Fiscal Impact

To the extent that transport projects produce public goods, the beneficiaries either cannot or should not be charged directly by the government for the benefits received. The costs of transport projects must, therefore, be recovered through taxes. As discussed in older article, for every tax dollar collected, society incurs an extra cost that is likely to be in the neighborhood of 30 percent. This marginal cost of public funds reduces the net benefits of transport projects and needs to be added to the cost of projects. If any of these costs can be recovered directly from the beneficiaries through user charges, it would be preferable to do so rather than relying on the tax system. Note that if a road is partially financed from tolls, the 30 percent premium on public funds applies only to that portion of the project that is not financed from tolls.

Risk and Sensitivity Analysis

Project outcomes necessarily depend on uncertain future events. The basic elements in the cost and benefit streams of projects—such as input and output prices and quantities—seldom represent certain, or almost certain, events in the sense that they can be reasonably represented by single values. Uncertainty and risk are present whenever a project has more than one possible outcome. The measurement of economic costs and benefits, therefore, inevitably involves explicit or implicit probability judgments.

Take the example of someone who wants to buy coffee today, hold it for a year, and then sell it. Because commodity prices are extremely variable, the outcome of this simple project is uncertain and the person undertaking the project is taking a risk. Such a project would have made money in 12 out of 23 years between 1970 and 1993, lost money in 10 out of 23 years, and broken even in 1 out of 23 years. If we use the past as a guide to the future, we would recognize the possibility of at least three outcomes, each with a different probability of occurring. If the project entailed renovating coffee plantations, uncertainty about yields and costs would be added to uncertainty about coffee prices. As a result, the number of possible outcomes would increase dramatically.

The preferred approach to sensitivity analysis uses switching values. The switching value of a variable is that value at which the project’s NPV becomes zero or the IRR equals the discount rate. We usually present switching values in terms of the percentage change in the value of variable needed to turn the project’s NPV equal to zero. We may use switching values to identify which variables have the greatest effect on project outcomes. We may also present the switching values of the relatively more important variables in order of declining sensitivity

In this example, the most critical variable is yield. A decrease of more than 25 percent in the posited expected yield will make the NPV negative if other values remain as expected. If experience suggests that yield can easily be that much less than expected, perhaps because of poor quality.