Critic calls OMB program evaluations biased

The Bush administration's rating tool for evaluating federal programs is subject to political bias and should be reformed to bring in more subject expertise, according to a former federal employee.

In a recent white paper circulated within the federal management community, Eric Bothwell, a 26-year veteran of the Indian Health Service and now a performance management consultant, said the Office of Management and Budget's Program Assessment Rating Tool used to grade most federal programs would itself be rated poorly on its design and effectiveness.

Bothwell said if the PART evaluation process underwent its own assessment, it would earn poor responses on two key criteria: being free from major design flaws and signs that a different approach would better accomplish the intended goals.

OMB, as an arm of the White House, is inherently ill-suited to make objective assessments of federal programs, Bothwell argued. "OMB is a politically focused organization and based on this alone, the role of evaluating federal programs is at risk of being biased by the political agenda of the administration," he wrote.

Citing his own interactions with OMB during examinations of the Indian Health Service, Bothwell said some PART questions are subjective, opening the door to bias, and sometimes generic criteria don't make sense when applied to programs with long-term or complex goals like health and social programs. When used to justify changes to program budgets, he said, the PART's skewed measures of efficiency or effectiveness can lead to unfair recommendations for cuts.

Despite his dispute with fundamental aspects of the program, though, Bothwell's recommendations to reform it work within the existing system.

OMB should modify the process to include at least three parties, he said -- adding a subject matter expert from a similar federal program and an expert on evaluation techniques from the Government Accountability Office to each review team in addition to the OMB examiner. Including GAO's perspective as the legislative branch's accountability arm would boost congressional confidence in the system, he said, potentially turning around the disinterest that appropriators have shown in PART evaluations.

"OMB has a critical and legitimate role in this effort, but they must perform it legitimately and model the standards that they hold other[s] to," Bothwell said. "It will also require them to release exclusive control of the process, collaborate with stakeholders far more, and hopefully embrace a more holistic model of performance management."

Asked to discuss Bothwell's recommendations, Robert Shea, OMB's associate deputy director for management, said, "I'm not interested in contributing to this."

But John Kamensky, a senior fellow with the IBM Center for the Business of Government who worked at GAO for years before becoming a top official in Vice President Al Gore's reinventing government initiative, said many of Bothwell's points were worth discussing.

Kamensky said fine-tuning the PART process could increase its credibility, with positive effects for agencies, the administration and Congress.

Setting PART in context, Kamensky said that before it was introduced in 2001, the OMB budgeting process relied on far less transparent evaluations. "Before the PART it was more ad hoc, it wasn't systematic, and it was secret. With the PART it has a framework, it's something the examiner has to publish publicly, and it's something the agency had a hand in," he said.

He also defended the professional objectivity of most rank-and-file OMB employees, noting that despite the office's White House link, career employees tend to keep private their political leanings.

Kamensky said that after the 1993 passage of the Government Performance and Results Act, a major milestone in the federal government's march toward program accountability, officials considered developing a separate, objective evaluation system. But one of the problems with using a more academic approach was that the results tended to be heavy with caveats and unwieldy when directing policy.

Kamensky also cited a potential barrier to implementing Bothwell's recommendations, noting the constitutionally separate roles of the executive and legislative branches in federal budget-making. If GAO were included in the program evaluations that inform the president's budget request, he said, it could be seen as congressional interference with the executive branch's budget development and decision-making process.

If the links between program evaluation and budgeting were completely severed that would resolve the separation of powers concern, Kamensky noted, but could also decrease OMB and appropriators' interest in relying on PART results.

Congress could take steps to fine-tune PART, Kamensky said, through legislation like a bill introduced in 2005 by Rep. Todd Platts, R-Pa., which would have codified much of the existing process into law. That measure was never enacted.