Bank Systems & Technology is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Core Systems

05:45 PM
Connect Directly
RSS
E-Mail
50%
50%

AUTOMATED VALUATION MODELS GROW ON GMAC-RFC

GMAC-RFC relies on automated valuation models to streamline certain aspects of the mortgage process.

For GMAC-RFC, America's largest private-label issuer of mortgage-backed securities and a leading warehouse lender, it's not enough to accurately estimate the value of the properties in its extensive portfolio. It also has to do so quickly and inexpensively.

GMAC-RFC, based in Minneapolis, Minn., recently began using automated valuation models, or AVMs, on every loan coming through its AssetWise decision support engine. The company has a contract with Basis100, based in Toronto, to use its Solimar AVM technology, and with Mortgage Risk Assessment Corporation, Jersey City, N.J., for its Home Price Indices and HPA2000 model.

AVMs use real estate transaction information to calculate the percentage change in the value of a particular property. The calculation is performed through the examination of local market factors (the "index" method) or by using information about comparable properties to estimate value based on statistical regression tools (the "hedonic" method). These extrapolations help to determine whether a property in early payment default needs a review appraisal.

"Sometimes we see properties put down as being worth $300,000 but the AVM comes back and says the property's worth $50,000. It could have been a 'flip,' it could have been a misrepresentation of the appraisal, or the appraiser could have been in cahoots with the loan officer or with the lender," said Brian Broomfield, director of modeling and analysis for GMAC-RFC. "You don't have to go out and do review work when the AVM value indicates the seller gave us a good appraisal value. It's likely that we're not going to be able to find a big value deviation to force repurchase on that loan."

Broomfield estimates AVMs save $1.5 million to $2.0 million per year by eliminating unneeded review appraisals. "Ultimately, it's a loss mitigation technique," he said.

Also, compared to human appraisers, AVMs provide relatively unbiased indicators of value. This, in turn, makes underwriters more efficient since they only revisit cases where the automated appraisal differs significantly from an expert opinion, rather than having to review a larger sample.

These advantages have made AVMs a staple among secondary market participants, including government-sponsored enterprises Fannie Mae and Freddie Mac. "The whole premise is like FICO Fair, Isaac & Co. credit scores--AVMs can improve the efficiency, effectiveness and objectivity of mortgage origination," said Broomfield. "And like FICO, the adoption of AVMs by Fannie and Freddie will improve the acceptance in the mortgage industry of AVMs."

Going forward, GMAC-RFC hopes to use AVM technology to provide investors with valuations of their entire portfolio; to create AVM hybrids serving so-called "gap" markets that are over-served by a full appraisal but under-served by a regular AVM; and to automatically select the most appropriate appraisal option for a given property.

Comment  | 
Print  | 
More Insights
Register for Bank Systems & Technology Newsletters
Slideshows
Video
Bank Systems & Technology Radio
Archived Audio Interviews
Join Bank Systems & Technology Associate Editor Bryan Yurcan, and guests Karen Massey and Jerry Silva from IDC Financial Insights, for a conversation about the firm's 11th annual FinTech rankings.