This paper analyses the effects of relaxing one of the critical underlying assumptions of the textbook approach to investment under uncertainty for partial equilibrium models. Most textbook models assume that either the potential investor has access to a single project or she can consider competing (or complementary) projects independently (for example, Dixit and Pyndick (1994)). This paper investigates the effect of relaxing these assumptions. For a multi-product monopolist, the decision on producing a new good (such as the introduction of new financial instruments, new software, etc.) will typically affect the profits obtained from the existing good. Then, conditional on having first invested in the production of one good, the optimal decision on producing an additional good will crucially be affected by the nature of the goods that are produced. Furthermore, once we allow the cashflows from the products to be interrelated and consider the decision problem ex-ante, we find that, in general, not only is the standard rule for optimal investment not valid, but also the sequence of investment decisions might be different from the textbook rule. More particularly, in the case of two goods, we find that the optimal entry thresholds depend crucially on the degree of substitutability or complementarity between the goods. When the goods are substitutes, the investment strategy is usually sequential and the entry threshold for the second good cannot be derived using the textbook approach, that is, without taking into consideration the parameters linking the demand functions of the two goods. When the goods are complements, depending on the degree of complementarity, the investment strategy may be simultaneous or sequential. We also find that when goods are extreme substitutes in the sense that the demand for one completely wipes out the demand for the other, it might be optimal for the monopolist to first invest in one good, and then at some later date exit the production of this good and enter the production of the other good (even though there is no option value to exit). In section 2 we present our model. The textbook case of isolated investment projects is illustrated in section 3. In section 4 we consider the investment strategies open to the monopolist and the timing of investment. In section 5 we analyze the problem of choosing the optimal investment strategy. We summarize our results in the final section.
|Publisher||Universidad Torcuato Di Tella, Department of Economics Working Papers|
|Publication status||Published - 2003|