Formal Models of Legislative/Administrative Interaction: A Survey of the Subfield

Abstract

The purpose of this article is to review an important developing subfield of public administration: formal models of legislative/administrative interaction.(1) Formal modeling was originally an approach confined to voting behavior, public policy analysis, and occasionally legislative behavior. Scholarship on institutions generally took more sociologically oriented approaches (Weber, 1946; Simon, 1947, 1957; March & Simon, 1958; and Wilson, 1975). Change was precipitated by the introduction of economic tools in the form of social choice theory. The core idea of this approach is that complex organizations and their outputs can be analyzed as sums of discrete, rational subunits that are modeled and predicted with formalistic language and structures. Unfortunately, the technical complexity of some the contributions diminishes their ability to reach a wider audience. Behind the hurdle of economic formulas, mathematical abbreviations, and statistical constructs are a collection of powerful insights into the study of bureaucratic behavior. In this article, I briefly review and summarize a set of these formal models of legislative/administration interaction that have contributed substantially in this area. My intent is to discuss these models in nontechnical and nonmathematical language that highlights the concepts and insights rather than the detailed methodology. In order to provide a brief history of the field, I have omitted some of the interesting subtleties. Although this is unfortunate, it is necessary to provide a general overview while highlighting the major contributions of each work in such a way that readers who are not familiar with this literature can consider its dominant themes. I first review models of legislative/administrative interaction that focus on the budgetary relationship. These models are rooted in the Niskanen (1971, 1975) tradition. Then I discusses contributions that are not focused on the budgetary aspects of the relationship. This is an eclectic sampling that highlights innovative contributions that are formal models or models incorporating some formal component. Formal models are logical, rigorous constructs that attempt to explain some aspect of administrative behavior in formalistic language that can be tested against observed phenomenon. A formal model is evaluated by its ability to explain known behavior or events in precise, symbolic terms. Such models are most useful when their application produces conclusions that would not otherwise be intuitive or immediately clear. The symbology of this type of model allow a rigorous manipulation of variables to produce results that typically are not possible through standard analysis. This type of modeling shows how a set of simple rules applied to a hypothetical situation can illuminate less obvious characteristics. The mathematical and statistical complexity of these models varies considerably. The development of this approach is related to the increased use of economic and mathematical tools in political science, most notably in the analysis of voting behavior, that occurred in the 1950s and 1960s. The fields of political science, policy studies, and public administration acquired the techniques of game theory, decision theory, various stochastic processes, advanced statistical tests, linear programming, and probabilistic analysis. The adaptation of many of these tools originate with the work of political scientists such as Simon (1947), Downs (1957) Luce and Raiffa (1957) Buchanan and Tullock (1962) Tullock (1967) Riker and Ordeshook (1973) and Ferejohn and Fiorina (1974). Many ideas are imported from economists such as Bhagwati (1965, 1969) or political economists such as Olson (1965). Allison (1969, 1971) attempted to model bureaucratic decision making using three different models which drew upon the works of Downs (1967) and March and Simon (1958). Although Allison's three models are a simplification, they remain important and are often cited because they offer the idea that governmental decision making can be categorized and analyzed through relatively simple constructs. …