How Resilient Are Mortgage Backed Securities to Collateralized Debt Obligation Market Disruptions?
Joseph R. Mason, Associate Professor of Finance, Drexel University Joshua Rosner, Managing Director, Graham Fisher & Co.
Criterion Economics L.L.C.
1
1965Q1 1967Q1 1969Q1 1971Q1 1973Q1 1975Q1 1977Q1 1979Q1 1981Q1 1983Q1 1985Q1 1987Q1 1989Q1 1991Q1 1993Q1 1995Q1 1997Q1 1999Q1 2001Q1 2003Q1 2005Q1
62%
63%
64%
65%
66%
67%
68%
69%
70%
Source: Bureau of Census
Criterion Economics L.L.C. 2
Figure 1: U.S. Home Ownership 1965-2006
Figure 2: Home Equity Extracted and Available for Extraction 1995-2005
1400 1200
Home Equity Cashed out at Refi ($Billions)
1000
Home Equity Loans ($Billions) Home Equity Cashed Out at Refi and Home Equity Loans ($Billion)
$Billions
800
600
400
200
0 1995 1996 1997 1998 1999 2000 Year 2001 2002 2003 2004 2005
Source: Joint Center for Housing Studies of Harvard University (2006).
Criterion Economics L.L.C. 3
Figure 3: Refinancing Behavior 1995-2005
90 80 70
% of Refinancings
60 50 40 30
% Refinancing Resulting in Lo wer Lo an A mo unt
% Refinancing Resulting in 5% o r Higher Lo an A mo unt
20 10 0 1995 1996 1997 1998 1999 2000 Year 2001 2002 2003 2004 2005
Sources: Joint Center for Housing Studies of Harvard University (2006).
Criterion Economics L.L.C. 4
Figure 4: Piggyback Lending for Home Purchase 2001-2004
45 40 35 30
Percent
Percent Piggybacks by Number of Loans Percent Piggybacks by Dollar Volume
25 20 15 10 5 0 2001 2002 Year 2003 2004
Source: SMR Research Corporation (2004).
Criterion Economics L.L.C. 5
Figure 5A: Investment Homes Share of Total Homes Purchased
2004 2005 Vacation Home, 13.00% Vacation Home, 12.20%
Investment Home, 23% Primary Residence, 64% Primary Residence, 60.10%
Investment Home, 27.70%
Source: National Association of Realtors
Criterion Economics L.L.C. 6
Figure 5B: Prime Market Investor Shares
18 16 14 12 10 8 6 4 2 0 1999 2000 2001 2002 2003 2004 2005
Second-Home Share Investor Share
Source: Harvard’s Joint Center for Housing Studies (2006)
Criterion Economics L.L.C. 7
100
200
300
400
500
600
700
800
900
0
1/7/2003 3/7/2003 5/7/2003 7/7/2003 9/7/2003 11/7/2003 1/7/2004 3/7/2004 5/7/2004 7/7/2004 9/7/2004 11/7/2004 1/7/2005 3/7/2005 5/7/2005 7/7/2005 9/7/2005 11/7/2005 1/7/2006 3/7/2006 5/7/2006 7/7/2006 9/7/2006 11/7/2006 1/7/2007
Source: Bureau of Census
Figure 6: Five Year BBB- Floating Home Equity Spread to One-Month
Criterion Economics L.L.C. 8
The Complexity of Securitization
Mortgages and Mortgage-backed Securities Are Complex and Difficult to Value
Complexity of Default vs Prepay Risk Additional Complexity in MBS
Deep Tranching Unique Securities
Examples
Criterion Economics L.L.C.
9
The Complexity of Securitization
Mortgages and Mortgage-backed Securities Are Complex and Difficult to Value
Complexity of Default vs Prepay Risk Additional Complexity in MBS
Deep Tranching Unique Securities
Examples
Criterion Economics L.L.C.
10
Default Models are Well-accepted Throughout the Mortgage Industry
Criterion Economics L.L.C.
11
Figure 7: Prepayment and Default Rates as a Function of Mortgage Duration
6 5
Percent of Loans
4 3 2 1 0
21 42 63 84 0
Prepay Default
105
126
147
168
189
210
231
252
273
294
315
336
Months
Source: Calomiris and Mason (2007).
Criterion Economics L.L.C. 12
357
Typically Adjust for Prepayment Speeds with PSA Models
Prepayment and Principal and Interest Cash flows arising from Selected Prepayment Scenarios
Source: Mathworks.com
Criterion Economics L.L.C. 13
Table 1: Regression Results (Dependent Variable Is Prepayments)
Parameter Intercept EPS Score % Change in Mtg. Rates Standard Deviation of % Change in Mtg. Rates DF 1 1 1 1 Estimate -51.0834 0.5294 -3.1048 2236.297 Std. Error 0.1388 0.0010 0.0225 5.5339 Pr > ChiSq < .0001 < .0001 < .0001 < .0001
Source: Calomiris and Mason (2007)
Criterion Economics L.L.C. 14
The Complexity of Securitization
Mortgages and Mortgage-backed Securities Are Complex and Difficult to Value
Complexity of Default vs Prepay Risk Additional Complexity in MBS
Deep Tranching Unique Securities
Examples
Criterion Economics L.L.C.
15
Figure 8: Tranches Issued in European Securitizations 1987-2003
Source: Firla-Cuchra and Jenkinson (2005)
Criterion Economics L.L.C. 16
Table 2: Issues with the Given Number of Tranches as a Percentage of All Issues Per Type (Mean Number of Tranches Per Issue)
Tranches per issue 1 2 3 4 5 6 7 8 9 10+ Total issues RMBS 20.5% 26.5% 19.6% 13.4% 10.2% 3.9% 2.0% 0.8% 0.2% 2.9% 596 EQUIP 24.4% 41.5% 14.6% 12.2% 7.3% CARDS 58.1% 17.6% 20.3% 2.7% 1.4% AUTO 27.8% 55.7% 13.9% 1.3% 1.3% OTHER 53.8% 25.5% 9.0% 6.9% 2.1% 1.4% 1.4%
41
74
79
145
Source: Firla-Cuchra and Jenkinson (2005)
Criterion Economics L.L.C. 17
The Complexity of Securitization
Mortgages and Mortgage-backed Securities Are Complex and Difficult to Value
Complexity of Default vs Prepay Risk Additional Complexity in MBS
Deep Tranching Unique Securities
Examples
Criterion Economics L.L.C.
18
The Complexity of Securitization
Mortgages and Mortgage-backed Securities Are Complex and Difficult to Value
Complexity of Default vs Prepay Risk Additional Complexity in MBS
Deep Tranching Unique Securities
Examples
Criterion Economics L.L.C.
19
Figure 9: Examples of Actual MBS Funding Structures
Source: ABSnet
Criterion Economics L.L.C. 20
The Complexity of Securitization
Point: Mortgages are complex and difficult to value, and MBS are built upon that complexity. Fundamental changes to underwriting and servicing standards are not easily identifiable in the inherent complexity of mortgages and MBS, posing risk to mortgage funding.
Criterion Economics L.L.C.
21
CDOs Add Complexity to MBS and Other Constituent Credit Instruments
Complexity of CDO Structures Growth of the CDO Market
Criterion Economics L.L.C.
22
Figure 10: Basic CDO Security Structure
Source: JP Morgan, CDO Handbook 5 (2001).
Criterion Economics L.L.C. 23
Figure 11: Typical Tranche Sizes and Coupon Rates
Tranche Class A Class B Class C Class D Preferred Shares Percent of Capital Structure 77.5 9 2.75 2.75 8 Rating AAA A BBB BB NR Coupon LIBOR + 26 LIBOR + 75 LIBOR + 180 LIBOR + 475 Residual Cash Flow
Source: Lucas, Goodman and Fabozzi (2005)
Criterion Economics L.L.C. 24
Key Differences between MBS and CDOs
CDO pools are actively managed. CDO transactions close before the pool of underlying assets is fully formed. CDOs are heterogeneous with respect to granularity. CDOs illustrate more ratings volatility than ABS or MBS. CDO collateral heterogeneity increases opacity. Secondary market trading (liquidity) is limited.
Criterion Economics L.L.C.
25
Figure 12: Examples of Actual CDO Funding Structures
Source: Bennett and Mason (forthcoming 2007)
Criterion Economics L.L.C. 26
CDOs Add Complexity to MBS and Other Constituent Credit Instruments
Complexity of CDO Structures Growth of the CDO Market
Criterion Economics L.L.C.
27
Figure 13: Annual Cash CDO Issuance
500 450 400
350
300
250
200
150
100
50
0 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005
Source: Lucas, Goodman, and Fabozzi (2006)
Criterion Economics L.L.C. 28
Fast Growth: Lessons from 1999-2003
The CDO market is opportunistic in the way it drops collateral types that are out of favor with investors and picks up collateral types that are in favor with investors. The best example of this is the switch out of poor-performing high-yield bonds and into well-performing high yield loans between 2001 and 2003. Also, certain types of ABS present in SF CDOs from 1999 through 2001 disappeared from later vintages: manufactured housing loans, aircraft leases, franchise business loans, and 12b-1 mutual fund fees. All of these assets had horrible performance in older SF CDOs. In their place, SF CDOs have recently focused more on RMBS and CMBS. (Lucas, Goodman and Fabozzi 2006, at 5).
Criterion Economics L.L.C.
29
The Link from Mortgages to MortgageBacked Securities to Collateralized Debt Obligations
CDO rating changes follow other ratings changes. An extremely high proportion of current CDO collateral is MBS. 1999-2003 redux?
Criterion Economics L.L.C.
30
Figure 14: Number of CDO, RMBS, and ABS Ratings Changes, Three-month MA
Criterion Economics L.L.C.
31
The Link from Mortgages to MortgageBacked Securities to Collateralized Debt Obligations
CDO rating changes follow other ratings changes. An extremely high proportion of current CDO collateral is MBS. 1999-2003 redux?
Criterion Economics L.L.C.
32
How much and what kind of MBS are in CDOs?
The FDIC reports that 81 percent of the $249 billion of CDO collateral pools issued in 2005, or $200 billion, was made up of residential mortgage products. (FDIC Outlook, Fall 2006)
Criterion Economics L.L.C.
33
How much and what kind of MBS are in CDOs?
The FDIC reports that 81 percent of the $249 billion of CDO collateral pools issued in 2005, or $200 billion, was made up of residential mortgage products. (FDIC Outlook, Fall 2006) Moody’s reports that among RMBS, 70%-75% below AAArated
Table 3: Top Collateral Types in Resecuritization CDOs, 2005 (All Vintages)
Criterion Economics L.L.C.
34
How much and what kind of MBS are in CDOs?
The FDIC reports that 81 percent of the $249 billion of CDO collateral pools issued in 2005, or $200 billion, was made up of residential mortgage products. (FDIC Outlook, Fall 2006) Moody’s reports that among RMBS in CDOs, 70%-75% below AAA-rated. (Moody’s CDO Asset Exposure Report, October 2006) $200 billion of RMBS in CDOs x 70% below AAA = $140 billion of lower-tranche RMBS in CDOs
Criterion Economics L.L.C.
35
How much and what kind of MBS are in CDOs?
The FDIC reports that 81 percent of the $249 billion of CDO collateral pools issued in 2005, or $200 billion, was made up of residential mortgage products. (FDIC Outlook, Fall 2006) Moody’s reports that among RMBS in CDOs, 70%-75% below AAA-rated. (Moody’s CDO Asset Exposure Report, October 2006) $200 billion of RMBS in CDOs x 70% below AAA = $140 billion of lower-tranche RMBS in CDOs SIFMA puts total RMBS issued in 2005 at $1,326 billion.
Criterion Economics L.L.C.
36
How much and what kind of MBS are in CDOs?
The FDIC reports that 81 percent of the $249 billion of CDO collateral pools issued in 2005, or $200 billion, was made up of residential mortgage products. (FDIC Outlook, Fall 2006) Moody’s reports that among RMBS in CDOs, 70%-75% below AAA-rated. (Moody’s CDO Asset Exposure Report, October 2006) $200 billion of RMBS in CDOs x 70% below AAA = $140 billion of lower-tranche RMBS in CDOs SIFMA puts total RMBS issued in 2005 at $1,326 billion. Assuming 90% AAA in all 2005 RMBS, that leaves a total of $133 billion in lower-tranche RMBS issued in 2005
Criterion Economics L.L.C.
37
The Link from Mortgages to MortgageBacked Securities to Collateralized Debt Obligations
CDO rating changes follow other ratings changes. An extremely high proportion of current CDO collateral is MBS. 1999-2003 redux?
Criterion Economics L.L.C.
38
1999-2003 Redux (but with MBS)?
Point: CDOs are buying a lot of lowertranche RMBS Point Point: The 10% of lower-tranche RMBS provide crucial credit support for the 90% of AAA securities.
If you can’t sell the 10%, can’t sell the other 90%!
Point Point Point: If CDOs exit RMBS as they did with other collateral after 1999-2003, that will leave a substantial hole in the market for crucial lower-tranche RMBS.
Criterion Economics L.L.C. 39
Table 5: CDO Risk Premiums and Credit Spreads and Macroeconomic Performance
Source: D’Amato (2005)
Criterion Economics L.L.C. 40
Figure 16: CDO Issuance and U.S. Home Equity ABS Indices
Source: BIS Quarterly Review, Dec. 2006
Criterion Economics L.L.C. 41
Overview of Public Policy Implications: Information Problems across Financial Markets CDO Market and Institutions Want to move is here
it this way
Adapted from Merton, “A Functional Perspective of Financial Intermediation,” Financial Management, Summer 1995. Criterion Economics L.L.C. 42
Policy Implications
Transparency Stability Continued Innovation
Criterion Economics L.L.C.
43
Policy Implications
MBS Pioneer Lewis S. Ranieri, in an interview published in the American Banker last June, said: "When you start divorcing the creator of the risk from the ultimate holder of the risk, it becomes an issue of, 'Does the ultimate holder truly understand the nature of the risk that you've redistributed?' " Mr. Ranieri said. "By cutting it up in so many ways or complicating it by so many levels, do you still have clarity ... on the nature of the underlying risk?" "It's not clear that we haven't gone, in some ways, too far -- that we have not gone beyond the ability to have true transparency," he said. "That is a fair question many of us in the business and people in the regulatory regime are wrestling with."
Criterion Economics L.L.C. 44