Book section

Decision Making in Public Administration

Handbook of Public Administration (1989) pp. 225-251 Published by Marcel Dekker

  • 12


    Mendeley users who have this article in their library.
  • N/A


    Citations of this article.
Sign in to save reference


Herbert Simon and Charles Lindblom are the most influential proponents. Decision-making is an interdisciplinary field: political science, public administration, economics, statistics, social psychology, etc. The pre-world war II period From its origin, the decision-making literature was preoccupied with the feasibility of rationality in the context of conflicting political values. Classical public administration introduced POSDCORB (planning, organizing, staffing, directing, coordinating, reporting, budgeting) as an acronym for the work of the chief executive (Gulick and Urwick 1937) and was of prescriptive nature. The first th have an interest in the actual process of decision making was Barnard (1938). He marked a transition between classical and behavioural apporaches to organizational decision making. The 1945-1970 period A. Simon and Bounded Rationality In "Administrative Behavior" (1947), Simon declared efficiency as the primary objective of the administration. His main question is how to achieve efficiency under the limits of human cognition and the consequent limits on individuals and organizations to behave rationally. This research on "bounded rationality" has become the so-called Carnegie School. While individuals fall short of achieving rationality, organizations can achieve it by means of specialization, selective attention and limitation of the values to be considered. Behavior falls chort of rationality in at least 3 ways: (1) knowledge of the consequences of choice is always incomplete (2) individuals' valuation of future consequences is imperfect; (3) only a few and not all possible alternatives are considered. Simon distinguished between value statement (i.e. statements of ends that express the popular will and belong to the political realm) and factual statements, which identify the empirical relationship between a desired end and the most efficient means to achieve it and belong to the realm of administration. However, every value is only a means to attain some higher value. "Rationality has to do with the construction of means-ends chains of this kind" (Simon 1947: 62). Later, March and Simon (1958) distinguish between optimal and satisfactory alternatives. In the first case (model of economic rationality), all possible alternatives are judged on a set of criteria and the optimal alternative is the one that scores best. Satisfactory alternatives (model of 'intended rationality') in contrast meet or exceed a set of minimal criteria, but it might not be the best alternative that exists. This leads to the well-known distinction between maximizing and 'satisficing' behavior. B. Lindblom and disjointed incrementalism Incrementalism by Lindblom (also called the method of successive limited comparisons, the branch method or disjointed incrementalism) also questions comprehensive rationality and stresses the limits of human cognition. The selection of value goals and means is seen to be interdependent, because goals are selected on the basis of what is attainable. Like Simon, ends become means to still more remote ends. But Lindblom says more clearly that it is not possible to distinguish ends and means. Valuas cannot be specified in the abstract, because inividuals disagree about them (there are even contradictions in one's own value set). It is only in the concrete decision situation that values are clarified and chosen and traded off against one another, because what is desired depends on the specific situation (e.g. existing level of unemployment and inflation). The number of alternatives that are compared is limited and some outcomes are neglected. The alternatives considered are the ones that are similar to the status quo and that are only incrementally different from each other. In a normative perspective, good policy decisions are those which achieve a broad political consensus ('partisan mutual adjustement', Lindblom 1965). Given humans limitations, Lindblom claimed the incremental process to be superior. These normative aspects of the theory, in particular, have been questioned. Policy decision, according to Lindblom, are (1) serial (most policy problems are treated several times and cannot be solved once and for all), (2) remedial (as they always address a problem), (3) marginal, as they differ little from the status quo and from one another. C. Wildavsky and budgetary decision making Incrementalism has dominated research on budgeting decisions with computer simulations. Disciples of Simon, too, have applied their 'process models' modeling sequences of decisions. The 1970s Brought a lot of criticisms of Lindblom's theory, not only of its normative but also of its descriptive claims. The impossibility of planned rational decisions and large or fundamental change was criticized by Ethioni and Dror, who both proposed their own descriptive and normative models (mixed-scanning and normative optimum model, respectively). The 1980s Research on incrementalism and bounded rationality continues, but reaches disparate results as to the quality of the decisions and predictions. One of the guiding questions in decision making research is the role knowledge, intelligence or information and analysis play in organizational decision making. In 1967, Wilensky identified hierarchy (conductive to concealment), specialization (conductive to rivalry and parochialism) and centralization (out of touch) to impair a smooth flow of information. With the eyplosion in the amount of information, the role of the expert has gained importance. In spite of many shortcomings concerning the use of information, the rational model remains the ideal in the literature.

Author-supplied keywords

  • bounded rationality
  • decision making
  • incrementalism
  • lindblom
  • public administration
  • simon

Get free article suggestions today

Mendeley saves you time finding and organizing research

Sign up here
Already have an account ?Sign in

Find this document

There are no full text links


  • John R Gist

Cite this document

Choose a citation style from the tabs below

Save time finding and organizing research with Mendeley

Sign up for free