5 - CMBS Primer2

SECOND EDITION
SPRING 2002
Transforming
Real Estate Finance
A CMBS Primer
Editors: Howard Esaki, Marielle Jan de Beur, Masumi Pearl
This book is an overview of the commercial mortgage-backed securities (CMBS) market. In some sense, this book has been over five years in the making, as it includes excerpts from research publications from as early as 1997. We also include new material on property types backing the loans in CMBS and a glossary of real estate and CMBS terminology. Among the many contributors to this book are: Howard Esaki, Marielle Jan de Beur, Masumi Pearl, Molly Paganin, James Hiatt, Robert Karner, Ryan Marshall, Scott Stelzer, Jonathan Strain, Robert Gillman, Joseph Philips, Robert Restrick, Lisa Schroer, Steven LHeureux and Mark Snyderman. We hope you find this book useful and welcome comments so that we can improve future editions.
Transforming Real Estate Finance
A CMBS Primer
1
NEW
INTRODUCTION 2 9
To begin, we give a brief history of the development of the CMBS market. We then discuss simple CMBS structures and review the rating process for a CMBS.
2 3
A CREDIT TEST
10 27
This chapter is our most recent review of CMBS data, market events, and relative value. We will update this section twice a year.
MAJOR PROPERTY TYPES IN CMBS 27 45
In this chapter, we give an overview of the property types backing CMBS and give rating agency views of each real estate sector. NEW
4 5
R E TA I L C O L L AT E R A L I N C M B S
46 55
This chapter describes the methodology of our recent retail study, which examines Morgan Stanley underwritten CMBS deals for exposure to 86 risky retail tenants. NEW
H O T E L C O L L AT E R A L I N C M B S 56 83
Hotels have often been viewed by rating agencies and investors as one of the riskiest property types. This chapter discusses the credit exposure of CMBS to hotels, recent hotel performance, the history of the hotel industry, and hotel branding.
CALL PROTECTION
84 95
Commercial mortgages, unlike residential mortgages, usually have some form of prepayment penalty. In this section, we describe the various forms of call protection on mortgages in CMBS.
LARGE LOANS
96 107
Loans of greater than $50 million are a growing part of the CMBS market. This chapter describes the large loan market and uses an option-adjusted spread model to evaluate relative value. REVISED
M U LT I FA M I LY M B S
108 115
In addition to private-label securities, the CMBS market also includes multifamily agency securities issued by Ginnie Mae, Fannie Mae, and Freddie Mac. This chapter gives a brief overview of the major agency programs. REVISED
CMBS IOs
116 123
CMBS IOs are interest-only securities stripped off of premium coupon mortgages. Unlike residential IOs, the main risk element of CMBS IOs is not prepayments, but defaults.
10 11
REVISED
C O M M E R C I A L M O R T G A G E D E FA U LT S
124 131
This chapter reports on the history of commercial mortgage defaults at insurance companies.
EUROPEAN CMBS 132 161
We detail the growing European CMBS market in this section.
12
TRANSACTION MONITORING
162 179
As the real estate cycle turns and CMBS age, more problem loans will appear in seasoned CMBS deals. This chapter gives examples of Morgan Stanleys quarterly tracking report (detailing specially serviced loans in Morgan Stanley issued transactions) and monthly CMBS delinquency report. NEW
13
C M B S D E L I N Q U E N C I E S B Y O R I G I N AT O R
180 191
Commercial banks, finance companies, investment banks and life insurance companies are the four major types of CMBS loan originators. This chapter examines delinquencies on CMBS loans by issuer type.
14
FA C T O R S T O C O N S I D E R B E F O R E I N V E S T I N G I N C M B S G L O S S A RY
192 197 198 203
The final section is a glossary of terms frequently used in the CMBS market.
Transforming Real Estate Finance chapter 1
Introduction
Commercial Mortgage Backed Securities (CMBS) are bonds backed by pools of mortgages on commercial and multifamily real estate. As of January 2002, the market capitalization of the CMBS market was slightly over $350 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. In this introductory chapter, we first review the historical development of the CMBS market and the structure of CMBS. In the next chapter, we present an overview of the current state of the markets. We then discuss the types of real estate backing commercial mortgage loans, forms of call protection used in CMBS, default risk, large loans in CMBS, and CMBS IOs. Next is a chapter on evaluating commercial mortgage default risk. We follow with a description of the rapidly growing European CMBS market. We provide examples of our various CMBS monitoring reports. We end with the questions an investor should ask before buying a CMBS.
DEVELOPMENT OF THE MARKET
Before the mid-1990s the US 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.
C M B S : D I V E R S I F I E D P U B L I C R E A L E S TAT E
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 mid-1990s 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-government, or privatelabel, issuers. Issuance of CMBS in the US grew rapidly in the mid-1990s, peaking at $78 billion of US issuance in 1998. Since then, annual domestic issuance has stabilized in the $50-70 billion range. Outside the US, CMBS has also taken hold as a financing vehicle, with $13 billion issued in Europe in 2001. 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 2001, about $5 billion of CMBS came to market in Asia.
S T R U C T U R E S A N D R AT I N G A G E N C I E S
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 6), 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 9) 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.
IO
exhibit 1
% Subo rdination 22% AAA (78% ) AA (4%) A (5.5%) BBB (3.5%) BB (4.5%) B (2.5%) UR (2%)
NEW ISSUE MARKET CMBS STRUCTURE
18%
12.5%
9%
4.5%
2%
0%
Source: Morgan Stanley
Please refer to important disclosures at the end of this report.
Introduction
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. 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 partys 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.
R AT I N G A G E N C Y M E T H O D O L O G Y
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 cashflow (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 Property type Loan-to-value ratio Debt service coverage ratio Fixed/floating rate Loan seasoning Direct versus correspondent lending Real Estate Outlook Deal Specific Number of loans in the deal Loan size Degree of subordination Balloon extension risk Quality of master servicer and special servicer
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

Credit Enhancement
OFFICE Individual PROPERTIES Loan Coverage
LTV
30 40 50 60 65 70 75 80
DSCR
2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15
AAA
3.9 7.8 13.1 20.9 27.2 35.8 47.1 60.2
AA
2.9 5.8 9.7 15.5 20.2 26.6 34.9 44.6
A
2.2 4.3 7.2 11.5 14.9 19.6 25.8 33.0
BBB
1.5 3.0 5.0 8.0 10.4 13.7 18.0 23.0
BB
1.0 1.6 2.7 4.4 5.7 7.5 9.9 12.6
B
0.5 1.0 1.1 1.7 2.3 3.0 3.9 5.0
REGIONAL Individual MALLS Loan Coverage
Credit Enhancement
LTV
30 40 50 60 65 70 75 80
DSCR
2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15
AAA
2.0 3.9 6.5 10.5 13.6 17.9 23.5 30.1
AA
1.5 2.9 4.8 7.8 10.1 13.3 17.5 22.3
A
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
BB
0.5 1.0 1.4 2.2 2.8 3.8 4.9 6.3
B
0.5 0.5 1.0 1.0 1.1 1.5 2.0 2.5
Source: The Rating of Commercial Mortgage-Backed Securities, Duff and Phelps Credit Rating Co.
Introduction
exhibit 3
continued
BASE-CASE CREDIT ENHANCEMENT GUIDELINES FOR VARIOUS PROPERTY TYPES
INDUSTRIAL/ ANCHORED RETAIL Individual PROPERTIES Loan Coverage
Credit Enhancement
LTV
30 40 50 60 65 70 75 80
DSCR
2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15
AAA
2.5 5.0 8.4 13.4 17.4 22.9 30.1 38.5
AA
1.9 3.7 6.2 9.9 12.9 17.0 22.3 28.5
A
1.4 2.7 4.6 7.3 9.5 12.6 16.5 21.1
BBB
1.0 1.9 3.2 5.1 6.7 8.8 11.5 14.7
BB
0.5 1.1 1.8 2.8 3.6 4.8 6.3 8.1
B
0.5 0.5 1.0 1.1 1.4 1.9 2.5 3.2
MULTIFAMILY Individual PROPERTIES Loan Coverage
Credit Enhancement
LTV
30 40 50 60 65 70 75 80
DSCR
2.50 2.00 1.75 1.50 1.45 1.35 1.25 1.15
AAA
2.6 5.2 8.7 13.9 18.1 23.9 31.4 40.1
AA
1.9 3.9 6.5 10.3 13.4 17.7 23.3 29.7
A
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
0.5 1.1 1.8 2.9 3.8 5.0 6.6 8.4
B
0.5 0.5 1.0 1.2 1.5 2.0 2.6 3.3
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-support levels to the extent that Fitch uses discrete DSC and LTV buckets while the other rating agencies use a continuous variation of creditsupport 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 3 shows the addition and subtractions that rating agencies apply to the credit support level.
exhibit 4
SUBJECTIVE ADJUSTMENTS TO CREDIT SUPPORT LEVELS

Special Adjustment1 Additions Reductions Other Factors Applied on a Case-By-Case Basis Floating Interest Rates Asset Quality Market Barriers to Entry Cash Flow Volatility Loan Size Location Assessment Concentration
Amortization Fully Interest Only Less Than 5 Years Seasoning Servicer/Special Servicer Assessment Quality Assessment Origination Information Cash Control Reserves
20% 10-20% 10% 20% 20%
5% 10% 5% 5% 10-20% 5-10%
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).
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-support 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.
Introduction
R AT I N G A G E N C I E S A N D I N V E S T O R S M O N I T O R C O N D U I T Q U A L I T Y
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 pass through the filters of both rating agencies and investors. If rating agencies recognize a slide in credit-quality rating, they will increase credit support 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 support 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 Moodys have AAA credit support 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 support 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 support levels. The rating agencies thus act as a first level of defense against a potential unobserved fall in loan credit quality.
A Credit Test
The terrorist attacks in September cast a pall over financial markets, and had reverberations that continue to shake fixed-income markets. The CMBS market was also affected, as worries increased about large loan deals, hotels, and retail properties. For the most part, however, the CMBS sector has thus far been relatively unscathed. Spreads on investment-grade bonds have widened by only a few basis points and global issuance set a new record. But falling rents and rising delinquencies could be the first serious test of CMBS credit since the explosive growth in the market in the mid-1990s. The outlook for 2002 holds more uncertainty than in past years, stemming from doubts about the course of the economy next year. With most forecasters believing a rebound in the second half will occur, the fundamentals for real estate should not pose a problem for investment-grade CMBS. We think that total delinquencies on seasoned CMBS will top 2.5% next year, but remain less than half of the peak reached on insurance company collateral in 1993. At these levels, AAAs would be unaffected, but we could see further widening of noninvestment grade CMBS. The major risk, in our view, is that the economy suffers another shock in 2002. This is not a low probability eventthere has been a major dislocation in the economy or fixed-income markets in each of the last five years. Our opinion continues that securitized products in general, and CMBS in particular, are well suited as a defensive investment against exogenous shocks or a further deteriorating economy. For 2002, we see slowly deteriorating real estate fundamentals, but believe that CMBS spreads will be relatively stable compared with other sectors. Investmentgrade CMBS should pass the credit test of rising delinquencies with flying colors. Our main forecasts are: Benchmark 10-year AAA spreads will mostly trade in a range of 45 bp to 65 bp to swaps. US issuance volume will decline to $60 billion. Non-US issuance will exceed 2001s record $20 billion. Delinquencies on commercial mortgages in seasoned CMBS will approach 3% and will lag behind the recovery expected in the second half. Spreads on non-investment grade CMBS will widen further as delinquencies increase. Congress will resolve problems with terrorist insurance, but many investors will still shy away from concentrated property risk. The rest of this chapter reports issuance, market capitalization, spread and total return data for 2001 and includes our projections for 2002. We also review recent trends in the major US commercial real estate markets.
10
ISSUANCE: A GLOBAL RECORD
Global CMBS issuance set a record in 2001. The $82 billion issued by early December exceeded the previous record of $78 billion in 1998. By the end of the year, Commercial Mortgage Alert expects global issuance to total $91 billion. We expect total global issuance to reach $85 billion in 2002. US issuance should total about $71 billion by year-end, the second busiest year ever. We are forecasting $60 billion in domestic issuance in 2002, assuming the 10-year UST remains around 5%.
exhibit 1
$bn
100
US CMBS ISSUANCE
80 60
40.4 28.8
77.7 71.0 58.5 48.9

1
60.0
40 20 0
0.2 0.1 2.6 2.9 4.1 3.4 7.6 17.2 17.4 17.8
14.0
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 ( Projected)
1 2
$65.3bn through 12/6/01. $8.0bn through 1Q2002.
Source: Morgan Stanley, Commercial Mortgage Alert
Non-U.S issuance will set a record of $20 billion this year, up from $12 billion in 2000. Non-US issuance is at the level of US CMBS issuance in 1995. We forecast that non-US issuance in 2002 will be close to $25 billion.
exhibit 2
$bn 25
20 15
12.0 9.4 20.0
1
25.0
NON-US CMBS ISSUANCE
10 5
2.8 0.4 1.5 0.0 0.2 0.8 1.4
3.6 1.1 0.9 0.6
0.6 0.0 0.3
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 ( Projected)
$16.4 through 12/6/01.
11
A Credit Test
Fixed-rate CMBS conduits accounted for 53% of US issuance in 2001, up from 44% in 2000, but below the peak of 68% in 1998. In the US, floating-rate deals made up 13% of the total in 2001, down slightly from the previous year. Single borrower transactions made up 20% of deals in the US, and over half of the deals outside the US. Large loan deals have been about a fifth of issuance in each of the last three years. Because of investor concerns over trophy properties, we may see a decline in this percentage in 2002.
exhibit 3 2001 CMBS ISSUANCE BY DEAL
TYPE (IN %)
Global Conduit Single Borrower Seasoned Collateral Short-Term/Floating Rate Other 44.9 26.7 6.2 11.9 10.3
US 53.3 20.2 3.1 13.0 10.4
Non-US 11.4 52.7 18.7 7.6 9.6
Insurance company loan commitments totaled $19.6 billion through the end of September 2001. Commitments have declined for two consecutive years, but remain at healthy levels compared to ten years earlier.
exhibit 4
$bn 30
LIFE INSURANCE COMPANY COMMERCIAL MORTGAGE COMMITMENTS, 1996-2001
20
10
0
1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 1
1
Through 9/30/01.
Source: ACLI
After declining for two straight years, average deal size increased slightly this year to over $500 million. There were 24 deals over $1 billion, up from ten in 2000. If large loans are combined into conduit pools for diversity, we could see a further increase in deal size next year.
12
exhibit 5
$mm 1000
825
AVERAGE DEAL SIZE
800 600
507 419 484
556
400
259 268 141 167 191
200 0
120 63
101
134
161
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 20011
1
Through 12/06/01.
exhibit 6 Date 11/14/01 10/11/01 5/25/01 5/10/01 7/20/01 5/14/01 3/22/01 8/03/01 12/5/01 5/22/01
TEN LARGEST CMBS DEALS IN 2001

Issuer GGP Mall Properties Trust 2001 C1 COMM 2001-J2 TIAA CMBS 1 Trust 2001 C1 TrizecHahn Office Properties Trust 2001-TZH LB-UBS Commercial Mortgage Trust 2001-C3 LB-UBS Commercial Mortgage Trust 2001-C2 First Union National Bank Bank of America, Commercial Mortgage Trust, 2001-C1 Lehman Brothers Floating Rate Commercial Mortgage Trust 2001 LLFC4 LB-UBS Commercial Mortgage Trust 2001-C7 GS Mortgage Securities Corp II 2001- ROCK Amount ($ mm) 2,550 1,506 1,465 1,440 1,382 1,319 1,308 1,267 1,214 1,210
Our current estimate of the market capitalization of CMBS is close to $350 billion, up from $280 billion last year. We estimate that the total market capitalization of CMBS will exceed $400 billion by the end of 2002.
13
A Credit Test
exhibit 7
$bn
400
ESTIMATED CMBS MARKET CAPITALIZATION, 1987-2002P
350 300 250 200 150 100 50 0

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002P
1
Through 12/6/01
AAA SPREADS WIDEN TO TREASURIES
We think that benchmark 10-year AAA CMBS spreads will mostly be in a 45-65 bp range over swaps in 2002. After the shocks of 1998 and 2001, we believe that the trading range will be both higher and wider than in the 1999 to mid2001 period. Ten-year AAA CMBS spreads ended 2001 about 12 bp wider to swaps than at the beginning of the year. During the year, 10-year AAA spreads were at their narrowest (43 bp) in January and the widest (62 bp) in October and November. The 19 bp range was twice that of 2000, but less than half the range of 1998 and 1999. To the UST, 10-year spreads tightened by 8 bp during the year.
bp 120
100 80 60 40 20
Pre-Shock Trading Range
33 23 60 47 37 50
exhibit 8
10-YEAR AAA CMBS SPREADS VS. LIBOR
Post-1998 Shock Trading Range
0 12/16/97
8/31/98
3/1/99 8/24/99
6/27/00
3/28/01
12/12/01
14
After the September 11 attacks, five-year AAA spreads moved above a 28-48 bp range to swaps for the first time in three years.
exhibit 9
bp 120
100 80 60 40 20 0 01/09/98
28 146 Weeks 54 48
5-YEAR AAA CMBS SPREADS TO LIBOR
01/02/99
12/26/99
12/18/00
12/12/01
Despite increases in delinquencies and a weakening economy, investmentgrade CMBS spreads tightened to UST by 10 bp to 20 bp during the year. The CDO bid drove some of this spread tightening. With rising delinquencies next year, we think investment-grade spreads will be flat to slightly wider versus UST in 2002.
exhibit 10
Spread (bp) 350

300 250 200 150 100 50 0 01/03/97
AAA AA A BBB
INVESTMENT GRADE SPREADS TO UST
221 175 150 132
03/30/98
06/24/99
09/17/00
12/12/01
15
A Credit Test
exhibit 11
bp
250 200 150
132 AAA CMBS 10-Year Swap Spreads Difference
10-YEAR AAA CMBS SWAPPED TO LIBOR
100 50 0 1/3/1997
76 56
3/30/1998
6/24/1999
9/17/2000
12/12/2001
Source: Morgan Stanley, Bloomberg
Non-investment grade CMBS spreads widened by 30 bp to 175 bp in limited trading. We expect further widening in 2002 as delinquency rates rise above subordination levels for many single-Bs.
exhibit 12
bp
1,000 800 600 400 200 0 9/12/1997
BB 10-Year B 10-Year
1000
NON-INVESTMENT GRADE SPREADS TO UST
555
10/5/1998
10/28/1999
11/19/2000
12/12/2001
16
exhibit 13
%
16 14 12 10 8 6 4 2 0 12/31/93
Single-B CMBS Yield 10-yr UST Yield
Average Single-B CMBS Yield (Excluding 4/9710/98)
SINGLE-B CMBS YIELDS
Average 10-Year UST Yield
12/26/95
12/21/97
12/17/99
12/12/01
Through December 12, YTD total returns on investment-grade CMBS ranged from +8.5% to +9.6%. The 10-year UST was almost unchanged from the start of the year, so most of the return was from coupon payments and modest spread tightening. Non-investment grade CMBS underperformed, with spreads widening as much as 175 bp for single-Bs.
exhibit 14
CMBS TOTAL RETURNS AS OF 12/12/2001

Nominal (%) (Yrs) Spread (bn) (%) (%) Since Mod (%) (Yrs) (to UST (%) Market MTD YTD Inception Dur Coupon AvgLife in bp) Yield Value TotalRet TotalRet 1/1/00
CMBS (by Rating) CMBS AAA CMBS AA CMBS A CMBS BBB (by Maturity) CMBS 1-3.5 yr AL CMBS 3.5-6 yr AL CMBS 6-8.5 yr AL CMBS 8.5+ yr AL
5.0 4.8 6.0 6.2 6.3 2.7 4.1 5.4 7.1
6.7 6.7 6.8 7.0 7.1 6.2 6.7 6.8 7.0
6.5 6.0 8.0 8.5 8.9 3.1 5.0 6.9 10.3
76.6 63.7 86.1 111.9 164.7 66.5 67.9 71.6 118.8
6.1 5.9 6.4 6.7 7.3 4.9 5.7 6.1 6.9
34.4 28.0 2.2 2.3 2.0 3.7 5.9 21.8 3.0
-0.8 -0.7 -1.0 -1.1 -1.1 -0.2 -0.5 -0.8 -1.3
9.4 9.5 9.0 8.5 9.6 9.3 8.9 9.5 8.7
24.9 24.3 26.7 27.3 29.0 20.0 21.0 25.3 26.5
R E L AT I V E V A L U E : T I G H T E R T H A N B A N K A N D F I N A N C E
AAA CMBS continued to trade 20 bp to 40 bp tighter than single-A bank and finance paper. CMBS spreads were flat to slightly wider than bank and finance spreads from 1996 through early 2000.
17
A Credit Test
exhibit 15
bp 300
10-YEAR AAA CMBS VS. SINGLE A BANK & FINANCE
AAA CMBS Single A Bank and Finance Difference
200
157 132
100
0
-25
-100 7/5/1996
11/13/1997
3/25/1999
8/2/2000
12/12/2001
BBB CMBS were about 20 bp wider than BBB unsecured corporates in December after reaching near parity in late 2000 and after September 11, 2001.
Spread to UST (bp)
350 300 250
221 217
BBB CMBS REITs
exhibit 16
BBB CMBS AND REIT SPREADS
200 150 100 4/10/98
3/11/99
2/10/00
1/10/01
12/12/01
At the end of the year, BBB unsecured REITs were about 5 bp wider than BBB CMBS. We think that these spreads will remain close to parity in 2002.
R E G U L AT O RY: S FA S B 1 4 0 A N D B I S
A near crisis in the CMBS market was averted in July when issuers agreed to modify language in new issue CMBS documents to clarify the actions of servicers with regard to specially serviced loans. Before the clarification there was concern that the SFASB 140 would interrupt CMBS deal flow. The BIS capital reforms will not be fully implemented until 2005. However, the FDIC recently adopted new capital reserve requirements, effective after January 1, 2002. The new guidelines require higher reserves for below-investment grade securities.
18
D E L I N Q U E N C Y R AT E S R E V E R S E T R E N D
Commercial mortgage delinquency rates reported by the American Council of Life Insurance (ACLI) dropped to an all-time low of 19 bp in the third quarter. The decline was a surprise given the weakening of most real estate markets. We expect this number to rise throughout 2002, but remain less than 2.5% throughout the year.
Length in Quarters
exhibit 17
COMMERCIAL MORTGAGE LOAN DELINQUENCY RATES
% Delinquent 8
26
22
42
36 7.53%
4
0.19%
0 1Q65
4Q69
2Q76
4Q81
2Q92
3Q01
Source: ACLI
Delinquencies on loans backing CMBS were at 1.16% in October, based on November remittance reports. This is the highest rate since we began tracking the data in 1999 and represents a 97 bp increase over the past 12 months. For deals seasoned at least one year, the delinquency rate was 1.81% of current balances, an increase of 37 bp over the past 12 months. We expect delinquencies to rise by an additional 1%-2% over the next year.
exhibit 18
2.4% 2.0% 1.6% 1.2% 0.8% 0.4% 0.0% 8/1/1999 12/16/1999 5/1/2000
Total Seasoned
CMBS TOTAL VS. SEASONED CMBS DELINQUENCIES
9/15/2000 1/30/2001 6/16/2001 11/1/2001
Source: Morgan Stanley, Intex
In CMBS, senior housing properties have the highest delinquency rate (6.58%), followed by hotels (3.49%), retail (1.21%) and industrial (0.93%). Among large states, Florida has the highest delinquency rate (2.52%). The delinquency rate on loans in California is 0.28%, the lowest among large states.
19
A Credit Test
exhibit 19
3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 8/1/99
CA FL
CMBS DELINQUENCIES BY CURRENT BALANCE (FL AND CA)
12/16/99
5/1/00
9/15/00
1/30/01
6/16/01
11/1/01
R E A L E S TAT E : I N F U L L D E C L I N E
Real estate markets, which started to soften in 2000, fell into decline in almost every area. Rents fell in some sectors, such as San Francisco office, by a third or more. Hotel occupancies fell sharply after September 11, but recovered somewhat at the end of the year. Within all property types vacancy rates rose during 2001, rising off very low levels during 2000. We anticipate that vacancy rates will rise moderately during 2002. On the bright side, new construction has also slowed over the past several months in response to the overall slowdown in the economy. Although the US economy is experiencing an official recession, most industry analysts anticipate this cycle will be more moderate than the early 1990s. Lack of speculative development, lower leverage within transactions, and significantly greater public ownership of real estate, are all factors indicating that real estate is better positioned going into this downturn.
OFFICE PROPERTIES: DEVELOPMENT PIPELINE DECLINES
Nationwide office vacancy rates increased 170 bp during the third quarter to 12% (CB Richard Ellis). Over the past 12 months the vacancy rate has increased 430 bp. Downtown vacancy rates were 10.4% during the quarter, versus 13.0% for suburban space. Another data source, REIS, reported a 130 bp increase in office vacancies in the third quarter, to 11.4%. Office vacancy rates are the highest since 1996. The metropolitan statistical areas (MSAs) with the highest vacancy rates were Columbus (20.2%), Dallas/Fort Worth (19.2%), Indianapolis (17.6%), Orlando (17.6%) and Jacksonville (17.5%). The MSAs with the lowest vacancy rates were Washington DC (3.5%), Wilmington (5.8%), Sacramento (7.8%), Oakland (8.3%) and Manhattan (9.0%).
20
exhibit 20
200,000
24% 20%
NATIONAL OFFICE COMPLETIONS & VACANCY RATES (1981-2002P)
150,000 16% 100,000 12% 8% 50,000 4% 0

1981 1984 1987 1990 1993 1996 1999 2002P
0%
Source: REIS
According to REIS, third quarter office completions slowed to 18 million sf from 22 million sf in the second quarter. REIS also projects a slowdown in the fourth quarter. New estimates for office development in 2001 are down 20% from previous projections. Office properties in CMBS have the lowest delinquency rates of any property type. Based on November data, office properties posted a 0.37% delinquency rate. Over the past 12 months delinquencies have declined 15 bp.
exhibit 21
1.5% 1.2%
Total
CMBS TOTAL VS. OFFICE DELINQUENCIES
0.9% 0.6% 0.3%

Office
0.0% 8/1/1999 12/16/1999 5/1/2000
9/15/2000 1/30/2001 6/16/2001 11/1/2001
I N D U S T R I A L P R O P E R T I E S : V A C A N C I E S R I S E M O D E R AT E LY
During the quarter, national vacancy rates on industrial properties increased 70 bp to 9.5% (CB Richard Ellis). The MSAs with the highest vacancy rates were Jacksonville (26%), Baltimore (17%), Washington DC (15%), Tucson (14%), and Columbus (13%).
21
A Credit Test
Markets with the lowest vacancy rates were Albuquerque (2%), Long Island (4%), Palm Beach (5%), San Diego (5%), and Kansas City (6%). Industrial development during the quarter totaled about 22 million sf, less than half the development in the second quarter.
Square Feet (000s)
250,000 200,000
Completions Estimated Annual Total Vacancy Rates
exhibit 22
Vacancy Rate
16%
NATIONAL INDUSTRIAL COMPLETIONS & VACANCY RATES 1980-2002P
12%
150,000 8% 100,000 4%
50,000 0
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002P
0%
Source: REIS, CB Commercial
Based on November data, delinquencies on industrial loans in CMBS were 0.58% of current balances. This was an increase of 35 bp over the past year.
1.5% 1.2% 0.9% 0.6% 0.3% 0.0% 8/1/1999 12/16/1999 5/1/2000
Industrial
exhibit 23
CMBS TOTAL VS. INDUSTRIALWAREHOUSE DELINQUENCIES
Total
9/15/2000 1/30/2001 6/16/2001 11/1/2001
R E TA I L : D E V E L O P M E N T D E C L I N E S
Retail vacancy rates increased 10 bp during the third quarter to 6.3% (REIS). Development of retail space during the third quarter totaled about 5 million sf. REIS revised its 2001 projection down to 27 million sf from its previous estimate of 39 million sf. During 2002 REIS anticipates development of 24 million sf.
22
Analysts report that mall traffic has been flat to slightly up after Thanksgiving. The International Council of Shopping Centers reported that during the first week of the holiday season, specialty (non-anchor) store sales declined 2.6% from last year.
exhibit 24
Square Feet (000s)
45,000 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002P
Completions Estimated Annual Total Vacancy Rates
Vacancy Rate
12%
NATIONAL RETAIL COMPLETIONS & VACANCY RATES 1992-2002P
8%
4%
0%
Source: REIS
exhibit 25
1.5% 1.2% 0.9% 0.6% 0.3% 0.0% 8/1/1999 12/16/1999 5/1/2000
CMBS TOTAL VS. RETAIL DELINQUENCIES
Retail
Total
9/15/2000 1/30/2001 6/16/2001 11/1/2001
Delinquencies on retail loans within CMBS transactions were 1.21% of current balances as of November. Over the past 12 months, this delinquency rate increased 47 bp.
M U LT I FA M I LY: S TA B L E A S A L W AY S
Rental housing vacancy rates increased 10 bp during third quarter to 8.4% (Census Bureau). Based on permit data, about 270,000 multifamily units should enter the market in 2001.
23
A Credit Test
exhibit 26
Permits (000s)
550 500 450 400 350 300 250 200 150 100 50 1977
1
Vacancy Rate
12%
Completions Vacancy Rates
NATIONAL MULTIFAMILY COMPLETIONS & VACANCY RATES 1979-2002P
9%
6%
3%
1980
1983
1986
1989
1992
1995
0% 1998 2001P 1
Vacancy is as of 3Q01, completions is average year to date annualized as of September 2001.
Source: US Census Bureau
Delinquencies on multifamily loans within CMBS were 0.35% of current balances. Over the past 12 months delinquencies declined 15 bp.
exhibit 27
1.5% 1.2% 0.9% 0.6% 0.3% 0.0% 8/1/1999 12/16/1999 5/1/2000
Multifamily
CMBS TOTAL VS. MULTIFAMILY DELINQUENCIES
Total
9/15/2000 1/30/2001 6/16/2001 11/1/2001
H O T E L S : D E M A N D D E T E R I O R AT E S
Nationwide revenue per available room (RevPAR) has declined about 19% between mid-September and late November (Smith Travel).
24
exhibit 28
HOTEL OCCUPANCY RATES BY PRICE CATEGORY AS OF OCTOBER 2001 (IN %)

2001 YTD 68.1 64.1 60.6 58.2 60.1 2000 71.8 65.1 61.5 58.5 59.5 1999 71.9 64.9 61.1 58.0 58.7 1998 72.6 65.9 62.0 58.4 58.7 1997 78.4 70.8 67.9 61.2 59.9
Luxury Upscale Mid-Price Economy Budget
Source: Smith Travel Research
The recovery from the current recession will have a significant impact on the performance of hotels next year, as hotel demand is highly correlated to GDP growth. Positive news is that we anticipate hotel development to decline significantly in 2002.
exhibit 29
HOTEL ROOM RATES BY PRICE CATEGORY AS OF OCTOBER 2001 (IN $)

2001 YTD 146.91 94.24 70.45 54.71 41.46 2000 144.99 90.88 68.23 52.44 42.20 1999 138.88 87.37 64.89 49.23 89.60 1998 133.58 85.33 62.15 47.14 37.43 1997 136.01 92.07 66.97 49.93 40.60
Luxury Upscale Mid-Price Economy Budget
Forecasters anticipate the industry will end 2001 with a 7% decline in RevPAR. We estimate that 2002 will be flat to 2001. Tourist destinations such as Boston, New York, San Francisco, Miami, Oahu Island, and Orlando have been the hardest hit since mid-September. Luxury hotels at the highest price levels have also suffered more than lower-priced properties.
exhibit 30
200 160 5% 120 80 40 1% 0 0%
1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002P
7% 6%
HOTEL ADDITIONS TO SUPPLY 1969-2002P
4% 3% 2%
Source: PriceWaterhouseCoopers
25
A Credit Test
exhibit 31
HOTEL MARKET DATA AS OF YTD OCTOBER 2001

Occupancy(%) 2001 2000 %Chg. 72.0 67.5 78.1 72.6 66.7 70.3 68.6 63.5 74.1 70.7 71.3 61.9 71.1 83.6 61.2 76.7 74.4 67.6 63.3 75.6 83.2 71.2 65.3 66.2 76.1 -4.9 -6.4 -13.8 -11.7 -11.5 -7.0 -10.3 4.9 -6.5 -5.0 -8.0 -6.3 -6.3 -11.7 -0.8 -7.2 -10.5 -4.3 -4.7 -4.4 -18.5 -7.7 -3.5 -2.0 -8.3 2001 91.85 80.37 138.90 113.68 80.62 80.05 81.82 75.89 97.01 109.84 86.69 75.04 120.00 183.83 74.81 117.33 89.41 99.02 101.21 112.78 146.32 98.88 73.05 87.56 118.65 Avg Room Rate($) 2000 87.90 80.87 145.32 117.38 82.61 80.26 81.64 75.78 96.59 106.63 84.73 75.43 120.53 199.08 73.96 115.78 89.80 102.34 100.62 110.28 149.09 97.33 71.98 84.90 116.22 %Chg. 4.5 -0.6 -4.4 -3.2 -2.4 -0.3 0.2 0.1 0.4 3.0 2.3 -0.5 -0.3 -7.7 1.1 1.3 -0.4 -3.2 0.6 2.3 -1.9 1.6 1.5 3.1 2.1
AnaheimSanta Ana, CA Atlanta, GA Boston, MA Chicago, IL Dallas, TX Denver, CO Detroit, MI Houston, TX Los Angeles Long Beach, CA Miami-Hialeah, FL Minneapolis St. Paul, MN Nashville, TN New Orleans, LA New York, NY Norfolk Virginia Beach, VA Oahu Island, HI Orlando, FL Philadelphia, PA-NJ Phoenix, AZ San Diego, CA San Francisco, CA Seattle, WA St. Louis, MO-IL Tampa St. Petersburg, FL Washington DC, MD-VA
68.5 63.2 67.3 64.1 59.0 65.4 61.5 66.6 69.3 67.2 65.6 58.0 66.6 73.8 60.7 71.2 66.6 64.7 60.3 72.3 67.8 65.7 63.0 64.9 69.8
Based on November 2001 data, hotel delinquencies were 3.49% of current balances. Over the past 12 months delinquencies have increased 191 bp.
exhibit 32
4.0%
CMBS TOTAL VS. HOTEL DELINQUENCIES
3.0%
Hotel
2.0%
1.0%
0.0% 8/1/1999 12/16/1999 5/1/2000
Total
9/15/2000 1/30/2001 6/16/2001 11/1/2001
26
R AT I N G A C T I O N S
CMBS continues to outperform corporate bonds in terms of rating actions. Upgrades on transactions increased in the latter part of the year. Based on rating actions through early December 2001 the upgrade/downgrade ratio approached 15 to 1, up from the mid-year ratio of 10 to 1. We anticipate that CMBS transactions will experience more downgrades next year due to deteriorating credit overall. However, we do not anticipate downgrades to outnumber upgrades in 2002. During the year AA bonds experienced the greatest number of upgrades, with 102 upgrades of various tranches from AA to AAA. BBB bonds experienced the greatest number of rating actions this year. The rating agencies issued 180 rating actions resulting in 87 full rating category upgrades. Downgrades were most common on B bonds with 10 full rating category downgrades.
exhibit 33
700 600 500 400 300 200 100 0 1999
1
1999, 2000 AND 2001 CMBS RATING ACTIONS
Upgrades Downgrades
580
395 263
27 2000
61
40 2001 YTD
1
Through December 7, 2001
Source: Morgan Stanley, Fitch, Moodys and Standard & Poors
exhibit 34
2001 RATING ACTIONS THROUGH DECEMBER 7, 2001 (SHADED AREAS ARE DOWNGRADES OF AT LEAST A FULL CATEGORY)1
TO
AAA AA A BBB BB B CCC and Below Total 0 158 140 180 78 59 5 620
AAA AA A FROM BBB BB B CCC and Below Total

1
0 102 17 10 2 1 0
132
0 56 64 24 5 0 0
149
0 0 59 53 9 1 0
122
0 0 0 88 29 2 0
119
0 0 0 5 26 24 0
55
0 0 0 0 6 21 1
28
0 0 0 0 1 10 4
15
Does not include rating affirmations.
Source: Morgan Stanley, Fitch, Moodys and Standard & Poors
27
Major Property Types in CMBS

While CMBS investors dont 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.
C L A S S I F I C AT I O N S
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.

W H Y D O P R O P E R T Y C L A S S I F I C AT I O N S M AT T E R ?
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.
28
class A
class B
class C
29

Conduit CMBS transactions provide property type and geographic diversity. Typically retail properties make up about 30% of a transaction, multifamily compose about 20%, office properties make up about 20%, hotels compose about 10%, and the remaining 20% is composed of manufactured housing, industrial, self-storage, and senior housing.
Retail Loans
Typical Loan Terms
1.4 DSCR 65% LTV (based on a 910% capitalization rate) $75$125 Loan Per Square Foot Value
R E L AT E D T E R M S
Credit Tenant Lease
All payments guaranteed by credit of tenant (i.e., WalMart or Kmart).

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-Tenancy Provisions
Permits the tenant to cancel its lease if another major store closes.
Recapture Provisions
Permits the owner to cancel a lease and to regain control of space after a tenant closes its store.
T Y P E S O F R E TA I L P R O P E R T I E S
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-Line Store
Smaller store within a center (i.e., Foot Locker or Hallmark Cards store).
Community Center
Over 100,000275,000 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.
30
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 INVESTOR WISDOM
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 support 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.
R AT I N G A G E N C Y V I E W
Generally received favorable treatment from rating agencies as a result of strong historical performance. Preference for longer leases; nationally or regionally rated credit tenants.
31

Multifamily Loans
Typical Loan Terms
1.25x DSCR 75% LTV (based on a 89% capitalization rate) $20,000$60,000 Loan Per Unit
Co-op Loans/Blanket Loans
Very low loan to value loans. Loans senior to coop share loans. Typically 10-40% of value.
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.
T Y P E S O F M U LT I FA M I LY P R O P E R T I E S
Garden Apartments
Multiple buildings; usually no more than 2-3 stories.

High Rise Apartments
Over three stories; usually located in downtown areas.

garden apartments
high rise apartments
32
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.
A must have property type for diversity. Lower default tendency due to constant mark-to-market of rental rates (1-year-lease terms). 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.
33

Office Loans
Typical Loan Terms
1.4x DSCR 70% LTV (910% cap rate) $50100 Loan Per Square Foot for Suburban Properties $70150 Loan Per Square Foot for Downtown Central Business District (CBD)
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 pass this expense through to the tenant in terms 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 $24 per square foot at lease signings. Landlords bear this expense.
Rollover
Term used to describe expiration of tenant lease. Lease terms generally 510 years; credit tenants often longer. It is preferred to not have rollover concentrations, which would expose an owner to uncertain rental market or potentially reduce NOI below debt service.
TYPES OF OFFICE BUILDINGS
downtown suburban
34
Above market rents are a concern if the market rent is insufficient to support 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.
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 13%.
35

Hotel Loans
Typical Loan Terms
1.5x DSCR 65% LTV (1012% cap rate) $20,00060,000 Loan per room $80,000 + Full Service or Luxury
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)
Total rooms revenue divided by the available rooms for a given period. It is the revenue per available room.
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 47% of gross revenue.
TYPES OF HOTEL PROPERTIES
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.
36
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.
Very conservative because use of average historical income often reduces income to levels significantly below current highs. Maximum occupancy they will underwrite is 6575%, 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
37

Manufactured Housing Community Loans
Typical Loan Terms
1.4x DSCR 70% LTV (910% cap rate) $10,000$20,000 per pad
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 supports each manufactured home.

Double Wide/Single Wide
Describes the size of manufactured home that a given slab will support. The double wide segment has experienced the fastest growth due to the growing acceptance of manufactured homes as a single family housing alternative.
T Y P E S O F M A N U FA C T U R E D H O U S I N G C O M M U N I T I E S
3-Star Park
Older park lacking the amenities of a 5-star park; higher proportion of single wide pads. Offer limited amenities and services.
4-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-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.
38
Lack of familiarity; confusion with asset-backed manufactured housing loans to individuals. Insurance companies typically didnt 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 ($3,0005,000 a year) exceeds pad rental ($300500 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 agencies favorable on credit. Low volatility of cash flows. Physical turnover rate 35% in manufactured housing versus; 5060% in multifamily. Few capital reserves required. Often better than multifamily.
5-star park
39

Industrial Loans
Typical Loan Terms
1.4x DSCR 7075% LTV (810% Cap Rate) $1025 PSF $3050 PSF (if high office component)
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 tenant 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-Up Construction
This is the preferred construction type for industrial buildings. It includes pre-cast concrete panels that are tilted-up on a steel frame. Tilt-up is preferable to corrugated metal exteriors for maintenance reasons.
distribution space
40
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.
Like industrial for diversity. Review environmental issues from manufacturing uses. Very low cost to re-lease if tenant leaves, but short lease terms create rollover risk.
flex space/office warehouse
41

Self Storage Loans
Typical Loan Terms
1.3x, 1.4x DSCR 6570% LTV $2050 PSF
Self-Storage Facility
Commercial property that lease storage space to individuals or businesses on a month-to-month basis. 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 users 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
42
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.
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.
43

HealthCare/Senior Housing Loans
Typical Loan Terms
1.22.0X DSCR (depending on complexity of service) 50--80% Loan to Value $30,000-80,000 loan per unit
T Y P E S O F H E A LT H C A R E / S E N I O R H O U S I N G P R O P E R T I E S
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 seniors 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
This product is 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. They provide fulltime 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 registered nurses on call. Licenses by 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.
44
CONVENTIONAL INVESTOR VIEW
Recent changes in Medicaid reimbursements have negatively affected skilled nursing and continuing care facilities. Excess supply of assisted living facilities have 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.
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.
45
Retail Collateral in CMBS

B U Y I N G R E TA I L
Over the past year, several US retailers have announced bankruptcies or store closings. These announcements are usually followed by an e-mail from CMBS data providers detailing the number of CMBS deals affected by the closing. In order to determine exposure, most analysts look at databases of the top three tenants of each loan in all CMBS transactions. How accurate is this evaluation of retail risk? While the 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 deals only list the top tenant.) (2) It fails to look at the cumulative risk of all unhealthy tenants. (3) It does not account for the state of the local real estate market. In this study, we address the first two of these shortfalls. Using a database from the National Research Bureau, we are able to capture 75% of the loan balances of retail tenants in a sample of CMBS deals. This compares with only 29% covered by a database of the top three tenants. In addition, we are able to look at the cumulative risk of a list of risky retailers. Although the new database does greatly extend the coverage of retail tenant risk, it does not address the effect of the health of local retail markets and general real estate conditions on loan risk. We leave this issue to a future paper. Because of the large amount of data analysis involved, we limit this study to 31 Morgan Stanley underwritten deals, and we examine the deals for exposure to 86 risky retail credits that were identified by the REIT equity research team. We hope to expand the study to all CMBS deals at a later date. Our main findings are: Morgan Stanley CMBS deals have limited exposure to the risky tenants. However, all of the deals, with the exception of MSC 1997-LB1, have some exposure to risky tenants. About 14% of the loans in the Morgan Stanley deals have exposure to at least one of the 86 risky tenants. On a risk-weighted basis, only 0.8% of Morgan Stanley CMBS collateral has exposure to the risky tenants. Risk-weighting accounts for the portion of a loan that can be attributed to the retailer, based on its gross leasable area. MSDWC 2001-PGM had the largest weighted risk exposure as a percentage of the entire deal at 4.7%.
46
exhibit 1
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 XLF MSDWC 2000-PRIN MSDWC 2001-PGMA MSDWC 2001-PPM MSDWC 2001-TOP1 MSDWC 2001-TOP3 MSDWC 2001-FRMA MSDWC 2001-SGMA Total Delinquencies (%) 2.7 0.3 0.0 1.7 0.0 0.0 1.7 0.0 7.2 0.9 1.3 0.5 0.0 0.0 0.0 0.0 1.8 0.0 0.4 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0
47

R E TA I L P R O P E R T Y O V E R V I E W
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 midmarket malls. Strong regional malls and grocery-anchored community centers typically have lower defaults and cash flow volatility than other retail property types within CMBS. As we anticipate credit to deteriorate in the coming year, the importance of monitoring retailers increases. In a weaker credit environment, retail properties may be adversely affected by lower consumer confidence and spending. Since retail properties usually make up a substantial portion of collateral in CMBS deals, we thought it would be useful to analyze the exposure of risky retailers to CMBS. 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. Altmans 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 Altmans findings, Morgan Stanleys 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 86 risky tenants are listed in Exhibit 2. 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.
48
exhibit 2
RISKY TENANTS
Name Z-Score Mayors Jewelers 1.91 Monfried Optical -5.93 Montgomery Ward Optical 2.25 Moto Photo 2.44 National Record Mart 1.63 Noodle Kidoodle 1.65 Overland Trading Co. 2.71 Pamida 2.39 Party City 3.76 Payless Cashaways 2.51 Pearle Vision 2.25 Pep Boys 2.08 Perfumania 1.32 Play Co. Toys 0.12 Reeds Jewelers 2.25 Repp Big & Tall 1.68 Restoration Hardware 2.12 Samuels 0.49 Schubach 0.49 Schucks Auto Supply 2.00 Sears Optical 2.25 Shopko Dept. Store 2.39 Silvermans 0.49 Silversmiths and Mission Jewelers 0.49 Singer/Specs Discount Vision -5.93 Site for Sore Eyes -5.93 Southern Optical -5.93 Sports Authority 2.58 Sterling Optical -5.93 Stone & Thomas 2.37 Superior Optical -5.93 Swisher Maids 0.61 Things Remembered 2.25 Todays Man 2.02 Toy Co. 0.12 Toys International 0.12 Track n Trail 2.71 U.S. Vision 1.73 Wards Optical 2.25 Wayne Jewelers 0.49 Weiners 2.59 West Coast Video -1.60 Wickes -1.53
Name Z-Score A. Hirsch & Son 0.49 Ames Dept. Store 1.88 B & T Factory 1.68 Benson Optical -5.93 Bikers Dream -0.90 Biozhem -22.71 BJs Optical 2.25 Blockbuster Video 1.47 Bookland 2.33 Books - A - Million 2.33 Books & Co. 2.33 Calloways Nursery 2.08 Casual Male Big & Tall 1.68 CD Exchange 1.25 CD Warehouse 1.25 Checker Auto Parts 2.00 Cole Vision Center 2.25 Disc-Go-Round 1.25 Discount Auto Parts 2.32 DIY Home Warehouse 1.93 Duling Optical -5.93 Elder-Beerman 2.37 Elegant Illusions 5.53 Elegant Pretenders 1.14 Factory Card Outlet 1.95 Famous Brands Housewares Outlet 3.09 Furrows 2.51 Hatfield Jewelers 0.49 Hearex Hearing Centers -1.82 Hills Department Store 1.88 Hollywood Video -1.75 HomeBase 2.31 Imposters 1.14 IPCO Optical -5.93 Jennifer Convertibles 3.26 Jennifer Leather 3.26 Joe Muggs Newstand 2.33 Kidoodle Theater 1.65 Kindy Optical -5.93 Kragen Auto Parts 2.00 Lechters 3.09 Loehmanns 1.81 Lumberjack Building Materials 2.51
49

METHODOLOGY
In order to assess CMBS exposure to the 86 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 38,000 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. 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, we can track 75% of retail balances in Morgan Stanley deals. If this study were performed using data solely from the S&P Conquest database, only 29% of the retail loan balances would have complete tenant information. In total, we had some level of tenant information for 92% of the retail loans within the transactions that we examined. Sixty percent of the loans were covered through information from the National Research Bureau Shopping Center Directory, and had complete tenant information for the shopping centers. Thirty-two 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 8% 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.
50
S H O R T FA L L S / O T H E R FA C T O R S T O C O N S I D E R
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 account for this in our study. For the purposes of this study, we also examined and included all adjacent or adjoining retailers within a specific shopping center, regardless of whether the tenants 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 cashflows 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.
51

KMART RISK
Kmart was not highlighted in our recent study on risky retailers because the companys Z-score was higher than our 2.4 cut off. Any retailer with a Z-score below 2.4 was deemed risky by Morgan Stanleys equity research REIT team. In Altmans study, scores below 1.8 are categorized as unhealthy. It is worth noting that the Z-score attributable to Kmart declined during 2001. As of first quarter 2001, the companys Z-score was 3.76; based on second quarter data it declined to 3.42, and by year end, it declined to 3.02, indicating fundamentals have been deteriorating.
M O R G A N S TA N L E Y C M B S E X P O S U R E T O K M A R T
In our January 4, 2002 issue of CMBS Perspectives, we published an article discussing CMBS exposure to Kmart. In CMBS transactions where Kmart was listed as one of the top three tenants, the weighted average exposure to loans with Kmart was 2.25% of current balances. We did a detailed search using the National Research Bureau (NRB) Shopping Center directory to look for Kmart exposure in Morgan Stanley CMBS transactions. The NRB database lists the tenants of 38,000 shopping centers in the United States, allowing us to search for both large and small retailers.
T O TA L R I S K E X P O S U R E B A S E D O N L O A N B A L A N C E
Thirty-three loans within 17 Morgan Stanley CMBS transactions have exposure to Kmart. Morgan Stanley CMBS exposure to loans with Kmart as a tenant is 3.18% of current balances, 93 bp higher than the CMBS universe exposure. A large $157 million loan in MSC 1998 XL1 is responsible, in part, to this high exposure. The entire balance of the loan is included in the total exposure, despite there being only two Kmarts in the entire loan backed by 44 shopping centers.
RISK-WEIGHTED EXPOSURE BASED ON GROSS LEASABLE AREA
On a risk-weighted basis, Morgan Stanley CMBS has less than 1% exposure to Kmart. Risk-weighted exposure only accounts for the portion of a loan that can be attributed to the retailer, based on its gross leasable area.
52
exhibit 3
MORGAN STANLEY CMBS EXPOSURE TO KMART

Kmart Exposure (%)
TOTAL
Current Deal Balance ($)
Loans with # of Kmart ($) loans
MSC 1998-XL1 MSC 1998-CF1 MSC 1999-CAM1 MSDWC 2001-PPM MSC 1996-C1 MSDWC 2001-TOP3 MSC 1996-WF1 MSC 1999-RM1 MSC 1997-HF1 MSC 1998-WF1 MSDWC 2001-TOP1 MSC 1998-HF1 MSC 1998-WF2 MSC 1998-HF2 MSDWC 2000-LIFE MSC 1997-C1 MSC 1997-WF1 TOTAL
884,972,546 1,037,585,367 745,511,272 606,109,806 224,956,904 1,024,673,148 502,087,492 810,161,563 511,108,048 1,303,029,623 1,148,473,606 1,181,297,510 1,013,190,631 1,016,789,579 679,533,489 501,404,646 483,978,193 13,674,863,422
205,449,431 51,114,711 25,425,911 17,867,583 5,726,985 22,056,943 10,314,127 15,794,842 9,423,334 22,008,117 14,115,740 12,521,469 9,325,272 6,853,826 3,418,490 1,529,916 1,584,514 434,531,212
2 4 3 2 1 3 1 3 1 4 1 2 2 1 1 1 1 33
23.2 4.9 3.4 2.9 2.5 2.2 2.1 1.9 1.8 1.7 1.2 1.1 0.9 0.7 0.5 0.3 0.3 3.2
Source: Morgan Stanley, S&P Conquest, National Research Bureau Shopping Center Directory
53

SAMPLE RETAIL ANALYSIS
exhibit 4
MSC 1996-C1
MSC 1996-C1
Retail Exposure
Loans Outstanding Original Balance ($MM) Current Balance ($MM) Retail 14 $75.2 $69.9 Entire Deal 99 $340.5 $225.0 % Retail 14.1% 22.1% 31.1% Retail Loans in Deal Number of Loans 14 14 14 Average GLA Center 223,750 285,000
Study Results
National Research Bureau Prospectus Total Risky Tenant Exposure 1 Blockbuster Video 2 National Record Mart
Retail Loans Included Number Current of Loans Balance 9 $50.8 0 $0.0 9 $50.8 Number of Stores 2 1
Store 5,850 5,200
Shopping Center Exposure 1 Constant Friendship Shopping Center 2 Lilac Mall
Total Stores 11 20
Risky Stores 1 2
Store 4,700 12,200
54
Statistics
Exposure by Current Balance ($MM) $0.5 W td. Risk Balance 1.1% Included Retail $50.8 Total Retail $69.9 0.8% 0.2% Entire Deal $225.0 Total Risk Balance $16.1 Included Retail $50.8 31.7% 23.0% Total Retail $69.9 Entire Deal $225.0 7.2% Exposure by GLA (sq ft) 16,900 Store At-Risk Sq. Ft. 3.8% Affected Centers 447,500 Total Centers 1,542,411 1.1% Shop. Ctr. At-Risk Sq. Ft. 447,500 Total Centers 1,542,411 29.0%
Current Balance $69.9 $69.9 $69.9
% Retail Loans Included Number Current of Loans Balance 64.3% 72.7% 0.0% 0.0% 64.3% 72.7%
% 2.6% 1.8%
Z-Score 1.5 1.6
Types of Loans ($)
Retail Loan Coverage ($)
Excluded Other Retail Included
Total GLA Center 162,500 285,000
% 2.9% 4.3%
Largest Risky Store 2.9% 2.5% TOTAL:
Shopping Center Current Balance 10.4 5.7 16.1
Weighted Risk Balance 0.3 0.2 .5
55
Hotel Collateral in CMBS

Rating agencies and many investors have always viewed hotels as one of the riskiest property types. Unlike most other commercial real estate, which have long-term leases, hotels have daily changes in occupancy and rental rates. The slowing economy and recent events have increased CMBS investor concerns about exposure to hotels. 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 has declined since September 11. Supply of new hotel rooms built during the 1990s is 25% lower than the supply built in the 1980s. Current forecasts indicate the hotel industry will post declining year-over-year revenue until mid-2002, recovering in late 2002. The hotel industry is currently very profitable although it is a cyclical business. During 2001, CMBS transactions with heavy hotel concentrations have experienced rating upgrade/downgrade ratios, in line with the CMBS universe performance. The rating agencies have recently commented on the corporate ratings of most public hotel companies and have revised some underwriting standards for new transactions. Conduit CMBS transactions average about 10% exposure to hotels. Leverage levels are more conservative on hotel loans than other property types, providing cushion for the downturn. Even if delinquencies on hotel loans double from current levels, only unrated bonds and B bonds would be affected. Two large loan transactions (100% exposure to hotels) we reviewed can withstand more severe declines in operating performance than are projected and still generate sufficient cash for debt service payments.
exhibit 1
High Onsite Management Intensity
Healthcare
PROPERTY TYPE MATRIX

Hotel
Low Operating Importance
Retail
High Operating Importance
Multifamily
Office
Industrial
Low Onsite Management Intensity
56
G R E AT E R S H O C K T O D E M A N D T H A N E A R LY 1 9 9 0 S
Hotel demand is highly correlated to changes in GDP. When GDP is growing, consumer confidence rises resulting in greater discretionary income. Recent events and a slowing economy prior to September 11 have resulted in a significant decline in hotel demand. Industry projections anticipate 2001 will post declines in demand of about 2.8% this year. During the Gulf crisis of 1991 demand declined just over 1%. Over the past decade 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.
exhibit 2
5%
NATIONWIDE HOTEL SUPPLY AND DEMAND GROWTH
Supply Change Demand Change
3%
1%
-1%
-3%
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001E 2002E
H O T E L S U P P LY S I G N I F I C A N T LY L O W E R I N 1 9 9 0 S T H A N 1 9 8 0 S
The typical time frame for hotel development is 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 during the previous downturn.
57

exhibit 3
Room Starts
1,500 1,175 1,000
TOTAL CONSTRUCTION: 1980S VS. 1990S
887
500
0 1980-1989 1990-1999
Source: Morgan Stanley, PriceWaterhouseCoopers

(000's)
200 160 5% 120 80 4% 3% 2% 40 1% 0
1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001P
Net Additions to Existing Supply % of Existing Supply
exhibit 4
7%
Projected
HOTEL ADDITIONS TO SUPPLY
6%
0%
Source: Morgan Stanley, PriceWaterhouseCoopers
R E V PA R P E R F O R M A N C E A N D P R O J E C T I O N S
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).
58
PriceWaterhouseCoopers recently issued a reforecast of RevPAR projections for 2001 and 2002. The firm published both a baseline and prolonged weakness scenario as outlined in Exhibit 5.
exhibit 5
PRICEWATERHOUSECOOPERS US NATIONWIDE REVPAR PROJECTIONS

Baseline -12.4 -15.0 -7.4 -3.4 5.8 8 -7.1 0.5 Prolonged Weakness -12.4 -20.3 -12.7 -8.9 -1.3 5.9 -8.5 -4.7
Quarter 3Q 2001 4Q 2001 1Q 2002 2Q 2002 3Q 2002 4Q 2002 Year 2001 Year 2002
The firm noted that neither scenario takes into account future terrorist acts, expansion of the military effort beyond the air campaign and limited ground troops, and military acts beyond Afghanistan. During the Gulf War, 1991 nationwide RevPAR declined about 2.5%, while current projections anticipate RevPAR to decline about 7% in 2001. PriceWaterhouseCoopers uses it own GDP estimates and other macroeconomic inputs to forecast RevPAR. The forecast assumes that GDP growth will be moderately positive in 2001 and 2002 with a recession in the second half of 2001 and a recovery in the first half of 2002.
exhibit 6
10%
6.6% 5.6%
NATIONWIDE REVENUE PER AVAILABLE ROOM (REVPAR) GROWTH
5%
0%
(2.5%)
-5%
(6.9%)
-10% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002E
Source: PriceWaterhouseCoopers, Smith Travel Research
59

H I G H E R P R O F I TA B I L I T Y H I G H E R A N D L O W E R D E B T S E R V I C E
Currently, the hotel industry is better positioned to withstand a decline in revenues from a profitability and leverage perspective than it was 10 years ago. In aggregate, hotel profitability was negligible between 1980 and 1989. In contrast, the industry should post profits exceeding $20 billion this year. From a leverage perspective, 1991 debt service payments accounted for over 14% of nationwide hotel revenues, while 2001 debt service payments should equal about 4% of hotel revenues.
exhibit 7
Profits ($bn) Interest Rate
Profits Interest Rate
30
14% 12% 10% 8%
NATIONWIDE HOTEL PROFITABILITY AND 10-YEAR TREASURY RATES
20
10 6% 0 4% 2%
1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
-10
0%
R AT I N G A G E N C Y V I E W S O N H O T E L S
In late 2001, rating agencies commented on the corporate ratings of several hotel companies, and revised rating guidelines for future CMBS transactions. In late September, all three rating agencies commented on the ratings of several hotel operating companies and hotel REITs. See Exhibit 8 for details. In addition to the comments on corporate ratings, Moodys and Fitch outlined changes to underwriting guidelines for hotel loans within CMBS transactions. Standard & Poors has not made any changes to its underwriting guidelines for hotels in new issue transactions.
60
Fitch noted in a recently released report that it is concerned about the performance of hotels over the next 12-24 months. Therefore, for new issue 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 12-month 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. Moodys 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.
61
62
chapter 5
exhibit 8
Moody's Rating Baa1 Baa3 BBBBBBBBB+ Rating Watch Negative Rating Watch Negative Rating Watch Negative BBBBaa3 Baa1 Review for possible Downgrade Review for possible Downgrade Review for possible Downgrade Review for possible Downgrade Moody's Comments S&P Rating S&P Comment BBB+
RATING AGENCY VIEWS ON HOTEL COMPANIES, AS OF NOVEMBER 2001

Fitch Ratings Fitch Comments Rating Watch Negative
Hotel Operating and Franchise Companies
Equity Ticker
Hotel Brands
Cendant Corporation Choice Hotels International
CUC CHH
Hilton Hotels Corporation
HLT
Rating Watch Negative
Marriott International
MAR
Starwood
HOT
Ba1
Review for possible Downgrade
BBB-
BBB-
Four Seasons
FS
Ramada, Howard Johnson Comfort, Quality, Clarion, Sleep Roadway, Econo Lodge, MainStay Hilton, Hilton Garden Inn, Hilton Suites, Doubletree, Guest Suites, Hampton Inn Marriott, Renaissance, Courtyard, Residence Inn, Fairfield Inn, TownePlace SpringHill Suites, Ritz-Carlton Westin, Sheraton, St.Regis/Luxury Collection Four Points, W Hotels Four Seasons Baa3 Rating confirmed negative outlook BBBAffirmed
Hotel REITs B3 (preferred) Ba2 (senior unsec) B1 (preferred) Baa3 (senior unsec) Ba1 (preferred) Ba2 (senior debt) Ba3 (preferred) Ba3 (senior debt) Ba3 (senior debt) Affirmed Review for possible Downgrade Review for possible Downgrade Review for possible Downgrade Lowered rating from Ba2, under review Affirmed Review for possible Downgrade Review for possible Downgrade Review for possible Downgrade
Equity Ticker Moody's Rating Moody's Comments
Property Type of Ownership
S&P Rating BBBB
S&P Comment Rating Watch Negative Rating Watch Negative
Fitch Ratings
Fitch Comments
Equity Inns
ENN
Mid-scale Limited Service Hotels
FelCor Lodging Trust
FCH
Upscale Full Service Hotels
Hospitality Properties Trust
HPT
Upscale Extended Stay
BBB-
Lowered rating watch from positive to stable
Host Marriott Corporation
HMT
Luxury Full Service Hotels
BB
LaQuinta Properties
LQI
Limited Service
BBBB-
Rating Watch Negative Revised rating watch from positive to rating watch developing
BB-
MeriStar Hospitality Corporation
MHX
Full Service Hotels
Winston Hotels
WHX
Limited Service, Extended Stay
B3 (preferred)
Review for possible downgrade
Source: Morgan Stanley, Moodys, Fitch, S&P
R AT I N G A C T I O N S I N 2 0 0 1
CMBS transactions are typically viewed on an annual basis by the rating agencies. Of the 70 CMBS transactions that have more than 15% hotel exposure, the rating agencies have taken some sort of rating action on various tranches within 16 transactions. In total, 59 rating actions were issued, 54 upgrades and 5 downgrades. The upgrade/downgrade ratio on these transactions with greater than average hotel exposure was 11:1, slightly better than the 10:1 ratio for the entire CMBS universe performance during the same period.
HOTEL DELINQUENCIES IN CMBS TRANSACTIONS
Based on our September 2001 remittance reports data, delinquencies on hotel loans accounted for 1.63% of current balances. Hotel delinquencies have trended 55 bp higher than our universe average since we began tracking the data in late 1999. Hotel delinquencies spiked to 1.93% in May 2001 and have moderated over the past several months. We anticipate delinquencies will continue to rise through the remainder of the year and into early 2002. If we assume that delinquencies on hotel loans double over the next 12 months, hotel delinquencies would rise to about 3.25% of current balances within conduit transactions. When we apply recovery assumptions based on the experience from a longterm study conducted by Howard Esaki, the average loss on defaulted loans is about 35%. If the defaulted loans experience the average recoveries, the loss to a conduit transaction would be about 1.20% (3.25 x .35). Based on 2001 subordination levels (2% subordination to B and 4.5% subordination to BB), a 1.20% loss to a typical conduit transaction would only impact the unrated bonds and possibly the B bonds.
63

exhibit 9
2.00%
Hotel Total/Average
CMBS DELINQUENCIES BY HOTEL AND TOTAL
1.50%
1.00%
0.50%
0.00% Aug-99
Feb-00
Aug-00
Feb-01
Sep-01
CMBS TRANSACTIONS WITH HOTEL EXPOSURE
In light of the declining operating performance of hotels since September 11, we looked at CMBS transactions to determine which had the greatest exposure to hotel properties. On average, conduit transactions have 10% hotel exposure. A search of the Trepp database revealed 70 transactions had hotel exposure exceeding 15%. Of the 70 that we examined, 17 were transactions with 100% hotel concentration. In addition to looking at hotel exposure,we also examined DSCR. Typically, a transactions DSCR increases over time as the property operations stabilize and cash flows increase. In addition to the DSCR for the entire transaction we also examined the DSCR for the hotel assets within the transactions. Reporting of financial statements is typically done on a quarterly basis so the DSCR reported maybe be somewhat outdated. A declining DSCR from issuance should be a flag to investors since nationwide hotels have posted very solid RevPAR growth numbers in 1999 and 2000. Since RevPAR is anticipated to decline for the remainder of this year and early next year hotel assets that were posting declining DSCR prior to September 11 will be under particular pressure.
CONDUIT TRANSACTIONS
On average, the DSCR for deals with more than 15% hotel concentration but less than 100% were 2 bp higher than at issuance. Five transactions had DSCR more than 2 bp below issuance levels. The transaction that showed the greatest DSCR decline was Starwood Asset Receivables Trust 2000-1 with a 21 bp decline. The transaction had no delinquencies as of the most current remittance report. The DSCR attributable to just the hotel assets within the transactions we examined averaged 2.10x, 30 bp higher than the average for the entire transaction. The 17 transactions with 100% hotel exposure had DSCR that averaged 2.86x, which was 61 bp higher than at issuance.
64
LARGE LOAN TRANSACTIONS
None of the large loan transactions had any delinquent loans as of the reporting dates noted. The average loan-to-value ratios (LTV) was 48% on the large loan transactions, significantly below the typical 65%-75% average leverage in CMBS conduit transactions. The low leverage and improving DSCR indicate that there is significant cushion in these deals to withstand predicted declines in operating performance. Since RevPAR growth peaked in 2000, the most vulnerable transactions to a significant decline in operating performance and DSCR would be the transactions issued in 2001 which may have taken into account 2000 operating performance. The rating agencies anticipate that 1999 should be considered a stabilized operating year for hotels. Transactions issued prior to 1999 have significant cushion so if revenues drop from 2001 levels to 1999 levels the transaction will still produce DSCR above underwritten levels. We examined hotel concentrations in CMBS transactions over time. Specifically, we examined diversified CMBS transactions with greater than 10% hotel exposure. Issuance of deals with these characteristics peaked at 28 in 1998 and declined to 18 in 1999, 13 in 2000 with only 7 deals issued in 2001. Since the rating agencies consider 1999 to be long term stabilized hotel performance, it appears that only 21 of the 103 we examined may have been issued off of greater than stabilized performance. It is also worth noting that the rating agencies have always been cautious about the underwriting of hotel loans, applying haircuts to above-market performance. An average of several years of historical performance has always been considered stabilized. These underwriting standards are intended to provide cushion within the loans to withstand recession-like downturns.
STRESS SCENARIOS FOR LARGE LOAN DEALS
Morgan Stanley has issued only one CMBS transaction in the past two years with hotel concentration that exceeds 15% the 2000 Hilton Hotels transaction (HHPT 2000-HLT). In order to assess how that transaction and the 1999 Host Marriott transaction (HMPT 1999-HMTX) may perform in a declining RevPAR environment, we analyzed the most recent financial statements on the deals and applied various stress scenarios to the transactions. In the six weeks after September 11, 2001, RevPAR declines nationwide have averaged about 20%, moderating to about 17% in the last three weeks. As highlighted earlier in this report declines are anticipated to moderate further through the remainder of the year and recover in 2002. In 2001 nationwide RevPAR is anticipated to decline 7% with flat to slightly positive growth in 2002.
65

exhibit 10
DEALS WITH MORE THAN 15%, BUT LESS THAN 100% HOTEL EXPOSURE
Deal Name
Deal Type
Lead Manager
Salomon Brothers Mortgage Securities VII, 2000-FL1 Nomura Asset Securities Corp., 1996-MD5 Credit Suisse First Boston Mortgage Securities Corp., 1998-FL2 Asset Securitization Corp., 1996-MD6 COMM, 2000-FL2 GS Mortgage Securities Trust II, 2001-GSFL4 Asset Securitization Corp., 1997-MD7 DLJ Commercial Mortgage Corp., 1998-STF1 Credit Suisse First Boston Mortgage Securities Corp., 1998-FL1 Asset Securitization Corp., 1995-MD4 Credit Suisse First Boston Mortgage Securities Corp., 2000-FL2 Allied Capital Commercial Mortgage Trust, 1998-1 DLJ Commercial Mortgage Corp., 1999-STF1 LB Commercial Conduit Mortgage Trust, 1996-C2 GS Mortgage Securities Corp., 1998-GL2 BTR2 Trust, 1999-S1 LB Commercial Mortgage Conduit Trust, 1995-C2 Commercial Mortgage Acceptance Corp., 1997-ML1 Commercial Mortgage Acceptance Corp., 1996-C2 Asset Securitization Corp., 1996-D3 Asset Securitization Corp., 1995-D1 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2001-LLF4 Chase Commercial Mortgage Securities Corp., 1998-SN1 Criimi Mae CMBS Corp., 1998-1 Credit Suisse First Boston Mortgage Securities Corp., 1995-AEW1 Lehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 COMM, 2001-FL4 DLJ Commercial Mortgage Corp., 1998-STF2 Asset Securitization Corp., 1996-D2 Credit Suisse First Boston Mortgage Securities Corp., 2000-FL1 COMM, 2001-J1 Lehman Large Loan, 1997-LL1 SASCO Floating Rate Commerical Mortgage Trust, 1999-C3 Nomura Asset Securities Corp., 1994-C3 DLJ Commercial Mortgage Corp., 1998-CF1 Salomon Brothers Mortgage Securities VII, 1996-C1 Blackrock Capital Finance, 1997-C1 Chase Commercial Mortgage Securities Corp., 1998-1 LB Commercial Conduit Mortgage Trust, 1998-C4 Morgan Stanley Capital I Inc., 1998-XL1 Credit Suisse First Boston Mortgage Securities Corp., 1997-C1 Morgan Stanley Capital I Inc., 1997-XL1 Credit Suisse First Boston Mortgage Securities Corp., 1997-C2 GS Mortgage Securities Corp., 1998-C1 COMM, 2000-C1 Credit Suisse First Boston Mortgage Securities Corp., 1998-C1 Starwood Asset Receivables Trust, 2000-1 Nomura Asset Securities Corp., 1995-MD3 Mortgage Capital Funding, 1998-MC3 First Union - Chase Commerical Mortgage Trust, 1999-C2 Morgan Stanley Capital I Inc., 1998-WF1 Nomura Asset Securities Corp., 1998-D6 J.P. Morgan Commercial Mortgage Finance Corp., 1998-C6 Total/Weighted Average
Conduit Large Loan Short Term Large Loan Short Term Short Term Large Loan Short Term Short Term Large Loan Short Term Conduit Short Term Conduit Large Loan Short Term Conduit Large Loan Conduit Conduit Conduit Short Term Seasoned Conduit Seasoned Conduit Short Term Short Term Conduit Short Term Private Large Loan Short Term Seasoned Conduit Conduit Seasoned Conduit Conduit Large Loan Conduit Large Loan Conduit Conduit Conduit Conduit Conduit Large Loan Conduit Conduit Conduit Conduit Conduit
Salomon Smith Barney Nomura Credit Suisse First Boston Nomura Deutsche Bank Alex. Brown Goldman Sachs & Co. Nomura Donaldson, Lufkin & Jenrette Credit Suisse First Boston Nomura Credit Suisse First Boston Morgan Stanley Donaldson, Lufkin & Jenrette Lehman Brothers Goldman Sachs & Co. Bankers Trust Lehman Brothers Merrill Lynch Goldman Sachs & Co. Nomura Nomura Lehman Brothers Chase Citibank Credit Suisse First Boston Lehman Brothers Deutsche Bank Alex. Brown Donaldson, Lufkin & Jenrette Nomura Credit Suisse First Boston Deutsche Bank Alex. Brown Lehman Brothers Lehman Brothers Nomura Donaldson, Lufkin & Jenrette Salomon Smith Barney Goldman Sachs & Co. Chase Lehman Brothers Morgan Stanley Credit Suisse First Boston Morgan Stanley Credit Suisse First Boston Goldman Sachs & Co. Deutsche Bank Alex. Brown Credit Suisse First Boston Merrill Lynch Nomura Salomon Smith Barney Chase Morgan Stanley Nomura JP Morgan
Source: Trepp, Realpoint, S&P
66
Current Deal Balance ($mm)
Issue Date % Hotel
Hotel Original DSCR DSCR
Current DSCR
% % For. Change From 30/60/90 and Iss. Del. REO
Total Del.
Top % Top Data As of State State
LTV
216.0 726.7 999.3 836.4 759.5 346.7 456.1 50.5 791.9 867.6 619.1 119.6 170.8 332.5 1,315.9 219.2 170.6 804.0 86.9 724.2 164.6 1,242.5 19.0 422.2 48.1 1,469.2 952.8 79.6 802.0 328.0 790.0 1,350.8 1,001.3 46.8 795.3 107.3 30.8 763.5 1,959.8 884.0 1,245.4 677.4 1,400.5 1,785.8 888.9 2,367.3 779.2 403.3 857.6 1,147.9 1,293.3 3,576.3 748.0 41,041.9
1/31/2000 4/2/1996 11/11/1998 12/17/1996 7/17/2000 2/22/2001 3/27/1997 4/27/1998 6/29/1998 10/30/1995 9/4/2001 1/23/1998 11/30/1999 10/30/1996 5/21/1998 3/25/1999 10/12/1995 12/30/1997 12/30/1996 10/22/1996 8/7/1995 8/16/2001 6/22/1998 6/4/1998 10/30/1995 12/18/2000 6/28/2001 12/28/1998 3/1/1996 12/27/2000 3/29/2001 10/14/1997 10/28/1999 11/29/1994 3/2/1998 2/29/1996 10/7/1997 5/15/1998 11/24/1998 6/11/1998 6/30/1997 10/17/1997 12/19/1997 10/29/1998 9/20/2000 6/25/1998 5/17/2000 3/31/1995 12/18/1998 5/24/1999 3/5/1998 3/30/1998 3/10/1998
53.7 47.9 46.3 43.5 42.9 42.3 41.1 40.0 39.7 39.3 37.8 37.3 35.1 33.9 33.3 30.0 29.0 27.2 26.8 26.7 26.0 25.4 25.1 24.2 23.7 23.4 23.3 23.2 22.7 22.2 21.9 21.4 21.2 20.8 20.1 19.4 17.6 17.6 17.6 17.6 17.5 17.0 17.0 16.7 16.6 16.5 16.3 15.9 15.9 15.6 15.3 15.0 15.0 23.5
NA 2.07 1.59 3.02 2.25 2.70 1.32 0.92 1.26 1.52 1.60 1.55 1.06 1.82 2.16 2.32 1.31 2.01 2.25 1.96 2.65 2.80 1.67 2.13 1.52 1.82 3.22 1.99 1.63 2.48 1.60 2.16 1.55 1.49 1.42 1.15 3.69 2.52 1.66 1.89 2.26 2.27 2.03 1.85 2.20 1.99 1.96 1.93 1.87 1.52 1.96 3.17 2.59 2.10
NA 1.90 1.41 1.96 1.49 2.45 1.74 1.38 1.22 1.77 1.55 1.38 1.16 1.46 1.86 1.69 1.44 1.68 1.22 1.44 1.53 2.90 1.30 1.59 1.26 1.72 2.31 1.35 1.62 1.50 1.73 1.89 1.11 1.65 1.35 1.37 1.42 1.65 1.64 1.90 1.38 2.00 1.43 1.53 1.48 1.50 1.29 1.60 1.54 1.34 1.54 1.58 1.54 1.64
NA 1.89 1.30 2.57 1.56 2.51 1.75 1.41 1.18 1.79 1.55 1.52 0.99 1.48 1.98 1.74 1.51 2.22 1.84 1.72 1.97 2.90 1.94 1.71 1.52 1.72 2.31 1.75 1.60 2.04 1.72 2.16 1.23 1.71 1.57 1.58 2.00 1.84 1.85 2.08 1.65 2.11 1.55 1.72 1.55 1.65 1.50 1.59 1.68 1.44 1.74 1.96 2.09 1.81
NA -0.01 -0.11 0.61 0.07 0.06 0.01 0.03 -0.04 0.02 0.00 0.14 -0.17 0.02 0.12 0.05 0.07 0.54 0.62 0.28 0.44 0.00 0.64 0.12 0.26 0.00 0.00 0.40 -0.02 0.54 -0.01 0.27 0.12 0.06 0.22 0.21 0.58 0.19 0.21 0.18 0.27 0.11 0.12 0.19 0.07 0.15 0.21 -0.01 0.14 0.10 0.20 0.38 0.55 0.17
0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.69 6.69 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 38.59 0.00 38.59 0.00 6.72 6.72 0.00 0.00 0.00 0.00 0.00 0.00 5.64 5.59 11.23 0.00 0.00 0.00 1.02 0.00 1.02 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 8.99 8.99 3.23 0.46 3.69 2.91 0.00 2.91 0.00 0.00 0.00 0.00 0.00 0.00 2.10 0.00 2.10 5.64 0.00 5.64 0.00 0.00 0.00 0.00 0.00 0.00 50.88 22.39 73.27 6.51 3.61 10.12 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 16.97 16.97 0.00 0.18 0.18 5.86 0.00 5.86 2.61 12.59 15.20 0.00 0.00 0.00 1.19 0.00 1.19 0.00 0.00 0.00 2.32 4.90 7.22 0.00 0.00 0.00 0.57 0.89 1.46 1.54 0.00 1.54 0.00 0.00 0.00 2.36 0.13 2.49 0.00 0.00 0.00 0.00 0.00 0.00 2.08 0.00 2.08 4.64 0.00 4.64 0.16 0.00 0.16 0.24 0.00 0.24 1.89 0.00 1.89 1.00 0.67 1.66
10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 9/1/2001 10/1/2001 9/1/2001 10/1/2001 8/1/2001 9/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 9/1/2001 9/1/2001 9/1/2001 9/1/2001 10/1/2001 10/1/2001 9/1/2001 10/1/2001 10/1/2001 10/1/2001 9/1/2001 10/1/2001 10/1/2001 9/1/2001 9/1/2001 10/1/2001 10/1/2001 10/1/2001 9/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 9/1/2001 9/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001 10/1/2001
NY TX HI TX NY CA CA GA CA TX FL FL NY GA CA FL TX CA CA CA TX CA NY MI CA CA RI VA TX CA NY CA TX GA NJ GA NY NY CA CA NY NY CA CA MI CA CA NY CA CA CA CA VA
53.7 14.7 27.5 21.6 31.3 37.6 28.0 38.1 18.5 31.2 23.2 15.4 38.6 11.3 17.1 43.2 20.4 16.2 44.7 26.8 16.1 19.1 26.8 22.1 61.7 32.0 24.9 50.9 16.0 34.6 23.1 24.9 26.6 20.8 13.5 25.4 29.6 19.5 20.1 39.0 26.5 28.9 31.5 15.9 20.4 14.4 30.2 39.7 11.0 13.0 35.0 23.2 21.9
NA 61.0 67.0 51.2 66.1 46.6 55.8 72.8 83.8 56.4 61.7 61.9 52.2 61.1 60.7 60.9 60.9 60.9 NA 62.7 58.9 51.3 44.0 67.7 69.0 55.8 52.5 82.5 70.2 58.4 53.5 56.2 NA 55.6 62.8 61.1 75.2 67.1 64.4 55.9 64.5 52.1 65.1 66.7 65.6 70.3 66.4 61.5 70.3 71.3 65.7 67.7 62.4 63.1
67

exhibit 11
DEALS WITH 100% HOTEL EXPOSURE
Deal Name
Deal Type
Lead Manager
Original Deal Balance ($mm)
Banc of America Large Loan Inc., 2000-WSF-1 Bear Stearns Commercial Mortgage Securities Inc., 1999-WYN1 Bear Stearns Commercial Mortgage Securities Inc., 2000-LCON BTR1 Trust, 1998-S1 CBM Funding Corp., 1996-1 German American Capital Corp., 1996-2 Hotel First Mortgage Trust, 1993-A Hilton Hotels Pool Trust, 2000-HLT Host Marriott Pool Trust, 1999-HMT Meristar Commercial Mortgage Trust, 1999-C1 Merrill Lynch Mortgage Investors Inc., 1998-H1 Nomura Asset Capital Corp., 1993-1 Opryland Hotel Trust, 2001-OPRY Lehman Brothers Floating Rate Commercial Mortgage Trust, 2001-C6 Strategic Hotel Capital Inc., 1999-C1 Strategic Hotel Capital LLC, 2001-SHC1 Starwood Asset Receivables Trust, 1999-C1
Total/Weighted Average
Single Asset Single Asset Single Asset Short Term Single Asset Single Asset Single Asset Single Asset Large Loan Single Asset Single Asset Single Asset Single Asset Single Asset Single Asset Single Asset Single Asset
Banc of America Bear Stearns & Co. Bear Stearns & Co. Bankers Trust Lehman Brothers Deutsche Morgan Grenfell Nomura Morgan Stanley Morgan Stanley Lehman Brothers Merrill Lynch Nomura Merrill Lynch Lehman Brothers Goldman Sachs & Co. Goldman Sachs & Co. Lehman Brothers
177.0 346.0 115.8 422.8 406.2 99.2 67.5 499.6 665.0 330.0 102.3 51.4 275.0 300.0 422.0 455.0 541.3
5,276.1
Source: Trepp, Realpoint, S&P
68
Current Deal Balance ($mm)
Issue Date
% Hotel Original Hotel DSCR DSCR
Current DSCR
Change % From 30/60/90 % For. Iss. Del. and REO
Total Top % Top Del. Data As of State State
LTV
176.2 113.3 22.2 325.8 72.7 55.7 494.0 631.1 320.6 99.4 42.9 271.0 300.0 454.1 518.5
4,588.7
8/30/2000 100.0 3.43 8/22/2000 100.0 2.93 2/13/1998 100.0 2.50 1/24/1996 100.0 2.42 4/24/1996 100.0 2.24 8/13/1993 100.0 2.77 11/9/2000 100.0 2.11 8/18/1999 100.0 2.51 8/30/1999 100.0 2.22 7/23/1998 100.0 2.57 9/30/1993 100.0 1.05 4/19/2001 100.0 1.90 9/27/2001 100.0 3.78 5/8/2001 100.0 2.96 3/16/1999 100.0 4.11
100.0 2.86
1.90 1.72 2.32 1.65 1.82 1.96 2.19 2.22 1.84 2.34 2.19 NA 2.02 NA 3.23 2.82 2.41
2.25
3.43 4.00 2.93 2.50 2.42 2.24 2.77 2.11 2.51 2.22 2.57 1.05 1.90 3.78 2.74 2.96 4.11
2.86
1.53 2.28 0.61 0.85 0.60 0.28 0.58 -0.11 0.67 -0.12 0.38 NA -0.12 NA -0.49 0.14 1.70
0.61
0.00 0.00 0.00 0.00 NA 0.00 NA 0.00 0.00 NA 0.00 NA 0.00 NA 0.00 0.00 0.00
0.00
0.00 0.00 0.00 0.00 NA 0.00 NA 0.00 0.00 NA 0.00 NA 0.00 NA 0.00 0.00 0.00
0.00
0.00 10/1/2001 0.00 10/1/2001 0.00 10/1/2001 0.00 NA NA 9/1/2001
CA CA PR NJ CA CA NY CA FL TN TN FL CA NY CA
100.0 62.3 20.3 46.4 100.0 43.8 100.0 59.6 100.0 NA 34.7 52.8 22.2 42.0 50.0 44.3 52.0 45.3 37.5 52.9 100.0 59.5 64.5 NA 100.0 45.1 100.0 42.0 40.2 54.9 50.0 47.4 38.7 46.4
48.0
307.8 11/10/1999 100.0 4.00
10/1/2001 Multi 8/1/2001 WA
0.00 10/1/2001 0.00 11/1/2000 0.00 10/1/2001 NA NA NA 0.00

0.00
10/1/2001 9/1/2001 9/1/2001 9/1/2001
0.00 10/1/2001 0.00 10/1/2001 0.00 10/1/2001 0.00 10/1/2001
383.5 10/21/1999 100.0 2.74
69

H H P T 2 0 0 0 - H LT
Since 1997, the DSCR on this transaction has steadily increased. The transaction was underwritten based on adjustments to trailing twelve-month financial statements as of June 2000 resulting in a 2.31x DSCR. The year-end 2000 financial statements resulted in a 2.54 DSCR. Year-to-date September financial statements produced a 2.11 DSCR. In order to anticipate future operating performance, several stress scenarios were applied to the HLT transaction. Stress scenario 1 applies a 7% decline in revenues to 2000 operating numbers while holding expenses constant. The decline in revenues is in line with PriceWaterhouseCoopers nationwide projections for 2001. Although variable expenses decline with revenues (typically 35%) we used a conservative assumption that there would be no cost savings associated with the decline in revenues. Stress scenario 1 results in a 1.98 DSCR. Stress scenario 2 applies a 19% reduction to year end 2000 revenues and holds operating expenses constant resulting in a 1.01 DSCR. Scenario 3 applies a 24% reduction to year-end 2000 revenues and reduces nonfixed expenses by 8% resulting in a 1.0 DSCR. The 8% reduction in operating expenses is based on the industry trend of 35% variable expenses associated with changes in revenues (.24 x .35 =.08). Each of these stress scenarios indicates the transaction can withstand a significant decline in revenues and still generate enough cash flow to cover debt service payments. Since the rating agencies view 1999 as a stabilized year, it is also worth noting that based on 1999 numbers, the transaction generates a DSCR of 2.13x. The parent companys financial position and credit rating are also important when considering how large loan transactions might perform. Hilton is rated Baa3 by Moodys and BBB- by S&P and Fitch. The Morgan Stanley analyst that covers the Hilton corporate debt anticipates that the company will generate over $200 million in free cash flow in 2002. These estimates take into account third quarter earnings and significant declines in RevPAR through the end of 2001. The corporate debt carries an Outperform rating with the Morgan Stanley analyst.
70
HMPT 1999-HMTX
The historical operating numbers on this transaction also show significant improvements since 1997. Between 1997 and year-end 2000, DSCR increased from 1.99x to 2.71x. The trailing twelve month (TTM) operating numbers through June 2001 (TTM June 2001) resulted in a 2.51 DSCR. The third quarter operating statements should be provided by the servicer in the near future. We will provide those numbers in a separate piece once they are available. We also applied stress scenarios to cash flows from the Host Marriott transaction. Scenario 1 applies a 7% decline to TTM June 2001 numbers while holding all operating expenses constant. These assumptions are based on the 2001 PriceWaterhouseCoopers RevPAR projections. Stress scenario 1 results in a 1.92 DSCR. Scenario 2 applies a 19% reduction to TTM June revenues while holding all departmental revenues constant. These assumptions produce a 1.01 DSCR, indicating revenues can decline 18% with no reduction in operating expenses without putting debt service payments in jeopardy. Stress scenario 3 assumes revenues decline 22% with a 7.7% reduction in only non-fixed operating expenses resulting in a 1.0x DSCR. The 7.7% reduction in expenses is based on the assumption that about 35% of the expenses associated with the revenue are attributable to variable expenses (.22 x .35 = 7.7). Stress scenario 1 is likely the closest to what may occur within the following 12 months. However, each of these scenarios indicates that the transaction can experience severe declines in revenues without placing debt service payments in jeopardy. The DSCR based on 1999 numbers is 2.10x, this performance is considered a stabilized year by the rating agencies. Host Marriott has a corporate credit rating of Ba2 by Moodys and BB by S&P. According to the Morgan Stanley debt analyst, the company currently has about $200 million in cash and an additional capacity of $315 million on its line of credit. The corporate debt carries an Outperform rating at Morgan Stanley.
71

exhibit 12
HILTON HOTELS POOL TRUST 2000 HLT OPERATING STATEMENTS

TTM Jun-00 Revenue (%) ($000s) 367,288 142,319 225,163 (78,002) 147,140 (27,525) 119,616 (17,300) 100% 40% 60% 21% 40% 7% 32% 5% Underwritten ($000s) 367,288 (142,294) 224,994 (75,571) 149,423 (26,329) 123,094 (16,625)
1997 Revenue ($000s) (%) Total Departmental Revenues Total Departmental Expenses Total Departmental Profits Total Undistributed Expenses Income Before Fixed Expenses Total Fixed Expenses Net Operating Income FF&E Reserve 306,414 (127,788) 178,626 (65,955) 112,671 (24,474) 88,197 (14,399) 100% 42% 58% 22% 37% 8% 29% 5%
1998 Revenue ($000s) (%) 338,300 (134,889) 203,411 (70,229) 133,182 (23,628) 109,554 (15,901) 100% 40% 60% 21% 39% 7% 32% 5%
1999 Revenue ($000s) (%) 353,367 (139,744) 213,623 (72,596) 141,027 (26,329) 114,698 (16,625) 100% 40% 60% 21% 40% 7% 32% 5%
Net Cash Flow Debt Service Debt Service Coverage Ratio
73,798 46,126 1.60
24% 15%
93,653 46,126 2.03
28% 14%
98,073 46,126 2.13
28% 13%
102,215 46,126 2.22
28% 13%
106,469 46,126 2.31
Portfolio Statistics Occupancy Average Daily Rate (ADR) Rev Per Available Room (RevPAR) 75% $144.90 $109.19 74% $161.75 $119.18
Annual Growth -1.0% 12% 9% 76% $167.01 $126.07
Annual Growth -1.0% 3% 6% 76% $174.32 $131.99
Annual Growth -1.0% 4% 5% 76% $174.30 $132.14
Stress Scenarios Scenario 1 7% decline in revenues, all expenses held constant at December 2000 levels Scenario 2 19% decline in revenues, all expenses held constant at December 2000 levels Scenario 3 24% decline in revenues, non-fixed expenses at 92% of December 2000 levels
Source: Morgan Stanley, Hilton Hotels Corporation
72
Revenue (%) 100% 39% 61% 21% 41% 7% 34% 5%
2000 Revenue ($000s) (%) 389,400 (147,600) 241,800 (78,200) 163,600 (27,100) 136,500 (19,470) 100% 38% 62% 20% 42% 7% 35% 5%
First 9 Months 2001 Revenue ($000s) (%) 256,300 (102,300) 154,000 (54,300) 99,700 (21,200) 78,500 (12,815) 100% 39% 61% 20% 41% 9% 32% 5%
Stress Scenario 1 362,142 (147,600) 214,542 (78,200) 136,342 (27,100) 109,242 (18,107)
Revenue (%) 100% 41% 59% 22% 38% 7% 30% 5%
Stress Scenario 2 315,414 (147,600) 167,814 (78,200) 89,614 (27,100) 62,514 (15,771)
Revenue (%) 100% 47% 53% 25% 28% 9% 20% 5%
Stress Scenario 3 295,944 (135,792) 160,152 (71,944) 88,208 (27,100) 61,108 (14,797)
Revenue (%) 100% 46% 54% 24% 30% 9% 21% 5%
29% 13%
117,030 46,126 2.54
30% 12%
65,685 31,090 2.11
27% 13%
91,135 46,126 1.98
25% 13%
46,743 46,126 1.01
15% 15%
46,311 46,126 1.00
16% 16%
U/W Assumption -1.0% 0% 0%
73

exhibit 13
HOST MARRIOTT 1999 HMT OPERATING STATEMENTS
1997 Revenue ($000s) (%)
1998 Revenue ($000s) (%)
1999 Revenue Underwritten ($000s) (%) ($000s)
Total Departmental Revenues Operating Expenses Total Departmental Profits Fixed Charges Income Before Fixed Expenses FF&E Reserve Net Cash Flow Debt Service DSCR (NCF/Debt Service)
449,345 (273,150) 176,195 (29,395) 146,800 (18,783) 128,017 64,197 1.99
100% 501,798 61% (295,149) 39% 206,649 7% (39,400) 33% 167,249 4% (22,198) 28% 14% 145,051 64,197 2.26
100% 530,473 59% (311,104) 41% 219,369 8% (64,964) 33% 154,405 4% (19,450) 29% 13% 134,955 64,197 2.10
100% 59% 41% 12% 29% 4% 25% 12%
484,636 (293,570) 191,066 (53,642) 137,424 (19,450) 117,974 64,197 1.84
Annual Growth
Annual Growth
Annual Growth
Occupancy Average Daily Rate Revenue Per Available Room (RevPAR)
81.4% $174.00 $142.00
N/A N/A N/A
83.1% $192.00 $159.00
1.7% 10% 12%
82.5% $195.00 $161.00
-0.6% 1.6% 1.3%
81.5% $184.00 $150.00
Stress Scenarios Scenario 1 7% decline in revenues, all expenses held constant at June 2001 TTM levels Scenario 2 18% decline in revenues, all expenses held constant at June 2001 TTM levels Scenario 3: 22% decline in revenues, expenses at 93% June 2001 TTM levels
Source: Morgan Stanley, Host Marriott
74
Revenue (%)
2000 Revenue ($000s) (%)
TTM Jun-01 Revenue ($000s) (%)
Scenario 1
Revenue (%)
Scenario 2
Revenue (%)
Scenario 3
Revenue (%)
100% 590,894 61% (332,225) 39% 258,669 11% (58,770) 28% 199,899 4% (25,941) 24% 13% 173,958 64,197 2.71
100% 56% 44% 10% 34% 4% 29% 11%
572,297 (322,505) 249,792 (59,771) 190,021 (28,615) 161,407 64,197 2.51
100% 56% 44% 10% 33% 5%
532,236 (322,505) 209,732 (59,771) 149,961 (26,612) 123,349 64,197 1.92
100% 61% 39% 11% 28% 5%
446,392 (299,929) 146,462 (59,771) 86,691 (22,320) 64,372 64,197 1.00
100% 67% 33% 13% 19% 5%
441,241 (295,092) 146,149 (59,771) 86,378 (22,062) 64,316 64,197 1.00
100% 67% 33% 14% 20% 5%
UW Assumption
-1.0% -5.6% -6.8%
75
Recent Performance of Hotels

W H AT W A S H O T I S N O W N O T
Prior to the significant decline in air traffic after September 11, 2001, full-service urban hotels were considered the most desirable properties because of strong corporate demand and high barriers to entry. Hotels in secondary and suburban markets were less desirable because of significant new development in those markets. Recently, the urban properties in major metro areas have experienced greater declines in RevPAR than the nationwide average, while secondary markets and suburban properties are benefiting. This is explained by the desire of business and leisure travelers to drive rather than fly to locations. In terms of geographic distribution, major cities suffered the most since midSeptember. On average, most of the major 25 metropolitan markets have underperformed the nationwide RevPAR trends. The markets hardest hit since midSeptember are Boston, Miami, New York, Orlando and San Francisco. Houston, Nashville, Norfolk, Philadelphia, St. Louis, and Tampa have produced results better than the national average. Smith Travel data shows that nationwide RevPAR was down about 4.0% year to date through September 2001. The Persian Gulf War, which occurred during the first quarter of 1991, resulted in a 2.4% decline in RevPAR. It took a full year for lodging demand to recover after the Persian Gulf War. Demand will likely be harder hit over the next 6-12 months than occurred during the Persian Gulf War, but additional supply is much lighter than it was in the early 1990s.
76
exhibit 14
HOTEL MARKET DATA AS OF YTD SEPTEMBER 2001

Occupancy(%) Avg Room Rate($) 2001 92.45 80.34 139.00 113.01 80.72 79.99 82.05 75.73 98.21 111.33 86.76 74.88 119.02 185.58 75.39 118.00 88.31 98.70 101.80 113.58 147.05 99.19 72.86 88.40 119.06 2000 87.75 80.76 142.46 116.00 81.62 79.97 81.35 75.48 97.23 106.99 84.30 74.92 118.29 194.85 74.52 115.70 88.70 101.77 100.14 110.18 147.32 96.97 71.67 85.38 114.88 %Chg. 5.4 -0.5 -2.4 -2.6 -1.1 0.0 0.9 0.3 1.0 4.1 2.9 -0.1 0.6 -4.8 1.2 2.0 -0.4 -3.0 1.7 3.1 -0.2 2.3 1.7 3.5 3.6
Top 25 Markets AnaheimSanta Ana, CA Atlanta, GA Boston, MA Chicago, IL Dallas, TX Denver, CO Detroit, MI Houston, TX Los AngelesLong Beach, CA Miami-Hialeah, FL MinneapolisSt. Paul, MN Nashville, TN New Orleans, LA New York, NY NorfolkVirginia Beach, VA Oahu Island, HI Orlando, FL Philadelphia, PA Phoenix, AZ San Diego, CA San Francisco/ San Mateo, CA Seattle, WA St. Louis, MO TampaSt. Petersburg, FL Washington, DC
2001 70.2 63.7 67.4 64.2 59.5 66.3 61.6 66.9 70.7 69.0 66.0 57.9 66.8 74.2 61.1 73.2 68.1 64.4 60.6 73.6 69.1 66.7 62.9 66.1 70.5
2000 %Chg. 72.1 67.5 77.0 72.1 66.0 70.7 68.5 63.4 74.2 70.9 71.0 61.3 70.6 83.3 61.6 76.7 74.8 66.9 63.0 75.9 82.9 71.3 65.1 66.9 75.5 -2.6 -5.6 -12.5 -11.0 -9.8 -6.2 -10.1 5.5 -4.7 -2.7 -7.0 -5.5 -5.4 -10.9 -0.8 -4.6 -9.0 -3.7 -3.8 -3.0 -16.6 -6.5 -3.4 -1.2 -6.6
exhibit 15
HOTEL RATES BY PRICE CATEGORY AS OF SEPTEMBER 2001 (IN %)

2001 YTD 68.5 64.4 60.9 58.5 60.5 147.43 94.68 70.77 54.92 41.46 2000 71.8 65.1 61.5 58.5 59.5 144.99 90.88 68.23 52.44 42.20 1999 71.9 64.9 61.1 58.0 58.7 138.88 87.37 64.89 49.23 89.60 1998 72.6 65.9 62.0 58.4 58.7 133.58 85.33 62.15 47.14 37.43 1997 78.4 70.8 67.9 61.2 59.9 136.01 92.07 66.97 49.93 40.60
Luxury Occupancy Rates Upscale (%) Mid-Price Economy Budget

Room Luxury Rates Upscale ($)
Mid-Price Economy Budget
77
Hotel Branding and Segmentation

BRANDING
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 companys 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 16
HOTEL BRANDING AND SEGMENTATION

Moderate Service
More Expensive
y om s on en Ec th As reng St
Midscale (With F&B) Holiday Inn Four Points Midscale (Without F&B) Hampton Inn Fairfield Residence Inn Comfort Inn
Upscale Doubletree Embassy Suites Wyndham Crowne Plaza Courtyard
Upper-Upscale Hilton Marriott Hyatt Four Seasons Ritz-Carlton W Hotels
High Service
Economy Days Inn Red Roof Inn Travel Lodge Super 8 Motel 6
As
y om on s Ec aken e W
Less Expensive
UPPER UPSCALE
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, threeor 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.
78
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-SCALE WITH FOOD AND BEVERAGE
These hotels encompass 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-SCALE WITHOUT F&B CHAINS
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 SEGMENT
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 budgetconscious business or pleasure traveler. Most bookings are made through the reservation system and repeat business is not significant.
79
History of the Hotel Industry

Lodgings 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.
17901920 GROWTH OF AMERICA AND FINANCIAL BOOM
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.
1 9 3 0 1 9 4 0 s T H E G R E AT D E P R E S S I O N
The boom of the20s 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 companys 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%.
19501960s BOOM AFTER WWII
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 Johnsons 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.
80
1970s QUICK CYCLE
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 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.
1980s
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 passive 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.
E A R LY 1 9 9 0 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.
81

exhibit 17
15% 12% 9% 6% 3% 0% 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001E
Cap-Rate Supply Interest Rate
CAPITALIZATION RATES AND 10YEAR TREASURY RATES
Source: PriceWaterhouseCoopers, Smith Travel Research
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 well-below 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.
M I D - T O L AT E 1 9 9 0 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 accountable to shareholders, reducing hotel leverage levels. In general hotel REITs did not exceed 60% leverage. Capital flow into hotel REITs declined significantly in 1998 to just over $500 million. This was due to Congress enacting legislation limiting acquisitions by pairedshare 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.
82
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.
83
Call Protection
E X E C U T I V E S U M M A RY
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 borrowers 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 terms of the commercial mortgage markets. This chapter will: 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.
84
DEFINITION OF TYPES OF LOAN LEVEL CALL PROTECTION
Several different types of call protection are found in commercial mortgage loans.
HARD LOCKOUT
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 MAINTENANCE
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 Loan Maturity Date: 12/31/07
Original Loan Balance: $10,000,000 Mortgage Rate: 7.25%
Start of Yield Maintenance Period: 1/01/98 Reference Treasury Rate: 5.75%
85
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 also decreases in importance to the lender as the remaining penalty period declines. As the term to maturity decreases, the remaining loan payments represent a lower percentage of the lenders 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 10-YR 11.18 9.20 7.27 5-YR 6.36 5.27 4.18 3-YR 4.03 3.34 2.66
Reference Rate Treasuries Flat Treasuries + 25 bp Treasuries + 50 bp

F I X E D P E R C E N TA G E P E N A LT Y P O I N T S
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 terms 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.
86
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 borrowers new mortgage rate would have to be 110 bp lower than the existing mortgage rate to justify paying the penalty.
exhibit 2 Penalty Points 10 9 8 7 6 5 4 3 2 1
ANNUAL BASIS POINTS OF SAVINGS REQUIRED TO PAY FOR PREPAYMENT PENALTY POINTS (bp)
10-YR 150 140 120 110 90 80 60 50 30 20 8-YR 180 160 140 130 110 90 70 60 40 20 Loan Term to Maturity 6-YR 4-YR 2-YR 220 200 170 150 130 110 90 70 50 20 300 270 240 210 180 150 120 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 investors 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 US 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 US 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 US Treasury strips that would replicate the payments due on the mortgage loan. For the US Treasury strips, we assumed the following rates and prices:
87
Call Protection
1-Year 2-Year 3-Year Rates(%) 5.65 5.68 5.73 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 US Treasury strips and the payments to the CMBS trust from the US Treasury strips.
exhibit 3
CASH FLOWS TO CMBS TRUST UNDER DEFEASANCE

Payment Due From Borrower on Loan $700,000 $700,000 $10,700,000 Cost to Borrower to Purchase US Treasury1 $660,520 $624,400 $9,010,470 Payments From US Treasury Strips to CMBS Trust $700,000 $700,000 $10,700,000
Time Year 1 Year 2 Year 3

1
At time of prepayment.
Nomura Asset Capital Corporation was the first originator to introduce loans with a defeasance option in CMBS pools. All of the loans backing the recently issued $3.7 billion Nomura Asset Securities Corporation 1998-D6 are either locked out from prepayments or prepayable through the exercise of the defeasance option. 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 US 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 US Treasury securities. 3) The servicer is responsible for purchasing the US 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 US 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
88
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.
A L L O C AT I O N O F P R E PAY M E N T P E N A LT I E S
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 025%. 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 6575% of the penalty while the current principal paying bond receives the remainder making it whole to the bonds 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 investors return.
Prepayment Distribution Formula = Bond Pass Through Rate - Reference Treasury Rate Mortgage Loan Rate - Reference Treasury Rate
Bond Pass 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
89
Call Protection
R E L AT I V E V A L U E A N A LY S I S O F D I F F E R E N T T Y P E S O F C A L L P R O T E C T I O N
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 0100% CPR 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.
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.
90
S H O R T A A A ( 1 S T PAY M E N T P R I O R I T Y ) B O N D S W I T H S H A R E D Y I E L D M A I N T E N A N C E P E N A LT I E S
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 premium dollar price on the bond. As an example, assume: Bond Dollar Price: 106 Mortgage Loan Rate: 8.80% Bond Coupon: 6.85% Reference Treasury Rate: 5.65% Yield Maintenance Penalty Period:
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. Premium Dollar Price of Short AAA = 6% Advantage = 2.3% (of current balances)
9 years
91
Call Protection
The fact that the prepayment penalty allocated to the investor is higher than the premium 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.
S H O R T A A A B O N D S W H I C H D O N O T S H A R E P R E PAY P E N A LT I E S
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 premium 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
EFFECT ON YIELD TO MATURITY AS PREPAYMENT SPEEDS INCREASESHORT AAAs

300 bp Decreases 100 bp Decreases Unchanged Decreases Interest Rates +100 bp +300 bp Decreases Increases
Bond Class Short AAA (IO Receives All Penalties) Short AAA (Make Bonds Whole)
Increases
Increases
Increases
Increases
Increases
AA BONDS
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 premium. 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% CPR 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.
92
exhibit 5
EFFECT ON YIELD TO MATURITY AS PREPAYMENT SPEEDS INCREASE (AA)

300 bp Decreases 100 bp Decreases Unchanged Decreases Interest Rates +100 bp +300 bp Increases Increases
Bond Class AA (IO Receives All Penalties) AA (Make Bonds Whole)

Varies
Varies
Increases
Increases
Increases
I N T E R E S T O N LY B O N D S
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
EFFECT ON YIELD TO MATURITY AS PREPAYMENT SPEEDS INCREASE (IO)

300 bp Increases 100 bp Increases Unchanged Increases Interest Rates +100 bp +300 bp Increases Decreases
Bond Class IO (IO Receives All Penalties) IO (Make Bonds Whole)

Increases
Increases
Increases
Increases
Decreases
TRENDS IN CMBS CALL PROTECTION
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.
93
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: THE BORROWERS PERSPECTIVE
The following example compares a prepayment under defeasance to a prepayment under yield maintenance from the borrowers 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, US 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: $9,189,718 $779,020 (8.48%) $9,970,826 $781,108 (8.50%) ($9,970,826$9,189,718) -$2,088 (0.02%)
Defeasance Advantage to Borrower:
So, the defeasance option in this case is more expensive to the borrower by 0.02% or $2,088.
94
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.
95
Large Loans
INTRODUCTION
Recently, the securitization of stand-alone large loans has re-emerged as a significant portion of the CMBS market. By stand-alone large loans, we 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 creditworthy borrowers than conduit loans. As of March 2002, AAA securities from 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. In this paper, we explore the credit quality of stand-alone large loans in the current market and how credit classes of stand-alone large loan deals are priced in the capital markets. Our main findings are: Stand-alone large loans generally have much lower LTVs and higher debt service coverage ratios than the average conduit loan. Rating agencies are wary of the concentration risk inherent in singleasset and single-borrower transactions and therefore often assign higher credit support levels (for a given LTV) to these deals than conduits. Our analysis shows that the volatility of NOI must be 1.5 to 3 times higher on large loans than on conduit loans to equate AAA optionadjusted spreads on large loan and conduit transactions. For BBBs and BBs the conclusions are less clear and are highly sensitive to assumptions about volatility of property NOI.
W H Y H A V E S TA N D - A L O N E L A R G E L O A N S R E - E M E R G E D ?
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. 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 contains at least one loan of $25 million or more. Fusion deals accounted for about 12% of CMBS in 1998 and 1999. Some of the large dollar balance loans in fusion deals would qualify as stand-alone large loans and others resemble conduit loans.
96
Recently, the number of single-borrower large loan deals has increased. We attribute some of this to the reluctance of many originators to aggregate portfolios of loans over a long period of time. This unwillingness stems from losses on loans waiting for securitization suffered by many originators during the spread widening of 1998. For a single asset, the period from origination to securitization can be half that for a diversified pool of loans. According to Commercial Mortgage Alert, single asset transactions constituted nearly 20% of all CMBS issued in 1999, up from only 2.4% in the same period in 1998. 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 $1 billion, issuers received better prices by issuing separate large loan deals rather than tainting an entire pool with concentration risk.
exhibit 1
($Bn)
15 12 9
4.517
13.048 20.096
SHARES OF CMBS MARKET BY DEAL TYPE
7.765
Single Borrower Large Loan Fusion
11.425
1.924 8.757
12.331
4.322 4.319 3.046
0.410
1.703
3 0
3.922
1.502 1.544 2.616
4.984
3.041 1.249
1994
1995
1996
1997
1998
1999
Source: Morgan Stanley, Commercial Mortgage AlertSource: REIS, CB Commercial
exhibit 2
Office Hotel
PROPERTY TYPE
26.7
Multifamily
13.1
Other
3.5
13.9 40.1
2.7
Industrial
Retail
Source: Morgan Stanley, Intex, Commercial Mortgage Alert
97
Large Loans
Also, a new large loan structure emerged toward the end of last year. In the new structure, lenders split large loans into two separate notes. The senior portion of the loan (the A Note) is sold into a CMBS trust while the junior portion (the B Note) is placed privately outside of the trust. The market coined the term A-B structure to describe this splitting of loan interests. The A-B structure reduces some of the concentration risk of large loans in a diversified conduit deal, but more importantly, the portion of the large loan that is sold as CMBS has very low leverage. The reduced pool leverage results in lower subordination levels from the rating agencies. Finally, because of weakness in the REIT capital markets, REITs have turned to alternative funding sources. REITs had been the primary source of funding for large properties before last year. With equity and unsecured corporate debt financing becoming more difficult, REITs and other owners of high-value properties started turning to the CMBS market for funding. Many of these loans were bridge loans by large REITs awaiting a recovery in the equity markets. The rally in REIT share prices in the first half of 1999 was short-lived. A more sustained increase in share prices could result in a decline in demand from REITs for secured financing.
C H A R A C T E R I S T I C S O F S TA N D - A L O N E L A R G E L O A N S
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 70%. Conduit transactions typically have weighted average LTVs in the 70% to 80% 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.
98
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 borrowers 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. 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 borrowers equity stake is typically larger than for conduits and may in some cases reach over $100 million. According to a Moodys 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 Morgan Stanley research publication, Call Protection in CMBS, May 1998.
Information Availability
The best property-level information on a CMBS transaction is available at the time of issuance. For a stand-alone large loan, this data includes audited financials and detailed lease information. For smaller conduit loans, data may be unaudited and lease information in many cases is either non-existent or less comprehensive.
99
Large Loans
Although the situation has been improving, it is often difficult to obtain individual property cash flow and DSCR data after issuance. Some market participants have argued that data on stand-alone large loans will be more readily available than on standard conduit loans because of requirements in the loan securitization documents. Since most of the deals with stricter reporting requirements have not been outstanding for very long, it remains to be seen whether property cash flow information will be better on stand-alone large loan deals than on standard conduits. In any case, it will probably be easier to monitor information and performance on a single large property than a collection of 300 small to medium-sized properties.
EMPIRICAL STUDIES
Two recent studies addressed the relative risk of large loans. A default study by Esaki, LHeureux, 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.
R AT I N G A G E N C Y A P P R O A C H T O L A R G E L O A N S
In rating single-borrower or single-asset transactions, the largest concern of rating agencies is concentration risk. For example, Duff and Phelps states, The distinguishing characteristic of [single-asset] transactions is the concentration of riskThis 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 support 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
Howard Esaki, Stephen LHeureux, and Mark Snyderman, Commercial Mortgage Defaults: An Update, Real Estate Finance, Spring 1999. Barry Ziering and Willard McIntosh, Property Size and Risk: Why Bigger Is Not Always Better, Prudential Real Estate Investors Research, April 1999. The Rating of Commercial Mortgage Backed Securities, Duff and Phelps, November 1994.
100
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.
D O R AT I N G S C O R R E C T LY A D J U S T F O R R I S K ?
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. 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, especially when the product area is relatively new. To compare the credit risks of stand-alone large loan and conduit transactions, we apply our newly-developed CMBS credit model to analyze default-adjusted spreads, or yields to investors net of expected credit losses. Before we do that, we first turn to a discussion of segmentation in the CMBS investor base.
INVESTOR PREFERENCES
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.
101
Large Loans
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.
exhibit 3
INVESTOR BASE: CONDUITS VS. LARGE LOANS

Conduit
Investor Type
Insurance Pension
Large Loan
Investor Type
Pension
25%
15% 10%
Bank Insurance
20% 35% 10%

Bank
50%
35%
Money Manager
Money Manager
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. Traditional conduit CMBS investors seem willing to accept high loan concentrations in deals with an A-B structure and the B notes appeal to insurance companies. We believe that we will see increased use of the A-B structure in future securitizations.
102
In the next section, we evaluate whether the additional levels of subordination on stand-alone large loan transactions are enough to compensate investors for concentration risk and potential higher volatility of cash flows on these properties.
O P T I O N - A D J U S T E D S P R E A D A N A LY S I S
We used option adjusted spread (OAS) methodology to compare representative single borrower and large loan transactions to a typical diversified conduit deal. We found that, under reasonable assumptions, AAAs from large loans have good relative value to conduit AAA CMBS. For BBBs and BBs the conclusions are less clear and are highly sensitive to assumptions about volatility of property NOI. OAS models are used to value the default and prepayment options imbedded in credit sensitive callable securities. CMBS bonds are typically quoted as nominal spreads to the relevant US Treasury rate. Unlike the nominal spread, an OAS incorporates variations in cash flows from credit and prepayment events. The spread to the risk free rate that equates these option adjusted cashflows to the market price of the bond is the option-adjusted spread. The lack of historical data on commercial mortgage performance to derive a statistical model of commercial mortgage performance led us to develop a rulebased methodology for both defaults and prepayments. Two conditions must be met for a borrower to default: net operating income (NOI) is less than the scheduled mortgage payment for at least six months and the current loan to value (LTV) is greater than 110%. With respect to prepayments, there are minimum debt service coverage ratios and maximum loan to value ratios based on property type that the cash flow and property value must meet for a borrower to prepay. Additionally, the benefit of refinancing must be greater than the costs, which are inclusive of penalty points and/or yield maintenance and transactions costs. In comparing stand-alone large loan and single asset deals to conduit transactions, we evaluated deals with defeasance call protection. In the following examples, the OAS thus measures the impact of defaults and not prepayments. The key assumption in the model is NOI volatility. Both DSC and LTV tests are ultimately dependent on the projected levels of NOI. Future property values are calculated by dividing the projected NOI by a capitalization rate. The initial capitalization rate varies through time based on an assumed correlation between interest rates and capitalization rates. We used Morgan Stanley proprietary models to generate future values for interest rates. The three deals that we analyzed differ in terms of leverage, loan diversity and credit enhancement to different classes of bonds. The debt service coverage ratio is the highest for the large loan deal and lowest for the conduit while the LTV ratio is lowest for the single asset deal and highest for the conduit deal. At the AAA level, credit enhancement is the highest for the single asset deal and lowest for the conduit deal. At the BBB level, however, the conduit deal has the highest credit support and the single asset deal the lowest. Exhibit 6 provides statistics on each of the different deals.
103
Large Loans
Exhibit 4 shows cumulative default rates for each of the deals based on a given level of NOI volatility. At the same level of NOI volatility, the single asset deal has the lowest level of cumulative defaults and the conduit deal the highest level of defaults. This result is expected, as the single asset deal has the lowest leverage. The key question is what NOI volatility equates the returns of similarly rated securities backed by the different collateral types with different credit support levels.
exhibit 4
Cumulative Default Rate
40%
CUMULATIVE DEFAULT RATES
30%
Single Asset Large Loan Conduit Deal
20%
10%
0% 0 6 8 10 12 14 16 18 20
NOI Volatility

OAS (bp)
exhibit 5
136
10-YEAR AAA OAS
134 132 130 128 126 124 122 120 0 6 8 10 12 14 16 18 20

NOI Volatility
Exhibit 5 shows the OAS for the 10-year AAA bond from each of the deals at increasing levels of NOI volatility. Based on market pricing as of February, 2000, we assumed a nominal spread 10 bp wider for both the single asset and large loan deal than for the conduit deal. The resulting option-adjusted spreads show little option cost for any of the 10-year AAA bonds even at very high levels of NOI volatility.
104
exhibit 6
CHARACTERISTICS OF SINGLE-ASSET, LARGE LOAN, AND CONDUIT DEALS

Number of Loans: 1 DSCR: 1.85X LTV: 54% Bonds A1 A2 B C D Coupon 6.16 6.54 6.78 6.88 6.96 Avg Life 5.51 9.93 9.93 9.93 9.93 Rating AAA AAA AA A BBB Credit Enhancement 45.0% 45.5% 31.5% 20.0% 0.0%
Single Asset Deal
Large Loan Deal
Number of Loans: 11 DSCR: 1.90X LTV: 58% Bonds A1 A2 B C D E F Coupon 6.22 6.48 6.70 6.77 7.15 7.15 7.15 Avg Life 5.39 9.52 9.89 9.96 9.96 9.96 9.96 Rating AAA AAA AA A BBB BB B Credit Enhancement 27.50% 27.50% 21.00% 14.00% 7.75% 4.50% 1.50%
Conduit Deal
Number of Loans: 219 DSCR: 1.57X LTV: 67% Bonds A1 A2 B C D E F Coupon 6.34 6.54 6.63 6.77 7.07 6.34 6.34 Avg Life 5.42 9.72 9.96 10.00 11.47 15.95 19.42 Rating AAA AAA AA A BBB BB B Credit Enhancement 26.00% 26.00% 21.00% 16.50% 11.50% 5.25% 2.75%
A 12% NOI volatility results in an 18% cumulative default rate for the conduit deal. The 18% default rate is approximately the long-term average default rate found in life insurance companies commercial mortgage loan portfolios in the ELS default study. There is no default option cost for the 10-year AAA bond from the conduit deal at this level of NOI volatility. To reach equivalent AAA option adjusted spreads, the single asset deal and large loan deals would have to reach volatilities of 38% and 21%, respectively, or 3.2 to 1.8 times the assumed conduit NOI volatility.
105
Large Loans
exhibit 7
OAS (bp)
BBB OAS
300 200 100 0 (100) (200) (300) (400) 0 6 8 10 12 14 16 18 20

NOI Volatility
exhibit 8
OAS (bp)
600
BBB OAS
400 200 0 (200) (400) 0 6 8 10 12 14 16 18 20
Large Loan Conduit Deal
NOI Volatility
Moving down the credit curve to the BBB bonds shows a slightly different result. We assumed the same spread for the single asset, large loan deal, and the conduit deal, reflecting market pricing as of February, 2000. Unlike the 10-year AAA bond all of the deals show a decline in yield at the historical average cumulative default rate. Exhibit 7 shows the OAS for the BBB bonds at different levels of NOI volatility. The conduit deal BBB OAS declines more rapidly than both the single asset and large loan deal. NOI volatility of 12% for the conduit deal results in an option-adjusted spread of 87 bp. To reach an equivalent OAS for the large loan and single asset deal NOI volatility would have to be between 13%14% and 15%16%, respectively.
106
At the BB level, the conduit deal has a lower OAS than the large loan deal at low levels of NOI volatility and a higher OAS than the large loan deal at higher levels of NOI volatility. (The single asset deal does not have a BB bond). The difference between the breakeven NOI volatilities for large loan deals versus conduits is narrower for the BB bonds than for the BBB and AAA bonds. The BBB from the large loan deal withstands 1%2% more NOI volatility than the BBB from the conduit deal and achieves the same OAS. At the BB level, the difference in NOI volatility that results in the same OAS ranges from 0% to 1%. Exhibit 8 shows the OAS for the BB bonds as NOI volatility varies.
C O N C L U S I O N : R E L AT I V E V A L U E
Interpreting the OAS results calculated from our model is difficult in the absence of historical benchmarks for NOI and price volatility. However, it would appear that the wider nominal 10-year AAA spreads for single asset and large loan transactions are not based on differences in projected NOI volatility alone. There is so little default option cost for any of the 10-year AAAs that single asset and large loan AAAs appear to offer good relative value to conduit AAAs. We attribute current spread differences to a combination of liquidity concerns and the aversion of many money managers to what they perceive as real estate risk. For BBB bonds, the single asset and large loan deals can also withstand higher levels of NOI volatility than the conduit deal and still have the same OAS. For BBBs, however, the margin for error is much less than for AAAs. While BBBs and BBs from large loan deals remain attractive to real-estate savvy investors, we believe that their appeal will remain limited for money managers.
107
Multifamily MBS
FA N N I E M A E
In addition to private-label CMBS, the CMBS market also encompasses 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).
T H E FA N N I E M A E D U S P R O G R A M
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 OF DUS LOANS
Loans originated under the DUS program are generally either fixed-rate balloon mortgages with 5-, 7-, 10-, or 15-year terms or fixed-rate fully amortizing loans with 25- or 30-year terms. 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 Maes purchase.
P R E PAY M E N T P R O T E C T I O N
One of the main advantages of multifamily securities over residential MBS is the prepayment protection on multifamily loans. Most DUS loans have yield maintenance premiums in the event of an early prepayment. The premium is usually yield maintenance calculated at the relevant treasury rate, or Treasuries flat. Common yield maintenance terms are: Balloon Term (years) 5 7 10 15 30 Yield Maintenance Term (years) 3 or 4.5 5 or 6.5 7 or 9.5 10 10
108
After the yield maintenance period ends, the borrower is still required to pay a 1% premium on prepayment which is retained by Fannie Mae. This premium 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 passed through to the investor by Fannie Mae only to the extent they are received from the borrower. Fannie Maes obligation extends only to the outstanding principal balance of the security, i.e., if an MBS DUS defaults as a premium security, the investor receives a minimum of par but may lose some or all of the premium. Most DUS loans can be assumed by a new, and creditworthy, borrower on payment of a 1% assumption fee that is not passed through to the investor. Given that the pricing speed assumption of DUS is usually 0% CPR, 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 lenders share of the loss is limited to 20-40% of the UPB. The yield maintenance premium is also part of the loss computation giving the lender a vested interest in enforcing payment of the premium. 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

Minimum Debt Service Coverage Ratio 1.15 1.25 1.35 1.55 Maximum Loan-to-Value Ratio 80% 80% 65% 55%
Source: Fannie Mae
Tier Tier Tier Tier
1 2 3 4
Note: Most DUS loans are in Tier 2 or Tier 3.
109
Multifamily MBS
FA N N I E M A E D M B S
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 CHARACTERISTICS
DMBS are Fannie Maes 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.
($mm)
10,000
8,627
exhibit 2
DMBS ISSUANCE
8,000
6,234
6,000 4,000
2,215
2,000
681 45 58 143
0 1996 1997 1998 1999 2000 2001 2002 1
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.
FA N N I E M A E G U A R A N T E E
DMBS, like all other Fannie Mae securities, carry Fannie Maes 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 passu with other senior debt of Fannie Mae.
110
LOAN CHARACTERISTICS
While single-asset executions are available, DMBS are most often secured by a pool of cross-collateralized, cross-defaulted, first-lien mortgages on income-producing 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 terms 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 supporting 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
ALL VARIABLE RATE
Assume: $100MM 50% LTV Facility (50% LTV); $100MM DMBS Mortgages on Assets ($200MM value) Mortgages on Assets ($200MM value)
MULTIFAMILY CREDIT FACILITY
Lender Borrower
$99.5MM1
Fannie Mae
$100MM DMBS
$99.5MM1
$100MM DMBS
Investors
Assumes $99.5MM discounted proceeds from sale of DMBS
Source: Fannie Mae
111
Multifamily MBS
At maturity, the investor will receive $100 million, via proceeds from new DMBS issuance, or a payoff by the borrower.
P R E PAY M E N T S A N D D E FA U LT S
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 LEVELS
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 has 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. DMBS are currently trading 35 bp wider than agency discount notes, largely because of liquidity, and provide an attractive investment for money market buyers.
exhibit 4
Imputed Pass-through vs. LIBOR
0.15% 0.05% -0.05% -0.15% -0.25% -0.35% -0.45% -0.55% -0.65% Apr-99 Dec-99
DMBS TRADES: THREE-MONTH MATURITY
Base line represents comparable LIBOR as of security trade date.
Jun-00 Nov-00 Mar-01

Issue Dates
Jul-01
Nov-01
Jan-02
Source: Fannie Mae
INVESTOR BASE
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 Treasurys, in that they are permitted investments for Federally supervised institutions and for trusts and funds invested under the authority of the US. DMBS can also be purchased in unlimited amounts by national banks, federally chartered credit unions and federal savings and loans associations.
112
GINNIE MAE/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 Administration (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:
exhibit 5
CHARACTERISTICS OF GINNIE MAE AND FHA PROJECT LOANS

Ginnie Mae Explicit 100% 44 PTC Yes FHA Implicit 99% 54 Physical No
Government guarantee Principal payment in case of default Delay days Delivery Data on Bloomberg
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
See inside MBS & ABS, May 1, 1997, p.3
113
Multifamily MBS
exhibit 6 Issuer Issue Type Underlying Mortgages Pool Types
A PROFILE OF GINNIE MAE MULTIFAMILY MBS

Ginnie Mae approved mortgage lender GNMA I FHA Insured multifamily mortgages 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 $1M 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)
Securities Interest Rate Guaranty Guarantor Principal and Interest Maturity Minimum Certificate Size Transfer Agent
Source: Reprinted from Ginnie Mae website, www.ginniemae.gov, 1997
FREDDIE MAC
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.
PROGRAM PLUS
Freddie Macs Program Plus is similar to Fannie Maes 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: Terms 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 terms of the loans are: Term(years)/Yield Maintenance(years) 7/6.5 10/9.5 15/14 20/15
114
115
CMBS IOs
INTRODUCTION
The CMBS interest-only securities (IO) market has both grown and matured since its inception in the mid-1990s. Unfortunately, the derivative taint of the name IO has caused some buyers to shy away from the sector. 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% CPR pricing assumption or if loans are extended at the balloon date. CMBS IO investors may benefit from prepayment penalties during yield maintenance periods.
W H AT I S A C M B S 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% par-priced securities. Both residential and CMBS IOs are typically rated AAA/Aaa. These ratings are based on priority of cashflow 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 IOs 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 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.
116
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 9% to 28% over ten-year periods between 1972 and 1997.
STRUCTURE
A traditional structure for a CMBS IO is shown in exhibit 1. Note that most of the IOs 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
WAC-AAA Coupon WACAA WACA WACBBB WACBB WACB/NR
TYPICAL IO STRUCTURE
AAA Classes
AA
BBB
BB
B/NR
Fixed Bond Coupons

Class X Coupon Strips
I N V E S T O R B A S E 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 instrumentsthe duration of a CMBS IO is about 4 years, compared to 7 years for classes rated BBB or BB.
C M B S S P R E A D S A N D P R E PAY M E N T P R I C I N G A S S U M P T I O N S
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.
117
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% CPY assumption, meaning that all loans prepay immediately after the lockout or yield maintenance period. Some investors also examine yields under a 0% CPR1 assumption. The difference in spreads using the 0% and 100% CPR 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 passed 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% CPR assumptions are unrealistic. Actual prepayment speeds will vary on a monthly basis, with several months of 0% CPR 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 todays 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.
D E FA U LT A S S U M P T I O N S
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 CPR 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.
E S A K I , L H E U R E U X , A N D S N Y D E R M A N : A N E W D E FA U LT S T U D Y
Our default assumptions are based on a study recently completed by Esaki, LHeureux, and Snyderman (ELS) of default rates on commercial mortgages held by life insurance companies between 1972 and 1997.2 Therefore, the standard stress scenario run on CMBS IO is a 2% CDR. The study showed that over this 261
CPR stands for conditional prepayment rate. It is an annualized prepayment rate expressed as a percentage of the remaining balance of the pool. See Commercial Mortgage Defaults: An Update, Howard Esaki, Steven LHeureux, and Mark Snyderman, Real Estate Finance, forthcoming.
118
year period, lifetime cumulative default rates on commercial mortgage loans originated by insurance companies ranged from a low of 9% for 1977 originations to a high of about 28% for the 1986 cohort. These cumulative default rates translate into annual CDRs of just over 1% and about 4%, respectively, assuming 10-year collateral and average 2%. Defaults, however, do not occur at an even rate over the life of a mortgage pool. The typical pattern is very low default rates shortly after origination, rising to a peak in years 3 to 7 after origination, and then tailing off thereafter. The magnitude of the peak default rate varies based on the real estate environment after the loans are originated.
I M PA C T O F P R E PAY M E N T S O N I O Y I E L D
Exhibit 2 shows the effect of increasing prepayments on 5 different CMBS IOs.
exhibit 2
PREPAYMENT RISK:YTM OF CONDUIT IOs IN PERCENT

CPR 0 10.63 8.83 10.36 9.95 9.27 CPR 20 9.97 8.82 9.85 9.50 9.18 CPR 50 9.48 8.80 9.46 9.19 9.08 CPR 100 8.59 8.68 8.59 8.68 8.67
DMARC 98-C1 NASC 98-D6 LBCMT 98-C1 FULBA 98-C2 MSC 98-WF2
Priced at +350 at 100 CPR; yield maintenance equals lockout
The current pricing convention for CMBS IOs, however, is 100% CPR 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% CPR rather than 100% CPR. 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 terms of the loans in a given transaction.
119
CMBS IOs
Y I E L D M A I N T E N A N C E P E N A LT I E S A N D 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 =
(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. 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 %)

CPR 0 10.63 8.83 10.36 9.95 9.26 CPR 20 13.51 8.82 11.44 10.72 10.85 CPR 50 15.42 8.80 12.48 11.37 11.63 CPR 100 17.66 8.68 14.31 12.10 13.48
Priced at +350 at 100 CPR; 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% CPR to 100% CPR ranges from 15 bp to more than 1500 bp.
120
exhibit 4
PREPAYMENT RISK: YTM OF RECENT CONDUIT IOs: INTEREST RATES UP 300 BP (IN %)
CPR 0 10.63 8.83 10.36 9.95 9.26 CPR 20 6.89 8.82 6.13 8.20 5.22 CPR 50 3.32 8.80 1.48 6.96 1.41 CPR 100 -1.03 8.68 -5.15 5.51 -2.31
Treasury rates up 300 bp; priced at +350 at 100 CPR; 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. 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 cashflow 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.
PA C & C O M PA N I O N 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 MM in IO would have been created on a $1 billion transaction. If 10 year Treasuries rallied just 25 basis points after the mortgage pool was set, but prior to pricing the bonds, the mortgage pool WAC would be 65 basis points greater than the bond WAC, resulting in approximately $43 MM in IO proceeds. This incremental increase in IO proceeds resulted in the development of the PAC IO / Companion 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 Companion IO.
Some CMBS transactions in the early & mid 1990s had multiple IO classes, with individual IOs stripped off of individual bond classes.
121
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 Companion IO is stripped off of the more senior and subordinate classes (See Exhibit 5). Recall, given the underlying call protection of todays 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 Companion IO has marginally more exposure to defaults in the mortgage pool. In that regard, the Companion IO is also referred to as the Levered IO as its yield profile has more exposure to defaults in the underlying mortgage pool. In recent transactions, the PAC IO has priced at T+145. This is a AAA rated security that can survive a 6 CDR stress scenario before giving up any yield. This stress is 3 times the standard commercial mortgage default rate per the ELS Study. The Levered IO is currently priced at T+475 and 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) resulting in a yield give up or slope of approximately 260 basis points. While the Levered IO gives yield in a stressed environment, the investor is still buying a loss adjusted AAA security at T+215.
exhibit 5
SERIES 2001 - TOP 4 CAPITAL STRUCTURE

Class X- 1, X -2
Class A- 1 Class A- 2 Class A- 3 Class B Class C Class D Class E Class F Class G -N NR = N ot rated
Aa a/AAA 5.06% Aa a/AAA 1-mo. LIBOR + 0 .45% Aa a/AAA 5.61% Aa 2/AA 5.80% A2 /A 6.05% A3 /A6.17% Ba a2/BB B 6.47% Ba a3/BB B6.91% Ba 1/BB+ to NR 6.00% C lass X -1 C lass X -2 through N ov. 2008
$389.2M M $60.0M M $320.2M M $24.8M M $24.8M M $9.0M M $20.3M M $9.0M M $45.1M M C lass X -1 after N ov. 2008
122
123
Commercial Mortgage Defaults

This chapter was co-authored by Mark Snyderman and Steven LHeureux and appeared in a slightly different form in Real Estate Finance, Spring 1999.
INTRODUCTION
We think that the best method to quantify the credit risks of a real estate portfolio is to analyze the historical performance of pools of commercial mortgage loans over their lifetime. Mark P. Snyderman has completed two pioneering studies of lifetime commercial mortgage performance. These studies tracked the lifetime defaults of commercial mortgages held by life insurance companies.1 His most recent article tracked defaults on 8 large insurance companies through 1991. The companies originated the loans between 1972 and 1986. Using the same insurance companies and data sources as the original studies, we updated the default data through 1997 for originations from 1972 to 1992. We examined the credit performance of over 15,000 individual loans, an increase of about 4,000 loans from the 1994 study. Preliminary results of an extension of the study through 2000 show no change in the major conclusions of this study. The main results of the updated study are: The average cumulative default rate for loans originated in 1987 or earlier (that is, with at least 10 years of seasoning) was 18.1% of the original pool balance. This average is slightly higher than in previous studies, but in line with earlier projections. The cumulative default rate of 28% on loans originated in 1986 was the highest for any cohort. The lowest cumulative default rate was 9% for loans originated in 1982. The severity of loss on liquidated loans averaged about 37.7%, slightly higher than the 36% reported in the 1991 study. The severity on loans liquidated between 1992 and 1997 was 44%. Annual defaults rose to about 2% in years 3 through 7 after origination, and then fell off for the remaining life. Defaults did not increase in balloon years. About half of the defaulting loans were liquidated and half restructured. The number of liquidated loans rose to about 60% in the 19921997 period. Most investment grade CMBS, as typically structured today, do not suffer a principal loss when subjected to the default rates of even the worst origination cohort since 1972. The study encompasses a period covering one of the worst real estate depressions in the past century.
Commercial Mortgages: Default Occurrence and Estimated Yield Impact, Journal of Portfolio Management, Fall 1991 and Update on Commercial Mortgage Defaults, Real Estate Finance Journal, Summer 1994.
124
N U M B E R O F L O A N S B Y O R I G I N AT O R A N D Y E A R
We tracked commercial mortgage default rates from life insurance company annual statements from 1992 to 1997, aggregating our data with the two previous Snyderman studies.2 The insurance companies and number of loans tracked are shown in exhibit 1.
exhibit 1
NUMBER OF LOANS BY ORIGINATOR, 1972-1992

Originator Aetna Life Insurance Company Connecticut Mutual Life Insurance Company1 Equitable Life Insurance Company John Hancock Mutual Life Insurance Company New England Mutual Life Insurance Company* The Northwestern Mutual Life Insurance Company The Prudential Insurance Company of America The Travelers Insurance Company Total Number of Loans 2,974 863 1,768 1,714 1,338 719 3,522 2,211 15,109
*Loans were tracked at the new firm. Source: Morgan Stanley
Merged into other firms since 1991.
exhibit 2
1,400 1,200 1,000 800 600 400 200 0 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
NUMBER OF LOANS BY ORIGINATION YEAR
Loan originations in the study peaked in 1978-1979 and 1985-1988, and reached lows in 1982 and 1992. Of the 15,109 total loans originated, 2663 (17.6%), had defaulted by 1997.3
As in the earlier studies, the sample comes from publicly available life insurance company annual statements that are filed with state insurance regulatory offices. If a loan is more than 90 days delinquent, it is counted as defaulted. For more details on the data, see Snyderman (1991). While insurance company loans are very diverse geographically and by property type, they may differ in some respects from the entire universe of commercial mortgages. For example, in the period we studied, insurance company portfolios tended to have a higher concentration of office properties than all commercial mortgages and included a higher concentration of large trophy properties. For more details on insurance company originations, see Snyderman (1994).
125

SIZE OF LOANS
The median loan size was about $4 million. About 75% of the loans were less than $8 million. The $4 million to $8 million loan category had the highest default rate, followed closely by the greater than $8 million category. The smallest loans had the lowest default rate.
GEOGRAPHIC DISTRIBUTION
exhibit 3 LOANS ORIGINATED
AND DEFAULT RATES BY PRINCIPAL AMOUNT

Total Default Rate(%) 13.5 17.4 20.1 19.8 18.1
The loans in the study are well diversified geographically, with the largest percentages in the West (24%), Northeast (22%), and South Central (19%). Within a given origination cohort, cumulative lifetime defaults have varied widely by geographic region. For example in the 1980 cohort, loans originated in the Northeast had a lifetime default rate of 6.0%, while loans on properties in the South Central region had a cumulative default rate of 31.0%. For the period 1972-1992, the South Central region had the highest average cumulative default rate, while Canada and the West Coast had the lowest average default rates. The higher default rates in the South Central region reflect the deep oil-state recession of the 1980s.
Loan Amount Loans (millions of $) Originated 02 3,842 24 3,949 48 3,459 >8 3,859 Total 15,109
exhibit 4 NUMBER OF LOANS
ORIGINATED BY GEOGRAPHIC REGION

Number of Loans 3,679 2,928 3,295 2,113 2,309 785 15,109 Percent of Total 24.3 19.4 21.8 14.0 15.3 0.5 100.0
West Coast South Central Northeast Mid-Central Southeast Canada/Other Total
exhibit 5 AVERAGE DEFAULT
RATE BY GEOGRAPHIC REGION

Default Rate 12.3 27.4 15.1 17.3 19.7 11.7
West Coast South Central Northeast Mid-Central Southeast Canada/Other

126
exhibit 6
30 25 20 15 10 5 0
LIFETIME DEFAULT RATES BY ORIGINATION COHORT
72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
exhibit 7
2.5
AVERAGE TIMING OF DEFAULTS (AS A PERCENT OF ORIGINAL BALANCE)
2.0
1.5
1.0
0.5
0.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Year Since Origination
127

D E FA U LT S B Y O R I G I N AT I O N C O H O R T
Cumulative lifetime default rates for cohorts with at least ten years of history ranged from 9.2% for 1977 originations to 27.6% for 1986 originations. The average default rate for cohorts with a minimum of ten years seasoning was 18.1%.
SEVERITY OF LOSS
The average severity of loss on liquidated loans for the entire 1972 though 1997 period was 37.7%, slightly higher than the 36% severity reported in the 1991 Snyderman study. The severity calculation includes foregone interest and expenses, as well as lost principal. The weighted average severity of loss on loans that were foreclosed and subsequently sold in the period 1992 through 1997 was 43.8%. The higher severity in this period reflects the steep downturn in real estate prices in the early 1990s. For individual insurance companies, the weighted average severity rates ranged from 26% to 72% in the 1992-1997 period. For the period 1992-1997, about 60% the loans entering foreclosure were liquidated, compared to 46% in the 1972-1991 period. There is no data available on the non-liquidated loans, which presumably were restructured by the lender.
T I M I N G O F D E FA U LT S
Averaged over all cohorts, commercial mortgage defaults rise over the first 3 years after origination to about 2% of original balances, remain near 2% for years 3 through 7, and then fall off over the remaining life of the cohort. Although we do not have data on the amortization schedules of the loans in the study, there does not appear to be any increase in default rates at balloon dates. The default rate curve shown in exhibit 7 masks individual cohort patterns, which can vary substantially from the average. Default timing for individual origination cohorts varies depending on the state of the commercial real estate market at the time of origination and the subsequent years. For example, the 1973 cohort had the highest default rates in year 3 after origination (nearly 5%), as this cohort was originated near the peak of a real estate cycle followed by a sharp downturn in real estate prices. Not a single cohort, however, had a peak in default rates in years 7, 10, or 15, which are typically balloon payment dates. (See the Appendix for yearly default data for each origination cohort.) Many of the loans originated in the 1970s and 1980s, however, were fully amortizing and not balloon mortgages.
128
C O M PA R I S O N T O P R E V I O U S S T U D I E S
The new study extends from 1991 to 1997 the two previous commercial mortgage default studies by Snyderman. These studies covered 7,205 loans and 10,955 loans, respectively, while the current study increases the coverage to over 15,000 loans. The methodology and data source for all three studies was the same. The new study incorporates a period of sharp decline in commercial real estate markets. The average default rate for all cohorts rose from 13.8% in the second Snyderman study to 16.2% in the current study. For cohorts with at least ten years of data, the average default rate in the new study was 18.1%, with a range of 9% to 23%. The previous study did not report the average default rate for cohorts with a minimum of ten years of seasoning. Although the new study shows higher default rates than previous studies, the new average for all cohorts with a minimum of ten years of history is very close to Snydermans projected default rates made for these cohorts before the new study.
I M PA C T O F D E FA U LT S O N B O N D S T R U C T U R E S
The new study found that the worst origination cohort since 1972 had a cumulative default rate of 27.6%. Applying a 37.7% severity rate to this default rate gives an estimated cumulative loss of 10.4%. The current subordination levels for BBB CMBS range from 8% to 12%, implying that most investment grade CMBS would withstand the equivalent of the worst real estate downturn of the post-World War II era. The estimated 10.4% loss rate is very conservative, since it implies that all defaulting loans are liquidated. In reality, we conservatively estimate that only about 50% of defaulting loans are liquidated, and the rest are restructured. In the insurance company data, we are able to ascertain the severity of loss only on liquidated loans, and not restructured loans. If, as in the earlier studies, we assume that the severity on restructured loans is half that on liquidated loans, the average severity rate on defaulting loans drops to 28.2%. Applying this severity to the highest default cohort gives a cumulative loss of 7.8%, well below the BBB subordination level. While estimated default and loss rates should leave the principal of investment grade CMBS untouched, they can have an impact on non-investment grade securities. Applying the estimated severity of 28.2% to the average cohort gives a cumulative loss of 4.4%. The best cohort has an estimated loss of 2.6%. In comparison, single-B and BB CMBS subordination levels currently average about 3% and 6%, respectively.
129

CONCLUSION
Our update of the two earlier commercial mortgage default studies authored by Mark P. Snyderman includes the effects of the commercial real estate recession of the early 1990s. The results of the current paper raise the average expected cumulative default rate of an origination cohort, but do not significantly change expectations of severity of loss on liquidated loans. In applying the results of this study to current CMBS collateral, analysts should be careful to note the potential differences between insurance company and mortgage conduit originations. Loan size, property concentrations, LTVs, debt service coverage, and geographic distribution of a given conduits originations may vary significantly from the life insurance company average. In addition, the procedures taken by a life insurance company on problem loans may differ from a CMBS servicer. For example, life insurance companies generally operate under regulatory constraints that give preference in terms of capital charges to restructured loans rather than foreclosed loans, which are treated as real estate equity. Finally, we have yet to fully evaluate the impact of the growing public nature of commercial real estate finance on real estate credit cycles. Some analysts believe that the growth of the CMBS and REIT markets will dampen the traditional boom/bust cycle of commercial real estate. If these analysts are correct, future updates of commercial default studies may show less volatility in commercial mortgage default rates.
130
appendix
TIMING OF DEFAULTS BY COHORT

6 0.19 0.00 0.14 0.39 0.21 0.14 0.09 0.18 2.25 2.00 0.00 0.78 3.71 3.28 8.46 3.65 2.75 2.66 1.09 0.48 0.14 1.62 0.35 1.28 1.28 0.29 2.08 1.04 0.31 0.73 2.26 1.28 0.49 1.08 0.29 1.68 1.12 0.32 0.48 0.16 0.16 1.77 2.30 1.59 0.53 0.35 0.00 0.00 0.47 1.88 0.78 1.10 0.16 0.16 0.00 0.00 1.38 3.67 0.00 0.92 0.92 0.00 0.46 0.00 0.00 2.49 2.49 1.25 1.75 2.24 0.75 0.25 0.00 0.50 0.00 1.54 2.01 1.54 2.01 2.01 1.42 0.83 0.12 0.24 0.36 0.24 2.13 1.07 1.96 1.16 1.16 1.51 1.42 0.80 0.36 0.18 0.36 0.00 0.66 1.80 1.14 1.52 0.85 1.42 0.66 0.95 0.95 0.66 0.09 0.00 0.00 0.71 0.56 1.55 0.99 0.56 0.14 0.56 0.42 0.56 0.85 0.00 0.28 0.14 0.00 0.00 0.00 1.03 0.62 1.23 0.41 0.21 0.21 1.85 0.82 0.41 0.62 0.62 0.21 0.00 0.00 0.39 0.39 0.00 1.17 1.36 0.58 0.78 0.78 0.78 0.58 0.39 0.00 0.00 0.00 0.00 0.00 0.28 0.28 0.14 0.14 0.42 0.71 0.42 0.71 0.71 0.28 0.28 0.14 0.00 0.00 0.00 0.00 0.14 0.14 0.14 0.14 0.00 0.82 0.95 0.54 0.95 0.68 0.14 0.27 0.82 0.27 0.27 0.00 0.14 0.00 0.38 0.00 0.00 0.00 0.19 0.19 0.38 0.75 0.94 0.56 0.19 0.19 0.75 0.38 0.00 0.19 0.00 0.00 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Years Since Origination
AVG. 0.38 1.34 1.89 2.03 1.97 2.12 1.89 1.13 1.01 0.68 0.67 0.50 0.46 0.36 0.26 0.34 0.26 0.11 0.08 0.10 0.03 0.01 0.01 0.01 0.00
1972
0.00
2.44
3.75
3.56
3.19
1.31
1973
0.14
2.99
4.76
3.40
1.63
0.54
1974
0.71
5.67
2.97
2.27
0.57
0.71
1975
2.33
2.33
1.94
0.58
0.58
0.58
1976
0.21
0.82
0.41
1.23
0.21
0.00
1977
0.00
0.85
0.14
0.28
0.14
0.28
1978
0.28
0.19
0.09
0.00
0.38
0.00
1979
0.27
0.18
0.09
0.62
0.18
0.36
1980
0.24
0.47
0.24
0.00
0.59
1.30
1981
0.75
0.00
1.00
0.50
0.75
2.74
1982
1.38
0.92
2.29
2.75
5.05
1.83
1983
1.26
0.31
1.10
4.40
2.51
3.30
1984
0.00
1.41
3.18
1.94
4.59
2.12
1985
0.08
2.40
2.08
2.48
2.56
3.68
1986
0.00
1.18
2.16
2.06
3.44
4.92
1987
0.21
0.31
1.77
2.71
4.27
5.63
1988
0.10
0.29
1.87
4.72
4.13
4.52
1989
0.35
0.92
3.12
3.35
2.54
2.08
1990
0.41
3.13
4.63
2.18
2.18
3.00
1991
0.72
1.20
1.68
1.68
1.20
0.24
1992
0.93
0.31
1.87
1.25
0.00
0.00
131
European CMBS
KEY POINTS
EUROPEAN CMBS RELATIVE VALUE European CMBS are attractively priced against relevant benchmarks. We anticipate spread tightening as investor sponsorship grows. EUROPEAN COMMERCIAL MORTGAGE SECURITIZATION We forecast issuance of European CMBS will rise markedly due to the attraction of the CMBS product to both issuers and investors. EUROPEAN COMMERCIAL REAL ESTATE Fundamental measures of European real estate markets all indicate that the markets are in good health.
INTRODUCTION
Although a significant number of securitization transactions can claim to include real estate components, we have adopted a relatively narrow interpretation of what constitutes CMBS transactions. For the purposes of this report our definition includes: Debt issuance sponsored by real estate companies, backed by leased real estate. Debt issuance by real estate conduits, backed by loans to real estate borrowers which are in turn backed by real estate assets subject to lease. Debt issuance by banks, backed by portfolios of bank loans which are in turn backed by real estate which is subject to lease. Debt issuance where proceeds are used to acquire real estate from corporate sellers to whom the real estate is then leased back for an extended period. These transactions rely on corporate credit covenants backed by real estate assets. This categorization results in the omission of, for example, operating company transactions which have an element of real estate backing, such as pub securitizations, as well as commercial real estate-related non-performing loans. Additionally, although a number of nursing home transactions rely on property leases to nursing home operators, the analytical approach employed has tended to rely on a detailed understanding of the economics of the individual underlying nursing home businesses. This contrasts with a more traditional real estate securitization analysis which focuses on the credit quality of the lessee at a somewhat higher level and sensitizes the transaction with allowance for lease breakage, vacancy rates and replacement lease rates. For these reasons we classify these nursing home issues as operating company transactions.
132
R E L AT I V E V A L U E C O N S I D E R AT I O N S
The following charts show the relative spread performance of CMBS compared to equivalently rated corporate and supranational debt.
exhibit 1
bp 200 180 160 140 120 CMBS 100 80 60 40 Jan-00 Supranationa
FIXED AAA CMBS TRADE WIDER THAN FIXED SUPRANATIONAL BENCHMARKS
Mar-00
May-00
J ul-00
Sep-00
Nov -00
Jan-01
Mar-01
Broadgate A2
Canary Wharf A
Trafford Centre
EIB 12/28
EIB 04/14
We have mapped the Broadgate A2, Canary Wharf I and Trafford Centre AAA bonds, which have average lives of 19, 16 and 25 years, respectively, against the EIB 04/14 and 12/28 issues. The CMBS issues have recently been offering a spread pick-up which has tended to narrow from 50-60 bp to 30-40 bp over the last 6 months against the similarly rated and comparable average life supranationals.
exhibit 2
bp 350 300 250 200 150 100 50 0 -50 -100 Jan-00
AAA CMBS COMPARED TO SUPRANATIONAL AND REAL ESTATE COMPANY DEBT
BBB Prop. Co s
AAA CM BS Supranational
Ma r-00
May-00
J ul-00
Sep-00
Nov-00
J an-01
Ma r-01
Broadgate EI B 6/28 Slough BB B 09/15
Canary Wharf A EI B 4/14 British Land BB B+ 09/23
Trafford Centre ELOC AAA
Note: Fixed Rate Spreads (all except ELoC which is a floating rate issue) are shown on a swapped to LIBOR basis Source: Morgan Stanley
In exhibit 2 we show the bonds in exhibit 1 swapped to LIBOR against the ELoC AAA floater and two BBB swapped bonds issued by two leading UK real estate companies. The AAA CMBS have generally been trading in a range of 20-40 bp over LIBOR over the last 6 months.
133
European CMBS
exhibit 3
bp 60 50 40 30 20 10 0 -10 -20 -30 Jan-00
AAA CMBS SWAPPED TO EURIBOR
Mar-00
May-00
Jul-00
Sep-00
Nov-00 Trafford Centre
Jan-01
Mar-01 ELOC
Broadgate
Canary Wharf A
Note: Broadgate, Canary Wharf and Trafford are Fixed Rate issues: ELoC is floating rate. Source: Morgan Stanley
Exhibit 3 shows the selected fixed and floating AAA CMBS swapped to Euribor. European CMBS provide yield spread advantages over comparable assets. Whilst this differential has narrowed with increasing investor sponsorship of the sector, European CMBS remains an attractive asset class. We believe spreads will tighten further as liquidity improves, with growing supply and investor participation.
CMBS CREDIT RISKS
Although the previous section of this survey suggested that there is relatively attractive pricing available to investors in CMBS, investors need to understand whether this arises as a result of materially higher credit risks. This section suggests that investors in CMBS arguably face less risks than in unsecured corporate debt and are rewarded for participating in a sector that has yet to reach full maturity and investor sponsorship. Although no hard data is available on the European market at this point due to its developing state, there has been a large scale study of US CMBS performance conducted by Fitch. Key data on the Fitch study is presented in exhibit 4. Average Annual Default Rate %
Corporate Bonds CMBS
Investment grade Non-investment grade All Bonds
0.08 3.07 1.51
0.06 0.14 0.07
The reported performance history broadly indicates that investment grade CMBS performance was ahead of corporate bond investment grade performance (although we have no data on final recoveries on defaulted CMBS, which may possibly have been superior to corporate debt in view of the formers collateralized status).
134
exhibit 4
FITCH CMBS DEFAULT STUDY: 31 DEC 1990 TO 31 DEC 1999

Initial Cumulative Amount Cumulative Issuance Number default by defaulted default rate by ($bn) defaulted number (%) ($m) amount (%)
Number of Rated Rating Bonds
Investment grade Non-investment grade AAA AA A BBB BB B CCC

Total
Source: Fitch
2,582 893 902 585 462 633 449 407 37

3,475
211.6 20.4 139.4 34.6 18.0 19.6 11.9 7.5 0.9

231.9
9 3 0 5 2 2 1 2 0
12
0.35 0.34 0 0.85 0.43 0.32 0.22 0.49 0

0.35
419.9 83.3 0 314.4 63.8 41.1 42.6 41.3 0

503.2
0.20 0.41 0 0.91 0.35 0.21 0.36 0.55 0

0.22
The most significant variation in performance is in the non-investment grade area, in which non-investment grade CMBS have displayed a materially superior performance to similarly rated corporate debt. The reasons attributed for this outperformance have been mainly related to the perception that CMBS issues, particularly as they became larger, increasingly benefited from underlying diversification, and managed to avoid undue concentrations to particularly high risk real estate lending sectors, such as retail, supermarkets, drugstores, gaming, lodging and restaurants. There is additional US evidence available supporting the view that CMBS default performance is relatively favorable. In early 1999 a study by Esaki, LHeureux and Snyderman was produced detailing defaults on a sample of 15,000 commercial mortgage loans originated between 1972 and 1997 by US life insurance companies that have historically been major lenders to commercial real estate in the US The findings indicated that, for loans with 10 or more years of seasoning, 18.21% of the original pool balance had defaulted, with the severity of loss on liquidated loans averaging 37.7% and therefore resulting in an average loss of 6.82%. Interestingly, this loss level compares to credit enhancement of 12-15% that typically accompanies BBB rated CMBS. Therefore, credit enhancement structures of CMBS should, on average, comfortably absorb average losses on commercial mortgage loans at the BBB level. In fact the study went one step further and calculated that the worst-performing origination cohort in the study (1986) would result in estimated cumulative losses of 10.4%, again well within the 12-15% credit enhancement typically observed in CMBS structures at the BBB level.
See Commercial Mortgage Defaults: An Update in Real Estate Finance (Spring 1999)
135
European CMBS
HISTORIC ISSUANCE
Based on the definitions set out in the introduction, we find that there have to date been very few true CMBS transactions completed in Europe and the UK has provided the highest share of transactions. We find that, according to our categorization, only 30 European CMBS transactions were completed from the beginning of 1995 through to the end of 2000. Appendix 1 provides a listing with key economic features of the transactions that have been completed to date. Exhibit 5 shows historic European CMBS issuance over the period 1996-2000 and contrasts this with US CMBS issuance over the same period.
exhibit 5
e bn
90 80 70 60 50 40 30 20 10
0.3 2.3 0.2 31.6 44.4
85.4
E urope U.S.
US VS EUROPEAN CMBS ISSUANCE, 19962000
64.3 54.4
4.2
5.9
0 1996 1997 1998 1999 2000
The European market is currently just a fraction of the US market, which originally developed in the early 1990s. The US market developed partly as a method of resolving the asset disposition issues raised by the savings and loan crisis, and also as a means of providing new sources of funding following the drastic reduction in the appetite of banks and insurance companies to lend to the real estate sector. A further eqe 3 bn of European CMBS transactions have either been completed, or are in the course of completion, through Q1 2001, via the following transactions: Canary Wharf Finance PLC 120m (tap of original November 1997 issuance) ELoC No. 5 PLC 524.9m Europa Two Limited
e1,500m
Silver No. 1 eq 258m
136
exhibit 6
Continent 22%
EUROPEAN CMBS: SPLIT BETWEEN CONTINENTAL AND UK TRANSACTIONS, 19962000

UK 78%
Exhibit 6 shows for the European CMBS market the relative split of issuance since the beginning of 1996 between UK and continental transactions (UK transactions are defined as involving both UK originators and also primarily UK domiciled collateral). We believe there are 3 primary reasons for the dominance of U.K. transactions within total European issuance. The U.K. tends to follow on more closely to US-originated financial innovation than countries on the continent. The prevalence of long-term, fully repairing and insuring lease structures in the U.K., together with the relatively creditor-friendly nature of the U.K. insolvency regime, are important factors that facilitate securitization. The relative significance of quoted real estate investment companies in the U.K. is a further differentiating factor, and this sector has been the source of a significant number of transactions.
exhibit 7
Corporate/ Other 22% Conduit 16%
ORIGINATION SOURCE OF EUROPEAN CMBS ISSUANCE, 19962000

Bank 19% Real Estate Co 43%
Exhibit 7 breaks out the origination or sponsoring source of historic issuance since the beginning of 1996.
The differences between UK and European leases are set out in Appendix 2.
137
European CMBS
The relative dominance of real estate company transactions reflects the recent needs of the quoted real estate company sector in the U.K. in requiring substantial, long-dated financing against specific developments that have been more easily accommodated via securitization than by the domestic debenture or bank markets. By contrast, in the US in 2000, 45% of all issuance was from conduits.
CMBS RISK AND TRANSACTION STRUCTURES
From exhibit 6 it is clear that the majority of European transactions that have been completed to date involve U.K. assets. We have set below an outline of 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. We show in Appendix 2 the main differences in landlord and tenant relationships between the main European centers.
exhibit 8
TYPICAL CMBS ISSUANCE STRUCTURE
Bank
Bank
Liquidity facility
Interest Rate Hedge
Bonds/Collateral
Issuing SPV
Cash
Bondholders
Loan Collateral
Loan
Collateral
Borrower A
Cross-Collateralisation (Potential)
Borrower B
Lease
Rentals
Lease
Rentals
Lessee 1
Lessee 2
138
TRANSACTION STRUCTURES
Exhibit 8 sets out in a simplified schematic form some of the key relationships that commonly feature in CMBS transactions. The following table highlights the common structural features of transactions.
exhibit 9
KEY STRUCTURAL FEATURES OF CMBS TRANSACTIONS
Legal Structure
Exhibit 8 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-users. 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.
Collateral Single asset/corporate transactions
These transactions will feature collateral pledges from Borrowers A and B to the issuer over the benefit of their interests (freehold and leasehold) in the underlying properties and leases, and all of the cashflows therefrom. Issuers will in turn pledge to the security trustee for bondholders their interest in the collateral pledged by borrowers via the pledge of the secured loans made to the individual borrowers.
Conduit transactions
Similar to single asset transactions, conduit transactions will feature pledges from Borrowers A and B to the issuer over their interests (freehold and leasehold) in the underlying properties and leases, and all of the cashflows there
139
European CMBS
exhibit 9
continued
from. Issuers will in turn pledge to the security trustee for bondholders their interest in the collateral pledged by borrowers via the pledge of the secured loans made to the individual borrowers. Fixed and floating charges over their assets are generally granted by borrowers to the security trustee of the securitization, via the issuer, which should enable the trustee to control any insolvency of a borrower through the appointment of an administrative receiver to deal with the secured assets and minimize any potential delay to cashflow. The exercise of, specifically, the floating charge, and the appointment of an administrative receiver to avoid a potentially lengthy court process in insolvency is a particular feature of the collateral structure under English law.
Cross-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 terms 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.
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 Support
Although legal structures of UK transactions are arranged in the expectation that underlying loans will not become embroiled in extended insolvency proceedings, liquidity facilities from highly-rated banks provide support for timely payment of interest and principal in the event of borrower insolvency or temporary disruptions to cashflows 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 support junior classes.
140
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 cashflows are by their nature fixed flows and cashflows 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 accounts to avoid commingling and associated corporate bankruptcy risks where applicable.
Reserve Accounts
Reserve accounts 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. 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 10
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 cashflow losses arising from both the time taken to re-lease the property and the associated potential cashflow 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
141
European CMBS
exhibit 10
continued
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 terms at the outset - in excess of 15 years - are a strong support to a transaction as they improve the relative assuredness of cashflows. Similarly, the existence of average remaining lease terms 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 account 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. 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-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 re-lettings, 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
142
transactions have been able to support 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 BBB- levels at LTVs of 70%, BBB at 78%, BB at 76%. The underlying structure and nature of the assets will impact these levels.
Debt Service Coverage (DSC)
This is defined as the amount of net rental or operating income generated in a specified period from a leased property by a borrower, divided by the amount of interest (and principal if applicable) due in that period on the borrowed money against that property. Meaningful comparisons between transactions at closing can be distorted by situations where properties are still renting space that is in the process of completion. A further complication occurs if differing amortization requirements are included in the denominator. Direct DSC comparisons should also take account of the ratings levels down to which securities are issued it is more useful and accurate to compare coverage levels on equivalent rated classes of securities. In general, properties with perceived high-quality, stable, revenue streams will be permitted to operate on a lower debt service coverage level than borrowers with less stable revenue streams for any given ratings level. 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.
143
European CMBS
TRANSACTION CASE STUDIES
We have set out below some of the key highlights of five selected CMBS issues. We find that, in the examples we have examined, the underlying real estate assets are able to support high levels of AAA securities issuance. For example, over 80% of the Europa One transaction comprises AAA securities, whilst for the European Loan Conduit programme as a whole the figure is 75%. The Canary Wharf II and Trafford Centre transactions both comprise 63% of AAA securities, whereas the MS Mortgage Finance (Broadgate) transaction has 51% of its issuance in AAA-rated classes. We note that ratings at these levels are not always readily achievable by some other classes of asset-backed securities. For example, in the absence of third party support, issuance from the operating company sectors has generally been capped-out at a level of A.
exhibit 11 Rating (Moodys/S&P/Fitch)
MORGAN STANLEY MORTGAGE FINANCE (BROADGATE) PLC

Amount (m) Interest Basis WAL
Aaa/AAA/AAA Aaa/AAA/AAA Aaa/AAA/AAA Aa2/AA/AA A2/A/A A3/A/A Baa2/BBB/BBB

325 310 150 225 175 175 180
Floating Fixed Fixed Fixed Floating Fixed Floating
8.5 19.3 31.1 29.1 9.1 31.1 3.5
This transaction represents the refinancing of part of the British Land Companys interest in the Broadgate office complex in the City of London. The transaction is particularly notable for the high credit quality of the underlying tenant base and also the long average periods before leases can be broken. The very large size of this transaction should help to promote secondary market liquidity. Total 1,540m LTV 73.1% Weighted average tenant credit rating of AAverage period to first tenant break of 15 years Fully amortizing Asset substitution provisions 15m reserve/200m liquidity facility
144
exhibit 12 Rating (Moodys/S&P/Fitch)
CANARY WHARF FINANCE II PLC

Amount (m) Interest Basis WAL
Aaa/AAA/AAA Aaa/AAA/AAA Aa2/AA/AA A2/A/A Baa2/BBB/BBB

240 e100 85 45 45
Fixed Floating Fixed Fixed Floating
20.2 6.1 23.9 25.1 5.0
This transaction represents the financing of part of the Canary Wharf complex in Londons Docklands. Similar to Broadgate, the transaction features very high average tenant credit ratings and long term leases. Canary Wharf was notable for the issuance of a substantial Euro denominated tranche. Total 475m 68.4% of rentals from two tenants (with average credit rating of AA) 100m financing balloon
exhibit 13
LTV 70.7% Debt Service Coverage 1.14x (Year 1); Debt Service Coverage 1.39x (Year 2)
TRAFFORD CENTRE FINANCE LTD

Interest basis WAL
Rating (Moodys/S&P/Fitch) Amount(m)
Aaa/AAA/AAA Aaa/AAA/AAA Aa2/AA/AA Baa2/BBB/BBB Baa2/BBB/BBB

50 340 120 50 50
Floating Fixed Fixed Floating Fixed
9.8 25.6 19.8 7.6 21.0
The Trafford Centre is the third largest regional retail shopping mall in the U.K. Similar to Broadgate and Canary Wharf, the transaction is supported by a large percentage of rentals derived from long-dated leases, but differs from the two office transactions in that investors enjoy the benefit of very considerable tenant diversity; there are over 230 separate tenants that lease space in the complex although the underlying credit ratings of the tenant base is materially lower than for the two office transactions described above. For example, the credit ratings for the three anchor tenants, who account for 30% of the lettable space, do not exceed BBB+. Key transaction features are: Total 610m LTV 69.3% 50.4% of rentals from unexpired leases of 20+ years 93.1% of rentals from unexpired leases of 10+ years 15m reserve fund 55m liquidity facility Diversified tenant mix
145
European CMBS
exhibit 14 Rating (S&P)
EUROPEAN LOAN CONDUIT (ELOC) PROGRAMME

ELoC 1 ELoC 2 ELoC 3 ELoC 4
AAA AA A BBB BBBBB Total LTV No. loans Debt Service Coverage (Interest Only) Final maturity
All amounts m Source: Morgan Stanley
135.1 10.1 6.8 16.9 168.9 80.7% 13 1.64x 2008
280.4 26.9 19.8 21.6 10.8 359.5 80.0% 11 1.48x 2008
197.3 17.8 15.3 15.9 8.3 254.6 76.0% 16 1.34x 2009
328.1 40.4 39.3 42.8 11.6 462.2 70.0% 6 1.18x 2007
The European Loan Conduit programme represents the periodic securitization of smaller mortgage loans originated by Morgan Stanley. In these transactions investors generally get the benefit of considerable borrower and tenant diversity. For example, in the ELoC 2 transaction there are 105 separate commercial properties (in addition to 272 pubs). The ELoC transactions are shorter-dated floating rate transactions for which repayment relies on the ability to refinance the underlying loans at maturity. The ELoC 5 transaction, totalling 524.9m, is currently in the course of completion. Europa One is sponsored by Germanys Rheinische Hypothekenbank and is a credit linked note structure (with tranches being linked to both an underlying reference pool and also to either the banks Pfandbriefe or its senior unsecured notes). The transaction is designed to remove the risk of a portfolio of real estate loans from Rheinhyps balance sheet. Although nominally a German transaction, Europa One is in reality the first pan-European CMBS transaction as is evident from the geographic dispersion of the underlying real estate collateral.
S&P rated all four ELoC transactions. Fitch rated ELoC 1 and 4 only.
146
exhibit 15 Rating(Moodys/S&P)
EUROPA ONE LIMITED

Amount (em) Interest Basis WAL
Aaa/AAA Aaa/AAA Aaa/AAA Aa1/AAA Aa2/AA A2/A Baa2/BBB Ba2/BB N/R

290.0 400.0 400.0 30.0 37.0 80.5 57.0 30.3 20.2
Floating Floating Floating Floating Floating Floating Floating Fixed Fixed
1.3 4.1 7.9 11.9 12.7 14.1 14.5 14.5 14.5
Total e1,345m Credit linked note structure 75 mortgage loans Office (35%); retail (24%); mixed (18%); hotel (14%) Spain (36%); France (30%); Netherlands (17%); Germany (9%) Weighted Average Loan To Value 68% Weighted Average Debt Service Cover 1.59x Remaining Average Term: 13 years Seasoning: 1.9 years A second, similarly-structured transaction, Europa Two Limited, totalling e1,500m is currently in the course of completion.
E U R O P E A N R E A L E S TAT E M A R K E T F U N D A M E N TA L S
In line with steady economic growth in the European region since the mid-1990s, it is clear from the core statistics available that real estate market conditions have gradually improved.
R E A L E S TAT E Y I E L D S
One leading indicator of improvement is the performance of reported real estate yields. For these purposes yield is very broadly defined as the rentals achieved from a property divided by its purchase cost. There are technical differences between practices in different national markets concerning whether, for example, the purchase and running costs of real estate are taken into account when calculating the numerator. If these costs are not taken into account then this yield is described as a gross yield basis, whereas the practice in some markets is to quote on a triple-net basis, which means that all purchase and non-recoverable costs of running the building are taken into account, reflecting more accurately a real estate investors actual return.
147
European CMBS
Irrespective of these differences, we can partially discern the health of real estate markets by analysis of real estate yield behavior. The general proposition is that in relatively stable economic conditions, declining yields in real estate markets will, prima facie, be evidence of improving demand for real estate assets. In summary, declining yields are evidence of investor willingness to pay more than previously for a series of rental cashflows. Also, as inflation and long term interest rates fall, we might normally expect real estate yields to move in broadly the same direction. Clearly, in buoyant real estate markets there is likely to be a relative deficiency of supply for end-users from the core office, retail and industrial/warehouse sectors and this will naturally tend to drive up lease rental rates. Therefore, it is entirely conceivable that both rentals and real estate values will rise simultaneously, although rentals rising more slowly than real estate acquisition costs causes yields to decrease. We would also expect yields to display a relationship to alternative investments. Given that contracted rental flows are, in a broad sense, analogous to bond cashflows, then real estate yields might also correlate to changes in a suitable benchmark, such as government securities yields. However, this relationship will be by no means perfect as real estate yields are not risk-free investments and will also be affected by investor perceptions of probable future real estate performance and inflation risks.
OFFICE MARKETS
Exhibit 16 shows composite yield performance in the main European markets across the 12 month period Autumn 1999 to Autumn 2000.
exhibit 16 All Figures %
EUROPEAN PRIME OFFICE YIELDS: INDICATE IMPROVED MARKET CONDITIONS

Autumn 1999 Autumn 2000
Paris Frankfurt Berlin Munich Dublin Milan London West End London City London Canary Wharf
Source: FPD Savills, Jones Lang Lasalle, Healey & Baker
5.75 5.00 5.25 4.50 5.00 6.00 5.50 6.00 6.50
5.50 4.90 5.00 4.75 4.75 5.75 5.25 6.25 6.50
148
In most markets office yields have continued to tighten marginally over the 12 month period. Market reports identify limited availability of space in many markets, driven by the health of the financial services industry and also the continuing influx of telecom, media and technology companies to many traditional central business district locations and their immediate fringes. The other major pattern that has emerged is the limited current supply of new space. The reasons for this vary, ranging from certain physical constraints (Dublin), to tight central area planning restrictions (Milan, Paris) and reasonably conservative development pipelines. Particular pressure is noticeable in German office markets. It should be noted that data quality in the office sector is generally superior to that in the more fragmented retail and industrial sectors and this is why most real estate market studies often tend to use office markets as a proxy bellwether for general real estate market health.
R E TA I L M A R K E T S
Exhibit 17 shows the behavior of retail yields over the 12 month period.
EUROPEAN PRIME RETAIL YIELDS: INDICATE STABLE MARKET CONDITIONS

Paris Frankfurt Berlin Munich Dublin Milan London West End

6.00 4.75 5.05 4.00 3.75 5.00 6.00
6.00 5.00 5.00 4.25 3.75 5.00 5.50
In the retail sector yields have held generally held constant in the major markets with the lowest absolute levels seen in some German markets. We believe that investment sentiment towards retail real estate investments is not, in general, negatively affected by perceptions of threats to the health of retail real estate via extreme competition, importation of US retailing methodologies and threats to traditional retail outlets from internet shopping and alternative non high street locations.
149
European CMBS
WAREHOUSE AND LOGISTICS MARKETS
Exhibit 18 shows the changes in yields in warehouse and logistics markets over the last 12 months.
exhibit 18
EUROPEAN PRIME WAREHOUSE/LOGISTICS YIELDS: INDICATE MORE VOLATILE MARKET CONDITIONS

All Figures %
Paris Germany (composite) Dublin Milan London West End

10.00 7.50 7.00 9.00 6.50
9.50 7.75 6.50 8.50 7.25
These markets are materially more fragmented than the office and retail sectors and yields are generally wider to take account of these factors and the associated risks.
V A C A N C Y R AT E S
Another key leading indicator of market health is represented by vacancy rate performance. Vacancy rates will tend to rise when overall economic conditions deteriorate and businesses stop expanding or even close down and vacate premises. Conversely, vacancy rates will tend to fall as economic conditions improve and businesses require more space for operations. Vacancy rates will also be crucially affected by the relationship between the supply of new space and demand for that space. It is entirely conceivable that vacancy rates can rise during buoyant economic conditions if the supply of new space exceeds the rise in demand. More usually, however, vacancy rates will tend to decline in buoyant business conditions, encouraging the supply of new space, which can easily create a glut when business conditions deteriorate. Exhibit 19 displays the progress of vacancy rates across the main European office markets, including data for the last two years and also noting the high points for those markets in the 1990s. The data is consistent with an explanation that the relevant markets are currently in good health with strong demand for office space driving down vacancy rates. High demand on the part of occupiers is one component of the strength of the market for investment property, which has seen yields decline and hold at relatively low levels, as shown in exhibits 16-18.
150
EUROPEAN OFFICE VACANCY RATES

High (Year) 1999 2000
Paris Frankfurt Dublin Milan London City Composite (14 cities)

Source: Jones Lang LaSalle, DTZ
9.0 (1994) 9.0 (1995) 14.0 (1992) 18.0 (1992) 9.0% (1993)
4.3 5.0 1.0 7.0 4.8%
4.0 4.5 1.0 6.1 6.0 4.0%
The data needs to be interpreted with some care because the relatively high rates for Milan and London could be misinterpreted. Reports indicate, for example, that the supply of suitable modern space in Central Milan is severely limited due to planning restrictions and this has resulted in a number of businesses moving out to business parks on the edge of, or outside, the city. In turn this will raise the reported vacancy rate in the central business district although this masks the fact that an amount of the space in the centre will increasingly become obsolete in the absence of an ability to redevelop. In London, we believe that the sheer size of the market is able to support a higher overall level of vacancies on the basis that the higher numbers of buyers and sellers of real estate space in London will improve the chances of unlet space being absorbed. Also, we think that the gradual drift of some major City tenants to Canary Wharf has had a partial impact on City vacancy rates, although the continuing level of overall high demand has easily enabled the City office market to continue to perform very strongly.
R E N TA L T R E N D S
A further leading indicator of real estate market health is provided by trends in rental levels. Exhibit 20 shows the performance of rental growth in key European office market locations in the 12 months up to the end of Q3 2000.
exhibit 20 All Figures % Rental Growth
EUROPEAN OFFICE MARKET RENTAL GROWTH Q3 99 Q3 00

All Figures % Rental Growth
Paris Frankfurt Berlin Munich Dublin

Source: Jones Lang LaSalle
29.4 12.5 9.1 15.4 33.3
Milan LondonCity LondonWest End LondonCanary Wharf
10.0 9.6 33.3 6.9
151
European CMBS
The strength of rental growth in many city office markets is entirely consistent with the strength of real estate markets recorded in our other data sources.
C O M PA R AT I V E P E R F O R M A N C E B E N C H M A R K S GOVERNMENT SECURITIES
We believe that the conclusions from this data are reasonably transparent. The main European office markets are in good health, with strong increases observed in rental growth, which has been associated with declining vacancy rates and continuing downward pressure on yields. The retail and industrial/logistics sectors are subject to less comprehensive availability of data but we believe that similar general trends to the office sector are discernible. However, we also believe that there may be evidence of some circumspection on the part of investors who, despite observing continuing impressive increases in rental growth, have over the last 12 months been reluctant to continue to chase down yields to materially lower levels. In this regard, we note that whereas our discussion of real estate yields, as detailed above, indicated a position of relative stability with some downwards trend in some areas over the last twelve months, this contrasts with the story as it applies to government securities in the main European markets. Exhibit 21 shows that government securities yields have moved markedly downwards through the year 2000. This has not been fully reflected to date in real estate yields, although there will tend to be data collection and market reaction lags in the real estate markets compared to the bond markets. To some extent certain government securities markets have been impacted by technical factors unduly affecting their demand and supply patterns. An example of this may be the unusually limited supply of government debt which could have artificially depressed yield levels. However, to the extent that government securities yields represent a proxy for the base cost of debt funding for real estate investment, then our comparison may be useful. The fact that these proxy base funding costs have continued to reduce compared to real estate yields must make real estate easier to finance, and therefore prima facie more attractive, for debt-based investors. Of course, the risk and return factors relating specifically to commercial real estate will tend to reduce the degree of correlation between changes in bond yields and changes in real estate yields. However, the general
exhibit 21
10 YEAR EUROPEAN GOVERNMENT BOND YIELDS

All Figures % Yield at 30/12/00 2000 average
France Germany Ireland Italy U.K.

Source: Bloomberg
5.050 4.901 5.137 5.300 4.897
5.403 5.264 5.520 5.600 5.265
152
proposition is that reducing real estate yields reflect correspondingly lower risk perceptions by investors, and the failure of real estate yields to chase bond yields down actually strengthens the market to the extent that real estate investment becomes easier to finance.
E U R O P E A N O F F I C E M A R K E T S A R I S K O F O V E R S U P P LY ?
Experienced investors in European real estate and their related securities will be conscious that, notwithstanding currently positive market conditions, too much development activity could easily disrupt supply/demand balances and put pressure on underlying real estate yields. This was the scenario that developed between 1989 and 1992 when an economic expansion was accompanied by excessive speculative development activity and an overhang of office property that needed the economic recovery in the mid-1990s to clear the market. A recent study, by Jones Lang Lasalle, completed in October 2000, of European office development across 12 major centers has revealed the following conclusions relating to forecast supply. Total actual and forecast completions of new office space, based on announced building projects, are shown in exhibit 22.
exhibit 22
FORECAST SUPPLY OF NEW OFFICE SPACE

Year Square Metres (,000)
1999 2000 2001 2002 2003 2004

Source: Jones Lang LaSalle
actual actual forecast forecast forecast forecast
2,904 4,230 4,520 4,550 3,560 4,250
Clearly, actual new building volumes can change relatively quickly as projects that are put on hold are reactivated and the total of 8.75m square metres anticipated for 2000-01 compares to only 5.0m that was anticipated 12 months ago. Current estimates suggest that a total of 17m square metres of new office supply will be brought on stream over the period 2001-04 and this compares to the 16m square metres of new supply that was brought on stream between 1989 and 1992. While this may at first sight appear to be a cause for concern, there is a reasonable case to support the view that underlying macroeconomic conditions are now materially more stable than in the prior period.
153
European CMBS
Supporting evidence for this proposition is provided by the following factors: The demand and supply drivers appear superior to 1989-92, with more pre-let space and lower volumes of speculative space being brought to the market. Funding fundamentals are currently better than in the earlier period, with 1989 short term interest rates averaging 9.8% (peaking at 10.7% in 1990), compared to current short term rates of below 5%. EU CPI is currently 2.9%, compared to 5% in 1989 (peaking at 5.4% in 1990). By most estimates, the European economies were running a position of excess demand in the earlier period compared to the current underutilization of capacity. This scenario can support growth without overheating. The broad conclusion is that, provided economic fundamentals remain stable then a similar level of new office supply over the 2001-04 period compared to 1989-92 should be comfortably absorbed. Clearly, the structure of the real estate markets is now different compared to the earlier period and some asset types (for example, buildings without modern cabling) will be subject to rental and occupancy pressures, although the broader picture appears to be more stable.
O W N E R S H I P A N D F U N D I N G O F E U R O P E A N R E A L E S TAT E
Although the ownership of real estate is of course highly fragmented, in recent years we believe that investment patterns in European real estate have changed materially. There are a number of reasons for this.
Corporates
European corporates are being affected by globalization trends. This means that increased competition, and merger and acquisition activity, has resulted in managements increasingly scrutinizing their asset bases with the intention of ensuring assets are deployed with maximum efficiency. There is growing evidence that corporates are increasingly moving to a perception that real estate is a non-core asset. Examples of this trend over the last 2 years involving leading corporates include: France: sales by Societe Generale to a property company of 76 office buildings for FF 4.3 bn France: sale by Carrefour of shopping centre properties for
e4.3
bn.
Germany: advertised sales by Deutsche Bahn and Deutsche Telekom of non-core properties. Italy: sale of office properties by Mont Edison and also sale by ENI of its real estate subsidiary, Immobiliare Metanopoli. United Kingdom: proposed sale and leaseback of operating assets by supermarket chain, J Sainsbury plc and telcoms operator, BT.
154
Internationalization
The single currency has prompted a considerable internationalization of European real estate investment. It is estimated that in 2000 a record e20 bn of cross border real estate investment will have been made. Approximately 50% of direct French commercial real estate transactions in 2000 are estimated to have been conducted by non-French investors, with many trades actually being entirely between non-French investors. Even the U.K., which is outside of the immediate scope of the Euro, saw 18% of its commercial real estate investment being undertaken by non-U.K. domiciled investors in 2000.
Real Estate Companies
Although dedicated real estate companies have a naturally high profile, their impact in terms of overall ownership of underlying real estate can easily be overestimated. Although measures such as the relative share of the equity of listed property companies of their national stock markets are not necessarily good indicators of underlying ownership of real estate (which can of course be materially affected by leveraging policies), it is nevertheless an interesting indicator of relative size. The data shows that the average weighted share of dedicated commercial real estate companies of European equity market capitalization is just 0.85% (this figure excludes the capitalization of open-ended German funds). One of the highest shares is in the United Kingdom where quoted real estate companies comprise approximately 1.5% of the market. Interestingly, this share is down from 4.5% in the late 1980s. The story behind this decline is one of equity market perceptions being increasingly that real estate is a relatively low growth sector which results in unattractive returns compared to other areas in low inflation environments. This is particularly true of real estate investment companies which have limited speculative activity which, although carrying more risk, is also potentially higher revenue producing. Perceptions towards equity real estate investment in the U.K. have never quite recovered from the value erosion registered in the early 1990s. The declining equity market share has been exaggerated more recently by considerable corporate finance activity in the sector which, prompted by equity valuations trading materially below the net asset valuations of companies, has seen a number of companies being taken private. In addition, real estate companies are beginning to look more closely at new business models which increasingly separate out the ownership and management of real estate assets. Although Germany is increasingly developing a quoted real estate sector, assisted by tax reforms that make it easier to sell real estate investments held in subsidiaries, the larger share of direct real estate is held by open-ended real estate funds with an estimated DM 100 bn of assets.
155
European CMBS
Pension Funds and Life Insurance Companies
One trend that has been constant across Europe has been for the relative direct ownership of real estate by long-term investors, such as pension funds and insurance companies, to decline. Reasons for this include the increasing perception that the need for real estate investment as a partial hedge against inflation is diminished in a low inflation environment. Additionally, pension funds and insurance companies have gone the way of equity investors in perceiving real estate to be a relatively low return asset class. Also, smaller pension funds are realizing that it is expensive to undertake active management of small real estate portfolios and minimum portfolio size is required to obtain a reasonable level of diversity. Available data suggests that in Germany insurance companies and pension funds have reduced their share of assets in direct real estate from 9.9% in 1980 to 3.4% in 1999. In the United Kingdom insurance companies had reduced their share of real estate from 10% of their investment portfolios in 1988 to 4% in 1999. Similarly, U.K. pension funds had reduced their share of investments in real estate from 8% in 1988 to 3.6% in 1999. One general exception to the role of insurance companies and pension funds in funding real estate is in Italy where almost 50% of pension fund assets are in real estate (compared to a European average, excluding Italy, of 9.3%). In Italy the share of insurance company assets that are dedicated to real estate investment fell from 10% in 1996 to 5.5% in 1998.
Bank Funding for Real Estate
Patterns relating to bank funding of real estate show material differences compared to the peak of the previous real estate cycle in 1989-92. We believe that banks have become more sophisticated over that period and have in place better controls over aggregate volumes of real estate lending and, in particular, over speculative property lending to real estate development. However, DTZs annual survey (March 2000) of bank lending to real estate indicated that real lending volume was equivalent to real lending volumes at the 1989-92 peak, although separate parts of the survey show bank lenders in the U.K. to be generally holding a more disciplined line compared to the last real estate boom concerning maximum loan-to-value ratios, maximum term of loans and also in attitudes to speculative development.
Conclusion
Overall, we conclude that real estate has become a relatively less attractive asset for equity investors in real estate and also for direct investors in real estate in the form of insurance companies and pension funds. To some extent, alternative sources of funding, such as unitized property vehicles in the UK and open-ended funds in Germany, have grown to partly fill any funding gap. At the same time, although there remains strong evidence of bank support for the real estate sector, it seems clear to us that lending appetite is more controlled than at the peak of the last real estate boom. Our general conclusion is that there will be considerable volumes of real estate continuing to change hands over the foreseeable future and there will be substantial requirements for associated funding. We believe that new CMBS issuance is well-placed to play a major part in that process.
156
exhibit 23
THE ATTRACTIVENESS OF CMBS TO ISSUERS AND INVESTORS
Investor Attractions CMBS Feature Issuer Attractions Liquidity of issue likely to be Large size Ability to raise funds in higher enhanced for larger transactions volumes than syndicated bank market or local debenture/bond markets; deal with single underwriter/small group and not large bank syndicate Long term debt Available maturities generally longer than bank lenders will absorb Tranched bond Lowers average cost of funding issuance Rated bond Assists marketing via transparency issuance Mix of Match underlying cashflows to fixed/floating bond coupons (with swaps rate coupons where applicable) Varying Matches cashflow from maturity profiles underlying assets Mix of currency Can ensure optimum distribution tranches Multiple real Large scale execution estate assets Ability to substitute/sell assets Cashflow driven financial covenants and real estate adviser provisions Synthetic/credit derivative structures Yield pick-up compared to comparably-rated securities
Matching of liabilities (for insurance companies and pension funds) Bonds available with risk/return profile according to investor preferences Clear risk differentiation/should enable easier surveillance Accommodates differing coupon requirements Accommodates differing maturity requirements Accommodates regional investor requirements Improves risk profile of exposure across a number of real estate assets and tenants Substitution/repayment provisions tightly drawn to protect original credit characteristics Exercise strong control via debt service triggers, reserve funds and real estate adviser control provisions Widens investor choices
Flexibility in managing portfolios
N/A
Suitable for issuers primarily seeking risk reduction rather than funding N/A
Improved risk/reward balance for an asset secured position compared to equivalent rated securities
157
European CMBS
PREVAILING FUNDING CONDITIONS AND THE ROLE OF CMBS
Whilst the relative share of real estate in the asset base of key investors may have declined over a 10 year period, there is of course no necessary suggestion that the supply of investment funds for real estate are showing an absolute shortage. However, we believe that some of the trends indicated in the investment and funding market for real estate may assist both potential issuers of, and investors in, commercial mortgage-backed securities. In general terms, we believe that with declining inflation trends the attractiveness of real estate as an equity-type investment may have declined. However, we also believe that carefully structured real estate assets can make very attractive, yieldenhanced, debt investments. Although CMBS in the U.K. and Europe have exhibited a number of different structures, we have set out below some of the typical features of the asset class that enable us to conclude that there are a number of attractions to both issuers and investors that bode well for future market growth. Clearly, different CMBS transactions will vary in terms of their characteristics according to the grid below but many of these characteristics will be present in individual transactions.
CONCLUSIONS
We draw the following core conclusions from this survey of European CMBS. We have demonstrated above that relative spreads on AAA CMBS are favorable against relevant comparables and this yield pick-up largely holds true for lower-rated tranches. To some extent this may be due to CMBS being a relatively small market, with associated historically limited liquidity, as well as relative unfamiliarity on the part of investors with products and structures. We also believe that transactions that have been completed to date, particularly the large transactions that have been featured over the last 2-3 years, have involved very strong assets and structures. Accordingly, European CMBS is a sector whose relative merits we currently recommend to investors. Underlying real estate market conditions in Europe are currently in robust health. We believe that the balance of probabilities is that these markets can remain in good health. Whilst we do not underestimate the risks arising from a slowdown being imported from the US, we believe that the European economies have a reasonable chance of avoiding a steep recession. In this context, when compared to the US, the European economies are less affected by changes to consumer behavior arising from stock market conditions, there has been less of a high technology investment surge (and therefore no risk of a rapid slowdown) and Eurozone exports to the US only account for 2.5% of total GDP.
158
We believe that, against this broad economic background, the more circumspect approach of both real estate developers and providers of funding to the real estate sector compared to the last boom should cause the current expansion in the underlying real estate market to result in more benign consequences than was the case during the previous major expansion starting in the late 1980s. The patterns of ownership and funding for real estate are changing rapidly. There will be considerable opportunities for financing real estate in these circumstances and we believe that well-structured CMBS transactions can play a key part in this process.
R AT I N G A G E N C Y U . K . C M B S M U LT I - B O R R O W E R M O D E L
The following generic rating analysis is employed by the rating agencies for real estate transactions similar to ELoC and Europa, which consist of a pool of commercial mortgage loans. The analysis simulates lease termination and tenant default in order to assess stressed debt service coverage on a loan by loan basis. The basic analysis can equally be applied to transactions featuring commercial real estate companies with a limited number of properties under lease to a concentrated group of tenants. Initially rating agencies will require not only forward-looking statements but detail on historic cashflows, ideally audited data for 3 years. Loan collateral is broken down by individual lease and projected forward over the life of the underlying loan. When leases are timed to terminate, or a break option exists, void periods are assumed based on the perception of the quality of the property. Historically very low vacancy rates and void periods may not be sustainable going forward depending on the nature and quality of the real estate asset and the market in which it operates. Stress runs are conducted depending on the rating level sought. For example, significantly higher proportions of tenants are assumed to vacate the property under a AAA stress than under a BBB stress. At lease rollover, rent is reduced based on the type of property and the stress run being carried out. Assumed declines in rental value in U.K. transactions are based on historical reference points that may be obtained, for example, from the Investment Property Dealers databank where the 1990-93 recession has been identified as an A class recession in the U.K. real estate markets. Refurbishment needs are assumed, varying with the property. Because European leases are generally shorter than U.K. equivalents, the cashflow disruption from lease rollover will be more significant than for U.K. leases.
159
European CMBS
In addition to tenant rollover, tenants are also randomly defaulted according to a corporate default matrix, where the probability of default is determined by the credit rating of the tenant and the stress run being carried out. The default rate for tenants will be a function of tenant credit rating and also of the period of reliance on each tenant under a specific lease. Void periods and re-letting costs and fees are assumed for each defaulting tenant. There are important differences between, for example, U.K. fully repairing and insuring leases, where it is the obligation of the tenant to repair and refurbish the property, and leases where it is the responsibility of the landlord to effect repair and refurbishment. In situations where the landlord is responsible, the rating analysis must further take into account the ability of the real estate assets cashflows to meet refurbishment, repair and other operating costs. The result of this analysis is a set of underwritten cashflows which can be compared with projected interest and scheduled principal payments of each loan. Where the stress run shows a DSCR dropping below 1.0x for an extended period then it is assumed that the loan will default. It is also assumed that scheduled principal payments take place where there is sufficient stressed excess cashflow to do so, otherwise the loan defaults. The DSCR default test will be adjusted where there is direct recourse to an investment grade borrower. If a loan default occurs, then loss severity is calculated by reference to the difference between the outstanding principal at the point of default and the stressed recovery value from the properties. Recovery values are adjusted for sales costs and delays in refinance or sale. Although the analysis described here is fundamentally cashflow-driven, property valuations play a significant role in the analysis. When loans default, realizations of underlying real estate securing the loan are used to reduce or eliminate loan losses. For this reason the rating agencies analyze original valuations to create stabilized initial valuations, against which market value declines will be applied when assessing the realization value of collateral following a loan default. Rating agency stabilized valuations may be lower than the external valuations if, for example, the rating agency believes the rental flows underpinning the external valuation may be difficult to sustain, possibly via excessive reliance on one or more credit tenants with impending lease rolloffs or alternatively if rental levels are regarded as above-market.
160
If real estate assets need to be foreclosed on, then the rating agency analysis will use longer foreclosure periods for higher desired ratings levels. A large number of data runs are conducted for each stress level and an average level of credit enhancement required together with standard deviation at both a pool and an individual loan level. Judgmental add-ons are also made for other factors that the model cannot reflect, such as loan, borrower and property concentration, loan servicing, loan origination, environmental and disaster issues and excess spread (where applicable). Maturity profiles are also considered, especially in relation to the tail where a single outstanding loan will concentrate the risk in relation to notes then outstanding. Using this loan by loan analysis, the rating agency is able to generate required credit enhancement for each class of rated notes which is sensitive to likely timing of default. In addition to the financial modeling, rating agencies will perform additional due diligence in terms of the review of third party valuations, property visits, review of the loan servicer and legal and property title due diligence.
161
Transaction Monitoring
Once CMBS bonds have been issued, the rating agencies, trustees, master servicers, and special servicers monitor the transactions. Each rating agency that originally rated the transaction provides an annual review of the deal and either upgrades, downgrades or affirms the current rating. As part of the annual review, the rating agencies examine the current LTV ratios, DSCR, prepayments, and status of specially services loans within each transaction. Over the past three years upgrades on CMBS transaction have significantly outnumbered downgrades. During 2001, CMBS upgrades outnumber downgrades 15 to one (see chapter 2). On a monthly basis, the trustees publish remittance and special servicing reports that detail payments made to bondholders, delinquent loan information, and specially serviced loan details. In order to assist investors with monitoring their CMBS investments, we publish monthly CMBS delinquency data and quarterly CMBS tracking reports, which outline loan level details. The first part of this chapter is an example of our monthly CMBS delinquency report. The report outlines overall delinquency trends in the CMBS universe and provides trends by property type and origination year. The second part of this chapter is an example of our quarterly CMBS tracking report, which provides details on specially serviced loans within Morgan Stanley sponsored transactions.
I. CMBS Delinquencies February 2002

Based on February remittance reports, seasoned CMBS delinquencies rose 8 bp to 2.09%. Delinquencies on all CMBS increased 7 bp to 1.30%. Senior housing and hotel delinquencies continue to trend above the universe average. ACLI delinquencies fell to 0.12% in the fourth quarter of 2001. Seasoned CMBS delinquencies continued to trend upward in January, but the rate of increase has moderated since December, when delinquencies rose by 16 bp. Based on February remittance reports, CMBS delinquencies on seasoned transactions (aged over 1 year) increased 8 bp to 2.09% of current balances. Seasoned delinquencies are now 102 bp higher than the 4-year rolling average of 1.07%. We believe that total delinquencies will trend up to 3% or more over the next 18 months. The CMBS delinquency rate on our entire universe of transactions increased by 7 bp in January to 1.30%. Sixty+ day delinquencies also increased 7 bp to 1.00% of current balances.
162
exhibit 1
% 2.5
Total Delinquency 4 Y ear Average
DELINQUENCIES IN SEASONED CMBS DEALS AND 4-YEAR AVERAGE
2.0 1.5 1.0 0.5 0.0 Dec-96
Mar-98
Jul-99
Oct-00
Feb-02
O R I G I N AT I O N Y E A R
Of the cohorts with current balances greater than $10 billion, the 1996 cohort had the highest delinquency rate increase for the month (16 bp), and the 1997 cohort had the highest delinquency rate (2.62%).
exhibit 2
CMBS DELINQUENCIES BY YEAR OF ORIGINATION (IN %) (AS OF FEBRUARY REMITTANCE REPORTS)

Current Balance 30/60/90+ ($bn) days (%) Forc. & REO (%) Change From Last Total ( %) Month (%)
Year
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 T otal/Avera ge
0. 3 0. 2 0. 3 0. 9 2. 1 6. 0 14. 4 37. 4 65. 6 38. 1 39. 4 42. 5 247. 1
0.75 0.04 0.00 0.40 1.95 3.08 1.96 1.93 0.92 0.78 0.47 0.06 0.95
1.46 0.00 0.00 0.30 2.01 1.18 0.53 0.70 0.31 0.46 0.05 0.00 0.35
2.20 0.04 0.00 0.70 3.96 4.26 2.49 2.62 1.23 1.23 0.53 0.06 1.30
-1. 18 -0. 01 -0. 07 -0. 55 0.86 0.72 0.16 0.15 -0. 08 0.12 0.12 -0. 08 0.07
W H AT L I E S A B O V E
Senior housing and hotels are the two property types with delinquency rates that are above the universe average of 1.30%. Retail delinquencies have remained under the universe average since November. Senior housing properties continued to post the highest delinquency rate (6.30%) this month, while hotels posted the largest delinquency rate increase (42 bp) of the property types.
163
exhibit 3
CMBS DELINQUENCIES BY PROPERTY TYPE (IN %) (AS OF FEBRUARY REMITTANCE REPORTS)

Current Balance 30/60/90+ ($bn) days (%) Forc. & REO (%) Change From Last Total ( %) Month (%)
Year
Hotel-Motel Industrial-W arehouse Mixed Mobile Home Multifamily Office Retail Self-Storage Senior Housing Warehouse Other Total/Average
25.3 13. 5 11.4 4.1 47.7 54.9 72.6 2.9 4.5 0. 8 9.3 247.1
3.8 8 0.76 0.4 2 0.4 8 0.4 9 0.2 9 0.8 4 0.1 1 4. 19 0.00 0. 00 0.95
0.8 5 0.11 0.1 4 0.0 3 0.2 6 0.1 7 0.4 1 0.0 3 2. 11 0.00 0. 00 0.35
4.7 3 0.87 0.5 5 0.5 1 0.7 5 0.4 6 1.2 4 0.1 5 6. 30 0.00 0. 00 1.30
0.4 2 -0. 11 -0.11 0.1 1 -0.03 0.0 7 0.0 3 -0.04 0. 28 0.00 0. 00 0.07
P E N N S Y LV A N I A D E L I N Q U E N C I E S T I C K U P
Although Florida properties continue to post the highest delinquency rate, Pennsylvania properties had the greatest delinquency rate increase. Pennsylvania delinquencies rose 53 bp to 1.62% in January. Ninety-day delinquencies comprised 95 bp of Pennsylvanias delinquencies, 30-day delinquencies accounted for another 60 bp of delinquencies, and foreclosures accounted for 7 bp. This month, about half of the new 30-day Pennsylvania delinquencies were attributable to hotel properties.
exhibit 5
CMBS DELINQUENCIES BY PROPERTY TYPE AND YEAR (IN %) (AS OF FEBRUARY REMITTANCE REPORTS)
1990 1991 1992 1993 1994
Product Type
Hotel-Motel Industrial-Warehouse Mixed Mobile Home Multifamily Office Retail Self-Storage Senior Housing Warehouse Other
0.00 0. 00 0.0 0 N/A 0. 00 1. 88 6. 06 N/A N/A 0. 00 0.0 0
N/A 0. 00 7.4 7 N/A 0. 00 0. 00 0. 00 N/A 0. 00 0. 00 0.0 0
0. 00 0. 00 0.0 0 0. 00 0. 00 0. 00 0. 00 N/A 0. 00 0. 00 0.0 0
0. 00 2. 95 1.2 9 0. 00 1. 34 1. 37 0. 00 0.0 0 5. 56 0. 00 0.0 0
4. 55 18. 70 0.0 0 0. 00 0. 23 0. 04 7. 60 0.0 0 6. 65 0. 00 0.0 0
Source: Commercial Mortgage Alert
164
ACLI DELINQUENCIES DECLINE IN 4Q2001
This week, the American Council of Life Insurers (ACLI) reported that delinquency rates on commercial mortgages originated by the life insurance industry fell to 0.12%, as of December 31, 2001. The previous mark was 0.19%, at the end of September 2001. Although delinquencies on industrial and hotel/motel properties increased during the fourth quarter, delinquencies on the other major property types declined. The economic downturn has not yet translated into an overall higher delinquency rate, as ACLI delinquencies typically have a 9-month lag in reflecting the effects of weak economic conditions. ACLI expects commercial mortgage delinquencies to rise in 2002. Part of the fall in the ACLI delinquency rate is also attributable to an increased number of foreclosures in 2001, as compared to 2000. In 2001, $519 million of loans were foreclosed, while $437 million of foreclosures occurred in 2000.
exhibit 4
CMBS DELINQUENCIES BY STATE (IN %) (AS OF FEBRUARY REMITTANCE REPORTS)

Current Balance ($bn) % T otal Delinq. Change From Last Month
State
CA NY TX FL NJ IL VA PA GA MA Total/Average
38.9 24.9 17.1 13.5 8.1 7.3 7. 3 6.4 6. 3 5.9 135.8
0.2 9 0.7 9 1.9 4 3.6 0 0.8 7 1.3 6 0.91 1.6 2 1.94 0.4 5 1.19
0.0 3 0.1 0 0.0 6 0.1 7 0.0 3 -0.46 -0. 08 0.5 3 -0. 46 0.2 7 0.04
1995
1996
1997
1998
1999
2000
2001
12. 94 0. 00 0.0 0 2. 41 0. 49 1. 93 5. 64 0.0 0 12. 96 0. 00 0.0 0
10. 25 0. 94 1.9 1 1. 43 0. 71 0. 43 2. 17 0.2 1 4. 10 0. 00 0.0 0
9. 91 1. 36 1.9 1 0. 58 0. 96 0. 91 2. 04 0.4 5 12. 31 N/A 0.0 0
4. 22 1. 17 0.6 1 0. 21 0. 42 0. 85 0. 90 0.0 0 4. 93 0.00 0.0 0
2. 14 0. 92 0.4 1 1. 33 1. 21 0. 54 1. 69 0.0 0 4. 70 0.00 0.0 0
2. 06 0. 14 0.3 9 0. 00 1. 53 0. 18 0. 18 0.5 0 0. 36 N/A 0.0 0
0. 39 0. 09 0.0 0 0. 00 0. 13 0. 01 0. 04 0.0 0 0. 00 N/A 0.0 0
165
exhibit 6
SEASONED MULTI-BORROWER CMBS DELINQUENCIES (%) (AS OF FEBRUARY REMITTANCE REPORTS)

Issue Original Balance ($mm)
Aetna Commercial Mortgage Trust, 1995-C5 American Southwest Financial Securities Corp., 1995-C1 AMRESCO Commercial Funding I Corp., 1997-C1 Asset Securitization Corporation, 1995-MD4 Asset Securitization Corporation, 1996-D2 (Nomura) Asset Securitization Corporation, 1996-D3 Asset Securitization Corporation, 1996-MDVI Asset Securitization Corporation, 1997-D4 Asset Securitization Corporation, 1997-D5 Asset Securitization Corporation, 1997-MDVII Nomura Asset Securities Corp., 1994-MD1 Nomura Asset Securities Corp., 1995-MD3 Nomura Asset Securities Corp., 1996-MD5 Nomura Asset Securities Corp., 1998-D6 Capco America Securitization Corporation, 1998-D7 Commercial Mortgage Asset Trust, 1999-C1 Commercial Mortgage Asset Trust, 1999-C2 Banc of America Commercial Mortgage, 2000-1 Banc of America Commercial Mortgage, 2000-2 Bear Stearns Commerical Mortgage Securities Inc., 1998-C1 Bear Stearns Commerical Mortgage Securities Inc., 1999-C1 Bear Stearns Commerical Mortgage Securities Inc., 1999-WF2 Bear Stearns Commerical Mortgage Securities Inc., 2000-WF1 Bear Stearns Commerical Mortgage Securities Inc., 2000-WF2 CBA Mortgage Corp., 1993-C1 Chase Commercial Mortgage Securities Corp, 1996-1 Chase Commercial Mortgage Securities Corp, 1996-2 Chase Commercial Mortgage Securities Corp, 1997-1 Chase Commercial Mortgage Securities Corp, 1997-2 Chase Commercial Mortgage Securities Corp, 1998-1 Chase Commercial Mortgage Securities Corp, 1998-2 Chase Commercial Mortgage Securities Corp, 1999-2 Chase Commercial Mortgage Securities Corp, 2000-1 Chase Commercial Mortgage Securities Corp, 2000-2 Chase Commercial Mortgage Securities Corp, 2000-3 Chase Manhattan Bank - First Union National Bank, 1999-1 Commercial Mortgage Acceptance Corporation, 1997-ML1 Commercial Mortgage Acceptance Corporation, 1998-C1 Commercial Mortgage Acceptance Corporation, 1998-C2 Commercial Mortgage Acceptance Corporation, 1999-C1 CS First Boston Mortgage Securities, 1995-WF1 CS First Boston Mortgage Securities Corp., 1997-C1 CS First Boston Mortgage Securities Corp., 1997-C2 CS First Boston Mortgage Securities Corp., 1998-C1 CS First Boston Mortgage Securities Corp., 1998-C2
443.3 294.1 511.5 968.3 879.5 784.2 896.3 1437.1 1788.3 502.6 410.7 536.7 774.0 3722.7 1249.0 2375.0 775.2 795.2 896.5 716.1 479.6 1088.0 892.6 842.8 541.5 445.4 263.2 540.6 817.8 820.2 1270.8 783.4 698.3 739.6 769.4 1402.9 850.9 1192.2 2900.1 736.5 245.0 1363.5 1468.1 2488.1 1923.1
Note: Shaded deals have a total delinquency rate of more than 2%. New deals added this month are in bold.
166
Current Balance ($mm)
Factor
30, 60 & 90+
Forc. & REO
Total
Change from Last Month
129.3 109.3 401.0 746.8 779.8 709.6 831.3 1297.1 1676.8 452.2 107.9 400.7 723.0 3552.4 1178.6 2313.9 756.9 728.9 877.2 676.7 457.7 1035.7 867.6 823.4 65.7 328.4 216.4 433.5 679.0 748.3 1223.9 768.1 687.5 730.6 759.4 1366.6 799.3 1071.8 2733.4 709.8 88.4 1229.8 1393.9 2350.6 1846.7
0.29 0.37 0.78 0.77 0.89 0.90 0.93 0.90 0.94 0.90 0.26 0.75 0.93 0.95 0.94 0.97 0.98 0.92 0.98 0.94 0.95 0.95 0.97 0.98 0.12 0.74 0.82 0.80 0.83 0.91 0.96 0.98 0.98 0.99 0.99 0.97 0.94 0.90 0.94 0.96 0.36 0.90 0.95 0.94 0.96
0.00 4.08 0.00 0.00 9.26 2.21 0.00 0.00 2.92 0.00 16.69 0.00 0.00 0.28 0.26 0.13 0.00 0.99 0.63 0.15 0.00 0.26 0.45 0.00 0.00 1.67 1.09 0.43 1.13 0.00 0.00 0.00 1.85 0.81 0.00 1.01 9.68 0.77 3.41 0.92 0.00 0.96 0.54 1.52 0.48
0.00 1.96 0.00 0.00 2.25 1.09 0.00 0.45 0.82 0.00 29.36 0.00 0.00 0.00 1.02 0.23 0.63 0.00 0.00 0.00 1.12 0.53 0.00 0.00 0.00 1.99 1.46 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.42 0.00 0.00 5.16 1.74 1.44 0.17
0.00 6.04 0.00 0.00 11.50 3.29 0.00 0.45 3.75 0.00 46.04 0.00 0.00 0.28 1.28 0.36 0.63 0.99 0.63 0.15 1.12 0.79 0.45 0.00 0.00 3.67 2.55 0.43 1.13 0.00 0.00 0.00 1.85 0.81 0.00 1.01 9.68 0.77 3.82 0.92 0.00 6.12 2.28 2.97 0.65
0.00 3.14 0.00 0.00 -5.03 -0.41 0.00 0.00 0.00 0.00 12.43 0.00 0.00 0.13 0.00 0.00 0.63 0.28 0.17 0.00 0.00 0.00 0.00 0.00 0.00 -2.17 0.00 0.00 0.62 0.00 0.00 0.00 0.00 0.00 0.00 0.18 -0.01 -0.19 1.52 -0.02 0.00 0.00 -0.91 -0.13 0.48
167
exhibit 6 (cont.)

Issue Original Balance ($ mm)
CS First Boston Mortgage Securities Corp., 1999-C 1 CS First Boston Mortgage Securities Corp., 2000-C 1 CS First Boston Mortgage Securities Corp., 1997-SPICE Deuts che Mortgage & Asset Receiving Corp., 1998 -C1 C OMM 1999-1 C OMM 2000-C1 DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., DLJ Mortgage Acceptance Corp., First Union-Lehman 199 7-C 1 First Union-Lehman 199 7-C 2 First Union-Lehman Bank of America 1998 -C 2 First Union Commercial Mortgage Trust, First Union Commercial Mortgage Trust, First Union Commercial Mortgage Trust, First Union Commercial Mortgage Trust, 1999-C1 1999-C4 2000-C1 2000-C2 1994 -MF 11 1995 -CF2 1996 -CF1 1996 -CF2 1997 -CF1 1997 -CF2 1998 -CF1 1998 -CG1 1998 -CF2 1999 -CG1 1999 -CG2 1999 -CG3 2000 -CF1 2000-CKP1
1175 .9 1118 .5 352. 3 1821 .8 1318 .8 908. 4 210. 4 510. 2 472. 1 510. 2 449. 0 663. 1 841. 1 1567 .8 1111 .1 1243 .2 1554 .0 901. 2 887. 6 1294 .5 1308 .8 2209 .5 3417 .4 1171 .2 890. 0 779. 9 1149 .4 1188 .3 459. 8 1716 .1 1071 .9 1444 .1 2530 .6 1338 .3 976. 5 1154 .8 882. 3 775. 9 1320 .0 554. 1 1868 .8 897. 5 1016 .5 960. 7 320. 2 297. 4
First Union Chase Commercial Mortgage Trust, 1999-C2 GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., GMAC Commercial Mortgage Securities Inc., 1996-C1 1997-C1 1997-C2 1998-C1 1998-C2 1999-C1 1999-C2 1999-C3 2000-C1 2000-C2 2000-C3
GS Mortgage Securities Corp. II, 1996-PL GS Mortgage Securities Corp. II, 1998-C1 GS Mortgage Securities Corp. II, 1999-C1 Heller Financial Commercial Mortgage Asset Corp., 1999 -PH1 Heller Financial Commercial Mortgage Asset Corp., 2000 -PH1 Impac Commercial Holdings, Inc., 1998-C1 ICCMAC Multifamily and Commercial Trust, 1999-1
168
Current Balance ($ mm)
Factor
30, 60 & 90+
Forc. & REO
Total
1137 .0 1097 .0 155. 0 1716 .6 1266 .7 886. 3 44. 7 398. 7 395. 5 403. 4 384. 5 611. 4 790. 3 1465 .9 1052 .8 1200 .9 1508 .9 880. 7 876. 5 1276 .4 1135 .8 2018 .2 3232 .8 1112 .9 870. 5 764. 1 1130 .8 1142 .7 231. 6 1410 .1 1005 .0 1305 .2 2353 .9 1279 .1 946. 9 1127 .1 867. 4 765. 2 1306 .7 271. 9 1777 .3 846. 9 971. 7 940. 7 274. 8 150. 7
0. 97 0. 98 0. 44 0. 94 0. 96 0.98 0.21 0. 78 0. 84 0. 79 0. 86 0. 92 0.94 0. 94 0. 95 0. 97 0. 97 0.98 0.99 0. 99 0. 87 0. 91 0. 95 0. 95 0.98 0.98 0. 98 0. 96 0. 50 0. 82 0. 94 0. 90 0. 93 0. 96 0. 97 0. 98 0.98 0. 99 0. 99 0.49 0. 95 0. 94 0.96 0.98 0.86 0.51
2. 57 0. 33 3. 22 3. 23 0. 83 0.48 0.00 19. 53 5. 57 0. 90 0. 90 11. 02 0.90 1. 47 3. 23 1. 39 0. 35 0.73 0.89 0. 00 5. 14 3. 22 1. 57 1. 72 1.20 0.00 1. 43 1. 39 2. 05 3. 51 1. 87 0. 95 1. 51 1. 36 3. 59 0. 99 0.70 2. 01 2. 19 0.00 4. 36 3. 37 0.32 1.75 1.04 1.13
0. 18 0. 00 0. 00 1. 71 0. 00 0.00 0.00 1. 12 0. 00 1. 60 6. 77 0. 00 0.00 0. 00 0. 00 0. 49 0. 04 0.00 0.00 0. 09 0. 67 0. 00 0. 07 0. 00 0.00 0.56 0. 00 0. 23 0. 00 0. 00 0. 00 0. 65 0. 00 1. 65 0. 00 0. 15 0.00 0. 00 0. 00 0.00 0. 00 0. 35 0.00 0.24 0.23 0.00
2. 75 0. 33 3. 22 4. 94 0. 83 0.48 0.00 20. 66 5. 57 2. 50 7. 67 11. 02 0.90 1. 47 3. 23 1. 88 0. 39 0.73 0.89 0. 09 5. 80 3. 22 1. 64 1. 72 1.20 0.56 1. 43 1. 62 2. 05 3. 51 1. 87 1. 60 1. 51 3. 00 3. 59 1. 14 0.70 2. 01 2. 19 0.00 4. 36 3. 72 0.32 1.99 1.27 1.13
-0. 13 0. 33 0. 04 0. 00 0. 00 0.00 0.00 9. 70 0. 00 0. 00 -0. 01 0. 68 0.00 0. 69 0. 76 0. 00 -0. 22 0.30 0.31 0. 04 2. 45 1. 20 0. 46 0. 00 0.00 -0. 31 0. 00 -0. 25 0. 82 0. 59 -0. 52 0. 00 0. 41 0. 42 0. 00 0. 00 -5. 68 0. 67 1. 46 0.00 -0. 21 -3. 30 0.20 -0. 05 0.00 -0. 73
169
exhibit 6 (cont.)

Original Balance ($ mm)
Issue
JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., JP Morgan Commercial Mortgage Finance Corp., LB Commercial Mortgage Trust LB Commercial Mortgage Trust LB Commercial Mortgage Trust LB Commercial Mortgage Trust LB Commercial Mortgage Trust LB Commercial Mortgage Trust 1995-C2 1996-C2 1998-C1 1998-C4 1999-C1 1999-C2
19 96-C 2 19 96-C 3 19 97-C 5 19 98-C 6 19 99-C 7 19 99-C 8 20 00-C 9 20 00-C10
354. 3 451. 3 1039 .5 800. 9 808. 8 735. 9 817. 5 745. 4 261. 9 399. 5 1731 .1 2028 .5 1584 .9 893. 5 1305 .7 1001 .6 998. 9 310. 1 405. 0 211. 6 1075 .7 644. 7 648. 3 1170 .4 844. 5 687. 3 1093 .8 751. 0 640. 8 594. 4 373. 5 513. 8 272. 3 340. 5 640. 7 1107 .3 867. 1 632. 1 594. 0 689. 0 776. 8
LB-UBS Commercial Mortgage Trust, 2000-C3 LB-UBS Commercial Mortgage Trust, 2000-C4 LB-UBS Commercial Mortgage Trust, 2000-C5 Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage Merrill Ly nch Mortgage I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, I nves tors, 199 4-C 1 199 4-M1* 199 5-C 1 199 5-C 2 199 5-C 3 199 6-C 1 199 6-C 2 199 7-C 1 199 7-C 2 199 8-C 2 199 8-C 1-C T L 199 8-C 3 199 9-C 1
Midland Realty Acceptance Corp., 1996-C1 Midland Realty Acceptance Corp., 1996-C2 Morgan Stanley Capital I Inc., 1995-GAL1 Morgan Stanley Capital I Inc., 1996-C1 Morgan Stanley Capital I Inc., 1997-C1 Morgan Stanley Capital I Inc., 1998-CF1 Morgan Stanley Capital I Inc., 1999-RM1 Morgan Stanley Capital I Inc., 1999-FNV1 Morgan Stanley Capital I Inc., 1999-LIFE1 Morgan Stanley Capital I Inc., 2000-LIFE1 Morgan Stanley Dean Witter Capital I Inc., 2000-LIFE2
170
Factor
30, 60 & 90+
Forc. & REO
Total
164. 6 268. 3 834. 3 743. 5 765. 0 708. 9 788. 7 731. 1 141. 1 326. 6 1615 .2 1951 .6 1532 .7 874. 0 1289 .6 988. 7 974. 1 34. 8 50. 5 54. 1 214. 0 445. 8 444. 2 872. 2 701. 3 645. 3 994. 3 598. 5 595. 5 571. 2 247. 8 384. 8 119. 5 216. 3 488. 1 1030 .3 805. 9 610. 6 580. 7 676. 6 754. 2
0. 46 0.59 0. 80 0.93 0.95 0. 96 0. 96 0.98 0. 54 0. 82 0. 93 0. 96 0. 97 0.98 0. 99 0.99 0.98 0.11 0.12 0.26 0. 20 0. 69 0.69 0. 75 0. 83 0.94 0. 91 0. 80 0. 93 0. 96 0. 66 0.75 0. 44 0. 64 0. 76 0. 93 0.93 0.97 0.98 0.98 0.97
4. 21 0.00 2. 48 1.50 0.00 4. 99 4. 99 0.51 6. 51 11. 66 2. 38 0. 00 0. 36 0.00 0. 04 0.00 0.00 0.00 0.00 0.00 0. 00 4. 33 0.00 8. 62 8. 44 1.23 1. 34 1. 78 0. 00 3. 94 2. 76 1.34 0. 00 2. 76 2. 33 4. 02 0.42 0.36 0.00 0.00 0.00
0. 00 0.00 0. 87 0.00 0.00 0. 30 0. 00 0.00 0. 00 0. 00 0. 00 0. 00 0. 49 0.00 0. 00 0.00 0.00 0.00 0.00 0.00 19. 63 0. 84 1.95 2. 20 0. 96 0.46 1. 20 0. 38 4. 27 2. 62 2. 53 0.00 6. 49 0. 00 0. 00 3. 65 0.00 1.31 0.00 0.00 0.00
4. 21 0.00 3. 35 1.50 0.00 5. 29 4. 99 0.51 6. 51 11. 66 2. 38 0. 00 0. 84 0.00 0. 04 0.00 0.00 0.00 0.00 0.00 19. 63 5. 17 1.95 10. 81 9. 40 1.69 2. 54 2. 16 4. 27 6. 56 5. 29 1.34 6. 49 2. 76 2. 33 7. 68 0.42 1.67 0.00 0.00 0.00
0. 00 0.00 0. 10 0.52 0.00 0. 29 -0. 21 -1. 15 4. 85 7. 25 0. 00 0. 00 0. 00 0.00 -0. 05 -0. 08 0.00 0.00 -1. 98 0.00 -3. 04 2. 55 -5. 69 0. 09 1. 43 0.21 0. 31 -0. 75 -2. 05 0. 14 0. 00 0.00 0. 01 0. 00 0. 00 -0. 05 0.00 0.22 0.00 0.00 0.00
171
exhibit 6 (cont.)

Original Balance ($ mm)
Issue
Morgan Stanley Capital I Inc., 1996-W F 1 Morgan Stanley Capital I Inc., 1997-W F 1 Morgan Stanley Capital I Inc., 1998-W F 1 Morgan Stanley Capital I Inc., 1998-W F 2 Morgan Stanley Capital I Inc., 1999-W F 1 Morgan Stanley Capital I Inc., 1997-H F 1 Morgan Stanley Capital I Inc., 1998-H F 1 Morgan Stanley Capital I Inc., 1998-H F 2 Mortgage Capital Funding Inc., 19 95-M C 1 Mortgage Capital Funding Inc., 19 96-M C 1 Mortgage Capital Funding Inc., 19 96-M C 2 Mortgage Capital Funding Inc., 19 97-M C 1 Mortgage Capital Funding Inc., 19 97-M C 2 Mortgage Capital Funding Inc., 19 98-M C 1 Mortgage Capital Funding Inc., 19 98-M C 2 Mortgage Capital Funding Inc., 19 98-M C 3 Nations Link Funding Corp. 1996-1 Nations Link Funding Corp. 1998-1 Nations Link Funding Corp. 1998-2 Nations Link Funding Corp. 1999-1 Nations Link Funding Corp. 1999-2 Paine Webber Mortgage Acceptance Corp. V, 1999 -C1 Penn Mutual Life Insurance Co. Series, 1996-PML PNC Mortgage Acceptance Corp., 1999-CM1 PNC Mortgage Acceptance Corp., 2000-C1 PNC Mortgage Acceptance Corp., 2000-C2 Prudential Securities Financing Corp., 1995-MCF 2 Prudential Securities Financing Corp., 1998-C1 Prudential Securities Secured Financing Corp., 1999-NRF1 Prudential Securities Secured Financing Corp., 1999-C2 Prudential Securities Secured Financing Corp., KEY 2000-C1 Salomon Brothers Mortgage Securities VII Inc., Salomon Brothers Mortgage Securities VII Inc., Salomon Brothers Mortgage Securities VII Inc., Salomon Brothers Mortgage Securities VII Inc., Salomon Brothers Mortgage Securities VII Inc., Salomon Brothers Mortgage Securities VII Inc., 1996-C 1 1999-C 1 2000-N L1 2000-C 1 2000-C 2 2000-C 3
605. 4 559. 2 1392 .2 1062 .0 968. 5 622. 4 1283 .7 1066 .3 231. 4 490. 9 460. 3 660. 4 872. 3 1298 .7 1012 .5 915. 0 324. 2 1024 .1 1592 .1 1228 .4 1167 .6 700. 5 781. 6 763. 4 808. 6 1083 .3 222. 8 1154 .6 934. 5 875. 1 821. 2 213. 2 744. 7 340. 0 719. 9 787. 1 923. 0 277. 0 454. 7 525. 0 234. 8 280. 7 $161,258.2
Southern Pacific Thrift & Loan Association 1996-C 1 Structured Assets Securities Corp., Structured Assets Securities Corp., Structured Assets Securities Corp., Structured Assets Securities Corp., Total/Weighted A verage 1994-C 1 1995-C 1 1995-C 4 1996-C 3
172
Factor
30, 60 & 90+
Forc. & REO
Total
484. 2 465. 5 1282 .5 1002 .2 919. 5 476. 8 1173 .5 1009 .2 56. 9 354. 2 364. 3 589. 2 799. 4 1224 .0 940. 6 853. 4 171. 3 908. 6 1487 .3 1177 .5 968. 2 663. 0 286. 2 745. 4 786. 9 1063 .8 119. 6 1044 .8 874. 7 835. 4 804. 1 94. 3 707. 3 296. 9 700. 3 771. 0 904. 1 64. 3 51. 6 148. 7 51. 6 60. 0 $143,271.1
0.80 0.83 0. 92 0. 94 0.95 0.77 0. 91 0. 95 0.25 0.72 0. 79 0. 89 0.92 0. 94 0.93 0.93 0. 53 0. 89 0. 93 0. 96 0.83 0. 95 0.37 0.98 0.97 0. 98 0. 54 0. 90 0.94 0.95 0.98 0. 44 0. 95 0.87 0.97 0. 98 0.98 0.23 0. 11 0. 28 0.22 0.21 0.89
0.42 0.00 0. 74 0. 00 0.52 0.00 1. 40 0. 90 0.00 0.29 6. 05 3. 67 0.17 0. 54 1.80 1.36 6. 12 3. 80 1. 45 0. 00 0.00 2. 04 0.00 1.03 1.25 0. 38 5. 67 0. 50 0.00 0.60 0.00 21. 78 2. 02 1.21 0.40 2. 26 1.31 0.10 0. 00 9. 91 0.22 0.00 1.60
0.33 1.71 0. 00 0. 00 0.00 0.00 0. 16 0. 00 0.00 0.00 0. 00 0. 91 0.29 0. 95 0.00 0.00 0. 00 0. 17 0. 54 0. 36 0.00 0. 42 1.22 0.00 0.00 0. 00 6. 96 0. 58 0.25 0.32 1.18 6. 63 0. 00 0.00 0.00 0. 00 0.00 0.00 11. 02 0. 00 0.00 0.00 0.48
0.75 1.71 0. 74 0. 00 0.52 0.00 1. 56 0. 90 0.00 0.29 6. 05 4. 58 0.46 1. 48 1.80 1.36 6. 12 3. 97 1. 99 0. 36 0.00 2. 46 1.22 1.03 1.25 0. 38 12. 63 1. 08 0.25 0.93 1.18 28. 41 2. 02 1.21 0.40 2. 26 1.31 0.10 11. 02 9. 91 0.22 0.00 2.09
0.00 0.00 0. 00 0. 00 0.00 -0. 22 -0. 22 -2. 30 0.00 0.29 3. 92 0. 02 -0. 90 0. 20 -0. 53 -0. 25 3. 80 0. 07 1. 06 0. 00 0.00 0. 19 0.01 0.00 0.00 -0. 04 0. 01 0. 00 0.00 0.00 0.00 -0. 03 0. 66 0.00 -0. 19 0. 41 0.00 -1. 07 0. 84 9. 91 0.00 -19. 01 0.08
173
II. More Details
The Morgan Stanley Quarterly Tracking report provides details on specially serviced loans in Morgan Stanley sponsored transactions. In this issue of our Tracking Report, we expand our coverage of specially serviced loans within Morgan Stanley sponsored transactions. Details on 58 loans accounting for $355 million in current balances are provided in this report. We also add an appendix that shows the delinquency history on each highlighted transaction. Specially serviced loans are loans that have been transferred to the special servicer because of delinquent payments or non-compliance with loan document requirements. Typically, specially serviced loans are 45-60 days delinquent before they are transferred for late payment. This report focuses on specially serviced loans and does not include loans that may be delinquent but not yet transferred to the special servicer. Loans may move in and out of the 30-day delinquency category and not become specially serviced. The information in this report is based on October remittance report data and conversations with the special servicers over the past several weeks. The universe of transactions examined is contained in Exhibit 1.
exhibit 1
SPECIALLY SERVICED LOAN CATEGORY

Number of Loans Current Balance ($mm) % of Total Current Balance
Property Type
Office Retail Senior Housing Hotel Multifamily Industrial Mobile Home Self Storage Total
8 9 20 10 3 5 2 1 58
80.0 84.0 119.4 38.5 9.2 20.8 2.7 1.0 355.6
22.5 23.6 33.6 10.8 2.6 5.9 0.8 0.3 100
174
Based on investor inquiries, this report provides more details on various loans than our previous publication. The following criteria were used to determine which specific loans would be included in the report: All specially serviced loans within transactions issued off Morgan Stanleys shelf where the total current balance of specially serviced loans was greater than $2 million. All specially serviced loans over $8 million, within transactions with current balances over $300 million, not issued off Morgan Stanleys shelf but where Morgan Stanley served as an underwriter or placement agent. All specially serviced loans originated by Morgan Stanley, regardless of the loan balance. Based on the above criteria, 67% of the specially serviced balances within the 31 transactions were analyzed. Fifteen transactions examined were issued off Morgan Stanleys shelf and this report covers 96% of the specially serviced balances within those transactions. In terms of property type concentration, senior housing properties accounted for the largest dollar balance of loans examined, $119 million in current balances, followed by retail at $84 million and office at $80 million. Twenty senior housing loans are contained in this report, the largest number by loan count, followed by hotel properties, which total 10 of the 58 loans we examine.
175
exhibit 2
MORGAN STANLEY TRANSACTIONS (OCTOBER 2001 REMITTANCE REPORTS)

Original Balance ($mm) Current Balance ($mm)
Issue
Transactions Issued Off Morgan Stanley's Shelf Morgan Stanley Capital I, 1995-GAL1 Morgan Stanley Capital I, 1996-C1 Morgan Stanley Capital I, 1996-MBL1 Morgan Stanley Capital I, 1996-WF1 Morgan Stanley Capital I, 1997-ALIC Morgan Stanley Capital I, 1997-C11 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, 1999-FNV1 Morgan Stanley Capital I, 1999-RM1 Morgan Stanley Capital I, 1999-WF1 Subtotal Other Morgan Stanley Sponsored Transactions Allied Capital Commercial Mortgage Trust, 1998-1 Bear Stearns Commercial Mortgage Securities, 1999-WF2 CAPCO America Securitization Corp, 1998-D7 Commercial Mortgage Acceptance Corp., 1996-C2 1 Commercial Mortgage Acceptance Corp., 1998-C1 Commercial Mortgage Acceptance Corp., 1999-C1 Deutsche Mortgage and Asset Receiving Corp., 1998-C1 First Union Commercial Mortgage Trust, 1999-C1 Heller Financial Commercial Mortgage Asset Corp., 2000-PH1 IMPAC CMB Trust, 1998-C1 JP Morgan Commercial Mortgage Finance Corp., 1997-SPTL-C1 KS Mortgage Capital, 1995-1 Merrill Lynch Mortgage Investors, 1995-C2 Nomura Asset Securities Corp., 1998-D6 Prudential Securities Secured Financing Corp., 1999-NRF1 Southern Pacific Thrift and Loan, 1996-C1 Subtotal Total
1
272.3 340.5 128.3 605.4 802.7 640.7 622.4 559.2 1,107.3 1,283.7 1,066.3 1,392.2 632.1 867.1 968.5
122.1 224.6 13.5 500.3 502.5 498.7 505.6 483.2 1,036.0 1,179.7 1,015.4 1,293.3 614.1 811.1 924.8
11,288.6 9,724.9
334.7 1,088.0 1,249.0 184.5 1,192.2 736.5 1,821.8 1,171.2 960.7 320.2 204.5 165.5 1,075.7 3,722.7 934.5 277.0
122.0 1,041.4 1,185.9 86.9 1,090.5 712.8 1,726.0 1,118.7 943.6 281.6 66.4 80.6 229.2 3,576.3 880.7 73.1
15,438.7 13,215.5 26,727.3 22,940.4
Previously reported specially serviced loans that have paid off. Loan level detail is not provided for the shaded transactions. Source: Intex, Remittance Reports
176
Delinquency Data (%)
Specially Serviced Data ($mm) Spec. Serv. Loans Spec. Serv. Loans % in this Report Covered
Factor
30, 60, Forc. & 90+ REO
Total
0.45 0.66 0.11 0.83 0.63 0.78 0.81 0.86 0.94 0.92 0.95 0.93 0.97 0.94 0.95 0.86
0.00 6.36 6.36 1.96 0.71 2.68 0.00 53.54 53.54 0.00 0.32 0.32 0.00 0.00 0.00 0.00 0.74 0.74 0.21 0.00 0.21 0.00 1.65 1.65 4.87 1.63 6.50 1.39 0.16 1.56 1.17 0.09 1.26 0.16 0.00 0.16 0.00 1.46 1.46 0.42 0.00 0.42 0.53 0.00 0.53 0.97 0.60 1.57
7.8 5.7 10.0 15.9 40.9 0.0 1.1 8.0 93.7 22.1 0.9 9.8 9.1 3.4 5.0 233.3
7.8 1.6 7.1 15.9 40.9 0.0 0.0 8.0 93.7 22.1 0.0 9.8 9.1 3.4 5.0 224.4
100 28 71.0 100 100 0 0 100 100 100 0 100 100 100 100 96.2
0.36 0.96 0.95 0.47 0.91 0.97 0.95 0.96 0.98 0.88 0.32 0.49 0.21 0.96 0.94 0.26 0.86 0.86
5.52 5.42 10.95 0.26 0.53 0.79 1.28 0.00 1.28 0.00 8.99 8.99 0.94 0.00 0.94 0.94 0.00 0.94 4.16 0.57 4.72 1.08 0.00 1.08 0.68 0.06 0.73 1.60 0.22 1.83 3.10 0.00 3.10 0.00 0.00 0.00 0.53 19.14 19.67 0.24 0.00 0.24 0.00 0.24 0.24 3.91 0.00 3.91 1.14 1.07 0.58 0.59 1.72 1.66
13.4 37.9 15.1 0.0 10.3 7.5 97.8 12.1 18.0 11.1 3.9 3.6 43.9 18.2 2.7 5.0 300.4 533.7
0.0 18.0 12.1 0.0 5.7 0.0 45.6 0.0 10.0 0.0 0.0 0.0 30.2 9.7 0.0 0.0 131.2 355.6
0 48 80 0 56 0 47 0 55 0 0 0 69 53 0 0 43.7 66.6
177
SAMPLE MONITORING REPORT
Des Peres Cinema
ORIGINAL BALANCE: CURRENT BALANCE: PERCENT OF POOL BALANCE: MASTER SERVICER: SPECIAL SERVICER: TRUSTEE: $8,700,000 $8,179,861 0.79% WELLS FARGO BANK GMAC COMMERCIAL MORTGAGE CORP. WELLS FARGO BANK
property description
valuation information
Status: Current Paid Through Date: October 2001 Location: Des Peres, Missouri Size: 89,870 sf Year Built: 1997 Property Type: Retail Movie Theater Major Tenants (% of sf): Des Peres Cinema (100%) Originator: Wells Fargo Bank
Appraisal Value: $12,500,000 Appraisal Date: June 17, 1998 Appraisal Cap. Rate: Not Available Market Cap. Rate: Not Available Original LTV: 68.9% Original Loan per Unit: $97 Current LTV: 65.4% Current Loan per Unit: $91 Underwritten DSCR: 1.40 Current DSCR: 1.56 (December 1999)
market data
Closest MSA: Market Average Occupancy for Property Type: Market Average Rent for Property Type:
St. Louis, Missouri 15.8 Miles NA $16
escrow information
Reserves in Place (Type): Replacement Reserves Replacement Reserve Balance: $0.28/sf Taxes and Insurance Balance: Not Required Leasing Commission/Tenant Improvements Balance: Not Available Other Reserve Balance: Defeasance Reserve Required at $500,000 Annually for Crossed Pool.
178
REASON FOR SPECIAL SERVICING TRANSFER
This loan is cross-collateralized with loans on four other cinemas within the pool: St. Charles Cinema, OFallon 15 Cinema, St. Clair 10 Cinema, and Halls Ferry 14 Cinema. The combined current balance of these assets is $29,614,954, which makes up about 2.8% of the pool. All the cross-collateralized loans are in special servicing. The loans were transferred to special servicing because the borrower, Wehrenberg Theatres, filed for bankruptcy protection in February 2001.
U P D AT E
The loan is current as of the October remittance report. The borrower is expected to be out of bankruptcy on November 30, 2001. All of the cross-collateralized leases have been affirmed by the bankruptcy hearings.
C U R R E N T S TAT U S O F R E S O L U T I O N
A modification on the lease for the Halls Ferry property is in the documentation process. Under the modified terms, the Halls Ferry property, which is underperforming, will pay a zero-base rent with any cash flow going toward debt service. The shortfall of the base rent from the Halls Ferry property will be made up from the other cross-collateralized locations resulting in the same cash flows to the trust as the original loan. Additionally, the borrower is required to pay $500,000 annually into a reserve to defease the Halls Ferry loan. The balance on the Halls Ferry loan is $2.4 million. Once the $2.4 million is escrowed, the Halls Ferry property will be defeased.
A S S E T S U M M A RY
Wehrenberg Theatres currently operates 29 movie cinemas in Illinois, Missouri, and the Southeast. The company was founded in 1906 and filed for bankruptcy protection in order to seek relief from leases at properties unrelated to this pool. Recent site inspections indicate that Des Peres Cinema, St. Charles Cinema, OFallon 15 Cinema, and St. Clair 10 Cinema are in better condition than the Halls Ferry property.
M AT U R I T Y D AT E
January 1, 2019
179
CMBS Delinquencies by Originator

In addition to monitoring delinquencies by property type and geographic distribution, it is also helpful to track delinquency trends by originator. We update our study examining delinquencies on CMBS loans by issuer type Life insurance company loans posted a 19 bp decline in delinquencies, and the lowest rate by issuer type Our previous study was based on August 2001 remittance report data. In this study, we examined January 2002 remittance reports to identify any changes or trends from our previous study. Since the last study, delinquencies on our entire universe have increased 41 bp to 1.23% of current balances. Delinquencies on seasoned collateral (aged over one year) increased 71 bp to 2.01% of current balances. During the same period, 60+ days delinquent loans increased 27 bp to 0.93% of current balances. The review produced the following conclusions: The only originator type to post lower delinquency rates was life insurance companies, declining 19 bp to 0.16% of current balances. Finance company collateral posted the greatest increase in delinquencies, rising 51 bp since our last study to 1.40% of current balances. Investment bank collateral reported the overall highest delinquencies, rising 44 bp to 1.46% of current balances. Commercial bank and finance company loans had the greatest dispersion in delinquencies. In total, the study examines $228 billion of CMBS collateral within 434 different transactions. Seventy-seven different companies contributed collateral to the Intex database. Originators with less than $300 million of collateral in the database were screened from the data set.
L I M I TAT I O N S
Our study of delinquencies is only a snapshot at one point in time and does not include cumulative loss data. Our analysis is based on information from Intexs database. Typically, if a loan is paid off within the database, the original balance remains, and the current balance shows a zero balance. However, in a few cases we found that some loans which were paid off were completely eliminated from the database. Lastly, our overall delinquency numbers include 30-day delinquencies. Often loans move in and out of 30 days due to timing of payments rather than fundamental deterioration.
180
exhibit 1
%
1.60
DELINQUENCIES BY INSTITUTION TYPE (BASED ON JANUARY REMITTANCE REPORTS)
1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00
% 30 Days Delinq % 60 Days Delinq % 90 Days Delinq % Foreclosure % REO
1.40% 1.12%
1.46%
0.16%
Life Insurance Company
Commercial
Finance Company
Investment Bank
exhibit 2
$mm
140 120 100 80 60 40 20 0 Life Insurance Company Commercial Bank Finance Company Investment Bank
1971-1994 1995 1996 1997 1998 1999 2000 2001
CURRENT BALANCE BY INSTITUTION TYPE AND COHORT YEAR (BASED ON JANUARY REMITTANCE REPORTS)
181
exhibit 3
DELINQUENCIES BY INSTITUTION TYPE (BASED ON JANUARY REMITTANCE REPORTS)

Number of Issuers Total Current Balance (%) Original Balance ($bn)
Originator Type
Commercial Bank Finance Company Investment Bank Life Insurance Company Total/Weighted Average
13 29 22 13 77
19.9 23.0 51.6 5.5 100.0
50.9 59.6 133.0 16.6 260.1
For the purposes of this study, each deal is broken down to the loan level. Therefore, the origination year provided is the year in which the loan was originated not the year in which the CMBS transaction was issued. In addition, the data is sorted by loan originator, so if several originators contributed collateral to one CMBS transaction the delinquencies attributable to each specific loan are assigned to the respective originator.
L I F E C O L L AT E R A L P O S T S A D E C L I N E
In the January remittance report data we examined, collateral originated by insurance companies continued to post the lowest delinquency rate. The total delinquency rate for life insurance collateral was 0.16%, followed by commercial bank collateral at 1.12%, finance company collateral at 1.40%, and investment bank collateral at 1.46%. In aggregate we found that although delinquencies in our CMBS universe rose 41 bp since the last study, delinquencies on life insurance collateral declined 19 bp. This was the only originator type to post a decline. Our findings agree with the data released on March 5, 2002 by the American Council of Life Insurers (ACLI). They reported that as of December 31, 2001, delinquencies on life insurance company collateral declined 7 bp to 0.12% of current balances. ACLI defines delinquent loans as over 60+days past due and does not include loans that are in the process of foreclosure in their definition.
182
Current Balance ($bn)
30 Days Del (%)
60 Days Del (%)
90 Days Foreclosure Del (%) (%) REO (%)
Total (%)
45.4 52.4 117.6 12.6 228.0
0.30 0.35 0.31 0.04 0.30
0.09 0.20 0.15 0.04 0.14
0.49 0.57 0.56 0.01 0.52
0.07 0.07 0.09 0.05 0.08
0.17 0.21 0.35 0.02 0.27
1.12 1.40 1.46 0.16 1.31
L O W E S T D E L I N Q U E N C Y R AT E S
Twelve originators had delinquency rates below 0.25%. Four were commercial banks, three were investment banks, three were life insurance companies, and two were finance companies. All have been in the business of originating loans for at least ten years with the exception of UBS which began originating in 1999.
exhibit 4
LARGE ORIGINATORS WITH DELINQUENCIES LESS THAN 0.25% (BASED ON JANUARY REMITTANCE REPORTS)
Originator Type Total Del. (%) Current Earliest Balance Orig ($mm) Year
Originator Name
Bank of America Capital Lease Chase Goldman Sachs Keybank Merrill Lynch Canada Principal Prudential Secore TIAA UBS Wells Fargo
Commercial Bank Finance Company Commercial Bank Investment Bank Commercial Bank Investment Bank Life Insurance Company Life Insurance Company Finance Company Life Insurance Company Investment Bank Commercial Bank
0.17 0.00 0.25 0.08 0.07 0.00 0.00 0.19 0.23 0.00 0.05 0.24
7,326.0 1,182.3 6,360.0 3,785.2 1,792.1 1,076.1 2,434.6 3,786.1 7,397.9 1,121.4 2,101.2 5,961.7
1973 1995 1992 1997 1997 1997 1987 1998 1996 1979 1999 1990
183

H I G H E S T D E L I N Q U E N C Y R AT E S
There were thirteen originators that had delinquency rates above 2.5%. One life insurance company, one commercial bank, five investment banks, and six finance companies.
exhibit 5
ORIGINATORS WITH DELINQUENCIES OVER 2.50% (BASED ON JANUARY REMITTANCE REPORTS)

Most Current Current Orig Balance Year ($mm)
Originator Name
Originator Type
Total Del. (%)
Allied Amresco Conti CSFB DLJ Conduit Legg Mason Merrill Lynch Mutual Benefit Life PaineWebber Provident RMF Smith Barney WMF
Finance Company Finance Company Finance Company Investment Bank Investment Bank Finance Company Investment Bank Life Insurance Company Investment Bank Commercial Bank Finance Company Investment Bank Finance Company
10.94 120.1 3.90 1,989.2 3.20 2,067.9 3.59 10,080.1 4.23 3,217.5 2.62 665.6 4.16 6,767.8 7.11 41.7 3.63 1,592.6 22.35 221.3 22.19 403.0 3.11 329.3 5.22 867.4
2001 1998 1999 2000 1998 2000 2001 1995 2000 1999 1998 1997 1998
Exhibits 6-9 show our data sorted by originator type. Exhibit 10 contains an alphabetical listing of delinquencies by originator broken out by year.
184
exhibit 6
COMMERCIAL BANK DELINQUENCIES BASED ON JANUARY REMITTANCE REPORTS

Current Balance ($mm) 30 Days Del (%) 60 Days Del (%) 90 Days Del (%) Foreclosure (%) REO (%) Total (%) Change From Aug (%)
Name Originator
Originator Type
Original Balance ($mm)
Commercial Bank Commercial Bank Commercial Bank Commercial Bank Commercial Bank 432.73 425.76 12,496.72 11,740.12 1,822.33 1,792.13 1,169.10 1,153.63 5,395.81 4,899.46 986.18 975.09 1,078.87 221.32 6,676.63 5,961.70 50,854.38 45,373.02 0.00 0.43 0.00 1.19 0.71 0.00 3.35 0.11 0.30 0.00 0.10 0.00 0.00 0.23 0.00 0.00 0.00 0.09 0.00 1.33 0.00 0.88 0.50 0.00 0.00 0.00 0.49 0.00 0.10 0.07 0.00 0.00 0.00 2.00 0.13 0.07 0.00 0.14 0.00 0.00 0.37 0.00 17.00 0.00 0.17 0.00 2.10 0.07 2.08 1.81 0.00 22.35 0.24 1.12
1,277.66 8,364.93 1,003.18 6,877.08 3,273.14
948.44 7,326.02 702.72 6,360.00 2,866.63
0.26 0.11 0.00 0.00 0.45
0.15 0.04 0.00 0.20 0.00
0.06 0.02 0.00 0.00 0.98
0.00 0.00 0.00 0.00 0.15
0.00 0.00 0.00 0.05 0.11
0.47 0.17 0.00 0.25 1.69
0.03 0.12 0.00 0.20 0.59 0.00 0.43 0.03 2.08 0.56 0.00 3.15 0.10 0.26
Banc One Bank Of America CDC Chase Citicorp Commercial Mortgage Origination Company of Canada First Union Keybank Lasalle Nationsbank PNC Provident Wells Fargo Total/Weighted Average
Commercial Bank Commercial Bank Commercial Bank Commercial Bank Commercial Bank Commercial Bank Commercial Bank Commercial Bank
185

exhibit 7
FINANCE COMPANY DELINQUENCIES BASED ON JANUARY REMITTANCE REPORTS

Original Balance ($mm) Current Balance ($mm)
Name Originator
Originator Type
Allied Amresco Artesia Boston Capital Bridger Capital Lease CBA CDPQ Central Park Conti Criimi Mae Finova GE Capital GMAC Greenwich Heller Impac Legg Mason LTC Midland National Realty NCB Residential Funding RMF Secore Southern Pacific Starwood Value Line WMF Total/Weighted Average
Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company Finance Company
341.30 2,518.58 666.62 315.92 680.52 1,276.91 581.58 508.14 516.04 2,505.73 805.26 683.09 7,256.57 13,606.49 3,732.12 3,582.75 320.17 821.23 356.95 4,185.55 1,500.13 470.34 1,515.72 425.08 7,975.44 778.89 452.02 335.60 901.99 59,616.71
120.14 1,989.16 652.43 296.29 670.44 1,182.27 103.51 353.52 493.75 2,067.90 684.27 662.12 7,035.50 12,094.71 3,526.18 3,280.60 275.15 665.61 212.74 3,451.56 1,396.25 316.55 1,458.87 402.97 7,397.91 280.98 382.65 126.63 867.37 52,448.01
186
30 Days Del (%)
60 Days Del (%)
90 Days Del (%)
Foreclosure (%)
REO (%)
Total (%)
Change From Aug (%)
0.00 0.63 0.00 0.00 0.00 0.00 0.00 0.00 0.65 0.44 0.00 0.00 0.12 0.65 0.44 0.16 0.53 0.67 0.00 0.20 0.00 0.04 1.14 0.00 0.23 0.90 0.00 0.00 0.00 0.35
0.00 2.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.52 0.00 0.00 0.08 0.26 0.06 0.03 0.00 0.00 0.00 0.17 0.04 0.00 0.00 0.00 0.00 0.47 0.00 0.00 0.00 0.20
5.59 0.88 0.00 2.27 0.00 0.00 0.00 0.00 0.64 1.42 0.53 0.00 0.08 0.59 0.29 0.07 0.51 1.74 0.95 0.79 0.64 0.00 0.13 21.27 0.00 0.41 0.00 0.00 0.00 0.57
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.00 0.08 0.00 0.14 0.00 0.23 0.00 0.00 0.21 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.19 0.07
5.35 0.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.77 0.00 1.34 0.00 0.19 0.00 0.06 0.00 0.21 0.00 0.57 0.00 0.00 0.03 0.92 0.00 0.07 0.00 0.00 3.03 0.21
10.94 3.90 0.00 2.27 0.00 0.00 0.00 0.00 1.29 3.20 0.53 1.34 0.37 1.69 0.93 0.32 1.27 2.62 0.95 1.95 0.68 0.04 1.30 22.19 0.23 1.85 0.00 0.00 5.22 1.40
-2.14 3.34 0.00 2.27 -0.32 0.00 0.00 0.00 1.06 1.52 -1.23 -0.12 -0.01 0.46 0.30 0.16 -0.64 1.22 0.09 0.51 0.05 0.04 0.91 0.06 -0.01 -0.45 0.00 -1.47 0.00 0.38
187

exhibit 8
INVESTMENT BANK DELINQUENCIES BASED ON JANUARY REMITTANCE REPORTS

Original Balance Current Balance ($mm) ($mm)
Name Originator
Originator Type
Archon Bankers Trust Bear Stearns CIBC Column CSFB Daiwa DLJ Conduit GACC Goldman Sachs Goldman Sachs/Bank Of America Lehman Merrill Lynch Merrill Lynch Canada Morgan Guaranty Morgan Stanley Nomura Nomura Conduit Painewebber Salomon Brothers Smith Barney UBS Total/Weighted Average
Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank Investment Bank
2,172.00 1,265.21 4,783.56 2,531.14 12,904.50 15,207.11 1,090.14 3,541.53 11,749.29 4,094.52 455.00 23,321.62 7,427.86 1,114.71 8,001.47 9,456.34 4,195.41 12,131.51 1,759.07 3,159.05 473.67 2,119.37 132,954.06
2,119.48 817.38 4,601.38 2,479.08 12,065.75 10,080.10 813.99 3,217.46 10,374.62 3,785.24 454.05 20,441.67 6,767.82 1,076.05 7,580.71 8,920.50 3,441.84 11,506.82 1,592.55 3,017.43 329.34 2,101.20 117,584.45
188
30 Days Del (%)
60 Days Del (%)
90 Days Foreclosure Del (%) (%)
REO (%)
Change Total From Aug (%) (%)
0.55 0.00 0.06 0.07 0.53 0.69 0.23 0.42 0.10 0.08 0.00 0.10 0.54 0.00 0.33 0.56 0.52 0.08 1.23 0.00 2.04 0.03 0.31
0.08 0.00 0.07 0.00 0.51 0.04 0.43 1.14 0.04 0.00 0.00 0.03 0.14 0.00 0.22 0.07 0.00 0.00 0.33 0.68 0.00 0.00 0.15
0.93 0.00 0.06 0.19 0.40 0.29 0.09 1.66 0.37 0.00 0.00 0.35 2.90 0.00 0.77 0.39 0.18 0.61 0.96 0.28 1.07 0.02 0.56
0.00 0.00 0.00 0.07 0.02 0.56 0.93 0.00 0.08 0.00 0.00 0.03 0.09 0.00 0.03 0.00 0.00 0.00 0.00 0.31 0.00 0.00 0.09
0.00 0.00 0.23 0.00 0.17 2.01 0.60 1.01 0.00 0.00 0.00 0.01 0.48 0.00 0.00 0.20 0.92 0.39 1.12 0.00 0.00 0.00 0.35
1.56 0.00 0.43 0.33 1.63 3.59 2.27 4.23 0.59 0.08 0.00 0.51 4.16 0.00 1.34 1.21 1.63 1.08 3.63 1.28 3.11 0.05 1.46
0.92 0.00 0.25 0.16 0.67 0.89 0.26 2.50 0.18 0.08 0.00 0.08 2.98 0.00 0.77 0.68 0.22 0.09 1.02 0.88 1.72 0.03 0.57
189

exhibit 9
LIFE INSURANCE COMPANY DELINQUENCIES BASED ON JANUARY REMITTANCE REPORTS
Name Originator
Originator Type
Original Balance ($mm)
Current Balance 3 ($mm) D
Aetna CIGNA Confederation Life General American Jackson National Life John Hancock Mass Mutual Mutual Benefit Life Penn Mutual Principal Protective Life Prudential TIAA Total/Weighted Average
Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company Life Insurance Company
1,245.12 629.88 377.00 375.53 1,926.60 427.47 777.50 606.24 623.57 592.07 1,419.52 1,342.12 709.29 667.93 526.52 41.70 781.56 288.25 2,500.59 2,434.57 554.13 278.69 3,873.88 3,786.06 1,314.50 1,121.38 16,629.78 12,591.90
190
30 Days Del (%)
60 Days Del (%)
90 Days Del (%)
Foreclosure (%)
REO (%)
Total (%)
Change From Aug (%)
0.00 0.00 0.00 0.00 0.00 0.34 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.04
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.13 0.00 0.04
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.01
0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.11 1.21 0.00 0.00 0.00 0.00 0.05
0.00 0.00 0.63 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02
0.00 0.00 0.63 0.00 0.00 0.34 0.00 7.11 1.21 0.00 0.00 0.19 0.00 0.16
0.00 0.00 -3.47 0.00 0.00 0.34 0.00 -10.21 0.07 0.00 -1.34 0.04 0.00 -0.19
191
Factors to Consider
MACROECONOMIC
Economic/Interest rate outlook
Investors should be aware of the growth outlook for the US 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 spreads
Investment-grade CMBS and swap spreads are highly correlated. The swap spread represents the price of exchanging a fixed-rate cash flow for a floating-rate 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 US Treasury (UST) yield plus 80 bp to the counterparty to receive a floating-rate (LIBOR) cashflow. 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 risk
CMBS investors should also have a view of the global economy and potential effects of global credit risk on US fixed-income markets. In 1998, the Russian debt crisis had a major impact on US markets, including CMBS.
R E A L E S TAT E FA C T O R S
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; flowof-funds data on commercial and multifamily originations)
192
US Census bureau (construction) Torto Wheaton Research (vacancy data by property type and market) 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) Moodys, Standard & Poors, Fitch (periodic reports on real estate and CMBS)
R E L AT I V E V A L U E
Investors use several benchmarks for CMBS spreads: Single-A bank and financeFormerly used as a benchmark for AAA CMBS; now used for both AAAs, AAs, and single-As. Unsecured REITsBenchmark for BBBs and BBB- CMBS. Single-A corporate industrialsBenchmark for investment-grade CMBS. Mortgage pass-through OASComparison for AAAs. ABSBenchmark 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.
BOND-SPECIFIC
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 US market are Moodys, Standard and Poors (S&P), and Fitch. Some investors require that either Moodys 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.
193
Factors to Consider
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 issuers 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 issuers 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. They cost about $2500/month.
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 premium on the security.
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.
194
ERISA-eligibility
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. 30-day delinquency rates are of lesser concern than 60 day+ rates.
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 support 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.
exhibit 1
CREDIT ENHANCEMENT PNCMA 2001-C1 A2

Cpn 5.91 6.36 6.58 6.80 2.57 6.93 5.91 5.91 5.91 5.91 Type Fixed Rate Fixed Rate Fixed Rate Fixed Rate Fixed Rate IO, Other Non-Fi Fixed Rate WAC/Pass Thru WAC/Pass Thru WAC/Pass Thru Fixed Rate Fixed Rate Fixed Rate Fixed Rate At Issuance Rating 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 Credit Enhancement 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%
Class A1 A2 B C1 C2 C2X D E F G H J K L
Current Balance 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
The sources for exhibits 1-6 are cashflow runs on Trepp LLC analytics on Bloomberg.
195
Factors to Consider
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 No. 1 2 3 4 5 6 7 8 9 10 11 12 Date 8/01 8/02 8/03 8/04 8/05 8/06 8/07 8/08 8/09 8/10 8/11 8/12
PREPAYMENT PENALTY MATRIX PNCMA 2001-C1 A2

LO 97.42% 96.51% 94.24% 92.85% 92.90% 90.87% 92.46% 91.90% 91.17% 71.60% 100.00% 100.00% YM 2.58% 3.49% 5.76% 7.15% 5.88% 7.41% 7.17% 6.39% 6.48% 2.40% 0.00% 0.00% PP>=5% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% PP>=4% PP>=3% PP>=2% PP>=1% 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% 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% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% None 0.00% 0.00% 0.00% 0.00% 1.22% 1.72% 0.37% 1.71% 2.35% 26.01% 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. CPR or sometimes CPY, 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

0.00 6.172 9.35 +45.02 15.00 6.172 9.34 +45.06 30.00 6.172 9.33 +45.11 45.00 6.171 9.31 +45.17 60.00 6.171 9.30 +45.24
CPR when PP <= X% 101-221 4 Price WAL Sprd Swaps
196
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

0.00 7.073 9.62 +131.98 15.00 7.073 9.62 30.00 7.073 9.61 45.00 7.073 9.59 +132.21 60.00 7.073 9.57 +132.33
CPR when PP <= X% 100-1 Price WAL Sprd Swaps
+132.02 +132.10
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 premium security.
exhibit 5 Default Rate
PRICE/YIELD TABLE PNCMA 2001-C1 A2

0.00 6.172 9.35 +45.02 0.50 6.171 9.31 +45.19 1.00 6.170 9.25 +45.43 2.00 6.166 9.10 +46.06 3.00 6.162 8.91 +46.92
101-22 4 Price
1
WAL Sprd Swaps
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 premium bond.
exhibit 6 Default Rate 100-1 Price WAL Sprd Swaps
PRICE/YIELD TABLE PNCMA 2001-C1 F

0.00 7.070 9.62 +132.01 0.50 7.070 9.62 1.00 7.070 9.62 2.00 7.070 9.62 +132.02 3.00 7.159 14.13 +107.51
+132.01 +132.01
197
Glossary
Anchored Strip Center A grocery or discount retailer (the anchors) attract tenants
to small stores that are adjacent.

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.
Asset Mix The mix of property types, loan concentration and diversity in the
number of loans.
Assisted Living Facilities This product is 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.
Call Protection What percentage of the pool is locked-out/defeased and for how
long? Defeasance and lockout provide for the most stable cash flows.The only cash flow variability will be a result of credit events. There are scenarios under which yield to maturity increases as prepayments increase in deals with yield maintenance loans and penalty points.
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. Class A Properties Trophy quality properties; 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 containing multiple anchors but not enclosed. Congregate Senior Housing These are independent living facilities that also 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. Continuing Care Retirement Communities These facilities offer the entire continuum
of senior housing from independent living to skilled nursing facilities. Residents move within the facility depending on the level of care required. Licensed operator.
Co-op Loans/Blanket Loans Very low loan to value loans. Loans senior to coop share.
198
Co-tenancy Provisions Permits the tenant to close its store if another major store closes. Coupon Bonds In high interest rate environments with discount dollar price bonds, faster prepayments during yield maintenance result in a higher yield to maturity for the investor. In a higher interest rate environment, the borrower is not required to pay a yield maintenance penalty. The borrower is required to prepay the loan at par. The investor has purchased the bond at a discount and been repaid at par. This opportunity does not exist with locked-out or defeased deals. In a rates unchanged or lower scenario, the prepayment sharing arrangement between the coupon bonds and the IO significantly influences yield to maturity. Older deals generally have less advantageous sharing formulas for the coupon bonds. Credit Tenant Lease All payments guaranteed by credit of tenant (i.e., Wal-Mart
or Kmart).
DSCR Debt Service Coverage Ratio. Generally, the higher the better. 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 US government securities in an amount such that the lender receives the same yield as if the borrower had not prepaid the loans. From the lenders perspective, a defeased loan is positive. The yield is the same, but there is a credit upgrade from commercial real estate credit to US Government credit.
Distribution Space A property type whose principal use is distribution or light assembly. It typically has minimal office space as a percentage of total space, 010%. Twenty-four feet ceiling heights are the minimum for modern distribution buildings. Higher 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. Dollar price bonds above $105 tend to trade 23 bp cheaper for every dollar increment up to $110 where liquidity is limited.
Double Wide/Single Wide Describes the size of a manufactured home that a given
slab will support. The double wide segment has been the fastest growing.
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.
199
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 percentage of office space as a percentage of
total space. This results in higher tenant improvement costs upon lease renewals. These tenants are less sensitive to clear heights and more concerned with access to 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. Geography Are there state concentrations greater than 10%? Go Dark Provisions Prevents tenants from vacating the space while continuing to pay rent; landlords like this because vacant space may harm sales of other stores located in 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%. Independent Living Facilities Multifamily apartment complexes catering to senior
citizens. They supply few services beyond building and ground maintenance. These facilities are unregulated.
In-Line Store Smaller store within a center (e.g., Foot Locker or Hallmark Cards). Interest Only Bonds (IOs) IOs generally benefit when prepayments increase during yield maintenance. Older CMBS deals (1996 and prior) generally allocate the entire yield maintenance penalty to the IO. More recent CMBS IOs from deals with yield maintenance also generally benefit if prepayments increase during yield maintenance periods. The prepayment penalty is generally higher than the present value of the interest strip lost as a result of the prepayment. The exception to this occurs when interest rates rise to the level where no prepayment penalty is due on the underlying loan. LTV Loan to Value. Generally, the lower the better. Leasehold Interest Building owner leases land. Lender wants ground lease term to exceed loan amortization period. Leasing Commissions Fees paid to brokers for bringing new leases signed by tenants to the owner of the building. Landlords pay this expense.
200
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. Location in Structure Mezzanine bonds have the most structural protection. They are protected from losses by subordinate bonds and protected from prepays by senior bonds. Short AAAs have most volatile spread due to prepayment and default exposure. Lockout Borrower prohibited from prepaying; same as a non-call in corporate bonds. MAI (Member, Appraisal Institute) Designation given to certified appraisers. Manufactured Housing Communities The land, streets, utilities, landscaping, and
concrete pads under the homes that 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 borrowers equity interest in the property. There is no lien on the property. NOI (Net Operating Income) Property revenues minus property expenses. 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-Adjusted Spread) A measure of spread that adjusts for default and call
option costs.
Occupancy Rate Number of occupied units/total number of units. Pad Concrete slab that supports each manufactured home. 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.
Rating in Tranches Before and Behind Bond As the CMBS market has matured, rating gradations have become finer (e.g., the creation of BBB- in early 1996). Seasoned deals may have imbedded upgrade potential as they continue to age (prepay and amortize). Recapture Provisions Permits the owner of a retail center to cancel a lease and to regain control of space after a tenant closes its store.
201
Glossary
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 now owns the property. Reserves Money set aside by the borrower for payment of certain expenses such
as capital expenditures, retenanting costs (tenant improvements and leasing commissions), real estate taxes and insurance.
RevPAR (Revenue Per Available Room) Average Daily Rate x Occupancy Rate. Rollover Term used to describe expiration of tenant lease. Lease terms are typically
5-10 years; credit tenant leases may be longer. It is preferable not to have rollover concentrations, which expose the owner to an uncertain rental market, or reduce NOI below debt service.
Self-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 center but the anchor is not part of collateral for loan. Shadow Rating A rating on an individual loan in a large loan pool given by the
rating agencies.
Skilled Nursing Facilities Independent nursing homes or a designated wing in a
hospital. They provide full-time licensed skilled nursing, medical and rehabilitative services. Average length of stay can range from 2 months to 2 years, or more. 24hour care is provided with doctors and registered nurses on call. Licenses by state; operator must obtain a certificate of need from 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.
Stressed DSCR Rating Agency Adjusted Cash Flow/Debt Service Payment using an
assumed debt service constant.

Super Regional Mall Over 1 MM SF with Multiple Department (4-5) stores as anchors. Tenant Improvements Cost to build walls, ceilings, install carpeting for a new
tenant. Typically $5 or 40% per square foot. The landlord typically incurs this expense. In strong demand markets, the landlord can pass this expense through to the tenant in terms of a higher rental rate. In weak markets, landlords must absorb the cost.
202
Tilt-Up Construction This is the preferred construction type for industrial buildings. It includes pre-cast concrete panels that are tilted-up on a steel frame. Tilt-up is preferable to corrugated metal exteriors for maintenance reasons. Triple Net Lease The tenant pays rent plus real estate taxes, insurance, and maintenance. UCF (Underwritten Cash Flow) The cash flow number used by the lender to estab-
lish a desired debt service coverage ratio. Typically includes a standardized deduction from net operating income to account for expenses need to maintain the property. NOI-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.
Unit Mix How many 1-bedroom versus 2-bedroom apartments are in a multi-family building. Older complexes have higher proportions of 1-bedrooms; higher percentages of 2-bedrooms are now preferred, as they are more flexible to families. WAC (Weighted Average Coupon) The higher the WAC, the greater the prepayment
risk; the lower the WAC, the greater the extension risk.
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.
203
CMBS Information
World Wide Web Sources
TRUSTEES
LaSalle StateStreet Wells Fargo Chase Bankers Trust
www.etrustee.net www.corporatetrust.statestreet.com www.ctslink.com www.jpmorgan.com/absmbs www.gis.deutsche-bank.com/gis/newclientcenter/ productcatalog/frameset_ap.asp?group=apir
SERVICERS
Amresco Midland GMAC First Union
www.amresco.com www.midlandls.com www.gmaccm.com www.firstunion.com/strprod/
R AT I N G A G E N C I E S
Fitch Moodys S&P

AGENCIES
www.fitchratings.com www.moodys.com www.standardandpoors.com
Fannie Mae Ginne Mae
www.fanniemae.com www.ginniemae.gov
Contacts
TRADING
Jim Hiatt, james.hiatt@morganstanley.com Scott Stelzer, scott.stelzer@morganstanley.com David Lehman, david.lehman@morganstanley.com

RESEARCH
212.761.2057 212.761.2055 212.761.2104
Howard Esaki, howard.esaki@morganstanley.com Marielle Jan de Beur, marielle.jandebeur@morganstanley.com Masumi Pearl, masumi.pearl@morganstanley.com Molly Paganin, molly.paganin@morganstanley.com
C A P I TA L M A R K E T S
212.761.1446 212.761.1454 212.761.1080 212.761.1448
Jon Strain, jonathan.strain@morganstanley.com Joel Friedman, joel.friedman@morganstanley.com

PRODUCT SPECIALIST
212.761.2270 212.761.2076
Bob Karner, robert.karner@morganstanley.com
212.761.1605
Disclaimer
The information and opinions in this report were prepared by Morgan Stanley & Co. Incorporated (Morgan Stanley). Morgan Stanley does not undertake to advise you of changes in its opinion or information. Morgan Stanley and others associated with it may make markets or specialize in, have positions in and effect transactions in securities or instruments of companies mentioned and may also perform or seek to perform investment banking services for those companies. Within the last three years, Morgan Stanley & Co. Incorporated, Morgan Stanley DW Inc. and/or their affiliates managed or co-managed a public offering of the securities of Advanta, Aetna, Artesia, BancOne, Bank of America, Bear Stearns, CDC, Chase, CIBC, Cigna, Commercial Mortgage Securities Corp., ContiMortgage, Credit Suisse First Boston, CSFB, Deutsche Bank, DLJ, Fannie Mae, FelCor Lodging Trust, Finova, First Union Bank, First Union Bancorp, First Union National Bank, GMAC, Goldman Sachs, Heller Financial, Hilton Hotels Corp., Hospitality Properties, Host Marriott, Jackson National Life, John Hancock, JP Morgan Chase, Keybank, LaSalle, LaSalle National Bank, Lehman Brothers, Midland Bank, Morgan Stanley Capital I Inc 1999-FNV1, Morgan Stanley Capital I Inc 1999-RM1, Morgan Stanley Capital I Inc 1999-LIFE1, Morgan Stanley Capital I Inc 1999-WF1, Nationsbank, New England Mutual Life, Nomura, Nomura Securities, PNC, Prudential, Residential Funding, Southern Pacific, TIAA, TrizecHahn Office, UBS and Wells Fargo. Morgan Stanley & Co. Incorporated, Morgan Stanley DW Inc. and/or their affiliates will deal as principal in the securities recommended herein. Morgan Stanley & Co. Incorporated, Morgan Stanley DW Inc. and/or their affiliates or their employees have or may have a long or short position or holding in the securities, options on securities, or other related investments of issuers mentioned herein. An employee or director of Morgan Stanley & Co. Incorporated or Morgan Stanley DW Inc. and/or their affiliates is a director of Cigna or an affiliate thereof. Calculation of the MSHY indices are based on sampling techniques developed by Morgan Stanley for the relevant market or market segments using indicative prices provided by Morgan Stanley for the companies included in the index. Actual trades may not have been consummated at these prices. The indicative prices do not necessarily reflect Morgan Stanleys internal bookkeeping and may vary significantly from prices available from other sources. No representation is made that the indices described above or the formula used to calculate them are accurate or complete. Past performance is not an indication of future performance. The MSHY indices, including the index composition and data, are the exclusive property of Morgan Stanley and may not be reproduced or redisseminated in any form and may not be used as a basis for any financial instrument or product without the permission of Morgan Stanley. The investments discussed or recommended in this report may not be suitable for all investors. Investors must make their own investment decisions based on their specific investment objectives and financial position and using such independent advisors as they believe necessary. Where an investment is denominated in a currency other than the investors currency, changes in rates of exchange may have an adverse effect on the value, price of, or income derived from the investment. Past performance is not necessarily a guide to future performance. Income from investments may fluctuate. The price or value of the investments to which this report relates, either directly or indirectly, may fall or rise against the interest of investors. Price and availability are subject to change without notice. To our readers in the United Kingdom: This publication has been issued by Morgan Stanley & Co. Incorporated and approved by Morgan Stanley & Co. International Limited solely for the purposes of section 21 of the Financial Services and Markets Act 2000. Morgan Stanley & Co. International Limited and/or its affiliates may be providing or may have provided significant advice or investment services, including investment banking services, for any company mentioned in this report. NOT FOR DISTRIBUTION TO PRIVATE CUSTOMERS AS DEFINED BY THE U.K. FINANCIAL SERVICES AUTHORITY LIMITED. This publication is disseminated in Japan by Morgan Stanley Japan Limited and in Singapore by Morgan Stanley Asia (Singapore) Securities Pte Ltd. To our readers in Australia: This publication has been issued by Morgan Stanley & Co. Incorporated but is being distributed in Australia by Morgan Stanley Dean Witter Australia Limited, a licensed dealer, which accepts responsibility for its contents. Any person receiving this report and wishing to effect transactions in any security discussed in it may wish to do so with an authorized representative of Morgan Stanley Dean Witter Australia Limited. To our readers in Canada: This publication has been prepared by Morgan Stanley & Co. Incorporated and is being made available in certain provinces of Canada by Morgan Stanley Canada Limited. Morgan Stanley Canada Limited has approved of, and has agreed to take responsibility for, the contents of this publication in Canada. To our readers in Spain: Morgan Stanley Dean Witter, S.V., S.A., a Morgan Stanley group company, supervised by the Spanish Securities Markets Commission (CNMV), hereby 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. Past performance is not indicative of future returns. Certain assumptions may have been made in this analysis which have resulted in any returns detailed herein. No representation is made that any returns indicated will be achieved. Transaction costs (such as commissions) are not included in the calculation of returns. Changes to the assumptions may have a material impact on any returns detailed.
THE AMERICAS
1585 Broadway New York, NY 10036-8293 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
ASIA PACIFIC
Three Exchange Square Hong Kong Tel: (852) 2848-5200

5 - CMBS Primer2

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

5 - CMBS Primer2

Transféré par

Droits d'auteur :

Formats disponibles

SECOND EDITION

Editors: Howard Esaki, Marielle Jan de Beur, Masumi Pearl

Transforming Real Estate Finance

We detail the growing European CMBS market in this section.

192 197 198 203

Transforming Real Estate Finance chapter 1

NEW ISSUE MARKET CMBS STRUCTURE

Source: Morgan Stanley

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 1

BASE-CASE CREDIT ENHANCEMENT GUIDELINES FOR VARIOUS PROPERTY TYPES

OFFICE Individual PROPERTIES Loan Coverage

REGIONAL Individual MALLS Loan Coverage

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 1

BASE-CASE CREDIT ENHANCEMENT GUIDELINES FOR VARIOUS PROPERTY TYPES

INDUSTRIAL/ ANCHORED RETAIL Individual PROPERTIES Loan Coverage

MULTIFAMILY Individual PROPERTIES Loan Coverage

SUBJECTIVE ADJUSTMENTS TO CREDIT SUPPORT LEVELS

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

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

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 1

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 2

ISSUANCE: A GLOBAL RECORD

77.7 71.0 58.5 48.9

$65.3bn through 12/6/01. $8.0bn through 1Q2002.

Source: Morgan Stanley, Commercial Mortgage Alert

NON-US CMBS ISSUANCE

3.6 1.1 0.9 0.6

0.6 0.0 0.3

$16.4 through 12/6/01.

Source: Morgan Stanley, Commercial Mortgage Alert

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 2

US 53.3 20.2 3.1 13.0 10.4

Non-US 11.4 52.7 18.7 7.6 9.6

Source: Morgan Stanley, Commercial Mortgage Alert

LIFE INSURANCE COMPANY COMMERCIAL MORTGAGE COMMITMENTS, 1996-2001

AVERAGE DEAL SIZE

Source: Morgan Stanley, Commercial Mortgage Alert

TEN LARGEST CMBS DEALS IN 2001

Source: Morgan Stanley, Commercial Mortgage Alert

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 2

ESTIMATED CMBS MARKET CAPITALIZATION, 1987-2002P

350 300 250 200 150 100 50 0

Source: Morgan Stanley, Commercial Mortgage Alert

AAA SPREADS WIDEN TO TREASURIES

10-YEAR AAA CMBS SPREADS VS. LIBOR

Post-1998 Shock Trading Range

Source: Morgan Stanley

5-YEAR AAA CMBS SPREADS TO LIBOR

Source: Morgan Stanley

Spread (bp) 350

INVESTMENT GRADE SPREADS TO UST

221 175 150 132

Source: Morgan Stanley

Please refer to important disclosures at the end of this report.

Transforming Real Estate Finance chapter 2

10-YEAR AAA CMBS SWAPPED TO LIBOR