THE CONDITIONS FOR THE IMPLEMENTATION OF THE STRATEGY

The analysed strategy cannot be implemented in any context: in fact, it needs two basic conditions to be effective. First, the tender procedure must refer to an open design. A distinction among the different forms of tender can be made on the basis the nature of design. In a first case, the client issues an invitation to tender on an unmodifiable design: in this case all the competing companies must submit just an economic offer, assuming that their tasks are fully and completely defined by the detailed design. In a second case, the competing firms submit not only an economic offer but they can also propose a variation to the basic design. This type of tender can get different forms: for instance, the client can also propose several alternatives design or can allow modification to the basic design just for well specified portion of it. The relevant point is that the administration does not consider its own design as the optimal one, but launches a learning process to find out if some different technical solutions may contribute to the efficiency of the process. In fact, it is possible to argue that no engineering company can get to an ex ante optimal design, but just to a satisficing one. For two reasons, at least. We previously argued that some know how is fully embedded in production and that the industrial activity and the production of knowledge cannot be split. If the knowledge that is in the production activity of the supplier of the construction process can become part of the design, of course the level of efficiency of the design rises to a superior degree. The second one - fully intertwined - concerns the well known issue of the bounded rationality of the economic agents. Since Simon’s seminal contribution (Simon 1959), the cognitive sciences have argued that the activity of problem resolution aims only satisfaction standards and not optimal ones; more specifically, it becomes evident that actors such as the clients technical service do not reach for an optimal ex ante design, but only to a satisficing one. Second, the tender procedure must be split for the different types of works. If we consider the possible procurement strategies a client can implement, a distinction can be made between procedures that consider the works as a whole, and those which see it as the sum of different interconnected performances. In the first case the client selects one economic agent (normally a general contractor) in charge to carry out the works for a certain price. In the second case, the client invites tenders for each type of work. The client with its own engineering services is charged with the integration of the different winning subcontractors that contribute to the quality and to the efficiency of the final design. A procedure based on the combination of an open design and of a set of multiple tenders represents the device which allows the administration to learn about all the opportunities the market can provide. Efficiency, then, is no more the traditional static goal, but is transformed into a dynamic concept, based on learning. Actually, this learning process is possible only under a third condition: the capacity of the client to impose common procedures and standards. A client able to detect an opportunity to cut costs by integrating into the basic design variations proposed by a competing firm, must, on one hand, be able to control the effectiveness of such a variation and its impact on overall costs; on the other, it needs to control the cooperation process among all the contractors. Of course, this coordination represents a cost that must be taken into account; more precisely, the client must evaluate the trade off between the advantages due to this learning process with the extra costs related to the coordination of all the actors involved in the project. Only under these conditions the strategy implemented can succeed. In the other cases, the firms limit themselves to the traditional role of subcontractor and do not act as co-designers. The client does not exploit the opportunities of such a procurement approach, not taking profit of the advantages of the learning process previously discussed.