About


Take On Payments, a blog sponsored by the Retail Payments Risk Forum of the Federal Reserve Bank of Atlanta, is intended to foster dialogue on emerging risks in retail payment systems and enhance collaborative efforts to improve risk detection and mitigation. We encourage your active participation in Take on Payments and look forward to collaborating with you.

Take On Payments

« Pssst…Have You Heard about PSD2? | Main | The Five-Star That Flops »

February 6, 2017


ACH: No Trace Left Behind

In my payments research role, I believe that one problem with ACH is the lack of any definitive method for identifying a payment and any associated return, dishonored return, or contested-dishonored return using only the existing 15-digit trace number. Ideally, the trace number alone should facilitate the correct retrieval of payment or return details even if other payments contain duplicate payment details, such as for recurring payments.

This PDF file contains an image that outlines the complex web of relationships that can be used to trace back returns to the original payment. Without the benefit of a unique trace number, the identification of the original payment could involve using common data elements to minimize misidentifying the payment.

A unique trace number would offer the following advantages:

  • Unambiguously identify a specific payment
  • Facilitate tracking features similar to what is available from package delivery services such as transmittal, settlement and receipt date/time, and similar tracking of any associated return(s)
  • Enhance risk-monitoring capability
  • Simplify reconciliation and auditing
  • Flag or prevent a return from settling before its associated forward payment
  • Identify "orphan" returns sent across the public network when the original payment was sent privately between financial institutions (FI)
  • Link together forward and return payments for certain international payment applications that are not possible today

Under NACHA rules, the FI originating the payment assigns a unique 15-digit trace number; the trace number's uniqueness is necessary to differentiate each payment in the batch. Uniqueness is not mandated across payments in other batches in the same payments file. Consequently, a trace number could be repeated in multiple payment files on the same day or across many days—and, even more troublesome, within the same payments file. NACHA strives for uniqueness by mating the trace number with an associated batch number, transmission (file creation) date, and a file ID modifier. Unfortunately, any return of a payment only passes along the original trace number without the benefit of the mated data.

A possible solution that could overcome the current limitations of the trace number would be a one-time-use, ACH-operator-assigned, 15-character alphanumeric trace number. When the originating network operator receives a file, the operator would replace the FI trace number with a unique trace number that he or she would forward to the receiving FI. Any return sent back to the originating FI would have the unique operator trace number converted back to the original FI trace number. For convenience, a cross-reference file associating operator trace numbers with FI trace numbers could help facilitate non-network communication between originating and receiving banks.

Operators could guarantee uniqueness by allowing an operator trace number to contain digits and upper and lowercase letters. Expanding to a 62-character set results in over 3.5 trillion distinct values using the last seven characters of the trace number (the first eight characters are the originating FI's routing and transit number). Further requiring at least one non-numeric character allows differentiation with FI numeric-only trace numbers.

What are your views on the benefits and disadvantages of non-repeatable trace numbers?

Photo of Steven Cordray  By Steven Cordray, payments risk expert in the Retail Payments Risk  Forum at the Atlanta Fed

February 6, 2017 in ACH , payments | Permalink

Comments

If the unique trace number could be assigned on the FI side, it would eliminate the extra step of forwarding of a unique number (which has its own chance of failing to forward) and some possible non-repudiation risks.

Perhaps this could be done by assigning each FI their own identifier, and pair that identifier with a unique number which is never used across batches, file IDs or dates. (A unique ID which is never reused since the FI Identifier would always make it unique across all FIs).

This would mean changes on the FI side and so some analysis would have to be done to find the cost benefits for NACHA, FI and FRS.

Posted by: B. Guhanick | February 8, 2017 at 09:40 AM

I like this idea. It would also make it extremely easy for an FI to research a transaction within their records by using the unique trace number. You are looking at around 20 billion transactions per year so the 3.5 trillion should easily cover the 6 year record retention requirement.

Posted by: David L Payne | February 7, 2017 at 06:58 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search



Recent Posts


July 2017


Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

Archives


Categories


Powered by TypePad