New York University Salomon Center Leonard N. Stern School of Business. Special Report On

Transcription

1 New York University Salomon Center Leonard N. Stern School of Business Special Report On Defaults and Returns in the High-Yield Bond and Distressed Debt Market: The Year 2011 in Review and Outlook By Edward I. Altman And Brenda J. Kuehne February 03, 2012

2 Contents Executive Summary... 3 Defaults, Default Rates, and Recoveries... 4 Bankruptcies... 8 Industry Defaults Age of Defaults Fallen Angel Defaults Default Losses and Recoveries Distressed Exchanges Subsequent Performance of Distressed Exchange Companies Forecast Recovery Versus Actual Related Recovery Statistics Mortality Rates and Losses Returns and Spreads A Continuing Investment Dilemma European Sovereign Debt Crisis New Issues and Other Changes in Size of the High-Yield Market The Leveraging of Corporate America Proportion and Size of the Distressed and Defaulted Public and Private Debt Markets Forecasting Default Rates and Recoveries Performance of Defaulted Debt Securities Appendix A: Quarterly Default Rate Comparison ( ) Appendix B: Defaulted Corporate Straight Debt Issues Appendix C: Distressed Exchanges Appendix D: Leveraged Loan Defaults Appendix E: Chapter 11 Filings by Liability Size Appendix F: Defaults by Industry Appendix G: Emergences from Bankruptcy Acknowledgments Dr. Altman is the Max L. Heine Professor of Finance and Director of the Credit and Debt Markets Research Program at the NYU Salomon Center, Leonard N. Stern School of Business and a member of the Advisory Board of Paulson & Co. Brenda Kuehne is a Credit and Debt Markets Research Specialist at the NYU Salomon Center. We appreciate the assistance of Alex Dai and Vin Morada of the NYU Salomon Center and the several market makers who provided us with price quotations. We offer a special thanks to the various rating agencies, Oleg Melentyev of Bank of America Merrill Lynch, Daniel Sweeney of Credit Suisse, Steven Miller of S&P LCD, Kerry Mastroianni of New Generation Research and Ty Wallach and Sheru Chowdhry of Paulson & Co. 2

3 From a performance perspective, the year 2011 proved to be a challenging one for investors in high-yield bonds. Despite a continuation of extremely low default rates and record amounts of new issuance, absolute annual returns as well as returns versus 10-yr Treasuries were well below historical averages. Additionally, 17.9% of the high-yield market was classified as distressed by year-end compared to only 7.6% one year earlier. The default rate rose slightly to 1.31%, a scant 18bp higher than last year and the second lowest annual rate since The fourth-quarter 2011 default rate was 0.67%, however, the highest quarterly default rate since the fourth-quarter Default losses on high-yield bonds came in at 0.58%, based on a weighted average recovery rate of 60.3% just after default, a level significantly higher than the historical average, and not seen since The weighted average recovery on bankruptcy and payment defaults was somewhat lower at 57.9%, compared to 79.5% for distressed exchange default recoveries. Returns on high-yield bonds were considerably lower than last year, ending the year at 5.52% (Citi Index). The excess return versus 10-yr US Treasuries was a %, inferior compared to a 2.35% historic average, and the sixth lowest in our 34-year time series. Yield-to-maturity spreads versus 10-yr US Treasuries increased to 6.54% by year-end 2011, 196bp higher than year-end 2010, and above the historical average of 5.25%. Defaulted bonds and bank loans also lost ground in 2011, with a combined annual return of -3.02%. The distress ratio of bonds yielding more than 1,000bp over comparable duration treasuries, measured by number of issues, decreased to 17.9% as of the end of 2011 from 22.4% three months earlier, but increased significantly from 7.2% at year-end The distress ratio at year-end was very close to the historic year-end average of 20.6%. Estimates of the face value size of the distressed and defaulted debt markets increased to $1.46 trillion as of December 31, 2011, up 38% from $1.06 trillion one year earlier, completely due to the increase in distressed debt both public and private. The market value estimate also increased to approximately $836 billion from $597 billion one year earlier. Based on three different methodologies, the 2012 default rate forecasts range from 3.93% (distressed ratio method) to 4.80% (yield-spread method), with a consensus average rate of 4.28%. 3

4 Defaults, Default Rates, and Recoveries High-yield bond default rates increased slightly in 2011, but remained well below historical averages. The rate increased from 1.13% at year-end 2010 to 1.31% for all of Defaults include straight corporate bonds whose firms went bankrupt, missed an interest payment and did not cure it within the grace or forbearance period, or completed a distressed exchange. The 2011 rate is based on a mid-year market size of $1.35 trillion, up by a sizeable $133 billion from a year earlier. In all, $17.8 billion of defaults were recorded in 2011 (Figure 1). Note in Figure 1 that the historical weighted-average annual default rate is 3.99% over the 41 year period ( ). This weighted-average rate is down compared to 4.25% at the end of Our weights are based on the par value of high-yield bonds outstanding in each year. The arithmetic annual average default rate dropped to 3.23% from 3.28% one year earlier. The fourth-quarter 2011 default rate was 0.67%, larger than one quarter earlier (0.44%), and indeed the highest quarterly default rate since the fourth-quarter Realizing a quarterly default rate above 0.5% broke the seven-quarter streak of default rates below that level. Since 1989, there have been two equally long or longer, consecutive quarterly periods of default rates also below 0.5% -- seven from Q to Q and nine from Q to Q (Figure 2 and Appendix A). Eighteen issuers defaulted in the fourth quarter on 53 issues. These constitute 58% of all defaulting issuers and 68% of all issues defaulting in In all, 31 issuers constituting 78 issues defaulted in 2011 (Appendix B), compared to 34 issuers and 50 issues in The average dollar amount of defaulting bonds per defaulting issuer in 2011 was $575 million, compared to $406 million in 2010, and $1.04 billion in The most sizeable defaults during the year were those of Dynegy Holdings LLC ($3.6 billion), NewPage Corp. ($3.2billion), Opti Canada, Inc. ($2.6 billion) and AMR Corp. ($1.9 billion), all attributable to bankruptcy filings. Excluding Angiotech Pharmaceuticals, Inc., with $575 million in total bond defaults, the remaining issuers to default in 2011 have bond default totals of less than $500 million. In our default statistics, we include those bonds from distressed exchanges actually tendered. For example, in the Dune Energy exchange, $297 million of bonds were exchanged of the $300 million outstanding and subject to the exchange offer. In 2011, there were eight distressed exchanges, involving as many companies, comprising $1.71 billion of defaults (9.6% of the total). See Appendix C for the list of 2011 distressed exchanges and later our discussion of these restructurings. In 2011, S&P and Moody s issuer-denominated default rates were 1.98% and 1.82%, respectively. Moody s 1.12% dollar-denominated default rate was lower than its issuer-denominated rate, as is usually the case when credit markets are in a strong, low default, benign state. 1 Fitch s dollar-denominated default rate for 2011 was 1.5%. 1 High Yield Bonds: Default and Loss Rate Comparison Mid-Cap Versus Large-Cap Issuers, M. Verde, P. Mancuso and E. Altman November 11, 2005, Fitch. 4

5 The issuer-based default rate for the last 12 months in the US leveraged loan market was 0.62% (Figure 3), and 0.17% based on amount of issuance, according to S&P s LCD compilations. This is in stark contrast to the significantly higher rates of 2.86% and 1.87%, respectively, at the end of Leveraged loans, according to S&P, are secured loans issued by non-investment grade companies. As such, their index does not include the rare case of a fallen-angel secured loan unless that loan became secured in a distressed exchange, but was unsecured when originally issued. As with bond defaults, issuer-based default rates tend to be higher than dollardenominated rates during benign credit periods for leveraged loans, and the reverse is true during stressed periods. During the latter periods we tend to observe not only more defaults, but larger firms are less likely to survive. Four leveraged loan issuers defaulted in 2011 (Appendix D), compared to 20 in According to our comparison between high-yield bond defaults (Appendix B) and leveraged loan defaults (Appendix D), only one firm, Sbarro, Inc., had both bonds and leveraged loans default in See Figure 4 for the association between dollar-denominated bond default rates and economic recessions in the U.S. since the early 1970 s, including the recession that recently ended in mid As usual, we see the default rate peaking at or near the end of the recession, although we observed the peak before it was confirmed that the recession had indeed ended in June, Our forecast for 2012 is for a high-yield bond default rate of 4.28%. If the extremely liquid markets of 2010 and most of 2011 persist, and the proportion of low-rated companies that are able to tap both the debt and equity markets for refinancing continues, our forecast will likely be on the high-side. We will explore at a later stage the statistical associations that support our relatively high expected defaults, which are somewhat above that of other forecasts. Associated to this forecast, there are several important risks on the horizon (see discussion later) that have increased the required yield on high-yield bonds beyond what the miniscule recent default rates imply. 5

6 Figure 1. Historical Default Rates Straight Bonds Only, Not Including Defaulted Issues From Par Value Outstanding, (Dollars in Millions) Par Value Year Outstanding a ($) Defaults ($) Default Rates (%) ,354,649 17, ,221,569 1,152,952 13, , ,091,000 50, ,075,400 5, ,600 7, ,073,000 36, ,100 11, ,000 38, ,000 96, ,000 63, ,200 30, ,400 23, ,500 7, ,400 4, ,000 3, ,000 4, ,000 3, ,907 2, ,000 5, ,600 18, ,000 18, ,258 8, ,187 3, ,557 7, ,243 3, , , , , , , , , , , , , , , , Arithmetic Average Default Rate Standard Deviation (%) 1971 to to to Weighted Average 1971 to Default Rate b Median Annual Default Rate 1978 to to to a As of midyear. b Weighted by par value of amount outstanding for each year. Source: NYU Salomon Center. 6

7 Dec-98 Apr-99 Aug-99 Dec-99 Apr-00 Aug-00 Dec-00 Apr-01 Aug-01 Dec-01 Apr-02 Aug-02 Dec-02 Apr-03 Aug-03 Dec-03 Apr-04 Aug-04 Dec-04 Apr-05 Aug-05 Dec-05 Apr-06 Aug-06 Dec-06 Apr-07 Aug-07 Dec-07 Apr-08 Aug-08 Dec-08 Apr-09 Aug-09 Dec-09 Apr-10 Aug-10 Dec-10 Apr-11 Jan-00 Dec-11 Quarterly Default Rate 4 - Quarter Moving Average February 03, 2012 Figure 2. Quarterly and the Four-Quarter Moving Average Default Rate % 5.0% 4.0% 3.0% Quarterly Moving 16.0% 14.0% 12.0% 10.0% 8.0% 2.0% 1.0% 0.0% 6.0% 4.0% 2.0% 0.0% Source: NYU Salomon Center. Figure 3. S&P Leveraged Loan Index 12-Month Moving Average Default Rate (Number of Issuers) 9.00% 8.00% 7.00% 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% Source: Standard & Poor s LCD. 7

8 $ Billion February 03, 2012 Figure 4. Historical Default Rates and Recession Periods in the US High-Yield Bond Market, % 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% Periods of Recession: 11/73 3/75, 1/80 7/80, 7/81 11/82, 7/90 3/91, 4/01 12/01, 12/07 6/09. Sources: Figure 1 of this report and National Bureau of Economic Research. Bankruptcies As can be seen in Figure 5, the amount of total liabilities for Chapter 11 bankruptcies in 2011 was $109.1 billion, based on 84 filings. Even though the total number of filings, for bankruptcies with liabilities greater than $100 million, decreased from year-end 2010, and was the lowest since 2007, the total amount of liabilities almost doubled from the prior year. MF Global Holdings Ltd. was the largest bankruptcy filing in 2011, with $39.7 billion in liabilities, followed by AMR Corp. ($29.6 billion). Appendix E lists this year s large Chapter 11 bankruptcies. Figure 5. Total Filings and Liabilities a for Chapter 11 Bankruptcy, of Public Companies Filing Pre- Petition Liabilities, in $ billions (left axis) Number of Filings (right axis) $800 $700 $600 $500 $400 $300 $200 $100 $ filings and liabilities of $56.9 billion filings and liabilities of $109.1 billion a Minimum $100 million in liabilities. Sources: Appendix E and the NYU Salomon Center Bankruptcy Filings Database. The number of billion-dollar bankruptcies in 2011 decreased by half from 14 in the prior year to seven, and was lower than the average over the 31 year period (1980-8

9 2011) of 11. No single industry seemed to account for the majority of the larger bankruptcies, as transportation, manufacturing, utility, retail, finance and leisure and entertainment companies are all represented in the short list of billion-dollarbankruptcies (Appendix F). According to New Generation Research, the number of public companies bankruptcy filings in 2011 was 73. The average of total liabilities for our public and private large company bankruptcy filings (84) in 2011 was $1.30 billion, up considerably from $500 million one year earlier, but still below the historical average of $1.65 billion (Figure 6). Figure 6. Historical Bankruptcy Filings Total Filings b (>$100 Total Filings Total Liabilities b ($ MN) Average Liabilities b ($ MN) Year Total Filings a Million) ( $1 Billion) (>100 Million) (>100 Million) Total ,036, ,652.1 a Represents both Chapter 7 and 11 public company filings; 73 Chapter 11 Filings in 2011 (Source: New Generation Research). b Filings with Total Liabilities greater than $100 million (Source: NYU Salomon Center Bankruptcy Filings Database). C Filings with Total Liabilities greater than $1 billion (Source: NYU Salomon Center Bankruptcy Filings Database and New Generation Research). In Figure 7, we compare the date of default with the Chapter 11 filing date for firms that defaulted on bonds and also went bankrupt, going back to Based on 946 observations from the NYU Salomon Center Master Default and Bankruptcy Databases, both events occurred on the same date in 482 instances (51%). In the remaining 49% of the cases, the lag between the default date and bankruptcy date 9

10 % of the Total Observations February 03, 2012 varied considerably, with decreasing levels as the two dates became further separated from each other. Of course, some defaulting issuers never formally file for bankruptcy as their problems are settled out of court or the default comes as a result of a distressed exchange (DE), and they do not file for bankruptcy in subsequent years (many (almost half) DEs do, however see our discussion at a later point). Figure 7. Time Differential Between Default and Bankruptcy Filing a ( ) 60% (482) 50% 40% 30% 20% 10% 0% Number of Months Lag a Based on 946 observations. Source: NYU Salomon Center Default and Bankruptcy Filings Databases. Industry Defaults Figure 8 lists the number of high-yield bond defaults by industry. Of the 31 defaulting issuers in 2011, communications and media, energy, retailing and transportation were industries in which a total of 19 firms defaulted. The remaining 12 defaulting issuers were spread over various industries. Appendix F presents a more detailed breakdown of all 31 defaulting issuers. Figure 9 shows high-yield corporate bond defaults across industries per dollar amount since Although only two issuers defaulted in the utilities sector, it boasted the largest dollar amount of defaults in 2011, almost exclusively attributable to the Dynegy Holdings default. As in the past, we observe that the communications and media sector far outdistanced all other sectors in the dollar amount of defaulting issues over the last 21 years, primarily the result of the telecom meltdown during , as well as 2009 s defaults due to large-scale bankruptcies. 10

11 Figure 8. Corporate Bond Defaults by Industry (Number of Companies) Industry Total Auto/Motor Carrier Conglomerates Energy Financial Services Leisure/ Entertainment 107 General Manufacturing Health Care Miscellaneous Industries RealEstate/ Construction 77 REIT Retailing Comm. & Media Transportation (Non Auto) 68 Utilities Total ,400 Source: NYU Salomon Center. 11

12 Figure 9. Corporate Bond Defaults by Industry (Dollars in Millions) Industry Auto/Motor Carrier ,737 Conglomerates Energy , ,200 Financial Services ,968 5,062 Leisure/ Entertainment 498 1, ,100 2,891 3,437 General Manufacturing 2,675 3, ,092 2,507 3,138 Health Care 18 1, ,214 1, Miscellaneous Industries 1,968 4,911 1,378 1,373 1, ,290 7,615 8,352 9,715 Real Estate/ Construction 2, ,110 Retailing 4,443 2,937 1,489 2, ,504 1,241 2,052 3,081 1,586 Communications & Media ,549 2,980 5,983 34,827 Transportation (Non Auto) 1,028 1, , ,890 1,430 Utilities 1, Total 14,631 18,021 4,883 5,649 4,536 3,465 4,200 6,994 23,440 29,976 68,934 Industry Total Auto/Motor Carrier ,573 2,692 1,382 16, ,717 Conglomerates ,065 Energy 2,734 7,399 8, ,511 1,993 3,414 35,163 Financial Services 3,803 1, ,973 29,274 2, ,926 Leisure/ Entertainment 21, ,286 6, ,022 10,395 1, ,542 General Manufacturing 2,455 2, ,396 1,486 2,379 3,747 26, ,205 60,263 Health Care 115 3, ,016 Miscellaneous Industries 5,594 4,494 1, ,396 1,505 1, ,369 Real Estate/ Construction 1, , ,158 4, ,351 Retailing 4, , ,412 1, ,693 34,841 Communications & Media 47,953 7,603 2, , ,904 30,954 2, ,935 Transportation (Non Auto) 4,711 2,086 2,421 12, ,268 34,196 Utilities 2,501 5,875 3,594 4,011 18,686 Total 96,673 36,764 11,657 35,954 7,559 5,473 50, ,878 13,809 17, ,071 Source: NYU Salomon Center. Age of Defaults Figure 10 shows the age distribution of defaults in 2011 and for the period Defaults in 2011 did not, for the most part, closely follow the normal pattern of low defaults in the first year after issuance, followed by high relative default rates in years through two to five. In fact, only 23% of the defaults occurred within four years after issuance, while 50% took place within five to eight years after issuance, and the majority of the remaining 27% occurred more than 10 years after issuance. This anomaly, with a relatively large proportion of defaults occurring later after issuance than is typical, was primarily attributable to the default of many older bonds issued by AMR Corp.. Figures 10 and 11 show the long-term historical pattern, highlighting the most vulnerable years as two through five. 12

13 Figure 10. Distribution of Years to Default From Original Issuance Date (By Year of Default), / / Years to No. of % of No. of % of No. of % of No. of % of No. of % of No. of % of No. of % of Default Issues Total Issues Total Issues Total Issues Total Issues Total Issues Total Issues Total Total /2007 Years to No. of % of No. of % of No. of % of No. of % of No. of % of No. of % of No. of % of Default Issues Total Issues Total Issues Total Issues Total Issues Total Issues Total Issues Total Total Years to No. of % of No. of % of No. of % of No. of % of No. of % of Default Issues Total Issues Total Issues Total Issues Total Issues Total Total , Source: NYU Salomon Center. 13

14 % of All Defaulted Issues February 03, 2012 Figure 11. Distribution of Years to Default From Original Issuance Date: Summary Chart, % 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% # of Years to Default Since Issued Source: NYU Salomon Center. Fallen Angel Defaults Five issuers were responsible for 23 defaulting issues that were investment grade at some time prior to default. This results in a fallen-angel issuer default rate of 3.36%, in-line with the historical average of 3.78% (Figure 12). The fallen-angel default rate for is slightly below the historical average annual rate for original issue defaults in the high-yield bond market (4.73%). This differential (3.78% vs. 4.73%), however, is not statistically significant due to a relatively high standard deviation of around 270bp per year between the two rates. Figure 13 shows the fallen angel proportion of defaults from 1977 to the present. In 2011, 29% of defaulted issues were originally rated investment grade, slightly more than the historical average of 27%. 14

15 Figure 12. Fallen Angels Versus Original(S&P) Issue and All High Yield Default Rates a (In Percent), Year Fallen Angel Average 12-Mo. Default Rate Original Issue Speculative Grade Default Rates All Speculative Grade Bond Default Rates Altman Dollar Weighted Annual Default Rates a 1.96 a 2.26 a Arithmetic Average Standard Deviation a Issue based. b All S&P issuer based except for Altman rates and Sources: NYU Salomon Center and S&P. 15

16 Figure 13. Fallen Angel Defaulted Issues by Original Rating, Year Total No. Defaulted Issues a Grade (%) Originally Rated Investment Total 2,836 27% a Where we could find an original rating from either S&P or Moody's. Sources: Moody's, NYU Salomon Center, and S&P. Default Losses and Recoveries The weighted-average recovery rate (based on market prices just after defaults) on high-yield bond defaults in 2011 increased to 60.3%, considerably above the historic average ( ) of 45.3%. This is higher than the recovery rate of 46.6% in 2010 and is the highest rate since The default loss rate in 2011, without an adjustment for fallen angels, and including the loss of 0.060% (6.0bp) from lost coupons, was approximately 58.2bp (Figure 14). If we remove fallen angel defaults (23 issues) the loss would have been 49.9bp on original issue high-yield bonds. Of note is the fact that for a second consecutive year the recovery rate on fallen angels, contrary to the norm, was lower than on original non-investment grade bonds. This was the result of the AMR Corp. defaults, which recovered only 17% on average. Over the 34-year period from 1978 to 2011, the arithmetic average annual loss rate on high-yield bond defaults is 2.34%, 2.68% on a weighted-average basis (Figure 15). 16

17 Figure Default Loss Rate Unadjusted for Fallen Angels (%) Only Fallen Angels (%) All Except Fallen Angels (%) Price Adjusted for Fallen Angels (%) Background Data Average Default Rate Average Price At Default a Average Price At Downgrade b Average Recovery Average Loss Of Principal Average Coupon Payment Default Loss Computation Default Rate X Loss Of Principal Default Loss of Principal Default Rate X Loss of 1/2 Coupon Default Loss of Coupon Default Loss of Principal and Coupon a If default date price is not available, end-of-month price is used. b Downgrade to noninvestment grade. Note: Average Default Rate of Only Fallen Angels is based on number of issuers. Sources: NYU Salomon Center and various dealer quotes. 17

18 Figure 15. Default Rates and Losses, a (Dollars in Millions) Par Value Outstanding a Par Value of Default ($) Default Rate (%) Weighted Price After Default ($) Weighted Default Loss Coupon (%) (%) b Year ($) ,354,469 17, ,221,569 13, ,152, , ,091,000 50, ,075,400 5, ,600 7, ,073,000 36, ,100 11, ,000 38, ,000 96, ,000 63, ,200 30, ,400 23, ,500 7, ,400 4, ,000 3, ,000 4, ,000 3, ,907 2, ,000 5, ,600 18, ,000 18, ,258 8, ,187 3, ,557 7, ,243 3, , , , , , , , , Arithmetic Average Weighted Average a Excludes defaulted issues. b Default loss rate adjusted for fallen angels is 9.3% in 2002, 1.82% in 2003, 0.59% in 2004, 1.56% in 2005, 0.039% in 2006, 0.20% in 2007, 3.42% in 2008, 7.38% in 2009,0.66% in 2010, and 0.58% in Source: NYU Salomon Center. Figure 16 lists the average recovery rate by seniority for In 2011, 15 of the defaulting issues were senior secured with an average recovery rate of 59.0%, compared to a historical average of 57.6% (57.4% median). Forty-five of the issues were senior unsecured with an average recovery rate of 64.0%, compared to a historical average of 38.8% (46.8% median). The large discrepancy in the recovery rate on this seniority, versus the historical average, was partially attributable to the fact that 11 of the 45 defaults occurred as part of a distressed exchange (see below). There were three issues that were senior subordinated with an average recovery of 42.8%, compared to an historical average of 30.6% (32.7% median). Two were subordinated with an average recovery of 20.0%, compared to a historical average of 30.6% (27.5% median). There were no priced defaults in 2011 in the discount and zero coupon category. Thirteen issues could not be priced. The historic 34-year median for all high-yield bond defaults rose slightly to 42.1%, while the arithmetic average increased as well to 38.2%. These latter statistics are based on a sample of over 2,700 defaults. 18

19 Figure 16. Weighted Average (by Issue) Recovery Rates on Defaulted Debt by Seniority per $100 Face Amount, Senior Secured Senior Unsecured Senior Subordinated Subordinated Discount and Zero Coupon All Seniorities Default Year No. % $ No. % $ No. % $ No. % $ No. % $ No. $ Total/Avg , , Median Standard Dev a a Standard deviations are calculated based on the yearly averages. Sources: NYU Salomon Center from various dealer quotes. Distressed Exchanges in 2011 Distressed exchanges (DEs) in 2011 accounted for 25.8% of the defaulted issuers (8 out of 31), but only 9.6% of the defaulted dollar amount. From 1984 through 2011, DEs accounted for about 11.3% of all defaulting issuers and 12.4% of all defaulted dollar amounts (Figure 17). Relatively speaking, 2011 was an above average year for DE activity (third highest annual number in our 27 year time-series) from a number of issuers perspective, but the value of the dollar amounts exchanged was relatively low. Figure 16 indicates the popular re-emergence of DEs in as compared to the last 27 years. Indeed, during this four year period, more than 56% (74 of 131) of all DEs took place. However, the pace at which DEs were being sought as a restructuring alternative slowed considerably, at least in absolute terms, in the second-half of 2009 and into 2010, as other avenues of refinancing became available. Still, 26% in 2011 is a relatively high proportion 19

20 of total defaulting issuers (second highest ever), albeit based on a small total number of defaults. The concept of a DE has taken on an added level of importance and urgency of late, especially as to whether such events will trigger a default in the credit default swap (CDS) market. Since early 2009, such events in the U.S. corporate bond market do not constitute a default event, as per the typical I.S.D.A. specification. Voluntary DEs in Europe, especially in the now crucial sovereign debt markets, will probably not be considered a default e.g. for Greece. However, if write-downs escalate to 50% or more, as is now indicated, such agreements are, in fact, equivalent to a default in our opinion. Important too is the performance of a firm subsequent to completing a DE. As discussed in an earlier study 2, data would appear to indicate that a DE is oftentimes just a short-term fix, unable to prevent future bankruptcy filings or acquisitions. Please see below for our updated discussion of events subsequent to completing a DE. 2 The Re-emergence of Distressed Exchanges in Corporate Restructurings, E. Altman and B. Karlin, NYU Salomon Center Working Paper, 2009 (see E. Altman s website, and published in The Journal of Credit Risk, Summer

21 Figure 17. High Yield Bond Distressed Exchange (D/E) Default and Recovery Statistics, D/E Defaults ($) Total Defaults ($) D/E Defaults (%) to Total $ D/E Defaults (No. of Issuers) Total Defaults (No. of Issuers) D/E Defaults (%) to Total No. of Issuers D/E Recovery Rate a All Default Recovery Rate a Difference Between D/E & All Default Recovery Rate Year , , , , , , , , , NA NA NA , NA NA NA , , , , , , , , , , (23.12) , NA NA NA , NA NA NA , NA NA NA , NA NA NA , NA NA NA , NA NA NA , (9.37) , , , , (15.05) , (25.93) , (6.50) Totals/ Averages $75, $611, % 131 1, % b b a Weighted-average recovery rates for each year. b Arithmetic average of the weighted-average annual recovery rates; only those years with DEs counted. The arithmetic average of each individual DE (131) for the entire sample period was 49.24% and the average for the non-de defaults (1,033 observations) was 36.80%. Source: NYU Salomon Center. Recovery Rates on Distressed Exchanges Because DEs are not as dramatic a reflection of a firm s distressed status as a bankruptcy or nonpayment of cash interest on debt, one might expect the recovery rate on DE defaults to be higher than other, more serious distressed situations. Of course, one reason for the larger recoveries in DEs is lenders need to be offered a premium in order to be persuaded to participate in the exchange. Figure 17 shows the arithmetic average recovery rate on all DE defaults was 53.8% for , compared to 43.1% for all defaults, and 36.8% for all non-de defaults (not shown in Figure 17). In 2011, DEs recovered 79.5%, while all defaults recovered only 60.3%. The historical spread widened slightly from one year earlier (10.7% versus 10.3%). In Figure 18, we calculate a difference in means test between the arithmetic average recovery rate (49.2%) 3 on the 131 DE issuers (286 issues) during and the 3 Please note that the weighted average recovery rate (53.6%) on our total sample of 131 DEs is slightly lower than simply averaging the annual DE recovery rates over the period (53.8%) in Figure

22 average recovery rate on all non-de defaults (36.8%) of the same period. We found that given the above, the DE recovery rate is significantly higher (t = 8.23) at the 1% confidence level. It is not surprising that bondholders will choose, in many instances, to accept a recovery with certainty from a DE, rather than take the chance of holding out for an uncertain and likely lower recovery in bankruptcy (see below). Our results do not include data for situations where a DE offer is rejected. Figure 18. Difference in Means Test Between Recovery Rates: All Nondistressed Exchange Defaults Versus Distressed Exchanges (D/E), (Based on Issues) All Defaults Excluding D/E (Issues) Distressed Exchange (Issues) Sample Size Mean Recovery Rate Standard Deviation Variance t-test a X DE X NDE a t = Var X DE Var X NDE + N DE N NDE Sources: NYU Salomon Center, and authors compilation. In Figure 19, we calculate a difference in means test between the weighted average recovery rates on the announcement date of a DE versus the completion date for the recent period 2008 through Of the 42 defaulted issues in 2008 due to a DE, in which prices were available for both the announcement and completion dates, the weighted average recovery was approximately 13 percentage points higher on the completion date, while in 2009 it was less so, with only a 4.9% difference in pricing between the two dates (based on 106 observations). The reverse was true in 2010, with the weighted average recovery on five issues being 4.6% higher on announcement than completion date witnessed a return to the completion date price being higher with a 10.86% difference in pricing between the two dates (based on 11 observations). Overall, the completion date price was higher than the announcement date in 86 of the 164 issues, just over 50% (52.4%) of the cases. For the entire four-year sample period ( ), the difference between the price at completion of the DE vs. at the time of the announcement was 8.24%, significant at the.01 level. Of course, market conditions in general can change between the two dates. 4 The 2008 difference was significant at the.01 level; the 2009 difference at the.05 level; the 2010 difference was not significant; the 2011 difference was significant at the.05 level, and the Difference for all four years was significant at the.01 level.. 22

23 Figure 19. Distressed Exchange Weighted Average Recovery Rates: Announcement Date versus Completion Date, # of Observations Completion Date Difference t-test Year Announcement Date a b c b Total a a Significant at the.01 level. b Significant at the.05 level. c Not significant. Source: NYU Salomon Center. Subsequent Performance of Distressed Exchange Companies For the first time in several years, we are able to update our initial study 5 which tracked the performance of those firms which had achieved a successful Distressed Exchange (DE), in most cases to avoid a bankruptcy filing. Our new sample involves all corporate bond DEs over the period for which we are able to confidently ascertain the current status of the firms, including those that filed for bankruptcy subsequent to the DE. We document the status of 72 DEs, and our post- DE experience covers at least three years. This updated sample includes 14 DEs from the class of 2008, a very active year, accounting for almost 20% (14/72) of the total. Our primary interest in this study is to document the success, or not, of the DE with respect to providing an effective means for firms to restructure their debt so as to avoid the usually more drastic default experience of a Bankruptcy filing either Chapter 7 liquidation or Chapter 11 reorganization. Since liquidation or reorganization under the Federal Bankruptcy Code results in statistically significant lower recoveries to creditors than do DEs (see earlier Figure17), and almost assuredly results in greater numbers of lost jobs, revenues and taxes than for firms which effectively restructure and survive outside the court, it is extremely relevant to observe if the DE helps to preserve the going concern value of the enterprise. If the DE only postpones the firms bankruptcy, then we argue that the DE was not a genuine success. In addition, the subsequent performance of DE companies has important implications for those investors whose original bonds are exchanged for new securities. We have arbitrarily chosen a minimum of a three-year post-de period to observe the performance of the companies in our sample. This period, we believe, gives ample time to assess whether or not the firm s DE has resulted in a continuing entity. At the same time, we believe that subsequent bankruptcies beyond the three year period are also clearly problematic and therefore we continue to track the performance of the DEs from before 2008, as well as the latest class being analyzed the class of Our post-de experience is broken down into the following categories: (1) Still Operating 5 The Re-emergence of Distressed Exchanges in Corporate Restructurings, E. Altman and B. Karlin, The Journal of Credit Risk, Summer

24 (2) Acquired (3) Bankrupt Chapter 11 (4) Bankrupt Chapter 7 (5) Bankrupt After Being Acquired The first two categories constitute our depiction of a successful DE, either in terms of the complete continuity of the old firm or as a part of a merged entity that is still operating (some of these successful DEs are perhaps questionable if a second DE took place to avoid a subsequent bankruptcy). The last three categories constitute a failed DE, at least in terms of a subsequent bankruptcy filing. Figure 20 shows the post-de results for our sample of 72 DEs from First, we describe the type of DE in terms of the securities, or cash, used in the exchange. The most popular mechanism is a Debt for Debt exchange with 24 (33%) of the 72 firms using a new issue of debt to substitute for the old debt. An additional 18 (25%) utilized new debt combined with either cash, equity, or both. Therefore, 58% of DEs resulted in the firm having at least some proportion of newly exchanged debt as part of its capital structure. Twenty-one (29%) utilized equity as all or part of the DE and another 21 (29%) utilized cash in the exchange. Ten (14%) could not be determined. In terms of success, or not, of the DE, we found that 33 of the 72 DEs in our sample (45.8%) ultimately filed for bankruptcy 27 Chapter 11 s (two after being acquired) and six Chapter 7 s. These we label unsuccessful DEs, although it may have taken a long time for the bankruptcy filing (in one case, the eventual filing took almost 19 years). The average time between the DE date and the subsequent filing was 2.62 years, while the median time was only 1.67 years. When we observed the sample of DEs from the earlier period, 27 (46.5%) ultimately filed for bankruptcy, very similar to our updated, larger sample. As expected, we found that very few DEs that utilized lower risk equity in the DE ultimately filed for bankruptcy. Out of the 33 bankrupt DEs, only 7 (21%) utilized either all equity or a combination of equity and cash (Figure 21). Indeed, 18 (58%) utilized debt alone or some combination of securities that included new debt, and if we exclude the undetermined category (5), the percent using new debt swells to 64%. In conclusion, we believe that while DEs are an effective mechanism to avoid an imminent bankruptcy filing, in almost half of the cases, the reprieve was only temporary, and the firms problems continued to persist. Of course, since perhaps (some DEs eventual fate is undetermined) more than half of our sample resulted in success, the DE effect was worthwhile, in those cases. Indeed, 39 of the 72 firm DE sample are classified as successes, though six endured a subsequent, second DE. 24

25 Figure 20. Subsequent Performance of Bond Distressed Exchanges, Exchange Type Subsequent Development Years from DE to Bankruptcy Debt 24 33% Bankruptcy Ch % Count 33 Cash 7 10% Bankruptcy Ch % Mean 2.62 Equity 10 14% Acquired 19 26% Median 1.67 Debt/Equity 7 10% Still Operating 17 24% Maximum Debt/Cash 10 14% Acquired Subsequent Ch % Minimum 0.02 Equity/Cash 3 4% Other 3 4% Debt/Equity/Cash 1 1% Undetermined 10 14% Total % Total % Still Operating Subsequent DE 6 8% Source: Altman-Kuehne Default Database NYU Salomon Center. Figure 21. Distressed Exchanges Resulting in Bankruptcy, Chapter 11 Exchange Type Debt Cash Equity Debt/Equity Debt/Cash Equity/Cash Undetermined Total % 7% 22% 4% 15% 4% 15% 100% Chapter 7 Exchange Type % 17% 0% 17% 33% 0% 17% 100% Total % 9% 18% 6% 18% 3% 16% 100% Source: Altman-Kuehne Default Database NYU Salomon Center. Forecast Recovery Versus Actual The 2011 weighted-average recovery rate of 60.3% was significantly above our linear or non-linear regression default/recovery rate forecasting models predictions (Figure 22). 6 We would have expected the average recovery rate to be closer to the 51%-53% range, based upon the model that we developed. 6 The Link Between Default and Recovery Rates: Theory, Empirical Results and Implications, Altman, Brady, Resti and Sironi, Journal of Business, November