Cmbs Primer 5th Edition 4d1f2w

FIFTH EDITION

2005

Transforming Real Estate Finance A CMBS Primer

Primary Analysts:

Howard Esaki Marielle Jan de Beur Masumi Goldman

This book is an overview of the Commercial Mortgage-Backed Securities (CMBS) market. The contents of this publication are over eight years in the making and include excerpts of research reports from as early as 1997. In this fifth edition of our primer, we have reorganized the chapters to highlight the different investment options within CMBS. New material since our last edition includes sections on the various types of AAA CMBS classes, total rate of return swaps, floating rate large loan transactions, and an updated version of the commercial mortgage default study. We hope you find this book useful and welcome comments so that we can improve future editions.

FIFTH EDITION

2005

Transforming Real Estate Finance A CMBS Primer

Primary Analysts: Howard Esaki Marielle Jan de Beur Masumi Goldman The Primary Analyst(s) identified above certify that the views expressed in this report accurately reflect his/her/their personal views about the subject securities/instruments/issuers, and no part of his/her/their compensation was, is or will be directly or indirectly related to the specific views or recommendations contained herein. This report has been prepared in accordance with our conflict management policy. The policy describes our organizational and istrative arrangements for the avoidance, management and disclosure of conflicts of interest. The policy is available at www.morganstanley.com/institutional/research. Please see additional important disclosures at the end of this report.

450

Morgan Stanley & Co. Incorporated December 16, 2004

Transforming Real Estate Finance

A CMBS Primer 1

INTRODUCTION T O C MBS

History and Structure Rating Agency Methodology

2

3

MAJOR P ROPERTY T YPES I N C MBS

CALL P ROTECTION

41 – 5 3

AAA F IXED R ATE C MBS

TOTAL R ATE O F R ETURN S WAPS

Mechanics of the Swap Break-even Analysis

6

CMBS I O s

Overview PAC & Levered IO Age-Adjusted Default Analysis Pricing Benchmarks: E and J

7

FIXED R ATE L ARGE L OANS

Fusion vs. Single Asset Characteristics Rating Agency Approach

8

BBB C MBS A ND R EIT s

Overview Performance

9

CMBS C ONDUIT S UBORDINATION L EVELS

The Right Subordination Projecting Losses on BBB- CMBS Projecting Losses on BB CMBS Projecting Losses on B CMBS

10

9 – 39

10 12 14 16 18 20 22 24 26

Tight Window Bonds Multifamily Directed Bonds Amortizing Bonds Dollar Priced Bonds

5

2 4

Property Classifications Retail Multifamily Office Manufactured Housing Industrial Self Storage Health Care Hotel Types of Call Protection Allocation of Prepayment Penalties Relative Value Analysis

4

1 – 8

MULTIFAMILY M BS

Fannie Mae DUS and DMBS Ginnie Mae and Freddie Mac Programs

43 47 48 55 – 7 0

56 62 63 68 71 – 7 4

72 73 75 – 9 1

76 81 84 88 93 – 9 9

94 96 98 101 – 1 22

102 109 123 – 1 77

124 128 146 162 179 – 1 87

180 185

11

FLOATING R ATE L ARGE L OAN C MBS

Extension Risk Ratings Drift Available Funds Cap Mechanics

12

13

COMMERCIAL M ORTGAGE D EFAULTS

190 194 204 209 – 2 21

Characteristics of Defaulted Loans Loss Severity Study Results and Impact on CMBS

211 212 214

TRANSACTION M ONITORING

223 – 2 65

Monthly CMBS Delinquency Report Conduit Tracking Report Retail Study CMBS Rating Actions CMBS Loan Originators

14 15 16 17

189 – 2 07

224 228 234 242 252

EUROPEAN C MBS

267 – 2 75

JAPANESE C MBS

277 – 2 90

FACTORS T O C ONSIDER B EFORE I NVESTING I N C MBS

291 – 2 97

GLOSSARY

299 – 3 06

This Page Intentionally Left Blank

Chapter 1

Transforming Real Estate Finance

Introduction to CMBS

Please see additional important disclosures at the end of this report.

Transforming Real Estate Finance chapter 1

Introduction to CMBS Commercial Mortgage Backed Securities (CMBS) are bonds backed by pools of mortgages on commercial and multifamily real estate. As of December 2004, the global market capitalization of the CMBS market was slightly over $541 billion. About 1 in 5 commercial and multifamily mortgages are in CMBS, compared to a securitization rate of about 50% for residential mortgages. CMBS offer several advantages over commercial whole loans. Securitization allows for the division of the loan into credit classes so that an investor may buy a class rated from AAA to single-B and unrated. In addition, CMBS are marked to market on a daily basis and hence are more liquid than whole loans. CMBS appeal to a wide array of investors because of attractive relative spreads and stronger call protection than residential mortgage securities. DEVELOPMENT O F T HE M ARKET

Before the mid-1990s the U.S. real estate business was predominately a private market. Lending was dominated by a handful of banks, life insurance companies, and pension funds. Real estate ownership was regionally focused with ownership concentrated in a few hands. Families and private partnerships were the largest owners. Small and diversified investments in real estate were almost nonexistent. Real estate is a cyclical business moving through various stages of expansion, equilibrium, slow growth, and recovery. Private ownership of real estate and lack of public information fostered extreme real estate cycles. Development and lending was done on a regional basis with little national information. The steep real estate recession of the late 1980s and early 1990s was the worst since the Depression of the 1930s. Prices of commercial real estate fell by 50% or more in some areas and delinquency rates on loans soared to all-time highs. Losses on commercial loan portfolios led to the exit of many traditional lenders from the commercial mortgage market. Regulators and rating agencies turned more negative on commercial mortgage holdings, so that the remaining lenders became less willing to extend credit. CMBS: D IVERSIFIED P UBLIC R EAL E STATE

Low real estate values combined with the failure or exit of traditional lenders provided innovation opportunities and a shift from private to public ownership. REITs began buying undervalued real estate portfolios funded through public stock and bond offerings. REIT shares provided an opportunity for small, diversified investments in real estate. As with REITs, CMBS provide investment opportunities in diversified real estate pools. Investment banks started to apply securitization legal structures, and technology developed during the 1970s and 1980s for residential mortgage-backed securities to commercial mortgages. In the mid- to late-1980s, issuers securitized a few loans on single properties into CMBS. Packaging of diversified pools of mortgages into CMBS developed in the mid1990s when the Resolution Trust Corporation (RTC) pooled non-performing loans from failed institutions. Some of these transactions exceeded $1 billion and led to the growth in the investor base for CMBS. After the success of the RTC transactions, CMBS gained wider acceptance with investors and non-gov-

2

ernment, or “private-label”, issuers. Issuance of CMBS in the U.S. grew rapidly in the mid-1990s. In 2004, U.S. CMBS issuance is expected to reach an all-time high of nearly $90 billion. Outside the U.S., CMBS has also taken hold as a financing vehicle, with $18.5 billion issued in Europe in 2004. Most of the transactions are out of the United Kingdom, but deals have been done in several other countries. In Asia, Japan is facing an RTC-like situation, with distressed properties and lenders. In 2004, about $5 billion of CMBS came to market in Asia. STRUCTURES A ND R ATING A GENCIES

CMBS have very simple structures compared to their residential mortgage counterparts. The bonds are almost always sequential pay, with amortization, prepayments, and default recoveries paid to the most senior remaining class. The lowest-rated remaining class absorbs losses. Since commercial mortgages almost always have some form of prepayment penalty (Chapter 3), credit analysis plays a more important role than prepayment analysis. For residential MBS, prepayments are a much more important factor. CMBS are static pools of commercial real estate loans divided into tranches with varying subordination levels and credit ratings. A typical transaction has about 90% investment grade bonds concentrated in AAA securities, with the remaining 10% non-investment grade. Interest only (IO) bonds (Chapter 6) can be stripped off all or part of the structure. Structure

A typical fusion structure consists of sequential pay, fixed rate bonds. The AAA bonds are time-tranched and include a front-pay wide window bond. In addition to the front-pay class, the AAA CMBS market also includes tight window bullets, multifamily directed classes, amortizing bonds, super senior and junior AAAs. (See Chapter 4 for more information.) In an environment of steadily declining subordination levels, the most recent trend in CMBS issuance is the super senior deal structure. Prior to the emergence of this structure, which includes multiple super senior AAA bonds supexhibit 1

SAMPLE 20041 NEW ISSUE MARKET CMBS STRUCTURE

Based on Commercial Mortgage Default Study. See Chapter 12.

1

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

3

Transforming Real Estate Finance chapter 1

Introduction to CMBS ported by a 10-year junior AAA bond, AAA subordination levels in 2004 ranged in the low- to mid-teens. Issuers have been able to structure super senior AAA classes with credit enhancement as high as 20% by carving out a smaller, subordinate AAA class. The subordinate 10-year AAA has a lower credit enhancement level than the other AAA-rated bonds. In addition to the mortgage collateral, credit enhancements may be in the form of reserve funds, guarantees, letters of credit, cross-collateralization and crossdefault provisions. Loans within the pool may have certain cash control provisions such as a “lock box” that requires payments from tenants to go directly to the trust instead of through the borrower if certain default triggers occur. Virtually all loans within CMBS are bankruptcy remote. The Trustee, Master Servicer and Special Servicer each play an ongoing role in the transaction. The Pooling and Servicing Agreement, Prospectus, and other legal documents outline each party’s responsibilities and fees. Typically, the Trustee is responsible for reporting monthly payments and collateral performance data to certificate holders. The Master Servicer is responsible for servicing all performing loans and monitoring loan document requirements. The Special Servicer resolves defaulted or delinquent loan issues. RATING A GENCY M ETHODOLOGY

Before each CMBS is issued, the rating agencies review the collateral in the transaction and determine the tranche ratings and pool sizing. During the process the agencies review the property level cash flows, perform physical inspections, and run various stress analyses on the underlying cash flows. This section examines the rating process for conduits, or diversified pools of mortgages. A conduit originates loans for sale or securitization, and not for holding in a portfolio. When a conduit deal comes to market, the rating agency performs due diligence on a subset of the properties (typically between 35% and 75%). The larger loans in the deal are always underwritten while the remaining properties are chosen such that they provide a representative cross section of the deal. To determine credit enhancement levels, the underwritten cash flow (UCF) produced by each property is then assigned a “haircut” based on various subjective parameters. Following is a list of parameters that rating agencies consider when evaluating a CMBS conduit1: • Loan Specific

• Deal Specific

Property type

Number of loans in the deal

Loan-to-value ratio

Loan size

Debt service coverage ratio

Degree of subordination

Fixed/floating rate

Balloon extension risk

Loan seasoning Direct versus correspondent lending

Quality of master servicer and special servicer

• Real Estate Outlook

1

4

The parameters are similar to those considered for a non-conduit deal except for adjustments for loan diversification.

Loans collateralized by different property types are generally ranked in the following order (best to worst): regional malls, multifamily, anchored community retail, industrial, office and, finally, hotel. These rankings are based on historical defaults and cash flow volatility of the different property types. The variation of required credit enhancement levels with debt service coverage (DSC) and loanto-value (LTV) ratios is shown in Exhibit 2 for common property types. (The credit enhancement levels are from Duff and Phelps, which merged with Fitch in 2000. The levels are indicative only, and are different for each rating agency.) For example at 80% LTV and 1.15 DSC, a regional mall requires 30.1% credit enhancement for a AAA rating while an office property requires double that figure. The credit enhancement levels shown in Exhibit 2 are somewhat “sticky” with respect to a credit/real estate cycle. It is therefore sometimes possible to “arb” this fact by, say, buying conduit CMBS backed by office properties in an environment where the fact that office properties are doing well is not reflected in credit enhancement levels. Obviously, AAA securities are much less conducive to this kind of play than lower-rated classes.

exhibit 2

BASE-CASE CREDIT ENHANCEMENT GUIDELINES FOR VARIOUS PROPERTY TYPES

OFFICE Individual PROPERTIES Loan Coverage

Credit Enhancement

LTV

DSCR

AAA

AA

A

30 40 50 60 65 70 75 80

2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15

3.9 7.8 13.1 20.9 27.2 35.8 47.1 60.2

2.9 5.8 9.7 15.5 20.2 26.6 34.9 44.6

2.2 4.3 7.2 11.5 14.9 19.6 25.8 33.0

REGIONAL Individual MALLS Loan Coverage

BBB 1.5 3.0 5.0 8.0 10.4 13.7 18.0 23.0

BB

B

1.0 1.6 2.7 4.4 5.7 7.5 9.9 12.6

0.5 1.0 1.1 1.7 2.3 3.0 3.9 5.0

BB

B

0.5 1.0 1.4 2.2 2.8 3.8 4.9 6.3

0.5 0.5 1.0 1.0 1.1 1.5 2.0 2.5

Credit Enhancement

LTV

DSCR

AAA

AA

A

30 40 50 60 65 70 75 80

2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15

2.0 3.9 6.5 10.5 13.6 17.9 23.5 30.1

1.5 2.9 4.8 7.8 10.1 13.3 17.5 22.3

1.1 2.2 3.6 5.7 7.5 9.8 12.9 16.5

BBB 1.0 1.5 2.5 4.0 5.2 6.9 9.0 11.5

Source: The Rating of Commercial Mortgage-Backed Securities, Duff and Phelps Credit Rating Co.

Please see additional important disclosures at the end of this report.

5

Transforming Real Estate Finance chapter 1

Introduction to CMBS exhibit 2

BASE-CASE CREDIT ENHANCEMENT GUIDELINES FOR VARIOUS PROPERTY TYPES (CONTINUED)

INDUSTRIAL/ ANCHORED Individual RETAIL Loan Coverage PROPERTIES

Credit Enhancement

LTV

DSCR

AAA

AA

A

30 40 50 60 65 70 75 80

2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15

2.5 5.0 8.4 13.4 17.4 22.9 30.1 38.5

1.9 3.7 6.2 9.9 12.9 17.0 22.3 28.5

1.4 2.7 4.6 7.3 9.5 12.6 16.5 21.1

MULTIFAMILY Individual PROPERTIES Loan Coverage

BBB 1.0 1.9 3.2 5.1 6.7 8.8 11.5 14.7

BB

B

0.5 1.1 1.8 2.8 3.6 4.8 6.3 8.1

0.5 0.5 1.0 1.1 1.4 1.9 2.5 3.2

Credit Enhancement

LTV

DSCR

AAA

AA

A

30 40 50 60 65 70 75 80

2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15

2.6 5.2 8.7 13.9 18.1 23.9 31.4 40.1

1.9 3.9 6.5 10.3 13.4 17.7 23.3 29.7

1.4 2.9 4.8 7.6 9.9 13.1 17.2 22.0

BBB 1.0 2.0 3.3 5.3 6.9 9.1 12.0 15.3

BB

B

0.5 1.1 1.8 2.9 3.8 5.0 6.6 8.4

0.5 0.5 1.0 1.2 1.5 2.0 2.6 3.3

Source: The Rating of Commercial Mortgage-Backed Securities, Duff and Phelps Credit Rating Co.

Investors should also be concerned with the dispersion of DSC and LTV in the entire deal; that is, having all loans with a DSC of 1.5x is better than having 50% of the loans at 1.0x and the remainder at 2.0x. There may be some element of “gaming” credit- levels to the extent that Fitch uses discrete DSC and LTV buckets while the other rating agencies use a continuous variation of credit with DSC and LTV. However, this is usually mitigated by the fact that at least two rating agencies rate the investment grade classes of a CMBS. Given the volatility of short-term interest rates, an adjustable-rate loan is underwritten under an interest rate scenario that is substantially higher than current rates. Loans without a track record of consistent payments are also rated more conservatively than those seasoned at least five years. The origin of a loan, whether direct or via a correspondent, matters less than it previously did. Many subjective assessments also go into the rating process and Exhibit 2 shows the addition and subtractions that rating agencies apply to the credit level.

6

exhibit 3

Amortization Fully Interest Only Less Than 5 Years Seasoning Servicer/Special Servicer Assessment Quality Assessment Origination Information Cash Control Reserves

SUBJECTIVE ADJUSTMENTS TO CREDIT LEVELS Special Adjustment1 Additions Reductions

Other Factors Applied on a Case-By-Case Basis

20%

5% —

Floating Interest Rates Asset Quality

10-20%

—

10%

10%

Market Barriers to Entry Cash Flow Volatility Loan Size

20% 20% — —

5% 5% 10-20% 5-10%

Location Assessment Concentration

1 Adjustments are applied as a percentage of base-case credit enhancement levels; i.e., if base-case credit enhancement is 10%, and the adjustment factor is 20%, the adjusted credit enhancement is 12% (10 x 1.2).

Source: The Rating of Commercial Mortgage-Backed Securities, Duff and Phelps Credit Rating Co.

On a deal level basis, conduits have diverged most significantly from the traditional CMBS model in the degree of diversification provided by the large number of underlying loans. It is not uncommon to see 200 or more loans in a conduit. The rating agencies like to see at least 50 loans underlying a deal with no more than 5% of the deal (by dollar amount) in any one state. Any single loan should not constitute more than 5% of the deal. Credit- levels are often tested by defaulting the three largest loans. At the triple-A rating level, subordination levels are likely to be very similar across rating agencies. However, some degree of rating shopping is likely to occur for lower-rated pieces, given the many subjective aspects of the rating process. One of these aspects is the quality of the master and special servicers. The rating agency looks for a special servicer with a proven track record of real estate workouts. The rating agency would be concerned if a large number of loans came due at the same time. This is because loan documents typically allow for three one-year extensions, and this is not necessarily long enough to get through a credit downturn. Finally, the rating agencies evaluate the real estate environment and where we are in the credit cycle.

Please see additional important disclosures at the end of this report.

7

Transforming Real Estate Finance chapter 1

Introduction to CMBS RATING A GENCIES A ND I NVESTORS M ONITOR C ONDUIT Q UALITY

Some investors have expressed concerns that commercial mortgage conduits have become too aggressive in lending and that the quality of loans in CMBS conduit pools will deteriorate. Other investors worry that conduits will chase after loans backed by “B” or “C” -quality properties, leaving the top-tier loans to insurance companies. In our opinion, these fears are misplaced, since securitized loans must ultimately through the filters of both rating agencies and investors. If rating agencies recognize a slide in credit-quality rating, they will increase credit levels (or lower ratings). If investors perceive increased risk, they will demand wider spreads on the securities. Investors can make adjustments to a recognized drop in loan quality; it is the unrecognized slippage that is dangerous. As discussed in the section above, rating agencies apply published standards to loans pooled into a CMBS and adjust the result by making qualitative assessments. Almost all CMBS carry at least two, and many have three, ratings. Different rating agencies assign different levels of credit to obtain a given rating level. In order to avoid a split rating, an issuer must go with the most conservative collateral assessment. For example, if Fitch and Moody’s have AAA credit requirements of 23% and 25%, respectively, an issuer must use 25% enhancement to attain a dual AAA rating. (In a few instances, an issuer will go with the lower credit level and receive a split rating. Non-investment grade classes, however, are often rated by only one agency). If an issuer starts to raise LTVs, reduce DSC ratios, or lend on more risky property types or in more risky areas, rating agencies will respond by lifting credit levels. The rating agencies thus act as a first level of defense against a potential unobserved fall in loan credit quality.

8

Chapter 2

Transforming Real Estate Finance

Major Property Types in CMBS

Please see additional important disclosures at the end of this report.

9

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS While CMBS investors don’t need to be real estate experts, particularly at the AAA level, a general understanding of various real estate property types and terminology is helpful. In this chapter we explain the various types of real estate properties and fundamentals that impact their performance. Information in a transaction prospectus may contain descriptions of various assets as Class A, B, or C. Below is a description of those asset classes. CLASSIFICATIONS Class A

Newly built; higher quality finishes and prominent locations. Class B

Generic real estate; 10-20 years old, well maintained, average locations, fewer amenities. Class C

Older properties needing frequent capital investment; uncertain future. WHY D O P ROPERTY C LASSIFICATIONS M ATTER?

• Generally, it is believed that Class A properties are the least risky from a cash flow volatility standpoint. • The Class A properties set the rental rates in a market. They attract the most credit worthy tenants and, by definition, typically have the least deferred maintenance and the lowest risk of functional obsolescence. • In a market downturn these properties will have the highest demand for space, albeit at a lower rental rate.

10

class A

class B

class C

Please see additional important disclosures at the end of this report.

11

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Conduit CMBS transactions provide property type and geographic diversity. Currently, retail properties for about 30% of outstanding CMBS in the market, multifamily s for about 18%, office properties for about 24%, hotels for about 8%, and the remaining 20% is composed of manufactured housing, industrial, self-storage, and senior housing.

Retail Loans Typical Loan

• 1.4 DSCR • 65% LTV (based on a 9–10% capitalization rate) • $75–$125 Loan Per Square Foot Value

RELATED T ERMS Credit Tenant Lease

All payments guaranteed by credit of tenant (i.e., WalMart). Triple Net Lease

Tenant pays rent, real estate taxes, expenses, and maintenance. Go Dark Provisions

Prevents tenant from vacating the space while continuing to pay rent; landlords like this because this vacant space is a detriment to other stores’ sales at the center. Co-T Tenancy Provisions

Permits the tenant to cancel its lease if another major tenant closes. Recapture Provisions

Permits the owner to cancel a lease and to regain control of space after a tenant closes its store. TYPES O F R ETAIL P ROPERTIES Super Regional Mall

Over 1 million square feet with multiple department (4-5) stores as anchors. Regional Mall

Over 750,000 square feet with several department stores (2-3) as anchors. In-L Line Store

Smaller store within a center (i.e., Foot Locker or Hallmark Cards store). Community Center

Over 100,000–275,000 square feet of space; multiple anchors but not enclosed. Anchored Strip Center

Grocery or discount retailer attracts tenants to small stores that are adjacent. Shadow Anchored Strip

Same as the anchor strip, but the anchor is not part of collateral for the loan.

12

community center

anchored strip

power center/big box

unanchored strip

Unanchored Strip

No major destination type tenant. Usually smaller local tenants. A particular location must have natural traffic to be successful. The best assets are located in highly developed areas with little vacant land. Power Center/Big Box

Anchor tenants and some small stores. Typically big discounters or mass retailers. Examples include Circuit City, Best Buy, and Target. CONVENTIONAL I NVESTOR W ISDOM

• Retail stores in general are under competitive pressures from alternative distribution channels. • Aggressive competition for market share leading to construction of big box/ large store formats. • Super-regional malls are the least affected by the negative factors listed above. These malls are anchored by strong department stores that the in-line tenants. • Mid-market malls are viewed as under the most pressure from alternative retailing concepts. Anchor stores in these malls are less of a draw and compete directly with value-based retailers. • Grocery/drug anchor strips are life insurance company favorites. RATING A GENCY V IEW

• Generally received favorable treatment from rating agencies as a result of strong historical performance. • Preference for longer leases; nationally or regionally rated credit tenants.

Please see additional important disclosures at the end of this report.

13

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Multifamily Loans Typical Loan

• 1.25x DSCR • 75% LTV (based on a 8–9% capitalization rate) • $20,000–$60,000 Loan Per Unit

RELATED T ERMS Co-o op Loans/Blanket Loans

Very low loan to value loans. Loans senior to co-op share loans. Unit Mix

Desirable ratio of 1-bedroom versus 2-bedroom apartments depends on the market. Older complexes had higher proportions of 1-bedrooms; higher percentages of 2-bedrooms are now preferred, since they provide more flexibility to families and lifestyle renters. Reserves

Underwriting usually includes $200-300 per unit per year for new paint, carpet, appliances, etc.; as apartments may be remarketed annually. TYPES O F M ULTIFAMILY P ROPERTIES Garden Apartments

Multiple buildings; usually no more than 2-3 stories. High Rise Apartments

Over three stories; usually located in downtown areas.

garden apartments

14

high rise apartments

CONVENTIONAL I NVESTOR W ISDOM

• Potential overbuilding in high growth markets of the Southeast and Southwest — Houston, Atlanta, Las Vegas. • Multifamily was the first sector to recover from the real estate recession. • Birth dearth has resulted in fewer households in prime renting years of 25-34. This fact somewhat mitigated by the “lifestyle” renter. • A lifestyle renter is someone who can afford a home but chooses to rent for convenience: divorcees, empty-nesters. • Healthy apartment properties have occupancies of 93% and higher. Occupancies below 88% for existing properties are worrisome. • Government sponsored entities desire transactions with high concentrations in multifamily properties. RATING A GENCY V IEW

• A “must have” property type for diversity. • Lower default tendency due to constant mark-to-market of rental rates (1-year-lease ). • Basic need housing. • Tolerated lower DSCR and higher LTV than other commercial property types. • Basket of individual home owner credits; not specific business risks. • Concerns of military or single employer concentrations.

Please see additional important disclosures at the end of this report.

15

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Office Loans Typical Loan

• 1.4x DSCR • 70% LTV (9–10% cap rate) • $50–100 Loan Per Square Foot for Suburban Properties • $70–150 Loan Per Square Foot for Downtown Central Business District (CBD)

RELATED T ERMS Tenant Improvements

Costs to build walls, ceilings, carpet for a new tenant, typically $5-40 per square foot. The landlord usually incurs this expense. In strong demand markets the landlord can this expense through to the tenant in of a higher rental rate. In weak markets, landlords must take tenant improvements out of net income, which reduces cash flow. Leasing Commissions

Fees paid to brokers to bring tenants, typically $2–4 per square foot at lease gs. Landlords bear this expense. Rollover

Term used to describe expiration of a tenant lease. Lease are generally for 5-10 years; credit tenants may be longer. It is preferable to not have rollover concentrations, which would expose an owner to uncertain rental market or potentially reduce NOI below debt service. TYPES O F O FFICE B UILDINGS downtown

16

suburban

CONVENTIONAL I NVESTOR W ISDOM

• Above market rents are a concern if the market rent is insufficient to the debt service. Rating agencies usually underwrite to market rents. • Overbuilding: Investors that lived through the last real estate depression are nervous about current levels of development. • Downtown versus suburban: Suburban office has suffered recently. The recent pop in the “tech bubble” has increased the supply of subleased space. RATING A GENCY V IEW

• Very conservative approach makes underwriting these loans difficult. • Very difficult to allow rollovers without cash reserves. • Slow to accept market improvements in rental rates and values without many other market comparable transactions. • Want higher DSCR because of income volatility during lease rollover. • Only give credit in underwriting for lesser of historical market rents or in place rents. • Tenant improvements/leasing commissions reserved for in escrow or excluded from underwriting income, which reduces the amount of potential loan. • Management fees of 5% used by rating agency underwriters are believed to be above market rate of 1-3%.

Please see additional important disclosures at the end of this report.

17

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Manufactured Housing Community Loans Typical Loan

• 1.4x DSCR • 70% LTV (9–10% cap rate) • $10,000–$20,000 per pad

RELATED T ERMS Manufactured Housing Communities

The land, streets, utilities, landscaping, and concrete pads under the homes comprise a manufactured housing community. The homes are independently financed. Homeowners pay monthly rent for the pad to the manufactured housing community owner. Pad

Concrete slab that s each manufactured home. Double Wide/Single Wide

Describes the size of manufactured home that a given slab will . The double wide segment has experienced the fastest growth due to the growing acceptance of manufactured homes as a single family housing alternative. TYPES O F M ANUFACTURED H OUSING C OMMUNITIES 3-S Star Park

Older park lacking the amenities of a 5-star park; higher proportion of single wide pads. Offer limited amenities and services. 4-S Star Park

Usually double-wide units in good condition. Features may include concrete patios or raised porches. Streets are generally paved. Many 4-star parks were formerly 5-star parks that are now showing their age by their dated look and type of improvements. 5-S Star Park

Curvilinear streets (streets that have curbs); neighborhood feel; well-landscaped; high proportion of double wide pads. Located in desirable neighborhood with convenient access to retail. Trailer Park

This is a lower-end asset class, often confused with manufactured housing communities. Trailer parks are highly transient, dense communities of homes on wheels. Tenants are provided with no amenities other than simple utility hookups. Not typically seen in conduit pools.

18

CONVENTIONAL I NVESTOR W ISDOM

• Lack of familiarity; confusion with asset-backed manufactured housing loans to individuals. • Insurance companies typically didn’t lend on this product, and it has somewhat of a stigma attached to it. • Some investors have trouble distinguishing the relative investment stability of manufactured housing communities from the credit of individual homeowners. The credit characteristics are very different. • Performance on these loans has been very strong. The security for loans requires virtually no maintenance and very few manufactured homes are ever moved from the pad. • Cost to move ($3,000–5,000 a year) exceeds pad rental ($300–500 a month). Upon homeowner foreclosure, bank will usually pay rent instead of moving the foreclosed unit. • Manufactured housing communities are also difficult to build as many communities have restrictive zoning ordinances against them. RATING A GENCY V IEW

• Rating agencies favorable on credit. • Low volatility of cash flows. • Physical turnover rate 3–5% in manufactured housing versus 50–60% in multifamily. • Few capital reserves required. • Often better than multifamily.

5-star park

Please see additional important disclosures at the end of this report.

19

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Industrial Loans Typical Loan

• 1.4x DSCR • 70–75% LTV (8–10% Cap Rate) • $10–25 PSF • $30–50 PSF (if high office component)

RELATED T ERMS Distribution Space

Principal use is distribution or light assembly. Minimal office space as a percentage of total space (typically 0-10%). “Clear heights”, 24´ ceiling heights, are the minimum for modern distribution buildings. Higher clear heights are more economical for tenants as they stack goods vertically and rent fewer square feet. Flex Space/Office Warehouse

Higher amount of office space as a percentage of total space. This results in higher tenant improvement costs necessary upon lease renewals. These tenants are less sensitive to clear heights and more sensitive to accessibility of qualified labor pools. Tilt-U Up Construction

This is the preferred construction type for industrial buildings. It includes precast concrete s that are “tilted-up” on a steel frame. Tilt-up is preferable to corrugated metal exteriors for maintenance reasons.

distribution space

20

CONVENTIONAL I NVESTOR W ISDOM

• Frequently exposed to single tenant credit; but generally lower tenant improvement costs makes rollovers more palatable than office. • Older buildings with less competitive clear heights becoming functionally obsolete. Constant roof repair is major expense item. • Concerns over environmental contamination if property has had heavy industrial use. • Construction cycle for industrial properties generally shorter than other property types (six to nine months versus two years for office); resulted in less overbuilding in last downturn. RATING A GENCY V IEW

• Like industrial for diversity. • Review environmental issues from manufacturing uses. • Very low cost to re-lease if tenant leaves, but short lease create rollover risk.

flex space/office warehouse

Please see additional important disclosures at the end of this report.

21

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Self Storage Loans Typical Loan

• 1.3x, 1.4x DSCR • 65–70% LTV • $20–50 PSF

RELATED T ERMS Self-S Storage Facility

Commercial property that leases storage space to individuals or businesses on a month-to-month basis. The average self-storage facility has between 40,000 and 10,000 square feet of rentable space divided among 400 to 1,000 individual units. Management

Half of all self-storage facilities have a manager living in an on-site apartment. The management team is important to solicit new business and to monitor any delinquencies. Tenant Profile/Turnover

Residential s make up more than two-thirds of self-storage facility renters. The average length of self-storage rental is 12 months. Occupancy rates tend to be in low-to mid-90s after long initial lease up period.

self-storage facility

22

CONVENTIONAL I NVESTOR W ISDOM

• Overbuilding and new competition are concerns, which are mitigated by limited sites zoned for storage in populated areas. • Properties are management-intensive. • Population shifts affect demand; population inflow creates new demand but population outflow may result in an increase in demand in the short term as homeowners store excess items during relocation. RATING A GENCY V IEW

• Approve of it as an additional diversifier. • Higher volatility because of monthly rental contracts but rental base has been relatively inert in moving stored items. • Preference for infill or dense urban locations.

Please see additional important disclosures at the end of this report.

23

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Health Care/Senior Housing Loans Typical Loan

• 1.2–2.0X DSCR (depending on complexity of service) • 50-80% Loan to Value • $30,000-80,000 Loan Per Unit

TYPES O F H EALTH C ARE/SENIOR H OUSING P ROPERTIES Independent Living Facilities

Multifamily apartment complexes catering to senior citizens. They supply few services beyond building and ground maintenance. These facilities are unregulated. Congregate Senior Housing

These are independent living facilities that provide a common dining facility and other services. Congregate senior housing has no medical component, but may provide access to emergency medical care through call buttons. Not licensed as a nursing home. Assisted Living Facilities

A product type targeted to elderly needing assistance, but not full time medical care. These facilities typically consist of apartment style units with a kitchenette. The operator of the facility provides three meals a day and assists residents with daily activities such as feeding, bathing, dressing, and medication reminders. Skilled Nursing Facilities

Independent nursing homes or a designated wing in a hospital. The facilities provide full-time licensed skilled nursing, medical and rehabilitative services. Average length of stay can range from two months to two years, or more. Twenty-four hour care is provided, with doctors and ed nurses on call. Facilities are licensed by the state; operator must obtain a certificate of need from state before beginning operation. Continuing Care Retirement Communities

These facilities offer the entire continuum of seniors housing from independent living to skilled nursing facilities. Residents move within the facility depending on the level of care required. Licensed operator.

24

CONVENTIONAL I NVESTOR V IEW

• Recent changes in Medicaid reimbursements have negatively affected skilled nursing and continuing care facilities. • Excess supply of assisted living facilities has resulted in higher vacancy rates and poor performance. • Few recent additions to supply due to changes in Medicaid reimbursements. • Concerns exist over the quality of management of the licensed facilities. The license is owned by the manager; not the property owner. Many life companies avoided the senior sector because of confusion with licensed “nursing homes.” Foreclosure could result in license forfeiture and force economic vacancy of asset until replacement manager is found and new license granted. • Not easily converted to multifamily if license lost. RATING A GENCY V IEW

• Positive demographics; aging of the baby boomers. • They prefer densely populated areas with heavier capital reserves. • Skilled nursing requires higher DSCR because high ratio of income derived from medical services that are not derived from location of property but rather need of specific patient.

Please see additional important disclosures at the end of this report.

25

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Hotel Loans Typical Loan

• 1.5x DSCR • 65% LTV (10–12% cap rate) • $20,000–60,000 Loan Per Room • $80,000 + Full Service or Luxury

RELATED T ERMS Average Daily Rate

Total guest room revenue divided by total number of occupied rooms. Occupancy Rate

Number of occupied rooms divided by total number of rooms. Revenue Per Available Room (RevPAR)

The revenue per available room is the total rooms revenue divided by the available rooms for a given period. FF&E

Furniture, fixtures, & equipment; standard hotel underwriting includes a deduction as an operating expense for the ongoing replacement of FF&E, typically 4% to 5% of gross revenue. This differentiates hotel underwriting from apartment underwriting where some of those same expenses are considered capital expenditures and are not an operating expense deducted from NOI. Typical refurbishment of common areas of the hotel should occur every seven years. Franchise Fee

Fee paid to hotel company that allows hotel owner to “fly the flag” of a hotel company (i.e., Marriott, Sheraton, etc.) and benefit from advertising and reservation network. Ranges from 4–7% of gross revenue. TYPES O F H OTEL P ROPERTIES Full Service

A hotel that offers banquet and convention services; one or more full service restaurants. Limited Service

No food service other than continental breakfast; minimal public space and small staff.

26

CONVENTIONAL I NVESTOR W ISDOM

• Hotels have the highest cash flow volatility of the four major property types as they reprice rooms on a daily basis. • The full service downtown hotels are more protected from new supply because of high building costs and limited site availability. • Limited service construction is up in many areas of the country. • Full service hotels have higher fixed costs and lower operating margins than limited service hotels. RATING A GENCY V IEW

• Very conservative because use of average historical income often reduces income to levels significantly below current highs. • Maximum occupancy they will underwrite is 65–75%, despite actual figures higher than that level. • Use very conservative FF&E, franchise fee, management fees, instead of market fees. • Bias toward national brands even if hotel is a niche segment that has strong history. full service

limited service

Please see additional important disclosures at the end of this report.

27

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Hotel Collateral in CMBS Rating agencies and many investors have often viewed hotels as one of the riskiest property types. Unlike most other commercial real estate, which have longterm leases, hotels have daily changes in occupancy and rental rates. In this chapter, we examine the credit exposure of CMBS to hotels and discuss the history of the hotel industry, hotel branding, and recent performance of hotels. • Hotel demand is highly correlated with GDP and can be volatile. • Supply of new hotel rooms built during the 1990s is 25% lower than the supply built in the 1980s. • The hotel industry is typically profitable although it is a cyclical business. • After September 11, 2001, the rating agencies commented on the corporate ratings of most public hotel companies and revised some underwriting standards for new transactions. • The aggregate CMBS market has about 8% exposure to hotels. Leverage levels are more conservative on hotel loans than other property types, providing cushion. exhibit 1

PROPERTY TYPE MATRIX

Source: Morgan Stanley

GREATER S HOCK T O D EMAND T HAN E ARLY 1 990 S

Hotel demand is highly correlated to changes in GDP. When GDP is growing, consumer confidence rises resulting in greater discretionary income. The events surrounding September 11, 2001, resulted in a significant decline in hotel demand. Industry data shows hotel demand declined significantly in late 2001, 2002 and 2003. In 2004, hotel demand recovered significantly with nationwide revenue per available room (RevPAR) growth over 7% through August.

28

HOTEL S UPPLY S IGNIFICANTLY L OWER I N 1 990 S THAN 1 980 S

Over the past 15 years hotel ownership has experienced a significant shift from private to public markets. The public capital markets have more efficiently matched supply with demand than was the case in previous cycles. The typical time frame for hotel development ranges between 12 and 36 months. Between 1990 and 1991, demand declined significantly as the United States entered a recession, at the same time supply continued to enter the market through projects that were started in the late 1980s. Currently, the industry is better suited from a new supply standpoint than it was in the early 1990s.

exhibit 2

TOTAL CONSTRUCTION: 1980s VS. 1990s

Source: Morgan Stanley, PricewaterhouseCoopers

exhibit 3

HOTEL ADDITIONS TO SUPPLY 1972–2004P

Source: PriceWaterhouseCoopers

Please see additional important disclosures at the end of this report.

29

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS Rev PAR P ERFORMANCE A ND P ROJECTIONS

The hotel industry measures performance in revenue per available room (RevPAR). RevPAR is defined as average daily rate (ADR) at a hotel multiplied by its occupancy rate. For example, if a hotel has an ADR of $100 and an occupancy of 80% its RevPAR is $80.00 ($100 x .80 =$80). RevPAR has declined 3 out of the last 14 years, 1991 (Gulf War), 2001 (U.S. terrorist attack), and 2002 (U.S. economic decline). HOTEL F UNDAMENTALS C ONTINUE T O I MPROVE

Healthy economic growth and limited new hotel development fueled solid nationwide RevPAR growth in 2004. The I numbers released for March and April noted the “lodging away from home” component posted year-over-year growth of 7.0% and 8.7%, respectively. Year to date through August, nationwide RevPAR was up 7.1% over the same period last year. Hotel demand is highly correlated to GDP. Since our economists anticipate strong GDP growth in the second half of 2004, we anticipate hotel fundamentals to remain strong. Annual 2005 RevPAR growth is projected to be 4.5% by PriceWaterhouseCoopers. PricewaterhouseCoopers uses its own GDP estimates and other macroeconomic inputs to forecast RevPAR. exhibit 4

NATIONWIDE REVENUE PER AVAILABLE ROOM (RevPAR) GROWTH

Source: PricewaterhouseCoopers, Smith Travel Research

RATING A GENCY V IEWS O N H OTELS

After the events of September 11, 2001, rating agencies commented on the corporate ratings of several hotel companies, and revised rating guidelines for future CMBS transactions. All three rating agencies commented on the ratings of several hotel operating companies and hotel REITs.

30

In addition to the comments on corporate ratings, Moody’s and Fitch outlined changes to underwriting guidelines for hotel loans within CMBS transactions. Standard & Poor’s has not made any changes to its underwriting guidelines for hotels in new issue transactions. Fitch noted it was concerned about the performance of hotels. Therefore, for transactions it will use the trailing 12-month RevPAR after August 31, 2001, reduced by 20%, or the 1999 RevPAR number, whichever is lower. In place of the significant RevPAR reduction, Fitch stated that the issuer could set up a 12month debt service reserve for hotel loans. Fitch also intends to make adjustments to expenses based on any recently increased costs. Fitch stated that for surveillance of CMBS transactions already issued, it will not apply the severe haircut to hotels but will compare the current operating numbers to the performance when the transaction was issued. Moody’s released a matrix outlining reserve requirements on hotel loans within new issue CMBS transactions. Reserve requirements range from 1-13 months depending on the type of hotel property, the anticipated stress environment, and the underwritten debt service coverage ratio (DSCR). For example, a full-service hotel with a 1.35 DSCR at issuance would require 13 months P&I reserves in a high stress environment while a limited service hotel with a 1.35x DSCR would require 5 months P&I reserves in a high stress environment. The rating agency also noted that it would consider 1998-99 as stabilized operating performance for hotels going forward.

Please see additional important disclosures at the end of this report.

31

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS exhibit 5

RATING AGENCY VIEWS ON HOTEL COMPANIES, AS OF NOVEMBER 2004

Source: Morgan Stanley, Moody’s, Fitch, S&P

32

HOTEL D ELINQUENCIES I N C MBS T RANSACTIONS

Based on our October 2004 remittance reports data, delinquencies on hotel loans ed for 2.44% of current balances. During 2004, hotel collateral within CMBS experienced significant improvement with delinquencies declining 3.08%. We anticipate hotel delinquencies will continue to improve based on our current GDP forecast for 2005.

exhibit 6

CMBS DELINQUENCIES BY HOTEL AND TOTAL

1

30/60/90/Foreclosure/REO.

Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

33

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS BRANDING A ND S EGMENTATION

As the hotel industry has evolved, companies such as Marriott have segmented the market in order to meet customers’ needs, increase profitability, and diversify its customer base. Branding has been used to attract guests to various price points and increase the parent company’s exposure. Positive brand identity is beneficial for the hotel manager and the hotel owner. If the brand is effective in producing demand, management will spend less time marketing nationally and focus on local operations and solicitations. The owner also benefits financially from increased business through the reservation system and through potentially lower marketing costs. As the economy weakens consumers move down one or two positions in service and price from say a Hilton to an Embassy Suites. As the economy strengthens guests typically move up in service level and price.

exhibit 7

HOTEL BRANDING AND SEGMENTATION

Source: Morgan Stanley

UPPER U PSCALE

These hotels are exclusive properties within major metropolitan markets that provide extensive amenities and high levels of service. A hotel within this segment will have the highest ADR in its market and often the highest RevPAR in the market. Amenities at these properties usually include health club/spa facilities, three- or four-star food and beverage outlets, concierge service, 24-hour room service, valet, and retail spaces. These properties are located in prime downtown and resort real estate locations. In general because of higher construction costs and high barriers to entry, luxury hotel construction tends to lag behind the general trend of the market cycle. While this segment is the last to see new supply, during a recession it is the first to experience a decline in occupancy as travelers become more budget-conscious and move down the service-level curve to less expensive accommodations.

34

UPSCALE

Upscale hotels are full-service hotels that cater to individual business travelers, groups, and conventions. Typically the price points at these hotels are between upper upscale hotels and mid-scale hotels. Amenities at upscale hotels include several food and beverage outlets, extensive meeting space, laundry service, concierge, and exercise facilities. Most of these properties enjoy downtown locations near convention centers or strong suburban locations. These locations may have some barriers to entry because of limited availability of land. The guest profile for these hotels is transient business travelers, extensive corporate business, and meeting/convention business. MID-S S CALE W ITH F OOD A ND B EVERAGE

These hotels encom a broad range of brands and product types depending on the market and the age of the property. Amenities at these hotels usually include one restaurant, limited meeting space, and an exercise facility. Typically, these hotels cater to the more budget-conscious business travelers or “road warriors.” Older properties in this segment can be somewhat outdated and suffer from inefficiencies. Concerns about this segment center around older properties in need of renovation and a tendency for travelers to prefer newer properties that are entering many markets at similar price points. MID-S S CALE W ITHOUT F &B C HAINS

Properties in this segment typically have secondary locations such as major highway intersections, airports, or suburban locations. The properties do not offer food and beverage amenities except for continental breakfast in some locations, and may or may not have small exercise facilities. Food and beverage is provided by nearby fast food or chain restaurants. The typical guest is a budget-conscious business traveler, or transient guest from the highway. Extensive development has occurred in this segment. Construction costs for these properties are significantly lower than full-service hotels because of their secondary locations and low rise nature. ECONOMY/BUDGET S EGMENT

Smaller roadside motels with very limited amenities characterize these properties. Construction is inexpensive, low barriers to entry exist, and the development timetable is short. The typical guest at these properties is a transient budget-conscious business or pleasure traveler. Most bookings are made through the reservation system and repeat business is not significant. HISTORY O F T HE H OTEL I NDUSTRY

Lodging’s long history shows that demand for hotels and development of hotels move in cycles. The time period of these cycles depends on the demand (affected by the overall state of the economy) and new supply.

Please see additional important disclosures at the end of this report.

35

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS 1790–1920 G ROWTH O F A MERICA A ND F INANCIAL B OOM

The first American hotel, the 73-room City Hotel, was built in New York City in 1794, and was much larger than its predecessors, the colonial inns. Between the late 1700s and the mid-1800s, similar hotels were built in major metropolitan cities such as Philadelphia, Boston, and Baltimore. During the 1800s, hotels followed the railroads westward and hotels were built in Chicago, Denver, and San Francisco. Now famous historic spas such as The Greenbrier in West Virginia were built during this period. The Buffalo Statler hotel, built in 1908, was revolutionary because it offered baths in each guest room. During the Roaring ’20s, as with many other areas of commerce, hotel demand and hotel development boomed. The 3,000-room Hotel Stevens (currently the Chicago Hilton) was built in 1927. Hotels were built through community funded bonds and often without regard for feasibility. Conrad Hilton began his chain of Hilton hotels during the 1920s by purchasing and developing eight hotels. So many hotels entered the market nationwide in the 1920s that occupancy went from 86% in 1920 to 68% in 1928, while room rates remained fairly constant. 1930–1940 s THE G REAT D EPRESSION

The boom of the ’20s was followed by the bust or Great Depression of the 1930s. Due to overbuilding and a significant decline in demand, more than 80% of American hotels went into foreclosure. Nationwide occupancy declined to 50%. Enterprises that had cash during this period were able to obtain hotels at deep discounts to the original costs. Ernest Hendersen founded the Sheraton hotel chain in 1937 and was able to build the company’s portfolio during the 1930s and 1940s. The hotel industry did not recover from the overbuilding of the 1920s until the early 1940s. World War II stimulated the economy and generated demand for hotel rooms as troops and military personnel were moved domestically and globally, an effect that continued after the war. By the mid-1940s nationwide occupancy had increased to 90%. 1950–1960 s BOOM A FTER W WII

With World War II over, the United States entered a stage of economic growth. During the 1950s, transportation via the automobile became very popular. Most Americans that had previously traveled by train were now using cars. This increased prosperity and altered mode of transportation resulted in the development of the first roadside “motor hotels” or motels. Kemmons Wilson developed the first Holiday Inn in 1952, and by 1960 there were more than 100 Holiday Inns. Marriott and Hyatt were founded in 1957, Howard Johnson’s in 1959, and Ramada and Radisson in 1962. Tax laws, highway development, and an increase in franchising fueled the development of both hotels and motels during this period. 1970 s QUICK C YCLE

With several chains established and the motel concept fully developed, the early 1970s became another boom for the hotel industry. At the same time, companies such as Holiday Inn and Marriott were looking to expand their presence through

36

franchising. The extensive capital available in the markets and ability to franchise brands encouraged development of hotels, which resulted in overbuilding. In 1974, inflation caused construction costs and interest rates to rise. The energy crisis reduced road travel and the recession prohibited business travel. This was another period of foreclosures and excess hotel supply. Although the overbuilding of the 1970s was similar to the 1920s the recovery to reasonable real estate prices came about more quickly than in the 1920s. By the end of the late-1970s hotel prices had recovered due to the lack of building during the end of the decade. By the end of the 1970s supply and demand were more in balance than five years prior and occupancy levels were rising. 1980 s

By 1983, inflation had slowed and new tax policies created a favorable environment for hotel development. The Savings and Loans (S&Ls) were in the process of deregulation and were an eager source of financing for real estate development. Lack of commercial real estate experience by the S&Ls resulted in loose underwriting standards and liberal lending policies. During this time, syndicated Limited Partnerships were the source of equity capital. Hotel partnerships were sold to wealthy individuals attracted by the aggressive growth projections and the trophy real estate. In order to capture the greatest tax benefits for these investments, this real estate was highly leveraged at 90%–100%. Investors benefited from ive tax losses in the short term and real estate appreciation in the longer term. A change in the tax laws in the late 1980s resulted in the revision of the allowance for these losses, slowing new development, but overbuilding was already under way, particularly in full-service hotels.

exhibit 8

CAPITALIZATION RATES AND 10YEAR TREASURY RATES

Source: PricewaterhouseCoopers, Smith Travel Research

Please see additional important disclosures at the end of this report.

37

Transforming Real Estate Finance chapter 2

Major Property Types in CMBS EARLY 1 990 s

The late 1980s to 1991 was a period of overbuilding in virtually all sectors of real estate, and hotels were no exception. Profitability nationwide for the hotel industry was negative between 1986 and 1992. Demand sharply declined from its significant growth in late 1989, to a cycle low in 1991. Unfortunately, development was already in the pipeline and supply growth was high. During this time, many hotels defaulted on loans because they were so highly leveraged and unable to meet debt payments. In the early 1990s, the federal government formed the Resolution Trust Corporation (RTC) to resolve the problems of the S&Ls, who were suffering from defaulted real estate loans. Development during the early 1990s was virtually halted and banks were unwilling to lend on real estate in general, hotels in particular. Capitalization rates during this period climbed, as investors demanded extremely high returns. Because of a lack of debt funding, most hotel purchases were made by companies with cash reserves. The real estate devaluation and lock up in capital reduced the number of individual hotel transactions from 130 in 1990 to 56 in 1991. Individual hotel transactions did not return to 108 again until 1994. Purchase prices per room were wellbelow development costs. Nationwide, the average purchase price per room in 1990 was $136,000. This figure declined significantly in 1991 to $96,000 and reached a bottom of $80,000 in 1995. MID- T O L ATE 1 990 s

By late 1995, the RTC disposed of most real estate previously held by the defunct S&Ls. Demand for hotel rooms outpaced supply between late 1992 and mid-1996. From 1993, increases in the average daily room rate (ADR) outpaced the consumer price index. Beginning in 1994, the increase in profitability began to attract investors. Prices began to increase as demand was outpacing supply and transactions increased. Beginning in 1993, hotel REITs began accessing the public capital markets. In 1993, $34 million was raised in the capital markets for hotel REITs. This increased to $600 million in 1994, doubled to approximately $1.2 billion in 1995 and 1996, and reached a peak of almost $1.6 billion in 1997. This inflow of public capital fueled supply growth and transaction activity for hotels as REITs acquired more properties and consolidation occurred. The consolidation of hotel ownership from private partnerships to public companies increased operating efficiencies and held owners able to shareholders, reducing hotel leverage levels. In general, hotel REITs did not exceed 60% leverage.

38

Capital flow into hotel REITs declined significantly in 1998 to just over $500 million. This was due to Congress enacting legislation limiting acquisitions by paired-share REITs, REIT investors’ fear of overbuilding in all real estate sectors, slower growth in REIT earnings than the broader market, and the capital flight to quality in the fall of 1998. The change in the paired-share legislation significantly affected Starwood and Patriot American (now Wyndham) the two largest paired-share hotel REITs, stopping their growth initiatives. During the mid- to late-1990s, as the CMBS market grew, hotels were included in pools of diversified mortgages. The nightly rents and intense on-site management combined with poor performance prior to the RTC clean-up resulted in lower LTV ratios for hotel loans than other types of real estate within CMBS transactions.

Please see additional important disclosures at the end of this report.

39

This Page Intentionally Left Blank

40

Chapter 3

Transforming Real Estate Finance

Call Protection

Please see additional important disclosures at the end of this report.

41

Transforming Real Estate Finance chapter 3

Call Protection EXECUTIVE S UMMARY

• Commercial mortgage loans differ from residential mortgage loans in that they are call protected. Commercial mortgage loans typically contain provisions that either prohibit or economically penalize the borrower for prepaying the loan before maturity. • There are four main categories of call features in commercial mortgage loans: hard or legal lockout, yield maintenance, fixed percentage penalty points and defeasance. Most commercial mortgage loans contain at least one of these forms of call protection, and many contain some combination of these penalties. • Allocation of the loan level prepayment penalties to bond classes in a CMBS differs across deals and is an important characteristic in determining relative value. • Lockout and defeasance provide investors with the greatest average life stability. Credit events are the only source of cash flow variability in CMBS deals where the underlying loans are locked out or defeased. With yield maintenance, some investors benefit from faster prepayments under some interest environments. • A number of major loan originators have gravitated to defeasance because of favorable pricing in the CMBS market. INTRODUCTION

Commercial mortgage loans differ from single-family residential loans in a very important aspect: call protection. Unlike single-family loans, commercial loans typically are not fully prepayable at the borrower’s option. Call protection in commercial loans stems from the life insurance companies’ historical position as the primary provider of capital to the real estate industry. As the dominant long-term lender in the market, the life insurance companies required call protection as a standard feature of the commercial loan market. Call protected loans were an attractive asset for an insurer to match against long duration liabilities. The growth of the CMBS market has altered the form of call protection found in newly originated commercial loans. As an increasing percentage of commercial loans are being securitized, the preferences of bond investors are playing a large role in defining the of the commercial mortgage markets. This chapter will:

42

1)

Define the various types of call protection found in commercial mortgage loans;

2)

Discuss different prepayment penalty allocation structures found in CMBS transactions;

3)

Explore relative value of various forms of call protection; and

4)

Discuss our expectation for future trends in CMBS call protection.

DEFINITION O F T YPES O F L OAN L EVEL C ALL P ROTECTION

Several different types of call protection are found in commercial mortgage loans. HARD L OCKOUT

The most straightforward form of call protection is a “lockout” provision. The lockout provision legally prohibits a borrower from prepaying a loan prior to scheduled maturity. The majority of commercial mortgage loans are not “locked out” for the entire term of the loan. Lockout periods are usually three to five years and are followed by penalty periods. During the penalty period, the borrower is allowed to prepay the loan, but the borrower must compensate the lender for the early termination right. The two forms of penalty are yield maintenance and fixed percentage penalty points. Further, most lockout provisions allow a borrower to assign the loan if the property is sold during the lockout period. YIELD M AINTENANCE

A yield maintenance penalty is designed to compensate the lender for the interest lost as a result of prepayments. The yield maintenance penalty in commercial mortgage loans is analogous to the “make whole” provisions found in corporate bond markets. Formulas used to calculate yield maintenance vary. Generally, the formulas provide a present value calculation of the positive interest differential between the remaining mortgage payments due on the original loan and the payments that would be due on a reinvestment of that repaid loan. The yield maintenance penalty is designed to make the lender indifferent to a prepayment. If interest rates are significantly higher at the time of prepayment than at the time of loan origination, the borrower would not be required to make a penalty payment. The lender does not suffer an opportunity cost in this situation as the lender can reinvest the prepaid proceeds at higher market interest rates. The key variable in calculating yield maintenance penalties is the discount rate or reference rate. The reference rate is compared to the existing mortgage loan rate to calculate the prepayment penalty. Reference rates are usually a comparable maturity Treasury rate, (referred to as “Treasuries flat”), or a comparable maturity Treasury rate plus a spread. Clearly, the lender/investor prefers Treasuries flat as the reference rate. Treasuries flat results in a higher present value for the yield maintenance calculation. The following is an example of a yield maintenance calculation at Treasuries flat. Assumptions

• Loan Origination Date: 1/01/98 • Original Loan Balance: $10,000,000 • Loan Maturity Date: 12/31/07 • Mortgage Rate: 7.25% • Start of Yield Maintenance Period: 1/01/98 • Reference Treasury Rate: 5.75%

Please see additional important disclosures at the end of this report.

43

Transforming Real Estate Finance chapter 3

Call Protection To simplify the math, we will assume the loan prepays on the origination date. Yield Maintenance Formula (PRESENT VALUE FACTOR) x (MORTGAGE

RATE – TREASURY RATE) x (REMAINING LOAN BALANCE)

-Penalty Period Present Value Factor =1-(1+Reference Treasury Rate) Reference Treasury Rate = 1-(1+5.75%)-10 5.75% = 7.45%

Yield Maintenance = (7.45)x(7.25%-5.75%)x($10,000,000)=$1,117,500 = 11.18% of the remaining loan balance

As the term to maturity for the loan shortens, the yield maintenance penalty as a percentage of the remaining balance decreases. Yield maintenance penalties provide less of a disincentive to the borrower to prepay as the remaining penalty period declines. The importance of the yield maintenance penalty to the lender also decreases as the remaining penalty period declines. As the term to maturity decreases, the remaining loan payments represent a lower percentage of the lender’s total return. The following table illustrates the difference between reference rates of Treasuries flat, Treasuries +25 bp, and Treasuries +50 bp as the penalty period shortens from 10 years to three using the same assumptions as the prior example. exhibit 1

PREPAYMENT PENALTY AS A PERCENTAGE OF OUTSTANDING LOAN BALANCE Penalty Period

Reference Rate

10-YR

5-YR

3-YR

Treasuries Flat Treasuries + 25 bp Treasuries + 50 bp

11.18 9.20 7.27

6.36 5.27 4.18

4.03 3.34 2.66

Source: Morgan Stanley

FIXED P ERCENTAGE P ENALTY P OINTS

Fixed percentage penalty points are potentially the weakest form of call protection. The prepayment penalty is a fixed percentage of the remaining loan balance. Fixed percentage penalty points typically decline over the life of the loan. A representative example of the of a loan with fixed percentage penalty points would include a lockout period of five years followed by declining penalty points of 5% in year 6, 4% in year 7, 3% in year 8, 2% in year 9, and 1% in year 10. Large interest rate moves and large increases in property values may overwhelm these fixed economic disincentives to prepay a fixed percentage penalty loan as the penalties do not change with interest rates.

44

The following exhibit indicates the required drop in the mortgage rate that compensates a borrower for fixed percentage penalty points. For example, if a borrower wanted to prepay a loan with a term to maturity of six years and a 5% fixed penalty, the borrower’s new mortgage rate would have to be 110 bp lower than the existing mortgage rate to justify paying the penalty.

exhibit 2

ANNUAL BASIS POINTS OF SAVINGS REQUIRED TO PAY FOR PREPAYMENT PENALTY POINTS (bp)

Penalty Points

10-YR

8-YR

6-YR

10 9 8 7 6 5 4 3 2 1

150 140 120 110 90 80 60 50 30 20

180 160 140 130 110 90 70 60 40 20

220 200 170 150 130 110 90 70 50 20

Loan Term to Maturity 4-YR 2-YR 300 270 240 210 180 150 120 90 60 30

540 490 430 380 330 280 220 170 110 60

Note: Calculation assumes a 10% initial mortgage rate, annual coupon payment, and cash flows discounted at new mortgage rate. Table entries are rounded to the nearest 10 bp. Source: Morgan Stanley

DEFEASANCE

Defeasance, a mainstay of the municipal market, has found its way into the commercial mortgage market. From the investor’s perspective, a loan with a defeasance appears “locked-out” from prepayment. The borrower may prepay the loan, but the cash flows to the investor will not change as a result of the prepayment. In exercising the defeasance option, the borrower replaces a mortgage loan with a series of U.S. Treasury strips which match the payment stream of the mortgage loan as collateral for the loan. Not only is an investor indifferent to a prepayment in a defeased loan, the investor actually prefers it. If the borrower exercises a defeasance option, the investor receives the benefit of improved credit quality on the collateral without a corresponding decline in return. Before the prepayment, the investor was exposed to commercial real estate credit risk. Following the prepayment, the investor is exposed to U.S. Treasury securities credit risk. As an example, consider a $10 million non-amortizing commercial mortgage loan with a 7% coupon, three-year term to maturity, and annual payments. If the borrower wanted to defease the loan, the borrower would purchase three U.S. Treasury strips that would replicate the payments due on the mortgage loan. For the U.S. Treasury strips, we assumed the following rates and prices:

Please see additional important disclosures at the end of this report.

45

Transforming Real Estate Finance chapter 3

Call Protection Rates(%) 5.65 5.68 5.73

• 1-Year • 2-Year • 3-Year

Dollar Price 94.36 89.20 84.21

The chart shows the payments due from the borrower on the commercial mortgage loan, the cost to purchase the U.S. Treasury strips and the payments to the CMBS trust from the U.S. Treasury strips. exhibit 3

CASH FLOWS TO CMBS TRUST UNDER DEFEASANCE

Time

Payment Due From Borrower on Loan

Cost to Borrower to Purchase U.S. Treasury1

Payments From U.S. Treasury Strips to CMBS Trust

Year 1 Year 2 Year 3

$700,000 $700,000 $10,700,000

$660,520 $624,400 $9,010,470

$700,000 $700,000 $10,700,000

1

At time of prepayment.

Source: Morgan Stanley

The following steps describe the impact of a defeased prepayment on a CMBS transaction. When a borrower wants to prepay a loan: 1) The borrower buys multiple U.S. Treasury strips in amounts that replicate the remaining principal and interest payments due on the mortgage loan. 2) The borrower delivers a legal agreement that designates the CMBS trust as having the first priority on the U.S. Treasury securities. 3) The servicer is responsible for purchasing the U.S. Treasury securities on behalf of the borrower. The borrower pays the execution costs. In summary, the original loan remains an asset of the trust, but the mortgage, which is the lien on the property to secure the loan, is removed, and the collateral securing the loan is now U.S. Treasury securities. A prepayment through the exercise of a defeasance option provides no disruption in cash flow to the bond investor whether interest rates are higher or lower. The amount of defeased collateral in CMBS deals has increased dramatically. Today CMBS transactions generally have > 90% defeasance. Investors are valuing interest only bonds (“IOs”) from CMBS deals backed by defeased collateral at tighter spreads than those backed by collateral with yield maintenance penalties. The growth in CMBS issuance has resulted in an increase in both the number and type of CMBS investors. CMBS backed by defeased mortgage loans offer investors the opportunity to avoid the complexities of commercial mortgage prepayment analysis. The CMBS bonds that are created from defeased loans resemble corporate bullet securities and have attracted corporate crossover buyers.

46

ALLOCATION O F P REPAYMENT P ENALTIES

For loans with yield maintenance penalties or fixed percentage penalty points, the allocation of these penalties to the various securities in the CMBS structure differs on a deal-by-deal basis. In older CMBS transactions (primarily 1996 and earlier), the prepayment penalties generally were allocated 75-100% to the IO bonds while the amount of penalty paid to the coupon bondholders was capped at some percentage, typically 0–25%. More recent deals generally allocate the prepayment penalties in a way that makes the currently paying bond class “whole” and distributes the remaining penalty to the IO. This more recent allocation method is analogous to the calculation of the yield maintenance penalty on the underlying loan. The currently paying bond investor receives compensation for the early return of principal in a lower interest rate environment. The IO holder generally receives 65–75% of the penalty while the current principal paying bond receives the remainder making it whole to the bond’s coupon, not “Treasuries flat.” The following is a representative example of a yield maintenance, penalty sharing formula found on post-1996 deals. The class that is currently receiving principal would receive an amount equal to the ratio of the difference between the bond rate and the reference Treasury rate and the difference between the mortgage loan coupon rate and the reference Treasury rate. As discussed earlier, the yield maintenance penalty is calculated based on the difference between the commercial mortgage loan rate and the reference Treasury rate. The CMBS bond coupon is generally lower than the mortgage loan rate, so the investor needs a fraction of the entire yield maintenance penalty to be “made whole.” The IO class would receive the remaining 69.3% of the penalty. Both fixed rate and IO investors need to be aware of the type of prepayment penalty sharing agreement found on a particular CMBS transaction as it influences an investor’s return.

Prepayment Distribution Formula =

Bond Through Rate - Reference Treasury Rate Mortgage Loan Rate - Reference Treasury Rate

Bond Through Rates = 6.45% Mortgage Coupon Rate = 8.25% Reference Treasury Rate = 5.65% Currently Paying = 6.45%-5.65% Bond Receives 8.25%-5.65% = .80 2.60 = 30.7% of penalty

Please see additional important disclosures at the end of this report.

47

Transforming Real Estate Finance chapter 3

Call Protection RELATIVE V ALUE A NALYSIS O F D IFFERENT T YPES O F C ALL P ROTECTION

The relative value analysis of yield maintenance versus lockout/defeasance is an analysis of the benefits of cash flow certainty versus the potential opportunities of cash flow uncertainty. With respect to locked-out or defeased deals, the only cash flow variability in the bond will be related to credit events. Movements in interest rates, spread levels, and the credit performance of the underlying loans will determine returns. Locked out and defeased CMBS deals are not exposed to commercial mortgage prepayment risk. The relative value analysis of yield maintenance securities, on the other hand, is not as straightforward. Earlier, we compared yield maintenance to “make whole” provisions found in the corporate market. Even if yield maintenance leaves the commercial mortgage lender indifferent to prepayments, various securities within a given CMBS transaction and across different CMBS deals will perform differently. In addition to the credit performance of the loans, the interest rate environment, prepayment speeds in various interest rate environments, and the allocation of prepayment penalties within a particular deal structure will determine the yield in a bond with yield maintenance loans. An investor in yield maintenance backed CMBS must analyze all of these variables in assessing relative value. Market convention in CMBS scenario analysis is to assume that loans do not prepay during their yield maintenance periods. If 100% of the loans in the pool are yield maintenance protected for their entire term, the assumption of “yield maintenance equals lockout” provides the same result as defeasance or lockout. For the sake of this analysis, we assumed that loans in yield maintenance do prepay in order to ascertain under which scenarios an investor would prefer to own bonds with yield maintenance over bonds with defeased collateral. We analyzed several classes of bonds from different generations of CMBS transactions with yield maintenance as the predominant form of call protection on the underlying loans1. The purpose of this analysis was to determine which bonds benefit in various interest rate and prepayment scenarios. We performed prepayment and interest rate scenario analysis on CMBS deals that allocate all of the prepayment penalties to the IO and on CMBS deals that allocate prepayment penalties that make the currently paying bonds whole. We analyzed each bond at prepayment speeds from 0–100% R in the following interest rate scenarios: 1) Interest rates 300 bp lower 2) Interest rates 100 bp lower 3) Interest rates unchanged 4) Interest rates 100 bp higher 5) Interest rates 300 bp higher The results were based on interest rate levels and prices as of April 1998, and may vary in different interest rate environments.

1

48

Not all of these deals were 100% protected by yield maintenance for their entire term. We chose actual securities to analyze rather than a hypothetical 100% yield maintenance for life bond, as we are not aware of such a CMBS transaction in the marketplace. We also made the simplifying assumption that pricing spread levels remain unchanged at different bond dollar prices.

SHORT A AA ( 1ST P AYMENT P RIORITY) B ONDS W ITH S HARED Y IELD MAINTENANCE P ENALTIES

Because the short AAA bonds are at a discount dollar price in the interest rates higher scenario, faster prepayments during penalty periods result in a higher yield to maturity for the investor. The borrower is not required to pay a prepayment penalty in an interest rates higher scenario, but the borrower is required to pay the par amount of the loan. The investor has effectively bought the loan at a discount and receives principal at a par dollar amount. Why would a borrower prepay in a higher interest rate environment? There are two potential reasons to prepay: 1) Sale of the property 2) Refinancing driven prepayment a) Equity take-out refinancing b) Fixed to floating rate refinancing In the interest rates unchanged and lower scenarios, the yield to maturity also increases as prepayments increase during yield maintenance penalty periods. The amount of the prepayment penalty received by the short AAA holder exceeds the dollar price on the bond. As an example, assume: • Bond Dollar Price: • Mortgage Loan Rate: • Bond Coupon: • Reference Treasury Rate: • Yield Maintenance Penalty Period:

106 8.80% 6.85% 5.65% 9 years

The yield maintenance penalty in this example would equal: Yield Maintenance Formula =Present Value Factor)x(Mortgage Rate-Treasury Rate) Present Value Factor =1-(1+Reference Treasury Rate)-Penalty Period Reference Treasury Rate =1-(1+5.65%)-9 5.65% =6.91% Yield Maintenance (6.91)x(8.80%-5.65%) =21.8% Allocation to Short AAA =Bond Coupon Rate-Reference Treasury Rate Loan Coupon-Reference Treasury Rate =6.85%-5.65% 8.80%-5.65% Allocation =38.1% Penalty Points Allocated to Short AAA =(38.1%)(21.8%) =8.3% vs. Dollar Price of Short AAA =6% Advantage =2.3% (of current balances)

Please see additional important disclosures at the end of this report.

49

Transforming Real Estate Finance chapter 3

Call Protection The fact that the prepayment penalty allocated to the investor is higher than the dollar percentage results in the prepayment benefiting the investor. While the yield to maturity is higher in this scenario, the average life of this bond decreases. For some investors, the shortening of the average life of the bond mitigates the benefits of a higher yield to maturity. SHORT A AA B ONDS W HICH D O N OT S HARE P REPAY P ENALTIES

The yield to maturity on short AAA bonds from earlier CMBS deals generally does not increase as prepayments increase during yield maintenance periods. These bonds do not receive any of the prepayment penalties. When they are trading at a dollar price, they lose yield as they receive prepayments at par. If interest rates were to increase enough so that the bonds were trading at a discount dollar price, the yield to maturity would increase as prepayment speeds increase during yield maintenance periods. exhibit 4

Bond Class

EFFECT ON YIELD TO MATURITY AS PREPAYMENT SPEEDS INCREASE—SHORT AAAs –300 bp

Short AAA Decreases (IO Receives All Penalties) Short AAA Increases (Make Bonds Whole)

–100 bp

Unchanged

Interest Rates +100 bp +300 bp

Decreases

Decreases

Decreases

Increases

Increases

Increases

Increases

Increases

Source: Morgan Stanley

AA B ONDS

The performance of the AA bonds generally follows the same pattern as the short AAA bond. The AA bond, however, has a higher duration and a resulting higher price sensitivity to interest rates. As a result, the AA bond is more likely to be trading at a discount dollar price at smaller increases in interest rates than the short AAA bond. So, in an interest rates up 100 bp scenario on an older deal, the yield to maturity on the short AAA bond decreases as prepayments increase, but the yield to maturity on the AA bond increases. The AA bond is priced at a discount in the rates up 100 bp scenario while the short AAA bond is a . For newer bonds with shared yield maintenance penalties, the actual speed of prepayment will determine whether the investor receives a higher yield to maturity than the 0% R case. Prepayment speeds must be fast enough to retire the AAA bonds during the yield maintenance period in order for the yield to maturity to increase on the AA bond. If the AAA securities are retired and the AA bond becomes the currently paying bond, the AA bond receives a portion of the prepayment penalties. If the AA bond does not become the currently paying bond during the yield maintenance period, it does not receive any of the prepayment penalties. The average life shortens in either case.

50

exhibit 5

Bond Class

EFFECT ON YIELD TO MATURITY AS PREPAYMENT SPEEDS INCREASE (AA) –300 bp

AA Decreases (IO Receives All Penalties) AA Varies (Make Bonds Whole)

–100 bp

Unchanged

Interest Rates +100 bp +300 bp

Decreases

Decreases

Increases

Increases

Varies

Increases

Increases

Increases

Source: Morgan Stanley

INTEREST O NLY B ONDS

IOs from CMBS deals with yield maintenance loans generally benefit as prepayment speeds increase during yield maintenance periods. The present value of the penalty paid to the IO holder is usually greater than the present value of the foregone interest that would have been received from the loan that prepaid. When interest rates rise to the level where no prepayment penalty is due from the borrower, the IO holder does not benefit from faster prepayments. In this case, the IO bond loses the income stream from the prepaid loan and does not receive a compensating prepayment penalty. Unlike a defeased IO, an IO from a CMBS deal backed by yield maintenance loans offers investor(s), the potential for higher returns with faster prepayments under certain interest rate scenarios. exhibit 6

Bond Class

EFFECT ON YIELD TO MATURITY AS PREPAYMENT SPEEDS INCREASE (IO) –300 bp

IO Increases (IO Receives All Penalties) IO Increases (Make Bonds Whole)

–100 bp

Unchanged

Interest Rates +100 bp +300 bp

Increases

Increases

Increases

Decreases

Increases

Increases

Increases

Decreases

Source: Morgan Stanley

TRENDS I N C MBS C ALL P ROTECTION

Conduit loan originators have established defeasance as the standard call protection in their origination programs. Which forms of prepayment protection do borrowers prefer? In some scenarios, yield maintenance is more expensive for borrowers and in some scenarios defeasance is more expensive for borrowers. We present three numerical examples of the costs to the borrower of defeasance versus yield maintenance in the Appendix. Yield maintenance and defeasance have very similar penalty calculations in current to lower interest rate environments although defeasance may be slightly more expensive in steeper yield curve environments. In rising interest rate environments, yield maintenance and defeasance prepayment disincentives tend to decline, but yield maintenance is eventually disadvantageous to the borrower. We think the increasing influence of the CMBS market in commercial real estate finance will result in the dominance of defeasance in commercial mortgage loans.

Please see additional important disclosures at the end of this report.

51

Transforming Real Estate Finance chapter 3

Call Protection Investors in CMBS have different sets of opportunities and investment decisions with defeased deals versus deals with yield maintenance. The defeased deal eliminates the cash flow volatility resulting from commercial mortgage prepayments. Investors in CMBS with yield maintenance call protection have exposure to the benefits and risks resulting from prepayment driven cash flow variability. Investors should carefully analyze the effect of interest rate movements and prepayments on the yield of individual CMBS classes. APPENDIX: T HE B ORROWER’S P ERSPECTIVE

The following example compares a prepayment under defeasance to a prepayment under yield maintenance from the borrower’s perspective. Assume the underlying loan has the following characteristics: • Amount: $10,000,000 • Term: 10 years • Amortization: 25 years • Coupon: 7.50% We will assume that the reference rate on the yield maintenance penalty is Treasuries flat and that a minimum one percentage point fee is applicable to any prepayment. Scenario 1:

In five years, U.S. Treasury Rates are unchanged, but the borrower wants to sell the property for personal reasons. • Loan Balance at the End of Year 5: • Yield Maintenance Penalty: • Cost of Treasury Strips: • Total Defeasance Cost: • Defeasance Advantage to Borrower:

$9,189,718 $779,020 (8.48%) $9,970,826 $781,108 (8.50%) ($9,970,826–$9,189,718) -$2,088 (0.02%)

So, the defeasance option in this case is more expensive to the borrower by 0.02% or $2,088.

52

Scenario 2

In five years, rates have risen just above the borrowers mortgage coupon, but the borrower wants to refinance to monetize the equity appreciation in his property. The 5-year Treasury is at 7.79%. • Loan Balance at the End of Year 5: • Yield Maintenance Penalty (min 1%): • Cost of Treasury Strips: • Total Defeasance Cost: • Defeasance Advantage to Borrower:

$9,189,718 $91,897 (1.00%) $9,172,606 ($17,112) (0.19%) $109,009 (1.19%)

The defeasance option is cheaper for the borrower by 1.19% or $109,009. Scenario 3:

Five years from now, rates will have risen to their highest levels in a decade, with 5-year Treasury rates at 9%. The borrower wants to sell the property. • • • • •

Loan Balance at the End of Year 5: Yield Maintenance Penalty (min 1%): Cost of Treasury Strips: Total Defeasance Cost: Defeasance Advantage to Borrower:

$9,189,718 $91,897 (1.00%) $8,741,211 $(448,507) (4.88%) $540,404 (5.88%)

In this scenario, the borrower effectively has the option of prepaying the loan on a discounted basis. This results in a cost savings of 5.88%, or $540,404 versus the amount paid under yield maintenance. Our example assumes a fairly flat yield curve. All else being equal, the defeasance option gets more expensive to the borrower as the yield curve steepens. Under the defeasance option, each mortgage loan payment is discounted at its corresponding zero coupon Treasury rate. When there is a larger spread between 1-year rates and 10-year rates, the discounted present value of the Treasury strip payments is higher, resulting in a greater cost to the borrower. Even in a steep yield curve environment, however, the defeasance option still allows the borrower the opportunity to prepay the loan at a discount. This option is not available under yield maintenance since the best a borrower can do is prepay at par.

Please see additional important disclosures at the end of this report.

53

This Page Intentionally Left Blank

54

Chapter 4

Transforming Real Estate Finance

AAA Fixed Rate CMBS

Please see additional important disclosures at the end of this report.

55

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS This chapter discusses some of the different types of AAA classes that have evolved in the fixed-rate CMBS market over the past few years. Today, transactions are issued with multiple AAA classes, and the characteristics of AAAs are different from the classes issued in the past. In addition to front-pay bonds and tight window bullets, the AAA CMBS market also includes multifamily directed classes, amortizing bonds, wide window bonds, super senior and junior AAAs as well as the traditional 10-year AAA bullet. In this chapter, we discuss these bonds, as well as dollar price 10-year AAAs.

Tight Window AAA Bonds In the late 1990s, the AAA CMBS world was a simpler place. Most conduit transactions were issued with only two AAA classes: a current pay 5-year wide window bond and a locked-out, 10-year bond. The more recent breed of 5-year AAAs is typically not structured as the frontpay class. With shorter classes ahead of the 5-year bond in the capital structure, many 5-year AAAs resemble bullets. They are initially locked out from receiving principal payments and ultimately pay down in a short period of time.

exhibit 1

CHARACTERISTICS OF 2004 5-YR AAA (TIGHT WINDOW) CLASSES

Tight-Window 5-Year AAA Bond CGCMT 2004-C1 A2 BACM 2004-2 A2 LBUBS 2004-C6 A2 COMM 2004-LB2A A2 CSFB 2004-C3 A3 LBUBS 2004-C2 A2 WBCMT 2004-C14 A2 BSCMS 2004-T14 A2 GCCFC 2004-GG1 A3 JPMCC 2004-PNC1 A2 LBUBS 2004-C4 A2 LBUBS 2004-C7 A2 MSC 2004-HQ4 A3 WBCMT 2004-C11 A2 COMM 2004-LB3A A2

Principal Window (Months) 6 7 3 4 6 3 8 6 6 6 4 3 5 2 12

Source: Morgan Stanley, Trepp, Annex A Files

56

# of Maturing Loans Tied to Payoff of Class 29 26 17 13 9 9 9 7 5 5 5 5 4 3 2

Original WAL (yr) 4.8 4.8 4.9 4.8 4.8 4.8 5.2 4.6 4.7 4.7 4.7 4.9 5.0 4.9 4.9

In order to gain a better understanding of the risk within short, locked out AAA classes, we analyzed a group of tight-window AAA bonds. We examined the Trepp universe of 2004 conduit transactions and captured all 5-year average life AAA bonds with tight windows1. Our analysis captures 15 deals that include 5year tight-window classes, most of which exhibit a strong dependence on the performance of a small percentage of loans in the deal. The timely repayment of principal on tight-window, locked out AAA classes may be dependent on the ability of a few loans to pay off as scheduled. That is, a transaction may be backed by more than 100 loans, but the payoff of the tight-window AAA bond may be tied to the cash flows of only two or three loans. This concentrated risk is not necessarily consistent with the level of risk an investor may expect from a diversified pool of loans in a conduit/fusion transaction. In our analysis, we found that the majority of 5-year, tight-window AAAs issued in 2004 will be paid down by fewer than 10 maturing loans. (See Exhibit 1)

Property Type Concentration of Maturing Loans Tied to Tight Window Payoff (Based on Original Balance)

Office 15% 21% 1% 44% 46% 66% 22% 62% 34% 0% 69% 30% 44% 33% 0%

Retail 0% 51% 51% 19% 11% 18% 77% 26% 66% 87% 30% 70% 0% 31% 92%

Multifamily 10% 26% 0% 34% 43% 15% 1% 0% 0% 0% 0% 0% 52% 0% 8%

Lodging 0% 0% 8% 0% 0% 0% 0% 9% 0% 13% 0% 0% 0% 0% 0%

Manufactured Housing 15% 0% 24% 3% 0% 0% 0% 0% 0% 0% 0% 0% 4% 36% 0%

Self Storage 60% 2% 16% 0% 0% 1% 0% 3% 0% 0% 1% 0% 0% 0% 0%

1

We define a tight principal window as being 12 months or less

Please see additional important disclosures at the end of this report.

57

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS RISK I S T IED T O T HE N UMBER O F M ATURING L OANS

Within the examined universe, COMM 2004-LB3A A2 has the most concentrated risk of the 2004 tight-window 5-year AAAs, with only two loans tied to its principal repayment. Although five loans actually mature while the A2 principal window is open, three of these loans are tied to the repayment of the multifamily directed class and do not contribute to the repayment of the A2 class. The class diverts the cash flows of some multifamily loans away from the tight-window class. Therefore, the impact of a few loans on the tight-window class is intensified. The COMM bond has concentrated risk in the performance of two loans, however, it is important to note a mitigating factor. Of the 15 bonds that we analyzed, this class has the longest principal window. Therefore, 15% of the bond’s original balance is paid down through amortization of the collateral pool prior to the two balloon maturities. CGCMT 2004-C1 A2 has the least concentrated risk, in of number of loans linked to final payoff of the bond. Principal cash flows from 29 loans are involved in the payoff of the A2 class. However, there are other factors to consider in addition to the number of loans involved in the payoff. For example, 60% of the loans (based on original balance) that will pay down the A2 class are backed by self storage properties. Although CGCMT 2004-C1 A2 offers the exhibit 2

EXPOSURE TO LARGEST MATURING LOANS INVOLVED IN PRINCIPAL REPAYMENT OF 5-YR AAA (TIGHT WINDOW) CLASSES

Class CGCMT 2004-C1 A2 BACM 2004-2 A2 COMM 2004-LB2A A2 CSFB 2004-C3 A3 LBUBS 2004-C6 A2 LBUBS 2004-C2 A2 WBCMT 2004-C14 A2 BSCMS 2004-T14 A2 GCCFC 2004-GG1 A3 WBCMT 2004-C11 A2 JPMCC 2004-PNC1 A2 LBUBS 2004-C4 A2 MSC 2004-HQ4 A3 LBUBS 2004-C7 A2 COMM 2004-LB3A A2 Source: Morgan Stanley, Trepp, Annex A Files

58

# of Maturing Loans Involved In Payoff of Class 29 26 13 9 17 9 9 7 5 3 5 5 4 5 2

Original Balance of Maturing Loans Involved In Payoff of Class ($MM) 163.0 242.2 162.4 244.7 328.5 317.1 234.9 176.2 280.5 69.1 126.1 404.4 46.8 278.0 81.7

most diversified risk based on the number of loans tied to pay off, investors need to decide if they are comfortable with heavy exposure to self storage assets. CONSIDER L ARGE L OAN P APER A S A N A LTERNATIVE

We urge investors to examine the loans that are tied to the repayment of the tight-window bond. An investor’s analysis of tight-window 5-year AAAs should be in-line with the type of monitoring and analysis required for a large loan deal. In fact, tight-window, AAA investors should consider fixed-rate large loan paper as an alternative investment. Currently, 5-year, tight-window AAAs are trading in the swaps + high teens area in the secondary market. In comparison, 5-year bullets on clean, single asset deals are trading in the low 30s to swaps.

Concentration Risk of Largest Maturing Loans Involved in Principal Repayment (Based on Orig. Bal.)

Largest Loan 15% 19% 16% 31% 24% 59% 64% 42% 32% 36% 44% 46% 52% 54% 92%

2nd-Largest Loan 9% 17% 15% 22% 21% 17% 19% 24% 30% 33% 33% 29% 24% 24% 8%

3rd Largest Loan 7% 10% 15% 11% 21% 7% 5% 20% 17% 31% 10% 23% 20% 16% NA

4th Largest Loan 5% 8% 13% 10% 11% 6% 4% 9% 16% NA 9% 1% 4% 5% NA

5th Largest Loan 4% 8% 10% 9% 8% 4% 3% 2% 4% NA 4% 1% NA 1% NA

Please see additional important disclosures at the end of this report.

Total Exposure to Largest 5 Loans 40% 62% 69% 83% 85% 93% 95% 97% 99% 100% 100% 100% 100% 100% 100%

59

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS RISK I S A LSO T IED T O T HE S IZE O F M ATURING L OANS

In analyzing 5-year, tight-window AAAs, investors should also examine a bond’s exposure to the largest loans involved in paying down that class. Simply counting the number of loans involved in the payoff of a class may not necessarily capture the concentration risk accurately. For example, at first glance, BACM 2004-2 A2 appears to offer more diversification than COMM 2004-LB2A A2. With 26 loans tied to principal payoff, the BACM bond is affected by the cash flows of twice as many loans as the COMM bond, which is tied to 13 loans. However, on closer examination, the concentration risk of these bonds is more similar than the number of loans implies. In the case of the BACM bond, the largest five loans (of the 26 total loans) for about 62% of the loan balances available for pay down of the A2 class. The largest five loans affecting the COMM bond for 69% of loans tied to the principal repayment of the A2 class. EXTENSIONS O N 5 -Y Y R T IGHT-W W INDOW B ONDS

Principal loss is not a primary concern at the AAA level, but extension risk may be. The CMBS market has not yet experienced a wave of loan maturities, but we can look to the extension experience of floating rate large loans as a conservative proxy for the extensions that may occur in the fixed rate market. We view the extension experience of floating rate large loans to be a conservative proxy, as we believe these loans are riskier and more likely to extend than fixed rate loans. Floating rate loans are often made on transitional properties and are structured with extension options, while fixed rate loans are typically backed by stabilized assets. Some fixed rate loans may be ARD loans, but we would only expect credit impaired loans to extend beyond their anticipated repayment dates. We recently exhibit 4

2004 5-YEAR AAA (TIGHT WINDOW) CLASSES LOANS SCHEDULED TO MATURE BEFORE OPENING OF TIGHT WINDOW

Class BACM 2004-2 A2 COMM 2004-LB2A A2 CSFB 2004-C3 A3 CSFB 2004-C3 A3 GCCFC 2004-GG1 A3 LBUBS 2004-C2 A2 LBUBS 2004-C2 A2 LBUBS 2004-C2 A2 LBUBS 2004-C7 A2 MSC 2004-HQ4 A3 WBCMT 2004-C11 A2 Source: Morgan Stanley, Trepp, Annex A Files

60

Property Name Cargill Office Building Eldorado Mobile Home Estates One Crown Center Corona Plaza Shopping Center DDR Portfolio Brittany Square Fairview Apartments Walgreens - Cheyenne Avenue Executive Plaza - Waco 387-403 Fulton Avenue Pirate Plaza Shopping Center

exhibit 3

EXTENSION RATES BY PROPERTY TYPE (BASED ON ORIGINAL BALANCE)

Property Type Hotel Office Multifamily Retail Other Total

Extension Rate (%) 40 30 18 8 13 23

Extension rate represents % of loans that exercised extension options at maturity. Source: Morgan Stanley, Intex, Trepp

published a piece that examined extension rates on floating rate large loans.2 Extension rates for floating rate loans are shown by property type in Exhibit 3. Extension risk on a 5-year tight-window class may also be dependent on loans that mature outside of its principal window. For example, if extensions occur on loans that are scheduled to mature prior to the principal window opening, the 5-year AAA class may be affected. Of the 15 tight-window bonds that we examined, eight contain loans that are scheduled to mature prior to the opening of the tight window.3 Therefore, in addition to examining the loans that mature within the tight window, investors concerned with extension should also examine the loans that mature prior to the opening of the tight window.

Original Loan Balance ($MM) 18.0 2.3 6.0 4.3 48.8 4.3 4.1 4.3 3.4 3.3 0.8

Current Loan Balance ($MM) 18.0 2.3 6.0 4.3 48.4 4.3 4.1 4.3 3.4 3.3 0.8

Maturity Date 9/1/2008 10/1/2008 12/11/2008 2/11/2009 3/1/2008 10/11/2008 11/11/2008 12/11/2008 5/1/2009 3/1/2009 10/11/2008

Property Type Office Mobile Home Office Retail Retail Retail Multifamily Retail Office Retail Retail

2

For more information, see chapter 11 of this primer.

3

Does not include LBUBS 2004-C6, in which 2 maturing loans are tied to the paydown of the multifamilydirected class.

Please see additional important disclosures at the end of this report.

61

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS Multifamily Directed Classes Multifamily directed classes, often labeled as A-1A bonds in the structure, are collateralized by multifamily commercial mortgages. If a deal has a multifamily directed class, cash flows of most of the collateral pool’s multifamily mortgages are tied exclusively to paying down that class. The multifamily directed class is intended to encourage GSE participation. According to the Commercial Mortgage Alert database of new issue transactions, 66% of the conduit/fusion deals issued in 2004 contain multifamily directed (A-1A) classes. Despite the widespread use of A-1A tranches, all else being equal, we prefer transactions that do not contain multifamily directed classes. The use of A-1A classes within deal structures essentially has the effect of reducing collateral diversity for non-A-1A holders. At first glance, bondholders in deals without A-1A classes look no better off than bondholders in deals containing A-1A classes – that is, the collateral composition of transactions containing A-1A classes looks as diverse as the collateral composition of transactions without A-1A classes (see Exhibit 5).

exhibit 5a

2004 FIXED RATE CMBS COLLATERAL COMPOSITION: DEALS WITH A-1A CLASSES

Source: Commercial Mortgage Alert

exhibit 5b

2004 FIXED RATE CMBS COLLATERAL COMPOSITION: DEALS WITHOUT A-1A CLASSES

Source: Commercial Mortgage Alert

62

However, the majority of multifamily loans in transactions containing A-1A classes are tied to multifamily directed classes. As a result, the existence of an A-1A class results in a less diversified collateral pool for non-A-1A bondholders (in of extension risk) and exposes these investors to higher concentrations of fewer property types. For example, in JPMCC 2004-CBX, multifamily loans comprise 23.6% of the collateral pool. However, the majority of these loans are tied to the A-1A class and do not contribute cash flows to the rest of the capital structure. If we calculate property type exposures for the non-A-1A classes based on the loans that contribute cash flows to these bonds, we find that the non A-1A classes have less than 1% exposure to multifamily cash flows. Exposure to office properties jumps to 38.4% from 29.6%, and retail exposure increases to 35.5% from 27.4%. Four other recent fixed rate CMBS with A-1A classes exhibit a similar trend. AAA classes with tight principal windows are particularly sensitive to the presence of an A-1A class. Since the principal repayment of tight window classes is typically tied to a handful of loans, the A-1A class further reduces diversification by diverting most multifamily cash flows away from the tight window bonds. Multifamily cash flows that would have otherwise been used to pay down bonds in the absence of a multifamily directed class are directed to pay down A-1A instead. In 2004, the IQ, HQ, TOP and PWR shelves were among the new issue programs that did not utilize multifamily directed classes.

Amortizing AAA Bonds In November 2004, the BACM 04-5 transaction priced with 2 tight-window AAA classes in addition to its 10-year AAA classes. The 5-year and 7-year tight window classes were created by incorporating a unique class (class A-AB) into the deal’s structure.

exhibit 6 Class

BACM 2004-5 AAA BONDS

Size ($MM)

Subordination (%)

A/L (Years)

Principal Window

Priced

57.600

20.00

2.89

1-56

S+26

A-1A

241.609

20.00

6.34

1-118

NA

A-2

250.910

20.00

4.77

56-60

S+17

A-3

305.377

20.00

6.77

80-84

S+24

45.540

20.00

6.99

56-110

S+32

A-4

188.667

20.00

9.64

110-118

S+29

A-J

90.241

13.38

9.80

118

S+33

A-1

A-AB

Source: Commercial Mortgage Alert, Trepp

Please see additional important disclosures at the end of this report.

63

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS In this section, we refer to Class A-AB as an amortizing bond. The A-AB bond is structured to absorb the monthly P&I payments from loan amortization while balloon payments are directed to pay down the tight window classes. CONCLUSIONS

We believe that class A-AB is cheap relative to one other recently issued class with a wide principal window, comparable average life and identical subordination level. We recommend this new type of bond for investors whose primary market concern is extension risk. Under no reasonable scenario were we able to extend class A-AB. We did find, however, that the class A-AB principal window shortens with minimal default stresses, although the bond will not pay off sooner than 2011 under our most severe scenarios. Early repayment of principal will provide upside for investors if they purchase this bond at a discount in the future. With rates on the rise, it is likely that this bond will trade at a discount in the future. LIMITED F UTURE I SSUANCE

We expect future issuance of amortizing bonds to be limited since its existence is dependent upon having enough 5-year and 7-year loans in the collateral pool to create tight window classes. If the market remains predominantly a 10-year balloon mortgage market, and deals are issued with low percentages of 5- and 7-year loans, we will not see heavy issuance of amortizing bonds. The BACM 2004-5 collateral pool contains 30% 5-year loans, 24% 7-year loans and 46% 10+ year loans. Even with this seemingly heavy concentration of 5and 7-year loans, the size of the A-AB class is small ($45.5 million) compared to other AAA classes.

exhibit 7

BACM 2004-5: LOAN CONCENTRATION

Source: Morgan Stanley, Trepp

64

EXTENSION P ROTECTION

As extension risk is a concern for many investors, the extension protection offered by the A-AB class is attractive. We found the A-AB class to have virtually no extension risk. This extension protection exists for two reasons: • Payments to A-AB bond holders are not tied to balloon maturities. • Class A-AB is at the top of the principal distributions waterfall, so its balance is reduced to the planned principal amount prior to other classes receiving principal. Under reasonable assumptions, we would expect the A-3 class to pay down prior to the A-AB class. This is because the A-AB class will not receive additional payments greater than its planned principal amount until classes A-1, A-2 and A-3 are fully paid off. However, even in a scenario where all 15 loans tied to the timely payoff of the A-3 class extend, the timing of the A-AB cash flows will be unaffected. The A-AB class will continue to receive payments from the monthly loan amortization of the remaining loans and pay off in 2014.

exhibit 8

Scenario Base Case Extension 1 Extension 2

BACM 2004-5 A-AB: EXTENSION STRESS SCENARIOS

Prepayment Default Details Speed 0% CDR 0 R 12-Month Extension of All Collateral 0 R 12-Month Extension of Loans Tied to the Payoff of A-3 0 R

Principal Window 07/09-01/14 7/09-1/14 7/09-1/14

A/L (Years) 6.998 6.998 6.998

Note: Loss severity in all scenarios is assumed to be 35% with a 12-month recovery period. Source: Morgan Stanley, Trepp

We were only able to extend the average life of class A-AB under a very severe, unrealistic scenario. When we defaulted and liquidated the largest five loans in the transaction, which for 29.8% of the deal, the A-AB class extended by less than one month.

Please see additional important disclosures at the end of this report.

65

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS EARLY R EPAYMENT O F A -A AB

Despite the lack of extension risk, the A-AB bond is sensitive to principal window contraction. The base case scenario of no prepayments or defaults predicts payoff in 2014, but the bond’s principal window shortens with minimal default stresses. The amortizing bond is sensitive to an earlier payoff because of its small class size. This class was originally structured to receive loan amortization payments, not balloon payments. Therefore, the small A-AB class can be reduced to zero prior to its scheduled maturity date if a large balloon payment becomes tied to its repayment. This can occur in a default scenario where liquidation proceeds are ed through the top of the structure, thereby promoting the payoff of the A1, A2 and A3 classes with unscheduled principal that would not otherwise be tied to these classes. Any leftover principal cash flow that was originally scheduled to pay down the A3 class in the base case scenario is now directed to the A-AB class.

exhibit 9

Scenario Base Case Default 1 Default 2 Default 3 Default 4 Default 5

BACM 2004-5 A-AB: DEFAULT STRESS SCENARIOS

Default Details 0% CDR 0% CDR for 24 months; 0.5% CDR thereafter 0% CDR for 24 months; 2% CDR thereafter 0% CDR for 24 months; 5% CDR thereafter 3% CDR starting immediately Default Bank of America Center on 12/06

Prepayment Speed 0 R 0 R 0 R 0 R 20 R 0 R

Principal Window 07/09-01/14 07/09-07/13 07/09-06/12 07/09-11/11 07/09-11/11 07/09-01/14

Note: Loss severity in all scenarios is assumed to be 35% with a 12-month recovery period. Source: Morgan Stanley, Trepp

66

A/L (Years) 6.998 6.782 6.472 6.408 6.351 6.998

In default scenario 1, we assume no defaults in the first 24 months and then a constant annual default rate of 0.5% for the remainder of the deal. Loss severity is assumed to be 35% with a 12-month recovery period. Rather than paying off in 2014, the bond pays off in 2013, reducing the length of the principal window by six months. Using a more severe scenario, with 3% annual defaults starting immediately, and a 20% R prepayment assumption, the final cash flow on the A-AB bond is paid in 2011. The shortening of this class is limited to a 2011 payoff, since there are not enough 5-year loans in the pool to pay off the A-AB class prior to 2011. BACM 2 004-5 5 A -A A B L OOKS C HEAP

The BACM 2004-5 transaction priced on November 8, 2004. Class A-AB, a 6.99 year bond with 20% credit enhancement priced at swaps + 32 bp. One week later, on November 15th, the JPMCC 2004-CBX deal priced. The JPMCC transaction does not have an amortizing bond, but does contain a comparable average life AAA bond (6.61 yr) with a wide principal window and a subordination level of 20%. This bond (class A-4) priced at swaps + 25 bp, 7 bp tighter than the BACM amortizing A-AB class. Although there may be some spread concession for the small size of the A-AB class and for the wider principal window, we believe the BACM amortizing class, with its top priority in the principal distribution waterfall and stable cash flows, offers value.

Please see additional important disclosures at the end of this report.

67

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS CMBS Bonds A priced CMBS bond can be examined for value by breaking it into two components: a par priced fixed-rate bond and an IO. By examining the spread on the embedded IO and comparing it to IOs in the market, we can determine whether the bond is undervalued. To analyze our sample of bonds, we used the Bond Splitter function on Trepp. We examined a sample of second-pay AAA bonds with dollar prices. For our analysis, we examined 2001 and 2002 conduit transactions. We examined all embedded IOs under a 100Y prepayment assumption. We chose to analyze transactions from 2001 and 2002, as recent vintages tend to have stronger call protection than deals issued in 1997 or 1998. Analyzing older transactions using a 100Y scenario may be too aggressive and would result in less attractive embedded IOs than those found in bonds of more recent vintages. We performed this analysis in June 2003, after interest rates rallied significantly, resulting in priced CMBS trading at even higher dollar prices. At the time, our trading desk estimated that on average, 10-yr AAA bonds off of conduit deals were trading 3-4 points higher than they were about 1 month earlier. We found that some bonds offered 5-10 bp of upside potential because their dollar prices rose high enough, such that the concession offered in spread for each point of made the embedded IO look cheap in comparison to market priced PAC IOs. In the sample of AAAs that we examined for this analysis, the dollar prices ranged between $112 and $119. Besides very high dollar prices, an investor should also consider the size of the front-pay AAA class. Second-pay AAAs that are preceded by larger

exhibit 10

ANALYSIS FOR CMBS AAA BONDS

Deal Name SBM7 2001-C2 BAFU 2001-3 JPMCC 2001-CIB2 MSDWC 2001-TOP1 GECMC 2001-1 CSFB 2002-CKP1 GMACC 2001-C2 WBCMT 02-C1 LBUBS 2002-C1 Source: Morgan Stanley, Trepp

68

Class A3 A2 A3 A4 A2 A3 A2 A4 A4

Spread to Swaps on Bond (Assuming 0,0,3CDR) 46 43 49 47 49 50 51 50 52

Dollar Price 118-30.240 112-3.750 118-4.750 119-15.875 118-26.375 118-12.375 119-15.875 117-10.875 118-18.875

front- pay AAAs may look attractive compared to second-pay AAAs that are preceded by smaller front-pay AAAs. A larger front-pay class can shield the priced second-pay AAA bond from recoveries on defaults, thereby preserving yield on the embedded IO. One of the bonds that we analyzed was GMACC 2001-C2 A2. This class currently has 23.4% credit enhancement and a dollar price of $119-15.875. GMACC 2001-C2 A2 has an average life of 7.84, so we assumed that the embedded par bond has a spread of swaps + 32 bp. New issue 10-year AAA bonds from conduit transactions trade around swaps + 36 bp. We arrived at swaps+32 bp for the 7.84-year embedded par bond by making the market assumption that each year of seasoning is worth 2 bp of spread. We then calculated spreads for the embedded IO under different default scenarios. In each scenario, we applied defaults after 24 months. Assuming no defaults for 24 months and 3% CDR thereafter, the embedded IO yield was T+177 bp. Running 6% CDR instead resulted in an IO yield of T+119 bp. We assumed a 34% loss severity on defaults and a 12-month recovery period. We then compared spreads on the embedded IO to new issue PAC IO spreads. The PAC IO is similar to the embedded IO in that it is well shielded from losses. The PAC IO is typically structured to withstand up to 6% annual defaults. Currently, new issue PAC IOs are pricing at T+95 bp. The embedded IO in GMACC 2001-C2 A2 looks cheap in comparison, as yields range from T+119 bp to T+177 bp, depending on the default scenario.

Embedded IO Spread to UST Assuming 3% CDR after 24 Months 118.1 172.6 161.6 168.3 163.4 178.4 176.5 192.6 204.1

Embedded IO Spread to UST Assuming 6% CDR after 24 Months 55.7 122.4 108.7 136.8 104.3 122.3 119.1 164.6 159.3

Spread to Swaps on Bond when IO Spread is T+95 bp 44.2 38.2 43.1 40.1 42.7 42.4 43.3 41.5 42.0

Please see additional important disclosures at the end of this report.

Cheapness to Hypothetical PAC IO & AAA Combination 1.8 4.8 5.9 6.9 6.3 7.6 7.7 8.5 10.0

69

Transforming Real Estate Finance chapter 4

AAA Fixed Rate CMBS HOW C HEAP I S G MACC 2 001-C C 2 A 2?

The analysis above shows that the GMACC 2001-C2 A2 bond is cheap, but does not address the question of how cheap. We ran a breakeven analysis to figure out how much the A2 bond would have to tighten before an investor would be indifferent between owning the bond versus owning a par bond and PAC IO at T+95 bp. We found that if we assume the embedded IO is priced at T+95 bp under a 3% CDR scenario, the A2 bond would yield swaps+43 bp. This implies about 8 bp of value, since the A2 bond currently trades at swaps +51 bp. Using this methodology, we found that all of the bonds in our sample offer value under a 3% CDR scenario. Under 6% annual defaults starting in 24 months, the embedded IO in SBM7 2001-C2 A3 class yields T+56 bp. The PAC IO, in comparison, would not break under this default scenario and continue to yield T+95 bp. For investors that consider the 6% CDR scenario to be a reasonable assumption, the SBM7 2001-C2 A2 class would not provide good value (See Exhibit 10).

70

Chapter 5

Transforming Real Estate Finance

Total Rate of Return Swaps

Please see additional important disclosures at the end of this report.

71

Transforming Real Estate Finance chapter 5

Total Rate of Return Swaps While Total Rate of Return Swaps (TRS) are not a new innovation in the CMBS market, they became an area of increased interest during the tight spread environment in 2004. With AAA CMBS trading at swaps +25 bp in January 2004, we received a number of questions on the TRS market during our roadshow, as investors searched for spread outside of the cash market. WHAT I S A T RS

A TRS is a method of synthetically replicating the returns from the CMBS market. A TRS is a contract in which one party agrees to pay the total return on a reference index or a basket of CMBS, while the other party pays a floating rate, based on 1-month LIBOR. Total return is comprised of two components: income and price appreciation or depreciation. The income component includes any interest that is paid during the life of the TRS contract, as well as interest that has accrued during that time. TRS are short-term contracts, typically three to twelve months in length. However, they reference a basket/index with durations typically found in 5-yr or 10-yr average life CMBS cash bonds. Investors may choose to be TRS payers to hedge long CMBS or loan positions, or choose to be receivers to obtain a diversified investment in CMBS through a single trade.

exhibit 1

TOTAL RATE OF RETURN SWAP (TRS)

Source: Morgan Stanley

72

BREAK-E E VEN A NALYSIS

How much CMBS spread widening can a TRS receiver tolerate before the trade loses money? In order to answer this question, we first need to isolate the TRS from other sources of profit and loss, such as interest rate or swap rate movements. In the diagram below, we have created a hypothetical example where an investor agrees to receive the total return on a basket of 10-year AAA CMBS, and simultaneously enters into an interest rate swap to mitigate interest rate risk. Let’s assume that the average yield on the basket of AAA securities is swaps +30 bp, and the average duration on these bonds is about seven years. In exchange for the total return on the AAA index, we will assume the investor pays 1-month LIBOR - 40 bp. For simplicity, we will perform the calculation assuming a 1-year time frame.

exhibit 2

THE RECEIVER PAYS FIXED ON INTEREST RATE SWAP

Source: Morgan Stanley

We can calculate the investor’s income by examining the cash flows from the two swaps. The investor receives a yield of swap rate + 30 bp on one leg of the TRS and hedges the interest rate exposure by paying the fixed leg of an interest rate swap. The investor also receives LIBOR from the floating rate leg of the interest rate swap and pays LIBOR - 40 bp on the TRS. Overall, the investor’s income stream is (swap rate + 30 bp) - swap rate + LIBOR - (LIBOR-40 bp). Therefore, the total income from these 2 trades is equal to 70 bp.

Please see additional important disclosures at the end of this report.

73

Transforming Real Estate Finance chapter 5

Total Rate of Return Swaps The investor’s break-even point on the trade occurs if the price of the AAA CMBS bonds in the index declines by 0.70%, due to spread widening versus swaps. A 0.70% decline in price corresponds to approximately 10 bp of spread widening over the course of 1 year. (In a typical 3- or 6-month TRS contract, an investor’s break-even point would be reached after 2.5 bp or 5 bp of CMBS widening, respectively.) ∆P/P = -0.70% ∆P/P= Y*Duration ∆Y*Duration = -0.70% ∆Y = 0.70% ÷ 7 ∆Y = ~10 bp An investor who chooses to buy the same basket of AAA bonds in the cash market could withstand about 4 bp of spread widening over the course of a year (30÷7 = 4.3 bp). The TRS receiver can withstand more widening on the CMBS basket because there is an additional 40 bp of cushion from the TRS that helps mitigate any CMBS spread widening that occurs. The 40 bp of cushion in the TRS fluctuates over time with the amount of hedging demand. Hedging demand comes from dealers hedging long secondary trading positions, loan originators hedging conduit pipelines, and basis traders selling CMBS versus other asset classes. Historically, the floating leg of the TRS has traded between LIBOR-70 bp and LIBOR-10 bp. In 2003 and during the first half of 2004, the market ranged between LIBOR-70 bp and LIBOR-35 bp. During the second half of 2004, increased investor interest in the TRS market removed most of the spread advantage of receiving the index. FACTORS T O C ONSIDER

Total rate of return swaps may not be ideal for all investors. TRS require investors to have necessary swap documents before entering a trade. A consideration for non-mark-to-market investors is that TRS are effectively marked to market, since monthly payments are made based on the return of the index. Therefore, a TRS investor would experience regular realized gains or losses. Liquidity and counterparty risk are other factors to consider.

74

Chapter 6

Transforming Real Estate Finance

CMBS IOs

Please see additional important disclosures at the end of this report.

75

Transforming Real Estate Finance chapter 6

CMBS IOs The first five pages of this chapter are excerpted from a piece originally written in 1998. INTRODUCTION

The CMBS interest-only securities (IO) market has both grown and matured since its inception in the mid-1990s. In this chapter, we trace the growth and development of the CMBS IO market and factors affecting issuance. We examine current pricing conventions for IOs and look at several case studies to evaluate the impact of changing prepayment and default scenarios on IO returns. Some of our major findings are: • IOs offer an opportunity to ERISA constrained investors to buy credit sensitive CMBS. • CMBS IO investors will benefit if prepayments fall short of the 100% R pricing assumption or if loans are extended at the balloon date. • CMBS IO investors may benefit from prepayment penalties during yield maintenance periods. WHAT I S A C MBS I O?

As in the single-family residential market, CMBS IOs receive a coupon stripped from an underlying pool of mortgages or bond classes. Stripping a coupon allows an issuer to sell par (or near par) priced securities, even if the coupon on the underlying mortgages is well above the bond coupons. For example, a 1% IO strip may be created off of collateral with a 7% coupon in order to sell 6% parpriced securities. Both residential and CMBS IOs are typically rated AAA/Aaa. These ratings are based on priority of cash flow rather than credit, and are meaningless except for regulatory purposes. Any loan default affects the yield of an IO, but from the rating agency perspective, the default does not affect the IO’s senior priority in receiving existing cash flow. CMBS IOs are ERISA eligible investments because of their non-subordinated position in the structure. Since the priority of the IO never changes, it is unlikely to be downgraded even if the credit quality of the collateral declines. In a declining credit environment, principal and interest bonds are more vulnerable to a downgrade. Owners of these bonds may be forced to liquidate if they have minimum rating requirements. In such an environment, spreads on IOs could widen, but the rating for regulatory purposes would probably not change.

76

For IOs backed by residential mortgages, prepayments are the most important risk factor. Historically, defaults on residential mortgages have been a relatively minor factor. Over the life of a residential mortgage pool, defaults have typically totaled less than 5% of the original balance compared to prepayments that range from 5% to 30% per year. For commercial IOs, the situation is reversed, with defaults potentially having a greater impact than prepayments. Most securitized commercial mortgages have either a prepayment lockout, yield maintenance or defeasance. Historic cumulative lifetime commercial mortgage default rates at insurance companies, however, have ranged from 4% to 32% for cohorts with at least 10 years of seasoning (1972-1992). STRUCTURE

A traditional structure for a WAC CMBS IO is shown in Exhibit 1. Note that most of the IO’s cash flow is off of the AAA-rated classes, which typically constitute 70% or more of the principal balances of a CMBS. In addition, the coupons on the AA, single A, and BBB securities are higher than on the AAA bonds, so less IO is stripped off of these classes.

exhibit 1

TYPICAL WAC IO STRUCTURE

Source: Morgan Stanley

INVESTOR B ASE

Current buyers of CMBS IOs include life insurance companies and bank portfolios to enhance portfolio yield. In addition, CMBS IOs appeal to money managers who seek a high-yielding AAA-rated asset. IOs also appeal to investors seeking shorter duration CMBS instruments – the duration of a CMBS IO is about 4 years, compared to 7 years for classes rated BBB or BB. CMBS S PREADS A ND P REPAYMENT P RICING A SSUMPTIONS

Before analyzing the relative value of IOs versus other CMBS sectors, we discuss some of the current prepayment pricing conventions for CMBS IOs and then turn to credit analysis in the next section.

Please see additional important disclosures at the end of this report.

77

Transforming Real Estate Finance chapter 6

CMBS IOs The market currently prices CMBS IOs assuming that loans with yield maintenance prepayment penalties are equivalent to loans with absolute prepayment lockout provisions. Participants in the CMBS market examine IO yields and spreads using a 100% Y assumption, meaning that all loans prepay immediately after the lockout or yield maintenance period. Some investors also examine yields under a 0% R1 assumption. The difference in spreads using the 0% and 100% R assumptions is often referred to as the “slope” or drop in spread from the slowest to the fastest prepayment speed. The actual prepayment experience of a pool of commercial loans backing a CMBS can be quite different from the pricing assumptions. Loans can and have prepaid in the yield maintenance period. Since a share of the penalty is ed on to the IO holder, this can prove beneficial to a CMBS IO investor, depending on the level of interest rates when the loan prepays. In addition, both the 0% and 100% R assumptions are unrealistic. Actual prepayment speeds will vary on a monthly basis, with several months of 0% R possibly followed by a sharp one-month increase if a loan prepays. Unlike the residential mortgage market, there is very little historical data on commercial mortgage prepayments. Due to the high percentage of loans with defeasance or yield maintenance in today’s CMBS transactions, the primary investor analysis on CMBS IOs is default driven. Recoveries and losses on defaulted loans are now the primary factor affecting early payment of principal in a CMBS transaction. DEFAULT A SSUMPTIONS

Aside from prepayments, defaults are the major factor influencing CMBS IO yields. Since default eventually may remove a loan from a pool, it is detrimental to the return on a CMBS IO. The IO investor no longer receives the interest strip after a defaulted loan is finally liquidated. A default, then, has a similar effect on a CMBS IO investor as a prepayment. However, default at a balloon date and subsequent extension of a mortgage, which lengthens its life, is beneficial to the IO investor. In pricing a CMBS IO, CMBS IO market participants examine a variety of default assumptions, ranging from 0% CDR to 5% CDR or higher. “CDR” stands for conditional default rate and is analogous to R for prepayments (see footnote 1). Unlike prepayment assumptions, CDRs ignore any call protection feature on the loans. Given the historical pattern of commercial mortgage default rates, we believe that the expected average annual default rate on a newly originated pool of commercial mortgages lies between 1% CDR and 4% CDR.

1

78

R stands for “conditional prepayment rate.” It is an annualized prepayment rate expressed as a percentage of the remaining balance of the pool.

IMPACT O F P REPAYMENTS O N I O Y IELD

Exhibit 2 shows the effect of increasing prepayments on 5 different CMBS IOs. exhibit 2

PREPAYMENT RISK: YTM OF CONDUIT IOs IN PERCENT R 0

R 20

R 50

R 100

10.63

9.97

9.48

8.59

8.83

8.82

8.80

8.68

LBCMT 98-C1

10.36

9.85

9.46

8.59

FULBA 98-C2

9.95

9.50

9.19

8.68

MSC 98-WF2

9.27

9.18

9.08

8.67

DMARC 98-C1 NASC 98-D6

Priced at +350 at 100 R; yield maintenance equals lockout. Source: Morgan Stanley

The current pricing convention for CMBS IOs, however, is 100% R after yield maintenance or lockout. Reading exhibit 2 from right to left shows the potential upside to an investor if prepayments are slower than the pricing assumption. In the five examples cited, an investor has from 14 bp to 140 bp of upside if actual prepayments are 20% R rather than 100% R. The greater the slope of the increase in yield as prepayments decline, the greater the potential benefit to the investor. The magnitude of this slope is determined by the length of the open period between the end of the prepayment penalty period and maturity. This window period can range from as little as one month to as much as three years or more, depending on the of the loans in a given transaction. YIELD M AINTENANCE P ENALTIES A ND I O s

In most CMBS structures, IOs receive a share of any prepayment penalties paid by borrowers. In early structures (pre-1997), the IO could receive as much as 100% of a yield maintenance penalty. More recently, CMBS IOs receive a share of the penalty according to a formula such as: P = % of penalty paid to IO =

1 –

(Bond Coupon – UST) (Mortgage Coupon – UST)

In this example, if the principal and interest bond coupon were 6.0%, the mortgage coupon 7.5%, and the UST 4.75%, the IO would receive 55% of the yield maintenance penalty. If the mortgage coupon is less than the UST, the borrower does not pay a penalty. If a penalty is paid, and the bond coupon is less than the UST, then the IO receives all of the penalty.

Please see additional important disclosures at the end of this report.

79

Transforming Real Estate Finance chapter 6

CMBS IOs For non-defeasance loans, prepayments during the yield maintenance period are beneficial to the IO investor in a stable rate environment. Exhibit 3 shows that in 4 out of the 5 sample deals, yields to maturity increase as prepayments increase if interest rates remain unchanged. The sole exception, NASC 98-6, is a CMBS backed by loans with defeasance. exhibit 3

PREPAYMENT RISK: YTM OF CONDUIT IOs: INTEREST RATES UNCHANGED (IN %) R 0

R 20

R 50

R 100

10.63

13.51

15.42

17.66

8.83

8.82

8.80

8.68

LBCMT 98-C1

10.36

11.44

12.48

14.31

FULBA 98-C2

9.95

10.72

11.37

12.10

MSC 98-WF2

9.26

10.85

11.63

13.48

DMARC 98-C1 NASC 98-D6

Priced at +350 at 100 R; yield maintenance equals lockout Source: Morgan Stanley

A sharp rise in rates coupled with high prepayments, however, is detrimental to the yield to maturity of non-defeased IOs. Exhibit 4 shows the effect of an instantaneous 300 bp increase in interest rates during the yield maintenance period. Under this scenario, the drop in yield from 0% R to 100% R ranges from 15 bp to more than 1500 bp. exhibit 4

PREPAYMENT RISK: YTM OF RECENT CONDUIT IOs: INTEREST RATES UP 300 bp (IN %) R 0

R 20

R 50

R 100

10.63

6.89

3.32

-1.03

8.83

8.82

8.80

8.68

LBCMT 98-C1

10.36

6.13

1.48

-5.15

FULBA 98-C2

9.95

8.20

6.96

5.51

MSC 98-WF2

9.26

5.22

1.41

-2.31

DMARC 98-C1 NASC 98-D6

Treasury rates up 300 bp; priced at +350 at 100 R; yield maintenance does not equal lockout; dollar price has not been adjusted for rate change. Source: Morgan Stanley

While IO bond yields decline greatly when prepayments and interest rates are both high, we believe that these two factors are typically inversely correlated. Higher rates clearly dampen the refinance incentive. On the other hand, if high rates are coupled with increases in property prices, borrowers may be able to refinance and take proceeds out of a property.

80

The most notable feature of exhibits 3 and 4 is the stable profile of CMBS IOs backed by defeasance loans, as demonstrated by NASC 98-D6. The loans in this transaction must be defeased upon prepayment by substituting Treasury securities for the mortgage in an amount such that the cash flow from the Treasuries matches that of the prepaid loan. IOs from defeased transactions present a more stable yield profile in rising rate/high prepayment environments, but less upside in low prepayment scenarios or from penalty windfalls. PAC & L EVERED I O S

The creation of two separate IO strips from one CMBS transaction is a more recent structural nuance designed to accommodate the creation of larger IOs while appealing to alternative IO buyers.3 The rapidly declining interest rates resulting from a recessionary economy in early 2001 resulted in the creation of much larger IOs, since the IO proceeds are directly proportional to the difference in the weighted average coupon (WAC) of the underlying mortgages vs. the WAC of the bonds created. Ten year treasuries rallied over 100 basis points from October of 2000 (5.80%) to March of 2001 (4.80%). If the typical CMBS transaction in the fall of 2001 had a 40 basis point differential in the WAC of the underlying mortgage pool (8.20%) vs. the WAC of the bonds (7.80%), approximately $25 million in IO would have been created on a $1 billion transaction. If 10-year Treasuries rallied just 25 bp after the mortgage pool was set, but prior to pricing the bonds, the mortgage pool WAC would be 65 bp greater than the bond WAC, resulting in approximately $43 million in IO proceeds. This incremental increase in IO proceeds resulted in the development of the PAC IO/Levered IO structure, appealing to an investor base of risk adverse IO buyers drawn to the very stable yield profile of the PAC IO, with the traditional IO investor base buying the higher yielding but less stable Levered IO.

3

Some CMBS transactions in the early & mid-1990s had multiple IO classes, with individual IOs stripped off of individual bond classes.

Please see additional important disclosures at the end of this report.

81

Transforming Real Estate Finance chapter 6

CMBS IOs The traditional single class IO was stripped off of all (or most all) bond classes in the CMBS structure. The PAC IO is only stripped off of mezzanine bond classes for a finite time; subsequently a Levered IO is stripped off of the more senior and subordinate classes (See Exhibit 5). Recall, given the underlying call protection of today’s commercial mortgages, the performance of any IO is primarily a function of credit. If a mortgage pool is comprised of 100% defeased collateral, the only unscheduled payment of principal is via default and recovery. Because recoveries paydown the A1 class first, and losses are applied to the most subordinate bond class, the yield profile on the PAC IO is insulated from defaults in the underlying mortgage pool. Subsequently, the yield on the Levered IO has more exposure to defaults in the mortgage pool. A PAC IO is a AAA rated security that can survive a 6 CDR stress scenario before giving up any yield. This stress is three times the standard commercial mortgage default rate per the ELS Study. The Levered IO is also rated AAA. This levered IO can be stressed at a 3 CDR (1.5 times the average default rate per the ELS study). While the Levered IO gives yield in a stressed environment, the investor is still buying a loss adjusted AAA.

82

exhibit 5

SERIES 2001 - TOP 4 CAPITAL STRUCTURE

Please see additional important disclosures at the end of this report.

83

Transforming Real Estate Finance chapter 6

CMBS IOs ANALYZING I OS: A PPLYING A GE-A A DJUSTED D EFAULTS

Seasoning is an important factor to consider when comparing older WAC IOs to more recent levered IOs. Traditional default analysis of CMBS IOs usually ignores the impact of aging collateral. Default rates are often applied to transactions in the same manner, whether the deal is newly issued or has seasoned for a few years. We believe using an age-adjusted default curve that s for collateral seasoning is a more accurate method of analyzing IOs. Since defaults tend to be fairly low within the first couple of years after loan origination, the results of an IO analysis may look quite different, depending on whether the default curve is applied to a transaction beginning in year 1 or in year 4 or 5. Examining IOs with an age-adjusted default curve reveals that there is a pricing discrepancy between recently issued levered IOs and some seasoned WAC IOs. In our analysis, we use the commercial mortgage default curve developed by Snyderman (1991) and later updated by Esaki (2001). The default curve reflects the average default experience of life insurance company mortgages that were originated between 1972 and 2000. In the commercial mortgage default study, the peak years for loan defaults occurred within 3 to 7 years after loan origination. In Esaki’s study, all points on the default curve are expressed as percentages of original loan balances. In order to run scenarios in TreppAnalytics, we converted these default rates into numbers that mimic constant default rates (CDRs). exhibit 6

DEFAULT CURVE BASED ON ESAKI COMMERCIAL MORTGAGE DEFAULT STUDY

Source: Morgan Stanley

84

APPLYING T HE D EFAULT C URVE

Under a base case pricing scenario (no defaults, 100Y, yield to call), we see that there are several 1999 WAC IOs that are currently trading in the low-to-mid 300 bp range over Treasuries. If we perform a default analysis on the WAC IOs, using the default curve in Exhibit 6 and assuming no seasoning of the collateral, the bonds lose about 80100 bp of yield. Although the loans in these CMBS deals have seasoned for a minimum of 3 years, we applied the default curve beginning in year 1.

exhibit 7 AVERAGE RESULTS OF DEFAULT ANALYSIS

Analysis

Base Case Pricing Scenario (Spread to UST)

Pricing after Defaults (Spread to UST)

Change in Spread (bp)

308

220

-89

409

214

-195

308

112

-196

1999 WAC IO- Apply Default Curve, Beginning with Year 1 2002 Levered IO- Apply Default Curve, Beginning with Year 1 1999 WAC IO- Apply Default Curve, Beginning with Year 4 Source: Morgan Stanley, Trepp

exhibit 8

Deal Name

1999 WAC IO ANALYSIS: APPLY DEFAULT CURVE BEGINNING WITH YEAR 1 DEFAULT RATE Base Case

Apply Default

Change

Pricing Scenario

Curve, Beginning

in Spread

Class

CUSIP

(Spread to UST)

with Year 1

(bp)

BSCMS 1999-C1

X

07383FBB3

300

191

-109

CCMSC 1999-2

X

161505DL3

300

218

-82

JPMC 1999-C7

X

617059FC9

270

189

-81

LBCMT 1999-C2

X

501773DN2

300

208

-92

MSC 1999-WF1

X

61745MKW2

300

189

-111

NLFC 1999-1

X

63859CBT9

330

229

-101

NLFC 1999-2

X

63859CDR1

300

251

-49

PNCMA 1999-CM1

S

69348HAA6

375

289

-86

AEC1

74436JFH5

300

213

-87

308

220

-89

PSSF 1999-C2 Total/Simple Avg

Source: Morgan Stanley, Trepp

Please see additional important disclosures at the end of this report.

85

Transforming Real Estate Finance chapter 6

CMBS IOs Base case pricing on select levered IOs, issued in 2002, is in the low-to-mid 400 bp range over Treasuries. If we apply the same default curve to 2002 levered IOs, the bonds experience a yield loss of about 200 bp. 2002 LEVERED IO ANALYSIS: APPLY DEFAULT CURVE BEGINNING WITH YEAR 1 DEFAULT RATE

exhibit 9

Base Case Deal Name

Apply Default

Change in

Pricing Scenario Curve, Beginning

Spread from

Class

CUSIP

(Spread to UST)

with Year 1

BACM 2002-2

XC*

05947UHT8

400

209

-191

BSCMS 2002-PBW1

X1*

07383FMR6

420

208

-212

CSFB 2002-CKN2

AX*

22540VV33

400

220

-180

GECMC 2002-1A

X1*

36158YER6

400

210

-190

GMACC 2002-C2

X1*

361849UW0

400

221

-179

JPMCC 2002-CIB4

Base Case (bp)

X1*

46625MKQ1

400

222

-178

XCL*

52108HKY9

390

207

-183

MSDWC 2002-HQ

X1*

61746WNB2

400

213

-187

MSDWC 2002-TOP7

X1*

61746WPJ3

450

228

-222

IO1*

929766BY8

LBUBS 2002-C2

WBCMT 2002-C2 Total/Simple Avg

425

200

-225

409

214

-195

Source: Morgan Stanley, Trepp

Based on these analyses, the extra 100 bp of spread on the levered IO in the base case pricing scenario appears warranted, since levered IOs lose about 100 bp more of yield when subjected to defaults. Applying an aged default curve to the 1999 transactions, however, leads us to a different conclusion. We adjusted the default curve for seasoning and applied it beginning in year 4 for the 1999 WAC IOs. For each IO, we applied an immediate default rate of 2%, followed by defaults of 2.2%, 2.4%, 3.1%, etc., in each of the following years. On average, the bonds lost about 200 bp of yield. (These same bonds only lost about 89 bp of yield in the traditional default analysis, which did not for collateral seasoning.) The 2002 levered IOs also lost about 200 bp of yield when subjected to defaults (starting in year 1). Based on this analysis, both vintages should actually be trading at similar spreads to Treasuries.

86

1999 WAC IO ANALYSIS: APPLY AGE-ADJUSTED DEFAULT CURVE BEGINNING WITH YEAR 4 DEFAULT RATE

exhibit 10

Deal Name

Base Case

Apply Default

Change in

Pricing Scenario

Curve, Beginning

Spread from

Class

CUSIP

(Spread to UST)

with Year 4

Base Case (bp)

BSCMS 1999-C1

X

07383FBB3

300

95

-205

CCMSC 1999-2

X

161505DL3

300

115

-185

JPMC 1999-C7

X

617059FC9

270

74

-196

LBCMT 1999-C2

X

501773DN2

300

100

-200

MSC 1999-WF1

X

61745MKW2

300

66

-234

NLFC 1999-1

X

63859CBT9

330

118

-212

NLFC 1999-2

X

63859CDR1

300

167

-133

PNCMA 1999-CM1

S

69348HAA6

375

167

-208

AEC1

74436JFH5

PSSF 1999-C2 Total/Simple Avg

300

105

-195

308

112

-196

Source: Morgan Stanley, Trepp

OTHER F ACTORS T O C ONSIDER

From the analysis presented above, levered IOs appear to offer value relative to older WAC IOs. However, we have examined age-adjusted defaults without consideration for other factors, which could help explain why the older WAC IOs trade at tighter spreads. While we have taken into the effects of seasoning on expected default rates, we have not considered possible differences between the vintages, such as the quality of the underlying collateral, call protection, underwriting standards and investor comfort with real estate valuations/fundamentals in particular years.

Please see additional important disclosures at the end of this report.

87

Transforming Real Estate Finance chapter 6

CMBS IOs PAC IO Primer: E v. J Although PAC IO spreads are currently quoted to the J-curve, traders are increasingly examining these bonds to the E-curve. In this piece, we discuss the characteristics of “J” and “E”, and why the E-curve makes sense for PAC IO pricing. WHAT I S J 1 , A ND W HY I SN’T I T I DEAL F OR P AC I O S ?

The J-spread is defined as the difference between a bond’s yield and the interpolated Treasury rate that corresponds to the bond’s average life. By definition, an interest only strip does not have any principal cash flows. Therefore, the average life of a PAC IO is actually the average life of the underlying securities from which the IO is stripped. Although the average life of a bond that pays principal is a fair proxy of where the bond’s average interest rate risk lies, the average life of a PAC IO does not describe the bond’s average interest rate risk. As a result, the average life of the PAC IO can vary considerably from its duration. For example, a PAC IO may have an average life of four years, while its duration may be closer to two years. Since the duration of a PAC IO is often much shorter than its average life, it is somewhat arbitrary to compare a PAC IO yield to its average life point on the Treasury curve, which has considerably more interest rate risk. The current steepness of the front end of the Treasury curve is another reason why the J-curve is not an ideal pricing benchmark for PAC IOs. Currently, there is a 137 bp slope between the 2-year and 5-year Treasury rates. Since the curve is this steep, small average life variability between PAC IOs could translate into large spread differences to the J-curve. Therefore, relative value between two PAC IOs may not be readily apparent, since J-spreads can vary significantly with average lives. For example, in a recent bid list, LBUBS 2003-C7 X (avg life 4.90 yrs) traded near T+25 bp, while LBUBS 2003-C5 X (avg life 4.63 yrs) traded near T+38 bp. Although the LBUBS 2003-C5 PAC IO traded 13 bp wider than the LBUBS 2003-C7 PAC IO, does this mean that the 2003-C5 PAC IO offers significantly better value? Pricing these bonds to the E-curve will answer this question. WHAT I S E , A ND W HY I T M AKES S ENSE

The front end of the swap curve is constructed with eurodollar futures contracts, which are future contracts on 3-month LIBOR. When a PAC IO is priced with the E-curve, each bond cash flow is discounted by the E-spread plus the rate on the eurodollar futures curve that corresponds to the timing of that cash flow. Unlike the J-spread, which is a spread over the bond’s average life point on the Treasury curve, the E-spread is a single constant spread that is added to each relevant point on the Eurodollar futures curve. This method of pricing the PAC IO is more appropriate than discounting all cash flows with a single rate at an average life point that is somewhat arbitrary when looking at PAC IOs.

1

88

For more details on the J-curve as a pricing benchmark in CMBS, see the Appendix.

When using the E-curve, even if one PAC IO is slightly longer than another, the E-spreads are still comparable to assess relative value. This is because the discount rates for each cash flow on each bond are benchmarked off of the same rates on the eurodollar futures curve. Using the J-curve, however, two bonds with different average lives would be discounted with rates based on different interpolated Treasury rates. The J-spreads for each bond, as a result, are difficult to compare. If we revisit our example of LBUBS 2003-C7 X and LBUBS 2003-C5 X, we find that these bonds actually look very similar when compared to the E-curve. The LBUBS 2003-C7 PAC IO prices at E+57 bp, while the LBUBS 2003-C5 PAC IO prices at E+55 bp. This example shows that LBUBS 2003-C5 X, which had an extra 13 bp in J-spread over LBUBS 2003-C7 X, does not appear to offer better value using the E-spread methodology.

exhibit 11

Bond LBUBS 2003-C7 X LBUBS 2003-C5 X

PAC IOS: E-SPREAD VS J-SPREAD

Avg Life (yr) 4.90 4.63

J-Spread (bp) 25 38

E-Spread (bp) 57 55

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

89

Transforming Real Estate Finance chapter 6

CMBS IOs Appendix: More Details on J Until 2001, secondary desks quoted spreads off the interpolated Treasury curve, otherwise known as the I-curve to Bloomberg s. Since then, they moved to the convention of quoting CMBS spreads to the interpolated nominal Treasury curve, or the J-curve. This change is primarily relevant for investors who price bonds based on the Treasury curve. Any spread difference between the curves results from assumptions used while interpolating points on the Treasury curve. Pricing a 9.5-yr CMBS bond as a spread to Treasuries involves interpolating yields for the on-the-run five- and ten-year Treasury notes. The I-curve interpolates between the remaining maturity of the two notes. If both notes were issued three months ago, the interpolation would involve 4.75 years as the starting point and 9.75 years as the end point. In contrast, the J-curve ignores any seasoning that may have taken place in either issue. Therefore, no matter how long ago the on-the-run notes were issued, the interpolation is based on the original maturities of five and ten years. WHAT P ROMPTED T HE C HANGE?

Although this curve discrepancy has always existed, the issue for CMBS only surfaced in 2001. This is because of two technical factors present in the Treasury market at the time: longer intervals between new issues and a steep yield curve. • Longer intervals – Given the budget surplus, the Treasury Department slowed the issuance of five-year and ten-year notes from monthly to quarterly and also increased the frequency of reopenings. As a result, on-the-run notes could be up to six months shorter than their original maturity, compared to one month in earlier periods. • Steep yield curve – Ten-year notes in 2001 yielded about 50 bp more than the five-year notes, about 20 bp more than the historical average since 1980.

90

For example, consider pricing MSC 1999-LIFE A2, an 8.0-yr CMBS tranche, on 5/8/01 at LIBOR+45 bp:

exhibit 12

I VS. J SPREAD COMPARISON I-Curve

J-Curve

Maturity (yrs)

Yield (%)

Maturity (yrs)

Yield (%)

Start point (5-yr UST)

4.52

4.74%

5.00

4.74%

End point (10-yr UST)

9.77

5.24%

10.00

5.24%

Interpolated (8-yr UST)

8.00

5.08%

8.00

5.05%

CMBS

Spread

Spread

Spread

Spread

to LIBOR

to UST

to LIBOR

to UST

45

122

45

125

Source: Morgan Stanley, Bloomberg

Yield = Swap Rate + LIBOR Spread UST Spread = Yield – Interpolated UST The yield of the CMBS is set by the spread to its benchmark, the swap rate of 5.85%. Accordingly, the CMBS tranche would yield 6.30%, or 5.85% plus 45 bp. The I-curve results in an interpolated Treasury yield of 5.08% or CMBS spread at UST + 122 bp. The J-curve, however, results in an interpolated Treasury yield of 5.05% or CMBS spread at UST + 125 bp, a 3 bp disparity to the I-curve.

Please see additional important disclosures at the end of this report.

91

This Page Intentionally Left Blank

92

Chapter 7

Transforming Real Estate Finance

Fixed Rate Large Loans

Please see additional important disclosures at the end of this report.

93

Transforming Real Estate Finance chapter 7

Fixed Rate Large Loans INTRODUCTION

Since the terrorist attacks of September 11, 2001, investors have been apprehensive of the concentration risks associated with trophy properties. As a result, fixed rate large loan deals and single asset transactions have virtually disappeared from the new issue market place. While short-term large loan deals continue to be issued today, the new vehicle for securitizing fixed rate large loans has become the “fusion” transaction. FUSION

Loans with large dollar balances are often mixed with smaller loans in deals that the market labels as “fusion.” The definition of “fusion” varies, but for purposes of classification, Commercial Mortgage Alert defines a fusion deal as a transaction that has conduit style loans and either has one loan that is more than 10% of the pool balance or all loans of $50 million or more are at least 15% of the deal. Many market participants classify a fusion transaction by examining the top 10 loan concentration within a deal. The rule of thumb is to classify the deal as a fusion transaction if the top 10 loans for at least 40% of the collateral. STAND-A A LONE L ARGE L OANS

“Stand-alone large loans” refer to mortgages of $50 million or more on commercial properties with an institutional borrower. Stand-alone large loan CMBS have taken the form of single-asset or single-borrower deals or transactions backed by a small number of commercial mortgages averaging $50 million or more. For single asset transactions, the loan size may be as much as $500 million. Stand-alone large loans tend to have lower leverage and more credit-worthy borrowers than conduit loans. Since the September 11th attacks, AAA securities from fixed rate large loan transactions trade about 15 bp to 30 bp wider than AAA tranches from diversified conduit CMBS. At the lower investment grade level, large loan classes are also trading at wider spreads to similarly rated classes from conduit deals. HISTORY O F T HE F IXED R ATE L ARGE L OAN M ARKET

The CMBS market began in the 1980s with the securitization of commercial mortgages on large “trophy” assets. The market evolved in the 1990s to one that was dominated by large pools of small- to medium-sized loans. A typical pool might consist of 200 loans on properties well diversified by geographic region and property type. Mortgage conduits originate the loans and most range in size from $1 million to $20 million.

94

exhibit 1

U.S. CMBS ISSUANCE

1

Through November 1, 2004.

Source: Morgan Stanley, Commercial Mortgage Alert

As issuers sought to get loans off their books more quickly, deal size decreased in line with loan accumulation periods. Smaller deal size meant that a large loan made up a greater percentage of the transaction. In 1998, several CMBS transactions exceeded $2 billion. For transactions that large, a $50 million loan is only 2.5% of the total balance and the deal does not receive a high concentration penalty from the rating agencies. As deal sizes dropped below $1 billion, issuers received better prices by issuing separate large loan deals rather than tainting an entire pool with concentration risk. exhibit 2

AVERAGE U.S. DEAL SIZE

1200

1000

800

600

400

200

0

1

As of November 1, 2004 pricing.

Does not include Agency CMBS or resecuritizations. Source: Commercial Mortgage Alert

Please see additional important disclosures at the end of this report.

95

Transforming Real Estate Finance chapter 7

Fixed Rate Large Loans CHARACTERISTICS O F S TAND-A A LONE L ARGE L OANS Credit

To date, the average stand-alone large loan has had a lower LTV and higher DSCR than the average conduit loan. Although LTVs for stand-alone large loans may range from 40% to 100%, most fall in the range of 45% to 60%. Conduit transactions typically have weighted average LTVs in the 60% to 75% range. DSCRs for stand-alone large loans are frequently in excess of 1.50x, while for conduits, the weighted average DSCR is generally in the 1.25x to 1.40x range. Some analysts believe that the reason for the lower leverage and higher debt service coverage ratios for stand-alone large loans is that prices of large properties are more volatile than for smaller properties. There is very little confirmation of this volatility, aside from anecdotal evidence about the fall in prices of “trophy properties” in the early 1990s. That higher priced properties have more volatile prices makes some intuitive sense, because there is a much smaller base of buyers than for average or low-priced properties. In the residential home market, for example, there is much evidence showing that high value homes both rise and fall in price much faster than average value homes. Stand-alone large loans tend to be on properties located in major markets, in contrast to smaller conduit properties, which are more evenly spread out in both larger and smaller metropolitan areas. Location in larger markets has both advantages and disadvantages. On the positive side, property markets in large markets are probably more liquid and have more potential buyers than those in small markets. On the other hand, the concentration of properties in large urban centers, for example, can create volatility in prices should the market suffer a downturn. Structural Features

Large commercial loans generally have structural protections that are often not a feature of conduit loans. First, stand-alone large loans have “lock box” features that separate the cash flow of the property from the borrower’s income. Second, the loans are usually placed in a “special purpose entity,” or SPE, which separates the loan from the other assets of the borrower in the case of bankruptcy. (In a conduit CMBS, the SPE may be less well defined at the borrower level.) Third, a stand-alone large loan often provides for removal of management, or “kick-out,” should the cash flow of the property deteriorate beyond a specified target level. These features, while important in insulating investors against potential borrower problems, are not a protection against a general deterioration in real estate credit. For example, if a fall in operating income is caused by a regional economic downturn, the replacement of management probably will not improve the credit risk of the property. Credit Strength of Borrower

Offsetting the risks of potentially more volatile asset prices is the better credit of stand-alone large loan borrowers. These borrowers are often rated entities, compared to most conduit lenders, who are not rated. A bankruptcy of a borrower leading to default on a mortgage is thus more likely in the case of a conduit borrower than a stand-alone large loan borrower.

96

Many of the borrowers in stand-alone large loan transactions are well-capitalized firms that have access to other sources of capital. This provides some protection against default in a real estate downturn. Given the lower LTVs on most standalone large loans, the borrower’s equity stake is typically larger than for conduits and may in some cases reach over $100 million. According to a Moody’s study of corporate defaults between 1970 and 1997, less than 5% of investment grade corporations defaulted within 10 years. The default rate is 20% for companies in the Ba category, and almost 50% for firms rated in the single-B category. It should be noted, however, that the default of a borrower does not necessarily mean a default or loss on a mortgage made by the borrower. The CMBS is secured by the property, not the credit of the borrower. As long as the value of the property is greater than the mortgage, or the cash flow greater than the debt payments, the borrower is unlikely to default on the mortgage. Prepayment Protection

Stand-alone large loans and conduit loans have very similar prepayment protection. Most currently have defeasance provisions. Earlier transactions had lockout or yield maintenance periods followed by penalty points. For more details on types of call protections, see the “Call Protection” chapter in this book. Information Availability

In 1999, it was often difficult to obtain individual property cash flow and DSCR data after issuance. The situation has improved since 1999 and stand-alone transactions are now closely monitored after closing. Depending on the of the borrower’s loan agreement, the following documents are available to investors on a quarterly and/or annual basis from the servicer: property operating statements, rent rolls, sales reports for retail properties and property inspections. EMPIRICAL S TUDIES

Two recent studies addressed the relative risk of large loans. A default study by Esaki, L’Heureux, and Snyderman (1999)1 found that default rates for small loans at insurance companies over the period 1973 to 1997 were slightly less than for large loans. The study, however, only had four categories of loans, of which the largest size was “over $8 million.” Another study published in April 1999 by Ziering and McIntosh2 examined the risk-return profiles of properties by size. The study also looked at properties in four categories, but the categories ranged from “less than $20 million” to “over $100 million.” Over the period 1981 through 1998, the authors found that larger properties generally had higher average returns, but also higher volatility of returns. The authors cite the thin market for trophy properties as the reason for higher volatility.

1

Howard Esaki, Stephen L’Heureux, and Mark Snyderman, “Commercial Mortgage Defaults: An Update,” Real Estate Finance, Spring 1999.

2

Barry Ziering and Willard McIntosh, “Property Size and Risk: Why Bigger Is Not Always Better,” Prudential Real Estate Investors Research, April 1999.

Please see additional important disclosures at the end of this report.

97

Transforming Real Estate Finance chapter 7

Fixed Rate Large Loans RATING A GENCY A PPROACH T O L ARGE L OANS

In rating single-borrower or single-asset transactions, the largest concern of rating agencies is concentration risk. For example, one agency stated, “The distinguishing characteristic of [single-asset] transactions is the concentration of risk…This is in contrast to pooled transactions, in which the diversification of properties and leases mitigates this risk.”3 To adjust for concentration risk, rating agencies have very conservative standards for single-borrower and single-asset transactions. For example, to obtain a AAA rating, a large office property must have a debt service coverage ratio in the range of 2.30x to 2.75x and an LTV not exceeding 40%. At the rating agencies, the advantage of diversified small to medium loan pools is demonstrated in the level of credit needed to obtain a higher rating level than the collateral on a stand-alone basis. For example, consider a pool of small multifamily properties, all with DSCRs of 1.35x and LTVs of 70%. Such a pool would need 20% to 25% of subordination for the AAA class. A large loan with equivalent characteristics would be shadow-rated BB. In order for a senior portion of this loan to attain a rating of AAA, we estimate that a subordination level of 35% is required. This subordination level would lower the effective LTV of the senior tranche of the large loan to 46%, which is the standard for a stand-alone AAA. The 10% to 15% difference in subordination levels reflects the “penalty” for concentration. In underwriting conduit transactions, rating agencies typically underwrite a sample of loans in the pool. The sample usually is about 30%, and includes most of the largest loans. In contrast, a large loan or single-asset transaction will have 100% underwriting by the agencies. In stand-alone large loan deals, full environmental and engineering reports are provided for each property. This is not the case for all conduit loans. It is not clear if the rating agencies adjust conduit subordination levels upward to adjust for the uncertainty of less than 100% underwriting. Based on their underwriting, rating agencies will adjust the cash flow and valuation on a property, deriving “adjusted” LTVs and DSCRs. In almost all cases, the LTV is adjusted upward (and NOI downward) from the reported number. Based on our conversations with the rating agencies, these adjustments are usually in the range of 10% to 20% for both stand-alone large loans and conduit loans. DO R ATINGS C ORRECTLY A DJUST F OR R ISK?

Rating agencies universally state that a AAA rating means the same amount of credit risk across all types of fixed-income securities. Theoretically, a AAA-rated corporate bond should have the same degree of credit risk as a AAA-rated CMBS. Within the CMBS sector, rating agencies state that they make adjustments for collateral quality so that a AAA conduit transaction has the same amount of risk as a AAA stand-alone large loan deal.

3

98

The Rating of Commercial Mortgage Backed Securities,” Duff and Phelps, November 1994.

In practice, we believe that there can be substantial differences in risk among securities with the same rating. We have long maintained that rating agencies have more conservative standards in rating structured securities than corporate bonds. INVESTOR P REFERENCES

Some CMBS investors prefer diversified pools of loans because they feel they do not have the real estate expertise to evaluate commercial mortgage credit risk on a single property. Some of these investors rely on the rating agencies to analyze default risk; others become comfortable with the general underwriting guidelines of an originator. Investors preferring pool diversity tend to be money managers with limited experience in underwriting real estate. They tend to view CMBS as a fixed-income asset with good call protection, rather than a direct investment in real estate. Other investors prefer single-asset transactions, or those backed by only a few loans, because they can then use their real estate expertise to underwrite each loan in a CMBS pool. These investors tend to be insurance companies with large underwriting staffs. These investors focus more on the real estate aspects of CMBS investments and often purchase stand-alone large loan CMBS as a compliment to their whole loan portfolio. Because of this split in investor preferences, stand-alone large loan transactions tend to trade differently than diversified conduit pools. Money managers tend to shy away from single asset deals because of the lack of diversification and a perceived lack in liquidity. Insurance companies re-underwrite a stand-alone large loan, and if they are comfortable with the asset, prefer to buy the lower investment grade classes. The yield on the BBB class, for example, most closely matches the yield on whole loans, the alternative investment for insurers. Insurance companies purchase about 85% of the mezzanine classes from standalone large loan transactions and only 45% from conduit deals. In contrast, money managers buy 50% of the AAA bonds from conduit deals, but only about a third of the AAA-rated securities from stand-alone large loan transactions. As a consequence of these preferences, the AAA class from a stand-alone large loan transaction tends to trade at a wider spread to Treasuries than a conduit deal. The difference has historically been in the range of 5 bp to 10 bp. The BBB class of a stand-alone large loan deal, however, trades at a narrower spread difference to conduit BBBs and, in some instances, actually trades at a tighter spread than in a diversified pool. If insurers deem an asset as high quality, the small size of the BBB class often results in excess demand at the initial pricing level.

Please see additional important disclosures at the end of this report.

99

This Page Intentionally Left Blank

100

Chapter 8

Transforming Real Estate Finance

BBB CMBS and REITs

Please see additional important disclosures at the end of this report.

101

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs BBB CMBS and REIT unsecured corporate bonds are often used as proxies for one another, as both are backed by real estate and have similar ratings. In this chapter, we examine the volatility, ratings history, liquidity, structural differences, the issuer’s perspective, the rating agencies’ perspective and performance of each. We found that while differences exist among the security types, there is no apparent advantage of one over the other. Since the credit is similar and spreads have historically converged, we recommend that BBB CMBS investors consider REIT debt as an alternative, particularly when spreads are wider on REIT debt.

exhibit 1

Rating Trading Range Since April 1998 Spread Volatility in Standard Deviation (bp) Rating Actions in 2004 Market Capitalization Bid Ask Spread Monthly Trading Volume Leverage Structural Differences Management

CHARACTERISTICS OF BBB CMBS AND REIT UNSECURED DEBT CMBS BBB+/BBB/BBB114-325 bp

REITs BBB+/BBB/BBB97-321 bp

22 35 upgrades to 25 downgrades $28 billion 2 bp for on-the-run names $750 million to $1 billion 70% LTV at BBB level Secured by specific properties Inactive management within the trust

36 1 upgrade and 7 downgrades $58 billion 2-3 bp for large cap names $1 billion 40% debt to market cap Corporate debt covenants Active management of portfolio

Source: Intex, Remittance Reports

We also advise investors who have focused on REIT debt to consider BBB CMBS as an alternative, particularly when CMBS spreads are wider than REIT debt. SIMILAR V OLATILITY

Historically, BBB CMBS and REIT paper have traded close to parity, with divergence resulting in very short-term opportunities. Since the end of 1998, the average spread difference between both has been only one bp. Since we began tracking the spreads in early 1998, the longest period that the spread differential between BBB CMBS and REITs remained wider than 5 bp was between August 2003 and May 2004. The average difference in spread during that period was 17 bp. The standard deviation of historical BBB CMBS spreads to Treasuries over the past several years is 22 bp; it is 36 bp for REITs. BBB CMBS stability is similar to that of the BBB Industrial Index and is less volatile than REIT debt (Exhibits 2 and 3).

102

exhibit 2

BBB CMBS VS. REITS TO UST (APRIL 1998 NOVEMBER 20041)

1

Through November 9, 2004.

Source: Morgan Stanley

exhibit 3

Mean (bp) Standard Deviation (bp)

SPREAD VOLATILITY: APRIL 1998-NOVEMBER 2004

BBB CMBS 116 22

REITS 133 36

BBB Industrials 104 22

Source: Morgan Stanley, Bloomberg

RATINGS H ISTORY Rating Actions Best for BBB CMBS

Through September 2004, the upgrade/downgrade ratio for CMBS rated BBB+/BBB/BBB- has been better than that of REIT paper. The rating agencies have changed their ratings on 372 BBB+/BBB/BBB- CMBS tranches since January, with 311 upgrades and 61 downgrades. The 2004 ratio of 5.1 upgrades to 1 downgrade is a better record than the 3.3 to 1 ratio for all credit classes within CMBS for the same period. Appendix A at the end of this report details all rating actions for CMBS rated BBB+/BBB/BBB- in 2004.

Please see additional important disclosures at the end of this report.

103

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs exhibit 4

2004 YTD RATING CHANGES1 FOR REITS

Company Brandywine Realty Trust Crescent Real Estate Equities Equity Office Properties Glimcher Realty Trust iStar Financial, Inc National Health Investors Omega Healthcare Prime Property Funding II, Inc. Sun Communities United Dominion Realty Trust

From [P]Ba3 Ba3 Baa1 Ba3 Ba1 B1 B2 A2 Baa3 Baa3

To [P]Ba2 B1 Baa2 B1 Baa3 Ba3 B1 A3 Ba1 Baa2

1

Based on Moody’s data only through October. Source: Moody’s

Throughout the first 10 months of 2004, REITs experienced a 1 to 1 upgrade/downgrade ratio. Both CMBS and REITs have had a better ratings record in 2004 than corporate bonds which experienced a 0.9 to 1 upgrade/ downgrade ratio. exhibit 5

2004 RATING ACTIONS

BBB+/BBB/BBB- CMBS

REITs1

311 61

5 5

2004 Upgrade Downgrade 1

Based on Moody’s data only. Source: Moody’s

LIQUIDITY Liquidity Fairly Even for REITs & BBB CMBS

Market capitalization, trading volume, deal size and bid-ask spread all contribute to liquidity within a sector. REIT paper has a market capitalization of about $58 billion. The market capitalization of CMBS rated BBB is less than half that of REIT paper, with about $28 billion currently outstanding. On average, BBB bonds for 4% of U.S. issuance volume. We anticipate that $3 billion of BBB CMBS was issued in 2004.

104

Several factors reduce liquidity for both CMBS and REIT debt. For instance, the class sizes of BBB CMBS tranches have been shrinking over time, reducing liquidity. Additionally, structured vehicles (CDOs) have been big buyers of BBB CMBS and REIT paper. Recently insurance companies have been significant buyers of BBB CMBS. Both CDOs and insurance companies tend to buy and hold rather than trade for total return, reducing the amount of paper traded. We anticipate that the street trades about $750 million to $1 billion of BBB CMBS paper on a monthly basis. The inclusion of BBB CMBS as ERISA eligible securities contributed slightly to liquidity. Liquidity in a particular name may vary depending on the seasoning and delinquencies within a transaction. Large blocks of on-the-run BBB CMBS with few or no delinquencies trade with bid-ask spread of 2 bp or less. Although the market capitalization is greater for REIT paper, REIT trading volume is similar to CMBS. On average, the street trades about $1 billion of REIT paper per month, with heavier trading over the last several months. The most liquid names are those with the most debt outstanding. The more-liquid REIT names (EOP, SPG) trade with a 2-3 bp bid-ask spread in on-the-run maturities. The remaining names in the universe are less liquid and typically trade with a 10 bp bid-ask spread. CDO Market Effect

The CDO market has had a significant impact on BBB CMBS and REIT debt. On the positive side, the CDO bid keeps spreads in check. Liquidity, however, may be limited by the buy-and-hold nature of CDOs. About 7% of REIT unsecured debt has been put into various CDO deals, effectively removing about $3.8 billion from REIT unsecured debt secondary trading. Nevertheless, we do not consider the 7% of outstanding REIT debt locked up in CDOs to be significant, despite the recent focus of CDOs on this sector. At the peak of the CDO bid in the high yield market, many market participants noted 25-30% of high yield new issue flows into CDOs. In the credit derivatives markets today, flows from synthetic CDOs for a similar percentage of that market, as well. The amount of BBB CMBS within CDOs is more significant than for REIT debt. We estimate that about $5.5 billion BBB CMBS, which is about 18% of the market, is contained in real estate CDOs. STRUCTURAL D IFFERENCES

REIT paper and CMBS have very similar credit characteristics in that both are backed by payments generated by commercial real estate. However, significant differences arise in the structures of both credits. Both CMBS and REIT paper benefit from the inherent diversity within a portfolio of properties. However, diversification within CMBS conduit transactions is greater than within REIT paper. Conduit CMBS transactions are made up of 100-200 smaller commercial loans providing diversity in of geography, borrowers and property types. Typically, REIT management teams have expert-

Please see additional important disclosures at the end of this report.

105

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs ise in one property type that may be concentrated in several regions rather than diversified across the country. However, the balance sheets and credit characteristics of REITs are usually significantly stronger than the credit of borrowers within CMBS conduit transactions. In CMBS, a default on one asset within a conduit has little or no impact on pricing and liquidity. There has never been a default on REIT unsecured paper. THE I SSUER’S P ERSPECTIVE

Although REIT debt often trades close to parity with BBB CMBS, REITs as issuers of either secured or unsecured financing take into the overall cost of capital for the transactions, as well as the flexibility of the debt relative to their portfolios. Most REITs choose to issue CMBS in the form of large loan transactions. Currently, approximately $12 billion of CMBS outstanding has been issued by REITs (Appendix B). Typically, these transactions contain about 65% AAA bonds, 11% AA bonds, 10% A bonds, 10% BBB bonds and 4% BBB- bonds. While the actual cost of funding is lower for a large loan REIT CMBS transaction than for unsecured REIT debt, issuers give up some financial flexibility because securitization requires mandatory capex reserves and limits refinancing and the sale or substitution of the assets. Large loan CMBS may have specific covenants that require minimum debt service coverage ratios, lock boxes for cash flows and specific insurance coverage. Lastly, issuing CMBS requires a rather long lead-time for marketing and sale of the transaction, while unsecured debt deals may be completed virtually overnight. Documentation is much simpler and execution is much quicker in the unsecured debt market. RATING A GENCY P ERSPECTIVES Unsecured REIT Debt

Rating agencies consider a number of factors when it comes to rating unsecured REIT debt. These factors include depth of management and experience, quality of property, geographic and tenant diversification, stability of cash flows, profitability and financial flexibility. As Moody’s stated in its September 2000 Special Comment report, The Qualitative Factors - It’s Not All in the Numbers, “Financial ratios, although not the dominant element in Moody’s rating process, are a critical part of our rating methodology.” As none of the rating agencies will disclose average ratios per category, we reviewed all the companies we cover and attempted to break out the ratios for each rating class. With the many outliers, we found it difficult to discern specific ratios, as this varies for different sectors, with retail clearly exhibiting the most aggressive financial fundamentals. Generically, debt to gross real estate is in the low 40% range for high BBBs, in the mid-to-high 40% range for mid-BBBs and in the low-to-mid 50% range for low-BBBs. Typically high BBBs have a fixed charge coverage of over 2.5x, with mid BBBs and low BBBs slightly lower than that. Again, there are many outliers; our review serves to demonstrate that looking at the financial fundamentals alone is only one piece of the rating process. Higher ratings equate to greater availability on bank lines. Many high BBBs were not

106

reliant on their bank lines, while many low BBBs had drawn down significantly on their bank lines. Please see Appendix C for a breakdown of the rating agencies’ analyses of REITs. CMBS Debt

Before each CMBS is issued, the rating agencies review the collateral in the transaction and determine the tranche ratings and pool sizing. The rating agencies apply published standards to loans pooled into a CMBS and adjust the result by making qualitative assessments. Almost all CMBS carry at least two ratings, and many have three. Different rating agencies assign different levels of credit to obtain a given rating level. During the process, the agencies review the property level cash flows, perform physical inspections and run various stress analyses on the underlying cash flows in order to simulate a worst-case recession scenario. When a conduit deal comes to market, the rating agency performs due diligence on a subset of the properties (typically between 35% and 75%). The larger loans in the deal are always underwritten, while the remaining properties are chosen such that they provide a representative cross-section of the deal. To determine credit enhancement levels, the underwritten cash flow (UCF) produced by each property is then assigned a “haircut” based on various subjective parameters. Exhibit 6 details a list of parameters that rating agencies consider when evaluating a CMBS conduit. The parameters are similar to those considered for a non-conduit deal except for adjustments for loan diversification. exhibit 6

CMBS EVALUATION PARAMETERS

Loan Specific Property type Loan-to-value ratio Debt service coverage ratio Fixed/floating rate Loan seasoning Direct versus correspondent lending

Deal Specific Number of loans in the deal Loan size Degree of subordination Balloon extension risk Quality of master servicer and special servicer Real estate outlook

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

107

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs exhibit 7

Positives Active Management Better Disclosure Covenants assets Lower Leverage

ATTRIBUTES OF REITS RELATIVE TO CMBS

Negatives May be more geographically concentrated Must pay out 90% of earnings subject to refinancing risk Backed by corporate guarantee, as opposed to explicitly secured by Corporate entities subject to event risk

Source: Morgan Stanley

Loans collateralized by different property types are generally ranked in the following order (best to worst): regional malls, multifamily, anchored community retail, industrial, office and hotel. These rankings are based on historical defaults and cash flow volatility of the different property types. Rating agencies review the dispersion of DSC and LTV in the entire deal; that is, having all loans with a DSC of 1.5x is better than having 50% of the loans at 1.0x and the remainder at 2.0x. Given the volatility of short-term interest rates, an adjustable-rate loan is underwritten using an interest rate scenario that is substantially higher than current rates. Loans without a track record of consistent payments are also rated more conservatively than those seasoned at least five years. Lower Leverage and Better Borrower Credit in REITs

Typically, conduit CMBS transactions have a 75% loan to value (LTV) ratio, and BBB bonds have subordination levels averaging about 5%, translating to a 70% LTV for the BBB classes. Large loan transactions have lower leverage, with LTV ratios ranging between 55-65%. Usually the BBB tranches within these transactions are the lowest-rated classes with minimal subordination. REITs have debt to market capitalization ratios that average 41% and debt to gross real estate ratios of about 49%, below the leverage within CMBS transactions. Most REIT debt covenants require maintenance of a 60% debt to assets ratio based on the book value without depreciation. Additionally, typically no more than 40% of a REIT’s assets can be encumbered by secured financing. REITs have several avenues for financing, including lines of credit, unsecured corporate debt, secured financing, selling properties and t ventures. Most borrowers within conduit transactions only have access to capital via secured financing.

108

PERFORMANCE 2004 CMBS Performance

Through October, CMBS has outperformed both corporate bonds and treasuries. Within CMBS, BBBs have produced the best performance. According to the MSCI CMBS Index, investment grade CMBS produced a 4.24% total return through October, and BBBs produced a total return of 6.02% during the same period. Based on October data, delinquencies on deals with more than one year of seasoning also remain low at 1.79% of current balances. When we examined cumulative loss data through September 2004, we find the 1995 vintage had the highest loss rate (1.05%), well below the 5% average credit enhancement on BBB CMBS. Since new issue CMBS have about 5% subordination to the BBB class, there is significant cushion for current levels of delinquencies.

Please see additional important disclosures at the end of this report.

109

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix A

2004 RATING ACTIONS THROUGH SEPTEMBER 30, 2004 Fitch

CMBS Transactions 1211 Avenue of the Americas Trust, 2000-C7 1211 Avenue of the Americas Trust, 2000-C7 1251 Avenue of the Americas Trust, 1999-XL 1251 277 Park Avenue Finance Corp, 1997-C1 Aetna Commercial Mortgage Trust, 1997-ALIC American Southwest Financial Securities Corp, 1995-C1 AMRESCO Commercial Mortgage Funding I Corp, 1997-C1 AMRESCO Commercial Mortgage Funding I Corp, 1997-C1 Annapolis Mall Mortgage Trust 2000-C1B Artesia Mortgage CMBS, 1998-C1 Asset Securitization Corp, 1995-A5 Asset Securitization Corp, 1995-MD4 Asset Securitization Corp, 1995-Md4 Asset Securitization Corp, 1996-D3 Asset Securitization Corp, 1996-MD6 Asset Securitization Corp, 1997-D4 Asset Securitization Corp, 1997-D4 Asset Securitization Corp, 1997-D4 Asset Securitization Corp, 1997-D5 Asset Securitization Corp, 1997-D5 Asset Securitization Corp, 1997-D5 Banc of America Large Loan, 2001-7WTC Banc of America Large Loan, 2001-7WTC Banc of America Large Loan, 2001-7WTC Banc of America Large Loan, 2001-7WTC Banc of America Commercial Mortgage Inc., Series 2000-2 Banc of America Commercial Mortgage Inc., Series 2000-2 Banc of America Commercial Mortgage Inc., Series 2000-2 Banc of America Commercial Mortgage, 2001-PB1 Banc of America Commercial Mortgage, 2001-PB1 Banc of America Commercial Mortgage, 2001-PB1 Banc of America Large Loan, 2002-FLT1 Banc of America Large Loan, 2002-FLT1 Banc of America Large Loan, 2002-FLT2 Banc of America Large Loan, 2002-FLT2 Banc of America Large Loan, 2002-FLT2 Banc of America Structured Notes, 2002-1 Banc One/FCCC Commercial Mortgage Loan Trust, 2000-C1 Banc One/FCCC Commercial Mortgage Loan Trust, 2000-C1 Bear Stearns Commercial Mortgage Securities, 2000-WF1 Bear Stearns Commercial Mortgage Securities, 2000-WF1 Chase Commercial Mortgage Securities Corp, 1996-1 Chase Commercial Mortgage Securities Corp, 1997-1 Chase Commercial Mortgage Securities Corp 1998-1 Chase Commercial Mortgage Securities Corp 1998-1 Chase Commercial Mortgage Securities Corp, 1998-2 Chase Commercial Mortgage Securities Corp, 1998-2 Chase Commercial Mortgage Securities Corp, 2000-2 Chase Commercial Mortgage Securities Corp, 2001-245 Park Avenue Commercial Mortgage Asset Trust, 1999-C2 Commercial Mortgage Asset Trust, 1999-C2 COMM 1999-1 COMM 2000-FL3 COMM 2000-FL3 COMM 2001-FL4 COMM 2001-FL4 COMM 2001-FL5 COMM 2001-FL5 COMM 2001-FL5 COMM 2002-FL6 COMM 2002-FL6 COMM 2002-FL6 COMM 2002-FL6 COMM 2002-FL6 COMM 2002-FL6 COMM 2002-FL7 COMM 2002-FL7 COMM 2002-FL7 COMM 2002-FL7 COMM 2002-FL7 COMM 2002-FL7 COMM 2002-FL7 COMM 2002-FL7 Commercial Capital Access One, Series 2

Source: Fitch, Moody’s and S&P

110

Class

Old Rating

Action

Current Rating

Date of Rating Action

D E E B-2 H B-3 E F B-4 E A-5 A-4 A-5 A-4

BBB BBBBBB BBB+ BBB BBB BBB+ BBB BBB+ BBB BBB BBB+ BBB BBB

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

BBB+ BBB BBB+ A A A A AABBB+ AA AAA AA A-

4/7/2004 4/7/2004 5/26/2004 6/9/2004 3/18/2004 2/3/2004 2/3/2004 2/3/2004 7/29/2004 6/15/2004 8/25/2004 2/10/2004 8/25/2004 3/25/2004

A-6 A-7 A-8

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

A+ A BBB+

3/24/2004 3/24/2004 3/24/2004

G H

BBB+ BBB

Downgrade Downgrade

B+ B

4/8/2004 4/8/2004

A

BBB-

Upgrade

BBB+

1/8/2004

E F

BBB BBB-

Upgrade Upgrade

BBB+ BBB

6/30/2004 6/30/2004

E

BBB+

Upgrade

A+

6/15/2004

D

BBB

Upgrade

BBB+

4/13/2004

E F

BBB BBB-

Upgrade Downgrade

BBB+ BB+

4/13/2004 6/16/2004

E K-SR L-SR

BBB BBB BBB-

Upgrade Downgrade Downgrade

BBB+ B B-

4/27/2004 4/2/2004 4/2/2004

G

BB

Downgrade

B+

9/30/2004

K-JP M-JP

BBB BBB-

Downgrade Downgrade

BBBBB+

6/10/2004 6/10/2004

M-GR

BBB-

Upgrade

BBB

5/27/2004

D

BBB

Upgrade

A-

10/5/2004

Moody’s Class

E

Old Rating

Baa2

Action

Upgrade

S&P Current Rating

A3

Date of Rating Action

Baa3

Upgrade

Aa3

7/20/2004

E F

Baa2 Baa3

Downgrade Downgrade

Ba2 Ba3

2/3/2004 2/3/2004

D E

Baa1 Baa2

Baa2 Baa3

Upgrade Upgrade

Upgrade Upgrade

A1 A2

Aa3 A1

9/7/2004 9/7/2004

Baa1 Baa3

Downgrade Downgrade

Ba1 Ba3

10/20/2004 10/20/2004

K- L-

Baa2 Baa2

Downgrade Downgrade

Baa3 Baa3

3/24/2004 3/24/2004

Baa1 Baa1 Baa1 Baa2 Baa2 Baa3 Baa3 Baa3

Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

Baa3 Baa2 Baa3 Baa3 Ba1 Ba1 Ba1 Ba2

Current Rating

F

Date of Rating Action

BBB

Upgrade

AA+

7/19/2004

A-6

BBB

Upgrade

AA

1/12/2004

A-8Z A_2 A-3

BBBBBB BBB-

Upgrade Upgrade Upgrade

AAAA-

7/9/2004 7/9/2004 7/9/2004

F G H G H J

BBB+ BBB BBBBBB+ BBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

ABBB+ BBB ABBB+ BBB

10/8/2004 10/8/2004 10/8/2004 7/2/2004 7/2/2004 7/2/2004

J-OM K-OM L-OM

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

AA AAA+

3/25/2004 3/25/2004 3/25/2004

E F

BBB+ BBB

Upgrade Upgrade

AAA AAA

9/28/2004 9/28/2004

E E D E D E

BBB+ BBB+ BBB BBBBBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAA A AABBB+

5/14/2004 10/19/2004 10/7/2004 10/7/2004 9/27/2004 9/27/2004

E F

BBB BBB-

Upgrade Upgrade

AAA

G

BBB

Downgrade

B-

9/22/2004

K-WP L-WP M-WP L-DC M-DC

BBB+ BBB BBBBBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade

ABBB+ BBB A A-

6/28/2004 6/28/2004 6/28/2004 6/28/2004 6/28/2004

K-GR L-GR L-NM M-NM M-GR

BBB+ BBB BBB BBBBBB-

Upgrade Upgrade Upgrade Upgrade Upgrade

A ABBB+ BBB BBB+

6/2/2004 6/2/2004 6/2/2004 6/2/2004 6/2/2004

2/10/2004 2/10/2004

K-CH M-CH

K-CS K-RP L-CS L-RP M-CS M-RP N-CS O-CS

Action

5/28/2004

A-2

F G

Class

Old Rating

1/23/2004 1/23/2004 1/23/2004 1/23/2004 1/23/2004 1/23/2004 1/23/2004 1/23/2004

Please see additional important disclosures at the end of this report.

111

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix A

2004 RATING ACTIONS THROUGH SEPTEMBER 30, 2004 (CONTINUED) Fitch

CMBS Transactions

Class

Commercial Mortgage Acceptance Corp, 1996-C1 G Commercial Mortgage Acceptance Corp, 1996-C2 F Commercial Mortgage Acceptance Corp, 1998-C1 Commercial Mortgage Acceptance Corp, 1998-C1 Commercial Mortgage Acceptance Corp, 1998-C2 D Commercial Mortgage Acceptance Corp, 1998-C2 E Commercial Mortgage Acceptance Corp, 1999-C1 Commercial Mortgage Acceptance Corp, 1999-C1 Commercial Mortgage Acceptance Corp, 1999-C1 Commercial Mortgage Acceptance Corp, 1999-C1 E Commercial Mortgage Acceptance Corp, 1999-C1 F Commercial Mortgage Asset Trust, 1999-C1 Commercial Mortgage Asset Trust, 1999-C1 Commercial Mortgage Asset Trust, 1999-C2 E Commercial Mortgage Asset Trust, 1999-C2 F Credit Suisse First Boston Mortgage Securities Corp, 1997-C1 D Credit Suisse First Boston Mortgage Securities Corp, 1997-C2 D Credit Suisse First Boston Mortgage Securities Corp, 1998-C2 D Credit Suisse First Boston Mortgage Securities Corp, 1998-C2 E Credit Suisse First Boston Mortgage Securities Corp, 2001-CF2 E Credit Suisse First Boston Mortgage Securities Corp, 2001-CF2 F Credit Suisse First Boston Mortgage Securities Corp, 2001-CF2 G Credit Suisse First Boston Mortgage Securities Corp, 2001-CK6 Credit Suisse First Boston Mortgage Securities Corp, 2001-CK6 Credit Suisse First Boston Mortgage Securities Corp, 2001-CK6 Credit Suisse First Boston Mortgage Securities Corp, 2001-FL1 Credit Suisse First Boston Mortgage Securities Corp, 2001-FL1 H Credit Suisse First Boston Mortgage Securities Corp, 2001-TFL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-CKP1 Credit Suisse First Boston Mortgage Securities Corp, 2002-CKP1 Credit Suisse First Boston Mortgage Securities Corp, 2002-CKP1 Credit Suisse First Boston Mortgage Securities Corp, 2002-3 E Credit Suisse First Boston Mortgage Securities Corp, 2002-3 F Credit Suisse First Boston Mortgage Securities Corp, 2002-FL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-FL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-FL2 Credit Suisse First Boston Mortgage Securities Corp, 2002-FL2 Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 F-WBC Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 G-WBC Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 Credit Suisse First Boston Mortgage Securities Corp, 2002-TFL1 H-ALH CRIIMI MAE Trust I, 1996-C1 B CRIIMI MAE Trust I, 1996-C1 C CRIIMI MAE CMBS Corp, 1998-1 CRIIMI MAE CMBS Corp, 1998-1 CRIIMI MAE CMBS Corp, 1998-1 CRIIMI MAE CMBS Corp, 1998-1 DLJ Mortgage Acceptance Corp, 1996-CF1 DLJ Mortgage Acceptance Corp, 1996-CF2 B-3 DLJ Mortgage Acceptance Corp, 1997-CF2 DLJ Commercial Mortgage Corp, 1998-CF1 B-1 DLJ Commercial Mortgage Corp, 1998-CF1 DLJ Mortgage Acceptance Corp, 1998-CG1 B-1 DLJ Commercial Mortgage Corp, 1999-CG2 B-1 DLJ Commercial Mortgage Corp, 1999-CG2 B-2 DLJ Commercial Mortgage Corp, 2000-CKP1 B-1 Fashion Valley Mall Mortgage Trust, 2002-C1A First Union National Bank Commercial Mortgage Trust, 1999-C1 First Union National Bank Commercial Mortgage Trust, 1999-C1 First Union National Bank Commercial Mortgage Trust, 1999 C-4 E First Union National Bank Commercial Mortgage Trust, Series 2001-C4 First Union National Bank Commercial Mortgage Trust, Series 2001-C4 First Union National Bank Commercial Mortgage Trust, Series 2001-C4 First Union National Bank-Lehman Brothers Commercial Mortgage Trust, 1997-C1 D First Union National Bank-Lehman Brothers Commercial Mortgage Trust, 1997-C1 E First Union National Bank-Lehman Brothers Commercial Mortgage Trust,1997-C1 First Union National Bank-Lehman Brothers Commercial Mortgage Trust,1997-C1 First Union National Bank-Lehman Brothers Commercial Mortgage Trust, 1997-C2 D First Union National Bank-Chase Manhattan Bank Commercial Mortgage Trust, 1999-1E First Union National Bank-Chase Manhattan Bank Commercial Mortgage Trust, 1999-1F GE Capital Commercial Mortgage Corp, 2001-2 E

Source: Fitch, Moody’s and S&P

112

Old Rating

Action

Current Rating

Date of Rating Action

BBB BBB+

Upgrade Upgrade

A AAA

3/18/2004 3/15/2004

BBB+ BBB-

Upgrade Upgrade

A+ A-

5/18/2004 5/18/2004

BBB+ BBB

Upgrade Upgrade

A A-

7/27/2004 7/27/2004

BBB BBBBBB+ BBB+ BBB BBBBBB+ BBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

A+ AAA+ A BBB+ ABBB+ BBB

5/24/2004 5/24/2004 5/5/2004 7/19/2004 5/5/2004 5/5/2004 6/30/2004 6/30/2004 6/30/2004

BBB+

Upgrade

AAA

3/25/2004

BBB+ BBB

Upgrade Upgrade

ABBB+

7/27/2004 7/27/2004

BBB+

Downgrade

BB+

3/9/2004

BBB-

Downgrade

BB-

3/9/2004

BBBBBB BB+

Downgrade Upgrade Upgrade

BB+ AAA AA

3/9/2004 6/15/2004 6/15/2004

BBB-

Upgrade

BBB+

3/2/2004

BBB

Upgrade

A

5/5/2004

BBB+ BBB BBBBBB+

Upgrade Upgrade Upgrade Upgrade

AA BBB+ A-

9/28/2004 5/18/2004 5/18/2004 6/16/2004

BBB

Upgrade

BBB+

5/18/2004

BBB BBB-

Upgrade Upgrade

AAA

2/13/2004 2/13/2004

BBB+ BBB BBBBBB+

Upgrade Upgrade Upgrade Upgrade

ABBB+ BBB A

7/15/2004 4/27/2004 4/27/2004 7/20/2004

Moody’s Class

Old Rating

Action

S&P Current Rating

Date of Rating Action

E F G

Baa2 Baa3 Baa3

Upgrade Upgrade Upgrade

A3 Baa2 Baa2

5/19/2004 5/19/2004 5/19/2004

D E

Baa2 Baa3

Upgrade Upgrade

Baa1 Baa2

7/8/2004 7/8/2004

D E

Baa2 Baa3

Upgrade Upgrade

A3 Baa1

2/11/2004 2/11/2004

D E

D E D E F-ALH

Baa1 Baa2

Baa2 Baa3 Baa2 Baa3 Baa2

Upgrade Upgrade

Aa2 A2

Upgrade Upgrade Downgrade Downgrade Downgrade

Aaa Aa2 Ba2 Ba3 Ba1

10/20/2004 10/20/2004 5/11/2004 5/11/2004 6/7/2004

F-POR

Baa1

Downgrade

Baa3

6/7/2004

Baa2 Baa3 Baa3

Downgrade Downgrade Downgrade

Ba2 Ba2 Ba3

6/7/2004 6/7/2004 6/7/2004

B-5

D E

Baa1 Baa3

Baa1

Baa2 Baa3

Upgrade Upgrade

Upgrade

Upgrade Upgrade

Aa1 A1

A2

A2 A3

Action

Current Rating

Date of Rating Action

D E D E

BBB BBBBBB+ BBB-

Upgrade Upgrade Upgrade Upgrade

A BBB+ A+ A-

5/27/2004 5/27/2004

F G H

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

ABBB+ BBB

8/26/2004 8/26/2004 8/26/2004

K-CR F G H

BBBBBB+ BBB BBB-

Downgrade Upgrade Upgrade Upgrade

D A BBB+ BBB

10/5/2004 8/27/2004 8/27/2004 8/27/2004

D E F-ALH

BBB BBBBBB

Downgrade Downgrade Downgrade

BBBBB+ BBB-

6/7/2004 6/7/2004 9/3/2004

G-ALH

BBB-

Downgrade

BB+

9/3/2004

B C B C

BBB+ BBB BBB+ BBB

Upgrade Upgrade Upgrade Upgrade

AAA AAA AAA AAA

8/27/2004 8/27/2004 9/22/2004 9/22/2004

1/9/2004 1/9/2004

G-WBC G-POR H-WBC

D E

Class

Old Rating

6/8/2004 6/8/2004 B-2

BBB+

Upgrade

AAA

8/25/2004

B-2TB B-1 B-2

BBBBBB BBB-

Upgrade Upgrade Upgrade

BBB A BBB+

1/30/2004 7/12/2004 7/12/2004

D E

BBB BBB-

Upgrade Upgrade

BBB+ BBB

4/16/2004 4/16/2004

G H J

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

A ABBB+

9/2/2004 9/2/2004 9/2/2004

D E

BBB BBB-

Upgrade Upgrade

ABBB+

8/2/2004 8/2/2004

5/11/2004

3/9/2004 3/9/2004

Please see additional important disclosures at the end of this report.

113

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix A

2004 RATING ACTIONS THROUGH SEPTEMBER 30, 2004 (CONTINUED) Fitch

CMBS Transactions GE Capital Commercial Mortgage Corp, 2001-2 GE Capital Commercial Mortgage Corp, 2001-2 GGP Mall Properties Trust, 2001-GGP1 GGP Mall Properties Trust, 2001-GGP1 GGP Mall Properties Trust, 2001-GGP1 GGP Mall Properties Trust, 2001-GGP1 GMAC Commercial Mortgage Securities, 1997-C1 GMAC Commercial Mortgage Securities, 1997-C1 GMAC Commercial Mortgage Securities, 1998-C2 GMAC Commercial Mortgage Securities, 1998-C2 GMAC Commercial Mortgage Securities, 1999-C1 GMAC Commercial Mortgage Securities, 1999-C1 GMAC Commercial Mortgage Securities, 1999-C2 GMAC Commercial Mortgage Securities, 1999-C2 GMAC Commercial Mortgage Securities, 1999-C3 GMAC Commercial Mortgage Securities, 2000-C3 GMAC Commercial Mortgage Securities, 2000-C3 GMAC Commercial Mortgage Securities, 2000-FL1 GMAC Commercial Mortgage Securities, 2000-FL1 GMAC Commercial Mortgage Securities, 2000-Fl-F GMAC Commercial Mortgage Securities, 2000-Fl-F GMAC Commercial Mortgage Securities, 2001-FL1 GMAC Commercial Mortgage Securities, 2001-FL1 GMAC Commercial Mortgage Securities, 2001-FLA GMAC Commercial Mortgage Securities, 2001-FLA GS Mortgage Securities Corp II, 1997-GLl GS Mortgage Securities Corp II, 1997-GLl GS Mortgage Securities Corp II, 2001-GLIII GS Mortgage Securities Corp II, 2001-GLIII GS Mortgage Securities Corp II, 2001-GLIII GS Mortgage Securities Corp II, 2001-GLIII GS Mortgage Securities Corp II, 2001-GLIII GS Mortgage Securities Corp II, 2002-GSFL V Heller Financial Mortgage Asset Corp, 1999-Ph1 Heller Financial Mortgage Asset Corp, 1999-Ph1 iStar Asset Receivables (Stars) Trust, 2002-1 iStar Asset Receivables (Stars) Trust, 2002-1 iStar Asset Receivables (Stars) Trust, 2002-1 iStar Asset Receivables (Stars) Trust, 2002-1 iStar Asset Receivables (Stars) Trust, 2002-1 JP Morgan Commercial Mortgage Finance Corp, 1997-C4 JP Morgan Commercial Mortgage Finance Corp, 1997-C5 JP Morgan Commercial Mortgage Finance Corp, 1997-C5 JP Morgan Commercial Mortgage Finance Corp, 1999-PLS1 JP Morgan Commercial Mortgage Finance Corp, 2000-Fl1 JP Morgan Commercial Mortgage Finance Corp, 2000-Fl1 JP Morgan Commercial Mortgage Finance Corp, 2000-Fl1 JP Morgan Chase Commercial Mortgage Finance Corp, 2002-WHALE1 JP Morgan Chase Commercial Mortgage Finance Corp, 2002-WHALE1 JP Morgan Chase Commercial Mortgage Finance Corp, 2002-WHALE1 JP Morgan Chase Commercial Mortgage Securities Corp, 2001-A JP Morgan Chase Commercial Mortgage Securities Corp, 2001-A JP Morgan Chase Commercial Mortgage Securities Corp, 2001-Fl1 JP Morgan Chase Commercial Mortgage Securities Corp, 2001-Fl1 JP Morgan Chase Commercial Mortgage Securities Corp, 2001-Fl1 JP Morgan Chase Commercial Mortgage Securities Corp, 2001-Fl1 J.P. Morgan Chase Commercial Mortgage Securities Corp, 2002-FL1 Lakewood Mall Finance Co., Series 1995-C1 LB Commercial Conduit Mortgage Trust II, 1996-C2 LB Commercial Mortgage Trust, 1998-C1 LB Commercial Mortgage Trust, 1998-C1 LB Commercial Mortgage Trust, 1998-C4 LB Commercial Mortgage Trust, 1998-C4 LB Commercial Mortgage Trust, 1999-C1 LB-UBS Commercial Mortgage Trust, Series 2000-C4 LB-UBS Commercial Mortgage Trust, Series 2000-C4 LB-UBS Commercial Mortgage Trust, Series 2000-C4 LB-UBS Commercial Mortgage Trust, 2000-C3 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust 2002-LLF C3 Lehman Brothers Floating Rate Commercial Mortgage Trust 2002-LLF C3 LTC Commercial Mortgage -Through Certificates Series 1996-1 LTC Commercial Mortgage -Through Certificates Ser 1998-1 Meristar Commercial Mortgage Trust, 1999-C1

Source: Fitch, Moody’s and S&P

114

Class

Old Rating

Action

Current Rating

Date of Rating Action

F G D-2 D-3

BBB BBBBBB BBB

Upgrade Upgrade Upgrade Upgrade

BBB+ BBB BBB+ BBB+

7/20/2004 7/20/2004 3/30/2004 3/30/2004

E F

BBB BBB-

Upgrade Upgrade

A BBB+

2/18/2004 2/18/2004

D E

BBB+ BBB

Upgrade Upgrade

A A-

5/18/2004 5/18/2004

E

BBB

Upgrade

BBB+

5/24/2004

D E

BBB BBB-

Upgrade Upgrade

AA A

2/25/2004 2/25/2004

F G

BBB+ BBB

Upgrade Upgrade

A BBB+

2/4/2004 2/4/2004

F G

BBB BBB-

Upgrade Upgrade

A A-

6/17/2004 6/17/2004

L M F

BBB+ BBBBBB+

Upgrade Upgrade Upgrade

ABBB A

10/27/2004 10/27/2004 10/26/2004

D D E F J K L D E E F G G

BBB+ BBB+ BBB BBBBBBBBB BB BBB BBBBBB+ BBB BBBBBB

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAAAAA AAA ABBB+ BBBAA A AAA AAA AA AAA

2/25/2004 2/10/2004 3/17/2004 3/17/2004 8/24/2004 8/24/2004 8/24/2004 3/16/2004 3/16/2004 1/7/2004 1/7/2004 1/7/2004 7/1/2004

E D E

BBB BBB BBB-

Upgrade Upgrade Upgrade

A A+ A-

3/25/2004 6/23/2004 6/23/2004

E K

BBB+ BBB

Upgrade Upgrade

AAA

7/27/2004 8/17/2004

D

BBB-

Upgrade

AAA

5/10/2004

Moody’s Class

Old Rating

Action

S&P Current Rating

Date of Rating Action

Class

Old Rating

Action

Current Rating

Date of Rating Action

D-2 D-3 E-2 E-3

Baa2 Baa2 Baa3 Baa3

Upgrade Upgrade Upgrade Upgrade

A2 A2 A3 A3

8/18/2004 8/18/2004 8/18/2004 8/18/2004

D-2 D-3 E-2 E-3

BBB BBB BBBBBB-

Upgrade Upgrade Upgrade Upgrade

AA+ AA+ AA AA

8/20/2004 8/20/2004 8/20/2004 8/20/2004

D E

Baa2 Baa3

Upgrade Upgrade

A3 Baa1

2/23/2004 2/23/2004

D E D E E F

BBB+ BBBBBB BBBBBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

ABBB+ BBB+ BBB A A-

10/18/2004 10/18/2004 3/18/2004 3/18/2004 5/21/2004 5/21/2004

S-MAC-2 S-MAC-3 D E

Baa2 Baa3 Baa2 Baa3

Downgrade Downgrade Downgrade Downgrade

Ba2 Ba3 Ba1 Ba2

9/30/2004 9/30/2004 1/26/2004 1/26/2004

D

Baa1

Upgrade

Aaa

4/6/2004

D E D E

BBB BBBBBB BBB-

Upgrade Upgrade Upgrade Upgrade

AA A AA AA-

2/25/2004 2/25/2004 6/3/2004 6/3/2004

F-NFC F-GGP G-NFC G-GGP H-GGP E

Baa1 Baa1 Baa2 Baa2 Baa3 Baa2

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

A1 A3 A2 Baa1 Baa2 A3

8/13/2004 8/13/2004 8/13/2004 8/13/2004 8/13/2004 8/12/2004

H J K

Baa1 Baa2 Baa3

Upgrade Upgrade Upgrade

A3 Baa1 Baa2

1/26/2004 1/26/2004 1/26/2004

H J K

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

A ABBB

1/26/2004 1/26/2004 1/26/2004

D E

Baa2 Baa3

Upgrade Upgrade

A2 Baa1

8/17/2004 8/17/2004

J K L

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

A+ A BBB

5/12/2004 5/12/2004 7/22/2004

E C

BBB+ BBB+

Upgrade Upgrade

AAA

8/19/2004 10/13/2004

D E

BBB BBB-

Upgrade Upgrade

BBB+ BBB

8/9/2004 8/9/2004

E F G

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

ABBB+ BBB

9/16/2004 9/16/2004 9/16/2004

J L D D

BBB+ BBBBBB BBB

Upgrade Downgrade Upgrade Upgrade

AA BB+ A+ A+

8/23/2004 8/23/2004 9/15/2004 9/15/2004

D E

Baa2 Baa3

Upgrade Upgrade

A3 Baa1

1/22/2004 1/22/2004

D

Baa2

Upgrade

Baa1

6/21/2004

C

Baa1

Downgrade

Ba1

10/15/2004

Please see additional important disclosures at the end of this report.

115

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix A

2004 RATING ACTIONS THROUGH SEPTEMBER 30, 2004 (CONTINUED) Fitch

CMBS Transactions Merrill Lynch Mortgage Investors, 1995-C2 Merrill Lynch Mortgage Investors, 1996-C1 Merrill Lynch Mortgage Investors, 1996-C2 Merrill Lynch Mortgage Investors, 1996-C2 Merrill Lynch Mortgage Investors, 1997-C1 Merrill Lynch Mortgage Investors, 1997-C1 Merrill Lynch Mortgage Investors, 1997-C2 Merrill Lynch Mortgage Investors, 1998-C2 Merrill Lynch Mortgage Investors, 1998-C2 Morgan Stanley Capital I, 1996-C1 Morgan Stanley Capital I, 1996-WF1 Morgan Stanley Capital I, 1997-C1 Morgan Stanley Capital I, 1997-HF1 Morgan Stanley Capital I, 1997-RR Morgan Stanley Capital I,1997-WF1 Morgan Stanley Capital I,1997-WF1 Morgan Stanley Capital I, 1997-XL1 Morgan Stanley Capital I, 1998-CF1 Morgan Stanley Capital I, 1998-HF2 Morgan Stanley Capital I, 1998-HF2 Morgan Stanley Capital I, 1998-WF1 Morgan Stanley Capital I, 1998-WF1 Morgan Stanley Capital I, 1998-XL1 Morgan Stanley Capital I, 1998-XL1 Morgan Stanley Capital I, 1999-CAM1 Morgan Stanley Capital I, 1999-CAM1 Morgan Stanley Capital I, 1999-LIFE1 Morgan Stanley Capital I, 1999-LIFE1 Morgan Stanley Capital I, 1999-RM1 Morgan Stanley Capital I, 1999-RM1 Morgan Stanley Capital I, 1999-RM1 Morgan Stanley Capital I, 2001-XLF Morgan Stanley Capital I, 2001-XLF Morgan Stanley Capital I, 2001-XLF Morgan Stanley Dean Witter Capital I, 2001-PPM Morgan Stanley Dean Witter Capital I, 2001-PPM Morgan Stanley Dean Witter Capital I, 2002-XLF Mortgage Capital Funding, 1996-MC2 Mortgage Capital Funding, 1996-MC2 Mortgage Capital Funding, 1997-MC1 Mortgage Capital Funding, 1997-MC1 Mortgage Capital Funding, 1998-MC1 Mortgage Capital Funding, 1998-MC1 Mortgage Capital Funding, 1998-MC1 N-45 First CMBS Issuer Corp, 2000-1 NationsLink Funding Corp, 1998-2 NationsLink Funding Corp, 1998-2 NationsLink Funding Corp, 1999-1 NationsLink Funding Corp, 1999-1 Nomura Asset Securities Corp, 1996-MD5 Nomura Asset Securities Corp, 1996-MD5 Nomura Asset Securities Corp, 1998-D6 Nomura Asset Securities Corp, 1998-D6 NSS Mortgage Securities II Corp Opryland Hotel Trust, 2001-OPRY Opryland Hotel Trust, 2001-OPRY Park Square Mortgage Trust. 2000-C5 Park Square Mortgage Trust. 2000-C5 Penn Mutual Life Insurance Company, 1996-PML PNC Mortgage Acceptance Corp, 1999 CM-1 PNC Mortgage Acceptance Corp, 1999 CM-1 PNC Mortgage Acceptance Corp, 2000-C1 PNC Mortgage Acceptance Corp, 2000-C2 PNC Mortgage Acceptance Corp, 2000-C2 PNC Mortgage Acceptance Corp, 2000-C2 Prudential Mortgage Capital Funding, ROCK 2001-C1 Prudential Mortgage Capital Funding, ROCK 2001-C1 Prudential Mortgage Capital Funding, ROCK 2001-C1 Prudential Securities Secured Financing Corp, 1998-C1 Prudential Securities Secured Financing Corp, 1998-C1 Prudential Securities Secured Financing Corp, KEY 2000-C1 Red Mountain Funding LLC, 1997-1 Red Mountain Funding LLC, 1997-1 Salomon Brothers Mortgage Securities VII, 1993-9

Source: Fitch, Moody’s and S&P

116

Class

Old Rating

Action

Current Rating

Date of Rating Action

E

BBB-

Upgrade

BBB+

3/15/2004

D E

BBB+ BBB

Upgrade Upgrade

AA A+

4/23/2004 4/23/2004

D

BBB+

Upgrade

AA-

5/11/2004

E

BBB

Upgrade

BBB+

6/23/2004

F E C

BBB+ BBB+ BBB

Upgrade Upgrade Upgrade

AAA AAA

6/30/2004 2/23/2004 2/17/2004

E

BBB

Upgrade

BBB+

2/5/2004

E F D

BBB BBBBBB+

Upgrade Upgrade Upgrade

BBB+ BBB A-

5/18/2004 5/18/2004 6/30/2004

E F E F E F G

BBB BBBBBB BBBBBB+ BBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

ABBB+ BBB+ BBB A BBB+ BBB

6/15/2004 6/15/2004 6/30/2004 6/30/2004 7/15/2004 7/15/2004 7/15/2004

E F

BBB BBB-

Upgrade Upgrade

BBB+ BBB

8/3/2004 8/3/2004

D

BBB+

Upgrade

A+

4/21/2004

G

BBB

Upgrade

BBB+

6/22/2004

D

BBB+

Upgrade

A+

7/20/2004

A-4 A-5 A-4 A-5 C

BBB BBBBBB BBBBBB+

Upgrade Upgrade Upgrade Upgrade Upgrade

AA A+ ABBB+ A-

7/21/2004 7/21/2004 1/6/2004 1/6/2004 3/10/2004

M

BBB

Upgrade

BBB

3/17/2004

E

BBB

Upgrade

BBB+

5/18/2004

E F G

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

ABBB+ BBB

8/10/2004 8/10/2004 8/10/2004

D E

BBB BBB-

Downgrade Downgrade

BB+ BB

4/21/2004 4/21/2004

Moody’s Class

D E

Old Rating

Baa1 Baa3

Action

Upgrade Upgrade

S&P Current Rating

A2 Baa2

Date of Rating Action

4/29/2004 4/29/2004

E

Baa2

Upgrade

Aa1

9/16/2004

D E

Baa2 Baa3

Upgrade Upgrade

Aa3 A2

2/26/2004 2/26/2004

E

Baa2

Upgrade

Baa1

Baa2 Baa3 Baa3

Downgrade Downgrade Downgrade

Ba1 Ba3 Ba2

5/17/2004 5/17/2004 5/17/2004

G1 D E D E

Baa3 Baa1 Baa2 Baa2 Baa3

Downgrade Upgrade Upgrade Upgrade Upgrade

Ba2 Aaa Aa2 A2 Baa2

10/20/2004 9/30/2004 9/30/2004 10/6/2004 10/6/2004

D D

Baa2 Baa2

Upgrade Upgrade

A2 A2

7/30/2004 1/22/2004

D E

Baa2 Baa3

Upgrade Upgrade

A2 A3

5/17/2004 5/17/2004

C D B-2 B-3

A3 Baa3 Baa2 Baa3

Upgrade Upgrade Downgrade Downgrade

A2 Baa2 Baa3 Ba1

5/19/2004 5/19/2004 4/15/2004 4/15/2004

Baa2 Baa3

Upgrade Upgrade

A3 Baa1

D D E D E

BBB BBB BBBBBB BBB-

Upgrade Upgrade Upgrade Upgrade Upgrade

A AA AAAA-

10/19/2004 4/7/2004 4/7/2004 4/7/2004 4/7/2004

D E

BBB BBB-

Upgrade Upgrade

ABBB+

10/14/2004 10/14/2004

C

BBB

Upgrade

AA-

7/12/2004

E F

BBB+ BBB

Upgrade Upgrade

AA+ AA

4/20/2004 4/20/2004

E F

BBB BBB-

Upgrade Upgrade

BBB+ BBB

8/25/2004 8/25/2004

E F

BBB+ BBB

Upgrade Upgrade

A+ A

4/17/2004 4/17/2004

D E

BBB+ BBB-

Upgrade Upgrade

A BBB+

4/21/2004 4/21/2004

F

BB

Upgrade

BBB-

7/19/2004

C D

BBB B

Upgrade Upgrade

A+ BB+

5/26/2004 5/26/2004

B-1 B-2

BBB+ BBB

Upgrade Upgrade

A A-

9/7/2004 9/7/2004

E F G

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

A ABBB+

8/6/2004 8/6/2004 8/6/2004

D E E

BBB BBBBBB+

Upgrade Upgrade Upgrade

A AA-

6/29/2004 6/29/2004 2/12/2004

B-4

BBB-

Upgrade

BBB+

6/15/2004

Action

Current Rating

Date of Rating Action

8/5/2004

F G1 G7

D E

Class

Old Rating

2/4/2004 2/4/2004

Please see additional important disclosures at the end of this report.

117

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix A

2004 RATING ACTIONS THROUGH SEPTEMBER 30, 2004 (CONTINUED) Fitch

CMBS Transactions Salomon Brothers Mortgage Securities VII, 1999-AQ1 Salomon Brothers Mortgage Securities VII, 1999-C1 Salomon Brothers Mortgage Securities VII, 1999-C1 Salomon Brothers Mortgage Securities VII, 1999-LB1 Salomon Brothers Mortgage Securities VII, 2000-1 Salomon Brothers Mortgage Securities VII, 2000-NL1 Salomon Brothers Mortgage Securities VII, 2000-NL1 Salomon Brothers Mortgage Securities VII, 2000-NL1 Salomon Brothers Mortgage Securities VII, 2000-NL1 Salomon Brothers Mortgage Securities VII, 2000-NL1 Salomon Brothers Mortgage Securities VII, 2001-CDC Salomon Brothers Mortgage Securities VII, 2001-CDC Salomon Brothers Mortgage Securities VII, 2001-CDC Salomon Brothers Mortgage Securities VII, 2001-CDC Salomon Brothers Mortgage Securities VII, 2002-CDC Salomon Brothers Mortgage Securities VII, 2002-CDC Salomon Brothers Mortgage Securities VII, 2002-CDC Salomon Brothers Mortgage Securities VII, 2002-CDC SASCO Floating Rate Commercial Mortgage Trust, 2001-C8 SASCO Floating Rate Commercial Mortgage Trust, 2001-C8 SASCO Floating Rate Commercial Mortgage Trust, 2001-C8 SASCO Floating Rate Commercial Mortgage Trust, 2001-C8 Sawgrass Mills Trust, 2001-XLSGM Sawgrass Mills Trust, 2001-XLSGM Scottsdale Fashion Square Trust Scottsdale Fashion Square Trust Scottsdale Fashion Square Trust Scottsdale Fashion Square Trust Strategic Hotel Capital LLC SHC, 2003-1 Strategic Hotel Capital LLC SHC, 2003-1 Strategic Hotel Capital LLC SHC, 2003-1 Structured Asset Securities Corp, 1995-C1 Structured Asset Securities Corp, 1997-LL1 SunAmerica Mortgage Trust, 1999-C2B SunAmerica Mortgage Trust, 1999-C2B TrizecHahn Office Properties Trust, 2001-TZH TrizecHahn Office Properties Trust, 2001-TZH TrizecHahn Office Properties Trust, 2001-TZH TrizecHahn Office Properties Trust, 2001-TZH UBS 400 Atlantic Street Mortgage Trust, 2002-C1A UBS 400 Atlantic Street Mortgage Trust, 2002-C1A UBS 400 Atlantic Street Mortgage Trust, 2002-C1A Vornado Finance LLC, 2000-VNO Vornado Finance LLC, 2000-VNO Wisconsin Avenue Securities, 1995-M4 Wisconsin Avenue Securities, 1996-M3 Wisconsin Avenue Securities, 1996-M5

Source: Fitch, Moody’s and S&P

118

Class

Old Rating

Action

Current Rating

Date of Rating Action

E F

BBB BBB-

Upgrade Upgrade

A+ A

6/9/2004 6/9/2004

E F G G H

BBB+ BBB BB+ BBB+ BBB

Upgrade Upgrade Upgrade Upgrade Upgrade

AA AABBB+ AA A

2/3/2004 2/3/2004 2/3/2004 3/25/2004 3/25/2004

G-DS

BBB-

Downgrade

BB+

2/25/2004

J K L

BBB+ BBB BBB-

Upgrade Upgrade Upgrade

AAA AAA AAA

5/14/2004 5/14/2004 5/14/2004

D-1 D-2 E-1 E-2 H I J F

BBB BBB BBBBBBBBB+ BBB BBBBBB-

Upgrade Upgrade Upgrade Upgrade Downgrade Downgrade Downgrade Upgrade

A+ A+ BBB BBB BBB BBBBB+ AAA

5/6/2004 5/6/2004 5/6/2004 5/6/2004 4/19/2004 4/19/2004 4/19/2004 5/26/2004

D-3 D-4 E-3 E-4

BBB BBB BBBBBB-

Upgrade Upgrade Upgrade Upgrade

AABBB BBB

7/29/2004 7/29/2004 7/29/2004 7/29/2004

B C B

BBB BBB BBB-

Upgrade Upgrade Upgrade

AAA AAA A

10/14/2004 10/14/2004 1/15/2004

Moody’s Class E F

E-DS F-DS G-DS H-DS X-2B H-DEN J-DEN K-DEN H J K

Old Rating Baa2 Baa3

Baa1 Baa2 Baa3 Baa3 Baa3 Baa1 Baa2 Baa3 Baa1 Baa2 Baa3

Action Upgrade Upgrade

Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Upgrade Upgrade Upgrade

S&P Current Rating A1 A3

Ba1 Ba1 Ba3 B3 Ba1 Baa2 Baa3 Ba1 Aaa Aaa Aaa

Date of Rating Action

Class

Old Rating

M-3

BBB

Upgrade

A

6/15/2004

M-3 B-3

BBB BBB

Upgrade Upgrade

A AA+

6/15/2004 6/15/2004

G

BBB

Upgrade

AA

9/20/2004

E F

BBB BBB-

Upgrade Upgrade

BBB+ BBB

6/24/2004 6/24/2004

D-3 D-4 E-3 E-4

BBB BBB BBBBBB-

Upgrade Upgrade Upgrade Upgrade

AA AA A A

7/19/2004 7/19/2004 7/19/2004 7/19/2004

Action

Current Rating

Date of Rating Action

10/28/2004 10/28/2004

6/25/2004 6/25/2004 6/25/2004 6/25/2004 8/26/2004 8/26/2004 8/26/2004 8/26/2004 5/27/2004 5/27/2004 5/27/2004

E B-4 B-5 D-3 D-4

Baa1 Baa2 Baa3 Baa3 Baa3

Upgrade Downgrade Downgrade Upgrade Upgrade

A2 Baa3 Ba1 Baa2 Baa2

10/15/2004 1/23/2004 1/23/2004 9/16/2004 9/16/2004

B-2 B-3 B-4 E F

Baa1 Baa2 Baa3 Baa2 Baa3

Downgrade Downgrade Downgrade Upgrade Upgrade

Baa2 Baa3 Ba1 Baa1 Baa2

9/14/2004 9/14/2004 9/14/2004 9/14/2004 9/14/2004

Please see additional important disclosures at the end of this report.

119

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix B

LARGE LOAN CMBS WITH REITS AS BORROWERS

REIT

Property Type

Arden Realty Inc. Associated Estates Realty Associated Estates Realty BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. BRE Properties Inc. Crescent Real Estate Equities Crescent Real Estate Equities Crescent Real Estate Equities Equity Office Properties Trust Equity Office Properties Trust Getty Realty Corp. Getty Realty Corp. Getty Realty Corp. Getty Realty Corp. Getty Realty Corp. Host Marriott Corp. HRPT Properties Trust Kimco Realty Corp. Macerich Co. MeriStar Hospitality MeriStar Hospitality Prentiss Properties Trust Prentiss Properties Trust Ramco-Gershenson Properties Reckson Associates Realty Corp Simon Property Group Inc. Simon Property Group Inc. Simon Property Group Inc. Simon Property Group Inc. Tanger Factory Outlet Centers Town & Country Trust Vornado Realty Trust

Office - Suburban Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Apartments Diversified / Misc. Diversified / Misc. Diversified / Misc. Office - CBD Office - CBD Triple-Net Lease Triple-Net Lease Triple-Net Lease Triple-Net Lease Triple-Net Lease Hotel Office - Suburban Shopping Center Regional Mall Hotel Hotel Office - Suburban Office - Suburban Shopping Center Office - Suburban Regional Mall Regional Mall Regional Mall Regional Mall Outlet Apartments Diversified / Misc.

Source: Morgan Stanley, Commercial Mortgage Alert and Intex

120

Deal Name

Amount (in $MM)

Lehman Large Loan, Series 1997-LL1 Chase Commercial Mortgage Securities Corp, 2000-1 Chase Commercial Mortgage Securities Corp, 2000-2 COMM, 2002-FL7 GS Mortgage Securities Corp II, 2003-GSFL VI Credit Suisse First Boston Mortgage Securities Corp, 2004-TFL1 Bear Stearns Commercial Mortgage Securities Inc., 2004-ESA Bear Stearns Commercial Mortgage Securities Inc., 2004-HS2 Bear Stearns Commercial Mortgage Securities Inc., 2002-HOME Structured Asset Securities Corp, 1998-C3 Greenwich Capital Commercial Funding Corporation, 2004-FL2 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2003-LLF C2 Bear Stearns Commercial Mortgage Securities Inc., 2003-BBA1 Banc of America Large Loan, 2001-7WTC Structured Asset Securities Corp, 1997-L1 iStar Asset Receivables Trust, 2000-1 Asset Securitization Corp, 1995-MD4 Nomura Asset Securities Corp, 1996-MD5 Columbia Center Trust, 2000-CCT LB Commercial Mortgage Trust, 1999-C2 Thirteen s of General Growth Properties GGP Ala Moana, 1999-C1 GGP/Homart Commercial Mortgage-Backed Securities, 1999-C1 GGP - Ivanhoe Commercial Backed Securities, 1999-C1 Morgan Stanley Capital I Inc., 2003-XLF Host Marriott Pool Trust, 1999-HMT Office Portfolio Trust, 2001-HRPT First Union - Chase Manhattan Commercial Mortgage Trust, 1999-C2 Morgan Stanley Dean Witter Capital I Inc., 2000-XLF MeriStar Commercial Mortgage Trust, 1999-C1 Structured Asset Securities Corp, 1998-C3 Lehman Large Loan, 1997-LL1 Structured Asset Securities Corp, 1998-C2 Morgan Stanley Capital I, 1998-XL1 Morgan Stanley Dean Witter Capital I Inc., 2000-XLF Chase Commercial Mortgage Securities Corp, 1999-2 Credit Suisse First Boston Mortgage Securities Corp, 2001-SPG1 iStar Asset Receivables Trust, 2000-1 Morgan Stanley Dean Witter Capital I Inc., 2003-HQ2 Morgan Stanley Capital I Inc., 2003-XLF GS Mortgage Securities Corp II, 2004-GG2 Morgan Stanley Dean Witter Capital I Inc., 2000-XLF

175.0 98.5 98.8 123.0 83.0 77.2 2,050.0 660.0 400.0 64.8 235.0 69.4 160.0 383.0 275.0 97.1 239.0 161.0 195.0 146.5 560.0 500.0 590.0 700.2 53.0 665.0 260.0 120.9 60.0 330.0 250.0 180.1 120.0 50.0 50.0 89.0 277.0 54.2 61.3 115.0 71.5 105.3

Please see additional important disclosures at the end of this report.

Closing Date 10/15/1997 3/28/2000 6/28/2000 10/21/2002 12/17/2003 4/6/2004 6/29/2004 4/1/2004 2/22/2002 10/29/1998 11/23/2004 6/9/2003 5/6/2003 8/9/2001 10/15/1997 5/17/2000 10/30/1995 4/2/1996 6/21/2000 10/14/1999 11/25/1997 8/26/1999 11/30/1999 9/30/1999 7/9/2003 8/18/1999 2/28/2001 5/20/1999 8/9/2000 8/30/1999 10/29/1998 10/15/1997 4/29/1998 6/11/1998 8/9/2000 11/23/1999 8/22/2001 5/17/2000 3/27/2003 7/9/2003 8/12/2004 8/9/2000

121

Transforming Real Estate Finance chapter 8

BBB CMBS and REITs appendix C

TOP RATING AGENCY CRITERIA FOR REITS

Moody’s breaks down its five key credit characteristics as follows: strength and stability of cash flow, profitability and operational efficiency, liquidity, asset quality, and capital adequacy and structure. Within those, the rating agency focuses on earnings momentum, fixed charge coverage, operating margins, funding capacity and structure, unencumbered assets to total assets, portfolio diversification, development focus, asset market value, capital structure, stock market valuation and bond pricing. Fitch breaks the same characteristics down quite clearly into three main components: financial analysis, a review of the management team and a review of the property portfolio. Fitch considers the ability to adapt to changing business environments one of the most important aspects of a management team. When examining financials, Fitch looks at the following: • Cash flow coverage • Property level data • Unencumbered/Encumbered cash flow • Refinance risk • Constant refinance rate • Debt maturity schedule • Capital Structure • Leverage - Capitalization of Cash Flow - Debt/EBITDA - Debt to Book Value (not as effective) - Debt to Market Capitalization (not as effective) Though the list is long, Fitch’s most important financial tests are unencumbered debt service coverage, total debt service coverage and total fixed charge coverage. Within Fitch’s review of the property portfolio, investment diversity and asset quality are examined. Market, location, site plan, building, property management, tenant breakdown and lease expiration schedule can all affect the quality of a portfolio. S&P defines its two main components as business profile and financial profile, quite similar to Fitch. Within the business profile, S&P examines a REIT’s market position, its asset quality, diversification/stability of operations and operating strategy/management evaluation. S&P’s analysis of the financial profile breaks down into five separate parts: • Financial Policy (dividend, leverage, ing, bank line usage, acquisitions/development, etc.) • Profitability (efficiency measures, returns, property level performance, etc.) • Capital Structure (debt/capital, character of debt, cost/flexibility of debt, fixed/variable rate debt, debt maturity schedule, etc.) • Cash flow protection (coverage measures, liquidity, etc.) • Financial flexibility (degree to which holdings are encumbered, leveragability of assets, sources of financing, quality of ownership base, etc.) All in all, there are no hard and fast rules used when rating a REIT, but the main factors examined by the different rating agencies are quite similar.

122

Chapter 9

Transforming Real Estate Finance

CMBS Conduit Subordination Levels

Please see additional important disclosures at the end of this report.

123

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels Subordination: The Right Range Until the super senior structure emerged in fixed-rate CMBS in October 2004, AAA credit enhancement levels continued to shrink. In the first five months of 2004, AAA CMBS subordination levels fell to an average of 14%. Five years ago, sub levels of 14% were consistent with single-A CMBS.

AAA AA A BBB BB B

exhibit 1

AVERAGE SUBORDINATION FOR CONDUIT/FUSION TRANSACTIONS

1998 29% 24% 18% 13% 6% 3%

1999 27% 22% 17% 12% 6% 3%

2000 23% 19% 14% 11% 5% 3%

2001 21% 17% 13% 9% 4% 2%

2002 20% 16% 12% 8% 4% 2%

2003 17% 14% 10% 7% 3% 2%

20041 14% 12% 9% 5% 3% 2%

1

As of August 19, 2004.

Source: Commercial Mortgage Alert, Morgan Stanley

Particularly at the top of the capital structure, the rating agencies have been reducing the subordination levels required to obtain a CMBS rating. Each year since 1995, subordination levels on investment grade CMBS have declined in an effort to recalibrate CMBS bond ratings with corporate bond ratings. In this section, we estimate what the original subordination levels should have been on 1997-2001 conduit deals in order to make their original ratings consistent with the credit of unsecured corporate bonds. TODAY’S S UB L EVELS F ALL W ITHIN O UR C ALCULATED R ANGES

After running two scenarios on the seasoned transactions, we found that the “right” original subordination levels for AAA conduit CMBS should have been between 9% and 15%. In our estimate, those levels would have resulted in identical default rates for AAA CMBS and AAA corporates. Our hypothetical “right” subordination level range is much lower than the 21%-29% range used by the rating agencies during 1998-2000. For BBB CMBS, our calculated “right” range is about 4% to 8%. That is, 4% to 8% subordination would have resulted in the same percentage of BBB CMBS defaults as historical BBB corporate defaults.

124

The results of this study show that today’s average subordination levels on new issue transactions are in a reasonable range, assuming commercial mortgage performance in the next decade is similar to the loan loss experience of seasoned deals to date. Based on our estimates, however, there is still some room for subordination levels to decline further. At the June 2004 CMSA conference, the rating agencies said that they will exercise caution and refrain from lowering sub levels further (for now), even though empirical data s lower levels. exhibit 2

Aaa Aa2 A2 Baa2 Ba2 B2

CMBS SUBORDINATION LEVELS CONSISTENT WITH CORPORATE BOND RATINGS Avg Default Timing 34% Severity 9.1% 8.3% 4.9% 3.8% 2.1% 1.1%

1986 Default Timing 43% Severity 15.4% 14.3% 9.8% 7.8% 4.1% 1.6%

Source: Morgan Stanley

METHODOLOGY

In our recent Projecting Losses studies, we used the Esaki-Snyderman default timing curves to project expected losses for conduits issued from 1997 to 2001. We compared those losses to remaining subordination levels to estimate lifetime default rates for bonds issued during that period. We now calculate the hypothetical original subordination levels that result in expected default rates that match historical corporate bond default rates. In the Projecting Losses pieces, we calculated expected future losses on 159 conduit transactions that were issued between January 1997 and December 2001. After calculating future losses for each transaction, we compared these numbers to current subordination levels on BBB-, BB and single-B bonds. A subordinationto-loss ratio that is less than one implies a bond default prior to maturity.

Please see additional important disclosures at the end of this report.

125

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels For this chapter, we adjusted subordination levels on the 159 bonds until the number of bonds with ratios below one was consistent with the corporate bond default rate for a particular rating. For example, according to Moody’s, the average 10-year cumulative default rate for single-A corporate bonds is 1.88%. Since we are examining a universe of 159 deals, this 1.88% default rate corresponds to 2.99 CMBS bonds out of 159 defaulting. We rounded the 2.99 bond defaults to 3 defaults, and kept lowering subordination levels across all deals until the universe had 3 bond defaults and was on the verge of experiencing 4 bond defaults. We repeated this procedure for all ratings from Aaa to B2 under two different scenarios that we used in the Projecting Losses articles to forecast future losses. One scenario assumed the average default timing curve from Esaki’s commercial mortgage default study and the average severity rate of 34%. The other scenario assumed the default timing curve of the 1986 origination cohort from the Esaki Study, and a 43% severity rate, which is the severity experienced by liquidated CMBS loans. For more details on these scenarios and the assumptions we used to forecast losses, see our Projecting Losses pieces1.

1

Projecting Losses on BBB- CMBS, April 16, 2004; Projecting Losses on BB CMBS, May 21, 2004; Projecting Losses on Single-B CMBS, May 28, 2004.

126

exhibit 3

Aaa Aa2 A2 Baa2 Ba2 B2

CMBS DEFAULTS CALIBRATED TO CORPORATE BOND DEFAULTS

Moody’s Average 10-Year Cumulative Default Rates for Corporate Bonds, 1983-2002 (%) 0.38 0.73 1.88 5.71 15.81 46.22

Corresponding # of CMBS Defaults Necessary to be Consistent With Corporates 0.60 1.16 2.99 9.08 25.14 73.49

Source: Moody’s Investors Service, Morgan Stanley

Please see additional important disclosures at the end of this report.

127

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels Section I: Projecting Losses on BBB- CMBS SUBORDINATION V S. E XPECTED L OSSES

In order to test the hypothesis that current credit enhancement levels continue to provide adequate for mezzanine and subordinate CMBS, we examine the ratio of BBB- subordination to expected losses on all conduit deals issued between 1997 and 2001. (We describe the methodology for predicting expected losses later in this section.) A ratio that is less than 1 means that we expect the class to default before maturity; a ratio of greater than 1 means we expect no default. Depending on the severity assumption and timing curve used to project future losses, between 0 and 5 bonds have subordination-to-loss ratios of less than 1. (See Exhibit 4) Under the most severe assumptions, about 11% of classes have a ratio between 1 and 2. We believe that classes with a ratio between 1 and 2 may be vulnerable to a downgrade. The majority of BBB- bonds, however, have a ratio of greater than 5. In our view, this shows that there is currently minimal default risk in the investment grade CMBS sector. If the majority of BBB-s are well protected from real estate risk, higher rated classes are even more insulated.

exhibit 4

BBB- SUBORDINATION-TO-EXPECTED-LOSS RATIOS FOR 1997-2001 CONDUITS

Ranges for BBBSubordination to Expected Loss Ratio 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 >8

Average Default Timing Curve; 34% Severity # of Deals 0 4 5 8 6 4 4 5 123

Source: Morgan Stanley, Intex, Trepp

128

% of Deals 0.0 2.5 3.1 5.0 3.8 2.5 2.5 3.1 77.4

Average Default Timing Curve; 43% Severity # of Deals 0 4 9 9 4 6 5 4 118

% of Deals 0.0 2.5 5.7 5.7 2.5 3.8 3.1 2.5 74.2

In Exhibit 8 of this section, we rank our universe of BBB- CMBS based on subordination to expected loss ratios. We believe that the bonds at the bottom of the list (with the highest ratios) have the highest level of principal protection and greatest chance for rating agency upgrade. Likewise, we believe that the bonds at the top of the list (with the lowest ratios) have a higher chance of default and downgrade. SCOPE O F T HE S TUDY

This study analyzes and ranks bonds based on the level of principal protection from losses. Our analysis does not assess the risk of potential downgrades that may occur for other reasons, such as interest shortfalls. For example, while ASC 1997-D5 A2 has a high subordination to expected loss ratio,1 this class was downgraded by Moody’s in September 2003 to Baa3 from its original A1 rating. The downgrade was prompted by interest shortfalls caused by servicer reimbursements of non-recoverable advances. Likewise, S&P downgraded this class to BBB from A+.

1986 Default Timing Curve; 34% Severity # of Deals 4 14 8 11 7 6 8 8 93

% of Deals 2.5 8.8 5.0 6.9 4.4 3.8 5.0 5.0 58.5

1986 Default Timing Curve; 43% Severity # of Deals 5 17 10 12 6 10 10 9 80

% of Deals 3.1 10.7 6.3 7.5 3.8 6.3 6.3 5.7 50.3

1 Ratio of 60.8, assuming average default timing curve from Esaki study and 34% severity rate; Ratio of 96.2, assuming 1986 default timing curve and 43% severity rate.

Please see additional important disclosures at the end of this report.

129

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels A high subordination to expected loss ratio1 does not mean a BBB- class necessarily offers good value. The market may have already priced in the probability of an upgrade. There may be cases, however, where the spread has not fully adjusted for the credit. For example, JPMC 1997-C4 F is currently rated BBBby S&P. With a current subordination level of 13.09% and a subordination-toloss ratio of 1581.4, we would argue that this class is a good upgrade candidate. This class currently has a $109 price and trades around S+100 bp. PROJECTING L OSSES - M AY 2 003

In order to assess the strength of this study as a predictor of rating actions, we reviewed rating changes on BBB- bonds from the prior version of this study.2

exhibit 5

PROJECTING LOSSES STUDY - MAY 2003 RATING AGENCIES UPGRADED 13 OF THE 20 CLASSES WITH THE HIGHEST SUBORDINATION-TO-LOSS RATIOS

Subordination to Expected Deal Loss Ratio1 CCMSC 2000-3 NA2 MSC 1997-WF1 1090.0 JPMC 1999-C7 326.0 JPMC 1997-C4 309.4 MSC 1998-WF2 159.8 MSC 1997-HF1 150.8 NLFC 1999-2 120.0 CCMSC 1997-1 118.5 PNCMA 2000-C2 113.6 MSDWC 2000-LIF2 110.1 GMACC 1999-C2 91.5 MCFI 1997-MC2 87.9 FULB 1997-C1 77.1 CMAC 1999-C1 72.6 NASC 1998-D6 69.9 ACMF 1997-C1 66.0 GMACC 1997-C1 62.6 JPMC 1997-C5 53.6 DLJCM 1999-CG2 48.4 CCMSC 1998-2 47.9

Ratings Actions Since May 2003 Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded Upgraded -

1

As of May 2003. Expected loss ratio equals zero. *S&P Rating Source: Morgan Stanley, Rating Agencies 2

2

See Projecting Losses: BBB- Upgrade Potential, May 2003

130

Old Moody’s Rating1 Baa3 Baa2 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 -

Old Fitch Rating1 BBBBBB-* BBB+ BBBBBB+ BBB BBB BBBBBBBBBBBBBBBBBBBBBBBB-* BBBBBB-

Current Moody’s Rating A2 A3 Baa3 Baa3 Baa3 Baa3 A3 Baa3 Baa3 Baa3 Baa3 Baa3 -

Over the course of a year, we would expect the bonds with the highest subordination-to-loss ratios to experience the greatest number of upgrades. We found that 65% of the 20 BBB- classes with the highest subordination-to-loss ratios were upgraded during the past year (Exhibit 5). The upgrade percentage declined for the next highest 20 (40%). We would also expect the BBB- bonds with the lowest subordination-to-loss ratios to experience fewer upgrades and potentially experience some downgrades. We found that 2 of the 20 BBB- classes with the lowest ratios were downgraded during the past year. Only 1 class was upgraded.

Date of Moody’s Upgrade 2/26/2004 7/23/2003 3/9/2004 -

Current Fitch Rating BBBBBB* AAA BBBAA ABBB+ BBBBBB A BBB BBB+ AA+ BBB+ BBB* BBBBBB-

Date of Fitch Upgrade 7/22/2003* Two Upgrades: 11/12/2003 and 3/30/2004 2/23/2004 9/10/2003 9/10/2003 7/7/2003 2/13/2004 11/25/2003 Two Upgrades: 11/18/2003 and 1/06/2004 Three Upgrades: 7/11/2003, 2/3/2004 and 3/17/2004 2/18/2004 9/23/2003* -

Please see additional important disclosures at the end of this report.

131

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels METHODOLOGY

In this section, we analyze the lowest rated investment grade class of 159 conduit transactions tracked by Trepp. Each conduit transaction was issued between January 1997 and December 2001. 140 of 159 classes are currently rated BBB- by Moody’s, Fitch or S&P; the other 19 classes are rated single-A, A-, BBB+ or BBB. For each transaction, we calculate a future expected loss rate based on historical losses and losses implied by the current level of 30, 60, and 90-day delinquencies, REO, foreclosures, performing specially serviced loans and servicer watchlisted loans. We do not for the impact of defeased collateral within the deals. We begin by asg a probability of liquidation for each delinquency category, based on results from Esaki’s commercial mortgage default study. Assumptions for Probabilities of Liquidation

In Esaki’s study, about 55% of loans that became more than 90 days delinquent were ultimately liquidated. For this section, we borrow the results from the Esaki study and assume that 55% of 90+ day delinquent loans will be liquidated. We then assume that 60-day delinquent loans will have a 25% liquidation rate, and 30-day delinquent loans will have a 10% liquidation rate. To be conservative, we assume that all foreclosed and REO loans will be liquidated. We assign a 2.5% liquidation rate to servicer watchlisted loans and a 5% liquidation rate to performing specially serviced loans. We assume a low probability of liquidation for the performing specially serviced loans because these loans could have been transferred to the special servicer for technical rather than monetary defaults. Calculating Loan Losses: Average Timing of Defaults

We then multiply the projected liquidated loan total by a severity rate to compute each deal’s expected loss rate for the year. For our analysis, we consider two loss severity rates: 34% and 43%3. The first severity rate that we assume is 34%. Liquidated life insurance company loans in the Commercial Mortgage Default Study: 1972-2000, by Howard Esaki, experienced a 34% severity rate. The average severity rate on liquidated CMBS loans is higher, at 43%3. The calculated loss rates for the year are then used in conjunction with historical losses to project future loss rates. Since the calculated losses are based on current delinquency levels, we assume that the resulting losses will occur within one year from today. For example, with a 34% severity rate assumption, DLJCM 2000-CKP1, is expected to have a 0.4% loss rate based on current levels of delinquencies, specially serviced loans and watchlisted loans. This transaction has seasoned for 3 years, so we assume losses will occur in year 4. The projected loss based on current delinquencies (0.4%) is then added to the deal’s historical losses of 2.5%. To project future losses, we assume that the timing of CMBS loan losses in this study mirrors the average timing of defaults on the life insurance company loans from Esaki’s study.

3

See special servicer severities, February 6, 2004.

132

According to the timing of defaults presented in Esaki’s study, about 33% of loans default by the end of year 4. The DLJCM 2000-CKP1 transaction is expected to have a 2.9% loss rate by year 4, based on historical losses and projected losses of currently delinquent loans. If we apply the results of Esaki’s study, this 2.9% loss rate should correspond to 33% of total loan losses. Therefore, we conclude that this transaction may suffer from a 5.9% loss rate (based on original balances) during its remaining lifetime, since 67% of loan losses have yet to occur. (2.9*(1-33%)¸33%=5.9% loss rate) A 5.9% loss rate based on original balances corresponds to a 6.3% loss rate, based on current balances.

exhibit 6

Year 1 2 3 4 5 6 7 8 9 10

AVERAGE TIMING OF DEFAULTS FROM ESAKI’S COMMERCIAL MORTGAGE DEFAULT STUDY1

Default Rates from Esaki’s Study (%) 0.2 1.0 1.7 2.0 2.1 2.3 2.8 1.5 0.9 0.5

Total Defaults That Occur Each Year (%) 1.5 6.9 11.5 13.1 13.9 15.2 18.9 9.7 5.7 3.6

Cumulative Level of Defaults Through Time (%) 1.5 8.4 19.9 33.0 46.9 62.1 81.0 90.7 96.4 100.0

1

Real Estate Finance, Commercial Mortgage Defaults: 1972-2000, by Howard Esaki, Winter 2002 Edition.

Source: Morgan Stanley

The timing of defaults presented in Exhibit 6 is fairly evenly distributed, with 47% of loan losses occurring in the first five years after loan origination, and the remaining 53% of losses occurring in years 6 through 10. If the timing of defaults in CMBS deviates from this average, it is possible that our analysis would yield different results.

Please see additional important disclosures at the end of this report.

133

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels We repeated our analysis, applying the default timing curve experienced by life insurance company loans that were originated in 1986. The 1986 origination cohort experienced the highest default rate of any origination year between 1972 and 1995. For the purposes of this analysis, we do not focus on the absolute level of defaults, but we apply the 1986 default timing curve as a proxy for the timing of CMBS loan losses in a real estate downturn. Life insurance company loans originated in 1986 experienced 70% of all defaults in years 6 through 10 after origination. Since most of the CMBS transactions in our study have seasoned for less than six years, the 1986 default timing curve assumes that the majority of losses have not yet occurred. Using this back-loaded timing curve, the projected future losses for these CMBS transactions are higher than in our initial analysis, resulting in lower subordination-to-loss ratios.

134

exhibit 7

Year 1 2 3 4 5 6 7 8 9 10

TIMING OF DEFAULTS FOR 1986 COHORT FROM ESAKI’S COMMERCIAL MORTGAGE DEFAULT STUDY1

Default Rates from Esaki’s Study (%) 0.0 0.9 2.3 2.1 3.3 5.8 10.3 2.5 1.3 0.2

Total Defaults That Occur Each Year (%) 0.0 3.2 7.9 7.5 11.3 20.2 35.8 8.6 4.6 0.8

Cumulative Level of Defaults Through Time (%) 0.0 3.2 11.1 18.6 29.9 50.2 86.0 94.6 99.2 100.0

1

Real Estate Finance, Commercial Mortgage Defaults: 1972-2000, by Howard Esaki, Winter 2002 Edition.

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

135

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 8

Deal Name DLJCM 2000-CKP1 BACM 2001-PB1 CCMSC 2000-1 JPMCC 2001-C1 BACM 2001-1 BAFU 2001-3 SBM7 2000-C2 SBM7 2001-C1 GMACC 2001-C1 JPMC 2000-C9 FUNBC 2001-C4 GECMC 2000-1 SBM7 2000-C3 GMACC 2000-C2 MLMI 1999-C1 MSDWC 2001-TOP1 CSFB 2001-CF2 BSCMS 2001-TOP2 GMACC 2000-C3 LBUBS 2000-C5 CSFB 2001-CKN5 JPMC 1999-C8 GECMC 2001-3 CSFB 1999-C1 KEY 2000-C1 JPMCC 2001-CIBC MLMI 1998-C3 SBM7 2001-C2 COMM 2000-C1 GMACC 2000-C1 CSFB 2000-C1 FUNBC 2000-C2 GECMC 2001-2 LBUBS 2000-C3 FUBOA 2001-C1

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

B1 J F G H J G G G F J F G F E F G F F G H F G F G F D H F F F G G G H

Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3

BBB+ BBBBBBBBBBBBBBBBBBBBBBBBBBBBBB-* BBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBB-

0.00 0.00 0.84 0.18 0.35 0.00 0.00 0.13 0.77 0.00 0.00 1.69 2.41 1.18 0.40 0.00 0.11 0.00 0.33 0.00 0.00 0.67 0.00 0.00 0.00 0.00 0.63 1.44 0.00 0.00 0.00 0.00 0.00 0.74 1.14

0.02 0.00 1.64 0.00 0.00 0.00 0.00 0.00 1.25 0.24 0.00 1.37 0.11 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.14 0.69 0.00 0.50 0.00 0.00

0.08 1.71 7.95 1.59 4.83 0.54 0.00 3.62 3.98 0.78 0.48 2.41 0.00 3.16 6.26 2.86 1.27 2.65 0.86 2.40 0.00 1.69 0.00 0.10 1.11 1.31 1.88 0.79 0.00 4.56 0.83 0.00 0.00 1.28 0.00

0.25 0.00 1.21 0.00 0.11 0.00 5.46 0.00 0.00 0.00 1.03 0.00 0.00 0.00 1.71 0.00 0.00 0.00 0.00 0.00 0.50 0.00 0.85 0.58 0.74 0.00 0.00 0.00 1.48 0.00 0.00 1.21 0.00 0.16 0.00

0.20 0.00 7.45 1.87 1.05 0.00 1.93 1.77 1.66 0.00 0.33 0.00 1.63 0.00 2.69 0.00 0.33 0.00 0.54 0.47 0.00 0.30 0.00 2.91 3.08 0.57 0.00 0.00 0.00 0.50 2.05 0.07 0.50 0.08 0.00

Source: Morgan Stanley, Trepp * S&P Rating

136

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE)

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.00 0.00 0.00 0.00 0.40 0.00 2.67 4.57 0.00 0.00 0.09 0.00 0.00 0.00 1.12 0.80 0.06 2.45 0.00 0.65 0.00 0.75 0.00 8.63 0.00 4.81 0.00 0.00 0.35 0.73 0.56 0.53 0.00 1.12 3.53

25.88 12.35 43.93 11.34 25.85 12.76 25.66 14.68 24.00 37.22 19.13 29.26 6.58 27.93 13.68 13.34 19.47 12.86 30.38 21.28 17.34 17.69 19.98 17.57 16.12 26.45 31.19 16.31 47.64 19.01 27.67 22.56 29.92 25.53 22.94

0.4 0.4 4.8 1.0 1.5 0.2 2.5 1.4 1.6 0.4 0.7 0.8 0.7 0.8 2.6 0.6 0.5 0.6 0.6 0.8 0.3 0.5 0.4 1.4 1.5 0.7 0.5 0.3 0.8 1.2 1.1 0.6 0.5 0.6 0.3

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BBBSubordination (or lowest investment grade class subordination)

Ratio of BBBSubordination to Future Expected Loss

2.5 0.6 0.0 0.0 0.1 0.6 0.6 0.1 0.0 1.8 0.0 0.5 0.4 1.0 0.0 0.0 0.4 0.0 0.2 0.0 0.1 2.5 0.0 0.4 0.2 0.1 2.3 0.0 0.3 0.0 0.0 0.0 0.1 0.2 0.2

6.3 4.4 5.6 4.2 3.4 3.3 3.9 3.2 3.3 2.9 2.9 2.9 2.3 2.1 3.2 1.3 2.0 1.3 1.6 1.6 1.7 2.1 1.9 2.1 2.1 1.6 2.1 1.4 1.4 1.5 1.3 1.3 1.2 0.9 1.1

9.8 8.0 10.3 7.7 8.7 8.5 10.0 8.9 9.2 8.9 9.3 9.2 8.1 7.9 12.4 5.2 7.8 5.3 7.4 7.4 8.1 9.8 9.2 10.8 10.8 9.1 12.6 8.6 9.2 9.8 9.1 9.3 8.9 6.7 8.5

1.6 1.8 1.8 1.8 2.5 2.6 2.6 2.7 2.8 3.1 3.2 3.2 3.5 3.7 3.9 3.9 3.9 4.0 4.5 4.5 4.7 4.8 5.0 5.0 5.2 5.5 6.0 6.4 6.7 6.7 6.8 7.2 7.2 7.5 7.7

Please see additional important disclosures at the end of this report.

137

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 8

Deal Name FUNBC 1999-C4 PMCF 2001-ROCK PMAC 1999-C1 BSCMS 2001-TOP4 MSDWC 2000-LIFE HFCMC 2000-PH1 GSMS 1999-C1 GMACC 1999-C3 FUNBC 2001-C2 BACM 2000-2 BSCMS 2000-WF2 LBUBS 2001-C2 PNCMA 2000-C1 JPMC 2000-C10 MSC 1999-FNV1 FUNBC 2000-C1 SBM7 2000-C1 BSCMS 1999-C1 CCMSC 2000-3 DLJMA 1997-CF1 LBCMT 1999-C2 LBUBS 2000-C4 LBCMT 1999-C1 CSFB 1997-C2 MSC 1998-HF2 MSC 1998-CF1 DLJCM 2000-CF1 NLFC 1998-2 PNCMA 1999-CM1 BSCMS 1999-WF2 CSFB 2001-CK6 GMACC 2001-C2 GMACC 1999-C2 CSFB 1998-C2 GMACC 1999-C1

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

F G E F F F E F J H F G F F F F G E F A3 F G E E F C B2 E B2 F H H G E E

Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 A2 Baa3 Baa3 Baa3 Baa1 -

BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBA BBBBBBBBBBBBBBBBBB BBBBBBBBBBBB

1.31 0.00 0.33 0.00 0.00 3.47 1.09 5.88 0.00 0.00 0.00 0.17 1.31 1.05 3.07 0.24 1.68 0.00 0.00 0.00 0.00 0.00 0.41 0.94 0.00 0.98 3.35 0.39 0.17 0.00 0.02 0.00 0.00 0.00 0.23

0.00 0.00 0.00 0.23 0.00 0.10 2.21 0.22 0.00 0.00 0.00 0.00 1.43 0.55 0.00 0.00 0.00 0.00 0.35 0.26 0.00 0.84 0.00 5.98 0.00 2.82 1.98 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

2.50 0.00 0.84 0.26 1.55 0.72 2.68 1.34 0.22 0.00 2.01 0.53 1.27 0.55 0.36 0.00 1.82 1.37 1.43 4.38 0.35 0.18 0.24 0.00 0.82 0.74 0.43 1.32 0.00 1.68 0.07 0.56 5.66 0.00 2.62

0.00 0.00 1.63 0.00 0.00 0.00 0.51 0.00 0.00 0.00 0.00 0.00 0.00 0.38 1.58 0.00 0.00 0.00 0.50 1.93 2.06 0.08 0.00 0.00 0.00 3.91 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.85 0.00

1.21 0.00 1.19 0.00 0.00 0.29 0.65 1.32 0.32 0.17 0.00 0.00 0.12 0.00 2.59 0.00 0.64 0.00 0.00 1.00 0.00 0.52 0.00 1.25 0.48 4.25 0.12 0.58 0.00 0.49 0.00 0.00 0.00 0.00 0.00

Source: Morgan Stanley, Trepp

138

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

2.18 0.49 1.58 0.00 0.33 1.86 0.31 0.00 0.75 0.00 0.49 0.00 0.70 1.38 0.00 1.86 0.81 0.00 0.65 2.91 0.00 0.75 2.11 0.82 4.14 3.63 2.59 1.01 1.09 0.00 0.70 0.00 0.00 4.11 0.11

23.17 5.56 17.47 10.07 15.90 18.35 25.07 16.49 27.91 35.16 12.49 30.31 20.83 10.43 21.24 33.46 16.40 14.10 25.74 0.00 22.15 25.33 33.21 18.98 27.21 19.52 20.83 8.22 16.95 4.96 14.07 25.68 12.30 21.05 21.46

1.1 0.1 1.1 0.1 0.4 0.5 1.2 1.0 0.4 0.3 0.5 0.3 0.6 0.4 1.6 0.3 0.7 0.3 0.7 1.2 0.9 0.5 0.3 1.0 0.5 2.8 0.6 0.5 0.2 0.5 0.1 0.3 1.1 0.5 0.6

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BBBSubordination (or lowest investment grade class subordination)

Ratio of BBBSubordination to Future Expected Loss

0.0 0.4 0.9 0.0 0.4 0.5 0.8 0.7 0.0 0.1 0.0 0.0 0.2 0.3 0.1 0.4 0.0 0.6 0.0 8.1 0.0 0.0 0.4 2.1 0.5 2.5 0.0 0.8 0.5 0.4 0.0 0.0 0.0 0.8 0.8

1.3 1.0 1.4 0.6 1.0 1.2 1.4 1.1 0.8 1.0 0.6 0.7 0.9 0.9 1.1 0.9 0.9 0.7 0.8 1.5 0.6 0.6 0.5 0.8 0.7 1.5 0.7 0.9 0.8 0.6 0.6 0.7 0.7 0.9 0.9

10.4 8.0 11.8 5.2 8.1 10.8 13.1 10.5 8.0 9.8 5.7 7.6 9.5 9.6 12.6 10.1 10.5 8.1 9.9 18.7 7.4 8.0 6.6 11.2 10.0 20.5 10.0 12.4 11.0 8.0 8.4 9.4 10.6 12.3 13.7

7.9 8.1 8.2 8.3 8.4 9.1 9.6 9.8 9.8 9.9 10.2 10.3 10.4 10.6 11.0 11.4 12.1 12.4 12.6 12.6 12.6 13.1 13.2 13.3 13.5 13.6 13.8 13.8 13.9 14.0 14.3 14.3 14.3 14.4 14.7

Please see additional important disclosures at the end of this report.

139

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 8

Deal Name CSFB 2001-CK1 MSDWC 2001-TOP5 FUNCM 1999-C2 BSCMS 2000-WF1 GECMC 2001-1 JPMCC 2001-CIB2 DLJCM 1999-CG1 LBUBS 2001-C7 CSFB 2001-CK3 CMAT 1999-C1 DLJCM 1999-CG2 CMAT 1999-C2 FUNBC 2001-C3 COMM 1999-1 MSDWC 2000-LIF2 MLMI 1998-C2 MSDWC 2001-TOP3 MSC 1999-LIFE1 MCFI 1998-MC3 CSFB 1998-C1 CCMSC 1999-2 PSSF 1999-C2 DLJCM 1998-CF2 LBUBS 2001-C3 GMACC 1997-C2 CSFB 2001-4 PSSF 1999-NRF1 FUCMT 1999-C1 DLJCM 1999-CG3 MCFI 1998-MC2 DMARC 1998-C1 GSMS 1998-C1 CCMSC 2000-2 BACM 2000-1 CCMSC 1998-2

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

G F F F G F B2 H G1 E B2 F H F F E F G E E F F B-2 G D G E E B2 E E E F F E

Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa1 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 -

BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB+ BBBBBB BBBBBBBBBBBBBBB-

0.12 0.00 0.00 0.00 0.00 0.00 0.00 0.28 0.00 0.00 1.35 0.00 0.16 0.00 0.00 0.56 0.27 0.00 1.06 0.00 2.65 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.79 0.00 0.48 0.00 0.86 3.19 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.94 0.00 0.10 0.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.14 0.00 0.00 0.00 0.63 0.00 0.00 0.00

0.00 0.00 1.33 0.45 0.20 0.51 0.00 0.10 0.00 1.41 0.72 0.67 0.18 0.33 0.00 1.34 0.41 0.00 0.30 1.27 0.00 0.00 0.77 0.00 2.04 0.00 0.00 0.00 0.37 0.00 0.40 1.33 0.00 0.97 0.69

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.25 0.00 0.00 0.00 1.76 0.00 0.95 0.00 0.93 0.00 0.00 0.68 0.00 0.48 0.24 1.10 0.00 1.61 1.15 0.00 0.00 0.00

0.21 0.00 0.00 0.46 0.00 0.00 1.22 0.00 0.00 1.10 1.84 0.00 0.00 1.19 0.00 1.52 0.00 0.00 1.39 0.97 0.00 0.57 0.00 0.00 3.73 0.00 0.00 0.33 0.00 0.47 4.53 1.50 0.00 0.00 0.00

Source: Morgan Stanley, Trepp

140

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.15 0.00 0.64 1.52 1.42 0.62 0.26 0.00 0.76 1.13 0.62 0.30 1.36 3.17 0.00 2.90 0.00 0.00 0.15 5.21 6.92 0.00 1.34 0.00 2.95 2.17 0.21 1.59 2.61 0.80 5.08 0.58 0.00 2.37 0.00

27.57 10.01 32.57 8.74 26.57 15.05 0.00 6.40 24.75 15.61 26.24 10.07 24.65 26.77 19.88 0.00 6.60 9.56 26.88 31.63 31.85 21.36 27.45 18.08 32.86 14.03 21.65 19.83 31.39 25.57 16.14 17.22 43.17 8.68 17.37

0.3 0.1 0.5 0.3 0.3 0.2 0.5 0.1 0.2 0.8 1.0 0.2 0.3 1.1 0.2 0.6 0.1 0.6 0.7 1.1 0.5 0.6 0.4 0.2 1.7 0.2 0.3 0.4 0.7 0.3 2.0 1.2 0.4 0.4 0.3

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BBBSubordination (or lowest investment grade class subordination)

Ratio of BBBSubordination to Future Expected Loss

0.0 0.0 0.6 0.1 0.0 0.0 0.6 0.0 0.0 0.2 0.1 0.4 0.0 0.0 0.0 1.5 0.0 0.0 0.3 1.2 0.0 0.3 0.5 0.0 1.5 0.0 0.5 0.5 0.1 1.2 0.6 0.9 0.0 0.1 0.4

0.6 0.3 0.7 0.5 0.6 0.5 0.7 0.3 0.5 0.7 0.7 0.7 0.6 0.7 0.3 0.6 0.3 0.4 0.7 0.6 0.5 0.6 0.6 0.3 1.0 0.4 0.6 0.6 0.5 0.4 0.7 0.6 0.4 0.5 0.5

9.5 5.2 11.0 7.4 9.0 8.5 10.8 5.5 8.0 11.5 11.5 11.5 9.6 13.0 6.2 11.7 5.3 7.9 12.8 11.7 10.6 12.4 11.9 6.3 19.7 8.2 12.3 11.5 10.9 8.9 15.2 12.3 9.5 11.6 10.3

14.9 15.0 15.1 15.3 15.4 15.6 15.8 16.5 16.7 17.0 17.1 17.2 17.3 17.4 18.0 18.1 18.4 18.9 19.0 19.2 19.5 19.7 19.8 20.1 20.1 20.3 20.4 20.5 20.7 20.7 20.7 21.2 21.2 21.9 21.9

Please see additional important disclosures at the end of this report.

141

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 8

Deal Name AMC 1998-C1 PNCMA 2000-C2 FULBA 1998-C2 CASC 1998-D7 MLMI 1997-C2 FULB 1997-C2 MCFI 1998-MC1 NLFC 1999-1 HFCMC 1999-PH1 JPMC 1998-C6 MSC 1999-RM1 LBCMT 1998-C4 NLFC 1998-1 LBCMT 1998-C1 SBM7 1999-C1 JPMCC 2001-CIB3 JPMC 1999-C7 CCMSC 1997-2 MSC 1998-HF1 JPMC 1997-C5 MLMI 1997-C1 PNCMA 2001-C1 SBM7 2001-MMA CMFUN 1999-1 MSC 1997-C1 MSC 1998-WF1 GMACC 1998-C1 DLJMA 1997-CF2 GMACC 1997-C1 CCMSC 1998-1 PSSF 1998-C1 CMAC 1998-C2 MCFI 1997-MC1 FULB 1997-C1 BSCMS 1998-C1

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

E G F A-5 E E G E G E F E E E F F E E E E E G E1-E8 F G E F B2 F E F E E F E

Baa3 Baa3 Ba1 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa1 Baa3 Baa3 Baa3 Baa3 Baa2 Ba3 Baa3 Baa3 Baa3 Baa3 Baa3 Baa3

BBB BBB-* BBBBBB BBB BBBBBBBBB BBBBBBBBBBBBABBB+ BBBBBB BBB BBBBBB+ BBBBBB BBB-

0.00 0.00 0.56 0.27 0.71 1.16 0.00 0.00 0.00 0.76 0.54 0.00 0.75 0.76 0.00 0.00 0.00 0.56 0.00 1.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.84 0.00 0.87 0.00

0.00 0.00 0.00 0.00 0.40 0.00 0.00 0.00 0.14 0.00 0.00 0.51 0.00 1.07 0.00 0.00 0.00 0.00 1.02 1.45 1.12 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.16 3.68 0.23 0.00

1.24 0.19 0.00 1.20 3.11 1.39 1.80 0.00 1.68 8.09 0.00 0.13 3.98 0.64 2.64 0.00 0.00 4.11 0.20 1.73 11.91 0.00 0.00 0.22 0.00 4.41 1.18 15.24 0.22 0.00 0.24 3.13 3.62 1.58 0.49

0.00 0.00 0.66 2.30 1.15 0.18 0.00 0.45 0.00 0.00 0.00 0.35 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.77 0.00 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 1.32 0.00

0.96 0.00 0.74 0.48 0.00 1.92 0.00 0.00 0.00 0.00 0.56 0.70 0.00 0.15 0.00 0.00 1.11 0.00 0.00 0.00 1.97 0.00 0.00 0.00 3.87 0.00 0.00 0.00 0.80 0.00 0.00 0.35 1.94 0.69 1.31

Source: Morgan Stanley, Trepp * S&P Rating

142

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.27 0.00 5.53 0.13 1.69 2.76 0.00 0.75 1.47 0.00 0.00 0.90 0.63 0.53 0.00 0.00 0.00 0.95 0.91 0.00 0.00 0.84 0.00 0.00 0.00 0.27 17.21 0.94 2.86 0.00 0.50 1.08 2.11 2.53 1.46

22.13 20.03 18.54 12.26 31.49 20.40 26.50 26.10 18.41 18.96 19.60 16.37 16.89 32.99 16.88 6.54 27.41 23.29 15.48 20.57 28.57 8.68 13.76 18.61 26.25 26.28 32.87 21.47 24.29 24.60 14.07 11.93 36.31 25.84 17.87

0.6 0.2 0.6 1.1 1.1 0.9 0.5 0.3 0.5 1.5 0.3 0.5 0.7 0.5 0.6 0.1 0.5 0.7 0.2 0.4 1.8 0.1 0.1 0.3 0.8 0.8 0.7 2.4 0.4 0.2 0.2 0.8 1.4 0.8 0.7

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BBBSubordination (or lowest investment grade class subordination)

Ratio of BBBSubordination to Future Expected Loss

1.4 0.0 0.7 0.6 0.7 0.8 1.3 0.3 0.1 0.0 0.3 0.0 0.8 1.0 0.0 0.0 0.0 0.5 1.0 2.0 0.9 0.0 0.0 0.2 0.8 0.2 0.3 0.2 1.8 0.7 0.6 0.3 0.7 0.2 0.0

0.6 0.4 0.4 0.5 0.5 0.6 0.5 0.5 0.4 0.4 0.4 0.4 0.5 0.4 0.4 0.2 0.4 0.4 0.3 0.4 0.5 0.2 0.2 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.2 0.3 0.3 0.2 0.2

14.2 10.1 9.1 11.7 12.6 13.7 12.7 13.2 11.3 11.5 12.2 10.5 14.3 13.7 12.8 7.4 13.1 14.2 12.6 14.7 20.7 7.6 9.3 10.6 12.3 12.3 11.1 14.1 13.9 11.0 10.5 12.9 15.7 9.0 9.0

23.2 24.1 24.2 24.7 24.7 24.8 27.2 27.2 28.2 28.9 29.4 29.6 31.1 31.6 33.0 33.2 35.2 36.5 36.7 37.1 38.2 38.6 39.0 39.1 39.3 39.6 40.7 41.1 43.1 43.2 43.4 46.4 50.6 52.1 52.1

Please see additional important disclosures at the end of this report.

143

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 8

Deal Name DLJCM 1998-CG1 GMACC 1998-C2 ACMF 1997-C1 CSFB 1997-C1 CCMSC 1997-1 ASC 1997-D5 ASC 1997-D4 MCFI 1997-MC2 CMAC 1998-C1 MSC 1997-HF1 NLFC 1999-2 CMAC 1999-C1 DLJCM 1998-CF1 NASC 1998-D6 MSC 1999-WF1 MSC 1998-WF2 JPMC 1997-C4 MSC 1997-WF1 IFUND 2001-A

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

B3 E G E E A2 B-1 E E F E G B2 A-5 G G F E D

Baa1 Baa3 Baa3 Baa3 Baa2 Baa3 Baa3 Baa3 Ba1 A2 -

BBB BBBBBBBBB+ AA BBBBBB-* BBBABBB+ BBB BBBBBB-* BBB*

0.00 0.78 0.00 0.00 0.00 0.12 0.16 0.60 0.64 1.96 0.00 0.41 0.00 0.06 0.53 0.00 0.00 0.00 0.00

0.44 0.49 0.00 0.00 0.00 0.21 0.00 0.42 0.00 0.00 0.00 0.88 0.72 0.05 0.00 0.00 0.00 0.00 0.00

0.44 1.02 6.74 3.28 0.96 0.00 2.98 0.98 0.36 1.44 0.00 0.00 0.00 1.33 0.00 0.00 0.43 0.00 0.00

0.00 0.09 0.00 0.18 0.00 2.86 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.39 0.00 0.00 0.00 4.15 1.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Source: Morgan Stanley, Trepp * S&P Rating

144

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

1.90 0.21 2.74 8.78 0.00 0.61 9.61 2.20 0.16 0.00 1.44 1.49 0.93 1.29 0.00 0.00 0.00 2.11 0.00

25.98 21.97 33.56 19.64 14.91 15.00 7.56 14.33 20.31 25.93 18.44 15.67 16.98 16.44 2.13 11.18 0.00 5.17 0.00

0.4 0.4 0.9 0.7 1.2 1.5 0.6 0.3 0.2 0.3 0.1 0.2 0.2 0.4 0.0 0.1 0.0 0.1 0.0

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BBBSubordination (or lowest investment grade class subordination)

Ratio of BBBSubordination to Future Expected Loss

0.3 0.3 0.0 0.9 0.4 1.3 0.6 0.4 0.5 0.4 0.0 0.0 0.3 0.1 0.1 0.0 0.0 0.0 0.0

0.2 0.2 0.2 0.2 0.2 0.3 0.2 0.2 0.2 0.1 0.2 0.2 0.1 0.1 0.1 0.0 0.0 0.0 0.0

10.8 11.3 9.3 13.4 14.4 20.2 10.0 13.5 12.9 8.6 14.4 10.8 11.6 11.3 7.9 6.1 13.1 17.0 12.5

52.3 56.8 57.0 59.0 60.2 60.8 61.5 63.4 63.5 66.5 70.2 71.4 90.9 97.1 116.5 271.9 1581.4 2072.7 NA

Please see additional important disclosures at the end of this report.

145

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels Section II: Projecting Losses on BB CMBS In Section I, we ranked BBB- CMBS bonds based on their level of principal protection from losses. In this section, we apply the same methodology from Part I to rank the universe of BB conduit CMBS. We measure the level of principal protection on a bond by comparing the future expected losses for the deal’s collateral to the current credit for the bond. The rankings by subordination to expected loss ratio are provided in Exhibit 13 of this section. RATIOS A ND R ANKINGS

We believe that the bonds at the bottom of the list (with the highest ratios) have the highest level of principal protection and the greatest chance for rating agency upgrade. Likewise, we believe that the bonds at the top of the list (with the lowest ratios) have a higher chance of default and downgrade. A ratio that is less than 1 means that we expect the class to default before maturity; a ratio of greater than 1 means we expect no default. Depending on the severity assumption and timing curve used to project future losses, between 2 and 15 bonds have subordination-to-loss ratios of less than 1 (See Exhibit 9). Between 9 and 21 bonds have a ratio between 1 and 2. We believe that classes with a ratio between 1 and 2 may be vulnerable to a downgrade.

exhibit 9

BB SUBORDINATION-TO-EXPECTED-LOSS RATIOS FOR 1997-2001 CONDUITS

Ranges for BB Subordination to Expected Loss Ratio 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 >8

Average Default Timing Curve; 34% Severity # of Deals 2 9 8 13 4 13 10 7 85

Source: Morgan Stanley, Trepp, Bloomberg

146

% of Deals 1.3 6.0 5.3 8.6 2.6 8.6 6.6 4.6 56.3

Average Default Timing Curve; 43% Severity # of Deals 2 12 11 10 12 13 7 12 72

% of Deals 1.3 7.9 7.3 6.6 7.9 8.6 4.6 7.9 47.7

SCOPE O F T HE S TUDY

This study analyzes and ranks bonds based on the level of principal protection from losses. Our analysis does not assess the risk of potential downgrades that may occur for other reasons, such as interest shortfalls. For example, while ASC 1997-D5 A5 has a high subordination to expected loss ratio,2 this class was downgraded by Fitch in September 2003 to BB from BBB. The downgrade was prompted by interest shortfalls caused by servicer reimbursements of non-recoverable advances. S&P also downgraded this tranche to D from BBB. PROJECTING L OSSES - M AY 2 003

In order to assess the strength of this study as a predictor of rating actions, we reviewed rating changes on the 112 BB bonds from the prior version of this study.3 Over the course of a year, we would expect the bonds with the lowest subordination-to-loss ratios to experience the greatest number of downgrades. Likewise, we would expect the bonds with the highest subordination-to-loss ratios to experience the greatest number of upgrades.

1986 Default Timing Curve; 34% Severity # of Deals 12 18 16 15 13 5 6 1 65

% of Deals 7.9 11.9 10.6 9.9 8.6 3.3 4.0 0.7 43.0

1986 Default Timing Curve; 43% Severity # of Deals 15 21 20 16 9 3 3 3 61

% of Deals 9.9 13.9 13.2 10.6 6.0 2.0 2.0 2.0 40.4

2

Ratio of 35.8, assuming average default timing curve from Esaki study and 34% severity rate.

3

See Projecting Losses: Are BBs Safe? May 2003.

Please see additional important disclosures at the end of this report.

147

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels We found this to be true when we compared the number of rating actions on the top half of the ranked list of bonds to the rating actions on the bottom half of the ranked list. The 56 bonds with the highest ratios experienced 13 upgrades and 5 downgrades, while the 56 bonds with the lowest ratios experienced 13 downgrades and 1 upgrade. We also found that the rating actions were concentrated at the extremes of the list. The 20 bonds with the highest subordination-to-loss ratios experienced the majority of the upgrades (8), and the 20 bonds with the lowest subordination-toloss ratios experienced half of the downgrades (9). METHODOLOGY

In this section, we analyze non-investment grade classes of 151 conduit transactions tracked by Trepp. Each conduit transaction was issued between January 1997 and December 2001. 142 of 151 classes are currently rated BB by Moody’s, Fitch or S&P; the other 9 classes are rated BB+ or BB-. For each transaction, we calculate a future expected loss rate based on historical losses and losses implied by the current level of 30, 60, and 90-day delinquencies,

exhibit 10

PROJECTING LOSSES STUDY - MAY 2003 # OF UPGRADES AND DOWNGRADES ON 112 BB BONDS1

Rankings2 1-56 57-112

# of Classes Upgraded 13 1

# of Classes Downgraded 5 13

1

Since May 2003.

2

Highest Ratio: Rank of 1; Lowest Ratio: Rank of 112

Source: Morgan Stanley, Trepp, Bloomberg

REO, foreclosures, performing specially serviced loans and servicer watchlisted loans. We do not for the impact of defeased collateral within the deals. We begin by asg a probability of liquidation for each delinquency category, based on results from Esaki’s commercial mortgage default study.

3

See Projecting Losses: Are BBs Safe? May 2003.

148

Assumptions for Probabilities of Liquidation

In Esaki’s study, about 55% of loans that became more than 90 days delinquent were ultimately liquidated. For this section, we borrow the results from the Esaki study and assume that 55% of 90+ day delinquent loans will be liquidated. We then assume that 60-day delinquent loans will have a 25% liquidation rate, and 30-day delinquent loans will have a 10% liquidation rate. To be conservative, we assume that all foreclosed and REO loans will be liquidated. We assign a 2.5% liquidation rate to servicer watchlisted loans and a 5% liquidation rate to performing specially serviced loans. We assume a low probability of liquidation for the performing specially serviced loans because these loans could have been transferred to the special servicer for technical rather than monetary defaults. Calculating Loan Losses: Average Timing of Defaults

We then multiply the projected liquidated loan total by a severity rate to compute each deal’s expected loss rate for the year. For our analysis, we consider two loss severity rates: 34% and 43%. The first severity rate that we assume is 34%. Liquidated life insurance company loans in the Commercial Mortgage Default Study: 1972-2000, by Howard Esaki, experienced a 34% severity rate. The average severity rate on liquidated CMBS loans is higher, at 43%4. The calculated loss rates for the year are then used in conjunction with historical losses to project future loss rates. Since the calculated losses are based on current delinquency levels, we assume that the resulting losses will occur within one year from today. For example, with a 34% severity rate assumption, DLJCM 2000-CKP1, is expected to have a 0.3% loss rate based on current levels of delinquencies, specially serviced loans and watchlisted loans. This transaction has seasoned for 4 years, so we assume losses will occur in year 5. The projected loss based on current delinquencies (0.3%) is then added to the deal’s historical losses of 2.5%. To project future losses, we assume that the timing of CMBS loan losses in this study mirrors the average timing of defaults on the life insurance company loans from Esaki’s study. According to the timing of defaults presented in Esaki’s study, about 47% of loans default by the end of year 5. The DLJCM 2000-CKP1 transaction is expected to have a 2.8% loss rate by year 5, based on historical losses and projected losses of currently delinquent loans. If we apply the results of Esaki’s study, this 2.8% loss rate should correspond to 47% of total loan losses. Therefore, we conclude that this transaction may suffer from a 3.2% loss rate (based on original balances) during its remaining lifetime, since 53% of loan losses have yet to occur. (2.8%*(1-47%)¸47%=3.2% loss rate) A 3.2% loss rate based on original balances corresponds to a 3.5% loss rate, based on current balances.

4

See Special Servicer Severities, February 6, 2004.

Please see additional important disclosures at the end of this report.

149

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels

exhibit 11

Year 1 2 3 4 5 6 7 8 9 10

AVERAGE TIMING OF DEFAULTS FROM ESAKI’S COMMERCIAL MORTGAGE DEFAULT STUDY1

Default Rates from Esaki’s Study (%) 0.2 1.0 1.7 2.0 2.1 2.3 2.8 1.5 0.9 0.5

Total Defaults That Occur Each Year (%) 1.5 6.9 11.5 13.1 13.9 15.2 18.9 9.7 5.7 3.6

Cumulative Level of Defaults Through Time (%) 1.5 8.4 19.9 33.0 46.9 62.1 81.0 90.7 96.4 100.0

1

Real Estate Finance, Commercial Mortgage Defaults: 1972-2000, by Howard Esaki, Winter 2002 Edition.

Source: Morgan Stanley

The timing of defaults presented in Exhibit 11 is fairly evenly distributed, with 47% of loan losses occurring in the first five years after loan origination, and the remaining 53% of losses occurring in years 6 through 10. If the timing of defaults in CMBS deviates from this average, it is possible that our analysis would yield different results. We repeated our analysis, applying the default timing curve experienced by life insurance company loans that were originated in 1986. The 1986 origination cohort experienced the highest default rate of any origination year between 1972 and 1995. For the purposes of this analysis, we do not focus on the absolute level of defaults, but we apply the 1986 default timing curve as a proxy for the timing of CMBS loan losses in a real estate downturn. Life insurance company loans originated in 1986 experienced 70% of all defaults in years 6 through 10 after origination. Since most of the CMBS transactions in our study have seasoned for less than 6 years, the 1986 default timing curve assumes that the majority of losses have not yet occurred. Using this back-loaded timing curve, the projected future losses for these CMBS transactions are higher than in our initial analysis, resulting in lower subordination-to-loss ratios.

150

exhibit 12

Year 1 2 3 4 5 6 7 8 9 10

TIMING OF DEFAULTS FOR 1986 COHORT FROM ESAKI’S COMMERCIAL MORTGAGE DEFAULT STUDY1

Default Rates from Esaki’s Study (%) 0.0 0.9 2.3 2.1 3.3 5.8 10.3 2.5 1.3 0.2

Total Defaults That Occur Each Year (%) 0.0 3.2 7.9 7.5 11.3 20.2 35.8 8.6 4.6 0.8

Cumulative Level of Defaults Through Time (%) 0.0 3.2 11.1 18.6 29.9 50.2 86.0 94.6 99.2 100.0

1

Real Estate Finance, Commercial Mortgage Defaults: 1972-2000, by Howard Esaki, Winter 2002 Edition

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

151

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 13

Deal Name DLJCM 2000-CKP1 JPMCC 2001-C1 JPMC 2000-C9 SBM7 2000-C3 SBM7 2000-C2 BACM 2001-1 SBM7 2001-C1 GECMC 2000-1 FUNBC 2001-C4 MSDWC 2001-TOP1 GMACC 2001-C1 GMACC 2000-C2 GMACC 2000-C3 BSCMS 2001-TOP2 CSFB 2001-CF2 BACM 2001-PB1 KEY 2000-C1 PNCMA 2001-C1 BAFU 2001-3 GMACC 2000-C1 LBUBS 2000-C5 FUBOA 2001-C1 COMM 2000-C1 GMACC 1999-C3 LBUBS 2000-C3 FUNBC 1999-C4 JPMC 1999-C8 GECMC 2001-2 PMCF 2001-ROCK CSFB 2000-C1 MSDWC 2000-LIFE JPMCC 2001-CIBC PNCMA 2000-C1 MSC 1998-CF1 CSFB 1999-C1

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

B5 J H J J K J H L H J G J H J L J J L H H L G G J H G I J H J H H D H

B3 B1 Ba3 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba1 Ba2 Ba2 Ba2 Ba3 Ba2 Ba2 Ba2 Ba2 -

BB BB BB BB* BB BB BBBB BB BB* BB BB BB BB BB BB* BB BB BB BB BB+ BB BB BB BB BB BB BB

0.00 0.10 0.00 2.43 0.00 0.35 0.40 0.00 0.00 0.00 1.20 5.02 0.17 0.00 0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.29 0.00 0.35 0.00 0.00 0.64 0.00 0.00 0.19 0.60 0.67 0.29 1.37 0.13

0.00 0.00 0.00 2.14 0.00 0.00 0.00 1.37 0.00 0.36 0.00 0.00 0.00 0.00 0.07 0.00 0.00 0.00 0.00 1.46 0.00 0.17 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.85 0.10 1.51

0.10 0.54 0.46 0.00 2.14 4.66 3.62 2.41 0.00 2.50 5.32 3.16 0.86 2.67 1.08 0.00 1.11 0.00 0.54 5.18 2.40 0.69 0.00 7.18 0.44 0.17 1.69 0.50 0.00 1.33 0.38 1.31 2.69 0.92 0.10

0.25 0.00 0.56 0.00 4.01 0.11 0.00 0.00 1.51 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.74 0.00 0.00 0.00 0.00 0.00 0.98 0.00 0.16 1.46 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.07 0.58

0.20 3.10 0.00 1.63 1.93 1.05 1.77 0.00 0.33 0.00 0.00 0.00 0.54 0.00 0.21 0.00 3.09 0.00 0.00 0.50 0.47 0.00 0.00 0.00 0.92 1.25 0.30 0.50 0.00 2.05 1.19 0.57 0.36 0.26 2.91

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

152

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.00 0.00 0.66 0.00 1.97 0.85 4.57 1.69 0.09 1.49 0.00 0.00 0.00 1.72 0.01 1.24 0.00 0.84 0.00 1.20 0.65 3.53 0.36 0.00 1.12 1.02 0.67 0.00 0.25 0.56 0.00 6.50 0.83 6.11 7.00

23.89 12.79 35.79 11.36 25.19 28.86 15.74 29.03 19.70 23.71 29.06 24.86 33.43 12.96 0.00 12.22 17.31 12.75 15.60 21.17 19.95 26.32 44.34 19.70 32.68 22.95 18.81 25.26 5.57 20.04 16.04 0.00 23.23 21.83 17.35

0.3 1.2 0.5 0.9 2.4 1.5 1.5 0.8 0.8 0.7 1.2 0.9 0.6 0.6 0.3 0.1 1.6 0.1 0.2 1.4 0.7 0.4 0.7 1.4 0.7 1.1 0.5 0.5 0.0 1.1 0.6 0.6 0.9 1.6 1.5

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BB Subordination (%)

Ratio of BB Subordination to Future Expected Loss

2.5 0.0 2.0 0.6 0.6 0.1 0.1 0.5 0.0 0.0 0.0 1.0 0.2 0.0 0.5 0.6 0.2 0.5 0.6 0.0 0.0 0.2 0.7 0.7 0.2 0.1 2.5 0.1 0.4 0.0 0.4 0.1 0.2 5.7 0.4

3.5 5.1 3.3 3.2 3.7 3.4 3.3 2.8 3.2 1.5 2.6 2.3 1.7 1.3 1.7 1.6 2.1 1.4 1.7 1.7 1.6 1.4 1.6 1.4 1.1 1.5 2.1 1.2 1.0 1.3 1.2 1.3 1.2 2.2 1.2

2.2 4.6 4.1 4.1 5.0 4.7 4.8 4.7 5.5 2.8 5.0 4.5 3.8 3.0 3.9 4.3 6.0 4.0 5.1 5.5 5.3 4.7 5.5 4.7 3.7 5.2 7.3 4.8 3.8 5.1 4.7 5.1 5.5 10.3 5.8

0.6 0.9 1.3 1.3 1.3 1.4 1.5 1.7 1.7 1.9 1.9 2.0 2.3 2.3 2.3 2.7 2.9 2.9 3.0 3.3 3.3 3.3 3.4 3.4 3.4 3.5 3.5 3.8 3.8 3.9 3.9 3.9 4.4 4.7 4.7

Please see additional important disclosures at the end of this report.

153

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 13

Deal Name GSMS 1999-C1 LBUBS 2000-C4 CSFB 2001-CK6 SBM7 2001-C2 DLJCM 2000-CF1 FUNBC 2001-C2 HFCMC 2000-PH1 CSFB 2001-CKN5 MLMI 1999-C1 GECMC 2001-3 JPMC 2000-C10 MSC 1999-FNV1 BSCMS 2000-WF2 FUNBC 2000-C1 BSCMS 1999-C1 DLJCM 1999-CG1 LBCMT 1999-C1 BSCMS 2000-WF1 DMARC 1998-C1 CSFB 2001-CK1 LBCMT 1999-C2 SBM7 2000-C1 CCMSC 2000-3 LBUBS 2001-C2 JPMCC 2001-CIB2 GMACC 1999-C2 GMACC 2001-C2 GECMC 2001-1 DLJCM 1999-CG2 BSCMS 1999-WF2 FUNCM 1999-C2 FUNBC 2000-C2 GMACC 1999-C1 BSCMS 2001-TOP4 FUNBC 2001-C3

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

F J K K B4 L H K F I H H H H G B4 G H F J H J H J H H K I B4 H H H F H K

Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 -

BB BB* BB BB BB BB* BB BB BB BB+ BB BB* BB BB BB BB BB BB BB BB BB BB

0.23 0.00 0.19 1.67 0.30 0.00 1.31 0.00 1.14 0.00 0.00 0.27 2.04 0.00 0.81 0.93 0.00 0.00 0.10 0.15 0.00 0.55 0.35 0.00 0.00 0.00 0.00 0.86 0.07 0.00 0.00 0.34 0.00 0.00 0.00

0.00 0.00 0.97 0.79 6.34 0.00 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.64 0.00 0.12 0.00 0.00 0.00 0.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

4.91 0.08 0.01 0.00 0.22 0.00 0.72 0.00 2.92 0.00 1.10 3.08 2.01 0.00 1.37 0.95 0.67 0.00 0.00 0.00 0.36 1.83 1.43 0.33 0.00 5.66 0.56 0.20 1.88 1.01 1.33 0.00 2.47 0.49 0.18

0.52 0.00 0.00 0.00 0.00 0.22 0.00 0.50 4.82 0.85 0.00 1.58 0.00 0.00 0.00 0.00 0.00 0.00 2.17 0.00 0.98 0.00 0.50 0.00 0.00 0.00 0.00 0.00 0.13 0.00 0.00 1.21 0.00 0.00 0.00

0.25 1.55 0.00 0.00 0.23 0.32 0.29 0.00 3.15 0.00 0.38 2.58 0.00 0.00 0.00 1.22 0.00 0.46 1.02 0.21 1.10 0.64 0.00 0.00 0.51 0.00 0.00 0.00 1.67 0.49 0.00 0.07 0.00 0.00 0.00

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

154

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.38 3.46 0.70 0.00 2.26 0.75 0.91 0.00 0.18 0.00 2.19 0.36 0.00 2.11 0.00 0.26 1.76 1.07 1.74 0.00 0.00 2.12 0.65 0.17 0.62 0.00 0.00 1.42 0.60 0.00 0.63 0.53 1.72 0.00 1.37

28.35 19.85 15.07 12.43 23.10 25.88 21.30 19.13 18.95 24.70 12.83 18.54 13.53 33.34 13.92 32.28 29.61 7.50 23.83 31.82 22.92 16.74 24.90 29.40 15.11 15.12 30.04 27.03 28.22 6.62 33.13 21.55 0.00 10.81 27.50

1.3 0.7 0.2 0.2 0.9 0.4 0.4 0.3 3.1 0.5 0.4 2.0 0.5 0.3 0.4 0.8 0.4 0.3 1.0 0.3 0.9 0.7 0.7 0.3 0.3 1.1 0.4 0.3 1.1 0.4 0.5 0.6 0.4 0.2 0.3

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BB Subordination (%)

Ratio of BB Subordination to Future Expected Loss

0.9 0.0 0.0 0.0 0.0 0.0 0.5 0.1 0.0 0.0 0.3 0.1 0.0 0.4 0.6 0.6 0.4 0.2 3.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.0 0.8 0.0 0.0

1.5 0.9 0.9 0.9 1.0 0.8 1.1 0.9 2.1 1.0 1.0 1.4 0.6 0.9 0.7 0.9 0.5 0.6 1.4 0.7 0.6 0.9 0.8 0.7 0.7 0.8 0.7 0.7 0.8 0.5 0.7 0.7 0.8 0.4 0.6

7.2 4.4 4.7 4.8 5.4 4.4 5.7 4.7 11.2 5.4 5.3 7.7 3.7 5.1 4.0 5.7 3.3 3.8 9.0 4.9 4.0 5.8 5.2 4.6 5.0 5.4 5.4 5.0 6.6 4.0 5.9 5.8 6.7 3.1 5.4

4.8 5.0 5.0 5.1 5.2 5.2 5.2 5.3 5.3 5.3 5.5 5.6 5.8 5.8 6.0 6.1 6.3 6.3 6.6 6.7 6.7 6.8 6.8 6.8 7.1 7.2 7.4 7.7 7.8 7.8 7.9 8.1 8.3 8.3 9.0

Please see additional important disclosures at the end of this report.

155

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 13

Deal Name CSFB 2001-CK3 MSDWC 2001-TOP5 JPMCC 2001-CIB3 MCFI 1998-MC2 MSDWC 2001-TOP3 BACM 2000-2 COMM 1999-1 CSFB 2001-4 MSC 1999-LIFE1 LBUBS 2001-C3 GSMS 1998-C1 PSSF 1999-NRF1 PSSF 1999-C2 DLJCM 1999-CG3 LBUBS 2001-C7 BACM 2000-1 FUCMT 1999-C1 MLMI 1998-C3 MLMI 1997-C2 CSFB 1997-C2 CCMSC 2000-2 GMACC 1997-C2 AMC 1998-C1 CASC 1998-D7 HFCMC 1999-PH1 PNCMA 1999-CM1 JPMC 1998-C6 FULBA 1998-C2 MCFI 1998-MC3 MSC 1998-HF1 JPMC 1997-C5 LBCMT 1998-C1 MCFI 1998-MC1 NLFC 1999-1 MSDWC 2000-LIF2

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

J H H G H K G J J J F G J B4 K H F E F F H E F B-2 J B4 F G F G F G H F J

Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba1 Ba2 Ba1 Ba2 Ba2 Ba1 Ba2 Ba2

BB BB BB BB BB BB BB BB BB+* BB BB* BB BBB-* BB BB Ba1 BB BB BB BB* BB BB* BB* BB BB BB+* BB

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.07 0.00 0.00 0.95 0.00 0.00 0.00 0.00 0.74 0.56 0.00 0.00 0.00 0.00 0.38 0.54 0.00 0.43 0.32 0.00 0.00 0.15 0.76 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.51 0.00 0.00 0.86 0.00 0.00 0.13 0.00 0.17 0.63 0.57 0.23 0.00 0.71 0.23 0.00 0.00 0.00

0.10 0.00 0.00 0.00 0.50 0.00 0.33 0.00 0.00 0.00 1.52 0.00 0.00 1.17 0.10 1.68 0.00 0.00 2.65 5.98 0.00 1.55 1.24 1.20 1.69 0.00 8.23 0.00 0.29 1.23 2.96 1.71 1.25 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00 1.25 0.00 1.75 0.00 1.15 0.48 0.93 0.30 0.00 0.00 0.24 0.00 1.02 0.00 0.00 0.00 0.00 2.30 0.00 0.00 0.00 0.20 0.00 0.00 0.00 0.00 0.48 0.45 0.00

0.00 0.00 0.00 0.48 0.00 0.17 1.19 0.00 0.00 0.00 1.50 0.00 0.57 0.37 0.00 0.00 0.32 0.00 0.22 1.25 0.00 3.86 0.97 0.48 0.00 0.00 0.00 1.09 1.39 0.00 0.00 0.16 0.07 0.00 0.00

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

156

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.76 0.00 0.00 0.80 0.00 0.00 2.74 3.27 0.00 1.01 0.94 0.21 0.00 0.92 0.00 4.88 1.73 0.00 1.71 1.13 0.00 3.05 0.00 0.00 0.43 0.90 0.00 5.37 0.66 0.91 0.40 1.46 0.00 0.75 0.00

23.92 9.77 14.56 26.08 7.88 34.89 31.76 15.48 22.61 18.90 17.17 23.83 22.47 36.30 6.31 8.70 21.73 28.53 31.03 20.39 34.33 35.75 23.48 11.08 17.49 15.41 18.99 17.25 26.23 16.57 22.67 31.78 26.40 25.83 20.34

0.2 0.1 0.1 0.3 0.2 0.3 1.1 0.2 0.7 0.2 1.2 0.3 0.6 0.8 0.1 0.4 0.4 0.4 1.0 1.5 0.4 1.5 0.6 1.1 0.4 0.2 1.5 0.6 0.7 0.3 0.5 0.6 0.6 0.3 0.2

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BB Subordination (%)

Ratio of BB Subordination to Future Expected Loss

0.0 0.0 0.0 1.3 0.0 0.1 0.0 0.0 0.0 0.0 1.4 0.5 0.3 0.1 0.0 0.1 0.5 2.8 0.7 2.1 0.0 1.9 1.4 0.6 0.1 0.5 0.0 0.7 0.3 1.0 2.0 1.1 1.3 0.3 0.0

0.5 0.3 0.5 0.5 0.3 0.5 0.8 0.5 0.5 0.4 0.7 0.6 0.6 0.5 0.3 0.6 0.6 0.9 0.5 0.4 0.4 1.1 0.6 0.5 0.4 0.4 0.4 0.4 0.7 0.4 0.4 0.5 0.5 0.5 0.2

4.5 3.1 4.6 4.3 3.0 5.1 7.3 4.6 4.7 3.6 7.0 6.4 7.1 6.1 3.3 7.0 6.6 11.1 6.0 5.4 5.2 13.2 8.3 6.4 5.4 6.2 5.9 5.6 10.2 5.8 6.7 7.6 8.2 8.1 3.3

9.2 9.3 9.3 9.3 9.3 9.5 9.5 9.8 9.8 9.9 10.1 10.5 11.1 11.2 11.4 11.5 11.7 12.0 12.1 12.4 12.4 12.5 13.4 13.5 13.8 14.1 14.3 14.9 15.2 15.6 16.0 16.3 16.6 16.7 16.8

Please see additional important disclosures at the end of this report.

157

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 13

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Deal Name

Class

Current Moody’s Rating

NLFC 1998-2 NLFC 1998-1 CSFB 1998-C2 SBM7 1999-C1 CMAC 1998-C2 CMFUN 1999-1 JPMC 1999-C7 GMACC 1997-C1 CMAT 1999-C2 CCMSC 1997-1 MCFI 1997-MC1 PNCMA 2000-C2 DLJMA 1997-CF2 DLJCM 1998-CG1 CCMSC 1998-1 CMAT 1999-C1 CCMSC 1999-2 GMACC 1998-C2 MSC 1998-WF1 DLJCM 1998-CF2 MSC 1999-RM1 FULB 1997-C2 PSSF 1998-C1 ASC 1997-D4 CMAC 1998-C1 BSCMS 1998-C1 MSC 1998-HF2 CCMSC 1998-2 CSFB 1997-C1 LBCMT 1998-C4 CCMSC 1997-2 ASC 1997-D5 GMACC 1998-C1 NASC 1998-D6 CMAC 1999-C1

F F F H G H F G J F F J B3TB B4 F G H G G B-3 H G H B-2 G G G F F G F A5 H B-2 J

Ba2 Ba2 Ba2 B1 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2 Ba2

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

158

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

BB* BB BB BB BB BB* BB BB+ BB* BB BB* BB BB BB BB BB BB+ BB* BB BB* BB BB BB BB BB BB BB -

0.52 0.25 0.53 0.00 0.78 0.00 0.12 2.11 0.00 0.00 0.00 0.00 0.00 0.66 0.00 0.66 2.65 0.11 0.00 0.08 0.00 1.17 0.00 0.91 0.24 0.95 0.00 0.00 0.00 0.51 0.56 0.00 0.00 0.12 0.96

0.07 0.37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.81 0.00 0.00 0.00 0.78 0.00 0.00 0.00 0.18 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.31 0.00 0.89

1.33 3.66 0.00 2.64 3.15 0.22 0.00 0.22 0.67 0.96 7.88 0.19 3.69 0.44 0.00 2.20 0.00 1.41 0.39 0.00 0.00 1.40 0.24 3.14 0.36 0.49 0.83 0.69 2.67 0.13 4.11 0.00 0.18 1.38 0.00

0.00 0.00 0.86 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 10.73 0.00 0.00 0.00 0.00 0.00 1.70 0.77 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.19 0.87 0.00 2.90 0.00 0.43 0.00

0.58 0.00 0.00 0.00 0.00 0.25 1.11 0.82 0.00 4.15 2.09 0.00 0.00 0.00 0.00 1.28 0.00 0.00 0.00 0.00 0.56 1.93 0.00 2.20 0.00 1.31 0.48 0.00 0.00 0.17 0.00 0.88 0.00 0.00 0.00

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

1.07 1.15 3.38 0.00 0.98 0.00 0.00 0.85 0.30 0.00 2.28 0.00 0.00 1.91 0.00 0.63 2.45 1.09 2.61 0.00 0.00 2.59 0.50 7.41 0.49 0.52 3.16 0.00 9.08 0.90 0.96 0.82 18.01 0.79 0.55

6.11 15.17 20.39 19.05 11.93 18.87 30.16 23.45 0.00 15.47 30.85 22.49 21.29 26.87 25.45 15.31 34.80 23.40 24.22 37.73 0.00 20.51 13.85 6.60 27.78 19.03 23.68 17.37 18.31 29.33 23.76 16.78 0.00 14.66 16.22

0.5 0.7 0.5 0.6 0.6 0.3 0.6 0.4 0.1 1.2 1.6 0.2 3.5 0.4 0.2 1.0 0.4 0.5 0.7 0.5 0.2 0.9 0.2 1.2 0.3 0.7 0.5 0.3 0.6 0.6 0.7 1.2 0.3 0.5 0.2

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BB Subordination (%)

Ratio of BB Subordination to Future Expected Loss

0.8 0.8 0.9 0.0 0.7 0.2 0.0 1.9 0.4 1.0 0.7 0.0 0.3 0.5 0.7 0.2 0.0 0.3 0.2 0.5 0.3 0.9 0.6 0.6 0.5 0.0 0.5 0.4 0.9 0.0 0.5 1.6 0.3 0.1 0.0

0.3 0.5 0.4 0.4 0.4 0.3 0.4 0.3 0.3 0.3 0.4 0.3 0.5 0.3 0.3 0.3 0.3 0.2 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.3 0.2 0.2 0.2 0.2 0.3 0.2 0.1 0.2

6.0 8.1 6.3 7.6 6.8 5.4 7.7 6.8 6.2 6.6 7.6 5.3 11.2 5.6 5.7 6.5 5.9 5.2 6.5 6.6 7.8 6.6 6.6 6.9 6.2 5.2 8.1 5.4 7.3 5.3 6.1 12.1 6.3 5.6 6.0

17.6 17.9 17.9 18.8 19.4 19.7 19.8 19.9 20.2 20.5 20.5 20.6 21.9 22.0 22.1 22.1 22.6 23.3 23.8 24.7 25.4 26.7 27.1 28.4 28.5 29.0 29.4 31.9 33.2 33.9 35.7 35.8 36.4 38.4 38.6

Please see additional important disclosures at the end of this report.

159

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 13

Deal Name DLJCM 1998-CF1 ACMF 1997-C1 FULB 1997-C1 NLFC 1999-2 MCFI 1997-MC2 MSC 1997-C1 MSC 1997-HF1 MSC 1999-WF1 MSC 1998-WF2 JPMC 1997-C4 MSC 1997-WF1

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

B4 G F G F F G H H G G

Ba2 Ba2 -

BB BB* BB* BB* BB BBB+ BB+ BB+ BB BBBB

0.00 9.58 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.74 0.00 0.00 0.00 1.17 0.00 0.00 0.00 0.00

0.72 6.74 1.35 0.00 1.41 0.00 1.52 0.00 0.00 0.43 0.00

0.00 0.00 1.40 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.73 0.00 0.00 2.00 0.00 0.00 0.00 0.00 0.00

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

160

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.93 0.00 2.69 1.45 2.21 1.92 0.00 0.00 1.18 0.00 2.12

20.71 32.57 23.65 19.41 19.53 23.76 0.00 2.90 11.26 0.00 6.03

0.3 1.1 0.8 0.1 0.4 0.5 0.2 0.0 0.1 0.0 0.1

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current BB Subordination (%)

Ratio of BB Subordination to Future Expected Loss

0.3 0.0 0.3 0.0 0.6 0.8 0.4 0.1 0.0 0.0 0.0

0.2 0.2 0.2 0.1 0.1 0.3 0.1 0.1 0.0 0.0 0.0

6.7 9.4 9.5 6.6 7.4 15.9 7.6 5.7 5.0 5.6 6.2

44.1 48.9 52.4 56.8 59.3 63.1 65.3 91.4 182.6 678.9 694.4

Please see additional important disclosures at the end of this report.

161

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels Section III: Projecting Losses on Single-B CMBS In Sections I and II, we ranked BBB- and BB CMBS bonds based on their level of principal protection from losses. In this section, we complete our Projecting Losses trilogy on subordinate CMBS by ranking the universe of single-B conduit bonds. We measure the level of principal protection on a bond by comparing the future expected losses for the deal’s collateral to the current credit for the bond. The rankings by subordination to expected loss ratio are provided in Exhibit 17 of this section. RATIOS A ND R ANKINGS

We believe that the bonds at the top of the list (with the lowest ratios) have the highest chance of default and downgrade. Likewise, we believe that the bonds at the bottom of the list (with the highest ratios) have a higher level of principal protection and a lower chance for rating agency downgrade.

exhibit 14

SINGLE-B SUBORDINATION TO EXPECTED LOSS RATIOS FOR 1997-2001 CONDUITS

Ranges for Single B Subordination to Expected Loss Ratio 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 >8

Average Default Timing Curve; 34% Severity # of Deals 12 20 20 18 10 11 10 7 44

Source: Morgan Stanley, Trepp, Bloomberg

162

% of Deals 7.9 13.2 13.2 11.8 6.6 7.2 6.6 4.6 28.9

Average Default Timing Curve; 43% Severity # of Deals 17 24 24 10 14 9 8 7 39

% of Deals 11.2 15.8 15.8 6.6 9.2 5.9 5.3 4.6 25.7

A ratio that is less than 1 means that we expect the class to default before maturity; a ratio of greater than 1 means we expect no default. Depending on the severity assumption and timing curve used to project future losses, between 12 and 39 bonds have subordination to loss ratios of less than 1 (See Exhibit 14). Between 20 and 32 bonds have a ratio between 1 and 2. We believe that classes with a ratio between 1 and 2 may be vulnerable to a downgrade. SCOPE O F T HE S TUDY

This study analyzes and ranks bonds based on the level of principal protection from losses. Our analysis does not assess the risk of potential downgrades that may occur for other reasons, such as interest shortfalls. For example, while ASC 1997-D5 A6 has a high subordination to expected loss ratio2, this class was downgraded by Fitch in September 2003 to B from BB+. The downgrade was prompted by interest shortfalls caused by servicer reimbursements of non-recoverable advances. S&P also downgraded this tranche to D from BBB-.

1986 Default Timing Curve; 34% Severity # of Deals 33 32 16 7 3 4 6 3 48

% of Deals 21.7 21.1 10.5 4.6 2.0 2.6 3.9 2.0 31.6

1986 Default Timing Curve; 43% Severity # of Deals 39 32 16 3 5 4 4 4 45

% of Deals 25.7 21.1 10.5 2.0 3.3 2.6 2.6 2.6 29.6

2 Ratio of 27.2, assuming average default timing curve from Esaki study and 34% severity rate; Ratio of 44.0, assuming 1986 default timing curve and 43% severity rate.

Please see additional important disclosures at the end of this report.

163

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels Projecting Losses - June 2003

In order to assess the strength of this study as a predictor of rating actions, we reviewed rating changes on the 117 single-B bonds from the prior version of this study. Over the course of a year, we would expect the bonds with the lowest subordination to loss ratios to experience the greatest number of downgrades. We found that 13 of the 20 single-B classes with the lowest ratios were downgraded during the past year. The number of downgraded classes declined as ratios increased, with only 4 of the next 20 ranked bonds being downgraded. Likewise, we would expect the bonds with the highest subordination to loss ratios to experience the fewest downgrades and/or the most upgrades. Of the 20 bonds with the highest ratios, we found that 2 were downgraded during the past year, while 2 were upgraded. The number of upgraded classes declined as ratios decreased, with only 1 of the next 20 ranked bonds experiencing an upgrade. No other classes were upgraded since the prior version of the study. METHODOLOGY

In this section, we analyze single-B classes of 152 conduit transactions tracked by Trepp. Each conduit transaction was issued between January 1997 and December 2001. 126 of the 152 classes are currently rated single-B by Moody’s, Fitch or S&P; the other 26 classes are rated B+ or B-. For each transaction, we calculate a future expected loss rate based on historical losses and losses implied by the current level of 30, 60, and 90-day delinquencies, REO, foreclosures, performing specially serviced loans and servicer watchlisted loans. We do not for the impact of defeased collateral within the deals. We begin by asg a probability of liquidation for each delinquency category, based on results from Esaki’s commercial mortgage default study. Assumptions for Probabilities of Liquidation

In Esaki’s study, about 55% of loans that became more than 90 days delinquent were ultimately liquidated. For this section, we borrow the results from the Esaki study and assume that 55% of 90+ day delinquent loans will be liquidated. We then assume that 60-day delinquent loans will have a 25% liquidation rate, and 30-day delinquent loans will have a 10% liquidation rate. To be conservative, we assume that all foreclosed and REO loans will be liquidated. We assign a 2.5% liquidation rate to servicer watchlisted loans and a 5% liquidation rate to performing specially serviced loans. We assume a low probability of liquidation for the performing specially serviced loans because these loans could have been transferred to the special servicer for technical rather than monetary defaults.

164

Calculating Loan Losses: Average Timing of Defaults

We then multiply the projected liquidated loan total by a severity rate to compute each deal’s expected loss rate for the year. For our analysis, we consider two loss severity rates: 34% and 43%. The first severity rate that we assume is 34%. Liquidated life insurance company loans in the Commercial Mortgage Default Study: 1972-2000, by Howard Esaki, experienced a 34% severity rate. The average severity rate on liquidated CMBS loans is higher, at 43%3. The calculated loss rates for the year are then used in conjunction with historical losses to project future loss rates. Since the calculated losses are based on current delinquency levels, we assume that the resulting losses will occur within one year from today. For example, with a 34% severity rate assumption, DLJCM 2000-CKP1, is expected to have a 0.3% loss rate based on current levels of delinquencies, specially serviced loans and watchlisted loans. This transaction has seasoned for 4 years, so we assume losses will occur in year 5. The projected loss based on current delinquencies (0.3%) is then added to the deal’s historical losses of 2.5%. To project future losses, we assume that the timing of CMBS loan losses in this study mirrors the average timing of defaults on the life insurance company loans from Esaki’s study. According to the timing of defaults presented in Esaki’s study, about 47% of loans default by the end of year 5. The DLJCM 2000-CKP1 transaction is expected to have a 2.8% loss rate by year 5, based on historical losses and projected losses of currently delinquent loans. If we apply the results of Esaki’s study, this 2.8% loss rate should correspond to 47% of total loan losses. Therefore, we conclude that this transaction may suffer a 3.2% loss rate (based on original balances) during its remaining lifetime, since 53% of loan losses have yet to occur. (2.8%*(1-47%)¸47%=3.2% loss rate) A 3.2% loss rate based on original balances corresponds to a 3.5% loss rate, based on current balances.

3

See Special Servicer Severities, February 6, 2004.

Please see additional important disclosures at the end of this report.

165

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels

exhibit 15

Year 1 2 3 4 5 6 7 8 9 10

AVERAGE TIMING OF DEFAULTS FROM ESAKI’S COMMERCIAL MORTGAGE DEFAULT STUDY1

Default Rates from Esaki’s Study (%) 0.2 1.0 1.7 2.0 2.1 2.3 2.8 1.5 0.9 0.5

Total Defaults That Occur Each Year (%) 1.5 6.9 11.5 13.1 13.9 15.2 18.9 9.7 5.7 3.6

Cumulative Level of Defaults Through Time (%) 1.5 8.4 19.9 33.0 46.9 62.1 81.0 90.7 96.4 100.0

1

Real Estate Finance, Commercial Mortgage Defaults: 1972-2000, by Howard Esaki, Winter 2002 Edition.

Source: Morgan Stanley

The timing of defaults presented in Exhibit 15 is fairly evenly distributed, with 47% of loan losses occurring in the first five years after loan origination, and the remaining 53% of losses occurring in years 6 through 10. If the timing of defaults in CMBS deviates from this average, it is possible that our analysis would yield different results. We repeated our analysis, applying the default timing curve experienced by life insurance company loans that were originated in 1986. The 1986 origination cohort experienced the highest default rate of any origination year between 1972 and 1995. For the purposes of this analysis, we do not focus on the absolute level of defaults, but we apply the 1986 default timing curve as a proxy for the timing of CMBS loan losses in a real estate downturn. Life insurance company loans originated in 1986 experienced 70% of all defaults in years 6 through 10 after origination. Since most of the CMBS transactions in our study have seasoned for less than six years, the 1986 default timing curve assumes that the majority of losses have not yet occurred. Using this back-loaded timing curve, the projected future losses for these CMBS transactions are higher than in our initial analysis, resulting in lower subordination to loss ratios.

166

exhibit 16

Year 1 2 3 4 5 6 7 8 9 10

TIMING OF DEFAULTS FOR 1986 COHORT FROM ESAKI’S COMMERCIAL MORTGAGE DEFAULT STUDY1

Default Rates from Esaki’s Study (%) 0.0 0.9 2.3 2.1 3.3 5.8 10.3 2.5 1.3 0.2

Total Defaults That Occur Each Year (%) 0.0 3.2 7.9 7.5 11.3 20.2 35.8 8.6 4.6 0.8

Cumulative Level of Defaults Through Time (%) 0.0 3.2 11.1 18.6 29.9 50.2 86.0 94.6 99.2 100.0

1

Real Estate Finance, Commercial Mortgage Defaults: 1972-2000, by Howard Esaki, Winter 2002 Edition.

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

167

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 17

Deal Name DLJCM 2000-CKP1 JPMCC 2001-C1 GECMC 2000-1 SBM7 2000-C3 MSDWC 2001-TOP1 SBM7 2001-C1 CSFB 2001-CF2 GMACC 2000-C3 GMACC 2001-C1 SBM7 2000-C2 BACM 2001-1 FUNBC 2001-C4 BSCMS 2001-TOP2 CCMSC 2000-1 GMACC 1999-C3 LBUBS 2000-C3 KEY 2000-C1 JPMC 2000-C9 PNCMA 2001-C1 GMACC 2000-C2 GMACC 2000-C1 PMCF 2001-ROCK BSCMS 1999-C1 MSDWC 2000-LIFE BAFU 2001-3 COMM 2000-C1 FUBOA 2001-C1 CSFB 2000-C1 GECMC 2001-2 LBCMT 1999-C1 PNCMA 2000-C1 JPMCC 2001-CIBC JPMC 1999-C8 FUNBC 1999-C4 LBUBS 2000-C4

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

B6 M K L L M M M M L N O L H K M M H M J L M I L O K O L L J L L H L M

Caa1 B3 B2 B2 B2 B2 B2 B2 B2 B2 B2 B1 B2 B2 B2 B2 B2 B2 B2 B2 B3 B2

B+ B B B* B B B B B B B B* BB B+* B B B B B* B B B B B BBB -

0.00 0.10 0.00 2.43 0.00 0.40 0.07 0.17 1.20 0.00 0.35 0.00 0.00 0.00 0.35 0.00 0.00 0.00 0.00 5.02 0.00 0.00 0.81 0.60 0.00 0.00 0.29 0.19 0.00 0.00 0.29 0.67 0.64 0.00 0.00

0.00 0.00 1.37 2.14 0.36 0.00 0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.84 0.00 0.00 0.00 0.00 0.00 0.00 1.46 0.00 0.00 0.00 0.00 0.00 0.17 0.00 0.00 0.00 0.85 0.00 0.00 0.00 0.00

0.10 0.54 2.41 0.00 2.50 3.62 1.08 0.86 5.32 2.14 4.66 0.00 2.67 9.59 7.18 0.44 1.11 0.46 0.00 3.16 5.18 0.00 1.37 0.38 0.54 0.00 0.69 1.33 0.50 0.67 2.69 1.31 1.69 0.17 0.08

0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.01 0.11 1.51 0.00 1.21 0.00 0.16 0.74 0.56 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.98 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.46 0.00

0.20 3.10 0.00 1.63 0.00 1.77 0.21 0.54 0.00 1.93 1.05 0.33 0.00 7.45 0.00 0.92 3.09 0.00 0.00 0.00 0.50 0.00 0.00 1.19 0.00 0.00 0.00 2.05 0.50 0.00 0.36 0.57 0.30 1.25 1.55

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

168

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.00 0.00 1.69 0.00 1.49 4.57 0.01 0.00 0.00 1.97 0.85 0.09 1.72 0.00 0.00 1.12 0.00 0.66 0.84 0.00 1.20 0.25 0.00 0.00 0.00 0.36 3.53 0.56 0.00 1.76 0.83 6.50 0.67 1.02 3.46

23.89 12.79 29.03 11.36 23.71 15.74 0.00 33.43 29.06 25.19 28.86 19.70 12.96 44.17 19.70 32.68 17.31 35.79 12.75 24.86 21.17 5.57 13.92 16.04 15.60 44.34 26.32 20.04 25.26 29.61 23.23 0.00 18.81 22.95 19.85

0.3 1.2 0.8 0.9 0.7 1.5 0.3 0.6 1.2 2.4 1.5 0.8 0.6 5.0 1.4 0.7 1.6 0.5 0.1 0.9 1.4 0.0 0.4 0.6 0.2 0.7 0.4 1.1 0.5 0.4 0.9 0.6 0.5 1.1 0.7

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current B Subordination (%)

Ratio of B Subordination to Future Expected Loss

2.5 0.0 0.5 0.6 0.0 0.1 0.5 0.2 0.0 0.6 0.1 0.0 0.0 0.0 0.7 0.2 0.2 2.0 0.5 1.0 0.0 0.4 0.6 0.4 0.6 0.7 0.2 0.0 0.1 0.4 0.2 0.1 2.5 0.1 0.0

3.5 5.1 2.8 3.2 1.5 3.3 1.7 1.7 2.6 3.7 3.4 3.2 1.3 5.9 1.4 1.1 2.1 3.3 1.4 2.3 1.7 1.0 0.7 1.2 1.7 1.6 1.4 1.3 1.2 0.5 1.2 1.3 2.1 1.5 0.9

1.4 2.5 2.0 2.3 1.1 2.5 1.3 1.4 2.2 3.3 3.0 3.2 1.3 6.0 1.4 1.2 2.2 4.1 1.8 2.9 2.3 1.4 1.0 1.7 2.5 2.7 2.3 2.2 2.2 1.0 2.4 2.5 4.2 3.1 1.8

0.4 0.5 0.7 0.7 0.7 0.8 0.8 0.8 0.8 0.9 0.9 1.0 1.0 1.0 1.0 1.1 1.1 1.3 1.3 1.3 1.4 1.4 1.4 1.5 1.5 1.6 1.7 1.7 1.7 1.9 1.9 1.9 2.0 2.1 2.1

Please see additional important disclosures at the end of this report.

169

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 17

Deal Name BSCMS 2000-WF1 BSCMS 2000-WF2 CSFB 2001-CKN5 LBUBS 2000-C5 LBCMT 1999-C2 GSMS 1999-C1 JPMC 2000-C10 CSFB 2001-CK1 HFCMC 2000-PH1 CSFB 2001-CK6 LBUBS 2001-C2 SBM7 2001-C2 GECMC 2001-3 DLJCM 1999-CG1 CMAT 1999-C1 GECMC 2001-1 CSFB 1999-C1 FUNBC 2001-C2 FUNBC 2000-C1 DLJCM 2000-CF1 MLMI 1999-C1 GMACC 1999-C2 MCFI 1998-MC2 CSFB 2001-CK3 FUNCM 1999-C2 PSSF 1999-NRF1 CCMSC 2000-3 MSC 1999-FNV1 JPMCC 2001-CIB2 BSCMS 2001-TOP4 GMACC 2001-C2 DLJCM 1999-CG2 MSDWC 2001-TOP3 MSDWC 2001-TOP5 SBM7 2000-C1

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

K L N K L G L M L N M N L B7 K L K O L B7 G K J M L K K K L L N B7 L L M

B2 B1 B2 B3 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2

B B B* B B B B B B B B B BBB B* B B B B B -

0.00 2.04 0.00 0.00 0.00 0.23 0.00 0.15 1.31 0.19 0.00 1.67 0.00 0.93 0.66 0.86 0.13 0.00 0.00 0.30 1.14 0.00 0.00 0.00 0.00 0.00 0.35 0.27 0.00 0.00 0.00 0.07 0.00 0.00 0.55

0.64 0.00 0.00 0.00 0.00 0.00 0.00 0.12 0.10 0.97 0.20 0.79 0.00 0.00 0.00 0.00 1.51 0.00 0.00 6.34 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.00 2.01 0.00 2.40 0.36 4.91 1.10 0.00 0.72 0.01 0.33 0.00 0.00 0.95 2.20 0.20 0.10 0.00 0.00 0.22 2.92 5.66 0.00 0.10 1.33 0.00 1.43 3.08 0.00 0.49 0.56 1.88 0.50 0.00 1.83

0.00 0.00 0.50 0.00 0.98 0.52 0.00 0.00 0.00 0.00 0.00 0.00 0.85 0.00 0.00 0.00 0.58 0.22 0.00 0.00 4.82 0.00 0.00 0.00 0.00 0.48 0.50 1.58 0.00 0.00 0.00 0.13 0.00 0.00 0.00

0.46 0.00 0.00 0.47 1.10 0.25 0.38 0.21 0.29 0.00 0.00 0.00 0.00 1.22 1.28 0.00 2.91 0.32 0.00 0.23 3.15 0.00 0.48 0.00 0.00 0.00 0.00 2.58 0.51 0.00 0.00 1.67 0.00 0.00 0.64

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

170

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

1.07 0.00 0.00 0.65 0.00 0.38 2.19 0.00 0.91 0.70 0.17 0.00 0.00 0.26 0.63 1.42 7.00 0.75 2.11 2.26 0.18 0.00 0.80 0.76 0.63 0.21 0.65 0.36 0.62 0.00 0.00 0.60 0.00 0.00 2.12

7.50 13.53 19.13 19.95 22.92 28.35 12.83 31.82 21.30 15.07 29.40 12.43 24.70 32.28 15.31 27.03 17.35 25.88 33.34 23.10 18.95 15.12 26.08 23.92 33.13 23.83 24.90 18.54 15.11 10.81 30.04 28.22 7.88 9.77 16.74

0.3 0.5 0.3 0.7 0.9 1.3 0.4 0.3 0.4 0.2 0.3 0.2 0.5 0.8 1.0 0.3 1.5 0.4 0.3 0.9 3.1 1.1 0.3 0.2 0.5 0.3 0.7 2.0 0.3 0.2 0.4 1.1 0.2 0.1 0.7

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current B Subordination (%)

Ratio of B Subordination to Future Expected Loss

0.2 0.0 0.1 0.0 0.0 0.9 0.3 0.0 0.5 0.0 0.0 0.0 0.0 0.6 0.2 0.0 0.4 0.0 0.4 0.0 0.0 0.0 1.3 0.0 0.6 0.5 0.0 0.1 0.0 0.0 0.0 0.2 0.0 0.0 0.0

0.6 0.6 0.9 1.6 0.6 1.5 1.0 0.7 1.1 0.9 0.7 0.9 1.0 0.9 0.8 0.7 1.2 0.8 0.9 1.0 2.1 0.8 0.5 0.5 0.7 0.6 0.8 1.4 0.7 0.4 0.7 0.8 0.3 0.3 0.9

1.2 1.3 2.0 3.7 1.4 3.5 2.3 1.8 2.8 2.5 1.8 2.6 2.8 2.7 2.3 1.9 3.6 2.6 2.8 3.3 6.8 2.4 1.5 1.6 2.5 2.1 2.6 4.8 2.4 1.3 2.6 3.1 1.2 1.3 3.4

2.1 2.1 2.2 2.3 2.3 2.3 2.4 2.5 2.5 2.7 2.7 2.8 2.8 2.9 3.0 3.0 3.0 3.1 3.1 3.2 3.2 3.2 3.2 3.2 3.3 3.4 3.4 3.4 3.4 3.5 3.5 3.6 3.7 3.9 4.0

Please see additional important disclosures at the end of this report.

171

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 17

Deal Name BSCMS 1999-WF2 GMACC 1999-C1 PMAC 1999-C1 DMARC 1998-C1 MSC 1999-LIFE1 MLMI 1998-C3 MSC 1998-CF1 DLJCM 1999-CG3 CSFB 2001-4 JPMCC 2001-CIB3 COMM 1999-1 PSSF 1999-C2 MLMI 1997-C2 MLMI 1998-C2 FUCMT 1999-C1 MSC 1998-HF1 FUNBC 2001-C3 CCMSC 2000-2 CASC 1998-D7 LBUBS 2001-C3 FUNBC 2000-C2 CSFB 1998-C1 BACM 2000-2 PNCMA 1999-CM1 LBUBS 2001-C7 MCFI 1998-MC1 BACM 2000-1 CSFB 1998-C2 AMC 1998-C1 MSDWC 2000-LIF2 GMACC 1997-C2 CCMSC 1998-1 JPMC 1998-C6 CMAC 1998-C2 LBCMT 1998-C1

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

J H F G M F D B7 M L J M H F G J N K B4 M K G N B7 N L L H G M F H G J J

B3 Ba2 B2 B3 B2 B2 B2 B2 B2 B2 B2 B3 -

B+ B B B B+* B* B B B B B* B B* B* B* B BB B* B* B* B* B B B BB B

0.00 0.00 0.00 0.10 0.00 0.00 1.37 0.95 0.00 0.00 0.00 0.00 0.74 0.29 0.00 0.00 0.00 0.00 0.00 0.00 0.34 1.85 0.00 0.54 0.00 0.76 0.00 0.53 0.00 0.00 0.00 0.00 0.00 0.78 0.15

0.00 0.00 0.00 0.00 0.00 2.51 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.27 0.00 0.00 0.00 0.86 0.13 0.00 0.00 0.78 0.00 0.17 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.63 0.00 0.23

1.01 2.47 0.51 0.00 0.00 0.00 0.92 1.17 0.00 0.00 0.33 0.00 2.65 1.34 0.00 1.23 0.18 0.00 1.20 0.00 0.00 1.18 0.00 0.00 0.10 1.25 1.68 0.00 1.24 0.00 1.55 0.00 8.23 3.15 1.71

0.00 0.00 1.64 2.17 1.75 0.00 4.07 0.30 0.00 0.00 1.25 0.93 1.02 0.00 0.24 0.00 0.00 0.00 2.30 0.00 1.21 1.05 0.00 0.00 0.00 0.48 0.00 0.86 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.49 0.00 1.20 1.02 0.00 0.00 0.26 0.37 0.00 0.00 1.19 0.57 0.22 1.53 0.32 0.00 0.00 0.00 0.48 0.00 0.07 0.97 0.17 0.00 0.00 0.07 0.00 0.00 0.97 0.00 3.86 0.00 0.00 0.00 0.16

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

172

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.00 1.72 2.21 1.74 0.00 0.00 6.11 0.92 3.27 0.00 2.74 0.00 1.71 1.91 1.73 0.91 1.37 0.00 0.00 1.01 0.53 2.61 0.00 0.90 0.00 0.00 4.88 3.38 0.00 0.00 3.05 0.00 0.00 0.98 1.46

6.62 0.00 21.36 23.83 22.61 28.53 21.83 36.30 15.48 14.56 31.76 22.47 31.03 0.00 21.73 16.57 27.50 34.33 11.08 18.90 21.55 29.66 34.89 15.41 6.31 26.40 8.70 20.39 23.48 20.34 35.75 25.45 18.99 11.93 31.78

0.4 0.4 1.1 1.0 0.7 0.4 1.6 0.8 0.2 0.1 1.1 0.6 1.0 0.6 0.4 0.3 0.3 0.4 1.1 0.2 0.6 1.2 0.3 0.2 0.1 0.6 0.4 0.5 0.6 0.2 1.5 0.2 1.5 0.6 0.6

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current B Subordination (%)

Ratio of B Subordination to Future Expected Loss

0.4 0.8 1.0 3.5 0.0 2.8 5.7 0.1 0.0 0.0 0.0 0.3 0.7 2.1 0.5 1.0 0.0 0.0 0.6 0.0 0.0 1.3 0.1 0.5 0.0 1.3 0.1 0.9 1.4 0.0 1.9 0.7 0.0 0.7 1.1

0.5 0.8 1.5 1.4 0.5 0.9 2.2 0.5 0.5 0.5 0.8 0.6 0.5 0.8 0.6 0.4 0.6 0.4 0.5 0.4 0.7 0.7 0.5 0.4 0.3 0.5 0.6 0.4 0.6 0.2 1.1 0.3 0.4 0.4 0.5

2.1 3.4 6.1 5.8 2.1 4.3 10.4 2.6 2.3 2.4 4.0 3.4 2.6 4.5 3.1 2.0 3.4 2.3 2.7 2.1 4.3 4.1 3.3 2.7 1.8 3.1 4.0 2.4 4.2 1.4 7.3 1.8 2.9 2.5 3.4

4.1 4.1 4.2 4.2 4.4 4.6 4.8 4.8 4.9 4.9 5.3 5.3 5.3 5.4 5.4 5.5 5.5 5.6 5.7 5.7 5.9 6.1 6.1 6.2 6.2 6.3 6.7 6.8 6.8 6.9 6.9 7.1 7.1 7.2 7.2

Please see additional important disclosures at the end of this report.

173

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 17

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Deal Name

Class

Current Moody’s Rating

HFCMC 1999-PH1 NLFC 1998-1 GSMS 1998-C1 CMAT 1999-C2 NLFC 1998-2 NLFC 1999-1 CMFUN 1999-1 SBM7 1999-C1 DLJCM 1998-CG1 MCFI 1998-MC3 JPMC 1999-C7 PNCMA 2000-C2 FULBA 1998-C2 MSC 1998-WF1 FULB 1997-C2 CCMSC 1999-2 GMACC 1998-C2 DLJCM 1998-CF2 CCMSC 1997-1 MSC 1998-HF2 CMAC 1998-C1 MSC 1999-RM1 CCMSC 1998-2 NASC 1998-D6 PSSF 1998-C1 JPMC 1997-C5 CMAC 1999-C1 LBCMT 1998-C4 DLJCM 1998-CF1 CCMSC 1997-2 GMACC 1997-C1 MLMI 1997-C1 DLJMA 1997-CF2 ASC 1997-D4 MSC 1997-C1

L H G M H H K K B6 H G M J J J K K B-5 H K K L H B-5 K F M K B6 H G F B3TB B3 H

B2 B2 B2 B2 B2 B1 B2 B2 B2 B1 B3

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

174

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

B B B+ B-* B B* B B B B* B-* B B B B-* B B* B B B B+ B+* B* B B+* B* B+* B+

0.38 0.25 0.07 0.00 0.52 0.00 0.00 0.00 0.66 0.32 0.12 0.00 0.43 0.00 1.17 2.65 0.11 0.08 0.00 0.00 0.24 0.00 0.00 0.12 0.00 0.00 0.96 0.51 0.00 0.56 2.11 0.00 0.00 0.91 0.00

0.00 0.37 0.00 0.00 0.07 0.00 0.00 0.00 0.81 0.23 0.00 0.00 0.57 0.00 0.18 0.00 0.78 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.71 0.89 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

1.69 3.66 1.52 0.67 1.33 0.00 0.22 2.64 0.44 0.29 0.00 0.19 0.00 0.39 1.40 0.00 1.41 0.00 0.96 0.83 0.36 0.00 0.69 1.38 0.24 2.96 0.00 0.13 0.72 4.11 0.22 5.97 3.69 3.14 0.00

0.00 0.00 1.15 0.00 0.00 0.45 0.00 0.00 0.00 0.00 0.00 0.00 0.20 1.70 0.00 0.00 0.00 0.77 0.00 0.00 0.00 0.00 0.00 0.43 0.22 0.00 0.00 0.87 0.00 0.00 0.00 0.75 10.73 0.00 0.00

0.00 0.00 1.50 0.00 0.58 0.00 0.25 0.00 0.00 1.39 1.11 0.00 1.09 0.00 1.93 0.00 0.00 0.00 4.15 0.48 0.00 0.56 0.00 0.00 0.00 0.00 0.00 0.17 0.00 0.00 0.82 1.68 0.00 2.20 2.00

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

0.43 1.15 0.94 0.30 1.07 0.75 0.00 0.00 1.91 0.66 0.00 0.00 5.37 2.61 2.59 2.45 1.09 0.00 0.00 3.16 0.49 0.00 0.00 0.79 0.50 0.40 0.55 0.90 0.93 0.96 0.85 7.13 0.00 7.41 1.92

17.49 15.17 17.17 0.00 6.11 25.83 18.87 19.05 26.87 26.23 30.16 22.49 17.25 24.22 20.51 34.80 23.40 37.73 15.47 23.68 27.78 0.00 17.37 14.66 13.85 22.67 16.22 29.33 20.71 23.76 23.45 29.11 21.29 6.60 23.76

0.4 0.7 1.2 0.1 0.5 0.3 0.3 0.6 0.4 0.7 0.6 0.2 0.6 0.7 0.9 0.4 0.5 0.5 1.2 0.5 0.3 0.2 0.3 0.5 0.2 0.5 0.2 0.6 0.3 0.7 0.4 1.2 3.5 1.2 0.5

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current B Subordination (%)

Ratio of B Subordination to Future Expected Loss

0.1 0.8 1.4 0.4 0.8 0.3 0.2 0.0 0.5 0.3 0.0 0.0 0.7 0.2 0.9 0.0 0.3 0.5 0.4 0.5 0.5 0.3 0.4 0.1 0.6 2.0 0.0 0.0 0.3 0.5 1.9 1.0 0.3 0.6 0.8

0.4 0.5 0.7 0.3 0.3 0.5 0.3 0.4 0.3 0.7 0.4 0.3 0.4 0.3 0.2 0.3 0.2 0.3 0.2 0.3 0.2 0.3 0.2 0.1 0.2 0.4 0.2 0.2 0.2 0.2 0.3 0.4 0.5 0.2 0.3

2.9 3.6 5.6 2.5 2.8 4.1 2.5 3.6 2.3 6.7 4.0 2.7 3.9 3.0 2.8 2.9 2.5 3.2 3.0 3.5 2.8 4.0 2.3 2.1 3.5 6.7 2.7 2.8 2.8 3.1 6.6 8.8 11.4 5.6 6.2

7.4 7.9 8.0 8.2 8.3 8.5 9.1 9.1 9.1 10.0 10.3 10.3 10.3 10.8 11.2 11.2 11.4 11.9 12.4 12.5 12.9 13.1 13.9 14.2 14.6 16.0 17.5 17.9 18.0 18.3 19.3 20.3 22.3 23.1 24.6

Please see additional important disclosures at the end of this report.

175

Transforming Real Estate Finance chapter 9

CMBS Conduit Subordination Levels exhibit 17

Deal Name FULB 1997-C1 ACMF 1997-C1 MCFI 1997-MC2 ASC 1997-D5 CSFB 1997-C1 GMACC 1998-C1 MSC 1997-HF1 MSC 1999-WF1 NLFC 1999-2 MSC 1998-WF2 MSC 1997-WF1 JPMC 1997-C4

1997-2001 CONDUIT DEALS RANKED BY SUBORDINATION-TO-LOSS RATIO (34% SEVERITY; AVERAGE DEFAULT TIMING CURVE) (CONTINUED)

Class

Current Moody’s Rating

Current Fitch Rating

30 Days Del (%)

60 Days Del (%)

90 Days Del (%)

FOR. (%)

REO (%)

H J H A6 G J H L H K H G

B2 -

B-* B B B B* B B B* B+ B* B*

0.00 9.58 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.74 0.00 0.00 0.00 0.00 0.31 1.17 0.00 0.00 0.00 0.00 0.00

1.35 6.74 1.41 0.00 2.67 0.18 1.52 0.00 0.00 0.00 0.00 0.43

1.40 0.00 0.00 2.90 0.19 0.00 0.00 0.00 0.00 0.00 0.00 0.00

0.73 0.00 0.00 0.88 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Source: Morgan Stanley, Trepp, Bloomberg * S&P Rating

176

Perf. Spec. Serv. (%)

Watchlist (%)

Expected Loss for Currently Delinquent, Watchlisted or Performing Specially Serviced Loans (% of orig. bal.)

2.69 0.00 2.21 0.82 9.08 18.01 0.00 0.00 1.45 1.18 2.12 0.00

23.65 32.57 19.53 16.78 18.31 0.00 0.00 2.90 19.41 11.26 6.03 0.00

0.8 1.1 0.4 1.2 0.6 0.3 0.2 0.0 0.1 0.1 0.1 0.0

Historical Cumulative Loss (% of orig. bal.)

Future Expected Losses, Based on Historical Losses and Exp. Loss for Current Delinquency Levels (% of curr. bal.)

Current B Subordination (%)

Ratio of B Subordination to Future Expected Loss

0.3 0.0 0.6 1.6 0.9 0.3 0.4 0.1 0.0 0.0 0.0 0.0

0.2 0.2 0.1 0.3 0.2 0.2 0.1 0.1 0.0 0.0 0.0 0.0

4.5 5.1 3.4 9.2 6.0 5.1 3.9 2.2 3.8 3.2 3.9 5.6

25.0 26.6 27.1 27.2 27.4 29.4 33.9 35.8 85.9 118.2 434.0 678.9

Please see additional important disclosures at the end of this report.

177

This Page Intentionally Left Blank

178

Chapter 10

Transforming Real Estate Finance

Multifamily MBS

Please see additional important disclosures at the end of this report.

179

Transforming Real Estate Finance chapter 10

Multifamily MBS FANNIE M AE

In addition to “private-label” CMBS, the CMBS market also encomes multifamily agency securities, issued by Fannie Mae, Ginnie Mae and Freddie Mac. The single largest issuer of multifamily MBS is Fannie Mae. The main Fannie Mae program is Delegated Underwriting and Servicing (DUS). THE F ANNIE M AE D US P ROGRAM

Fannie Mae created the DUS program in 1988 to streamline the underwriting process and help fulfill its commitment to multifamily housing. Under the program, specially approved lenders may underwrite, close, service and sell mortgages to Fannie Mae without prior review by Fannie Mae. DUS lenders benefit from this special relationship because they have greater autonomy in underwriting and servicing and can also be more competitive given that DUS loan rates are lower than in the prior approval program. Before this program, the process was lengthier given that the agency had to underwrite and approve the transaction in advance of purchase. CHARACTERISTICS O F D US L OANS

Loans originated under the DUS program are generally either fixed-rate balloon mortgages with 5-, 7-, 10-, or 15-year or fixed-rate fully amortizing loans with 25- or 30-year . Variations, such as 20-year fully amortizing loans, are also permissible. The loans are secured by mortgages on income-producing, multifamily rental or cooperative buildings with at least five units and with occupancy rates of at least 90%. The buildings may be existing or recently completed and may require moderate rehabilitation. Loan amounts are $1 million to $50 million. There is always a loss sharing agreement between Fannie Mae and DUS lenders in case of default. Loans must have been originated within 6 months of Fannie Mae’s purchase. PREPAYMENT P ROTECTION

One of the main advantages of multifamily securities over residential MBS is the prepayment protection on multifamily loans. Most DUS loans have yield maintenance s in the event of an early prepayment. The is usually yield maintenance calculated at the relevant treasury rate, or “Treasuries flat.” Common yield maintenance are: • Balloon Term (years) 5 7 10 15 30

180

Yield Maintenance Term (years) 3 or 4.5 5 or 6.5 7 or 9.5 10 10

After the yield maintenance period ends, the borrower is still required to pay a 1% on prepayment which is retained by Fannie Mae. This is waived during the last 90 days of the loan term to facilitate refinancing. Curtailments are not allowed and, consequently, the borrower is faced with the choice of either prepaying the entire loan balance or not at all. Prepayment fees are ed through to the investor by Fannie Mae only to the extent they are received from the borrower. Fannie Mae’s obligation extends only to the outstanding principal balance of the security, i.e., if an MBS DUS defaults as a security, the investor receives a minimum of par but may lose some or all of the . Most DUS loans can be assumed by a new, and creditworthy, borrower on payment of a 1% assumption fee that is not ed through to the investor. Given that the pricing speed assumption of DUS is usually 0% R, the assumability option does not add any negative convexity to the security. The prepayment fee actually due from the borrower is calculated by substituting the note rate for the coupon in the above calculation. The difference between the fee received and the fee paid to the investor is shared by FNMA and the lender. UNDERWRITING

DUS lenders have strong incentives to underwrite high quality loans. First and foremost, Fannie Mae monitors the performance of their DUS lenders. In addition, when a DUS loan defaults, losses up to the first 5% of the UPB are borne solely by the lender and losses in excess of 5% are shared by Fannie Mae and the lender according to a formula. The lender’s share of the loss is limited to 20-40% of the UPB. The yield maintenance is also part of the loss computation giving the lender a vested interest in enforcing payment of the . For pricing and underwriting purposes, Fannie Mae categorizes DUS loans into one of four credit “tiers” based on debt service coverage and loan-to-value ratios. Tier 4 loans have the highest credit quality, while Tier 1 loans have the lowest. Most DUS loans tend to fall into the middle two tiers. Tier 1 loans are extremely rare. Fannie Mae may also designate loans with a ‘+’ in each category, based on subjective criteria such as property location and management. A ‘+’ reduces the guarantee fee by about 10 bp. exhibit 1 DUS UNDERWRITING TIERS

Tier Tier Tier Tier

1 2 3 4

Minimum Debt Service Coverage Ratio 1.15 1.25 1.35 1.55

Maximum Loan-to-Value Ratio 80% 80% 65% 55%

Note: Most DUS loans are in Tier 2 or Tier 3. Source: Fannie Mae

Please see additional important disclosures at the end of this report.

181

Transforming Real Estate Finance chapter 10

Multifamily MBS FANNIE M AE D MBS

In 1996, Fannie Mae began issuing discount MBS (DMBS) as a means of selling multifamily loans in the secondary mortgage market. Fannie Mae started routinely securitizing multifamily loans on a programmatic basis in 1994, but these securities did not possess the characteristics of DMBS. DMBS C HARACTERISTICS

DMBS are Fannie Mae’s only short-term, non-interest bearing securities that are collateralized by mortgages. Since DMBS are non-interest bearing securities, they are sold to investors at a discount and repaid at par upon maturity. DMBS maturities generally range between three and nine months, with occasional exceptions. exhibit 2

DMBS ISSUANCE

Notes: 1As of February 1, 2002. Issuance volumes prior to 2000 only represent DMBS with 3-month maturities. Source: Fannie Mae

To date, almost all DMBS issuance has consisted of three-month securities. Since DMBS were first issued in 1996, more than $18 billion securities have been sold. During 2001, DMBS issuance totaled $8.6 billion, a 38% increase over issuance in 2000. Despite low interest rates, many borrowers chose to maintain financial flexibility by using short term financing. For 2002, Fannie Mae projects DMBS issuance to be close to 2001 levels. As of February 1st, 2002, Fannie Mae issued $681 million of DMBS. FANNIE M AE G UARANTEE

DMBS, like all other Fannie Mae securities, carry Fannie Mae’s guarantee of full and timely payment of principal. DMBS are not rated by the rating agencies but are equivalent to agency credits. In the event of a principal shortfall, payments on DMBS are pari u with other senior debt of Fannie Mae.

182

LOAN C HARACTERISTICS

While single-asset executions are available, DMBS are most often secured by a pool of cross-collateralized, cross-defaulted, first-lien mortgages on incomeproducing residential properties with at least 5 units. These multifamily mortgages are typically underwritten to conform to the Fannie Mae Delegated Underwriting and Servicing (DUS) requirements. DMBS loans tend to be more conservative than multifamily loans in conduit transactions, in of loan-to-value (LTV) ratios and debt service coverage ratios (DSCR). LTVs on DMBS collateral range between 50% and 75%, while DSCRs typically range between 1.65X and 1.30X. In typical CMBS conduit transactions, multifamily loans tend to have 1.25X DSCR and 75% LTVs. The loans ing DMBS are extended to borrowers, such as REITs and pension funds, under a credit facility agreement. The credit facility provides borrowers with short-term advances that are funded by the sale of DMBS. Borrowers receive proceeds from DMBS issuance, which are determined by market discount. These loans mimic variable-rate financing, as they may be “rolled” every few months when the DMBS mature. The facility term may be 5, 7 or 10 years in length. In Exhibit 3, we have illustrated an example where a borrower requests a shortterm advance of $100 million, with $200 million of multifamily properties as collateral. Fannie Mae issues $100 million of DMBS, and the market discount of 0.5% on the securities results in the borrower receiving $99.5 million.

exhibit 3

MULTIFAMILY CREDIT FACILITY

1

Assumes $99.5MM discounted proceeds from sale of DMBS.

Source: Fannie Mae

Please see additional important disclosures at the end of this report.

183

Transforming Real Estate Finance chapter 10

Multifamily MBS At maturity, the investor will receive $100 million, via proceeds from new DMBS issuance, or a payoff by the borrower. PREPAYMENTS A ND D EFAULTS

DMBS are locked out from prepayments for their entire term and, therefore, have no prepayment risk. Default risk is also non-existent to DMBS investors, as Fannie Mae guarantees payments of principal when due. To date, there have been no defaults, but in the case of default, Fannie Mae would pay DMBS investors par at maturity. TRADING L EVELS

Discounts on new issue three-month DMBS may be converted to annualized yields or spreads to LIBOR. Historically, three-month DMBS have yielded LIBOR less 15-16 bp. During the first few weeks of 2002, three-month DMBS have been issued with average yields of LIBOR – 6 bp. In the secondary market, Morgan Stanley traded nearly $5 billion of DMBS in 2001, and $700 million in 2002 YTD. We estimate that Morgan Stanley is involved in 20-25% of DMBS trades in the secondary market.

exhibit 4

DMBS TRADES: THREE-MONTH MATURITY

Source: Fannie Mae

INVESTOR B ASE

DMBS are purchased primarily by money market investors who view these securities as an attractive alternative to Treasury Bills. Fannie Mae DMBS are similar to Treasuries, in that they are permitted investments for federally supervised institutions and for trusts and funds invested under the authority of the U.S. DMBS can also be purchased in unlimited amounts by national banks, federally chartered credit unions and federal savings and loans associations.

184

GINNIE M AE/FHA

Within the agency multifamily market, the second largest issuer in the agency market is the Government National Mortgage Association (GNMA). Project loans may be made under a number of Housing and Urban Development Department (HUD) programs, including: • 221(d)4: • 223(f): • 223(a)7: • 232: • 241(f):

Construction or permanent financing Refinancing Accelerated refinancing Nursing home/assisted living Equity take-out second mortgage

All Ginnie Mae securities are backed by loans originated by the Federal Housing istration (FHA) and are either permanent loan certificates (PLCs) or construction loan certificates (CLCs). PLCs are usually 35-year fully amortizing fixed-rate mortgages. Prepayment protection is either (1) a 5-year lockout followed by declining penalty points (5, 4, 3, 2, 1) over the next five years or (2) a 10-year lockout. Many PLCs begin as CLCs and are converted to PLCs upon completion of the construction project. CLCs trade at wider spreads than PLCs because of liquidity and uncertainty associated with funding a construction loan. Exhibit 5 shows some of the major differences between Ginnie Mae and FHA project loans: CHARACTERISTICS OF GINNIE MAE AND FHA exhibit 5 PROJECT LOANS Government guarantee Principal payment in case of default Delay days Delivery Data on Bloomberg

Ginnie Mae Explicit

FHA Implicit

100% 44 PTC Yes

99% 54 Physical No

Source: Fannie Mae

Effective April 1, 1997, Ginnie Mae reduced pool processing time from 10 days to 5 days and added other features to streamline its multifamily MBS program.1

1

See inside MBS & ABS, May 1, 1997, p.3.

Please see additional important disclosures at the end of this report.

185

Transforming Real Estate Finance chapter 10

Multifamily MBS exhibit 6 Issuer

A PROFILE OF GINNIE MAE MULTIFAMILY MBS Ginnie Mae approved mortgage lender

Issue Type

GNMA I

Underlying Mortgages

FHA Insured multifamily mortgages

Pool Types

Securities Interest Rate Guaranty Guarantor Principal and Interest Maturity Minimum Certificate Size Transfer Agent

Construction Loan Securities (CL) Security rate remains constant with conversion to permanent loan (CS) Security rate changes with conversion to permanent loan Project Loan Securities (PL) Level payment permanent securities (PN) Non-level payment permanent securities (LM) Securities for Mature Loans. Loans pooled after more than 24 months of amortization (LS) Securities for Small Loans. Loans of no more than $1MM Fixed; at .25 to .50 percent below the interest rate of the underlying mortgage(s) Full and timely payment of principal and interest Ginnie Mae (full faith and credit of the United States) Paid monthly to securities holders Varies, typically 40 years $25,000 (may be less for aged securities) Chase (formerly Chemical Bank)

Source: Reprinted from Ginnie Mae website, www.ginniemae.gov, 1997

FREDDIE M AC

Freddie Mac, a large issuer in the 1980s, reduced its role in the multifamily securitization market in the 1990s. The agency has recently begun to increase its multifamily loan production.

186

PROGRAM P LUS

Freddie Mac’s Program Plus is similar to Fannie Mae’s DUS program. Under the program, Freddie Mac pre-approves multifamily lenders with “local market expertise.” Since 1993, Freddie Mac has financed $5.3 billion (1,400 properties) under Program Plus. To be eligible for Program Plus, loans must be between $5 million and $50 million and have the following characteristics: • of 7, 10, 15, 20, or 25 years • Amortization period of 20, 25, or 30 years • Maximum LTV of 75% • Minimum DSCR of 1.3 The yield maintenance of the loans are: • Term(years)/Yield Maintenance(years) 7/6.5 10/9.5 15/14 20/15

Please see additional important disclosures at the end of this report.

187

This Page Intentionally Left Blank

188

Chapter 11

Transforming Real Estate Finance

Floating Rate Large Loan CMBS

Please see additional important disclosures at the end of this report.

189

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS In this chapter, we examine three aspects of floating rate CMBS. In Section I, we examine the incidence of extensions in the floating rate large loan universe. In Section II, we examine whether floating rate CMBS deals are inclined to experience a negative ratings drift over time. In Section III, we describe the mechanics of available funds caps in the context of a hypothetical floating rate CMBS structure.

Section I: Floating Rate Large Loan Extensions MOTIVATIONS F OR E XTENDING

In this section, we examine the incidence of extensions in the floating rate large loan universe as well as the motivations for exercising extension options (i.e., credit vs. non-credit related reasons). Our main findings are that nearly a quarter of loans exercised extension options at maturity, and that the majority of extensions were motivated by credit issues. Floating rate loans within CMBS transactions typically have a substantial principal balance remaining at the time of maturity. In order to repay this balance, a borrower may refinance the loan or sell the underlying property. If the loan’s include an extension option, the borrower may instead choose to postpone repayment of the loan. Whereas a borrower’s incentive to prepay a loan in the residential mortgage market is heavily dependent on prevailing interest rates, a borrower’s incentive to extend a loan in the commercial mortgage market is predominantly driven by credit-related issues. A borrower’s decision to extend a loan could be a result of credit problems, such as the property’s weak financial condition or deterioration in the property’s operating performance. The borrower may also need to extend for a variety of other reasons, unrelated to credit. For example, the borrower may find refinancing prior to the loan’s maturity, but need to extend because the new loan’s closing date is beyond the original loan’s maturity date. Another possibility is that the borrower believes that the underlying property’s financial performance could improve over the next year and allow the borrower to refinance at lower rates than the property currently allows. Additionally, improved financials could result in higher valuations, which may result in additional proceeds to the borrower. If commercial lending is scarce, the borrower may also have a difficult time finding refinancing and need to extend.

190

There also may be some instances in which the level of interest rates motivates the borrower to extend. That is, the borrower holds the view that interest rates will decline in the coming year, allowing the borrower to access cheaper rates in the future. 23% O F L OANS E XERCISE E XTENSION O PTIONS A T M ATURITY

For our analysis, we examined the universe of floating rate large loan CMBS issued since January 2000. The majority of loans within this sample were issued with extension options. Extendable loans for 69% of the universe, based on loan count, and 86% of the universe, based on loan balance. To date, about half of the extendable loan balances have reached maturity. Of the extendable loan balances that have reached maturity, 23% have extended. Based on loan count, the results were similar, with 27% of loans exercising extension options. exhibit 1

LOAN EXTENSIONS WITHIN FLOATING RATE CMBS (BASED ON ORIGINAL BALANCE)

Source: Morgan Stanley, Intex, Trepp

COMPOSITION O F E XTENDED L OANS

In all, we found 112 matured loans that exercised extension options. Office loans ed for nearly half of the extended loan pool, while hotel loans ed for 16% of the loans, and multifamily loans ed for 14% of the loans. Although office loans comprise a large portion of the extended loan pool, they did not experience the highest extension rate of the major property types.

Please see additional important disclosures at the end of this report.

191

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS exhibit 2

PROPERTY TYPES OF EXTENDED LOANS BASED ON SECURITIZED LOAN BALANCES

1

Includes industrial, manufactured housing, mixed use, and senior housing loans.

Source: Morgan Stanley, Intex, Trepp

HOTEL L OANS E XPERIENCE H IGHEST E XTENSION R ATE

While we found that on average, 23% of loan balances exercised extension options, we found that this experience varied depending on collateral type. For example, based on loan balances, hotel loans exercised 40% of extension options, while retail loans exercised 8%. (See Exhibit 3) exhibit 3

EXTENSION RATES BY PROPERTY TYPE (BASED ON ORIGINAL BALANCE)

Property Type Hotel Office Multifamily Retail Other2 Total

Extension Rate (%)1 40 30 18 8 13 23

1

Represents percentage of loans that exercised extension options at maturity.

2

Includes industrial, manufactured housing, mixed use, and senior housing loans.

Source: Morgan Stanley, Intex, Trepp

This observation is not surprising since we also determined that the majority of extended loans (68%) chose to extend for credit-related reasons. Over the past couple of years, the retail sector has remained healthy, while the hotel sector has suffered a downturn. As a result, we found the higher incidence of hotel loan extensions to be linked to deterioration in operating performance and lower occupancy rates.

192

METHODOLOGY

In order to determine the rate at which borrowers exercise extension options on floating rate large loans, we examined floating rate large loan CMBS transactions that were issued since January 2000. We identified 76 floating rate loans within this universe that were flagged with a CMSA modification code of 1 (maturity date extension) or CMSA workout code of 4 (extension) in Trepp or Intex. Unfortunately, the servicer files, which are the source for this extension information, do not always flag all of the extended loans. In order to capture extended loans that were not flagged in the servicer files, we searched for loans that were still outstanding at least one year after the original maturity date. Using this methodology, we identified an additional 53 loans. In total, we were able to identify 129 floating rate loans that extended. We eliminated 17 of the 129 loans, which did not have extension options, and limited our analysis to the remaining 112 extendable loans. In order to determine the motivations for exercising extension options on these 112 loans, we examined changes in property performance (DSCR and occupancy) since loan origination. In addition, we considered servicer commentary in monthly watchlist files and consulted Realpoint analysts for their opinions. We found that 75 loans (68%) exercised extension options due to credit problems.

Please see additional important disclosures at the end of this report.

193

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS Section II: A Discussion of Drift In this section, we examine whether floating rate CMBS deals are inclined to experience negative ratings drift over time. Floating rate loans appeal to CMBS borrowers, in part, because they provide flexibility for prepayment or extension. A borrower can replace floating rate financing with longer term fixed rate financing, if a property has stabilized prior to maturity. Likewise, a borrower may exercise an extension option and buy more time to improve a property’s cash flows. These structural features, however, seem to imply that floating rate CMBS deals may ultimately be left with a pool of weak assets, if all of the stabilized assets prepay or pay off as scheduled on their maturity dates. We address this topic and examine whether floating rate CMBS deals are inclined to experience a negative ratings drift over time. We tackle this question by examining the ratings history of floating rate CMBS deals issued in 2000. We choose to examine the 2000 vintage, as we specifically want to focus on the performance of transactions after significant loan paydowns have occurred. Floating rate CMBS tend to be short transactions (~3 years). Therefore, the 2000 vintage should have experienced significant paydowns, as these deals have seasoned for 3-4 years. OUR F INDINGS

For our analysis, we examined the ratings drift of 224 classes. Since issuance, these 224 classes have experienced 118 downgrades by the rating agencies. With the exception of four downgrades on GSMS 2000-GSFL, which were related to inadequate terrorism insurance, all downgrades resulted from the deterioration of remaining loans. This seems to the observation that floating rate CMBS ratings deteriorate over time as loans pay down. We found, however, that these paydowns also increase credit enhancement on many classes. In the case of the 2000 floating rate cohort, paydowns resulted in 1401 upgrades. Over time, the 2000 floating rate vintage experienced a positive upgrade/downgrade ratio of 1.2 to 1. A positive ratio does not necessarily imply a positive ratings drift across all classes. In order to determine which classes have positive and negative ratings drift, we examined the rating actions on non-rake floating rate tranches issued in 2000. We did not to include rake classes, which are tied to the performance of a single loan. Almost all rating actions on rake classes have been downgrades (29 tranches downgraded and 1 tranche upgraded).

1

The 224 classes experienced a total of 144 upgrades (140 due to increased credit enhancement; 4 due to improved of terrorism insurance in GSMS 2000-GSFL).

194

exhibit 4

AVERAGE NOTCH DRIFTS FOR 2000 FLOATING RATE CMBS1

Original Rating AAA AA+ AA AAA+ A ABBB+ BBB BBBBB+ BB BBB+ B BCCC+ CCC CCCTotal

Average Notch Drift1 0.0 1.0 1.7 2.0 0.8 0.8 -3.7 3.3 3.1 3.6 4.5 1.2 3.3 3.0 1.9 0.3 NA 3.0 NA 1.6

1

Data set only includes non-rake tranches that have been upgraded/downgraded by at least one rating agency.

Source: Morgan Stanley, Moody’s, S&P, Fitch

We began by aggregating non-rake tranches into buckets, based on their original ratings. We then calculated the average notch drift of the tranches in each of these original rating buckets. If a tranche was upgraded/downgraded by more than one rating agency, we used the lowest ratings drift to get the most conservative result. Therefore, if a class was upgraded 2 notches by Moody’s and 3 notches by S&P, we used the 2 notch drift. Likewise, if a class was downgraded 2 notches by Moody’s and 3 notches by S&P, we used the -3 notch drift. CONCLUSION

We found that each category, with the exception of A-, experienced a positive average ratings drift since issuance. (see Exhibit 4)

Please see additional important disclosures at the end of this report.

195

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS The downward drift of the A- category was driven by downgrades on MSDWC 2000-XLF D, COMM 2000-FL2 F and COMM 2000-FL3 D. The latter two classes were the lowest rated non-rake tranches in those structures. We also found that although the ratings drift for floating rate CMBS classes has been positive as a whole, the ratings drift pattern differs at each rating agency. As shown in Exhibit 6, classes that were originally rated lower than BBB+ by S&P exhibit a negative ratings drift. Fitch rating actions, on the other hand, have resulted in positive ratings drift across almost every original rating category.

exhibit 5

NUMBER OF TRANCHES UPGRADED/ DOWNGRADED BY EACH RATING AGENCY1 Moody’s

Upgrades Full Rating Category2 AAA AA A BBB BB B CCC Total

# of Tranches 0 9 8 9 5 1 0 32

% of Tranches 0% 28% 25% 28% 16% 3% 0% 100%

# of Tranches 0 0 6 3 2 2 0 13

1

Does not include upgrades/downgrades on rake tranches.

2

Includes plus and minus designations at each rating level.

Source: Morgan Stanley, Moody’s, S&P, Fitch

196

S&P Downgrades % of Tranches 0% 0% 46% 23% 15% 15% 0% 100%

Upgrades # of Tranches 0 7 7 3 1 0 0 18

% of Tranches 0% 39% 39% 17% 6% 0% 0% 100%

exhibit 6

AVERAGE NOTCH DRIFTS FOR 2000 FLOATING RATE CMBS1 BY RATING AGENCY

Original Rating AAA AA+ AA AAA+ A ABBB+ BBB BBBBB+ BB BBB+ B BCCC+ CCC CCCTotal

Moody’s 0.0 NA 1.6 2.0 -2.0 0.3 -4.7 4.5 0.0 5.0 7.0 1.0 3.0 NA 2.0 -1.0 NA NA NA 1.1

S&P NA 1.0 1.8 3.0 2.0 0.3 -2.6 1.0 -4.0 -0.5 -1.0 -1.3 -2.0 -3.0 -2.0 -2.0 NA -1.0 NA -0.7

Fitch NA 1.0 2.0 2.0 2.3 3.1 -0.3 7.0 7.7 6.5 10.0 3.6 2.5 3.5 2.2 1.5 NA 7.0 NA 3.5

1

Data set only includes non-rake tranches that have been upgraded/downgraded by at least one rating agency.

Source: Morgan Stanley, Moody’s, S&P, Fitch

S&P

Fitch

Downgrades # of Tranches 0 0 5 3 4 4 1 17

% of Tranches 0% 0% 29% 18% 24% 24% 6% 100%

Upgrades # of Tranches 0 14 11 12 7 5 1 50

% of Tranches 0% 28% 22% 24% 14% 10% 2% 100%

Downgrades # of Tranches 0 0 2 1 3 4 0 10

Please see additional important disclosures at the end of this report.

% of Tranches 0% 0% 20% 10% 30% 40% 0% 100%

197

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS exhibit 7

2000 FLOATING RATE CMBS RATING ACTIONS SINCE ISSUANCE Moody’s Class

Old Rating

Action

COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1 COMM 2000-FL1

B C D E F G G H H H J J J K

Aa2 Aa2 A2 A2 Baa1 A3 Baa3 Aa2 Baa2 Baa3 A2 Baa3 Ba1 Ba2

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

Aaa Aaa Aa2 Aa2 A2 Aaa A3 Aaa Aa2 Baa2 Aa3 A2 Baa3 Ba1

12/13/2001 12/13/2001 12/13/2001 12/13/2001 12/13/2001 10/24/2002 12/13/2001 4/7/2003 10/24/2002 12/13/2001 4/7/2003 10/24/2002 12/13/2001 12/13/2001

COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL2 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3 COMM 2000-FL3

B D E E F F GCO GHM GLP GNS GNS GNS GWH HCO HHM HLP HNS HNS HNS HWH JCO JHM JLP JNS JNS JNS JWH KQA KSR KWC KWS LQA LSR LWS

Aa2 A1 Baa2 A2 Baa3 A3 Baa1 Baa1 Baa1 Ba3 Ba1 Baa1 Baa1 Baa2 Baa2 Baa2 B1 Ba2 Baa2 Baa2 Baa3 Baa3 Baa3 B2 Ba3 Baa3 Baa3 Baa1 Baa1 Baa1 Baa1 Baa2 Baa2 Baa2

Upgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

Aaa A3 Ba1 Baa2 Ba3 Baa3 Baa2 Baa3 Ba1 B1 Ba3 Ba1 Ba3 Baa3 Ba1 Ba2 B2 B1 Ba2 B1 Ba1 Ba2 Ba3 B3 B2 Ba3 B2 Ba1 Ba2 B1 Ba1 Ba2 Ba3 Ba2

9/11/2003 10/24/2002 9/11/2003 10/24/2002 9/11/2003 10/24/2002 10/24/2002 10/24/2002 10/24/2002 9/11/2003 6/18/2003 10/24/2002 10/24/2002 10/24/2002 10/24/2002 10/24/2002 9/11/2003 6/18/2003 10/24/2002 10/24/2002 10/24/2002 10/24/2002 10/24/2002 9/11/2003 6/18/2003 10/24/2002 10/24/2002 5/30/2003 5/30/2003 5/30/2003 5/30/2003 5/30/2003 5/30/2003 5/30/2003

B4 B5

Ba2 Ba3

Upgrade Upgrade

Baa2 Baa3

9/11/2003 9/11/2003

C

Aa2

Upgrade

Aaa

9/3/2003

CMBS Transactions

New Rating

Date of Rating Action

Chase Commercial Mortgage Securities Corp, 2000-FL1 Chase Commercial Mortgage Securities Corp, 2000-FL1 Chase Commercial Mortgage Securities Corp, 2000-FL1 Chase Commercial Mortgage Securities Corp, 2000-FL1 Chase Commercial Mortgage Securities Corp, 2000-FL1 Chase Commercial Mortgage Securities Corp, 2000-FL1

DLJ Commercial Mortgage Corp, 2000-STF1 DLJ Commercial Mortgage Corp, 2000-STF1 DLJ Commercial Mortgage Corp, 2000-STF1 DLJ Commercial Mortgage Corp, 2000-STF1 DLJ Commercial Mortgage Corp, 2000-STF1 DLJ Commercial Mortgage Corp, 2000-STF1 DLJ Commercial Mortgage Corp, 2000-STF1 GMAC Commercial Mortgage Asset Corp, 2000-FL-A

Source: Moody’s, S&P, Fitch, Intex

198

S&P Class

Old Rating

Action

Fitch New Rating

Date of Rating Action

C D E F G H

A ABBB BBBBB B

Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

BBB+ BBB BB BBB B-

11/13/2003 11/13/2003 11/13/2003 11/13/2003 11/13/2003 11/13/2003

B C C D F G H J K

AA AA+ AAA A ABBB BBBBB

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAA AAA AA+ AA AAA AAA A+ BBB+ BB+

10/16/2001 12/14/2001 10/16/2001 10/16/2001 8/16/2002 8/16/2002 8/16/2002 8/16/2002 8/16/2002

B D KHS KWC LHS

B

AA ABBB+ BBB+ BBB-

AA

Upgrade Downgrade Downgrade Downgrade Downgrade

Upgrade

AAA BBBBB+ B+ BB-

AA+

5/19/2003 5/19/2003 5/19/2003 5/19/2003 5/19/2003

5/27/2003

Class

Old Rating

Action

New Rating

B C C D GCO GWH GWH HCO HWH HWH JCO JWH JWH

AA A A+ A BBB+ BB+ BBB+ BBB BB BBB BBBBBBBB-

Upgrade Upgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

AAA AA A BBB+ BB+ B+ BB+ BB B BB BBBBB-

10/14/2003 10/14/2003 1/15/2003 1/15/2003 1/15/2003 1/15/2003 4/12/2002 1/15/2003 1/15/2003 4/12/2002 1/15/2003 1/15/2003 4/12/2002

B D D D KHS KQA KSR KSR LHS LQA LSR LSR MHS

AA BB BBBABBBBBB+ BBB BBB+ BBB BBB BBBBBB BBB-

Upgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

AAA BBBB BBBBB BB+ B BBB BB BB BBBBBB-

10/16/2003 4/2/2004 10/16/2003 4/22/2003 10/16/2003 10/16/2003 4/2/2004 10/16/2003 10/16/2003 10/16/2003 4/2/2004 10/16/2003 10/16/2003

A2 A3 B1 B2 B4 B5 B6

AA A BBB BBBBB BBB

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAA AAA AAA AAA A BBB BB-

8/14/2003 8/14/2003 8/14/2003 8/14/2003 8/14/2003 8/14/2003 8/14/2003

B

AA+

Upgrade

AAA

9/3/2002

Please see additional important disclosures at the end of this report.

Date of Rating Action

199

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS exhibit 7

2000 FLOATING RATE CMBS RATING ACTIONS SINCE ISSUANCE (CONTINUED) Moody’s Class

Old Rating

Action

D

A2

Upgrade

A1

9/3/2003

GMAC Commercial Mortgage Asset Corp, 2000-FL-F GMAC Commercial Mortgage Asset Corp, 2000-FL-F GMAC Commercial Mortgage Asset Corp, 2000-FL-F GMAC Commercial Mortgage Asset Corp, 2000-FL-F GMAC Commercial Mortgage Asset Corp, 2000-FL-F GMAC Commercial Mortgage Asset Corp, 2000-FL-F GMAC Commercial Mortgage Asset Corp, 2000-FL-F

B C

Aa2 A2

Upgrade Upgrade

Aa1 A1

4/14/2003 4/14/2003

GMAC Commercial Mortgage Securities, 2000-FL1 GMAC Commercial Mortgage Securities, 2000-FL1 GMAC Commercial Mortgage Securities, 2000-FL1 GMAC Commercial Mortgage Securities, 2000-FL1

C D E F

A2 Baa2 Baa3 Ba2

Downgrade Downgrade Downgrade Downgrade

Baa1 Ba1 Ba2 B2

1/26/2004 1/26/2004 1/26/2004 1/26/2004

A A F G3 X0 X0 X1 X1 X2 X2

Aa1 Aaa Baa2 Baa3 Aa1 Aaa Aa1 Aaa Aa1 Aaa

Upgrade Downgrade Upgrade Upgrade Upgrade Downgrade Upgrade Downgrade Upgrade Downgrade

Aaa Aa1 Baa1 Baa1 Aaa Aa1 Aaa Aa1 Aaa Aa1

10/29/2003 9/27/2002 10/29/2003 10/29/2003 10/29/2003 9/27/2002 10/29/2003 9/27/2002 10/29/2003 9/27/2002

B C G H J

Aa2 A2 Ba2 B2 B3

Upgrade Upgrade Downgrade Downgrade Downgrade

Aa1 A1 Ba3 B3 Caa1

1/7/2003 1/7/2003 1/7/2003 1/7/2003 1/7/2003

B B C C

A2 Aa2 Baa2 A3

Upgrade Downgrade Downgrade Downgrade

Aa2 A2 Ba2 Baa2

9/11/2003 6/5/2003 6/5/2003 1/27/2003

CMBS Transactions GMAC Commercial Mortgage Asset Corp, 2000-FL-A

New Rating

Date of Rating Action

GMAC Commercial Mortgage Asset Corp, 2000-FL-B GMAC Commercial Mortgage Asset Corp, 2000-FL-B GMAC Commercial Mortgage Asset Corp, 2000-FL-B GMAC Commercial Mortgage Asset Corp, 2000-FL-B GMAC Commercial Mortgage Asset Corp, 2000-FL-B GMAC Commercial Mortgage Asset Corp, 2000-FL-B

GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III GS Mortgage Securities Corp II, 2000-GSFL III JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 JP Morgan Commercial Mortgage Finance Corp, 2000-FL1 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF

Source: Moody’s, S&P, Fitch, Intex

200

S&P Class C

Old Rating A

Action Upgrade

Fitch New Rating A+

Date of Rating Action 5/27/2003

B C D E F G H JBO KBO LBO M N P Q S T U

AA+ AA AAA+ A ABBB+ BBB+ BBB BBBBB+ BB BBB+ B BCCC

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

AAA AAA AAA AA A+ A ABBB+ B BB B+ B CCC+ CCC CCC CCC-

3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003 3/31/2003

C C

BBB A

Downgrade Downgrade

BB BBB

7/9/2003 2/11/2003

Class

Old Rating

Action

New Rating

B B C D E E F

AA

Upgrade

AA+

2/6/2002

A BBB BBBAA BB B

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAA AAA AAA AAA AA BB+

10/17/2002 10/17/2002 10/17/2002 7/30/2003 10/17/2002 10/17/2002

B B C C D E F

AA+ AA A+ A BBB BBBBB

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAA AA+ AAA A+ AA A BBB-

2/25/2004 9/10/2002 2/25/2004 9/10/2002 2/25/2004 2/25/2004 2/25/2004

B B C C C C D D D E E F F G H J

AA+ AA AA+ AA A+ A ABBB+ BBB BBB BBBBBBBB+ BB B B-

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Downgrade Downgrade

AAA AA+ AAA AA+ AA A+ AAA ABBB+ AAA BBB AAA BBBBBB CCC CC

5/16/2003 8/28/2002 3/17/2004 2/10/2004 5/16/2003 8/28/2002 3/17/2004 2/10/2004 8/28/2002 3/17/2004 8/28/2002 3/17/2004 8/28/2002 3/17/2004 5/16/2003 5/16/2003

B C D E F G H L M N P Q S

AA+ AA AAA+ A AABBBBBBB BBB+ B

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

AAA AAA AAA AAA AA AAA+ BB+ B+ B+ B BB-

8/20/2003 8/20/2003 8/20/2003 8/20/2003 8/20/2003 8/20/2003 8/20/2003 3/10/2003 3/10/2003 12/2/2002 12/2/2002 12/2/2002 12/2/2002

Please see additional important disclosures at the end of this report.

Date of Rating Action

201

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS exhibit 7

2000 FLOATING RATE CMBS RATING ACTIONS SINCE ISSUANCE (CONTINUED) Moody’s Class

Old Rating

Action

C D D D D E E E E F1 F1 F1

A2 B2 Ba2 Baa2 A3 Caa2 B3 Ba3 Baa2 Caa2 B2 Baa3

Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

A3 Caa1 B2 Ba2 Baa2 Ca Caa2 B3 Ba3 Ca Caa2 B2

9/6/2002 9/11/2003 6/5/2003 1/27/2003 9/6/2002 9/11/2003 6/5/2003 1/27/2003 9/6/2002 6/5/2003 1/27/2003 9/6/2002

Salomon Brothers Mortgage Securities VII, 2000-FL1 Salomon Brothers Mortgage Securities VII, 2000-FL1 Salomon Brothers Mortgage Securities VII, 2000-FL1 Salomon Brothers Mortgage Securities VII, 2000-FL1 Salomon Brothers Mortgage Securities VII, 2000-FL1 Salomon Brothers Mortgage Securities VII, 2000-FL1 Salomon Brothers Mortgage Securities VII, 2000-FL1

B C D E F F G

Aa2 A2 Baa2 Baa3 Ba1 Ba2 B2

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

Aaa Aaa A1 Baa1 A3 Ba1 Baa3

6/29/2001 6/29/2001 6/29/2001 6/29/2001 4/30/2002 6/29/2001 4/30/2002

SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2 SASCO Floating Rate Commercial Mortgage Trust, 2000-C2

D E F G H H J J K K

Aa2 Aa3 A2 A3 A2 Baa1 A3 Baa2 Baa1 Baa3

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

Aaa Aa1 Aa3 A1 Aaa A2 Aaa A3 Aa2 Baa1

3/1/2002 3/1/2002 3/1/2002 3/1/2002 10/31/2002 3/1/2002 10/31/2002 3/1/2002 10/31/2002 3/1/2002

CMBS Transactions Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF Morgan Stanley Dean Witter Capital I, 2000-XLF

Source: Moody’s, S&P, Fitch, Intex

202

New Rating

Date of Rating Action

S&P Class D D D E E F1

Old Rating CCC B ACCC BBB BBB-

Action Downgrade Downgrade Downgrade Downgrade Downgrade Downgrade

Fitch New Rating D CCC B D CCC D

Date of Rating Action

Class

Old Rating

Action

New Rating

B C C D D E E F F G G H H H J J J K K K L L M M N N P P Q Q S T

AA+ AA+ AA AA AAAAA+ A+ A A AA ABBB+ ABBB+ BBB BBB+ BBB BBBBBBBB+ BB+ BB BB BBBBB+ B+ B BCCC

Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade Upgrade

AAA AAA AA+ AAA AA AA AAAAA+ A+ A AAA A AAAA ABBB+ AAA BBB+ BBB AAA BBBAA BB+ AABB A+ BBA B+ BBB BB+

Date of Rating Action

11/6/2003 7/9/2003 2/11/2003 7/9/2003 2/11/2003 2/11/2003

Please see additional important disclosures at the end of this report.

10/8/2001 7/3/2002 10/8/2001 7/3/2002 10/8/2001 7/3/2002 10/8/2001 7/3/2002 10/8/2001 7/3/2002 10/8/2001 11/21/2002 7/3/2002 10/8/2001 11/21/2002 7/3/2002 10/8/2001 11/21/2002 7/3/2002 10/8/2001 11/21/2002 7/3/2002 11/21/2002 7/3/2002 11/21/2002 7/3/2002 11/21/2002 7/3/2002 11/21/2002 7/3/2002 11/21/2002 11/21/2002

203

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS Section III: Available Funds Cap Mechanics Available funds caps are a commonly discussed topic in the home equity world, as investors are concerned with rising rates and the basis risk between fixed rate collateral and floating rate bonds. Although CMBS structures are not typically faced with an interest rate mismatch between collateral and bonds, the concept of available funds cap does exist in some CMBS deals. In this section, we will describe the mechanics of available funds cap in the context of a hypothetical floating rate CMBS structure. HOW D OES A N A VAILABLE F UNDS C AP W ORK?

An available funds cap limits the coupon on specified CMBS bonds to the collateral weighted average coupon (WAC).1 The available funds cap is triggered on a CMBS class if the bond coupon for that class exceeds the weighted average collateral coupon. When such a scenario occurs, the cap initially limits the coupon on the affected class to the collateral coupon, but typically, any additional cash that remains after paying interest on all classes is paid sequentially to each capped class. This additional cash makes up the difference between the capped exhibit 8a

Loan 1 2

Balance ($MM) 500 500 1,000

exhibit 8b

Class A B C4 D4 X5

AVAILABLE FUNDS CAP EXAMPLE: ASSETS IN MONTH 11 Interest Rate2 LIBOR+70 LIBOR+130 LIBOR+100

AVAILABLE FUNDS CAP EXAMPLE: LIABILITIES IN MONTH 1

Class Size ($MM)

-Through Rate2

600 200 100 100 1,000 1,000

LIBOR+30 LIBOR+60 LIBOR+90 LIBOR+120 WAC/IO LIBOR+51

1

Interest only loans, 36-month term, 24-month lockout period. 1-Month LIBOR=1.10% 3 Difference between bond’s stated coupon and collateral WAC. 4 Subject to WAC Cap 5 Notional Amount Source: Morgan Stanley 2

1

The available funds cap is also referred to as a WAC cap.

204

Interest Payment ($MM) 0.750 1.000 1.750

Accrued Interest ($MM)

Accrued Interest Capped at WAC ($MM)

0.700 0.283 0.167 0.192

0.700 0.283 0.167 0.175

coupon payment and the original specified coupon payment. Any remaining cash after paying all of the coupons is ed to the IO. To illustrate this concept, consider a simple, hypothetical floating rate CMBS deal that is backed by two loans. These two loans have a weighted average coupon of LIBOR+100 bp. The liabilities consist of four classes and an IO. The very senior classes of a CMBS deal are typically not subject to a WAC cap, but the mezzanine classes may be. In our example, classes C and D have specified coupons of LIBOR+90 bp and LIBOR+120 bp and are subject to an available funds cap. In this hypothetical transaction, the collateral WAC in month 1 will pay the stated coupons on classes A, B and C. Class D, however, will be capped at the collateral WAC of LIBOR+100 bp, since its stated coupon (LIBOR+120 bp) exceeds the collateral WAC. After paying the full coupons on classes A through C and the capped coupon on class D, $0.425 million in collateral interest is still available for distribution. This is enough cash to pay class D an additional $0.017 million, such that it receives its full coupon payment. The remaining $0.408 million is distributed to the IO. The available funds cap may also be triggered in a scenario where the higher coupon loans pay off, leaving the lower coupon loans in the pool.

WAC Cap Shortfall3 ($MM)

Additional Interest Amount ($MM)

0.017

0.017

Interest Payment ($MM)

Interest Shortfall ($MM)

0.700 0.283 0.167 0.192 0.408 1.750

-

Please see additional important disclosures at the end of this report.

205

Transforming Real Estate Finance chapter 11

Floating Rate Large Loan CMBS In Exhibits 9a & 9b, we revisit the same structure but assume that one of the loans pays off, leaving a single loan with a coupon of LIBOR+70 bp. In this scenario, class C as well as class D triggers the available funds cap. After paying the capped coupons, there is enough cash to pay the full class C coupon. Class D, however, does not receive its full coupon of $0.192 million and only receives $0.183 million. Therefore, this class experiences an interest shortfall of $0.009 million. The IO in this case does not receive any interest. exhibit 9a

Loan 1 2

Balance ($MM) 500 500

exhibit 9b

Class A B C4 D4 X5

1

AVAILABLE FUNDS CAP EXAMPLE: MONTH 26 ASSETS1 Interest Rate2 LIBOR+70 LIBOR+130 LIBOR+70

AVAILABLE FUNDS CAP EXAMPLE: MONTH 26 LIABILITIES

Class Size ($MM)

-Through Rate2

100 200 100 100 500 500

LIBOR+30 LIBOR+60 LIBOR+90 LIBOR+120 WAC/IO LIBOR+72

Interest only loans, 36-month term, 24 month lockout period. 1-Month LIBOR=1.10% 3 Difference between bond’s stated coupon and collateral WAC. 4 Subject to WAC Cap 5 Notional Amount Source: Morgan Stanley 2

206

Interest Payment ($MM) 0.750 0.750

Accrued Interest ($MM)

Accrued Interest Capped at WAC ($MM)

0.117 0.283 0.167 0.192

0.117 0.283 0.150 0.150

WAC Cap Shortfall3 ($MM)

Additional Interest Amount ($MM)

0.017 0.042

0.017 0.033

Interest Payment ($MM)

Interest Shortfall ($MM)

0.117 0.283 0.167 0.183 0.750

0.009

Please see additional important disclosures at the end of this report.

207

This Page Intentionally Left Blank

208

Chapter 12

Transforming Real Estate Finance

Commercial Mortgage Defaults

Please see additional important disclosures at the end of this report.

209

Transforming Real Estate Finance Chapter 12

Commercial Mortgage Defaults This chapter contains our commercial mortgage default study which covers the time period 1972-2002. Our main conclusion is that most investment grade CMBS (with the exception of BBBs from more recent vintages) are well protected against the most severe real estate downturn of the last 30 years. The addition of two years of new data alters the original conclusion only slightly. In the past, all investment-grade CMBS were protected from the magnitude of losses experienced by life insurance company loans. With continued reductions in subordination levels, a greater proportion of recently rated BBB classes would now be vulnerable to a downturn of the magnitude of the late 1980s and early 1990s. In our updated study: • We added 1,383 new loans, increasing the total to nearly 18,000. • The average lifetime cumulative default rate (based on loan balances) for cohorts with at least 10 years of seasoning decreased from 20.5% to 19.6% over the last two years. • 1986 remained the worst origination year, with nearly 32% of the total balance eventually defaulting. • The average severity on liquidated loans was about 33%. • Of the cohorts with at least 10 years of seasoning, the 1991 and 1992 origination years had the lowest cumulative default rates. • For a given cohort, on average, the peak years for defaults were years 3-7 after origination. • About 55% of the defaulted loans were liquidated. BACKGROUND

Mark Snyderman authored two pioneering studies in 1991 and 1994 on cumulative lifetime default rates on commercial mortgages held by life insurance companies. His 1994 article tracked defaults (90+ day delinquent loans) on eight large insurance companies through 1991. The studies were the first to track commercial mortgage credit through several complete real estate cycles. Snyderman was able to follow the performance of loans originated in a given year (cohort) until all of the loans had either matured, prepaid, or defaulted. In 1999, Snyderman, L’Heureux, and Esaki (ELS) used the same insurance companies and data sources as the original studies to update the default data through 1997. The period of the ELS extension included the final years of the worst real estate downturn since the Great Depression of the 1930s. One of the findings of the study was that the 1986 cohort of originations was the worst in the past 30 years, with a cumulative default rate of 28%. Investment-grade CMBS from the average conduit deal issued at that time, however, would not have lost principal if subjected to the default and loss rates experienced by the 1986 cohort. Two years ago (2002), Esaki updated the study, adding three years of data. The main finding was that declining subordination levels would put some BBB CMBS at risk if the default rates of the worst cohort were repeated.

210

COMPARISON T O P REVIOUS S TUDIES

Our update through 2002 shows little difference from the results of the previous study with data through 2000. Only 21 loans from the database defaulted during 2001 and 2002. This represents only 0.3% of the total number of loans originated in the last 10 years of the study. In the current update, we examined the credit performance of 17,978 individual loans, an increase of about 1,400 loans from the 2002 study. The average lifetime cumulative default rate for origination cohorts with at least ten years of history was 19.6%, slightly lower than the 20.5% in the prior default study published by Esaki in 2002. On average, about 91% of defaults occurred within the first 10 years of origination. The cumulative default rate of almost 32% for originations in 1986 was again the highest for any cohort. NUMBER O F L OANS B Y O RIGINATOR A ND O RIGINATION Y EAR

In the current study, we tracked commercial mortgage default rates from life insurance company annual statements for 2001 and 2002. We then aggregated our data with the two previous Snyderman studies, the ELS findings, and the Esaki (2002) report. As in the earlier studies, we chose to include cohorts with a minimum of five years of seasoning. Of the 17,978 total loans originated, about 2,700 (15.3%), had defaulted by 2002. SIZE O F L OANS

The median loan size was about $4.2 million, and the average loan size was about $8.5 million. Average loan size has trended up over time, increasing from $3 million in 1972 to $14 million in 1997. Over 70% of the loans were less than $8 million. The less-than-$2 million loan category had the lowest default rate, while the $4-8 million category had the highest default rate. This is the same as previous studies. GEOGRAPHIC D ISTRIBUTION

As in earlier studies, the loans are geographically well diversified, with the largest percentages in the West (23%) and Northeast (22%). The highest default rates were in the South Central region (25%), with the lowest in the West (10%). DEFAULTS B Y O RIGINATION C OHORT

The cumulative lifetime default rate (by loan balance) for cohorts with at least 10 years of history ranged from 4.0% for 1992 originations to 31.7% for 1986 originations. The average lifetime cumulative default rate (based on loan balance) for cohorts with at least 10 years of seasoning was 19.6%, down from 20.5% in the previous study. The drop resulted from the addition of 1991 and 1992, which had the two lowest cumulative default rates of any cohort.

Please see additional important disclosures at the end of this report.

211

Transforming Real Estate Finance Chapter 12

Commercial Mortgage Defaults TIMING O F D EFAULTS

On average, the annual default rate was low within the year of loan origination, but rose to about 1% in the first year following origination, then jumped to a range of 1.5% to 2.7% for the next six years. Default rates then declined to less than 1% for the next three years, and tailed off gradually. These results are nearly identical to the Esaki (2002) study. As in that study, there is no spike in defaults at balloon dates. Some research analysts have noted that loan restructures result in the appearance of low default rates in balloon years, but there is no evidence to this in our study. The default timing pattern for individual cohorts can vary widely from the average. The timing and total defaults of a cohort are highly dependent on its position in the real estate cycle. For almost all cohorts, however, the peak in defaults is in years three through seven after origination. LIQUIDATIONS A ND R ECOVERIES

As in previous studies, not all defaulting (90+ days delinquent) loans liquidate. We found that only about half of the loans we recorded as entering default for the first time went straight through to liquidation. Another quarter of the loans were restructured, while the rest became current again. Of the loans becoming current, about 60% were eventually restructured, and another 30% defaulted again. We estimate that about 55% of loans entering default are eventually liquidated, 40% are restructured, 3% become delinquent again, and only 2% fully recover. Our finding that 20% of loans entering default initially recover is similar to a Fitch (2001) CMBS default study finding that 22% of defaulted loans in CMBS return to current status. Our study goes one step further, however, and finds that only about 9% of these loans remain current. This means that only about 2% (20% x 9%) of loans entering default return permanently to current status. SEVERITY O F L OSS

The loss severity on liquidated loans for the loans added to the previous study was about 31%, the same as in the previous study and less than the 36% in the original Snyderman studies. The severity calculation includes foregone interest and expenses, as well as lost principal. In Exhibit 1, we outline the components that are used in the loss severity calculation. For loans that were liquidated within the last 10 years of the study, foregone interest ed for a large portion of loss severity. On average, the loans liquidated within this 10-year time period experienced 24 months between default and liquidation.

212

exhibit 1

SEVERITY OF LOSS ON LIQUIDATED LOANS (1992-2002)

Source: Morgan Stanley

For the entire 1972-2002 period, the average severity of loss on foreclosed loans is about 33%. As in earlier studies, the range of severities for individual loans is large. Some loans had severities of over 100%, while others recorded no loss.

Please see additional important disclosures at the end of this report.

213

Transforming Real Estate Finance Chapter 12

Commercial Mortgage Defaults IMPACTS O F D EFAULTS O N C URRENT B OND S TRUCTURES

In the Esaki (2002) study, we found that the average cohort with at least 10 years of seasoning lost about 5.4% of its original balance through defaults. In this study, the average cohort with at least 10 years of seasoning lost slightly less, about 4.9% of its original balance. AVERAGE SUBORDINATION FOR CONDUIT/ FUSION TRANSACTIONS

exhibit 2

AAA AA A BBB BB B

1998 29% 24% 18% 13% 6% 3%

1999 27% 22% 17% 12% 6% 3%

2000 23% 19% 14% 11% 5% 3%

2001 21% 17% 13% 9% 4% 2%

2002 20% 16% 12% 8% 4% 2%

2003 17% 14% 10% 7% 3% 2%

20041 14% 12% 9% 5% 3% 2%

1

As of August 19, 2004.

Source: Commercial Mortgage Alert, Morgan Stanley

This calculation assumes that restructured loans have half of the severity (16.5%) of liquidated loans. Since the average conduit/fusion transaction today is being issued with BBB subordination levels of about 5%, most investmentgrade CMBS are still protected against the average loss of origination cohorts of the last 30 years. The loss on the worst cohort, 1986, has an estimated loss of 8.1% of its original balance. This exceeds the average BBB subordination level on conduit and fusion CMBS transactions being issued today, and would result in the default of even some single-A classes. It is, however, still below the lowest credit levels for AAA CMBS. exhibit 3

NUMBER OF LOANS BY ORIGINATOR, 1972-1997

Originator

Percentage Change from Esaki (2002)

Total Loan Amount ($BN) 26.6

Aetna Life Insurance Company

3,042

1.1%

Connecticut Mutual Life Insurance Company

1,062

15.9%

5.7

Equitable Life Insurance Company

1,838

1.1%

17.2

John Hancock Mutual Life Insurance Company

2,538

13.0%

18.1

New England Mutual Life Insurance Company

1,736

25.0%

11.5

The Northwestern Mutual Life Insurance Company

1,108

21.9%

13.6

The Prudential Insurance Company of America

4,349

7.6%

42.6

The Travelers Insurance Company Total

Source: Morgan Stanley

214

Number of Loans

2,305

1.7%

17.4

17,978

8.3%

152.8

Our loss estimates are based on a number of assumptions. The most important assumption for loss calculation is that the severity of restructured loans is half that for liquidated loans. If we assume, as some market participants believe, that restructured loans have close to 0% severity, the loss for the worst cohort drops to 6.1%. On the other hand, if we assume that restructured and liquidated loans have the same severity of 33%, the loss rate estimate rises to 10.6%. CONCLUSIONS

The results from our update of commercial mortgage defaults through the year 2002 do not significantly change the previous findings on cumulative default and loss rates, the severity of losses on liquidated loans, or the shape of the loss curve. REFERENCES

Esaki, Howard. “Commercial Mortgage Defaults: 1972-2000.” Real Estate Finance, Winter 2002. Esaki, Howard, Steven L’Heureux, and Mark P. Snyderman. “Commercial Mortgage Defaults: An Update.” Real Estate Finance, Spring 1999. Lans, Diane M. and Noel Cain. “Dissecting Defaults and Losses: 2001 CMBS Conduit Loan Default Study.” Fitch IBCA Special Report, August 2001. Snyderman, Mark P. “Commercial Mortgages: Default Occurrence and Estimated Yield Impact.” Journal of Portfolio Management, Fall 1991. Snyderman, Mark P. “Update on Commercial Mortgage Defaults.” Real Estate Finance, Summer 1994.

Please see additional important disclosures at the end of this report.

215

Transforming Real Estate Finance Chapter 12

Commercial Mortgage Defaults exhibit 4

NUMBER OF LOANS BY ORIGINATION YEAR

Source: Morgan Stanley

exhibit 5

AVERAGE LOAN AMOUNT BY ORIGINATION YEAR

Source: Morgan Stanley

exhibit 7

LOANS ORIGINATED AND DEFAULT RATES BY REGION

Number of Loans

Amount of Loans ($BN)

West Coast

4,120

9.2

37.0

South Central

3,042

7.3

19.6

Northeast

3,891

6.7

40.1

Mid-Central

3,317

8.0

25.4

Southeast

3,069

8.8

27.4

539

20.9

3.3

17,978

8.3

152.8

Canada/Other Total Source: Morgan Stanley

216

Percentage Change from from Esaki (2002)

exhibit 6

LOANS ORIGINATED AND DEFAULT RATES BY PRINCIPAL AMOUNT Loans Originated

Percentage Change from Esaki (2002)

Percent of Total

Loan Count Default Rate (%)

1-2MM

4,083

3.3%

22.7%

12.9%

2-4 MM

4,473

5.2%

24.9%

15.6%

Loan Amount ($MM)

4-8 MM

4,326

9.8%

24.1%

16.7%

>8MM

5,096

14.5%

28.3%

15.7%

17,978

8.3%

100.0%

15.3%

Total Source: Morgan Stanley

By Loan Count

By Loan Amount

Percentage Change from Esaki (2002)

Default Rate (%)

Percent of Total

Percentage Change

Default Rate (%)

22.9

0.2

10.2

24.2

-0.4

11.6

16.9

-0.2

24.6

12.8

0.2

23.7

21.6

-0.3

13.3

26.3

-0.4

13.9

18.5

0.0

15.8

16.7

0.0

16.6

17.1

0.1

15.7

17.9

0.5

14.4

3.0

0.3

9.8

2.1

0.1

15.7

100.0

NA

15.3

100.0

NA

15.2

Percent of Total

Please see additional important disclosures at the end of this report.

217

Transforming Real Estate Finance Chapter 12

Commercial Mortgage Defaults exhibit 8a

LIFETIME DEFAULT RATES BY ORIGINATION COHORT (BY PRINCIPAL BALANCE)

Source: Morgan Stanley

exhibit 8b

LIFETIME DEFAULT RATES BY ORIGINATION COHORT (BY LOAN COUNT)

Source: Morgan Stanley

exhibit 9a

AVERAGE TIMING OF DEFAULTS – LOAN COUNT

Source: Morgan Stanley

218

exhibit 9b

AVERAGE TIMING OF DEFAULTS – LOAN AMOUNT

Source: Morgan Stanley

exhibit 10

LIQUIDATED, RESTRUCTURED, AND RECOVERED LOANS

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

219

220

0.00

0.14

0.71

2.33

0.21

0.00

0.28

0.27

0.24

0.75

1.38

1.25

0.00

0.08

0.00

0.21

0.10

0.35

0.41

0.72

0.94

0.21

0.43

0.00

0.00

0.00

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1

0.00

0.16

0.36

0.00

0.62

0.31

1.20

3.13

0.93

0.29

0.31

1.18

2.40

1.41

0.31

0.92

0.00

0.47

0.36

0.19

0.85

0.82

2.33

6.23

2.99

2.44

1.19

2

0.00

0.00

0.36

0.64

0.00

1.88

1.68

4.63

3.13

1.87

1.77

2.16

2.08

3.17

1.10

2.29

1.00

0.24

0.09

0.09

0.14

0.41

1.94

3.26

5.03

3.75

1.64

3

0.00

0.00

0.00

0.43

1.04

1.25

1.68

2.18

3.36

4.72

2.71

2.06

2.48

1.94

4.39

2.75

0.50

0.00

0.62

0.00

0.28

1.23

0.58

1.42

3.40

3.56

1.71

4

0.00

0.16

0.00

0.21

0.00

0.00

1.20

2.18

2.55

4.13

4.38

3.44

2.56

4.59

2.51

5.05

0.75

0.59

0.18

0.38

0.14

0.21

0.58

0.57

1.49

3.19

1.67

5

0.14

0.32

0.00

0.00

0.62

0.00

0.24

3.00

2.09

4.52

5.63

4.92

3.60

2.29

3.29

1.83

2.74

1.30

0.36

0.00

0.28

0.00

0.58

0.71

0.41

1.31

1.81

6

0.16

0.00

0.00

0.21

0.00

0.48

1.09

2.67

2.75

3.65

8.46

3.28

3.70

0.78

0.00

2.00

2.25

0.18

0.09

0.14

0.21

0.39

0.14

0.00

0.19

1.66

7

0.71

0.21

0.00

0.00

0.48

0.27

1.74

1.28

2.09

2.26

1.68

1.76

0.47

1.38

2.49

1.54

2.13

0.66

0.71

0.00

0.00

0.14

0.14

0.38

1.08

8

0.00

0.00

0.00

0.00

0.00

0.70

1.28

1.04

1.28

1.12

2.29

1.88

3.67

2.49

2.01

0.98

1.80

0.56

0.00

0.39

0.14

0.14

0.00

0.96

9

0.21

0.31

0.00

0.27

0.58

0.29

0.31

0.49

0.32

1.59

0.78

0.00

1.25

1.54

1.96

1.14

1.55

1.03

0.39

0.42

0.14

0.00

0.72

10

0.00

0.24

0.14

0.23

0.10

0.73

1.08

0.48

0.53

1.10

0.92

1.75

2.01

1.16

1.52

0.99

0.62

0.00

0.00

0.14

0.00

0.69

11

0.00

0.27

0.12

0.00

0.42

0.20

0.08

0.35

0.16

0.92

2.24

2.01

1.16

0.85

0.56

1.23

1.17

0.14

0.00

0.19

0.55

12

0.00

0.12

0.00

0.21

0.10

0.16

0.00

0.16

0.00

0.75

1.42

1.51

1.42

0.14

0.41

1.36

0.42

0.82

0.19

0.51

Source: Life Insurance company annual statements; Morgan Stanley

0.33

AVG

0.00

0.10

0.00

0.10

0.08

0.18

0.00

0.46

0.25

0.83

1.42

0.66

0.56

0.21

0.58

0.71

0.95

0.38

0.43

13

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.12

0.80

0.95

0.42

0.21

0.78

0.42

0.68

0.75

0.31

14

0.00

0.00

0.00

0.00

0.00

0.00

0.50

0.24

0.36

0.95

0.56

1.85

0.78

0.71

0.82

0.94

0.43

15

0.00

0.08

0.00

0.00

0.00

0.25

0.36

0.18

0.57

0.85

0.82

0.78

0.71

0.68

0.56

0.37

16

0.00

0.00

0.00

0.46

0.25

0.24

0.36

0.09

0.00

0.41

0.58

0.28

0.14

0.19

0.18

17

TIMING OF DEFAULTS BY COHORT – LOAN COUNT 18

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.28

0.62

0.39

0.28

0.27

0.19

0.14

19

0.00

0.00

0.00

0.12

0.00

0.00

0.14

0.62

0.00

0.14

0.82

0.75

0.20

20

0.00

0.00

0.24

0.00

0.19

0.00

0.21

0.00

0.00

0.27

0.38

0.12

21

0.00

0.00

0.00

0.00

0.14

0.00

0.00

0.00

0.27

0.00

0.04

22

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.19

0.01

23

0.00

0.00

0.00

0.00

0.00

0.00

0.14

0.00

0.02

24

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

25

0.00

0.00

0.19

0.00

0.00

0.00

0.03

26

0.00

0.00

0.00

0.00

0.00

0.00

27

0.00

0.00

0.00

0.00

0.00

28

0.00

0.00

0.00

0.00

29

0.00

0.00

0.00

30

0.00

0.00

0.14

0.80

1.42

1.93

2.90

4.69

7.93

17.30

18.31

21.34

22.84

27.34

20.10

23.28

17.87

20.18

15.71

12.20

8.27

6.16

5.65

4.72

9.51

13.74

13.99

14.82

13.02

10 Year Total

0.14

0.80

1.42

1.93

2.90

4.69

7.93

17.57

18.54

21.44

23.46

27.73

20.50

23.81

18.18

22.02

19.95

17.77

14.04

11.85

9.32

11.29

16.12

17.56

19.84

19.51

15.27

Lifetime Total

100.00

100.00

100.00

100.00

100.00

100.00

100.00

98.45

98.75

99.54

97.33

98.58

98.05

97.78

98.28

91.67

78.75

68.67

58.86

52.00

60.61

41.82

59.04

78.23

70.55

75.96

85.25

% in 10 yrs

Chapter 12

0

appendix A-1

Transforming Real Estate Finance

Commercial Mortgage Defaults

Please see additional important disclosures at the end of this report.

221

0.00

0.26

0.81

1.91

0.32

0.00

0.14

0.11

0.08

0.69

0.73

0.82

0.00

0.05

0.00

0.13

0.09

0.10

0.09

0.50

1.05

0.27

0.34

0.00

0.00

0.00

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1

0.00

0.10

0.09

0.00

0.37

0.18

1.22

2.47

1.28

0.28

0.20

0.92

2.33

1.27

0.16

1.54

0.00

0.25

0.32

0.07

1.44

1.35

3.04

7.10

2.24

1.68

0.92

2

0.00

0.00

0.34

0.69

0.00

1.39

0.85

5.95

2.38

1.85

1.23

2.28

1.83

1.99

1.09

1.77

0.57

0.16

0.97

0.04

0.04

0.15

1.18

4.55

4.59

2.52

1.51

4 1.84

0.00

0.00

0.00

0.17

1.81

0.80

1.11

1.80

3.55

3.65

2.86

2.14

2.45

1.22

3.81

1.44

0.50

0.00

0.57

0.00

2.76

1.69

0.54

1.02

2.46

0.00

0.08

0.00

0.23

0.00

0.00

1.27

2.87

1.85

3.47

4.60

3.25

2.92

2.52

2.46

2.66

0.53

0.53

0.43

0.29

0.04

0.15

0.20

0.74

1.06

2.99 10.20

1.72

3

6

2.86

7.76

1.53

0.00

0.83

1.69

0.14

0.03

0.18

0.07

0.15

0.07

0.00

0.06

2.68

0.11

0.14

0.00

0.00

0.37

0.00

0.37

3.26

1.84

3.43

4.73

0.13

0.00

0.00

0.10

0.00

0.37

1.02

4.33

4.25

4.63

5.82 10.30

2.90

1.14

2.99

1.31

2.21

2.16

0.28

0.00

2.83

0.00

0.23

0.85

0.40

1.25

2.00

5

0.78

0.03

0.00

0.00

0.18

0.27

1.50

1.42

2.84

2.47

2.24

4.79

0.80

0.53

2.05

1.22

2.62

0.47

3.33

0.00

0.00

0.12

0.05

0.23

1.49

7

0.00

0.00

0.00

0.00

0.00

0.49

1.16

0.49

1.31

1.30

2.75

1.84

3.19

3.44

1.22

0.98

1.48

0.40

0.00

0.44

0.11

0.15

0.00

0.91

8

0.11

0.57

0.00

0.10

0.56

0.13

0.18

0.24

0.38

2.27

0.65

0.00

0.92

1.53

2.27

1.64

1.76

0.88

0.67

0.96

0.04

0.00

0.61

9

0.00

0.28

0.03

0.73

0.08

1.07

2.32

0.93

0.17

0.79

4.42

2.11

2.05

0.89

1.63

0.44

0.73

0.00

0.00

0.11

0.00

0.87

10

0.00

0.46

0.49

0.00

0.41

0.23

0.07

1.18

0.14

0.68

0.91

1.54

0.90

0.80

0.53

0.72

1.11

0.07

0.00

0.22

0.43

11

0.00

0.15

0.00

0.10

0.31

0.27

0.00

0.04

0.00

0.48

0.92

1.32

1.13

0.03

0.58

1.58

0.41

0.42

0.08

0.30

12

0.00

0.03

0.00

0.06

0.32

0.10

0.00

2.39

0.04

1.27

3.13

0.46

0.33

0.10

0.19

0.42

0.80

0.23

0.36

13

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.80

0.42

0.83

2.22

0.07

0.59

0.38

0.50

0.69

0.24

14

0.00

0.00

0.00

0.00

0.00

0.00

0.43

0.25

0.18

0.61

1.00

1.79

0.25

0.60

0.72

0.54

0.23

15

0.00

0.46

0.00

0.00

0.00

0.15

0.50

0.38

0.49

0.46

0.39

0.39

0.52

0.49

0.70

0.28

16

TIMING OF DEFAULTS BY COHORT – LOAN SIZE

Source: Life Insurance company annual statements; Morgan Stanley

0.19

AVG

0

appendix A-2 17

0.00

0.00

0.00

0.15

0.73

0.11

0.55

0.14

0.00

0.35

0.41

0.15

0.05

0.14

0.15

18

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.09

1.10

0.36

0.74

0.12

0.10

0.14

19

0.00

0.00

0.00

0.03

0.00

0.00

0.04

0.35

0.00

0.27

0.36

0.47

0.09

20

0.00

0.00

0.08

0.00

0.16

0.00

0.11

0.00

0.00

0.40

0.16

0.08

21

0.00

0.00

0.00

0.00

0.08

0.00

0.00

0.00

0.44

0.00

0.05

22

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.35

0.02

23

0.00

0.00

0.00

0.00

0.00

0.00

0.38

0.00

0.04

24

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

25

0.00

0.00

0.11

0.00

0.00

0.00

0.02

26

0.00

0.00

0.00

0.00

0.00

0.00

27

0.00

0.00

0.00

0.00

0.00

28

0.00

0.00

0.00

0.00

29

0.00

0.00

0.00

30

0.00

0.00

0.11 14.58

12.48

0.44

1.20

1.47

3.03

4.00

6.14

18.31

19.24

19.83

23.46

31.66

21.29

27.16

17.13

20.80

16.58

16.38

16.47

10.40

18.00

10.90

13.36

19.90

16.03

22.63

15.20

Lifetime Total

0.11

0.44

1.20

1.47

3.03

4.00

6.14

17.85

18.60

19.81

22.95

31.05

20.17

25.88

16.95

17.59

13.84

10.89

9.59

5.78

13.21

5.34

8.36

16.33

11.35

18.94

13.89

10 Year Total

85.60

100.00

100.00

100.00

100.00

100.00

100.00

100.00

97.49

96.70

99.87

97.82

98.09

94.75

95.28

98.92

84.53

83.48

66.50

58.21

55.64

73.41

48.98

62.55

82.08

70.83

83.71

85.81

% in 10 yrs

This Page Intentionally Left Blank

222

Chapter 13

Transforming Real Estate Finance

Transaction Monitoring

Please see additional important disclosures at the end of this report.

223

Transforming Real Estate Finance chapter 13

Transaction Monitoring In this chapter, we provide samples of transaction monitoring pieces that we publish regularly. Section I contains an example of our monthly delinquency article which reports delinquencies for the entire universe as well as for seasoned, multiborrower CMBS. Section II contains an excerpt from our conduit tracking report of specially serviced loans in Morgan Stanley transactions. Section III includes an excerpt from our retail study analyzing risky tenant exposure in Morgan Stanley deals. Section IV provides an example of our quarterly tracking of CMBS rating actions. Section V contains our most recent report on the credit performance of CMBS originators.

Section I. CMBS Delinquency Report Based on October 2004 remittance reports, delinquencies on all CMBS transactions declined 4 bp to 1.13% of original balances. Most of the improvement was attributable to office collateral. Sixty-plus day delinquencies on all CMBS declined 2 bp to 0.97%. After peaking at 2.62% in November 2003, seasoned delinquencies have declined consistently over the past eight months. Based on October remittance reports, delinquencies on seasoned transactions (aged over one year) declined 10 bp to 1.79%. The liquidation of a $33.4 million loan in MSC 1997-ALIC (Cambridge Park 2) ed for 3 bp of the decline. The remaining 7 bp of improvement is attributable to a large number of small loans. Sixty-plus day delinquencies on seasoned deals declined 4 bp to 1.55%.

exhibit 1

DELINQUENCIES IN SEASONED CMBS DEALS AND LIFETIME AVERAGE

Source: Morgan Stanley, Intex

224

exhibit 2

CMBS DELINQUENCIES BY YEAR OF ORIGINATION (IN %) (AS OF OCTOBER 2004 REMITTANCE REPORTS)

Current Balance Year ($BN) 1990 0.2 1991 0.1 1992 0.2 1993 0.4 1994 0.7 1995 3.1 1996 9.2 1997 26.4 1998 57.0 1999 34.2 2000 32.6 2001 52.3 2002 48.9 2003 82.0 2004 59.0 Total/Average 406.3

30/60/90+ Days 19.46 0.00 0.00 1.33 1.21 0.89 1.01 1.65 0.84 1.35 0.94 0.71 0.24 0.09 0.06 0.60

Forc. & REO 0.00 0.00 0.00 0.95 3.39 2.21 0.89 1.46 1.04 1.01 1.22 0.34 0.12 0.00 0.00 0.53

Total 19.46 0.00 0.00 2.27 4.60 3.09 1.90 3.11 1.88 2.36 2.15 1.04 0.36 0.09 0.06 1.13

Change From Last Month 0.04 0.00 0.00 0.02 1.18 -0.80 0.13 -0.35 -0.07 -0.19 0.06 0.06 0.04 0.03 0.05 -0.04

Source: Morgan Stanley, Intex

ORIGINATION Y EAR

Of the cohorts with more than $20 billion in collateral outstanding, the 1997 cohort posted the greatest improvement, declining 35 bp to 3.11%.

exhibit 3

CMBS DELINQUENCIES BY PROPERTY TYPE (IN %) (AS OF OCTOBER 2004 REMITTANCE REPORTS)

Product Current Balance Type ($BN) Hotel-Motel 29.0 Industrial-Warehouse 21.9 Mixed 16.3 Mobile Home 8.3 Multifamily 70.4 Office 100.0 Retail 124.7 Self-Storage 5.6 Senior Housing 3.9 Other 26.2 Total/Average 406.3

30/60/90+ Days 1.06 0.73 0.21 0.45 1.15 0.46 0.45 0.25 1.55 0.01 0.60

Forc. & REO 1.39 0.83 0.46 0.13 0.58 0.49 0.35 0.03 1.06 0.36 0.53

Total 2.44 1.56 0.67 0.58 1.73 0.95 0.80 0.28 2.61 0.37 1.13

Change From Last Month -0.13 -0.01 -0.11 0.14 0.10 -0.14 -0.02 0.03 -0.65 0.00 -0.04

Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

225

Transforming Real Estate Finance chapter 13

Transaction Monitoring The liquidation of the Cambridge Park 2 loan was responsible for 13 bp of the improvement. MULTIFAMILY D ELINQUENCIES R ISE

Delinquencies on multifamily collateral rose 10 bp during the month to 1.73%. Although multifamily fundamentals are showing signs of improvement (vacancies nationwide declined 20 bp in 3Q04) we continue to maintain a cautious view on the sector. OFFICE C OLLATERAL P OST 1 4 B P I MPROVEMENT

Office delinquencies declined 14 bp to 0.95% of current balances. Eight basis points of the decline is attributable to the liquidation of three office properties, Allen Center in GMACC 1997-C2, the Bell Atlantic Building in CMAC 1998-C2 and Cambridge Park 2 in MSC 1997-ALIC. The weighted average loss severity on those loans was 28.6%. CUMULATIVE L OSSES

Cumulative losses for the seasoned universe rose 3 bp to 0.59% of original balances. MCFI 1997-MC1 experienced the greatest rise in cumulative losses (104 bp) due to the liquidation of the Radisson Hotel & Suites-Buffalo. The loan’s loss severity was 100%.

exhibit 4

Product Type Hotel-Motel Industrial-Warehouse Mixed Mobile Home Multifamily Office Retail Self-Storage Senior Housing Other Source: Morgan Stanley, Intex

226

CMBS DELINQUENCIES BY PROPERTY TYPE AND YEAR (IN %) (AS OF OCTOBER 2004 REMITTANCE REPORTS)

1990 0.00 0.00 0.00 N/A 0.00 38.92 0.00 N/A N/A 0.00

1991 N/A 0.00 0.00 N/A 0.00 0.00 0.00 N/A N/A 0.00

1992 N/A 0.00 0.00 0.00 0.08 0.00 0.00 N/A N/A 0.00

1993 0.00 0.00 0.00 0.00 1.47 0.00 9.85 N/A N/A 0.00

1994 1.26 12.97 0.00 0.00 0.00 0.00 10.11 N/A 1.62 0.00

1995 21.59 4.53 0.00 11.46 2.20 0.00 2.96 0.00 12.90 0.00

1996 8.97 1.97 0.00 0.00 2.34 0.37 0.94 0.00 5.93 0.00

exhibit 5

State CA NY TX FL IL NJ VA MA PA MD Total/Average

CMBS DELINQUENCIES BY STATE (IN %) (AS OF OCTOBER 2004 REMITTANCE REPORTS)

Current Balance ($BN) 59.3 45.3 25.9 23.0 12.6 12.7 12.6 10.5 10.2 9.9 221.9

% Total Delinq. 0.53 0.43 2.79 1.70 2.11 0.76 0.14 1.39 1.75 0.27 1.06

Change From Last Month -0.06 -0.07 0.24 -0.06 -0.10 0.03 -0.08 -0.49 -0.22 -0.11 -0.06

Source: Morgan Stanley, Intex

1997 7.40 3.67 4.43 0.54 1.50 3.92 3.63 0.41 4.55 6.41

1998 3.64 3.62 1.09 0.86 1.01 2.99 1.42 0.55 3.05 2.59

1999 2.15 2.24 2.60 0.39 3.66 3.82 1.46 0.66 2.33 2.04

2000 1.44 3.02 4.20 0.69 3.44 2.41 1.25 1.75 2.33 0.00

2001 1.52 0.76 0.62 0.33 3.74 0.24 0.53 0.00 2.09 0.00

2002 0.00 0.25 0.00 1.16 1.19 0.22 0.09 0.00 0.00 0.00

Please see additional important disclosures at the end of this report.

2003 0.00 0.00 0.00 0.00 0.26 0.04 0.04 0.00 0.00 0.00

2004 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 N/A 0.00

227

Transforming Real Estate Finance chapter 13

Transaction Monitoring Section II. Summer Update In this summer 2004 issue of our tracking report, we provide details on 73 specially serviced loans ($447.6 million in current balances) in Morgan Stanley conduit transactions. Loans in this report are either currently in special servicing or are resolved loans that were covered in our previous report. Conduit transactions issued off Morgan Stanley’s shelf that do not have any loans in special servicing are not in this report. The information provided is based on July 2004 and August 2004 remittance reports and recent conversations with special servicers.

exhibit 6

MORGAN STANLEY TRANSACTIONS (JULY 2004 REMITTANCE REPORTS)

Deal Name Morgan Stanley Capital I, 1996-C1 Morgan Stanley Capital I, 1996-WF1 Morgan Stanley Capital I, 1997-ALIC Morgan Stanley Capital I, 1997-C1 Morgan Stanley Capital I, 1997-HF1 Morgan Stanley Capital I, 1997-WF1 Morgan Stanley Capital I, 1998-CF1 Morgan Stanley Capital I, 1998-HF1 Morgan Stanley Capital I, 1998-HF2 Morgan Stanley Capital I, 1998-WF1 Morgan Stanley Capital I, 1998-WF2 Morgan Stanley Capital I, 1999-FNV1 Morgan Stanley Capital I, 1999-LIFE Morgan Stanley Capital I, 1999-RM1 Morgan Stanley Capital I, 1999-WF1 Morgan Stanley Dean Witter Capital, 2000-LIFE Morgan Stanley Dean Witter Capital, 2001-TOP1 Morgan Stanley Dean Witter Capital, 2001-TOP3 Morgan Stanley Dean Witter Capital, 2002-TOP7 Morgan Stanley Capital I, 2003-TOP11 Total/Weighted Average Source: Intex, Remittance Reports

228

Original Balance ($MM)

Current Balance ($MM)

Factor

340.5 605.4 802.7 640.7 622.4 559.2 1,107.3 1,283.7 1,066.3 1,392.2 1,062.0 632.1 594.0 867.1 968.5 689.0 1,172.2 1,031.2 976.6 1,194.9 17,608.0

128.3 234.2 312.0 315.6 270.7 356.7 850.8 992.2 898.2 1,069.6 898.1 582.6 559.2 739.1 798.0 641.8 1,080.2 989.9 931.4 1,180.4 13,829.0

0.38 0.39 0.39 0.49 0.43 0.64 0.77 0.77 0.84 0.77 0.85 0.92 0.94 0.85 0.82 0.93 0.92 0.96 0.95 0.99 0.83

TRANSACTION S TATISTICS

Based on July remittance reports, delinquencies on the transactions we reviewed were 2.20% of current balances, 14 bp higher than the average for seasoned CMBS deals (aged over one year). Cumulative losses on the transactions we covered were 64 bp, versus 54 bp for seasoned CMBS. LOAN S TATISTICS

Twelve loans ing for about $88.5 million in current balances are new specially serviced loans since our last report. Twelve problem loans ing for about $42.7 million in current balances have been resolved since our last report, either through liquidation, payoff or by being returned to the master servicer.

Delinquency Data (%)

30, 60 & 90+

Forc. & REO

Total

Cum Loss

2.65 0.00 13.62 0.00 2.94 0.00 1.26 1.27 0.83 2.93 0.00 4.49 0.00 0.55 0.00 0.00 2.15 0.50 0.00 0.00 1.26

3.21 0.00 0.00 0.00 0.00 0.00 4.75 0.00 0.88 1.70 0.00 3.19 1.76 0.57 0.00 1.57 1.13 0.00 0.47 0.00 0.94

5.85 0.00 13.62 0.00 2.94 0.00 6.01 1.27 1.71 4.63 0.00 7.69 1.76 1.12 0.00 1.57 3.28 0.50 0.47 0.00 2.20

1.02 1.32 0.39 0.81 0.36 0.00 5.70 0.96 0.52 0.21 0.00 0.10 0.00 0.27 0.06 0.39 0.01 0.06 0.00 0.00 0.64

Specially Serviced Data Specially Serviced or % of Specially Resolved Serviced or Loans ($MM) Resolved Loans 7.5 18.3 42.5 6.6 7.9 7.6 105.5 21.9 13.3 50.5 10.7 45.8 9.9 4.2 4.2 12.2 40.2 6.8 4.4 27.6 447.6

Please see additional important disclosures at the end of this report.

1.7 4.1 9.5 1.5 1.8 1.7 23.6 4.9 3.0 11.3 2.4 10.2 2.2 0.9 0.9 2.7 9.0 1.5 1.0 6.2 100.0

229

Transforming Real Estate Finance chapter 13

Transaction Monitoring Based on current balances, office properties ed for 26.4% of the loans reviewed, followed by retail (19.6%) and multifamily (15.6%). In of loan count, retail properties had the greatest representation (16 loans), followed by multifamily (15 loans), office (13 loans), industrial-warehouse (12 loans) and hotel (7 loans). RATING A CTIONS

Year-to-date through August 25, 2004, tranches on deals covered in this report experienced a 16.5 to 1 upgrade/downgrade ratio. Over the same period, the CMBS universe experienced a 3.4 to 1 upgrade/downgrade ratio.

230

exhibit 7 SPECIALLY SERVICED LOANS

IN MORGAN STANLEY TRANSACTIONS (JULY 2004 REMITTANCE REPORTS)

Number of Loans 13

Specially Serviced or Resolved Loans ($MM) 118.3

% of Specially Serviced or Resolved Loans 26.4

Retail

16

87.6

19.6

Multifamily

15

69.6

15.6

Industrial-Warehouse

12

68.3

15.3

Hotel

7

64.7

14.5

Health Care

5

29.1

6.5

Mobile Home

3

6.4

1.4

Mixed Use

1

2.6

0.6

Self Storage

1

1.0

0.2

73

447.6

100

Property Type Office

Total Source: Intex, Trepp, Remittance Reports

Please see additional important disclosures at the end of this report.

231

Transforming Real Estate Finance chapter 13

Transaction Monitoring Sample Loan Summary: HILLSIDE MOBILE HOME PARK • Original Balance:

$4,839,717

• Current Balance:

$4,115,841

• Percent of Pool Balance:

Not Applicable

• Master Servicer:

GMAC COMMERCIAL MORTGAGE CORP.

• Special Servicer:

LENNAR PARTNERS.

• Trustee:

LASALLE NATIONAL BANK

Status: Paid Though Date: Location: Size: Year Built: Property Type: Originator:

property description

valuation information

Discounted Payoff Not Applicable Stillwater, New York 383 Units 1971 Mobile Home Morgan Stanley

Appraisal Value: $2,650,000 Appraisal Date: November 2003 Original LTV: 57.6% Original Loan per Unit: $12,636 Current LTV: Not Applicable Current Loan per Unit: Not Applicable Underwritten DSCR: 1.53 Current DSCR: 0.72 (July 2004)

market data

Closest MSA: Market Average Occupancy for Property Type: Market Average Rent for Property Type:

Source: Intex, Trepp, REIS, Special Servicers, Remittance Reports

232

Albany, New York - 23 Miles Not Available Not Available

REASON F OR S PECIAL S ERVICING T RANSFER

Payment default. UPDATE

According to the August 2004 remittance report, there was a discounted payoff of the loan on August 16, 2004, with a $1,979,805 loss to the trust. As of July 2004, the DSCR was 0.72. An updated appraisal in November 2003 valued the property at $2,650,000. PREVIOUS S TATUS O F R ESOLUTION

In our last report, escrows and principal and interest payments were past due. According to the special servicer, the borrower had fallen behind on principal and interest payments due to the increase in escrow requirements and its recent increase in expenditures. The borrower fell behind in escrow payments when the state required a water and sewer upgrade. The special servicer had indicated that the borrower was in the process of funding approximately $150,000 in upgrades to connect the park to the county sewer system. In addition, there had been a general increase in estimated taxes and insurance. According to the special servicer, the DSCR was 1.13 as of July 9, 2003. The special servicer indicated that a forbearance agreement was being negotiated. of the agreement stipulated that the borrower would pay $5,000 per month for the next 10 months to bring the escrows current, and the special servicer would waive the default interest. ASSET S UMMARY

The subject property is a 383-unit mobile home park, located in Stillwater, New York. The mobile home park has several amenities, including a clubhouse and a swimming pool. The property was approximately 74% occupied as of May 2002. The special servicer indicated that a 2001 property inspection found the property to be in good condition. MATURITY D ATE

February 1, 2006

Please see additional important disclosures at the end of this report.

233

Transforming Real Estate Finance chapter 13

Transaction Monitoring Section III. Retail Detail In 2001, we published our Retail Detail study, which assessed the exposure of risky retailers within Morgan Stanley underwritten CMBS. At the time, several U.S. retailers announced bankruptcies or store closings, and the health of the retail sector was deteriorating. Today, the economy is showing signs of improvement and the retail sector as a whole has weathered the economic downturn well. According to Reis, vacancy rates for retail properties have remained in a stable range of 6.8% to 7.1% since the end of 2001. As of the end of 2Q03, the retail vacancy rate was 6.9%. Although we are less concerned with the retail sector than we are with other property sectors, we still recognize the importance of monitoring deals for exposure to weak retailers. In this update of the retail study, we examine 42 Morgan Stanley deals for exposure to 116 risky retail credits. The risky retail credits were identified by the REIT equity research team using the Z-score methodology. All Morgan Stanley transactions underwritten prior to July 2003 with outstanding retail loans were included in this analysis (Exhibit 8). The equity research risky tenant list is based on 4Q02 financial results (Exhibit 9). Our main findings are: • Morgan Stanley CMBS deals have limited exposure to risky tenants. On a risk-weighted basis, 1.2% of Morgan Stanley CMBS collateral has exposure to the retailers. Risk-weighting s for the portion of a loan that can be attributed to the retailer, based on its gross leasable area. • On average, Morgan Stanley deals have 18% total exposure to the risky tenants. Total exposure includes the full balance of all loans with exposure to any of the 116 risky tenants. • With the exception of MSC 1997-LB1, all Morgan Stanley deals (with retail loans) have some exposure to the risky tenants. • Retail loans within large loan transactions have slightly more risk exposure than retail loans within conduit deals. On a risk-weighted basis, 3.7% of the retail loans within the large loan deals had exposure to the risky tenants. Within conduit deals, the retail loans had 3.4% exposure. While it is somewhat surprising that the risky tenant exposure is slightly higher in large loan deals, we do not believe that the regional and super-regional malls backing large loan transactions pose more risk than the strip centers and smaller shopping centers that are typically found in conduit transactions. This study only captures risky public companies and does not assess the risk of local, non-public retailers, which are often found in strip centers. • MSDWC 2001-PGM had the highest risk-weighted exposure as a percentage of the entire deal at 5.7%.

234

exhibit 8

MORGAN STANLEY CMBS DEALS Deal MSC 1996-C1 MSC 1996-WF1 MSC 1997-ALIC MSC 1997-C1 MSC 1997-HF1 MSC 1997-LB1 MSC 1997-WF1 MSC 1997-XL1 MSC 1998-CF1 MSC 1998-HF1 MSC 1998-HF2 MSC 1998-WF1 MSC 1998-WF2 MSC 1998-XL1 MSC 1998-XL2 MSC 1999-CAM1 MSC 1999-FNV1 MSC 1999-LIFE MSC 1999-RM1 MSC 1999-WF1 MSC 2000 HG MSDWC 2000-LIFE MSDWC 2000-LIFE2 MSDWC 2000-PRIN MSDWC 2001-PGMA MSDWC 2001-PPM MSDWC 2001-TOP1 MSDWC 2001-TOP3 MSDWC 2001-TOP5 MSDWC 2001-FRMA MSDWC 2001-SGMA MSDWC 2001-DFMA MSDWC 2001-IQA MSDWC 2001-XLF MSDWC 2002-HQ MSDWC 2002-IQ2 MSDWC 2002-IQ3 MSDWC 2002-XLF MSDWC 2002-TOP7 MSDWC 2003-TOP9 MSDWC 2003-HQ2 MSC 2003-IQ4

Aug 2003 Total Delinquencies (%) 2.53 2.27 1.53 1.85 0.29 0.00 2.01 0.00 11.09 1.14 2.68 2.76 0.00 0.00 0.00 1.51 4.72 1.77 1.53 0.63 0.00 0.82 0.00 0.00 0.00 0.00 1.39 0.41 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

235

Transforming Real Estate Finance chapter 13

Transaction Monitoring MOTIVATION F OR T HE S TUDY

Store closing announcements and retailer bankruptcies are often followed by a notification from CMBS data providers detailing the number of CMBS deals affected by the closing. In order to calculate the deal exposure to a troubled tenant, most CMBS data providers use databases containing the top three tenants for each loan. While this analysis is informative, we think it falls short in at least three aspects: (1) The procedure misses all but the top three tenants within a loan (Some loans only list the top tenant.). (2) The tenant information that is used by the data providers is often not updated after deal issuance. (3) The analysis does not look at the cumulative risk of all unhealthy tenants. We were prompted to publish a comprehensive study that addresses the shortcomings listed above. Using a database from the National Research Bureau, we are able to capture 82% of the loan balances of retail tenants in Morgan Stanley underwritten CMBS deals. This compares with only 32% covered by a database of the top three tenants. In addition, we look at the cumulative exposure of a list of risky retailers, rather than deal exposure to just one retailer. RETAIL P ROPERTY O VERVIEW

In aggregate, retail properties typically constitute 27%-30% of the collateral in CMBS transactions and historically have been among the most common property types within CMBS. Within retail properties, grocery-anchored community centers and super-regional malls are considered more desirable than box centers or mid-market malls. Strong regional malls and grocery-anchored community centers typically have lower defaults and cash flow volatility than other retail property types within CMBS.

236

exhibit 9

RISKY TENANTS AND Z-SCORES BASED ON FOURTH QUARTER 2002 FINANCIAL STATEMENTS

Deal A. Hirsch & Son Allied Tires Ames Babies ’R Us Benson Optical BestPrice! Fashions BestPrice! Kids Big Bear Big Bear Plus Big Kmart Bi-Lo BJ’s Optical Blockbuster Video Stores BOGO’s Food and Deals Books and Co Books-A-Million Cadilac Calloway’s Nursery Candies Outlet Carrow’s Casual Male Big & Tall CD Exchange CD Warehouse Champps Americana Checkers Chesapeake Bagel Bakery Cinema Ride Coco’s Cole Vision Center dELiA*s Denny’s Disc-Go-Round Dockers Outlet by Designs Duling Optical Eagle Country Market Eagle Discount Foods Ecko Unltd Eddie Bauer Einstein Bros Elegant Illusions Elegant Pretenders FAO Schwartz Florsheim Foodco Goodyear Auto Service Center Harold’s Harry’s In A Hurry Hatfield Jewelers Hearex Hearing Centers Hill’s Department Store Homeland Grocery Hometown Auto Retailers Imaginarium Imposters IPCO Optical Jennifer Convertibles Jennifer Leather Joe Mugg’s Newstand Joe’s Crab Shack Just Tire

Z-Score (1.8) 1.2 2.5 (26.6) 2.7 2.7 2.6 2.6 2.4 2.6 2.2 1.4 3.4 2.3 2.3 2.1 1.9 2.2 0.1 2.2 0.9 0.9 2.1 1.6 0.4 (11.9) 0.1 2.2 0.5 0.1 0.9 2.2 (26.6) 3.4 3.4 2.2 0.2 0.4 5.4 (1.4) 0.9 (0.4) 3.4 1.2 0.7 (4.3) (1.8) (1.0) 0.6 3.0 4.0 2.5 (1.4) (26.6) 4.7 4.7 2.3 2.1 1.2

Deal Z-Score Kids ’R Us 2.5 Kindy Optical (26.6) Kmart 2.4 Landry’s Seafood House 2.1 Levi’s Outlet by Designs 2.2 Maier & Berkele (0.9) Manhattan Bagel 0.4 Mayor’s Jewelers (0.9) Monfried Optical (26.6) Montgomery Ward Optical 2.2 National Vision Centers 1.7 Nevada Optical (26.6) Newport News 0.2 Noah’s New York Bagels 0.4 O’Charleys 2.0 One Price & More! 2.7 P&C 2.6 Paper Warehouse 1.5 Party Universe 1.5 Pearle Vision 2.2 Pearle Vision Express 2.2 Pep Boys 2.2 Perfumania 1.5 Quality 2.6 Rainforest Café 2.1 Rally’s 1.6 Right Start 0.9 Rite Aid 2.2 Rite Aid Pharmacy 2.2 Samuel’s (1.8) Schubach (1.8) Sears Optical 2.2 Service Merchandise 0.8 Silverman’s (1.8) Silversmiths and Mission Jewelers (1.8) Singer/Specs Discount Vision (26.6) Site for Sore Eyes (26.6) Sizes Unlimited 2.4 Souplantation 2.3 Southern Optical (26.6) Spiegel 0.2 Sterling Optical (26.6) Superior Optical (26.6) Sweet Tomatoes 2.3 Target Optical 2.2 The Avenue 2.4 The Avenue Plus 2.4 The Crab House 2.1 Things ed 2.2 Today’s Man 0.8 Tommy Hilfiger 2.1 Toys ’R Us 2.5 Vista Optical 1.7 Ward’s Optical 2.2 Wayne Jewelers (1.8) Wickes Lumber 2.6 Willie G’s 2.1 Wing Foot 1.2 Xando Cosi (2.8) Zany Brainy 0.9

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

237

Transforming Real Estate Finance chapter 13

Transaction Monitoring Risky Retailers

Risky tenants were determined from the Z-Score methodology, which was developed by Professor Edward Altman of NYU and implemented by the Morgan Stanley REIT equity research team. The Z-score methodology is a statistical technique, which attempts to predict financial distress of corporations using financial ratios. Altman’s methodology uses multiple discriminant analysis (MDA) and a sample of 66 firms to derive the best linear combination of the firms’ financial ratios. The discriminant function that results from the MDA considers five financial ratios, which are multiplied by different coefficients to produce a score: Z-Score:1.2* (Working Capital / Total Assets)+ 1.4* (Retained Earnings / Total Assets) + 3.3* (EBIT / Total Assets) + 0.6 * (Market Value of Equity / Liabilities) + 1.0 * (Sales / Total Assets) Based on Altman’s findings, Morgan Stanley’s equity research REIT group used a cutoff Z-score of 2.4 to identify unhealthy companies. Any publicly traded company with a Z-score below 2.4 is included as one of the risky tenants. In addition, the REIT group also included companies with stock prices below $1.00, even if their Z-scores were above 2.4. Companies that trade on the NASDAQ bulletin board or companies that had been delisted because of delinquency in filings were also included. The resulting 116 risky tenants are listed in Exhibit 10. For further discussion of the Z-score methodology and the financial ratios, see Predicting Financial Distress of Companies: Revisiting the Z-score and Zeta Models by Edward I. Altman. METHODOLOGY

In order to assess CMBS exposure to the 116 risky retailers, we used the National Research Bureau Shopping Center Directory to locate risky tenants within the CMBS deals. The Shopping Center Directory lists all tenants within 40,000 shopping centers nationwide, giving us the ability to locate many tenants that would be undetected in conventional CMBS tenant searches. The National Research Bureau Shopping Center Directory is updated semi-annually. This report was based on data in the first 2003 version released in June. CMBS data providers and prospectus material typically only list the largest three tenants within a shopping center. Therefore, the information provided in the directory allows us to look at smaller tenants within various retail properties. With both the shopping center directory and S&P Conquest data, we can track 82% of retail balances in Morgan Stanley deals. If this study were performed using data solely from the S&P Conquest database, only 32% of the retail loan balances would have complete tenant information. In total, we had some level of tenant information for 97% of the retail loans within the transactions that we examined. Sixty-nine percent of the loans were

238

covered through information from the National Research Bureau Shopping Center Directory, and had complete tenant information for the shopping centers. Twenty-eight percent of the loans were covered through data obtained from the S&P Conquest database, which only provided information on one to three tenants per shopping center. Tenant information was not available for about 3% of the retail loans. For each deal, we calculated a “weighted-risk” retail exposure (in $). We define weighted-risk retail exposure as the sum of all weighted-risk retail balances for each shopping center with risky tenants. The weighted-risk retail balance for each shopping center is computed in the following way: [Current balance of shopping center loan with risky tenant exposure]*[Sum of gross leasable areas (GLA) of risky tenants in the shopping center] / [GLA of entire shopping center] We also computed total risk exposure for each deal, which is equal to the sum of all current shopping center loan balances with risky tenant exposure. In our previous study, we did not consider the retail risk of any multiproperty loans. In this publication, we have increased the scope of our study to include multiproperty loans, most of which are found within the XL deals. To calculate weighted-risk retail balances for multiproperty loans, we considered each shopping center individually by using allocated loan amounts. SHORTFALLS/OTHER F ACTORS T O C ONSIDER

As with most studies, there are a few caveats that should be highlighted. Riskweighted exposure is not necessarily a true indicator of the riskiness of a shopping center. There are financially distressed retailers that are not captured in this study, as we only cover publicly traded companies. Risk-weighted exposure also implies that only a portion of the loan is at risk when stores become distressed. This may be true if a couple of small retailers go out of business. However, the entire loan may be at risk if a number of stores in the shopping center become distressed. We did not for this in our study. For the purposes of this study, we also examined and included all adjacent or ading retailers within a specific shopping center, regardless of whether the tenant’s space is collateral in the securitization. It is possible that we are including out-parcels or portions of shopping centers that are not part of the Morgan Stanley securitizations. However, inclusion is important since retailers are affected by the health of the shopping center as a whole. In addition, we computed risk-weighted exposures based on the size of the retailer, rather than by the portion of property cash flows represented by the retailer. An anchor store, for example, with a large GLA, would be weighted heavily, even though anchors typically pay lower rents than other smaller tenants within a mall. For example, within a newly built center, a grocery store anchor may pay rent of only $10 per square foot but occupy 70% of the GLA, while the in-line tenants may pay $20 per square foot and only occupy 30% of the GLA.

Please see additional important disclosures at the end of this report.

239

Transforming Real Estate Finance chapter 13

Transaction Monitoring SAMPLE DEAL SUMMARY exhibit 10

Source: Morgan Stanley

240

MSC 1996-C1

Please see additional important disclosures at the end of this report.

241

Transforming Real Estate Finance chapter 13

Transaction Monitoring Section IV. CMBS Ratings Actions • Rating changes at each of the three rating agencies followed similar patterns during the first three quarters of the year. At least 84% of each agency’s upgrades were on mezzanine CMBS, while more than 59% of each agency’s downgrades were on subordinate tranches.1 • Fitch was the most active, ing for more than half of all rating actions. Moody’s was the least active but ed for almost half of the downgrades. Fitch had the highest upgrade/downgrade ratio (6.4 to 1), while Moody’s had a more balanced ratio (1.3 to 1) and S&P had a ratio closer to the ratio of the overall universe (3.8 to 1). exhibit 11

2004 YTD1 RATING ACTIONS BY RATING AGENCY

Rating Agency Moody’s S&P Fitch Total

Upgrades 198 348 554 1,100

Downgrades 156 91 87 334

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

• As the rating agencies moved in concert, so did the direction of the bonds. Upgrades outnumbered downgrades on investment-grade CMBS by a ratio of 8.8 to 1, while downgrades outnumbered upgrades on non-investment grade tranches. The upgrade/downgrade ratio for non-investment grade CMBS was 0.7 to 1. This ratio was lower for non-investment grade floating rate CMBS (0.2 to 1). exhibit 12

2004 YTD1 RATING CHANGES BY CREDIT CLASS

Fixed Rate CMBS Credit Class Upgrades Downgrades Investment Grade 790 44 Non-Investment Grade 146 190 Total 936 234

Floating Rate CMBS All CMBS Upgrades Downgrades Upgrades Downgrades 155 63 945 107 9 37 155 227 164 100 1100 334

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

1

Mezzanine CMBS includes AA+ through BBB-; Subordinate CMBS includes BB+ and below.

242

• The overall upgrade/downgrade ratio was positive for both fixed and floating rate CMBS. Fixed rate classes experienced a 4.0 to 1 ratio, while floating rate classes experienced a lower ratio of 1.6 to 1. • In total, rating agencies upgraded 1,100 CMBS classes and downgraded 334, resulting in a 3.3 to 1 ratio. This is an improvement over the 1.8 to 1 ratio for year-end 2003. exhibit 13

HISTORICAL CMBS RATING ACTIONS

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

• The CMBS upgrade/downgrade ratio remains more favorable than the corporate bond ratio over the same period. Through the third quarter, corporate bonds experienced an upgrade/downgrade ratio of 0.9 to 1.

Please see additional important disclosures at the end of this report.

243

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 14

2004 YTD1 RATING CHANGES BY COHORT YEAR

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

• The 2000 cohort persists as the worst performing issuance year in of rating actions, ing for 24% of all CMBS downgrades (81) in 2004YTD. In 2003, about 14% of all tranches issued in 2000 were downgraded. Through 3Q2004, about 8% of tranches issued in 2000 were downgraded. • While almost half of the downgrades on the 2000 cohort were on floating rate classes and rake tranches in 2003, over 85% of the downgrades in 2004YTD were due to deteriorating credit fundamentals on fixed rate classes.

exhibit 15

2004 YTD1 DOWNGRADES BY CMBS ISSUANCE YEAR

Issuance Year 1998 1999 2000 2001 2002 2003 1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P, Commercial Mortgage Alert

244

% of Total Unique Tranches Downgraded 3.0 4.7 7.9 4.1 3.9 0.2

This Page Intentionally Left Blank

245

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 16 AAA AAA 0 AA+ 71 AA 154 AA19 A+ 27 A 31 A12 BBB+ 11 BBB 10 BBB4 BB+ 3 BB 0 BB0 B+ 0 B 0 B0 CCC+ 0 CCC 0 CCC0 CC+ 0 CC 0 CC0 C+ 0 C 0 C0 D 0 Total 342

AA+ 7 0 49 15 13 20 4 2 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 115

2004 YTD1 RATING CHANGES MATRIX FOR ALL CMBS AA 0 1 0 17 11 21 7 7 13 3 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 84

AA0 0 3 0 21 29 2 3 9 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 69

A+ A 0 3 0 1 3 2 0 0 1 0 46 0 30 33 11 16 9 28 4 10 1 2 1 3 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 106 100

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

246

A0 0 0 0 2 4 0 22 22 16 2 1 2 0 1 0 0 0 0 0 0 0 0 0 0 0 72

BBB+ 0 1 0 0 1 2 3 0 43 22 6 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 81

BBB 0 0 1 0 0 3 2 4 1 39 11 4 0 1 2 0 0 0 0 0 0 0 0 0 0 0 68

BBB0 0 0 1 0 0 2 3 11 0 19 9 0 0 0 0 1 0 0 0 0 0 0 0 0 0 46

BB+ 0 0 0 0 1 1 1 2 6 15 0 20 9 1 5 1 0 2 0 0 0 0 0 0 0 0 64

BB 0 0 0 0 0 1 0 0 3 5 7 0 8 2 1 2 0 0 0 0 0 0 0 0 0 0 29

BB0 0 0 0 0 0 0 0 0 6 2 14 0 2 2 0 0 0 0 0 0 0 0 0 0 1 27

B+ 0 0 0 0 0 0 0 1 0 0 0 10 10 0 10 0 0 1 0 0 0 0 0 0 0 0 32

B 0 0 0 0 0 0 0 0 2 0 0 4 6 12 0 3 0 2 0 0 0 0 0 0 0 0 29

B0 0 0 0 0 0 0 0 1 2 1 2 4 15 18 0 0 1 0 0 0 0 0 0 0 1 45

CCC+ 0 0 0 0 0 0 0 0 0 0 0 0 1 2 13 12 0 1 0 0 0 0 0 0 0 1 30

CCC 0 0 0 0 0 0 0 0 0 0 1 3 0 1 9 28 0 0 0 0 0 0 0 0 0 1 43

CCC0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 3 0 3 0 0 0 0 0 0 0 0 7

CC+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

CC 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 4 1 13 1 0 0 0 0 0 0 0 21

CC0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

C+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Please see additional important disclosures at the end of this report.

C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 2 0 0 0 0 0 4

C0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

D Total 0 10 0 74 0 212 0 52 0 77 0 158 0 96 0 82 0 163 0 128 0 57 0 74 0 43 1 40 2 64 4 57 1 3 6 31 1 2 0 0 0 2 0 0 0 0 5 5 0 0 0 4 20 1434

247

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 17 AAA AAA 0 AA+ 54 AA 125 AA14 A+ 22 A 25 A10 BBB+ 8 BBB 6 BBB2 BB+ 3 BB 0 BB0 B+ 0 B 0 B0 CCC+ 0 CCC 0 CCC0 CC+ 0 CC 0 CC0 C+ 0 C 0 C0 D 0 Total 269

AA+ 1 0 42 10 10 19 4 2 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 92

2004 YTD1 RATING CHANGES MATRIX FOR FIXED RATE CMBS TRANCHES AA 0 0 0 12 9 17 5 4 9 1 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 61

AA0 0 1 0 17 27 2 3 8 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 59

A+ 0 0 1 0 1 44 25 10 8 3 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 94

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

248

A 3 1 2 0 0 0 32 15 23 8 2 3 1 1 0 0 0 0 0 0 0 0 0 0 0 0 91

A0 0 0 0 2 1 0 18 20 14 2 1 2 0 1 0 0 0 0 0 0 0 0 0 0 0 61

BBB+ 0 1 0 0 1 0 2 0 40 21 6 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 74

BBB 0 0 1 0 0 2 1 2 1 35 10 3 0 1 2 0 0 0 0 0 0 0 0 0 0 0 58

BBB0 0 0 1 0 0 1 0 5 0 17 7 0 0 0 0 1 0 0 0 0 0 0 0 0 0 32

BB+ 0 0 0 0 1 0 1 0 1 6 0 20 9 0 4 1 0 2 0 0 0 0 0 0 0 0 45

BB 0 0 0 0 0 1 0 0 1 1 7 0 8 2 1 1 0 0 0 0 0 0 0 0 0 0 22

BB0 0 0 0 0 0 0 0 0 1 0 12 0 2 2 0 0 0 0 0 0 0 0 0 0 1 18

B+ 0 0 0 0 0 0 0 1 0 0 0 6 9 0 10 0 0 1 0 0 0 0 0 0 0 0 27

B 0 0 0 0 0 0 0 0 1 0 0 3 5 11 0 3 0 2 0 0 0 0 0 0 0 0 25

B0 0 0 0 0 0 0 0 0 0 0 1 2 14 17 0 0 1 0 0 0 0 0 0 0 1 36

CCC+ 0 0 0 0 0 0 0 0 0 0 0 0 0 1 11 12 0 1 0 0 0 0 0 0 0 1 26

CCC 0 0 0 0 0 0 0 0 0 0 0 3 0 1 7 25 0 0 0 0 0 0 0 0 0 1 37

CCC0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 3 0 0 0 0 0 0 0 0 5

CC+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

CC 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 3 1 13 1 0 0 0 0 0 0 0 19

CC0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

C+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Please see additional important disclosures at the end of this report.

C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 1 0 0 0 0 0 3

C0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

D Total 0 4 0 56 0 172 0 37 0 63 0 136 0 83 0 63 0 127 0 93 0 50 0 63 0 37 0 34 1 57 2 49 1 3 6 31 1 2 0 0 0 1 0 0 0 0 5 5 0 0 0 4 16 1170

249

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 18

AAA AA+ AA AAA+ A ABBB+ BBB BBBBB+ BB BBB+ B BCCC+ CCC CCCCC+ CC CCC+ C CD Total

AAA 0 17 29 5 5 6 2 3 4 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 73

AA+ 6 0 7 5 3 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 23

2004 YTD1 RATING CHANGES MATRIX FOR FLOATING RATE CMBS TRANCHES AA 0 1 0 5 2 4 2 3 4 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 23

AA0 0 2 0 4 2 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10

A+ 0 0 2 0 0 2 5 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 12

1

Through September 30, 2004.

Source: Morgan Stanley, Fitch, Moody’s, S&P

250

A 0 0 0 0 0 0 1 1 5 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9

A0 0 0 0 0 3 0 4 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11

BBB+ 0 0 0 0 0 2 1 0 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7

BBB 0 0 0 0 0 1 1 2 0 4 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10

BBB0 0 0 0 0 0 1 3 6 0 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 14

BB+ 0 0 0 0 0 1 0 2 5 9 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 19

BB 0 0 0 0 0 0 0 0 2 4 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 7

BB0 0 0 0 0 0 0 0 0 5 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9

B+ 0 0 0 0 0 0 0 0 0 0 0 4 1 0 0 0 0 0 0 0 0 0 0 0 0 0 5

B 0 0 0 0 0 0 0 0 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 4

B0 0 0 0 0 0 0 0 1 2 1 1 2 1 1 0 0 0 0 0 0 0 0 0 0 0 9

CCC+ 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 0 0 0 0 0 0 0 0 0 0 0 4

CCC 0 0 0 0 0 0 0 0 0 0 1 0 0 0 2 3 0 0 0 0 0 0 0 0 0 0 6

CCC0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 2

CC+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

CC 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 2

CC0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

C+ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Please see additional important disclosures at the end of this report.

C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1

C0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

D 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 0 0 0 0 0 0 0 0 0 0 4

Total 6 18 40 15 14 22 13 19 36 35 7 11 6 6 7 8 0 0 0 0 1 0 0 0 0 0 264

251

Transforming Real Estate Finance chapter 13

Transaction Monitoring Section V. Originator Delinquencies We revisited our periodic report examining the credit performance of various CMBS loan originators using September remittance report data. Delinquencies on the CMBS universe that we examined declined 29 bp to 1.38% of current balances since we last analyzed the data (January 2004 remittance reports). Cumulative losses on CMBS remained low, rising 7 bp since our last report to 37 bp. When we examined the data by originator type, as we have done in the past, we found each originator type posted delinquencies and cumulative losses fairly close to the mean (Exhibit 19). CREDIT C OORDINATES M ATRIX

In order to provide a visual representation encoming more variables (delinquencies, cumulative losses, origination year, and originator), we applied the credit coordinates matrix developed by our ABS home equity research team to this study. The average cumulative loss rate and delinquency rate for a given sample of data provide the boundary point separating the quadrants within the graphic. The lower left hand quadrant is the most desirable or “sweet spot”, with cumulative losses and delinquencies below the average. The upper right hand quadrant is the least desirable with losses and delinquencies above the average. In order to put all of the originators on the same playing field in of loan seasoning, we produced a matrix for each origination year. We focused our analysis on originations that occurred between 1995 and 2002, since CMBS issuance was limited prior to 1995, and loans originated after 2002 have low delinquencies and no cumulative losses. We also limited our analysis to the most active originators based on current balances outstanding1.

exhibit 19

Originator Type Commercial Bank Finance Company Investment Bank Insurance Company Total/Weighted Average

DELINQUENCIES AND CUMULATIVE LOSSES BY INSTITUTION TYPE

Number of Issuers 22 29 28 20 99

Original Balance ($MM) 91,563.9 79,366.7 249,928.2 25,628.6 446,487.2

Current Balance ($MM) 74,441.3 59,229.2 189,848.4 18,819.1 342,338.0

% 30 Days Del 0.26 0.27 0.19 0.04 0.21

Source: Morgan Stanley, Intex

1

For investment bank, commercial bank and finance company originators we used a $3 billion outstanding balance cut off. Since life company collateral s for only $18.8 billion in current balances, we used a $1 billion cut off.

252

exhibit 20 High losses and low delinquencies

Cumulative Losses

CREDIT COORDINATES MATRIX

High losses and High delinquencies

Average

Low losses and low delinquencies

Low losses and high delinquencies

"Sweet Spot"

Delinquencies

Source: Morgan Stanley

Regardless of the origination year, the “sweet spot” was not dominated by one originator type over time. Investment banks, commercial banks, finance companies and life companies were all represented in the lower left hand quadrant during each year. Very few originators were located in the least desirable quadrant (upper right); however, between 1995 and 1999, investment banks ed for 8 out of the 11 originators in that quadrant. Although these graphics provide a quick view of five important variables, other factors outside the scope of this analysis will affect the performance of a particular transaction. Subordination levels, structure and collateral mix will also play a role in the performance of bonds within specific deals.

% 60 Days Del 0.04 0.09 0.07 0.20 0.07

% 90 Days Del 0.34 0.77 0.41 0.66 0.47

% Foreclosure 0.12 0.15 0.15 0.10 0.14

% REO 0.42 0.41 0.56 0.23 0.48

% Total Del 1.17 1.68 1.38 1.23 1.38

Please see additional important disclosures at the end of this report.

Change Since 01/04 0.09 -0.29 -0.52 0.54 -0.29

% Cum Loss 0.30 0.48 0.35 0.48 0.37

253

Transforming Real Estate Finance chapter 13

Transaction Monitoring Most transactions are comprised of collateral from several different originators. Therefore, the loan performance of a single loan contributor should not be used to determine the credit quality of a particular transaction. Investors need to assess the performance of all originators that contribute to a transaction. For example, TOP transactions contain collateral originated by five originators: Bear Stearns, John Hancock, Morgan Stanley, Principal, and Wells Fargo. Examining the loans of any one of these entities is not necessarily an indicator of how the TOP program will fare over time. Examining the long term loan performance of all the originators contributing to TOP transactions should be a good benchmark for the credit quality within the deals. In addition to contributing collateral to TOP transactions each of these originators also contributes to other dealer programs. In the eight years of data examined, loans from these originators delivered solid credit performance. Bear Stearns appeared in the lower left hand quadrant in all eight years, Morgan Stanley, Principal, and Wells Fargo appeared in six out of eight years, while John Hancock appeared in four years. The credit coordinate matrices for 1995-2002 are contained in Exhibits 26 through 33 at the end of this piece. ORIGINATION Y EAR P ERFORMANCE

Investors are often interested in how particular loan vintages perform over time. We examined CMBS loan origination years between 1986 and 2004 and found no origination year with cumulative losses over 1.06%. Although subordination levels have been declining over time, cumulative losses of each vintage year remain well below the 5% credit enhancement on BBB CMBS issued in 2004. The 1995 vintage experienced the highest cumulative loss rate (1.06%). Approximately one-quarter of the cumulative losses on 1995 collateral were due to the $23.0 million Columbia Mall loan liquidation in DLJMA 1995-CF2 and the $17.3 million Perimeter Square Shopping Center loan in MLMI 1995-C2. Vintages with at least $20 billion of collateral outstanding followed a seasoning curve with cumulative losses rising each year between 1997 and 2004. However, vintages prior to 1997 do not follow this trend. For example, 1996 collateral had cumulative losses of 0.63% while 1997 collateral had cumulative losses of 1.00%. This may be explained by small original balances in years prior to 1997 since CMBS was in its infancy.

254

exhibit 21

CUMULATIVE LOSSES BY ORIGINATION YEAR (%)

Source: Morgan Stanley, Intex

exhibit 22

DELINQUENCIES AND CUMULATIVE LOSSES BY ORIGINATION YEAR

Original Year Pre-1986 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Total/Weighted Average*

Original Balance ($MM) 1,832.7 752.3 1,099.3 1,046.1 1,201.2 1,202.2 945.5 2,054.9 5,347.5 9,239.7 11,252.5 21,924.5 48,154.4 81,813.9 49,198.6 47,407.6 69,041.0 58,111.7 86,370.3 21,309.8 519,305.7

Current Balance ($MM) 211.4 27.2 145.4 99.5 148.4 244.4 83.4 217.3 546.0 805.2 3,372.0 9,923.6 29,418.8 58,709.3 35,046.5 33,892.5 53,291.1 50,566.9 82,974.9 21,188.7 380,912.5

% Total Del 0.16 0.00 2.26 0.00 0.00 15.60 0.00 0.03 1.51 3.40 3.82 1.91 3.28 2.00 2.50 2.03 0.97 0.32 0.06 0.00 1.27

% Cum Loss 0.10 0.30 1.01 0.29 0.77 0.37 0.80 0.32 0.32 0.67 1.06 0.63 1.00 0.57 0.42 0.28 0.07 0.02 0.00 0.00 0.33

*Total and Weighted Average of the entire universe. See the methodology section for more detail. Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

255

Transforming Real Estate Finance chapter 13

Transaction Monitoring PROPERTY T YPE

Finally, we examined cumulative losses by property type to assess the effect on an originator’s delinquency and cumulative loss performance. For example, if an originator type had an above average concentration in unconventional property types it might explain the originator’s higher than average cumulative loss rate. To evaluate how an originator’s average cumulative loss rate is influenced by property type, we calculated property type concentrations for each originator type. Our data the conventional wisdom that life insurance companies tend to be more conservative by lending less often to non-conventional property types while investment banks are more active in non-conventional property types. Life companies originate more retail loans (37%) than the average and fewer hotel loans (4%) than the average. Investment banks issue the greatest percentage of hotel loans (10%) and the lowest percentage of multifamily loans (15%). In addition to property type concentrations, we examined originator cumulative loss rates by property type. Investment bank collateral had cumulative loss rates below the average for all property types except multifamily. Finance company loans had cumulative loss rates above the average for office, hotel, and other collateral. Hotel collateral experienced the highest cumulative loss rate by property type (1.19%), but comprised only 8% of the CMBS universe we examined. exhibit 23

Commercial Bank Office 23 Hotel 5 Retail 30 Multifamily 26 15 Other1 Total 100

PROPERTY TYPE CONCENTRATION BY ORIGINATOR GROUP (%) (BY ORIGINAL BALANCE) Finance Company 22 7 25 21 25 100

Investment Bank 28 10 33 15 14 100

Insurance Company 25 4 37 15 19 100

Average 26 8 31 19 17 100

1

Other includes industrial, self storage, senior housing, and mixed use.

Source: Morgan Stanley, Intex

METHODOLOGY

This analysis of CMBS collateral was based on the Intex database, which contains $519.3 billion in original balances from 796 transactions. In the previous version of this study, our data did not contain paid down transactions within Intex, limiting our analysis to outstanding deals. In this update, we added data from paid down deals and provided cumulative loss information by originator.

256

exhibit 24

Commercial Bank Office 0.19 Hotel 1.98 Retail 0.24 Multifamily 0.16 0.23 Other1 Total 0.30

CUMULATIVE LOSSES BY INSTITUTION AND PROPERTY TYPE (%) Finance Company 0.25 1.54 0.28 0.20 0.83 0.48

Investment Bank 0.20 1.01 0.31 0.25 0.32 0.35

Insurance Company 0.30 0.22 0.95 0.08 0.18 0.48

Average 0.21 1.19 0.34 0.21 0.43 0.37

1

Other includes industrial, self storage, senior housing, and mixed use.

Source: Morgan Stanley, Intex

For the purposes of this study, we analyzed all originators with more than $100 million in original balances outstanding. After eliminating the smaller originators, our universe contained $446.5 billion in original balances from 99 different originators. All CMBS collateral was analyzed including large loans and floating rate loans. Delinquencies and cumulative losses were evaluated at the loan level. If several originators contributed to one transaction, delinquencies for each loan were assigned to the respective originator. Over the past several years, some originators have merged, so certain entities such as DLJ no longer exist. For this study, we do not combine data for originators that have merged over time. Maintaining the original underwriter’s name allows investors to benchmark their portfolios. We also grouped the collateral by originator type (commercial bank, finance company, investment bank or insurance company). Within the insurance company category, 18 of the 20 originators are life insurance companies. The other two originators (Nationwide and State Farm) predominantly provide casualty insurance. In addition to analyzing CMBS collateral performance by originator, we assessed collateral performance by origination year. We included the entire Intex universe for this analysis. Since the data is evaluated at the loan level, the origination year provided is the year in which the loan was originated, not the year in which the CMBS transaction was issued. The overall delinquency rate we provided includes 30-day delinquencies. However, it is important to note that loans often move in and out of the 30-day bucket due to timing of payments rather than fundamental deterioration. Secore is a temporary shelf used by originators that are not ed in a particular state where the property is located. For our analysis, we only had sufficient data to assign Secore originated loans to Morgan Stanley where appropriate. We did not have sufficient information on the remaining Secore collateral to assign them to their respective originators.

Please see additional important disclosures at the end of this report.

257

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 25

LOANS ORIGINATED IN 1995: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

258

exhibit 26

LOANS ORIGINATED IN 1996: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

259

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 27

LOANS ORIGINATED IN 1997: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

260

exhibit 28

LOANS ORIGINATED IN 1998: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

261

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 29

LOANS ORIGINATED IN 1999: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

Cumulative Losses (%)

2.0

TIAA (Total Delinquencies =0.00, Cumulative Losses =4.08)

1.8 1.6

Morgan Guaranty*

Morgan Stanley

1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0

Lasalle PrincipalWells Fargo Prudential GMAC Keybank Greenwich Column Merrill Lynch Bear Stearns First Union* GE Capital Lehman CIBC Bank of America John Hancock UBS Chase*Archon

0.0

2.0 NationsBank Nomura Nomura Conduit* Secore Wachovia

*Inactive originators Source: Morgan Stanley, Intex

262

4.0

GACC

Salomon Brothers

CSFB

Goldman Sachs

6.0 8.0 Total Delinquencies (%)

10.0

12.0

Investment Bank -- Plain Text Commercial Bank -- Italic Text Finance Company -- Gray Text Insurance Company -- Gray Italic Text

exhibit 30

LOANS ORIGINATED IN 2000: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

263

Transforming Real Estate Finance chapter 13

Transaction Monitoring exhibit 31

LOANS ORIGINATED IN 2001: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

264

exhibit 32

LOANS ORIGINATED IN 2002: CUMULATIVE LOSSES AND CURRENT DELINQUENCIES

*Inactive originators Source: Morgan Stanley, Intex

Please see additional important disclosures at the end of this report.

265

This Page Intentionally Left Blank

266

Chapter 14

Transforming Real Estate Finance

European CMBS

Please see additional important disclosures at the end of this report.

267

Transforming Real Estate Finance chapter 14

European CMBS Poised for Growth • CMBS poised for growth in Europe. • European CMBS remain attractive. • The CMBS advantage could close further. • Move up in rating and pick up spread. EUROPEAN M ARKET P OISED F OR G ROWTH

Growth in the European CMBS market picked up in 2004, and most analysts project that the sector is entering a period of steady increases in issuance volumes. The market saw $23 billion in issuance in 2004, a 50% increase from 2003. The slowdown in 2003 resulted from a number of factors including the ramping up of several new conduits, the delay of a large Italian deal, and a downturn in the U.K. office market.

exhibit 1

EUROPEAN CMBS ISSUANCE (IN BILLIONS OF U.S. DOLLARS)

1

Forecast by Morgan Stanley Research; SCIP is an Italian government mixed residential and commercial mortgage transaction. Source: Morgan Stanley and Commercial Mortgage Alert (CMA)

In of issuance, the European CMBS market is nine to ten years behind the U.S. The pattern of issuance growth in Europe between 1999 and 2003 parallels U.S. growth from 1990-1994. While we do not expect the same surge in European CMBS growth that occurred in the U.S. in 1998, we expect steady growth over the next five years. European CMBS has the potential to grow to a market with $40-50 billion of annual issuance within a few years. Two forces that could lead to substantial growth are the implementation of the Basel II accords in 2006 and the age of a “true sale initiative” in that would ease the process of securitization in Europe’s largest economy.

268

exhibit 2

EUROPEAN CMBS MARKET…THE NEXT U.S.?

1

Estimate by Morgan Stanley Research. F=Forecast by Morgan Stanley Research.

Source: Morgan Stanley

Historically, half or more of all European CMBS issuance has been out of the the U.K. Recently, deals have been issued in several other countries, including , Italy, Sweden, and Poland. There has also been a proliferation of “panEuropean” deals with collateral from several countries.

exhibit 3

2003 EUROPEAN CMBS ISSUANCE BREAKDOWN (BY COUNTRY)

Source: Morgan Stanley

RELATIVE V ALUE: E UROPEAN C MBS R EMAINS A TTRACTIVE VERSUS C ORPORATES

Floating-rate European CMBS spreads source moved to historically tight levels in 2004, yet remain wider than corporate and some U.S. alternative investments. For example, as of December 2004, 5-year AAA floating-rate European CMBS

Please see additional important disclosures at the end of this report.

269

Transforming Real Estate Finance chapter 14

European CMBS traded at LIBOR + 20 bp, 40 bp tighter than at the start of the year. Spreads have also tightened further down the credit curve.

exhibit 4

AAA FLOATINGRATE CMBS SPREADS, DEC. 1999- DEC 2004 (bp)

Source: Morgan Stanley

With the investor base for European CMBS increasing, the gap between CMBS and European corporates has narrowed by 40-100 bp in 2004 and could close further, in our view. In addition, the increasing amount of information on European CMBS should be another force pushing in bid-ask spreads. RATING A DVANTAGE F OR E UROPEAN C MBS

In addition to the spread advantage over corporate bonds, European CMBS are rated more conservatively than both European corporates and U.S. CMBS. European AAA subordination levels are at 20-30%, close to the U.S. levels of several years ago and substantially higher than today’s U.S. AAA levels of 12% to 18%.

exhibit 5 (%)

AAA

AA

A

BBB

BB

B

100.00

0

0

0

0

0

3.33

95.00

1.67

0

0

0

A

0

5.08

93.22

1.69

0

0

BBB

0

1.82

1.82

94.55

1.82

0

BB

0

0

0

0

100.00

0

B

0

0

0

0

0

0

CCC-C

0

0

0

0

0

0

AAA AA

Source: Fitch

270

WESTERN EUROPE – 1-YEAR RATING TRANSITION MATRIX, 1993-2002

European CMBS have been one of the best performing asset classes in of the ratio of upgrades to downgrades. Exhibit 5 shows that in the period 19932002, for example, twice as many BBB CMBS (2 x 1.82%) were upgraded as downgraded. CMBS R ISK A ND T RANSACTION S TRUCTURES

We outline below some of the key features of transaction structures and associated risks that need to be considered by investors when considering investment in CMBS. Although the description is based heavily on our experience of U.K. transactions, the general principles are applicable to all real estate markets, although legal processes will vary between jurisdictions. TRANSACTION S TRUCTURES

Exhibit 6 sets out in a simplified schematic form some of the key relationships that commonly feature in CMBS transactions.

exhibit 6

TYPICAL CMBS ISSUANCE STRUCTURE

Source: Morgan Stanley

Please see additional important disclosures at the end of this report.

271

Transforming Real Estate Finance chapter 14

European CMBS The following table highlights the common structural features of transactions. exhibit 7

KEY STRUCTURAL FEATURES OF CMBS TRANSACTIONS

Legal Structure

Exhibit 6 describes a simplified typical structure and can be used to illustrate both single-asset/single corporate sponsor and conduit-style transactions. Single asset/corporate transactions

An issuing vehicle will make a loan of bond proceeds to one or more borrowers, who in turn will be property-owning subsidiaries within a company group structure (Borrowers A and B) ultimately owned by a holding company. These borrowers will in turn be the owners of specific properties that will be leased to underlying lessees. The leases generate the rental flows to the borrowers that in turn will enable the borrowers to service the loans from the issuer. The issuing SPV can be included within the corporate structure with the same ultimate holding company owner as borrowers A and B. Conduit transactions

Conduit transactions are broadly similar except that Borrowers A and B will be unconnected and the loans to these borrowers will instead usually have been originated by a commercial lender prior to being assigned to the issuer at the time of the bond issue. The assets of Borrowers A and B will be owned property assets leased to end-s. For both single asset and conduit transactions, Borrowers A and B will normally be limited or special purpose companies with contractually limited other activities. These borrowers will fulfill no material role other than to borrow from the issuer (or intermediate borrower) and to own and lease the relevant property and to enter the asset pledge and collateral agreements. Cross-C Collateralization

An important distinction between conduit and non-conduit transactions is that, in the case of conduit transactions, the underlying borrowers are not normally part of the same corporate entity and therefore underlying properties cannot be cross-collateralized. However, excess interest on the pool of loans (in simple the difference between the interest earned on the loans and the interest costs of the bonds) can be used to help absorb losses in those conduit transactions that are not cross-collateralized. Although the absence of cross-collateralization may be a negative feature in conduit transactions, compensating features usually include borrower and property diversity

1

272

See “Commercial Mortgage Defaults: An Update” in Real Estate Finance (Spring 1999).

Asset Sales

Particularly in the case of corporate transactions, borrowers may want to preserve the option of selling specific assets from the portfolio. Structures can permit this, but for concentrated portfolios there is normally a requirement for proceeds to be generated to the extent that they raise 115%-130% of the underlying financing allocated to that property which is then used to pay down the financing. This technique avoids the potential negative adverse selection through the cherry-picking of properties and reduces the LTV of the remaining financing. Clearly, this mechanism will not apply to non cross-collateralized conduit transactions. Liquidity

Although legal structures of U.K. transactions are arranged in the expectation that underlying loans will not become embroiled in extended insolvency proceedings, liquidity facilities from highly-rated banks provide for timely payment of interest and principal in the event of borrower insolvency or temporary disruptions to cash flows due to re-tenanting. Liquidity facility providers are ultimately senior to bond investors, and such facilities usually contain restrictions on how much liquidity can be advanced to junior classes.Liquidity facilities also normally contain borrowing base restrictions which require underlying assets to provide minimum coverage levels for liquidity drawings. Interest Rate Hedging

Bond marketing considerations may require at least partial issuance in floating rate instruments. Underlying lease cash flows are by their nature fixed flows and cash flows on underlying conduit loans can be fixed. This mismatch is covered by interest rate hedging with bank counterparties. Cash Control

Rentals due on underlying leases are ideally directed to special trustee s to avoid commingling and associated corporate bankruptcy risks where applicable. Reserve s

Reserve s are funded (at the expense of the equity holder in the transaction) for various reasons. These may typically include reserves required if specified debt service coverage levels are breached or special reserves that may be required, for example, if a defined share of underlying leases are due to terminate within a certain period prior to a refinancing date. Debt service reserves are alternatively seen pending achievement of certain rental levels in lease-up situations.

Please see additional important disclosures at the end of this report.

273

Transforming Real Estate Finance chapter 14

European CMBS The following table describes the key credit risk concerns in CMBS transactions and some of the ways in which bondholder protection is structured to overcome the credit issues. exhibit 8

KEY CREDIT RISKS OF CMBS TRANSACTIONS

Tenant Quality

The credit quality of transactions will benefit from a preponderance of investment grade tenants. This is a rare feature outside of ‘trophy’-type securitizations. The value of higher rated tenants is that they are less likely to default on leases in times of economic stress, which will reduce the exposure of the transaction to re-leasing on tenant default with associated cash flow losses arising from both the time taken to re-lease the property and the associated potential cash flow losses that might arise from the need to re-lease the property at lower rental rates in difficult economic circumstances. Tenant Diversity

High tenant quality is frequently combined with limited tenant diversity. This is a feature of city office developments but less conspicuous with large-scale retail developments, where the tenant base is usually more widely spread, although the tenants may be in the same industry and retail operators may each suffer similar business stresses at the same time. Risk to transactions can therefore become concentrated: a single lessee default can cause material disruption. For example, in the Canary Wharf II transaction 65% of closing rentals was derived from 2 tenants – yet in ELoC 4 there are 440 separate tenants with a maximum rental of 4% from a single tenant. Lease Maturities

Long lease at the outset - in excess of 15 years - are a strong to a transaction as they improve the relative assuredness of cash flows. Similarly, the existence of average remaining lease exceeding 10 years at the point of any assumed refinancing also helps to provide confidence that debt can be refinanced. Lease Break Clauses and Maturities

Key analytical assumptions when reviewing real estate financings involve taking of the pattern of lease maturities and break clauses within leases. Transactions are assessed according to how well they may perform if lease termination (or lessee default) coincides with recessionary conditions, which require material discounts in rentals or rent-free periods to entice new tenants. Transactions are required to withstand progressively higher levels of discount in line with higher desired ratings on underlying securities. Rental decline assumed in recessions may range from 20-35% depending on the credit rating desired. An even distribution of lease maturities is preferable to bunching as this reduces the risk of lease maturities coinciding with economic stresses.

274

Generally, the assumptions for rated transactions are that 65% of maturing leases are renewed by tenants, with the balance re-leased, after an interval, at lower rental levels. Re-L Leasing Periods

Assumptions for time periods required for re-leasing vacated properties or finding new lessees post default will depend upon the desired rating of the underlying bonds with assumed periods normally between 15 and 18 months. The nature of real estate collateral is also key here as the assumed re-leasing periods may be lower in the case of high quality leading city developments or retail park sites that are multi-tenanted rather than isolated, specialized, single tenant sites. Macro-perceptions on the durability of certain underlying industries may also affect views on re-leasing prospects. For example, long-term perceptions concerning the viability of London as a financial centre will affect views on City office relettings, whereas views as to the threat to retail locations from home/internet shopping could impact perceptions of the Trafford Centre transaction. Valuations

Valuations at closing are obtained from leading valuation companies and transactions have been able to the issuance of BBB securities at LTVs of 70.2% (Canary Wharf Finance II), 73.1% (MS Mortgage Finance (Broadgate)) and 69.3% (Trafford Centre). The ELoC conduit transactions have been able to issue variously at BBBlevels at LTVs of 70%, BBB at 78%, BB at 76%. The underlying structure and nature of the assets will impact these levels. For example, at the BBB level Canary Wharf Finance II had an initial DSC of 1.14X (interest only) – but this was forecast to rise to 1.39x in Year 2 as discounted rents expired – whereas initial interest cover for similarly-rated tranches on the ELoC transactions ranged between 1.18x and 1.64x. Amortization Structures

Arrangements vary, although it is common for transactions to require balloon refinancing even at the end of extended debt tenors – Canary Wharf II features a £100m (21% of original principal) refinancing requirement after 30 years. The assumption is, notwithstanding the extended maturity, that the real estate can be comfortably re-financed at these levels. Trafford Centre, however, is scheduled to amortize fully over its 22-year life. In contrast, the ELoC transactions are much shorter term with final maturities not exceeding 9 years. Amortization reducing the underlying loan LTVs compared to LTVs at closing of approximately 10% occurs in all transactions and it is assumed that due to the high degree of interim amortization (principal is paid down faster for the transaction life than would be required under a 30 year mortgage-style amortization profile), and also with the benefit of projected strong interest cover levels at the refinancing date, then refinancing opportunities for the underlying loans would be available.

Please see additional important disclosures at the end of this report.

275

This Page Intentionally Left Blank

276

Chapter 15

Transforming Real Estate Finance

Japanese CMBS

Please see additional important disclosures at the end of this report.

277

Transforming Real Estate Finance chapter 15

Japanese CMBS A JPY 25 Trillion Market? The Japanese commercial mortgage-backed securities market is still in the early stages of development. In 2004, issuers brought close to US$7 billion in Japanese CMBS to market, roughly the size of the U.S. market in 1991. Does the Japanese CMBS market have the potential to grow at the rate of the U.S. market over the past decade? What are the risks and opportunities for investors in this relatively new fixed-income product? Our main findings are: Growth Potential

• The Japanese CMBS market has the potential to grow over the next decade to a market cap of JPY 25 trillion (US$200 billion) or more. Conduit Market

• If a conduit market develops, annual issuance could reach JPY 3 trillion (US$25 billion) over the next 4 to 6 years. Japanese Economy

• Long term growth in the Japanese CMBS market is dependent on the recovery in the Japanese economy and stabilization or growth in real estate values. Expanding Investor Base

• Growth of a Japanese CMBS conduit market will likely lead to an expansion of the investor base and greater liquidity. In the next section, we first review the historical development of the Japanese CMBS market. We then discuss the current structural features of Japanese CMBS and rating agency methodology. Following is a section on pricing of Japanese CMBS and relative value. Then we present data on recent Japanese CMBS and the Japanese real estate market and end with the questions an investor should ask before buying a Japanese CMBS. REASONS F OR J APANESE C MBS E MERGENCE

Much as in the U.S. CMBS market, Japanese CMBS are evolving out of distressed conditions for commercial real estate in Japan. The steep real estate recession of the late 1980s and early 1990s in the U.S. was the worst since the Depression of the 1930s. Prices of commercial real estate fell by 50% or more in some areas and delinquency rates on loans soared to all-time highs. Japan in the 1990s has seen real estate values fall as much as 90% after the bubble period of the 1980s.

278

Before the mid-1990s the U.S. real estate business was predominately a private market. Lending was dominated by a handful of banks, life insurance companies, and pension funds. Losses on commercial loan portfolios led to the exit of many traditional lenders from the commercial mortgage market. Regulators and rating agencies turned more negative on commercial mortgage holdings, so that the remaining lenders became less willing to extend credit. As the chart below shows, thrifts and insurance companies went from supplying over $40 billion in commercial and multifamily loans in 1985 to decreasing their holdings by $46 billion by 1992. exhibit 1

NET SUPPLY BY INSURANCE COMPANIES AND THRIFT INSTITUTIONS TO THE U.S. COMMERCIAL/ MULTIFAMILY MORTGAGE MARKET (IN BILLIONS OF $) Source: Federal Reserve Board, Flow-of-Funds

In Japan, we could see a similar process growing out of the current downturn. Many banks and insurance companies are in financial distress and are less willing to extend long-term credit to real estate borrowers. As in the U.S., the Japanese CMBS market is a natural substitute for traditional originators of commercial mortgages. A B RIEF H ISTORY O F C MBS: U .S. A ND E UROPE

In the U.S., investment banks started to apply securitization legal structures, and technology developed during the 1970s and 1980s for residential mortgage backed securities to commercial mortgages. In the mid- to late-1980s, issuers securitized a few loans on single properties into CMBS. Packaging of diversified pools of mortgages into CMBS developed in the U.S. in the early 1990s when the Resolution Trust Corporation (RTC) pooled nonperforming loans from failed institutions. Some transactions exceeded $1 billion and led to the growth in the investor base for CMBS. After the success of the RTC transactions, CMBS gained wider acceptance with investors and nongovernment, or “private-label conduit,” issuers.

Please see additional important disclosures at the end of this report.

279

Transforming Real Estate Finance chapter 15

Japanese CMBS Issuance of CMBS in the U.S. grew rapidly in the mid-1990s, reaching $78 billion in 1998. In Europe, CMBS has also taken hold as a financing vehicle, with $13 billion issued in 2001. Most of the transactions are out of the United Kingdom, but deals have been done in several other countries. The Second Stage

As the table below shows, we believe that the Japanese CMBS market is currently in the second stage of development, comparable to the early 1990s in the U.S. This phase comes at the end of a period of distress for real estate. This period lasted for about five years in the U.S. – in Japan it could go on for longer. Toward the middle or end of this period we would expect to see the growth of conduit deals and a decrease in CMBS backed by distressed deals. exhibit 2

CMBS GROWTH: U.S. AND JAPAN (BILLIONS OF U.S. $)

Year Stage 1 – Early Development 1989 1990 Stage 2 – Distressed Assets 1991 1992 1993 1994 1995 Stage 3 – Conduit Growth 1996 1997 1998

RTC

United States NonRTC Total

Japan Year

Total

0 0

4.1 3.4

4.1 3.4

1998 1999

0.0 1.1

2.7 8.8 3.1 3.0 1.2

5.0 5.2 14.1 14.4 16.7

7.6 14.0 17.2 17.4 17.8

2000 2001 2002p 2003 2004

5.2 4.9 9.0 TBD TBD

0.0 0.0 0.0

28.8 40.4 77.7

28.8 40.4 77.7

2005 2006 2007

TBD TBD TBD

p= Morgan Stanley Forecast; TBD= to be determined Source: Morgan Stanley, Commercial Mortgage Alert

Japanese CMBS: JPY 25 Trillion?

In its next phase, we believe that the Japanese CMBS market has the potential to grow in the next decade to a market capitalization of JPY 25 trillion (US$200 billion) or more. In order for this growth to occur, in our view, a Japanese version of the U.S. conduit market must develop. A conduit originates loans solely for the purpose of securitization and not for portfolio holding. The development of CMBS conduits spurred growth of the U.S. CMBS market. Conduit originations for more than half of all CMBS collateral in the U.S. In Europe as well, conduits have been a main contributor to the growth of the CMBS market.

1

Moody’s Investors Service, “Japanese Securitization Market: 2001 Year in Review and 2002 Outlook,” May 9, 2002.

280

exhibit 3

RATIO OF CMBS ISSUANCE TO GDP

% of GDP

1.0 0.8 0.6 0.4

US

0.2 Japan 0.0 US: 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 Japan: 1998 1999 2000 2001 2002

Source: Morgan Stanley, Commercial Mortgage Alert

Conduits do not want to hold collateral for a long period. They therefore tend to originate standardized pools of mortgages since this speeds the securitization process. This standardization attracts money managers who are not real estate experts. These investors do not have to spend a lot of time analyzing a pool and are attracted by the commodity-like nature of conduit pools. The broadened investor base has been one of the main factors behind the growth of the CMBS market in the U.S. Although still in the early stages, conduits are starting to develop in Japan. It is our belief that conduits will not start to grow at a rapid pace until the Japanese economy begins a sustained recovery. In the U.S., conduit originations did not grow rapidly until the real estate recovery was under way. In the initial stages of the market, growth in conduit originations depends on both a high level of real estate transactions and investor willingness to buy securities related to real estate. Another important element is the willingness of some investors to hold the unrated and lower-rated classes of CMBS, which are a levered real estate investment. If the conduit market does eventually take off, we believe the Japanese CMBS market could eventually become as large a share of the mortgage market as in the U.S. The size of the U.S. market is rapidly approaching $400 billion dollars, or about 20% of the value of all commercial and multifamily mortgages. Although there are not exact numbers available, we estimate that there is the potential for the commercial and multifamily mortgage market to grow to JPY 125 trillion (US$1 trillion). This assumes a slightly lower mortgage to GDP ratio than in the U.S. If we also assume that the Japanese mortgage market continues to grow and the Japanese CMBS share of mortgages approaches the U.S. ratio, the size of the Japanese CMBS market has the potential to grow to JPY 25 trillion (US$200 billion). We think that annual issuance could approach JPY 3 trillion (US$25 billion). This would put the ratio of Japanese CMBS to Japanese GDP roughly in line with the current ratio in the U.S.

Please see additional important disclosures at the end of this report.

281

Transforming Real Estate Finance chapter 15

Japanese CMBS STRUCTURAL F EATURES

CMBS have very simple structures compared to their residential mortgage counterparts. Bonds are almost always sequential pay, with amortization, prepayments, and default recoveries paid to the most senior remaining class. The lowest rated remaining class absorbs losses, after equity and reserves are reduced to zero. In the U.S., commercial mortgages almost always have some form of prepayment penalty, so credit analysis plays a more important role than prepayment analysis. In Japan, call protection has not become a standard feature yet, but we believe that this is necessary for a conduit market to develop. CMBS are static pools of commercial real estate loans divided into tranches with varying subordination levels and credit ratings. In the U.S., a typical transaction has about 90% investment grade bonds, concentrated in AAA securities, with the remaining 10% noninvestment grade. Interest only (IO) bonds can be stripped off all or part of the structure. A typical structure consists of sequential pay, fixed rate bonds. The AAA bonds are time-tranched with a 5-year AAA bond ahead of a 10-year AAA bond. The subordination level for a AAA conduit deal ranges from 15% to 25%. In Japan, since the market is still in its early stages, no standard structure has yet developed. Most performing loan deals, however, have the sequential structure described above. The tranches are typically 5 to 7 year bullets, with an additional 2 to 3 years until the legal final maturity. The extra time until the legal final is to allow for workouts of loans that default at the bullet payment date. CALL P ROTECTION, S ERVICING, A ND O THER F EATURES

Call protection on Japanese CMBS is not as strong as in the U.S, where defeasance is the most common form of prepayment penalty. For a defeased loan, the prepaying borrower must place Treasury securities into the trust to generate cash flow that matches the mortgage payments. For European CMBS, fixed-rate loans usually have call protection. Most floating-rate loans in Europe have weak or no call protection. In Japan, defeasance is rarely used, although some loans have prepayment penalties. The Trustee, Master Servicer and Special Servicer each play an ongoing role in the transaction. The Pooling and Servicing Agreement, Prospectus, and other legal documents outline each party’s responsibilities and fees. Typically, the Trustee is responsible for reporting monthly payments and collateral performance data to certificate holders. The Master Servicer is responsible for servicing all performing loans and monitoring loan document requirements. The Special Servicer resolves defaulted or delinquent loan issues. In both Japan and the U.S., in addition to the mortgage collateral, credit enhancements may be in the form of reserve funds, guarantees, letters of credit, cross-collateralization and cross-default provisions. Loans within the pool may have certain cash control provisions such as a “lock box” that requires payments from tenants to go directly to the trust instead of through the borrower if certain default triggers occur. Virtually all loans within CMBS are bankruptcy remote.

282

Under new ing rules issued by the Japan Association of Certified Public ants, an originator cannot move a securitized asset off balance sheet unless its holding of the subordinate class is less than 5%. The impact of this change has been minimized by the development of a sub-investment grade investor base. THE M YCAL C HALLENGE

In 2001, the Japanese corporation Mycal, a major retailer, went into bankruptcy. A bankruptcy court challenged the bankruptcy-remoteness of the collateral in two CMBS backed by shopping centers leased by Mycal. Although the specific case is still being negotiated, most observers believe that it should not have any negative long-term effects on securitization in Japan. Rating agencies have downgraded some classes of the deals because of collateral performance, but have not taken any negative rating actions because of the case. They also have not changed rating criteria. In June 2002, Emeritus Professor Shindo of Tokyo University published an article ing the position of investors in the Mycal CMBS case. The opinion was a counter-argument to the opinion of Professor Yamamoto of Kyoto University, which Mycal’s made public in May. Shindo believes that receivables in a securitization should not be part of bankruptcy reorganization under the Japanese Corporate Rehabilitation Law. In Japan, the rating agencies have published limited credit enhancement guidelines for CMBS deals. We believe, however, that the rating agencies will apply a similar methodology to the rating of Japanese conduit CMBS. The table below shows Standard and Poor’s published criteria for Japanese CMBS. exhibit 4 JAPANESE CMBS BASE-CASE CMBS RATING CRITERIA Rating AAA AA A BBB BB B

DSCR 2.25-2.5 2.0-2.25 1.8-2.0 1.6-1.8 1.4-1.6 1.2-1.4

LTV (in %) 35-40 40-47.5 47.5-55 55-60 60-65 65-75

Source: Standard and Poor’s

Early on, some rating agencies looked at the volatility of real estate values compared to the U.S. as a basis for Japanese CMBS ratings criteria. Recently, however, most have used a cash flow approach to evaluating credit risk. This methodology tends to lead to more stable valuations than using individual property appraisals.

Please see additional important disclosures at the end of this report.

283

Transforming Real Estate Finance chapter 15

Japanese CMBS MONITORING

After a transaction is issued, rating agencies monitor deals for changes in credit risk. As the Japanese CMBS market is relatively new and many deals are private, it is often difficult for investors to receive detailed monthly monitoring reports. In the more developed U.S. market, private vendors such as Trepp, RealPoint, and Conquest provide loan level monitoring details on every transaction. As the Japanese CMBS market develops, we expect to see a growth in the number of readily available monitoring reports and the amount of detail in each report. Improvement in monitoring systems will help the investor base for Japanese CMBS to grow, as less time will be needed to keep track of investments in the sector. LACK O F J APANESE D EFAULT D ATA

In Japan there is scant default data and not enough deals or history to evaluate rating changes. How, then, can we assess the expected default rates on Japanese CMBS? The short answer is we can’t, at least to the degree of confidence as in the U.S. What we can do, though, is look at what sort of scenario it would take to cause default on a AAA Japanese CMBS. We can then attempt to judge the probability of such an event occurring. We estimate that the combination of subordination and equity in current Japanese CMBS exceeds 50% and can be as much as 65% on performing loan deals. Rating agencies (and originators) also usually reduce or “haircut” the current cash flows of a property in their underwriting. Given these assumptions, we estimate that it would take a fall in current property values of 70% or more to expose the AAA-rated classes to losses. BBB classes could withstand a fall of prices in excess of 40% in most cases. PRICES D OWN 5 3%

The Japan Real Estate Institute (JREI) has the longest time series of property values in Japan. The JREI publishes a commercial land price index for urban areas that dates back to the mid-1950s. In Japan, a high percentage of property value is in land since there is a limited amount of space for development. The national commercial land price index has fallen 53% from March 1990 through exhibit 5

120

COMMERCIAL LAND PRICES IN JAPAN

100

Nationwide 6 Largest Cities

80 60 40 20 0 Mar-55

Source: JREI

284

Oct-66

Jun-78

Feb-90

Sep-01

March 2002. This is by far the worst period of decline in the 47-year history of the index. In large cities, prices have fallen even more. The commercial index for large cities has fallen by 84% from 1990 to 2002. While it’s possible that the index could decline further, a drop of the magnitude to threaten AAA CMBS seems unlikely. A further 70% fall in the land price index would put its value in nominal close to the level in March 1967. In real , the fall in prices would put values close to the level of the early 1960s, or almost 90% below the peak value in 1990. In large cities, an additional drop of 60% would put values at less than 5% of the peak. A similar percentage drop in another bubble market, the NASDAQ, equates to a decline from the peak of 5000 to 250, less than 20% of its value as of June 2002. INSULATION

In the U.S. CMBS market, the high subordination levels for investment grade CMBS insulated those securities from the recent downturn in U.S. real estate markets. In 2001 and 2002, vacancies in many areas of the U.S. were moving to double-digit levels as real estate markets declined, yet investment grade CMBS spreads were flat to tighter. Subordination had the effect of removing much of the direct real estate risk from the higher-rated classes of CMBS. In Japan, it is too early to say whether subordination levels are of the same degree of conservatism as in the early days of U.S. CMBS. Under current rating criteria, however, the Japanese real estate market could undergo further significant declines without causing defaults on investment grade Japanese CMBS. The severe downturn of the past ten years could be repeated without causing a default of most AAA-rated Japanese CMBS. Since most Japanese CMBS have a maturity of 5 to 7 years, the decline would also have to occur in a much shorter time frame than the current downturn to affect investment grade bonds.

Please see additional important disclosures at the end of this report.

285

Transforming Real Estate Finance chapter 15

Japanese CMBS Case Studies: Selected Japanese CMBS EAST R EALTYCO., L TD.

Issuer Amount Issue date Final maturity Collateral Ratings

East Realty Kabushiki Kaisha JPY 83.9 billion (US$670 million) March 2002 2009 Expected maturity: 2007 Tobu Dept. Store (Ikebukuro) and adjacent buildings; Tobu Dept. Store (Funabishi) and adjacent buildings; land in Harajuku. Standard and Poor’s, Fitch

STRUCTURE/PRICING Class A B C D E

Amount (Bil. of JPY) 44.1 10.2 10.4 11.2 8.0

Rating AAA AA A BBB BB+

RATING A GENCY P OSITIVES

• • • • • • •

Location of stores Prime location of Harajuku land Geographical diversification High alternative use Tobu as creditworthy tenant Liquidity facility Strong trustee (Yasuda Trust)

Spread to LIBOR (in bp) 60 80 100 160 350

Coupon Fixed Fixed Fixed Fixed Fixed

RATING A GENCY C ONCERNS

• • • •

Concentration risk in Ikebukuro store Single tenant risk Outlook for domestic retail Illiquid property types (railway facilities and department stores) • Balloon risk (nonamortizing loans)

LTV AND DSC RATIOS Class A B C D E

Rating AAA AA A BBB BB+

S&P LTV 36 44 52 61 68

Fitch LTV 35.8 44.1 52.6 61.7 68.2

S&P DSC 2.77 2.25 1.89 1.61 1.46

Fitch DSC 2.53 2.05 1.72 1.47 1.33

Source: S&P Presale Report, 2/18/2002; Fitch New Issue Report, 3/26/2002

Reserves Payments Underwriter Property Manager Trustee Rating Analyst Source: S&P and Fitch Presale Reports

286

JPY 3 billion capital expenditure reserve; JPY $1.5 billion operational expense reserve; JPY 4.2 billion security reserve Semi-annual Mizuho Securities (lead), Daiwa, Kokusai Tobu Railway Yasuda Trust S&P: Tomoyoshi Omuro 3593-8584; Fitch: Masaaki Kudo 3288-2830

J-C C MBS-1 1 L TD.

Issuer Amount Issue date Final maturity Collateral Ratings

J-CMBS-1 Ltd. JPY 21 billion (US$171 million) May 2000 2007 Expected maturity: 2004 13 properties in central Tokyo. Largest loan: Akasaka 1, an office building (27%) Standard and Poor’s/Moody’s/Fitch

STRUCTURE/PRICING Class A B C D X

Rating (Moody’s/ S&P/Fitch) Aaa/AAA/AAA Aa2/AA/AA A2/A/A Baa2/BBB/BBB Aaa/AAA/AAA

Amount (Bil. of JPY) 13.0 2.5 2.5 3.0 N/a

Coupon Floating Floating Floating Floating Floating

RATING A GENCY C ONCERNS

RATING A GENCY P OSITIVES

• • • • • •

Spread to LIBOR (in bp) 30 45 70 130 N/a

Cross-collateralized portfolio Location in central Tokyo Diversified tenant base Conservatively underwritten cash flow Cash trap trigger at DSCR of 1.50x Well capitalized sponsors

• Low occupancies in some buildings • Supply concerns in Tokyo office market • No amortization

LTV AND DSC RATIOS Class A B C D X

Rating AAA AA BBB BBB AAA

S&P LTV 36 43 50 58 NA

Fitch LTV

S&P DSC 2.6 2.1 1.9 1.6 1.46

Fitch DSC

Source: S&P Presale Report, 5/26/00

Reserves Payments Underwriter Property Manager Trustee Servicer Rating Analyst

Holdback of JPY 3 billion for 6 months Semi-annual Salomon Brothers Mitsui Fudosan Cititrust GMAC Commercial Mortgage S&P: Kenji Kondo 3593-8590

Source: S&P Presale Report

Please see additional important disclosures at the end of this report.

287

Transforming Real Estate Finance chapter 15

Japanese CMBS JLOC 2 001-II I L TD.

Issuer Amount Issue date Final maturity Collateral Ratings

Centennial Funding JPY 71 billion (US$590 million) November 2001 2008 Expected maturity: 2006 20 trust certificates backed by 313 properties (80% office) Moody’s/Standard and Poor’s/Fitch

STRUCTURE/PRICING Class A B C D

Rating (Moody’s/ S&P/Fitch) Aaa/AAA/AAA Aa2/AA/AA A2/A/A Baa2/BBB/BBB

RATING A GENCY P OSITIVES

Amount (Bil. of JPY) 50.0 8.0 7.0 6.0

Spread to LIBOR (in bp) 35 55 80 125

Coupon Floating Floating Floating Floating

RATING A GENCY C ONCERNS

• Potential volatility of prices in • Geographical diversity small cities • Property locations in central business • Office concentration districts • Asset manager has established disposition plan • 10% concentration in one office building • Completed liquidation of some assets • Higher vacancy rates than • Assets formerly owned by Daihyaku Mutual market Life • Fast-pay triggers dependent on debt service coverage • Asset manager experience Other: Interest and principal payments dependent on cash flow from buildings and proceeds from sale of properties. Proceeds from sale of properties used to pay down classes sequentially, starting with the AAA class. No scheduled amortization. Reserves Payments Underwriter Property Manager Trustee Asset Manager

Cash reserves of JPY 4.4 billion Quarterly Morgan Stanley Nomura Real Estate Development Co., MF Building, Sumitomo Real Estate, Pacific Development Bankers Trust Morgan Stanley Properties, Japan

Source: Moody’s/S&P/Fitch Presale Reports

288

The Japanese Real Estate Market It has long been my contention that spreads on investment-grade CMBS (in the U.S. market) have little correlation with real estate fundamentals. In 1998, the U.S. commercial real estate market was booming, but spreads on CMBS moved to the widest level ever after the Russian default and Long Term Capital failure. In 2002, spreads tightened 20 bp or more for all investment-grade conduit ratings even though the real estate market was weakening and office vacancies exceeded 20% in many cities. I believe that in Japan, CMBS spreads will similarly be detached from real estate fundamentals. I think that this will be especially true for diversified conduit deals. Rating agencies have required high levels of subordination for investment-grade rated CMBS, providing insulation from further moderate declines in real estate. Despite this insulation, it is nonetheless instructive to examine the current state of the Japanese real estate market. Lower-rated CMBS could be affected by real estate trends and even highly rated-classes might feel the impact of a sharp downturn. Ten-Y Year Decline

The Japanese real estate market has been in almost steady decline for the last decade. After the bubble period of the 1990s, prices have declined in every region in Japan. On a nominal basis, land prices in Tokyo and Osaka have declined to near the pre-bubble levels of the early 1980s. In real , or as measured as a ration to GDP, prices have fallen even further.

exhibit 6

REGIONAL LAND PRICES

160

Kanto (Tokyo) Kinki (Osaka) Chubu- Tok (Nagoya) Tohoku (Sendai) Kyushu (Fukuoka)

140 120 100 80 60 40 Mar-85

Mar-89

Mar-93

Mar-97

Mar-01

Source: JREI

Please see additional important disclosures at the end of this report.

289

Transforming Real Estate Finance chapter 15

Japanese CMBS exhibit 7

RATIO OF LAND PRICE INDEX TO GDP

Source: JREI

290

Chapter 16 Transforming Real Estate Finance

Factors to Consider Before Investing in CMBS

Please see additional important disclosures at the end of this report.

291

Transforming Real Estate Finance chapter 16

Factors to Consider Before Investing in CMBS ECONOMIC/INTEREST R ATE O UTLOOK

Investors should be aware of the growth outlook for the U.S. economy. Investment-grade bonds tend to do well on a total return basis when GDP growth is slow and the Federal Reserve is in an easing mode. Credit spreads, however, might widen and defaults rise during an extended slowdown. SWAP S PREADS

Investment-grade CMBS and swap spreads are highly correlated. The swap spread represents the price of exchanging a fixed-rate cash flow for a floatingrate one. Swap spreads are also a proxy for overall credit risk. A ten-year swap spread of Treasuries + 80 bp means that one party must pay the 10-year U.S. Treasury (UST) yield plus 80 bp to the counterparty to receive a floating-rate (LIBOR) cash flow. If a CMBS has a yield of the UST + 130 bp and the swap spread is 80 bp, then the CMBS is said to trade at Swaps + 50 bp. That is, the purchaser of the CMBS receives UST +130 bp, can pay out UST + 80 bp, and receive Libor plus the 50 bp difference between 130 bp and 80 bp. Investors should be aware of the historical trading ranges of CMBS to swaps for each rating category. CMBS buyers should put current spreads in the context of historical data and be able to explain circumstances that might drive spreads outside of trading ranges. GLOBAL R ISK

CMBS investors should also have a view of the global economy and potential effects of global credit risk on U.S. fixed-income markets. In 1998, the Russian debt crisis had a major impact on U.S. markets, including CMBS. REAL E STATE F ACTORS Real estate cycle

High subordination levels insulate most investment-grade CMBS investors against default during real estate downturns of the magnitude experienced over the past 30 years. Non-investment grade buyers are more vulnerable to weakening real estate conditions. All investors, however, should monitor changes in macro real estate trends to judge if spread widening could occur versus Treasuries, swaps, or other sectors. Real estate data

There are a myriad of sources from which to obtain data on real estate conditions. Providers include: • American Council of Life Insurers (commercial mortgage delinquencies and originations). • Federal Reserve Board (Beige Book on regional economic conditions; flow-offunds data on commercial and multifamily originations). • U.S. Census bureau (construction). • Torto Wheaton Research (vacancy data by property type and market).

292

• CB Commercial (vacancies and rents). • F.W. Dodge (construction). • REIS (property market overviews). • PricewaterhouseCoopers (property market overviews). • Smith Travel Research (hotel occupancies and room rates). • Moody’s, Standard & Poor’s, Fitch (periodic reports on real estate and CMBS). RELATIVE V ALUE

Investors use several benchmarks for CMBS spreads: • Single-A bank and finance – Formerly used as a benchmark for AAA CMBS; now used for both AAAs, AAs, and single-As. • Unsecured REITs – Benchmark for BBBs and BBB- CMBS. • Single-A corporate industrials – Benchmark for investment-grade CMBS. • Mortgage -through OAS – Comparison for AAAs. • ABS – Benchmark for short AAAs. CMBS investors examine the historical relationships among CMBS spreads and those from each of these sectors. “Relative value” analysis involves judging whether the divergence of spreads represents a buying opportunity.

Please see additional important disclosures at the end of this report.

293

Transforming Real Estate Finance chapter 16

Factors to Consider Before Investing in CMBS BOND-S S PECIFIC

In analyzing a specific class of CMBS, an investor should consider the following factors: Ratings

Most CMBS have at least two ratings. The rating agencies in the U.S. market are Moody’s, Standard and Poor’s (S&P), and Fitch. Some investors require that either Moody’s or S&P rate the bond. A potential investor should check if bond purchased in the secondary market is on ratings watch. Rating agencies state that they continually monitor outstanding ratings. Fitch and S&P conduct annual reviews and are more likely to change a rating at that time. Rating agency analysts are available to answer credit questions about new issues or bonds in the secondary market. With a new issue, it is sometimes valuable to call the rating agency that has not rated the bond, since that agency is likely to have analyzed the credit more conservatively. Subordination levels

An investor should compare the subordination level of the bond in question to others in the market and to older transactions. Lower subordination is not necessarily an indication of lesser credit quality. Issuers with highest quality collateral obtain the lowest credit enhancement levels, and will often have the lowest delinquency rates. Subordination levels have trended down over time. An investor should be comfortable that the current enhancement levels are appropriate for the expected future default rates. Issuer quality

Spreads in the CMBS market have tiered based on the perceived quality of the issuer’s collateral. As a general rule, bank and insurance companies are viewed as having the highest credit quality mortgages. Investors should consider the possibility that an issuer will exit the CMBS business. Even though CMBS are bankruptcy remote, the failure of an issuer might lead to spread widening because a market perception of reduced liquidity. One quantitative check on the quality of an issuer’s collateral is the performance of seasoned transactions. Morgan Stanley and other dealers publish delinquency rates by issuer. Cash flows

Investors can model simple cash flows on Bloomberg for almost all CMBS. More detailed modeling services include Trepp LLC and Conquest. These services allow for loan-by-loan default modeling and also provide detailed monitoring information. Liquidity

Investors should try to determine how many dealers make a market in a particular CMBS. If only one or two dealers trade a bond, the market will place a liquidity on the security.

294

Property/Regional concentration

Concentrations of various property types or within regions are important factors in analyzing CMBS. Property type or regional concentrations above 40% of a pool may raise a warning flag. Non-standard property types such as cold storage, health care, or manufactured housing communities also draw increased scrutiny. This is not to say that these property types should be avoided. Instead, an investor should make sure that the rating agencies have made proper adjustments and that the pricing reflects any potential risk. Deals with high (greater than 30%) concentrations of multifamily loans are often viewed favorable, since Federal agencies are more likely to purchase these deals. ERISA-eeligibility

Effective August 23, 2001, the Dept. of Labor ruled that CMBS rated as low as BBB- are eligible as pension fund amendments under ERISA. For most deals closed before that date, the issuer needs to amend the original documents to achieve ERISA-eligibility for bonds rated less than AAA. As of August 2001, Morgan Stanley, Nomura, and Bank of America had amended deals issued from their shelves before the effective date. Remittance reports

For seasoned transactions, an investor should obtain the most current remittance report and special servicing report. Every month, the trustee compiles the report, which details distributions and loan delinquencies. Special servicing reports highlight loans transferred from the servicer to the special servicer. Total delinquency rates of less than 2% are usually not a concern for AAA investors. Thirty-day delinquency rates are of lesser concern than 60 day+ rates. exhibit 1 CREDIT ENHANCEMENT

PNCMA 2001-C1 A2

Class A1 A2 B C1 C2 C2X D E F G H J K L

Current Balance

n

Type

141,114 560,781 33,060 18,856 12,000 12,000 11,020 8,816 13,224 7,714 16,530 14,326 5,510 8,816

5.91 6.36 6.58 6.80 2.57 — 6.93 — — — 5.91 5.91 5.91 5.91

Fixed Rate Fixed Rate Fixed Rate Fixed Rate Fixed Rate IO, Other Non-Fi Fixed Rate WAC/ Thru WAC/ Thru WAC/ Thru Fixed Rate Fixed Rate Fixed Rate Fixed Rate

At Issuance Credit Rating Enhancement AAA/Aaa AAA/Aaa AA/Aa2 A/A2 A/A2 A/A2 A–/A3 BBB+/Baa1 BBB+/Baa2 BBB–/Baa3 BB+/Ba1 BB/Ba2 BB–/Ba3 B+/B1

19.75% 19.75% 16.00% 12.50% 12.50% — 11.25% 10.25% 8.75% 7.88% 6.00% 4.38% 3.75% 2.75%

The sources for exhibits 1-6 are cash flow runs on Trepp LLC analytics on Bloomberg.

Please see additional important disclosures at the end of this report.

295

Transforming Real Estate Finance

Factors to Consider Before Investing in CMBS

chapter 16

Price/Yield tables

The following tables show details for a new issue conduit transaction, PNCMA 2001-C1. The deal represents an average CMBS transaction. The tables show the resilient nature of both AAA and BBB bonds under a number of default and prepayment scenarios. The first table shows the capital structure of the transaction, with credit and ratings. The AAA bond has 19.75% subordination and the BBB 8.75%. These levels were close to the average for conduit transactions as of August 2001. The following table shows the types of call protection on the loans for each year after issuance. In the first year, 97.42% of the loans have defeasance or are locked-out (LO). exhibit 2 PREPAYMENT PENALTY MATRIX

PNCMA 2001-C1 A2

No.

Date

LO

YM

PP>=5% PP>=4% PP>=3% PP>=2% PP>=1%

None

1

8/01

97.42%

2.58%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

2

8/02

96.51%

3.49%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

3

8/03

94.24%

5.76%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

4

8/04

92.85%

7.15%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

5

8/05

92.90%

5.88%

0.00%

0.00%

0.00%

0.00%

0.00%

1.22%

6

8/06

90.87%

7.41%

0.00%

0.00%

0.00%

0.00%

0.00%

1.72%

7

8/07

92.46%

7.17%

0.00%

0.00%

0.00%

0.00%

0.00%

0.37%

8

8/08

91.90%

6.39%

0.00%

0.00%

0.00%

0.00%

0.00%

1.71%

9

8/09

91.17%

6.48%

0.00%

0.00%

0.00%

0.00%

0.00%

2.35%

10

8/10

71.60%

2.40%

0.00%

0.00%

0.00%

0.00%

0.00%

26.01%

11

8/11

100.00%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

12

8/12

100.00%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

The yield table below shows the effects of changing prepayments on the yield of the AAA bond, priced at 101-22 1/4. Note that the yield, average life, and spread are almost unchanged across a wide range of prepayments. R or sometimes “Y,” stands for the annual prepayment rate on loans not in lockout or defeasance. Before year 10, no more than 8.83% (100% – 91.17% in the previous table) of the loans are able to prepay. In addition, the 5-year AAA class is absorbing the effects of prepayments. exhibit 3 PRICE/YIELD TABLE

PNCMA 2001-C1 A2

R when PP <= X% 101-22¼ Price WAL Sprd Swaps

296

0.00

15.00

30.00

45.00

60.00

6.172

6.172

6.172

6.171

6.171

9.35

9.34

9.33

9.31

9.30

+45.02

+45.06

+45.11

+45.17

+45.24

The yield table below shows the effects of changing prepayments on the yield of the BBB bond, priced at 101-1. As with the AAA, note that the yield, average life, and spread are almost unchanged across a wide range of prepayments. exhibit 4 PRICE/YIELD TABLE

PNCMA 2001-C1 F

R when PP <= X% 100-1 Price WAL

0.00

15.00

30.00

45.00

60.00

7.073

7.073

7.073

7.073

7.073

9.62

9.61

9.59

9.57

+132.02 +132.10

+132.21

+132.33

9.62

Sprd Swaps

+131.98

The following yield table shows the effects of defaults on the AAA class. The default rate is an annual default rate of the remaining balance, abbreviated as CDR. The yield changes very little up to 3% CDR, an extremely high default rate by historical standards. The change in yield at higher CDRs is caused by the reduction of the average life of a security.

exhibit 5 PRICE/YIELD TABLE

PNCMA 2001-C1 A2

Default Rate 101-22¼ Price WAL Sprd Swaps

0.00

0.50

1.00

2.00

3.00

6.172

6.171

6.170

6.166

6.162

9.35

9.31

9.25

9.10

8.91

+45.02

+45.19

+45.43

+46.06

+46.92

The yield table below shows the effects of defaults on the BBB class. The yield is unchanged up to 2% CDR. At a default rate of 3% CDR, principal is not affected, but the yield decreases slightly. The yield decrease is from the shortening of a bond.

exhibit 6 PRICE/YIELD TABLE

PNCMA 2001-C1 F

Default Rate

0.00

0.50

1.00

2.00

3.00

100-1 Price

7.070

7.070

7.070

7.070

7.159

9.62

9.62

9.62

14.13

+132.01 +132.01

+132.02

+107.51

WAL Sprd Swaps

9.62 +132.01

Please see additional important disclosures at the end of this report.

297

This Page Intentionally Left Blank

298

Transforming Real Estate Finance

Glossary

Please see additional important disclosures at the end of this report.

299

Transforming Real Estate Finance

Glossary Anchor Store A major grocery, discount retailer, or department store that attracts shoppers to the mall. Anchored Strip Center A shopping center with at least one anchor store. All else being equal, an anchored strip center is more valuable than an unanchored strip because of the extra flow of shoppers. ASERS (Appraisal Subordinate Entitlement Reductions)/CVA (Collateral Valuation Adjustments) Structural deal feature that estimates unrealized losses on defaulted

loans and prevents payment of current interest on the estimated losses. Designed to prevent conflicts between the interests of subordinate and senior classes. Assisted Living Facilities A property type targeted to elderly needing assistance, but not full time medical care. These facilities typically consist of apartment style units with a kitchenette. The operator of the facility provides three meals a day and assists residents with daily activities such as feeding, bathing, dressing, and medication reminders. Average Daily Rate Total guest room revenue divided by total number of

occupied rooms. B-p piece See Subordinated Tranche. Call Protection The main forms of call protection in CMBS loans are lockout, penalty points, yield maintenance and defeasance. The purpose of call protection is to protect the lender from borrower prepayment. Defeasance and lockout provide for the most stable cash flows. The only cash flow variability will be a result of credit events. Capital Expenditures (Cap Ex) Extraordinary expense items necessary to maintain the property. Examples would include repairing a roof, repaving a parking lot, or replacing heating and air conditioning systems. Capitalization Rate Rate of interest (or return) used to convert a series of property cash flows into a building value. The higher the cap rate, the riskier the asset. CDR Annualized default rate expressed as a percentage of the remaining pool balance; a default is assumed to be a liquidation. Class A Properties Trophy quality properties with higher quality finishes and prominent locations. Class B Properties Generic real estate; 10-20 years old, well maintained, average locations, fewer amenities. Class C Properties Older properties needing renovation; uncertain future. Community Center Over 100,000 - 275,000 square feet of space; multiple anchors

but not enclosed. Congregate Senior Housing These are independent living facilities that provide a common dining facility and other services. Congregate senior housing has no medical component, but may provide access to emergency medical care through call buttons. Not licensed as a nursing home.

300

Continuing Care Retirement Communities These facilities offer the entire continuum of seniors housing from independent living to skilled nursing facilities. Residents move within the facility depending on the level of care required. Licensed operator. Co-o op Loans/Blanket Loans Very low loan to value multifamily loans. Loans senior to

co-op share loans. Co-T Tenancy Provisions A provision in a retail lease which permits the tenant to cancel its lease if another major tenant closes. R The constant rate of the outstanding collateral principal expected to be prepaid in a year. Also see Constant Prepayment Yield (Y). Y A modified R that assumes prepayments to be zero until all yield maintenance provisions are expired. CMBS IOs are priced assuming a 100 Y speed. Credit Tenant Lease A type of retail lease. All payments guaranteed by credit of tenant (i.e., Wal-Mart). Debt Service Constant Annual Principal and Interest Payment/Original Loan Amount. Debt Service Coverage Ratio (DSCR) Net Income/Annual Debt Service. An indicator of protection from cash flow volatility. Defeasance Upon prepayment of the loan, the borrower is required to provide the lender with U.S. government securities in an amount such that the lender receives the same yield as if the borrower had not prepaid the loans. From the lender’s perspective, a defeased loan is positive. The yield is the same, but there is a credit upgrade from commercial real estate credit to U.S. Government credit. Distribution Space Structure in an industrial property. Principal use is distribution or light assembly. Minimal office space as a percentage of total space (typically 0–10%). “Clear heights”, 24’ ceiling heights, are the minimum for modern distribution buildings. Higher clear heights are more economical for the tenant as they stack goods vertically and rent fewer square feet. Dollar Price The price at which bonds are currently trading, with a $100 dollar pricing being a par bond. Double Wide/Single Wide Describes the size of manufactured home that a given slab will . The double wide segment has experienced the fastest growth due to the growing acceptance of manufactured homes as a single family housing alternative. E-C Curve The front end of the swap curve is constructed with Eurodollar futures contracts, which are future contracts on 3-month LIBOR. Unlike the J-spread, which is a spread over the bond’s average life point on the Treasury curve, the E-spread is a single constant spread that is added to each relevant point on the Eurodollar futures curve. EAs (Extension Advisor) Provides representation for the senior classes in a loan modification process. Typically the transaction documents require the special servicer to get approval from the EA to grant any extensions to a loan beyond a certain date. Fee Simple Ownership of both land and building in perpetuity.

Please see additional important disclosures at the end of this report.

301

Transforming Real Estate Finance

Glossary FF&E Furniture, Fixtures, and Equipment; standard hotel underwriting includes a deduction as an operating for the ongoing replacement of FF&E, typically 4% to 5% of gross revenue. This differentiates hotel underwriting from apartment underwriting where some of those same expenses are considered capital expenditures and are not an operating expense deducted from NOI. Typical hotel refurbishment should occur at least every 7 years. Flex Space/Office Warehouse Higher amount of office space as a percentage of total space. This results in higher tenant improvement costs necessary upon lease renewals. These tenants are less sensitive to clear heights and more sensitive to accessibility of qualified labor pools. Franchise Fee Fee paid to hotel company that allows hotel owner to “fly the flag” of the hotel company (i.e., Marriott, Sheraton, etc). Fee ranges from 4% to 7% of gross revenue. Full Service Hotel A hotel that offers banquet and convention services; one or more full service restaurants. Fusion Transaction Commercial Mortgage Alert defines a fusion deal as a transaction that has conduit style loans and either has one loan that is more than 10% of the pool balance or all loans of $50 million or more are at least 15% of the deal. Market participants often classify a deal as a fusion transaction if the top 10 loans for at least 40% of the collateral. Garden Apartments Multiple buildings; usually no more than 2-3 stories. Go Dark Provisions A provision in a retail lease. Prevents tenant from vacating the space while continuing to pay rent; landlords like this because this vacant space is a detriment to other stores’ sales at the center. Ground Lease Building owner leases land from land owner. Haircut Difference between the loan underwriter cash flow and the cash flow used by the rating agency to size the loan. Examples of haircuts given by the rating agencies include above market rents and apartment occupancies greater than 95%. High Rise Apartments Over three stories; usually located in downtown areas. I-C Curve Interpolated Treasury curve. Pricing a 9.5-year CMBS bond as a spread to the I-Curve involves interpolating yields for the remaining maturities of the on-therun 5-year and 10-year Treasury notes. If both notes were issued three months ago, the interpolation would involve 4.75 years as the starting point and 9.75 years as the ending point. For more details, see pages 90-91. Independent Living Facilities Multifamily apartment complexes catering to senior citizens. They supply few services beyond building and ground maintenance. These facilities are unregulated. In-L Line Store Smaller store within a center (i.e.., Foot Locker or Hallmark Cards

store). Investment Grade Tranches Tranches in a CMBS transaction that are rated between

AAA and BBB-.

302

J-C Curve Interpolated nominal Treasury curve. Pricing a 9.5-year CMBS bond as a spread to the J-Curve involves interpolating yields for the original maturities of the on-the-run 5-year and 10-year Treasury notes. The J-Curve ignores any seasoning that may have taken place in either issue. For more details, see pages 90-91. Junior AAA Tranches See Super Senior AAA Tranches. LTV See Loan to Value. Leasehold Interest Building owner leases land. Lender wants ground lease term to exceed loan amortization period. Leasing Commissions Fees paid to brokers to bring tenants for office leases, typically $2-4 per square foot at lease gs. Landlords bear this expense. Levered IO (Interest Only) The Levered IO is stripped off of the senior and surbordinate classes of the CMBS structure. The levered IO has more exposure to collateral defaults than the PAC IO, which is stripped off of the mezzanine bonds in the capital structure. The levered IO is AAA rated because of its top priority in receiving cash flows. Limited Service Hotel No food service other than continental breakfast, minimal public space and small staff. Loan to Value (LTV) Loan Amount/Appraised Value. A 70% LTV is the average for commercial loans in CMBS transactions. Lockout Borrower prohibited from prepaying; same as a non-call in corporate bonds. Loss Severity Realized loan loss (in dollars)/remaining outstanding loan balance at the month of write-off. MAI (Member, Appraisal Institute) Designation given to certified appraisers. Manufactured Housing Communities The land, streets, utilities, landscaping, and concrete pads under the homes comprise a manufactured housing community. The homes are independently financed. Homeowners pay monthly rent for the pad to the manufactured housing community owner. Master Servicer Receives a 2-10 bp fee to collect monthly mortgage payments and reserve payments required by the loan documents. Mezzanine Debt Debt secured by borrower’s equity interest in the property. There is no lien on the property. Usually this refers to debt in addition to the first mortgage. Mezzanine Tranches Typically the CMBS tranches rated between AA and BBB. They are not as senior as the AAA tranches, but are senior to the B-pieces. NOI (Net Operating Income) Property revenues minus property expenses.

Please see additional important disclosures at the end of this report.

303

Transforming Real Estate Finance

Glossary OA (Operating Advisor) Majority owner of the first loss class controls certain aspects of the special servicing; generally preferred by rating agencies as an enhancement to credit. OAS (Option-A Adjusted Spread) A measure of spread that adjusts for default and call

option costs. Occupancy Rate Number of occupied units/total number of units. PAC IO See Levered IO. Pad Concrete slab that s each manufactured home in a manufactured housing community. Pari u Loan Pari u loans are created when an issuer splits a large loan into smaller pieces. These smaller pieces are often securitized in separate transactions and are paid monthly principal and interest simultaneously and on a pro-rata basis. Power Center/Big Box Predominantly anchor tenants and few small stores. Typically big discounters or mass retailers. Prepayment Penalty Points A prepayment penalty that is equal to a percentage of the remaining loan balance (i.e., 5%, 4%, 3%, etc). Generally, the least preferable form of call protection. Recapture Provisions A provision in a retail lease. Permits the owner to cancel a lease and to regain control of space after a tenant closes its store. Regional Mall Over 750,000 square feet with several department stores (2-3) as

anchors. REO (Real Estate Owned) A property becomes REO upon foreclosure of the loan. The lender or trust now owns the property. RevPAR (Revenue Per Available Room) Average Daily Rate x Occupancy Rate. Used mainly for analyzing hotel/lodging properties. Rollover Term used to describe expiration of a tenant lease in an office property. Lease are generally for 5-10 years; credit tenants may be longer. It is preferable not to have rollover concentrations, which would expose an owner to uncertain rental market or potentially reduce NOI below debt service. Self-S Storage Facility Commercial property that leases storage space to individuals or businesses on a month-to-month basis. The average self-storage facility has between 40,000 and 10,000 square feet of rentable space divided among 400 to 1,000 individual units. Shadow Anchored Strip Similar to an anchored strip except that the shopping center does not own the anchor store. Shadow Rating A rating on an individual loan in a large loan pool given by the rat-

ing agencies.

304

Skilled Nursing Facilities Independent nursing homes or a designated wing in a hospital. The facility provides full-time licensed skilled nursing, medical and rehabilitative services. Average length of stay can range from 2 months to 2 years, or more. 24-hour care is provided, with doctors and ed nurses on call. Facilities are licensed by the state; operator must obtain a certificate of need from the state before beginning operation. Special Servicer Handles workout situations of delinquent or defaulted loans for a fee. Usually, the special servicer owns the most subordinate bonds in the transaction. Real estate expertise is key for special services. Stressed DSCR Rating Agency Adjusted Cash Flow/Debt Service Payment using an assumed debt service constant. Sub Servicer A servicer who performs specialized tasks for the master or special servicers. The master or special servicer that contracts out the work remains legally responsible. Super Regional Mall Over 1 million square feet with multiple department (4-5) stores

as anchors. Super Senior AAA Tranches The most senior tranches in a CMBS transaction. They are structured with higher credit levels than the junior AAA class. Tenant Improvements Costs to build walls, ceilings, carpet for a new office tenant. Typically $5-40 per square foot. The landlord usually incurs this expense. In strong demand markets the landlord can this expense through to the tenant in of a higher rental rate. In weak markets, landlords must take tenant improvements out of net income, which reduces cash flow. Tilt-U Up Construction This is the preferred construction type for industrial buildings. It includes pre-cast concrete s that are “tilted-up” on a steel frame. Tilt-up is preferable to corrugated metal exteriors for maintenance reasons. Trailer Park This is a lower-end asset class, often confused with manufactured housing communities. Trailer parks are highly transient, dense communities of homes on wheels. Tenants are provided with no amenities other than simple utility hookups. Not typically seen in conduit pools. Triple Net Lease A type of retail lease. Tenant pays rent, real estate taxes, expenses, and maintenance. Trustee The trustee represents all investors in a CMBS transaction and acts as the trust that holds the title to the collateral. The trustee is responsible for enforcing the pooling and servicing agreement, supervising the master and special services, and distributing the monthly proceeds to the bond holders.

Please see additional important disclosures at the end of this report.

305

Transforming Real Estate Finance

Glossary UCF (Underwritten Cash Flow) The cash flow number used by the lender to establish a desired debt service coverage ratio. Typically includes a standardized deduction from net operating income to for expenses need to maintain the property. Net Operating Income less reserves. Unanchored Strip No major destination type tenant. Usually smaller local tenants. Location needs natural traffic to be successful. Properties perform best if they are located in a highly developed area with little vacant land. WAC (Weighted Average Coupon) The higher the WAC, the greater the prepayment risk; the lower the WAC, the greater the extension risk. WAL (Weighted Average Life) The average number of years until all mortgage principal is expected to be paid off, weighted by the dollar balance of the mortgages. A more widely used measure than duration for CMBS. Yield Maintenance A prepayment penalty that requires a borrower to make a penalty payment to the lender if the lending rate at the time of prepayment is lower than the mortgage rate on the loan. Yield maintenance formulas vary, but generally, they discount the difference between the remaining contractual mortgage payments and a hypothetical mortgage payment at current market rates. Most yield maintenance formulas use the Treasury rate as the discount rate to determine the present value of the difference in the two payments. Using the Treasury rate is advantageous to the lender, since the lender made the loan at the Treasury rate plus a spread, but recovers the remaining cash flows at the Treasury rate. Yield maintenance is generally assumed to equal lock-out in modeling CMBS transactions. Z-S Score A calculation developed by Professor Edward Altman of NYU to assess financial risk of a company; based on data contained in the company’s financial statements. Z-Score = 1.2* (Working Capital / Total Assets) + 1.4* (Retained Earnings / Total Assets) + 3.3* (EBIT / Total Assets) + 0.6* (Market Value of Equity / Liabilities) + 1.0* (Sales / Total Assets)

306

Transforming Real Estate Finance

Disclosures

Please see additional important disclosures at the end of this report.

307

Transforming Real Estate Finance

Disclosures CREDIT PRODUCTS RATING DISTRIBUTION TABLE (AS OF DECEMBER 1, 2004) Rating Overweight Equal-weight Underweight Total

# of Ratings in Coverage Universe 193 466 210 869

% Total Fixed Income Instrument Coverage 22% 54% 24% 100%

Coverage includes all companies that we currently rate.

ANALYST R ATINGS D EFINITIONS

Overweight (O) Over the next 6 months, the fixed income instrument’s total return is expected to exceed the average total return of the relevant benchmark, as described in this report, on a risk adjusted basis. Equal-weight (E) Over the next 6 months, the fixed income instrument’s total return is expected to be in line with the average total return of the relevant benchmark, as described in this report, on a risk adjusted basis. Underweight (U) Over the next 6 months, the fixed income instrument’s total return is expected to be below the average total return of the relevant benchmark, as described in this report, on a risk adjusted basis. More volatile (V) The analyst anticipates that this fixed income instrument is likely to experience significant price or spread volatility in the short term. Important Disclosures on Subject Companies

The information and opinions in this report were prepared by Morgan Stanley & Co. Incorporated and its s (collectively, "Morgan Stanley") and the research analyst(s) named on page one of this report. Morgan Stanley policy prohibits research analysts from investing in securities/instruments in their MSCI sub industry. Analysts may nevertheless own such securities/instruments to the extent acquired under a prior policy or in a merger, fund distribution or other involuntary acquisition. Morgan Stanley is involved in many businesses that may relate to companies or instruments mentioned in this report. These businesses include market making, providing liquidity and specialized trading, risk arbitrage and other proprietary trading, fund management, investment services and investment banking. Morgan Stanley trades as principal in the securities/instruments (or related derivatives) that are the subject of this report. Morgan Stanley may have a position in the debt of the Company or instruments discussed in this report. Other Important Disclosures

The securities/instruments discussed in this report may not be suitable for all investors. This report has been prepared and issued by Morgan Stanley primarily for distribution to market professionals and institutional investor clients. Recipients who are not market professionals or institutional investor clients of Morgan Stanley should seek independent financial advice prior to making any investment decision based on this report or for any necessary explanation of its

308

contents. This report does not provide individually tailored investment advice. It has been prepared without regard to the individual financial circumstances and objectives of persons who receive it. Morgan Stanley recommends that investors independently evaluate particular investments and strategies, and encourages investors to seek the advice of a financial advisor. The appropriateness of a particular investment or strategy will depend on an investor's individual circumstances and objectives. You should consider this report as only a single factor in making an investment decision. Morgan Stanley fixed income research analysts, including those principally responsible for the preparation of this research report, receive compensation based upon various factors, including quality, accuracy and value of research, firm profitability or revenues (which include fixed income trading and capital markets profitability or revenues), client and competitive factors. Analysts' compensation is not linked to investment banking or capital markets transactions performed by Morgan Stanley or the profitability or revenues of particular trading desks. This report is not an offer to buy or sell any security/instrument or to participate in any trading strategy. In addition to any holdings disclosed in the section entitled "Important Disclosures on Subject Companies," Morgan Stanley and/or its employees not involved in the preparation of this report may have investments in securities/instruments or derivatives of securities/instruments of companies mentioned in this report, and may trade them in ways different from those discussed in this report. Derivatives may be issued by Morgan Stanley or associated persons. Morgan Stanley makes every effort to use reliable, comprehensive information, but we make no representation that it is accurate or complete. We have no obligation to tell you when opinions or information in this report change. With the exception of information regarding Morgan Stanley, reports prepared by Morgan Stanley research personnel are based on public information. Facts and views presented in this report have not been reviewed by, and may not reflect information known to, professionals in other Morgan Stanley business areas, including investment banking personnel. The value of and income from investments may vary because of changes in interest rates, foreign exchange rates, default rates, prepayment rates, securities/instruments prices, market indexes, operational or financial conditions of companies or other factors. There may be time limitations on the exercise of options or other rights in securities/instruments transactions. Past performance is not necessarily a guide to future performance. Estimates of future performance are based on assumptions that may not be realized. This report may include research based on technical analysis. Technical analysis is generally based on the study of trading volumes and price movements in an attempt to identify and project price trends. Technical analysis does not consider the fundamentals of the underlying issuer or instrument and may offer an investment opinion that conflicts with other research generated by Morgan Stanley. Investors may consider technical research as one input in formulating an investment opinion. Additional inputs should include, but are not limited to, a review of the fundamentals of the underlying issuer/security/instrument.

Please see additional important disclosures at the end of this report.

309

Transforming Real Estate Finance

Disclosures To our readers in Taiwan: Information on securities/instruments that trade in Taiwan is distributed by Morgan Stanley & Co. International Limited, Taipei Branch (the “Branch”). Such information is for your reference only. The reader should independently evaluate the investment risks and is solely responsible for their investment decisions. This publication may not be distributed to the public media or quoted or used by the public media without the express written consent of Morgan Stanley. Information on securities/instruments that do not trade in Taiwan is for informational purposes only and is not to be construed as a recommendation or a solicitation to trade in such securities/instruments. The Branch may not execute transactions for clients in these securities/instruments. Certain information in this report was sourced by employees of the Shanghai Representative Office of Morgan Stanley Dean Witter Asia Limited for the use of Morgan Stanley Dean Witter Asia Limited. This publication is disseminated in Japan by Morgan Stanley Japan Limited; in Hong Kong by Morgan Stanley Dean Witter Asia Limited; in Singapore by Morgan Stanley Dean Witter Asia (Singapore) Pte., regulated by the monetary authority of Singapore, which accepts responsibility for its contents; in Australia by Morgan Stanley Dean Witter Australia Limited A.B.N. 67 003 734 576, a licensed dealer, which accepts responsibility for its contents; in Canada by Morgan Stanley Canada Limited, which has approved of, and has agreed to take responsibility for, the contents of this publication in Canada; in Spain by Morgan Stanley, S.V., S.A., a Morgan Stanley group company, which is supervised by the Spanish Securities Markets Commission (Cnmv) and states that this document has been written and distributed in accordance with the rules of conduct applicable to financial research as established under Spanish regulations; in the United Kingdom by Morgan Stanley & Co. International Limited, authorised and regulated by the financial services authority; and in the united states by Morgan Stanley & Co. Incorporated and Morgan Stanley DW Inc., which accept responsibility for its contents. In Australia, this report, and any access to it, is intended only for “wholesale clients” within the meaning of the Australian Corporations Act. NOT FOR DISTRIBUTION TO PRIVATE CUSTOMERS AS DEFINED BY THE UK FINANCIAL SERVICES AUTHORITY. The trademarks and service marks contained herein are the property of their respective owners. Third-party data providers make no warranties or representations of any kind relating to the accuracy, completeness, or timeliness of the data they provide and shall not have liability for any damages of any kind relating to such data. This report may not be sold or redistributed without the written consent of Morgan Stanley. Morgan Stanley research is disseminated and available primarily electronically, and, in some cases, in printed form. Additional information is available upon request.

310

This book is an overview of the Commercial Mortgage-Backed Securities (CMBS) market.

Transforming Real Estate Finance

The contents of this publication are over eight years in the making and include excerpts of research reports from as early as 1997. In this fifth edition of our primer, we have reorganized the chapters to highlight the different investment options within CMBS. New material since our last edition includes sections on the various types of AAA CMBS classes, total rate of return swaps, floating rate large loan transactions, and an updated version of the commercial mortgage default study.

World Wide Web Sources

We hope you find this book useful and welcome comments so that we can improve future editions.

CMBS Information

TRUSTEES

LaSalle StateStreet Wells Fargo Chase

www.etrustee.net www.statestreet.com www.wellsfargo.com/com/comintro.jhtml www.jpmorgan.com/absmbs

SERVICERS

Amresco Midland GMAC

www.amresco.com www.midlandls.com www.gmaccm.com

RATING AGENCIES

Fitch Moody’s S&P

www.fitchratings.com www.moodys.com www.standardandpoors.com

AGENCIES

Fannie Mae Ginnie Mae

www.fanniemae.com www.ginniemae.gov

s U.S. CMBS TRADING/CAPITAL MARKETS

Jon Strain, [email protected]

212.761.2270 212.761.2164 212.761.2055 212.761.2177 212.761.2090 212.761.2106 212.761.1317

EUROPEAN CMBS RESEARCH

Howard Esaki, [email protected]

44.20.7677.7592 44.20.7677.8670

U.S. CMBS RESEARCH

Marielle Jan de Beur, [email protected]

212.761.1454 212.761.1080

U.S. PRODUCT SPECIALISTS

Bob Karner, [email protected]

212.761.1605 212.761.2040

THE AMERICAS 1585 Broadway New York, NY 10036-89293 Tel: (1) 212-761-4000 EUROPE 25 Cabot Square, Canary Wharf London E14 4QA, England Tel: (44 20) 7425-8000 JAPAN 20-3, Ebisu 4-chome, Shibuya-ku Tokyo 150-6008, Japan Tel: (813) 5424-5000 AISA PACIFIC Three Exchange Square Hong Kong Tel: (852) 2848-5200

© 2005 Morgan Stanley