Investment Decisions with Sampling

Abstract

* Conventional capital budgeting techniques, such as those used by Bierman and Smidt [1] or Merrett and Sykes [3] lead to a rejection of investments with a negative net present value (leaving out portfolio effects). This paper argues that one should not always reject an investment with a negative net present value in situations of uncertainty and when there is a possibility of replicating the investment. Multi-plant firms have an opportunity to innovate sequentially that is frequently not available to firms with single plants (unless the single plant has multiple production lines). Consider the development of a new type of equipment in a multi-plant company. The analysis for a single unit of equipment 'indicates a negative net present value. But there is some probability that the equipment would be successful and would have a positive present value in any subsequent use. In other words, there is uncertainty about the outcome but there is some probability that it would be a desirable investment. In such a situation, the possibility that the firm may miss out on a technological break-through may be sufficient motivation for trying the equipment as a sample investment. If one unit of the equipment has a positive net pres-