Indiana Court Autopsies Welfare Privatization Effort

Guest Blogger

David Super

Lively debates continue in both the academy and the political arena about the potential and risks of privatization. One of the most hotly-contested areas has been the administration of social welfare programs.

On July 18, a Superior Court in Indiana ruled on the lawsuits between the State and IBM concerning that state’s path-breaking attempt to privatize public assistance program administration. The story it tells provides a fascinating cautionary tale that can inform in evaluating similar proposals in the future.

The court provides a nice summary of its entire 66-page opinion in the first paragraph: “Neither party deserves to win this case. This story represents a ‘perfect storm’ of misguided government policy and overzealous corporate ambition. Overall, both parties are to blame and Indiana’s taxpayers are left as apparent losers.”

Shortly after being elected governor in 2004, Mitch Daniels declared his intention to transfer administration of Temporary Assistance to Needy Families (TANF), food stamps (now the Supplemental Nutrition Assistance Program, or SNAP), Medicaid, and other programs for low-income people to private contractors. They made their determination to privatize broadly clear from the start “then proceeded methodically, resisting accommodations that might have compromised their goals.”

Part of the rationale for the change was improving client service, in particular by reducing the number of mandatory trips to the office. In fact, a great many of these trips had been deliberately required by the Department to increase the burden of receiving benefits and drive down caseloads. Others resulted from policies seeking rapid reductions in food stamp error rates. The rapid decline in Indiana’s food stamp participation by eligible working poor families resulting from the error-reduction drive had already demonstrated how vulnerable people can be affected by hasty, radical changes in program administration. Unfortunately, Indiana failed to heed that lesson.

Although other companies initially expressed interest, in the final stage of the procurement process only a group led by IBM submitted a bid. Several months before the final contract was signed, Texas stopped the roll-out of its own, similar human services privatization initiative “because applications were not being processed.” Nonetheless, Indiana proceeded to sign a ten-year contract for $1.3 billion, with the contractors’ work front-loaded and their payments backloaded, much of it falling in years after Gov. Daniels would be term-limited out of office. The State claimed that savings from privatization ultimately would make the contract cost-effective but was publicly vague about how those savings would be obtained. The State then negotiated eleven separate contract modifications over the next two-and-a-half years, adding $178 million to the contract price. The court finds a massive waste of taxpayers’ dollars and is sharply critical of state officials for charging into this “super-sized” undertaking without more of a basis for believing it would work. It finds the State officials’ “competence … in this project is sometimes open to question.” The court implies that the economics of the contract never made any sense for either side.

The State began implementation of privatization just three months after concluding the contract. It approved roll-outs to many additional counties before the results in the pilot counties could be evaluated in meaningful detail. The court finds that “[t]he parties saw implementation issues immediately.” So, I might add, did food pantry operators and other human services providers, who reported that the privatized system was in chaos. The system required households to contract private eligibility workers by phone for most matters related to receipt of benefits, but the contractors’ telephone lines were persistently overloaded. Two years after implementation began, the State began to question the continued viability of the privatized system. A few months later, it terminated the contract with IBM.

The court details remarkable patterns of inept, often chaotic misadministration by the contractors. It also paints a stunningly incompetent picture of state officials’ design and implementation of the contract. Among the problems with the contract were that the State failed to anticipate the disruptive effects of natural disasters and the surge in need that the recession caused. Both of these omissions are astounding: tornados and floods are common enough in the Midwest, with the Food Stamp Act providing for special aid to disaster victims since 1977. And every recession since the 1970s has brought large surges in demand for food stamps and other benefits. Yet when the contractors’ resources were diverted to addressing these problems, program administration fell apart and the State had not given itself effective recourse.

Additional problems resulted from implementation of the kind of policy changes that are the common outcome of the political process. When Indiana launched a new health care program for low-income people, it had little bargaining leverage with IBM on how much extra it would have to pay to have the program administered or what resources IBM would provide.

The contract was so badly written that, although the programs were clearly being administered horribly – with high error rates causing federal penalties and massive delays in providing benefits to households – IBM was nonetheless meeting the objective criteria set out in the contract. (The contract expressly disclaimed any warranties of “uninterrupted or error-free operations”.) The State therefore was unable to establish a material breach justifying ter­mi­na­tion of the contract and was held responsible for paying many of IBM’s losses from the termination. The result is that Indiana must pay more than $52 million to IBM as a consequence to its termination of the contract. It is not clear how much of that payment reflects actual value received and how much is deadweight loss, but the figure nonetheless seems high enough to mark this litigation as a disaster for the State. And it easily could have been (and in many other states likely would have been) much higher.

Moreover, the State agreed to numerous steep liquidated damage provisions should it terminate the contract early. To get out of paying some of those, the State had to convince the court not to enforce the contract it had signed, specifically to disallow many of the terms as unenforceable penalties. To a significant degree, it succeeded in this effort. If the State anticipated that these provisions could not be enforced when it negotiated them, it would seem to have been in bad faith. If not, it was extremely lucky.

IBM’s performance is also troubling. For example, it claimed that Indiana owed it for the cost of four reprogrammings to accommodate subsequent changes in law but could only find two of those work orders, and both of those were to bring the system into compliance with legal rules in place long before the contract was signed.

In addition, IBM’s subcontractor ACS, whose performance the court particularly scorns, apparently spent much of the contract period lobbying the State to terminate IBM and turn the responsibility over to ACS. In addition to directly violating ACS’s contract with IBM, this also raises questions about how effec­tive even lucrative contracts are at getting the parties to focus on program administration – or whether such lucrative contracts may induce contractors to seek even greater gains for themselves.

The State made matters worse by continually interacting with the subcontractors behind IBM’s back and in violation of its contract with IBM (and by disregarding IBM’s repeated complaints about this). The urge to talk with those administering the program on the ground is certainly understandable, but if the State was going to do so it should never have contracted away that right.

The State was hampered severely in its efforts to enforce, and ultimately to terminate, the contract by its own politically motivated self-congratulatory public statements. The State would point to this or that failure of IBM’s administration, but IBM would counter with public statements Governor Daniels and other senior officials were making at the same time about how well privatization was working. By the time the State terminated the contract, it had already effectively exonerated IBM for much of its worst performance. This suggests that politics can hamper contract negotiation and enforcement at least to the same extent that it can compromise direct public administration. More generally, it suggests that privatization pursued for ideological purposes is likely to be very bad business for the taxpayers.

One final lesson from all of this is that these kinds of contracts take relatively little time to initiate (especially when the State is rushing to meet a political timetable) but they take long enough to fall apart that the political process provides no effective corrective. Governor Daniels, who pushed this through and presided over the State’s incompetent contract administration, the waste of vast sums of public funds, and the collapse of public administration is already in his last months in office. He thus is beyond the voters’ reach, even if they felt the same outrage that Judge Dreyer clearly does.

David Super is Professor of Law at Georgetown University Law Center. You can reach him by e-mail at das62 at law.georgetown.edu