See the Disclosure Appendix for the Analyst Certification and Other Disclosures. Modeling of Mortgage Prepayments and Defaults Lakhbir Hayre Managing Director Fixed Income Quantitative Analysis September 25, 2006
Topics An Overview of the Mortgage Market Challenges in Prepayment and Default Modeling Implications for Valuation of Mortgage- Backed Securities 2
The US Mortgage Market -- Colossus of the Bond World All Mortgage Debt Single-Family Mortgage Debt Mortgage-Backed Securities Asset-Backed Securities $12.3 trillion $9.5 trillion $6.2 trillion $2.0 trillion US Treasuries $4.2 trillion Corporate Bonds $5.2 trillion Municipals $2.3 trillion Sources: Federal Reserve System, Bond Market Association. 3
What are Mortgage Securities? A number of mortgage loans - from a few dozen to more than 10,000 - are pooled; Each loan pays interest and principal until it matures, is prepaid, or goes into default; Cashflows from the loans are paid to investors, after subtraction of administrative (or servicing) fees; Cashflows are either simply passed on to investors (passthrough securities) or allocated according to specified rules (structured securities, such as Collateralized Mortgage Obligations (CMOs). 4
Basic Security Features Cashflows are monthly, unlike Treasures or corporate bonds, which pay semi-annually; Amortizing assets => principal paid out over a period of time; For pass-throughs, each monthly payment will tend to include some principal; For structured MBS/ABS, principal paid out over a principal window Prepayment of principal by borrowers Þ call risk key property of many MBS/ABS durations much shorter than similar maturity bullet security. 5
Basic MBS is the Pass-Through Issued by FHLMC, FNMA, GNMA and Private Entities Many mortgages with similar characteristics collected into a pool Investor receives pro-rata share of monthly payments Interest and principal payments are guaranteed by the issuing agency, or through credit enhancements (for private issuers) 6
Structure of a Pass-Through Borrower pays 6.5% + principal payments Loan servicer receives servicing fee of 0.35% Investor receives coupon payments of 6% + principal payments Fannie/Freddie/Ginnie receives a guarantee fee of 0.15% Source: Citigroup. Actual numbers may vary from pool to pool 7
Valuation of Mortgage Securities MBSs are bonds with embedded options; More complex than standard callable bonds: Each $1 is a separate option Option-exercise is inefficient High degree of path dependence Prepayment models key to valuation; Prepayment models combined with Term Structure Models to obtain option-adjusted spreads (OAS). 8
Basic Steps in Mortgage Valuation Generate a large number of interest rate paths, both for discounting and for cash flow generation; On each path, call a prepayment model/default model to calculate mortgage cash flows; Calculate average PV of cash flows, using benchmark rates plus a spread; Spread that equates average PV to market price is the option-adjusted spread (OAS). 9
Prepayment Rates Are Critical in Determining MBS Value A. Cashflows Assuming No Prepayments Cash Flow Per $100,000 10,000 8,000 6,000 4,000 2,000 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Interest Principal Servicing B. Cashflows Assuming a More Realistic Prepayment Rate 20,000 Source: Citigroup. Cash Flow Per $100,000 15,000 10,000 5,000 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Interest Principal Servicing 10
Some Difficulties in Developing Prepayment Models A large number of important variables Continual innovations in mortgage financing implies constantly changing regimes Diverse and changing range of mortgage loan types A high degree of path dependence Unpredictable and inefficient borrower behavior Limited historical prepayment data and incomplete information 11
A Large Number of Factors Impact Prepayment and Default Rates Economic: Mortgage Rates, Housing Inflation, Consumer Confidence, Unemployment, etc. Loan: Coupon rate, original term, remaining term, type (Fixed, ARM, Hybrid), loan size, geographical location, etc. Borrower: Credit, Socio-Economic Status, Personal Situation Other: Past exposure to refinancing opportunities, mortgage origination and servicing process, etc. 12
Changing Environment Key determinants change over time: closing costs, choice of loan types, mortgage lending industry, loan origination process,etc. Borrowers have become more savvy over the years Borrower sentiment (or psychology) plays an important role 13
Changes in the Mortgage Market over Time CPR (%) 90 80 70 60 50 40 30 20 10 0 Speeds on 2001 Coupons in Jul 2003 Speeds on 1992 Coupons in Nov 1993 0 50 100 150 200 250 300 WAC - "No Point" Mtg Rate (bp) Speeds on 2000 Coupons in Dec 2001 Speeds on 1996 Coupons in Dec 1998 Sources: Fannie Mae, Freddie Mac and Citigroup.. 14
The Media Effect Measures Psychological Impact of Multi-Year Lows in Rates Rate (%) 9.0 8.5 8.0 7.5 7.0 6.5 6.0 Rates drop 100bp from April to Year- End 97, but nothing happens Rates hit multi-year low Rates back to early 1998 lows but nothing happens 5.5 Rates fall significantly 5.0 below early 1998 lows 4 Apr 97 4 Jul 97 3 Oct 97 2 Jan 98 3 Apr 98 3 Jul 98 2 Oct 98 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 Mortgage Rate (Left Axis) MBA Refinancing Index (Right Axis) Source: Mortgage Bankers Association, Freddie Mac, Citigroup 15
Loan Type Variation: Term 30-Year: Most common type in the US; 15-Year: Higher monthly payments, so few 1 st time home buyers => slower turnover and seasoning ramp. Also common refi vehicle for 30-year mortgages; 20-Year. Attracts borrowers who want a 15-year loan, but cannot afford the higher monthly payments; 10-Year. Mostly people refinancing out of a 15-year loan; 40-Year. A newer product, popular with borrowers stretching to buy a house and who want to minimize the monthly payment. Notes: 1. For a given difference between the coupons on the current and a new mortgage, the shorter the term, the lower the refinancing incentive; 2. Regardless of the shape of the yield curve, the shorter the term, the lower the mortgage rate. 16
Loan Type Variation: Coupon Fixed Rate. Basic mortgage in the US; Adjustable Rate. Coupon resets periodically at a stated margin over a specified index (typically 1-year Treasury or 6-month LIBOR). Initial coupon often teasered and much lower than on a fixed-rate loan, so ARMs attract lot of 1 st time buyers or other people with short time horizons; Hybrid. Coupon fixed for the first 3, 5, 7 or 10 years, then adjusts like a standard ARM. The shorter the fixed rate period, the shorter the typical borrower horizon, and the faster the speeds. 17
Loan Type Variation: Credit Jumbo. Prime quality loans that are too large for agency pools; Fannie May/Freddie Mac. Generally prime quality loans that fall below the conforming limit ; Ginnie Mae. Loans insured by the FHA or the VA. Relative to FN/FH, poorer credit and lower loan balances. Alternative (Alt) A. Borrowers are generally moderate to good credit (hence the A ), but lack full documentation Sub-Prime. Borrowers with poor credit histories. 18
Loan Type Variation: Other Features Loan Size. Has a big impact on speeds. Geographical Location. Ditto. Prepayment Penalties. Uncommon in prime loans, but prevalent fro sub-primes. Amortization Schedule. Traditional mortgages in the US have been fully amortizing. However, strong growth in recent years in loans which pay interest only for a number of years (eg 10/20), or can even have negative amortization (ie. loan balance can increase), such as Option ARMs. 19
Prepayment Speeds on Prime and Subprime Loans Prime Subprime 80 70 60 50 40 30 20 10 0 December-02 June-03 January-04 July-04 February-05 August-05 March-06 October-06 20
Effect of Loan Balance Loan Loan Balance (LLB) pools are less reactive to refinancing opportunities, but little difference in turnover speeds 60 50 Generic 6s of 2003 LLB 6s of 2003 40 CPR (%) 30 20 10 0 200301 200304 200307 200310 200401 200404 200407 200410 Source: Citigroup. 21
Significant Differences in Speeds by State Factors include loan size, closing costs, taxes, home price appreciation, and local economic conditions 60 50 40 CA IL TX FL NY US CPR (%) 30 20 10 0 Apr '03 Jun '03 Aug '03 Oct '03 Dec '03 Feb '04 Apr '04 Jun '04 Aug '04 Oct '04 Source: Citigroup. 22
High Degree of Path Dependence Borrowers will differ in their propensity and ability to refinance As a pool of borrowers experiences refinancings, most able borrowers leave the pool at higher rates Remaining borrowers less responsive (burnout) Hence prepayment rates depend on complete history of interest rates 23
Burnout Freddie Mac 9s of 1991 Freddie Mac 8.5s of 1991 30 Year Mortgage Rate 80 70 60 50 40 30 20 10 10 9.5 9 8.5 8 7.5 7 6.5 6 5.5 0 Aug-91 May-93 Feb-95 Nov-96 Aug-98 May-00 Feb-02 Nov-03 5 Sources: Freddie Mac and Citigroup. 24
Inefficient Exercise of the Prepayment Option 90 Burnout and Media Effect CPR (%) Speeds on 2001 Coupons in Jul 2003 80 70 60 50 40 Speeds on 2001 Coupons in Jan 2004 30 20 10 0 0 50 100 150 200 250 300 WAC "No Point" Mtg Rate (bp) 25
History of Prepayment Modeling First Generation Models (Salomon, 1985) - Standard Multiple Regression Models - Many Variables good historical fit, but not robust over time Diversity of Collateral and Borrowers and Continuing Changes in Prepayment Environment Suggests More Fundamental Approach (Salomon, 1995) - Sources of Prepayments (Modular Approach) - Flexible and Dynamic Inputs and Relationships 26
Why are Mortgages Prepaid? Housing Turnover - the sale of a home triggers a prepayment Refinancings - the loan is refinanced Defaults - foreclosure on the house leads to the loan being paid off Curtailments (or partial prepayments) - borrowers make more than their scheduled payment Full Payoffs - the loan is paid off: for example, due to a natural disaster This is true for all loans, regardless of type of loan, country/region etc. However, the magnitude of each component will depend on cultural, demographic, collateral and economic factors. 27
Structure and Key Features of the Model Modular Approach: Projected Speed = Sum of speeds due to 1. Housing Turnover 2. Refinancings - Rate, Cash-Outs and Credit Driven 3. Curtailments 4. Defaults 28
Overall Housing Turnover Rate 10.0% -200bp 7.5% Housing Turnover Rate (in%) 5.0% 2.5% Actual No shift +200bp Projected 0.0% Jul-78 Jul-81 Jul-84 Jul-87 Jul-90 Jul-93 Jul-96 Jul-99 Jul-02 Jul-05 Jul-08 Source: National Association of Realtors, US Census Bureau, Citigroup 6 29
Housing Turnover-Related Speeds Dominant in high-interestrate environment Strong seasonal component Seasoning: brand-new pools tend to prepay more slowly Lock-in: higher coupons typically have higher turnover rates CPR (%) 8 7 6 5 4 3 2 1 0 Source: Citigroup Prepayment Rates Fannie Mae 6s of 1993 Oct-93 Feb-94 Jun-94 Oct-94 Feb-95 Jun-95 Oct-95 Feb-96 Jun-96 Oct-96 Feb-97 Jun-97 Oct-97 30
Seasoning Depends on Loan Age and Home Price Appreciation 14 125% PSA 12 CPR (%) 10 Age-related Seasoning is a critical dimension of Turnover 8 6 4 2 0 0 5 10 15 20 25 30 35 40 45 50 55 Loan Age (Months) However, Seasoning is Modulated by Appreciation in Home Prices 31
Actual and Projected Home Price Appreciation 16% Annual US Home Price Appreciation (in %) 12% 8% 4% 0% Feb-71 Projected Actual Feb-74 Feb-77 Feb-80 Feb-83 Feb-86 Feb-89 Feb-92 Feb-95 Feb-98 Feb-01 Feb-04 Feb-07 Feb-10 Source: Fannie Mae, Freddie Mac, Citigroup 32
Basic Dynamics of Refinancing Model We assume that there are several classes of borrowers, ranging from slowest to fastest, each class having its own refi curve Refi Rate Source: Citigroup. Refi Incentive 33
Evolution of Population of Mortgagors The mix of borrowers changes each month, as faster refinancers leave the pool at a faster rate 60% % in Class 50% 40% 30% 20% 10% 0% Initial Distribution Post-Refi Distribution Slowest Slow Fast Fastest Source: Citigroup. The model keeps track of the population mix, and the overall refi rate for month is the aggregate over the remaining borrowers: Refi Rate = Fraction of pool in class 1 * Refi rate for class 1 +...... + Fraction of pool in class k * Refi rate for class k 34
Other Aspects of the Refinancing Model No consensus on how to calculate the refinancing incentive - A common approach is to compare PVs of new and old mortgages - Another approach: # of months to recoup costs of refinancing Mortgage rates used to calculate refi incentive need to depend on loan type eg sub-prime rates much higher than prime rates Loan balance is an important factor in determining incentive A seasoning curve can be introduced using transient costs of refinancing; Reactivity to refinancing opportunities depends on FICO, LTV and other loan features. 35
A Simple Default Model Multiple of the SDA Curve 1.2 The Standard Default Assumption (SDA Curve) 1.0 0.8 Pe CPR (%) 0.6 0.4 0.2 Tail 0.0 0 30 60 90 120 150 180 210 240 270 300 330 360 Mortgage Age (Months) 36
A Default Model Framework Probability of Default = Probability(LTV > Threshold) *Probability (Trigger Event) Likelihood of Trigger Events depends on FICO, Debt-to-Income ratio, unemployment rates, payment shock, etc. Default Probability 0.5% 0.4% 0.3% 0.2% 0.1% 0.0% AR with Affordability AR(1) 0 3 6 9 12 15 18 21 24 27 30 Age (years) 37
Key Determinants of Trigger Events Unemployment or Curtailment of Income 38.98 % Illness or Death of Mortgagor 16.25 Excessive Obligation 9.70 Marital Difficulties 9.45 Illness or Death in Family 7.31 Extreme Hardship 4.06 Business Failure 2.68 Property Problem or Casualty Loss 2.00 Inability to Sell or Rent Properties 1.83 Employment Transfer or Military Service 0.97 All other Reasons 6.77 Source: Freddie Mac. 38
Historical Monthly Transition Rates for Sub-Prime Loans Fixed-Rate Loans To Non-delinq Delinqent Foreclosure REO Payoff From (%) (%) (%) (%) (%) Non-delinq 97.2 1.0 0.1 0.0 1.7 Delinquent 10.8 73.5 14.4 0.0 1.3 Foreclosure 2.9 1.7 88.2 5.2 2.0 REO 0.1 0.1 0.2 84.9 14.7 2/28 Hybrids Non-delinq Delinquent Foreclosure REO Payoff From (%) (%) (%) (%) (%) Non-delinq 96.6 1.2 0.1 0.0 2.1 Delinquent 10.0 72.5 16.0 0.0 1.5 Foreclosure 3.3 1.8 87.8 4.8 2.3 REO 0.1 0.0 0.2 83.7 16.0 Source: Citigroup. To 39
Subprime Collateral Default Rates by LTV 25 20 15 10 5 0 200208 200211 200302 200305 200308 200311 200402 200405 200408 200411 200502 200505 200508 200511 200602 200605 L,10x,75 L,75x,85 L,85x,+ 40
Summary and Implications for MBS Valuation Modeling of prepayments and defaults as much art as science; MBS cashflow generation depends on these models; Hence little consensus on valuations, especially for complex MBS derivatives; Work by non-practitioners of little value in deciding, say, how much more to pay for a $60,000 average balance Texas pool vs. a $80,000 Illinois pool; On a positive note, great employment opportunities for good prepayment modeler; A book from John Wiley & Sons, SSB Guide to Mortgage- and Asset- Backed Securities, edited by L. Hayre, provides a great introduction to this area. 41
Disclaimer ANALYST CERTIFICATION Each individual contributing to this report hereby certifies that, with respect to the material contributed by him or her, all of the views expressed in this report accurately reflect his or her personal views about such subject securities, issuers, currencies, commodities, futures, options, economies or strategies. Further, each of these individuals also certifies that no part of his or her compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report. ADDITIONAL INFORMATION AVAILABLE UPON REQUEST Citibank, N.A., London Branch and Inc, including its parent, subsidiaries, divisions and/or affiliates ( the Firm ), may make a market in the securities discussed in this report and may sell to or buy from customers, as principal, securities recommended in this report. The Firm may have a position in securities or options of any issuer recommended in this report. The Firm may be regular issuers of, and trade in (including position taking), financial instruments linked to securities, which may have been reported on in this research report. The Firm may perform or solicit investment banking or other services from any issuer recommended in this report. An employee of the Firm may be a director of an issuer recommended in this report. Within the past three years, the Firm may have acted as manager or co-manager of a public offering of the securities of any issuer recommended in this report. Securities recommended, offered, or sold by the Firm : (i) are not insured by the Federal Deposit Insurance Corporation; (ii) are not deposits or other obligations of any insured depository institution (including Citibank); and (iii) are subject to investment risks, including the possible loss of the principal amount invested. Past performance is not a guarantee of future results. This report does not take into account the investment objectives, financial situation or particular needs of any particular person. Investors should obtain advice based on their own individual circumstances before making an investment decision. Investing in non-u.s. securities, including ADR s entails certain risks. The securities of non-u.s. issuers may not be registered with, nor be subject to the reporting requirements of, the U.S. Securities and Exchange Commission. There may be limited information available on foreign securities. Foreign companies are generally not subject to uniform audit and reporting standards, practices and requirements comparable to those in the U.S. Securities of some foreign companies may be less liquid and their prices more volatile than securities of comparable U.S. companies. In addition, exchange rate movements may have an adverse effect on the value of an investment in a foreign securities and its corresponding dividend payment for U.S. investors. Net dividends to ADR investors are estimated, using withholding tax rates conventions, deemed accurate, but investors are urged to consult their tax advisor for exact dividend computations. Although information has been obtained from and is based upon sources the Firm believes to be reliable, we do not guarantee its accuracy and it may be incomplete or condensed. All opinions and estimates constitute the Firm 's judgement as of the date of the report and are subject to change without notice. This report is for informational purposes only and is not intended as an offer or solicitation for the purchase or sale of a security. This research report does not constitute an offer of securities. Any decision to purchase securities mentioned in this research must take into account existing public information on such security or any registered prospectus. Investing in non-us securities by US persons may entail certain risks. Investors who have received this report from the Firm may be prohibited in certain US States from purchasing securities mentioned in this report from the Firm; please ask your Financial Consultant for additional details. This report is distributed in the United Kingdom by Citibank, N.A. London Branch or Limited, Citigroup Centre, Canada Square, Canary Wharf, London E14 5LB, UK. This material is directed exclusively at market professional and institutional investor customers in the United Kingdom and is not for distribution to private customers in the United Kingdom, as defined by the rules of the Financial Services Authority, who should not rely on this material. Moreover, any investment or service to which the material may relate will not be made available to such private customers. This material may relate to investments or services of a person outside of the United Kingdom or to other matters which are not regulated by the Financial Services Authority and further details as to where this may be the case are available upon request in respect of this material. If this publication is being made available in certain provinces of Canada by (Canada) Inc. ("The Firm Canada"), the Firm Canada has approved this publication. If this report was prepared by the Firm (excluding Nikko Citigroup Limited) and distributed in Japan by Nikko Citigroup Limited, it is being so distributed under license. This report is made available in Australia, to non-retail clients through Australia Pty Limited (ABN 64 003 114 832 and AFSL No. 240992), a participant of the ASX Group and a participant of the Sydney Futures Exchange Limited and to retail clients through Smith Barney Citigroup Australia Pty Ltd (ABN 10 009 145 555 and AFSL No. 240813), a participant of the ASX Group. In New Zealand it is made available through New Zealand Limited, a member firm of the New Zealand Stock Exchange. (Pty) Limited is incorporated in the Republic of South Africa (company registration number 2000/025866/07) and its registered office is at Citibank Plaza, 145 West Street, Sandown, Sandton, 2196, Republic of South Africa. The investments and services contained herein are not available to private customers in South Africa. This publication is made available in Singapore through Singapore Pte Ltd, a Capital Markets Services license holder. This report is being distributed in Hong Kong by or on behalf of, and is attributable to Asia Limited, 20th Floor, Three Exchange Square, Hong Kong. Inc. is a member of the Securities Investor Protection Corporation (SIPC). Inc., 2006. All rights reserved Smith Barney is a division and service mark of Inc. and its affiliates and is used and registered throughout the world. Citigroup and the Umbrella Device are trademarks and service marks of Citicorp and its affiliates and are used and registered throughout the world. CitiFx is a service mark of Citicorp. Any unauthorized use, duplication or disclosure is prohibited by law and may result in prosecution. Nikko is a service mark of Nikko Cordial Corporation. 42