By Ronald Jasgur
The mortgage industry is rapidly being reshaped and restructured by the current housing crisis, but I’m continuing to see a blatant problem that is receiving almost no attention…a problem that is keeping values artificially depressed, enabling fraud in the REO and short-sale space, and leaving a wake of failed real estate transactions across the country.
It’s appraisals. And BPOs. And not for the reasons you probably think.
The issue here is data integrity. Or, more specifically the misrepresentation of pertinent facts about a particular property in its MLS profile. Some of this occurs due to laziness. Much of it is intentional, though. And all of it forms the basis for each and every near-future valuation of real estate in markets and neighborhoods across the country.
MLS information is the largest (and often only) source of local market activity and provides the data that populates every appraisal and BPO relied upon by lenders…both for new mortgage approvals and the decisioning process that moves short sales and REO sales through the pipeline. Heck, even local tax assessors use this data for their market valuations.
The data in the MLS is only as accurate as the listing agent wants it to be. REO listings often don’t include accurate room counts, square footage, condition information, and more. And so many REO listings still don’t include multiple photos that give a “feel” for the property.
In short sales we see even more egregious errors: intentional misrepresentation of school districts and municipalities, staged photos with intent to show more distress than there really is and most importantly, false listing statuses that prevent properties from being found at all.
Here are a few examples of how bad data affects the valuation process:
- A 4-bedroom, 3.5-bath home of 3,000 sq ft gets listed as a 3-bedroom, 2.5-bath home of 2,980 sq ft. A subsequent search of properties matching the actual criteria fails to show information relating to this sale.
- A lakefront home shows no indication of waterfront characteristics. Subsequent MLS searches for lakefront properties miss this one altogether.
- A property in top-notch school district “A” is listed showing it is in nearby school-district “B”. Families wanting to move their kids into district “A” never see the listing, and appraisers looking for comps in the same district easily overlook it.
What’s the big deal?
For starters, every lender uses some type of automated valuation software that aggregates all of the local sale information and spits out a target price or range based on the lender’s requirements (i.e. NPV, REO, etc.) Flawed data in the MLS affects future values, especially in markets where there aren’t a large number of sales in an area or price range.
Appraisers have been taking a beating over low valuations, killing deals. But a close look at the underlying data identifies that the agents who populate the MLS with information may be as much to blame as the appraisers themselves.
Agents need to police their local MLS. They need to report inaccurate information, so MLS staff can investigate and levy penalties against agents who willfully manipulate housing characteristics for their (or their seller’s) personal gain.
Appraisers also need to report inaccuracies that they uncover during due-diligence and field work. If the assessor says a property is 2,401 sq feet but the MLS listing shows 1,900 sq ft (or 2,800 sq ft), it must be reported and adjusted accordingly.
Multiple Listing Services need to scrub new listings against archived listings to see if there are any material data point discrepancies that should flag for a review. And multiple photos should be mandatory.
In meeting with banks and servicers about our offer processing and negotiation platforms, OfferSubmission and VerifiedShortSale, almost all of them seem to feel that the valuation products they rely upon are often flawed. It isn’t the product that’s bad, though; it’s the data that drives the reports. Garbage in, garbage out.
With so much attention focused on AMCs, new appraisal forms and data requirements, and the state of housing in general, it’s time to look squarely at the root of the problem and solve it once and for all.