Statement_of_Work.docx

Statement of Work Deal, Pool and Tranche Level Data for Asset-Backed Securities

The US Securities and Exchange Commission (SEC) requires deal, pool, and tranche level data for asset-backed securities (ABS) sold and distributed in the U.S.

The data shall provide historical coverage of all ABS and CMBS transactions sold in the U.S. since at least 2000, including registered and 144A/private deals. The data shall be made available through FTP or SFTP download or via a secure web portal. The data may consist of several tables that must be linked via a set of key variables. ABS tranches must be identifiable by CUSIP in order to link with external datasets.

Specifically, the data must include, at a minimum:

· Extensive coverage of all ABS and CMBS transactions sold in the U.S. since at least 2000, including registered and 144A/private deals. This would include all ABS by collateral type, including consumer, corporate, commercial and whole business. It would also include all CMBS and non-agency RMBS deals and would total 16,000 deals at minimum.

· Deal level data should include: aggregate deal characteristics (e.g., issuer name, amount issued, relevant dates: closing/pricing/cutoff, etc.), involved parties (e.g., all underwriters, bookrunners, sponsors, trustees, servicers, etc.), aggregate pool characteristics (number of loans in the pool, loan types, other characteristics dependent on asset class, e.g., WAC, WAM, etc.).

· Tranche level data should include: security identifiers (CUSIP or ISIN), tranche characteristics (e.g., types, face values, coupons, subordination levels, credit enhancement), initial spreads/prices, etc.

· Post-issue performance data should include the following. At the tranche level: distributions, impairments, defaults, missed payments, modifications. At the pool level: percentage of delinquencies in the pool, defaults of underlying assets, prepayment history.

Technical requirements for data delivery

· Data delivery in flat text or .csv files with the ability to easily move the information into statistical software packages like SAS, Stata, or R.

· Availability for multiple users across the organization.

· Ability to link tranche and deal level information.

· Quarterly updates at minimum.

· Reference materials such as data dictionaries or data mapping files.