It has been recommended that the Bank develop and implement a monitoring program to analyze pricing trends involving pricing comparison for prohibited groups to ensure pricing consistency.

There are a lot to considered when conducted such and would like to see if anyone would like to recommend any best practices to conduct this type of review or how to implement a monitoring program. There is no much to go on other then what was recommended. We do have a tier pricing for consumer loans and we do have a loan tracking log that all are loan officers complete with all information to included everything other than national origin. e.g APPLICATION DATE ,TYPE (PL/VISA/MC,APPLICATION AMOUNT,PURPOSE CRA ELIGIBLE, RED FLAG ALERT DETAILS (IF APPLICABLE), CREDIT HISTORY (SATISFACTORY / DEROGATORY) CHARGE-OFFS / COLLECTION ITEMS, MONTHLY NET INCOME, TOTAL DEBT SERVICE, TOTAL DEBT SERVICE, DI%, NET EXCESS, INCOME EMPLOYMENT, COLLATERAL DESCRIPTION, "DECISION: APPROVAL (A) DECLINED (D) COUNTER-OFFER (CO) CANCELED (C)", DATE OF ACTION, COMMENTS: REASONS FOR DECLINE/TERMS OF APPROVAL OR COUNTER OFFER...etc.

I guess how this is conducted is left up to the Bank and I'm just trying to get a handle of this task and trying to work in the right direction, if there is such one.

This is left for interpretation in regards to which direction to go.

Any help would be greatly appreciated. Any samples, templates, resources that anyone may have would also be greatly appreciated. Thank you every one for your kind assistance.

Praying for an answer

What you are trying to do can be either very simple, or complicated. There are many variables with pricing, as you indicated: tiered pricing, risk ratings, loan to value, loan size, etc. That is complicated to build a program to identify any and all deviations, especially since the cost of funds can change daily.

The simple way is to not allow ANY deviation from the pricing sheet! Discretionary pricing is what causes fair lending issues!!
Factors such as risk rating, ltv, account relationship, etc. can be built into the loan policy and pricing sheet. If the lender allows NO deviations, review is easy - pick the highest rate loan and lowest rate loan and compare to the pricing sheets for the respective days the pricing was set. If they match, you're OK.

Would it be feasible to just look at all loans (Denied and Approved) or just approved loans or just denied loans. I'm thinking of looking at 3 months just to get a handled of this review.

For pricing, you would only look at approved/funded.

Thanks

If there are any other that would like to provide their input I would greatly appreciate it. Thanks Everyone for your help

I'm confused. If you have a pricing matrix - would you not only have to look at loans granted that were exceptions?

Otherwise, everyone gets the same price based on your matrix and credit underwriting criteria. What would you be comparing, regardless of the prohibited basis you were trying to use??

And who made this recommendation and why? Just because they could???

Hi Rlcarey,

Yes we have a tier loan pricing matrix - Risk based pricing structure for consumer loans with a low range of X% to a high of XX%which started about two years ago. The tier pricing is based on the several criteria's ranging from the credit score to delinquencies to net excess income.

As for what I would be comparing that is exactly what I'm trying to get a feel of. In regards to who made the recommendation, well it was the Feds.

So to ensure that at least something is done before their next visit, I am trying to work on what would be the best avenue to proceeds. Again everything is left to interpretation and how it is done is a matter for discussion at that point in time.

Thank you for your assistance and any advise would be greatly appreciated.

I would probably take a two-pronged approach. First, review the criteria in your risk-based pricing structure. Are they reasonable? Do they tend to have disparate impact? If so, is there a robust business justification? If the risk-based pricing involves a model, is it empirically derived and demonstrably and statistically sound? Has the model been validated?

Next, I would review any pricing exceptions. How are those distributed by race/ethnicity/gender? Are there particular loan officers that are more prone to grant exceptions? Do you have controls around exceptions?