Performance budgeting still a work in progress

The Office of Management and Budget is preparing to release another tool designed to help agencies measure the performance of information technology programs, but agencies are still struggling with how to link performance goals to budget requests.

The Performance Reference Model is expected out soon. It is one of the five reference models of the federal enterprise architecture. The model measures how well processes, such as the delivery of services to citizens, are working governmentwide.

The model is designed to work with the Program Assessment Rating Tool (PART), which OMB developed to help rate program performance when making budget decisions.

Officials at OMB are working carefully to make sure that the model and PART are complementary, which is why the Performance Reference Model has been in development for a while, said Robert Haycock, acting manager of the Federal Enterprise Architecture Program Management Office. The reference model "doesn't compete with it, it actually helps facilitate it," he said April 3 at a conference sponsored by the Digital Government Institute.

PART builds on the agencywide performance issues raised by the 1993 Government Performance and Results Act, which seeks to link program funds to agencies' performance.

PART assigns each program — including IT programs — a grade. It was used by OMB in formulating the fiscal 2004 budget request. But in the first year, at least half of the programs reviewed did not have adequate measures, and OMB officials hope that the differentiation between input, output and outcome metrics identified in the performance reference model will help agencies learn how to develop good measures for specific programs.

PART is still a "work in progress," but it will help agencies improve program performance over time, said Donna McLean, chief financial officer at the Transportation Department. As the guidelines for the ratings become clearer, agencies can incorporate them in program planning and restructuring, she testified April 1 at an oversight hearing of the House Government Efficiency and Financial Management Subcommittee.

"In an effort to make the relationship between funding and performance more transparent and understandable, several agencies have begun to modify the preparation and presentation of their budgets to clarify how proposed funding relates to performance goals and outcomes," McLean said.

However, if the guidelines for PART change significantly, agency officials won't be able to predict what they need to do to receive a favorable rating on a program, she said.

Until the measures are in place, efforts to increase or decrease funding for federal programs depending on their performance will be in vain, said Robert Shea, counsel to the deputy director for management at OMB, at the Digital Government conference.

The Bush administration has made it very clear that, for now, PART will only be used to evaluate individual programs, not to compare programs. But others believe programs should be reviewed in groups based on common missions.

Rather than grade each program alone, PART should review projects with similar goals to determine which program is working and which ones will provide the maximum benefit, suggested Maurice McTigue, director of the Government Accountability Project at George Mason University's Mercatus Center.

"The complaint I have on PART at this moment is that it's not comparing program with program," McTigue testified at the April 1 hearing. "It's looking at programs on a stand-alone basis. If you have that [comparable] evidence in front of you, it becomes possible to cut funding from the programs that don't work.

"Congress needs to de-fund the programs that are shown to have no beneficial effect," McTigue said.