I would like to know more about where the task of inputing the HMDA Data resides in most banks. Also, when there is the need for a HMDA scrub, is it typically done by the area that input the information into the data collection software or by the area where the information is input into the loan system that is used to download into the data collection software. Thankl you for your help.

Our HMDA data is input from various sources within the bank. We have problems with that and recognize the need to centralize it. The Compliance Department scrubs 100% of the data, but if centralized where there are not so many "I don't know what I'm doing" hands in it, we could drop back to sampling the data. At my former bank, the data was uploaded to HMDA/CRA software from the bank's mainframe system and dropped into correct buckets according to collateral and purpose codes. All of the HMDA fields, such as GAI and GMI were in the mainframe system. The Credit Administration Department assigned AND second-reviewed the codes for accuracy. And they entered and second-reviewed all the data into the mainframe system. Our data hit the LAR with very few errors. The short of it....IMHO, you can't audit your own work. Therefore, the data needs to enter the LAR, not by the Compliance Department, whose duty it is to audit or scrub the data, either 100% or by sample.

Thank you so much for your insight.

Front line processors enter the HMDA data into the loan processing system.
Comparison of application to info on system has been incorporated into the FNMA prefunding process.
Finally, compliance (me) pulls the LAR and scrubs it quarterly and again at year end.
I use pivot tables to do my scrub.