We are getting ready to do our redlining analysis by MSA and our vendor has asked for the five top producing MSA's for each year. They told us that by # of loans or by $ volume of loans; it up to us. I have the list sorted by total $ amount of Loan volume because I think that is more meaningful for fair lending purposes. Is that what everyone else is doing?

I would use number of loans rather than dollar. It would reflect the penetration into all areas by outreach. An inner city loan might be $150K, while one in a more affluent tract 10 times that amount. Usually applications and loans in majority-minority and LMI tracts are smaller, so identifying the outreach could give the bank a better perception of how their outreach is going.

I see your point. Makes good sense. Thank you.

One other thing, I happen to like using applications rather than loans.

The bank can determine the origination rate, and how it compares with their peers. Sometimes, in the lower income tracts/persons, the denial rate is legitimately higher - young being new (or bad) credit, DTI, making and being able to afford less, and LTV, not having reserves for a down payment. Using applications reflects the outreach.

For redlining, also look at the housing demograohics of the tracts. I've seen some inner city tracts with 67% rental rather than OO housing. The opportunities to lend in those areas are severely restricted. Document what you find.

I'd go with # of loans. When regulatory agencies conduct a redlining analysis, they use the # of applications/originations to determine the bank's "material" markets aka the markets they are going to review. This is primarily because they rely on HMDA data for the analysis and use that data to compare lending performance against adjusted aggregate lenders (similarly situated lenders within the same market).

I would use number of applications and number of loans. Dollar volumes are really meaningless for redlining analysis.