Securitization queries

[Roshan Ramdas]

Hi David,

I have a very basic question with regards to securitization,....need your help pls.

I am trying to understand the monetary benefits of securitization from the context of the "Originator".

I understand some of the angles in terms of banks being able to monetize assets (eg loans) // freeing up capital rather quickly // reducing credit risk // reduction in assets in turn implies reduction in the regulatory capital requirement.

However, when it comes to actually being able to make money,... I am not fully clear on how securitization helps an originator. For eg - a bank advances a loan of $100000 to a subprime borrower and applies a high interest rate of 15% (risk based pricing). The fact that the bank sells the loan (via securitization) implies that it no longer receives the interest related cash flows. Going by the logic of "excess spread", even if we assume that investors were paid a coupon of 12%, that difference of 3% would go towards meeting other expenses and credit and liquidity enhancement structures,....and there are some passages in Christopher Culps chapters that seem to suggest that the originator should not have any access to this excess spread (which establishes some form of dependency b/w the Originator and the SPE and may require the Originator to consolidate the asset right back into their balance sheet).

Using my example,....could you let me know how an Originator could go about in making money pls.

Thank you

 

[Roshan Ramdas]

Hi David,

The below question is directly from GARPs Credit Risk text book and the answer has been highlighted.

Question - An investor has sold default protection on the most senior tranche of a CDO. If the default correlation between assets held in the CDO decreases sharply, assuming everything else is unchanged, the investors position Answer - will gain significantly since the probability of exercising the protection falls.

I understand the directional impact of default correlation on equity and senior tranches. I nonetheless finds GARP's solution explanation a bit brief and need you to validate my reasoning please.

For an investor who sells protection via (eg CDS), the CDS insurance premium (which he receives) is going to be negotiated and fixed at the time the contract is entered into. If default correlation between assets were to reduce after the trade has been entered, the senior tranche risk reduces and the associated CDS premium would reduce. This implies a gain for existing protection seller positions, who are receiving a higher CDS premium versus the current lowered CDS premium and a loss for existing protection buyer positions who are paying a higher premium versus what is required.

Thank you

 

[David Harper CFA FRM]

Hi Roshan,

Right this is always confusing, IMO; e.g., related discussion here including Malz plot (bottom of page 326) of Senior tranche LOSSES versus correlation @ https://forum.bionicturtle.com/threads/basket-tranches-value-and-risk.5901/#post-26713 (i.e., the solid line indicates far fewer losses for rho = 0 than rho = .90)

I agree with your reason, yes that is exactly correct. Say the credit protection seller (the "investor" who is short the CDS and therefore long the credit risk) enters the CDS contract, on the senior tranche, when the the CDS spread is, say, 100 basis points. As the credit protection seller, she is receiving this fixed spread. Then the default correlation decreases such that the senior tranche becomes less risky; to me, the easiest way to think about the consequence of less risk is a lower credit spread (this works for bond, CDS, etc). Tranche risk goes down --> spread goes down. So consequently, say, spread decreases to 75 basis points. The CDS buyer is now paying more than the market's current spread; and the investor is now receiving more than the market. The CDS buyer, who was short the tranche, experiences a loss as spreads narrow; but the investor, who was long the tranche , experiences a gain as spreads narrow (receives 100 when market is only 75 --> is a positive gain for this investor). So I agree with your reasoning, it's the same!

I need to bookmark your first query for later, I am not immediately comprehending the issue .... Culp's cash securitization is a "true sale" (removal from the balance sheet), so we don't require that the originator increase the PV in order for the monetization benefit to occur, right? e.g., if a bank extends a sub-prime loan then, they have an asset (amortizing) which is bond-like: they are collecting interest and principle over an T-year time horizon. It's just like being long a bond with T-year maturity; there is an opportunity cost to the funded investment such that, sometimes, it's better to sell the bond to somebody else and do something better with the proceeds. The "monetization benefit" of cash securitization is firstly simply the sale of that bond (asset) to somebody else: the future stream of credit risky cash flows is immediately realized as a PV lump sum today. Not only does the originator not wait, they transfer the risk to the buyer. So, even if the PV is unchanged, there is a huge benefit to "translating" the stream of expected (but risky) cash flows into a current lump sum. (plus the credit transfer). Thanks,