Contents - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

1. Eco-restructuring: The transition to an ecologically sustainable economy

Introduction: On sustainability
The need for holistic systems analysis
Environmental threats and (un)sustainability indicators
Sharpening the debate
Non-controversial issues: Population, resources, and technology
Controversial issues: Pollution, productivity, and biospheric stability
Finding the least-cost (least-pain) path
Concluding comments
Notes
References

Robert U. Ayres

The paper that follows was originally prepared for the United Nations University (UNU) as a background "white" paper in support of the development of UNU's long-term research programme (in fact, the UNU version was based on an earlier version that I prepared for a workshop held at IIASA in January 1991). As such, it was also the main background paper for the UNU Eco-Restructuring Conference held in Tokyo in July 1993, at which most of the papers in this book were first presented.

For various reasons, the preparation of this book from those conference papers has been an unusually long and difficult process. The review and revision process has been very slow. A few of the original papers had to be dropped because the authors were too busy to undertake necessary revisions and submit revised versions. Some that were topical at the time have become a little dated. To fill gaps (some of which were evident from the start) several additional papers have been solicited and included. But because these authors were not present at the conference they lacked the common background and required more than the usual amount of "editorial guidance." I think, however, that the final result is useful and interesting, if not the "last word" on the subject (which, in any case, will never be written).

The objective of the 1993 Tokyo conference was to explore the technical and economic feasibility of long-term sustainability. The conference did not totally neglect the political and institutional issues, but they were deliberately given secondary status. Social and cultural issues were set aside altogether, as being outside our collective ken. The overarching issue of population control was discussed only in my background paper, which follows, and only in terms of generalities. The bulk of the book deals with technological issues. Economics should have been given more attention than it was, but few economists are prepared to take on the practical aspects of long-range restructuring. This remains an open subject for future research.

Because of the background sketched above, I have since revised the paper only moderately to respond to reviewers' comments, while retaining much of its original logical structure. I recognize that the structure is not ideal for the intended audience of the book. (It is probably not ideal for any audience). But, as I say, the paper was essentially the "terms of reference" for all the other authors. To change it fundamentally would be a little unfair.

It must be acknowledged that, in many ways, the terms of the discussion have changed since the paper was written. In some ways, as on the problem of global climate change, there has been clear progress. One can only applaud this fact. On the other hand, thanks to the too vague definition offered in the Brundtland Commission Report Our Common Future, the term "sustainability" has been popularized and made virtually meaningless in recent years. It has been consistently misused, in particular, by the World Bank and other economic development agencies. These institutions are inclined to interpret "sustainable development" as "perpetual growth," which is an extreme perversion of the original sense of the phrase. But sustainability is also now an icon of generalized political correctness, where it is carelessly applied to a variety of attributes from culture to democracy.

Holding to the original sense of the word, this chapter seeks to sort out the questions about sustainability on which there is substantial scientific agreement from those unresolved questions that are still subject to considerable controversy. In this context, several testable theses are proposed. The first thesis is that there are limits to the capacity of the natural environment to accommodate anthropogenic disturbances. The earth is finite. Second, there are also limits to the substitutability of conventional market goods and services for environmental services. Third, there are limits to the extent to which technology can repair or replace environmental resources that are irreversibly damaged. For instance, only the most naive technological optimist can imagine undertaking to substitute positive engineering control systems, designed by humans, for natural means of climate control and stabilization.

With respect to some controversial questions we cannot expect firm answers. For instance, there is a possibility that nature may be exceedingly adaptable, resilient, and resistant to anthropogenic disturbance. Or nature may not be so resilient. It is conceivable, too, that human ingenuity could invent engineering alternatives to natural processes being threatened, or that technology could offer means of "adaptation" to ecological and climatic stress. However, the argument is that these possibilities are not probabilities. The limits of resilience are probably not very distant, and it is very difficult to justify a high degree of confidence that "business as usual" can continue without risk for even a few more decades.

In brief, the underlying problem is that many current demographic, economic, and industrial trends currently seem to point unmistakably in the wrong direction, i.e. away from sustainability. To achieve sustainability, and to minimize ecological risk, it will be necessary to reverse most of these trends. Indeed, some aggregated measures of material and energy use may have to be reduced by large factors (four to ten). Such a reversal will entail very fundamental changes in the economic system. The directions and magnitudes of these changes are assessed briefly, and various approaches to their implementation are analysed.

Introduction: On sustainability

Before the beginning of the industrial revolution, some two centuries ago, human activities - on the average - were not really incompatible with a healthy and sustainable biosphere. The vast majority of humans lived and worked on farms. Land was the primary source of wealth. Horses and other animals, supplemented by windmills, sails, and waterwheels, provided virtually all power for ploughing, milling, mining, and transport. The sun, either directly or through products of photosynthesis, provided virtually all energy except in a few coal mining regions. Metals were mined and smelted (primarily by means of charcoal), but their uses were almost exclusively metallic rather than chemical. Recycling was normal. Precisely because wealth was derived exclusively from the land, Thomas Malthus worried at the end of the eighteenth century about the propensity of human population to grow exponentially, in view of the limited amount of potentially arable land available for human cultivation.

As we approach the end of the twentieth century, humans are far more numerous and also wealthier (on average) than they were two centuries ago when Malthus wrote. In particular, those countries that industrialized first are now comparatively rich. In the rich countries most people live in cities. Land is no longer the primary source of wealth. Energy (except food) is largely derived from the combustion of fossil fuels (coal, oil, gas). Power for machines is obtained mainly from engines driven by heat from (internal or external) combustion of fossil fuels. (Nuclear and hydroelectric power, together, account for a relatively small percentage of the total.) However, one key attribute of this recent rise to wealth is critical for the future of humankind: what we have achieved so far has been done by exploiting an endowment of natural capital, especially topsoil and minerals. For some material resources technology can offer viable substitutes. For other resources in the natural endowment - notably the biosphere and its functions - no substitute is likely.

The report of the United Nations' World Commission on Environment and Development (WCED) - known as the Brundtland Commission - was published in 1987 under the title Our Common Future. This triggered the Global Environmental Summit at Rio de Janeiro in June 1992 and its major product, Agenda 21. Since that time it has been widely recognized that there is a very real conflict between meeting the needs and desires of the 5 billion people now alive and the possibility of satisfying the 10 billion or so people expected by the middle of the twenty-first century. It will be exceedingly difficult simultaneously to satisfy the objectives of environmental preservation, on the one hand, and accelerated economic development of the third world, based on current population trends and energy/material-intensive technologies, on the other. The implications of this conflict have been delineated eloquently in the Commission's report (Brundtland 1987; McNeill 1989). They need not be spelled out in detail again here.

Experts can and do disagree on the probabilities and timing of environmental threats relative to other problems facing the human race. Some ideologues have even argued that the threats are figments of the fevered imaginations of the "Greens." I think not. Arguments on these matters will probably continue for some time to come. But there is increasing evidence to suggest that major changes in the global economic and industrial system may be needed if the world is to achieve a sustainable state before the middle of the twenty-first century. Even though there is not yet a scientific consensus on the extent of the needed changes, it is clear that they will involve significant technological elements, as well as major investments.

The population problem comes to mind first, especially in the context of the 1994 Cairo Conference on Population and the Status of Women. It is unlikely that the other problems of the global environment can be solved if the world's population is not stabilized. Experts now generally agree that education and the status of women are central issues here. This implies that a world of relatively stable population must be one in which social patterns are significantly different from those now encountered in many parts of the world.

The kinds of techno-economic changes envisaged as necessary conditions for long-term sustainability also include a sharp reduction in the use of fossil fuels (especially coal) to minimize the danger of global "greenhouse warming." Alternatives to increasing use of fossil fuels include a return to nuclear power, large-scale use of photovoltaics, intensive biomass cultivation, large-scale hydroelectric projects (in some regions), and major changes in patterns of energy consumption and conservation. Again, there are disputes over which of these energy alternatives is the most (least) desirable, feasible, etc. However, the future of energy, from both the supply (technology) and the demand perspective, is a critical topic (to which several chapters of this book are devoted).

Again, the broad question addressed in this book is how to shift from a techno-economic "trajectory" based on exploiting natural resources - soil, water, biodiversity, climate - that, once lost, can never be replaced, to one that could lead to a future society that preserves and conserves these resources. To facilitate this search, this chapter approaches the problem in three stages. First, it attempts to identify the most pressing questions, especially with regard to the severity of the threat and the technical feasibility of solutions. Next, it attempts to distinguish those questions on which there is little or no scientific disagreement from those on which the evidence itself is disputed. Thirdly, it raises the most fundamental question of all: how to get from "where we are" to "where we need to be."

However, before plunging into the argument, some subsidiary topics are worthy of brief mention. These are discussed in the next two sections.

The need for holistic systems analysis

Since the early 1970s, the environmental movement has become increasingly professionalized and bureaucratized. As a consequence, largely, of the latter development, the "environment" is seen no longer in a holistic sense but in terms of a number of specific, essentially independent issues. Nowadays, the "causes" of pollution are attributed, for the most part, to narrowly defined actions (or failures to act) of equally narrowly defined "polluters." The responsibilities for abatement or clean-up are correspondingly narrow. Solid wastes, hazardous or toxic wastes, liquid wastes, and airborne wastes are likely to be allocated to different government departments, ranging from public health agencies to water/sewerage authorities, whose regulatory powers are controlled by different kinds of legislation framed in different circumstances, sometimes based on quite different regulatory philosophies. "The right hand does not know what the left hand is doing," and vice versa.

Activities of different arms of the same agency can interfere with each other. For instance, incineration can reduce the solid waste disposal problem, and even produce useful energy as a by-product, but it creates an air pollution problem. On the other hand, to reduce the emissions of particulates and sulphur oxides from power plants creates solid wastes that must be disposed of somewhere on land. There is nobody with a global view of the problem to mediate among the parochial interests. There is nobody with the responsibility or the authority to induce competing offices, departments, and bureaux to cooperate.

Yet the environment is, by its very nature, unsuited to incremental control strategies. It is equally unsuited for reductionist "bottom-up" modes of analysis. The problem is that scientific insights are now, and will continue to be, insufficient for predicting the detailed environmental consequences of any change or perturbation. To take a concrete instance, nobody can predict the exact physiological effects of ingesting any chemical from knowledge of its structure. Still less can the genetic or ecological consequences of its dispersion be predicted. This uncertainty is multiplied by the enormous number of different chemicals, materials, and mixtures simultaneously manufactured and used by man (natural and synthetic alike), not to mention the variety (type and intensity) of possible reaction modes and interaction effects.

Setting aside carcinogens and highly toxic or radioactive substances, one important environmental problem has as yet been predicted in advance from the creation or displacement of any particular material stream. This single exception was Rowland's chance recognition of the reactive potential of chlorofluorocarbons (CFCs) in the stratosphere, and the resulting possibility of stratospheric ozone depletion. This potential hazard, derided by chemical industry spokesmen in the 1970s as "speculative," has turned out to be real.

In speaking of the environment it is literally true that "everything depends upon everything else." A holistic "top-down" perspective is essential to identifying the most important underlying factors and relationships. It is equally important to adopt a very broad perspective for seeking and - it is hoped - finding effective global strategies to save the planet.

Environmental threats and (un)sustainability indicators

There has been a good deal of academic debate in recent years on the exact meaning that should be ascribed to the term "sustainability." For instance, Repetto states that "current decisions should not impair the prospects for maintaining or improving future living standards" (Repetto 1985, p. 16). The WCED paraphrased the same general idea; sustainable development "meets the needs of the present without compromising the ability of future generations to meet their own needs" (Brundtland 1987). Tietenberg phrases it in utility terms, and defines sustainability as non-declining utility (Tietenberg 1984, p. 33). Pezzey goes further and insists that it is the discounted present value of utility that should not decline (Pezzey 1989). Mainstream economists have concerned themselves with replacing depleted natural resource stocks. For instance, Nobel Laureate Robert Solow proposed that "an appropriate stock of capital - including the initial endowment of resources - [be] maintained intact" (Solow 1986). More recently Solow has said: "If 'sustainability' is anything more than a slogan or an expression of emotion, it must amount to an injunction to preserve productive capacity for the indefinite future. That is compatible with the use of non-renewable resources only if society as a whole replaces used-up resources with something else" (Solow 1992).

All of these definitions (and others) essentially agree on a single economic measure of welfare (GNP). They fundamentally assume unlimited substitutability between conventional economic goods and services that are traded in the market-place and unpriced environmental services, from stratospheric ozone to the carbon cycle. However, virtually all environmentalists and an increasing number of economists explicitly reject the unlimited substitutability view as simplistic (e.g. Boulding 1966; Ayres and Kneese 1971; Ayres 1978; Daly 1990). Similar critiques have been articulated by David Pearce and his colleagues (Pearce 1988; Pearce et al. 1989).

The "ecological" criterion for sustainability admits the likelihood that some of the important functions of the natural world cannot be replaced within any realistic time-frame - if ever - by human technology, however sophisticated. The need for arable land, water, and a benign climate for agriculture is an example; the role of reducing bacteria in recycling nutrient elements in the biosphere is another; the ozone layer of the stratosphere is a third. The ecological criterion for long-run sustainability implicitly allows for some technological intervention: for example, methods of artificially accelerating tree growth may compensate for some net decrease in the area devoted to forests. But, absent any plausible technological "fixes," this definition does not admit the acceptability of major climate changes, widespread desertification, deforestation of the tropics, accumulation of toxic heavy metals and non-biodegradable halogenated organics in soils and sediments, or sharp reductions in biodiversity, for instance.

Having said this, it is obviously easier to find indicators of unsustainability than of sustainability. In work for the Advisory Council for Research on Nature and the Environment (Netherlands), preparing for the UNCED Conference in Rio de Janeiro, 1992, Dutch researchers proposed a taxonomy of sustainability indicators (Weterings and Opschoor 1992). Their taxonomy has three dimensions:

1. Pollution of natural systems with xenobiotic substances or natural substances in unnatural concentrations. The results include acidification and "toxification" of the environment.

2. Depletion of natural resources: renewable, non-renewable, and semi-renewable. In fact, biodiversity can be regarded as a depletable resource, though not one that is commonly thought of as such. Of course, it also differs from other depletable resources that are exchanged in (and priced by) well-developed markets. There is no such market for biodiversity, or for its complement, genetic information. Nevertheless, I regard this as a market failure and argue that loss of biodiversity is an aspect of depletion.

3. Encroachment (human intervention) affecting natural systems, e.g. loss of groundwater or soil erosion.

Based on this taxonomy, Weterings and Opschoor prepared the summary table of quantifiable sustainability indicators shown in table 1.1. The notion of "sustainable level" in regard to pollution, toxification, acidification, greenhouse gas build-up, and so on is predicated on the idea that natural processes will compensate for some of the damage. For instance, natural weathering of rocks generates some alkaline materials that can neutralize acid. (Increased acidity will, however, increase the rate of weathering.) Similarly, it is assumed that some of the excess carbon dioxide produced by combustion processes may be absorbed in the oceans or taken up by accelerated photosynthetic activity in northern forests (this process is called "CO₂ fertilization").

Regarding depletion, it is assumed that some minerals (such as aluminum) can be mined more or less indefinitely, even though the highest-quality ores will be exhausted first. Other depletable ores could in effect be exhausted, in the sense that recovery from minable ores would be too expensive to be worthwhile except for very specialized and limited uses. Copper might be an example of this kind (though many geologists are more optimistic than Weterings and Opschoor). As regards renewable resources such as fisheries and groundwater, it has long been known that there is a level of exploitation that can be sustained indefinitely by scientific management, but that beyond that level harvesting pressures can drive populations down to the point where recovery may take decades, or may never occur at all. Many fisheries appear to be in this situation at present, notwithstanding the fact that sustainable levels are not very precisely known. Granted some uncertainty, it is nevertheless clear that, in all three dimensions, "sustainability" would require significant reductions in current levels of impact.

In recognition of the fact that both soil erosion and groundwater loss overlap considerably with the "depletion" category, the later version of their work substituted "loss of naturalness," namely loss of integrity, diversity, absence of disturbance (Weterings and Opschoor 1994). What remains in category (3) is the notion of "disturbance of natural systems" as such. Most environmentalists think of "systems" in terms of ecosystems and biomes. The sum total of such disturbances is indeed a significant environmental problem, though individual cases tend to be geographically localized. However there are also global systems that are being dangerously disturbed by anthropogenic activity. Examples of global systems include the hydrological cycle, ocean currents, the climate, the global radiation balance (including the ozone layer that protects the earth's surface from lethal ultraviolet radiation), the carbon/oxygen cycle, the nitrogen cycle, and the sulphur cycle.² This problem is discussed in detail later.

Table 1.1 Sustainable vs. expected level of environmental impact for selected indicators

Source: Weterings and Opschoor (1992), table 6, p. 25.

Holistic analysis presupposes that it is possible to classify variables by degree of importance and derive significant and defensible results by judicious simplification. A universal measure to estimate and compare the relative environmental impact of different activities, goods, services, 7and regulatory policies would be of great value.

Such a measure should satisfy the following conditions:

- it should be based on measurable quantities;

- it should relate to the most significant environmental impact potentials of human activities;

- it should allow transparent, cost-efficient, and reproducible estimates of the environmental impact potentials of all kinds of plans, processes, goods, and services;

- it must be applicable on the global level as well as regional and local levels.

Choosing a single indicator to compare the environmental impact intensities of all kinds of present and future processes, goods, and services might seem to be a daring step, precisely because it constitutes a vast reduction of complexity. Simplification cannot be proven to be "correct" in scientific terms.³ Only its plausibility in a variety of circumstances can be established.

For several reasons it can be argued that aggregate resource productivity, the ratio of GNP (or a better unit of economic welfare) to an index of total renewable-but-unrenowned or non-renewable resource inputs, in physical units, might be a plausible measure of sustainability. At least the two are correlated: the greater the resource productivity, the nearer to long-term sustainability. Obviously, the inverse of resource productivity - non-renewable or non-renewed resource use per unit of welfare output - is a measure of unsustainability.

Regrettably, neither this measure nor anything similar is currently computed at the national level by statistical agencies, and the required data are not readily available even to them, still less to non government organizations. However, note that the corresponding measure can be computed in principle for a sector (industry), a firm, a region with well-defined boundaries, or even a single product. Something like the inverse of resource productivity, materials intensity per unit service (MIPS), has been calculated for a number of specific cases at the Wuppertal Institute.⁴

Sharpening the debate

It is important now to confront three basic questions:

(1) Is continued economic growth (appropriately defined) compatible in principle with long-run ecological sustainability?

(2) If so, is our current mix of technologies and economic instruments consistent in practice with this goal?

(3) If not, what is the "least-cost" (and "least-pain") political/ institutional path from where we are now to a sustainable world economy? Will it be very expensive, as claimed by many conservatives, or are there enough opportunities for energy and material savings by intelligent use of "clean technology" to compensate for many of the costs?

This trio of central questions, as stated, currently elicits passionately opposed positions. Fortunately, several of these questions can be restated in a way that leads toward an answer. The first question above can be restated:

(1') Bearing in mind that most economists have been trained to believe that substitution of capital and/or technology for natural resources is virtually always possible, one can ask: is there any class of environmental assets or services both that is essential to human life (or to the biosphere) and for which there are no plausible substitutes?

If substitutability (e.g. of capital for environmental resources) is more or less without limit, or if the limits are very remote, then it can be argued that present trends are sustainable, or could become sustainable (depending on one's exact definition of sustainability) with a few marginal changes in policy.

A majority of business and political leaders appear to assume that only minor changes in current technology and/or regulatory policy would suffice to overcome any environmental threat. In fact, even most so-called "environmentalists" appear to believe that the most serious environmental threats we face are direct threats to human health (contaminated water or food, skin cancer) or loss of amenity (forest die-back, oil spills, dirty beaches, litter, haze, bad smells, etc.) No doubt ax-president Bush truly saw himself as an "environmentalist" because of his long-standing love of hunting, boating, and fishing. It has to be said, at the outset, that the problems that appear on most lists of "priority concerns" are localized, not global, problems. Even the rising public concern about loss of "endangered species" is limited to birds, fish, whales, and mammals - especially large mammals such as pandas and tigers. These are not the environmental problems of greatest concern from the standpoint of long-term survival of the earth as a habitable planet.

The second question can also be restated:

(2') If there are environmental assets and services that are both essential and non-substitutable (i.e. the answer to the first question is "yes") - as I believe - a second question follows: are any of these environmental assets or services now threatened by irreversible and/or irreparable damage? Is there a credible - not necessarily probable - threat to the long-term survival of life on this planet?

The answer to this question is obviously critical for what follows. It can be further broken down into several subsidiary questions, for example:

(2.1) Is continued global population growth compatible with long-run eco-sustainability? Can the most densely populated countries (China, India, Indonesia, Bangladesh) continue to feed themselves as their numbers increase? If not, what is the relationship between demographic variables and economic growth potential in various regions (notably China, India, and Africa)?
(2.2) Do industrial activity and its associated demand for raw materials and depletion of high-quality deposits of natural mineral or other environmental resources constitute a major constraint on continued environmentally sustainable economic growth? If so, how and why?
(2.3) Do waste and pollution (including acidification and environmental accumulation of toxic elements) constitute a direct threat to human welfare or to the habitability of the planet? For example, do they constitute a constraint on food production? If so, do they constitute a constraint on economic growth? If so, how and why?
(2.4) Does anthropogenic disturbance of balanced environmental systems (including ecosystems) constitute a major threat in the above sense? If so, how and why?

A brief digression is appropriate in connection with (2.4): the earth system depends on several balanced, biologically controlled recycling systems for nutrient elements that are required by living organisms in forms or amounts greater than would be found in the earth's crust or the prebiotic atmosphere or hydrosphere. Nitrogen, for instance, constitutes the major part of the atmosphere, but molecular nitrogen (N₂) is so stable that it is virtually unusable by plants or animals. It is only when this strong nitrogen bond is split by some external agency (yielding nitrogen compounds such as ammonia, ammonium, nitrates, or nitrogen oxides) that the nitrogen becomes a nutrient element. Free oxygen would not exist at all without living organisms; it would all be combined with other elements as water, carbonates, silicates, sulphates, etc. Carbon, too, would be tied up (mostly as insoluble carbonates) and unavailable. Thus, a truly "dead" planet (such as Mars or Venus) is literally uninhabitable.

Destruction of the earth's nutrient recycling systems would probably be the surest way of destroying all life on earth. To be sure, human intervention at present can better be characterized as "eutrophication", in the sense of sharply increasing the availability of these nutrients. Yet eutrophication in a lake or stream can be disastrous if it leads to an unbalanced and explosive growth of a few species, which exhaust the supply of some other nutrient (e.g. oxygen) resulting in a "crash" that destroys the whole food web. What we do not (and probably cannot) predict is the probability or imminence of such a threat at the global level. (I do suspect that it is more likely than being struck by a comet!)

Turning to the third major question, which concerns strategies for change and their cost, here too a breakdown into subsidiary questions is helpful. For example:

(3.1) Are there any feasible strategies, and implementable means, of bringing population growth to an end without government coercion, war, or epidemic? Which of them would involve the least economic cost and/or the least conflict with deeply held religious beliefs?

(3.2) Is there any fundamental technological limit (other than the second law of thermodynamics) to the energy and materials productivities that can be achieved in the long run? Is there any fundamental limit to the long run efficiency of materials recycling? To put it another way, is there a plausible set of technological "fixes"? We seek, in effect, an "existence proof" that solutions are possible.

(3.3) Among the technological "fixes" postulated above, is there one (or more) that is inexpensive, even profitable? Can the needed technology be harnessed at modest, or even negative, cost? From the macroeconomic perspective, the question is: can continued economic growth be achieved simultaneously with environmental improvement by increasing resource productivity - thereby reducing the need for resource inputs and the generation of wastes - without significantly decreasing labour and capital productivity? To put it another way, is it feasible to find ways to increase all factor productivities simultaneously, i.e. without substituting energy or material resources for labour? In simpler words, is there a mother-lode of "win-win" possibilities - "free lunches" - for reducing pollution and increasing the value of output at the same time?

It is interesting to note that affirmative answers to (3.2) and, especially, (3.3) - the existence of possible technological "fixes" at low (or no) cost imply a high degree of technological optimism. Curiously, most economists adopt an extremely optimistic stance in regard to questions of resource availability (2.2) but become pessimists when it comes to eliminating or repairing damage caused by pollution (3.3). It would seem logical that a Malthusian pessimist would be entitled to be pessimistic about the existence of "win-win" opportunities, but an opponent of the neo-Malthusian position should also be optimistic with regard to finding low-cost or profitable solutions to the growth problem. Simple consistency would seem to require that both question (2.2) - resource substitutability - and question (3.3) - technological "fixes" - be answered the same way: either "yes" to both or "no" to both.

Contents - Next