Throughout the last century, state and local governments were laboratories for budget reform. Many reforms were heralded but never implemented; others were implemented initially but did not survive changes in administration. Others at least partially succeeded, only to be followed by still more reforms. The result is a landscape of hybrid budgets in which fragments of the latest budget reform are overlaid and integrated with the most useful parts of previous budget reforms.
In 1994, the Florida legislature passed the Government Performance and Accountability Act (chapter 94-249, FS), requiring a statewide, phased-in conversion to a budgeting approach it called "performance-based program budgeting" (P[B.sup.2]). Because Florida is one of only two states that apply budgetary incentives and disincentives for performance results (Melkers and Willoughby 1998), studying the state's experience should provide valuable insights. Judged against the prescriptions of rational budgeting theory, performance-based program budgeting would be given high marks as a good budget reform. P[B.sup.2] requires government spending to be classified by program; programs must have missions and objectives; and input, output, and outcome measures must be linked to these missions and to appropriation levels. Programs commit to achieving a specified level of performance for each output and outcome in exchange for a specified level of funding. Programs that attain these performance levels may benefit from a range of incentives that the legislature may award. Penalties (called "disincentives" in the legislation) may be applied to programs that fall short of their performance levels.
The purpose for this article is twofold: (1) to introduce system dynamics as a technique for use in applied research; and (2) to model the Florida experience to gain a deeper understanding of what is required for successful budget reform--an understanding that can be applied when designing future budget reforms and when fine-tuning existing budgeting practices to make them more effective. To provide a framework for assessing the Florida experience, we present an implementation model that combines insights from the policy implementation, budgeting, and system-dynamics literature.
While many frameworks might be used to identify critical factors in implementing public policy (Matland 1995), Edwards' (1980) model proves most useful for helping us understand what is involved in implementing the Florida budget reform. His is a top-down model that allows us to look at the reform across departments and programs to gain a general impression of the reform as a whole.
Edwards's model classifies critical implementation factors into four categories: Communication, which should provide clear, accurate, and consistent implementation orders without being too precise; resources, including staff and facilities, as well as information about how to implement the reform and the authority to insure implementation as intended; the dispositions and attitudes of the implementers, which may be grounded in the organizational culture; and bureaucratic structure, whose two main features, for our review, are organizational fragmentation and standard operating procedures, or routines for doing budgeting and planning. Edwards suggests these four categories have a direct effect on implementation success and interact in ways that indirectly affect implementation success.
System Dynamics as a Research Framework
The systems approach to evaluating a phenomenon evolved from the disciplines of operational research and general systems theory. Operations research teams, to solve war-related problems during World War II, popularized the approach. Urban planners, industrial engineers and the technology disciplines have been using system dynamics as a modeling tool for decades (Forrester 1969, 1971; Stephanou 1982; Richardson 1990). …