appropriate for categories of risk that are (a) controllable by the contractor, (b) controllable by the client, and (c) not controllable by either.
Samuelson (1986) shows that under 'general conditions', some level of risk sharing, with 0 < b < 1, should be preferred by the client to either CPFF or fixed price contracts. The general nature of Samuelson's analysis does not lead to any specific optimal values for b, but the optimum value of b increases the more risk averse the client and the more costs are controllable by the contractor. A further complication is that contractor risk aversion affects the actual level of contractor effort on controlling costs once the sharing rate b has been negotiated. The greater the perceived risk of loss in a contracting situation the more vigorously contractors strive to reduce costs for the sake of avoiding loss as well as for the sake of gaining increments of profit (Scherer, 1964). A similar difficulty exists in considering the efficiency of sharing client-controllable risk. However, in specific situations, the client may be able to identify and cost various options available and evaluate these under different risk sharing arrangements.
In the case of risk that is not controllable by either client or contractor, a plausible variation to the risk sharing arrangement in (16.1) and (16.2) is to set F = K + b2(V — E), where K is the fee required by the contractor if a cost reimbursement CPFF contract were to be agreed and V is set to the value R or T referred to in the previous section. Assuming variance (Var) is an appropriate measure of risk, this risk sharing arrangement reduces the contractor's risk from Var(E — C) to Var b(E — C) = b2 Var(E — C). Therefore the contractor's risk premium should be reduced from (V — E) to b2 (V — E) under this risk sharing arrangement. With agreement between client and contractor on the value of the expected cost E, this arrangement is risk efficient for both client and contractor for a wide range of circumstances (Chapman and Ward, 1994).
Of course, difficulties in specifying an optimum, risk efficient level for the sharing rate need not preclude the use of incentive contracts and pragmatic definition of sharing rates by the client. In practice, incentive contracts, or target cost contracts, often specify different sharing rates for costs above, below, and close to the target cost, as in the case of FPI and CPIF contracts noted earlier. This provides substantial flexibility to design incentive contracts that can reflect the particular project context and the relative willingness and ability of client and contractor to bear financial risk. Broome and Perry (2002) describe several examples of incentive contracts and the different rationales underlying each one, usefully illustrating many of the considerations involved in effective and efficient allocation of risk. The underlying principle, as Broome and Perry put it, is:
the alignment of the motivations of the parties so as to maximize the likelihood of project objectives being achieved, taking into account the constraints and risks that act on the project and the strengths and weaknesses of the parties to it.
Was this article helpful?