In the wake of the White House's attempt to put the Community Development Block Grant (CDBG) program on the chopping block, a House subcommittee held a hearing to determine whether a program as diverse and flexible as CDBG could be evaluated using OMB's one-size-fits-all performance measurements. Factors Influencing the Hearing The hearing focused on CDBG in isolation from other programs that fared poorly under OMB's performance measurement process, but the hearing did not address the systematic problems of that process. The popularity and importance of CDBG do not alone explain the form the hearing ultimately took. Instead, there are political factors at play that prevented the hearing from addressing these larger problems. Political Context Foremost among the many heated disputes inspired by the White House's fiscal year (FY06) budget submission was the proposal to eliminate CDBG in its current form and to combine it, at significantly reduced funding levels, with several other programs to be administered by the Department of Commerce rather than Housing and Urban Development (HUD). The White House justified its proposal by claiming that CDBG failed to produce results, as measured by the Program Assessment Rating Tool. Although Congress ultimately rejected that proposal, the problem remains that the White House failed CDBG in the FY 06 PART and could do so again in the future. The consolidation proposal proved controversial for lawmakers on both sides of the aisle, because CDBG is an enormously popular program. Administered by HUD's Office of Community Planning and Development (CPD), CDBG gives federal money to state and local governments to support a variety of activities aimed to assist low- and moderate-income communities. State and local governments are given great flexibility in developing a program that best meets the needs of their given community. CDBG money has been used to enforce housing codes, build sidewalks and sewer systems, clean up and redevelop brownfields, and build affordable housing, among other activities. The proposal to cut CDBG based on its supposed ineffectiveness put Republican lawmakers in a difficult bind. On the one hand, constituent pressures prevented them from agreeing with PART's assessment of CDBG. On the other hand, PART is a powerful tool that enables this administration to use the notional objectivity of quantified performance measurement to justify predetermined political outcomes; Republican lawmakers were thus constrained from attacking PART in its entirety. Moreover, a Republican member has sponsored a bill, recently reported out of committee, that would essentially codify the PART. When the House Government Reform Committee's Subcommittee on Federalism and the Census held its May 24 hearing, GOP members resolved this tension by seeking inputs on improvements to PART that could allow it to accommodate the special characteristics of CDBG, while ignoring the larger problems of PART. CDBG and the Broader Context of PART Although the House hearing focused entirely on CDBG, that program fared poorly in its PART assessment for the same reason that many other block grants -- and, for that matter, many programs important to the public interest--also fared poorly. PART is, by design and in practice, a one-size-fits-all test focused less on meaningful assessments of actual effectiveness than on the political whims of the budget examiners conducting the assessment. The White House's Office of Management and Budget (OMB) has used PART since 2002. Under the guise of a neutral scientific tool, PART evaluates programs using questionable criteria, some of which conflict directly with programs' authorizing statutes. Although OMB purports to have specialized sets of questions for different types of programs (research and development, regulatory, block grants, and credit programs), in practice the questions for each type of program are essentially indistinguishable. Within each type of program, OMB does not even pretend to particularize its performance inquiry depending on the specific characteristics of a given program. After conducting these problem-ridden assessments, OMB then uses the PART scores to justify changes in program budgets. PART is particularly ill-suited for all block grant programs, not just CDBG. The basic purpose of block grant programs is to send funds to the states with minimal strings attached. Imposing nationwide performance measurement requirements would force the states to gather uniform data, whether or not their specific programs were designed with those data end-points in mind. In contrast with the states' rights agenda that drove the development of many block grants, federal performance measurement (in particular a one-size-fits-all test like PART) aggressively trumps the states' own performance measurement processes and could have the effect of holding states accountable for consequences beyond compliance with basic legal requirements. Holding the block grant program itself accountable for failing to gather uniform performance data counters Congress's intent for the program. In fact, the evidence indicates that the White House has used PART in a systematic attack on block grants, including CDBG. Grant programs rate significantly lower than in PART reviews than all other programs on average. For example, in the FY05 PART reviews, OMB scored no block grants as effective even though it gave that rating to 11% of all programs, and it rated only 5% of all programs as ineffective but gave that failing score to 43% of all block grants. This trend continued in the FY06 PARTs, in which only 27% of block grants were deemed effective while 47% of all other programs received the highest score. The Hearing as a Balancing Act Although CDBG was not alone in being slated for deep cuts justified by a failing PART score, the May 24 hearing focused on CDBG in isolation. The House has already rejected the controversial proposal in the White House's FY06 budget submission to combine CDBG with 17 other programs from five different agencies as the new Strengthening America's Communities Initiative, which would be relocated to the Department of Commerce and funded at $3.7 billion -- a 34 percent cut, without adjusting for inflation, from the programs' FY05 $5.6 billion budget. As to be expected from the political context, the hearing avoided the deeper systematic problems of PART and instead attempted to find a middle ground between saving CDBG from the ax while not rejecting PART altogether. Disputing the CDBG PART Assessment Instead of addressing the larger problems of PART, participants in the hearing disputed the basis for the program's rating of "ineffective." Critics within the agency and in the hundreds of communities served by CDBG have argued that the performance measures used by OMB do not adequately capture this flexible and dynamic program. Witnesses at the hearing focused in particular on OMB's decision to score CDBG a zero for clear programmatic purpose, based on the argument that "the program does not have a clear and unambiguous mission. Both the definition of 'community development' and the role CDBG plays in that field are not well defined." HUD deputy secretary Roy Bernardi refuted OMB's PART rating and defended the performance of CDBG, saying the program has a purpose clearly outlined in the Housing and Community Development (HCD) Act of 1974, which established CDBG. Bernardi further asserted that HUD follows the intent of the law in its administration of the program. Congress intentionally designed the law governing CDBG to minimize HUD's role and allow communities to develop programs that meet their specific needs. In fact, in 1981, Congress specifically modified the act to reduce HUD's role from making qualitative assessments of grantee programs to simply assuring that grantees complied with the governing statute. In its hands-off approach to administering CDBG, HUD is following the directive from Congress. Neither Bernardi nor the other witnesses used this occasion to emphasize that CDBG was not alone in being penalized under PART for following Congress's stated intent. For example, OMB penalized the Consumer Product Safety Commission, Occupational Safety and Health Administration, and Mine Safety and Health Administration for failing to use cost-benefit analysis in rulemakings, even though these agencies operated under statutory authority and, in the case of OSHA and MSHA, Supreme Court precedent forbidding the practice. Also not discussed is an additional constraint on HUD's ability to collect performance information: the Paperwork Reduction Act. The PRA limits an agency's ability to collect any information from 10 or more people -- in other words, almost every occasion that an agency would be collecting information -- by requiring OMB approval. Agency information collections, no matter how important to the public interest, are collectively subject to a fictional budget of "burden hours," or the estimated time required to complete information collection. That budget in turn is subject to periodic reduction goals. The PRA mandates an annual reduction in burden hours, putting pressure on agencies to minimize the collection of information notwithstanding the need to gather more information about performance. Saving Face for PART All the participants in the hearing walked a political tightrope when the hearing shifted its focus to question of applying performance measures to CDBG. The performance measurement movement has been so successful in altering the mindset of many in government, the nonprofit sector, and elsewhere that there is little critical thinking about the underlying assumptions of performance measurement - namely, that it is possible to think of "performance" as a discrete set of activities susceptible of measurement in politically neutral, quantified data, and that these performance measures should be central rather than subsidiary to high-quality fiscal and management decisions. Moreover, the political context constrained any in-depth discussion of the problems of PART itself. Accordingly, Bernardi and state and local groups at the hearing argued that performance measures can be established for CDBG and that PART simply used the wrong measures. For Bernardi and many of CDBG's proponents, the solution is that performance measures should be developed by the local communities rather than handed down by OMB. The "genius" of the program, as one state group called it, is its flexibility; it can cater to the specific needs of a given community. Since programs are developed by the community and tailored to community needs, that community knows best what the outcomes of the program should be and how they should be measured. As Bernardi stated in his testimony before the subcommittee, "because the CPD formula block grant programs promote maximum flexibility in program design and since the use of these funds is driven by local choices, HUD believed that performance based measurement systems should be developed at the state and local level." In fact, HUD has been working with OMB and stakeholders since 2003 to develop better performance measures for the program. In September 2003, CPD issued a notice to all program grantees outlining its efforts to improve performance measures and encouraging grantees to develop their own local systems for measuring program outcomes. In response, 43 percent of grantees reported using performance measures or that they are working to develop such systems. Since that time, a working group of representatives from an array of national housing and community development associations came together to develop performance measures that reflected the objectives and outcomes of their programs. The resulting system allowed grantees to determine their own objectives "based on the intent of the project and activity. While program flexibility is maintained, the system offers a specific menu of objectives, outcomes and indicators so that reporting can be standardized and the achievements of these programs can be aggregated at the national level," according to Bernardi's testimony. The witnesses sidestepped the systematic problem built into PART itself: namely, that the OMB budget examiners applying PART look only for a couple of measures applied across the board in a program. Despite HUD's work on developing performance measures that reflect the diverse objectives of the program, OMB still chose to evaluate CDBG using the one-size-fits-all PART measurements that not only failed to capture the intricacies of the program but also evaluated CDBG grantees on criteria that fell outside the scope of the authorizing statute. The one-size-fits-all reductiveness of PART cannot accommodate the multiple, project-specific outcome measures envisioned by the witnesses. Forcing PART to accommodate these program-specific concerns for CDBG alone, moreover, would do nothing for all the other programs suffering under the crudely mechanistic rigidities of PART. The witnesses also danced around the problem of using such a flawed tool in making management and budget decisions. OMB budget examiners appear to launch PART assessments with predetermined political outcomes in mind, and the White House then uses the good government rhetoric of "results" to justify slashing the budgets of programs that fail the assessment. Though even CDBG's advocates noted problems with the programs, none of the witnesses suggested that cutting the budget was a solution. In fact, Sheila Crowley, president of the National Low Income Housing Coalition, argued that HUD is often crippled by lack of proper resources. Over the past decade, HUD has continued to administer the same public services with an ever-diminishing budget. PART is, naturally, oblivious to these resource constraints. The witnesses argued for an alternative role for performance measurement: using performance measures as a management tool rather than a guide to budget decisions. Subcommittee Chair Michael Turner (R-OH) asked witnesses repeatedly if and how performance measures can be used not just to prove the effectiveness of a program but also to improve program effectiveness. Turner's questions prodded the witnesses to find a role for performance measurement as a vehicle for determining best practices and quality case examples to be circulated among CDBG grantees (but not imposed upon them). In Turner's alternative vision, HUD would gather project-specific performance information and then share the success stories of one grantee with others. Currently, HUD gathers performance information only to ensure compliance with laws and regulations. In the mid-1990s, HUD introduced the Integrated Disbursement and Information (IDIS) reporting system, which allowed grantees to input data about program activities. While IDIS tracks the financial status of CDBG grantees, it does little to integrate that information with the various performance reports required of grantees. Turner seemed to see HUD's emphasis on compliance not as the result of the HCD Act and the 1981 follow-up minimizing HUD's role but, rather, as a missed opportunity for this kind of best practices consultation. HUD is currently working to improve the IDIS system in a way that will make aggregating data possible. This interesting alternative approach is not possible in the current performance measurement climate. PART itself still demands an over-all assessment of the performance of CDBG and other programs in their entirety, without regard for the separate successes of individual projects or even the block-grant philosophy that the states themselves, as the fabled laboratories of democracy, should be experimenting on a state-by-state basis. The stated purpose of PART and the larger government performance movement is that budgetary and management decisions should both be determined by a shared universe of "performance" data; the alternative approach of performance measurement that determines neither but results only in nonbinding suggestions runs counter to that purpose. Finally, a bill pending in the House, the Program Assessment and Results Act, would give OMB a blank check to continue with its current approach and would do nothing to reshape performance measurement in the more palatable ways suggested by Turner and the witnesses at the hearing. Despite the impossibility of Turner's alternative approach, the hearing concluded on that note. Several witnesses stated that, with better performance data, HUD could provide this kind of best practice information to grantees. Further representatives from state and local groups said that they would benefit from best practices guidelines and technical assistance from HUD.