Jun 30

FRM 2008 Episode #8: Credit B (Counterparty risk & securitization)

by David Harper, CFA, FRM, CIPM

counterparty_thumb

About Episode #8 (Credit Risk: Counterparty and Securitization)

Please find links to earlier episodes (#1 to #7) at the end of this note.

We follow the sequence of GARP's 2008 FRM Study Guide. This episode #8 (Credit B) reviews the assigned readings for counterparty risk and securitization. I review securitization before credit derivatives (that's in two weeks, Credit C) because it is more natural to start with Culp's reading. First we look at the basic securitized structure in Culp before examining the cash/synthetic collateralized debt obligation (CDO) structure in Meissner. A few tips:

Monte Carlo simulation

mcs_cycle1

Until we reach counterparty risk in the FRM, we often treat exposures with local valuation methods. For example, bond duration and the option Greeks are single factor sensitivities: if a single factor changes (e.g., yield in the case of duration; stock price in the case of option delta), what is the change in the instrument's price? Put another way, the instrument's exposure is measured by its sensitivity to movements in a single underlying risk factor. This is the simple method. But we do this for a trade-off: we sell accuracy and buy convenience. For example, in using bond duration, we assume the entire yield curve shifts in parallel (a single factor) knowing full well that price/yield curve shifts are more complex.

For counterparty exposure to OTC derivative contracts, you'll see we really cannot use the simple local valuation (delta normal). We must instead use Monte Carlo simulation (full valuation). If the implementation of a Monte Carlo is difficult, the idea is simple (please take a look at the interest rate swap example on the member page):

  • Design the engine (specify the parametric model); e.g., assume the interest rate path follows a particular flavor of one-factor model
  • Run many trials by randomizing inputs in the engine; e.g., give the engine random numbers to simulate an interest rate path that is random but still behaves according to our model "engine"
  • Evaluate the instrument or portfolio exposure and store the result; e.g., for each randomized interest rate path, what is our exposure under if we are a counterparty in an interest rate swap?
  • Analyze; e.g., what is my average exposure? what is my worst expected exposure?

Stochastic Process

vasicek

Canabarro profoundly says "different market instruments require the specification of different stochastic processes to characterize their evolution through time."

The load-bearing work, and the bulk of the model risk, lies in the engine construction (i.e., which stochastic process will be used?). For example, we can use Monte Carlo to model the Brownian motion for the price path of a stock (the same that underlies Black Scholes). But, even before we start, as the price engine already contains the truth of its behavior, we have a pretty good idea of the future lognormal distribution.

So Canabarro's point--that different instruments warrant different stochastic processes--is important. Conclusions drawn from simulated outcomes are only as good as the underlying engine.

Securitization

securitize

As usual, I suggest you size up the big picture before getting stuck in details. Per my graphics on this topic, I like to remind we have two big, basic things going on here:

  • On the left: Somebody (the originator) is monetizing credit-sensitive assets; e.g., receivables or mortgages into cash.
  • On the right: Investors are buying tailored claims on the cash flows created by the credit sensitive assets (tranches that participate variously in the orderly division of the cash flow waterfall).
  • Much of the rest is either (i) entities and parties that facilitate the monetization (financing objective) and transfer (risk transfer or reinsurance objective) of credit sensitive assets to investor who will buy the risk for extra yield or (ii), very importantly, the mechanisms that make the structure more palatable to the investors by providing credit/liquidity support

Why securitize? As the reading is focused on traditional, sturdier structures, the first three motives are emphasized:

  • Monetize credit sensitive-assets: originator raises cash
  • Transfer risk to willing investors
  • Accounting motive: remove loans from the balance sheet, perhaps to superficially meet regulatory criteria (Basel I) without matching economic substance
  • Arbitrage motive: Starts from the investor side, leads to more dubious structures ("chasing yield"). 

The back half of Culp is about credit enhancement and liquidity support. The subprime case study illustrates (makes concrete) several of these:

  • Senior tranches in the subprime trust are given credit enhancement with overcollateralization plus subordination
  • Structural cash flow mismatch (i.e., mortgage loan ARMs will adjust per their own pattern which does not necessarily match the investor coupons)

Net Excess Spread

waterfall

Culp gives an example that summarizes the excess spread; the net excess spread also provides credit enhancement, in addition to overcollateralization and subordination.

I uploaded a simple spreadsheet version, of his example, to the member page. It may be worth your time as it illustrates a few of his ideas.

New Learning Spreadsheets Added

learningxls

For this episode (Credit Risk B: Counterparty risk and securitization), I added five new learning spreadsheets to the member page.

  • Monte Carlo: Interest Rate Swap for Counterparty Exposure (Vasicek & CIR)
  • Loss distribution: Credit Rating & PD maps to credit enhancement (CE)
  • Net Excess Spread in simple securitization structure (Culp Chapter 16)
  • Capital structure of GSAMP Trust (subprime securitization)

Please note: none of these are market on the member page with a yellow highlight. That means I view them as optional and not critical to passing the exam.

However, if you are new to counterparty risk and the Monte Carlo simulation, you might frankly find my interest rate swap simulation spreadsheet an easier introduction than the readings, which are technical at times.

Screencast Tutorial

intro_thumb

Paid member access the screencast in the member section. In addition to the viewable screencast:

  • You can downloadable the underlying Power Point slides (in PDF format). For this episode, there is a single 91 page deck.
  • An ipod format (.m4v)
  • A downloadable version of the screencast in a .zip file. (Save to new directory on local and launch the .html file.)

Non-members can sample the start of the screencast tutorial here.

Practice Questions

As usual, I wrote a few practice questions to engage your participation in the screencasts.

Question 1 (counterparty risk)

The Picoult reading states, "Economic Capital (EC) is a measure of risk. It is the potential unexpected loss (UL) of economic value of a portfolio or business, over some long time horizon (e.g., one year), at some high confidence level (e.g., 99.97%)."

(i) Compare this to Michael Ong's definition of Unexpected Loss (UL) in his assigned reading (Internal Credit Risk Models). Are the ULs different/same?
(ii) Is this EC similar to the capital requirement in Basel II?
(iii) What are the generic steps to computing unexpected loss (UL)?
(iv) How will the computation of EC differ for market risk, credit risk, and operational risk?
(v) Does UL include expected losses (EL)? If not, why not?
(vi) Where does UL fit into risk adjusted return on capital (RAROC)?

Question 2 (counterparty risk)

You are the buyer of a put option (i.e., you are taking a long position) in an OTC transaction. Your counterparty is a public company writing a put option on its own stock. Let S0 = stock price, K = strike price, and V = value of the company.

(i) Is this a right-way or wrong-way exposure?
(ii) This is an option with default risk. What does Stulz call this (Reading: Credit Risks and Credit Derivatives) and what is the formula) for the payoff of this option with default risk?
(iii) If your entire portfolio consists of options on publicly-traded equities, like this option, and you wished to employ Monte Carlo simulation to estimate your counterparty credit risk, according to Canabarro & Duffie what (general) sort of stochastic process might you use for the simulation engine?

Question 3 (counterparty risk)

You are the fixed-rate receiver in a plain vanilla interest rate swap. Your counterparty pays a floating rate of LIBOR plus 100 basis points.

(i) For purposes of credit risk (exposure), why don't we treat this as a bond with "lending risk" (Picoult's term)?
(ii) What is the current exposure (CE) at swap inception?
(iii) What are the counterparty credit risk metrics we are likely to analyze (including the VaR-like metric)?
(iv) Let's say we have four swaps with the same counterparty. The current market value of these positions are: +10, +6, -4 and -7 million. What is the total current exposure all are under legally enforceable netting agreements? if none are covered by netting agreements?
(v) If we use Monte Carlo simulation to analyze the exposure profile of the swap over the swap's tenor, which "effect" increases our exposure over time and which "effect" decreases our exposure over time such that we know the peak exposure will be somewhere about one-third through the tenor?
(vi) What mechanisms/modifiers can be used to decrease our counterparty exposure?

Question 4 (Culp on securitization)

Adapting Culp's example, assume a bank wants to securitize a $100 million portfolio of credit-sensitive assets that earns an interest rate of 100 basis points over LIBOR. Assume LIBOR is 5%. The senior expenses of the SPE are 10 basis points. The SPE issues three classes of securities: The senior tranche with a face value of $60 million and a coupon of LIBOR + 50 basis points; a mezzanine tranche with a face value of $30 million a coupon of LIBOR + 100 basis points; the equity tranche constitutes the rest of the capital structure and receives the realized excess spread.

(i) Where is the overcollateralization (O/C)?
(ii) What is the internal credit enhancement (internal C/E) provided to the senior (tranche) debt?
(iii) How much is the net excess spread?
(iv) List examples of EXTERNAL credit support
(v) Give an example of structural liquidity risk
(vi) Give examples of liquidity support

Question 5 (subprime securitization)

In regard to the case study on subprime securitization:

(i) Which of the seven frictions, to at least some degree, are impacted by asymmetric information (information asymmetries) and/or moral hazard?
(ii) Why is the interest rate swap an important feature of the GSAMP Trust?
(iii) In the securitized structure, credit enhancement is a function of the target credit rating (e.g., BBB) for a tranche. Where is model risk located?
(iv) With regard especially to the role of systemic risk and correlation, how does an ABS structure credit rating differ from a corporate bond?
(v) How did increases in interest rates, starting in June 2004, impact securitized subprime structures?
(vi) The reading is quite critical of rating agencies (and, ironically, offers the tamest remedies for this key friction). Cite the failures related to the rating agencies, according to the reading.

 

My answers to these questions

Previous newsletters

Here are links to Episodes #1 through #6:

Thanks very much.

David Harper, CFA, FRM, CIPM
Founder
www.bionicturtle.com

David-Harper_100w

Comments

  1. Be the first to leave a comment!

Leave a Comment