An Essential Guide to the ‘Stat’ Revolution in Government

A new book provides a comprehensive picture of the performance-management practice that has taken the public sector by storm.
August 27, 2014
By Stephen Goldsmith  |  Contributor
Professor of practice at the Harvard Kennedy School and director of the Innovations in American Government Program

The concept of the "Stat" program" -- created by Bill Bratton and Rudy Giuliani in New York City and taken to new levels by Martin O'Malley as mayor of Baltimore and governor of Maryland -- appears in a variety of forms. While some state and local governments merely appropriate the "-Stat" suffix as an inoculation against public complaints, others pursue this performance-management technique rigorously in a way that enhances public accountability and drives organizational improvement.

PerformanceStat book coverTranslating performance measurement into true productivity gains is the subject of a new book of intensely relevant case studies by my Harvard Kennedy School colleague Robert Behn, The PerformanceStat Potential: a Leadership Strategy for Producing Results. Each of its 16 crisply composed chapters stands alone -- indeed, this book is a veritable encyclopedia on the management practice that has come to saturate the public sector since its first incarnation two decades as the New York Police Department's CompStat. Then and now, as Behn points out, its success depends on four fundamental principles: "accurate and timely intelligence shared by all"; "rapid deployment of resources"; "effective tactics and strategies"; and "relentless follow-up and assessment." Behn is aggressively comprehensive in teasing out the key operational components that spur these aphorisms into action -- and produce results.

The thread holding these concepts and practices together is dynamic leadership. As cities become more responsive and practitioners turn to digital governance, an effective leader remains the driving force that puts digital tools and smart solutions to work. Behn explains that "in trying to understand any effective PerformanceStat, too many people seek to catalog the explicit components of the organization's system. Thus, they fail to appreciate the leadership team's subtle tacit behaviors." Such tacit knowledge evades explanation; it can only be shown, not told. Accordingly, a chief executive -- whether a program coordinator, department head or city manager -- must produce what Behn identifies as the "human components of organizational competence." These include a results-focused mindset, an analytical-learning bent, a diversity of leadership roles and purpose-driven persistence.

A competent organization is formed by competent members. In the traditional model, public employees are given a set of prescribed responses to situations they may encounter. But these rules, when applied to specific contexts, often defy intuition. Common sense and innovative thinking cannot be codified. Government in general, and Stat programs in particular, are most effective when the bureaucratic overseers allow civil servants to use discretion. This makes it possible to identify "positive deviants" -- those individuals who are doing their work better, whether they know it or not, and who hold the key to discovering what works in the highly contextual frame of a specific jurisdiction. Likewise, the poorest performers can be identified and enabled to improve. The Stat process reframes accountability, too often thought of as a finger-pointing game of benchmarking performance, into a model of personal responsibility, organizational learning and productivity growth. "To be truly effective, to help all subunits improve their results," Behn argues, "a PerformanceStat has to be more about learning than about accountability.

Such learning requires thorough employee feedback. When managers deftly blend criticism and encouragement, feedback is motivating. Motivation boosts effort, and effort improves outcomes. These outcomes, in Stat programs, are in turn subject to evaluation as action, feedback and motivation are streamlined into a permanent cycle. This loop fosters efficiency, but it must be maintained until it becomes habit. "Ratcheting up performance is an addiction strategy," Behn tells us. For that addiction to catch on, it needs a committed champion who will work with and through the entirety of the organization.

The word "potential" in the book's title is an apt disclaimer for the promise of PerformanceStat. Behn details some of its successful iterations -- as CitiStat in Baltimore, BlightStat in New Orleans and MentalHealthStat in Los Angeles, for instance -- but is also careful to explore instances in which it did not add value. In fact, Behn does not describe the idea of PerformanceStat so much as he performs it. With a compelling mix of context-heavy story-telling, integrated analysis of civic data, and insistent discrimination of cause and effect, his delivery follows the leadership strategy he outlines from the outset.

The result is a guide for understanding PerformanceStat in totality, from its theoretical framework all the way down to the arrangement of the meeting room, so that practitioners can learn from examples and apply the strategy to governments large and small, agencies national and local, and everywhere in between.