The Financial Information System for California (FI$Cal) project – now in its 17th year (not a typo) of implementation – is slogging along with a revised project plan that has reduced functionality and obscured project costs. FI$Cal is due to miss yet another planned project completion this summer, and will have to extend the project timeline by another two years.
That’s the woeful top-line news from the California State Auditor’s Office (SAO) report on the project released this week.
Citing the repeated nature of its concerns regarding FI$Cal – which the SAO is statutorily required to report on each year to the governor and legislature – this latest appraisal could have been cut and pasted for the most part from the prior year’s effort.
At the bottom line, the SAO stated that the FI$Cal team would not complete the project by its latest scheduled end date of June 2022.
Making Numbers Fit Budgets
As to the FI$Cal budget situation, some background will help.
The original project budget prior to kickoff was $1.6 billion. However, neither the governor nor the legislature was inclined to go along with such a bold investment and scaled it back to $600 million when it was finally approved and funded in 2005.
The state subsequently issued a series of changed orders, what the state refers to as Special Project Reports (SPR) that further modified the project’s scope, schedule, and budget, and increased the project timeline and budget correspondingly. The most current of these SPRs – the 9th version issued in September 2020 – extended the project’s official end date to June 2022 with a projected total cost of $1.06 billion.
The state auditor maintained that this was hardly the true cost of the project due to some not-so-subtle accounting legerdemain discussed below.
A recurring theme to these SAO reports, in addition to the budget overruns and missed deadlines, has been state employee project staff vacancies as well as FI$Cal’s propensity for causing delays in critical state financial reporting.
While the former may be a bureaucratic nuisance it nonetheless contributes significantly to the latter. As the transition from vendor-supplied staffing to state staff has progressed, the state folks must assume a greater portion of the project workload. However, as the SAO points out, the project has had a state staff vacancy rate of more than 15 percent for the past five years, and the project team may struggle to successfully manage this transition.
As to the latter issue of financial reporting, for the third consecutive year the state will issue late financial statements, resulting in part from state agencies’ challenges in using FI$Cal. Timely financial reports are important for the state to maintain a high credit rating and access to Federal funding. The SAO reports that these financial reporting delays may ultimately prove costly. The state’s ability to publish accurate and timely financial statements is important for the state to sustain the trust of financial markets and maintain a high credit rating ensuring access to low-interest debt. If due to financial reporting tardiness, the state suffers a downgraded credit rating, it could substantially increase borrowing costs, affecting the state’s ability to pay for debt-financed projects such as schools and levees.
State agencies’ struggles with using FI$Cal have contributed to these delays in publishing the state’s annual financial statements. According to the SAO, some of the explanations that agencies provided for this lapse included unfamiliarity with the FI$Cal system and its complexity.
State officials have regaled me over the years with anecdotes about the challenges of using the new system, the need to run parallel operation of the legacy financial system CalSTARS to determine the accuracy of FI$Cal, and related concerns.
Playing the Accounting Game
However, perhaps the most troubling aspect of the SAO report is its reiteration of the state’s puzzling admission of its decision to “cook the books” in a manner of speaking. The auditor maintains that there are ongoing concerns about significant project features and processes that will still need to be implemented even after the project is officially considered complete – whenever that is.
As California’s CIO some 20-plus years prior, I’ll admit I employed a similar tactic. I was presented with a departmental IT budget augmentation request to increase the Health and Welfare department’s Statewide Automated Welfare System (SAWS) IT project by $30 million in the following year’s budget. As this project had already tripled in costs over the previous five years to nearly half a billion dollars, and we were under greater and greater legislative and media scrutiny involving such escalations, I was quite reluctant to approve, and therefore have to publicly justify before the legislature another project implementation budget increase.
However, I also believed that the project was finally on track for successful completion by the following year, so I wanted to provide the budget necessary to make that happen. So, I devised a strategy to avoid yet again a major budget increase, which would have been a blatant indication of a potential problem and that would not be missed by the legislature, media or the state auditor.
As we know, IT projects are traditionally composed of various stages. They will have both a development stage, when the system is being “built,” and a much lower costing maintenance and operation (M&O) stage once the system “built” or completed. The cost of the latter does not add to the original project’s total publicized budget, and thus we had an opportunity to avoid boosting project cost at least as far as any potential outside oversight scrutiny was concerned. So, I agreed to approve the $30 million for M&O, and the project’s development stage was officially over.
It was a trick we employed in rare circumstances and which, while I wasn’t proud of it, at least it kept the wolves at bay until we could get the project successfully completed. For obvious reasons it was done sub rosa in agreement between me and the project manager. I thought I had come up with a clever and unique way to continue funding a project and keep any legislative, media or auditor watchdog from braying about another IT project running over budget. This budgetary legerdemain was never intended to be formally adopted as an officially acceptable state of California project management and funding mechanism, but that’s just what has happened with FI$Cal.
My accounting sleight of hand nearly a quarter century ago was never discovered; however, FI$Cal was not so lucky.
As recently retired State Auditor Elaine Howle wrote in her annual report two years ago, “Among our concerns is the manner in which the 2019 project plan update sets a formal end date for the project even though the FI$Cal project will not have implemented promised functionality.” She went on to criticize the project’s continuing strategy of “removing key features from the project’s scope, increasing the budget, and developing unrealistic schedules, resulting in a product that will lack crucial features.” And she concluded by charging that the true cost of FI$Cal has been significantly understated.