Hedge Funds - 5

CDO market named as villain of the crisis

"[N]ow the Securities and Exchange Commission has found the target - and should be congratulated." • "Several contenders have been presented for the role of the bête-noire." • "Credit default swaps have been most often targeted by regulators and politicians, but the much-vilified products had little to do with the crisis. Neither were hedge funds to blame, or shorting, or the lack of clearing houses." • "It was collateralised debt obligations that were to blame. Without the CDO market, the collapse of the US mortgage market would have been a localised turndown"

Ratings and risks

BBB tranche of a CDO = BBB-rated bond? - This was a common assumption • The BBB tranche can have the same probability of loss or the same expected loss as a BBB bond • What are we missing here? • The probability distribution of loss is different • It is much more likely that an investor in the CDO tranche will lose 100% of the principal than that this will happen for a BBB-rated bond - "Cliff risk"

Why adjust ratings?

Credit enhancements that are not inputs in the CRA model: • Experience of the collateral manager • Structure of the cash flow waterfall (overcollateralization) • Insurance • Excess spread (ratio of total collateral income over CDO notes coupon payment) • Liquidity considerations

Two-asset example

Two identical securities - Both have probability of default equal to - Both pay $0 conditional on default and $1 otherwise • Pool these securities in a portfolio - Total notional value of the underlying fund is $2 - Issue two tranches against this fund, each of which will pay $1 • Junior tranche bears the first $1 of losses to the portfolio - Junior tranche pays $1 if both securities avoid default, zero if either one defaults • Senior tranche bears losses if the capital of the junior tranche is exhausted • If the defaults of the two securities are imperfectly correlated, the senior tranche will pay $1 or $0 - Exactly like the underlying securities • However: Will be less likely to default than either of the underlying securities • If both securities have 10% default probability and defaults are uncorrelated: - Probability of default of senior tranche is 1% • Junior tranches have lower prices and higher promised returns • Using a larger number of securities in the pool, a progressively larger fraction of issued tranches can end up with higher credit ratings • Add a third (identical) security to the previous example and issue three tranches: - First tranche defaults if any security defaults - Second tranche defaults if two or more securities default - Third tranche defaults if all three securities default Default probabilities: - First: 27.1% - Second: 2.8% - Third: 0.1% Both (2/3 of capital) less risky than the underlying securities

Limits of arbitrage?

• "For the last nine months the main reason to not buy something for 72 cents that was really worth a dollar was the idea that there was this guy with infinite capacity on the other side [of the trade]," lamented one prominent hedge fund manager

Underlying pool structure

• Any asset with a stable stream of cash flow can in principle be included in the reference portfolio • Static pool - Asset pool fixed without substitution - Loan balances fixed without adjustable credit limits - Pre-defined amortization schedule • Revolving pool - Asset pool varies, allows substitution - Loan balances adjustable up to maximum limits • Substituting pool - Allow substitution of new loans within defined credit parameters - For example, corporate bonds, mortgages/consumer loans

Collateralized debt obligation (CDO)

• Asset-backed security whose underlying collateral is typically a portfolio of bonds - Corporate, mortgage-backed securities etc. • Cash flow CDO (we focus on these) - Collateral portfolio not subject to active trading by the CDO manager • Market value CDO - Tranche payments largely determined by the trading performance of the CDO manager

Structured finance ratings

• By mid-2007, there were 37 000 AAA-rated structured finance issues in the US alone • Roughly 60% of all global structured products were rated AAA - Less than 1% of corporate issues are AAA-rated • In 2006, Moody's reported that 44% of its revenue came from rating structured finance products • In 2008, 27 of the 30 tranches of CDOs underwritten by Merrill Lynch originally rated as AAA were downgraded to junk

Effects of Adjustments

• Credit enhancements do not explain adjustments • Cross-sectional variation in adjustments negatively related to model-implied AAA proportions - For the smallest quintile of AAA implied by the CRA model, the model-based fraction is 42.6% of AAA but adjustments add another 26.8% - Total fraction of AAA is thus 69.4% • Larger adjustments positively related to future downgrades • In 92.4% of cases, AAA tranches only met AA default standard

Pricing of systematic risk

• Credit ratings only provide an assessment of the risks of the security's expected payoff - Systematic risk plays no role - Example: Bonds with equal credit ratings can trade with different spreads because of the differences in systematic risk • Example of two (extreme) types of securities: I. Catastrophe bonds (no exposure to systematic risk) - Deliver the promised payoff if there is no natural disaster II. Digital call option on the market (maximum exposure to systematic risk) - By selecting the appropriate strike price, the probability of "default" can be tuned to match any credit rating • Pooling and tranching effectively create securities whose payoff profiles resemble those of a digital call option on the market index - Pooling allows for broad diversification of idiosyncratic risk - Losses driven entirely by systematic risk exposures • Investors in senior tranches of CDOs bear enormous systematic risk - Increasingly likely to suffer significant losses when the overall economy or market goes down - Defaults of single-name bonds affected by firm-specific bad luck • In effect, structured products have enabled investors to write insurance against large declines in the aggregate economy • Investors in senior tranches: - Likely to earn a yield that appears attractive relative to securities with a similar credit rating - Yields well below the return they could have earned by writing the insurance directly (e.g. through options) • Coval, Jurek, and Stafford (2009, AER): Yield spread for senior CDO tranches do not compensate for the systematic risk they bear

Structured finance and subprime

• Government agencies (Fannie Mae, Freddie Mac, Ginnie Mae) were chartered to buy mortgages originated by local banks - Aim was to ensure a continuous supply of credit to home buyers - Size and credit quality requirements • Mortgages were then repackaged by the agencies and resold in capital markets - Implicit guarantee of the US government • Subprime mortgages: mortgages given to those below the credit standards for the government sponsored enterprises - In 2006 accounted for 22% of all mortgages issued that year

The future?

• Greater emphasis on transparency and standardization - Standardization of CDS terms and central clearing of CDS contract • Rating agencies and other gatekeepers have become more conservative • Some asset securitizations such as credit card and car loan-backed where the collateral pool is fairly trustworthy will be likely to prevail • Some structured finance products including resecuritizations (ABS CDO, CDO^2) may no longer be viable

Caveats of securitization

• How accurate are estimates of default risk • More importantly: How likely are defaults to be correlated! • Issuing a capital structure amplifies errors in evaluating the risk of the underlying securities - Modest imprecision in the parameter estimates can have severe consequences • Effectively, securitization process substitutes risks that are largely diversifiable for risks that are highly systematic - Effect on the chances of survival during crisis vs. traditional securities of equal rating?

Default correlation

• Lower the default correlation, more improbable it is that all securities default simultaneously • Consider our previous example of two assets: - Depending on the default correlation, the default probability of the senior tranche can be anything between 1% and 10% • In the single name rating business, the agencies could ignore default correlation • Structure of CDOs magnifies the effect of imprecise estimates of default likelihoods, recovery rates, and correlations - Problem accentuated further through sequential structuring (e.g. CDO-squared)

Securitization: What and Why?

• Main idea: Pool economic assets and issue a prioritized capital structure of claims (tranches) against this collateral pool • As a result of the prioritization scheme, may of the tranches are (should be) safer than the average asset in the underlying pool • Able to get (good) ratings from the rating agencies • Issuers of structured securities were eager to get their products rated on the same scale as bonds - Able to attract investors subject to rating-based constraints

Cash flows (waterfall)

• More junior tranches can only be paid after the more senior ones have received their promised return • Equity last to receive and first to absorb losses • Previous example: first 5% of losses borne by the equity tranche (loses 100% of principal)

What is Securitization>

• Securitization is the pooling and tranching of credit risk • Assets of a securitization are a pool of fixed income securities with embedded credit risk: - Corporate bonds and loans - Mortgages - Credit card receivables - Other structured products • Liabilities are structured in tranches in order of priority - Senior: lowest risk and lowest promised return - Mezzanine: more risk and higher promised return - Equity: bears the most risk

Goals of securitization

• Separate debt funding from risk bearing - Senior tranche investors put up most of the capital to fund assets - Junior tranche investors bear most of the credit risk • Risk diversification - A structured credit may be backed by hundreds or thousands of loans - Cash flows for large asset pools may be easier to predict • Tailor risk/return characteristics to market demand - Risk/return profile of structured securities is different from that of the underlying collateral • Regulatory arbitrage - Assets can be moved off balance sheet while continuing to be exposed to those assets' credit risk

Why Stop Here?

• Synthetic CDO: A CDO where the underlying portfolio consists of (short) credit default swaps (CDS) - Long position in a corporate bond carries credit risk equal to a short position in a CDS • CDO-squared: A CDO where the underlying portfolio includes (usually mezzanine) tranches of other CDOs • CDO-cubed: A CDO where the underlying portfolio includes CDO-squared tranches • CDOn : You get the picture...

Securitization process

• Tailor cash flow risk to satisfy guidelines set forth by the credit rating agencies • Step 1: Assemble a large collection of credit-sensitive assets in a portfolio - Special purpose vehicle, (SPV), separate from originator's balance sheet • If the SPV issued claims were not prioritized: passthrough securitization - Portfolio's credit rating would be given by the average rating of the securities in the pool - No credit enhancement • Step 2: Issue a capital structure of prioritized claims ("tranches") against the underlying pool - Tranches prioritized in how they absorb losses from underlying portfolio - Senior tranches only absorb losses after junior claims have been exhausted • Senior tranches can obtain credit ratings better than the average rating in the pool • This overcollateralization plays key role in determining the credit rating for the more senior tranche

J.P. Morgan to Pay $153.6 Million to Settle SEC Charges

• The Securities and Exchange Commission today announced that J.P. Morgan Securities LLC will pay $153.6 million to settle SEC charges that it misled investors in a complex mortgage securities transaction just as the housing market was starting to plummet. • The SEC alleges that J.P. Morgan structured and marketed a synthetic collateralized debt obligation (CDO) without informing investors that a hedge fund helped select the assets in the CDO portfolio and had a short position in more than half of those assets. As a result, the hedge fund was poised to benefit if the CDO assets it was selecting for the portfolio defaulted

JPMorgan unit has $100bn of risky bonds

• The unit at the centre of JPMorgan Chase's $2bn trading loss has built up positions totalling more than $100bn in assetbacked securities and structured products - the complex, risky bonds at the centre of the financial crisis in 2008. • The unit, the chief investment office (CIO), has been the biggest buyer of European mortgage-backed bonds and other complex debt securities such as collateralised loan obligations in all markets for three years • "I can't see how they could unwind these positions because no one can replace them in terms of size. It's a bit of the same problem they face with the derivatives trade," said a credit trader at a rival bank. "They pretty much are the market."