Affiliations: [a] Department of Systems and Institutions for the Economy, University of L'Aquila, Italy. E-mail: biggiero@ec.univaq.it | [b] Department of Computer Science and Systems, University of Rome, Italy. E-mail: laise@dis.uniroma1.it
Abstract: Two main paradigms split the core of the debate on governance structures: the maxi-minimizing and the satisficing paradigms. Some differences between the two has been hidden by the (only apparently) similarity of the “maximum” and “maximal” concepts. If the maximizing behavior refers to the maximum of the utility function and not to the maximal of a binary preference relation, then satisficing does not correspond to maximizing behavior. Besides such differences, from an empirically grounded point of view the maxi-minimizing approach to choose and to evaluate governance structures has relevant limitations. The remarkable limitations of the maxi-minimizing approach become further evident and crucial when one acknowledges that the comparison between alternative governance structures is – like almost any other human choice – unavoidably multicriteria. The absence of tradeoffs among the different criteria prevents the generalized use of a single real-valued function (utility, profit, cost, etc.) to represent decision maker's goals. However, multicriteria problems can be solved when properly dealt with outranking methods. These are heuristic choice algorithms fully consistent with the satisficing paradigm and with the behavioral economic theory. In order to show its effectiveness, a tutorial example is discussed in detail comparing the three main forms of governance structures: market, hierarchy and network. They are scrutinized by four evaluation criteria: profitability, effectiveness, accountability and organizational capability. It will be demonstrated that the outranking algorithm can solve multicriteria decision making problems under conflicting criteria of evaluation. Outranking solutions are at the same time satisficing solutions, that take into account and operationalize the Simonian “levels of aspiration” through concrete parameters.
