In the previous post we explored the concept of funds transfer pricing (FTP), but only on the surface level. Now for the good stuff: how should the FTP rate be assigned? Well, in a number of ways from the simple to the complex. Like a lot of things in life, simple means fast, easy, less data, good enough; whereas complex requires lots of patience and data for a little more accuracy.

Methods of FTP can be placed in one of two categories: Pool Pricing and Matched Maturity Pricing. In the pool methods, loans and deposits are summarized, or pooled, into categories. Generally speaking, pool pricing leans towards simple, but there are degrees of simple and are identified as:

The table below compares the results you might find in choosing one method over the others. The term of the treasury rate chosen should roughly match the average life of the category or product. In the single term method, we have to compromise by using a term that is likely lower than the average loan life and higher than the deposits. With the double rate method, we start to see a funding center gap, and in the multiple rate method, a more precise result.

Source US Treasury 03/21/2015

Matched Maturity Goes to Eleven

 “Where can you go from there? Where? Nowhere. Exactly. What we do is, if we need that extra push over the cliff, you know what we do? Put it up to eleven.” Nigel Tufnel

Matched Maturity requires a download of all customer accounts at the instrument level. With all of the security concerns, this might not be desirable unless customer account numbers can be masked or changed. In Matched Maturity, each loan and deposit is evaluated and assigned it’s very own funds transfer rate, which stays with it until the account matures or reprices.

Single Term/Bullet

A fixed rate loan made on March 21, 2010 would have a different rate from a loan made on March 22, 2016. Assuming this is a 30-year fixed rate mortgage, in a portfolio that has an average life of 7 years, match the average life to a point on the yield curve to find the transfer rate. The 7-year rate on 3/22/2010 was 3.16%. The same rate today is 1.70%. The 30-year Freddie rate in March 2010 was 4.97%, and is 3.73% today. The match funded transfer pricing looks like this:

If this were an ARM or floating rate loan, the average life used would simply be the next reprice date.
What this tells you is that loans made today are more profitable by 22 bps.

But loans don’t pay off in one lump sum like a bullet bond. They amortize over time, so we have to introduce Matched Multiple Term/Amortizing.

Matched Multiple Term/Amortizing

Simply stated, it treats each contractual payment as its own maturity and matches an interpolated yield curve rate with each payment. This is far too much to illustrate here.

Comparing Results

If we look at the results of the different methods, we see a variety of spreads:

So depending on your choice of methods, you will have very different results – and that’s where discussions within the organization become a challenge. What will you do with this data? Are loans originated in different locations more profitable and will you take some action as a result? Will you stop offering certain unprofitable products?

Further, if you aren’t careful about the method used, can’t explain it thoroughly, and don’t have agreement from all corners of the organization, then the results are useless. It simply becomes a report for the finance office.

My personal belief, after years of teaching FTP, cost accounting, and profitability, is that keeping it simple allows for an easier discussion of the results and a lot less headaches in implementing and maintaining an FTP system.

There is a lot more to the discussion than allowed by a couple pages in a blog, so feel free to reach out to me with any questions.