IRS Is Ignoring Red Flags in Its Love Affair With Direct File

The Inflation Reduction Act gave the funds to study the cost of developing and running a direct e-file system. The agency then produced a 2023 report offering an unsurprising conclusion. Most taxpayers (72%, in a survey that failed to allow for neutral answers) would be very or somewhat interested in free IRS-provided tax software.

With that finding, the Treasury Department greenlit the program. And within six weeks of the filing season, amid breathless announcements from supporters, the agency decided to make Direct File a permanent option.

Not everyone is enamored. That’s because there’s a belief that the IRS (or Treasury) is putting its thumb on the scales.

To disprove that belief, IRS officials should answer three questions: Would we take on this product without external pressure, how did we determine success metrics, and does the program accurately weigh all the cost?

The IRS’s job is to help taxpayers meet their filing, reporting, and payment requirements, and to provide compliance services to those who fail to meet them.

OMB Circular A-76 establishes federal policy for the performance of commercial activities. It states that “the longstanding policy of the federal government has been to rely on the private sector for needed commercial services” and that “agencies shall identify all activities performed by government personnel as either commercial or inherently governmental.”

After several taxwriters asked the basis of the IRS’s authority, IRS Commissioner Danny Werfel testified last month that Section 7803 of the tax code provides him the authority to run Direct File. Such an assertion runs afoul of the principle that Congress doesn’t hide elephants in mouseholes.

Many argue that tax return preparation should be free because one is obligated to file a tax return. But this argument is difficult to take seriously. The IRS doesn’t charge taxpayers a user fee to file their returns. It provides paper forms. It supports (if inadequately) Free File. It supports the Volunteer Income Tax Assistance and Tax Counseling for the Elderly programs, as well as Low Income Taxpayer Clinics.

In none of those cases does it build, maintain, and provide customer service for its own software. Our tax system relies on self-assessment. Taxpayers, not the IRS, are required to determine taxable income.

The IRS has a statutory authority to prepare a tax return in two instances: when a taxpayer fails to file and when the submitted return is fraudulent.

Does the IRS want to enter the tax preparing business full time? And if so, should it provide software and support to prepare all federal tax returns? Before asserting it should provide Direct File, the IRS should be certain it’s able to say yes to both those questions.

Concept Stage

The IRS just ran a high-profile proof of concept. From my time in the IRS Research Division, determining whether something is technically feasible is a valid research question. But it isn’t the only one.

The agency on several occasions stated publicly that volume wasn’t a key performance metric, yet also provided (and continually adjusted downward) volume expectations. Assessing in a post-tax-season press release that interest was “strong” wasn’t a great look.

The public should know how the IRS concluded what metrics it used to assess the pilot and exactly when the agency set those metrics.

As the IRS considers what’s next, it should know whether it’s comparing apples to apples. The IRS provided white glove service to the 140,000 pilot participants. It trained 400 customer service representatives to support the program. Less than 10% of users sought customer assistance, with an average wait time of 45 seconds. The average customer support volume was 568 calls per day.

Ninety percent of touchpoint respondents who used customer support rated experience as excellent or above average. This isn’t surprising.

While all is glorious in the data IRS discloses, the costs are murky. The agency reported spending $13 million on pilot development and implementation. At the same time, an April 2024 government watchdog report stated the IRS budgeted $114 million in fiscal 2024 for Direct File, or nearly nine times the agency’s public accounting.

The Government Accountability Office report raises many issues, including:

• Direct File cost estimates omit startup costs plus costs (approximately $7.2 million) of 29 US Digital Service staff.
• Insufficient documentation to provide reasonable assurance the pilot will capture necessary data to inform more complete cost estimates.

For context, the IRS in 2022 estimated its enrolled agent application and renewal costs (direct and overhead) to be roughly $10.4 million for a volume of 74,000. For the uninitiated, enrolled agent renewal doesn’t require IRS to write tax prep software from scratch.

There are serious questions about the IRS’s costing and the agency should show its math, in detail.

Unenviable Task

It seems exceedingly unlikely the agency would have prioritized a Direct File pilot while onboarding a new commissioner and taking on investment of once-in-a-generation funding. Yet it began a pilot that required evaluation under immense political pressure.

The cost isn’t free, the decision to compete with private sector isn’t trivial, and a significant target of the analysis isn’t those who believe IRS-provided tax returns are a right but those who are skeptical it is fitting for the agency to absorb the enterprise.

Werfel has stated he wants to build trust, a laudable goal greatly aided by transparency. Direct File expansion presented the IRS a perfect opening to be transparent in its decision making—an opening the agency apparently declined to or, possibly, found itself unable to take.

This article does not necessarily reflect the opinion of Bloomberg Industry Group, Inc., the publisher of Bloomberg Law and Bloomberg Tax, or its owners.

Author Information

Bob Kerr is principal of Kerr Consulting LLC, a tax administration and tax policy advising and consulting firm.

Write for Us: Author Guidelines