With last week's release of the Home Mortgage Disclosure Act data for 2012, many reporting financial institutions are likely to have HMDA on the agenda of their next compliance committee meeting.
Item 1 on that agenda is likely to be technical compliance. Does the institution have its HMDA disclosure statement ready to go? "Ready to go" means having its disclosure statement available at its home office and, if it has offices in other MSAs, either available at one branch within each MSA or ready to provide a copy upon receiving a written request.
For many financial institutions, item 1 may be as far as the HMDA discussion goes. But in an era of increasing regulatory scrutiny of HMDA data to identify institutions to single out for greater Fair Lending analysis, not to mention growing interest and increasingly sophisticated analytical tools used by activist groups, it is prudent to extend the HMDA discussion beyond item 1.
For the prudent financial institution, item 2 on that agenda ought to be "How'd we do?" Of course, that question is broad and can be tackled in many different ways, but at a fundamental level may include an analysis of the institution's denial rates to minority applicants relative to non-minority applicants. That analysis can be further extended beyond a borrower ethnic/racial characteristic to include a geographic component by analyzing the institution's activity to High-Minority Census Tracts (HMCTs), which are census tracts where the percentage of the population that considers itself minority exceeds the percentage of those that consider themselves non-minority.
Item 3 on that agenda then becomes "How do we compare?" If the HMDA 2012 data show, for example, that your financial institution denies 46% of the applications it receives from minority applicants, compared to 36% from non-minority applicants, is it time to panic? The first place to go to provide context for those numbers is to compare them to the same numbers on an aggregate basis for other lenders in your MSAs or counties. For further context, it can be helpful to compare your institution's performance to just those financial institutions that have the same regulator as your financial institution.
The above three items will put your compliance committee in a good position to address the all-important fourth item: "Do we need to dig deeper?" If your numbers of themselves could be perceived as potentially unfavorable, and context provided by a preliminary comparison to the performance of other lenders either on an aggregate basis for all institutions as a whole or just those regulated by the same agency doesn't resolve the matter, then it is time to explore ways of digging deeper.
There are many number of ways this deeper digging can go. You could seek to isolate to specific combinations of loan types, property types, and loan purposes. You might look at amount applied for relative to income. The key is to seek to narrow down the number of variables in the records to a result in as close to an apples-to-apples comparison as possible, short of conducting a matched pair analysis of files. However, if digging deeper doesn't answer enough questions, then it may be time to conduct a matched pair analysis of files, where declines of applicants from a "prohibited basis" group are compared to the closest matching approvals from applicants from a non-protected group.
We have prepared a sample of what a basic HMDA report that an institution might prepare for its compliance committee meeting might look like. You can view it here: http://www.nbrisk.com/reports/hmda/hmda/hmda.html. It was prepared using commonly available spreadsheet tools and data that are freely available from the FFIEC's website. The report you prepare may look different; this is only one example of how you might take a closer look your numbers.
However you choose to do it, the key is to be sure to take a closer look at your numbers, and to do so ideally before your regulator does.