A milestone for measuring results

The Government Performance and Results Act marked a significant milestone last year — its 10-year anniversary. This law should be seen as part of the single most important recent development in public management — the view that the crucial lever for improving the government's performance is to measure and manage according to performance metrics.

The idea is that such metrics can play a role in driving better public-sector performance analogous to industry's profit metric.

Attention to measuring performance is not limited to the federal government; it is a growing movement in state and city governments as well. More interestingly, performance measurement is growing internationally. Most dramatically, British Prime Minister Tony Blair fought his last election campaign on the theme of improving delivery of public services, and those efforts have centered on establishing performance targets.

The progress of efforts to make performance measurement part of federal management is uneven. The act successfully survived the transition from President Clinton to President Bush because they both made it a management priority, with the current administration emphasizing its link to budgeting rather than its use as a management tool. However, at many agencies, performance measurement is a paper drill undertaken by staff isolated from line management.

In teaching performance management in executive education, I recently had an experience that made me feel more optimistic that the practice is taking root. In a three-class sequence I teach, I always start by saying: "Imagine an employee says, 'We're short on budget. Measuring our performance costs money. Why should we be spending money measuring what we do rather than doing more of what we do?' What answer would you give?"

The evolution of answers has been interesting. When I first taught this material several years ago, government employees gave almost exclusively what I would call an accountability answer: "We're spending taxpayer money, so taxpayers have a right to know how well we are spending it." Gradually, a larger proportion has begun to give a performance-improvement answer: "Managers can use performance measures to improve the performance of their organization."

There's a big difference between these answers. The first assumes performance is a constant and that the job of performance measurement is to record the value for that constant — and then perhaps punish an agency if performance is poor. The second assumes performance is not fixed and that if managers use performance measures to motivate, teach and

focus people, performance will improve.

The last time I taught this class to government folks, virtually everyone talked about performance improvement rather than accountability. That's great news. If the only purpose of performance measurement is "gotcha," line managers will resist using it. If they see it as a tool for improvement, they will embrace it.

Kelman is a professor of public management at Harvard University's Kennedy School and former administrator of the Office of Federal Procurement Policy. He can be reached at steve_kelman@harvard.edu.