Needless to say, this is a big mess. While not all of the return filing backlog could have been avoided, given the months the IRS mail processing was either fully shut down or impeded because of pandemic conditions, there was a solution that would have sped up the processing of returns significantly. That solution is 2-D Barcoding or some variation thereof.
This is not a new idea. In fact, in the “most serious problem” discussion about e-filing in the 2004 National Taxpayer Advocate Annual Report to Congress (pages 89-109), I noted that 31% of individual returns were prepared on a computer and subsequently printed out and mailed. The IRS calls these tax returns “V-coders” because they are coded with a “V” when the IRS processes them. At the time, the IRS was struggling to achieve the e-filing mandate of 80% of all returns by 2007, established in the IRS Restructuring and Reform Act of 1998 and codified in IRC § 6011(f)(1) and (2). By the 2004 filing season, only 61 million of the 128 million individual returns were e-filed, or 48%.
We proposed the IRS implement 2-D barcoding, by which computer-prepared returns would generate a horizontal and vertical code containing all information on the return upon printing. Upon receiving the paper return, the IRS could scan the code, which would convert the information into digital form and allow the return to be treated the same as an e-filed return. Although it would not eliminate the need for human beings to open returns, this approach eliminates the need for manual keystroking of data from the paper return and quality review of that keystroking; it also allows all data to be captured, rather than select fields, improving audit selection and other downstream compliance operations.
In a September 2004 memorandum responding to our inquiry about such implementation, the Director of Customer Account Services, Wage & Investment Operating Division responded that the IRS was not pursuing 2-D bar coding for individual returns because “promoting this method of paper filing would slow the growth of e-filing.” (See footnote 72 on page 102 of the 2004 NTA Annual Report to Congress.) Note that at the time IRS did have this technology and was using it to scan paper Form 1065 and Form 1120S Schedule K1s, as well as paper Form 941s. And note that in 2004, at least 17 states were using 2-D barcoding for paper return filings.
In its formal response to the report’s preliminary recommendation that the IRS initiate processing of 2-D barcodes on paper returns, the IRS stated “… the IRS believes that offering taxpayers a paper alternative to e-file is counterproductive to Congressional e-file goals and sends taxpayers a mixed message as to what Service policy is with respect to e-file.”
In response to this myopic position, TAS responded:
While 2-D bar coded returns may not produce as many benefits as e-filed returns, the IRS must acknowledge that a certain population of taxpayers will always refuse to e-file for one reason or another. Even when the IRS meets the 80 percent goal, it will still need to make paper returns available to the 20 percent of taxpayers who continue to file in such manner. We disagree with the IRS’s point that offering this technology to individual taxpayers would send mixed signals. If taxpayers are properly informed about the benefits exclusive to e-file, it is doubtful that 2-D bar coding technology will attract current or prospective e-filers. However, such technology will still benefit those unpersuaded paper filers and the IRS by avoiding transcription errors and reducing processing costs.
2004 NTA Annual Report to Congress, P. 108, citations omitted.
I reiterated this recommendation in 2006 and 2008 testimony before the Senate Finance Committee and the House Ways and Means Oversight Subcommittee, respectively, and other hearings over the years, as well as in many subsequent Annual Reports, including the 2019 Purple Book, which compiles the NTA’s legislative recommendations. The 2020, 2021, and 2022 Purple Books continue to reiterate this recommendation.
The idea gained traction despite the IRS opposition: section 2104 of a bipartisan early version of the Taxpayer First Act actually included a mandate for 2-D bar coding. The proposed mandate would have been effective for returns filed after December 31, 2019. Ummm … let’s see: the pandemic arose after December 31, 2019. So if that mandate had passed or been included in the Taxpayer First Act ultimately enacted, IRS would have the capability of scanning those millions of returns backlogged during the pandemic and get them into the e-filing system.
It now looks like the IRS may finally be coming around to this idea. The 2022 Purple Book reports “[t]he IRS has indicated an interest in adding 2-D barcodes on all IRS forms and outgoing correspondence due to the industry-proven efficiencies associated with extracting machine-readable data from paper returns and correspondence. It is exploring both 2-D barcode and OCR [optical character recognition] technology with the software industry as part of a pilot program.”
Now we come full circle. On December 17, 2021, IRS Chief Counsel issued Program Manager Technical Advice (PMTA) 2022-02 regarding the IRS’s authority to mandate that tax-software developers embed a 2-D barcode in all tax returns. The memo notes that the current IRS process of manually entering paper return information is “resource intensive and error prone.” In a very cursory (one-paragraph!) analysis, Chief Counsel concluded the IRS cannot require tax software developers to embed barcodes on returns prepared on their software, because “Section 6011(e)(1), which provides the Treasury Department and the IRS authority to prescribe regulations to determine ‘which returns must be filed on magnetic media or in other machine-readable form,’ does not apply here, because the proposed mandate is not directed to taxpayers or other filers.”
One has to wonder what would have happened if, in 2004, IRS has embraced the idea of 2-D barcoding and sought Chief Counsel advice at that time. Having the legal opinion that a mandate required legislation, would Congress have enacted such a mandate, given the significant taxpayer benefits and resource benefits of such a mandate and the minimal burden on software developers? At the very least, would the mandate of the earlier version of the Taxpayer First Act have made it into the final version?
We’ll never know for sure, but one thing we do know is that had the IRS responded to our 2004 recommendation about 2-D barcoding in a positive fashion rather than one-size-fits-all approach it adopts so often, we would not be in the mess we are in now. This sad history lesson demonstrates that resources is not always the reason for lack of progress.