Human Asset Being Finance

Google+ Pinterest LinkedIn Tumblr +

Many asset types are securitized: corporate bonds, credit card debt, auto loans, student loans, residential and commercial mortgages, and so on. The purpose of this chapter is to introduce residential mortgages. Although mortgages are in general more complex than the “average” securitized asset, they still demonstrate how assets are placed into a securitization and how the asset characteristics affect the structure. Once the reader has learned about residential mortgages, it is a relatively small step in complexity to analyze other asset-backed securities (ABSs).

Consider credit card debt, for example. Credit card securitization operates on the same principles as mortgage securitization in that excess spread and over-collateralization give the credit enhancement needed to protect the structure. Debt holders have the option to prepay their loans, just as in mortgages. Like mortgages, credit cards usually charge floating rates. The trust into which the debt is placed may have caps or swaps to protect it against basis risk. Unlike a generic mortgage, credit card debt revolves; that is, it can grow and shrink. There are mortgages that exhibit similar behavior, for example, a home-equity line of credit (“second mortgage”) and an option ARM (an adjustable-rate mortgage that can grow in balance). A simple credit card model is given in section B of Appendix B.

In this chapter, first mortgage assets are introduced, then a basic collateral cash flow model, followed by a discussion of how rating agencies differ in their modeling approaches.


Simple loans were introduced in the previous chapter as an example of object-oriented modeling. In this chapter, the loan definition is expanded to be more realistic. These loans are still abstractions—actual loans as underwritten contain much more information than listed here (see, for example, Pratt ). For example all of the credit quality information of the borrower is absent here. This information is certainly needed to determine the expected losses. The abstraction here is sufficient to flesh out the mortgage collateral model, given loan loss characteristics from another model.

A repline (“representative line”) is the loan abstraction we will use. A repline can represent a single mortgage or a group of mortgages. Both cases refer to the weighted-average coupon (WAC), the weighted-average maturity (WAM), and so on. A fixed-rate repline is comprised of all fixed-rate loans. Typically, fixed and floating loans are not aggregated into the same repline. However, different types of floating-rate loans (e.g., loans with different initial reset periods) can be aggregated into a single repline. Historically, replines were introduced because it was computationally burdensome to model individual loans. These days, it is not a hardship to model upward of thousands of loans, but this abstraction is retained nonetheless.

Table summarizes the fields in a simplified repline. Each of these fields is expounded in the following notes:

  1. Type: string specifying fixed-rate (FRM), floating-rate (ARM), or option ARM (OPTARM). Option ARMs are addressed in section —they require additional loan parameters that are not listed in Table .

  2. Balance: initial balance in dollars.

  3. WAC: initial balance weighted average (gross) coupon of constitutent loans comprising the repline. For a fixed-rate repline, the WAC is static over the life of the repline. For a floating-rate repline, this coupon changes every reset period.

  4. ExpRate: these fees (stated as annualized rates on the loan balance) must be paid out from the repline before interest flows are paid.

  5. WAM: initial balance weighted average maturity of constituent loans comprising the repline, in months. This is the remaining amortization term as opposed to the remaining term to maturity. The remaining term to maturity plus age must equal the original term.

  6. Age: months since settlement.

  7. PrepayPenaltyPeriod: the number of months following settlement during which the borrower is penalized should the loan be prepaid.

  8. PrepayPenalty: this string encodes the prepayment penalty conditions. For example, “36 C 105” means that prior to month 36, the penalty is 5% of the current loan balance. “24 IP 4.8” means that prior to month 24, the penalty is 4.8 times the current monthly periodic WAC times the current loan balance. The 4.8 is essentially 80% of six months. The penalty can be more complex, for instance “12 C 105 12 C 104” indicates a 5% penalty for the first year and a 4% penalty for the second year.

  9. BalloonPeriod: if this is positive, the loan pays off its remaining balance as a principal paydown on this month.

  10. IOPeriod: up to and including this month, the loan pays “gross” interest only, that is, WAC x Balance. After the interest-only period, the loan pays a standard mortgage payment, that is, the constant periodic payment needed to pay off the balance in the remaining life of the loan at the current coupon.

  11. PrefundPeriod: if positive, the loan balance represents cash that has not yet been converted into assets as of the settlement date. Up to PrefundPeriod, the cash earns interest at PrefundRate. At PrefundPeriod, loans are purchased with the characteristics specified in the repline. These loans begin to pay interest, principal, prepayments, and so on, after the PrefundPeriod.

  12. PrefundRate: up to and including the prefunding period, the repline pays interest only at this rate.


About Author

Leave A Reply