FHA to Require UAD Compliance

Appraisers are getting up to speed on the Uniform Mortgage Data Program (UMDP), Fannie Mae and Freddie Mac’s joint effort to standardize loan delivery and appraisal data. Specifically, Uniform Appraisal Dataset (UAD), which will standardized appraisal data, including data formats, abbreviations, ratings and definitions, and other instructions in the appraisal report forms.

The Federal Housing Administration (FHA) is preparing appraisers for how UAD will impact FHA roster appraisers. According to FHA’s Spring 2011 FHA Appraiser newsletter, appraisers will be required to submit UAD-compliant appraisal reports when using the URAR and the individual condominium unit report form. FHA does not allow the use of the exterior inspection forms.

FHA will announce further guidance later this year probably through a Mortgagee Letter on the use of UAD for FHA-insured mortgages. In the meantime, appraisers can find more information on UMDP and UAD here.

  1. Accuracy of data entry continues to be of high priority within the industry. The mythical words of “Junk in – Junk out” will become increasingly troublesome. The industry as a whole should emphasize upon the accuracy of data entry and upon a format of uniformed definitions so as to prevent from the final results within valuations and statistical reviews being obscured. The various listing services across America need to operate upon a uniformed basis so that their definitions and categories can be understood by the users of this data, regardless of the regions differences.

    Although a posting within a listing service is primarily issued to promote and aid in the marketing of a property, other readers such as appraisers and broker price opinion providers utilize this identical data to evaluate properties like the subject and its peers for value. Customs within many regions of America have created a wide variety of styles or methods of reporting. The interpretations of such styles and methods are viewed differently by other regions of America. The results are that underwriters and review appraisers from other regions across the globe interpret the information differently than as intended within the subject’s region. These differences could ultimately result in trouble for your seller or purchaser. Perhaps it has already happen to you once or twice.

    The industry should understand that the data is utilized within appraisals, automated valuation modules (AVMs) and reviews that are obtained primarily from local multiple listing services and county assessor records. Statistics are becoming a method of evaluation by more users than every before for residential valuations. Statistics rely upon consistent data entry and consistency for defined categories. However, the data is either directly being viewed by someone outside of the region or transferred by appraisers, reviewers and broker price opinion (BPO) providers to someone outside of the region. Than that data is being utilized within statistical algorithms that may be non supportive of your market’s interpretation of value.

    As more users of AVMs and statistical algorithms appear as unbiased evaluations they are only as good as the data that is being entered. For now we are seeing AVMs and statistical findings being based upon poor data sets, yet their findings are being viewed as a tool to approve or not approve a transaction or at the least be supportive of an appraisal or BPO. If you explore the basis of many statistical modules, you may find the initial data that is being relied upon may not be accurate. For example: what if the county assessor records indicates a 300 variance in square footage and the bedroom count reads 2 in place of 3. This error for one dwelling may be from the subject or from a comparable being utilized within the algorithm. Regardless of its use as a subject or as a comparable, if these errors are found to be in multiple numbers within the data entries than you have a flawed result. That algorithm and other supporting statistical data may become troublesome within the underwriting process of your transaction. Because no one discovers the fine details of the faulty data from which the supporting AVMs and or statistical algorithms are being based upon.

    The importance upon the initial data entry position needs to be highlighted and emphasized. Consistency across the various listing services should be a goal of the industry. If various regions want to have the ability to promote unique attributes of a property, than that could be accomplished within additional remarks. But the basic categories should be consistent across the listing services.

    John Wayne Glasener
    Approved Appraisal Inc
    Dardenne Prairie, Missouri

  2. Consistency and accuracy of data are key elements of a good listing service. It would really be beneficial if all the listing services will follow a uniformed approach so as to help the data users better and easily understand the definitions no matter where they are located in the country.