In private sector investment appraisal, the textbook approach advocates discounted cash flow (DCF), which reduces a stream of benefits and costs to a single net present value (NPV), the difference between discounted benefits and costs. Being a difference, this measure is affected by project size. A project may have an NPV twice an alternative project with a higher rate of return just because of its size.
Sometimes discounted costs and benefits are expressed as a ratio such as the financial internal rate of return (FIRR). The private sector ignores externalities, or costs and benefits that do not affect the private investor. Externalities include environmental damage, pollution, and effects on others not reflected in the markets. A financial analysis ignores the fact that market prices are distorted by government intervention since market prices determine private profit.
A public project appraisal is less interested in FIRR and more in the net benefit of a project for the nation state as a whole. Social or economic benefit must account for externalities by incorporating social or shadow prices, i.e., prices that would occur in the market, were it free from intervention. Sometimes these are based on international or accounting prices in World Bank (WB) and Asian Development Bank (ADB) manuals, and sometimes on domestic prices in distinguished text books such as Mishan (1971).
When externalities and price distortion are allowed for in the cost and benefit stream, the social criteria are the same, i.e., NPV and internal rate of return (IRR), generally called economic internal rate of return (EIRR) to distinguish it from FIRR. Sometimes the benefit cost ratio (BCR) or benefits less costs divided by costs is used. Since the EIRR and BCR criteria are ratios, they are scale-neutral, whereas the NPV being a difference increases with the size of the project. NPV, EIRR (or sometimes IRR), and BCR are related in Table 1.
TABLE 1: Relationships Between NPV, BCR, and EIRR
These criteria can give conflicting results in ranking projects, and in deciding whether to accept or reject a project. Two major problems arise; choice of discount rate (or cost of capital), which is not needed to calculate IRR, but which is needed to interpret it; and the project time horizon. At normal discount rates of 10-12 percent, costs or benefits arising after 20 years are too small to have much influence on NPV or to affect ranking of projects. Projects with less than 10 percent EIRR, the usual threshold for acceptance, may be accepted because of special social contributions now fashionable and politically correct, such as employment of women or ethnic minorities. Different risk between projects does not seem to be a factor taken into account of in relation to rates of return. The marginal cost of capital is assumed to be the same for all projects-standard but dubious practice in business. A complicated decision matrix exists for selecting projects ranked differently by NPV or rate of return, depending on whether there is more than one project and if projects are mutually exclusive. In practice, projects are selected severally in relation to an acceptable in-house rate of return in both public and private sector.
A perennial problem is distribution of income over time and space. The discount rate takes account of the first and is discussed further below. Distribution of benefits between income groups can in theory be dealt with by giving different weights to benefits according to which income group is expected to enjoy them. But this is seldom attempted in practice. Discounting to allow for distribution over time and between generations excites much debate among philosophers, social scientists and economists who are not agreed even within disciplines. This is especially relevant in projects affecting the environment with effects beyond the life of a normal project assessment. In principle this may be dealt with by including terminal environment effects due to a project as a capital sum in the last year of project costs or benefits. This is unlikely to satisfy environmentalists, because of the diminishing effect of discounting at usual rates on any capital allowance in the last year of any project.
Economists usually depend on a utilitarian ethic that does not discriminate between generations. As in political voting, each individual is weighted equally. There remains the problem of the period over which to maximize utility plus the capital stock to be available at the end of the period not to mention how to decide distribution between generations. And should economists maximize total utility or utility per head, i.e., average utility? The philosophers do not help: Barry argues justice cannot be contracted between generations. Even Rawls seems to think there is no answer to the problem of sharing burdens and benefits between generations.
Choosing an appropriate discount rate and calculating social prices are major challenges to be added to the specific difficulties of including environment factors in project valuation. Economists demand discounting: Philosophers reject it (Beckerman, 1995). Rawls (1972), who rejects rationality of time preference, accepts the use of a discount rate to ration investment funds. Some economists following Pigou advocate a lower or even zero discount rate for public or environmental projects. As Beckerman says, if discounting is not in the interests of future generations, it would not be in ours either; and so it is not unfair to them. Further, economists discount not so much utility or welfare but rather the income flow available for consumption.
Discounting raises consumption of all generations by discriminating against poor investments. The implications of the alternative to discounting, a zero discount rate, would minimize consumption in favor of profitless investment. A lower discount rate for environmental projects differentiates wrongly between environmental beauty and other productive investments. Moreover it could be double-counting, as the environmental benefits should be included in the cost and benefit stream of any project ideally measured at prices appropriate for the time and value preferences in society (McFarquhar, 1996).
In practice if not in principle, the choice of discount rate is relatively easy except in the Courts in England where a recent case ended inconclusively in the House of Lords (Castle and McFarquhar, 1999). In the private sector, it is simply the future income stream net of tax discounted at the risk free rate of interest plus an appropriate risk premium. This discount rate may be assessed in the circumstances of each investment, or be prescribed by a private company.
This is robust because the discount rate can be used in principle, if seldom in practice, to reflect risk differences between projects and differences in perception of risk between analysts, borrowers, or lenders. Most textbooks advise using the weighted average cost of capital (WACC), now under increasing criticism, in the private sector. In reality a company probably applies a house rate that provides a threshold for acceptance of new investment. In the public sector, most institutions have a house rate that can be compared with a project EIRR. In US water projects, the acceptable rate it is determined by statute!
With increasing private finance initiative (PFI) in the public sector and policies for cost recovery in urban infrastructure projects, FIRR has become more important. Of course it was always necessary for assessing implications of cash flow and the subsidy component of public projects. In the private sector, FIRR is compared with the weighted average cost of capital in real terms. Whereas the opportunity cost of capital may be fixed for any country, the cost of capital may differ between projects depending on sources of capital (Hufschmit, et al., 1983). See Table 2.
TABLE 2: FIRR vs. WACC, Five Possibilities
This analysis reveals the subsidy requirements of projects which are deemed socially desirable but do not cover costs.
Cost effective analysis (CEA)
This is much simpler and therefore more practical than CBA. It seeks the least cost method for a given objective or the maximum benefit from given expenditure. To the extent that cost estimation uses opportunity costs, CEA suffers the same problems of valuation as CBA but benefits are intuitively assessed in physical terms.