Incentives are widely used in construction contracts to incentivise the contractor to reduce construction cost. The optimal risk-sharing rate suggested by the Principal-Agent theory has been influential in shaping project investors' choice of incentive intensity. However, this theory is deficient due to failure to acknowledge the role of contract breakup cost. Based on the Principal-Agent theory, this research develops a model of construction procurement with the aim of demonstrating that failing to consider contract breakup hazards might lead project investors to offload too much risk onto the contractor. The intrinsic problem of under-pricing of risks in the Principal-Agent model makes it crucial for us to start seeking a new theoretical foundation, whereby risk-bearing capacity can be accommodated into contract design.