FORUM PROFILE

HMDA – Telephone – App Sex – when documented incorrectly, but signed by App?

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #268693
    pparks
    Participant

    We are currently using an in-house software application that not only produces the regulatory edits as part of its process, it also has some of it’s own enhanced edits for all of its users. One of those edits uses some kind of proxying ideology that looks at the first name of an applicant, and if it appears to be a possible name usually applicable to the opposite gender, the edit produces for one to review. Note that we have not currently adopted proxying for any of our analysis processes. However, I found this edit to be a good operational edit to ensure the data was carrying over correctly from the source URLA document field. During a recent internal audit, they reviewed our edit report and noted a particular application. During their review, they looked at the license of the applicant and the pronouns used by the MLO in the 1008 comments. They then took it further by having a manager speak to that MLO. It was during that conversation where I am sure the MLO was presented the evidence that the MLO stated they had populated that field in error. The applicant was not female. He meant to check off male. If one is to believe that the MLO’s recollection of the telephone conversation was not inappropriately influenced by the evidence presentment and this was truly a mistake, should this be considered a HMDA error in light of the following:

    1) The applicant was presented with their documented responses during initial disclosure and at closing, and they signed off both times without correcting what was documented (attesting to it).
    2) The enhanced edit specific to the in-house software application is not a regulatory edit and not an edit we had created ourselves.
    3) Looking at the driver license and making our own judgments is discriminatory in nature as one would be presuming how the applicant identifies, or how they should identify.
    4) Calling all bankers involved in applications that involve this edit is not beneficial or reliable. How can we trust their memory? How can we trust that we are not inappropriately changing/influencing their mind/response? Could us choosing to change the gender now without the applicant’s acknowledgement of its legitimacy, and with it is not being supported by the file, risk damaging the reporting integrity of the application process that occurred?

    I do understand and agree with Internal Audit’s desire to report everything correctly, but is this situation truly considered to be a HMDA error as the LAR truly reflected the signed application and credit decision, or is this a battle to be fought elsewhere…or is the resulting discovery by internal audit inadmissible and not to be considered due to the methodology and path they chose to get there, and that’s even if the recollection of the MLO is correct? What is your perspective of this situation? HMDA Error, no error, or observation? Should we really take our reviews as far as internal audit did in this case? Thanks for looking into this and getting back to me.

    #271198
    jholzknecht
    Keymaster

    In my opinion, this is not a violation.
    * If the borrower completed a written application and indicates sex as male and transcription error occurred when the data was loaded into the reporting system – an error has occurred and should be corrected.
    * In an oral application, if the borrower indicates sex as male and the MLO records it as female – an error has occurred. If the error is detected immediately, it should be corrected. When the application is signed by the applicant, if the error is detected, it should be corrected.
    * When auditing at a later date, assuming the entry was incorrect based on the name of the applicant is incorrect. A male applicant named Bill might check sex as female because she identifies as female.

Viewing 2 posts - 1 through 2 (of 2 total)
  • You must be logged in to reply to this topic.