have, prior to assignment of any rating, agreed to pay to MOODY S for appraisal and rating services rendered by it fees ranging from $1,500 to $2,400,000. Moody s Corporation (MCO) and its wholly-owned credit rating agency subsidiary, Moody s Investors Service (MIS), also maintain policies and procedures to address the independence of MIS s ratings and rating processes. Information regarding certain affiliations that may exist between directors of MCO and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in MCO of more than 5%, is posted annually on Moody s website at under the heading Shareholder Relations Corporate Governance Director and Shareholder Affiliation Policy. Moody s Investors Service Pty Limited does not hold an Australian financial services licence under the Corporations Act. This credit rating opinion has been prepared without taking into account any of your objectives, financial situation or needs. You should, before acting on the opinion, consider the appropriateness of the opinion having regard to your own objectives, financial situation and needs. 16 Moody s Special Comment