Tech Fails: A Big Bet That Cost the IRS Billions
Two out of three software development projects end with partial or total failure, and the biggest projects fail most often, effectively wasting billions of dollars and thousands of person-hours in value every single year (1). In this series, we look at preeminent failures in big tech projects, try to uncover the factors that led to the disaster, and figure out the lessons to be learned.
Today, we’re investigating the demise of the U.S. Tax Systems Modernization (TSM) project of 1986. The project was estimated at the time to cost between $8 and $10 billion and was scheduled for completion in 2001.
The Project
The U.S. government initiated the TSM in 1986 because its tax processing system was outdated and in desperate need of repair. At that moment, the processing system had remained virtually unchanged since it was automated in the early 1960s. For instance, most of the 200 million returns that the Internal Revenue Service received each year in 1986 were still submitted in paper form, and only part of the information from these forms was keyed into computers.
The TSM project intended to change all this by creating a new tax processing system that virtually eliminated the reliance on paper, and that made taxpayer information available to IRS employees wherever and whenever it was needed. Specifically, the agency envisioned a paperless work environment in which tax returns can be filed electronically and where information on paper returns and other documents is quickly converted to electronic data. In its 1986 Annual Report, the IRS boasted encouraging initial results, and laid out plans to expand this program nation-wide:
Once the system is fully in place in the early 1990s, possibly more than 30 million taxpayers a year may choose to file returns electronically. (2)
The project consisted of a number of massive redesigns to critical legacy components. The components were depended-on by thousands of citizens daily, making this an extremely complicated, risky, and ambitious challenge. The Integrated Case Processing (ICP) and the Integrated Collection System (ICS) components would become stations for “one-stop” processing of taxpayer requests, plugging into the Corporate Accounts Processing database, a modern, nation-wide database of taxpayer records.
The Demise
Fast-forward to 1997 — just four years ahead of the initially planned release, the IRS officially announced that it was scrapping the entire TSM project, after conceding “that it had spent $4 billion developing modern computer systems that do not work in the real world.” According to the press at the time, “a spokesman for the commission said the cost of shutting down the project “would be astronomical.” He said 12 other systems were under review to determine if they also should be killed.” (4)
As noted in a 1995 report on the TSM by the GAO, “after 8 years and an investment of almost $2 billion, the IRS has realized only marginal improvements in its operations.”
The systems that the IRS has delivered to date have marginally improved IRS’ current tax processing and compliance operations. For instance, the systems have shortened processing time and reduced the amount of paper needed for some processes. However, they were not built to be an integrated part of the comprehensive TSM program and they have not delivered the large increases in capability and customer service that IRS hopes to have in the future. (5)
What Happened?
As admitted at the time by Assistant Commissioner of the IRS Arthur Gross, the primary reason for scrapping the latest IRS modernization effort and starting over was that, in a nutshell, the systems then under development “did not work in the real world.”
A couple of official reports shed more light on the specific issues that brought this mammoth project to its knees. The GAO report from 1995 mentioned above initially surfaced several deficiencies with respect to TSM technical activity:
- There was no official implementation schedule for TSM projects (this had been one of the long-standing problems plaguing the project).
- Discrepancies were found between multiple Master Plan documents, while specific projects, activities, or timelines were inconsistent from one document to another.
- The IRS had not prioritized projects based on their relative contribution to operational improvement and insistently held all modernization projects to be equally important.
- The measurement of project performance indicators was not implemented at any level. Without links to performance measurement, mistakes may not be discovered and might be repeated in subsequent planning.
- The IRS did not have the proper mix of skilled and experienced technical and management personnel to implement the TSM. However, as noted, the need for experienced technicians and managers was extensive, and it could take years before the IRS had the proper skill mix to complete the project successfully.
A 1996 report on the TSM by the National Research Council also identified some key points: (3)
- Large-scale, multiyear development projects require the right technology, but more importantly, they need proper management practices and procedures.
- Although the IRS had competent senior operational managers, it lacked senior technical managers, especially senior technical managers, with experience in very large systems development.
- The IRS had failed to create a concise overall TSM architecture plan that would allow it to establish specific performance goals or metrics against which progress could be measured and reported. “A detailed technical understanding of TSM, as represented by precise technical specifications and accurate system models, has not been demonstrated by the IRS over the past five years.”
In 1998, the IRS started work on yet another modernization project; this time dubbed the Business System Modernization program, funded by another $3 billion, and supposed to finish around 2013. This new attempt was also killed in 2009, after delivering only 15 percent of the capability intended — but that’s a story for another episode!
If you enjoyed this article, press the “Follow” button below for more content that will help you become a better technology professional. You can also subscribe to our YouTube channel, where you’ll find our best articles made into nice videos you can enjoy.
Telepat North is an invitation-only club for outstanding coders, offering members free books, courses, learning resources and swag! If you’re interested, you can learn more by visiting north.telepat.io.
If you’re interested in why software projects fail, make sure to check out “Software Development Failures” by Kweku Ewusi-Mensah, an empirically-based study of why software development failures happen, and the lessons we can learn from them.
(1) The Chaos Report 2015, The Standish Group International, Inc — https://www.standishgroup.com/sample_research_files/CHAOSReport2015-Final.pdf
(2) 1986 Annual Report, the Internal Revenue Service — https://www.irs.gov/pub/irs-soi/86dbfullar.pdf
(3) National Research Council. 1996. Continued Review of the Tax Systems Modernization of the Internal Revenue Service: Final Report. Washington, DC: The National Academies Press. https://doi.org/10.17226/10771.
(4) IRS admits its $4 billion modernizing is a failure — www.baltimoresun.com › bs-xpm-1997–01–31–1997031030-story
(5) TAX SYSTEMS MODERNIZATION: Unmanaged Risks Threaten Success — https://www.gao.gov/products/T-AIMD-95-86
