8.0 Introduction

The Iroquois Indians are said to have lived under a traditional directive to consider the impact of their decisions on the next seven generations. What kind of world are we leaving to our children, our grandchildren, as well as our great-great-great-great-great grandchildren? In this chapter and the next two, we move beyond our efficiency versus safety debate over pollution-control standards and consider these long-run impacts. In the process, our focus shifts from allowable standards for pollution to questions of the maintenance of environmental quality and natural resources over the long term.

Many current decisions involving pollution and the environment have long-lasting impacts. A pollutant that has long-term consequences is called a stock pollutant. Each one of us carries residues of the pesticide DDT in the fat cells of our body, even though it was banned from use in this country in the early 1970s. Chlorofluorocarbons released into the atmosphere contribute to ozone depletion for decades. Certain types of high-level nuclear waste retain their toxicity for tens of thousands of years. And current ${\text{CO}}_2$ emissions from burning oil, natural gas, and coal will continue warming the planet for more than 100 years. In contrast, flow pollutants do their damage relatively quickly and are then either diluted to harmless levels or transformed into harmless substances. Examples include smog, noise, and heat pollution.

People also have long-lasting impacts through their use of natural resources. Natural resources are inputs to the economy—both renewable (water, wood, fish, and soil) and nonrenewable (minerals, oil). Exploiting natural resources at rates faster than their rate of generation will reduce resource stocks, making them less available for future generations. As long as the rate of extraction of renewable resources is balanced by regeneration, resource use will not reduce future availability. However, any use of nonrenewable resources reduces their stock. For example, extracting oil and burning it reduces the stock of oil that remains. We discuss natural resources in greater depth in Chapter 10. Economists refer to natural resource stocks, the state of the environment, and the ability of the environment to assimilate pollutants and provide valuable services such as filtration of nutrients to provide clean water as natural capital. Changes in natural capital are one important way in which current decisions affect the future.

But current decisions also affect future generations through other channels. Since the start of the Industrial Revolution in the late 1700s and early 1800s, scientific discoveries, new technologies, and new ways of doing things have led to rapid economic growth. Innovations have transformed sectors of the economy such as health care (vaccines, pharmaceutical drugs, and X-rays), transportation (automobiles, high-speed trains, and jet airplanes), food production (synthetic fertilizers, pesticides, and high-yield crop varieties), and communications (television, cell phones, and Internet). Building new industrial plants and machinery (manufactured capital) and increasing the level of education and experience of workers (human capital) have made the global economy more productive. Establishing trust among members of society or having functioning institutions (social capital) is also important for economic productivity. Increases in various forms of capital have led to large improvements in the standard of living for many (but not all) people over the past 200 years. But these improvements have also come at a cost, in the form of pollution and declines in natural resource stocks.

The main question of this chapter is what do all of these changes mean for the future? Due to the dramatic increase in the pace of environmental, social, and technological changes, it is difficult for us today to predict what the future will look like 100 or 200 years from now. Just imagine what someone living in the 1800s would think if they were to be suddenly transported to the modern day and see cities with 10 million people, computers, cell phones, and airplanes for the first time. Yet thinking about the next seven generations requires us to try to think through the long-term consequences of current actions.

We start here by defining sustainability generally, and then we’ll quickly see how achieving sustainability in practice requires determining the extent to which increases in manufactured, human, or social capital can substitute for lost natural capital. This in turn will help determine whether “growth” in the economy is actually leading to increases in human welfare—or, put another way, whether investments in other forms of capital at the expense of natural capital loss are really making us better off.

If, on the one hand, growth is making us better off, then we can explore how best to trade-off investments between preserving natural capital and expanding manufactured capital, through a modified form of benefit–cost analysis that brings future benefits and costs of today’s actions into the equation. If, on the other hand, natural capital is unique and irreplaceable with no substitutes among manufactured, human, or social capital, what policies can we put in place so that developmental decisions protect much of our remaining natural capital and ensure sustainability? This chapter and the next are the most difficult conceptual chapters in the book—they get us deep into a debate between neoclassical and ecological economists—so, fasten your seat belts.

8.1 Sustainability: Neoclassical and Ecological Approaches

Most people accept that we have a responsibility to manage the environment and the planet’s natural resources to provide future generations with a high quality of life: the idea of sustainability. The most commonly used definition of sustainability comes from a report by the World Commission on Environment and Development (1987), or Brundtland Commission, named after former Norwegian Prime Minister Gro Harlem Brundtland, the Commission Chair. The Bruntland Commission defined sustainable development as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” Economists find it difficult to precisely define “needs” and generally think instead about well-being. So, in economics, sustainability is typically defined as follows.

A sustainable economy is one in which the well-being of a typical (median) member of all future generations will be higher, or at least the same, as the well-being of the typical member of the current generation. Put another way, sustainability means non-declining utility for the typical member of all future generations.¹

While the definition is precise, putting it into operation is controversial. Every time we drive our car, and burn up a gallon of petroleum, aren’t we in fact impoverishing future generations by depriving them of access to this amazing gift from nature, this unique, nonrenewable, high-energy fuel? How can oil consumption possibly be sustainable? Well, it might be sustainable if, at the same time, technology is advancing to provide future generations with alternative energy sources (solar and wind) that can provide a comparable energy supply at roughly the same price. The idea here is that we do not need to sustain any particular resource as long as we can find alternatives that provide equal or better outcomes. Up to a point, manufactured, human, or social capital can substitute for natural capital in improving human welfare. Whether we have reached that point is the key question.

If substitution is still generally possible, then achieving sustainability does not require that any particular form of capital be conserved. This definition of sustainability, known as weak sustainability, requires only that the overall capital stock—manufactured, human, social, and natural—needed to support a high quality of life is maintained. For example, weak sustainability would be achieved in food production if increases in manufactured capital, in the form of manufactured fertilizers and pesticides, can replace the loss of natural soil nutrients and pest control. In other words, weak sustainability can be achieved even with the loss of natural capital as long as other forms of capital are adequate substitutes.

Two inputs are substitutes in production if one input can replace the other and maintain production levels the same. (If you have taken intermediate microeconomics, let your mind drift back to a smooth, bow-shaped isoquant; now recall how two inputs such as capital and labor can be substituted for each other while maintaining the total output level the same.)

Weak sustainability also does not imply that the same mix of products needs to be produced. It may also be possible to substitute in consumption as well as in production, if increased consumption of one good replaces the loss of another and keeps people equally satisfied. (Again, if you have taken intermediate microeconomics, think about an indifference curve where good $X$ can be substituted for good $Y$ and achieve the same level of utility.) If people are happy to substitute video games and amusement parks for spending time in nature, the loss of natural areas does not necessarily reduce satisfaction from recreation and tourism. As a case in point, consider the number of tourist visits in Florida. In 2010, approximately 17 million people visited Disney World’s Magic Kingdom while less than 1 million visited Everglades National Park.

Knowing whether the economy is on a path to satisfy weak sustainability is not easy. As we have seen, material growth often comes at the expense of environmental quality. Judging whether the future is better with improvements in some areas and declines in others requires making value judgments about whether the gains really do outweigh the losses. And not all people have the same values. While the typical person might be willing to have more Disney Worlds and less Everglades, the avid bird-watcher or nature lover may be made worse off if this were to happen. Of course, in judging sustainability, it is really the value judgments of future generations that count. Would future generations consider themselves better off with the world we give them as compared to the world as it is now? If we give them more material goods but more pollution, or more video games and amusement parks but fewer chances to walk in nature, will they think that it is a good trade? Because they are not here yet to ask we cannot know for sure.

Such difficulties have led some economists to take a different tack to assessing sustainability. Rather than focus on an overall measure of well-being, strong sustainability requires maintaining the total stock of each form of capital (natural, manufactured, human, and social). In particular, strong sustainability focuses on measures of natural capital. Declines in natural capital violate strong sustainability regardless of whether there are increases in manufactured, human, or social capital. Strong sustainability tends to emphasize irreplaceable assets, those things for which there is no substitute. If a forest or wetland were lost, how could all of the complex set of ecological functions that occur in such systems be replaced? At least some components of natural capital may be essential in that they constitute the life-support system for humans and other species. People need oxygen to breathe but we can’t manufacture oxygen in the volumes required; we depend on trees and other plants to produce oxygen as a by-product of photosynthesis. If all plants went extinct people would follow suit.

Irreplaceable assets may also refer to uniqueness. In the 1960s, the U.S. Bureau of Reclamation contemplated constructing a dam on the Colorado River that would have flooded part of the Grand Canyon. The dam was never built, in large part because critics pointed out that the Grand Canyon is a unique and incomparably beautiful place for which the benefits of a dam would not make up for the loss.

Discussions about weak and strong sustainability have divided economists into two broad groups: neoclassical and ecological. (For the purposes of argument, we draw very sharp distinctions between these two groups, but in reality, many economists have one foot in both camps.) Neoclassical economists tend to view natural and other forms of capital as substitutes in production. They also tend to be technological optimists, believing that, as resources become scarce, prices will rise, and human innovation will find high-quality substitutes, lowering prices once again. Neoclassical economists also tend to view nature as highly resilient; pressure on ecosystems will lead to steady, predictable degradation, but no surprises. This view of small (marginal) changes and smooth substitution between inputs is at the heart of the neoclassical tradition in economics.

From a broader perspective, neoclassical economists generally believe that the global spread of market-based economies provides a powerful foundation for achieving a sustainable future. While still seeing a need for government regulations to control pollution and resource depletion, neoclassical economists generally believe that as markets spread, living standards around the globe will continue to increase. This, in turn, will lead to declining population growth rates and demands for increased environmental protection that will keep environmental degradation within an acceptable range. This is not to say that neoclassical economists believe there are no trade-offs, only that sustainability is most likely to be achieved through a well-functioning and properly regulated market system.

By contrast, ecological economists argue that other forms of capital cannot be used to substitute for natural capital; natural capital is essential and irreplaceable (L-shaped isoquants for microeconomics fans). Ecological economists emphasize the fundamental importance of nature as a critical and irreplaceable support system on which the economy is built: the economy is a wholly owned subsidiary of nature. Ecological economists view the global ecosystem in which our economy is embedded as fragile, in the sense that accumulating stresses may lead to catastrophic changes in the system itself. For example, climate change from unchecked global warming or increased ultraviolet exposure from unchecked ozone depletion could radically alter the natural environment.

Ecological economists also worry about abrupt changes in nature that might lead to catastrophic problems. For example, climate change may lead to disintegration of ice sheets in Greenland and the Antarctic, leading to rapid rise in sea levels or abrupt changes in ocean currents and weather patterns. Of course, ecosystems are subject to catastrophe all the time—for example, when lightning causes a fire in a forest. Catastrophe, when severe enough, in fact forms grist for the mill of evolution by favoring individuals within a species better adapted to new conditions. So, nature itself is not threatened. “Saving the planet” is not the issue. But preserving the conditions that support human life and well-being is the issue.

Local and global ecosystem services—water purification, nutrient cycling, regional climate regulation, waste processing, soil stabilization, pest and disease control, and crop pollination—provide a critical foundation for our economic well-being. Ecological economists argue that our economy depends in countless, often subtle ways on the complex web of services provided by our natural environment as it exists today. As a result, major environmental changes brought on by the stress of doubling and then redoubling population and consumption levels could lead to very large and sudden declines in human welfare.

Ecological economists also tend to be technological pessimists. They view recent rapid increases in living standards as a temporary phenomenon made possible only by the unsustainable use of natural capital. This ongoing degradation of natural capital will likely lead to the decline of human well-being. New innovations may not save us as they often have unintended consequences that result in increased pollution and resource depletion, thereby intensifying the sustainability problems.

In contrast to neoclassical economists, ecological economists consider the current patterns of globalization and economic growth to be unsustainable and that further ecological pressure in the form of population and consumption growth intensifies the problems and is likely to lead to disaster. While not hostile to the spread of markets or to incentive-based approaches to resolving environmental problems, ecological economists see the need for an expanded role of government in aggressively protecting our dwindling stock of natural capital.

We will return to the debate between neoclassical and ecological economists on their views in Chapter 9, where we discuss how to measure sustainability and what the empirical record indicates. The remainder of this chapter discusses the fundamentals of sustainability—the allocation of resources and changes in capital stocks through time. We begin by considering the treatment of future versus present benefits and costs and the role and meaning of discounting.

8.2 Future Benefits, Costs, and Discounting

In this section, we discuss how to value the benefits reaped by future generations that result from our actions today. Consider the following:

This puzzle illustrates a crucial point: investment is productive. As a result, at least from an individual’s point of view, $100 today is worth more than $100 next year or $100, 50 years from now. From a social point of view as well, resources today are worth more than the same resources in the future because resources today can be invested to yield more resources in the future.

How does this observation relate to the environment? Consider a simplified example: a decision today to clean up a hazardous waste site costing taxpayers $1.3 million. If we fail to clean up, then in 40 years, one individual (call her Joanne) who lives close to the site will contract a nonfatal cancer, thus imposing costs on Joanne—including medical bills, lost wages, and pain and suffering—equivalent to $1.5 million. Assume that this is the only damage that the waste will ever cause. Further assume that Joanne will feel fully compensated if paid $1.5 million. In other words, that she is indifferent between not having cancer and contracting cancer plus receiving compensation of $1.5 million. (Whether monetary compensation can really substitute for good health is an important question, one that is similar to the debate on strong versus weak sustainability and how easy it is to find adequate substitutes.) With these assumptions, the issue on the table is this: should we spend $1.3 million this year to prevent $1.5 million in damages in 40 years’ time?

An efficiency advocate would say no. To see why, consider the following response: We could more than compensate the future pollution victim for the damage caused by setting aside the money today that would be available for Joanne’s health care and compensatory damage payments in 40 years. However, we wouldn’t need to put aside $1.3 million. We could put a much smaller amount of money in the bank today, and by earning interest, it would grow over time to equal $1.5 million. In fact, $1.5 million in 40 years from now is equivalent to only $459,000 today at a 3 percent real (inflation-adjusted) rate of interest. Putting money aside into a bank account specifically to compensate future generations for pollution damages or depletion of natural capital is known as posting an environmental bond.

The Alaska Permanent Fund, which sets aside money from current oil revenues for the benefit of current and future generations of Alaska residents, provides one example of a real-world environmental bond. A second example is a bottle deposit. Some states require a refundable deposit of 5 to 10 cents per can or bottle of soda or juice purchased. If you choose to throw your container out the window of your car, the deposit money is then available for cleanup. (The deposit also provides an incentive for third parties to clean up the mess.) On a larger scale, mining companies and generators of hazardous waste sometimes must post bonds sufficient to allow regulatory agencies to assure reclamation if the property is abandoned.

The potential for posting an environmental bond offers us a rationale for spending less on prevention today than the full value of the damage we inflict on future generations. When future benefits are not weighted as heavily as current benefits, we say that the future benefits have been discounted. The basic rationale behind discounting is that because investment is productive, resources on hand today are more valuable, both to individuals and to society as a whole, than are resources available at a later date.

The figure of $459,000 mentioned earlier is known as the present discounted value (PDV) of $1.5 million received in 40 years. Formally, the PDV of $$X$ received in $T$ years is the amount of money one would need to invest today to receive $$X$ in $T$ years, at a specified rate of interest, or discount rate, $r$. We use the term interest rate to refer to the rate at which money increases going into the future and discount rate as the rate at which future values are decreased to find an equivalent present value. But, they are identical concepts. There is a simple formula for calculating the value of a future benefit in terms of today’s dollars. If the discount rate is $r$, then the present discounted value of $$X$ received in $T$ years is

In the case just described,

Let us consider a real-world private investment decision to see how discounting works.

8.3 An Example of Discounting: Light Bulbs

A recent technological innovation has the potential to dramatically reduce electricity use throughout the world. It may lead to lessened reliance on fossil and nuclear fuels and, thus, substantial improvements in environmental quality. The product is the humble light bulb. New light-emitting diode (LED) light bulbs can replace standard incandescent bulbs, use around one-tenth as much electricity, and last around 40 times as long. So what’s the catch? The up-front investment is substantially higher.

Suppose that you are the physical plant manager at a new hotel with 1,000 light fixtures and have to decide whether investing in these new bulbs is a good idea. How do you proceed? The first step is to marshal the information that you have on the costs and benefits of the two options over their lifetimes. This is carried out in Table 8.1.

The numbers in Table 8.1 consider a simplified problem with a 4-year time horizon. Suppose that the initial investment to buy 1,000 LED bulbs is $40,000 and to buy 1,000 incandescent bulbs is $1,000. Now because LED bulbs last 40 times as long, assume that the LED bulbs last for 4 years so there is a one-time expense of $40,000 in year 1. Assume that incandescent bulbs have to be replaced 10 times each year. That means that every year there is an investment cost of $10,000 for incandescent bulbs. Suppose that the electricity bill for LED bulbs is $500 per year and that for incandescent bulbs is $3,000. We also assume, as we will throughout the book, zero inflation (or, alternatively, that the figures are inflation-adjusted to reflect real purchasing power). On the face of it, buying LED bulbs appears to be a good idea as it leads to savings of $10,000 over the 4 years.

Unfortunately, choosing between the options is not that simple. The reason: the LED bulbs require an extra initial investment in the first year. If instead of buying LED bulbs you bought the incandescent bulbs, you could take the money saved in year 1 and put it in the bank and earn interest. At reasonable interest rates you might still do better by investing in LED bulbs. But at a high enough interest rate you will actually do better by investing in incandescent bulbs.

Things have suddenly become more complex. Fortunately, there is a simple way out: calculate the PDV of the net savings from investing in LED bulbs. The PDV will tell you how much, in today’s dollars, investment in LED bulbs saves over incandescent bulbs. To calculate the PDV of the investment, simply apply the formula for each year and add up the values:

Table 8.2 provides the relevant calculations. How do we interpret the numbers in the table? Consider $9,391.43, the number in the fourth-year column and the 0.10 discount rate row. This is the PDV of $12,500 received in 4 years when the discount rate is 10 percent. In other words, one would need to bank $9,391.43 today at a 10 percent interest rate to have $12,500 on hand in 4 years. If the interest rate were 0.00 (the first row), then one would have to have the full $12,500 on hand today to have $12,500 4 years from now.

The last column of the table illustrates that if the interest rate is zero, the LED bulbs do save $10,000. However, as the interest rate climbs to 5 percent and 10 percent, the investment looks less attractive but overall is still a good deal. However, when the interest rate rises to 20 percent, the LED bulbs become the more expensive option. With this high interest rate, it is better to save the money upfront on the initial investment and earn interest while paying out larger sums in terms of electricity and investment costs in later years. The main lessons to be learned here are that the higher the discount rate, (1) the less important are the benefits earned down the road and (2) the more important are the initial expenses.

Private for-profit decision-makers, such as hotel managers, do not have to decide what discount rate to use when comparing future benefits with present costs. Businesses seeking to maximize profits should use the market rate of interest on investments of similar risk, that is, their opportunity cost of capital. A rate of 5 percent, 10 percent, or 20 percent will be determined in the financial marketplace. This begs the question of how markets determine interest rates, which is discussed in the next section. However, government policymakers do not need to rely on the market interest rate. Instead they must choose a discount rate for analyzing policy decisions such as deciding how much stock pollutants to allow. What is the right discount rate for public choices, such as the cleanup of hazardous waste sites or the level of legal carbon dioxide emissions? After we understand the factors that influence market interest rates, we can consider the right choice for social decision-making.

8.4 Savings, Investment, and Market Interest Rates

Private-sector discount rates—or equivalently, market rates of interest—are determined by the aggregate investment and savings decisions of individuals, businesses, and governments, much the same way that market prices are determined by the choices of producers and consumers. In fact, the discount rate really functions as a price; it determines the relative value of future benefits and costs compared with present benefits and costs, just as prices determine the relative values of various products.

A simple way to understand what determines an interest rate, say the interest rate paid on savings accounts in banks, is through supply and demand analysis in the market for “loanable funds.” Figure 8.1 illustrates this.

The supply of loanable funds is determined by bank deposits—money in individual and business savings accounts that, in turn, form the basis for loans by banks. The supply curve slopes up, because, in order to induce people to save more, and consume less, people must be offered a higher price or return for their savings. What are the motivations for saving? People save money to purchase goods and services in the future, primarily for retirement. It is a bit of a puzzle that people do not save more of their income—the U.S. savings rate is typically 5 percent or less. Part of the reason has to do with what economists call positive rate of time preference: the well-known desire to consume more today, regardless of the consequences for tomorrow. Positive time preference may have a biological basis. Our hunter-gatherer ancestors did not have secure means of saving, and a good meal today might be the best strategy to ensure genetic survival. Regardless of whether the basis is cultural or biological, a positive rate of time preference is a clear deterrent to savings. As a result, advanced capitalist countries all have forced savings programs such as social security to overcome a natural tendency for people to save too little for retirement. This positive rate of time preference also partially explains why higher interest rates are needed to induce people to save more.

On the other side of the market, the demand curve for loanable funds reflects the opportunities for productive investment projects in the economy—in manufactured, natural, human, or social capital. It slopes down, reflecting the idea that, at any given moment there are limited opportunities for the high rate of return investments that can pay a high interest rate, but more opportunities for low rate of return investments, supporting payback to lenders at a lower interest rate. (Formally, this reflects the declining marginal productivity of capital.) At equilibrium in the market—in this diagram, 2 percent—the marginal dollar invested by borrowers will generate real gains to borrowers at the same rate demanded by savers to supply the dollars for investment. (In markets with risky investments, the supply curve for savings will shift up to the left, raising the interest rate that savers will require to provide their dollars to investors or governments.)

The underlying point here is that market interest rates, or equivalently, discount rates, will be determined by two key factors. First is the rate of return on investment, which is the rate at which the economy is able to make real gains through productive investment in all forms of capital. Higher productivity will shift the demand curve for loans out, thus raising the interest rate. Second is the rate of time preference—the premium that savers must be offered to induce them to save and keep them from consuming part of their income today. A higher rate of time preference will shift the supply curve of savings up, also raising interest rates.

8.5 The Social Discount Rate and Dynamic Efficiency

Understanding how savings and investment decisions interact to generate market discount rates also provides the key to understanding discounting for social decisions. We can now tackle the question as to what discount rate a government should choose for analyzing decisions about how much to spend to clean up hazardous waste sites, how much to regulate emissions of greenhouse gases, or how fast to use up natural resource stocks.

In 1928, a young mathematician named Frank Ramsey wrote a paper that sought to answer the question of how much of a country’s income should be saved and invested to maximize the present value of benefits to the nation over time. John Maynard Keynes (of Keynesian macroeconomic fame) who was a contemporary of Ramsey, said that the paper was “one of the most remarkable contributions to mathematical economics ever made, both in respect of the intrinsic importance and difficulty of its subject…” (Keynes 1933). Although it was not written about environmental issues, Ramsey’s analysis is directly relevant to addressing the question regarding the determination of the discount rate to use in the analysis of environmental issues such as climate change, stock pollution, and natural resource use. What is perhaps all the more remarkable for someone who made central contributions to the analysis of issues involving the long-term consequences of policies, and who has had such a lasting influence on economics, is that he died at the young age of 26.

Ramsey derived the discount rate that should be used in order to maximize the present value of net benefits over time to society as a whole. Although the mathematics that Ramsey used in his analysis is a bit beyond what can be presented here (he used the “calculus of variations,” which could be thought of as calculus on steroids), the logic of the result mirrors what we learned from the supply and demand analysis of interest rate determination discussed in the previous section. The Ramsey equation breaks the discount rate $\left(r\right)$ into two distinct components.

The first component is the rate of time preference, which we will label as $\delta$ (delta). If $\delta =2\text{percent}$, then out of pure “short-sighted” desire to consume more today than in the future, people would discount future consumption at that rate. With a 2 percent rate of time preference, they would trade $100 5 years from now for somewhat less money today. To be precise, $\$100-{\left(1+0.02\right)}^5=\$90.57$.

The second component of the discount rate comprises two terms, $\eta$ (eta) and $g$. The term $g$ refers to the growth rate of the economy. If $g>0$, the economy is growing and people in the future will generally be richer and will be able to consume more than people today. The term $g$ also captures the idea discussed in the previous section. The higher the payoff of productive investment $g$, the higher is the opportunity cost of forgoing that investment, the more valuable dollars are today, and thus, the higher is the discount rate.

The term $\eta$ refers to the elasticity of the marginal utility of consumption. Economists often assume that there are “declining marginal benefits” of consumption: that is, a dollar to a wealthier individual is worth less than a dollar to a poorer individual. So, if people will be wealthier in the future and consume more, they will tend to have lower marginal benefits from additional consumption. In other words, a dollar in the future is discounted because it is worth a lot less in terms of the benefits it generates than a dollar today.²

To address environmental and resource issues, we must modify Ramsey’s model. Rather than have $g$ represent the rate of growth of the economy, we have to consider it as the rate at which productive investment is paying off in terms of real increases in human welfare. We will look more at this issue in the next chapter, where we will see that GDP growth is generally not a good measure of real improvements in well-being.

To get some sense of what the Ramsey formula implies for the discount rate, we can plug in reasonable values for the rate of time preference, the economic growth rate, and the elasticity of the marginal utility of consumption. The rate of time preference for individuals is usually taken to be positive (sooner is better than later in terms of enjoying benefits) but generally felt to be fairly low, so let’s say 1 percent. The long-term growth rate of the U.S. economy has been between 1 and 3 percent, so let’s say 2 percent. Estimates for the elasticity of the marginal utility of consumption generally center around 2. In this case, the social rate of discount would be

However, others have argued that the social discount rate should be far lower. First, some economists have argued that the rate of time preference for society as a whole should be 0. The famous economist A. C. Pigou (1920; of “Pigouvian tax” fame) thought that a positive rate of time preference was the result of a “faulty telescopic facility,” and Frank Ramsey (1928) thought that a positive rate of time preference, which means that future generations count less than the present generation, was “ethically indefensible.” On the growth side, some ecological economists have argued that long-term economic growth is doubtful, largely because of declining environmental quality and limited natural resources. In addition, it is not clear whether growth, as conventionally measured, is actually improving human well-being today, an issue we will explore in Chapters 9 and 11. If long-term economic growth stops, then $g=0$, and the second term in the Ramsey equation would also be 0. So, without a positive rate of time preference and no long-term growth in per capita human welfare, the social discount rate should be 0!

For now, assume that the neoclassical economists are right, that $g$ is positive. It is easy to get lost in the math of discounting, but pause for a minute to recognize one very important fact: if $g$ in the Ramsey equation is greater than zero, then our economy is, by definition, (weakly) sustainable! That is because $g$ is defined as the rate of growth of human well-being, and if it is positive going forward, this means that people’s welfare, on average, is improving. This reflects, again, the neoclassical optimism about the ability of manufactured capital to substitute for natural capital. In the next chapter, we will look at some data to assess this assumption.

That said, neoclassical economists admit that in individual cases, businesses can “overexploit” natural capital, leading to inefficiently high damages to future generations. That is, pumping stock pollutants into the air or water or depleting a natural resource can impose greater damages on current and future generations compared to the benefits gained by current and future generations from exploiting the resource. How can this be avoided? If the economy is indeed weakly sustainable, then the neoclassical best policy is to maximize the present value of net benefits. If consuming a resource stock at present generates more benefits than saving it for the future, then the neoclassical answer is that it is best to consume the resource now. So, whether we should protect natural capital, clean up a hazardous waste site, or take action on climate change depends on whether the benefits of doing so exceed the costs (the efficiency standard).

When costs and benefits occur at different times, discounting is used to put everything into a common framework of present value and achieve what is known as dynamic efficiency. Under the dynamic efficiency approach, an action is justified if the present value of the benefits from the action exceeds the present value of the costs. Of course, this is true only if all costs and benefits (market and nonmarket) to all people (current and future generations) are included. This approach, in theory, appropriately weighs the trade-offs between protecting natural capital and investing in manufactured, social, or human capital.

In the earlier example of the hazardous waste site with cleanup costs of $1.3 million now and benefits of $1.5 million in 40 years, the present value of the costs ($1.3 million) exceed the present value of the benefits ($459,000) when using a 3 percent discount rate. Therefore, it wouldn’t make sense to clean up the waste site. The $1.3 million could be better used for some other form of investment that has a higher rate of return or to generate current benefits. In fact, with suitably high discount rates, say, because the current generation has many opportunities for productive investment in human or manufactured capital, the dynamic efficiency rule might cause us to choose to make very little investment in hazardous waste cleanup, or not take action on climate change, or protection of natural capital. This is the argument often made in many fast-growing developing countries, where raising living standards often trumps pollution concerns.

This section has highlighted the importance of the long-term economic growth rate, and the rate of time preference, for determining the proper social discount rate—a key parameter in using benefit–cost analysis when the costs and benefits of the action under scrutiny extend out over time. Unfortunately, because there is no agreement among economists over the values of the rate of time preference, or the long-term economic growth rate, there is also no agreement over the proper social discount rate that should be used. The right discount rate to use—particularly with respect to whether $g$ is actually greater than zero—is a major divide between more “ecological” and more “neoclassical” economists. In many ways, the choice gets back to the degree of substitutability between human-made and natural capital and the prospects for welfare improvements from economic growth. We will return to the questions of accurately measuring long-run growth and the correlation between growth and happiness, in Chapters 9 and 11. For now, recognize that this lack of agreement is critical as it has major consequences for recommendations about environmental policy, as we see in the next section.

8.6 Discounting Climate Change

The choice of discount rate can have a major influence on benefit–cost conclusions for policy. This is especially true in cases such as climate change, where the really large benefits of today’s emission reduction policies will not begin to be seen for 50 or even 100 years. As illustrated in Table 8.3, high rates of discount dramatically reduce the present value of future benefits. Even huge benefits 100 years from now are worth next to nothing with high discount rates. A future benefit of $1 million falls to $72.57 at a 10 percent discount rate and falls to an equivalent of one penny at a 20 percent discount rate.

Indeed, it was the choice of discount rate that was largely responsible for the dramatic differences in policy recommended by economist Nicolas Stern on the one hand, and Willaim Nordhaus on the other, which we discussed in Chapter 1. Recall that Stern’s benefit–cost analysis recommended 80 percent cuts in global warming pollution in developed countries by 2050, while Nordhaus called for allowing global emissions to grow by 40 percent—a lot but still below the business-as-usual trajectory. Stern’s analysis assumed a “low” discount rate of 1.4 percent, while Nordhaus opted for a much higher discount rate of 5.5 percent. Stern used the Ramsey equation to justify his discount rate. He assumed that the rate of time preference was near zero (0.1 percent) and the growth rate of the economy was 1.3 percent while the value of the elasticity of marginal utility was 1. These assumptions imply that the discount rate should be

Nordhaus believed that the correct discount rate to use was the market rate of return on manufactured capital investments, which is much higher. Nordhaus used a market rate of return of 5.5 percent. Using the Ramsey equation, Nordhaus could achieve a 5.5 percent rate of return by assuming that the rate of time preference is 1.5 percent, the growth rate is 2 percent, and the value of the elasticity of marginal utility is 2: $r=\delta +\eta g=1.5\%+2*2\%=5.5\%.$

The different discount rates mean that the same future damages—say, $100 billion of damages in 100 years—show up very differently in the different economists’ analyses. At the 1.4 percent rate (Stern), the $100 billion is still significantly reduced to 24.9 billion, but discounted at 5.5 percent (Nordhaus), the damages fall to only $0.47 billion—a huge difference. Given this, it is no surprise that Stern recommends much more aggressive action than does Nordhaus.

So who is right in this debate? From our perspective, it depends on the social opportunity cost of capital, which in turn depends on the elements in the Ramsey equation. Should society have a positive or a zero rate of time preference? What should be assumed about the long-term growth rate in income or welfare (and at what rate do marginal benefits decline)? Overall, what is the real rate of return on alternative investments in education, health care, and the economy more generally into which society could put its resources, rather than into rewiring the global energy system to stabilize the climate? Is it 5 percent? Or 1 percent? Or zero? Again, more discussion on the empirical evidence will be in the next chapter.

8.7 Ecological Economics, Strong Sustainability, and the Precautionary Principle

We began this chapter by thinking about the long-term consequences of current actions on the environment and future generations. Just as there are two ways of thinking about how much pollution is too much, the efficiency standard and the safety standard, there are also two ways of thinking about long-term consequences: weak sustainability and strong sustainability.

We have seen that neoclassical economists typically assume weak sustainability, and recommend benefit–cost analysis, with discounted future costs and benefits, to avoid penalizing future generations in individual cases of resource depletion or stock pollution. By contrast, ecological economists believe that in many important instances, there are no adequate substitutes for natural capital, and strong sustainability must be pursued. But obviously, not all natural capital can be protected—some forests will be converted to agriculture, some oil will be burned, and some copper will be mined. How should society prioritize?

A famous baseball player and announcer, Yogi Berra, once said that “It’s tough to make predictions, especially about the future.” At the beginning of this chapter, we began with the Iroquois directive to think about the impacts on the next seven generations. We noted that with the dramatic pace of environmental, social, and technological changes, it is difficult to predict what the future will look like 100 or 200 years from now. Thinking about future impacts and sustainability brings us face to face with uncertainty.

The underlying question from the ecological perspective is this: to what extent, if any, can we afford to further run down the existing stock of natural capital? In many cases, it is not clear whether a natural resource, or species, or ecosystem is essential to maintain welfare or whether there are adequate substitutes. When there is uncertainty, ecological economists argue for the application of the precautionary principle: actions that potentially pose a threat should be avoided unless there is sufficient evidence to show that the action is in fact safe. Applying the precautionary principle to natural capital can be summed up as follows: when in doubt, conserve.

The precautionary principle is often invoked in debates on the introduction of new chemicals. Some new chemicals might have unintended negative environmental consequences. Under the precautionary principle, a new chemical would not be allowed to be introduced until it was shown to be safe and not pose a threat to the environment.

A second example: when European nations rely on the precautionary principle to restrict the introduction of genetically engineered crops, they are doing so to protect natural capital services embodied in traditional agriculture from, for example, the potential development of “superweeds.”

The first thing to note about the precautionary principle is that it is like a commandment—it says, “Thou shalt protect natural capital without good substitutes” regardless of the cost of doing so. In other words, under the presumption that future generations will benefit from protecting resources, the precautionary principle requires that the current generation make potentially large sacrifices on behalf of the generations to come. In the case of genetically engineered crops, for example, by restricting development in this area—and thus preserving human health from potential degradation by, for example, new allergies—we may face higher prices for food. As a safety standard, at some point, the costs of protecting natural capital might simply become so high that political support for the precautionary principle can disappear.

Determination of the availability of substitutes requires focusing on uniqueness, irreversibility, and uncertainty. Do the services provided by natural capital in question have substitutes? Consider the case of species. Each year, many species become extinct due to loss of habitat. Many people would feel a tremendous loss if magnificent animals such as elephants or tigers became extinct. But, what about a relatively obscure and insignificant plant species that exists in a tropical rainforest? Deforestation has caused the extinction of many plant species. The primary use value from such plant species lies in their potential value to yield chemical compounds of medicinal or industrial value. An average of one in four pharmacy purchases contains rainforest-derived compounds. Suppose that at least one of the thousands of species alive in the forest contains a valuable chemical effective in curing many types of cancer. In such a case, that species and the rainforest that harbors it are unique natural capital.

An issue closely related to uniqueness is the technological potential for substitution. Each species, for example, represents a unique DNA blueprint that has evolved over millions of years. While we may be able to imagine substitutes for rainforest products generated by advanced technology, are such options feasible within the relevant time frame? In cases of unique natural capital—where good substitutes do not now, or will not soon, exist for the services flowing from the natural capital being destroyed—the stock should be protected, according to the precautionary principle.

It is possible, of course, that species extinction can proceed for some time at the current rate with no loss of unique medicinal or genetic value for humans. The remaining (millions) of species may provide an adequate resource base for developing medicines and biotechnologies on a sustained-yield basis. But, they may not. This example highlights another issue that arises in attempting to apply our sustainability criterion: the uncertainty of benefits flowing from natural capital. The genetic resources of the rainforest may yield tremendous improvements in health care and agricultural productivity, or they may prove to be relatively fruitless. In the case of the rainforest, beyond the major loss of pharmaceutical material, the ultimate consequences of destroying so vast an ecosystem are essentially unknowable. Similarly, as we saw in the introductory chapter, the atmosphere may be able to absorb current emissions of greenhouse gases with manageable changes in global temperature, but alternatively, the real possibility of major catastrophe exists.

While there is substantial uncertainty about the potential yield of some natural capital, decisions to degrade capital stocks such as rainforests or the atmosphere are often irreversible. Once these resources are depleted, they can be restored only at a high cost, if at all. Uncertainty combined with irreversibility provide what is known as an option value to natural capital (see Chapter 5). That is, preservation of the stock is valuable merely to keep our options open. The greater the uncertainty and the more severe the irreversibility, the more caution should be exercised in the exploitation of natural capital.

8.8 Strong Sustainability in Practice: Endangered Species, EIS, and Reach

There is currently only one major piece of U.S. legislation that mandates strong sustainability: the Endangered Species Act. We will review the functioning of the act in Chapters 13 and 14, but for now, recognize that the act requires the preservation of species designated as endangered, regardless of the cost. The rationale is consistent with our aforementioned analysis: species extinction is irreversible, and in many cases, species carry unique cultural and scientific value.

Beyond endangered species, the National Environmental Policy Act (NEPA) requires a “light” version of strong sustainability for many developmental decisions in the United States. One section of the 1970 act required that government agencies prepare an environmental impact statement (EIS) for “legislation or other major federal actions significantly affecting the quality of the human environment.” The statement must include:

1. any adverse environmental effects that cannot be avoided should the proposal be implemented;
2. alternatives to the proposed action;
3. the relationship between local short-term uses of man’s environment and the maintenance and enhancement of long-term productivity; and
4. any irreversible and irretrievable commitments of resources that would be involved in the proposed action should it be implemented.³

The law also requires that public comments be solicited in the drafting of the impact statement. More than half of the states have subsequently adopted their own versions of the EIS to assist them in their decision-making.

The basic philosophy behind the EIS is that the government should identify potential adverse environmental impacts of its actions and recommend superior alternatives. Further, by opening up the process to public comment, much broader perspectives on the likely effects can be incorporated into the decision-making process. As a planning tool, the EIS has had its share of success stories. When it works well, it encourages agency officials to respond to the environmental concerns of the public—environmentalists, developers, workers, and citizens—and hammer out an ecologically sound compromise on development issues.

However, even on its own terms, the process has its faults. The rather nebulous charge to identify adverse negative effects has led to a situation in which “the usefulness of the NEPA process to decision makers often is weakened by a persistent tendency to overload NEPA documents with a voluminous amount of irrelevant or highly technical data. No one wants to read such documents: not public citizens, not members of public interest groups, not judges, and certainly not decision-makers.”⁴ More importantly, there is no process by which “superior alternatives” are judged, nor effective procedures to ensure post-project compliance with the recommended mitigation measures are assured. Finally, the EIS process imposes costs on both government and the private sector. From an industry perspective, the EIS provides an opportunity for environmentalists to drag out the permitting process, engaging in what is known as paralysis by analysis.

The EIS is a form of government regulation that embodies the precautionary principle. Prior to the EIS, if a developer could show that the exploitation of natural capital would likely lead to an improvement in welfare, the development would generally be allowed to proceed. Conservationists were forced to prove the reverse in order to slow down the rate of resource depletion. The EIS has shifted some of that burden of proof to developers.

The most recent effort to implement precautionary regulation has occurred through the European Union (EU) system of chemical regulation, REACH (registration, evaluation, and authorization of chemicals). Unlike the chemical policy in the United States, the European system requires chemical manufacturers to carry out health and environmental safety tests prior to introducing new chemicals. The approach is comparable to the system of testing required for new drugs in the United States. REACH allows EU governments to restrict chemicals that pose unacceptable risks to health or the environment, through risk management measures, partial bans, or complete bans. Again, following the precautionary principle logic, regulation has shifted the burden of proof of safety to the manufacturers. Finally, REACH’s cost burden was reasonable, allowing this precautionary approach to make it through the EU legislative process (Ackerman 2005).

Applying the precautionary principle means augmenting market decisions in cases where endangered resources are unique, risks are high, or impacts are irreversible. These can be rather tame governmental regulations, such as the EIS, or much more aggressive regulations such as REACH, which, from a conservative perspective, begin to resemble dreaded “ecosocialism.” The ecological economic view undoubtedly requires much stronger government intervention than what we commonly observe to promote sustainability.

8.9 Summary

Similarly to the Iroquois Indians, we can recognize our ability to dramatically alter the welfare of our descendants, both for good and for bad. In the material sphere, we clearly have a responsibility to endow future generations with the productive capital—natural, human, manufactured, and social—sufficient to generate a quality of life at least comparable to our own. However, the means of achieving this goal of sustainable development remain a subject of intense debate.

This chapter examined two ways to think about the goal of sustainability. Weak sustainability requires that we leave to future generations various forms of capital (human, manufactured, natural, and social) that allow them to be at least as well off as we are today. If human, manufactured, or social capital can substitute for natural capital, then weak sustainability does not require that natural capital be conserved. Neoclassical economists tend to think that there are substitutes for much of natural capital and that technological progress will uncover these substitutes as natural capital becomes scarce. In general, neoclassical economists believe that well-functioning and properly regulated market economies are the best way to ensure weak sustainability so that future generations have a bright future with equal or better lives compared to the present generation.

Neoclassical economists recognize that, in particular cases, future generations can be made worse off by decisions to “overexploit” resources or the environment. They thus advocate the use of benefit–cost analysis for actions that run down the stock of natural capital, in order to build up other kinds of capital: manufactured, human, or social. This kind of analysis, if it includes all future costs and benefits, and discounts them at the rate prescribed by the Ramsey formula, is designed to ensure that net benefits will be maximized over time—that is, to ensure dynamic efficiency.

If manufactured capital can indeed substitute for natural capital, then maximization of the overall size of the economic pie for future and current generations, if the pie is growing, can be done using the tools of benefit–cost analysis and discounting. Yet, even within the neoclassical framework, discounting for social decision-making remains controversial. Economists do not agree on the appropriate discount rate for society as a whole. And this is critical, especially for projects yielding benefits more than a couple of decades out. In those cases, small changes in the discount rate chosen to evaluate a proposal can dramatically alter the outcome of a benefit–cost test.

In contrast to the neoclassical view, ecological economists argue that, in many instances, natural capital does not find adequate substitutes in manufactured, human, or social capital. For unique resources or sinks, where decisions that degrade the natural capital stock are irreversible, ecological economists argue for the application of the precautionary principle. Similarly to a safety standard, the precautionary principle demands as much protection for resources and sinks as is politically feasible, rather than attempting to identify some efficient level of continued degradation. Ecological economists argue, for example, that there is no good substitute for global climate stability. Further degradation of the atmosphere’s ability to safely store carbon dioxide is likely to lead to large-scale and extended human suffering. At the same time, the costs of a phase-out of fossil fuels over the next few decades are manageable—the obstacles are not technical or economic, rather political. Therefore governments should act much more aggressively to address the problem if we are to avoid catastrophic climate change.

In arenas ranging from chemical pollution to affordable energy and water supplies to biodiversity and agricultural potential, ecological economists are not optimistic about the prospects for sustainability under the current trajectories of population and consumption growth. The only solution will be a shift in the precautionary policies by the government and the widespread adoption of sustainable business practices (discussed in Chapter 3) by businesses.

As we have seen, a neoclassical benefit–cost approach (Stern) can reach the same aggressive policy conclusion about action on climate change as does an ecological, precautionary approach. But neoclassical economists can also reach a “go sort of slow” conclusion (Nordhaus), again significantly depending on the choice of discount rate.

Peering 200 years into the future, the neoclassical vision is fairly optimistic on sustainability. In fact, by asserting that $g$ in the Ramsey model will continue to be 1, 2, or 3 percent per year going forward for decades to come, neoclassical economists traditionally assume that weak sustainability is assured. Why? Market systems, they say, given the proper government regulation of environmental sinks and sources, will induce the technological innovation necessary to overcome any widespread shortages of natural capital. However, failure over the last decade by many national governments to cut global warming pollution at all—even at the go-slow levels recommended by Nordhaus’ benefit–cost analysis—is tempering that optimism.

Ecological economists, meanwhile, are downright gloomy about sustainability for the global economy. They argue that in the absence of a major sea change in both government policy (in the direction of precautionary principles) and business practice (in the direction of sustainability), the prospects for the seventh generation could be quite challenging. In the next chapter, we will look more closely at attempts to measure sustainability at the level of individual resources and sinks and at the economy-wide level. We will also examine the challenges faced in implementing the precautionary principle. This exploration will shed additional light on the ecological–neoclassical debate.

1. 8.0 We reduce the stock of natural capital available for future generations in two ways: first, via the emission of stock pollutants, which exhaust environmental sinks; and second, through the exploitation of natural resources, both renewable and nonrenewable. (Flow pollutants have no long-term impact.) Natural capital, along with manufactured, human, and social capital, comprises the basis for producing goods and services needed for human well-being. This chapter shows that a sustainable economy depends on the degree to which increases in manufactured capital, in particular, can substitute for reductions in the stock of natural capital.
2. 8.1 This section defines sustainability as “nondeclining utility” for the typical member of all future generations. Achieving weak sustainability does not require that any particular form of capital be preserved, reflecting the core belief of neoclassical economists that, practically speaking, manufactured capital can generally substitute for natural capital to produce the goods and services people need: we are not “running out of resources (or sinks).” Strong sustainability, by contrast, requires that many forms of natural capital (as well as human, social, and manufactured) must be preserved, if not enhanced, in order to protect the well-being of future generations. Advocated by ecological economists, strong sustainability as a goal is based on the assumption that manufactured capital is not generally a good substitute for natural capital.
3. 8.2 To compare costs and benefits over time, economists rely on a process called discounting. Discounting reflects the time value of money: because investment is productive, society (and individuals) is better off having dollars today, as they can be productively invested to increase the size of the economic pie for tomorrow. The present discounted value of $$X$ received in $T$ years at a discount rate of $r$ is the amount of money you would have to have today to put in the bank at that interest rate, to have $$X$ on hand in $T$ years.
4. 8.3 The use of present-value calculations for private benefit–cost analysis is illustrated in the light-bulb example. A higher discount rate—in this case, the market interest rate—means a lower present discounted value. High discount rates, thus, mean that current costs and benefits are weighed much more heavily than those occurring in the future. Private decision-makers use market rates of interest or profit for making decisions. But when social decisions are being made, a different discount rate is often chosen.
5. 8.4 The section illustrates how market interest rates are set in markets for loanable funds. One factor that affects the supply of loanable funds (deposits in savings accounts) is positive time preference. A higher time preference will lower the savings and thus raise market interest rates. On the demand side, the availability of higher rate of return investments will shift the demand for loans out by business and government investors, also raising market interest rates. Higher rates of return on investments occur when there are higher rates of growth for the economy. Thus, both time preference and higher economic growth rates will tend to raise interest rates—which, for businesses and consumers, are used as discount rates to evaluate alternative investments.
6. 8.5 The Ramsey equation is the basis for determining the social discount rate—the rate used by government when setting policies about pollution control and resource degradation. We focus on two key determinants: again, the rate of time preference $\left(\delta \right)$ and also the rate of growth of the human well-being ($g$, note: different than the rate of growth of the economy). As there is no widespread agreement on the magnitude of either of these two factors, the choice of discount rate is often controversial. Because neoclassical economists assert that $g$ is positive, they, in fact, assume that the economy is (weakly) sustainable. Under these circumstances, it is possible to under- or overinvest in protecting natural capital, so benefit–cost analysis (with discounting) is needed to determine the right trade-off between investment in the protection of natural capital versus creation of other types of capital. If used properly, benefit–cost analysis will ensure dynamic efficiency: maximizing the net benefits to future and current generations in the decisions to use up, or protect, natural capital.
7. 8.6 Nordhaus and Stern both use benefit–cost analysis in two different efforts seeking to identify the dynamically efficient level of global warming pollution reduction, reaching very different conclusions. Again, this is because of different assumptions about the value for the parameters of the Ramsey equation, leading to very different discount rates, leading to very different assessments of the benefits of pollution control 50 to 100 years out.
8. 8.7 Under weak sustainability, benefit–cost analysis with discounting can guide the decisions to protect, or not protect, natural capital. By contrast, achieving strong sustainability requires applying the Precautionary Principle. Similarly to a safety standard, the Precautionary Principle requires the maximum politically feasible protection for natural capital. In assessing which resources to protect under this standard, uncertainty, irreversibility, and uniqueness all factor in.
9. 8.8 Examples of applications of the Precautionary Principle include the U.S. Endangered Species Act and, to a lesser extent, the Environmental Impact Statement (EIS) process required under NEPA. The EIS requires negative environmental impacts and superior alternatives to be identified if federal agencies are involved in a project but has little ability to ensure precautionary outcomes. The European Union’s recent chemical regulation, REACH, is the latest example of a precautionary regulation.

REFERENCES

1. Ackerman, Frank. 2005. The unbearable lightness of regulatory costs. Fordham Urban Law Journal 33: 1071–96.
2. Bear, Dinah. 1987. Does NEPA make a difference? In Environmental Impact Assessment, Nicholas A Robinson (Ed.). Albany, NY: New York Bar Association.
3. Bromley, Daniel. 2007. Environmental regulations and the problem of sustainability: Moving beyond market failure. Ecological Economics 63: 676–83.
4. Glasson, John, Rika Therivel, and Andrew Chadwick. 2005. Introduction to Environmental Impact Assessment. London: Taylor & Francis.
5. Howarth, Richard. 2007. Towards an operational sustainability criterion. Ecological Economics 63: 656–63.
6. Keynes, John Maynard. 1933. Frank Plumpton Ramsey. In Essays in Biography. London: Macmillan and Co.
7. Pezzey, John. 1992. Sustainability: An interdisciplinary guide. Environmental Values 1(4): 321–62.
8. Pigou, Arthur. 1920. The Economics of Welfare. London: Macmillan and Co.
9. Ramsey, Frank. 1928. A mathematical theory of savings. The Economic Journal 38: 543–59.
10. The World Commission on Environment and Development. 1987. Our Common Future. New York: Oxford University Press.