Financial Incentives Can Improve Public Sector Performance

RAND Study Examines Performance-Based Accountability Across Education, Early Childhood, Transportation, Health Care, Health Sector.

Performance-based accountability systems can improve how employees deliver public services, but evidence demonstrating how effective these systems are at achieving their performance goals is rare, according to a new RAND Corporation study released today at a Capitol Hill briefing.

The study by RAND, a nonprofit research organization, examines whether performance-based accountability systems — which are enjoying growing popularity in the public sector — are having their desired effects. It examines nine performance-based accountability systems in five sectors in which public services play an important role: child care, education, health care, public health emergency preparedness and transportation.

Researchers found that in optimum circumstances, these systems — which link financial or other incentives to measured performance — are an effective way to provide better public services. But creating an effective performance-based accountability system requires careful attention to choosing the right design for the system, which must be monitored, evaluated and adjusted as needed to meet performance goals.

“While incentives tied to performance can shape behavior, it is very hard to build a formal accountability system in a public service area that gets it right the first time,” said Brian Stecher, lead researcher of the report and acting director of RAND Education. “You need to take time to develop the measures, identify the incentives, test them out and fine-tune the system to promote the desired outcomes.”

Researchers conclude that the most clear-cut example of an effective performance-based accountability system is cost-plus-time (A+B) contracting for highway construction, in which contractors receive a financial bonus for completing road and construction projects with an accelerated timeframe.

Unfortunately, many of the conditions that make A+B contracting work in highway construction are not found in other forms of public service.

For example, the No Child Left Behind Act of 2001 combined explicit expectations for student performance with well-aligned tests to measure achievement, along with strong consequences for schools that did not meet performance targets.

While the legislation led to important changes in teacher behavior, teachers appeared to respond narrowly in ways that improved measured outputs with less attention to long-term outcomes or goals.

Researchers also examined health care, including hospital and physician/medical group “pay for performance” programs, and found that small financial incentives — frequently combined with public reporting — have modestly improved the quality of care delivered.

The size and nature of the incentive is an important factor in the overall success of the program. For example, in many health care pay-for-performance programs, the potential financial rewards represent a small percentage of the overall physician pay and, as a result, may not serve as a strong incentive.

Stecher said there are several elements that contribute to an effective performance-based accountability systems, including establishing goals that are widely shared among the groups involved in the program, providing clear and observable measures, and providing incentives that apply to individuals or organizations that have control over the process.

“The incentives must be meaningful to those who are providing the service,” Stecher said. “Otherwise, their performance will not improve.”

The study, co-authored by 10 RAND researchers whose collective expertise spans all of the five sectors examined, makes several recommendations to developers of performance-based accountability systems, including:

• Realize that performance-based accountability systems are not always the best option for improving performance. Designers must consider those factors that may hinder or support a system's effectiveness.
• Determine if the performance measures are at the individual, department or organizational level.
• Make the performance rewards large enough to matter, but not larger than the actual benefit of the improved performance.
• Create measures that people can influence. Do not hold people accountable for problems outside of their control.
• Implement the program in stages to allow for opportunities to modify the program as needed, and to identify and fix shortcomings in the program.
• Monitor and evaluate the program. This is the only way to detect problems and improve the accountability system over time.

The report, “Toward a Culture of Consequences: Performance-Based Accountability Systems for Public Services,” can be found at www.rand.org.

The report was prepared by RAND Education, a leader in providing objective, reliable research and analysis on educational challenges that is used to improve educational access, quality and outcomes in the United States and throughout the world.