Off-MLS Transactions – Method for Local Evaluation

by

 

Members Need to Understand Potential Consumer Harm

Two years ago, at Clareity’s MLS Executive Workshop, MLSListings (the 16,000 member regional MLS serving the greater San Jose area) presented its study of Off-MLS Transactions, evaluating how common the practice was in their market and demonstrating the economic harm caused to consumers. At the 2014 MLS Executive Workshop, CoreLogic presented a wider study, although it still covered only 5% of the country, and the subsequent Inman News article raised awareness further. But everyone knows that “real estate is local,” so now MLSs around the country are beginning to consider replicating the original MLSListings study in their market. Understanding the scope of the practice and the harm that may be caused to consumers is necessary if an MLS is to discuss the trend with membership at the local market level. Such studies may also inform further discussions at the national level.

This article describes MLSListings’ method of performing that evaluation at the local MLS level:

The Process: 

  1. Pull all sales for SFR, Townhome and Condo from tax records within settlement date range.
  • Exclude all records with blank price field
  • Manual scrub for non-applicable transaction records
  • Format County APNs to match MLS data record APN (varies by County)
  1. Pull MLS data for same time period
  2. Match APNs, eliminate duplicate records. Output is properties sold off-MLS.
  3. Median prices for MLS and off MLS are compared by geographic data

Additional considerations:

Manual review of county records data can further improve quality of comparison. Known examples include:

  • Individual cases in which each unit in a multi-unit property is recorded with the price of the entire property (e.g. 200+ condos in San Jose for $118 million each). These were manually removed upon discovery following reasonable review.
  • Individual cases in which transactions are recorded as full grant deeds, yet upon further inspection, sale price is well below the reasonable lower limit of a comparable property, buyer/seller are similar entities, etc. Manual scrub for such cases yielded approximately 3 percent of transactions falling within these conditions, deemed within an acceptable range of deviation.

Accepted conditions/assumptions: 

  • County record data represents all sales within a time period and a given geographic area County classifications of SFR, Condo and Townhome correspond to same classifications within MLS system
  • County record data ‘settlement date’ is the best available match to MLS ‘COE date’
  • APN number is the best available match to compare identical records in county data and MLS system data

Where final sale price is withheld in the MLS record, last list price is substituted as the best available match