+ All Categories
Home > Documents > Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Date post: 12-Apr-2015
Category:
Upload: radical-8
View: 955 times
Download: 27 times
Share this document with a friend
Description:
Morgan Stanley - Real Estate Finance
208
Editors: Howard Esaki, Marielle Jan de Beur, Masumi Pearl Transforming Real Estate Finance A CMBS Primer SECOND EDITION SPRING 2002
Transcript
Page 1: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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

Transforming Real Estate FinanceA CMBS Primer

SECOND EDIT ION SPRING 2002

Page 2: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

This book is an overview of the commercial mortgage-backed securities (CMBS) market.

In some sense, this book has been over fiveyears in the making, as it includes excerpts from research publications from as early as1997. We also include new material on propertytypes backing the loans in CMBS and a glossaryof 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, RyanMarshall, Scott Stelzer, Jonathan Strain, RobertGillman, Joseph Philips, Robert Restrick, LisaSchroer, Steven L’Heureux and Mark Snyderman.

We hope you find this book useful and welcomecomments so that we can improve future editions.

Page 3: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

INTRODUCTION 2 – 9To 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.

A CREDIT TEST 10 – 27This chapter is our most recent review of CMBS data, market events, and relativevalue. We will update this section twice a year.

MAJOR PROPERTY TYPES IN CMBS 27 – 45In this chapter, we give an overview of the property types backing CMBS and giverating agency views of each real estate sector.

RETAIL COLLATERAL IN CMBS 46 – 55This chapter describes the methodology of our recent retail study, which examinesMorgan Stanley underwritten CMBS deals for exposure to 86 risky retail tenants.

HOTEL COLLATERAL IN CMBS 56 – 83Hotels have often been viewed by rating agencies and investors as one of the riski-est 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 – 95Commercial 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 – 107Loans of greater than $50 million are a growing part of the CMBS market. Thischapter describes the “large loan” market and uses an option-adjusted spread modelto evaluate relative value.

MULTIFAMILY MBS 108 – 115In 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.

CMBS IOs 116 – 123CMBS 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.

COMMERCIAL MORTGAGE DEFAULTS 124 – 131This chapter reports on the history of commercial mortgage defaults at insurance companies.

EUROPEAN CMBS 132 – 161We detail the growing European CMBS market in this section.

TRANSACTION MONITORING 162 – 179As the real estate cycle turns and CMBS age, more problem loans will appear inseasoned CMBS deals. This chapter gives examples of Morgan Stanley’s quarterlytracking report (detailing specially serviced loans in Morgan Stanley issued transac-tions) and monthly CMBS delinquency report.

CMBS DELINQUENCIES BY ORIGINATOR 180 – 191Commercial banks, finance companies, investment banks and life insurance com-panies are the four major types of CMBS loan originators. This chapter examinesdelinquencies on CMBS loans by issuer type.

FACTORS TO CONSIDER BEFORE INVESTING IN CMBS 192 – 197

GLOSSARY 198 – 203The final section is a glossary of terms frequently used in the CMBS market.

Transforming Real Estate Finance

A CMBS Primer

1

NEW 2

3

NEW 4

NEW 5

6

7

REVISED 8

REVISED 9

10

11

REVISED 12

NEW 13

14

Page 4: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Commercial Mortgage Backed Securities (CMBS) are bonds backed by pools of mortgages on commercial and multifamily real estate. As of January 2002, themarket 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 allowsfor the division of the loan into credit classes so that an investor may buy a classrated from AAA to single-B and unrated. In addition, CMBS are marked to marketon a daily basis and hence are more liquid than whole loans. CMBS appeal to awide 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 theCMBS market and the structure of CMBS. In the next chapter, we present anoverview of the current state of the markets. We then discuss the types of realestate 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 evaluatingcommercial mortgage default risk. We follow with a description of the rapidlygrowing European CMBS market. We provide examples of our various CMBSmonitoring 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 privatemarket. Lending was dominated by a handful of banks, life insurance companies,and pension funds. Real estate ownership was regionally focused with ownershipconcentrated in a few hands. Families and private partnerships were the largestowners. 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 worstsince 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 lendersfrom the commercial mortgage market. Regulators and rating agencies turnedmore negative on commercial mortgage holdings, so that the remaining lendersbecame less willing to extend credit.

CMBS: D IVERSIF IED PUBLIC REAL ESTATE

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 publicstock and bond offerings. REIT shares provided an opportunity for small, diversified investments in real estate.

Transforming Real Estate Finance

Introduction

2

chapter 1

Page 5: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

As with REITs, CMBS provide investment opportunities in diversified real estatepools. Investment banks started to apply securitization legal structures, and tech-nology developed during the 1970s and 1980s for residential mortgage-backedsecurities 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-1990swhen the Resolution Trust Corporation (RTC) pooled non-performing loans fromfailed institutions. Some of these transactions exceeded $1 billion and led to thegrowth in the investor base for CMBS. After the success of the RTC transactions,CMBS gained wider acceptance with investors and non-government, or “private-label”, issuers. Issuance of CMBS in the US grew rapidly in the mid-1990s, peakingat $78 billion of US issuance in 1998. Since then, annual domestic issuance has sta-bilized in the $50-70 billion range.

Outside the US, CMBS has also taken hold as a financing vehicle, with $13 billionissued 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 anRTC-like situation, with distressed properties and lenders. In 2001, about $5 billion of CMBS came to market in Asia.

STRUCTURES AND RATING AGENCIES

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 mortgagesalmost always have some form of prepayment penalty (Chapter 6), creditanalysis plays a more important role than prepayment analysis. For residentialMBS, prepayments are a much more important factor.

CMBS are static pools of commercial real estate loans divided into tranches withvarying subordination levels and credit ratings. A typical transaction has about90% investment grade bonds concentrated in AAA securities, with the remaining10% non-investment grade. Interest only (IO) bonds (Chapter 9) can be strippedoff all or part of the structure.

A typical structure consists of sequential pay, fixed rate bonds. The AAA bondsare time-tranched with a 5-year AAA bond ahead of a 10-year AAA bond.

Please refer to important disclosures at the end of this report. 3

NEW ISSUE MARKET CMBS

STRUCTURE

exhibit 122%

18%

12.5%

9%

4.5%

2%

0%

IO

AAA(78% )

AA(4%)

A(5.5%)

BBB(3.5%)

BB(4.5%)

B(2.5%)

UR(2%)

% Subo rdination

Source: Morgan Stanley

Page 6: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

In addition to the mortgage collateral, credit enhancements may be in the form ofreserve funds, guarantees, letters of credit, cross-collateralization and cross-defaultprovisions. 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 ofthrough the borrower if certain default triggers occur. Virtually all loans within CMBSare 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 legaldocuments outline each party’s responsibilities and fees. Typically, the Trustee isresponsible for reporting monthly payments and collateral performance data tocertificate holders. The Master Servicer is responsible for servicing all performing loans and monitoring loan document requirements. The Special Servicer resolvesdefaulted or delinquent loan issues.

RATING AGENCY METHODOLOGY

Before each CMBS is issued, the rating agencies review the collateral in the transaction and determine the tranche ratings and pool sizing. During the processthe agencies review the property level cash flows, perform physical inspections,and run various stress analyses on the underlying cash flows. This section exam-ines the rating process for conduits, or diversified pools of mortgages. A conduitoriginates loans for sale or securitization, and not for holding in a portfolio.

When a conduit deal comes to market, the rating agency performs due diligenceon a subset of the properties (typically between 35% and 75%). The larger loansin the deal are always underwritten while the remaining properties are chosensuch that they provide a representative cross section of the deal. To determinecredit enhancement levels, the underwritten cashflow (UCF) produced by eachproperty 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 SpecificProperty typeLoan-to-value ratioDebt service coverage ratioFixed/floating rateLoan seasoningDirect versus correspondent lending

• Real Estate Outlook

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

Transforming Real Estate Finance

Introduction

4

chapter 1

• Deal SpecificNumber of loans in the dealLoan sizeDegree of subordinationBalloon extension riskQuality of master servicer and special servicer

Page 7: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Loans collateralized by different property types are generally ranked in the following order (best to worst): regional malls, multifamily, anchored communityretail, industrial, office and, finally, hotel. These rankings are based on historicaldefaults and cash flow volatility of the different property types. The variation ofrequired credit enhancement levels with debt service coverage (DSC) and loan-to-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% creditenhancement for a AAA rating while an office property requires double that figure.The credit enhancement levels shown in Exhibit 2 are somewhat “sticky” withrespect to a credit/real estate cycle.

It is therefore sometimes possible to “arb” this fact by, say, buying conduit CMBSbacked by office properties in an environment where the fact that office propertiesare 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.

Please refer to important disclosures at the end of this report. 5

Individual CreditLoan Coverage Enhancement

LTV DSCR AAA AA A BBB BB B30 2.50 3.9 2.9 2.2 1.5 1.0 0.540 2.00 7.8 5.8 4.3 3.0 1.6 1.050 1.75 13.1 9.7 7.2 5.0 2.7 1.160 1.50 20.9 15.5 11.5 8.0 4.4 1.765 1.45 27.2 20.2 14.9 10.4 5.7 2.370 1.35 35.8 26.6 19.6 13.7 7.5 3.075 1.25 47.1 34.9 25.8 18.0 9.9 3.980 1.15 60.2 44.6 33.0 23.0 12.6 5.0

BASE-CASE CREDITENHANCEMENT GUIDELINES FORVARIOUS PROPERTY TYPES

exhibit 2

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

Individual CreditLoan Coverage Enhancement

LTV DSCR AAA AA A BBB BB B30 2.50 2.0 1.5 1.1 1.0 0.5 0.540 2.00 3.9 2.9 2.2 1.5 1.0 0.550 1.75 6.5 4.8 3.6 2.5 1.4 1.060 1.50 10.5 7.8 5.7 4.0 2.2 1.065 1.45 13.6 10.1 7.5 5.2 2.8 1.170 1.35 17.9 13.3 9.8 6.9 3.8 1.575 1.25 23.5 17.5 12.9 9.0 4.9 2.080 1.15 30.1 22.3 16.5 11.5 6.3 2.5

OFFICEPROPERTIES

REGIONALMALLS

Page 8: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Investors should also be concerned with the dispersion of DSC and LTV in theentire 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 LTVbuckets while the other rating agencies use a continuous variation of credit-support 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 under-written under an interest rate scenario that is substantially higher than currentrates. Loans without a track record of consistent payments are also rated moreconservatively than those seasoned at least five years. The origin of a loan,whether direct or via a correspondent, matters less than it previously did. Manysubjective assessments also go into the rating process and Exhibit 3 shows theaddition and subtractions that rating agencies apply to the credit support level.

Transforming Real Estate Finance

Introduction

6

chapter 1

Individual CreditLoan Coverage Enhancement

LTV DSCR AAA AA A BBB BB B30 2.50 2.5 1.9 1.4 1.0 0.5 0.540 2.00 5.0 3.7 2.7 1.9 1.1 0.550 1.75 8.4 6.2 4.6 3.2 1.8 1.060 1.50 13.4 9.9 7.3 5.1 2.8 1.165 1.45 17.4 12.9 9.5 6.7 3.6 1.470 1.35 22.9 17.0 12.6 8.8 4.8 1.975 1.25 30.1 22.3 16.5 11.5 6.3 2.580 1.15 38.5 28.5 21.1 14.7 8.1 3.2

BASE-CASE CREDIT ENHANCEMENT GUIDELINES FOR VARIOUS PROPERTY TYPES

exhibit 3

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

Individual CreditLoan Coverage Enhancement

LTV DSCR AAA AA A BBB BB B30 2.50 2.6 1.9 1.4 1.0 0.5 0.540 2.00 5.2 3.9 2.9 2.0 1.1 0.550 1.75 8.7 6.5 4.8 3.3 1.8 1.060 1.50 13.9 10.3 7.6 5.3 2.9 1.265 1.45 18.1 13.4 9.9 6.9 3.8 1.570 1.35 23.9 17.7 13.1 9.1 5.0 2.075 1.25 31.4 23.3 17.2 12.0 6.6 2.680 1.15 40.1 29.7 22.0 15.3 8.4 3.3

INDUSTRIAL/ANCHORED

RETAIL PROPERTIES

MULTIFAMILYPROPERTIES

continued

Page 9: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

On a deal level basis, conduits have diverged most significantly from the traditionalCMBS model in the degree of diversification provided by the large number ofunderlying loans. It is not uncommon to see 200 or more loans in a conduit. Therating agencies like to see at least 50 loans underlying a deal with no more than5% of the deal (by dollar amount) in any one state. Any single loan should notconstitute more than 5% of the deal. Credit-support levels are often tested bydefaulting the three largest loans.

At the triple A rating level, subordination levels are likely to be very similar acrossrating agencies. However, some degree of rating shopping is likely to occur forlower-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 agencylooks 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 thesame time. This is because loan documents typically allow for three one-year exten-sions, and this is not necessarily long enough to get through a credit downturn.

Finally, the rating agencies evaluate the real estate environment and where weare in the credit cycle.

Please refer to important disclosures at the end of this report. 7

Special Adjustment1 Other Factors Applied Additions Reductions on a Case-By-Case Basis

AmortizationFully 5% Floating Interest Rates

Interest Only 20% — Asset QualityLess Than

5 Years Seasoning 10-20% — Market Barriers to EntryServicer/Special Cash Flow Volatility

Servicer Assessment 10% 10% Loan SizeQuality Assessment

Origination 20% 5% Location AssessmentInformation 20% 5% Concentration

Cash Control — 10-20%Reserves — 5-10%

SUBJECTIVE ADJUSTMENTS TO CREDIT SUPPORT LEVELS

exhibit 4

1Adjustments are applied as a percentage of base-case credit enhancement levels; i.e., if base-case creditenhancement is 10%, and the adjustment factor is 20%, the adjusted credit enhancement is 12% (10 x 1.2).Source: The Rating of Commercial Mortgage-Backed Securities, Duff and Phelps Credit Rating Co.

Page 10: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

RATING AGENCIES AND INVESTORS MONITOR CONDUIT QUALITY

Some investors have expressed concerns that commercial mortgage conduits havebecome too aggressive in lending and that the quality of loans in CMBS conduitpools will deteriorate. Other investors worry that conduits will chase after loansbacked by “B” or “C” -quality properties, leaving the top-tier loans to insurancecompanies. In our opinion, these fears are misplaced, since securitized loans mustultimately pass through the filters of both rating agencies and investors. If ratingagencies recognize a slide in credit-quality rating, they will increase credit supportlevels (or lower ratings). If investors perceive increased risk, they will demandwider spreads on the securities. Investors can make adjustments to a recognizeddrop in loan quality; it is the unrecognized slippage that is dangerous.

As discussed in the section above, rating agencies apply published standards toloans pooled into a CMBS and adjust the result by making qualitative assess-ments. Almost all CMBS carry at least two, and many have three, ratings. Differentrating agencies assign different levels of credit support to obtain a given ratinglevel. In order to avoid a split rating, an issuer must go with the most conserva-tive collateral assessment. For example, if Fitch and Moody’s have AAA creditsupport 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 gowith the lower credit support level and receive a split rating. Non-investmentgrade classes, however, are often rated by only one agency). If an issuer starts toraise LTVs, reduce DSC ratios, or lend on more risky property types or in morerisky areas, rating agencies will respond by lifting credit support levels. The ratingagencies thus act as a first level of defense against a potential unobserved fall inloan credit quality.

Transforming Real Estate Finance

Introduction

8

chapter 1

Page 11: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 9

Page 12: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The terrorist attacks in September cast a pall over financial markets, and hadreverberations that continue to shake fixed-income markets. The CMBS marketwas also affected, as worries increased about large loan deals, hotels, and retailproperties. For the most part, however, the CMBS sector has thus far been rela-tively unscathed. Spreads on investment-grade bonds have widened by only afew basis points and global issuance set a new record. But falling rents and risingdelinquencies could be the first serious test of CMBS credit since the explosivegrowth 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 forecastersbelieving a rebound in the second half will occur, the fundamentals for realestate should not pose a problem for investment-grade CMBS. We think thattotal delinquencies on seasoned CMBS will top 2.5% next year, but remain lessthan half of the peak reached on insurance company collateral in 1993. At theselevels, AAAs would be unaffected, but we could see further widening of non-investment grade CMBS.

The major risk, in our view, is that the economy suffers another shock in 2002.This is not a low probability event—there has been a major dislocation in theeconomy or fixed-income markets in each of the last five years. Our opinioncontinues that securitized products in general, and CMBS in particular, are well suited as a defensive investment against exogenous shocks or a furtherdeteriorating economy.

For 2002, we see slowly deteriorating real estate fundamentals, but believe thatCMBS spreads will be relatively stable compared with other sectors. Investment-grade 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 bpto swaps.

• US issuance volume will decline to $60 billion. Non-US issuance will exceed2001’s 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 delinquenciesincrease.

• Congress will resolve problems with terrorist insurance, but many investors willstill shy away from concentrated property risk.

The rest of this chapter reports issuance, market capitalization, spread and totalreturn data for 2001 and includes our projections for 2002. We also review recenttrends in the major US commercial real estate markets.

Transforming Real Estate Finance

A Credit Test

10

chapter 2

Page 13: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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 tototal $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 yearever. We are forecasting $60 billion in domestic issuance in 2002, assuming the10-year UST remains around 5%.

• Non-U.S issuance will set a record of $20 billion this year, up from $12 billionin 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.

Please refer to important disclosures at the end of this report. 11

1$65.3bn through 12/6/01.2$8.0bn through 1Q2002. Source: Morgan Stanley, Commercial Mortgage Alert

0

20

40

60

80

100

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

$bn

0.2 0.12.6 2.9 4.1 3.4

7.6

14.017.4

28.8

40.4

77.7

71.0

17.2 17.8

60.058.5

1

48.9

( Projected)

2

US CMBSISSUANCE

exhibit 1

1$16.4 through 12/6/01. Source: Morgan Stanley, Commercial Mortgage Alert

0

5

10

15

20

25

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

$bn

0.41.5

0.0 0.20.8 1.4

0.6 0.0

2.8

0.9

3.6

0.6

9.4

12.0

0.31.1

20.0 1

( Projected)

25.0

NON-US CMBSISSUANCE

exhibit 2

Page 14: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• Fixed-rate CMBS conduits accounted for 53% of US issuance in 2001, up from44% 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 slightlyfrom the previous year.

• Single borrower transactions made up 20% of deals in the US, and over half ofthe deals outside the US. Large loan deals have been about a fifth of issuancein each of the last three years. Because of investor concerns over trophy prop-erties, we may see a decline in this percentage in 2002.

• Insurance company loan commitments totaled $19.6 billion through the end ofSeptember 2001. Commitments have declined for two consecutive years, butremain at healthy levels compared to ten years earlier.

• After declining for two straight years, average deal size increased slightly thisyear to over $500 million. There were 24 deals over $1 billion, up from ten in2000. If large loans are combined into conduit pools for diversity, we could seea further increase in deal size next year.

Transforming Real Estate Finance

A Credit Test

12

chapter 2

1Through 9/30/01. Source: ACLI

0

10

20

30

1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

$bn

1

LIFE INSURANCECOMPANY

COMMERCIALMORTGAGE

COMMITMENTS,1996-2001

exhibit 4

Source: Morgan Stanley, Commercial Mortgage Alert

2001 CMBS ISSUANCE BY DEALTYPE (IN %)

exhibit 3

Global US Non-US Conduit 44.9 53.3 11.4

Single Borrower 26.7 20.2 52.7Seasoned Collateral 6.2 3.1 18.7

Short-Term/Floating Rate 11.9 13.0 7.6Other 10.3 10.4 9.6

Page 15: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• Our current estimate of the market capitalization of CMBS is close to $350 bil-lion, up from $280 billion last year. We estimate that the total market capitaliza-tion of CMBS will exceed $400 billion by the end of 2002.

Please refer to important disclosures at the end of this report. 13

1Through 12/06/01. Source: Morgan Stanley, Commercial Mortgage Alert

63120

556

484507

825

419

268

191167141

259

161134

101

0

200

400

600

800

1000

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

$mm

1

AVERAGE DEAL SIZE

exhibit 5

Date Issuer Amount ($ mm)11/14/01 GGP Mall Properties Trust 2001 – C1 2,55010/11/01 COMM 2001-J2 1,5065/25/01 TIAA CMBS 1 Trust 2001 – C1 1,4655/10/01 TrizecHahn Office Properties Trust 2001-TZH 1,4407/20/01 LB-UBS Commercial Mortgage Trust 2001-C3 1,3825/14/01 LB-UBS Commercial Mortgage Trust 2001-C2 1,3193/22/01 First Union National Bank – Bank of America,

Commercial Mortgage Trust, 2001-C1 1,3088/03/01 Lehman Brothers Floating Rate Commercial

Mortgage Trust 2001 – LLFC4 1,26712/5/01 LB-UBS Commercial Mortgage Trust 2001-C7 1,2145/22/01 GS Mortgage Securities Corp II 2001- ROCK 1,210

TEN LARGEST CMBS DEALS IN 2001exhibit 6

Source: Morgan Stanley, Commercial Mortgage Alert

Page 16: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

AAA SPREADS WIDEN TO TREASURIES

• We think that benchmark 10-year AAA CMBS spreads will mostly be in a 45-65bp range over swaps in 2002. After the shocks of 1998 and 2001, we believethat the trading range will be both higher and wider than in the 1999 to mid-2001 period.

• Ten-year AAA CMBS spreads ended 2001 about 12 bp wider to swaps than atthe beginning of the year.

• During the year, 10-year AAA spreads were at their narrowest (43 bp) inJanuary and the widest (62 bp) in October and November. The 19 bp rangewas 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.

Transforming Real Estate Finance

A Credit Test

14

chapter 2

1Through 12/6/01 Source: Morgan Stanley, Commercial Mortgage Alert

$bn

0

50

100

150

200

250

300

350

400

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002P1

ESTIMATED CMBS MARKET

CAPITALIZATION,1987-2002P

exhibit 7

Source: Morgan Stanley

0

20

40

60

80

100

120

12/16/97 12/12/01

47

37

Post-1998 Shock Trading Range

Pre-ShockTradingRange

33

23

8/31/98 8/24/99 6/27/003/1/99

60

50

3/28/01

bp

10-YEAR AAACMBS SPREADS

VS. LIBOR

exhibit 8

Page 17: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• After the September 11 attacks, five-year AAA spreads moved above a 28-48 bprange to swaps for the first time in three years.

• Despite increases in delinquencies and a weakening economy, investment-grade CMBS spreads tightened to UST by 10 bp to 20 bp during the year. TheCDO bid drove some of this spread tightening. With rising delinquencies nextyear, we think investment-grade spreads will be flat to slightly wider versusUST in 2002.

Please refer to important disclosures at the end of this report. 15

5-YEAR AAA CMBS SPREADS

TO LIBOR

exhibit 9

0

20

40

60

80

100

120

01/09/98 01/02/99 12/26/99 12/18/00 12/12/01

28

48146 Weeks

bp

54

Source: Morgan Stanley

INVESTMENTGRADE SPREADS

TO UST

exhibit 10

0

50

100

150

200

250

300

350

01/03/97 03/30/98 06/24/99 09/17/00 12/12/01

Spread (bp)

221

175150132

A

AAAAA

BBB

Source: Morgan Stanley

Page 18: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

A Credit Test

16

chapter 2

Source: Morgan Stanley, Bloomberg

10-YEAR AAACMBS SWAPPED

TO LIBOR

exhibit 11

0

50

100

150

200

250

1/3/1997 3/30/1998 6/24/1999 9/17/2000 12/12/2001

bp

132

76

56

Difference

AAA CMBS10-Year Swap Spreads

• Non-investment grade CMBS spreads widened by 30 bp to 175 bp in limitedtrading. We expect further widening in 2002 as delinquency rates rise abovesubordination levels for many single-Bs.

Source: Morgan Stanley

NON-INVESTMENTGRADE SPREADS

TO UST

exhibit 12

0

200

400

600

800

1,000

9/12/1997 10/5/1998 10/28/1999 11/19/2000 12/12/2001

bp

1000

555

BB 10-YearB 10-Year

Page 19: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• Through December 12, YTD total returns on investment-grade CMBS rangedfrom +8.5% to +9.6%. The 10-year UST was almost unchanged from the start ofthe year, so most of the return was from coupon payments and modest spreadtightening. Non-investment grade CMBS underperformed, with spreads widen-ing as much as 175 bp for single-Bs.

RELATIVE VALUE: T IGHTER THAN BANK AND F INANCE

• AAA CMBS continued to trade 20 bp to 40 bp tighter than single-A bank andfinance paper. CMBS spreads were flat to slightly wider than bank and financespreads from 1996 through early 2000.

Please refer to important disclosures at the end of this report. 17

CMBS TOTAL RETURNS AS OF 12/12/2001exhibit 14

Source: Morgan Stanley

CMBS 5.0 6.7 6.5 76.6 6.1 34.4 -0.8 9.4 24.9(by Rating)CMBS AAA 4.8 6.7 6.0 63.7 5.9 28.0 -0.7 9.5 24.3CMBS AA 6.0 6.8 8.0 86.1 6.4 2.2 -1.0 9.0 26.7CMBS A 6.2 7.0 8.5 111.9 6.7 2.3 -1.1 8.5 27.3

CMBS BBB 6.3 7.1 8.9 164.7 7.3 2.0 -1.1 9.6 29.0(by Maturity)

CMBS 1-3.5 yr AL 2.7 6.2 3.1 66.5 4.9 3.7 -0.2 9.3 20.0CMBS 3.5-6 yr AL 4.1 6.7 5.0 67.9 5.7 5.9 -0.5 8.9 21.0CMBS 6-8.5 yr AL 5.4 6.8 6.9 71.6 6.1 21.8 -0.8 9.5 25.3CMBS 8.5+ yr AL 7.1 7.0 10.3 118.8 6.9 3.0 -1.3 8.7 26.5

Nominal (%)(Yrs) Spread (bn) (%) (%) SinceMod (%) (Yrs) (to UST (%) Market MTD YTD InceptionDur Coupon AvgLife in bp) Yield Value TotalRet TotalRet 1/1/00

Source: Morgan Stanley, Bloomberg

SINGLE-B CMBSYIELDS

exhibit 13

0

2

4

6

8

10

12

14

16

12/31/93 12/26/95 12/21/97 12/17/99 12/12/01

% Average Single-BCMBS Yield (Excluding

4/97–10/98)

Average 10-YearUST Yield

Single-B CMBS Yield10-yr UST Yield

Page 20: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

18

chapter 2

• BBB CMBS were about 20 bp wider than BBB unsecured corporates inDecember after reaching near parity in late 2000 and after September 11, 2001.

• At the end of the year, BBB unsecured REITs were about 5 bp wider than BBBCMBS. We think that these spreads will remain close to parity in 2002.

REGULATORY: SFASB 140 AND BIS

• A near crisis in the CMBS market was averted in July when issuers agreed tomodify language in new issue CMBS documents to clarify the actions of ser-vicers with regard to specially serviced loans. Before the clarification there wasconcern 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 afterJanuary 1, 2002. The new guidelines require higher reserves for below-invest-ment grade securities.

Source: Morgan Stanley, Bloomberg

BBB CMBS ANDREIT SPREADS

exhibit 16

100

150

200

250

300

350

4/10/98 3/11/99 2/10/00 1/10/01 12/12/01

Spread to UST (bp)

221217

REITsBBB CMBS

Source: Morgan Stanley, Bloomberg

10-YEAR AAACMBS VS.

SINGLE A BANK & FINANCE

exhibit 15

-100

0

100

200

300

7/5/1996 11/13/1997 3/25/1999 8/2/2000 12/12/2001

bp

157

132

AAA CMBSSingle A Bank and FinanceDifference

-25

Transforming Real Estate Finance

A Credit Test

Page 21: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 19

DELINQUENCY RATES REVERSE TREND

• Commercial mortgage delinquency rates reported by the American Council ofLife 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.

• Delinquencies on loans backing CMBS were at 1.16% in October, based onNovember remittance reports. This is the highest rate since we began trackingthe 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 currentbalances, an increase of 37 bp over the past 12 months. We expect delinquen-cies to rise by an additional 1%-2% over the next year.

• 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 delin-quency rate on loans in California is 0.28%, the lowest among large states.

Source: ACLI

COMMERCIALMORTGAGE LOAN

DELINQUENCYRATES

exhibit 17Length in Quarters

7.53%

0.19%

26 22 42 36

1Q65 4Q69 2Q76 4Q81 2Q92 3Q010

2

4

6

8

% Delinquent

Source: Morgan Stanley, Intex

CMBS TOTAL VS.SEASONED CMBS

DELINQUENCIES

exhibit 18

0.0%

0.4%

0.8%

1.2%

1.6%

2.0%

2.4%

8/1/1999 12/16/1999 5/1/2000 9/15/2000 1/30/2001 6/16/2001 11/1/2001

Seasoned

Total

Page 22: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

REAL ESTATE: IN FULL DECLINE

• Real estate markets, which started to soften in 2000, fell into decline in almostevery area. Rents fell in some sectors, such as San Francisco office, by a thirdor more. Hotel occupancies fell sharply after September 11, but recoveredsomewhat at the end of the year. Within all property types vacancy rates roseduring 2001, rising off very low levels during 2000. We anticipate that vacancyrates will rise moderately during 2002.

• On the bright side, new construction has also slowed over the past severalmonths in response to the overall slowdown in the economy.

• Although the US economy is experiencing an official recession, most industryanalysts anticipate this cycle will be more moderate than the early 1990s. Lackof speculative development, lower leverage within transactions, and significant-ly greater public ownership of real estate, are all factors indicating that realestate is better positioned going into this downturn.

OFFICE PROPERTIES: DEVELOPMENT P IPEL INE DECLINES

• Nationwide office vacancy rates increased 170 bp during the third quarter to12% (CB Richard Ellis). Over the past 12 months the vacancy rate has increased430 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 thethird quarter, to 11.4%. Office vacancy rates are the highest since 1996.

• The metropolitan statistical areas (MSAs) with the highest vacancy rates wereColumbus (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%).

Transforming Real Estate Finance

A Credit Test

20

chapter 2

Source: Morgan Stanley, Intex

CMBSDELINQUENCIES

BY CURRENTBALANCE

(FL AND CA)

exhibit 19

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

8/1/99 12/16/99 5/1/00 9/15/00 1/30/01 6/16/01 11/1/01

FL

CA

Page 23: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• According to REIS, third quarter office completions slowed to 18 million sf from22 million sf in the second quarter. REIS also projects a slowdown in the fourthquarter. New estimates for office development in 2001 are down 20% from pre-vious projections.

• Office properties in CMBS have the lowest delinquency rates of any propertytype. Based on November data, office properties posted a 0.37% delinquencyrate. Over the past 12 months delinquencies have declined 15 bp.

INDUSTRIAL PROPERTIES: VACANCIES R ISE MODERATELY

• During the quarter, national vacancy rates on industrial properties increased 70bp 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%).

Please refer to important disclosures at the end of this report. 21

Source: REIS

NATIONAL OFFICECOMPLETIONS &VACANCY RATES

(1981-2002P)

exhibit 20

0

50,000

100,000

150,000

200,000

1981 1984 1987 1990 1993 1996 1999 2002P0%

4%

8%

12%

16%

20%

24%

Source: Morgan Stanley, Intex

CMBS TOTAL VS.OFFICE

DELINQUENCIES

exhibit 21

0.0%

0.3%

0.6%

0.9%

1.2%

1.5%

8/1/1999 12/16/1999 5/1/2000 9/15/2000 1/30/2001 6/16/2001 11/1/2001

Total

Office

Page 24: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• 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 thanhalf the development in the second quarter.

• Based on November data, delinquencies on industrial loans in CMBS were0.58% of current balances. This was an increase of 35 bp over the past year.

RETAIL : DEVELOPMENT DECLINES

• 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 estimateof 39 million sf. During 2002 REIS anticipates development of 24 million sf.

Transforming Real Estate Finance

A Credit Test

22

chapter 2

CMBS TOTAL VS.INDUSTRIAL-WAREHOUSE

DELINQUENCIES

exhibit 23

0.0%

0.3%

0.6%

0.9%

1.2%

1.5%

8/1/1999 12/16/1999 5/1/2000 9/15/2000 1/30/2001 6/16/2001 11/1/2001

Total

Industrial

Source: Morgan Stanley, Intex

Source: REIS, CB Commercial

NATIONALINDUSTRIAL

COMPLETIONS &VACANCY RATES

1980-2002P

exhibit 22

0

50,000

100,000

150,000

200,000

250,000

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002P0%

4%

8%

12%

16%

Square Feet (000s) Vacancy Rate

CompletionsEstimated Annual TotalVacancy Rates

Page 25: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• Analysts report that mall traffic has been flat to slightly up after Thanksgiving.The International Council of Shopping Centers reported that during the firstweek of the holiday season, specialty (non-anchor) store sales declined 2.6%from last year.

• Delinquencies on retail loans within CMBS transactions were 1.21% of currentbalances as of November. Over the past 12 months, this delinquency rateincreased 47 bp.

MULTIFAMILY: STABLE AS ALWAYS

• 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 marketin 2001.

Please refer to important disclosures at the end of this report. 23

Source: REIS

NATIONAL RETAILCOMPLETIONS &VACANCY RATES

1992-2002P

exhibit 24

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002P0%

4%

8%

12%

Square Feet (000s) Vacancy Rate

CompletionsEstimated Annual TotalVacancy Rates

Source: Morgan Stanley, Intex

CMBS TOTAL VS.RETAIL

DELINQUENCIES

exhibit 25

0.0%

0.3%

0.6%

0.9%

1.2%

1.5%

8/1/1999 12/16/1999 5/1/2000 9/15/2000 1/30/2001 6/16/2001 11/1/2001

Retail

Total

Page 26: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• Delinquencies on multifamily loans within CMBS were 0.35% of current bal-ances. Over the past 12 months delinquencies declined 15 bp.

HOTELS: DEMAND DETERIORATES

• Nationwide revenue per available room (RevPAR) has declined about 19%between mid-September and late November (Smith Travel).

Transforming Real Estate Finance

A Credit Test

24

chapter 2

1Vacancy is as of 3Q01, completions is average year to date annualized as of September 2001.

Source: US Census Bureau

NATIONALMULTIFAMILY

COMPLETIONS &VACANCY RATES

1979-2002P

exhibit 26

50

100

150

200

250

300

350

400

450

500

550

1977 1980 1983 1986 1989 1992 1995 1998 2001P

0%

3%

6%

9%

12%

Permits (000s) Vacancy Rate

Vacancy RatesCompletions

1

Source: Morgan Stanley, Intex

CMBS TOTAL VS.MULTIFAMILY

DELINQUENCIES

exhibit 27

0.0%

0.3%

0.6%

0.9%

1.2%

1.5%

8/1/1999 12/16/1999 5/1/2000 9/15/2000 1/30/2001 6/16/2001 11/1/2001

Total

Multifamily

Page 27: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• The recovery from the current recession will have a significant impact on theperformance of hotels next year, as hotel demand is highly correlated to GDPgrowth. Positive news is that we anticipate hotel development to decline signifi-cantly in 2002.

• 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, OahuIsland, and Orlando have been the hardest hit since mid-September. Luxuryhotels at the highest price levels have also suffered more than lower-pricedproperties.

Please refer to important disclosures at the end of this report. 25

2001 YTD 2000 1999 1998 1997Luxury 68.1 71.8 71.9 72.6 78.4

Upscale 64.1 65.1 64.9 65.9 70.8Mid-Price 60.6 61.5 61.1 62.0 67.9Economy 58.2 58.5 58.0 58.4 61.2

Budget 60.1 59.5 58.7 58.7 59.9

HOTEL OCCUPANCY RATES BY PRICE CATEGORYAS OF OCTOBER 2001 (IN %)

exhibit 28

Source: Smith Travel Research

2001 YTD 2000 1999 1998 1997Luxury 146.91 144.99 138.88 133.58 136.01

Upscale 94.24 90.88 87.37 85.33 92.07Mid-Price 70.45 68.23 64.89 62.15 66.97Economy 54.71 52.44 49.23 47.14 49.93

Budget 41.46 42.20 89.60 37.43 40.60

HOTEL ROOM RATES BY PRICE CATEGORY AS OFOCTOBER 2001 (IN $)

exhibit 29

Source: Smith Travel Research

Source: PriceWaterhouseCoopers

HOTEL ADDITIONSTO SUPPLY

1969-2002P

exhibit 30

0

40

80

120

160

200

1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002P0%

1%

2%

3%

4%

5%

6%

7%

Page 28: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• Based on November 2001 data, hotel delinquencies were 3.49% of current balances. Over the past 12 months delinquencies have increased 191 bp.

26

chapter 2

HOTEL MARKET DATA AS OF YTD OCTOBER 2001exhibit 31

Source: Smith Travel Research

Occupancy(%) Avg Room Rate($)2001 2000 %Chg. 2001 2000 %Chg.

Anaheim-Santa Ana, CA 68.5 72.0 -4.9 91.85 87.90 4.5Atlanta, GA 63.2 67.5 -6.4 80.37 80.87 -0.6Boston, MA 67.3 78.1 -13.8 138.90 145.32 -4.4Chicago, IL 64.1 72.6 -11.7 113.68 117.38 -3.2Dallas, TX 59.0 66.7 -11.5 80.62 82.61 -2.4Denver, CO 65.4 70.3 -7.0 80.05 80.26 -0.3Detroit, MI 61.5 68.6 -10.3 81.82 81.64 0.2Houston, TX 66.6 63.5 4.9 75.89 75.78 0.1Los Angeles–Long Beach, CA 69.3 74.1 -6.5 97.01 96.59 0.4Miami-Hialeah, FL 67.2 70.7 -5.0 109.84 106.63 3.0Minneapolis–St. Paul, MN 65.6 71.3 -8.0 86.69 84.73 2.3Nashville, TN 58.0 61.9 -6.3 75.04 75.43 -0.5New Orleans, LA 66.6 71.1 -6.3 120.00 120.53 -0.3New York, NY 73.8 83.6 -11.7 183.83 199.08 -7.7NorfolkVirginia Beach, VA 60.7 61.2 -0.8 74.81 73.96 1.1Oahu Island, HI 71.2 76.7 -7.2 117.33 115.78 1.3Orlando, FL 66.6 74.4 -10.5 89.41 89.80 -0.4Philadelphia, PA-NJ 64.7 67.6 -4.3 99.02 102.34 -3.2Phoenix, AZ 60.3 63.3 -4.7 101.21 100.62 0.6San Diego, CA 72.3 75.6 -4.4 112.78 110.28 2.3San Francisco, CA 67.8 83.2 -18.5 146.32 149.09 -1.9Seattle, WA 65.7 71.2 -7.7 98.88 97.33 1.6St. Louis, MO-IL 63.0 65.3 -3.5 73.05 71.98 1.5TampaSt. Petersburg, FL 64.9 66.2 -2.0 87.56 84.90 3.1Washington DC,MD-VA 69.8 76.1 -8.3 118.65 116.22 2.1

Transforming Real Estate Finance

A Credit Test

Source: Morgan Stanley, Intex

CMBS TOTAL VS.HOTEL

DELINQUENCIES

exhibit 32

0.0%

1.0%

2.0%

3.0%

4.0%

8/1/1999 12/16/1999 5/1/2000 9/15/2000 1/30/2001 6/16/2001 11/1/2001

Hotel

Total

Page 29: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

RATING ACTIONS

• 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 anticipatedowngrades 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.

Please refer to important disclosures at the end of this report. 27

1Through December 7, 2001 Source: Morgan Stanley, Fitch, Moody’s and Standard & Poor’s

1999, 2000 AND2001 CMBS

RATING ACTIONS

exhibit 33

263

395

580

2761 40

0

100

200

300

400

500

600

700

1999 2000 2001 YTD 1

Upgrades

Downgrades

2001 RATING ACTIONS THROUGH DECEMBER 7, 2001(SHADED AREAS ARE DOWNGRADES OF AT LEAST AFULL CATEGORY)1

exhibit 34

1Does not include rating affirmations. Source: Morgan Stanley, Fitch, Moody’s and Standard & Poor’s

CCCand

AAA AA A BBB BB B Below Total

AAA 0 0 0 0 0 0 0 0AA 102 56 0 0 0 0 0 158

A 17 64 59 0 0 0 0 140BBB 10 24 53 88 5 0 0 180

BB 2 5 9 29 26 6 1 78B 1 0 1 2 24 21 10 59

CCC andBelow 0 0 0 0 0 1 4 5

Total 132 149 122 119 55 28 15 620

TO

FROM

Page 30: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

While CMBS investors don’t need to be real estate experts, particularly at the AAA level, a general understanding of various real estate property types and terminology is helpful. In this chapter we explain the various types of real estate properties and fundamentals that impact their performance.

Information in a transaction prospectus may contain descriptions of various assets as Class A, B, or C. Below is a description of those asset classes.

CLASSIF ICATIONS

Class ANewly built; higher quality finishes and prominent locations.

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

Class COlder properties needing frequent capital investment; uncertain future.

WHY DO PROPERTY CLASSIF ICATIONS MATTER?

• 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 mostcredit 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.

Transforming Real Estate Finance

Major Property Types in CMBS

28

chapter 3

Page 31: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 29

class A

class B

class C

Page 32: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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

RELATED TERMS

Credit Tenant Lease All payments guaranteed by credit of tenant (i.e., WalMart or Kmart).

Triple Net LeaseTenant pays rent, real estate taxes, expenses, and maintenance.

Go Dark ProvisionsPrevents 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 ProvisionsPermits the tenant to cancel its lease if another major store closes.

Recapture ProvisionsPermits the owner to cancel a lease and to regain control of space after a tenant closes its store.

TYPES OF RETAIL PROPERTIES

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

Regional MallOver 750,000 square feet with several department stores (2-3) as anchors.

In-Line StoreSmaller store within a center (i.e., Foot Locker or Hallmark Cards store).

Community CenterOver 100,000–275,000 square feet of space; multiple anchors but not enclosed.

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

Shadow Anchored StripSame as the anchor strip, but the anchor is not part of collateral for the loan.

Transforming Real Estate Finance

Major Property Types in CMBS

30

chapter 3

Typical Loan Terms

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

Page 33: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 31

Unanchored StripNo 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 BoxAnchor 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 bigbox/large store formats.

• Super-regional malls are the least affected by the negative factors listed above. Thesemalls 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.

RATING AGENCY V IEW

• Generally received favorable treatment from rating agencies as a result of strong historical performance.

• Preference for longer leases; nationally or regionally rated credit tenants.

unanchored strippower center/big box

anchored stripcommunity center

Page 34: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Multifamily Loans

RELATED TERMS

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.

TYPES OF MULTIFAMILY PROPERTIES

Garden ApartmentsMultiple buildings; usually no more than 2-3 stories.

High Rise ApartmentsOver three stories; usually located in downtown areas.

Transforming Real Estate Finance

Major Property Types in CMBS

32

chapter 3

Typical Loan Terms

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

high rise apartmentsgarden apartments

Page 35: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

CONVENTIONAL INVESTOR WISDOM

• Potential overbuilding in high growth markets of the Southeast and Southwest— Houston, Atlanta, Las Vegas.

• Multifamily was the first sector to recover from the real estate recession.

• Birth dearth has resulted in fewer households in prime renting years of 25-34.This fact somewhat mitigated by the “lifestyle” renter.

• A lifestyle renter is someone who can afford a home but chooses to rent for convenience: divorcees, empty-nesters.

• Healthy apartment properties have occupancies of 93% and higher.Occupancies below 88% for existing properties are worrisome.

• Government sponsored entities desire transactions with high concentrations in multifamily properties.

RATING AGENCY V IEW

• 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.

Please refer to important disclosures at the end of this report. 33

Page 36: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Office Loans

RELATED TERMS

Tenant ImprovementsCosts to build walls, ceilings, carpet for a new tenant. Typically $5– 40 per squarefoot. The landlord usually incurs this expense. In strong demand markets thelandlord can pass this expense through to the tenant in terms of a higher rentalrate. In weak markets, landlords must take tenant improvements out of net income, which reduces cash flow.

Leasing CommissionsFees paid to brokers to bring tenants, typically $2–4 per square foot at lease signings. Landlords bear this expense.

RolloverTerm used to describe expiration of tenant lease. Lease terms generally 5–10 years;credit tenants often longer. It is preferred to not have rollover concentrations, whichwould expose an owner to uncertain rental market or potentially reduce NOI belowdebt service.

TYPES OF OFFICE BUILDINGS

Transforming Real Estate Finance

Major Property Types in CMBS

34

chapter 3

Typical Loan Terms

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

Business District (CBD)

downtown suburban

Page 37: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

CONVENTIONAL INVESTOR WISDOM

• Above market rents are a concern if the market rent is insufficient to supportthe debt service. Rating agencies usually underwrite to market rents.

• Overbuilding: Investors that lived through the last real estate depression arenervous about current levels of development.

• Downtown versus suburban: Suburban office has suffered recently. The recent pop in the “tech bubble” has increased the supply of subleased space.

RATING AGENCY V IEW

• Very conservative approach makes underwriting these loans difficult.

• Very difficult to allow rollovers without cash reserves.

• Slow to accept market improvements in rental rates and values without manyother 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 excludedfrom underwriting income, which reduces the amount of potential loan.

• Management fees of 5% used by rating agency underwriters are believed to be above market rate of 1–3%.

Please refer to important disclosures at the end of this report. 35

Page 38: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Hotel Loans

RELATED TERMS

Average Daily RateTotal guest room revenue divided by total number of occupied rooms.

Occupancy RateNumber 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 rev-enue per available room.

FF&EFurniture, fixtures, & equipment; standard hotel underwriting includes a deductionas an operating expense for the ongoing replacement of FF&E, typically 4% to 5% of gross revenue. This differentiates hotel underwriting from apartment underwritingwhere some of those same expenses are considered capital expenditures and arenot an operating expense deducted from NOI. Typical refurbishment of commonareas of the hotel should occur every seven years.

Franchise FeeFee paid to hotel company that allows hotel owner to “fly the flag” of a hotelcompany (i.e., Marriott, Sheraton, etc.) and benefit from advertising and reserva-tion network. Ranges from 4–7% of gross revenue.

TYPES OF HOTEL PROPERTIES

Full ServiceA hotel that offers banquet and convention services; one or more full service restaurants.

Limited ServiceNo food service other than continental breakfast; minimal public space and small staff.

Transforming Real Estate Finance

Major Property Types in CMBS

36

chapter 3

Typical Loan Terms

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

Page 39: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

CONVENTIONAL INVESTOR WISDOM

• Hotels have the highest cash flow volatility of the four major property types asthey reprice rooms on a daily basis.

• The full service downtown hotels are more protected from new supply becauseof 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 thanlimited service hotels.

RATING AGENCY V IEW

• Very conservative because use of average historical income often reducesincome to levels significantly below current highs.

• Maximum occupancy they will underwrite is 65–75%, despite actual figureshigher 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.

Please refer to important disclosures at the end of this report. 37

full service

limited service

Page 40: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Manufactured Housing Community Loans

RELATED TERMS

Manufactured Housing CommunitiesThe land, streets, utilities, landscaping, and concrete pads under the homes comprise a manufactured housing community. The homes are independentlyfinanced. Homeowners pay monthly rent for the pad to the manufactured housing community owner.

PadConcrete slab that supports each manufactured home.

Double Wide/Single WideDescribes the size of manufactured home that a given slab will support. The double wide segment has experienced the fastest growth due to the growingacceptance of manufactured homes as a single family housing alternative.

TYPES OF MANUFACTURED HOUSING COMMUNIT IES

3-Star ParkOlder park lacking the amenities of a 5-star park; higher proportion of singlewide pads. Offer limited amenities and services.

4-Star ParkUsually double-wide units in good condition. Features may include concretepatios 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 ParkCurvilinear streets (streets that have curbs); neighborhood feel; well-landscaped;high proportion of double wide pads. Located in desirable neighborhood withconvenient access to retail.

Trailer ParkThis is a lower-end asset class, often confused with manufactured housing com-munities. Trailer parks are highly transient, dense communities of homes onwheels. Tenants are provided with no amenities other than simple utilityhookups. Not typically seen in conduit pools.

Transforming Real Estate Finance

Major Property Types in CMBS

38

chapter 3

Typical Loan Terms

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

Page 41: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

5-star park

CONVENTIONAL INVESTOR WISDOM

• Lack of familiarity; confusion with asset-backed manufactured housing loans to individuals.

• Insurance companies typically didn’t lend on this product, and it has somewhatof a stigma attached to it.

• Some investors have trouble distinguishing the relative investment stability ofmanufactured 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 loansrequires virtually no maintenance and very few manufactured homes are evermoved from the pad.

• Cost to move ($3,000–5,000 a year) exceeds pad rental ($300–500 a month).Upon homeowner foreclosure, bank will usually pay rent instead of movingthe foreclosed unit.

• Manufactured housing communities are also difficult to build as many communities have restrictive zoning ordinances against them.

RATING AGENCY V IEW

• Rating agencies favorable on credit.

• Low volatility of cash flows.

• Physical turnover rate 3–5% in manufactured housing versus; 50–60% in multifamily.

• Few capital reserves required.

• Often better than multifamily.

Please refer to important disclosures at the end of this report. 39

Page 42: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Industrial Loans

RELATED TERMS

Distribution SpacePrincipal 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 aremore economical for tenant as they stack goods vertically and rent fewersquare feet.

Flex Space/Office WarehouseHigher amount of office space as a percentage of total space. This results in higher tenant improvement costs necessary upon lease renewals. These tenantsare less sensitive to clear heights and more sensitive to accessibility of qualifiedlabor pools.

Tilt-Up ConstructionThis is the preferred construction type for industrial buildings. It includes pre-castconcrete panels that are “tilted-up” on a steel frame. Tilt-up is preferable to corrugated metal exteriors for maintenance reasons.

Transforming Real Estate Finance

Major Property Types in CMBS

40

chapter 3

Typical Loan Terms

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

distribution space

Page 43: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

flex space/office warehouse

CONVENTIONAL INVESTOR WISDOM

• Frequently exposed to single tenant credit; but generally lower tenantimprovement costs makes rollovers more palatable than office.

• Older buildings with less competitive clear heights becoming functionally obsolete. Constant roof repair is major expense item.

• Concerns over environmental contamination if property has had heavy industrial use.

• Construction cycle for industrial properties generally shorter than other property types (six to nine months versus two years for office); resulted in less overbuilding in last downturn.

RATING AGENCY V IEW

• 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.

Please refer to important disclosures at the end of this report. 41

Page 44: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Self Storage Loans

RELATED TERMS

Self-Storage FacilityCommercial property that lease storage space to individuals or businesses on amonth-to-month basis. Average self-storage facility has between 40,000 and 10,000square feet of rentable space divided among 400 to 1,000 individual units.

ManagementHalf 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 anydelinquencies.

Tenant Profile/TurnoverResidential users make up more than two-thirds of self-storage facility renters. Theaverage length of self-storage rental is 12 months. Occupancy rates tend to be inlow to mid 90s after long initial lease up period.

Transforming Real Estate Finance

Major Property Types in CMBS

42

chapter 3

Typical Loan Terms

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

self-storage facility

Page 45: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

CONVENTIONAL INVESTOR WISDOM

• 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 butpopulation outflow may result in an increase in demand in the short term ashomeowners store excess items during relocation.

RATING AGENCY V IEW

• Approve of it as an additional diversifier.

• Higher volatility because of monthly rental contracts but rental base has beenrelatively inert in moving stored items.

• Preference for infill or dense urban locations.

Please refer to important disclosures at the end of this report. 43

Page 46: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

HealthCare/Senior Housing Loans

TYPES OF HEALTH CARE / SENIOR HOUSING PROPERTIES

Independent Living FacilitiesMultifamily apartment complexes catering to senior citizens. They supply few services beyond building and ground maintenance. These facilities are unregulated.

Congregate Senior HousingThese are independent living facilities that provide a common dining facility andother services. Congregate seniors housing has no medical component, but mayprovide access to emergency medical care through call buttons. Not licensed as a nursing home.

Assisted Living FacilitiesThis product is targeted to elderly needing assistance, but not full time medicalcare. These facilities typically consist of apartment style units with a kitchenette.The operator of the facility provides three meals a day and assists residents withdaily activities such as feeding, bathing, dressing, and medication reminders.

Skilled Nursing FacilitiesIndependent nursing homes or a designated wing in a hospital. They provide full-time licensed skilled nursing, medical and rehabilitative services. Average lengthof stay can range from two months to two years, or more. Twenty-four hour careis provided, with doctors and registered nurses on call. Licenses by state; operatormust obtain a certificate of need from state before beginning operation.

Continuing Care Retirement CommunitiesThese facilities offer the entire continuum of seniors housing from independentliving to skilled nursing facilities. Residents move within the facility depending on the level of care required. Licensed operator.

Transforming Real Estate Finance

Major Property Types in CMBS

44

chapter 3

Typical Loan Terms

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

Page 47: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

CONVENTIONAL INVESTOR V IEW

• Recent changes in Medicaid reimbursements have negatively affected skillednursing and continuing care facilities.

• Excess supply of assisted living facilities have resulted in higher vacancy ratesand 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 lifecompanies avoided the senior sector because of confusion with licensed“nursing homes.” Foreclosure could result in license forfeiture and force eco-nomic vacancy of asset until replacement manager is found and new licensegranted.

• Not easily converted to multifamily if license lost.

RATING AGENCY V IEW

• Positive demographics; aging of the baby boomers.

• They prefer densely populated areas with heavier capital reserves.

• Skilled nursing requires higher DSCR because high ratio of income derivedfrom medical services that are not derived from location of property but rather need of specific patient.

Please refer to important disclosures at the end of this report. 45

Page 48: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

BUYING RETAIL

Over the past year, several US retailers have announced bankruptcies or storeclosings. These announcements are usually followed by an e-mail from CMBSdata providers detailing the number of CMBS deals affected by the closing. Inorder 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 fromthe National Research Bureau, we are able to capture 75% of the loan balances ofretail tenants in a sample of CMBS deals. This compares with only 29% coveredby a database of the top three tenants. In addition, we are able to look at thecumulative risk of a list of risky retailers. Although the new database does greatlyextend the coverage of retail tenant risk, it does not address the effect of thehealth of local retail markets and general real estate conditions on loan risk. Weleave this issue to a future paper.

Because of the large amount of data analysis involved, we limit this study to 31Morgan Stanley underwritten deals, and we examine the deals for exposure to 86risky retail credits that were identified by the REIT equity research team. We hopeto 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 someexposure to risky tenants.

• About 14% of the loans in the Morgan Stanley deals have exposure to at leastone of the 86 risky tenants.

• On a risk-weighted basis, only 0.8% of Morgan Stanley CMBS collateral hasexposure to the risky tenants. Risk-weighting accounts for the portion of a loanthat can be attributed to the retailer, based on its gross leasable area.

• MSDWC 2001-PGM had the largest weighted risk exposure as a percentage ofthe entire deal at 4.7%.

Transforming Real Estate Finance

Retail Collateral in CMBS

46

chapter 4

Page 49: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 47

Deal Total Delinquencies (%)MSC 1996-C1 2.7MSC 1996-WF1 0.3MSC 1997-ALIC 0.0MSC 1997-C1 1.7MSC 1997-HF1 0.0MSC 1997-LB1 0.0MSC 1997-WF1 1.7MSC 1997-XL1 0.0MSC 1998-CF1 7.2MSC 1998-HF1 0.9MSC 1998-HF2 1.3MSC 1998-WF1 0.5MSC 1998-WF2 0.0MSC 1998-XL1 0.0MSC 1998-XL2 0.0MSC 1999-CAM1 0.0MSC 1999-FNV1 1.8MSC 1999-LIFE 0.0MSC 1999-RM1 0.4MSC 1999-WF1 0.5MSC 2000 HG 0.0MSDWC 2000-LIFE 0.0MSDWC 2000-LIFE2 0.0MSDWC 2000 XLF 0.0MSDWC 2000-PRIN 0.0MSDWC 2001-PGMA 0.0MSDWC 2001-PPM 0.0MSDWC 2001-TOP1 0.0MSDWC 2001-TOP3 0.2MSDWC 2001-FRMA 0.0MSDWC 2001-SGMA 0.0

MORGAN STANLEY CMBS DEALSexhibit 1

Source: Morgan Stanley

Page 50: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Retail Collateral in CMBS

48

chapter 4

RETAIL PROPERTY OVERVIEW

In aggregate, retail properties typically constitute 27%-30% of the collateral inCMBS transactions and historically have been among the most common propertytypes within CMBS. Within retail properties, grocery-anchored community centersand super-regional malls are considered more desirable than box centers or mid-market malls. Strong regional malls and grocery-anchored community centers typ-ically have lower defaults and cash flow volatility than other retail property typeswithin CMBS.

As we anticipate credit to deteriorate in the coming year, the importance of moni-toring retailers increases. In a weaker credit environment, retail properties may beadversely affected by lower consumer confidence and spending. Since retail prop-erties usually make up a substantial portion of collateral in CMBS deals, wethought it would be useful to analyze the exposure of risky retailers to CMBS.

Risky RetailersRisky tenants were determined from the Z-Score methodology, which was devel-oped by Professor Edward Altman of NYU and implemented by the MorganStanley REIT equity research team. The Z-score methodology is a statistical tech-nique, which attempts to predict financial distress of corporations using financialratios. Altman’s methodology uses multiple discriminant analysis (MDA) and asample of 66 firms to derive the best linear combination of the firms’ financialratios.

The discriminant function that results from the MDA considers five financial ratios,which are multiplied by different coefficients to produce a score:

Z-Score: 1.2* (Working Capital / Total Assets)+

1.4* (Retained Earnings / Total Assets) +

3.3* (EBIT / Total Assets) +

0.6 * (Market Value of Equity / Liabilities) +

1.0 * (Sales / Total Assets)

Based on Altman’s findings, Morgan Stanley’s equity research REIT group used acutoff Z-score of 2.4 to identify unhealthy companies. Any publicly traded compa-ny 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 iftheir Z-scores were above 2.4. Companies that trade on the NASDAQ bulletinboard 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 byEdward I. Altman.

Page 51: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 49

Name Z-Score Name Z-ScoreA. Hirsch & Son 0.49 Mayor’s Jewelers 1.91Ames Dept. Store 1.88 Monfried Optical -5.93B & T Factory 1.68 Montgomery Ward Optical 2.25Benson Optical -5.93 Moto Photo 2.44Bikers Dream -0.90 National Record Mart 1.63Biozhem -22.71 Noodle Kidoodle 1.65BJ’s Optical 2.25 Overland Trading Co. 2.71Blockbuster Video 1.47 Pamida 2.39Bookland 2.33 Party City 3.76Books - A - Million 2.33 Payless Cashaways 2.51Books & Co. 2.33 Pearle Vision 2.25Calloway’s Nursery 2.08 Pep Boys 2.08Casual Male Big & Tall 1.68 Perfumania 1.32CD Exchange 1.25 Play Co. Toys 0.12CD Warehouse 1.25 Reeds Jewelers 2.25Checker Auto Parts 2.00 Repp Big & Tall 1.68Cole Vision Center 2.25 Restoration Hardware 2.12Disc-Go-Round 1.25 Samuel’s 0.49Discount Auto Parts 2.32 Schubach 0.49DIY Home Warehouse 1.93 Schuck’s Auto Supply 2.00Duling Optical -5.93 Sears Optical 2.25Elder-Beerman 2.37 Shopko Dept. Store 2.39Elegant Illusions 5.53 Silverman’s 0.49Elegant Pretenders 1.14 Silversmiths and Mission Jewelers 0.49Factory Card Outlet 1.95 Singer/Specs Discount Vision -5.93Famous Brands Housewares Outlet 3.09 Site for Sore Eyes -5.93Furrow’s 2.51 Southern Optical -5.93Hatfield Jewelers 0.49 Sports Authority 2.58Hearex Hearing Centers -1.82 Sterling Optical -5.93Hill’s Department Store 1.88 Stone & Thomas 2.37Hollywood Video -1.75 Superior Optical -5.93HomeBase 2.31 Swisher Maids 0.61Imposters 1.14 Things Remembered 2.25IPCO Optical -5.93 Today’s Man 2.02Jennifer Convertibles 3.26 Toy Co. 0.12Jennifer Leather 3.26 Toys International 0.12Joe Mugg’s Newstand 2.33 Track ‘n Trail 2.71Kidoodle Theater 1.65 U.S. Vision 1.73Kindy Optical -5.93 Ward’s Optical 2.25Kragen Auto Parts 2.00 Wayne Jewelers 0.49Lechters 3.09 Weiner’s 2.59Loehmann’s 1.81 West Coast Video -1.60Lumberjack Building Materials 2.51 Wickes -1.53

RISKY TENANTSexhibit 2

Source: Morgan Stanley

Page 52: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

METHODOLOGY

In order to assess CMBS exposure to the 86 risky retailers, we used the NationalResearch Bureau Shopping Center Directory to locate risky tenants within theCMBS deals. The Shopping Center Directory lists all tenants within 38,000 shop-ping centers nationwide, giving us the ability to locate many tenants that wouldbe undetected in conventional CMBS tenant searches. The National ResearchBureau Shopping Center Directory is updated semi-annually. CMBS data providersand prospectus material typically only list the largest three tenants within a shop-ping center. Therefore, the information provided in the directory allows us tolook at smaller tenants within various retail properties.

With both the shopping center directory and S&P Conquest, we can track 75% ofretail balances in Morgan Stanley deals. If this study were performed using datasolely from the S&P Conquest database, only 29% of the retail loan balanceswould have complete tenant information.

In total, we had some level of tenant information for 92% of the retail loanswithin the transactions that we examined. Sixty percent of the loans were covered through information from the National Research Bureau ShoppingCenter Directory, and had complete tenant information for the shopping cen-ters. Thirty-two percent of the loans were covered through data obtained fromthe S&P Conquest database, which only provided information on one to threetenants 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 defineweighted-risk retail exposure as the sum of all weighted-risk retail balances foreach 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 ofgross leasable areas (GLA) of risky tenants in the shopping center] / [GLA ofentire shopping center]

We also computed total risk exposure for each deal, which is equal to the sum ofall current shopping center loan balances with risky tenant exposure.

Transforming Real Estate Finance

Retail Collateral in CMBS

50

chapter 4

Page 53: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

SHORTFALLS/OTHER FACTORS TO CONSIDER

As with most studies, there are a few caveats that should be highlighted. Risk-weighted exposure is not necessarily a true indicator of the riskiness of a shop-ping center. There are financially distressed retailers that are not captured in thisstudy, as we only cover publicly traded companies.

Risk-weighted exposure also implies that only a portion of the loan is at riskwhen stores become distressed. This may be true if a couple of small retailers goout of business. However, the entire loan may be at risk if a number of stores inthe 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 oradjoining retailers within a specific shopping center, regardless of whether thetenant’s space is collateral in the securitization. It is possible that we are includingout-parcels or portions of shopping centers that are not part of the MorganStanley securitizations. However, inclusion is important since retailers are affectedby the health of the shopping center as a whole.

In addition, we computed risk-weighted exposures based on the size of the retail-er, 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, eventhough 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 ofonly $10 per square foot but occupy 70% of the GLA, while the in-line tenantsmay pay $20 per square foot and only occupy 30% of the GLA.

Please refer to important disclosures at the end of this report. 51

Page 54: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

KMART R ISK

Kmart was not highlighted in our recent study on risky retailers because thecompany’s Z-score was higher than our 2.4 cut off. Any retailer with a Z-scorebelow 2.4 was deemed “risky” by Morgan Stanley’s equity research REIT team. In Altman’s study, scores below 1.8 are categorized as “unhealthy”.

It is worth noting that the Z-score attributable to Kmart declined during 2001. Asof first quarter 2001, the company’s Z-score was 3.76; based on second quarterdata it declined to 3.42, and by year end, it declined to 3.02, indicating funda-mentals have been deteriorating.

MORGAN STANLEY CMBS EXPOSURE TO KMART

In our January 4, 2002 issue of CMBS Perspectives, we published an article dis-cussing CMBS exposure to Kmart. In CMBS transactions where Kmart was listedas one of the top three tenants, the weighted average exposure to loans withKmart was 2.25% of current balances.

We did a detailed search using the National Research Bureau (NRB) ShoppingCenter 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.

TOTAL R ISK EXPOSURE BASED ON LOAN BALANCE

Thirty-three loans within 17 Morgan Stanley CMBS transactions have exposure toKmart. 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 highexposure. The entire balance of the loan is included in the total exposure, despitethere 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 toKmart. Risk-weighted exposure only accounts for the portion of a loan that canbe attributed to the retailer, based on its gross leasable area.

Transforming Real Estate Finance

Retail Collateral in CMBS

52

chapter 4

Page 55: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 53

MORGAN STANLEY CMBS EXPOSURE TO KMARTexhibit 3

Source: Morgan Stanley, S&P Conquest, National Research Bureau Shopping Center Directory

TOTALCurrent Deal

Balance ($)Loans with

Kmart ($)# of

loans

KmartExposure

(%)

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

Page 56: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Retail Collateral in CMBS

54

chapter 4

MSC 1996-C1exhibit 4

Source: Morgan Stanley

MSC 1996-C1

Retail ExposureRetail Entire Deal % Retail

Loans Outstanding 14 99 14.1%Original Balance ($MM) $75.2 $340.5 22.1%Current Balance ($MM) $69.9 $225.0 31.1%

Study Results Retail Loans IncludedRetail Loans in Deal

Number Current Numberof Loans Balance of Loans

National Research Bureau 9 $50.8 14Prospectus 0 $0.0 14Total 9 $50.8 14

Risky Tenant Exposure Number Average GLAof Stores Store Center

1 Blockbuster Video 2 5,850 223,7502 National Record Mart 1 5,200 285,000

Shopping Center Exposure Total RiskyStores Stores Store

1 Constant Friendship Shopping Center 11 1 4,7002 Lilac Mall 20 2 12,200

SAMPLE RETAIL ANALYSIS

Page 57: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 55

StatisticsExposure by Current Balance ($MM)

W td. Risk Balance $0.5Included Retail $50.8 1.1%Total Retail $69.9 0.8%Entire Deal $225.0 0.2%

Total Risk Balance $16.1Included Retail $50.8 31.7%

% Retail Loans Included Total Retail $69.9 23.0%Current Number Current Entire Deal $225.0 7.2%Balance of Loans Balance Exposure by GLA (sq ft)

$69.9 64.3% 72.7% Store At-Risk Sq. Ft. 16,900$69.9 0.0% 0.0% Affected Centers 447,500 3.8%$69.9 64.3% 72.7% Total Centers 1,542,411 1.1%

Shop. Ctr. At-Risk Sq. Ft. 447,500Total Centers 1,542,411 29.0%

% Z-Score2.6% 1.51.8% 1.6 Types of Loans ($) Retail Loan Coverage ($)

Total GLA Largest Shopping Center WeightedCenter % Risky Store Current Balance Risk Balance162,500 2.9% 2.9% 10.4 0.3285,000 4.3% 2.5% 5.7 0.2

TOTAL: 16.1 .5

Included

Excluded

Retail

Other

Page 58: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Rating agencies and many investors have always viewed hotels as one of theriskiest property types. Unlike most other commercial real estate, which havelong-term leases, hotels have daily changes in occupancy and rental rates. Theslowing economy and recent events have increased CMBS investor concernsabout exposure to hotels. In this chapter, we examine the credit exposure ofCMBS 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-yearrevenue 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 standardsfor new transactions.

• Conduit CMBS transactions average about 10% exposure to hotels. Leveragelevels are more conservative on hotel loans than other property types, provid-ing cushion for the downturn.

• Even if delinquencies on hotel loans double from current levels, only unratedbonds and B bonds would be affected.

• Two large loan transactions (100% exposure to hotels) we reviewed can with-stand more severe declines in operating performance than are projected andstill generate sufficient cash for debt service payments.

Transforming Real Estate Finance

Hotel Collateral in CMBS

56

chapter 5

PROPERTY TYPEMATRIX

exhibit 1

Industrial

Low OnsiteManagement

Intensity

HighOperating

ImportanceLow Operating

Importance

High OnsiteManagement

Intensity

Office

Multifamily

Retail

Hotel

Healthcare

Source: Morgan Stanley

Page 59: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

GREATER SHOCK TO DEMAND THAN EARLY 1990S

Hotel demand is highly correlated to changes in GDP. When GDP is growing,consumer confidence rises resulting in greater discretionary income. Recent eventsand a slowing economy prior to September 11 have resulted in a significantdecline in hotel demand. Industry projections anticipate 2001 will post declines indemand of about 2.8% this year. During the Gulf crisis of 1991 demand declinedjust over 1%.

Over the past decade hotel ownership has experienced a significant shift fromprivate to public markets. The public capital markets have more efficientlymatched supply with demand than was the case in previous cycles.

HOTEL SUPPLY S IGNIF ICANTLY LOWER IN 1990S THAN 1980S

The typical time frame for hotel development is between 12 and 36 months.Between 1990 and 1991, demand declined significantly as the United Statesentered a recession, at the same time supply continued to enter the marketthrough projects that were started in the late 1980s.

Currently, the industry is better suited from a new supply standpoint than it wasduring the previous downturn.

Please refer to important disclosures at the end of this report. 57

NATIONWIDEHOTEL SUPPLYAND DEMAND

GROWTH

exhibit 2

-3%

-1%

1%

3%

5%

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001E 2002E

Supply ChangeDemand Change

Source: Smith Travel Research

Page 60: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

REVPAR PERFORMANCE AND PROJECTIONS

The hotel industry measures performance in revenue per available room(RevPAR). RevPAR is defined as average daily rate (ADR) at a hotel multiplied byits occupancy rate. For example, if a hotel has an ADR of $100 and an occupancyof 80% its RevPAR is $80.00 ($100 x .80 =$80).

Transforming Real Estate Finance

Hotel Collateral in CMBS

58

chapter 5

TOTALCONSTRUCTION:

1980S VS. 1990S

exhibit 3

887

1,175

0

500

1,000

1,500

1980-1989 1990-1999

Room Starts

Source: Morgan Stanley, PriceWaterhouseCoopers

HOTEL ADDITIONSTO SUPPLY

exhibit 4

0

40

80

120

160

200

1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001P0%

1%

2%

3%

4%

5%

6%

7%

(000's)

Net Additions to Existing Supply

% of Existing SupplyProjected

Source: Morgan Stanley, PriceWaterhouseCoopers

Page 61: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

PriceWaterhouseCoopers recently issued a reforecast of RevPAR projections for2001 and 2002. The firm published both a “baseline” and “prolonged weakness”scenario as outlined in Exhibit 5.

The firm noted that neither scenario takes into account future terrorist acts,expansion of the military effort beyond the air campaign and limited groundtroops, and military acts beyond Afghanistan.

During the Gulf War, 1991 nationwide RevPAR declined about 2.5%, while currentprojections anticipate RevPAR to decline about 7% in 2001.

PriceWaterhouseCoopers uses it own GDP estimates and other macroeconomicinputs to forecast RevPAR. The forecast assumes that GDP growth will be moder-ately positive in 2001 and 2002 with a recession in the second half of 2001 and arecovery in the first half of 2002.

Please refer to important disclosures at the end of this report. 59

ProlongedQuarter Baseline Weakness3Q 2001 -12.4 -12.44Q 2001 -15.0 -20.31Q 2002 -7.4 -12.72Q 2002 -3.4 -8.93Q 2002 5.8 -1.34Q 2002 8 5.9Year 2001 -7.1 -8.5Year 2002 0.5 -4.7

PRICEWATERHOUSECOOPERS US NATIONWIDE REVPAR PROJECTIONS

exhibit 5

Source: PriceWaterhouseCoopers

NATIONWIDEREVENUE PER

AVAILABLE ROOM(REVPAR) GROWTH

exhibit 6

-10%

-5%

0%

5%

10%

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

E

6.6%5.6%

(2.5%)

(6.9%)

Source: PriceWaterhouseCoopers, Smith Travel Research

Page 62: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

HIGHER PROFITABIL ITY H IGHER AND LOWER DEBT SERVICE

Currently, the hotel industry is better positioned to withstand a decline in rev-enues from a profitability and leverage perspective than it was 10 years ago.

In aggregate, hotel profitability was negligible between 1980 and 1989. In con-trast, 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 equalabout 4% of hotel revenues.

RATING AGENCY V IEWS ON HOTELS

In late 2001, rating agencies commented on the corporate ratings of several hotelcompanies, and revised rating guidelines for future CMBS transactions.

In late September, all three rating agencies commented on the ratings of severalhotel operating companies and hotel REITs. See Exhibit 8 for details.

In addition to the comments on corporate ratings, Moody’s and Fitch outlinedchanges to underwriting guidelines for hotel loans within CMBS transactions.Standard & Poor’s has not made any changes to its underwriting guidelines forhotels in new issue transactions.

Transforming Real Estate Finance

Hotel Collateral in CMBS

60

chapter 5

NATIONWIDEHOTEL

PROFITABILITYAND 10-YEAR

TREASURY RATES

exhibit 7

-10

0

10

20

30

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

20

01 0%

2%

4%

6%

8%

10%

12%

14%ProfitsInterest Rate

Profits ($bn) Interest Rate

Source: PriceWaterhouseCoopers

Page 63: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Fitch noted in a recently released report that it is concerned about the perform-ance of hotels over the next 12-24 months. Therefore, for new issue transactionsit will use the trailing 12-month RevPAR after August 31, 2001 reduced by 20%, orthe 1999 RevPAR number, whichever is lower. In place of the significant RevPARreduction, Fitch stated that the issuer could set up a 12-month debt servicereserve for hotel loans. Fitch also intends to make adjustments to expenses basedon any recently increased costs.

Fitch stated that for surveillance of CMBS transactions already issued, it will notapply the severe haircut to hotels but will compare the current operating numbersto the performance when the transaction was issued.

Moody’s released a matrix outlining reserve requirements on hotel loans withinnew issue CMBS transactions. Reserve requirements range from 1-13 monthsdepending on the type of hotel property, the anticipated stress environment, andthe underwritten debt service coverage ratio (DSCR). For example, a full-servicehotel with a 1.35 DSCR at issuance would require 13 months P&I reserves in ahigh stress environment while a limited service hotel with a 1.35x DSCR wouldrequire 5 months P&I reserves in a high stress environment. The rating agencyalso noted that it would consider 1998-99 as stabilized operating performance forhotels going forward.

Please refer to important disclosures at the end of this report. 61

Page 64: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Hotel Collateral in CMBS

62

chapter 5

Hot

el O

pera

ting

and

Fran

chis

e C

ompa

nies

Equi

tyTi

cker

Hot

el B

rand

sM

oody

's R

atin

gM

oody

's C

omm

ents

S&P

Rat

ing

S&P

Com

men

tFi

tch

Rat

ings

Fitc

h C

omm

ents

Cend

ant C

orpo

ratio

nCU

CRa

mad

a, H

owar

d Jo

hnso

nBa

a1Re

view

for

poss

ible

Dow

ngra

deBB

B+Ra

ting

Wat

ch N

egat

ive

Choi

ce H

otel

s In

tern

atio

nal

CHH

Com

fort

, Qua

lity,

Cla

rion,

Sle

epBa

a3Re

view

for

poss

ible

Dow

ngra

deBB

B-Ra

ting

Wat

ch N

egat

ive

Road

way

, Eco

no L

odge

, Mai

nSta

yHi

lton

Hote

ls C

orpo

ratio

nHL

THi

lton,

Hilt

on G

arde

n In

n, H

ilton

Sui

tes,

Baa3

Revi

ew fo

r po

ssib

le D

owng

rade

BBB-

Ratin

g W

atch

Neg

ativ

eBB

B-Ra

ting

Wat

ch N

egat

ive

Doub

letr

ee, G

uest

Sui

tes,

Ham

pton

Inn

Mar

riott

Inte

rnat

iona

lM

ARM

arrio

tt, R

enai

ssan

ce, C

ourt

yard

,Re

side

nce

Inn,

Fai

rfie

ld In

n, T

owne

Plac

eBa

a1Re

view

for

poss

ible

Dow

ngra

deBB

B+Ra

ting

Wat

ch N

egat

ive

Sprin

gHill

Suite

s, R

itz-C

arlto

nSt

arw

ood

HOT

Wes

tin, S

hera

ton,

St.R

egis

/Lux

ury

Colle

ctio

nBa

1Re

view

for

poss

ible

Dow

ngra

deBB

B-Ra

ting

Wat

ch N

egat

ive

BBB-

Ratin

g W

atch

Neg

ativ

e

Four

Poi

nts,

W H

otel

sFo

ur S

easo

nsFS

Four

Sea

sons

Baa3

Ratin

g co

nfirm

ed n

egat

ive

outlo

okBB

B-Af

firm

ed

Hot

el R

EITs

Equi

tyTi

cker

Prop

erty

Typ

e of

Ow

ners

hip

Moo

dy's

Rat

ing

Moo

dy's

Com

men

tsS&

PR

atin

gS&

P C

omm

ent

Fitc

hR

atin

gsFi

tch

Com

men

ts

Equi

ty In

nsEN

NM

id-s

cale

Lim

ited

Serv

ice

Hote

lsB3

(pre

ferr

ed)

Revi

ew fo

r po

ssib

le D

owng

rade

BB-

Ratin

g W

atch

Neg

ativ

e

FelC

or L

odgi

ng T

rust

FCH

Upsc

ale

Full

Serv

ice

Hote

lsBa

2 (s

enio

r un

sec)

Revi

ew fo

r po

ssib

le D

owng

rade

BBRa

ting

Wat

ch N

egat

ive

B1 (p

refe

rred

)Re

view

for

poss

ible

Dow

ngra

de

Hosp

italit

y Pr

oper

ties

Trus

tHP

TUp

scal

e Ex

tend

ed S

tay

Baa3

(sen

ior

unse

c)Af

firm

edBB

B-Lo

wer

ed r

atin

g w

atch

from

posi

tive

to s

tabl

eBa

1 (p

refe

rred

)Af

firm

ed

Host

Mar

riott

Corp

orat

ion

HMT

Luxu

ry F

ull S

ervi

ce H

otel

sBa

2 (s

enio

r de

bt)

Revi

ew fo

r po

ssib

le D

owng

rade

BBRa

ting

Wat

ch N

egat

ive

Ba3

(pre

ferr

ed)

Revi

ew fo

r po

ssib

le D

owng

rade

LaQ

uint

a Pr

oper

ties

LQI

Lim

ited

Serv

ice

Ba3

(sen

ior

debt

)Re

view

for

poss

ible

Dow

ngra

deBB

-Ra

ting

Wat

ch N

egat

ive

BB-

Ratin

g W

atch

Neg

ativ

e

Mer

iSta

r Ho

spita

lity

Corp

orat

ion

MHX

Full

Serv

ice

Hote

lsBa

3 (s

enio

r de

bt)

Low

ered

rat

ing

from

Ba2

, und

er r

evie

wBB

-Re

vise

d ra

ting

wat

ch fr

ompo

sitiv

e to

rat

ing

wat

chde

velo

ping

Win

ston

Hot

els

WHX

Lim

ited

Serv

ice,

Ext

ende

d St

ayB3

(pre

ferr

ed)

Revi

ew fo

r po

ssib

le d

owng

rade

RATI

NG

AG

ENCY

VIE

WS

ON

HO

TEL

COM

PAN

IES,

AS

OF

NO

VEM

BER

2001

exhi

bit

8

Sour

ce: M

orga

n St

anle

y, M

oody

’s, F

itch,

S&P

Page 65: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

RATING ACTIONS IN 2001

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 ratingagencies have taken some sort of rating action on various tranches within 16transactions. In total, 59 rating actions were issued, 54 upgrades and 5 down-grades. The upgrade/downgrade ratio on these transactions with greater thanaverage hotel exposure was 11:1, slightly better than the 10:1 ratio for the entireCMBS 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 havetrended 55 bp higher than our universe average since we began tracking thedata in late 1999.

Hotel delinquencies spiked to 1.93% in May 2001 and have moderated over thepast several months. We anticipate delinquencies will continue to rise throughthe remainder of the year and into early 2002.

If we assume that delinquencies on hotel loans double over the next 12months, hotel delinquencies would rise to about 3.25% of current balanceswithin conduit transactions.

When we apply recovery assumptions based on the experience from a long-term 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.

Please refer to important disclosures at the end of this report. 63

Page 66: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Hotel Collateral in CMBS

64

chapter 5

CMBSDELINQUENCIES

BY HOTELAND TOTAL

exhibit 9

0.00%

0.50%

1.00%

1.50%

2.00%

Aug-99 Feb-00 Aug-00 Feb-01 Sep-01

HotelTotal/Average

Source: Morgan Stanley, Intex

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 exposureto hotel properties. On average, conduit transactions have 10% hotel exposure.

A search of the Trepp database revealed 70 transactions had hotel exposureexceeding 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, atransaction’s DSCR increases over time as the property operations stabilize andcash flows increase. In addition to the DSCR for the entire transaction we alsoexamined the DSCR for the hotel assets within the transactions. Reporting offinancial statements is typically done on a quarterly basis so the DSCR reportedmaybe be somewhat outdated.

A declining DSCR from issuance should be a flag to investors since nationwidehotels have posted very solid RevPAR growth numbers in 1999 and 2000. SinceRevPAR is anticipated to decline for the remainder of this year and early nextyear hotel assets that were posting declining DSCR prior to September 11 will beunder particular pressure.

CONDUIT TRANSACTIONS

On average, the DSCR for deals with more than 15% hotel concentration but lessthan 100% were 2 bp higher than at issuance. Five transactions had DSCR morethan 2 bp below issuance levels. The transaction that showed the greatest DSCRdecline was Starwood Asset Receivables Trust 2000-1 with a 21 bp decline. Thetransaction had no delinquencies as of the most current remittance report.

The DSCR attributable to just the hotel assets within the transactions we exam-ined 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.

Page 67: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

LARGE LOAN TRANSACTIONS

None of the large loan transactions had any delinquent loans as of the report-ing dates noted. The average loan-to-value ratios (LTV) was 48% on the largeloan transactions, significantly below the typical 65%-75% average leverage inCMBS conduit transactions. The low leverage and improving DSCR indicate that there is significant cushion in these deals to withstand predicted declinesin operating performance.

Since RevPAR growth peaked in 2000, the most vulnerable transactions to a significant decline in operating performance and DSCR would be the transactionsissued in 2001 which may have taken into account 2000 operating performance.The rating agencies anticipate that 1999 should be considered a stabilized operat-ing 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 produceDSCR above underwritten levels.

We examined hotel concentrations in CMBS transactions over time. Specifically,we examined diversified CMBS transactions with greater than 10% hotel expo-sure. Issuance of deals with these characteristics peaked at 28 in 1998 anddeclined to 18 in 1999, 13 in 2000 with only 7 deals issued in 2001. Since therating agencies consider 1999 to be long term stabilized hotel performance, itappears that only 21 of the 103 we examined may have been issued off ofgreater than stabilized performance.

It is also worth noting that the rating agencies have always been cautious aboutthe underwriting of hotel loans, applying haircuts to above-market performance.An average of several years of historical performance has always been consideredstabilized. These underwriting standards are intended to provide cushion withinthe 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 withhotel concentration that exceeds 15% – the 2000 Hilton Hotels transaction (HHPT2000-HLT). In order to assess how that transaction and the 1999 Host Marriotttransaction (HMPT 1999-HMTX) may perform in a declining RevPAR environment,we analyzed the most recent financial statements on the deals and applied vari-ous stress scenarios to the transactions.

In the six weeks after September 11, 2001, RevPAR declines nationwide haveaveraged about 20%, moderating to about 17% in the last three weeks. As high-lighted earlier in this report declines are anticipated to moderate further throughthe 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.

Please refer to important disclosures at the end of this report. 65

Page 68: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Hotel Collateral in CMBS

66

chapter 5

DEALS WITH MORE THAN 15%, BUT LESS THAN 100%HOTEL EXPOSURE

exhibit 10

Source: Trepp, Realpoint, S&P

Deal Name Deal Type Lead Manager

Salomon Brothers Mortgage Securities VII, 2000-FL1 Conduit Salomon Smith BarneyNomura Asset Securities Corp., 1996-MD5 Large Loan NomuraCredit Suisse First Boston Mortgage Securities Corp., 1998-FL2 Short Term Credit Suisse First BostonAsset Securitization Corp., 1996-MD6 Large Loan NomuraCOMM, 2000-FL2 Short Term Deutsche Bank Alex. BrownGS Mortgage Securities Trust II, 2001-GSFL4 Short Term Goldman Sachs & Co.Asset Securitization Corp., 1997-MD7 Large Loan NomuraDLJ Commercial Mortgage Corp., 1998-STF1 Short Term Donaldson, Lufkin & JenretteCredit Suisse First Boston Mortgage Securities Corp., 1998-FL1 Short Term Credit Suisse First BostonAsset Securitization Corp., 1995-MD4 Large Loan NomuraCredit Suisse First Boston Mortgage Securities Corp., 2000-FL2 Short Term Credit Suisse First BostonAllied Capital Commercial Mortgage Trust, 1998-1 Conduit Morgan StanleyDLJ Commercial Mortgage Corp., 1999-STF1 Short Term Donaldson, Lufkin & JenretteLB Commercial Conduit Mortgage Trust, 1996-C2 Conduit Lehman BrothersGS Mortgage Securities Corp., 1998-GL2 Large Loan Goldman Sachs & Co.BTR2 Trust, 1999-S1 Short Term Bankers TrustLB Commercial Mortgage Conduit Trust, 1995-C2 Conduit Lehman BrothersCommercial Mortgage Acceptance Corp., 1997-ML1 Large Loan Merrill LynchCommercial Mortgage Acceptance Corp., 1996-C2 Conduit Goldman Sachs & Co.Asset Securitization Corp., 1996-D3 Conduit NomuraAsset Securitization Corp., 1995-D1 Conduit NomuraLehman Brothers Floating Rate Commercial Mortgage Trust, 2001-LLF4 Short Term Lehman BrothersChase Commercial Mortgage Securities Corp., 1998-SN1 Seasoned ChaseCriimi Mae CMBS Corp., 1998-1 Conduit CitibankCredit Suisse First Boston Mortgage Securities Corp., 1995-AEW1 Seasoned Credit Suisse First BostonLehman Brothers Floating Rate Commercial Mortgage Trust, 2000-LLF C7 Conduit Lehman BrothersCOMM, 2001-FL4 Short Term Deutsche Bank Alex. BrownDLJ Commercial Mortgage Corp., 1998-STF2 Short Term Donaldson, Lufkin & JenretteAsset Securitization Corp., 1996-D2 Conduit NomuraCredit Suisse First Boston Mortgage Securities Corp., 2000-FL1 Short Term Credit Suisse First BostonCOMM, 2001-J1 Private Deutsche Bank Alex. BrownLehman Large Loan, 1997-LL1 Large Loan Lehman BrothersSASCO Floating Rate Commerical Mortgage Trust, 1999-C3 Short Term Lehman BrothersNomura Asset Securities Corp., 1994-C3 Seasoned NomuraDLJ Commercial Mortgage Corp., 1998-CF1 Conduit Donaldson, Lufkin & JenretteSalomon Brothers Mortgage Securities VII, 1996-C1 Conduit Salomon Smith BarneyBlackrock Capital Finance, 1997-C1 Seasoned Goldman Sachs & Co.Chase Commercial Mortgage Securities Corp., 1998-1 Conduit ChaseLB Commercial Conduit Mortgage Trust, 1998-C4 Conduit Lehman BrothersMorgan Stanley Capital I Inc., 1998-XL1 Large Loan Morgan StanleyCredit Suisse First Boston Mortgage Securities Corp., 1997-C1 Conduit Credit Suisse First BostonMorgan Stanley Capital I Inc., 1997-XL1 Large Loan Morgan StanleyCredit Suisse First Boston Mortgage Securities Corp., 1997-C2 Conduit Credit Suisse First BostonGS Mortgage Securities Corp., 1998-C1 Conduit Goldman Sachs & Co.COMM, 2000-C1 Conduit Deutsche Bank Alex. BrownCredit Suisse First Boston Mortgage Securities Corp., 1998-C1 Conduit Credit Suisse First BostonStarwood Asset Receivables Trust, 2000-1 Conduit Merrill LynchNomura Asset Securities Corp., 1995-MD3 Large Loan NomuraMortgage Capital Funding, 1998-MC3 Conduit Salomon Smith BarneyFirst Union - Chase Commerical Mortgage Trust, 1999-C2 Conduit ChaseMorgan Stanley Capital I Inc., 1998-WF1 Conduit Morgan StanleyNomura Asset Securities Corp., 1998-D6 Conduit NomuraJ.P. Morgan Commercial Mortgage Finance Corp., 1998-C6 Conduit JP MorganTotal/Weighted Average

Page 69: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 67

CurrentDeal

Balance($mm) Issue Date % Hotel

HotelDSCR

OriginalDSCR

CurrentDSCR

ChangeFrom

Iss.

%30/60/90

Del.

% For.and

REOTotalDel. Data As of

TopState

% TopState LTV

216.0 1/31/2000 53.7 NA NA NA NA 0.00 0.00 0.00 10/1/2001 NY 53.7 NA726.7 4/2/1996 47.9 2.07 1.90 1.89 -0.01 0.00 0.00 0.00 10/1/2001 TX 14.7 61.0999.3 11/11/1998 46.3 1.59 1.41 1.30 -0.11 0.00 6.69 6.69 10/1/2001 HI 27.5 67.0836.4 12/17/1996 43.5 3.02 1.96 2.57 0.61 0.00 0.00 0.00 10/1/2001 TX 21.6 51.2759.5 7/17/2000 42.9 2.25 1.49 1.56 0.07 0.00 0.00 0.00 10/1/2001 NY 31.3 66.1346.7 2/22/2001 42.3 2.70 2.45 2.51 0.06 0.00 0.00 0.00 10/1/2001 CA 37.6 46.6456.1 3/27/1997 41.1 1.32 1.74 1.75 0.01 0.00 0.00 0.00 10/1/2001 CA 28.0 55.850.5 4/27/1998 40.0 0.92 1.38 1.41 0.03 38.59 0.00 38.59 10/1/2001 GA 38.1 72.8

791.9 6/29/1998 39.7 1.26 1.22 1.18 -0.04 0.00 6.72 6.72 10/1/2001 CA 18.5 83.8867.6 10/30/1995 39.3 1.52 1.77 1.79 0.02 0.00 0.00 0.00 10/1/2001 TX 31.2 56.4619.1 9/4/2001 37.8 1.60 1.55 1.55 0.00 0.00 0.00 0.00 10/1/2001 FL 23.2 61.7119.6 1/23/1998 37.3 1.55 1.38 1.52 0.14 5.64 5.59 11.23 9/1/2001 FL 15.4 61.9170.8 11/30/1999 35.1 1.06 1.16 0.99 -0.17 0.00 0.00 0.00 10/1/2001 NY 38.6 52.2332.5 10/30/1996 33.9 1.82 1.46 1.48 0.02 1.02 0.00 1.02 9/1/2001 GA 11.3 61.1

1,315.9 5/21/1998 33.3 2.16 1.86 1.98 0.12 0.00 0.00 0.00 10/1/2001 CA 17.1 60.7219.2 3/25/1999 30.0 2.32 1.69 1.74 0.05 0.00 0.00 0.00 8/1/2001 FL 43.2 60.9170.6 10/12/1995 29.0 1.31 1.44 1.51 0.07 0.00 0.00 0.00 9/1/2001 TX 20.4 60.9804.0 12/30/1997 27.2 2.01 1.68 2.22 0.54 0.00 0.00 0.00 10/1/2001 CA 16.2 60.986.9 12/30/1996 26.8 2.25 1.22 1.84 0.62 0.00 8.99 8.99 10/1/2001 CA 44.7 NA

724.2 10/22/1996 26.7 1.96 1.44 1.72 0.28 3.23 0.46 3.69 10/1/2001 CA 26.8 62.7164.6 8/7/1995 26.0 2.65 1.53 1.97 0.44 2.91 0.00 2.91 10/1/2001 TX 16.1 58.9

1,242.5 8/16/2001 25.4 2.80 2.90 2.90 0.00 0.00 0.00 0.00 10/1/2001 CA 19.1 51.319.0 6/22/1998 25.1 1.67 1.30 1.94 0.64 0.00 0.00 0.00 9/1/2001 NY 26.8 44.0

422.2 6/4/1998 24.2 2.13 1.59 1.71 0.12 2.10 0.00 2.10 9/1/2001 MI 22.1 67.748.1 10/30/1995 23.7 1.52 1.26 1.52 0.26 5.64 0.00 5.64 9/1/2001 CA 61.7 69.0

1,469.2 12/18/2000 23.4 1.82 1.72 1.72 0.00 0.00 0.00 0.00 9/1/2001 CA 32.0 55.8952.8 6/28/2001 23.3 3.22 2.31 2.31 0.00 0.00 0.00 0.00 10/1/2001 RI 24.9 52.579.6 12/28/1998 23.2 1.99 1.35 1.75 0.40 50.88 22.39 73.27 10/1/2001 VA 50.9 82.5

802.0 3/1/1996 22.7 1.63 1.62 1.60 -0.02 6.51 3.61 10.12 9/1/2001 TX 16.0 70.2328.0 12/27/2000 22.2 2.48 1.50 2.04 0.54 0.00 0.00 0.00 10/1/2001 CA 34.6 58.4790.0 3/29/2001 21.9 1.60 1.73 1.72 -0.01 0.00 0.00 0.00 10/1/2001 NY 23.1 53.5

1,350.8 10/14/1997 21.4 2.16 1.89 2.16 0.27 0.00 0.00 0.00 10/1/2001 CA 24.9 56.21,001.3 10/28/1999 21.2 1.55 1.11 1.23 0.12 0.00 0.00 0.00 9/1/2001 TX 26.6 NA

46.8 11/29/1994 20.8 1.49 1.65 1.71 0.06 0.00 16.97 16.97 10/1/2001 GA 20.8 55.6795.3 3/2/1998 20.1 1.42 1.35 1.57 0.22 0.00 0.18 0.18 10/1/2001 NJ 13.5 62.8107.3 2/29/1996 19.4 1.15 1.37 1.58 0.21 5.86 0.00 5.86 9/1/2001 GA 25.4 61.130.8 10/7/1997 17.6 3.69 1.42 2.00 0.58 2.61 12.59 15.20 9/1/2001 NY 29.6 75.2

763.5 5/15/1998 17.6 2.52 1.65 1.84 0.19 0.00 0.00 0.00 10/1/2001 NY 19.5 67.11,959.8 11/24/1998 17.6 1.66 1.64 1.85 0.21 1.19 0.00 1.19 10/1/2001 CA 20.1 64.4

884.0 6/11/1998 17.6 1.89 1.90 2.08 0.18 0.00 0.00 0.00 10/1/2001 CA 39.0 55.91,245.4 6/30/1997 17.5 2.26 1.38 1.65 0.27 2.32 4.90 7.22 9/1/2001 NY 26.5 64.5

677.4 10/17/1997 17.0 2.27 2.00 2.11 0.11 0.00 0.00 0.00 10/1/2001 NY 28.9 52.11,400.5 12/19/1997 17.0 2.03 1.43 1.55 0.12 0.57 0.89 1.46 10/1/2001 CA 31.5 65.11,785.8 10/29/1998 16.7 1.85 1.53 1.72 0.19 1.54 0.00 1.54 10/1/2001 CA 15.9 66.7

888.9 9/20/2000 16.6 2.20 1.48 1.55 0.07 0.00 0.00 0.00 10/1/2001 MI 20.4 65.62,367.3 6/25/1998 16.5 1.99 1.50 1.65 0.15 2.36 0.13 2.49 9/1/2001 CA 14.4 70.3

779.2 5/17/2000 16.3 1.96 1.29 1.50 0.21 0.00 0.00 0.00 9/1/2001 CA 30.2 66.4403.3 3/31/1995 15.9 1.93 1.60 1.59 -0.01 0.00 0.00 0.00 10/1/2001 NY 39.7 61.5857.6 12/18/1998 15.9 1.87 1.54 1.68 0.14 2.08 0.00 2.08 10/1/2001 CA 11.0 70.3

1,147.9 5/24/1999 15.6 1.52 1.34 1.44 0.10 4.64 0.00 4.64 10/1/2001 CA 13.0 71.31,293.3 3/5/1998 15.3 1.96 1.54 1.74 0.20 0.16 0.00 0.16 10/1/2001 CA 35.0 65.73,576.3 3/30/1998 15.0 3.17 1.58 1.96 0.38 0.24 0.00 0.24 10/1/2001 CA 23.2 67.7

748.0 3/10/1998 15.0 2.59 1.54 2.09 0.55 1.89 0.00 1.89 10/1/2001 VA 21.9 62.441,041.9 23.5 2.10 1.64 1.81 0.17 1.00 0.67 1.66 63.1

Page 70: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Hotel Collateral in CMBS

68

chapter 5

DEALS WITH 100% HOTEL EXPOSUREexhibit 11

Source: Trepp, Realpoint, S&P

Deal Name Deal Type Lead Manager

OriginalDeal

Balance($mm)

Banc of America Large Loan Inc., 2000-WSF-1 Single Asset Banc of America 177.0

Bear Stearns Commercial Mortgage Securities Inc., 1999-WYN1 Single Asset Bear Stearns & Co. 346.0

Bear Stearns Commercial Mortgage Securities Inc., 2000-LCON Single Asset Bear Stearns & Co. 115.8

BTR1 Trust, 1998-S1 Short Term Bankers Trust 422.8

CBM Funding Corp., 1996-1 Single Asset Lehman Brothers 406.2

German American Capital Corp., 1996-2 Single Asset Deutsche Morgan Grenfell 99.2

Hotel First Mortgage Trust, 1993-A Single Asset Nomura 67.5

Hilton Hotels Pool Trust, 2000-HLT Single Asset Morgan Stanley 499.6

Host Marriott Pool Trust, 1999-HMT Large Loan Morgan Stanley 665.0

Meristar Commercial Mortgage Trust, 1999-C1 Single Asset Lehman Brothers 330.0

Merrill Lynch Mortgage Investors Inc., 1998-H1 Single Asset Merrill Lynch 102.3

Nomura Asset Capital Corp., 1993-1 Single Asset Nomura 51.4

Opryland Hotel Trust, 2001-OPRY Single Asset Merrill Lynch 275.0

Lehman Brothers Floating Rate Commercial Mortgage Trust, 2001-C6 Single Asset Lehman Brothers 300.0

Strategic Hotel Capital Inc., 1999-C1 Single Asset Goldman Sachs & Co. 422.0

Strategic Hotel Capital LLC, 2001-SHC1 Single Asset Goldman Sachs & Co. 455.0

Starwood Asset Receivables Trust, 1999-C1 Single Asset Lehman Brothers 541.3

Total/Weighted Average 5,276.1

Page 71: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 69

CurrentDeal

Balance($mm) Issue Date

%Hotel

HotelDSCR

OriginalDSCR

CurrentDSCR

ChangeFrom

Iss.

%30/60/90

Del.% For.

and REOTotalDel. Data As of

TopState

% TopState LTV

176.2 8/30/2000 100.0 3.43 1.90 3.43 1.53 0.00 0.00 0.00 10/1/2001 CA 100.0 62.3

307.8 11/10/1999 100.0 4.00 1.72 4.00 2.28 0.00 0.00 0.00 10/1/2001 CA 20.3 46.4

113.3 8/22/2000 100.0 2.93 2.32 2.93 0.61 0.00 0.00 0.00 10/1/2001 PR 100.0 43.8

22.2 2/13/1998 100.0 2.50 1.65 2.50 0.85 0.00 0.00 0.00 9/1/2001 NJ 100.0 59.6

325.8 1/24/1996 100.0 2.42 1.82 2.42 0.60 NA NA NA 10/1/2001 Multi 100.0 NA

72.7 4/24/1996 100.0 2.24 1.96 2.24 0.28 0.00 0.00 0.00 10/1/2001 CA 34.7 52.8

55.7 8/13/1993 100.0 2.77 2.19 2.77 0.58 NA NA NA 8/1/2001 WA 22.2 42.0

494.0 11/9/2000 100.0 2.11 2.22 2.11 -0.11 0.00 0.00 0.00 11/1/2000 CA 50.0 44.3

631.1 8/18/1999 100.0 2.51 1.84 2.51 0.67 0.00 0.00 0.00 10/1/2001 NY 52.0 45.3

320.6 8/30/1999 100.0 2.22 2.34 2.22 -0.12 NA NA NA 10/1/2001 CA 37.5 52.9

99.4 7/23/1998 100.0 2.57 2.19 2.57 0.38 0.00 0.00 0.00 10/1/2001 FL 100.0 59.5

42.9 9/30/1993 100.0 1.05 NA 1.05 NA NA NA NA 9/1/2001 TN 64.5 NA

271.0 4/19/2001 100.0 1.90 2.02 1.90 -0.12 0.00 0.00 0.00 10/1/2001 TN 100.0 45.1

300.0 9/27/2001 100.0 3.78 NA 3.78 NA NA NA NA 9/1/2001 FL 100.0 42.0

383.5 10/21/1999 100.0 2.74 3.23 2.74 -0.49 0.00 0.00 0.00 10/1/2001 CA 40.2 54.9

454.1 5/8/2001 100.0 2.96 2.82 2.96 0.14 0.00 0.00 0.00 9/1/2001 NY 50.0 47.4

518.5 3/16/1999 100.0 4.11 2.41 4.11 1.70 0.00 0.00 0.00 10/1/2001 CA 38.7 46.4

4,588.7 100.0 2.86 2.25 2.86 0.61 0.00 0.00 0.00 48.0

Page 72: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

HHPT 2000-HLT

Since 1997, the DSCR on this transaction has steadily increased. The transactionwas underwritten based on adjustments to trailing twelve-month financial state-ments as of June 2000 resulting in a 2.31x DSCR. The year-end 2000 financialstatements resulted in a 2.54 DSCR. Year-to-date September financial statementsproduced a 2.11 DSCR.

In order to anticipate future operating performance, several stress scenarios wereapplied to the HLT transaction.

Stress scenario 1 applies a 7% decline in revenues to 2000 operating numberswhile holding expenses constant. The decline in revenues is in line withPriceWaterhouseCoopers nationwide projections for 2001. Although variableexpenses decline with revenues (typically 35%) we used a conservative assump-tion 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 holdsoperating expenses constant resulting in a 1.01 DSCR.

Scenario 3 applies a 24% reduction to year-end 2000 revenues and reduces non-fixed expenses by 8% resulting in a 1.0 DSCR. The 8% reduction in operatingexpenses is based on the industry trend of 35% variable expenses associated withchanges in revenues (.24 x .35 =.08).

Each of these stress scenarios indicates the transaction can withstand a signifi-cant decline in revenues and still generate enough cash flow to cover debtservice payments.

Since the rating agencies view 1999 as a stabilized year, it is also worth notingthat based on 1999 numbers, the transaction generates a DSCR of 2.13x.

The parent company’s financial position and credit rating are also important whenconsidering how large loan transactions might perform.

Hilton is rated Baa3 by Moody’s and BBB- by S&P and Fitch. The MorganStanley analyst that covers the Hilton corporate debt anticipates that the companywill generate over $200 million in free cash flow in 2002. These estimates takeinto account third quarter earnings and significant declines in RevPAR throughthe end of 2001. The corporate debt carries an Outperform rating with the Morgan Stanley analyst.

Transforming Real Estate Finance

Hotel Collateral in CMBS

70

chapter 5

Page 73: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

HMPT 1999-HMTX

The historical operating numbers on this transaction also show significantimprovements since 1997. Between 1997 and year-end 2000, DSCR increased from1.99x to 2.71x. The trailing twelve month (TTM) operating numbers through June2001 (TTM June 2001) resulted in a 2.51 DSCR. The third quarter operating state-ments should be provided by the servicer in the near future. We will providethose numbers in a separate piece once they are available.

We also applied stress scenarios to cash flows from the Host Marriott transac-tion. Scenario 1 applies a 7% decline to TTM June 2001 numbers while holdingall operating expenses constant. These assumptions are based on the 2001PriceWaterhouseCoopers RevPAR projections. Stress scenario 1 results in a 1.92 DSCR.

Scenario 2 applies a 19% reduction to TTM June revenues while holding alldepartmental revenues constant. These assumptions produce a 1.01 DSCR, indicating revenues can decline 18% with no reduction in operating expenseswithout putting debt service payments in jeopardy.

Stress scenario 3 assumes revenues decline 22% with a 7.7% reduction in onlynon-fixed operating expenses resulting in a 1.0x DSCR. The 7.7% reduction inexpenses is based on the assumption that about 35% of the expenses associatedwith 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 12months. However, each of these scenarios indicates that the transaction can experi-ence severe declines in revenues without placing debt service payments in jeopardy.

The DSCR based on 1999 numbers is 2.10x, this performance is considered astabilized year by the rating agencies.

Host Marriott has a corporate credit rating of Ba2 by Moody’s 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.

Please refer to important disclosures at the end of this report. 71

Page 74: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Hotel Collateral in CMBS

72

chapter 5

HILTON HOTELS POOL TRUST 2000 – HLT OPERATINGSTATEMENTS

exhibit 12

Source: Morgan Stanley, Hilton Hotels Corporation

1997($000s)

Revenue(%)

1998($000s)

Revenue(%)

1999($000s)

Revenue(%)

TTMJun-00($000s)

Revenue(%)

Under-written($000s)

Total Departmental Revenues 306,414 100% 338,300 100% 353,367 100% 367,288 100% 367,288

Total Departmental Expenses (127,788) 42% (134,889) 40% (139,744) 40% 142,319 40% (142,294)

Total Departmental Profits 178,626 58% 203,411 60% 213,623 60% 225,163 60% 224,994

Total Undistributed Expenses (65,955) 22% (70,229) 21% (72,596) 21% (78,002) 21% (75,571)

Income Before Fixed Expenses 112,671 37% 133,182 39% 141,027 40% 147,140 40% 149,423

Total Fixed Expenses (24,474) 8% (23,628) 7% (26,329) 7% (27,525) 7% (26,329)

Net Operating Income 88,197 29% 109,554 32% 114,698 32% 119,616 32% 123,094

FF&E Reserve (14,399) 5% (15,901) 5% (16,625) 5% (17,300) 5% (16,625)

Net Cash Flow 73,798 24% 93,653 28% 98,073 28% 102,215 28% 106,469

Debt Service 46,126 15% 46,126 14% 46,126 13% 46,126 13% 46,126

Debt Service Coverage Ratio 1.60 2.03 2.13 2.22 2.31

Portfolio StatisticsAnnualGrowth

AnnualGrowth

AnnualGrowth

Occupancy 75% 74% -1.0% 76% -1.0% 76% -1.0% 76%

Average Daily Rate (ADR) $144.90 $161.75 12% $167.01 3% $174.32 4% $174.30

Rev Per Available Room

(RevPAR)

$109.19 $119.18 9% $126.07 6% $131.99 5% $132.14

Stress ScenariosScenario 1 7% decline in revenues, all expenses held constant at December 2000 levelsScenario 2 19% decline in revenues, all expenses held constant at December 2000 levelsScenario 3 24% decline in revenues, non-fixed expenses at 92% of December 2000 levels

Page 75: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 73

Revenue(%)

2000($000s)

Revenue(%)

First 9Months

2001($000s)

Revenue(%)

StressScenario 1

Revenue(%)

StressScenario 2

Revenue(%)

StressScenario 3

Revenue(%)

100% 389,400 100% 256,300 100% 362,142 100% 315,414 100% 295,944 100%

39% (147,600) 38% (102,300) 39% (147,600) 41% (147,600) 47% (135,792) 46%

61% 241,800 62% 154,000 61% 214,542 59% 167,814 53% 160,152 54%

21% (78,200) 20% (54,300) 20% (78,200) 22% (78,200) 25% (71,944) 24%

41% 163,600 42% 99,700 41% 136,342 38% 89,614 28% 88,208 30%

7% (27,100) 7% (21,200) 9% (27,100) 7% (27,100) 9% (27,100) 9%

34% 136,500 35% 78,500 32% 109,242 30% 62,514 20% 61,108 21%

5% (19,470) 5% (12,815) 5% (18,107) 5% (15,771) 5% (14,797) 5%

29% 117,030 30% 65,685 27% 91,135 25% 46,743 15% 46,311 16%

13% 46,126 12% 31,090 13% 46,126 13% 46,126 15% 46,126 16%

2.54 2.11 1.98 1.01 1.00

U/WAssumption

-1.0%

0%

0%

Page 76: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Hotel Collateral in CMBS

74

chapter 5

HOST MARRIOTT 1999 – HMT OPERATING STATEMENTSexhibit 13

Source: Morgan Stanley, Host Marriott

1997($000s)

Revenue(%)

1998($000s)

Revenue(%)

1999($000s)

Revenue(%)

Underwritten($000s)

Total Departmental Revenues 449,345 100% 501,798 100% 530,473 100% 484,636Operating Expenses (273,150) 61% (295,149) 59% (311,104) 59% (293,570)Total Departmental Profits 176,195 39% 206,649 41% 219,369 41% 191,066Fixed Charges (29,395) 7% (39,400) 8% (64,964) 12% (53,642)Income Before Fixed Expenses 146,800 33% 167,249 33% 154,405 29% 137,424FF&E Reserve (18,783) 4% (22,198) 4% (19,450) 4% (19,450)

Net Cash Flow 128,017 28% 145,051 29% 134,955 25% 117,974Debt Service 64,197 14% 64,197 13% 64,197 12% 64,197DSCR (NCF/Debt Service) 1.99 2.26 2.10 1.84

AnnualGrowth

AnnualGrowth

AnnualGrowth A

Occupancy 81.4% N/A 83.1% 1.7% 82.5% -0.6% 81.5%Average Daily Rate $174.00 N/A $192.00 10% $195.00 1.6% $184.00Revenue Per Available Room(RevPAR)

$142.00 N/A $159.00 12% $161.00 1.3% $150.00

Stress ScenariosScenario 1 7% decline in revenues, all expenses held constant at June 2001 TTM levelsScenario 2 18% decline in revenues, all expenses held constant at June 2001 TTM levelsScenario 3: 22% decline in revenues, expenses at 93% June 2001 TTM levels

Page 77: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 75

Revenue(%)

2000($000s)

Revenue(%)

TTM Jun-01($000s)

Revenue(%) Scenario 1

Revenue(%) Scenario 2

Revenue(%) Scenario 3

Revenue(%)

100% 590,894 100% 572,297 100% 532,236 100% 446,392 100% 441,241 100%61% (332,225) 56% (322,505) 56% (322,505) 61% (299,929) 67% (295,092) 67%39% 258,669 44% 249,792 44% 209,732 39% 146,462 33% 146,149 33%11% (58,770) 10% (59,771) 10% (59,771) 11% (59,771) 13% (59,771) 14%28% 199,899 34% 190,021 33% 149,961 28% 86,691 19% 86,378 20%4% (25,941) 4% (28,615) 5% (26,612) 5% (22,320) 5% (22,062) 5%

24% 173,958 29% 161,407 123,349 64,372 64,31613% 64,197 11% 64,197 64,197 64,197 64,197

2.71 2.51 1.92 1.00 1.00

UWAssumption

-1.0%-5.6%-6.8%

Page 78: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

WHAT WAS HOT IS NOW NOT

Prior to the significant decline in air traffic after September 11, 2001, full-serv-ice urban hotels were considered the most desirable properties because ofstrong corporate demand and high barriers to entry. Hotels in secondary andsuburban markets were less desirable because of significant new developmentin those markets.

Recently, the urban properties in major metro areas have experienced greaterdeclines in RevPAR than the nationwide average, while secondary markets andsuburban properties are benefiting. This is explained by the desire of businessand leisure travelers to drive rather than fly to locations.

In terms of geographic distribution, major cities suffered the most since mid-September. On average, most of the major 25 metropolitan markets have under-performed the nationwide RevPAR trends. The markets hardest hit since mid-September are Boston, Miami, New York, Orlando and San Francisco.

Houston, Nashville, Norfolk, Philadelphia, St. Louis, and Tampa have producedresults better than the national average.

Smith Travel data shows that nationwide RevPAR was down about 4.0% year todate through September 2001. The Persian Gulf War, which occurred during thefirst quarter of 1991, resulted in a 2.4% decline in RevPAR. It took a full year forlodging demand to recover after the Persian Gulf War. Demand will likely beharder 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.

Transforming Real Estate Finance

Recent Performance of Hotels

76

chapter 5

Page 79: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 77

HOTEL MARKET DATA AS OF YTD SEPTEMBER 2001exhibit 14

Source: Smith Travel Research

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

2001 YTD 2000 1999 1998 1997Luxury 68.5 71.8 71.9 72.6 78.4

Upscale 64.4 65.1 64.9 65.9 70.8Mid-Price 60.9 61.5 61.1 62.0 67.9Economy 58.5 58.5 58.0 58.4 61.2

Budget 60.5 59.5 58.7 58.7 59.9

Luxury 147.43 144.99 138.88 133.58 136.01Upscale 94.68 90.88 87.37 85.33 92.07

Mid-Price 70.77 68.23 64.89 62.15 66.97Economy 54.92 52.44 49.23 47.14 49.93

Budget 41.46 42.20 89.60 37.43 40.60

HOTEL RATES BY PRICE CATEGORY AS OF SEPTEMBER 2001 (IN %)

exhibit 15

OccupancyRates

(%)

Source: Smith Travel Research

RoomRates

($)

Page 80: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

BRANDING

As the hotel industry has evolved, companies such as Marriott have segmentedthe market in order to meet customers’ needs, increase profitability, and diversifyits customer base. Branding has been used to attract guests to various price pointsand increase the parent company’s exposure.

Positive brand identity is beneficial for the hotel manager and the hotel owner. Ifthe brand is effective in producing demand, management will spend less timemarketing nationally and focus on local operations and solicitations. The owneralso benefits financially from increased business through the reservation systemand through potentially lower marketing costs.

As the economy weakens consumers move down one or two positions in serviceand price from say a Hilton to an Embassy Suites. As the economy strengthensguests typically move up in service level and price.

UPPER UPSCALE

These hotels are exclusive properties within major metropolitan markets that pro-vide extensive amenities and high levels of service. A hotel within this segmentwill have the highest ADR in its market and often the highest RevPAR in the mar-ket. Amenities at these properties usually include health club/spa facilities, three-or four-star food and beverage outlets, concierge service, 24-hour room service,valet, and retail spaces. These properties are located in prime downtown andresort real estate locations.

In general because of higher construction costs and high barriers to entry, luxuryhotel 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 firstto experience a decline in occupancy as travelers become more budget-consciousand move down the service-level curve to less expensive accommodations.

Transforming Real Estate Finance

Hotel Branding and Segmentation

78

chapter 5

HOTEL BRANDINGAND

SEGMENTATION

exhibit 16Upper-Upscale

HiltonMarriott

HyattFour SeasonsRitz-Carlton

W HotelsUpscaleDoubletree

Embassy SuitesWyndham

Crowne PlazaCourtyard

Midscale(Without F&B)

Hampton InnFairfield

Residence InnComfort InnEconomy

Days InnRed Roof InnTravel Lodge

Super 8Motel 6

MoreExpensive

LessExpensive

HighService

ModerateService Midscale

(With F&B)Holiday InnFour Points

As Economy

Strengthens

As Economy

Weakens

Source: Morgan Stanley

Page 81: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

UPSCALE

Upscale hotels are full-service hotels that cater to individual business travelers,groups, and conventions. Typically the price points at these hotels are betweenupper upscale hotels and mid-scale hotels. Amenities at upscale hotels includeseveral food and beverage outlets, extensive meeting space, laundry service,concierge, and exercise facilities. Most of these properties enjoy downtown loca-tions near convention centers or strong suburban locations. These locations mayhave some barriers to entry because of limited availability of land. The guest pro-file 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 dependingon the market and the age of the property. Amenities at these hotels usuallyinclude one restaurant, limited meeting space, and an exercise facility. Typically,these hotels cater to the more budget-conscious business travelers or “road war-riors.” Older properties in this segment can be somewhat outdated and sufferfrom inefficiencies. Concerns about this segment center around older properties inneed of renovation and a tendency for travelers to prefer newer properties thatare entering many markets at similar price points.

MID-SCALE WITHOUT F&B CHAINS

Properties in this segment typically have secondary locations such as majorhighway intersections, airports, or suburban locations. The properties do notoffer food and beverage amenities except for continental breakfast in somelocations, and may or may not have small exercise facilities. Food and bever-age 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 forthese 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 developmenttimetable is short. The typical guest at these properties is a transient budget-conscious business or pleasure traveler. Most bookings are made through thereservation system and repeat business is not significant.

Please refer to important disclosures at the end of this report. 79

Page 82: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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

1790–1920 GROWTH OF AMERICA AND F INANCIAL BOOM

The first American hotel, the 73-room City Hotel, was built in New York City in1794, and was much larger than its predecessors, the colonial inns. Between thelate 1700s and the mid-1800s, similar hotels were built in major metropolitan citiessuch as Philadelphia, Boston, and Baltimore. During the 1800s, hotels followedthe railroads westward and hotels were built in Chicago, Denver, and SanFrancisco. Now famous historic spas such as The Greenbrier in West Virginiawere built during this period. The Buffalo Statler hotel, built in 1908, was revolu-tionary because it offered baths in each guest room.

During the Roaring ‘20s, as with many other areas of commerce, hotel demandand hotel development boomed. The 3,000-room Hotel Stevens (currently theChicago Hilton) was built in 1927. Hotels were built through community fundedbonds and often without regard for feasibility. Conrad Hilton began his chain ofHilton hotels during the 1920s by purchasing and developing eight hotels. Somany hotels entered the market nationwide in the 1920s that occupancy wentfrom 86% in 1920 to 68% in 1928, while room rates remained fairly constant.

1930–1940s THE GREAT DEPRESSION

The boom of the’20s was followed by the bust or Great Depression of the 1930s.Due to overbuilding and a significant decline in demand, more than 80% ofAmerican hotels went into foreclosure. Nationwide occupancy declined to 50%.

Enterprises that had cash during this period were able to obtain hotels at deepdiscounts to the original costs. Ernest Hendersen founded the Sheraton hotelchain in 1937 and was able to build the company’s portfolio during the 1930s and1940s. The hotel industry did not recover from the overbuilding of the 1920s untilthe early 1940s. World War II stimulated the economy and generated demand forhotel rooms as troops and military personnel were moved domestically and glob-ally, an effect that continued after the war. By the mid-1940s nationwide occupan-cy had increased to 90%.

1950–1960s BOOM AFTER WWI I

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

Transforming Real Estate Finance

History of the Hotel Industry

80

chapter 5

Page 83: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

1970s QUICK CYCLE

With several chains established and the motel concept fully developed, the early1970s became another boom for the hotel industry. At the same time, companiessuch as Holiday Inn and Marriott were looking to expand their presence throughfranchising. The extensive capital available in the markets and ability to franchisebrands encouraged development of hotels, which resulted in overbuilding.

In 1974, inflation caused construction costs and interest rates to rise. The energycrisis reduced road travel and the recession prohibited business travel. This wasanother period of foreclosures and excess hotel supply. Although the overbuild-ing of the 1970s was similar to the 1920s the recovery to reasonable real estateprices came about more quickly than in the 1920s. By the end of the late-1970shotel prices had recovered due to the lack of building during the end of thedecade. By the end of the 1970s supply and demand were more in balance thanfive years prior and occupancy levels were rising.

1980s

By 1983 inflation had slowed and new tax policies created a favorable environ-ment for hotel development. The Savings and Loans (S&Ls) were in the processof deregulation and were an eager source of financing for real estate develop-ment. Lack of commercial real estate experience by the S&Ls resulted in looseunderwriting standards and liberal lending policies. During this time, syndicatedLimited Partnerships were the source of equity capital. Hotel partnerships weresold to wealthy individuals attracted by the aggressive growth projections and thetrophy real estate. In order to capture the greatest tax benefits for these invest-ments, this real estate was highly leveraged at 90%–100%. Investors benefitedfrom passive tax losses in the short term and real estate appreciation in the longerterm. A change in the tax laws in the late 1980s resulted in the revision of theallowance for these losses, slowing new development, but overbuilding wasalready under way, particularly in full-service hotels.

EARLY 1990s

The late 1980s to 1991 was a period of overbuilding in virtually all sectors of realestate, and hotels were no exception. Profitability nationwide for the hotel indus-try was negative between 1986 and 1992. Demand sharply declined from its sig-nificant growth in late 1989, to a cycle low in 1991. Unfortunately developmentwas already in the pipeline and supply growth was high. During this time, manyhotels defaulted on loans because they were so highly leveraged and unable tomeet debt payments. In the early 1990s, the federal government formed theResolution Trust Corporation (RTC) to resolve the problems of the S&Ls, whowere suffering from defaulted real estate loans. Development during the early1990s was virtually halted and banks were unwilling to lend on real estate in gen-eral, hotels in particular. Capitalization rates during this period climbed, asinvestors demanded extremely high returns. Because of a lack of debt funding,most hotel purchases were made by companies with cash reserves.

Please refer to important disclosures at the end of this report. 81

Page 84: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The real estate devaluation and lock up in capital reduced the number of individ-ual hotel transactions from 130 in 1990 to 56 in 1991. Individual hotel transactionsdid not return to 108 again until 1994. Purchase prices per room were well-belowdevelopment 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 bot-tom of $80,000 in 1995.

MID- TO LATE 1990s

By late 1995, the RTC disposed of most real estate previously held by thedefunct S&Ls. Demand for hotel rooms outpaced supply between late 1992 andmid-1996. From 1993, increases in the average daily room rate (ADR) outpacedthe consumer price index. Beginning in 1994, the increase in profitability beganto attract investors. Prices began to increase as demand was outpacing supplyand transactions increased.

Beginning in 1993, hotel REITs began accessing the public capital markets. In1993, $34 million was raised in the capital markets for hotel REITs. This increasedto $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 capitalfueled supply growth and transaction activity for hotels as REITs acquired moreproperties and consolidation occurred. The consolidation of hotel ownership fromprivate partnerships to public companies increased operating efficiencies and heldowners accountable to shareholders, reducing hotel leverage levels. In generalhotel REITs did not exceed 60% leverage.

Capital flow into hotel REITs declined significantly in 1998 to just over $500 mil-lion. This was due to Congress enacting legislation limiting acquisitions by paired-share REITs, REIT investors’ fear of overbuilding in all real estate sectors, slowergrowth in REIT earnings than the broader market, and the capital flight to qualityin the fall of 1998. The change in the paired-share legislation significantly affectedStarwood and Patriot American (now Wyndham) the two largest paired-sharehotel REITs, stopping their growth initiatives.

Transforming Real Estate Finance

Hotel Collateral in CMBS

82

chapter 5

CAPITALIZATIONRATES AND 10-

YEAR TREASURYRATES

exhibit 17

0%

3%

6%

9%

12%

15%

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001E

Interest Rate

Cap-RateSupply

Source: PriceWaterhouseCoopers, Smith Travel Research

Page 85: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

During the mid to late 1990s, as the CMBS market grew, hotels were includedin pools of diversified mortgages. The nightly rents and intense on-site man-agement combined with poor performance prior to the RTC clean-up resultedin lower LTV ratios for hotel loans than other types of real estate within CMBS transactions.

Please refer to important disclosures at the end of this report. 83

Page 86: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

EXECUTIVE SUMMARY

• Commercial mortgage loans differ from residential mortgage loans in that theyare call protected. Commercial mortgage loans typically contain provisions thateither prohibit or economically penalize the borrower for prepaying the loanbefore maturity.

• There are four main categories of call features in commercial mortgage loans:hard or legal lockout, yield maintenance, fixed percentage penalty points anddefeasance. Most commercial mortgage loans contain at least one of these formsof 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 determiningrelative value.

• Lockout and defeasance provide investors with the greatest average life stability. Credit events are the only source of cash flow variability in CMBSdeals where the underlying loans are locked out or defeased. With yield maintenance, some investors benefit from faster prepayments under someinterest 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 veryimportant aspect: call protection. Unlike single-family loans, commercial loanstypically are not fully prepayable at the borrower’s option. Call protection in com-mercial loans stems from the life insurance companies’ historical position as theprimary provider of capital to the real estate industry. As the dominant long-termlender in the market, the life insurance companies required call protection as astandard feature of the commercial loan market. Call protected loans were anattractive asset for an insurer to match against long duration liabilities. The growthof the CMBS market has altered the form of call protection found in newly origi-nated commercial loans. As an increasing percentage of commercial loans arebeing securitized, the preferences of bond investors are playing a large role indefining 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.

Transforming Real Estate Finance

Call Protection

84

chapter 6

Page 87: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

DEFINIT ION 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. Thelockout provision legally prohibits a borrower from prepaying a loan prior toscheduled maturity. The majority of commercial mortgage loans are not “lockedout” for the entire term of the loan. Lockout periods are usually three to fiveyears and are followed by penalty periods. During the penalty period, the borroweris allowed to prepay the loan, but the borrower must compensate the lender for theearly termination right. The two forms of penalty are yield maintenance and fixedpercentage penalty points. Further, most lockout provisions allow a borrower toassign 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 interestlost as a result of prepayments. The yield maintenance penalty in commercialmortgage loans is analogous to the “make whole” provisions found in corporatebond markets. Formulas used to calculate yield maintenance vary. Generally, theformulas provide a present value calculation of the positive interest differentialbetween 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 thanat the time of loan origination, the borrower would not be required to make apenalty 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 orreference rate. The reference rate is compared to the existing mortgage loan rateto calculate the prepayment penalty. Reference rates are usually a comparablematurity Treasury rate, (referred to as “Treasuries flat”), or a comparable maturityTreasury rate plus a spread. Clearly, the lender/investor prefers Treasuries flat asthe reference rate. Treasuries flat results in a higher present value for the yieldmaintenance calculation. The following is an example of a yield maintenance calculation at Treasuries flat.

Assumptions

• Loan Origination Date: 1/01/98 • Original Loan Balance: $10,000,000

• Loan Maturity Date: 12/31/07 • Mortgage Rate: 7.25%

• Start of Yield Maintenance Period: 1/01/98 • Reference Treasury Rate: 5.75%

Please refer to important disclosures at the end of this report. 85

Page 88: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

To simplify the math, we will assume the loan prepays on the origination date.

As the term to maturity for the loan shortens, the yield maintenance penalty as a percentage of the remaining balance decreases. Yield maintenance penaltiesprovide less of a disincentive to the borrower to prepay as the remaining penaltyperiod declines. The importance of the yield maintenance penalty also decreasesin importance to the lender as the remaining penalty period declines. As the termto maturity decreases, the remaining loan payments represent a lower percentageof the lender’s total return.

The following table illustrates the difference between reference rates of Treasuriesflat, Treasuries +25 bp, and Treasuries +50 bp as the penalty period shortens from10 years to three using the same assumptions as the prior example.

F IXED PERCENTAGE PENALTY POINTS

Fixed percentage penalty points are potentially the weakest form of call protection.The prepayment penalty is a fixed percentage of the remaining loan balance. Fixedpercentage penalty points typically decline over the life of the loan. A representativeexample of the terms of a loan with fixed percentage penalty points would includea 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 ratemoves and large increases in property values may overwhelm these fixed economicdisincentives to prepay a fixed percentage penalty loan as the penalties do notchange with interest rates.

Transforming Real Estate Finance

Call Protection

86

chapter 6

(PRESENT VALUE FACTOR) x (MORTGAGERATE – TREASURY RATE) x (REMAININGLOAN BALANCE)

Yield Maintenance Formula

1-(1+Reference Treasury Rate)-Penalty Period

Reference Treasury Rate1-(1+5.75%)-10

5.75%7.45%

(7.45)x(7.25%-5.75%)x($10,000,000)=$1,117,500

11.18% of the remaining loan balance

Present Value Factor =

Yield Maintenance

=

=

=

=

PREPAYMENT PENALTY AS A PERCENTAGE OF OUTSTANDING LOAN BALANCE

exhibit 1

Penalty PeriodReference Rate 10-YR 5-YR 3-YR

Treasuries Flat 11.18 6.36 4.03Treasuries + 25 bp 9.20 5.27 3.34Treasuries + 50 bp 7.27 4.18 2.66

Source: Morgan Stanley

Page 89: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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

DEFEASANCE

Defeasance, a mainstay of the municipal market, has found its way into thecommercial mortgage market. From the investor’s perspective, a loan with adefeasance appears “locked-out” from prepayment. The borrower may prepaythe 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 witha series of US Treasury strips which match the payment stream of the mortgageloan as collateral for the loan. Not only is an investor indifferent to a prepaymentin a defeased loan, the investor actually prefers it. If the borrower exercises adefeasance option, the investor receives the benefit of improved credit quality onthe 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 mortgageloan with a 7% coupon, three-year term to maturity, and annual payments. Ifthe borrower wanted to defease the loan, the borrower would purchase threeUS Treasury strips that would replicate the payments due on the mortgage loan. For the US Treasury strips, we assumed the following rates and prices:

Please refer to important disclosures at the end of this report. 87

Penalty Loan Term to MaturityPoints 10-YR 8-YR 6-YR 4-YR 2-YR

10 150 180 220 300 5409 140 160 200 270 4908 120 140 170 240 4307 110 130 150 210 3806 90 110 130 180 3305 80 90 110 150 2804 60 70 90 120 2203 50 60 70 90 1702 30 40 50 60 1101 20 20 20 30 60

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

exhibit 2

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

Page 90: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Rates(%) Dollar Price• 1-Year 5.65 94.36• 2-Year 5.68 89.20• 3-Year 5.73 84.21

The chart shows the payments due from the borrower on the commercial mort-gage loan, the cost to purchase the US Treasury strips and the payments to theCMBS trust from the US Treasury strips.

Nomura Asset Capital Corporation was the first originator to introduce loans witha 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 outfrom 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 collateralsecuring the loan is now US Treasury securities. A prepayment through the exercise of a defeasance option provides no disruption in cash flow to the bondinvestor whether interest rates are higher or lower.

The amount of defeased collateral in CMBS deals has increased dramatically. TodayCMBS transactions generally have > 90% defeasance. Investors are valuing interest

Transforming Real Estate Finance

Call Protection

88

chapter 6

CASH FLOWS TO CMBS TRUST UNDER DEFEASANCEexhibit 3

Payment Due Cost to Borrower Payments FromFrom Borrower to Purchase US Treasury Strips

Time on Loan US Treasury1 to CMBS Trust Year 1 $700,000 $660,520 $700,000Year 2 $700,000 $624,400 $700,000Year 3 $10,700,000 $9,010,470 $10,700,000

1At time of prepayment. Source: Morgan Stanley

Page 91: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

only bonds (“IOs”) from CMBS deals backed by defeased collateral at tighterspreads than those backed by collateral with yield maintenance penalties. Thegrowth in CMBS issuance has resulted in an increase in both the number andtype of CMBS investors. CMBS backed by defeased mortgage loans offer investorsthe opportunity to avoid the complexities of commercial mortgage prepaymentanalysis. The CMBS bonds that are created from defeased loans resemble corporate bullet securities and have attracted corporate crossover buyers.

ALLOCATION OF PREPAYMENT PENALTIES

For loans with yield maintenance penalties or fixed percentage penalty points, theallocation of these penalties to the various securities in the CMBS structure differson 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 whilethe amount of penalty paid to the coupon bondholders was capped at some percentage, typically 0–25%.

More recent deals generally allocate the prepayment penalties in a way thatmakes the currently paying bond class “whole” and distributes the remainingpenalty to the IO. This more recent allocation method is analogous to the calcu-lation of the yield maintenance penalty on the underlying loan. The currentlypaying bond investor receives compensation for the early return of principal in a lower interest rate environment. The IO holder generally receives 65–75% of the penalty while the current principal paying bond receives the remainder making it whole to the bond’s coupon, not “Treasuries flat.”

The following is a representative example of a yield maintenance, penalty sharingformula found on post-1996 deals. The class that is currently receiving principalwould receive an amount equal to the ratio of the difference between the bondrate and the reference Treasury rate and the difference between the mortgageloan 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 referenceTreasury rate. The CMBS bond coupon is generally lower than the mortgageloan 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 andIO investors need to be aware of the type of prepayment penalty sharing agreementfound on a particular CMBS transaction as it influences an investor’s return.

Please refer to important disclosures at the end of this report. 89

Bond Pass Through Rate - Reference Treasury RateMortgage Loan Rate - Reference Treasury Rate6.45%8.25%5.65%

6.45%-5.65%8.25%-5.65%.802.6030.7% of penalty

Prepayment Distribution Formula =

Bond Pass Through Rates

Mortgage Coupon Rate===

=

Reference Treasury Rate

Currently Paying Bond Receives

=

=

Page 92: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

RELATIVE VALUE ANALYSIS OF D IFFERENT TYPES OF CALL PROTECTION

The relative value analysis of yield maintenance versus lockout/defeasance is ananalysis of the benefits of cash flow certainty versus the potential opportunities ofcash flow uncertainty. With respect to locked-out or defeased deals, the only cashflow variability in the bond will be related to credit events. Movements in interestrates, 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, isnot as straightforward. Earlier, we compared yield maintenance to “make whole”provisions found in the corporate market. Even if yield maintenance leaves thecommercial mortgage lender indifferent to prepayments, various securities withina given CMBS transaction and across different CMBS deals will perform different-ly. In addition to the credit performance of the loans, the interest rate environ-ment, prepayment speeds in various interest rate environments, and the allocationof prepayment penalties within a particular deal structure will determine the yieldin a bond with yield maintenance loans. An investor in yield maintenance backedCMBS must analyze all of these variables in assessing relative value.

Market convention in CMBS scenario analysis is to assume that loans do not pre-pay during their yield maintenance periods. If 100% of the loans in the pool areyield maintenance protected for their entire term, the assumption of “yield main-tenance equals lockout” provides the same result as defeasance or lockout. Forthe sake of this analysis, we assumed that loans in yield maintenance do prepayin order to ascertain under which scenarios an investor would prefer to ownbonds with yield maintenance over bonds with defeased collateral.

We analyzed several classes of bonds from different generations of CMBS transac-tions with yield maintenance as the predominant form of call protection on theunderlying loans1. The purpose of this analysis was to determine which bonds benefit in various interest rate and prepayment scenarios. We performed prepaymentand interest rate scenario analysis on CMBS deals that allocate all of the prepaymentpenalties to the IO and on CMBS deals that allocate prepayment penalties that makethe currently paying bonds whole. We analyzed each bond at prepayment speedsfrom 0–100% 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, andmay vary in different interest rate environments.

Transforming Real Estate Finance

Call Protection

90

chapter 6

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

Page 93: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

SHORT AAA (1ST PAYMENT PRIORITY) BONDS WITH SHARED YIELDMAINTENANCE PENALTIES

Because the short AAA bonds are at a discount dollar price in the interest rateshigher scenario, faster prepayments during penalty periods result in a higher yieldto maturity for the investor. The borrower is not required to pay a prepaymentpenalty in an interest rates higher scenario, but the borrower is required to paythe 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 aretwo potential reasons to prepay:

1) Sale of the property

2) Refinancing driven prepayment

a) Equity take-out refinancingb) Fixed to floating rate refinancing

In the interest rates unchanged and lower scenarios, the yield to maturity alsoincreases as prepayments increase during yield maintenance penalty periods. Theamount of the prepayment penalty received by the short AAA holder exceeds thepremium 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: 9 years

Please refer to important disclosures at the end of this report. 91

The yield maintenance penaltyin this example would equal:

Present Value Factor)x(Mortgage Rate-Treasury Rate)1-(1+Reference Treasury Rate)-Penalty Period

Reference Treasury Rate1-(1+5.65%)-9

5.65%6.91%(6.91)x(8.80%-5.65%)21.8%Bond Coupon Rate-Reference Treasury RateLoan Coupon-Reference Treasury Rate6.85%-5.65%8.80%-5.65%38.1%

(38.1%)(21.8%)8.3%

6%2.3% (of current balances)

Yield Maintenance FormulaPresent Value Factor

Yield Maintenance

Allocation to Short AAA

AllocationPenalty Points Allocated

to Short AAA

vs. Premium Dollar Price of Short AAA

Advantage

=

=

=

=

=

=

=

=

=

=

==

Page 94: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The fact that the prepayment penalty allocated to the investor is higher than thepremium dollar percentage results in the prepayment benefiting the investor.

While the yield to maturity is higher in this scenario, the average life of this bonddecreases. For some investors, the shortening of the average life of the bond mitigates the benefits of a higher yield to maturity.

SHORT AAA BONDS WHICH DO NOT SHARE PREPAY PENALTIES

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 atpar. If interest rates were to increase enough so that the bonds were trading at adiscount dollar price, the yield to maturity would increase as prepayment speedsincrease during yield maintenance periods.

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 resultinghigher price sensitivity to interest rates. As a result, the AA bond is more likely tobe trading at a discount dollar price at smaller increases in interest rates than theshort AAA bond. So, in an interest rates up 100 bp scenario on an older deal, theyield to maturity on the short AAA bond decreases as prepayments increase, butthe yield to maturity on the AA bond increases. The AA bond is priced at a dis-count 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 maturitythan the 0% CPR case. Prepayment speeds must be fast enough to retire the AAAbonds during the yield maintenance period in order for the yield to maturity toincrease on the AA bond. If the AAA securities are retired and the AA bondbecomes the currently paying bond, the AA bond receives a portion of the prepay-ment penalties. If the AA bond does not become the currently paying bond duringthe yield maintenance period, it does not receive any of the prepayment penalties.The average life shortens in either case.

Transforming Real Estate Finance

Call Protection

92

chapter 6

Interest RatesBond Class –300 bp –100 bp Unchanged +100 bp +300 bp

Short AAA Decreases Decreases Decreases Decreases Increases(IO ReceivesAll Penalties)

Short AAA Increases Increases Increases Increases Increases(Make Bonds Whole)

EFFECT ON YIELD TO MATURITY ASPREPAYMENT SPEEDS INCREASE—SHORT AAAs

exhibit 4

Source: Morgan Stanley

Page 95: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

INTEREST ONLY BONDS

IOs from CMBS deals with yield maintenance loans generally benefit as prepay-ment 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 fromthe borrower, the IO holder does not benefit from faster prepayments. In thiscase, the IO bond loses the income stream from the prepaid loan and does notreceive a compensating prepayment penalty. Unlike a defeased IO, an IO from a CMBS deal backed by yield maintenance loans offers investor(s), the potentialfor higher returns with faster prepayments under certain interest rate scenarios.

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 doborrowers prefer? In some scenarios, yield maintenance is more expensive for bor-rowers and in some scenarios defeasance is more expensive for borrowers. Wepresent three numerical examples of the costs to the borrower of defeasance versusyield maintenance in the Appendix. Yield maintenance and defeasance have verysimilar penalty calculations in current to lower interest rate environments althoughdefeasance may be slightly more expensive in steeper yield curve environments. Inrising interest rate environments, yield maintenance and defeasance prepaymentdisincentives tend to decline, but yield maintenance is eventually disadvantageousto the borrower. We think the increasing influence of the CMBS market in commer-cial real estate finance will result in the dominance of defeasance in commercialmortgage loans.

Please refer to important disclosures at the end of this report. 93

Interest RatesBond Class –300 bp –100 bp Unchanged +100 bp +300 bp

AA Decreases Decreases Decreases Increases Increases(IO ReceivesAll Penalties)

AA Varies Varies Increases Increases Increases(Make Bonds Whole)

EFFECT ON YIELD TO MATURITY AS PREPAYMENTSPEEDS INCREASE (AA)

exhibit 5

Source: Morgan Stanley

Interest RatesBond Class –300 bp –100 bp Unchanged +100 bp +300 bp

IO Increases Increases Increases Increases Decreases(IO ReceivesAll Penalties)

IO Increases Increases Increases Increases Decreases(Make Bonds Whole)

EFFECT ON YIELD TO MATURITY AS PREPAYMENTSPEEDS INCREASE (IO)

exhibit 6

Source: Morgan Stanley

Page 96: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Investors in CMBS have different sets of opportunities and investment decisions withdefeased deals versus deals with yield maintenance. The defeased deal eliminates the cash flow volatility resulting from commercial mortgage prepayments. Investors inCMBS with yield maintenance call protection have exposure to the benefits and risksresulting from prepayment driven cash flow variability. Investors should carefully ana-lyze the effect of interest rate movements and prepayments on the yield of individualCMBS classes.

APPENDIX: THE BORROWER’S PERSPECTIVE

The following example compares a prepayment under defeasance to a prepayment under yield maintenance from the borrower’s perspective.

Assume the underlying loan has the following characteristics:

• Amount: $10,000,000• Term: 10 years• Amortization: 25 years• Coupon: 7.50%

We will assume that the reference rate on the yield maintenance penalty isTreasuries 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 sellthe property for personal reasons.

• Loan Balance at the End of Year 5: $9,189,718• Yield Maintenance Penalty: $779,020 (8.48%)• Cost of Treasury Strips: $9,970,826• Total Defeasance Cost: $781,108 (8.50%)

($9,970,826–$9,189,718)• Defeasance Advantage to Borrower: -$2,088 (0.02%)

So, the defeasance option in this case is more expensive to the borrower by0.02% or $2,088.

Transforming Real Estate Finance

Call Protection

94

chapter 6

Page 97: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Scenario 2In five years, rates have risen just above the borrowers mortgage coupon, but theborrower 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: $9,189,718• Yield Maintenance Penalty (min 1%): $91,897 (1.00%)• Cost of Treasury Strips: $9,172,606• Total Defeasance Cost: ($17,112) (0.19%)• Defeasance Advantage to Borrower: $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: $9,189,718• Yield Maintenance Penalty (min 1%): $91,897 (1.00%)• Cost of Treasury Strips: $8,741,211• Total Defeasance Cost: $(448,507) (4.88%)• Defeasance Advantage to Borrower: $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 versusthe amount paid under yield maintenance.

Our example assumes a fairly flat yield curve. All else being equal, the defea-sance option gets more expensive to the borrower as the yield curve steepens.Under the defeasance option, each mortgage loan payment is discounted at itscorresponding zero coupon Treasury rate. When there is a larger spread between1-year rates and 10-year rates, the discounted present value of the Treasury strippayments is higher, resulting in a greater cost to the borrower. Even in a steepyield curve environment, however, the defeasance option still allows the borrowerthe opportunity to prepay the loan at a discount. This option is not availableunder yield maintenance since the best a borrower can do is prepay at par.

Please refer to important disclosures at the end of this report. 95

Page 98: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

INTRODUCTION

Recently, the securitization of stand-alone large loans has re-emerged as a significantportion 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-borrowerdeals or transactions backed by a small number of commercial mortgages averaging$50 million or more. For single asset transactions, the loan size may be as much as$500 million. Stand-alone large loans tend to have lower leverage and more credit-worthy borrowers than conduit loans.

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 lowerinvestment 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 pricedin the capital markets. Our main findings are:

• Stand-alone large loans generally have much lower LTVs and higherdebt service coverage ratios than the average conduit loan.

• Rating agencies are wary of the concentration risk inherent in single-asset and single-borrower transactions and therefore often assign highercredit 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 timeshigher on large loans than on conduit loans to equate AAA option-adjusted spreads on large loan and conduit transactions.

• For BBBs and BBs the conclusions are less clear and are highly sensitiveto assumptions about volatility of property NOI.

WHY HAVE STAND-ALONE LARGE LOANS RE-EMERGED?

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 thatwas dominated by large pools of small- to medium-sized loans. A typical poolmight consist of 200 loans on properties well diversified by geographic regionand 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 thatthe market labels as “fusion.” The definition of “fusion” varies, but for purposes ofclassification, Commercial Mortgage Alert defines a fusion deal as a transaction thatcontains at least one loan of $25 million or more. Fusion deals accounted for about12% of CMBS in 1998 and 1999. Some of the large dollar balance loans in fusiondeals would qualify as stand-alone large loans and others resemble conduit loans.

Transforming Real Estate Finance

Large Loans

96

chapter 7

Page 99: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Recently, the number of single-borrower large loan deals has increased. We attributesome of this to the reluctance of many originators to aggregate portfolios of loansover a long period of time. This unwillingness stems from losses on loans waitingfor securitization suffered by many originators during the spread widening of1998. For a single asset, the period from origination to securitization can be halfthat for a diversified pool of loans. According to Commercial Mortgage Alert, single asset transactions constituted nearly 20% of all CMBS issued in 1999, upfrom 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 loanmade up a greater percentage of the transaction. In 1998, several CMBS transac-tions exceeded $2 billion. For transactions that large, a $50 million loan is only2.5% of the total balance and the deal does not receive a high concentrationpenalty from the rating agencies. As deal sizes dropped below $1 billion, issuersreceived better prices by issuing separate large loan deals rather than tainting anentire pool with concentration risk.

Please refer to important disclosures at the end of this report. 97

Source: Morgan Stanley, Commercial Mortgage AlertSource: REIS, CB Commercial

SHARES OF CMBS MARKET BY DEAL TYPE

exhibit 1

3.922

0.410

1.544

1.502

2.616

1.703

4.984

4.517

1.924

1.249

3.041

8.757

12.331

7.765

0

3

6

9

12

15

1994 1995 1996 1997 1998 1999

4.322

3.046

4.319

11.425

20.096

13.048

Single BorrowerLarge LoanFusion

($Bn)

Source: Morgan Stanley, Intex, Commercial Mortgage Alert

PROPERTY TYPE

exhibit 2

26.7

13.9

40.1

2.73.5

13.1

Office

Multifamily

Retail

Industrial

Other

Hotel

Page 100: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Also, a new large loan structure emerged toward the end of last year. In the newstructure, lenders split large loans into two separate notes. The senior portion ofthe loan (the A Note) is sold into a CMBS trust while the junior portion (the BNote) is placed privately outside of the trust. The market coined the term “A-Bstructure” to describe this splitting of loan interests. The A-B structure reducessome of the concentration risk of large loans in a diversified conduit deal, butmore importantly, the portion of the large loan that is sold as CMBS has very lowleverage. The reduced pool leverage results in lower subordination levels fromthe rating agencies.

Finally, because of weakness in the REIT capital markets, REITs have turned toalternative funding sources. REITs had been the primary source of funding forlarge properties before last year. With equity and unsecured corporate debtfinancing becoming more difficult, REITs and other owners of high-value proper-ties started turning to the CMBS market for funding. Many of these loans werebridge loans by large REITs awaiting a recovery in the equity markets. The rallyin REIT share prices in the first half of 1999 was short-lived. A more sustainedincrease in share prices could result in a decline in demand from REITs forsecured financing.

CHARACTERIST ICS OF STAND-ALONE LARGE LOANS

CreditTo date, the average stand-alone large loan has had a lower LTV and higherDSCR than the average conduit loan. Although LTVs for stand-alone large loansmay range from 40% to 100%, most fall in the range of 45% to 70%. Conduittransactions 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 forconduits, 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 debtservice coverage ratios for stand-alone large loans is that prices of large propertiesare more volatile than for smaller properties. There is very little confirmation ofthis volatility, aside from anecdotal evidence about the fall in prices of “trophyproperties” in the early 1990s. That higher priced properties have more volatileprices makes some intuitive sense, because there is a much smaller base of buy-ers than for average or low-priced properties. In the residential home market, forexample, there is much evidence showing that high value homes both rise andfall in price much faster than average value homes.

Stand-alone large loans tend to be on properties located in major markets, incontrast to smaller conduit properties, which are more evenly spread out inboth larger and smaller metropolitan areas. Location in larger markets has bothadvantages and disadvantages. On the positive side, property markets in largemarkets are probably more liquid and have more potential buyers than thosein small markets. On the other hand, the concentration of properties in largeurban centers, for example, can create volatility in prices should the marketsuffer a downturn.

Transforming Real Estate Finance

Large Loans

98

chapter 7

Page 101: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Structural FeaturesLarge commercial loans generally have structural protections that are often not a feature of conduit loans. First, stand-alone large loans have “lock box” featuresthat separate the cash flow of the property from the borrower’s income. Second,the loans are usually placed in a “special purpose entity,” or SPE, which separatesthe loan from the other assets of the borrower in the case of bankruptcy. (In aconduit 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 borrowerproblems, 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 economicdownturn, the replacement of management probably will not improve the credit risk of the property.

Credit Strength of BorrowerOffsetting 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-capitalizedfirms that have access to other sources of capital. This provides some protectionagainst default in a real estate downturn. Given the lower LTVs on most stand-alone large loans, the borrower’s equity stake is typically larger than for conduitsand may in some cases reach over $100 million.

According to a Moody’s study of corporate defaults between 1970 and 1997, lessthan 5% of investment grade corporations defaulted within 10 years. The defaultrate is 20% for companies in the Ba category, and almost 50% for firms rated inthe single-B category. It should be noted, however, that the default of a borrowerdoes 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 asthe value of the property is greater than the mortgage, or the cash flow greaterthan the debt payments, the borrower is unlikely to default on the mortgage.

Prepayment ProtectionStand-alone large loans and conduit loans have very similar prepayment protection.Most currently have defeasance provisions. Earlier transactions had lockout or yieldmaintenance periods followed by penalty points. For more details on types of callprotections, see the Morgan Stanley research publication, “Call Protection in CMBS,”May 1998.

Information AvailabilityThe best property-level information on a CMBS transaction is available at the timeof issuance. For a stand-alone large loan, this data includes audited financials anddetailed lease information. For smaller conduit loans, data may be unaudited andlease information in many cases is either non-existent or less comprehensive.

Please refer to important disclosures at the end of this report. 99

Page 102: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Although the situation has been improving, it is often difficult to obtain individualproperty cash flow and DSCR data after issuance. Some market participants haveargued that data on stand-alone large loans will be more readily available than onstandard conduit loans because of requirements in the loan securitization docu-ments. Since most of the deals with stricter reporting requirements have not beenoutstanding for very long, it remains to be seen whether property cash flow infor-mation 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 ona 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 byEsaki, L’Heureux, and Snyderman (1999)1 found that default rates for small loansat insurance companies over the period 1973 to 1997 were slightly less than forlarge loans. The study, however, only had four categories of loans, of which thelargest size was “over $8 million.”

Another study published in April 1999 by Ziering and McIntosh2 examined therisk-return profiles of properties by size. The study also looked at properties infour categories, but the categories ranged from “less than $20 million” to “over$100 million.” Over the period 1981 through 1998, the authors found that largerproperties generally had higher average returns, but also higher volatility ofreturns. The authors cite the thin market for trophy properties as the reason forhigher volatility.

RATING AGENCY APPROACH TO LARGE LOANS

In rating single-borrower or single-asset transactions, the largest concern of ratingagencies is concentration risk. For example, Duff and Phelps states, “The distin-guishing characteristic of [single-asset] transactions is the concentration ofrisk…This is in contrast to pooled transactions, in which the diversification ofproperties and leases mitigates this risk.”3

To adjust for concentration risk, rating agencies have very conservative standardsfor single-borrower and single-asset transactions. For example, to obtain a AAArating, a large office property must have a debt service coverage ratio in therange 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 isdemonstrated in the level of credit support needed to obtain a higher rating levelthan the collateral on a stand-alone basis. For example, consider a pool of smallmultifamily properties, all with DSCRs of 1.35x and LTVs of 70%. Such a poolwould need 20% to 25% of subordination for the AAA class. A large loan withequivalent characteristics would be shadow-rated BB. In order for a senior por-tion of this loan to attain a rating of AAA, we estimate that a subordination levelof 35% is required. This subordination level would lower the effective LTV of the

Transforming Real Estate Finance

Large Loans

100

chapter 7

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

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

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

Page 103: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

senior tranche of the large loan to 46%, which is the standard for a stand-aloneAAA. The 10% to 15% difference in subordination levels reflects the “penalty” for concentration.

In underwriting conduit transactions, rating agencies typically underwrite a sam-ple of loans in the pool. The sample usually is about 30%, and includes most ofthe largest loans. In contrast, a large loan or single-asset transaction will have100% underwriting by the agencies.

In stand-alone large loan deals, full environmental and engineering reports areprovided for each property. This is not the case for all conduit loans. It is notclear if the rating agencies adjust conduit subordination levels upward to adjustfor the uncertainty of less than 100% underwriting.

Based on their underwriting, rating agencies will adjust the cash flow and valua-tion on a property, deriving “adjusted” LTVs and DSCRs. In almost all cases, theLTV is adjusted upward (and NOI downward) from the reported number. Basedon our conversations with the rating agencies, these adjustments are usually inthe range of 10% to 20% for both stand-alone large loans and conduit loans.

DO RATINGS CORRECTLY ADJUST FOR RISK?

Rating agencies universally state that a AAA rating means the same amount ofcredit risk across all types of fixed-income securities. Theoretically, a AAA-ratedcorporate 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 col-lateral quality so that a AAA conduit transaction has the same amount of risk as aAAA stand-alone large loan deal.

In practice, we believe that there can be substantial differences in risk amongsecurities with the same rating. We have long maintained that rating agencieshave more conservative standards in rating structured securities than corporatebonds, 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-adjustedspreads,” 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 donot have the real estate expertise to evaluate commercial mortgage credit risk ona single property. Some of these investors rely on the rating agencies to analyzedefault risk; others become comfortable with the general underwriting guidelinesof an originator. Investors preferring pool diversity tend to be money managerswith 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.

Please refer to important disclosures at the end of this report. 101

Page 104: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Other investors prefer single-asset transactions, or those backed by only a fewloans, because they can then use their real estate expertise to underwrite eachloan in a CMBS pool. These investors tend to be insurance companies with largeunderwriting staffs. These investors focus more on the real estate aspects of CMBSinvestments and often purchase stand-alone large loan CMBS as a compliment totheir whole loan portfolio.

Because of this split in investor preferences, stand-alone large loan transactionstend to trade differently than diversified conduit pools. Money managers tend to shy away from single asset deals because of the lack of diversification and aperceived lack in liquidity. Insurance companies re-underwrite a stand-alone largeloan, and if they are comfortable with the asset, prefer to buy the lower investmentgrade classes. The yield on the BBB class, for example, most closely matches theyield on whole loans, the alternative investment for insurers.

Insurance companies purchase about 85% of the mezzanine classes from stand-alone large loan transactions and only 45% from conduit deals. In contrast, moneymanagers buy 50% of the AAA bonds from conduit deals, but only about a thirdof the AAA-rated securities from stand-alone large loan transactions.

As a consequence of these preferences, the AAA class from a stand-alone largeloan transaction tends to trade at a wider spread to Treasuries than a conduitdeal. The difference has historically been in the range of 5 bp to 10 bp. The BBBclass 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 tighterspread than in a diversified pool. If insurers deem an asset as high quality, the smallsize of the BBB class often results in excess demand at the initial pricing level.

Traditional conduit CMBS investors seem willing to accept high loan concentrationsin deals with an A-B structure and the B notes appeal to insurance companies. Webelieve that we will see increased use of the A-B structure in future securitizations.

Transforming Real Estate Finance

Large Loans

102

chapter 7

Source: Morgan Stanley

15%

10%

50%

25%

Insurance

Money Manager

Bank

Pension

Conduit

Investor Type

INVESTOR BASE: CONDUITS VS. LARGE LOANS

exhibit 3

20%

10%

35%

35%Insurance

Money Manager

Bank

Pension

Investor Type

Large Loan

Page 105: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

In the next section, we evaluate whether the additional levels of subordination on stand-alone large loan transactions are enough to compensate investors forconcentration risk and potential higher volatility of cash flows on these properties.

OPTION-ADJUSTED SPREAD ANALYSIS

We used option adjusted spread (OAS) methodology to compare representativesingle borrower and large loan transactions to a typical diversified conduit deal.We found that, under reasonable assumptions, AAAs from large loans have goodrelative value to conduit AAA CMBS. For BBBs and BBs the conclusions are lessclear and are highly sensitive to assumptions about volatility of property NOI.

OAS models are used to value the default and prepayment options imbedded incredit sensitive callable securities. CMBS bonds are typically quoted as nominalspreads to the relevant US Treasury rate. Unlike the nominal spread, an OASincorporates variations in cash flows from credit and prepayment events. Thespread to the risk free rate that equates these option adjusted cashflows to themarket 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 rule-based methodology for both defaults and prepayments. Two conditions must bemet for a borrower to default: net operating income (NOI) is less than the scheduledmortgage payment for at least six months and the current loan to value (LTV) isgreater than 110%. With respect to prepayments, there are minimum debt servicecoverage ratios and maximum loan to value ratios based on property type thatthe cash flow and property value must meet for a borrower to prepay.Additionally, the benefit of refinancing must be greater than the costs, which areinclusive of penalty points and/or yield maintenance and transactions costs. Incomparing 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 areultimately dependent on the projected levels of NOI. Future property values arecalculated by dividing the projected NOI by a capitalization rate. The initial capitalization rate varies through time based on an assumed correlation betweeninterest rates and capitalization rates. We used Morgan Stanley proprietary modelsto generate future values for interest rates.

The three deals that we analyzed differ in terms of leverage, loan diversity andcredit enhancement to different classes of bonds. The debt service coverage ratiois the highest for the large loan deal and lowest for the conduit while the LTVratio is lowest for the single asset deal and highest for the conduit deal. At theAAA level, credit enhancement is the highest for the single asset deal and lowestfor the conduit deal. At the BBB level, however, the conduit deal has the highestcredit support and the single asset deal the lowest. Exhibit 6 provides statistics on each of the different deals.

Please refer to important disclosures at the end of this report. 103

Page 106: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Exhibit 4 shows cumulative default rates for each of the deals based on a givenlevel of NOI volatility. At the same level of NOI volatility, the single asset deal hasthe lowest level of cumulative defaults and the conduit deal the highest level ofdefaults. 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 secu-rities backed by the different collateral types with different credit support levels.

Exhibit 5 shows the OAS for the 10-year AAA bond from each of the deals atincreasing 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 largeloan deal than for the conduit deal. The resulting option-adjusted spreads showlittle option cost for any of the 10-year AAA bonds even at very high levels ofNOI volatility.

Transforming Real Estate Finance

Large Loans

104

chapter 7

Source: Morgan Stanley

CUMULATIVEDEFAULT RATES

exhibit 4

0%

10%

20%

30%

40%

0 6 8 10 12 14 16 18 20

CumulativeDefault Rate

NOI Volatility

Single AssetLarge LoanConduit Deal

Source: Morgan Stanley

10-YEAR AAA OAS

exhibit 5

120

122

124

126

128

130

132

134

136

0 6 8 10 12 14 16 18 20

OAS (bp)

NOI Volatility

Single AssetLarge LoanConduit Deal

Page 107: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 105

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 inlife insurance companies’ commercial mortgage loan portfolios in the ELS defaultstudy. There is no default option cost for the 10-year AAA bond from the conduitdeal 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.

Single Asset Deal Number of Loans: 1 • DSCR: 1.85X • LTV: 54%Credit

Bonds Coupon Avg Life Rating EnhancementA1 6.16 5.51 AAA 45.0%A2 6.54 9.93 AAA 45.5%B 6.78 9.93 AA 31.5%C 6.88 9.93 A 20.0%D 6.96 9.93 BBB 0.0%

Large Loan Deal Number of Loans: 11 • DSCR: 1.90X • LTV: 58%Credit

Bonds Coupon Avg Life Rating EnhancementA1 6.22 5.39 AAA 27.50%A2 6.48 9.52 AAA 27.50%B 6.70 9.89 AA 21.00%C 6.77 9.96 A 14.00%D 7.15 9.96 BBB 7.75%E 7.15 9.96 BB 4.50%F 7.15 9.96 B 1.50%

Conduit Deal Number of Loans: 219 • DSCR: 1.57X • LTV: 67%Credit

Bonds Coupon Avg Life Rating EnhancementA1 6.34 5.42 AAA 26.00%A2 6.54 9.72 AAA 26.00%B 6.63 9.96 AA 21.00%C 6.77 10.00 A 16.50%D 7.07 11.47 BBB 11.50%E 6.34 15.95 BB 5.25%F 6.34 19.42 B 2.75%

CHARACTERISTICS OF SINGLE-ASSET, LARGE LOAN, AND CONDUIT DEALSexhibit 6

Source: Morgan Stanley

Page 108: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Large Loans

106

chapter 7

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 conduitdeal, reflecting market pricing as of February, 2000. Unlike the 10-year AAA bond allof 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 andlarge loan deal. NOI volatility of 12% for the conduit deal results in an option-adjust-ed spread of 87 bp. To reach an equivalent OAS for the large loan and single assetdeal NOI volatility would have to be between 13%–14% and 15%–16%, respectively.

Source: Morgan Stanley

BBB OAS

exhibit 7

(400)

(300)

(200)

(100)

0

100

200

300

0 6 8 10 12 14 16 18 20

Single AssetLarge LoanConduit Deal

OAS (bp)

NOI Volatility

Source: Morgan Stanley

BBB OAS

exhibit 8

(400)

(200)

0

200

400

600

0 6 8 10 12 14 16 18 20

NOI Volatility

Large Loan

Conduit Deal

OAS (bp)

Page 109: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

At the BB level, the conduit deal has a lower OAS than the large loan deal at lowlevels of NOI volatility and a higher OAS than the large loan deal at higher levelsof NOI volatility. (The single asset deal does not have a BB bond). The differencebetween 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 thelarge 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 NOIvolatility that results in the same OAS ranges from 0% to 1%. Exhibit 8 shows theOAS for the BB bonds as NOI volatility varies.

CONCLUSION: RELATIVE VALUE

Interpreting the OAS results calculated from our model is difficult in the absenceof historical benchmarks for NOI and price volatility. However, it would appearthat 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 andthe 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 higherlevels of NOI volatility than the conduit deal and still have the same OAS. ForBBBs, 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.

Please refer to important disclosures at the end of this report. 107

Page 110: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

FANNIE MAE

In addition to “private-label” CMBS, the CMBS market also encompasses multi-family agency securities, issued by Fannie Mae, Ginnie Mae and Freddie Mac. The single largest issuer of multifamily MBS is Fannie Mae.

The main Fannie Mae program is Delegated Underwriting and Servicing (DUS).

THE FANNIE MAE DUS PROGRAM

Fannie Mae created the DUS program in 1988 to streamline the underwritingprocess and help fulfill its commitment to multifamily housing. Under the program,specially approved lenders may underwrite, close, service and sell mortgages toFannie Mae without prior review by Fannie Mae. DUS lenders benefit from thisspecial relationship because they have greater autonomy in underwriting andservicing and can also be more competitive given that DUS loan rates are lowerthan 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.

CHARACTERIST ICS OF DUS LOANS

Loans originated under the DUS program are generally either fixed-rate balloonmortgages with 5-, 7-, 10-, or 15-year terms or fixed-rate fully amortizing loanswith 25- or 30-year terms. Variations, such as 20-year fully amortizing loans, arealso 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 musthave been originated within 6 months of Fannie Mae’s purchase.

PREPAYMENT PROTECTION

One of the main advantages of multifamily securities over residential MBS is theprepayment protection on multifamily loans. Most DUS loans have yield mainte-nance premiums in the event of an early prepayment. The premium is usuallyyield maintenance calculated at the relevant treasury rate, or “Treasuries flat.”Common yield maintenance terms are:

• Balloon Term (years) Yield Maintenance Term (years)5 3 or 4.57 5 or 6.510 7 or 9.515 1030 10

Transforming Real Estate Finance

Multifamily MBS

108

chapter 8

Page 111: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

After the yield maintenance period ends, the borrower is still required to pay a1% premium on prepayment which is retained by Fannie Mae. This premium iswaived during the last 90 days of the loan term to facilitate refinancing.Curtailments are not allowed and, consequently, the borrower is faced with thechoice of either prepaying the entire loan balance or not at all.

Prepayment fees are passed through to the investor by Fannie Mae only to theextent they are received from the borrower. Fannie Mae’s obligation extends onlyto the outstanding principal balance of the security, i.e., if an MBS DUS defaultsas a premium security, the investor receives a minimum of par but may lose someor all of the premium.

Most DUS loans can be assumed by a new, and creditworthy, borrower on pay-ment of a 1% assumption fee that is not passed through to the investor. Giventhat the pricing speed assumption of DUS is usually 0% CPR, the assumabilityoption does not add any negative convexity to the security.

The prepayment fee actually due from the borrower is calculated by substitutingthe note rate for the coupon in the above calculation. The difference between thefee received and the fee paid to the investor is shared by FNMA and the lender.

UNDERWRIT ING

DUS lenders have strong incentives to underwrite high quality loans. First andforemost, 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 solelyby the lender and losses in excess of 5% are shared by Fannie Mae and thelender according to a formula. The lender’s share of the loss is limited to 20-40%of the UPB. The yield maintenance premium is also part of the loss computationgiving the lender a vested interest in enforcing payment of the premium.

For pricing and underwriting purposes, Fannie Mae categorizes DUS loans intoone 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 extremelyrare. Fannie Mae may also designate loans with a ‘+’ in each category, based onsubjective criteria such as property location and management. A ‘+’ reduces theguarantee fee by about 10 bp.

Please refer to important disclosures at the end of this report. 109

DUS UNDERWRITING TIERSexhibit 1

Minimum Debt Maximum Service Coverage Ratio Loan-to-Value Ratio

Tier 1 1.15 80%Tier 2 1.25 80%Tier 3 1.35 65%Tier 4 1.55 55%

Note: Most DUS loans are in Tier 2 or Tier 3. Source: Fannie Mae

Page 112: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

FANNIE MAE DMBS

In 1996, Fannie Mae began issuing discount MBS (DMBS) as a means of sellingmultifamily loans in the secondary mortgage market. Fannie Mae started routinelysecuritizing multifamily loans on a programmatic basis in 1994, but these securi-ties did not possess the characteristics of DMBS.

DMBS CHARACTERIST ICS

DMBS are Fannie Mae’s only short-term, non-interest bearing securities that arecollateralized by mortgages. Since DMBS are non-interest bearing securities, theyare sold to investors at a discount and repaid at par upon maturity.

DMBS maturities generally range between three and nine months, with occasional exceptions.

To date, almost all DMBS issuance has consisted of three-month securities. SinceDMBS 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 in2000. Despite low interest rates, many borrowers chose to maintain financial flexi-bility by using short term financing. For 2002, Fannie Mae projects DMBS issuanceto be close to 2001 levels. As of February 1st, 2002, Fannie Mae issued $681 mil-lion of DMBS.

FANNIE MAE GUARANTEE

DMBS, like all other Fannie Mae securities, carry Fannie Mae’s guarantee of fulland timely payment of principal. DMBS are not rated by the rating agencies butare equivalent to agency credits. In the event of a principal shortfall, payments onDMBS are pari passu with other senior debt of Fannie Mae.

Transforming Real Estate Finance

Multifamily MBS

110

chapter 8

Notes:1As of February 1, 2002.Issuance volumes prior to 2000 only represent DMBS with 3-month maturities.Source: Fannie Mae

DMBS ISSUANCE

exhibit 2

45 58

2,215

6,234

8,627

681143

0

2,000

4,000

6,000

8,000

10,000

1996 1997 1998 1999 2000 2001 2002

($mm)

1

Page 113: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

LOAN CHARACTERIST ICS

While single-asset executions are available, DMBS are most often secured by a pool of cross-collateralized, cross-defaulted, first-lien mortgages on income-pro-ducing residential properties with at least 5 units. These multifamily mortgagesare typically underwritten to conform to the Fannie Mae Delegated Underwritingand Servicing (DUS) requirements.

DMBS loans tend to be more conservative than multifamily loans in conduittransactions, in terms of loan-to-value (LTV) ratios and debt service coverageratios (DSCR). LTVs on DMBS collateral range between 50% and 75%, whileDSCRs typically range between 1.65X and 1.30X. In typical CMBS conduit trans-actions, 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 providesborrowers with short-term advances that are funded by the sale of DMBS.Borrowers receive proceeds from DMBS issuance, which are determined bymarket 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 short-term advance of $100 million, with $200 million of multifamily properties as col-lateral. 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.

Please refer to important disclosures at the end of this report. 111

1Assumes $99.5MM discounted proceeds from sale of DMBS Source: Fannie Mae

MULTIFAMILYCREDIT FACILITY

exhibit 3

Borrower

Investors

Lender Fannie Mae

$99.5MM1 $100MM DMBS

Mortgages on Assets ($200MM value)

Mortgages on Assets ($200MM value)

$99.5MM1

$100MM DMBS

ALL VARIABLE RATEAssume: $100MM 50% LTV Facility (50% LTV); $100MM DMBS

Page 114: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

At maturity, the investor will receive $100 million, via proceeds from new DMBSissuance, or a payoff by the borrower.

PREPAYMENTS AND DEFAULTS

DMBS are locked out from prepayments for their entire term, and thereforehave no prepayment risk. Default risk is also non-existent to DMBS investors, as Fannie Mae guarantees payments of principal when due. To date, there havebeen no defaults, but in the case of default, Fannie Mae would pay DMBSinvestors par at maturity.

TRADING LEVELS

Discounts on new issue three-month DMBS may be converted to annualizedyields or spreads to LIBOR. Historically, three-month DMBS have yielded LIBORless 15-16 bp. During the first few weeks of 2002, three-month DMBS have beenissued with average yields of LIBOR – 6 bp.

In the secondary market, Morgan Stanley has traded nearly $5 billion of DMBS in2001, and $700 million in 2002 YTD. We estimate that Morgan Stanley is involvedin 20-25% of DMBS trades in the secondary market. DMBS are currently trading 3-5 bp wider than agency discount notes, largely because of liquidity, and providean attractive investment for money market buyers.

INVESTOR BASE

DMBS are purchased primarily by money market investors who view these securi-ties as an attractive alternative to Treasury Bills.

Fannie Mae DMBS are similar to Treasurys, in that they are permitted invest-ments for Federally supervised institutions and for trusts and funds investedunder the authority of the US. DMBS can also be purchased in unlimitedamounts by national banks, federally chartered credit unions and federal savingsand loans associations.

Transforming Real Estate Finance

Multifamily MBS

112

chapter 8

Source: Fannie Mae

DMBS TRADES:THREE-MONTH

MATURITY

exhibit 4

-0.65%

-0.55%

-0.45%

-0.35%

-0.25%

-0.15%

-0.05%

0.05%

0.15%

Apr-99 Dec-99 Jun-00 Nov-00 Mar-01 Jul-01 Nov-01 Jan-02

Issue Dates

Imputed Pass-through vs. LIBOR Base line representscomparable LIBOR as ofsecurity trade date.

Page 115: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

GINNIE MAE/FHA

Within the agency multifamily market, the second largest issuer in the agencymarket is the Government National Mortgage Association (GNMA). Project loansmay be made under a number of Housing and Urban Development Department(HUD) programs, including:

• 221(d) 4: Construction or permanent financing• 223(f): Refinancing• 223(a)7: Accelerated refinancing• 232: Nursing home/assisted living• 241(f): Equity take-out second mortgage

All Ginnie Mae securities are backed by loans originated by the Federal HousingAdministration (FHA) and are either permanent loan certificates (PLCs) or con-struction loan certificates (CLCs).

PLCs are usually 35-year fully amortizing fixed-rate mortgages. Prepayment pro-tection 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 CLCsand are converted to PLCs upon completion of the construction project. CLCstrade at wider spreads than PLCs because of liquidity and uncertainty associatedwith funding a construction loan.

Exhibit 5 shows some of the major differences between Ginnie Mae and FHAproject loans:

Effective April 1, 1997, Ginnie Mae reduced pool processing time from 10 days to5 days and added other features to streamline its multifamily MBS program.1

Please refer to important disclosures at the end of this report. 113

CHARACTERISTICS OF GINNIE MAE AND FHAPROJECT LOANS

exhibit 5

Ginnie Mae FHAGovernment guarantee Explicit Implicit

Principal paymentin case of default 100% 99%

Delay days 44 54Delivery PTC Physical

Data on Bloomberg Yes No

Source: Fannie Mae

1 See inside MBS & ABS, May 1, 1997, p.3

Page 116: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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 itsmultifamily loan production.

PROGRAM PLUS

Freddie Mac’s Program Plus is similar to Fannie Mae’s DUS program. Under the pro-gram, Freddie Mac pre-approves multifamily lenders with “local market expertise.” Since1993, 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

Transforming Real Estate Finance

Multifamily MBS

114

chapter 8

A PROFILE OF GINNIE MAE MULTIFAMILY MBSexhibit 6

Issuer Ginnie Mae approved mortgage lenderIssue Type GNMA IUnderlyingMortgages FHA Insured multifamily mortgagesPool Types 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

SecuritiesInterest Rate Fixed; at .25 to .50 percent below the interest rate of the

underlying mortgage(s)Guaranty Full and timely payment of principal and interest

Guarantor Ginnie Mae (full faith and credit of the United States)Principal

and Interest Paid monthly to securities holdersMaturity Varies, typically 40 years

MinimumCertificate Size $25,000 (may be less for aged securities)Transfer Agent Chase (formerly Chemical Bank)

Source: Reprinted from Ginnie Mae website, www.ginniemae.gov, 1997

Page 117: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 115

Page 118: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

INTRODUCTION

The CMBS interest-only securities (IO) market has both grown and matured sinceits inception in the mid-1990’s. Unfortunately, the derivative taint of the name “IO”has caused some buyers to shy away from the sector. In this chapter we trace thegrowth and development of the CMBS IO market and factors affecting issuance.

We examine current pricing conventions for IOs and look at several case studies toevaluate 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 creditsensitive 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 duringyield maintenance periods.

WHAT IS A CMBS IO?

As in the single-family residential market, CMBS IOs receive a coupon strippedfrom an underlying pool of mortgages or bond classes. Stripping a coupon allowsan issuer to sell par (or near par) priced securities, even if the coupon on theunderlying mortgages is well above the bond coupons. For example, a 1% IOstrip 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 arebased on priority of cashflow rather than credit, and are meaningless except forregulatory purposes. Any loan default affects the yield of an IO, but from the rating agency perspective, the default does not affect the IO’s senior priority inreceiving 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 ofthese bonds may be forced to liquidate if they have minimum rating require-ments. In such an environment, spreads on IOs could widen, but the rating for regulatory purposes would probably not change.

Transforming Real Estate Finance

CMBS IOs

116

chapter 9

Page 119: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

For IOs backed by residential mortgages, prepayments are the most important risk factor. Historically, defaults on residential mortgages have been a relativelyminor factor. Over the life of a residential mortgage pool, defaults have typicallytotaled less than 5% of the original balance compared to prepayments that rangefrom 5% to 30% per year. For commercial IOs, the situation is reversed, withdefaults potentially having a greater impact than prepayments. Most securitizedcommercial mortgages have either a prepayment lockout, yield maintenance ordefeasance. Historic cumulative lifetime commercial mortgage default rates atinsurance 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 theIO’s cash flow is off of the AAA-rated classes, which typically constitute 70% ormore 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 isstripped off of these classes.

INVESTOR BASE Current buyers of CMBS IOs include life insurance companiesand bank portfolios to enhance portfolio yield. In addition, CMBS IOs appeal tomoney managers who seek a high-yielding AAA-rated asset.

IOs also appeal to investors seeking shorter duration CMBS instruments—theduration of a CMBS IO is about 4 years, compared to 7 years for classes ratedBBB or BB.

CMBS SPREADS AND PREPAYMENT PRIC ING ASSUMPTIONS

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

Please refer to important disclosures at the end of this report. 117

Source: Morgan Stanley

TYPICAL IOSTRUCTURE

exhibit 1

Fixed Bond Coupons

AAA Classes AA A BBB BB B/NR

Class X Coupon Strips

WAC-AAA CouponWAC-

AAWAC-

AWAC-BBB

WAC-BB

WAC-B/NR

Page 120: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The market currently prices CMBS IOs assuming that loans with yield mainte-nance prepayment penalties are equivalent to loans with absolute prepaymentlockout provisions. Participants in the CMBS market examine IO yields andspreads using a 100% CPY assumption, meaning that all loans prepay immedi-ately after the lockout or yield maintenance period. Some investors also examineyields under a 0% CPR1 assumption. The difference in spreads using the 0% and100% CPR assumptions is often referred to as the “slope” or drop in spread fromthe slowest to the fastest prepayment speed.

The actual prepayment experience of a pool of commercial loans backing aCMBS can be quite different from the pricing assumptions. Loans can and haveprepaid in the yield maintenance period. Since a share of the penalty is passedon to the IO holder, this can prove beneficial to a CMBS IO investor, dependingon 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. Unlikethe residential mortgage market, there is very little historical data on commercialmortgage prepayments.

Due to the high percentage of loans with defeasance or yield maintenance intoday’s CMBS transactions, the primary investor analysis on CMBS IO’s is defaultdriven. Recoveries and losses on defaulted loans are now the primary factoraffecting early payment of principal in a CMBS transaction.

DEFAULT ASSUMPTIONS

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 thereturn on a CMBS IO. The IO investor no longer receives the interest strip after adefaulted loan is finally liquidated. A default, then, has a similar effect on a CMBS IOinvestor as a prepayment. However, default at a balloon date and subsequent exten-sion 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 defaultassumptions, ranging from 0% CDR to 5% CDR or higher. “CDR” stands for condi-tional default rate and is analogous to CPR for prepayments (see footnote 1).Unlike prepayment assumptions, CDRs ignore any call protection feature on theloans. Given the historical pattern of commercial mortgage default rates, webelieve that the expected average annual default rate on a newly originated poolof commercial mortgages lies between 1% CDR and 4% CDR.

ESAKI , L’HEUREUX, AND SNYDERMAN: A NEW DEFAULT STUDY

Our default assumptions are based on a study recently completed by Esaki,L’Heureux, and Snyderman (ELS) of default rates on commercial mortgages heldby life insurance companies between 1972 and 1997.2 Therefore, the standardstress scenario run on CMBS IO is a 2% CDR. The study showed that over this 26-

Transforming Real Estate Finance

CMBS IOs

118

chapter 9

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

2 See “Commercial Mortgage Defaults: An Update,” Howard Esaki, Steven L’Heureux, and Mark Snyderman,Real Estate Finance, forthcoming.

Page 121: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

year period, lifetime cumulative default rates on commercial mortgage loans origi-nated 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 translateinto annual CDRs of just over 1% and about 4%, respectively, assuming 10-yearcollateral 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 apeak in years 3 to 7 after origination, and then tailing off thereafter. The magni-tude of the peak default rate varies based on the real estate environment after theloans are originated.

IMPACT OF PREPAYMENTS ON IO Y IELD

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

The current pricing convention for CMBS IOs, however, is 100% CPR after yieldmaintenance or lockout. Reading exhibit 2 from right to left shows the potentialupside to an investor if prepayments are slower than the pricing assumption. Inthe five examples cited, an investor has from 14 bp to 140 bp of upside if actualprepayments are 20% CPR rather than 100% CPR.

The greater the slope of the increase in yield as prepayments decline, the greaterthe 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 penaltyperiod 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 agiven transaction.

Please refer to important disclosures at the end of this report. 119

CPR 0 CPR 20 CPR 50 CPR 100DMARC 98-C1 10.63 9.97 9.48 8.59

NASC 98-D6 8.83 8.82 8.80 8.68LBCMT 98-C1 10.36 9.85 9.46 8.59FULBA 98-C2 9.95 9.50 9.19 8.68MSC 98-WF2 9.27 9.18 9.08 8.67

PREPAYMENT RISK:YTM OF CONDUIT IOs IN PERCENTexhibit 2

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

Page 122: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS IOs

120

chapter 9

YIELD MAINTENANCE PENALTIES AND IOs

In most CMBS structures, IOs receive a share of any prepayment penalties paidby borrowers. In early structures (pre-1997), the IO could receive as much as100% of a yield maintenance penalty. More recently, CMBS IOs receive a share of the penalty according to a formula such as:

In this example, if the principal and interest bond coupon were 6.0%, the mort-gage 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, theborrower does not pay a penalty. If a penalty is paid, and the bond coupon isless than the UST, then the IO receives all of the penalty.

For non-defeasance loans, prepayments during the yield maintenance period arebeneficial to the IO investor in a stable rate environment. Exhibit 3 shows that in4 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 CMBSbacked by loans with defeasance.

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 aninstantaneous 300 bp increase in interest rates during the yield maintenance period. Under this scenario, the drop in yield from 0% CPR to 100% CPR rangesfrom 15 bp to more than 1500 bp.

P = % of penalty paid to IO =

(Bond Coupon – UST)(Mortgage Coupon – UST)1 –

CPR 0 CPR 20 CPR 50 CPR 100DMARC 98-C1 10.63 13.51 15.42 17.66

NASC 98-D6 8.83 8.82 8.80 8.68LBCMT 98-C1 10.36 11.44 12.48 14.31FULBA 98-C2 9.95 10.72 11.37 12.10MSC 98-WF2 9.26 10.85 11.63 13.48

PREPAYMENT RISK: YTM OF CONDUIT IOs:INTEREST RATES UNCHANGED (IN %)exhibit 3

Priced at +350 at 100 CPR; yield maintenance equals lockoutSource: Morgan Stanley

Page 123: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 121

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

The most notable feature of exhibits 3 and 4 is the stable profile of CMBS IOsbacked by defeasance loans, as demonstrated by NASC 98-D6. The loans in thistransaction must be defeased upon prepayment by substituting Treasury securitiesfor the mortgage in an amount such that the cashflow from the Treasuries match-es that of the prepaid loan. IOs from defeased transactions present a more stableyield profile in rising rate/high prepayment environments, but less upside in lowprepayment scenarios or from penalty windfalls.

PAC & COMPANION IO ’S

The creation of two separate IO strips from one CMBS transaction is a morerecent structural nuance designed to accommodate the creation of larger IO’swhile appealing to alternative IO buyers.3 The rapidly declining interest ratesresulting from a recessionary economy in early 2001 resulted in the creation ofmuch larger IO’s, since the IO proceeds are directly proportional to the differencein the weighted average coupon (WAC) of the underlying mortgages vs. the WACof the bonds created. Ten year treasuries rallied over 100 basis points fromOctober 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 billiontransaction. If 10 year Treasuries rallied just 25 basis points after the mortgage poolwas set, but prior to pricing the bonds, the mortgage pool WAC would be 65 basispoints greater than the bond WAC, resulting in approximately $43 MM in IO pro-ceeds. This incremental increase in IO proceeds resulted in the development of thePAC IO / Companion IO structure, appealing to an investor base of risk adverse IObuyers drawn to the very stable yield profile of the PAC IO, with the traditional IOinvestor base buying the higher yielding but less stable Companion IO.

CPR 0 CPR 20 CPR 50 CPR 100DMARC 98-C1 10.63 6.89 3.32 -1.03

NASC 98-D6 8.83 8.82 8.80 8.68LBCMT 98-C1 10.36 6.13 1.48 -5.15FULBA 98-C2 9.95 8.20 6.96 5.51MSC 98-WF2 9.26 5.22 1.41 -2.31

PREPAYMENT RISK: YTM OF RECENT CONDUIT IOs:INTEREST RATES UP 300 BP (IN %)exhibit 4

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 changeSource: Morgan Stanley

3 Some CMBS transactions in the early & mid 1990’s had multiple IO classes, with individual IO’s stripped offof individual bond classes.

Page 124: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The traditional single class IO was stripped off of all (or most all) bond classes inthe CMBS structure. The PAC IO is only stripped off of mezzanine bond classesfor a finite time; subsequently a Companion IO is stripped off of the more seniorand subordinate classes (See Exhibit 5). Recall, given the underlying call protec-tion of today’s commercial mortgages, the performance of any IO is primarily afunction of credit. If a mortgage pool is comprised of 100% defeased collateral,the only unscheduled payment of principal is via default and recovery. Becauserecoveries paydown the A1 class first, and losses are applied to the most subordi-nate bond class, the yield profile on the PAC IO is insulated from defaults in theunderlying mortgage pool. Subsequently, the yield on the Companion IO hasmarginally more exposure to defaults in the mortgage pool. In that regard, theCompanion IO is also referred to as the Levered IO as its yield profile has moreexposure to defaults in the underlying mortgage pool.

In recent transactions, the PAC IO has priced at T+145. This is a AAA rated securi-ty that can survive a 6 CDR stress scenario before giving up any yield. This stressis 3 times the standard commercial mortgage default rate per the ELS Study. TheLevered IO is currently priced at T+475 and is also rated AAA. This levered IOcan 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 theLevered IO gives yield in a stressed environment, the investor is still buying a lossadjusted AAA security at T+215.

Transforming Real Estate Finance

CMBS IOs

122

chapter 9

SERIES 2001 - TOP 4 CAPITAL STRUCTUREexhibit 5

Aa a/AAA5.06%

Aa a/AAA1-mo. LIBOR + 0 .45%

Aa a/AAA5.61%

Aa 2/AA5.80%

A2 /A6.05%

A3 /A-6.17%

Ba a2/BB B6.47%

Ba 1/BB+ to NR6.00%

Class A- 1

Class A- 2

Class A- 3

Class B

Class C

Class D

Class E

Class F

Class G -N

Class X- 1, X -2

$389.2M M

$60.0M M

$320.2M M

$24.8M M

$24.8M M

$20.3M M

$9.0M M

$45.1M M

NR = N ot rated

Ba a3/BB B-6.91% $9.0M M

C lass X -1 C lass X -2 through N ov. 2008 C lass X -1 after N ov. 2008

Page 125: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 123

Page 126: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

INTRODUCTION

We think that the best method to quantify the credit risks of a real estate portfoliois to analyze the historical performance of pools of commercial mortgage loansover their lifetime. Mark P. Snyderman has completed two pioneering studies oflifetime commercial mortgage performance. These studies tracked the lifetimedefaults of commercial mortgages held by life insurance companies.1 His mostrecent article tracked defaults on 8 large insurance companies through 1991. Thecompanies originated the loans between 1972 and 1986.

Using the same insurance companies and data sources as the original studies, weupdated the default data through 1997 for originations from 1972 to 1992. Weexamined the credit performance of over 15,000 individual loans, an increase ofabout 4,000 loans from the 1994 study. Preliminary results of an extension of thestudy through 2000 show no change in the major conclusions of this study. Themain results of the updated study are:

• The average cumulative default rate for loans originated in 1987 orearlier (that is, with at least 10 years of seasoning) was 18.1% of theoriginal pool balance. This average is slightly higher than in previousstudies, but in line with earlier projections.

• The cumulative default rate of 28% on loans originated in 1986 wasthe 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%, slightlyhigher than the 36% reported in the 1991 study. The severity on loansliquidated between 1992 and 1997 was 44%.

• Annual defaults rose to about 2% in years 3 through 7 after origina-tion, and then fell off for the remaining life. Defaults did not increasein balloon years.

• About half of the defaulting loans were liquidated and half restruc-tured. The number of liquidated loans rose to about 60% in the 1992-1997 period.

• Most investment grade CMBS, as typically structured today, do not suf-fer a principal loss when subjected to the default rates of even theworst origination cohort since 1972.

• The study encompasses a period covering one of the worst real estatedepressions in the past century.

1 “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.

Transforming Real Estate Finance

Commercial Mortgage DefaultsThis chapter was co-authored by Mark Snyderman and Steven L’Heureux and appeared in a slightly different form in Real Estate Finance, Spring 1999.

124

chapter 10

Page 127: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 125

NUMBER OF LOANS BY ORIGINATOR AND YEAR

We tracked commercial mortgage default rates from life insurance company annual statements from 1992 to 1997, aggregating our data with the two previousSnyderman studies.2

The insurance companies and number of loans tracked are shown in exhibit 1.

Loan originations in the study peaked in 1978-1979 and 1985-1988, and reachedlows in 1982 and 1992. Of the 15,109 total loans originated, 2663 (17.6%), haddefaulted by 1997.3

NUMBER OF LOANS BY ORIGINATOR, 1972-1992exhibit 1

Originator Number of LoansAetna Life Insurance Company 2,974Connecticut Mutual Life Insurance Company1 863Equitable Life Insurance Company 1,768John Hancock Mutual Life Insurance Company 1,714New England Mutual Life Insurance Company* 1,338The Northwestern Mutual Life Insurance Company 719The Prudential Insurance Company of America 3,522The Travelers Insurance Company 2,211Total 15,109

1Merged into other firms since 1991. *Loans were tracked at the new firm.Source: Morgan Stanley

Source: Morgan Stanley

NUMBER OFLOANS BY

ORIGINATION YEAR

exhibit 2

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 920

200

400

600

800

1,000

1,200

1,400

2 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).

3 While insurance company loans are very diverse geographically and by property type, they may differ in somerespects from the entire universe of commercial mortgages. For example, in the period we studied, insurancecompany 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).

Page 128: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Commercial Mortgage Defaults

126

chapter 10

SIZE OF LOANS

The median loan size was about $4million. About 75% of the loans wereless than $8 million. The $4 millionto $8 million loan category had thehighest default rate, followed closelyby the greater than $8 million cate-gory. The smallest loans had thelowest default rate.

GEOGRAPHIC D ISTRIBUTION

The loans in the study are well diversified geographically, with thelargest percentages in the West(24%), Northeast (22%), and SouthCentral (19%).

Within a given origination cohort,cumulative lifetime defaults have var-ied widely by geographic region. Forexample in the 1980 cohort, loansoriginated in the Northeast had a life-time default rate of 6.0%, while loanson properties in the South Centralregion had a cumulative default rateof 31.0%. For the period 1972-1992,the South Central region had thehighest average cumulative defaultrate, while Canada and the WestCoast had the lowest average defaultrates. The higher default rates in theSouth Central region reflect the deepoil-state recession of the 1980s.

Source: Morgan Stanley

LOANS ORIGINATEDAND DEFAULT RATESBY PRINCIPAL AMOUNT

exhibit 3

TotalLoan Amount Loans Default(millions of $) Originated Rate(%)

0 – 2 3,842 13.52 – 4 3,949 17.44 – 8 3,459 20.1

> 8 3,859 19.8Total 15,109 18.1

Source: Morgan Stanley

NUMBER OF LOANSORIGINATED BYGEOGRAPHIC REGION

exhibit 4

Number Percentof Loans of Total

West Coast 3,679 24.3South Central 2,928 19.4

Northeast 3,295 21.8Mid-Central 2,113 14.0Southeast 2,309 15.3

Canada/Other 785 0.5Total 15,109 100.0

Source: Morgan Stanley

AVERAGE DEFAULTRATE BY GEOGRAPHICREGION

exhibit 5

Default RateWest Coast 12.3

South Central 27.4Northeast 15.1

Mid-Central 17.3Southeast 19.7

Canada/Other 11.7

Page 129: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 127

Source: Morgan Stanley

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 920

5

10

15

20

25

30

LIFETIME DEFAULTRATES BY

ORIGINATIONCOHORT

exhibit 6

Source: Morgan Stanley

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 240.0

0.5

1.0

1.5

2.0

2.5

Year Since Origination

AVERAGE TIMINGOF DEFAULTS (AS

A PERCENT OFORIGINAL

BALANCE)

exhibit 7

Page 130: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

DEFAULTS BY ORIGINATION COHORT

Cumulative lifetime default rates for cohorts with at least ten years of historyranged from 9.2% for 1977 originations to 27.6% for 1986 originations. The aver-age 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 1997period was 37.7%, slightly higher than the 36% severity reported in the 1991Snyderman 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 subse-quently sold in the period 1992 through 1997 was 43.8%. The higher severity inthis period reflects the steep downturn in real estate prices in the early 1990s. Forindividual insurance companies, the weighted average severity rates ranged from26% to 72% in the 1992-1997 period.

For the period 1992-1997, about 60% the loans entering foreclosure were liquidat-ed, compared to 46% in the 1972-1991 period. There is no data available on thenon-liquidated loans, which presumably were restructured by the lender.

TIMING OF DEFAULTS

Averaged over all cohorts, commercial mortgage defaults rise over the first 3years after origination to about 2% of original balances, remain near 2% for years3 through 7, and then fall off over the remaining life of the cohort. Although wedo not have data on the amortization schedules of the loans in the study, theredoes not appear to be any increase in default rates at balloon dates.

The default rate curve shown in exhibit 7 masks individual cohort patterns, whichcan vary substantially from the average. Default timing for individual originationcohorts varies depending on the state of the commercial real estate market at thetime of origination and the subsequent years. For example, the 1973 cohort hadthe highest default rates in year 3 after origination (nearly 5%), as this cohort wasoriginated near the peak of a real estate cycle followed by a sharp downturn inreal estate prices. Not a single cohort, however, had a peak in default rates inyears 7, 10, or 15, which are typically balloon payment dates. (See the Appendixfor yearly default data for each origination cohort.) Many of the loans originated inthe 1970s and 1980s, however, were fully amortizing and not balloon mortgages.

Transforming Real Estate Finance

Commercial Mortgage Defaults

128

chapter 10

Page 131: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

COMPARISON TO PREVIOUS STUDIES

The new study extends from 1991 to 1997 the two previous commercial mortgagedefault studies by Snyderman. These studies covered 7,205 loans and 10,955loans, respectively, while the current study increases the coverage to over 15,000loans. The methodology and data source for all three studies was the same.

The new study incorporates a period of sharp decline in commercial real estatemarkets. The average default rate for all cohorts rose from 13.8% in the secondSnyderman study to 16.2% in the current study. For cohorts with at least ten yearsof 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 cohortswith a minimum of ten years of seasoning. Although the new study shows higherdefault rates than previous studies, the new average for all cohorts with a mini-mum of ten years of history is very close to Snyderman’s projected default ratesmade for these cohorts before the new study.

IMPACT OF DEFAULTS ON BOND STRUCTURES

The new study found that the worst origination cohort since 1972 had acumulative default rate of 27.6%. Applying a 37.7% severity rate to this defaultrate gives an estimated cumulative loss of 10.4%. The current subordinationlevels for BBB CMBS range from 8% to 12%, implying that most investmentgrade 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 alldefaulting loans are liquidated. In reality, we conservatively estimate that onlyabout 50% of defaulting loans are liquidated, and the rest are restructured. Inthe insurance company data, we are able to ascertain the severity of loss onlyon liquidated loans, and not restructured loans. If, as in the earlier studies, weassume that the severity on restructured loans is half that on liquidated loans,the average severity rate on defaulting loans drops to 28.2%. Applying thisseverity to the highest default cohort gives a cumulative loss of 7.8%, wellbelow the BBB subordination level.

While estimated default and loss rates should leave the principal of investmentgrade CMBS untouched, they can have an impact on non-investment gradesecurities. 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%. Incomparison, single-B and BB CMBS subordination levels currently average about 3% and 6%, respectively.

Please refer to important disclosures at the end of this report. 129

Page 132: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

CONCLUSION

Our update of the two earlier commercial mortgage default studies authored byMark P. Snyderman includes the effects of the commercial real estate recession of the early 1990s. The results of the current paper raise the average expectedcumulative default rate of an origination cohort, but do not significantly changeexpectations 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 andmortgage conduit originations. Loan size, property concentrations, LTVs, debtservice coverage, and geographic distribution of a given conduit’s originationsmay vary significantly from the life insurance company average.

In addition, the procedures taken by a life insurance company on problem loansmay differ from a CMBS servicer. For example, life insurance companies generallyoperate under regulatory constraints that give preference in terms of capitalcharges 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 ofcommercial real estate finance on real estate credit cycles. Some analysts believethat the growth of the CMBS and REIT markets will dampen the traditionalboom/bust cycle of commercial real estate. If these analysts are correct, futureupdates of commercial default studies may show less volatility in commercialmortgage default rates.

Transforming Real Estate Finance

Commercial Mortgage Defaults

130

chapter 10

Page 133: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 131

Year

s Si

nce

Ori

gina

tion

12

34

56

78

910

1112

1314

1516

1718

1920

2122

2324

25AV

G.

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

0.19

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

1973

0.14

2.99

4.76

3.40

1.63

0.54

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

1974

0.71

5.67

2.97

2.27

0.57

0.71

0.14

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

1975

2.33

2.33

1.94

0.58

0.58

0.58

0.39

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

1976

0.21

0.82

0.41

1.23

0.21

0.00

0.21

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

1977

0.00

0.85

0.14

0.28

0.14

0.28

0.14

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

1978

0.28

0.19

0.09

0.00

0.38

0.00

0.09

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

1979

0.27

0.18

0.09

0.62

0.18

0.36

0.18

2.13

1.07

1.96

1.16

1.16

1.51

1.42

0.80

0.36

0.18

0.36

0.00

1980

0.24

0.47

0.24

0.00

0.59

1.30

2.25

1.54

2.01

1.54

2.01

2.01

1.42

0.83

0.12

0.24

0.36

0.24

1981

0.75

0.00

1.00

0.50

0.75

2.74

2.00

2.49

2.49

1.25

1.75

2.24

0.75

0.25

0.00

0.50

0.00

1982

1.38

0.92

2.29

2.75

5.05

1.83

0.00

1.38

3.67

0.00

0.92

0.92

0.00

0.46

0.00

0.00

1983

1.26

0.31

1.10

4.40

2.51

3.30

0.78

0.47

1.88

0.78

1.10

0.16

0.16

0.00

0.00

1984

0.00

1.41

3.18

1.94

4.59

2.12

3.71

1.77

2.30

1.59

0.53

0.35

0.00

0.00

1985

0.08

2.40

2.08

2.48

2.56

3.68

3.28

1.68

1.12

0.32

0.48

0.16

0.16

1986

0.00

1.18

2.16

2.06

3.44

4.92

8.46

2.26

1.28

0.49

1.08

0.29

1987

0.21

0.31

1.77

2.71

4.27

5.63

3.65

2.08

1.04

0.31

0.73

1988

0.10

0.29

1.87

4.72

4.13

4.52

2.75

1.28

1.28

0.29

1989

0.35

0.92

3.12

3.35

2.54

2.08

2.66

1.62

0.35

1990

0.41

3.13

4.63

2.18

2.18

3.00

1.09

0.14

1991

0.72

1.20

1.68

1.68

1.20

0.24

0.48

1992

0.93

0.31

1.87

1.25

0.00

0.00

TIM

ING

OF

DEFA

ULTS

BY

COHO

RTap

pend

ix

Sour

ce: T

he R

atin

g of

Com

mer

cial

Mor

tgag

e-Ba

cked

Sec

uriti

es, D

uff a

nd P

help

s Cr

edit

Ratin

g Co

.

Page 134: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

KEY POINTS

•EUROPEAN CMBS RELATIVE VALUEEuropean CMBS are attractively priced against relevant benchmarks. We anticipatespread tightening as investor sponsorship grows.

•EUROPEAN COMMERCIAL MORTGAGE SECURITIZATIONWe forecast issuance of European CMBS will rise markedly due to the attractionof the CMBS product to both issuers and investors.

•EUROPEAN COMMERCIAL REAL ESTATEFundamental measures of European real estate markets all indicate that the mar-kets are in good health.

INTRODUCTION

Although a significant number of securitization transactions can claim to includereal estate components, we have adopted a relatively narrow interpretation ofwhat constitutes CMBS transactions.

For the purposes of this report our definition includes:

• Debt issuance sponsored by real estate companies, backed by leasedreal 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 fromcorporate sellers to whom the real estate is then leased back for anextended period. These transactions rely on corporate credit covenantsbacked by real estate assets.

This categorization results in the omission of, for example, operating companytransactions which have an element of real estate backing, such as pub securitiza-tions, as well as commercial real estate-related non-performing loans.Additionally, although a number of nursing home transactions rely on propertyleases to nursing home operators, the analytical approach employed has tendedto rely on a detailed understanding of the economics of the individual underlyingnursing home businesses. This contrasts with a more traditional real estate securi-tization analysis which focuses on the credit quality of the lessee at a somewhathigher level and sensitizes the transaction with allowance for lease breakage,vacancy rates and replacement lease rates. For these reasons we classify thesenursing home issues as operating company transactions.

Transforming Real Estate Finance

European CMBS

132

chapter 11

Page 135: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

RELATIVE VALUE CONSIDERATIONS

The following charts show the relative spread performance of CMBS compared toequivalently rated corporate and supranational debt.

We have mapped the Broadgate A2, Canary Wharf I and Trafford Centre AAAbonds, which have average lives of 19, 16 and 25 years, respectively, against theEIB 04/14 and 12/28 issues. The CMBS issues have recently been offering a spreadpick-up which has tended to narrow from 50-60 bp to 30-40 bp over the last 6months against the similarly rated and comparable average life supranationals.

In exhibit 2 we show the bonds in exhibit 1 swapped to £ LIBOR against theELoC AAA floater and two BBB swapped £ bonds issued by two leading UK realestate companies. The AAA CMBS have generally been trading in a range of 20-40bp over LIBOR over the last 6 months.

Please refer to important disclosures at the end of this report. 133

Note: Fixed Rate Spreads (all except ELoC which is a floating rate issue) are shown on a swapped to LIBOR basisSource: Morgan Stanley

AAA CMBSCOMPARED TO

SUPRANATIONALAND REAL ESTATE

COMPANY DEBT

exhibit 2

Source: Morgan Stanley

40

60

80

100

120

140

160

180

200

Jan-00 Mar-00 May-00 J ul-00 Sep-00 Nov -00 Jan-01 Mar-01

bp

CMBS

Supra-nationa

Broadgate A2 Canary Wharf A Trafford Centre EIB 12/28 EIB 04/14

-100

-50

0

50

100

150

200

250

300

350

Jan-00 Mar-00 May-00 J ul-00 Sep-00 Nov-00 J an-01 Mar-01

bp

BBBProp. Co s

AAACM BS

Supra-national

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

FIXED £ AAACMBS TRADEWIDER THAN

FIXED £SUPRANATIONAL

BENCHMARKS

exhibit 1

Page 136: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Exhibit 3 shows the selected fixed and floating AAA CMBS swapped to Euribor.

European CMBS provide yield spread advantages over comparable assets. Whilstthis differential has narrowed with increasing investor sponsorship of the sector,European CMBS remains an attractive asset class. We believe spreads will tightenfurther as liquidity improves, with growing supply and investor participation.

CMBS CREDIT R ISKS

Although the previous section of this survey suggested that there is relativelyattractive pricing available to investors in CMBS, investors need to understandwhether this arises as a result of materially higher credit risks. This section sug-gests that investors in CMBS arguably face less risks than in unsecured corporatedebt and are rewarded for participating in a sector that has yet to reach full matu-rity and investor sponsorship.

Although no hard data is available on the European market at this point due to itsdeveloping state, there has been a large scale study of US CMBS performanceconducted by Fitch.

Key data on the Fitch study is presented in exhibit 4.

Average Annual Default Rate %

Corporate Bonds CMBSInvestment grade 0.08 0.06Non-investment grade 3.07 0.14All Bonds 1.51 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 maypossibly have been superior to corporate debt in view of the former’s collateralized status).

Transforming Real Estate Finance

European CMBS

134

chapter 11

Note: Broadgate, Canary Wharf and Trafford are Fixed Rate issues: ELoC is floating rate.Source: Morgan Stanley

-30

-20

-10

0

10

20

30

40

50

60

bp

Broadgate Canary Wharf A Trafford Centre ELOC

Mar-00Jan-00 May-00 Jul-00 Sep-00 Nov-00 Jan-01 Mar-01

AAA CMBSSWAPPED TO

EURIBOR

exhibit 3

Page 137: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

1 See “Commercial Mortgage Defaults: An Update” in Real Estate Finance (Spring 1999)

Please refer to important disclosures at the end of this report. 135

The most significant variation in performance is in the non-investment grade area,in which non-investment grade CMBS have displayed a materially superiorperformance to similarly rated corporate debt.

The reasons attributed for this outperformance have been mainly related to theperception that CMBS issues, particularly as they became larger, increasinglybenefited from underlying diversification, and managed to avoid undueconcentrations 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 defaultperformance is relatively favorable. In early 1999 a study by Esaki, L’Heureux andSnyderman was produced detailing defaults on a sample of 15,000 commercialmortgage loans originated between 1972 and 1997 by US life insurancecompanies that have historically been major lenders to commercial real estate inthe 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 onliquidated loans averaging 37.7% and therefore resulting in an average loss of6.82%. Interestingly, this loss level compares to credit enhancement of 12-15%that typically accompanies BBB rated CMBS. Therefore, credit enhancementstructures of CMBS should, on average, comfortably absorb average losses oncommercial mortgage loans at the BBB level. In fact the study went one stepfurther 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.

FITCH CMBS DEFAULT STUDY: 31DEC 1990 TO 31 DEC 1999

exhibit 4

Number of Initial Cumulative Amount CumulativeRated Issuance Number default by defaulted default rate by

Rating Bonds ($bn) defaulted number (%) ($m) amount (%)

Investment grade 2,582 211.6 9 0.35 419.9 0.20

Non-investment grade 893 20.4 3 0.34 83.3 0.41

AAA 902 139.4 0 0 0 0AA 585 34.6 5 0.85 314.4 0.91A 462 18.0 2 0.43 63.8 0.35

BBB 633 19.6 2 0.32 41.1 0.21BB 449 11.9 1 0.22 42.6 0.36B 407 7.5 2 0.49 41.3 0.55

CCC 37 0.9 0 0 0 0Total 3,475 231.9 12 0.35 503.2 0.22

Source: Fitch

Page 138: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

HISTORIC ISSUANCE

Based on the definitions set out in the introduction, we find that there have todate been very few true CMBS transactions completed in Europe and the UK hasprovided the highest share of transactions. We find that, according to our catego-rization, only 30 European CMBS transactions were completed from the beginningof 1995 through to the end of 2000. Appendix 1 provides a listing with key eco-nomic features of the transactions that have been completed to date.

Exhibit 5 shows historic European CMBS issuance over the period 1996-2000 andcontrasts this with US CMBS issuance over the same period.

The European market is currently just a fraction of the US market, which originallydeveloped in the early 1990s. The US market developed partly as a method ofresolving the asset disposition issues raised by the savings and loan crisis, and alsoas a means of providing new sources of funding following the drastic reduction inthe 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, orare 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

Transforming Real Estate Finance

European CMBS

136

chapter 11

Source: Morgan Stanley

0.3 2.3 4.2 5.9

31.6

44.4

85.4

64.3

54.4

0.20

10

20

30

40

50

60

70

80

90

1996 1997 1998 1999 2000

e bn

U.S.

EuropeUS VS EUROPEANCMBS ISSUANCE,

1996–2000

exhibit 5

Page 139: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Exhibit 6 shows for the European CMBS market the relative split of issuancesince the beginning of 1996 between UK and continental transactions (UKtransactions are defined as involving both UK originators and also primarily UKdomiciled collateral).

We believe there are 3 primary reasons for the dominance of U.K. transactionswithin total European issuance.

• The U.K. tends to follow on more closely to US-originated financial innova-tion than countries on the continent.

• The prevalence of long-term, fully repairing and insuring lease structures inthe U.K., together with the relatively creditor-friendly nature of the U.K. insol-vency regime, are important factors that facilitate securitization.

• The relative significance of quoted real estate investment companies in theU.K. is a further differentiating factor, and this sector has been the source of asignificant number of transactions.

Exhibit 7 breaks out the origination or sponsoring source of historic issuancesince the beginning of 1996.

2 The differences between UK and European leases are set out in Appendix 2.

Please refer to important disclosures at the end of this report. 137

Source: Morgan Stanley

Continent22%

UK78%

EUROPEAN CMBS:SPLIT BETWEEN

CONTINENTALAND UK

TRANSACTIONS,1996–2000

exhibit 6

Source: Morgan Stanley

Corporate/Other22%

Conduit16%

Real Estate Co43%

Bank19%

ORIGINATIONSOURCE OF

EUROPEAN CMBSISSUANCE,

1996–2000

exhibit 7

Page 140: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The relative dominance of real estate company transactions reflects the recentneeds of the quoted real estate company sector in the U.K. in requiring substantial,long-dated financing against specific developments that have been more easilyaccommodated 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 havebeen completed to date involve U.K. assets.

We have set below an outline of some of the key features of transaction structuresand associated risks that need to be considered by investors when consideringinvestment in CMBS. Although the description is based heavily on our experienceof U.K. transactions, the general principles are applicable to all real estate markets,although legal processes will vary between jurisdictions. We show in Appendix 2the main differences in landlord and tenant relationships between the mainEuropean centers.

Transforming Real Estate Finance

European CMBS

138

chapter 11

Source: Morgan Stanley

Bank

Issuing SPV Bondholders

Borrower A Borrower B

Lessee 1 Lessee 2

Bank

Liquidity facilityInterest Rate Hedge

Bonds/Collateral

Cash

Loan

Lease Rentals Lease Rentals

Collateral

Collateral

Cross-Collateralisation(Potential)

Loan

TYPICAL CMBSISSUANCE

STRUCTURE

exhibit 8

Page 141: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

TRANSACTION STRUCTURES

Exhibit 8 sets out in a simplified schematic form some of the key relationshipsthat commonly feature in CMBS transactions.

The following table highlights the common structural features of transactions.

Legal StructureExhibit 8 describes a simplified typical structure and can be used to illustrateboth single-asset/single corporate sponsor and conduit-style transactions.

Single asset/corporate transactionsAn issuing vehicle will make a loan of bond proceeds to one or more borrow-ers, who in turn will be property-owning subsidiaries within a company groupstructure (Borrowers A and B) ultimately owned by a holding company. Theseborrowers will in turn be the owners of specific properties that will be leasedto underlying lessees. The leases generate the rental flows to the borrowersthat in turn will enable the borrowers to service the loans from the issuer. Theissuing SPV can be included within the corporate structure with the same ulti-mate holding company owner as borrowers A and B.

Conduit transactionsConduit transactions are broadly similar except that Borrowers A and B will beunconnected and the loans to these borrowers will instead usually have beenoriginated by a commercial lender prior to being assigned to the issuer at thetime of the bond issue. The assets of Borrowers A and B will be owned prop-erty assets leased to end-users.

For both single asset and conduit transactions, Borrowers A and B will normal-ly be limited or special purpose companies with contractually limited otheractivities. These borrowers will fulfill no material role other than to borrowfrom the issuer (or intermediate borrower) and to own and lease the relevantproperty and to enter the asset pledge and collateral agreements.

CollateralSingle asset/corporate transactionsThese transactions will feature collateral pledges from Borrowers A and B tothe issuer over the benefit of their interests (freehold and leasehold) in theunderlying properties and leases, and all of the cashflows therefrom. Issuerswill in turn pledge to the security trustee for bondholders their interest in thecollateral pledged by borrowers via the pledge of the secured loans made tothe individual borrowers.

Conduit transactionsSimilar to single asset transactions, conduit transactions will feature pledgesfrom Borrowers A and B to the issuer over their interests (freehold and lease-hold) in the underlying properties and leases, and all of the cashflows there

Please refer to important disclosures at the end of this report. 139

KEY STRUCTURAL FEATURES OF CMBS TRANSACTIONS

exhibit 9

Page 142: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

from. Issuers will in turn pledge to the security trustee for bondholders theirinterest in the collateral pledged by borrowers via the pledge of the securedloans made to the individual borrowers.

Fixed and floating charges over their assets are generally granted by borrowersto the security trustee of the securitization, via the issuer, which should enablethe trustee to control any insolvency of a borrower through the appointment ofan administrative receiver to deal with the secured assets and minimize anypotential delay to cashflow. The exercise of, specifically, the floating charge,and the appointment of an administrative receiver to avoid a potentiallylengthy court process in insolvency is a particular feature of the collateral struc-ture under English law.

Cross-CollateralizationAn important distinction between conduit and non-conduit transactions is that,in the case of conduit transactions, the underlying borrowers are not normallypart of the same corporate entity and therefore underlying properties cannotbe cross-collateralized.

However, excess interest on the pool of loans (in simple terms the differencebetween 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 notcross-collateralized.

Although the absence of cross-collateralization may be a negative feature inconduit transactions, compensating features usually include borrower andproperty diversity.

Asset SalesParticularly in the case of corporate transactions, borrowers may want to pre-serve the option of selling specific assets from the portfolio. Structures can per-mit this, but for concentrated portfolios there is normally a requirement forproceeds to be generated to the extent that they raise 115%-130% of the under-lying financing allocated to that property which is then used to pay down thefinancing. This technique avoids the potential negative adverse selectionthrough the cherry-picking of properties and reduces the LTV of the remainingfinancing. Clearly, this mechanism will not apply to non cross-collateralizedconduit transactions.

Liquidity SupportAlthough legal structures of UK transactions are arranged in the expectationthat underlying loans will not become embroiled in extended insolvency pro-ceedings, liquidity facilities from highly-rated banks provide support for timelypayment of interest and principal in the event of borrower insolvency or tem-porary disruptions to cashflows due to re-tenanting. Liquidity facility providersare ultimately senior to bond investors, and such facilities usually containrestrictions on how much liquidity can be advanced to support junior classes.

Transforming Real Estate Finance

European CMBS

140

chapter 11

continuedexhibit 9

Page 143: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 141

KEY CREDIT RISKS OF CMBSTRANSACTIONS

exhibit 10

Liquidity facilities also normally contain borrowing base restrictions whichrequire underlying assets to provide minimum coverage levels for liquiditydrawings.

Interest Rate HedgingBond marketing considerations may require at least partial issuance in floatingrate instruments. Underlying lease cashflows are by their nature fixed flowsand cashflows on underlying conduit loans can be fixed. This mismatch is covered by interest rate hedging with bank counterparties.

Cash ControlRentals due on underlying leases are ideally directed to special trusteeaccounts to avoid commingling and associated corporate bankruptcy riskswhere applicable.

Reserve AccountsReserve accounts are funded (at the expense of the equity holder in the trans-action) for various reasons. These may typically include reserves required ifspecified debt service coverage levels are breached or special reserves thatmay be required, for example, if a defined share of underlying leases are dueto terminate within a certain period prior to a refinancing date. Debt servicereserves are alternatively seen pending achievement of certain rental levels inlease-up situations.

The following table describes the key credit risk concerns in CMBS transactionsand some of the ways in which bondholder protection is structured to overcomethe credit issues.

Tenant QualityThe credit quality of transactions will benefit from a preponderance of invest-ment grade tenants. This is a rare feature outside of ‘trophy’-type securitiza-tions. The value of higher rated tenants is that they are less likely to default onleases in times of economic stress, which will reduce the exposure of the trans-action to re-leasing on tenant default with associated cashflow losses arisingfrom both the time taken to re-lease the property and the associated potentialcashflow losses that might arise from the need to re-lease the property at lowerrental rates in difficult economic circumstances.

Tenant DiversityHigh tenant quality is frequently combined with limited tenant diversity. This is a feature of city office developments but less conspicuous with large-scaleretail developments, where the tenant base is usually more widely spread,although the tenants may be in the same industry and retail operators may

Page 144: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

each suffer similar business stresses at the same time. Risk to transactions cantherefore become concentrated: a single lessee default can cause material dis-ruption. For example, in the Canary Wharf II transaction 65% of closing rentalswas derived from 2 tenants – yet in ELoC 4 there are 440 separate tenants witha maximum rental of 4% from a single tenant.

Lease MaturitiesLong 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 thatdebt can be refinanced.

Lease Break Clauses and MaturitiesKey analytical assumptions when reviewing real estate financings involve taking account of the pattern of lease maturities and break clauses within leas-es. Transactions are assessed according to how well they may perform if leasetermination (or lessee default) coincides with recessionary conditions, whichrequire material discounts in rentals or rent-free periods to entice new tenants.Transactions are required to withstand progressively higher levels of discount inline with higher desired ratings on underlying securities. Rental decline assumedin recessions may range from 20-35% depending on the credit rating desired.

An even distribution of lease maturities is preferable to bunching as thisreduces the risk of lease maturities coinciding with economic stresses.

Generally, the assumptions for rated transactions are that 65% of maturingleases are renewed by tenants, with the balance re-leased, after an interval, at lower rental levels.

Re-Leasing PeriodsAssumptions for time periods required for re-leasing vacated properties or findingnew lessees post default will depend upon the desired rating of the underlyingbonds with assumed periods normally between 15 and 18 months.

The nature of real estate collateral is also key here as the assumed re-leasing peri-ods may be lower in the case of high quality leading city developments or retailpark sites that are multi-tenanted rather than isolated, specialized, single tenant sites.

Macro-perceptions on the durability of certain underlying industries may alsoaffect views on re-leasing prospects. For example, long-term perceptions concern-ing 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/internetshopping could impact perceptions of the Trafford Centre transaction.

ValuationsValuations at closing are obtained from leading valuation companies and

Transforming Real Estate Finance

European CMBS

142

chapter 11

continuedexhibit 10

Page 145: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

transactions have been able to support the issuance of BBB securities at LTVsof 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- levelsat LTVs of 70%, BBB at 78%, BB at 76%. The underlying structure and natureof 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 theamount 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 ofcompletion. A further complication occurs if differing amortization require-ments 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 willbe permitted to operate on a lower debt service coverage level than borrowerswith 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 asdiscounted rents expired – whereas initial interest cover for similarly-ratedtranches on the ELoC transactions ranged between 1.18x and 1.64x.

Amortization StructuresArrangements vary, although it is common for transactions to require balloonrefinancing even at the end of extended debt tenors – Canary Wharf II featuresa £100m (21% of original principal) refinancing requirement after 30 years. Theassumption is, notwithstanding the extended maturity, that the real estate canbe comfortably re-financed at these levels. Trafford Centre, however, is sched-uled to amortize fully over its 22 year life.

In contrast, the ELoC transactions are much shorter term with final maturitiesnot exceeding 9 years. Amortization reducing the underlying loan LTVs com-pared 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 ispaid down faster for the transaction life than would be required under a 30year mortgage-style amortization profile), and also with the benefit of projectedstrong interest cover levels at the refinancing date, then refinancing opportuni-ties for the underlying loans would be available.

Please refer to important disclosures at the end of this report. 143

Page 146: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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 assetsare able to support high levels of AAA securities issuance. For example, over 80%of the Europa One transaction comprises AAA securities, whilst for the EuropeanLoan Conduit programme as a whole the figure is 75%. The Canary Wharf II andTrafford Centre transactions both comprise 63% of AAA securities, whereas the MSMortgage Finance (Broadgate) transaction has 51% of its issuance in AAA-ratedclasses. We note that ratings at these levels are not always readily achievable bysome other classes of asset-backed securities. For example, in the absence ofthird party support, issuance from the operating company sectors has generallybeen ‘capped-out’ at a level of A.

This transaction represents the refinancing of part of the British Land Company’sinterest in the Broadgate office complex in the City of London. The transaction isparticularly notable for the high credit quality of the underlying tenant base andalso the long average periods before leases can be broken. The very large size ofthis transaction should help to promote secondary market liquidity.

• Total £1,540m• LTV 73.1%• Weighted average tenant credit rating of A-• Average period to first tenant break of 15 years• Fully amortizing • Asset substitution provisions• £15m reserve/£200m liquidity facility

Transforming Real Estate Finance

European CMBS

144

chapter 11

Rating (Moody’s/S&P/Fitch) Amount (£m) Interest Basis WALAaa/AAA/AAA 325 Floating 8.5Aaa/AAA/AAA 310 Fixed 19.3Aaa/AAA/AAA 150 Fixed 31.1

Aa2/AA/AA 225 Fixed 29.1A2/A/A 175 Floating 9.1A3/A/A 175 Fixed 31.1

Baa2/BBB/BBB 180 Floating 3.5

MORGAN STANLEY MORTGAGEFINANCE (BROADGATE) PLC

exhibit 11

Source: Morgan Stanley

Page 147: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

This transaction represents the financing of part of the Canary Wharf complex inLondon’s Docklands. Similar to Broadgate, the transaction features very high aver-age tenant credit ratings and long term leases. Canary Wharf was notable for theissuance of a substantial Euro denominated tranche.

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 largepercentage of rentals derived from long-dated leases, but differs from the twooffice transactions in that investors enjoy the benefit of very considerable tenantdiversity; there are over 230 separate tenants that lease space in the complexalthough the underlying credit ratings of the tenant base is materially lower thanfor the two office transactions described above. For example, the credit ratings forthe three anchor tenants, who account for 30% of the lettable space, do notexceed BBB+. Key transaction features are:

Please refer to important disclosures at the end of this report. 145

Rating (Moody’s/S&P/Fitch) Amount (m) Interest Basis WALAaa/AAA/AAA £240 Fixed 20.2Aaa/AAA/AAA e100 Floating 6.1

Aa2/AA/AA £ 85 Fixed 23.9A2/A/A £ 45 Fixed 25.1

Baa2/BBB/BBB £ 45 Floating 5.0

CANARY WHARF FINANCE II PLCexhibit 12

Source: Morgan Stanley

Rating (Moody’s/S&P/Fitch) Amount(£m) Interest basis WALAaa/AAA/AAA 50 Floating 9.8Aaa/AAA/AAA 340 Fixed 25.6

Aa2/AA/AA 120 Fixed 19.8Baa2/BBB/BBB 50 Floating 7.6Baa2/BBB/BBB 50 Fixed 21.0

TRAFFORD CENTRE FINANCE LTDexhibit 13

Source: Morgan Stanley

• Total £475m

• 68.4% of rentals from two tenants(with average credit rating of AA)

• £100m financing balloon

• LTV 70.7%

• Debt Service Coverage 1.14x(Year 1); Debt Service Coverage 1.39x (Year 2)

• Total £610m

• LTV 69.3%

• 50.4% of rentals from unexpiredleases of 20+ years

• 93.1% of rentals from unexpiredleases of 10+ years

• £15m reserve fund

• £55m liquidity facility

• Diversified tenant mix

Page 148: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The European Loan Conduit programme represents the periodic securitization ofsmaller mortgage loans originated by Morgan Stanley. In these transactionsinvestors generally get the benefit of considerable borrower and tenant diversity.For example, in the ELoC 2 transaction there are 105 separate commercial proper-ties (in addition to 272 pubs). The ELoC transactions are shorter-dated floatingrate transactions for which repayment relies on the ability to refinance the under-lying loans at maturity. The ELoC 5 transaction, totalling £524.9m, is currently inthe course of completion.

Europa One is sponsored by Germany’s Rheinische Hypothekenbank and is acredit linked note structure (with tranches being linked to both an underlying ref-erence pool and also to either the bank’s Pfandbriefe or its senior unsecurednotes). The transaction is designed to remove the risk of a portfolio of real estateloans from Rheinhyp’s balance sheet. Although nominally a ‘German’ transaction,Europa One is in reality the first pan-European CMBS transaction as is evidentfrom the geographic dispersion of the underlying real estate collateral.

Transforming Real Estate Finance

European CMBS

146

chapter 11

1 S&P rated all four ELoC transactions. Fitch rated ELoC 1 and 4 only.

Rating (S&P) ELoC 1 ELoC 2 ELoC 3 ELoC 4AAA 135.1 280.4 197.3 328.1AA 10.1 26.9 17.8 40.4A 6.8 19.8 15.3 39.3

BBB 16.9 21.6 15.9 42.8BBB- - - - 11.6

BB - 10.8 8.3 -Total 168.9 359.5 254.6 462.2LTV 80.7% 80.0% 76.0% 70.0%

No. loans 13 11 16 6Debt Service

Coverage (Interest Only) 1.64x 1.48x 1.34x 1.18xFinal maturity 2008 2008 2009 2007

All amounts £ m

EUROPEAN LOAN CONDUIT (ELOC)PROGRAMME

exhibit 14

Source: Morgan Stanley

Page 149: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• 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.

EUROPEAN REAL ESTATE MARKET FUNDAMENTALS

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 havegradually improved.

REAL ESTATE Y IELDS

One leading indicator of improvement is the performance of reported real estateyields. For these purposes ‘yield’ is very broadly defined as the rentals achievedfrom a property divided by its purchase cost. There are technical differencesbetween practices in different national markets concerning whether, for example,the purchase and running costs of real estate are taken into account when calcu-lating the numerator. If these costs are not taken into account then this yield isdescribed as a ‘gross’ yield basis, whereas the practice in some markets is toquote on a triple-net basis, which means that all purchase and non-recoverablecosts of running the building are taken into account, reflecting more accurately a real estate investor’s actual return.

Please refer to important disclosures at the end of this report. 147

Rating(Moody’s/S&P) Amount (em) Interest Basis WALAaa/AAA 290.0 Floating 1.3Aaa/AAA 400.0 Floating 4.1Aaa/AAA 400.0 Floating 7.9Aa1/AAA 30.0 Floating 11.9Aa2/AA 37.0 Floating 12.7

A2/A 80.5 Floating 14.1Baa2/BBB 57.0 Floating 14.5

Ba2/BB 30.3 Fixed 14.5N/R 20.2 Fixed 14.5

EUROPA ONE LIMITEDexhibit 15

Source: Morgan Stanley

Page 150: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Irrespective of these differences, we can partially discern the health of real estatemarkets by analysis of real estate yield behavior. The general proposition is thatin relatively stable economic conditions, declining yields in real estate marketswill, prima facie, be evidence of improving demand for real estate assets. In sum-mary, declining yields are evidence of investor willingness to pay more than pre-viously for a series of rental cashflows. Also, as inflation and long term interestrates fall, we might normally expect real estate yields to move in broadly thesame direction.

Clearly, in buoyant real estate markets there is likely to be a relative deficiency ofsupply for end-users from the core office, retail and industrial/warehouse sectorsand this will naturally tend to drive up lease rental rates.

Therefore, it is entirely conceivable that both rentals and real estate values willrise simultaneously, although rentals rising more slowly than real estate acquisi-tion 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 cash-flows, then real estate yields might also correlate to changes in a suitable bench-mark, such as government securities yields. However, this relationship will be byno means perfect as real estate yields are not risk-free investments and will alsobe affected by investor perceptions of probable future real estate performanceand inflation risks.

OFFICE MARKETS

Exhibit 16 shows composite yield performance in the main European marketsacross the 12 month period Autumn 1999 to Autumn 2000.

Transforming Real Estate Finance

European CMBS

148

chapter 11

All Figures % Autumn 1999 Autumn 2000Paris 5.75 5.50

Frankfurt 5.00 4.90Berlin 5.25 5.00

Munich 4.50 4.75Dublin 5.00 4.75Milan 6.00 5.75

London – West End 5.50 5.25London – City 6.00 6.25

London – Canary Wharf 6.50 6.50

EUROPEAN PRIME OFFICE YIELDS: INDICATE IMPROVED MARKET CONDITIONS

exhibit 16

Source: FPD Savills, Jones Lang Lasalle, Healey & Baker

Page 151: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

In most markets office yields have continued to tighten marginally over the 12month period. Market reports identify limited availability of space in many mar-kets, driven by the health of the financial services industry and also the continuinginflux of telecom, media and technology companies to many traditional centralbusiness district locations and their immediate fringes.

The other major pattern that has emerged is the limited current supply of newspace. The reasons for this vary, ranging from certain physical constraints(Dublin), to tight central area planning restrictions (Milan, Paris) and reasonablyconservative development pipelines. Particular pressure is noticeable in Germanoffice markets.

It should be noted that data quality in the office sector is generally superior tothat in the more fragmented retail and industrial sectors and this is why most realestate market studies often tend to use office markets as a proxy bellwether forgeneral real estate market health.

RETAIL MARKETS

Exhibit 17 shows the behavior of retail yields over the 12 month period.

In the retail sector yields have held generally held constant in the major marketswith the lowest absolute levels seen in some German markets. We believe thatinvestment sentiment towards retail real estate investments is not, in general, nega-tively affected by perceptions of threats to the health of retail real estate via extremecompetition, importation of US retailing methodologies and threats to traditionalretail outlets from internet shopping and alternative non ‘high street’ locations.

Please refer to important disclosures at the end of this report. 149

All Figures % Autumn 1999 Autumn 2000Paris 6.00 6.00

Frankfurt 4.75 5.00Berlin 5.05 5.00

Munich 4.00 4.25Dublin 3.75 3.75Milan 5.00 5.00

London – West End 6.00 5.50

EUROPEAN PRIME RETAIL YIELDS:INDICATE STABLE MARKET CONDITIONS

exhibit 17

Source: FPD Savills, Jones Lang Lasalle, Healey & Baker

Page 152: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

WAREHOUSE AND LOGIST ICS MARKETS

Exhibit 18 shows the changes in yields in warehouse and logistics markets overthe last 12 months.

These markets are materially more fragmented than the office and retail sectors andyields are generally wider to take account of these factors and the associated risks.

VACANCY RATES

Another key leading indicator of market health is represented by vacancy rateperformance. Vacancy rates will tend to rise when overall economic conditionsdeteriorate and businesses stop expanding or even close down and vacate prem-ises. Conversely, vacancy rates will tend to fall as economic conditions improveand businesses require more space for operations.

Vacancy rates will also be crucially affected by the relationship between the sup-ply of new space and demand for that space. It is entirely conceivable that vacan-cy rates can rise during buoyant economic conditions if the supply of new spaceexceeds the rise in demand. More usually, however, vacancy rates will tend todecline 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 officemarkets, including data for the last two years and also noting the high points forthose markets in the 1990s.

The data is consistent with an explanation that the relevant markets are currentlyin 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 themarket for investment property, which has seen yields decline and hold at rela-tively low levels, as shown in exhibits 16-18.

Transforming Real Estate Finance

European CMBS

150

chapter 11

All Figures % Autumn 1999 Autumn 2000Paris 10.00 9.50

Germany (composite) 7.50 7.75Dublin 7.00 6.50Milan 9.00 8.50

London – West End 6.50 7.25

EUROPEAN PRIME WAREHOUSE/LOGISTICSYIELDS: INDICATE MORE VOLATILE MARKETCONDITIONS

exhibit 18

Source: FPD Savills, Jones Lang Lasalle, Healey & Baker

Page 153: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The data needs to be interpreted with some care because the relatively high ratesfor Milan and London could be misinterpreted. Reports indicate, for example, thatthe supply of suitable modern space in Central Milan is severely limited due toplanning restrictions and this has resulted in a number of businesses moving outto business parks on the edge of, or outside, the city. In turn this will raise thereported vacancy rate in the central business district although this masks the factthat an amount of the space in the centre will increasingly become obsolete inthe absence of an ability to redevelop.

In London, we believe that the sheer size of the market is able to support a high-er overall level of vacancies on the basis that the higher numbers of buyers andsellers of real estate space in London will improve the chances of unlet spacebeing absorbed. Also, we think that the gradual drift of some major City tenantsto Canary Wharf has had a partial impact on City vacancy rates, although the con-tinuing level of overall high demand has easily enabled the City office market tocontinue to perform very strongly.

RENTAL TRENDS

A further leading indicator of real estate market health is provided by trends inrental levels.

Exhibit 20 shows the performance of rental growth in key European office marketlocations in the 12 months up to the end of Q3 2000.

Please refer to important disclosures at the end of this report. 151

All Figures % High (Year) 1999 2000Paris 9.0 (1994) 4.3 4.0

Frankfurt 9.0 (1995) 5.0 4.5Dublin 14.0 (1992) 1.0 1.0Milan – – 6.1

London – City 18.0 (1992) 7.0 6.0Composite (14 cities) 9.0% (1993) 4.8% 4.0%

EUROPEAN OFFICE VACANCY RATESexhibit 19

Source: Jones Lang LaSalle, DTZ

All Figures % Rental GrowthParis 29.4

Frankfurt 12.5Berlin 9.1

Munich 15.4Dublin 33.3

EUROPEAN OFFICE MARKETRENTAL GROWTH Q3 99 – Q3 00

exhibit 20

Source: Jones Lang LaSalle

All Figures % Rental GrowthMilan 10.0

London–City 9.6London–West End 33.3

London–Canary Wharf 6.9

Page 154: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The strength of rental growth in many city office markets is entirely consistentwith the strength of real estate markets recorded in our other data sources.

COMPARATIVE PERFORMANCE BENCHMARKS – GOVERNMENT SECURIT IES

We believe that the conclusions from this data are reasonably transparent. Themain European office markets are in good health, with strong increases observedin rental growth, which has been associated with declining vacancy rates andcontinuing downward pressure on yields.

The retail and industrial/logistics sectors are subject to less comprehensiveavailability 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 chasedown 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 relativestability 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 mainEuropean markets. Exhibit 21 shows that government securities yields have movedmarkedly downwards through the year 2000. This has not been fully reflected todate in real estate yields, although there will tend to be data collection and marketreaction lags in the real estate markets compared to the bond markets.

To some extent certain government securities markets have been impacted bytechnical factors unduly affecting their demand and supply patterns. An exampleof this may be the unusually limited supply of government debt which couldhave artificially depressed yield levels. However, to the extent that governmentsecurities yields represent a proxy for the base cost of debt funding for realestate investment, then our comparison may be useful. The fact that these proxybase funding costs have continued to reduce compared to real estate yields mustmake real estate easier to finance, and therefore prima facie more attractive, fordebt-based investors. Of course, the risk and return factors relating specifically tocommercial real estate will tend to reduce the degree of correlation betweenchanges in bond yields and changes in real estate yields. However, the general

Transforming Real Estate Finance

European CMBS

152

chapter 11

All Figures % Yield at 30/12/00 2000 averageFrance 5.050 5.403

Germany 4.901 5.264Ireland 5.137 5.520

Italy 5.300 5.600U.K. 4.897 5.265

10 YEAR EUROPEAN GOVERNMENTBOND YIELDS

exhibit 21

Source: Bloomberg

Page 155: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

proposition is that reducing real estate yields reflect correspondingly lower riskperceptions by investors, and the failure of real estate yields to chase bondyields down actually strengthens the market to the extent that real estate invest-ment becomes easier to finance.

EUROPEAN OFFICE MARKETS – A R ISK OF OVERSUPPLY?

Experienced investors in European real estate and their related securities will beconscious that, notwithstanding currently positive market conditions, too muchdevelopment activity could easily disrupt supply/demand balances and put pres-sure on underlying real estate yields. This was the scenario that developedbetween 1989 and 1992 when an economic expansion was accompanied byexcessive speculative development activity and an overhang of office propertythat needed the economic recovery in the mid-1990s to clear the market.

A recent study, by Jones Lang Lasalle, completed in October 2000, of Europeanoffice development across 12 major centers has revealed the following conclu-sions relating to forecast supply.

Total actual and forecast completions of new office space, based on announcedbuilding projects, are shown in exhibit 22.

Clearly, actual new building volumes can change relatively quickly as projectsthat are put on hold are reactivated and the total of 8.75m square metres antici-pated 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 supplywill be brought on stream over the period 2001-04 and this compares to the 16msquare 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 reason-able case to support the view that underlying macroeconomic conditions are nowmaterially more stable than in the prior period.

Please refer to important disclosures at the end of this report. 153

Year Square Metres (,000)1999 actual 2,9042000 actual 4,2302001 forecast 4,5202002 forecast 4,5502003 forecast 3,5602004 forecast 4,250

FORECAST SUPPLY OF NEW OFFICE SPACEexhibit 22

Source: Jones Lang LaSalle

Page 156: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Supporting evidence for this proposition is provided by the following factors:

• The demand and supply drivers appear superior to 1989-92, with morepre-let space and lower volumes of speculative space being brought tothe 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 positionof excess demand in the earlier period compared to the currentunderutilization of capacity. This scenario can support growth without overheating.

• The broad conclusion is that, provided economic fundamentals remainstable then a similar level of new office supply over the 2001-04 periodcompared to 1989-92 should be comfortably absorbed. Clearly, thestructure of the real estate markets is now different compared to theearlier period and some asset types (for example, buildings withoutmodern cabling) will be subject to rental and occupancy pressures,although the broader picture appears to be more stable.

OWNERSHIP AND FUNDING OF EUROPEAN REAL ESTATE

Although the ownership of real estate is of course highly fragmented, in recentyears we believe that investment patterns in European real estate have changedmaterially. There are a number of reasons for this.

CorporatesEuropean corporates are being affected by globalization trends. This means thatincreased competition, and merger and acquisition activity, has resulted in manage-ments increasingly scrutinizing their asset bases with the intention of ensuring assetsare deployed with maximum efficiency. There is growing evidence that corporatesare 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 officebuildings 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 Telekomof non-core properties.

• Italy: sale of office properties by Mont Edison and also sale by ENI ofits real estate subsidiary, Immobiliare Metanopoli.

• United Kingdom: proposed sale and leaseback of operating assets bysupermarket chain, J Sainsbury plc and telcoms operator, BT.

Transforming Real Estate Finance

European CMBS

154

chapter 11

Page 157: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

InternationalizationThe single currency has prompted a considerable internationalization of Europeanreal estate investment. It is estimated that in 2000 a record e20 bn of cross bor-der real estate investment will have been made. Approximately 50% of directFrench commercial real estate transactions in 2000 are estimated to have beenconducted by non-French investors, with many trades actually being entirelybetween non-French investors. Even the U.K., which is outside of the immediatescope of the Euro, saw 18% of its commercial real estate investment being under-taken by non-U.K. domiciled investors in 2000.

Real Estate CompaniesAlthough dedicated real estate companies have a naturally high profile, theirimpact in terms of overall ownership of underlying real estate can easily be over-estimated. Although measures such as the relative share of the equity of listedproperty companies of their national stock markets are not necessarily good indi-cators of underlying ownership of real estate (which can of course be materiallyaffected by leveraging policies), it is nevertheless an interesting indicator of rela-tive size. The data shows that the average weighted share of dedicated commer-cial 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 estatecompanies comprise approximately 1.5% of the market. Interestingly, this share isdown from 4.5% in the late 1980s. The story behind this decline is one of equitymarket perceptions being increasingly that real estate is a relatively low growthsector which results in unattractive returns compared to other areas in low infla-tion environments. This is particularly true of real estate investment companieswhich have limited speculative activity which, although carrying more risk, is alsopotentially higher revenue producing.

Perceptions towards equity real estate investment in the U.K. have never quiterecovered from the value erosion registered in the early 1990s. The decliningequity market share has been exaggerated more recently by considerable corpo-rate finance activity in the sector which, prompted by equity valuations tradingmaterially below the net asset valuations of companies, has seen a number ofcompanies being taken private. In addition, real estate companies are beginningto look more closely at new business models which increasingly separate out theownership and management of real estate assets.

Although Germany is increasingly developing a quoted real estate sector, assistedby tax reforms that make it easier to sell real estate investments held in sub-sidiaries, the larger share of direct real estate is held by open-ended real estatefunds with an estimated DM 100 bn of assets.

Please refer to important disclosures at the end of this report. 155

Page 158: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Pension Funds and Life Insurance CompaniesOne trend that has been constant across Europe has been for the relative directownership of real estate by long-term investors, such as pension funds and insur-ance companies, to decline. Reasons for this include the increasing perception thatthe need for real estate investment as a partial hedge against inflation is diminishedin a low inflation environment. Additionally, pension funds and insurance companieshave gone the way of equity investors in perceiving real estate to be a relatively lowreturn asset class. Also, smaller pension funds are realizing that it is expensive toundertake active management of small real estate portfolios and minimum portfoliosize is required to obtain a reasonable level of diversity.

Available data suggests that in Germany insurance companies and pension fundshave 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 ofreal 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 infunding real estate is in Italy where almost 50% of pension fund assets are in realestate (compared to a European average, excluding Italy, of 9.3%). In Italy theshare of insurance company assets that are dedicated to real estate investment fellfrom 10% in 1996 to 5.5% in 1998.

Bank Funding for Real EstatePatterns relating to bank funding of real estate show material differences compared to the peak of the previous real estate cycle in 1989-92. We believethat banks have become more sophisticated over that period and have in placebetter controls over aggregate volumes of real estate lending and, in particular,over speculative property lending to real estate development. However, DTZ’sannual survey (March 2000) of bank lending to real estate indicated that reallending 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 boomconcerning maximum loan-to-value ratios, maximum term of loans and also inattitudes to speculative development.

ConclusionOverall, we conclude that real estate has become a relatively less attractive assetfor equity investors in real estate and also for direct investors in real estate in theform of insurance companies and pension funds. To some extent, alternativesources of funding, such as unitized property vehicles in the UK and open-endedfunds 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 ofthe last real estate boom.

Our general conclusion is that there will be considerable volumes of real estatecontinuing to change hands over the foreseeable future and there will be substan-tial requirements for associated funding. We believe that new CMBS issuance iswell-placed to play a major part in that process.

Transforming Real Estate Finance

European CMBS

156

chapter 11

Page 159: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 157

CMBS FeatureLarge size

Long term debt

Tranched bondissuance

Rated bondissuance

Mix of fixed/floating rate coupons

Varying maturity profilesMix of currency

tranchesMultiple real

estate assets

Ability to substitute/sell

assetsCashflow driven

financial covenants and real estate

adviser provisions Synthetic/credit

derivative structures

Yield pick-up compared to

comparably-ratedsecurities

THE ATTRACTIVENESS OF CMBSTO ISSUERS AND INVESTORS

exhibit 23

Source: Morgan Stanley

Issuer AttractionsAbility to raise funds in highervolumes than syndicated bankmarket or local debenture/bondmarkets; deal with single under-writer/small group and not largebank syndicate

Available maturities generallylonger than bank lenders willabsorbLowers average cost of funding

Assists marketing via transparency

Match underlying cashflows tobond coupons (with swapswhere applicable)Matches cashflow from underlying assetsCan ensure optimum distribution

Large scale execution

Flexibility in managing portfolios

N/A

Suitable for issuers primarilyseeking risk reduction ratherthan funding N/A

Investor AttractionsLiquidity of issue likely to beenhanced for larger transactions

Matching of liabilities (for insurancecompanies and pension funds)

Bonds available with risk/return pro-file according to investor preferences Clear risk differentiation/shouldenable easier surveillanceAccommodates differing couponrequirements

Accommodates differing maturityrequirementsAccommodates regional investorrequirementsImproves risk profile of exposureacross a number of real estateassets and tenantsSubstitution/repayment provisionstightly drawn to protect originalcredit characteristicsExercise strong control via debtservice triggers, reserve funds and real estate adviser control provisions Widens investor choices

Improved risk/reward balance foran asset secured position com-pared to equivalent rated securities

Page 160: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

European CMBS

158

chapter 11

PREVAIL ING FUNDING CONDIT IONS AND THE ROLE OF CMBS

Whilst the relative share of real estate in the asset base of key investors may havedeclined over a 10 year period, there is of course no necessary suggestion thatthe supply of investment funds for real estate are showing an absolute shortage.

However, we believe that some of the trends indicated in the investment andfunding market for real estate may assist both potential issuers of, and investorsin, commercial mortgage-backed securities.

In general terms, we believe that with declining inflation trends the attractivenessof real estate as an equity-type investment may have declined. However, we alsobelieve that carefully structured real estate assets can make very attractive, yield-enhanced, debt investments.

Although CMBS in the U.K. and Europe have exhibited a number of differentstructures, we have set out below some of the typical features of the asset classthat enable us to conclude that there are a number of attractions to both issuersand investors that bode well for future market growth.

Clearly, different CMBS transactions will vary in terms of their characteristicsaccording to the grid below but many of these characteristics will be present inindividual 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 favorableagainst relevant comparables and this yield pick-up largely holds true forlower-rated tranches. To some extent this may be due to CMBS being a rela-tively small market, with associated historically limited liquidity, as well as rela-tive unfamiliarity on the part of investors with products and structures. We alsobelieve that transactions that have been completed to date, particularly thelarge transactions that have been featured over the last 2-3 years, haveinvolved very strong assets and structures. Accordingly, European CMBS is asector whose relative merits we currently recommend to investors.

• Underlying real estate market conditions in Europe are currently in robusthealth. We believe that the balance of probabilities is that these markets canremain in good health. Whilst we do not underestimate the risks arising from aslowdown being imported from the US, we believe that the Europeaneconomies have a reasonable chance of avoiding a steep recession.

• In this context, when compared to the US, the European economies are lessaffected by changes to consumer behavior arising from stock market condi-tions, there has been less of a high technology investment surge (and thereforeno risk of a rapid slowdown) and Eurozone exports to the US only account for2.5% of total GDP.

Page 161: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 159

• We believe that, against this broad economic background, the more circumspectapproach of both real estate developers and providers of funding to the realestate sector compared to the last boom should cause the current expansion inthe 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 playa key part in this process.

RATING AGENCY U.K. CMBS MULTI -BORROWER MODEL

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 ofcommercial mortgage loans. The analysis simulates lease termination and tenantdefault in order to assess stressed debt service coverage on a loan by loan basis.The basic analysis can equally be applied to transactions featuring commercialreal 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 butdetail on historic cashflows, ideally audited data for 3 years.

• Loan collateral is broken down by individual lease and projected forward overthe life of the underlying loan.

• When leases are timed to terminate, or a break option exists, void periods areassumed based on the perception of the quality of the property. Historicallyvery low vacancy rates and void periods may not be sustainable going forwarddepending on the nature and quality of the real estate asset and the market inwhich it operates. Stress runs are conducted depending on the rating levelsought. For example, significantly higher proportions of tenants are assumed tovacate 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 stressrun being carried out. Assumed declines in rental value in U.K. transactions arebased on historical reference points that may be obtained, for example, fromthe Investment Property Dealer’s databank where the 1990-93 recession hasbeen identified as an ‘A’ class recession in the U.K. real estate markets.Refurbishment needs are assumed, varying with the property. BecauseEuropean leases are generally shorter than U.K. equivalents, the cashflow dis-ruption from lease rollover will be more significant than for U.K. leases.

Page 162: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• In addition to tenant rollover, tenants are also randomly defaulted according toa corporate default matrix, where the probability of default is determined bythe credit rating of the tenant and the stress run being carried out. The defaultrate for tenants will be a function of tenant credit rating and also of the periodof reliance on each tenant under a specific lease. Void periods and re-lettingcosts and fees are assumed for each defaulting tenant.

• There are important differences between, for example, U.K. fully repairing andinsuring leases, where it is the obligation of the tenant to repair and refurbishthe property, and leases where it is the responsibility of the landlord to effectrepair and refurbishment. In situations where the landlord is responsible, therating analysis must further take into account the ability of the real estate asset’scashflows to meet refurbishment, repair and other operating costs.

• The result of this analysis is a set of underwritten cashflows which can be com-pared with projected interest and scheduled principal payments of each loan.Where the stress run shows a DSCR dropping below 1.0x for an extended peri-od then it is assumed that the loan will default. It is also assumed that sched-uled principal payments take place where there is sufficient stressed excesscashflow to do so, otherwise the loan defaults. The DSCR default test will beadjusted where there is direct recourse to an investment grade borrower.

• If a loan default occurs, then loss severity is calculated by reference to the dif-ference between the outstanding principal at the point of default and thestressed recovery value from the properties. Recovery values are adjusted forsales costs and delays in refinance or sale.

• Although the analysis described here is fundamentally cashflow-driven, propertyvaluations play a significant role in the analysis. When loans default,realizations of underlying real estate securing the loan are used to reduce oreliminate loan losses. For this reason the rating agencies analyze originalvaluations to create stabilized initial valuations, against which market valuedeclines will be applied when assessing the realization value of collateralfollowing a loan default. Rating agency stabilized valuations may be lower thanthe external valuations if, for example, the rating agency believes the rentalflows underpinning the external valuation may be difficult to sustain, possiblyvia excessive reliance on one or more credit tenants with impending lease roll-offs or alternatively if rental levels are regarded as above-market.

Transforming Real Estate Finance

European CMBS

160

chapter 11

Page 163: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• If real estate assets need to be foreclosed on, then the rating agency analysiswill use longer foreclosure periods for higher desired ratings levels.

• A large number of data runs are conducted for each stress level and an averagelevel of credit enhancement required together with standard deviation at both apool 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 origina-tion, environmental and disaster issues and excess spread (where applicable).

• Maturity profiles are also considered, especially in relation to the tail where asingle outstanding loan will concentrate the risk in relation to notes then out-standing.

• Using this loan by loan analysis, the rating agency is able to generate requiredcredit enhancement for each class of rated notes which is sensitive to likelytiming of default.

• In addition to the financial modeling, rating agencies will perform additionaldue diligence in terms of the review of third party valuations, property visits,review of the loan servicer and legal and property title due diligence.

Please refer to important disclosures at the end of this report. 161

Page 164: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Transaction Monitoring

162

chapter 12

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 reviewof the deal and either upgrades, downgrades or affirms the current rating. As partof the annual review, the rating agencies examine the current LTV ratios, DSCR,prepayments, and status of specially services loans within each transaction. Overthe past three years upgrades on CMBS transaction have significantly outnum-bered downgrades. During 2001, CMBS upgrades outnumber downgrades 15 toone (see chapter 2).

On a monthly basis, the trustees publish remittance and special servicing reportsthat detail payments made to bondholders, delinquent loan information, and spe-cially serviced loan details.

In order to assist investors with monitoring their CMBS investments, we publishmonthly CMBS delinquency data and quarterly CMBS tracking reports, which out-line loan level details. The first part of this chapter is an example of our monthlyCMBS delinquency report. The report outlines overall delinquency trends in theCMBS universe and provides trends by property type and origination year.

The second part of this chapter is an example of our quarterly CMBS trackingreport, which provides details on specially serviced loans within Morgan Stanleysponsored 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 rateof increase has moderated since December, when delinquencies rose by 16 bp.Based on February remittance reports, CMBS delinquencies on seasoned transac-tions (aged over 1 year) increased 8 bp to 2.09% of current balances. Seasoneddelinquencies 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 next18 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.

Page 165: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 163

ORIGINATION YEAR

Of the cohorts with current balances greater than $10 billion, the 1996 cohort hadthe highest delinquency rate increase for the month (16 bp), and the 1997 cohorthad the highest delinquency rate (2.62%).

WHAT L IES ABOVE

Senior housing and hotels are the two property types with delinquency rates thatare above the universe average of 1.30%. Retail delinquencies have remainedunder the universe average since November. Senior housing properties continuedto post the highest delinquency rate (6.30%) this month, while hotels posted thelargest delinquency rate increase (42 bp) of the property types.

Source: Morgan Stanley, Intex

Year

CurrentBalance

($bn)30/60/90+days (%)

Forc. &REO (%) Total ( %)

ChangeFrom LastMonth (%)

1990 0. 3 0.75 1.46 2.20 -1. 181991 0. 2 0.04 0.00 0.04 -0. 011992 0. 3 0.00 0.00 0.00 -0. 071993 0. 9 0.40 0.30 0.70 -0. 551994 2. 1 1.95 2.01 3.96 0.861995 6. 0 3.08 1.18 4.26 0.721996 14. 4 1.96 0.53 2.49 0.161997 37. 4 1.93 0.70 2.62 0.151998 65. 6 0.92 0.31 1.23 -0. 081999 38. 1 0.78 0.46 1.23 0.122000 39. 4 0.47 0.05 0.53 0.122001 42. 5 0.06 0.00 0.06 -0. 08

T otal/Avera ge 247. 1 0.95 0.35 1.30 0.07

CMBS DELINQUENCIES BY YEAR OF ORIGINATION(IN %) (AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 2

Source: Morgan Stanley, Intex

0.0

0.5

1.0

1.5

2.0

2.5

Dec-96 Mar-98 Jul-99 Oct-00 Feb-02

Total Delinquency4 Y ear Average

%

DELINQUENCIESIN SEASONED

CMBS DEALS AND4-YEAR AVERAGE

exhibit 1

Page 166: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

PENNSYLVANIA DELINQUENCIES T ICK UP

Although Florida properties continue to post the highest delinquency rate,Pennsylvania properties had the greatest delinquency rate increase. Pennsylvaniadelinquencies rose 53 bp to 1.62% in January. Ninety-day delinquencies com-prised 95 bp of Pennsylvania’s delinquencies, 30-day delinquencies accounted foranother 60 bp of delinquencies, and foreclosures accounted for 7 bp. This month,about half of the new 30-day Pennsylvania delinquencies were attributable tohotel properties.

Transforming Real Estate Finance

Transaction Monitoring

164

chapter 12

CMBS DELINQUENCIES BY PROPERTY TYPE (IN%) (AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 3

Source: Morgan Stanley, Intex

Year

CurrentBalance

($bn)30/60/90+days (%)

Forc. &REO (%) Total ( %)

ChangeFrom LastMonth (%)

Hotel-Motel 25.3 3.8 8 0.8 5 4.7 3 0.4 2Industrial-W arehouse 13. 5 0.76 0.11 0.87 -0. 11Mixed 11.4 0.4 2 0.1 4 0.5 5 -0.11Mobile Home 4.1 0.4 8 0.0 3 0.5 1 0.1 1Multifamily 47.7 0.4 9 0.2 6 0.7 5 -0.03Office 54.9 0.2 9 0.1 7 0.4 6 0.0 7Retail 72.6 0.8 4 0.4 1 1.2 4 0.0 3Self-Storage 2.9 0.1 1 0.0 3 0.1 5 -0.04Senior Housing 4.5 4. 19 2. 11 6. 30 0. 28Warehouse 0. 8 0.00 0.00 0.00 0.00Other 9.3 0. 00 0. 00 0. 00 0. 00

Total/Average 247.1 0.95 0.35 1.30 0.07

CMBS DELINQUENCIES BY PROPERTY TYPE AND YEAR (IN %)(AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 5

Source: Commercial Mortgage Alert

Product Type 1990 1991 1992 1993 1994

Hotel-Motel 0.00 N/A 0. 00 0. 00 4. 55Industrial-Warehouse 0. 00 0. 00 0. 00 2. 95 18. 70Mixed 0.0 0 7.4 7 0.0 0 1.2 9 0.0 0Mobile Home N/A N/A 0. 00 0. 00 0. 00Multifamily 0. 00 0. 00 0. 00 1. 34 0. 23Office 1. 88 0. 00 0. 00 1. 37 0. 04Retail 6. 06 0. 00 0. 00 0. 00 7. 60Self-Storage N/A N/A N/A 0.0 0 0.0 0Senior Housing N/A 0. 00 0. 00 5. 56 6. 65Warehouse 0. 00 0. 00 0. 00 0. 00 0. 00Other 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0

Page 167: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

ACLI DEL INQUENCIES DECLINE IN 4Q2001

This week, the American Council of Life Insurers (ACLI) reported that delinquen-cy rates on commercial mortgages originated by the life insurance industry fell to0.12%, as of December 31, 2001. The previous mark was 0.19%, at the end ofSeptember 2001.

Although delinquencies on industrial and hotel/motel properties increased duringthe fourth quarter, delinquencies on the other major property types declined. Theeconomic downturn has not yet translated into an overall higher delinquency rate,as ACLI delinquencies typically have a 9-month lag in reflecting the effects ofweak economic conditions. ACLI expects commercial mortgage delinquencies torise in 2002.

Part of the fall in the ACLI delinquency rate is also attributable to an increasednumber of foreclosures in 2001, as compared to 2000. In 2001, $519 million ofloans were foreclosed, while $437 million of foreclosures occurred in 2000.

Please refer to important disclosures at the end of this report. 165

StateCurrent Balance

($bn) % T otal Delinq.Change From

Last Month

C A 38.9 0.2 9 0.0 3NY 24.9 0.7 9 0.1 0T X 17.1 1.9 4 0.0 6F L 13.5 3.6 0 0.1 7NJ 8.1 0.8 7 0.0 3IL 7.3 1.3 6 -0.46VA 7. 3 0.91 -0. 08P A 6.4 1.6 2 0.5 3GA 6. 3 1.94 -0. 46MA 5.9 0.4 5 0.2 7

Total/Average 135.8 1.19 0.04

CMBS DELINQUENCIES BY STATE (IN %)(AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 4

1995 1996 1997 1998 1999 2000 2001

12. 94 10. 25 9. 91 4. 22 2. 14 2. 06 0. 390. 00 0. 94 1. 36 1. 17 0. 92 0. 14 0. 090.0 0 1.9 1 1.9 1 0.6 1 0.4 1 0.3 9 0.0 02. 41 1. 43 0. 58 0. 21 1. 33 0. 00 0. 000. 49 0. 71 0. 96 0. 42 1. 21 1. 53 0. 131. 93 0. 43 0. 91 0. 85 0. 54 0. 18 0. 015. 64 2. 17 2. 04 0. 90 1. 69 0. 18 0. 040.0 0 0.2 1 0.4 5 0.0 0 0.0 0 0.5 0 0.0 0

12. 96 4. 10 12. 31 4. 93 4. 70 0. 36 0. 000. 00 0. 00 N/A 0.00 0.00 N/A N/A0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0

Source: Morgan Stanley, Intex

Page 168: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

166

chapter 12

SEASONED MULTI-BORROWER CMBS DELINQUENCIES (%)(AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 6

IssueOriginal

Balance ($mm)

Aetna Commercial Mortgage Trust, 1995-C5 443.3

American Southwest Financial Securities Corp., 1995-C1 294.1

AMRESCO Commercial Funding I Corp., 1997-C1 511.5

Asset Securitization Corporation, 1995-MD4 968.3Asset Securitization Corporation, 1996-D2 (Nomura) 879.5Asset Securitization Corporation, 1996-D3 784.2Asset Securitization Corporation, 1996-MDVI 896.3Asset Securitization Corporation, 1997-D4 1437.1Asset Securitization Corporation, 1997-D5 1788.3Asset Securitization Corporation, 1997-MDVII 502.6Nomura Asset Securities Corp., 1994-MD1 410.7Nomura Asset Securities Corp., 1995-MD3 536.7Nomura Asset Securities Corp., 1996-MD5 774.0Nomura Asset Securities Corp., 1998-D6 3722.7Capco America Securitization Corporation, 1998-D7 1249.0Commercial Mortgage Asset Trust, 1999-C1 2375.0Commercial Mortgage Asset Trust, 1999-C2 775.2

Banc of America Commercial Mortgage, 2000-1 795.2Banc of America Commercial Mortgage, 2000-2 896.5

Bear Stearns Commerical Mortgage Securities Inc., 1998-C1 716.1Bear Stearns Commerical Mortgage Securities Inc., 1999-C1 479.6Bear Stearns Commerical Mortgage Securities Inc., 1999-WF2 1088.0Bear Stearns Commerical Mortgage Securities Inc., 2000-WF1 892.6Bear Stearns Commerical Mortgage Securities Inc., 2000-WF2 842.8

CBA Mortgage Corp., 1993-C1 541.5

Chase Commercial Mortgage Securities Corp, 1996-1 445.4Chase Commercial Mortgage Securities Corp, 1996-2 263.2Chase Commercial Mortgage Securities Corp, 1997-1 540.6Chase Commercial Mortgage Securities Corp, 1997-2 817.8Chase Commercial Mortgage Securities Corp, 1998-1 820.2Chase Commercial Mortgage Securities Corp, 1998-2 1270.8Chase Commercial Mortgage Securities Corp, 1999-2 783.4Chase Commercial Mortgage Securities Corp, 2000-1 698.3Chase Commercial Mortgage Securities Corp, 2000-2 739.6Chase Commercial Mortgage Securities Corp, 2000-3 769.4

Chase Manhattan Bank - First Union National Bank, 1999-1 1402.9

Commercial Mortgage Acceptance Corporation, 1997-ML1 850.9Commercial Mortgage Acceptance Corporation, 1998-C1 1192.2Commercial Mortgage Acceptance Corporation, 1998-C2 2900.1Commercial Mortgage Acceptance Corporation, 1999-C1 736.5

CS First Boston Mortgage Securities, 1995-WF1 245.0

CS First Boston Mortgage Securities Corp., 1997-C1 1363.5CS First Boston Mortgage Securities Corp., 1997-C2 1468.1CS First Boston Mortgage Securities Corp., 1998-C1 2488.1CS First Boston Mortgage Securities Corp., 1998-C2 1923.1

Transforming Real Estate Finance

Transaction Monitoring

Note: Shaded deals have a total delinquency rate of more than 2%. New deals added this month are in bold.

Page 169: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 167

CurrentBalance ($mm) Factor 30, 60 & 90+ Forc. & REO Total

Change fromLast Month

129.3 0.29 0.00 0.00 0.00 0.00

109.3 0.37 4.08 1.96 6.04 3.14

401.0 0.78 0.00 0.00 0.00 0.00

746.8 0.77 0.00 0.00 0.00 0.00779.8 0.89 9.26 2.25 11.50 -5.03709.6 0.90 2.21 1.09 3.29 -0.41831.3 0.93 0.00 0.00 0.00 0.00

1297.1 0.90 0.00 0.45 0.45 0.001676.8 0.94 2.92 0.82 3.75 0.00

452.2 0.90 0.00 0.00 0.00 0.00107.9 0.26 16.69 29.36 46.04 12.43400.7 0.75 0.00 0.00 0.00 0.00723.0 0.93 0.00 0.00 0.00 0.00

3552.4 0.95 0.28 0.00 0.28 0.131178.6 0.94 0.26 1.02 1.28 0.002313.9 0.97 0.13 0.23 0.36 0.00756.9 0.98 0.00 0.63 0.63 0.63

728.9 0.92 0.99 0.00 0.99 0.28877.2 0.98 0.63 0.00 0.63 0.17

676.7 0.94 0.15 0.00 0.15 0.00457.7 0.95 0.00 1.12 1.12 0.00

1035.7 0.95 0.26 0.53 0.79 0.00867.6 0.97 0.45 0.00 0.45 0.00823.4 0.98 0.00 0.00 0.00 0.00

65.7 0.12 0.00 0.00 0.00 0.00

328.4 0.74 1.67 1.99 3.67 -2.17216.4 0.82 1.09 1.46 2.55 0.00433.5 0.80 0.43 0.00 0.43 0.00679.0 0.83 1.13 0.00 1.13 0.62748.3 0.91 0.00 0.00 0.00 0.00

1223.9 0.96 0.00 0.00 0.00 0.00768.1 0.98 0.00 0.00 0.00 0.00687.5 0.98 1.85 0.00 1.85 0.00730.6 0.99 0.81 0.00 0.81 0.00759.4 0.99 0.00 0.00 0.00 0.00

1366.6 0.97 1.01 0.00 1.01 0.18

799.3 0.94 9.68 0.00 9.68 -0.011071.8 0.90 0.77 0.00 0.77 -0.192733.4 0.94 3.41 0.42 3.82 1.52

709.8 0.96 0.92 0.00 0.92 -0.02

88.4 0.36 0.00 0.00 0.00 0.00

1229.8 0.90 0.96 5.16 6.12 0.001393.9 0.95 0.54 1.74 2.28 -0.912350.6 0.94 1.52 1.44 2.97 -0.131846.7 0.96 0.48 0.17 0.65 0.48

Source: Morgan Stanley, Intex

Page 170: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

168

chapter 12

SEASONED MULTI-BORROWER CMBS DELINQUENCIES (%)(AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 6 (cont.)

IssueOriginal

Balance ($ mm)

CS First Boston Mortgage Securities Corp., 1999-C 1 1175 .9 2000-C 1 1118 .5

1997-SPICE 352. 3

Deuts che Mortgage & Asset Receiving Corp., 1998 -C1 1821 .8C OMM 1999-1 1318 .8C OMM 2000-C1 908. 4

DLJ Mortgage Acceptance Corp., 1994 -MF11 210. 4 1995 -CF2 510. 2 1996 -CF1 472. 1 1996 -CF2 510. 2 1997 -CF1 449. 0 1997 -CF2 663. 1 1998 -CF1 841. 1 1998 -CG1 1567 .8 1998 -CF2 1111 .1 1999 -CG1 1243 .2 1999 -CG2 1554 .0 1999 -CG3 901. 2 2000 -CF1 887. 6

2000-CKP1 1294 .5

First Union-Lehman 199 7-C 1 1308 .8 199 7-C 2 2209 .5

Bank of America 1998 -C 2 3417 .4

First Union Commercial Mortgage Trust, 1999-C1 1171 .2 1999-C4 890. 0 2000-C1 779. 9 2000-C2 1149 .4

First Union Chase Commercial Mortgage Trust, 1999-C2 1188 .3

GMAC Commercial Mortgage Securities Inc., 1996-C1 459. 8 1997-C1 1716 .1 1997-C2 1071 .9 1998-C1 1444 .1 1998-C2 2530 .6 1999-C1 1338 .3 1999-C2 976. 5 1999-C3 1154 .8 2000-C1 882. 3 2000-C2 775. 9 2000-C3 1320 .0

GS Mortgage Securities Corp. II, 1996-PL 554. 11998-C1 1868 .81999-C1 897. 5

Heller Financial Commercial Mortgage Asset Corp., 1999 -PH1 1016 .5 2000 -PH1 960. 7

Impac Commercial Holdings, Inc., 1998-C1 320. 2ICCMAC Multifamily and Commercial Trust, 1999-1 297. 4

CS First Boston Mortgage Securities Corp.,

CS First Boston Mortgage Securities 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

First Union-Lehman

First Union Commercial Mortgage Trust,First Union Commercial Mortgage Trust,First Union Commercial Mortgage Trust,

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.,

GS Mortgage Securities Corp. II, GS Mortgage Securities Corp. II,

Heller Financial Commercial Mortgage Asset Corp.,

Transforming Real Estate Finance

Transaction Monitoring

Note: Shaded deals have a total delinquency rate of more than 2%. New deals added this month are in bold.

Page 171: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 169

CurrentBalance ($ mm) Factor 30, 60 & 90+ Forc. & REO Total

Change fromLast Month

1137 .0 0. 97 2. 57 0. 18 2. 75 -0. 131097 .0 0. 98 0. 33 0. 00 0. 33 0. 33

155. 0 0. 44 3. 22 0. 00 3. 22 0. 04

1716 .6 0. 94 3. 23 1. 71 4. 94 0. 001266 .7 0. 96 0. 83 0. 00 0. 83 0. 00

886. 3 0.98 0.48 0.00 0.48 0.00

44. 7 0.21 0.00 0.00 0.00 0.00398. 7 0. 78 19. 53 1. 12 20. 66 9. 70395. 5 0. 84 5. 57 0. 00 5. 57 0. 00403. 4 0. 79 0. 90 1. 60 2. 50 0. 00384. 5 0. 86 0. 90 6. 77 7. 67 -0. 01611. 4 0. 92 11. 02 0. 00 11. 02 0. 68790. 3 0.94 0.90 0.00 0.90 0.00

1465 .9 0. 94 1. 47 0. 00 1. 47 0. 691052 .8 0. 95 3. 23 0. 00 3. 23 0. 761200 .9 0. 97 1. 39 0. 49 1. 88 0. 001508 .9 0. 97 0. 35 0. 04 0. 39 -0. 22

880. 7 0.98 0.73 0.00 0.73 0.30876. 5 0.99 0.89 0.00 0.89 0.31

1276 .4 0. 99 0. 00 0. 09 0. 09 0. 04

1135 .8 0. 87 5. 14 0. 67 5. 80 2. 452018 .2 0. 91 3. 22 0. 00 3. 22 1. 20

3232 .8 0. 95 1. 57 0. 07 1. 64 0. 46

1112 .9 0. 95 1. 72 0. 00 1. 72 0. 00870. 5 0.98 1.20 0.00 1.20 0.00764. 1 0.98 0.00 0.56 0.56 -0. 31

1130 .8 0. 98 1. 43 0. 00 1. 43 0. 00

1142 .7 0. 96 1. 39 0. 23 1. 62 -0. 25

231. 6 0. 50 2. 05 0. 00 2. 05 0. 821410 .1 0. 82 3. 51 0. 00 3. 51 0. 591005 .0 0. 94 1. 87 0. 00 1. 87 -0. 521305 .2 0. 90 0. 95 0. 65 1. 60 0. 002353 .9 0. 93 1. 51 0. 00 1. 51 0. 411279 .1 0. 96 1. 36 1. 65 3. 00 0. 42

946. 9 0. 97 3. 59 0. 00 3. 59 0. 001127 .1 0. 98 0. 99 0. 15 1. 14 0. 00

867. 4 0.98 0.70 0.00 0.70 -5. 68765. 2 0. 99 2. 01 0. 00 2. 01 0. 67

1306 .7 0. 99 2. 19 0. 00 2. 19 1. 46

271. 9 0.49 0.00 0.00 0.00 0.001777 .3 0. 95 4. 36 0. 00 4. 36 -0. 21

846. 9 0. 94 3. 37 0. 35 3. 72 -3. 30971. 7 0.96 0.32 0.00 0.32 0.20940. 7 0.98 1.75 0.24 1.99 -0. 05

274. 8 0.86 1.04 0.23 1.27 0.00150. 7 0.51 1.13 0.00 1.13 -0. 73

Source: Morgan Stanley, Intex

Page 172: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

170

chapter 12

SEASONED MULTI-BORROWER CMBS DELINQUENCIES (%)(AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 6 (cont.)

IssueOriginal

Balance ($ mm)

JP Morgan Commercial Mortgage Finance Corp., 19 96-C 2 354. 3 19 96-C 3 451. 3 19 97-C 5 1039 .5 19 98-C 6 800. 9 19 99-C 7 808. 8 19 99-C 8 735. 9 20 00-C 9 817. 5 20 00-C10 745. 4

LB Commercial Mortgage Trust 1995-C2 261. 91996-C2 399. 51998-C1 1731 .11998-C4 2028 .51999-C1 1584 .91999-C2 893. 5

LB-UBS Commercial Mortgage Trust, 2000-C3 1305 .72000-C4 1001 .62000-C5 998. 9

Merrill Lynch Mortgage I nves tors , 199 4-C 1 310. 1Merrill Lynch Mortgage I nves tors , 199 4-M1* 405. 0Merrill Lynch Mortgage I nves tors , 199 5-C 1 211. 6Merrill Lynch Mortgage I nves tors , 199 5-C 2 1075 .7Merrill Lynch Mortgage I nves tors , 199 5-C 3 644. 7Merrill Lynch Mortgage I nves tors , 199 6-C 1 648. 3Merrill Lynch Mortgage I nves tors , 199 6-C 2 1170 .4Merrill Lynch Mortgage I nves tors , 199 7-C 1 844. 5Merrill Lynch Mortgage I nves tors , 199 7-C 2 687. 3Merrill Lynch Mortgage I nves tors , 199 8-C 2 1093 .8Merrill Lynch Mortgage I nves tors , 199 8-C 1-C T L 751. 0Merrill Lynch Mortgage I nves tors , 199 8-C 3 640. 8Merrill Lynch Mortgage I nves tors , 199 9-C 1 594. 4

Midland Realty Acceptance Corp., 1996-C1 373. 51996-C2 513. 8

Morgan Stanley Capital I Inc., 1995-GAL1 272. 3

340. 5640. 7

1107 .3

867. 1

632. 1

594. 0689. 0

Morgan Stanley Dean Witter Capital I Inc., 2000-LIFE2 776. 8

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 TrustLB Commercial Mortgage Trust

LB Commercial Mortgage TrustLB Commercial Mortgage Trust

LB-UBS Commercial Mortgage Trust,LB-UBS Commercial Mortgage Trust,

Midland Realty Acceptance Corp.,

Morgan Stanley Capital I Inc., 1996-C1Morgan 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-LIFE1Morgan Stanley Capital I Inc., 2000-LIFE1

Transforming Real Estate Finance

Transaction Monitoring

Note: Shaded deals have a total delinquency rate of more than 2%. New deals added this month are in bold.

Page 173: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 171

CurrentBalance ($ mm) Factor 30, 60 & 90+ Forc. & REO Total

Change fromLast Month

164. 6 0. 46 4. 21 0. 00 4. 21 0. 00268. 3 0.59 0.00 0.00 0.00 0.00834. 3 0. 80 2. 48 0. 87 3. 35 0. 10743. 5 0.93 1.50 0.00 1.50 0.52765. 0 0.95 0.00 0.00 0.00 0.00708. 9 0. 96 4. 99 0. 30 5. 29 0. 29788. 7 0. 96 4. 99 0. 00 4. 99 -0. 21731. 1 0.98 0.51 0.00 0.51 -1. 15

141. 1 0. 54 6. 51 0. 00 6. 51 4. 85326. 6 0. 82 11. 66 0. 00 11. 66 7. 25

1615 .2 0. 93 2. 38 0. 00 2. 38 0. 001951 .6 0. 96 0. 00 0. 00 0. 00 0. 001532 .7 0. 97 0. 36 0. 49 0. 84 0. 00

874. 0 0.98 0.00 0.00 0.00 0.00

1289 .6 0. 99 0. 04 0. 00 0. 04 -0. 05988. 7 0.99 0.00 0.00 0.00 -0. 08974. 1 0.98 0.00 0.00 0.00 0.00

34. 8 0.11 0.00 0.00 0.00 0.0050. 5 0.12 0.00 0.00 0.00 -1. 9854. 1 0.26 0.00 0.00 0.00 0.00

214. 0 0. 20 0. 00 19. 63 19. 63 -3. 04445. 8 0. 69 4. 33 0. 84 5. 17 2. 55444. 2 0.69 0.00 1.95 1.95 -5. 69872. 2 0. 75 8. 62 2. 20 10. 81 0. 09701. 3 0. 83 8. 44 0. 96 9. 40 1. 43645. 3 0.94 1.23 0.46 1.69 0.21994. 3 0. 91 1. 34 1. 20 2. 54 0. 31598. 5 0. 80 1. 78 0. 38 2. 16 -0. 75595. 5 0. 93 0. 00 4. 27 4. 27 -2. 05571. 2 0. 96 3. 94 2. 62 6. 56 0. 14

247. 8 0. 66 2. 76 2. 53 5. 29 0. 00384. 8 0.75 1.34 0.00 1.34 0.00

119. 5 0. 44 0. 00 6. 49 6. 49 0. 01

216. 3 0. 64 2. 76 0. 00 2. 76 0. 00488. 1 0. 76 2. 33 0. 00 2. 33 0. 00

1030 .3 0. 93 4. 02 3. 65 7. 68 -0. 05

805. 9 0.93 0.42 0.00 0.42 0.00

610. 6 0.97 0.36 1.31 1.67 0.22

580. 7 0.98 0.00 0.00 0.00 0.00676. 6 0.98 0.00 0.00 0.00 0.00754. 2 0.97 0.00 0.00 0.00 0.00

Source: Morgan Stanley, Intex

Page 174: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

172

chapter 12

SEASONED MULTI-BORROWER CMBS DELINQUENCIES (%)(AS OF FEBRUARY REMITTANCE REPORTS)

exhibit 6 (cont.)

IssueOriginal

Balance ($ mm)

Morgan Stanley Capital I Inc., 1996-W F1 605. 4 1997-W F1 559. 2 1998-W F1 1392 .2 1998-W F2 1062 .0 1999-W F1 968. 5

1997-H F 1 622. 4 1998-H F 1 1283 .7 1998-H F 2 1066 .3

Mortgage Capital Funding Inc., 19 95-M C 1 231. 4 19 96-M C 1 490. 9 19 96-M C 2 460. 3 19 97-M C 1 660. 4 19 97-M C 2 872. 3 19 98-M C 1 1298 .7 19 98-M C 2 1012 .5 19 98-M C 3 915. 0

Nations Link Funding Corp. 1996-1 324. 2 1998-1 1024 .1 1998-2 1592 .1 1999-1 1228 .4 1999-2 1167 .6

Paine Webber Mortgage Acceptance Corp. V, 1999 -C1 700. 5

Penn Mutual Life Insurance Co. Series, 1996-PML 781. 6

PNC Mortgage Acceptance Corp., 1999-CM1 763. 4 2000-C1 808. 6 2000-C2 1083 .3

Prudential Securities Financing Corp., 1995-MCF 2 222. 8 1998-C1 1154 .6

1999-NRF1 934. 5 1999-C2 875. 1KEY 2000-C1 821. 2

Salomon Brothers Mortgage Securities VII Inc., 1996-C 1 213. 2 1999-C 1 744. 7 2000-N L1 340. 0 2000-C 1 719. 9 2000-C 2 787. 1 2000-C 3 923. 0

Southern Pacific Thrift & Loan Association 1996-C 1 277. 0

Structured Assets Securities Corp., 1994-C 1 454. 7 1995-C 1 525. 0 1995-C 4 234. 8 1996-C 3 280. 7

Total/Weighted A verage $161,258.2

Morgan Stanley Capital I Inc.,

Morgan Stanley Capital I Inc.,Morgan Stanley Capital I Inc.,Morgan Stanley Capital I Inc.,

Morgan Stanley Capital I Inc.,

Morgan Stanley Capital I Inc.,Morgan Stanley Capital I Inc.,

Mortgage Capital Funding Inc.,Mortgage Capital Funding Inc.,

Mortgage Capital Funding Inc.,Mortgage Capital Funding Inc.,Mortgage Capital Funding Inc.,Mortgage Capital Funding Inc.,

Mortgage Capital Funding Inc.,

Nations Link Funding Corp.Nations Link Funding Corp.Nations Link Funding Corp.

Nations Link Funding Corp.

PNC Mortgage Acceptance Corp.,PNC Mortgage Acceptance Corp.,

Prudential Securities Financing Corp.,Prudential Securities Secured Financing Corp.,Prudential Securities Secured Financing Corp.,Prudential Securities Secured Financing Corp.,

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.,

Structured Assets Securities Corp.,

Structured Assets Securities Corp.,Structured Assets Securities Corp.,

Transforming Real Estate Finance

Transaction Monitoring

Note: Shaded deals have a total delinquency rate of more than 2%. New deals added this month are in bold.

Page 175: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 173

CurrentBalance ($ mm) Factor 30, 60 & 90+ Forc. & REO Total

Change fromLast Month

484. 2 0.80 0.42 0.33 0.75 0.00465. 5 0.83 0.00 1.71 1.71 0.00

1282 .5 0. 92 0. 74 0. 00 0. 74 0. 001002 .2 0. 94 0. 00 0. 00 0. 00 0. 00

919. 5 0.95 0.52 0.00 0.52 0.00

476. 8 0.77 0.00 0.00 0.00 -0. 221173 .5 0. 91 1. 40 0. 16 1. 56 -0. 221009 .2 0. 95 0. 90 0. 00 0. 90 -2. 30

56. 9 0.25 0.00 0.00 0.00 0.00354. 2 0.72 0.29 0.00 0.29 0.29364. 3 0. 79 6. 05 0. 00 6. 05 3. 92589. 2 0. 89 3. 67 0. 91 4. 58 0. 02799. 4 0.92 0.17 0.29 0.46 -0. 90

1224 .0 0. 94 0. 54 0. 95 1. 48 0. 20940. 6 0.93 1.80 0.00 1.80 -0. 53853. 4 0.93 1.36 0.00 1.36 -0. 25

171. 3 0. 53 6. 12 0. 00 6. 12 3. 80908. 6 0. 89 3. 80 0. 17 3. 97 0. 07

1487 .3 0. 93 1. 45 0. 54 1. 99 1. 061177 .5 0. 96 0. 00 0. 36 0. 36 0. 00

968. 2 0.83 0.00 0.00 0.00 0.00

663. 0 0. 95 2. 04 0. 42 2. 46 0. 19

286. 2 0.37 0.00 1.22 1.22 0.01

745. 4 0.98 1.03 0.00 1.03 0.00786. 9 0.97 1.25 0.00 1.25 0.00

1063 .8 0. 98 0. 38 0. 00 0. 38 -0. 04

119. 6 0. 54 5. 67 6. 96 12. 63 0. 011044 .8 0. 90 0. 50 0. 58 1. 08 0. 00

874. 7 0.94 0.00 0.25 0.25 0.00835. 4 0.95 0.60 0.32 0.93 0.00804. 1 0.98 0.00 1.18 1.18 0.00

94. 3 0. 44 21. 78 6. 63 28. 41 -0. 03707. 3 0. 95 2. 02 0. 00 2. 02 0. 66296. 9 0.87 1.21 0.00 1.21 0.00700. 3 0.97 0.40 0.00 0.40 -0. 19771. 0 0. 98 2. 26 0. 00 2. 26 0. 41904. 1 0.98 1.31 0.00 1.31 0.00

64. 3 0.23 0.10 0.00 0.10 -1. 07

51. 6 0. 11 0. 00 11. 02 11. 02 0. 84148. 7 0. 28 9. 91 0. 00 9. 91 9. 91

51. 6 0.22 0.22 0.00 0.22 0.0060. 0 0.21 0.00 0.00 0.00 -19. 01

$143,271.1 0.89 1.60 0.48 2.09 0.08

Source: Morgan Stanley, Intex

Page 176: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Transaction Monitoring

174

chapter 12

II. More DetailsThe 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 servicedloans within Morgan Stanley sponsored transactions. Details on 58 loans accountingfor $355 million in current balances are provided in this report. We also add anappendix 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 trans-ferred for late payment. This report focuses on specially serviced loans and does notinclude 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 becomespecially serviced. The information in this report is based on October remittancereport data and conversations with the special servicers over the past several weeks.

The universe of transactions examined is contained in Exhibit 1.

Source: Morgan Stanley, Intex

Property TypeNumber

of LoansCurrent Balance

($mm)

% of Total

Current Balance

Office 8 80.0 22.5Retail 9 84.0 23.6Senior Housing 20 119.4 33.6Hotel 10 38.5 10.8Multifamily 3 9.2 2.6Industrial 5 20.8 5.9Mobile Home 2 2.7 0.8Self Storage 1 1.0 0.3Total 58 355.6 100

SPECIALLY SERVICED LOAN CATEGORY exhibit 1

Page 177: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 175

Based on investor inquiries, this report provides more details on various loansthan our previous publication. The following criteria were used to determinewhich specific loans would be included in the report:

• All specially serviced loans within transactions issued off Morgan Stanley’sshelf where the total current balance of specially serviced loans was greaterthan $2 million.

• All specially serviced loans over $8 million, within transactions with current balances over $300 million, not issued off Morgan Stanley’s shelf but whereMorgan Stanley served as an underwriter or placement agent.

• All specially serviced loans originated by Morgan Stanley, regardless of theloan balance.

Based on the above criteria, 67% of the specially serviced balances within the 31transactions were analyzed. Fifteen transactions examined were issued off MorganStanley’s shelf and this report covers 96% of the specially serviced balances withinthose transactions.

In terms of property type concentration, senior housing properties accounted forthe largest dollar balance of loans examined, $119 million in current balances, fol-lowed by retail at $84 million and office at $80 million.

Twenty senior housing loans are contained in this report, the largest number by loancount, followed by hotel properties, which total 10 of the 58 loans we examine.

Page 178: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Transaction Monitoring

176

chapter 12

1 Previously reported specially serviced loans that have paid off.Loan level detail is not provided for the shaded transactions.Source: Intex, Remittance Reports

Issue

OriginalBalance

($mm)

CurrentBalance

($mm)

Transactions Issued Off Morgan Stanley's ShelfMorgan Stanley Capital I, 1995-GAL1 272.3 122.1Morgan Stanley Capital I, 1996-C1 340.5 224.6Morgan Stanley Capital I, 1996-MBL1 128.3 13.5Morgan Stanley Capital I, 1996-WF1 605.4 500.3Morgan Stanley Capital I, 1997-ALIC 802.7 502.5Morgan Stanley Capital I, 1997-C11 640.7 498.7Morgan Stanley Capital I, 1997-HF1 622.4 505.6Morgan Stanley Capital I, 1997-WF1 559.2 483.2Morgan Stanley Capital I, 1998-CF1 1,107.3 1,036.0Morgan Stanley Capital I, 1998-HF1 1,283.7 1,179.7Morgan Stanley Capital I, 1998-HF2 1,066.3 1,015.4Morgan Stanley Capital I, 1998-WF1 1,392.2 1,293.3Morgan Stanley Capital I, 1999-FNV1 632.1 614.1Morgan Stanley Capital I, 1999-RM1 867.1 811.1Morgan Stanley Capital I, 1999-WF1 968.5 924.8Subtotal 11,288.6 9,724.9

Other Morgan Stanley Sponsored TransactionsAllied Capital Commercial Mortgage Trust, 1998-1 334.7 122.0Bear Stearns Commercial Mortgage Securities, 1999-WF2 1,088.0 1,041.4CAPCO America Securitization Corp, 1998-D7 1,249.0 1,185.9Commercial Mortgage Acceptance Corp., 1996-C2 1 184.5 86.9Commercial Mortgage Acceptance Corp., 1998-C1 1,192.2 1,090.5Commercial Mortgage Acceptance Corp., 1999-C1 736.5 712.8Deutsche Mortgage and Asset Receiving Corp., 1998-C1 1,821.8 1,726.0First Union Commercial Mortgage Trust, 1999-C1 1,171.2 1,118.7Heller Financial Commercial Mortgage Asset Corp., 2000-PH1 960.7 943.6IMPAC CMB Trust, 1998-C1 320.2 281.6JP Morgan Commercial Mortgage Finance Corp., 1997-SPTL-C1 204.5 66.4KS Mortgage Capital, 1995-1 165.5 80.6Merrill Lynch Mortgage Investors, 1995-C2 1,075.7 229.2Nomura Asset Securities Corp., 1998-D6 3,722.7 3,576.3Prudential Securities Secured Financing Corp., 1999-NRF1 934.5 880.7Southern Pacific Thrift and Loan, 1996-C1 277.0 73.1Subtotal 15,438.7 13,215.5

Total 26,727.3 22,940.4

MORGAN STANLEY TRANSACTIONS(OCTOBER 2001 REMITTANCE REPORTS)

exhibit 2

Page 179: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 177

Delinquency Data (%) Specially Serviced Data ($mm)

Factor 30, 60,

90+ Forc. &

REO Total

Spec.Serv.

Loans

Spec. Serv.Loansin thisReport

%Covered

0.45 0.00 6.36 6.36 7.8 7.8 1000.66 1.96 0.71 2.68 5.7 1.6 280.11 0.00 53.54 53.54 10.0 7.1 71.00.83 0.00 0.32 0.32 15.9 15.9 1000.63 0.00 0.00 0.00 40.9 40.9 1000.78 0.00 0.74 0.74 0.0 0.0 00.81 0.21 0.00 0.21 1.1 0.0 00.86 0.00 1.65 1.65 8.0 8.0 1000.94 4.87 1.63 6.50 93.7 93.7 1000.92 1.39 0.16 1.56 22.1 22.1 1000.95 1.17 0.09 1.26 0.9 0.0 00.93 0.16 0.00 0.16 9.8 9.8 1000.97 0.00 1.46 1.46 9.1 9.1 1000.94 0.42 0.00 0.42 3.4 3.4 1000.95 0.53 0.00 0.53 5.0 5.0 1000.86 0.97 0.60 1.57 233.3 224.4 96.2

0.36 5.52 5.42 10.95 13.4 0.0 00.96 0.26 0.53 0.79 37.9 18.0 480.95 1.28 0.00 1.28 15.1 12.1 800.47 0.00 8.99 8.99 0.0 0.0 00.91 0.94 0.00 0.94 10.3 5.7 560.97 0.94 0.00 0.94 7.5 0.0 00.95 4.16 0.57 4.72 97.8 45.6 470.96 1.08 0.00 1.08 12.1 0.0 00.98 0.68 0.06 0.73 18.0 10.0 550.88 1.60 0.22 1.83 11.1 0.0 00.32 3.10 0.00 3.10 3.9 0.0 00.49 0.00 0.00 0.00 3.6 0.0 00.21 0.53 19.14 19.67 43.9 30.2 690.96 0.24 0.00 0.24 18.2 9.7 530.94 0.00 0.24 0.24 2.7 0.0 00.26 3.91 0.00 3.91 5.0 0.0 00.86 1.14 0.58 1.72 300.4 131.2 43.7

0.86 1.07 0.59 1.66 533.7 355.6 66.6

Page 180: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

SAMPLE MONITORING REPORTDes Peres Cinema

• ORIGINAL BALANCE: $8,700,000

• CURRENT BALANCE: $8,179,861

• PERCENT OF POOL BALANCE: 0.79%

• MASTER SERVICER: WELLS FARGO BANK

• SPECIAL SERVICER: GMAC COMMERCIAL MORTGAGE CORP.

• TRUSTEE: WELLS FARGO BANK

Transforming Real Estate Finance

Transaction Monitoring

178

chapter 12

Status: CurrentPaid Through Date: October 2001Location: Des Peres, MissouriSize: 89,870 sfYear Built: 1997Property Type: Retail – Movie Theater Major Tenants (% of sf): Des Peres

Cinema (100%)Originator: Wells Fargo Bank

Appraisal Value: $12,500,000Appraisal Date: June 17, 1998Appraisal Cap. Rate: Not AvailableMarket Cap. Rate: Not AvailableOriginal LTV: 68.9%Original Loan per Unit: $97Current LTV: 65.4%Current Loan per Unit: $91Underwritten DSCR: 1.40Current DSCR: 1.56 (December 1999)

Closest MSA: St. Louis, Missouri – 15.8 MilesMarket Average Occupancy for Property Type: NAMarket Average Rent for Property Type: $16

Reserves in Place (Type): Replacement ReservesReplacement Reserve Balance: $0.28/sfTaxes and Insurance Balance: Not RequiredLeasing Commission/Tenant Improvements Balance: Not AvailableOther Reserve Balance: Defeasance Reserve Required at $500,000

Annually for Crossed Pool.

property description valuation information

market data

escrow information

Page 181: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

REASON FOR SPECIAL SERVIC ING TRANSFER

This loan is cross-collateralized with loans on four other cinemas within the pool:St. Charles Cinema, O’Fallon 15 Cinema, St. Clair 10 Cinema, and Halls Ferry 14Cinema. The combined current balance of these assets is $29,614,954, whichmakes up about 2.8% of the pool. All the cross-collateralized loans are in specialservicing. The loans were transferred to special servicing because the borrower,Wehrenberg Theatres, filed for bankruptcy protection in February 2001.

UPDATE

The loan is current as of the October remittance report. The borrower is expectedto be out of bankruptcy on November 30, 2001. All of the cross-collateralizedleases have been affirmed by the bankruptcy hearings.

CURRENT STATUS OF RESOLUTION

A modification on the lease for the Halls Ferry property is in the documentationprocess. Under the modified terms, the Halls Ferry property, which is underper-forming, 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 fromthe other cross-collateralized locations resulting in the same cash flows to thetrust as the original loan. Additionally, the borrower is required to pay $500,000annually into a reserve to defease the Halls Ferry loan. The balance on the HallsFerry loan is $2.4 million. Once the $2.4 million is escrowed, the Halls Ferryproperty will be defeased.

ASSET SUMMARY

Wehrenberg Theatres currently operates 29 movie cinemas in Illinois, Missouri,and the Southeast. The company was founded in 1906 and filed for bankruptcyprotection in order to seek relief from leases at properties unrelated to this pool.Recent site inspections indicate that Des Peres Cinema, St. Charles Cinema,O’Fallon 15 Cinema, and St. Clair 10 Cinema are in better condition than the Halls Ferry property.

MATURITY DATE

January 1, 2019

Please refer to important disclosures at the end of this report. 179

Page 182: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Delinquencies by Originator

180

chapter 13

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 thelowest rate by issuer type

Our previous study was based on August 2001 remittance report data. In thisstudy, we examined January 2002 remittance reports to identify any changes ortrends from our previous study.

Since the last study, delinquencies on our entire universe have increased 41 bp to1.23% of current balances. Delinquencies on seasoned collateral (aged over oneyear) 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 insurancecompanies, declining 19 bp to 0.16% of current balances.

• Finance company collateral posted the greatest increase in delinquencies, rising51 bp since our last study to 1.40% of current balances.

• Investment bank collateral reported the overall highest delinquencies, rising 44bp to 1.46% of current balances.

• Commercial bank and finance company loans had the greatest dispersion indelinquencies.

In total, the study examines $228 billion of CMBS collateral within 434 differenttransactions. Seventy-seven different companies contributed collateral to the Intexdatabase. Originators with less than $300 million of collateral in the databasewere screened from the data set.

L IMITATIONS

Our study of delinquencies is only a snapshot at one point in time and does notinclude cumulative loss data.

Our analysis is based on information from Intex’s database. Typically, if a loan ispaid off within the database, the original balance remains, and the current bal-ance shows a zero balance. However, in a few cases we found that some loanswhich were paid off were completely eliminated from the database.

Lastly, our overall delinquency numbers include 30-day delinquencies. Oftenloans move in and out of 30 days due to timing of payments rather than funda-mental deterioration.

Page 183: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 181

Source: Morgan Stanley, Intex

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

Life InsuranceCompany

Commercial FinanceCompany

Investment Bank

%% 30 Days Delinq% 60 Days Delinq% 90 Days Delinq% Foreclosure% REO

0.16%

1.12%

1.40% 1.46%DELINQUENCIESBY INSTITUTION

TYPE

(BASED ONJANUARY

REMITTANCEREPORTS)

exhibit 1

Source: Morgan Stanley, Intex

0

20

40

60

80

100

120

140

Life InsuranceCompany

Commercial Bank Finance Company Investment Bank

$mm

1971-1994199519961997

1998199920002001

CURRENT BALANCEBY INSTITUTION

TYPE AND COHORTYEAR

(BASED ONJANUARY

REMITTANCEREPORTS)

exhibit 2

Page 184: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Delinquencies by Originator

182

chapter 13

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 origi-nated not the year in which the CMBS transaction was issued. In addition, thedata is sorted by loan originator, so if several originators contributed collateral toone CMBS transaction the delinquencies attributable to each specific loan areassigned to the respective originator.

L IFE COLLATERAL POSTS A DECLINE

In the January remittance report data we examined, collateral originated by insur-ance companies continued to post the lowest delinquency rate. The total delin-quency rate for life insurance collateral was 0.16%, followed by commercial bankcollateral at 1.12%, finance company collateral at 1.40%, and investment bank col-lateral at 1.46%.

In aggregate we found that although delinquencies in our CMBS universe rose 41bp 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 AmericanCouncil 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 cur-rent balances. ACLI defines delinquent loans as over 60+days past due and doesnot include loans that are in the process of foreclosure in their definition.

DELINQUENCIES BY INSTITUTION TYPE(BASED ON JANUARY REMITTANCE REPORTS)

exhibit 3

Source: Morgan Stanley, Intex

Originator TypeNumber of

IssuersTotal CurrentBalance (%)

OriginalBalance ($bn)

Commercial Bank 13 19.9 50.9Finance Company 29 23.0 59.6Investment Bank 22 51.6 133.0Life Insurance Company 13 5.5 16.6Total/Weighted Average 77 100.0 260.1

Page 185: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 183

LOWEST DELINQUENCY RATES

Twelve originators had delinquency rates below 0.25%. Four were commercialbanks, three were investment banks, three were life insurance companies, andtwo were finance companies. All have been in the business of originating loansfor at least ten years with the exception of UBS which began originating in 1999.

Originator Name Originator TypeTotal

Del. (%)

CurrentBalance

($mm)

EarliestOrigYear

Bank of America Commercial Bank 0.17 7,326.0 1973Capital Lease Finance Company 0.00 1,182.3 1995Chase Commercial Bank 0.25 6,360.0 1992Goldman Sachs Investment Bank 0.08 3,785.2 1997Keybank Commercial Bank 0.07 1,792.1 1997Merrill Lynch Canada Investment Bank 0.00 1,076.1 1997Principal Life Insurance Company 0.00 2,434.6 1987Prudential Life Insurance Company 0.19 3,786.1 1998Secore Finance Company 0.23 7,397.9 1996TIAA Life Insurance Company 0.00 1,121.4 1979UBS Investment Bank 0.05 2,101.2 1999Wells Fargo Commercial Bank 0.24 5,961.7 1990

Source: Morgan Stanley, Intex

LARGE ORIGINATORS WITH DELINQUENCIES LESSTHAN 0.25% (BASED ON JANUARY REMITTANCEREPORTS)

exhibit 4

CurrentBalance ($bn)

30 DaysDel (%)

60 DaysDel (%)

90 DaysDel (%)

Foreclosure(%) REO (%) Total (%)

45.4 0.30 0.09 0.49 0.07 0.17 1.1252.4 0.35 0.20 0.57 0.07 0.21 1.40

117.6 0.31 0.15 0.56 0.09 0.35 1.46 12.6 0.04 0.04 0.01 0.05 0.02 0.16

228.0 0.30 0.14 0.52 0.08 0.27 1.31

Page 186: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Delinquencies by Originator

184

chapter 13

HIGHEST DELINQUENCY RATES

There were thirteen originators that had delinquency rates above 2.5%. One life insurance company, one commercial bank, five investment banks, and sixfinance companies.

Exhibits 6-9 show our data sorted by originator type. Exhibit 10 contains analphabetical listing of delinquencies by originator broken out by year.

Originator Name Originator TypeTotal

Del. (%)

CurrentBalance

($mm)

MostCurrent

OrigYear

Allied Finance Company 10.94 120.1 2001Amresco Finance Company 3.90 1,989.2 1998Conti Finance Company 3.20 2,067.9 1999CSFB Investment Bank 3.59 10,080.1 2000DLJ Conduit Investment Bank 4.23 3,217.5 1998Legg Mason Finance Company 2.62 665.6 2000Merrill Lynch Investment Bank 4.16 6,767.8 2001Mutual Benefit Life Life Insurance Company 7.11 41.7 1995PaineWebber Investment Bank 3.63 1,592.6 2000Provident Commercial Bank 22.35 221.3 1999RMF Finance Company 22.19 403.0 1998Smith Barney Investment Bank 3.11 329.3 1997WMF Finance Company 5.22 867.4 1998

Source: Morgan Stanley, Intex

ORIGINATORS WITH DELINQUENCIES OVER 2.50%(BASED ON JANUARY REMITTANCE REPORTS)

exhibit 5

Page 187: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 185

COM

MER

CIAL

BAN

K DE

LIN

QUE

NCI

ESBA

SED

ON

JAN

UARY

REM

ITTA

NCE

REP

ORT

Sex

hibi

t 6

Sour

ce: M

orga

n St

anle

y, In

tex

Nam

e O

rigi

nato

rO

rigi

nato

r Ty

pe

Ori

gina

lB

alan

ce($

mm

)

Cur

rent

Bal

ance

( $m

m)

30 D

ays

Del

(%)

60 D

ays

Del

(%)

90 D

ays

Del

(%)

Fore

clos

ure

(%)

REO (%

)To

tal

(%)

Cha

nge

From

Aug (%

)

Banc

One

Com

mer

cial

Ban

k 1

,277

.66

948.

440.

260.

150.

060.

000.

000.

470.

03Ba

nk O

f Am

eric

aCo

mm

erci

al B

ank

8,3

64.9

3 7

,326

.02

0.11

0.04

0.02

0.00

0.00

0.17

0.12

CDC

Com

mer

cial

Ban

k 1

,003

.18

702

.72

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Chas

eCo

mm

erci

al B

ank

6,87

7.08

6,3

60.0

00.

000.

200.

000.

000.

050.

250.

20Ci

ticor

pCo

mm

erci

al B

ank

3,2

73.1

4 2

,866

.63

0.45

0.00

0.98

0.15

0.11

1.69

0.59

Com

mer

cial

Mor

tgag

e O

rigin

atio

nCo

mpa

ny o

f Can

ada

Com

mer

cial

Ban

k 4

32.7

3 4

25.7

60.

000.

000.

000.

000.

000.

000.

00Fi

rst U

nion

Com

mer

cial

Ban

k12

,496

.72

11,7

40.1

20.

430.

101.

330.

100.

142.

100.

43Ke

yban

kCo

mm

erci

al B

ank

1,82

2.33

1,7

92.1

30.

000.

000.

000.

070.

000.

070.

03La

salle

Com

mer

cial

Ban

k1,

169.

10 1

,153

.63

1.19

0.00

0.88

0.00

0.00

2.08

2.08

Nat

ions

bank

Com

mer

cial

Ban

k 5

,395

.81

4,8

99.4

60.

710.

230.

500.

000.

371.

810.

56PN

CCo

mm

erci

al B

ank

986.

18 9

75.0

90.

000.

000.

000.

000.

000.

000.

00Pr

ovid

ent

Com

mer

cial

Ban

k 1

,078

.87

221

.32

3.35

0.00

0.00

2.00

17.0

022

.35

3.15

Wel

ls F

argo

Com

mer

cial

Ban

k6,

676.

63 5

,961

.70

0.11

0.00

0.00

0.13

0.00

0.24

0.10

Tota

l/W

eigh

ted

Aver

age

50,8

54.3

8 4

5,37

3.02

0.30

0.09

0.49

0.07

0.17

1.12

0.26

Page 188: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Delinquencies by Originator

186

chapter 13

FINANCE COMPANY DELINQUENCIESBASED ON JANUARY REMITTANCE REPORTS

exhibit 7

Name Originator Originator Type

OriginalBalance

($mm)

CurrentBalance

($mm)

Allied Finance Company 341.30 120.14Amresco Finance Company 2,518.58 1,989.16Artesia Finance Company 666.62 652.43Boston Capital Finance Company 315.92 296.29Bridger Finance Company 680.52 670.44Capital Lease Finance Company 1,276.91 1,182.27CBA Finance Company 581.58 103.51CDPQ Finance Company 508.14 353.52Central Park Finance Company 516.04 493.75Conti Finance Company 2,505.73 2,067.90Criimi Mae Finance Company 805.26 684.27Finova Finance Company 683.09 662.12GE Capital Finance Company 7,256.57 7,035.50GMAC Finance Company 13,606.49 12,094.71Greenwich Finance Company 3,732.12 3,526.18Heller Finance Company 3,582.75 3,280.60Impac Finance Company 320.17 275.15Legg Mason Finance Company 821.23 665.61LTC Finance Company 356.95 212.74Midland Finance Company 4,185.55 3,451.56National Realty Finance Company 1,500.13 1,396.25NCB Finance Company 470.34 316.55Residential Funding Finance Company 1,515.72 1,458.87RMF Finance Company 425.08 402.97Secore Finance Company 7,975.44 7,397.91Southern Pacific Finance Company 778.89 280.98Starwood Finance Company 452.02 382.65Value Line Finance Company 335.60 126.63WMF Finance Company 901.99 867.37Total/Weighted Average 59,616.71 52,448.01

Page 189: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 187

Source: Morgan Stanley, Intex

30 DaysDel (%)

60 DaysDel (%)

90 DaysDel (%)

Foreclosure(%)

REO(%)

Total(%)

ChangeFrom Aug

(%)

0.00 0.00 5.59 0.00 5.35 10.94 -2.140.63 2.23 0.88 0.00 0.15 3.90 3.340.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 2.27 0.00 0.00 2.27 2.270.00 0.00 0.00 0.00 0.00 0.00 -0.320.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.00 0.00 0.000.65 0.00 0.64 0.00 0.00 1.29 1.060.44 0.52 1.42 0.05 0.77 3.20 1.520.00 0.00 0.53 0.00 0.00 0.53 -1.230.00 0.00 0.00 0.00 1.34 1.34 -0.120.12 0.08 0.08 0.08 0.00 0.37 -0.010.65 0.26 0.59 0.00 0.19 1.69 0.460.44 0.06 0.29 0.14 0.00 0.93 0.300.16 0.03 0.07 0.00 0.06 0.32 0.160.53 0.00 0.51 0.23 0.00 1.27 -0.640.67 0.00 1.74 0.00 0.21 2.62 1.220.00 0.00 0.95 0.00 0.00 0.95 0.090.20 0.17 0.79 0.21 0.57 1.95 0.510.00 0.04 0.64 0.00 0.00 0.68 0.050.04 0.00 0.00 0.00 0.00 0.04 0.041.14 0.00 0.13 0.00 0.03 1.30 0.910.00 0.00 21.27 0.00 0.92 22.19 0.060.23 0.00 0.00 0.00 0.00 0.23 -0.010.90 0.47 0.41 0.00 0.07 1.85 -0.450.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.00 0.00 -1.470.00 0.00 0.00 2.19 3.03 5.22 0.000.35 0.20 0.57 0.07 0.21 1.40 0.38

Page 190: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Delinquencies by Originator

188

chapter 13

INVESTMENT BANK DELINQUENCIESBASED ON JANUARY REMITTANCE REPORTS

exhibit 8

Name Originator Originator TypeOriginal Balance

($mm)Current Balance

($mm)

Archon Investment Bank 2,172.00 2,119.48Bankers Trust Investment Bank 1,265.21 817.38Bear Stearns Investment Bank 4,783.56 4,601.38CIBC Investment Bank 2,531.14 2,479.08Column Investment Bank 12,904.50 12,065.75CSFB Investment Bank 15,207.11 10,080.10Daiwa Investment Bank 1,090.14 813.99DLJ Conduit Investment Bank 3,541.53 3,217.46GACC Investment Bank 11,749.29 10,374.62Goldman Sachs Investment Bank 4,094.52 3,785.24Goldman Sachs/Bank Of America Investment Bank 455.00 454.05Lehman Investment Bank 23,321.62 20,441.67Merrill Lynch Investment Bank 7,427.86 6,767.82Merrill Lynch Canada Investment Bank 1,114.71 1,076.05Morgan Guaranty Investment Bank 8,001.47 7,580.71Morgan Stanley Investment Bank 9,456.34 8,920.50Nomura Investment Bank 4,195.41 3,441.84Nomura Conduit Investment Bank 12,131.51 11,506.82Painewebber Investment Bank 1,759.07 1,592.55Salomon Brothers Investment Bank 3,159.05 3,017.43Smith Barney Investment Bank 473.67 329.34UBS Investment Bank 2,119.37 2,101.20Total/Weighted Average 132,954.06 117,584.45

Page 191: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 189

Source: Morgan Stanley, Intex

30 DaysDel (%)

60 DaysDel (%)

90 DaysDel (%)

Foreclosure(%)

REO(%)

Total(%)

ChangeFrom Aug

(%)

0.55 0.08 0.93 0.00 0.00 1.56 0.920.00 0.00 0.00 0.00 0.00 0.00 0.000.06 0.07 0.06 0.00 0.23 0.43 0.250.07 0.00 0.19 0.07 0.00 0.33 0.160.53 0.51 0.40 0.02 0.17 1.63 0.670.69 0.04 0.29 0.56 2.01 3.59 0.890.23 0.43 0.09 0.93 0.60 2.27 0.260.42 1.14 1.66 0.00 1.01 4.23 2.500.10 0.04 0.37 0.08 0.00 0.59 0.180.08 0.00 0.00 0.00 0.00 0.08 0.080.00 0.00 0.00 0.00 0.00 0.00 0.000.10 0.03 0.35 0.03 0.01 0.51 0.080.54 0.14 2.90 0.09 0.48 4.16 2.980.00 0.00 0.00 0.00 0.00 0.00 0.000.33 0.22 0.77 0.03 0.00 1.34 0.770.56 0.07 0.39 0.00 0.20 1.21 0.680.52 0.00 0.18 0.00 0.92 1.63 0.220.08 0.00 0.61 0.00 0.39 1.08 0.091.23 0.33 0.96 0.00 1.12 3.63 1.020.00 0.68 0.28 0.31 0.00 1.28 0.882.04 0.00 1.07 0.00 0.00 3.11 1.720.03 0.00 0.02 0.00 0.00 0.05 0.030.31 0.15 0.56 0.09 0.35 1.46 0.57

Page 192: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Delinquencies by Originator

190

chapter 13

LIFE INSURANCE COMPANY DELINQUENCIESBASED ON JANUARY REMITTANCE REPORTS

exhibit 9

Name Originator Originator Type

OriginalBalance

($mm)

CurrentBalance

($mm)3D

Aetna Life Insurance Company 1,245.12 629.88CIGNA Life Insurance Company 377.00 375.53Confederation Life Life Insurance Company 1,926.60 427.47General American Life Insurance Company 777.50 606.24Jackson National Life Life Insurance Company 623.57 592.07John Hancock Life Insurance Company 1,419.52 1,342.12Mass Mutual Life Insurance Company 709.29 667.93Mutual Benefit Life Life Insurance Company 526.52 41.70Penn Mutual Life Insurance Company 781.56 288.25Principal Life Insurance Company 2,500.59 2,434.57Protective Life Life Insurance Company 554.13 278.69Prudential Life Insurance Company 3,873.88 3,786.06TIAA Life Insurance Company 1,314.50 1,121.38Total/Weighted Average 16,629.78 12,591.90

Page 193: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Please refer to important disclosures at the end of this report. 191

Source: Morgan Stanley, Intex

30 DaysDel (%)

60 DaysDel (%)

90 DaysDel (%)

Foreclosure(%)

REO (%)

Total(%)

ChangeFrom Aug

(%)

0.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.63 0.63 -3.470.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.00 0.00 0.000.34 0.00 0.00 0.00 0.00 0.34 0.340.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 7.11 0.00 7.11 -10.210.00 0.00 0.00 1.21 0.00 1.21 0.070.00 0.00 0.00 0.00 0.00 0.00 0.000.00 0.00 0.00 0.00 0.00 0.00 -1.340.01 0.13 0.05 0.00 0.00 0.19 0.040.00 0.00 0.00 0.00 0.00 0.00 0.00

0 0.04 0.04 0.01 0.05 0.02 0.16 -0.19

Page 194: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

MACROECONOMIC

Economic/Interest rate outlookInvestors 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 growthis slow and the Federal Reserve is in an easing mode. Credit spreads, however,might widen and defaults rise during an extended slowdown.

Swap spreadsInvestment-grade CMBS and swap spreads are highly correlated. The swap spreadrepresents 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 the10-year US Treasury (UST) yield plus 80 bp to the counterparty to receive a float-ing-rate (LIBOR) cashflow. If a CMBS has a yield of the UST + 130 bp and theswap 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, andreceive 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 foreach rating category. CMBS buyers should put current spreads in the context ofhistorical data and be able to explain circumstances that might drive spreads out-side of trading ranges.

Global riskCMBS investors should also have a view of the global economy and potentialeffects of global credit risk on US fixed-income markets. In 1998, the Russian debtcrisis had a major impact on US markets, including CMBS.

REAL ESTATE FACTORS

Real estate cycleHigh subordination levels insulate most investment-grade CMBS investors againstdefault during real estate downturns of the magnitude experienced over the past 30years. Non-investment grade buyers are more vulnerable to weakening real estate conditions. All investors, however, should monitor changes in macro real estate trendsto judge if spread widening could occur versus Treasuries, swaps, or other sectors.

Real estate dataThere are a myriad of sources from which to obtain data on real estate condi-tions. Providers include:

• American Council of Life Insurers (commercial mortgage delinquencies andoriginations)

• Federal Reserve Board (Beige Book on regional economic conditions; flow-of-funds data on commercial and multifamily originations)

Transforming Real Estate Finance

Factors to Consider

192

chapter 14

Page 195: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

• 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)

• Moody’s, Standard & Poor’s, Fitch (periodic reports on real estate and CMBS)

RELATIVE VALUE

Investors use several benchmarks for CMBS spreads:

• Single-A bank and finance—Formerly used as a benchmark for AAA CMBS;now used for both AAAs, AAs, and single-As.

• Unsecured REITs—Benchmark for BBBs and BBB- CMBS.

• Single-A corporate industrials—Benchmark for investment-grade CMBS.

• Mortgage pass-through OAS—Comparison for AAAs.

• ABS—Benchmark for short AAAs

CMBS investors examine the historical relationships among CMBS spreads andthose from each of these sectors. “Relative value” analysis involves judgingwhether the divergence of spreads represents a buying opportunity.

BOND-SPECIF IC

In analyzing a specific class of CMBS, an investor should consider the following factors:

RatingsMost CMBS have at least two ratings. The rating agencies in the US market areMoody’s, Standard and Poor’s (S&P), and Fitch. Some investors require that eitherMoody’s or S&P rate the bond. A potential investor should check if bond pur-chased in the secondary market is on ratings watch. Rating agencies state thatthey continually monitor outstanding ratings. Fitch and S&P conduct annualreviews and are more likely to change a rating at that time.

Rating agency analysts are available to answer credit questions about new issuesor bonds in the secondary market. With a new issue, it is sometimes valuable tocall the rating agency that has not rated the bond, since that agency is likely tohave analyzed the credit more conservatively.

Please refer to important disclosures at the end of this report. 193

Page 196: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Subordination levelsAn investor should compare the subordination level of the bond in question toothers in the market and to older transactions. Lower subordination is not neces-sarily an indication of lesser credit quality. Issuers with highest quality collateralobtain the lowest credit enhancement levels, and will often have the lowest delin-quency rates.

Subordination levels have trended down over time. An investor should be com-fortable that the current enhancement levels are appropriate for the expectedfuture default rates.

Issuer qualitySpreads in the CMBS market have tiered based on the perceived quality of theissuer’s collateral. As a general rule, bank and insurance companies are viewed ashaving the highest credit quality mortgages.

Investors should consider the possibility that an issuer will exit the CMBS busi-ness. Even though CMBS are bankruptcy remote, the failure of an issuer mightlead to spread widening because a market perception of reduced liquidity.

One quantitative check on the quality of an issuer’s collateral is the performanceof seasoned transactions. Morgan Stanley and other dealers publish delinquencyrates by issuer.

Cash flowsInvestors can model simple cash flows on Bloomberg for almost all CMBS. Moredetailed modeling services include Trepp LLC and Conquest. These services allowfor loan-by-loan default modeling and also provide detailed monitoring informa-tion. They cost about $2500/month.

LiquidityInvestors should try to determine how many dealers make a market in a particu-lar CMBS. If only one or two dealers trade a bond, the market will place a liquidi-ty premium on the security.

Property/regional concentrationConcentrations of various property types or within regions are important factorsin analyzing CMBS. Property type or regional concentrations above 40% of a poolmay 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, aninvestor should make sure that the rating agencies have made proper adjustmentsand that the pricing reflects any potential risk.

Deals with high (greater than 30%) concentrations of multifamily loans are oftenviewed favorable, since Federal agencies are more likely to purchase these deals.

Transforming Real Estate Finance

Factors to Consider

194

chapter 14

Page 197: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

ERISA-eligibilityEffective August 23, 2001, the Dept. of Labor ruled that CMBS rated as low asBBB- are eligible as pension fund amendments under ERISA. For most dealsclosed before that date, the issuer needs to amend the original documents toachieve ERISA-eligibility for bonds rated less than AAA. As of August 2001,Morgan Stanley, Nomura, and Bank of America had amended deals issued fromtheir shelves before the effective date.

Remittance reportsFor seasoned transactions, an investor should obtain the most current remittancereport and special servicing report. Every month, the trustee compiles the report,which details distributions and loan delinquencies. Special servicing reports high-light loans transferred from the servicer to the special servicer.

Total delinquency rates of less than 2% are usually not a concern for AAAinvestors. 30-day delinquency rates are of lesser concern than 60 day+ rates.

Price/Yield TablesThe following tables show details for a new issue conduit transaction, PNCMA2001-C1. The deal represents an average CMBS transaction. The tables show theresilient 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 supportand ratings. The AAA bond has 19.75% subordination and the BBB 8.75%. Theselevels were close to the average for conduit transactions as of August 2001.

Please refer to important disclosures at the end of this report. 195

CREDIT ENHANCEMENTPNCMA 2001-C1 A2

exhibit 1

Current At Issuance CreditClass Balance Cpn Type Rating Enhancement

A1 141,114 5.91 Fixed Rate AAA/Aaa 19.75%A2 560,781 6.36 Fixed Rate AAA/Aaa 19.75%B 33,060 6.58 Fixed Rate AA/Aa2 16.00%C1 18,856 6.80 Fixed Rate A/A2 12.50%C2 12,000 2.57 Fixed Rate A/A2 12.50%C2X 12,000 — IO, Other Non-Fi A/A2 —D 11,020 6.93 Fixed Rate A–/A3 11.25%E 8,816 — WAC/Pass Thru BBB+/Baa1 10.25%F 13,224 — WAC/Pass Thru BBB+/Baa2 8.75%G 7,714 — WAC/Pass Thru BBB–/Baa3 7.88%H 16,530 5.91 Fixed Rate BB+/Ba1 6.00%J 14,326 5.91 Fixed Rate BB/Ba2 4.38%K 5,510 5.91 Fixed Rate BB–/Ba3 3.75%L 8,816 5.91 Fixed Rate B+/B1 2.75%

The sources for exhibits 1-6 are cashflow runs on Trepp LLC analytics on Bloomberg.

Page 198: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The following table shows the types of call protection on the loans for each yearafter issuance. In the first year, 97.42% of the loans have defeasance or arelocked-out (LO).

The yield table below shows the effects of changing prepayments on the yield ofthe AAA bond, priced at 101-22 1/4. Note that the yield, average life, and spreadare almost unchanged across a wide range of prepayments. CPR or sometimes“CPY,” stands for the annual prepayment rate on loans not in lockout or defea-sance. 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 theeffects of prepayments.

Transforming Real Estate Finance

Factors to Consider

196

chapter 14

PREPAYMENT PENALTY MATRIXPNCMA 2001-C1 A2

exhibit 2

No. Date LO YM PP>=5% PP>=4% PP>=3% PP>=2% PP>=1% None

1 8/01 97.42% 2.58% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%2 8/02 96.51% 3.49% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%3 8/03 94.24% 5.76% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%4 8/04 92.85% 7.15% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%5 8/05 92.90% 5.88% 0.00% 0.00% 0.00% 0.00% 0.00% 1.22%6 8/06 90.87% 7.41% 0.00% 0.00% 0.00% 0.00% 0.00% 1.72%7 8/07 92.46% 7.17% 0.00% 0.00% 0.00% 0.00% 0.00% 0.37%8 8/08 91.90% 6.39% 0.00% 0.00% 0.00% 0.00% 0.00% 1.71%9 8/09 91.17% 6.48% 0.00% 0.00% 0.00% 0.00% 0.00% 2.35%10 8/10 71.60% 2.40% 0.00% 0.00% 0.00% 0.00% 0.00% 26.01%11 8/11 100.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%12 8/12 100.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%

PRICE/YIELD TABLEPNCMA 2001-C1 A2

exhibit 3

CPR when PP <= X% 0.00 15.00 30.00 45.00 60.00101-221 ⁄4 Price 6.172 6.172 6.172 6.171 6.171WAL 9.35 9.34 9.33 9.31 9.30Sprd Swaps +45.02 +45.06 +45.11 +45.17 +45.24

Page 199: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The yield table below shows the effects of changing prepayments on the yield ofthe 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.

The following yield table shows the effects of defaults on the AAA class. Thedefault rate is an annual default rate of the remaining balance, abbreviated asCDR. The yield changes very little up to 3% CDR, an extremely high default rateby historical standards. The change in yield at higher CDRs is caused by thereduction of the average life of a premium security.

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.

Please refer to important disclosures at the end of this report. 197

PRICE/YIELD TABLEPNCMA 2001-C1 F

exhibit 4

CPR when PP <= X% 0.00 15.00 30.00 45.00 60.00100-1 Price 7.073 7.073 7.073 7.073 7.073WAL 9.62 9.62 9.61 9.59 9.57Sprd Swaps +131.98 +132.02 +132.10 +132.21 +132.33

PRICE/YIELD TABLEPNCMA 2001-C1 A2

exhibit 5

Default Rate 0.00 0.50 1.00 2.00 3.00101-221 ⁄4 Price 6.172 6.171 6.170 6.166 6.162WAL 9.35 9.31 9.25 9.10 8.91Sprd Swaps +45.02 +45.19 +45.43 +46.06 +46.92

PRICE/YIELD TABLEPNCMA 2001-C1 F

exhibit 6

Default Rate 0.00 0.50 1.00 2.00 3.00100-1 Price 7.070 7.070 7.070 7.070 7.159WAL 9.62 9.62 9.62 9.62 14.13Sprd Swaps +132.01 +132.01 +132.01 +132.02 +107.51

Page 200: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

Glossary

198

Anchored Strip Center A grocery or discount retailer (the anchors) attract tenantsto small stores that are adjacent.

ASERS (Appraisal Subordinate Entitlement Reductions)/CVA (Collateral ValuationAdjustments) Structural deal feature that estimates unrealized losses on defaultedloans and prevents payment of current interest on the estimated losses. Designedto prevent conflicts between the interests of subordinate and senior classes.

Asset Mix The mix of property types, loan concentration and diversity in thenumber of loans.

Assisted Living Facilities This product is targeted to elderly needing assistance, butnot full-time medical care. These facilities typically consist of apartment-style unitswith a kitchenette. The operator of the facility provides three meals a day andassists residents with daily activities such as feeding, bathing, dressing, andmedication reminders.

Average Daily Rate Total guest room revenue divided by total number of occupiedrooms.

Call Protection What percentage of the pool is locked-out/defeased and for howlong? Defeasance and lockout provide for the most stable cash flows.The onlycash flow variability will be a result of credit events. There are scenarios underwhich yield to maturity increases as prepayments increase in deals with yieldmaintenance loans and penalty points.

Capital Expenditures (Cap Ex) Extraordinary expense items necessary to maintainthe property. Examples would include repairing a roof, repaving a parking lot, orreplacing heating and air conditioning systems.

Class A Properties Trophy quality properties; higher quality finishes and prominentlocations.

Class B Properties Generic real estate; 10-20 years old, well maintained, averagelocations, fewer amenities.

Class C Properties Older properties needing renovation; uncertain future.

Community Center Over 100,000 - 275,000 square feet of space containingmultiple anchors but not enclosed.

Congregate Senior Housing These are independent living facilities that also providea common dining facility and other services. Congregate senior housing has nomedical component, but may provide access to emergency medical care throughcall buttons. Not licensed as a nursing home.

Continuing Care Retirement Communities These facilities offer the entire continuumof senior housing from independent living to skilled nursing facilities. Residentsmove 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.

Page 201: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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 pricebonds, faster prepayments during yield maintenance result in a higher yield tomaturity for the investor. In a higher interest rate environment, the borrower isnot required to pay a yield maintenance penalty. The borrower is required toprepay the loan at par. The investor has purchased the bond at a discount andbeen repaid at par. This opportunity does not exist with locked-out or defeaseddeals. In a rates unchanged or lower scenario, the prepayment sharing arrange-ment between the coupon bonds and the IO significantly influences yield tomaturity. 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 indicatorof protection from cash flow volatility.

Defeasance Upon prepayment of the loan, the borrower is required to providethe lender with US government securities in an amount such that the lenderreceives the same yield as if the borrower had not prepaid the loans. From thelender’s perspective, a defeased loan is positive. The yield is the same, but thereis a credit upgrade from commercial real estate credit to US Government credit.

Distribution Space A property type whose principal use is distribution or lightassembly. It typically has minimal office space as a percentage of total space,0–10%. Twenty-four feet ceiling heights are the minimum for modern distributionbuildings. Higher heights are more economical for the tenant as they stack goodsvertically and rent fewer square feet.

Dollar Price The price at which bonds are currently trading, with a $100 dollarpricing being a par bond. Dollar price bonds above $105 tend to trade 2–3 bpcheaper for every dollar increment up to $110 where liquidity is limited.

Double Wide/Single Wide Describes the size of a manufactured home that a givenslab will support. The double wide segment has been the fastest growing.

EAs (Extension Advisor) Provides representation for the senior classes in a loanmodification process. Typically the transaction documents require the special servicer to get approval from the EA to grant any extensions to a loan beyond a certain date.

Fee Simple Ownership of both land and building in perpetuity.

Please refer to important disclosures at the end of this report. 199

Page 202: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

FF&E Furniture, Fixtures, and Equipment; standard hotel underwriting includes adeduction as an operating for the ongoing replacement of FF&E, typically 4% to5% of gross revenue. This differentiates hotel underwriting from apartment under-writing where some of those same expenses are considered capital expendituresand are not an operating expense deducted from NOI. Typical hotel refurbish-ment should occur at least every 7 years.

Flex Space/Office Warehouse Higher percentage of office space as a percentage oftotal space. This results in higher tenant improvement costs upon lease renewals.These tenants are less sensitive to clear heights and more concerned with accessto 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% ofgross revenue.

Full Service Hotel A hotel that offers banquet and convention services; one ormore full service restaurants.

Geography Are there state concentrations greater than 10%?

Go Dark Provisions Prevents tenants from vacating the space while continuing topay rent; landlords like this because vacant space may harm sales of other storeslocated in the center.

Ground Lease Building owner leases land from land owner.

Haircut Difference between the loan underwriter cash flow and the cash flow usedby the rating agency to size the loan. Examples of haircuts given by the ratingagencies include above market rents and apartment occupancies greater than 95%.

Independent Living Facilities Multifamily apartment complexes catering to seniorcitizens. 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 (IO’s) IOs generally benefit when prepayments increase duringyield maintenance. Older CMBS deals (1996 and prior) generally allocate theentire yield maintenance penalty to the IO. More recent CMBS IOs from dealswith yield maintenance also generally benefit if prepayments increase during yieldmaintenance periods. The prepayment penalty is generally higher than the pres-ent value of the interest strip lost as a result of the prepayment. The exception tothis occurs when interest rates rise to the level where no prepayment penalty isdue on the underlying loan.

LTV Loan to Value. Generally, the lower the better.

Leasehold Interest Building owner leases land. Lender wants ground lease term toexceed 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.

Transforming Real Estate Finance

Glossary

200

Page 203: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Limited Service Hotel No food service other than continental breakfast, minimalpublic space and small staff.

Loan to Value (LTV) Loan Amount/Appraised Value. A 70% LTV is the average forcommercial loans in CMBS transactions.

Location in Structure Mezzanine bonds have the most structural protection. Theyare protected from losses by subordinate bonds and protected from prepays bysenior bonds. Short AAAs have most volatile spread due to prepayment anddefault 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, andconcrete pads under the homes that comprise a manufactured housing communi-ty. The homes are independently financed. Homeowners pay monthly rent for thepad to the manufactured housing community owner.

Master Servicer Receives a 2-10 bp fee to collect monthly mortgage payments andreserve payments required by the loan documents.

Mezzanine Debt Debt secured by borrower’s equity interest in the property. Thereis 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 ofthe special servicing; generally preferred by rating agencies as an enhancement tocredit.

OAS (Option-Adjusted Spread) A measure of spread that adjusts for default and calloption 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 ofthe remaining loan balance (i.e., 5%, 4%, 3%, etc). Generally, the least preferableform 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 toregain control of space after a tenant closes its store.

Please refer to important disclosures at the end of this report. 201

Page 204: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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 suchas capital expenditures, retenanting costs (tenant improvements and leasing com-missions), 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 typically5-10 years; credit tenant leases may be longer. It is preferable not to have rolloverconcentrations, 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 individualsor businesses on a month-to-month basis. The average self-storage facility hasbetween 40,000 and 10,000 square feet of rentable space divided among 400 to1,000 individual units.

Shadow Anchored Strip Similar to an anchored strip center but the anchor is notpart of collateral for loan.

Shadow Rating A rating on an individual loan in a large loan pool given by therating agencies.

Skilled Nursing Facilities Independent nursing homes or a designated wing in a hospital. They provide full-time licensed skilled nursing, medical and rehabilitativeservices. Average length of stay can range from 2 months to 2 years, or more. 24-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.

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 anassumed 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 thisexpense. In strong demand markets, the landlord can pass this expense throughto the tenant in terms of a higher rental rate. In weak markets, landlords mustabsorb the cost.

Transforming Real Estate Finance

Glossary

202

Page 205: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

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 ispreferable 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 standardizeddeduction from net operating income to account for expenses need to maintainthe 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 arelocated in a highly developed area with little vacant land.

Unit Mix How many 1-bedroom versus 2-bedroom apartments are in a multi-familybuilding. Older complexes have higher proportions of 1-bedrooms; higher per-centages of 2-bedrooms are now preferred, as they are more flexible to families.

WAC (Weighted Average Coupon) The higher the WAC, the greater the prepaymentrisk; the lower the WAC, the greater the extension risk.

Yield Maintenance A prepayment penalty that requires a borrower to make a penaltypayment 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, theydiscount the difference between the remaining contractual mortgage payments and a hypothetical mortgage payment at current market rates. Most yield maintenanceformulas 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 tothe lender, since the lender made the loan at the Treasury rate plus a spread, butrecovers the remaining cash flows at the Treasury rate. Yield maintenance is generallyassumed to equal lock-out in modeling CMBS transactions.

Please refer to important disclosures at the end of this report. 203

Page 206: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

Transforming Real Estate Finance

CMBS Information

World Wide Web SourcesTRUSTEES

LaSalle www.etrustee.netStateStreet www.corporatetrust.statestreet.comWells Fargo www.ctslink.comChase www.jpmorgan.com/absmbsBankers Trust www.gis.deutsche-bank.com/gis/newclientcenter/

productcatalog/frameset_ap.asp?group=apir

SERVICERS

Amresco www.amresco.comMidland www.midlandls.comGMAC www.gmaccm.comFirst Union www.firstunion.com/strprod/

RATING AGENCIES

Fitch www.fitchratings.com Moody’s www.moodys.comS&P www.standardandpoors.com

AGENCIES

Fannie Mae www.fanniemae.comGinne Mae www.ginniemae.gov

ContactsTRADING

Jim Hiatt, [email protected] 212.761.2057Scott Stelzer, [email protected] 212.761.2055David Lehman, [email protected] 212.761.2104

RESEARCH

Howard Esaki, [email protected] 212.761.1446Marielle Jan de Beur, [email protected] 212.761.1454Masumi Pearl, [email protected] 212.761.1080Molly Paganin, [email protected] 212.761.1448

CAPITAL MARKETS

Jon Strain, [email protected] 212.761.2270Joel Friedman, [email protected] 212.761.2076

PRODUCT SPECIAL IST

Bob Karner, [email protected] 212.761.1605

Page 207: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

The information and opinions in this report were prepared by Morgan Stanley & Co. Incorporated (“MorganStanley”). Morgan Stanley does not undertake to advise you of changes in its opinion or information. MorganStanley and others associated with it may make markets or specialize in, have positions in and effect transac-tions in securities or instruments of companies mentioned and may also perform or seek to perform invest-ment banking services for those companies.

Within the last three years, Morgan Stanley & Co. Incorporated, Morgan Stanley DW Inc. and/or their affiliatesmanaged or co-managed a public offering of the securities of Advanta, Aetna, Artesia, BancOne, Bank ofAmerica, 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 UnionBank, First Union Bancorp, First Union National Bank, GMAC, Goldman Sachs, Heller Financial, Hilton HotelsCorp., 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 Inc1999-WF1, Nationsbank, New England Mutual Life, Nomura, Nomura Securities, PNC, Prudential, ResidentialFunding, 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 inthe securities recommended herein.

Morgan Stanley & Co. Incorporated, Morgan Stanley DW Inc. and/or their affiliates or their employees have ormay have a long or short position or holding in the securities, options on securities, or other related invest-ments of issuers mentioned herein.

An employee or director of Morgan Stanley & Co. Incorporated or Morgan Stanley DW Inc. and/or their affili-ates 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 rele-vant market or market segments using indicative prices provided by Morgan Stanley for the companies includ-ed in the index. Actual trades may not have been consummated at these prices. The indicative prices do notnecessarily reflect Morgan Stanley’s internal bookkeeping and may vary significantly from prices availablefrom other sources. No representation is made that the indices described above or the formula used to calcu-late 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 Stanleyand may not be reproduced or redisseminated in any form and may not be used as a basis for any financialinstrument or product without the permission of Morgan Stanley.

The investments discussed or recommended in this report may not be suitable for all investors. Investors mustmake their own investment decisions based on their specific investment objectives and financial position andusing such independent advisors as they believe necessary. Where an investment is denominated in a curren-cy other than the investor’s 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 per-formance. Income from investments may fluctuate. The price or value of the investments to which this reportrelates, either directly or indirectly, may fall or rise against the interest of investors. Price and availability aresubject 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 section21 of the Financial Services and Markets Act 2000. Morgan Stanley & Co. International Limited and/or its affili-ates may be providing or may have provided significant advice or investment services, including investmentbanking services, for any company mentioned in this report. NOT FOR DISTRIBUTION TO PRIVATE CUS-TOMERS 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 MorganStanley Asia (Singapore) Securities Pte Ltd.

To our readers in Australia: This publication has been issued by Morgan Stanley & Co. Incorporated but isbeing distributed in Australia by Morgan Stanley Dean Witter Australia Limited, a licensed dealer, whichaccepts responsibility for its contents. Any person receiving this report and wishing to effect transactions inany security discussed in it may wish to do so with an authorized representative of Morgan Stanley DeanWitter Australia Limited.

To our readers in Canada: This publication has been prepared by Morgan Stanley & Co. Incorporated and isbeing made available in certain provinces of Canada by Morgan Stanley Canada Limited. Morgan StanleyCanada Limited has approved of, and has agreed to take responsibility for, the contents of this publication inCanada.

To our readers in Spain: Morgan Stanley Dean Witter, S.V., S.A., a Morgan Stanley group company, supervisedby the Spanish Securities Markets Commission (CNMV), hereby states that this document has been written anddistributed in accordance with the rules of conduct applicable to financial research as established underSpanish regulations.

Past performance is not indicative of future returns. Certain assumptions may have been made in this analysiswhich have resulted in any returns detailed herein. No representation is made that any returns indicated willbe achieved. Transaction costs (such as commissions) are not included in the calculation of returns. Changesto the assumptions may have a material impact on any returns detailed.

Transforming Real Estate Finance

Disclaimer

Page 208: Morgan Stanley Transforming Real Estate Finance a CMBS Primer

THE AMERICAS1585 BroadwayNew York, NY 10036-8293Tel: (1) 212-761-4000

EUROPE25 Cabot Square, Canary WharfLondon E14 4QA, EnglandTel: (44 20) 7425-8000

JAPAN20-3, Ebisu 4-chome, Shibuya-kuTokyo 150-6008, JapanTel: (813) 5424-5000

ASIA PACIFICThree Exchange SquareHong KongTel: (852) 2848-5200


Recommended