Newsroom May 11 2017 - Shelley Metzenbaum

May 10, 2017 - Accountability Expectations: Getting Them Right to Enable Public Servants to Serve

By Shelley Metzenbaum

 

This week, we pause to celebrate and thank the millions of government workers engaged in public service. Let me suggest that we also commit to fixing government systems to make it easier for government employees to make meaningful progress on organizational mission, whether that mission is public safety, public health, discovery of new knowledge and places, or something else.

One way to do that is to adopt performance and evidence-informed management practices that inspire effort, wise action, and well-informed innovation.  For that, we must help people across government appreciate the value, and master the methods, of managing and communicating using goals, measurement, analysis, and well-designed trials. Just as important, we need to identify and promote adoption of reasonable accountability expectations. 

Over the past few decades, noteworthy strides have been made around the world, both at national and sub-national levels of government, embracing more sophisticated performance and evidence-informed management practices. These practices have resulted in greater beneficial government impact, often at lower cost. Progress has been made on outcomes as diverse as reducing crime, disease, teen pregnancy, and smoking, and improving air quality and the speed of pothole repairs.  Well-framed goals, measurement, and analysis also strengthen public understanding of what government does and why.  In situations where individuals need to choose among options, they enable better decisions to meet personal needs and preferences.

At the same time, problems can and do arise when unreasonable accountability expectations are established. For example, a robust body of evidence finds that specific measurable, ambitious targets motivate and inspire innovation provided the targets are reasonable relative to resources, knowledge, and skills. By definition, all stretch targets are not likely to be met. (A rule of thumb is to expect 75% attainment of stretch targets.) When government leaders ignore the evidence and manage or conduct oversight suggesting they expect 100% target attainment rather than trends moving in the right direction at a reasonable pace, it tempts selection of timid targets, measurement manipulation, and, on occasion, implosion of measurement systems. 

Similarly, evaluations that ask, “Does this program work?” make even the most skilled program manager wary of undertaking evaluations for fear a program will be cut even when the underlying need for the program persists. It is far more productive to frame evaluations to answer questions such as: “What practices and sets of practices will work better and where, when, and for whom?” and “How can we accomplish the same or more at a lower cost?”

What, then, are reasonable accountability expectations?  Let me suggest adapting the accountability expectations articulated by Jack Maple, a key architect of CompStat, the crime-reducing performance improvement system New York City used to drive down its crime rate. Specifically, let me suggest the following accountability principle.

No one gets in trouble when performance trends in the wrong direction provided he or she knows and clearly communicates, in terms the public understands: 

  • The goals and specific objectives (how much of what by when);
  • Trends and relevant related information;
  • Why a trend is moving in the wrong direction and plans to reverse the trend or better identify causal factors; and
  • Why a trend is moving in the right direction and plans to achieve a higher beneficial return on future government spending.

As we celebrate public service, let us also commit to embracing and promoting these reasonable accountability expectations to enable public servants to tap their own and others’ intelligence, ingenuity, and energy so they can better serve the public.

 

Shelley Metzenbaum is an Academy Fellow and Senior Fellow at the Volcker Alliance.