Plan to Improve Government Performance Would Actually Undercut Progress

Because they lack a profit-based bottom line, we often assume that federal agencies also lack a natural spur to innovate. But that misses one key source of change, which is that Congress and presidents tend to hold strong beliefs about the need to reform government. A permanent sense of dissatisfaction with government has resulted in the perpetual search for new performance systems: PPBS, MBO, ZBB, CSRA, GPRA, and PART. If you don’t know these acronyms, don’t worry; all you need to know is that they took a lot of time and effort and failed to achieve their goals, justifying a replacement that in turn proved to be no more successful. A new bill, the Taxpayer Right to Know, would not only continue this cycle of failure, but might also undercut existing practices that are actually doing some good. This bill has passed the House and is awaiting consideration on the Senate floor, having been passed out of the Senate Homeland Security and Governmental Affairs committee.

The current performance system in place is the GPRA Modernization Act of 2010. The Modernization Act addresses some of the weak points in the original 1993 Government Performance and Results Act (GPRA). Research I have undertaken with Alex Kroll suggests that the Modernization Act is actually having a positive effect on performance improvement efforts. This is reinforced by a recent Government Accountability Office report on the implementation of cross-agency priority goals, one of the requirements of the new law. We compared employees who were exposed to new routines created by the Modernization Act and those who were not, and compared the rates at which they used performance information while controlling for other factors. Increasing the use of performance data is an explicit goal of both the Modernization Act and its predecessors. We found that those who were exposed to Modernization Act routines were more likely to also report using performance data to make decisions. Further, the quality of those routines seems to matter. For example, if quarterly performance reviews mandated by the Act were well run, this was associated with higher rates of performance information use.

What makes this evidence more compelling is that when we applied the same test to the Modernization Act’s predecessors—GPRA and the Bush-era Program Assessment Rating Tool (PART)—we found no evidence that they were associated with purposeful performance information use. Its always difficult to assess the effect of government reforms. But, the evidence suggests that the Modernization Act is making a positive difference in a way its predecessors were not.

The Taxpayer Right to Know bill is modeled on PART. It would require a program-by-program breakdown including the cost of programs and evidence of performance, though it would skip the summary effectiveness-ineffectiveness rankings that made PART so scary for agencies. In the abstract this is not a bad idea, but it’s a mistake for three reasons.

First, the Act is based on the assumption that taxpayers are going to actively use this data. That is the same expectation that fueled the creation of the now defunct ExpectMore.gov website that was based on PART data, and the current Performance.Gov website that already provides plenty of performance data. Maintaining public transparency is a good idea, but ultimately the public does not routinely seek out detailed program-by-program performance data. They have lives to live, and should be able to expect that public officials are doing the primary work of using data to improve outcomes.

Second, there are real transaction costs that come with these reforms. They are a form of mandatory reporting that requires time and effort. PART was abandoned by the Obama administration partly because both agencies and the OMB found it extraordinarily onerous, distracting attention from policy analysis and performance improvement. The more time bureaucrats spend doing one thing, the less time they have for something else. A new reform will distract attention from the implementation of the Modernization Act, as well as other recently-enacted governmentwide reform legislation such as the Digital Accountability and Transparency Act and the Federal Information Technology Acquisition Reform Act.

Finally, reforms take time to succeed. They need to become part of the cultural DNA of an agency, a process already undermined by the rapid turnover of appointees. One way to think about performance improvement in government is that it is a 30-year war to change the norms and beliefs of employees. That war can be won only if each reform incrementally adapts and improves upon prior efforts. The parts of the performance system that work best have been sustained over time in statute. For example, the Modernization Act formalized the Bush-era practice of putting a Performance Improvement Officer in each agency. By contrast, chopping and changing reforms every couple of years will result in bureaucrats seeing each new model as a burden to be waited out until the next change.

Our analysis of the Modernization Act concluded by saying: “A new president may be tempted to look for another approach or to simply deprioritize the Modernization Act. This would be a mistake: the key policy implication of our study is that the Modernization Act should be given more time. Indeed, perhaps the best policy choice a president can make is to further invest in the implementation of existing routines.” The same advice applies to Congress. Indeed, one of the biggest reasons for failures of congressionally-mandated reforms is that Congress itself rarely uses the fruit of its labors to make decisions, reducing the credibility of every new initiative. The best action that Congress could take to improve performance is to commit to the system in place, rather than layer a duplicative set of reporting mandates on federal agencies.

Donald Moynihan is Professor of Public Affairs at the Robert M. La Follette School of Public Affairs, University of Wisconsin-Madison.