Efficient self-coordination in policy networks: A simulation study

The paper begins with a reexamination of claims regarding the welfare-theoretical efficiency of various modes of non-hierarchical policy coordination which Charles Lindblom (1965) had subsumed under the label of Partisan Mutual Adjustment. It is argued that these claims are implausible if Lindblom's mechanisms of horizontal self-coordination are examined one at a time. They either will not assure significant welfare gains in the general case, or the attempt to raise the level of general welfare through self-coordination will encounter rapidly escalating transaction costs. As Lindblom had pointed out, however, several coordination methods will often be combined in real-world policy processes. The intuition that this might significantly increase the welfare efficiency of self-coordination is explored in a computer simulation study based on the game-theoretical reformulation of five simple coordination mechanisms. We can show that, in a given population of interdependent actors, Positive Coordination within relatively small coalitions who are required to obtain the agreement of outside actors through Negative Coordination and Bargaining, are able to achieve relatively high welfare gains while economizing on transaction costs. This pattern is by no means unusual in real-world policy processes.