This is the capstone — one graded run across the whole course. The questions roam over everything: how a single-name CDS is an insurance contract where the protection buyer pays a premium leg and the seller pays out (one minus recovery) of notional on a credit event, and what the 2009 Big Bang fixed about coupons and settlement; how the credit triangle ties spread, hazard rate, and recovery together and why survival decays geometrically; how CDX and iTraxx package a basket of names with a standard coupon, a roll every March and September, and a basis to the names inside; how a true sale moves loans into a bankruptcy-remote SPV so the bonds can out-rate the originator, and why falling rates make mortgage pools negatively convex; how attachment and detachment points slice a pool into tranches that take losses bottom-up and cash top-down; how default correlation barely touches expected loss but reshapes the tail — the lesson AIG, the monolines, and the Gaussian copula learned the hard way in 2008; and how a modern CLO, built on leveraged loans with OC and IC tests, is a very different animal from a 2008 ABS CDO. There is no formula sheet and no second guess — read every option before you commit, because each wrong one is a trap that has caught a real trainee.
How this exam works
This is a graded exam. Questions arrive one at a time. Once you submit an answer it is final — there is no going back, no retries, and a wrong answer simply fails that question. Your score stays hidden until the very end, where you need 70% to pass. Slow down and read every option before you commit.
In a plain single-name CDS, who pays whom, and when?
Select an answer to continue.
Course Recap
Whatever your score reads, the framework you just stress-tested — the single-name CDS as insurance with a premium leg and a (one-minus-recovery) payout, standardized by the 2009 Big Bang; the credit triangle that ties spread, hazard, and recovery together with geometric survival and an RPV01 mark-to-market; the CDX and iTraxx indices with their fixed coupons, semiannual rolls, and basis to the names inside; the true sale into a bankruptcy-remote SPV that lets MBS and ABS out-rate the originator while falling rates make pools negatively convex; the attachment and detachment points that slice a pool into tranches taking losses bottom-up and cash top-down, cushioned by excess spread, overcollateralization, and subordination; the default correlation that barely touches expected loss but fattens the tail — the Gaussian copula, the base-versus-compound skew, the CDO-squared leverage, AIG, and the monolines that learned it in 2008; and the modern CLO built on actively managed leveraged loans with OC and IC tests, a very different animal whose senior tranches survived the GFC and COVID — is the working map of how credit risk is sliced, sold, and traded. Here is the whole course in one glance.
Big picture
Credit Derivatives & Securitization, in one glance
- Credit Derivatives & Securitization
- CDS mechanics & Big Bang
- Buyer pays premium leg; seller pays (1 minus recovery) on a credit event
- Triggers: bankruptcy, failure to pay, restructuring
- Big Bang 2009: fixed 100 bp IG / 500 bp HY coupons + upfront
- Auction settlement; ISDA 40% senior recovery
- Naked CDS = short credit, no bond needed
- Spreads & hazard rates
- Credit triangle: spread is about hazard times (1 minus recovery)
- Hazard is about spread over (1 minus recovery)
- 200 bp at 40% recovery gives about 3.33% intensity
- Survival = exp(minus hazard times t), geometric decay
- MtM is about (spread minus coupon) times RPV01
- CDX & iTraxx indices
- CDX.NA.IG 125 at 100 bp; HY 100 at 500 bp
- iTraxx Europe 125 at 100 bp; Crossover 75 at 500 bp
- Rolls every March and September; on/off-the-run
- Basis = traded spread minus intrinsic of the names
- A default wipes 1/N of the notional
- Securitization & MBS/ABS
- True sale to a bankruptcy-remote SPV; bonds out-rate originator
- Pass-through (pro-rata) vs CMO (tranched)
- Agency (Ginnie full-faith / Fannie-Freddie GSE) vs non-agency
- Rates fall, prepayments rise: negative convexity
- 100% PSA seasons at 6% CPR; originate-to-distribute weakened underwriting
- Tranching & waterfalls
- Attachment = losses start; detachment = fully wiped
- Losses bottom-up (equity first); cash top-down (senior first)
- Tranche loss = clamp(pool loss minus attach, 0, thickness)
- Credit enhancement: excess spread, OC, subordination, reserves
- Sequential vs pro-rata principal pay
- Correlation & 2008
- Correlation barely moves expected loss, fattens the tail
- Rising correlation: senior riskier, equity safer
- Gaussian copula (Li, 2000) modeled default-time correlation
- Base correlation monotone; compound non-unique for mezz
- CDO-squared, AIG super-senior, monolines, issuer-pays ratings
- Modern CLOs
- Leveraged-loan collateral, not subprime
- AAA to equity stack; equity about 8 to 10%
- Actively managed; reinvestment period about 4 to 5 years
- OC/IC test fails: divert cash from equity to pay down senior
- CLO 2.0/3.0 not 2008; AAA/AA never defaulted, survived GFC and COVID
- CDS mechanics & Big Bang
Key Takeaways
What you now own
You can read credit as a tradable, sliceable asset from end to end. You know a single-name CDS is insurance — the buyer pays a premium leg, the seller pays (one minus recovery) of notional on a bankruptcy, failure-to-pay, or restructuring event — and you can explain what the 2009 Big Bang fixed with its 100/500 bp coupons, upfront payments, and auction settlement, and why a naked CDS is just a short-credit bet. You can run the credit triangle in both directions, turn a 200 bp spread and 40% recovery into a 3.33% hazard, see that survival decays geometrically, and mark a CDS to market with RPV01. You know the CDX and iTraxx benchmarks cold — their constituents, their coupons, their March-and-September rolls, their basis, and what a single default does to the basket. You understand how a true sale into a bankruptcy-remote SPV lets MBS and ABS out-rate their originator, why falling rates make mortgage pools negatively convex, and how originate-to-distribute corroded underwriting. You can place attachment and detachment points, route losses bottom-up and cash top-down through a waterfall, size a tranche loss with the clamp formula, and name the enhancements — excess spread, overcollateralization, subordination, reserves — that protect the senior notes. You can explain why default correlation barely budges expected loss yet reshapes the tail, making the senior tranche riskier and the equity safer, and you can recount how the Gaussian copula, base-versus-compound correlation, CDO-squared leverage, AIG’s super-senior book, the monolines, and issuer-pays ratings combined to detonate in 2008. And — the payoff of the whole course — you can explain why a modern CLO, built on actively managed leveraged loans with OC and IC tests, is a fundamentally different animal whose senior tranches never defaulted through the GFC or COVID. That is the credit-derivatives-and-securitization toolkit, end to end.