Legacy of program assessment tool unclear
Even opponents of PART say it’s likely to have some lasting and positive effects.
While the Bush administration's Program Assessment Rating Tool has been a controversial pillar of the President's Management Agenda, elements of it -- such as transparency and solid performance measures -- are likely to survive the upcoming political transition, an Office of Management and Budget official and outside observer said this week.
OMB released its last round of performance ratings on Friday, providing detailed assessments of more than 1,000 federal programs. The final evaluation, which incorporated ratings for 67 programs that were reviewed for the first time or reassessed in the past year, found that 80 percent of initiatives were performing adequately or better.
Adam Hughes, director of federal fiscal policy at the nonprofit OMB Watch, credited the administration for making the assessments available online.
"The effort to publicize the data, particularly on the ExpectMore.gov Web site, the transparency initiatives and access they've given to the process and the data is positive … and that needs to continue," Hughes said. "There are parts of PART that could be more transparent and open, which is where we get to some of our criticisms, but the Web site and data are excellent."
Dustin Brown, OMB deputy assistant director for management, said increased transparency would be one of PART's most significant legacies. "The fact that we made …over 1,000 assessments available online with sometimes painstaking detail attached … to anyone with an Internet connection is a tremendous accomplishment," Brown said.
He added that he is passionate about ensuring the more than 5,000 performance measures developed and implemented through the tool remain an integral element of the next administration.
"We know so much more about outcomes and outputs and efficiency measures now, and that's what program managers use and live by," Brown said. "It's how we link higher level 'how is [Health and Human Services] doing?' assessments with 'how is Bob the program manager with HHS doing?' [It's] how we are making that linkage with organizational and individual performance. That's what I and performance managers across government will ensure live on."
Scores have improved significantly since OMB published evaluations for the first 20 percent of programs to be rated in February 2003. Of the programs assessed at that time, half could not demonstrate results and only 44 percent were adequate or better. Despite the progress agencies have made in evaluating their programs, Hughes and Brown said there is endlessly more potential in how performance data could be used.
Many critics point out that PART has limited value if Congress does not use it as a budgetary tool. Congress has been minimally, if at all, engaged in the assessments, Hughes said.
"If the next administration wants to … continue to make improvements they have to engage Congress far more," he said. "It's going to be difficult because Congress feels this is their jurisdiction, their job. They can make the argument that they have a good system already."
Brown said data mined through PART could be extremely helpful in improving interagency coordination and understanding federal investments in broad categories of programs, such as those to address homelessness or unemployment.
"We have horizontal problems and a vertical bureaucracy," Brown said. "Looking ahead there is a great opportunity that awaits career civil servants in a new administration to do cross-cutting reviews and analyses of issues that are important to next administration and the American public."
Regardless of the benefits, it is unlikely PART will continue by name. "Every administration has their own spin on how to evaluate programs," Hughes noted.
Brown said OMB has gone to great lengths to ensure the evaluations are as objective as possible, but Hughes argued that the results remain tainted because OMB is political and the performance levels are subjective.
"The simplistic nature of the tool is a problem because it allows for broad stroke manipulation -- 'effective' or 'ineffective,' that rating can mean tons of different things," Hughes said.
Brown acknowledged that "nothing's going to be 100 percent objective," but noted that OMB makes clear "the type of evidence that's required for each question" that contributes to the rating. "Looking through the questions PART asks, it's hard to argue any of them really are political in nature," he said.