Capital budgeting under uncertainty: An extended goal programming approach

Mathematical programming techniques have long been recognized as the most suitable approach to the complex problem of capital budgeting in corporations. This is because contrary to conventional single-project appraisal techniques such as the net present value method, financial planners do not normally engage in project-by-project analysis to determine the projects that would be included in capital budgets. In addition, the existing chance-constrained (both single and multiple objectives) models developed to deal with the problem of uncertainty in capital budgeting are grossly defective, since they, indeed, address the problem of risk, rather than uncertainty. This paper presents an extended goal programming methodology to address the problem of capital budgeting under uncertainty to overcome the defects of chance-constrained capital budgeting models. In particular, since financial planners frequently deal with the complex problem of capital budgeting by aggregating large numbers of small investment proposals into families of large projects, the paper presents necessary and sufficient conditions for the acceptance of a set of investment projects by a business enterprise. The results indicate that under uncertainty, firms faced with capital rationing are less economically efficient than others that are not so faced. Also, the results show that optimal allocation policy under uncertainty requires the actual discount rate to be greater than the market cost of capital, a finding which is consistent with corporate finance practice.