The Argument From Risk

In an important sense, the arguments from Utilitarianism and stewardship critique the moral foundations of conventional cost-benefit analysis, not the use of high discount rates per se. In contrast with these moral critiques, the third line of reasoning explored in this chapter - the argument from risk -accepts the premise that climate change policies should be analyzed based on the private preferences that people reveal through their market behavior. But while conventional discounting procedures rest on the assertion that individuals exhibit a high degree of impatience - and hence an unwillingness to exchange short-term costs for long-term benefits - the argument from risk holds that the use of high discount rates is in fact inconsistent with the empirical evidence.

To understand this point, it is useful to note that financial markets involve the purchase and sale of a large number of assets characterized by varying degrees of risk. Risky assets such as corporate stocks pay average long-run returns of roughly 6% per year in real (inflation-adjusted) terms (IPCC, 1996, chap. 5). As I noted in the introduction, authors such as Nordhaus (1994b) and Manne (1995) argue that public policies should be evaluated at a discount rate that reflects the typical returns investors demand of corporate stocks. In this view, the use of low discount rates might lead policymakers to approve low-return public projects that crowded out private investment that yielded higher returns to society.

Although this claim is intuitive, it runs up against an important body of theory and evidence from the finance literature. In particular, safe forms of investment such as U.S. Treasury Bills generated long-term yields of less than 1% per year between 1926 and 2000 (Ibbotson Associates, 2001). In financial economics, Treasury Bills are generally viewed as a risk-free asset since their returns are remarkably stable over time. In this respect, they resemble money market accounts and short-term certificates of deposit. By way of comparison, corporate bonds, which are characterized by an intermediate degree of risk, yield long-term returns of roughly 3% per year. In explaining observations of this nature, financial economists work with models in which the expected (or average) rate of return on a risky asset (rr) is determined by the return available on safe assets (rs) plus a risk premium (RPr) according to the equation:

In simplified terms, this equation captures the idea that people will invest in risky forms of wealth only if they expect to receive a rate of return that is higher than that available on safe investment options. The risk premium measures the effective reward an investor demands in exchange for accepting uncertainty.

Various formulae exist for determining the risk premium that investors demand based on the uncertainties that surround the potential returns achieved by a given investment (Cochran, 2001). In general, these formulae are derived from theoretical models in which people allocate investments between available assets to optimally balance the goals of maximizing returns and minimizing risks. These methods both indicate, however, that the risk premium is positive for investments that increase the degree of uncertainty surrounding an investor's overall economic welfare. Corporate stocks fall into this category because investors' incomes rise and fall with the market. On the other hand, insurance policies yield payoffs that - although uncertain - serve to reduce the risks that surround an investor's overall financial position. Since insurance policies protect buyers from the risk of incurring large (sometimes catastrophic) losses, people purchase them despite the fact that these policies will (on average) return less cash than money deposited in the bank. In terms of Eq. (2), this implies that insurance policies have a negative risk premium.

As I noted in the preceding section, climate change policies are designed to reduce the environmental risks faced by future generations. This is illustrated in Fig. 5, which shows that climate stabilization can forestall the risk that climate change will impose irreversible, catastrophic costs with a significant (though poorly measured) probability. More formally, Tol (2003) presents a quantitative analysis in which climate change completely devastates the economies of Eastern Europe and the former Soviet Union with a probability of 0.1%. This result occurs because of shifts in precipitation patterns that deprive this region of needed water resources. Less ominous catastrophes occur in Tol's model with greater levels of probability. Tol's study is important because it represents a serious attempt to integrate the scientific, technological, and economic uncertainties that surround global warming using a fully specified mathematical model.

What are the implications of these points for the choice of discount rates in cost-benefit analysis? One approach to answering this question is provided by Sandmo (1972) and Starrett (1988), who explore theoretical models in which public policies should be evaluated using discount rates that reflect the risks those policies impose on future society. According to these authors:

1. If a policy would involve risks that are similar to those posed by private investments, then it would be appropriate to discount its future net benefits based on the returns paid by corporate stocks.

2. If a policy were risk-free, then its net benefits should be discounted at the risk-free rate of return.

3. For policies that provide insurance benefits - i.e., that reduce the overall uncertainties faced by future society - the use of discount rates below the risk-free rate would be theoretically appropriate.

For the reasons described above, it is reasonable to presume that climate stabilization measures fall into this last category.

Alternatively, the general framework employed by Sandmo and Starrett implies that cost-benefit analysts may address questions of risk by (a) adjusting a standard measure of net benefits to account for the value of risk reduction; and (b) discounting adjusted net benefits at the risk-free rate of return (see Howarth, 2003). This approach is illustrated by Cline's (1992) analysis of the costs and benefits of climate change, which supports stabilizing greenhouse gas emissions at roughly half the year 2000 level under business-as-usual - a target that is even stricter than the climate stabilization scenario described above. Although Cline defends his use of a 1.3% annual discount rate based on Utilitarian moral reasoning, this discount is in line with the rates of return paid by safe investments.

It is important to note that this ''argument from risk'' does not assert that policy-makers should adopt ad hoc or ethically based discount rates that are below the returns paid by private-sector investments. Instead, the point is that climate change policies have risk characteristics that are quite unlike those pertaining to corporate stocks. According to economic theory, the choice of discount rates should reflect the risk characteristics of the policy or project under examination. The use of low discount rates is appropriate when evaluating policies that reduce risk.

0 0

Post a comment