Charta is on Google Play: practise on the go. Get the app
Charta

Explain the securitization process and the role of a special purpose vehicle (SPV), then judge how risky the various bank business models for securitized products are.

FRM® Part I · Foundations of Risk Management · Credit Risk Transfer Mechanisms · FRM1.FR.4.d

The idea

Securitization pools loans and sells claims on the pool's cash flows. The originator sells the assets in a true sale to a special purpose vehicle (SPV), a separate legal entity that is bankruptcy remote. Investors therefore depend on the pool and not on the originator's solvency. The SPV pays for the assets by issuing tranches in a waterfall. Losses hit the equity or first-loss tranche, then mezzanine, then senior, while cash flows go in reverse order. Credit enhancement comes from subordination, overcollateralization and excess spread. A tranche loses nothing until pool losses pass its attachment point. Beyond that, Tranche loss = (pool loss − attachment) ÷ tranche width, capped at 100%. Bank models differ in risk. Originate-to-distribute keeps little credit risk but has warehouse and incentive problems. Retaining senior tranches or re-securitizing carries concentrated tail risk, plus liquidity risk when funded short-term.

Formulas

Loss on a securitisation tranche

tranche loss rate = min[max(L − A, 0) ÷ (D − A), 1]

L
the loss on the underlying pool, as a share of the pool
A
the tranche's attachment point
D
the tranche's detachment point; D − A is its width

Nothing is lost until L passes A, and the tranche is wiped out once L reaches D. Divide by the tranche width, not by the pool.

A practice question

From the question bank, with the reason each wrong option is wrong.

What is the main purpose of a special purpose vehicle in a securitization?

  1. A bankruptcy-remote entity that buys the assets and issues tranches - correct
  2. To guarantee the investors' returns using the originator's capitalWhy not: Confuses the SPV with a guarantor; investors rely on the assets, not on the originator's balance sheet.
  3. To keep the loans on the originator's balance sheet so they can be servicedWhy not: Reverses the purpose; a true sale moves assets to the SPV, away from the originator.
  4. To pool the originator's own debt so it can issue senior bondsWhy not: Wrong direction; the SPV buys the loans and issues notes backed by that pool rather than the originator's own debt.

The originator sells assets in a true sale to a legally separate SPV, which funds the purchase by issuing tranched notes. Bankruptcy remoteness means the investors' claims depend on the pool and not on the originator's solvency.

A harder one

A securitization pool has three tranches: equity covering the first 5% of pool losses, mezzanine covering losses from 5% to 15%, and a senior tranche above 15%. Pool losses are 9%. The percentage of the mezzanine tranche's principal that is lost is closest to:

  1. 9%Why not: Applies the pool loss rate to the tranche and ignores subordination and tranche width.
  2. 40% - correct
  3. 80%Why not: Divides the excess loss (9 - 5 = 4) by 5, the equity width, rather than the 10 width of the mezzanine tranche.
  4. 4%Why not: Reports pool loss above the attachment point as a share of the pool, not of the mezzanine tranche.

Mezzanine loss = (pool loss - attachment) / width = (9 - 5) / (15 - 5) = 40%. Equity is wiped out and the senior tranche is untouched.

The hardest tier

A bank compares three securitization business models for mortgage loans: (1) originate and sell all tranches within weeks, keeping nothing but the loans temporarily held for the pipeline; (2) originate, securitize and retain only the small equity tranche; (3) buy mezzanine tranches of other issuers' mortgage securities, re-securitize them into CDOs, and keep the super-senior tranche, funded with short-term wholesale borrowing. In a severe, system-wide house price decline, which model is riskiest for the bank?

  1. Model 1, selling everything within weeksWhy not: Overstates the pipeline (warehouse) risk of a few weeks of loans, which is small compared with a large retained position.
  2. Model 2, retaining only the equity trancheWhy not: Reasons that first loss is riskiest per dollar but ignores that the equity slice is capped and small relative to model 3's leveraged tail position.
  3. Model 3, re-securitized mezzanine with a funded super-senior kept - correct
  4. All three are equally risky because each one is exposed to the mortgage marketWhy not: Ignores differences in retained size, leverage and funding, which drive the losses.

Model 3 keeps a large tail exposure that is highly sensitive to correlated losses in re-securitized mezzanine tranches, and it also carries funding (liquidity) risk. Model 1 has only warehouse risk and model 2 is limited by the small tranche size.

In the app: 6 practice questions on this outcome, spaced review that brings it back before you forget it.

Start free →

More in Credit Risk Transfer Mechanisms