The Housing Crisis: By T2 Partners

An Overview Of The Housing Crisis And Why There Is More Pain To Come T2 Accredited Fund, LP Tilson Offshore Fund, Ltd. T2 Qualified Fund, LP July 1, 2009

T2 Partners Management L.P. Is a Registered Investment Advisor 145 E. 57th Street, 10th Floor New York, NY 10022 (212) 386-7160 [email protected] www.T2PartnersLLC.com

The collapse of the U.S. housing market, the world’s largest debt market, is the defining economic event of our lifetimes. This presentation explores what happened and why, where we are today, and what the future likely holds. For more detailed information, please read our new book, More Mortgage Meltdown: 6 Ways to Profit in These Bad Times.

More Mortgage Meltdown Was Released in May

Tilson Mutual Funds

The Tilson Mutual Funds (www.tilsonmutualfunds.com) follow a longterm-oriented, value-based investment strategy, seeking to identify stocks that are trading at a substantial discount to their intrinsic value. The Tilson Focus Fund (TILFX) seeks maximum capital appreciation, while the Tilson Dividend Fund (TILDX), while also seeking capital appreciation, seeks to generate current income from dividends and selling covered calls. An investor should consider the investment objectives, risks, and charges and expenses of the Funds before investing. The prospectus contains this and other information about the Funds. A copy of the prospectus is available at www.tilsonmutualfunds.com/tilsonprospectus.pdf or calling the fund directly at 888-484-5766. The prospectus should be read carefully before investing. Investment in the Funds is subject to investment risks, including, without limitation, market risk, management style risk, sector focus risk, foreign securities risk, nondiversified fund risk, portfolio turnover risk, credit risk, interest rate risk, maturity risk, investment-grade securities risk, junk bonds or lower-rated securities risk, derivative instruments risk and real estate securities risk. The Tilson Mutual Funds are distributed by Capital Investment Group, Inc., Member FINRA/SIPC, 17 Glenwood Ave, Raleigh, NC, 27603. There is no affiliation between the Tilson Mutual Funds, including its principals, and Capital Investment Group, Inc.

Attend the Next Value Investing Congress on October 19-20 in New York City Register at www.valueinvestingcongress.com

Subscribe to Value Investor Insight and SuperInvestor Insight

Table of Contents 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

What happened Why did it happen? Background on the U.S. housing market Two waves of losses are behind us…but three are looming Current economic situation Interest rates Underwater homeowners Recent signs of stabilization are likely the mother of all head fakes The outlook for home prices Detailed Information on five types of mortgages: prime, Alt-A, jumbo prime, CES/HELOCs and Option ARMs 11. More on home prices 12. Total losses 13. Where we are finding opportunities Appendices • More background data on the housing market • Case study: Sacramento, CA • A closer look at mortgages that were securitized: quantity and quality • A closer look at mortgages that were securitized: defaults • Where did the securitized mortgages end up? A primer on ABSs and CDOs • What should the government do in the case of distressed financial institutions

Page 9 Page 16 Page 28 Page 41 Page 43 Page 51 Page 54 Page 55 Page 57 Page 58 Page 81 Page 92 Page 97 Page 99 Page 107 Page 115 Page 128 Page 139 Page 143

What Happened?

For the Second Half of the 20th Century, Housing Was a Stable Investment 300 Shiller Lawler Real Home Price Index (1890=100)

275 250 225 200 175

Trend Line 150 125

98 19

94 19

90 19

86 19

82 19

78 19

74 19

70 19

66 19

62 19

58 19

54 19

19

50

100

Sources: Robert J. Shiller, Irrational Exuberance, Princeton University Press 2000, Broadway Books 2001, 2nd edition, 2005, also Subprime Solution, 2008, as updated by the author at http://www.econ.yale.edu/~shiller/data.htm; Lawler Economic & Housing Consulting.

10

…And Then Housing Prices Exploded

300 Shiller Lawler Real Home Price Index (1890=100)

275 250 225

Housing Bubble

200 175

Trend Line 150 125

20 06

02 20

19 98

94 19

19 90

86 19

19 82

78 19

19 74

70 19

19 66

62 19

58 19

4 19 5

19 50

100

Sources: Robert J. Shiller, Irrational Exuberance: Second Edition, as updated by the author; Lawler Economic & Housing Consulting. 11

Prices Exploded Because the Borrowing Power of a Typical Home Purchaser More Than Tripled from 2000-2006 $400,000 Pre-Tax Income Borrowing Power

$300,000

$200,000

Factors contributing to the ability to borrow more and more were: 1. Lenders were willing to allow much higher debt-to-income ratios 2. Interest-only mortgages (vs. full amortizing) 3. No money down 4. Low- and no-documentation loans, which led to widespread fraud 5. Slowly rising income 6. Falling interest rates

9.2x in January 2006

$100,000

3.3x in January 2000 $0 Jan-95

Source: Amherst Securities.

Jan-96

Jan-97

Jan-98

Jan-99

Jan-00

Jan-01

Jan-02

Jan-03

Jan-04

Jan-05

Jan-06

Jan-07

Jan-08

12

Americans Have Borrowed Heavily Against Their Homes Such That the Percentage of Equity Has Fallen Below 50% for the First Time There still appears to be substantial equity protecting the debt – but this is misleading. Given that 1/3 of homes have no mortgage (i.e., 100% equity), the remaining homes have only 12% equity. A further 10% decline in home prices would, on average, leave no equity cushion left to support U.S. mortgage debt.

90%

$12,000

80% $10,000

60%

$8,000

$6,000

1945 Mortgage Debt: $18.6 billion Equity: $97.5 billion

$4,000

50% Q109 Mortgage Debt: $10.5 trillion Equity: $7.4 trillion

40% 30%

Equity as a % of Home Value

Mortgage Debt (Bn)

70%

20% $2,000 10% $0

0% 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

Source: Federal Reserve Flow of Fund Accounts of the United States. 13

Housing Became Unaffordable in Many Areas Using a Typical 30-Year Fixed-Rate Mortgage, Which Led Many Borrowers to Take Exotic Mortgages 80 Riverside, CA Los Angeles, CA San Diego, CA

70

Housing Opportunity Index

60 50 40 30 20 10

07 20

06 3 Q

Q

3

20

20 05 Q

3

20 04 Q

3

20 02 Q 1

1 20 0 Q 1

00 Q 1

20

99 Q 1

19

98 19 Q 1

19 Q 1

Q

1

19

96

97

0

Sources: NAHB/Wells Fargo Housing Opportunity Index, which measures percentage of households that could afford the average home with a standard mortgage. 14

There Was a Dramatic Decline in Mortgage Lending Standards from 2001 through 2006 Combined Loan to Value

100% Financing

86

18% 17% 84

84

16% 83

81

14%

14%

81

80

12% Percent of Originations

Combined Loan to Value (%)

82

78 76 76 74 74

74

10% 9% 8% 8%

6%

72

4%

70

2%

68

0% 2001

2002

2003

2004

2005

2006

2007

3%

1%

1%

2001

2002

2003

2004

2005

2007

100% Financing & Limited Doc

Limited Documentation

12%

70% 65%

11%

63% 60%

10%

56%

49%

50% 45%

39%

40%

8%

8% Percent of Originations

Percent of Originations

2006

33% 30%

6% 5% 4% 4%

20%

2%

1%

10%

0%

0%

2001

2002

0%

0% 2001

2002

2003

2004

2005

2006

2007

2003

2004

2005

2006

2007

Sources: Amherst Securities, LoanPerformance; USA Today (www.usatoday.com/money/economy/housing/2008-12-12-homeprices_N.htm).

• In 2005, 29% of new mortgages were interest only — or less, in the case of Option ARMs — vs. 1% in 2001 • In 1989, the average down payment for firsttime home buyers was 10%; by 2007, it was 2% • The sale of new homes costing $750,000 or more quadrupled from 2002 to 2006. The construction of inexpensive homes costing $125,000 or less fell by two-thirds 15

Why Did It Happen?

Among the Many Causes of The Great Housing Bubble, Two Stand Out 1.

The lenders making crazy loans didn’t care if the homeowner ended up defaulting for two reasons: –

–

They didn’t hold the loan, but instead sold it to someone else. It was eventually bought by a Wall Street firm, which packaged it with thousands of other mortgages in a Residential Mortgage-Backed Security (RMBS), which was then sliced into numerous tranches that were sold to investors around the world. This entire process was extraordinarily profitable for all involved, especially the Wall Street firms and the rating agencies; Or, if they did plan to hold the loan, they assumed home prices would keep rising, such that homeowners could either refinance before loans reset or, if the homeowner defaulted, the losses (i.e., severity) would be minimal.

2.

The entire system – real estate agents, appraisers, mortgage lenders, banks, Wall St. firms and rating agencies – became corrupted by the vast amounts of quick money to be made There were many other reasons, of course – a bubble of this magnitude requires what Charlie Munger calls “lollapalooza effects” –

– – – – –

Regulators and politicians were blinded by free market ideology and/or the dream that all Americans should own their homes, causing them to fall asleep at the switch, not want to take the punch bowl away and/or get bought off by the industries they were supposed to be overseeing Debt became increasingly available and acceptable in our culture Millions of Americans became greedy speculators and/or took on too much debt Greenspan kept interest rates too low for too long Institutional investors stretched for yield, didn’t ask many questions and took on too much leverage In general, everyone was suffering from irrational exuberance, driven by the money being made

17

When Home Price Appreciation Slows, Loss Severity Skyrockets

The assumption of perpetually high HPA led lenders to give virtually anyone a loan because even if they defaulted, the home could simply be resold with little or no loss.

Sources: LoanPerformance; OFHEO; Deutsche Bank; “Who's Holding the Bag?”, Pershing Square presentation, 5/23/07.

As Long As Home Prices Rise Rapidly, Even Pools of Bubble-Era Subprime Mortgages Perform Well – But If Home Prices Fall, Look Out Below! Cumulative Five-Year Loss Estimates for a Bubble-Era Pool of Subprime Mortgages 60%

50%

Cumulative Loss (%)

40%

30%

20%

10%

0% 20%

15%

10%

5%

0%

-5%

-10%

-15%

-20%

-25%

-30%

-35%

-40%

Home Price Appreciation

Source: T2 Partners estimates. 19

The Enormous Amounts of Money to Be Made Corrupted Our Financial System The #1 Immutable Law of the Universe If you offer people a lot of money to do something, no matter how foolish, unethical or illegal, a large number of them will do it – Corollary #1: The more money to be made, the more bad behavior that will occur – Corollary #2: The people engaged in such behavior will rationalize it such that they genuinely believe that what they’re doing isn’t foolish, unethical or illegal

The #2 Immutable Law of the Universe Bad behavior leads to bad consequences

Deregulation of the Financial Sector Led to a Surge of Compensation, Leverage and Profits Ratio of Financial Services Wages to Nonfarm Private-Sector Wages, 1910-2006

Source: Ariell Reshef, University of Virginia; Thomas Philippon, NYU; Wall St. Journal, 5/14/09.

21

Over the Past 30 Years, We Have Become a Nation Gorged in Debt – To The Benefit of Financial Services Firms Low Debt Era

Rising Debt Era

2.5%

350%

300%

2.0% 250%

Total Debt

1.5%

Financial Profits

200%

1.0% 0.5% 0.0% Dec- 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 96 99 02 05

150%

Total Debt as Percent of GDP

Financial Profits as Percent of GDP

3.0%

100%

Sources: Federal Reserve, BEA, as of Q2 2007, GMO presentation. 22

Profits and Wages in the Financial Sector Soared from 1980 Onward

Source: Moody’s Economy.com, NY Times, 12/18/08.

Wall Street Firms Were Making a Fortune Securitizing Loans • Among the most profitable areas for Wall Street firms was producing Asset-Backed Securities (ABSs) and Collateralized Debt Obligations (CDOs) • To produce ABSs and CDOs, Wall Street needed a lot of loan “product” • Mortgages were a quick, easy, big source • It is easy to generate higher and higher volumes of mortgage loans: simply lend at higher loan-to-value ratios, with ultra-low teaser rates, to uncreditworthy borrowers, and don’t bother to verify their income and assets (thereby inviting fraud) • There’s only one problem: DON’T EXPECT TO BE REPAID! 24

A Case Study of Wall Street Compensation Run Amok: Stan O’Neal, Dow Kim & the Mortgage Team at Merrill Lynch

Source: On Wall Street, Bonuses, Not Profits, Were Real, NY Times, 12/18/08.

The Rating Agencies Were Making a Fortune Rating Structured Finance Products Moody’s Stock Price Reflected Its Surge in Profits $70.00

$60.00

$50.00

$40.00

$30.00

$20.00

$10.00 2003

Source: Yahoo! Finance.

2004

2005

2006

Source: Federal Reserve Bank of St. Louis. Ja n09

Ja n08 Ju l- 0 8

Ja n07 Ju l- 0 7

Ja n06 Ju l-0 6

Ja n05 Ju l-0 5

Ja n04 Ju l- 0 4

Ja n03 Ju l- 0 3

Ja n02 Ju l-0 2

Ja n01 Ju l-0 1

Ja n00 Ju l-0 0

Ja n99 Ju l- 9 9

Interest Rate (%)

Greenspan Kept Interest Rates Too Low for Too Long, Which Fueled the Housing Bubble

7

6

5

4

3

2

1

0

Background on the U.S. Housing Market

There Was a Surge of Toxic Mortgages From 2000 to Mid-2007 $4,000

$3,500

Originations (Bn)

$3,000

Conforming, FHA/VA Jumbo Alt-A Subprime Seconds

$2,500

$2,000

$1,500

$1,000

$500

$0 1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Source: Inside Mortgage Finance, published by Inside Mortgage Finance Publications, Inc. Copyright 2009. 29

Private Label Mortgages (Those Securitized by Wall St.) Are 15% of All Mortgages, But Account for 51% of Seriously Delinquent Mortgages Approximately two-thirds of homes have mortgages and, of these, 56% are owned or guaranteed by the two government-sponsored enterprises (GSEs), Fannie & Freddie

Number of Mortgages (million)

Number of Seriously Delinquent Mortgages (000) Banks & Thrifts 397

Banks & Thrifts 8 Fannie Mae 18

15%

Fannie Mae 444 Freddie Mac 232

Private Label 8

Ginne Mae/FHA 378

Ginne Mae/FHA 6

Private Label 1734

Freddie Mac 13

51% Source: Freddie Mac, Q4 2008. 30

More Than 9% of Mortgages on 1-to-4 Family Homes Were Delinquent or in Foreclosure as of Q1 2009

10.0% 9.0%

Percentage of Home Loans

8.0% 7.0% 6.0% 5.0% 4.0% 3.0% 2.0% 1.0%

Q 4 Q 19 4 7 Q 19 9 4 80 Q 19 4 8 Q 19 1 4 82 Q 19 4 8 Q 19 3 4 84 Q 19 4 8 Q 19 5 4 86 Q 19 4 87 Q 19 4 88 Q 19 4 89 Q 1 99 4 0 Q 19 4 91 Q 19 4 92 Q 19 4 93 Q 19 4 9 Q 19 4 4 95 Q 19 4 9 Q 19 6 4 97 Q 19 4 9 Q 19 8 4 99 Q 20 4 0 Q 20 0 4 01 Q 20 4 0 Q 20 2 4 03 Q 20 4 0 Q 20 4 4 05 Q 20 4 06 Q 2 00 4 7 20 08

0.0%

Source: National Delinquency Survey, Mortgage Bankers Association; T2 Partners estimates. Note: Delinquencies (60+ days) are seasonally adjusted. 31

All Types of Loans Are Seeing a Surge in Delinquencies, Led by Subprime 45% 40%

Percent Noncurrent

35%

Alt A Option ARM Jumbo Subprime Prime Home Equity Lines of Credit

30% 25% 20% 15% 10% 5%

Q 1

19 Q 99 3 19 Q 99 1 20 Q 00 3 20 Q 00 1 20 Q 01 3 20 Q 01 1 20 Q 02 3 20 Q 02 1 20 Q 03 3 20 Q 03 1 20 Q 04 3 20 Q 04 1 20 Q 05 3 20 Q 05 1 20 Q 06 3 20 Q 06 1 20 Q 07 3 20 Q 07 1 20 Q 08 3 20 08

0%

Sources: Amherst Securities, LoanPerformance; National Delinquency Survey, Mortgage Bankers Association; FDIC Quarterly Banking Profile; T2 Partners estimates. Note: Prime is seasonally adjusted.

32

Foreclosure Filings Have Increased Dramatically

400,000 350,000

300,000

250,000 200,000

150,000

100,000

50,000 0 Ju n Au -05 gO 05 ct De 05 cFe 05 bAp 06 rJu 0 6 nAu 06 gO 06 ct D -06 ec F e 06 bAp 07 rJu 0 7 nA u 07 gO 07 ct D 07 ec Fe 07 bA p 08 rJu 0 8 n Au -08 gO 08 ct De 08 cFe 08 bAp 09 r- 0 9

•

Foreclosures in April rose 32% year-over-year, but only 1% sequentially April was the highest monthly total since RealtyTrac began issuing its report in January 2005 despite a decrease in bank repossessions (REOs) RealtyTrac estimates that over 1.5 million bank-owned properties are on the market, representing around a third of all properties for sale in the U.S.

Number of Foreclosures

• •

Note: Foreclosure filings are defined as default notices, auction sale notices and bank repossessions. Source: RealtyTrac.com U.S. Foreclosure Market Report.

33

Credit Suisse Predicts More Than 6 Million Additional Foreclosures by the End of 2012

Sources: Credit Suisse, 9/08; http://calculatedrisk.blogspot.com

Existing Homes Sales Are Falling and Foreclosures Are Rising, Leading to a Surge in Inventories

Months Supply

Existing Home Sales

12

7.5

11

7.0

4.0 million units, equal to 10.2 months as of the end of April 2009

10

6.5

6.0

Months

Millions

9

5.5

8 7 6

5.0 5

4.7 million units as of the end of April 2009

4.5

4

4.0 1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

3 1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Actual inventory levels are significantly higher due to “shadow inventory”, a combination of homes that banks own (REO) but have no yet listed for sale, as well as homeowners who want to sell but can’t because they’re underwater. Source: NATIONAL ASSOCIATION OF REALTORS® Existing Home Sales data series. 35

Inventory Levels Are Especially High for High-End Homes and Condos/Co-Ops Months Inventory in CA By Home Price

Another example: based on the rate of sales over the seven months ending in May 2009, there is five years of housing inventory in Greenwich, CT Sources: JP Morgan; CA Assoc. of Realtors (left chart); National Assoc. of Realtors via Moody’s Economy.com, WSJ 6/22/09 (right chart). 36

Nearly Half of Homes Sold Last Month Were Distressed Sales, Which Pressures Homes Prices

Sources: National Association of Realtors, NY Times, 6/9/09.

The Decline in Lending Standards Led to a Surge in Subprime Mortgage Origination

25%

$700

$600 20% 20%

20%

18%

% of T ota l

$500 Origina tions (Bn)

15% $400

$300 10%

9%

10%

10% 9%

9%

10%

10% 8% 7%

$200

8%

7%

5% $100

0%

$0 1994

1995

1996

1997

1998

1999

2000

2001

2002

Source: Reprinted with permission; Inside Mortgage Finance, published by Inside Mortgage Finance Publications, Inc. Copyright 2009.

2003

2004

2005

2006

2007

38

The Wave of Resets from Subprime Loans Is Mostly Behind Us $35

We are here

$30

Loans with Payment Shock (Bn)

$25

$20

$15

$10

$5

Sources: LoanPerformance, Deutsche Bank; slide from Pershing Square presentation, How to Save the Bond Insurers, 11/28/07.

Ju l-1 0 O ct -1 0

Ja n10 A pr -1 0

Ju l-0 9 O ct -0 9

O ct -0 8 Ja n09 Ap r09

Ap r-0 8 Ju l-0 8

Ju l-0 7 O ct -0 7 Ja n08

Ja n07 A pr -0 7

Ju l-0 6 O ct -0 6

Ja n06 A pr -0 6

$0

39

The Mortgage Meltdown Has Moved Beyond Subprime to Five Other Areas Prime Mortgage Commercial Real Estate Alt-A Other Corporate Commercial & Industrial Subprime High-Yield / Leveraged Loans Jumbo Prime Home Equity Credit Card Auto Option ARM Construction & Development Other Consumer CDO/ CLO

$0.0

$0.5

$1.0

$1.5

$2.0

$2.5

$3.0

$3.5

$4.0

$4.5

$5.0

Amount Outstanding (Trillions) Sources: Federal Reserve Flow of Funds Accounts of the United States, IMF Global Financial Stability Report October 2008, Goldman Sachs Global Economics Paper No. 177, FDIC Quarterly Banking Profile, OFHEO, S&P Leverage Commentary & Data, T2 Partners estimates.

40

Two Waves of Losses Are Behind Us… But Three Are Looming Losses Mostly Behind Us • Wave #1: Borrowers committing (or the victim of) fraud, as well as speculators, who defaulted quickly. Timing: beginning in late 2006 (as soon as home prices started to fall) into 2008. Mostly behind us. • Wave #2: Mostly subprime borrowers who defaulted when their mortgages reset due to payment shock. Timing: early 2007 (as twoyear teaser subprime loans written in early 2005 started to reset) to the present. Now tapering off as low interest rates mitigate payment shock. Losses Mostly Ahead of Us • Wave #3: Prime loans (most of which are owned or guaranteed by the GSEs) defaulting due to job loss and home price declines (i.e., underwater homeowners). Timing: started to surge in early 2008 to the present. • Wave #4: Jumbo prime, second lien and HELOCs (most of which are on banks’ books) defaulting due to job loss and home price declines/ underwater homeowners. Timing: started to surge in early 2008 to the present. • Wave #5: Losses among loans outside of the housing sector, the largest of which will be in the $3.5 trillion area of commercial real estate. Timing: started to surge in early 2008 to the present. 41

Future Losses Will Be Driven By Three Primary Factors 1. The economy •

Especially unemployment

2. Interest rates • •

Ultra-low rates have helped mitigate some of the damage But if the recent spike in rates continues, it could lead to an even greater surge in defaults and losses

3. Behavior of homeowners who are underwater • • • •

• •

Roughly one-fourth of homeowners with mortgages are currently underwater, some deeply so For many, it is economically rational for them to walk – leading to so-called “jingle mail” – but how many will actually do so? There is little historical precedent – we are in uncharted waters As home prices continue to fall and more and more homeowners become deeper and deeper underwater, they are obviously more likely to default, thereby creating a vicious cycle, but what exactly will the relationship be? Have millions of foreclosures led to a diminution of the stigma of defaulting and losing one’s home? Our best guess is that there will be rough symmetry: for homeowners 5% underwater, an additional 5% will default due to being underwater; 10% underwater will lead to 10% more defaults, and so forth… 42

Current Economic Situation

There Have Been 6 Million Jobs Lost So Far in This Recession, More Than 3 Million in the Past 6 Months

Change in Nonfarm Payroll Employment (000s)

600 400 200 0

-200 -400 -600 -800

There have been job losses every month since December 2007

Ja n9 Ja 0 n9 Ja 1 n9 Ja 2 n9 Ja 3 n9 Ja 4 n9 Ja 5 n9 Ja 6 n9 Ja 7 n9 Ja 8 n9 Ja 9 n0 Ja 0 n0 Ja 1 n0 Ja 2 n0 Ja 3 n0 Ja 4 n0 Ja 5 n0 Ja 6 n0 Ja 7 n0 Ja 8 n09

-1000

Source: Bureau of Labor Statistics. 44

The Unemployment Rate Jumped to 9.4% in May, the Highest Level Since 1983 If part-time and discouraged workers are factored in, the unemployment rate would have been 16.4% in May. In addition, the average work week was 33.1 hours, a record low. 11% 10%

Unemployment Rate

9% 8%

7% 6% 5% 4%

Source: Bureau of Labor Statistics.

Ja n09

Ja n03 Ja n06

Ja n97 Ja n00

Ja n94

Ja n88 Ja n91

Ja n82 Ja n85

Ja n79

Ja n73 Ja n76

Ja n70

3%

45

The Decline from Peak Employment Now Exceeds the Past Five Recessions 0.0% 1981 - 83

1980

1990 - 93

1974 - 76

-0.5%

2001 - 05

-1.0% -1.5% -2.0% -2.5% -3.0% -3.5% -4.0% 2007- present -4.5% 0

6

12

18

24

30

36

42

48

Months after pre-recession peak Source: Bureau of Labor Statistics 46

Consumer Confidence Rebounded in April

and May 160

140

Consumer Confidence Index

120

100

80

60

40

20

0 1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Note: 1985=100. Source: The Conference Board (www.pollingreport.com/consumer.htm) 47

Mortgage Equity Withdrawals of Roughly $400 Billion Annually During the Peak Bubble Years Fueled Unsustainable Consumer Spending

Source: www.calculatedriskblog.com/2009/05/mew-consumption-and-personal-saving.html.

Banks are Tightening Consumer Credit and New Household Borrowing Has Plunged

Percent of US Banks Tightening Consumer Credit 70% 60% 50% 40% 30% 20% 10% 0%

Household Borrowing 1990-2008 (Seasonally-Adjusted Annual Rate)

-10% $1,200

Credit Cards

Se p08

Ja n08

M ay -0 7

Se p06

Ja n06

M ay -0 5

Se p04

Ja n04

M ay -0 3

Se p02

Ja n02

M ay -0 1

($ billions)

Se p00

Ja n00

-20%

$1,000

Other Consumer Loans $800

$600

$400

$200

$0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: Federal Reserve Board. 49

The U.S. Savings Rate Has Spiked Upward in Recent Months to a 15-Year High This is good news in the long run, but could be a severe economic headwind in the short run, given that consumer spending is 2/3 of GDP

Source: Paul Kedrosky’s blog, 6/26/09; http://paul.kedrosky.com/archives/2009/06/the_black_swan.html.

Interest Rates

The Interest Rate on Conforming 30-Year FixedRate Mortgages Fell to All-Time Lows in Early 2009

10 Jumbo 30 Yr FRM 9

Jumbo 5/1 Hybrid ARM Conforming 30 Yr FRM Conforming 5/1 Hybrid ARM

8

10-Year Treasury

Rate (%)

7

6

5

4

3

Fe

b0 M 4 ay -0 Au 4 g0 N 4 ov -0 Fe 4 bM 05 ay -0 Au 5 g0 N 5 ov -0 Fe 5 bM 06 ay -0 Au 6 g0 N 6 ov -0 Fe 6 bM 07 ay Au 07 g0 N 7 ov -0 Fe 7 b0 M 8 ay -0 Au 8 g0 N 8 ov -0 Fe 8 bM 09 ay -0 9

2

Sources: HSH ASSOCIATES, Freddie Mac PMMS, Yahoo! Finance.

52

The Recent Spike in Treasury Yields Has Pushed Mortgages Rates Up, Which Crushes Refis and Threatens Any Hope of Stabilization of the Housing Market Roughly speaking, every 1% increase in mortgage rates reduces borrowing power by 10%

(Conventional or nonconforming)

Sources: HSHAssociates.com; Ryan ALM via WSJ Market Data Group; Mortgage Bankers Association via Thomson Reuters; in the Wall St. Journal, 6/11/09.

53

Underwater Homeowners

24% of Homeowners With a Mortgage Owe More Than the Home Is Worth, Making Them Much More Likely to Default Among people who bought homes in the past five years, 30%+ are underwater* In Bubble Markets, Far More Homeowners Are Underwater Price Drop Since Peak -15.2% -32.0% -21.8% -24.8% -36.6% -27.8% -10.4% -34.4% -37.7% -41.8%

* The actual figures are likely even worse, as this data doesn’t capture people who bought since 2003 and subsequently did a cash-out refi or after-the-fact second mortgage. 50% of all subprime and Alt-A loans in existence when the collapse happened were cash-out refis that carried a higher loan balance than the original purchase loan amount.

% of Last 5 Yrs Purchasers Who Are Under Water* 23.0% 56.4% 27.8% 25% 50.3% 65.1% 51.2% 23.2% 20% 63.9% 36.4% 61.4% Percent Underwater

Metro Area New York Los Angeles Boston Washington Miami San Francisco Atlanta San Diego Phoenix Las Vegas

Price Index Is at Lowest Level Since 2004-Q3 2003-Q4 2002-Q2 2004-Q1 2004-Q1 2003-Q3 2004-Q4 2002-Q4 2004-Q3 2003-Q4

There Has Been a Dramatic Rise in Homeowners Who Are Underwater 24%

20%

16%

15%

10%

6% 5%

4%

0% Dec-06

Dec-07

Sep-08

Source: Zillow.com Q4 08 Real Estate Market Report; Moody's Economy.com, First American CoreLogic, T2 Partners estimates

Dec-08

Mar-09

55

Certain Types of Loans Are Severely Underwater 80% 73% 70%

Percent Underwater

60% 50%

50% 45% 40%

30% 25% 20%

10%

0% Prime

Alt A

Subprime

Option ARM

Sources: Amherst Securities, LoanPerformance, Standard & Poor’s. 56

Recent Signs of Stabilization Are Likely the Mother of All Head Fakes Rather than representing a true bottom, recent signs of stabilization are likely due to five factors that are (or are likely to be) short-term: 1. Ultra-low interest rates 2. A shift from low-end to middle- and upper-end homes defaulting, which has the effect of raising average home prices – but it’s very bad news 3. Home sales and prices are seasonally strong in April, May and June due to tax refunds and the spring selling season 4. The new $8,000 tax credit for first-time homebuyers –

But this expires on November 30th

5. A temporary reduction in the inventory of foreclosed homes –

–

– –

Shortly after Obama was elected, his administration promised a new, more robust plan to stem the wave of foreclosures so the GSEs and many other lenders imposed a foreclosure moratorium Early this year, the Obama administration unveiled its plan, the Homeowner Affordability and Stabilization Plan, which is a step in the right direction – but even if it is hugely successful, we estimate that it might only save 20% of homeowners who would otherwise lose their homes The GSEs and other lenders are now quickly moving to save the homeowners who can be saved – and foreclose on those who can’t This is necessary to work our way through the aftermath of the bubble, but will lead to a surge of housing inventory later this year, which will further pressure home prices 57

Outlook for Housing Prices • •

We think housing prices will reach fair value/trend line, down 40% from the peak based on the S&P/Case-Shiller national (not 20-city) index, which implies a 5-10% further decline from where prices where as of the end of Q1 2009. It’s almost certain that prices will reach these levels The key question is whether housing prices will go crashing through the trend line and fall well below fair value. Unfortunately, this is very likely. In the long-term, housing prices will likely settle around fair value, but in the short-term prices will be driven both by psychology as well as supply and demand. The trends in both are very unfavorable – –

• • • •

•

Regarding the former, national home prices have declined for 33 consecutive months since their peak in July 2006 through April 2009 and there’s no end in sight, so this makes buyers reluctant – even when the price appears cheap – and sellers desperate. Regarding the latter, there is a huge mismatch between supply and demand, due largely to the tsunami of foreclosures. In March 2009, distressed sales accounted for just over 50% of all existing home sales nationwide – and more than 57% in California. In addition, the “shadow” inventory of foreclosed homes already likely exceeds one year and there will be millions more foreclosures over the next few years, creating a large overhang of excess supply that will likely cause prices to overshoot on the downside, as they are already doing in California.

Therefore, we expect housing prices to decline 45-50% from the peak, bottoming in mid-2010 We are also quite certain that wherever prices bottom, there will be no quick rebound There’s too much inventory to work off quickly, especially in light of the millions of foreclosures over the next few years While foreclosure sales are booming in many areas, regular sales by homeowners have plunged, in part because people usually can’t sell when they’re underwater on their mortgage and in part due to human psychology: people naturally anchor on the price they paid or what something was worth in the past and are reluctant to sell below this level. We suspect that there are millions of homeowners like this who will emerge as sellers at the first sign of a rebound in home prices Finally, we don’t think the economy is likely to provide a tailwind, as we expect it to contract the rest of 2009, stagnate in 2010, and only then grow tepidly for some time thereafter 58

Detailed Information on Five Types of Mortgages: Prime, Alt-A, Jumbo Prime, CES/HELOCs and Option ARMs

19 Q 99 3 19 Q 99 1 20 Q 00 3 20 Q 00 1 20 Q 01 3 20 Q 01 1 20 Q 02 3 20 Q 02 1 20 Q 03 3 20 Q 03 1 20 Q 04 3 20 Q 04 1 20 Q 05 3 20 Q 05 1 20 Q 06 3 20 Q 06 1 20 Q 07 3 20 Q 07 1 20 Q 08 3 20 08

Q 1

Percent Noncurrent (60+ days)

Delinquencies of Prime Mortgages Are Soaring 5.0%

4.5%

4.0%

3.5%

3.0%

2.5%

2.0%

1.5%

1.0%

0.5%

0.0%

Source: Mortgage Bankers Association National Delinquency Survey. 60

There Is a Surge of Notices of Default and Foreclosures Among the GSEs

Prime Notices of Default

Subprime Notices of Default

Prime Foreclosures

Subprime Foreclosures

Source: The Field Check Group. 61

15 States With the Highest Prime Mortgage Foreclosure Rates

Source: New York Times, 5/24/09. 62

Delinquencies of Prime and Alt-A Mortgages Are Soaring

Source: New York Times, 5/24/09. 63

There Are $2.4 Trillion of Alt-A Mortgages and Their Resets Are Mostly Ahead of Us $300

$10

We are here

$9

$6 $150

Amount (Bn)

$200

$7

$5 $4

$100

$3

$50

$2

Estimated Cumulative Reset Amount (Bn)

$250

$8

$1 $0

l-1 5 Ju

n15 Ja

4 l-1 Ju

n14 Ja

l-1 3 Ju

n13 Ja

2 l-1 Ju

n12 Ja

l- 1 1 Ju

Ja n11

0 l-1 Ju

Ja

n10

$0

Sources: Credit Suisse, LoanPerformance. NOTE: This chart only shows resets for a small fraction of Alt-A loans, but is representative of all of them.

64

n99 Ju l-9 Ja 9 n0 Ju 0 l-0 Ja 0 n01 Ju l-0 Ja 1 n0 Ju 2 l-0 Ja 2 n03 Ju l-0 Ja 3 n0 Ju 4 l-0 Ja 4 n05 Ju l-0 Ja 5 n0 Ju 6 l-0 Ja 6 n07 Ju l-0 Ja 7 n0 Ju 8 l-0 Ja 8 n09

Ja

Percent Noncurrent (60+ days)

Delinquencies of Securitized Alt-A Mortgages Are Soaring 25%

20%

15%

10%

5%

0%

Sources: Amherst Securities, LoanPerformance. 65

Alt-A Delinquencies By Vintage Show the Collapse in Lending Standards in 2006 and 2007 30% 2007

2006

Percent Noncurrent (60+ days)

25%

20%

15% 2005 10% 2004 5%

2003

0% 0

5

10

15

20

25

30

35

40

45

50

55

60

Months of Seasoning Sources: Amherst Securities, LoanPerformance. 66

n99 Ju l-9 Ja 9 n0 Ju 0 l-0 Ja 0 n01 Ju l-0 Ja 1 n02 Ju l-0 Ja 2 n03 Ju l-0 Ja 3 n04 Ju l-0 Ja 4 n05 Ju l-0 Ja 5 n06 Ju l-0 Ja 6 n07 Ju l-0 Ja 7 n08 Ju l-0 Ja 8 n09

Ja

Percent Noncurrent (60+ days)

Delinquencies of Securitized Jumbo Prime Mortgages Are Soaring 4.0%

3.5%

3.0%

2.5%

2.0%

1.5%

1.0%

0.5%

0.0%

Sources: Amherst Securities, LoanPerformance. 67

As the Credit Crisis Worsened, the Spread Between Jumbo and Conforming Loans Widened Dramatically

Sources: Mortgage Bankers Association via Thomson Reuters; HSHAssociates.com, in the WSJ 6/11/09. 68

HELOCs and Home Equity Loans Soared in Popularity During the Bubble $1,000 Closed-End Junior Lien Mortgages $900

Home Equity Lines of Credit

$800

Amount (Bn)

$700 $600 $500 $400 $300 $200 $100

08 20

07

1 Q

Q 1

20

06 Q 1

20

05 Q 1

20

4

20 03

20 0 Q 1

1 Q

Q 1

20 02

01 20 Q 1

Q 1

20

00

$0

Note: Does not include approximately $200 billion of securitized HELOCs and junior liens. Source: FDIC Quarterly Banking Profile. 69

Many Borrowers Used HELOCs to Buy New Cars • •

As home prices have declined and other funding sources have dried up, millions of consumers have maxed out on home equity debt. In hot markets like California and Florida, a significant percentage of all consumers tapped into the value of their homes to help finance their new cars, according to CNW Marketing Research.

•

Clearly this dynamic does not bode well for HELOC recovery rates or new car sales.

Source: New York Times 5/27/2008. 70

Delinquencies of HELOCs and CESs Are Soaring 4.5% Closed-End Junior Lien Mortgages Home Equity Lines of Credit

Percent Noncurrent (90+ days)

4.0% 3.5% 3.0% 2.5% 2.0% 1.5% 1.0% 0.5%

04 20 Q 04 4 2 Q 004 1 20 Q 05 2 2 Q 005 3 20 Q 05 4 20 Q 05 1 20 Q 06 2 2 Q 006 3 20 Q 06 4 2 Q 006 1 20 Q 07 2 20 Q 07 3 2 Q 007 4 20 Q 07 1 2 Q 008 2 20 Q 08 3 2 Q 008 4 20 Q 08 1 20 09 Q

3

20

2 Q

Q

1

20 0

4

0.0%

Source: FDIC Quarterly Banking Profile. 71

Pools of HELOCs and CESs Can Suffer Astronomical Losses Due to 100%+ Severities On one second lien deal, Ambac expected losses of 10-12% when it guaranteed the senior tranche. A year ago, Ambac admitted that the pool would likely lose 81.8% of its value – and based on the pool’s performance since then, this will almost certainly prove to be conservative. 3.0% Ambac Projection April 2008 Actual

Monthly Loss Rate (3m average)

2.5%

2.0%

1.5%

1.0%

From Ambac slide, April 2008: • Second lien deal that closed in April 2007 • Loss to date 9.9%; projected loss: 81.8% • Projected collateral loss as a % of current collateral: 86% • A reasonable estimate of projected collateral loss for the above transaction might have been 10-12%, with the transaction having an A+ rating at inception and being structured to withstand 28-30% collateral loss

0.5%

0.0% 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 Months Since Close Sources: Ambac Q1 08 presentation, Amherst Securities; funds managed by T2 Partners are short Ambac.

A Primer on Option ARMs

• • • •

• • • • •

An Option ARM is an adjustable rate mortgage typically made to a prime borrower Sold under various names such as “Pick-A-Pay” Banks typically relied on the appraised value of the home and the borrower’s high FICO score, so 83% of Option ARMs written in 2004-2007 were low- or no-doc (liar’s loans) Each month, the borrower can choose to pay: 1) the fully amortizing interest and principal; 2) full interest; or 3) an ultra-low teaser interest-only rate (typically 2-3%), in which case the unpaid interest is added to the balance of the mortgage (meaning it is negatively amortizing) Approximately 80% of Option ARMs are negatively amortizing Lenders, however, booked earnings as if the borrowers were making full interest payments A typical Option ARM is a 30- or 40-year mortgage that resets (“recasts”) after five years, when it becomes fully amortizing If an Option ARM negatively amortizes to 110-125% of the original balance (depending on the terms of the loan), this triggers a reset even if five years have not elapsed Upon reset, the average monthly payment can jump significantly, though the payment shock is currently mitigated by today’s low interest rates 73

Further Details on Option ARMs From Washington Mutual’s 2007 10K (emphasis added): “The Option ARM home loan product is an adjustable-rate mortgage loan that provides the borrower with the option each month to make a fully-amortizing, interest-only, or minimum payment. As described in greater detail below, the minimum payment is typically insufficient to cover interest accrued in the prior month and any unpaid interest is deferred and added to the principal balance of the loan. The minimum payment on an Option ARM loan is based on the interest rate charged during the introductory period. This introductory rate has usually been significantly below the fully-indexed rate. The fully-indexed rate is calculated using an index rate plus a margin. Once the introductory period ends, the contractual interest rate charged on the loan increases to the fully-indexed rate and adjusts monthly to reflect movements in the index. If the borrower continues to make the minimum monthly payment after the introductory period ends, the payment may not be sufficient to cover interest accrued in the previous month. In this case, the loan will "negatively amortize" as unpaid interest is deferred and added to the principal balance of the loan. The minimum payment on an Option ARM loan is adjusted on each anniversary date of the loan but each increase or decrease is limited to a maximum of 7.5% of the minimum payment amount on such date until a "recasting event" occurs. A recasting event occurs every 60 months or sooner upon reaching a negative amortization cap. When a recasting event occurs, a new minimum monthly payment is calculated without regard to any limits on the increase or decrease in amount that would otherwise apply under the annual 7.5% payment cap. This new minimum monthly payment is calculated to be sufficient to fully repay the principal balance of the loan, including any theretofore deferred interest, over the remainder of the loan term using the fullyindexed rate then in effect. A recasting event occurs immediately whenever the unpaid principal balance reaches the negative amortization cap, which is expressed as a percent of the original loan balance. Prior to 2006, the negative amortization cap was 125% of the original loan balance... For all Option ARM loans originated in 2006, the negative amortization cap was 110% of the original loan balance. For Option ARM loans originated in 2007, the negative amortization cap was raised to 115%... In the first month that follows a recasting event, the minimum payment will equal the fully-amortizing payment.

About $750 Billion of Option ARMs Were Written, Nearly All at the Peak of the Bubble 9%

$300 9%

8% 8%

$250

7% 6% 5%

5% 5%

$150

4%

Percent of Total

Originations (Bn)

$200

3%

$100

2% $50 1% $0

1% 0%

2004

2005

2006

2007

2008

Sources: 2008 Mortgage Market Statistical Annual, published by Inside Mortgage Finance Publications, Inc. Copyright 2008. T2 Partners estimates. 75

Options ARMs Were a Bubble State Phenomenon

Other 25%

Arizona 3% Nevada 3%

California 58%

Florida 10%

Sources: Amherst Securities, LoanPerformance. 76

Beginning in March 2005, High-FICO-Score Borrowers Opted for an Above-Market-Rate Option ARM in Exchange for the Low Teaser Rate 8.5 Fannie Mae 30 Year FRM Index Option ARM Index 8.0

7.5

Interest Rate (%)

7.0

Nearly all Option ARM borrowers during this period (when nearly all Option ARMs were written) can’t afford a fullyamortizing mortgage – otherwise they would have taken one

6.5

6.0

5.5

5.0

4.5

Ja n02 A pr -0 2 Ju l-0 2 O ct -0 2 Ja n03 A pr -0 3 Ju l-0 3 O ct -0 3 Ja n04 Ap r-0 4 Ju l-0 4 O ct -0 4 Ja n05 A pr -0 5 Ju l-0 5 O ct -0 5 Ja n06 A pr -0 6 Ju l-0 6 O ct -0 6 Ja n07 A pr -0 7 Ju l-0 7 O ct -0 7 Ja n08

4.0

Source: Amherst Securities, BloombergFinance, L.P.

Option ARMs are Recasting Much Faster Than Expected Due to Negative Amortization

$18 $16 $14

Original recast schedule (5 yrs from origination) Recast schedule based on current neg am

$12 $10

$8 $6 $4 $2

A

pr -0 Ju 8 n0 A 8 ug -0 O 8 ct0 D 8 ec -0 Fe 8 b0 A 9 pr -0 9 Ju n0 A 9 ug -0 O 9 ct0 D 9 ec -0 Fe 9 b1 A 0 pr -1 Ju 0 n1 A 0 ug -1 O 0 ct1 D 0 ec -1 Fe 0 b1 A 1 pr -1 Ju 1 n1 A 1 ug -1 O 1 ct1 D 1 ec -1 Fe 1 b1 A 2 pr -1 2 Ju n1 A 2 ug -1 2

$-

Source: Credit Suisse for BusinessWeek.

Delinquencies of Securitized Option ARMs Are Soaring 35%

Percent Noncurrent (60+ days)

30%

25%

20%

“My sense is that many Option ARM borrowers are in a worse position than subprime borrowers. They wind up owing more and the resets are more significant.” -- Kevin Stein, associate director of the California Reinvestment Coalition

15%

10%

5%

Ja

n99 Ju l-9 Ja 9 n0 Ju 0 l-0 Ja 0 n01 Ju l-0 Ja 1 n02 Ju l-0 Ja 2 n03 Ju l-0 Ja 3 n04 Ju l-0 Ja 4 n05 Ju l-0 Ja 5 n06 Ju l-0 Ja 6 n07 Ju l-0 Ja 7 n08 Ju l-0 Ja 8 n09

0%

Sources: Amherst Securities, LoanPerformance, T2 Partners estimates. 79

Comments From a Federal Senior Bank Examiner (3/08) “The next problem is with the Option ARM product. Approximately 80-90% are paying the minimum credit card payment and most loans are negatively amortizing. Here the payment shock is two-fold – rate and principal – and the increase in payments can be astronomical: 200% or higher, not the 10 to 100% that subprime has experienced. Also, the dollars exposed in Alt-A [many analysts group Option ARMs with Alt-A] are nearly 50% higher than subprime: the Alt-A average balance is $299,000 versus $181,000 for subprime. Also, 73% were underwritten with Low or No Doc. The Option ARM books of many lenders are already showing significant deterioration and they have not even recast yet. This is the next tsunami to hit the housing market. This will hit much higher price points $600,000 and above as this was the affordability product used by higher income/higher FICO score households to buy that dream home.”

More on Home Prices

Home Prices Are in an Unprecedented Freefall 220 S&P/Case-Shiller U.S. National Home Price Index S&P/Case-Shiller 20-City Composite OFHEO Purchase-Only Index NAR Median Sales Price of Existing Homes

200

180

160

140

120

3

Q

Q

1

20 0

0 20 Q 00 1 20 Q 01 3 20 Q 01 1 20 Q 02 3 20 Q 02 1 20 Q 03 3 20 Q 03 1 20 Q 04 3 20 Q 04 1 20 Q 05 3 20 Q 05 1 20 Q 06 3 20 Q 06 1 20 Q 07 3 20 Q 07 1 20 Q 08 3 20 Q 08 1 20 09

100

Sources: Standard & Poor’s, OFHEO Purchase-Only Index, NATIONAL ASSOCIATION OF REALTORS® Existing Home Sales data series. 82

Home Prices Need to Fall Another 5-10% to Reach Trend Line 300 Shiller Lawler Real Home Price Index (1890=100)

275 250 225

Housing Bubble

200 175

Trend Line 150 125

20 06

02 20

19 98

94 19

19 90

86 19

19 82

78 19

19 74

70 19

19 66

62 19

58 19

4 19 5

19 50

100

Sources: Robert J. Shiller, Irrational Exuberance: Second Edition, as updated by the author; Lawler Economic & Housing Consulting. 83

Tre nd Line

0.8 0.3

1.5 1.0

Tre nd Line

0.5 0.0

2.0 1.8

U.S. Dollar

U.K. Pound

1979-1992

1979-1985

1.6 1.4 1.2 1.0 0.8 79

81

83

2000

85

87

89

0.9 0.8 79

80

81

82

83

Gold

Crude Oil

1970-1999

1962-1999

80

1992-October 2008

1.5 1.0 0.5 0.0

Tre nd Line

2.4 2.0 1.6 1.2

1.4 1.3

92 94 96 98 00 02 04 06 08

Japanese Yen

Japanese Yen

1983-1990

1.2 1.1 1.0 0.9 0.8

Commodities 250

1.2 1.1 1.0 0.9 0.8 92

40 20 0

0 70

74

78

82

86

90

94

98

95

96

Cocoa 1970-1999

600 500

150 100 50 0

62 66 70 74 78 82 86 90 94 98

94

Nickel

Real Price

400

Real Price

Real Price

800

60

93

1979-1999

200

1200

1992-1998

1.4 1.3

83 84 85 86 87 88 89 90

84

Trend Line

0.8

81 83 85 87 89 91 93 95 97 99

Currencies

1.2 1.1 1.0

91

1600 Real Price

1.4 1.3

S&P 500

1981-1999

3.0 2.5 2.0

46 50 54 58 62 66 70 74 78 82

Cumulative Return

Cumulative Return

20 21 22 23 24 25 26 27 28 29 30 31

Cumulative Return

1.3

2.0

Japan vs. EAFE ex-Japan Detrended Real Price

1.8

2.5

Stocks

Cumulative Return

S&P 500 1946-1984

Relative Return

2.3

S&P 500 1920-1932 Detrended Real Price

Detrended Real Price

A Study of Bubbles Shows That All of Them Eventually Return to Trend Line

79 81 83 85 87 89 91 93 95 97

97

400 300 200 100 0 70 74

78 82 86 90

94 98

Source: GMO LLC. Note: For S&P charts, trend is 2% real price appreciation per year. Source: GMO. Data through 10//10/08. * Detrended Real Price is the price index divided by CPI+2%, since the long-term trend increase in the price of the S&P 500 has been on the order of 2% real.

84

The Biggest Danger is That Home Prices Overshoot on the Downside, Which Often Happens When Bubbles Burst S&P 500 1927-1954 2.50

Overrun: 59% Fair Value to Bottom: 1.5 Years Fair Value to Fair Value: 23 Years

2.00 1.75 1.50 1.25

S&P 500 1955-1986

1.00 2.25

0.75

2.00

0.50

-59%

0.25 0.00 1927

1930

1933

1936

1939

1942

1945

1948

1951

1954

Detrended Real S&P 500 Stock Price Index

Detrended Real S&P 500 Stock Price Index

2.25

Overrun: 45% Fair Value to Bottom: 7 Years Fair Value to Fair Value: 12 Years

1.75 1.50 1.25 1.00 0.75 0.50

-45% 0.25 0.00 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1976 1978 1980 1982 1984 1986

Source: GMO LLC, T2 Partners calculations. 85

In California, Prices Are Way Down, Driven By a Surge in the Number of Homes Sold Out of Foreclosure California $500

70%

60%

50% $300

40%

30%

$200

Foreclosure Resale %

Median Home Price (000s)

$400

20% $100 10%

ct -0 8 Ja n09 Ap r-0 9

O

l-0 8 Ju

r-0 8 Ap

7

-0 8 Ja n

ct -0

7

7 O

Ju l-0

r-0 Ap

07 Ja n-

ct -0 6 O

l-0 6 Ju

6

0% Ap r-0

Ja

n0

6

$0

Source: MDA Dataquick. Note: Includes new construction. 86

Case Study: San Diego County Resales

More than half of homes and condos sold in November in San Diego had been in foreclosure, contributing to an increase in home sales - - while prices fell 35%.

3,000

2,500

+42% Resale Homes

2,000 1,154

-12%

Normal Foreclosure

1,500

1,000

1,312

+232% 1,254 500 378

(22% Foreclosure)

(52% Foreclosure)

0 November 1/1/20072007 Source: San Diego Union-Tribune, MDA Dataquick. Note: Excludes new construction.

November 2008 1/1/2008

87

Home Prices Have Crashed Through Trend Line in California, But Stabilized in March and April

$600 Median Sales Price 4% Trend

Median Price ($000s)

$500

$400

$300

$200

$100

1 Ja n03 Ja n05 Ja n07 Ja n09

n0 Ja

-9 9 Ja n

5

97 Ja n-

n9 Ja

n93 Ja

9

91 Ja n-

n8 Ja

-8 7 Ja n

3

85 Ja n-

n8 Ja

-8 1 Ja n

Ja n-

79

$0

Source: California Association of REALTORS ® . All rights reserved. www.rebsonline.com, T2 Partners estimates. 88

The Housing Affordability Index Shows Houses Are Now Affordable

Mortgage Payment on Median Priced Home as % of Family Income

26

24

22

20

18 Before concluding that houses are cheap, however, there are three big caveats: first, low rates are only available to those who qualify for conforming mortgages, which doesn’t help millions of homeowners or potential homeowners who have spotty credit histories or are underwater on their current mortgages. Second, with low enough interest rates, almost anything looks affordable; if rates rise, houses won’t look so reasonably priced based on these metrics. Finally, in light of the severe economic downturn, average income may fall for quite some time.

16

Source: NATIONAL ASSOCIATION OF REALTORS® Housing Affordability Index.

09 20

08 20

07 20

06 20

05 20

04 20

03 20

02 20

01 20

00 20

99 19

98 19

97 19

96 19

95 19

94 19

93 19

92 19

91 19

90

19

19

89

14

89

The Home Price-to-Rent Ratio Has Returned to Normal Levels 27

Median Home Price to Median Gross Rent

25

23

21

19

17

15 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source NATIONAL ASSOCIATION OF REALTORS® Existing Home Sales data series, U.S. Census Bureau, T2 Partners estimates. 90

The Timing Indicates That We Are Still in the Middle Innings of the Bursting of the Great Housing Bubble • • • • • • •

Mortgage lending standards became progressively worse starting in 2000, but really went off a cliff beginning in early 2005 The worst loans were subprime ones, which generally had two-year teaser rates and are now defaulting at unprecedented rates Such loans made in Q1 2005 started to default in high numbers upon reset in Q1 2007, which not surprisingly was the beginning of the current crises The crisis has continued to worsen as even lower quality subprime loans made over the remainder of 2005 reset over the course of 2007, triggering more and more defaults It takes an average of 15 months from the date of the first missed payment by a homeowner to a liquidation (generally a sale via auction) of the home Thus, the Q1 2005 subprime loans that defaulted in Q1 2007 led to foreclosures and auctions in early 2008 Given that lending standards got much worse in late 2005 through 2006 and into the first half of 2007, and the many other types of loans that are now with longer reset dates that are now starting to default at catastrophic rates, there are sobering implications for expected defaults, foreclosures and auctions in 2009 and beyond, which promise to drive home prices down further

In summary, today we are only in the middle innings of an enormous wave of defaults, foreclosures and auctions that is hitting the United States. We predicted in early 2008 that it would get so bad that it would require large-scale federal government intervention – which has occurred, and we’re not finished yet. 91

Total Losses

Total Losses Are Now Estimated at $2.1-$3.8 Trillion – And Only a Fraction of This Has Been Realized To Date $4,000

$3,778

Corporate

$3,552 $3,500

Consumer Commercial Real Estate Residential Mortgages

$3,000

Amount (Bn)

$2,632 $2,500 $2,083 $2,000 $1,473 $1,500

$1,214

$1,000

GSEs Insurers

$500

Banks/ Brokers

$0 Goldman Sachs Jan 2009

Roubini Jan 2009

T2 Partners Mar 2009

IMF Apr 2009 Writedowns to Capital Raised Date

Sources: Goldman Sachs, International Monetary Fund, RGE Monitor, Bloomberg Finance L.P., T2 Partners estimates. 93

A Breakdown of Our Financial Sector Loss Estimates Amount (Bn) $0 CDO/ CLO Other Consumer Construction & Development Option ARM Auto Credit Card Home Equity Jumbo Prime High-Yield / Leveraged Loans Subprime Commercial & Industrial Other Corporate Alt-A Commercial Real Estate Prime Mortgage

Source: T2 Partners estimates.

$100

$200

$300

$400

$500

$600

$700

Total Estimated Financial Sector Losses = $3.8 trillion

$800

Institutions Have Been Able to Raise Capital to Mostly Keep Up With Writedowns, But This Will Likely Not Continue $1,500 Losses & Writedowns Capital Raised $1,250

Amount (Bn)

$1,000

$750

$500

$250

$0 Prior

Q3 2007 Q4 2007 Q1 2008 Q2 2008 Q3 2008 Q4 2008 Q1 2009 Q2 2009

Source: Bloomberg Finance L.P. 95

We Think We’re Likely in the Midst of a Secular Bear Market

Source: Ned Davis Research; WSJ Market Data Group; appeared in WSJ 6/16/09 96

Where We Are Finding Opportunities •

•

•

•

• •

Blue-chips. The stocks of some of the greatest businesses, with strong balance sheets and dominant competitive positions, are trading at their cheapest levels in years – due primarily to the overall market decline and weak economic conditions rather than any company-specific issues. In this category, we’d put Wal-Mart (which we own), Coca-Cola, McDonald’s, Altria, ExxonMobil, Johnson & Johnson, and Microsoft. Out of favor blue-chips. For somewhat more adventurous investors looking to buy great companies in the most out-of-favor sectors such as financials, retailers and healthcare, we own Berkshire Hathaway, Wells Fargo, American Express, Target and Pfizer. All are great businesses, but their stocks have suffered mightily thanks to the economic downturn. We think they’re good bets to rebound when things stabilize. Balance sheet plays. For investors who are comfortable with lower-quality businesses but want downside protection, there are many companies trading near or even below net cash on the balance sheet. Examples in our portfolio include digital media equipment company EchoStar Corp. and clothing retailer dELiA*s. Berkshire is the best of both worlds: a premier company but also a balance sheet play. Turnarounds. There are countless companies that have gotten clobbered by the economic downturn and are reporting dismal results – with stock prices to match. Investors in those that survive and return to anything close to former levels of profitability will be well rewarded – but picking these stocks isn’t easy. Among our holdings in this category are Wendy’s restaurants, Winn-Dixie supermarkets, Huntsman, a specialty chemical maker, Crosstex, a pipeline company, and Resource America, a specialty finance company. Special situations. This is somewhat of a catch-all category that, for us, includes Contango Oil & Gas, a stock that’s declined due to an aborted attempt to sell the company and the sharp drop in the price of natural gas. Mispriced options. Every once in a while we take a tiny position in a highly speculative situation – often where the stock price is below $1 – in which there’s a real chance that the outcome is zero, but also a decent chance, in our opinion, of making many multiples of our money. On an expected value basis, therefore, a small portfolio of such investments is attractive. Our holdings include General Growth Properties, TravelCenters of America, Ambassadors International, Borders Group and PhotoChannel Networks. 97

Appendix

More Background Data on the Housing Market

The Housing Bubble Helped Many People Achieve the Dream of Home Ownership – Which is Now Turning Into a Nightmare Percentage of Households Owning Homes

70 69 68 67

%

66 65 64 63 62 61

Source: Census Bureau.

20 07

20 05

20 03

20 01

19 99

19 97

19 95

19 93

19 91

19 89

19 87

19 85

19 83

19 81

19 79

19 77

19 75

19 73

19 71

19 69

19 67

19 65

60

Source: Census Bureau.

2008-3

2006-4

2005-1

2003-2

2001-3

1999-4

1998-1

1996-2

1994-3

1992-4

1991-1

1989-2

1987-3

1985-4

1984-1

1982-2

1980-3

1978-4

1977-1

1975-2

1973-3

1971-4

1970-1

1968-2

1966-3

1964-4

1963-1

1961-2

1959-3

1957-4

1956-1

Home Vacancies Are at an All-Time High Home Vacancy Rate 1965-2008

3.0%

2.5%

2.0%

1.5%

1.0%

0.5%

0.0%

Nearly 10% of Homes Built This Decade Are Vacant Vacancy Rate By Date of Construction 9.5%

2.2%

March 2000 or earlier Source: Census Bureau Source: Census Bureau.

April 2000 to present

Housing Starts, Completions and Sales Have Plunged to All-Time Lows 2000 Starts Comple tions N e w H ome s Sold

1800

1600

1400

1200

1000

800

600

400

200 1970

1973

1976

Source: Commerce Department.

1979

1982

1985

1988

1991

1994

1997

2000

2003

2006

2009

Unlike Past Housing Downturns, New Home Sales Have Fallen Far More Than Existing Home Sales, Leading to an Enormous Inventory Glut

Source: New York Times, 6/27/09.

So Far, Few Loan Modifications Are Working

Percent of Modified Loans 30+ Days Delinquent

70%

60%

Q1 2008 Q2 2008 Q3 2008 55.5% 49.7%

50% 43.0%

49.2%

44.3%

58.3%

55.9%

57.4%

53.4% 47.7%

58.7% 54.9%

53.0%

57.4%

50.9%

43.2%

40%

39.9% 33.6% 32.3%

30%

38.8%

33.5%

20% 18.2% 10%

0% 1

2

3

4 5 6 Months Since Modification

Source: Office of the Comptroller of the Currency and the Office of Thrift Supervision Mortgage Metrics

7

8

9

Most Loan Modifications Fail Because the Homeowner Is So Deep Underwater

Among foreclosures in California in May, the average amount owed was $412,000, yet the home was worth only $235,000. This means the homeowner was 43% underwater and the home price would have to rise 75% to equal the amount of the debt.

Source: The Field Check Group -- data provided by ForeclosureRadar.com.

Case Study: Sacramento, CA

Are We Seeing the Beginnings of a Bottom in Hard-Hit Markets?

Source: NY Times, 5/4/09. 108

Home Prices in Sacramento More Than Tripled in Six Years – And Have Now Fallen 47%

Source: Zillow.com. 109

The Vast Majority of Sacramento Homeowners Who Purchased During the Bubble Years Are Now Underwater

Source: Zillow.com. 110

In Sacramento County, Home Sales Have Rebounded – But Are Still Outweighed by Defaults

Monthly Notices-of-Default in Sacramento

Sources: MDA Dataquick; The Field Check Group -- data provided by ForeclosureRadar.com. Note: Includes new construction.

111

Home Prices Are Stabilizing in Sacramento Country, In Part Due to More Higher End Homes Being Sold Off

Sacramento House Prices at the Time of Foreclosure/REO

Sacramento Mix of Houses at the Time of Foreclosure/REO

Source: The Field Check Group -- data provided by ForeclosureRadar.com. 112

Comments From Mark Hanson (1) The Field Check Group, May 5, 2009 California housing – at the low end – is 'bottoming' mostly because: a) median prices are down 55% from their peak over the past two years, thereby making the low end affordable; b) foreclosures have temporarily been cut by 66% through moratoriums reducing supply; and c) demand is picking up going into the busy season. But the moratoriums are ending and the number of foreclosures in the pipeline is massive – they will start showing themselves as REO over the near to mid-term. The Obama plan held the foreclosure wave back, creating a huge backlog and now the servicers are testing hundreds of thousands of defaults against the new loss mitigation initiatives. We presently see the Notice of Defaults at record highs and Notice of Trustee Sales back up to nine-month highs – there is no reason for a loan to go to the Notice of Trustee Sale stage if indeed it wasn't a foreclosure. However, the new 'batch' are not only from the low end but a wide mix all the way up to several million dollars in present value. Because the majority of buyers are in ultra low and low-mid prices ranges, the supplydemand imbalance from foreclosures and organic supply will crush the mid-to-upper priced properties in 2009. We already have early seasonal hard data proving this. As the mid-to-upper end go through their respective implosions this year and the volume of sales in these bands increase as prices tumble, the mix shift will raise median and average house prices creating the ultimate in false bottoms. We also have data proving this phenomenon.

Comments From Mark Hanson (2) The Field Check Group, May 5, 2009 After a year or so the real pain will occur when the mid to upper bands are down 40% from where they are now, and the price compression has made the low to low-mid bands much less attractive – the very same bands that are so hot right now. Rents are tumbling and those that bought these properties for investment will be at risk of default (investors have been buying all the way down). Investors have just started to get taken to the woodshed from all of the supply and this will get much worse. Mid-toupper end rental supply is also flooding on the market making it much better to rent a beautiful million dollar house than putting $300,000 down and buying. After investors are punished -- and with move-up buyers gone for years – it will leave first-time homeowners to fix the housing market on their own. Good luck and good night. Five years from now when things look to be stabilizing, all of these terrible kick-the-can-down-the-road modifications that leave borrowers in 5-year-teaser, ultra-high-leverage, 150% LTV, balloon loans will start adjusting upward and it will be Mortgage Implosion 2.0. These loan mods will turn millions of homeowners into over-levered, underwater, renters and ensure housing is a dead asset class for years to come. Due to a confluence of events including a national foreclosure moratorium and near-zero sales in the mid to upper end during the off season, the broader housing data show signs of stabilization. Taken in context, it is a blip. There are no silver linings or green shoots in housing whatsoever other than by these first-time homeowners – former renters – who now find it cheaper to own than rent. This is a very good thing, but it only applies to a small segment of the population and will not be able to support the market. In addition, the first-time buyers who come out of the rental market put continuous pressure on rents. Our data shows that the mid-to-upper end housing market is on the precipice of the exact cliff that the market fell off of in 2007, led by new loan defaults. What happens to the economy when you hit the midto-upper end earners the same way the low-to-mid end was hit with the subprime implosion? We will find out soon enough. When we look back on housing at the end of 2009, anyone that made positive housing predictions this year will not believe how far off they were.

A Closer Look at Mortgages That Were Securitized: Quantity and Quality

Hundreds of Billions of Dollars of Mortgages Were Securitized, Many On Terms With No Historical Precedent Securitized First Liens – Origination Volume These are the worst loans: $828 billion worth

Green: Loans with historical precedent Yellow: Loans with limited historical precedent Red: Loans with no historical precedent

Source: Amherst Securities Group, L.P.

Tens of Billions of Dollars of 2nd Lien Mortgages Were Also Securitized, Many On Terms With No Historical Precedent Securitized Second Liens – Origination Volume Another $56 billion of even bigger problems

Green: Loans with historical precedent Yellow: Loans with limited historical precedent Red: Loans with no historical precedent

Source: Amherst Securities Group, L.P.

Volume of June 2005 Fixed Rate and 2/28 Full Doc Securitized Mortgage Loans Fixed Full Doc – June 2005 Production Total Volume: $ 8.1 billion Green: 70.0%; Yellow: 9.3%; Red: 5.4%

2/28 Full Doc – June 2005 Production Total Volume: $16.4 billion Green: 39.9%; Yellow: 25.2%; Red: 26.1% Loan-to-Value

FICO

Loan-to-Value

*

Note: Green: Loans with historical precedent; Yellow: Loans with limited historical precedent; Red: Loans with no historical precedent * 2-28 loans are those with two-year teaser interest rates that then reset, often to higher rates, causing payment shock and a surge in defaults. Because they offer the lowest monthly payments (for the first two years), they are generally the lowest-quality loans, preferred by speculators and the most over-stretched borrowers. Source: Amherst Securities Group, L.P.

Volume of June 2005 Fixed Rate and 2/28 Low Doc Securitized Mortgage Loans Fixed Low Doc – June 2005 Production Total Volume: $ 7.7 billion Green: 49.2%; Yellow: 25.8%; Red: 8.0%

Source: Amherst Securities Group, L.P.

2/28 Low Doc – June 2005 Production Total Volume: $14.1 billion Green: 17.0%; Yellow: 33.4%; Red: 31.1%

Origination Volume of Fixed Rate, Full Doc Securitized Mortgage Loans, January 2005 In the best category of loans (full doc, fixed rate), in January 2005, just before mortgage lending standards collapsed, nearly all securitized mortgages were green, meaning they had FICO and LTV characteristics with historical precedent.

Prime

Alt-A Subprime

Origination Volume of Fixed Rate, Full Doc Securitized Mortgage Loans, June 2005 Mortgage lending standards began to worsen by June 2005.

Origination Volume of Fixed Rate, Full Doc Securitized Mortgage Loans, January 2006 By January 2006, mortgage lending standards had deteriorated substantially, even more the best loans, with large percentages yellow and red, meaning they had FICO and LTV characteristics with little or no historical precedent.

Origination Volume of Fixed Rate, Full Doc Securitized Mortgage Loans, June 2006 By June 2006, mortgage lending standards had collapsed, even for the best loans, with large percentages yellow and red, meaning they had FICO and LTV characteristics with little or no historical precedent.

Origination Volume of 2/28, Low Doc Securitized Mortgage Loans, January 2005 For the worst category of loans (low/no doc with two-year teaser rates), mortgage lending standards were abysmal as early as January 2005 – and got worse from there.

Origination Volume of 2/28, Low Doc Securitized Mortgage Loans, June 2005

Origination Volume of 2/28, Low Doc Securitized Mortgage Loans, January 2006

Origination Volume of 2/28, Low Doc Securitized Mortgage Loans, June 2006 A very high percentage of these loans will never be repaid.

A Closer Look at Mortgages That Were Securitized: Defaults

Default Rates of June 2005 Fixed Rate and 2/28 Full Doc Securitized Mortgage Loans Fixed Full Doc – June 2005 Production Total Volume: $ 8.1 billion Green: 70.0%; Yellow: 9.3%; Red: 5.4%

Loan-to-Value

FICO

Loan-to-Value

2/28 Full Doc – June 2005 Production Total Volume: $16.4 billion Green: 39.9%; Yellow: 25.2%; Red: 26.1%

Unprecedented default rates – and lending standards got much worse subsequent to June 2005! Source: Amherst Securities, May 25th reports, reflecting payments through 4/30/09.

Default Rates of June 2005 Fixed Rate and 2/28 Low Doc Securitized Mortgage Loans Fixed Low Doc – June 2005 Production Total Volume: $ 7.7 billion Green: 49.2%; Yellow: 25.8%; Red: 8.0%

Loan-to-Value

FICO

Loan-to-Value

2/28 Low Doc – June 2005 Production Total Volume: $14.1 billion Green: 17.0%; Yellow: 33.4%; Red: 31.1%

Default rates are much higher for no/low doc “liars” loans Source: Amherst Securities, May 25th reports, reflecting payments through 4/30/09.

Monthly Default Rate for Fixed Rate Securitized Mortgage Loans (Green) Defaults are defined as loans that are 90 days or more delinquent. sTr measures the percentage of loans that become 90 days or more delinquent during the month, as a percentage of non-delinquent loans at the beginning of the month. This chart shows the performance of the very best (fixed rate, green) mortgages. Note that late 2004 and early 2005 vintage loans have sTrs of approximately 70 basis points, which translates into roughly an 8% cumulative default rate in one year, whereas more recent vintage loans are quickly spiking up to a 2% sTr, which translates into an 21.5% cumulative default rate in one year. 9/06 12/04

Source: Amherst Securities Group, L.P.

Monthly Default Rate for Fixed Rate Securitized Mortgage Loans (Yellow) In this chart, late 2004 and early 2005 vintage loans have sTrs of approximately 1.5%, which translates into a 16.6% cumulative default rate in one year, whereas more recent vintage loans are quickly spiking up to a 3% sTr, which translates into a 30.6% cumulative default rate in one year.

Source: Amherst Securities Group, L.P.

Monthly Default Rate for Fixed Rate Securitized Mortgage Loans (Red) In this chart, late 2004 and early 2005 vintage loans have sTrs of approximately 2%, which translates into a 21.5% cumulative default rate in one year, whereas more recent vintage loans are quickly spiking up to a 4-5% sTr, which translates into a 38-46% cumulative default rate in one year.

Source: Amherst Securities Group, L.P.

Monthly Default Rate for 2-28 Securitized Mortgage Loans (Green) 2-28 loans are those with twoyear teaser interest rates that then reset, sometimes to higher rates, which triggers payment shock and a surge in defaults. In this chart, note the surge in sTr shortly after the two-year reset, as well as the rapidly rising sTr even before the reset in more recent vintage loans – compare 12/04, 12/05 and 9/06 loans, for example. 9/06 (pre-reset)

Source: Amherst Securities Group, L.P.

12/05 (pre-reset) 12/04-6/05 (pre-reset)

Monthly Default Rate for 2-28 Securitized Mortgage Loans (Yellow) 2006 and 2007 loans are defaulting at 5-6% per month even before the reset

Source: Amherst Securities Group, L.P.

Monthly Default Rate for 2-28 Securitized Mortgage Loans (Red) For recent vintage 2-28 red loans, sTrs are jumping to 6-8% long before the reset

Source: Amherst Securities Group, L.P.

Current Trends Are Quickly Leading to Unprecedented Default Levels Three-Year Cumulative Defaults

Voluntary Prepayment Rate (VPR)

(1 yr):

Historical levels

2004 green, fixed

Late 2005 and thereafter, Green, fixed

Late 2005 and thereafter, Green, 2/28 Late 2005 and thereafter, Red, 2/28

Notes: MDR = sTr = monthly default rate. CDR = cumulative default rate, which represents the amount of loans in default after 36 months as a percentage of the original balance when keeping MDR and VPR constant for that time period. Source: Amherst Securities Group.

Bubble-Era Pools of Subprime Mortgages Are Performing Catastrophically The 20 RMBS Pools That Comprise the ABX 06-2 Annualized default rate Cumulative defaults

Monthly default rate

Monthly prepay rate

Annualized prepay rate

Total:

An average of 47.6% of the loans have already defaulted

On average, 5.1% of the performing loans in the pools defaulted during the month

Source: Amherst Securities, May 25th reports, reflecting payments through 4/30/09.

The monthly prepay rate only averaged 0.5%

Where Did the Securitized Mortgages End Up? A Primer on ABSs and CDOs

Wall Street Firms Bought Loans and Securitized Them, First Into Asset-Backed Securities Called Residential Mortgage Backed Securities (RMBSs) Quick Review: What is a Securitization?

Source: Deutsche Bank Securitization Research; “How to Save the Bond Insurers”, Pershing Square presentation, 11/28/07.

Mortgages Were Pooled into RMBSs, Tranches of Which Were Pooled into CDOs RMBS

CDO Composition

CDO Rating

100%

Loss rates of, say, 20%, in the underlying RMBS’s can lead to catastrophic losses for a CDO

AAA AA 90% A A 80% A BBB 70% BBB AAA

BBB

60% Super Senior

BBB BBB 50% BBB BBB 40% BBB BBB 30% BBB BBB

This is an example of a “Mezzanine CDO.” A “High-Grade CDO” would select collateral primarily from the A and AA tranches mixed with ~25% senior tranches from other, often mezzanine, CDOs

20% BBB

AA

BBB 10%

A

0%

Equity

AAA

BBB

BBB BBB-

Equity

A

AA

Equity

Note: Asset-based securities backed by home mortgages are called Residential Mortgage-Backed Securities (RMBS), those backed by commercial real estate loans are called Commercial Mortgage-Backed Securities (CMBS), etc. Source: Citigroup, All Clogged Up: What’s Ailing the Financial System, 2/13/08.

Trillions of Dollars of ABSs and CDOs Were Created and Distributed Throughout the Financial System

Note: This is all ABSs and CDOs, not just those related to mortgages Source: Lehman Brothers, 4/08; Carlyle presentation 10/15/08.

What Should the Government Do in the Case of Distressed Financial Institutions? Answer: The problem isn’t lack of capital, but lack of the right kind of capital. They have too much debt and not enough equity. Therefore the solution is to convert some of the debt to equity.

First, One Must Understand the Capital Structure of Financial Institutions – And How Highly Leveraged They Are

Taxpayers will suffer all losses above preferred stock

Source: Amherst Securities.

Freddie Mac

Taxpayers will suffer all losses above preferred stock

Source: Amherst Securities.

Lehman Brothers

Had the preferred stock, junior debt and a small fraction of the senior debt been converted to equity, Lehman could have become one of the best-capitalized financial institutions in the world and everyone would have benefitted.

Source: Amherst Securities.

Overview of Fannie, Freddie and Lehman

Source: Amherst Securities

The $1 Trillion Dollar Mistake • The U.S. financial system likely faces losses of at least $1 trillion, most of which is likely to be realized within the next 2-3 years • There is almost no equity cushion left to cover these losses, so somebody is going to have to come up with $1 trillion to save the system • THE QUESTION IS: WHO? • Current government policy (in the cases of Fannie, Freddie, AIG, Citigroup and Bank of America) is to invest (or guarantee assets) senior to the equity, but junior to all debt, even junior/unsecured/subordinated debt • Absent the government, debt holders would bear the $1 trillion cost (as they did with Lehman), but under current policy, taxpayers will bear this entire cost • This is unfair and unwise – Fairness: Debt holders were paid higher interest than, for example, buying Treasuries in exchange for knowingly taking more risk. These investors made bad decisions, lending to highly leveraged companies that made bad decisions, so why should they be protected? – Moral Hazard: The reckless behavior of debt investors was a major contributor to the bubble. It was low-cost debt with virtually no strings attached that allowed borrowers, especially the world's major financial institutions, to become massively overleveraged, fueling the greatest asset bubble in history. This was not an equity bubble – unlike the internet bubble, for example, stock market valuations never got crazy – it was a debt bubble, so it would be particularly perverse and ironic if government bailouts allowed equity holders to take a beating, yet fully protected debt holders.

• The solution: debt should be converted into equity

But What About Systemic Risk? • What about the lessons from Lehman? Unless we protect the debt holders, won’t the financial system shut down again, as it nearly did in the wake of the Lehman bankruptcy? No. The lessons from Lehman have been misunderstood. – The mistake wasn't the failure to protect the debt, but rather allowing Lehman to go bankrupt, which not only impacted Lehman's equity and debt holders, but also stiffed Lehman's countless clients and counterparties. It's the latter that caused the true chaos. – Lehman should have been seized and put into conservatorship, so that all of Lehman's clients and counterparties could have relied on Lehman (as was done with AIG) - but debt holders would have taken losses as they were realized (which is not being done with AIG).

• If debt holders take a hit, might other financial institutions who own the debt might become insolvent, creating a domino effect? And might debt markets freeze up such that even currently healthy banks might not be able to access debt and collapse? – Regarding the former, the debt is owned by a wide range of institutions all over the world: sovereign wealth funds, pension funds, endowments, insurance companies and, to be sure, other banks. Some of them would no doubt be hurt if they take losses on the debt they hold in troubled financial institutions - but that's no reason to protect all of them 100% with taxpayer money. – As for the latter concern that debt markets might freeze up, causing even healthy banks to collapse, it's important to understand that right now there is no junior debt available to any financial institution with even a hint of weakness - there's very high cost equity and government-guaranteed debt. Neither of these will be affected if legacy debt holders are forced to bear some of the cost of the failure of certain institutions.

Case Study: Bank of America • As of the end of 2008 (not including Merrill Lynch), BofA had $1.82 trillion in assets ($1.72 trillion excluding goodwill and intangibles), supported by a mere $86.6 billion in tangible equity – 5.0% of tangible assets or 20:1 leverage – and $48.9 billion of tangible common equity – 2.8% of tangible assets or 35:1 leverage • Among the company's loans are many in areas of great stress including: – $342.8 billion of commercial loans ($6.5 billion of which is nonperforming, up from $2.2 billion a year earlier) – $253.5 billion of residential mortgages ($7.0 billion of which is nonperforming, up from $2.0 billion a year earlier) – $152.5 billion of home equity loans (HELOCs; about $33 billion of which were Countrywide's) – $18.2 billion of Option ARMs (all Countrywide’s)

• BofA is acknowledging a significant increase in losses, but its reserving has actually become more aggressive over the past year. From the end of 2007 to the end of 2008, nonperforming assets more than tripled from $5.9 billion to $18.2 billion, yet the allowance for credit losses didn't even double, from $12.1 billion to $23.5 billion. As a result, the allowance for loan and lease losses as a percentage of total nonperforming loans and leases declined from 207% to 141% • BofA has $268.3 billion of long-term debt, $158.1 of commercial paper and other short-term borrowings – In addition, Merrill Lynch at the time of acquisition had $5.3 billion of junior subordinated notes, $31.2 billion of short-term debt and $206.6 billion of long-term debt

What the Government Should Do If a Company Blows Up and Can’t Find a Buyer? 1. Seize it and put it into conservatorship 2. The Conservator would replace board and management and suspend the voting rights and other decision rights of the shareholders 3. No dividends, share repurchases or other transfers of resources to the old shareholders could take place while the Conservatorship is in effect 4. The Conservator should be able to haircut all unsecured debt holders and other unsecured creditors, regardless of seniority 5. The Conservator would also be able to impose mandatory debt-toequity conversions on all unsecured creditors and debt holders, with or without first extinguishing the equity of the old shareholders 6. The Conservator would have full authority to sell assets and to restructure the balance sheet and the activities of the business in any way deemed appropriate and lawful 7. The Conservator would have the power to liquidate the company

Source: Time to Pull the Plug on AIG?, Willem Buiter, http://blogs.ft.com/maverecon/2008/11/time-to-pull-the-plug-on-aig

It Is Premature to Allow Banks to Repay TARP Money We agree with this Financial Times editorial: An escape too soon for US banks June 10 2009 The US Treasury’s decision to release 10 of its wards from the straitjacket of the troubled asset relief programme was presented – and widely taken – as good news. But in fact, it sacrificed sound financial policy for political need. Letting JPMorgan Chase, Goldman Sachs, Morgan Stanley and others repay $68bn to the US taxpayer helps placate the combustible public fury at bank bail-outs that almost escaped Barack Obama’s control this winter. The president even makes a point of noting that the government made a profit on the returned funds. The 10 financial groups’ escape from Tarp’s heavy-handed restrictions also weakens accusations of socialism by some Republicans, to which Mr Obama’s administration has been surprisingly thin-skinned. But letting the banks out of Tarp at this point was premature. Supposedly weaned off public support, the 10 had to demonstrate they could raise funds without government guarantees. The temporary liquidity guarantee programme, however, remains available to them. In effect subsidised by the government, but freed from Tarp rules on compensation and hiring, they can poach both staff and business from competitors still in Tarp. This is more regulatory discrimination than it is restitution of market discipline. Even worse is that, for these 10, Tarp has been terminated without addressing the problem that it was supposed to stave off. That problem – or so many thought in the chaos that followed Lehman Brothers’ collapse – was that some banks are too big to be allowed to fail, lest they drag the whole financial system down with them. This fundamental situation has not changed. The stress tests of the largest banks – now looking rather lenient – and the demand that some of them raise additional capital were designed to make failure less likely; they did nothing to make failure less intolerable to the public interest. A bank that was too large to fail in October is still too large to fail now, should another market freeze-up occur. These walking systemic dangers have every reason to restart their risky hunt for yield. There is much to criticise about the government’s use of Tarp to influence the banks. But returning to the status quo ante is no remedy. Tim Geithner, Treasury secretary, should first have secured the promised special resolution regime for bank holding companies. Only then would exits from Tarp have sent the right signal to the market: that the released banks were now free to fail.