It appears that the CFPB changed the APOR tables after publish. They updated the variable one year rate from 4.47 to 4.46 without notification. This is not a good sign.

That is concerning. Most systems are set to calculate this once and not go back and recalculate - assumes it is accurate. Not sure how without any communication you would know to go back and reassess. How did you happen to run across this?

I accidently stumbled into it when reviewing a client loan. I couldn't figure out why we had the APOR at 4.47. But we also archive the tables as they are released, so we have proof that it changed.

I'd say we have a pretty good argument for using the originally published rate for the spread if our LOS doesn't redo the calc. The rule says the spread is "the difference between the covered loan's annual percentage rate and the average prime offer rate for a comparable transaction as of the date the interest rate is set", and that we may "use the average prime offer rates published by the Bureau". So, in my opinion, if the table had the wrong APOR on the date the rate was set, we can still rely on that rate. Even if they change an old rate 3 weeks after it was originally published and the rate was set, that still isn't the APOR they had listed in the table on the date we set the rate.

Agree with you cgorham. However, that still doesn't simplify things for everyone from an audit or exam perspective when these things get questioned and they are left playing detective to try to determine what happened.

Or if you bet the bank on the loan not being an HPML or Section 32 loan.

And in a civil lawsuit, how do you plan to prove that?

Our LOS retains copies of the tables that are published each week. So I would get the one from the week the rate was set and point out the APOR the CFPB provided on their table.

My comment was just in relation to the HMDA spread. Section 32 and 35 are different issues that I haven't looked at. Our LOS is actually still using the APORs from 12/25/17 for 32/35 because the CFPB failed to give reasonable notice on the changes they were making to the table formats and location. Fortunately those APORs are all equal to or lower than the current APORs, so it is being overly cautious and shouldn't result in any violations. We are still planning to go back and recalculate them all once it is fixed. The HMDA spread in our system recalculates every time we pull a LAR, so theoretically it should automatically be corrected for any changes they make.

I'm definitely not saying that this it would be a best practice. If we know we used an incorrect value, then we obviously have a responsibility to get it corrected. I'm just saying that we could at least have an argument if we happen to miss one due to the CFPB's... technical issues in this area.