We hold a unique historic Downturn Default Database of Loss Given Default (LGD) data from the 1989-1991 “worst case” housing recession and from lesser downturns. We supply these data within our LGD Data service for lenders needing to supplement their internal data. Crucially, too, these data provide the parameter assumptions used for our “worst case” and “downturn” macroeconomic scenarios in our Stress and Scenario Testing solution.

Given the scarcity of publicly available historic LGD data within the mortgage lending industry, our LGD Data service is flexible to meet the specific requirements of individual clients. Since our Downturn Default Database comprises 44,319 repossessions during 1991-95 we can provide a variety of data extracts (e.g. forced sale discounts/Distress Factors) and time-series analysis (e.g. time from repossession to sale) to inform a client’s “downturn” or “worst case” LGD modelling.

In the context of Basel II, the Acadametrics Downturn Default Database and the analysis therefrom constitute “external data”. FSA guidance is that the use of external data/models is encouraged in the context of all relevant and available information, and that firms should use external data where doing so enhances their internal approach. "Use of External Models and External Data in the IRB Approach" was the subject of an FSA facilitated Expert Group report dated August 2005 which said "Used sensibly, external data and models enrich a firm’s understanding, risk processes and controls, often providing a reliable and cost effective supplement to internal data and models” (Note 1). The report (Note 2) also provides guidelines covering “accuracy, completeness and appropriateness”, “representativeness and comparability”, “consistency of definitions” and “data storage and IT requirements”.

The Expert Group paper “Loss Given Default Mortgages, November 2005” notes that, for mortgage risk, Basel II requires a downturn estimate (Note 3) as per the Basel Framework Document (Note 4) and subsequent guidance notes (Note 5). However, this requirement creates a real difficulty for those mortgage lenders whose data collection process did not start until after the UK housing market recession of 1989-1991 and who may well be able to develop an entirely satisfactory long run estimate but hold insufficient data from the “last housing recession” deemed appropriate to support the production of a downturn or worst case estimate.

In such an instance, Acadametrics can provide the following analyses:

Discounts (or Distress) Factors represented as a % difference over an indexed property value (using any chosen house price index), between the actual sale price achieved and the calculated value at date of sale, as assessed by indexation from the value at mortgage inception. Value, being an analysis of how the discounts vary by property value ratio bands where the ratio is of the indexed value of the property at its point of sale to the average house prices at a regional level. Days elapsed: an interrogation of the database to establish the time from possession to sale.

A lender can choose between two basic approaches towards calculating downturn LGD (Note 6). These are the “Scenario” approach and the “House Price Index” approach. The “Scenario” approach is used by the Acadametrics Stress and Scenario Testing model. Where a lender has already developed a “House Price Index” approach, Acadametrics can transform and augment the data to provide the model with the capability of calculating the requisite LGD downturn estimates.

An example of a typical Acadametrics LGD Data analysis is shown here. A prospectus is available here.

Notes.

1. Financial Services Authority Expert Group Report "Use of External Models and External Data in the IRB Approach" (Aug 2005). back

2. ditto, Paragraphs 22-25. back

3. Expert Group paper on Loss Given Default Mortgages, November 2005 page 1,“…the Mortgage EG concluded that Residential Real Estate is an asset class where downturn LGD is relevant and should be used.” back

4. Basel Committee on Banking Supervision “International Convergence of Capital Measurement and Capital Standards A Revised Framework” Bank for International Settlements (Updated November 2005) (download here) , para 468, “A bank must estimate an LGD for each facility that aims to reflect economic downturn conditions where necessary to capture the relevant risks. The LGD cannot be less than the long-run default weighted average loss rate. In addition, a bank must take into account the potential for the LGD of the facility to be higher than the default-weighted average during a period when credit losses are substantially higher.” back

5. Basel Committee on Banking Supervision “Guidance on paragraph 468 of the framework document" Bank for International Settlements (July 2005) (download here), page 3, “Principle 1. A bank must have a rigorous and well documented process for assessing the effects if any of economic downturn conditions on recovery rates and for producing LGD estimates consistent with downturn conditions.” back

6. Expert Group paper on Loss Given Default Mortgages, November 2005 page 4, Paragraph 13, “The EG agrees that there is value in UK industry developing some minimum requirements for the assessment and application of downturn conditions for this asset class in the UK. The first, the “House Price Index” approach proposes a methodology for the assessment and application of downturn conditions based on minimum requirements. The second, the “Scenario” approach is more statistical in nature. Here firms will model the impact and severity of downturn conditions through scenarios and simulations. Both approaches recognise that the last housing recession is a reasonable event to consider in understanding downturn conditions.”