The Roosevelt Project
Olivier Herbelot, March 1994
Standard discounted cash flow methods are not well suited to the valuation of investments whose characteristics can be modified by the decision-maker after the initial investment decision has been made (multistage decision investments). For some problems of this type the theory of financial options offers a better alternative.
The theory is applied here to an existing coal-fired power plant that is required to comply with the new SO2 emission limits introduced by the Clean Air Act Amendments of 1990. By assumption, the power plant operator can either purchase emission allowances from other utilities, or switch fuels to a lower-sulfur coal, or install an SO2 emission reduction system (scrubber). The two main sources of uncertainties (future price of SO2 allowances, and future difference between the price of high-sulfur and low-sulfur coals) are assumed to follow Wiener stochastic processes over time. A binomial model is developed to calculate the present value of the options to install the scrubber and/or switch coals. It is shown that the possibility of switching coal has little value to the utility in the case considered, but that the possibility of installing a scrubber reduces the net present cost of complying with the Clean Air Act SO2 requirements. A parametric study is performed to estimate the influence of various model variables on the option present values. Also, the effect of future scrubber technology improvements is investigated. Finally, the model is used to obtain an investment criterion that specifies, ex-ante, the future conditions under which the scrubber should be installed or the fuel switched.
The investment case considered shows how contingent claim analysis can be applied in practice to evaluate realistic flexible investments. The results underline the need to take investment flexibility into account and the practical advantages of option valuation. They show that investment criteria can be substantially modified by the value of flexibility. Also, the binomial model for two underlying variables developed here is found to be quite intuitive and easy to apply numerically. It can also be used to determine investment criteria.