Trust Preferred CDOs A Primer 11-11-04 (Merrill Lynch)
Trust Preferred CDOs A Primer 11-11-04 (Merrill Lynch)
Trust Preferred CDOs A Primer 11-11-04 (Merrill Lynch)
11 November 2004
Lang Gibson Director (1) 212 449-3942 lang_gibson@ml.com
CDO Research
In s u ra n c e T ru P S **
Source: Merrill Lynch * All bank or bank-dominant collateral (e.g., hybrids) / ** 100% insurance collateral
Merrill Lynch does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.
While the focus of this primer is bank TruPS, we do address insurance TruPS and surplus notes due to their increasing inclusion in bank deals.
Trust Preferred CDOs: A Primer 11 November 2004 n Bank TruPS TruPS benefit from the tax advantages associated with debt and capital relief associated with equity. Both the securities long-term nature (30 years) and the issuers ability to defer dividends for up to five years contribute to bank regulators allowing TruPS to account for up to 25% of Tier 1 capital. Bank issuers are typically wholly-owned subsidiaries of bank holding companies, or may be a subsidiary of the bank directly. The trust sells securities to investors and uses the proceeds to purchase subordinated debentures of its holding company. The trust uses the interest payments that it receives from the purchased debentures to make payments to the holders of its preferred securities. The terms of the subordinated debenture are identical to those of the issued TruPS (Chart 2).
Chart 2: Flows for a TruPS Security
100% o Contr
mi te
Li
Proceeds of Offering
n Insurance TruPS and Surplus Notes Both insurance TruPS and surplus notes, which are often also included in CDO pools, are hybrid instruments (much like bank TruPS) in the capital structure of an operating insurance company. Their interest is tax deductible and they receive partial credit for permanent equity capital by the rating agencies.
There are also some important distinctions between insurance TruPS and surplus notes. Insurance TruPS are typically issued at the holding company level; whereas insurance surplus notes are issued from operating insurance companies. Furthermore, on GAAP balance sheets, insurance TruPS are accounted for as long-term debt. By contrast, surplus notes are issued by the operating company and are generally less subordinated than TruPS. Typically, surplus notes are notched down 2x from the operating companys debt rating versus 4-5 notches for TruPS. Interest and principal payments on surplus notes are subject to prior regulatory approval. State regulators generally do not allow issuance of surplus notes to exceed 15% of issuer surplus.
Issuance ($mm) $6,000 $5,000 $4,000 $3,000 $2,000 $1,000 $0 2000 2001
Bank TruPS CDOs 2002 Insurance TruPS CDOs 2003 2004 YTD
TruPS CDOs provide investors with regionally diversified exposure to a new issuer base with a solid performance history, particularly with respect to banks whose peak historical failure experience is a fraction of the peak overall corporate default experience. As well, they provide debt investors with exposure to bank originated commercial real estate (CRE) risk with more subordination than a direct investment in CMBS, whose enhancement levels have significantly loosened as a result of several factors including better-than-expected performance. Performance for TruPS CDOs, like that of the underlying collateral, has been excellent, with no rating actions to date. Major risks include industry concentration, interest deferral potential, auction-related extension risk and prepayment risk, which are all key determinants of ultimate performance.
Trust Preferred CDOs: A Primer 11 November 2004 n Enter the insurers TruPS CDOs issued between 2000 and late-2002 were primarily backed by bank and thrift collateral with small buckets for insurers in some deals. The first TruPS CDO backed by 100% insurance collateral was issued in late 2002. Still, TruPS CDO issuance has been dominated by bank collateral. Pure insurance TruPS CDOs represent 15% of the total outstanding TruPS CDO market today. The share of insurers in hybrid bank/insurance pools could grow to 50% within the next year, enabling originators to reach their spread and diversity targets. However, we expect bank collateral to dominate pool composition in the shortand medium-term. Whereas rating agencies and investors seem to generally view bank collateral credit quality as stronger than insurers, they also recognize the diversity benefits of adding insurance credits to the pool. P&C insurers, particularly, tend to have disparate lines of businesses and geographic exposures. Robust TruPS CDO issuance over the past few years has driven small bank TruPS spreads to particularly tight, scarcity-driven levels. These tighter collateral spreads have, in turn, diminished CDO volumes this year because there are fewer collateral issuers with adequate spreads and diversity to enable the arbitrage (Chart 4).
Chart 4: Monthly TruPS CDO volume vs. bank and insurer TruPS spreads: 2000-present
450 400 350 Spreads (bps) 300 250 200 150 100 50 0 $2,000 $1,800 $1,600 $1,400 $1,200 $1,000 $800 $600 $400 $200 $-
The tightening in bank TruPS spreads has spurred originators to include a sizeable bucket for insurance TruPS, which provides several benefits, including a pool spread pickup and industry diversification. These mixed collateral pool deals, often referred to as hybrid TruPS CDOs, are becoming more prevalent. In addition to a blend of new issue small bank and insurance collateral, TruPS CDOs typically feature a bucket for secondary market TruPS, usually issued by larger bank and insurer issuers, as well as subordinated bank debt.
2.00%
1.50%
1.00%
0.50%
0.00%
46 54 50 86 82 58 78 90 42 66 19 74 38 70 94 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 19 98 34 02 62
Source: FDIC
Between 1987 and 1989, the annual failure rate ranged between 1.43% and 2.04%; the 70-year and 35-year average bank failure rate is a respective 0.24% and 0.38%. In order to develop a more tailored proxy for historical default rates for the typical bank found in a CDO, we refer now to Fitch data, which stratifies the FDIC data further by asset size.
Chart 6: Cumulative BIF-Insured Bank Failure Rates by Size
>$10 billion >$200 m illion and <$10 billion <$200 m illion 14 12 10 8 6 4 2 0 (% )
19 90
19 88
19 86
Note: By asset size of bank. Som e data m ay be estim ated. BIF Bank Insurance Fund.
Source: Fitch Ratings
The nations largest banks (>$10 billion assets) have historically outperformed the very smallest bank category (<$200 million assets), few of which are found in CDOs (Chart 6). The banks whose assets range between $200 million and $10 billion warrant our focus because these represent the majority of banks issuing into TruPS CDOs. Chart 5 illustrates that these mid-size banks have performed comparably if not better than the largest banks historically, with a cumulative gross default rate of 9.83%. In comparison, the chart demonstrates that the smallest banks have experienced cumulative default rates closer to 12%. Historically, troubled banks have been more likely to be acquired than default. Therefore, consolidation on the order of approximately 2% per year (net of start-ups) bodes well for future
19 94
19 96
19 84
19 98
19 92
performance in the sector, in our opinion. The commercial bank failure rate history (a conservative proxy for defaults) compares favorably with historical global defaults as they relate to corporate bonds (Chart 7). Additionally, both banks and insurers benefit from heavy regulation and the implicit government agency oversight that this implies. This regulation gives these sectors an advantage and contributes to their relatively low default rates versus non-bank corporates.
Chart 7: Commercial Bank Failures vs. Global Corp. Defaults
4.00% 3.50% 3.00% 2.50% 2.00% 1.50% 1.00% 0.50% 0.00%
78
76
74
72
70
80
82
84
86
88
90
92
94
96
98
00 20
19
19
19
19
19
Bank Failures
19
19
19
19
Source: FDIC and Moodys * Based on Moodys 12-month trailing global corporate default rate. Includes both IG and HY issuers.
n Insurance loss trends Property & Casualty (P&C) insurers generally command a higher premium in the market as compared to life & health (L&H) companies due to the higher risk associated with their lines of business. Actuarially, it is easier to predict death and health problems (L&H insurers) than the unexpected, multi-sigma events that are more common with P&C businesses. Historical insurer default rates reflect these different risk profiles: L&H annual defaults have averaged 59 bp; whereas P&C defaults have averaged 79 bp (Chart 8). The historical default paths of these two segments have generally been correlated, with the exception of the past few years when P&C defaults rose to 1.25% in 2002 and the L&H rate fell to zero in 2001, our most recent data point. The 1991 recession saw the peak default rate for the L&H and P&C segments simultaneously rise to their historical highs of 1.83% and 2.3%, respectively.
19
19
19
19
19
19
20
02
1.50%
1.00%
0.50%
0.00% 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
P&C
Source: AM Best
L/H
Structure
TruPS CDOs have historically been issued as static pools, with no discretionary trading or reinvestment of principal proceeds (a typical funding mix for a bank-
dominated deal is 63% AAA, 28% A, and 9% equity). AA and BBB tranches are relatively rare for TruPS CDOs. However, there are typically several AAA and Arated tranches both sequential and pro rata in the same deal. The structures used are similar to those found in most other static pool cash CDOs, with some exceptions. We summarize the key collateral and structural considerations in Table 1.
Table 1: Generic TruPS CDO Structure
Class A-1 A-2 A-3 B PS Amount($mm) $162.25 $ 52.42 $ 5.82 $ 98.49 $ 31.01 $350.00 % of Total 46% 15% 2% 28% 9% Rating AAA/Aaa AAA/Aaa AAA/Aaa A/A2 NR/NR Rated Coupon 3mL+52 3mL+70 3mL+70 3mL+160 NA Avg. Life 7.5 - 8.5 10 10 10
n Deferral of Interest An important feature of TruPS is that an issuer may defer interest payments for up to five years, and this can be done multiple times over their thirty-year life. Anecdotal evidence suggests that dividend or interest deferrals on TruPS in both sectors have been rare, although exact figures are not available. The one known case of interest deferrals within a TruPS CDO occurred in the Regional Diversified Funding I Ltd. transaction issued in March 2000. This was the first TruPS CDO ever, and was backed by bank and thrift TruPS. Following dividend deferral on two of the underlying securities in August 2004, Moodys put the senior notes on negative watch. Moodys subsequently took the senior notes off watch a month later when the first of the two deferring obligors received regulatory approval to resume payments, including all accrued and unpaid obligations. We believe that credit deterioration would have to be substantial before a bank or insurer would defer payment. The agencies point to the likelihood of high correlation between bank TruPS interest deferrals and FDIC intervention. We
note that if interest payments are deferred by the holding company, it must also refrain from paying common dividends an onerous structure that incents issuers to continue TruPS payments. For insurers, ongoing interest and principal payments on surplus notes must be approved by the state regulators. As policyholder confidence may be diminished by any perceived negative signals from a deferral, we believe that regulators would exercise extreme caution, only suspending payments under dire conditions. The ability of the obligors to defer interest clearly adds some uncertainty to the cash flows of the CDO. To offset this risk, TruPS in deferral are treated as defaulted securities for purposes of the coverage tests with an assumed recovery rate typically in the 2-5% range. Due to a lack of performance history, agencies use these conservative default and recovery assumptions. Additionally, some transactions feature a test that redirects cash flows to the senior-most tranches as a way to cure the amount of the deferring asset. This is done until either a percentage, or total par amount, of the deferred asset is covered or the security resumes interest payments. In addition, TruPS CDOs typically feature reserve accounts designed to cover potential deferred interest payments. n Pool Diversity TruPS CDOs are unique in that they are concentrated in only one or two industries. However, there is clearly some diversification benefit from aggregating pools from defined regions across the nation. As such, diversity within bank TruPS pools is examined by five distinct regions. Consequently, banks effectively represent five distinct industry categories for rating purposes. The assumption is that banks and insurers are highly correlated intra-regionally, but are somewhat uncorrelated interregionally. The agencies adjust their default rate assumptions, applying penalties to excess geographic concentration within the pool. In the case of hybrid TruPS CDOs, the issue of industry concentration is clearly lessened to the extent an insurance bucket is included. Furthermore, the business mix varies a great deal among middle market P&C issuers, which are more numerous than L&H companies in these deals, implying some standalone diversity benefit from their inclusion. The other consideration when examining pool concentration is obligor exposure. Large single obligor exposures can increase loss severity in any CDO, particularly for subordinate noteholders. To account for this, rating agencies impose single obligor limits and apply more punitive stresses to account for large exposures. n Prepayments & Auction Calls
Prepayments
While many CDO subtypes require consideration of prepayment, or call, risk, it is a particularly important factor in analyzing TruPS CDOs because the lack of observable history makes prepays in this asset type difficult to predict. In TruPS CDOs, prepayment risk is more credit-related as opposed to interest rate related because credits are sourced into the pool at par as Libor floaters. The TruPS issuers call option, combined with the otherwise non-amortizing nature of the TruPS collateral, makes it difficult to forecast the weighted average life (WAL) of the notes, CDO auction calls and redemption features notwithstanding. Prepays can be motivated by either opportunistic refinancing or regulatory changes: Opportunistic call: If the issuer no longer desires the TruPS capital, for whatever reason, it may decide to call the deal to reduce its funding cost after the non-call period typically five years for mid-market issues and ten years for TruPS issued by large issuers in the secondary bucket. As well, an issuer can call the bond if it were able to issue a new TruPS at a tighter spread, an unlikely event in the current tight spread environment. Finally, the acquisition of smaller banks by larger ones can lead to a call if the acquiring bank opts to use its own lower cost of funding. However, the acquired banks TruPS debt has historically gone unnoticed once it has been amalgamated into the larger banks balance sheet.
Regulatory call: Should regulatory rules be altered to remove the TruPSs Tier 1 status for bank and thrifts or remove the tax deductibility status, the issuer may call the bond.
All TruPS CDOs issued to date have been static pools, such that proceeds from prepayments cannot be used to reinvest in new collateral. Instead they immediately reduce the average lives of the senior-most notes. Prepayment risk is a particularly important consideration given the homogeneous nature of the pool with respect to asset type. In sizing credit enhancement, rating agencies stress CDO structure through various prepayment timing scenarios. Larger and earlier prepayments are positive because prepaid assets cannot default and are used to pay down senior notes, resulting in a higher subordination percentage for all of the notes. However, prepayments are more likely to occur in the better performing names. Such cherry-picking would negatively impact the credit quality of the remaining pool. This long-dated tail risk, combined with reduced absolute excess spread following prepayments, may serve to increase the risk profile for the subordinate-most noteholder. As such, it is important for investors in these classes to be comfortable with the credits in the pool. Early TruPS transactions were blind pools, where investors did not have access to collateral specifics. Although most TruPS CDO trustee reporting does not include specifics on the collateral, it is common for originators to provide collateral information and analysis to investors in their deals.
Auction Calls
Investors, particularly those lower in the capital structure, do bear the risk that the tranche average life extends past ten years if auctions repeatedly fail. However, the TruPS CDO structure mitigates the long maturity of the TruPS collateral with 1) the auction call and 2) sequential amortization in the case of a failed auction. The auction is designed to pay off the notes principal in its entirety starting in year ten. Should the auction fail, a re-attempted auction call occurs every payment period until the auction is successful. Furthermore, until such time that the auction is successful, a portion of excess spread (50% - 60%) is diverted to amortize the notes sequentially. Consequently, the senior-most tranche average life sensitivity is significantly less exposed to an auction failure than the junior tranches. As the notes are amortized with this excess spread, the likelihood for a successful auction call increases with each payment date because there are increasingly fewer liabilities to pay off. Furthermore, this feature creates additional excess spread, as the same collateral pool is now supporting fewer rated notes. An auction will only fail to the extent that the collaterals value is less than the principal amount of the notes. Therefore, a failed auction would likely only occur if there is substantial credit deterioration, which would include widespread TruPS collateral spread widening (systemic) and/or significant deferrals or defaults (idiosyncratic). Furthermore, any fixed rate secondary bonds purchased at a premium could decline in value if the issuers call option (which the investor is short) goes in-the-money due to a spread and/or rate rally. However, this call-related optionality risk is minimal in todays tight spread and low rate environment, making it unlikely for spreads to tighten or rates to rally enough to behoove issuers to call their TruPS issues. In the cash flow sections, we evaluate the average life sensitivities across the capital structure to determine relative extension risk in the tranches.
10
Trust Preferred CDOs: A Primer 11 November 2004 n Interest Rate Hedging Considerations To date, there has been minimal fixed/floating interest rate mismatch in TruPS CDOs, as the majority of both the assets and liabilities are floating rate. The mismatch is so minimal (or nonexistent) in some transactions that no hedge program exists. To the extent that there has been hybrid collateral (TruPS that are fixed for a specific period and then switch to floating), most CDOs have a similarly-priced subordinate hybrid liability, offsetting much of the interest rate mismatch. In general, the mismatch to hedge in these transactions has been small compared with other cash CDO sub-sectors, enabling some TruPS CDO issuers to easily afford hedges that are cancellable at the first auction call date. Cancellable calls are not economically feasible in other types of CDOs, such as structured finance CBOs, due to the need for a large hedge to account for the predominantly fixed rate assets versus floating rate liabilities. The need for limited hedging is a strong positive for a CDO, as out-of-the-money hedges and/or over-hedging has exacerbated poor performance within other CDO sub-sectors. n Credit Enhancement
Subordination
The conservative manner in which ratings agencies define bank defaults leads to a conservative sizing of CDO subordination. The analysis includes several factors; 1) defaults, 2) conservative recovery amounts and timing, 3) prepayment stresses, 4) geographic concentrations and 5) single obligor concentrations (see Appendix 1 for summary of Rating Agency Methodologies). Given the limited predictability of the collateral cash flows, the agencies stress their assumptions for collateral performance to address the potential variability in cash flow timing. Thus, TruPS CDOs can have fairly high required subordination levels, as these stresses account for the limited historical performance of the collateral. In our opinion, these conservative subordination levels will likely fall over time.
Excess Spread
Another unique aspect of TruPS CDOs is the amount of excess spread that can potentially be generated over the life of a transaction and can cushion against losses. The non-amortizing nature of the long-dated collateral can generate high excess spread relative to other CDO sectors. The same features mean that excess spread is diminished only by prepayments and defaults, which can be difficult to predict. This is where cash flow scenarios (our next section), with varying prepay and default assumptions, can add value to the analysis of a TruPS CDO. The rating agencies typically stress the CDO cash flows using front-loaded default stresses, which place pressure on excess spread early in the life of the transaction.
11
4.50%
Price 3mL +48 3mL +70 3mL +70** 3mL+1.65% 3mL+1.65%*** 3mL+1.65%****
*Subordination based on $400mm of collateral supporting the notes. ** Pays a FX coupon of 5.60% for the first 10 years. *** Pays a FX-rate coupon of 5.71% for the first 5 years. **** Pays a FX-rate coupon of 6.60% for the first 10 years.
The class A1 notes are first in priority for interest and principal. The A2 and A3 classes receive pro rata interest and principal distributions following the A1 notes. The class B notes are pro rata amongst themselves. n Default sensitivities The senior AAA class can withstand a constant annual default rate (CADR) of 11.9% before breaking its yield. The junior AAA and A classes can withstand CADRs of 8.5% and 2.4%, respectively. These CADRs compare favorably with the 1970-2003 historical average of 38 bp (peak of 204 bp). The coverage multiple above the historical average failure rate that the senior AAA- and A-rated tranches can withstand before breaking yield is 29.8x and 5.9x, respectively (Table 3).
Table 3: Breakeven Default Rates (based on a break in yield)
Tranche A1 A2 & A3 notes B notes
Source: Merrill Lynch
* CADR - Constant Annual Default Rate. Assumes a 10 year auction call, 10% CPR in year 6 and 2% therafter, 10% recoveries, 2-year recovery lag, forward LIBOR, no deferral of interest. ** Coverage of CADR over the 35-year annual average historic bank failure rate of 0.40%.
Multiple of 35- Year Historic Avg. Failure Rate** 29.8 21.3 6.0
n Average life sensitivities In addition to tranche sensitivities to collateral defaults, weighted average lives (WALs) can vary as a function of prepayments and defaults, as well as the timing of the auction call and other call and redemption features (e.g., extension risk). Our example deal has a feature beginning in year 10 that redirects 60% of the excess spread each period from the equity holder to amortize the senior-most outstanding class of notes. This amortization feature reduces the WAL of the notes in the event of a failed auction call in year 10 and greatly minimizes the extension risk in the senior AAAs. For example, the WAL of the AAAs increases only 0.43 years if we assume a 30-year auction call scenario instead of a successful auction in year 20 (Table 4).
12
Table 4: WAL with Base Case Prepayments and Varying Auction Call Dates*
Class A1 A2 B Auction Call Year 10 8.8 10.0 10.0 Auction Call Year 20 13.0 20.0 20.0 Auction Call Year 30 13.4 28.6 30.0
Source: Merrill Lynch * Assumes 10% CPR in year 6 and 2% thereafter, 40 bp CADR, 10% recoveries, 2-year recovery lag, forward LIBOR, no deferral of interest.
In addition to the auction timing scenarios above, we examine WAL sensitivity for various prepayment scenarios under both a 10-year and 30-year auction call scenario (Tables 5 and 6). We begin with our base case prepayment assumptions, including 10% prepayment rate in year 6, the first year following the non-call period. We then vary both the initial prepayment in year 6 and the ongoing annual prepayments over the life of the transaction. We chose a year 6 lump sum prepayment speed assumption as high as 50% to take into account the risk, however remote, that opportunistic refinancing could be the result of extensive market-wide spread tightening, thus affecting a large portion of the pool.
Table 5: WAL Sensitivity to Prepayments by Tranche Year 10 Auction Call*
Class A1 CPR** 2% 4% 6% Class A2 CPR** 2% 4% 6% Class B CPR** 2% 4% 6% 10% 10.0 10.0 10.0 10% 10.0 10.0 10.0 10% 8.8 8.5 8.3 Prepayment Speed in Year 6 20% 7.9 7.7 7.5 Prepayment Speed in Year 6 20% 10.0 10.0 10.0 Prepayment Speed in Year 6 20% 10.0 10.0 10.0 30% 10.0 10.0 10.0 40% 10.0 10.0 10.0 50% 10.0 10.0 10.0 30% 10.0 10.0 10.0 40% 10.0 10.0 9.9 50% 9.5 9.1 8.7 30% 7.1 6.9 6.7 40% 6.2 6.0 5.9 50% 5.5 5.5 5.5
Source: Merrill Lynch * Assumes 40 bp CADR. ** Assumed annual prepayment speed following year 6.
13
Source: Merrill Lynch * Assumes 40 bp CADR. ** Assumed annual prepayment speed following year 6.
Not surprisingly, moving down the capital structure, the tranche WALs are increasingly less responsive to prepayments. Although relatively insensitive in the year 10 auction call scenario, the WAL of the junior AAA class does shorten meaningfully with increases in the year 6 lump-sum prepayment assumption (Table 5). In contrast, the WAL of the subordinate-most note demonstrates limited sensitivity in the 30-year auction call scenario. Although a 30-year auction call is exceptionally unlikely, the stresses provide insight into how much more sensitive all three tranche WALs could be to calls at the extreme (Table 6). n Equity IRR Sensitivities Equity IRRs for TruPS investors have fallen since the products inception, similar to most other CDO asset classes, as collateral spreads tighten to historical lows. Despite tighter TruPS spreads, TruPS CDO equity currently provides the highest IRR of the three major cash CDO sub-sectors (see the relative value section). CDO equity returns are robust regardless of the assumptions made for the auction call timing (Chart 9). Even in our worst-case prepayment scenario for equity, where 100% of the pool prepays in year 6, the equity still has positive returns up to a 1.2% CADR, three times the historical average annual bank failure rate.
14
15.00%
10.00% Return
5.00%
0.00% 0.00% -5.00% 0.20% 0.40% 0.60% 0.80% 1.00% 1.20% 1.40%
Relative Value
n Debt Tranches TruPS CDOs have the highest spread among all new issue cash CDO subsectors, yet this four-year old asset class has yet to experience a single downgrade. Unlike competing CDO subsectors, we believe that investors should achieve a comfort level with the deals extension risk, more so than principal loss risk. In other words, significant collateral losses are far more likely to extend the tranches average life than to touch principal, a function of the relatively long-term non-amortizing nature of the collateral. In AAAs, TruPS CDOs have a 20 bp pickup to HY CLOs and a 29 bp pickup to the average of three non-CDO competing sectors (Chart 10). In As, the pickup is a respective 60 bp (versus HY CLOs) and 104 bp (versus our composite of three competing sectors) (Chart 11).
Chart 10: AAA-rated TruPS CDOs Versus Competing Spreads
100 Libor Spread (bps) 90
Libor Spread (bps)
80
80 70 60 50 40 30 May-02 Jul-02 May-03 Jul-03 Sep-03 Nov-03 Jan-04 Mar-04 May-04 Jan-02 Mar-02 Sep-01 Nov-01 Sep-02 Nov-02 Jan-03 Mar-03 Sep-04 Jul-01 Jul-04
60
40
20
Jul-01
May-02
Jul-02
May-03
Jul-03
May-04
Nov-01
Nov-02
Nov-03
Sep-01
Sep-02
Sep-03
Jul-04
7yr SL
HY CLO
Tailored IG Synth.
Multisector
Hybrid TruPS
10 yr CMBS
TruPS CDO
10yr CC
3yr HEQ
Sep-04
Jan-02
Jan-03
Mar-02
Mar-03
Jan-04
Mar-04
15
150
100
50
May-02
May-03
May-04
Mar-02
Jan-03 Mar-03
Nov-01
Nov-02
Nov-03
Sep-01
Sep-02
Sep-03
Mar-04
HY CLO
Tailored IG Synth.
Multisector
Hybrid TruPS
CMBS
TruPS CDO*
Corp
5yr HEQ
We attribute the TruPS CDO spread pickup to the following four factors: 1) Relative investor sponsorship: Few investors are familiar with this collateral relative to most CDO subsectors as it was not possible to invest in the debt of mid-market banks and insurers prior to 2000. In fact, other banks and insurers were, not surprisingly, the first investors across the capital structure of TruPS CDOs. 2) Extension risk: Although the average life is ten years at the CDO level, the thirty-year legal final of the TruPS collateral can exceed maturity limits for some investors. TruPS CDOs address the long maturity of the collateral with periodic auction calls and delevering of the structure in case of repeated failed auctions. 3) Prepayment risk: Prepayments (e.g., calls) could reduce excess spread resulting in a riskier pool to the extent that the best credits are prepaid first. Although this could have an adverse effect on the rating of the notes, we believe that the market is overly backward looking when it comes to predicting future prepayments. Given the current spread environment, interest rate driven prepayment risk is not significant. Instead, the main catalyst for prepayments would be credit driven opportunistic refinancing. However, as the pooled TruPS issuers have favorable credit quality when they enter the transaction and spreads are at historical tights, mergers are the more likely catalyst to cause prepayments. However, as bank failures often result in a merger arranged by the regulator, we would much prefer a called bond to default. 4) Issuers option to defer dividends: The issuer has the right to defer TruPS dividends for up to five years. Although the deferred dividends are capitalized and would eventually be paid should payments resume, deferral is clearly a negative for the issuers creditworthiness. However, three mitigating factors exist to protect debt holders against deferrals: Conservative Definition of Default: Debt holders are protected from this dividend deferral risk by the rating agencies conservative treatment of any deferral as a default in the O/C test. Consequently, if enough deferrals exist to trip the O/C test, the senior-most note holders could begin amortizing before any defaults occur. Recovery for O/C Test Purposes: For purposes of the O/C test, recoveries are assumed to be a conservative 2%-5%. Reserve Accounts: TruPS CDOs have reserve accounts that collect a portion of excess spread so as to continue paying debt coupons in case deferrals are significant enough to erode regular CDO debt coupon payments.
16
Sep-04
Jul-01
Jul-02
Jul-03
Jan-02
Jan-04
Jul-04
Trust Preferred CDOs: A Primer 11 November 2004 n Equity Tranche As with all CDO subsectors, TruPS CDO equity IRRs have fallen to historical lows over the past few years, although the week-to-week IRR volatility is relatively muted. Our CDO equity IRR barometer measures the relative attractiveness of buying the equity at any point in time. The barometer represents the excess spread between the sectors asset and liability yield adjusted for expected losses and expenses and then levers this excess spread to the gearing in a generic TruPS CDO. The main drivers of week-to-week changes in the barometer, therefore, are changes in asset and liability spreads. Over the products history, the TruPS CDO equity IRR barometers have fallen 7.8 percentage points to approximately 13% under base case assumptions of a 48 bp CADR (assuming 75% banks and 25% insurer collateral) and 10% recoveries and to about 7% with a stressed case assumption of a 95 bp CADR and 0% recovery rate (Chart 12).
Chart 12: TruPS CDO Equity IRR Barometer: Base Vs. Stressed Case
20
10
0 Apr-01 Apr-02 Apr-03 Jan-01 Jan-02 Jan-03 Jan-04 Apr-04 Jul-01 Jul-02 Jul-03 Oct-01 Oct-02 Oct-03 Jul-04
13.0
Source: Merrill Lynch, Cohen Brothers CADR = Constant Annual Default Rate. Note: Collateral = 75% Bank TruPS and 25% Insurance TruPS
We have seen a relatively greater tightening in asset spreads as compared to liability spreads over this period. Specifically, bank and insurer TruPS spreads have tightened a respective 105 bp and 75 bp; whereas AAA and A liability spreads have tightened a lesser 48 bp and 45 bp.
Chart 13: Monthly TruPS CDO IRRs Versus Volume
18.7
20
18.6 17.7 $1.3 $1.1 17.6 17.5 16.2 14.8 13.7 $0.9 13.0 12.9
Oct-04
15 10 5 Base case IRR
1.0
$0.9
0.5
$0.3
$0.4
$0.4
$0.3
0.0 Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov
17
The barometers are intended to help investors detect relative value amongst CDO sub-sectors as well as other equity-like securities and to forecast CDO volume in the relevant sector. The 13%-area base case IRR for TruPS CDO equity is significantly higher than IRRs in the two competing cash CDO sub-sectors, which have recently ranged between 11% and 12%. This years steady decline in TruPS CDO equity IRRs, particularly since May (17.5%), has been associated with a falloff in volume (Chart 13). As defaults in TruPS collateral, if any, are expected to be back-ended (we have seen only deferrals so far in this products history), investors receive particularly front-loaded cash flows in this CDO equity product. However, to the extent that there are enough deferrals in the collateral to trip an O/C test, equity cash flows would be cut off until collateral dividend payments resume sufficiently correcting the O/C test.
18
5.00% 4.50% 4.00% 3.50% 3.00% 2.50% 2.00% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Small Banks Large Banks
Source: FDIC
n Profitability Return on assets and equity ROA has stabilized from the early 1990s, with small banks outperforming larger banks until fairly recently, when ROA performance fell slightly behind. Large banks have more tools, such as securitization, to pare down their assets and the ability to do more fee income business, without crimping profitability. Thus, small banks often have higher NIMs but lower ROAs and ROEs (Chart 15). Furthermore, large banks tend to have particularly outsized ROEs because they focus more extensively on artificially deflating their equity base.
Chart 15: Bank ROA and ROE
1.40% Net Operating Income/Total Assets 1.30%
Net Income/Equity Capital 16.00% 15.00% 14.00% 13.00% 12.00% 11.00% 10.00% 9.00%
1.20% 1.10% 1.00% 0.90% 0.80% 0.70% 0.60% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Small Banks Large Banks
8.00% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 small banks large banks
Source: FDIC
Source: FDIC
19
Trust Preferred CDOs: A Primer 11 November 2004 n Loan performance Another key component of bank analysis involves an examination of the loan books performance. It is important that loans are growing in tandem with deposits, and that there are adequate loan loss reserves to shield against potential charge-offs. Examining the history of loan delinquencies and charge-offs provide insight into banks underwriting standards and can be used to judge the adequacy of loan loss reserves. Loan quality for small banks, as measured here by % net charge-offs, has stabilized since 1994, and has been superior to that of larger banks (Chart 16). Charge-offs at large banks can be adversely impacted by lumpier exposures to individual corporations, particularly during the 2001-2002 corporate recession. Additionally, the ratio of nonperforming loans to total assets can be used as a leading indicator of future losses. This ratio has improved significantly since the early 1990s, and has been stronger for small banks, who were not significantly effected by the large number of corporate defaults in the recent recession.
Chart 16: Charge-Offs/Assets and Nonperforming Loans/Assets
1.20% Net Charge-Offs/Total Assets 1.00% 0.80% 0.60% 0.40% 0.20% 0.00% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Small Banks Large Banks
2.50% 2.00% 1.50% 1.00% 0.50% 0.00% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Small Banks
Large Banks
Source: FDIC
Source: FDIC
n Liquidity and core deposits Liquidity, as measured by core deposits to total assets, has stabilized for banks in general since the early 1990s. Core deposits represent a significant portion of midsize banks borrower base, and thus tend to be higher for small banks (Chart 17).
Chart 17: Core Deposits/Assets
110% Core Deposits/Total Assets 100% 90% 80% 70% 60% 50% 40% 30% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Small Banks
Large Banks
Source: FDIC
20
Trust Preferred CDOs: A Primer 11 November 2004 n Capital adequacy Capital adequacy is an important gauge for banks, as a strong capital base provides stability to a financial institution. The ratio of equity capital to total assets has steadily improved for banks since the early 1990s (Chart 18). Historically, smaller banks have maintained greater capital adequacy versus large banks.
Chart 18: Capital/Assets
12%
11% 10% 9% 8% 7% 6% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Small Banks
Large Banks
Source: FDIC
n CRE Lending Finally, understanding the banks core lending business is a key consideration. For smaller banks, commercial real estate (CRE) lending comprises the bulk of the loan portfolios risk and return, although residential mortgages are the single largest loan category notionally. Bank failures of the late 1980s/early 1990s were primarily due to banks and thrifts aggressive underwriting standards for CRE lending. We believe that history has taught banks to rely more on cash flow generation than strictly collateral and to require larger upfront equity deposits. Additionally, the outlook for CRE, according to the forecasts of Torto Wheaton Research, is positive, in terms of both vacancy rates and rents (Chart 19).
Chart 19: Torto Wheaton Rent and Vacancy Rate Trends: Historical and Projected
20 Forecast 18 16 14 Vacancy rate 12 10 8 6 4 2 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2005 2007 2009 2011 2013 -1 -2 -3 -4 -5 3 2 1 0 Real rent growth (%) 4
Source: Torto Wheaton Research (TWR) Note: office and industrial two of the six CMBS property types are the only property types whose vacancy and rental data goes past the 1988-91 property recession.
21
One can view investing in bank TruPS as an indirect investment in CRE. Given this, we think there is an interesting comparison to be made with CMBS. Cumulative losses for early CMBS transactions were stressed conservatively, using default rates that are higher than the lofty ones experienced during the early1990s real estate recession. According to Fitch, excellent performance in the CMBS sector since inception, historically low mortgage rates, the inclusion of large loans in conduit transactions and improved collateral and sponsorship quality has led to lower rating agency credit enhancement requirements for CMBS, a trend that will likely continue. Similar to the higher subordination requirements at the inception of the CMBS market, subordination offered in the TruPS CDO market is arguably conservative. As agency-required subordination for CMBS has declined (e.g., in 1995, BBBs required roughly 16% of hard support versus 5% today) in response to lower-thanexpected defaults, we would expect TruPS CDO subordination to similarly decline over time, as losses remain nonexistent or insignificant. We believe that the geographic diversity offered in TruPS CDO bank pools significantly mitigates potential bubbles in certain commercial property markets, such as Atlanta, Dallas, and Los Angeles. Still, a 40% CRE loan portfolio in a small bank portfolio is generally riskier than the same 40% CRE concentration in a super-regional bank, which is more diversified across geography. In sum, commercial banks have experienced low historical failure rates, and small banks have experienced steady improvement in overall loan quality, liquidity and capital adequacy, particularly in comparison with larger banks. While there are many advantages for smaller bank TruPS issuers to seek financing via CDOs, there is a limited supply of issuers, and available collateral, in the market. The limited supply of available collateral only further exacerbates the pressure on bank TruPS spreads. The increasing scarcity of bank collateral, investor demand for diversification and need for yield in a generally tight spread environment has led to an increased inclusion of insurance TruPS in bank-dominated TruPS pools, as we discuss below. n Brief Overview of the US Insurance Sector Over the past four years, the nations highly regulated US insurance industry has faced significant obstacles including a recession, the terrorist attacks of 9-11, a weak stock market and a continuation of weak earnings fundamentals. Fortunately, since 2000 and particularly in the aftermath of 9-11, premiums have been increasing at a double-digit pace to keep pace with these costs. Prior to 2000, it was typical that premium rate increases would barely keep pace with inflation, much less loss rates. Still, P&C impairments have more than doubled from a 1994-1999 average of 54 bp to a 2000-2002 average of 122 bp, whereas L&H defaults have converged to zero. The most recent shocks for the insurance sector have been a wave of hurricanes in Florida and the initial announcement of the Spitzer probe, which caused the Merrill Lynch insurance index to widen 20 bp in the ensuing few weeks. The relatively low average P&C impairment rate (54 bp) in the 1994-99 period was buoyed by strong capital markets, a lack of major catastrophes and overcapitalization, which was the result of strengthened regulatory oversight and the introduction of risk-based capital standards. The insurance industry, particularly the commercial P&C segment, faces many challenges today that can best be summarized by rating agency outlooks by line of business (Table 7). For an overview of pure insurance TruPS CDOs and a general description of the collateral, we refer our readers to our June 2003 report, Insurance Trust Preferred CDOs.
22
Our Fixed Income Research insurance analyst, Angelo Graci, maintains an Overweight recommendation on the Insurance sector, which was reiterated in his October 25, 2004 report, Trade Ideas Amidst Insurance Sector Storm. In this report, Angelo writes, The Insurance sector is more likely to outperform the broad index over the next few months due to: 1) The most significant Spitzer allegations (and greatest risks) should remain concentrated with the large insurance brokers due to the nature of the conflicts of interest, 2) The rating implications on the insurance companies should remain modest due to our expectations for manageable financial impact and minimal impact on business models, 3) The sector, overall, is very likely to remain a high quality sector. The average rating in the Insurance sector is mid single-A. We expect the average rating to remain in the single-A category.
23
Conclusion
TruPS CDOs offer the highest spread among all cash CDO sectors. On an expected loss-adjusted basis, they are particularly attractive, as this sector has experienced no downgrades and no collateral defaults over its four year existence. While historical losses for middle market banks and insurers are more comparable to crossover corporate credits, subordination in TruPS CDOs is more comparable to that of a HY CBO backed by a BB/B portfolio of corporate bonds. Consequently, we think that there is more than adequate structural protection against losses in TruPS CDOs. Although the longer maturity of the collateral vis--vis the CDO tranche tenor implies average life uncertainty, various structural mitigants provide ample protection against extension risk. Significant collateral losses, if any, are far more likely to extend the tranches average life than to touch principal. TruPS CDOs offer a compelling and unique opportunity for investors to gain unlevered (AAA) or levered (AA or below) exposure to middle market banks and insurer debt. For a summary of each of the three rating agencys methodologies for TruPS CDOs, please see the appendix in our eminent release of our standalone TruPS CDO Primer. Furthermore, in this same standalone primer report, we will have full details on all TruPS CDO issued to date.
The author would like to thank Katie Lynch, CFA, Vice President, CDO Research Analyst for her contributions.
24
25
Each bank is run through Moodys Financial Institutions model, which considers various factors for each bank issuer, including earnings, capitalization, and concentration in commercial real estate assets, among other financial variables. Moodys assumes that the credit quality of the pool used to determine default probability is the higher of 1) a WARF of 480 (Baa2/Baa3), as corresponds to Moodys corporate bond default rate and their analysis of historical FDIC default data on insured commercial banks and 2) the portfolio average WARF determined by their model. The model is adjusted for barbelling, excessive issuer concentration, and potentially unstable default estimates. n Default and Recovery Assumptions The higher of the two WARFs, as described above, determines the pool-wide default rate. In order to arrive at a weighted average rating, recovery rates are then estimated. Moodys uses what they describe as a stressed recovery rate assumption of 5% to be applied to each issuer default, lagged five years, which is consistent with how long a TruPS can defer interest without causing an event of default. In modeling default timing, Moodys extends its conventional CDO default curve in order to account for the longer dated assets in the pool. Specifically, defaults are front-loaded over the first eight years of the transaction, with 35% in the first year and 9.3% in the remaining seven years. The exceptions are their 5-year 100% prepayment and 10-year auction call scenarios, where the default curve is compressed to fit into the shorter time frame. n Collateral Diversity Moodys considers diversity by region, citing FDIC historical default data that suggests strong regional patterns in bank default data. The Moodys Diversity Score Methodology is used, whereby the US is divided into five regions. The assumption is that banks are highly correlated intra-regionally but are somewhat uncorrelated inter-regionally. With respect to single-obligor concentration within the pool, Moodys typically limits exposure to any unrated bank to 3% or less. The agency will typically apply additional stresses to unrated banks with concentrations in excess of this. For rated banks, a higher concentration limit will be allowed, generally up to 5%. n Prepayments To account for the lack of history and potential across the pool for prepayments, Moodys assumes 100% pool prepays using 5-year, 10-year, and 30-year final maturities.
26
27
Appendix 2
TruPS CDO Issuance History
Date Offered 03/31/2000 09/07/2000 02/22/2001 03/28/2001 06/28/2001 07/31/2001 11/28/2001 12/18/2001 03/20/2002 03/26/2002 06/24/2002 07/01/2002 09/20/2002 10/16/2002 10/29/2002 11/18/2002 12/19/2002 12/19/2002 02/19/2003 03/18/2003 03/31/2003 04/23/2003 05/15/2003 05/23/2003 05/27/2003 06/26/2003 08/27/2003 09/12/2003 09/18/2003 9/25/2003 9/26/2003 10/29/2003 12/02/2003 12/05/2003 12/12/2003 02/05/2004 02/05/2004 02/26/2004 03/09/2004 03/18/2004 03/26/2004 04/08/2004 04/23/2004 04/30/2004 05/07/2004 06/17/2004 06/24/2004 08/19/2004 09/16/2004 10/01/2004 11/02/2004 Issuer Regional Diversified Funding I Preferred Term Securities Preferred Term Securities II MMCapS Funding I MM Community Funding Preferred Term Securities III MM Community Funding II Preferred Term Sec. IV Preferred Term Sec. V MM Community Funding III Preferred Term Sec. VI T-PREF Funding Preferred Term Sec. VII TPREF Funding CDO II Trapeza I I-Preferred Term Sec. I Preferred Term Sec. VIII T-PREF Funding III Trapeza II Preferred Term Sec. IX MM Community Funding IX Tropic CDO I I-Preferred Term Securities II InCaps Funding I Trapeza III Preferred Term Securities X Dekania CDO I Pref. Term Securities XI Alesco Preferred Funding I Tropic CDO II Trapeza IV I-Preferred Term Securities III Trapeza V InCaps Funding II Alesco Preferred Funding II Regional Diversified Funding 2004-1 Tropic III U.S. Capital Funding I Preferred Term Securities 13 Alesco III Dekania II Trapeza VI Icons Alesco Preferred Funding IV I-PRETSL IV Preferred Term Securities 14 U.S. Capital Funding II Alesco Preferred Funding V Preferred Term Securities XV Trapeza VII Tropic CDO IV Manager Salomon Smith Barney FT&KB First Tennessee Sandler ONeill Sandler ONeill First Tennessee Sandler ONeill First Tennessee FT&KB Sandler ONeill FT&KB Bear/Sandler/SSB FT&KB Bear/Sandler/SSB Trapeza Capital Mgt. FT&KB FT&KB Bear/Sandler/SSB Trapeza Capital Mgt FT&KB SSB/Sandler BS FT&KB Citigroup Trapeza FT&KB Dekania Capital Mgt. FT&KB Cohen Brothers BS Trapeza FT&KB Trapeza Citigroup Cohen Brothers Citigroup Bear Stearns StoneCastle Advisors, LLC FTN/KBW Cohen Brothers Dekania Capital Trapeza Capital Morgan Stanley Cohen Brothers FTN/KBW FTN/KBW StoneCastle Advisors, LLC Cohen Bros FTN/KBW Trapeza Bear Stearns Underwriter Salomon Smith Barney FT&KB First Tennessee Barclays Salomon Smith Barney First Tennessee Salomon Smith Barney First Tennessee FT&KB Salomon Smith Barney FT&KB Bear/Sandler/SSB FT&KB Bear/Sandler/SSB CSFB FT&KB FT&KB Bear/Sandler/SSB CSFB FT&KB SSB/Sandler BS FT&KB Citigroup CSFB FT&KB ML FT&KB Merrill Lynch/Sandler BS CSFB FT&KB CSFB Citigroup ML Citigroup Bear Stearns Sandler ONeill FTN/KBW Merrill Lynch/Sandler Merrill Lynch CSFB Morgan Stanley Merrill Lynch FTN/KBW FTN/KBW Sandler ONeill Merrill Lynch FTN/KBW CSFB Bear Stearns Collateral Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Insurance TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Insurance TruPS Insurance TruPS Bank TruPS Bank TruPS Insurance TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Insurance TruPS Bank TruPS Insurance TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Bank TruPS Insurance TruPS Bank TruPS Insurance TruPS Bank TruPS Insurance TruPS Bank TruPS Bank TruPS Bank and Insurance TruPS 84% Bank and 16% Insurance TruPS Bank TruPS Bank TruPS $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ Size 241.70 311.34 347.45 293.50 525.00 516.00 766.25 927.30 513.00 539.50 500.00 581.90 532.10 578.00 337.20 344.00 534.10 269.50 411.90 522.50 314.00 310.00 523.10 385.60 316.83 580.80 306.90 669.60 344.10 344.00 421.80 520.60 308.50 313.10 348.60 363.50 345.00 210.35 539.10 362.80 414.70 361.70 286.00 415.50 337.50 504.00 349.00 365.00 625.00 350.00 335.00
28
Appendix 3
Trust Preferred CDO Issuance Database
Date Offered 03/31/2000 Issuer Regional Diversified Funding I Ltd. Manager/ Underwriter SSB Collateral Bank TruPs Class A Sub. Income Notes Size $ 225.0 $ 16.7 $ 241.7 $ 201.1 $ 90.0 $ 20.3 $ 311.3 $ 227.8 $ 93.0 $ 26.7 $ 347.5 $ 191.5 $ 77.0 $ 25.0 $ 293.5 $ 333.0 $ 140.0 $ 52.0 $ 525.0 $ 365.8 $ 114.1 $ 36.1 $ 516.0 $ 480.0 $ 205.0 $ 81.3 $ 766.3 12.8 Avg Life Mdys A1 NR S&P Fitch A+ NR Pricing
09/07/2000
FT&KB
Bank TruPs
Aaa Baa2
AAA BBB
8.79% 9.74%
02/22/2001
Bank TruPs
A B Equity
Aaa Baa2
AAA BBB
3mL+100 9.55%
03/28/2001
Bank TruPs
A B C
Aaa Baa2
AAA BBB
8.03% 9.48%
06/28/2001
MM Community Funding
Bank TruPs
A B Equity
Aaa Baa2
AAA BBB
07/31/2001
Bank TruPS
A B Equity
Aaa Baa2
AAA BBB
3mL+100 3mL+290
11/28/2001
MM Community Funding II
A B Equity
Aaa A3
AA+ NR NR
AAA ANR
12/18/2001
Bank TruPs
A B Equity
12.8
Aaa A3
AAA NR
AAA A-
29
30
Refer to important disclosures on page 40.
03/26/2002
95% new issue TruPS 5% secondary TruPS from banks and thrifts
A B C
30.0 30.0
Aaa A3
AAA NR
AAA A
06/24/2002
FT&KB
A B Income Notes
Aaa A3
AAA
AAA A+ NR
09/20/2002
FT&KB
$ 189.0 N/A $ 120.0 $ 177.9 $ 45.2 $ 532.1 $ $ $ $ $ 220.0 100.0 196.0 62.0 578.0
AAA AAA NR
AAA AAA A+
10/16/2002
Bear/Sandler/SSB Bear/Sandler/SSB
90% new issue TruPS, 5% 2ndary TruPS and 5% sub debt from banks and thrifts
A-1 A-2 B PS
NA NA NA
Aaa Aaa A3
AAA AAA NR
AAA AAA A-
11/18/2002
FT&KB
10.0
AAA
10.0
A-
12/19/2002
FT&KB
80% new issue TruPS 9% new issue surplus notes 11% sub. Debt from banks thrifts
Aaa Aaa A1 A2 A3
AAA AAA NR NR NR
AAA AAA A A A
12/19/2002
89% new issue TruPS 8% secondary TruPS, 3% sub debt from banks and thrifts
Aaa Aaa A2 A2
AAA AAA
AAA AAA A A NR
31
32
Refer to important disclosures on page 40.
03/18/2003
FT&KB FT&KB
TruPS from various banks and thrifts and some sub debts
03/31/2003
SSB/Sandler
TruPS from various banks and thrifts and subordinated bank debt
Aaa Aaa A3 A3
AAA AAA NR NR
04/23/2003
Tropic CDO I
TruPS from various banks and thrifts and subordinated bank debt
AAA AAA NR NR NR NR
05/23/2003
Citigroup Citigroup
Insurance TruPS (51%), surplus notes (35%) , and senior notes (14%)
Aaa
AAA AABBB
05/27/2003
Trapeza III
Trapeza CSFB
TruPS from various banks and thrifts Baa2/Baa3 WAR; 3.3% per issuer
AAA AAA NR NR NR NR
06/26/2003
FT&KB FT&KB
33
34
Refer to important disclosures on page 40.
09/12/2003
FT&KB FT&KB
Aaa Aaa A2 A2 A2
AAA AAA A A A
AAA AAA A A A
09/18/2003
Bank TruPS
A1 A2 B1 B2 PS
Aaa Aaa A3 A3
AAA AAA NR NR
9/25/2003
AAA AAA NR NR NR NR
9/26/2003
Trapeza IV
Trapeza CSFB
Bank TruPS
AAA AAA NR NR NR NR
12/02/2003
Trapeza V
Trapeza CSFB
Bank Trust-Preferred
AAA AAA NR NR NR NR
12/05/2003
Citigroup Citigroup
Insurance TruPS (45%), surplus notes (45%) , and senior notes (10%)
AAA AAA NR NR NR
Bank TruPS
Aaa Aaa A2 A2
AAA AAA NR NR
AAA AAA A A NR
Citigroup Citigroup
Aaa Aaa A3 A3
AAA AAA NR NR
35
36
Refer to important disclosures on page 40.
02/26/2004
Aaa Aaa A3 A3 NR
AAA AAA NR NR NR
FT&KB FT&KB
Bank TruPS
Aaa Aaa A3 A3
AAA AAA NR NR
03/26/2004
Dekania II
Insurance TruPS
Aaa NR NR NR NR NR NR
3mL+65 3mL+100 Not Offered 3mL+135 3mL+190 S+190 (5.118%) 3mL+340 7.61%
04/23/2004
ICONS
Insurance TruPS
Aaa Aa1 A3 A3 A3 NR
AAA AA A A A BBB
AAA AA A A A BBB
04/30/2004
Alesco IV
A1 A2 A3 B1 B2 B3 Equity
05/07/2004
I-PRETSL IV
FTN/KBW FTN/KBW
Insurance TruPS
37
38
Refer to important disclosures on page 40.
06/24/2004
$ 171.0 $ 33.5 $ 70.0 $ 40.0 $ 34.1 $ 348.6 $ 189.0 $ 42.0 $ 10.0 $ 42.5 $ 37.6 $ 4.5 $ 6.3 $ 33.0 $ 364.8 $ 323.1 $ 63.4 $ 15.0 $ 114.5 $ 22.0 $ 36.0 $ 10.0 $ 41.4 $ 625.4 8.7 10.0 10.0 10.0 10.0 10.0 10.0
Aaa Aaa A3 A3 NR
AAA AAA NR NR NR
08/19/2004
Bank TruPS
AAA AAA AA NR NR NR NR
09/16/2004
FTN/KBW FTN/KBW
11/02/2004
Tropic CDO IV
Bear Stearns
39
Analyst Certification
I, Lang Gibson, hereby certify that the views expressed in this research report accurately reflect my personal views about the subject securities and issuers. I also certify that no part of my compensation was, is, or will be, directly or indirectly, related to the specific recommendations or view expressed in this research report.
Important Disclosures
Copyright 2004 Merrill Lynch, Pierce, Fenner & Smith Incorporated (MLPF&S). All rights reserved. Any unauthorized use or disclosure is prohibited. This report has been prepared and issued by MLPF&S and/or one of its affiliates and has been approved for publication in the United Kingdom by Merrill Lynch Pierce, Fenner & Smith Limited, which is authorized and regulated by the Financial Services Authority; has been considered and distributed in Australia by Merrill Lynch Equities (Australia) Limited (ABN 65 006 276 795), licensed under the Australian Corporations Act, AFSL No 235132; has been considered and distributed in Japan by Merrill Lynch Japan Securities Co, Ltd, a registered securities dealer under the Securities and Exchange Law in Japan; is distributed in Hong Kong by Merrill Lynch (Asia Pacific) Ltd, which is regulated by the Hong Kong SFC; and is distributed in Singapore by Merrill Lynch International Bank Ltd (Merchant Bank) and Merrill Lynch (Singapore) Pte Ltd (Company Registration No. 198602883D), which are regulated by the Monetary Authority of Singapore. The information herein was obtained from various sources; we do not guarantee its accuracy or completeness. Neither the information nor any opinion expressed constitutes an offer, or an invitation to make an offer, to buy or sell any securities or any options, futures or other derivatives related to such securities (related investments). Officers of MLPF&S or one of its affiliates may have a financial interest in securities of the issuer(s) or in related investments. This research report is prepared for general circulation and is circulated for general information only. It does not have regard to the specific investment objectives, financial situation and the particular needs of any specific person who may receive this report. Investors should seek financial advice regarding the appropriateness of investing in any securities or investment strategies discussed or recommended in this report and should understand that statements regarding future prospects may not be realized. Investors should note that income from such securities, if any, may fluctuate and that each securitys price or value may rise or fall. Accordingly, investors may receive back less than originally invested. Past performance is not necessarily a guide to future performance. The bonds of the company are traded over-the-counter. Retail sales and/or distribution of this report may be made only in states where these securities are exempt from registration or have been qualified for sale. MLPF&S usually makes a market in the bonds of this company. Foreign currency rates of exchange may adversely affect the value, price or income of any security or related investment mentioned in this report. In addition, investors in securities such as ADRs, whose values are influenced by the currency of the underlying security, effectively assume currency risk.
40