Economic Institutions and the Natural Environment - Oxford Scholarship Jump to ContentJump to Main Navigation
Human Well-Being and the Natural Environment$

Partha Dasgupta

Print publication date: 2001

Print ISBN-13: 9780199247882

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0199247889.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE ( (c) Copyright Oxford University Press, 2015. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see Subscriber: null; date: 28 September 2016

Economic Institutions and the Natural Environment

Economic Institutions and the Natural Environment

(p.107) 7 Economic Institutions and the Natural Environment
Human Well-Being and the Natural Environment

Partha Dasgupta (Contributor Webpage)

Oxford University Press

Abstract and Keywords

Three systems of rights to natural resources are studied in Ch. 7: private, communal, and state. Institutional failures are shown to be the cause of inefficiencies and inequities, both in momentary allocations of resources and in the inter‐generational transfer of resources. It is argued that in the world we have come to know, there is a bias in the use of the natural environment, in that use at any moment is excessive, not insufficient. Since observed prices frequently do not reflect the social worth of natural resources, use should be made of notional prices, called accounting prices.

Keywords:   accounting prices, efficiency, environment, equity, institutional failure, inter‐generational transfer, natural resources, resource allocation, resource rights, social worth

7.1. Markets

If you were to browse among leading Western journals on environmental and resource economics, you would discover that a recurrent activity in the field has been devising valuation methods. A question that would occur to you is, why? Why should there be a special need to determine the worth of natural resources; why not rely on market prices? You might also wonder why markets aren't an adequate set of institutions for our dealings with Nature.

The answer is that for many natural resources markets simply do not exist. In some cases they don't exist because the costs of negotiation and monitoring are too high. Economic activities affected by ecological interactions involving long geographical distances (the effects of uplands deforestation on downstream activities hundreds of miles away) form one class of examples. There are also interactions separated by large temporal distances (the effect of carbon emission on climate in the distant future, in a world where forward markets don't exist because future generations are not present today to negotiate with us). Then there are cases (the atmosphere, aquifers, the open seas) where the nature of the physical situation (the migratory nature of the resource) makes private property rights impossible and so keeps markets from existing; while in others, ill‐specified or unprotected property rights prevent markets from being formed (as is the case frequently with mangroves and coral reefs), or make them function wrongly even when they do get formed. Economists have a special name for the side‐effects of human activities when they are undertaken without mutual agreement: externalities. Not surprisingly, the study of externalities has loomed large in environmental economics.1 In elementary economics textbooks, it is (p.108) not uncommon to read that externalities lead to market failure. The way I have defined the term here, externalities are a form of institutional failure. One can go on to say that environmental problems are frequently symptoms of institutional failure, of which market failure is but one example.

It is convenient to classify externalities into two categories: unidirectional and reciprocal.2 Damage inflicted by upstream deforestation on downstream farmers without compensation is an example of the former. The famous ‘tragedy of the commons’ is a metaphor for the latter (see below). Carbon emissions into the atmosphere is one example of the tragedy; other examples are unregulated fishing in the open sea and groundwater withdrawal under free access.3

We will study unidirectional externalities in Chapter 12 (Section 12.4), where we identify certain weaknesses in the structural adjustment programmes that indebted countries were encouraged to put in place during the 1980s. Here I want to illustrate reciprocal externalities by discussing a mechanism that leads to the tragedy of the commons.

Consider a group of farmers who draw water from an underground basin. Although this isn't an example of open access to a resource base, it does harbour reciprocal externalities.4 While farmers may have titles to the land they cultivate, they can't have titles to the water below, for water is migratory. In view of this, communities have often instituted the doctrine, under which farmers have the right to extract as much water as they wish without regard to the effect of their withdrawals on others.

The problem with the doctrine is that it provides no protection to a well‐owner from the lowering of the water table under his land caused by his neighbour's action. In the absence of some form of collective action (say, a charge on water; see Section 10.2), the doctrine encourages farmers to extract at too fast a rate. Admittedly, if the farms are small, no single farmer can affect the water table significantly. But if there are many farmers, the aggregate effect can (p.109) be substantial. In extreme cases, over‐extraction ruins the basin, because, for example, of salt water intrusion.

Intergenerational externalities have a different structure. Because of the arrow of time, they are unidirectional. No doubt such externalities are to an extent ameliorated by the fact that we care about our children's well‐being and know that they, in turn, will care about theirs, and so on, in an intergenerational sequence. But because of institutional failure in other spheres of economic activity (borrowing, lending, purchasing insurance), our implicit concern for generations in the distant future may be inadequate. These considerations, among others, offer reasons why market rates of interest do not reflect social discount rates.5 So, market failure involves not only misallocation of resources in the present, but also misallocation across time. Even laissez‐faire economies aren't all that good at producing publicly observable signals of environmental scarcities. If there is open access to a fishery, for example, its market price, in situ, is zero. Being in limited supply, though, the scarcity value that ought to be attributed to it would be positive.6 There is a directional bias in environmental externalities: market failure usually leads to an excessive use of natural resources, not an insufficient one.

There is another reason why markets shouldn't be expected to function well for environmental resources. A major achievement of modern economics has been to show that prices in well‐functioning markets reflect social scarcities, but that markets are able to function well only if the processes governing the transformation of goods and services into further goods and services are ‘linear’. However, when someone talks of ‘ecosystem stress’ or ‘ecological thresholds’, as ecologists frequently do, they mean states of affair in systems governed by nonlinear processes.7 Such processes can be a feature not only of global ecosystems, but also of local ones. Even if a large‐scale ecosystem did not show signs of stress, local ones could, and often do, display signs, observable only by local inhabitants. Markets can't be expected to function well in such environments. For example, market prices may be unable to signal the impending collapse of a local resource base. This isn't to advocate command‐and‐control systems of management, but simply to say that markets need to be supplemented in such ways as to ensure there are additional public signals to accompany prices. Later we shall consider the kinds of supplements that would help matters.

(p.110) 7.2. The Local Community

In an oft‐quoted passage, Arrow (1974: 33) expressed the view that organizations are a means of achieving the benefits of collective action in situations where the price system fails. If interpreted literally, this formulation gets the historical chronology backward, but Arrow's intention was clear enough: there are institutions serving an economic purpose that are neither the market system nor the State.

Spatially localized resources are frequently neither private nor state property, but common property. In poor countries communal property rights to resources are most often based on custom and tradition: they aren't backed by the kinds of deeds that would pass scrutiny in courts of law. Therefore, tenure isn't secure, a vital problem, to which I return below.8 In poor countries the local commons include grazing lands, threshing grounds, swidden fallows, inland and coastal fisheries, rivers and canals, woodlands, forests, village tanks, and ponds. Being spatially localized, their use can be monitored by members of the community. This makes their management an easier matter than, say, fisheries in the open seas.

Are the local commons extensive? As a proportion of aggregate assets, their presence ranges widely across ecological zones. There is a rationale for this, based on the human desire to reduce risk. Communal property rights enable members of a group to reduce individual risks by pooling their risks. Moreover, the incentive to pool risks that are associated with the use of any particular resource depends on the other risks people face; it depends on their remaining sources of income, on transaction possibilities in other spheres of life, and so forth. An almost immediate empirical corollary is that the local commons are most prominent in arid regions, mountain regions, and unirrigated areas, and least prominent in humid regions and river valleys.9 Another corollary is that income inequalities are less in those locations where common‐property resources are more prominent. Aggregate income is a different matter, though; it is the arid and mountain regions and unirrigated areas that are the poorest.10 This needs to be borne in mind when public policy is devised. However, the dependence on common‐property resources even within dry regions declines with increasing wealth across households. Theory predicts this and case studies confirm it.11

Jodha (1986, 1995) studied evidence from over eighty villages in twenty‐one dry districts in India to conclude that, among poor families, the proportion of (p.111) income based directly on the local commons is in the range of 15–25 per cent. Moreover, as sources of income, they often complement private property (i.e. labour, milch and draft animals, agricultural land, and tools). Common‐property resources offer the rural poor some protection in times of unusual economic stress. For landless people they may be the only non‐human asset at their disposal. A number of resources, such as water and minor forest products (fuelwood and fodder, berries and nuts, herbs and spices, resin and gum) are the responsibility of women and children.

In a study of twenty‐nine villages in south‐eastern Zimbabwe, Cavendish (1998, 1999) has arrived at even larger estimates: the proportion of income based directly on the commons is 35 per cent, the figure for the poorest quintile being 40 per cent. A similar picture emerges from Hecht, Anderson, and May (1988), who offered qualitative descriptions of the importance of babassu products among the landless in Maranhão, Brazil. The products support the poorest of the poor, most especially the women among them. They are an important source of cash income in the period between agricultural crop harvests.12

Such evidence as Jodha and Cavendish have unearthed does not, of course, prove that the local commons in their samples were well managed, but it does show that rural households would have strong incentives to devise arrangements whereby they would be communally managed.

Are the local commons managed communally? Not invariably, but in many cases they are, or have been in the past. Where they are managed, the commons aren't open to all, but only to those having historical rights, through kinship ties and community membership. In short, the local commons usually aren't open to all; they aren't open access. Communal management of local resources makes connection with social capital, viewed as a complex of interpersonal networks and hints at the basis upon which cooperation has traditionally been built.13 A large empirical literature has confirmed that resource users in many cases cooperate, on occasion through democratic means. Case studies have shown that communal property‐rights and management have prevented rural and coastal communities from experiencing the tragedy of the commons. In his work on South Indian villages, Seabright (1997) has shown that milk producers' cooperatives are more prevalent in the drier districts. But the local commons are also more prevalent in drier districts. So, one way to interpret Seabright's finding is that cooperation in one sphere of life (managing the commons) makes cooperation in other spheres (marketing milk) that much easier. Cooperation begets cooperation. The literature on the local commons is valuable because it (p.112) has unearthed how institutions that are neither part of the market system, nor of the State, nor of the household, develop organically to cope with resource allocation problems.14

This is the good news. There are, however, two pieces of bad news. First, a general finding from studies on the management of common‐property resources is that entitlements to products of the commons is, and was, frequently based on private holdings: richer households enjoy a greater proportion of the benefits from the commons.15 This is consonant with cooperative game theory. In his classic formulation of the way rational people could be expected to bargain, Nash (1950) showed that those who enjoy better outside options would enjoy a greater share of the benefits of cooperation.16

The second piece of bad news is that the local commons have degraded in recent years in many parts of the poor world.17 Why should this happen now if the commons have previously been managed in a sustainable manner? A recent intellectual tradition goes something like this: the reason the poor degrade their local resource base is that, at the margin, income today is of the utmost urgency, meaning that poverty leads poor people to discount future incomes at unusually high rates relative to today's income.18

I don't know of many empirical studies testifying to this, but there are some. High rates of discount have been reported among the rural poor in a sample of villages in Indonesia and sub‐Saharan Africa. Rates as high as 100 per cent per year were inferred from responses to questionnaires the authors distributed among villagers.19

These findings are not easy to interpret. One would imagine that, if discount rates were as high as have been inferred, the local commons in those regions would have been depleted at a faster rate than the evidence, such as it is, suggests.20 Admittedly, limits on the availability of effective labour could be a reason why depletion rates can be low even when people discount the future at rates far higher than the natural rates at which the resource base regenerates. But (p.113) depleting the local commons doesn't require much capital equipment. Farmers in some of the regions in question until recently practised swidden agriculture, some practise it even today. Nor is a lack of people a notable problem in Indonesia and sub‐Saharan Africa. Moreover, poverty is not a recent phenomenon. If the thesis were correct, the local commons would have disappeared long ago. The matter remains a puzzle. So I consider a different explanation. It is based on the thought that institutional failure can mean that private returns on investment in the resource base are low. In the world as we know it, this is probably a common occurrence.

There are many ways in which institutional failure manifests itself. Uncertain property rights are a prime example. You may think that you own the land your forefathers cultivated and passed on to you, but if you do not possess a deed to the land, your rights to it are insecure. Political instability (at the extreme, civil war) is another source of uncertainty: your property could be taken away from you by force. In fact, political instability is a direct cause of environmental degradation: civil disturbance all too frequently expresses itself through the destruction of physical capital.

When people are uncertain of their rights to a piece of property, they are reluctant to make the investments necessary to protect and improve it, the expected returns being low. Often enough, the investments have to be of a collective kind, for example building irrigation ditches, terraces, and structures for controlling floods. If the probity of communitarian institutions is uncertain, the private returns expected from collective work are low. The influence would be expected to run the other way too, with growing resource scarcity contributing to political instability, as rival groups battle over resources. The feedback could be positive, exacerbating the problem for a time, reducing private returns on investment further. Groups fighting over spatially localized resources are a common occurrence.21

Rapid population growth can trigger resource depletion if institutional practices are unable to adapt to changing economic circumstances. In Côte d'Ivoire, for example, growth in rural population has been accompanied by increased deforestation and reduced fallows. Biomass production has declined, as has agricultural productivity.22

Management of the local commons has often relied on social norms of behaviour, which are founded on reciprocity. But institutions based on reciprocity are fragile in the face of growing markets. When traditional systems of (p.114) management collapse, and aren't replaced, the use of the local commons becomes unrestrained. The commons then deteriorate, leading to the proverbial ‘tragedy’. Later (Chapter 12) we will see why reciprocity is fragile.

7.3. The State

The State is often the culprit. There are cases where state appropriation of communal property (forests) removed the local population's incentives to protect the resource base. Less directly, management practices at the local level have been overturned by central fiat. A number of states in the Sahel imposed rules that in effect destroyed communal management practices in the forests. Villages ceased to have the authority to enforce sanctions on those who violated locally instituted rules. State authority damaged local institutions and turned the local commons into open‐access resources.23 There have also been cases where the ability of the State to enforce environmental laws diminished under structural adjustment programmes. As this was frequently not accompanied by the emergence of communal management systems, the resource base in effect became open to access by one and all.24

Recently, environmental degradation has been traced to government tax policy. In a notable paper, Binswanger (1991) argued that in Brazil the exemption from taxation of virtually all agricultural income, allied to the fact that logging is regarded as proof of land occupancy, has provided incentives to the rich to acquire forests and then to deforest the land.25 He argued that the subsidy enjoyed by the private sector has been so large that a reduction in deforestation by the removal of subsidies is in Brazil's own interests, let alone that of the rest of the world. This has implications for international negotiations. The current consensus is that Brazil has much to lose from reducing the rate of the deforestation it has been engaged in. If this is true, there would be a case for the rest of the world to subsidize Brazil, as compensation for losses the country would sustain if it were to exercise restraint. Binswanger's account shows that it is by no means clear that the consensus is correct.

Alston, Libecap, and Mueller (1999a,b) have taken the analysis farther. They have shown that accelerated deforestation in the Brazilian Amazon, followed by violent conflicts between landowners and squatters, has also occurred because of legal inconsistencies between the civil law, which supports the title held by landowners, and the constitutional law, which supports the right of squatters to claim land not in ‘beneficial use’ (e.g. farming or ranching). Ironically, the (p.115) latter right reflects the government's stated desire for land reform. The authors have shown that the vagueness of the ‘use’ criteria, and the uncertainty as to when a landowner's claim to a piece of land or a squatter's counter‐claim to it is enforced, are together an explosive force.26

I have stressed weaknesses in the workings of the State in poor countries for two reasons. First, if the natural environment had been managed well in such countries, I wouldn't be writing this book. Secondly, the State is of the utmost importance for environmental management. If markets are to work well, property rights must be clearly defined and protected and contracts have to be enforced. This only the State can do. If communitarian allocations are to be equitable, there has to be an assurance that the benefits from the local commons aren't expropriated by the more powerful within the community. This only the State can assure. If the rights to the benefits of investment in a resource base of those who invest in it are to be protected and promoted, the State is the agency to which they must turn. Moreover, there are infra‐structural investments (sewage systems, water purification plants) that in poor countries only the State is capable of financing. In urban centres, where communitarian systems are unlikely to get off ground (e.g. on air pollution), the State, with its powers to tax and regulate, is the pivotal agency of reform. Environmental economics in the North (I am using the term in its current geo‐political sense) has for the most part been developed as a branch of public economics, with its emphasis on optimal instruments for pollution control (e.g. taxes, tradable permits; see Chapter 10). On each such matter, the State has a key role to play.

State ownership of wells and mines is a commonplace in poor countries. Fossil fuels and minerals are a source of government revenue, used for development purposes in some countries and the accumulation of the rulers' personal wealth in others. Forests are often state property. They too are resources of commercial value. But there is a category of resources that, even though they may be perceived as state property, are actually public property. The State doesn't ‘own’ them: instead, it acts as a trustee on behalf of all generations. National parks and cultural monuments are examples of public goods whose preservation is the responsibility of the State as trustee.

(p.116) 7.4. Property Rights and Management: A Schemata

Property‐rights to a resource are the rights, restrictions, and privileges with regard to its use.27 Management of a resource base is a different, though related, matter. Property rights may be well defined, but a resource could nevertheless be managed badly, owing to disagreements among those holding the rights on how it ought to be used, corrupt practices regarding its use, a lack of understanding of ecosystem functions, and so forth.28 Ownership is yet another matter. Someone owns a resource by law, but transfers their right of use to another person for a limited period. Ownership doesn't change, but the right‐of‐use does.

Private, communal, and state property rights to resources were discussed in the previous three sections. Table 7.1 presents the classification. We can think of a property rights regime and the associated management structure as defining the institution governing the use of a resource base. In practice, every institution harbours externalities, but some harbour more than others. The relative efficiency of institutions depends on the character of the resource base (e.g. if its productivity is subject to especial risk, if the resource base is lumpy, if its use by someone is observable by other relevant parties); it depends on the efficiency of institutions governing the production and allocation of other goods and services (e.g. credit and insurance); it depends on the extent to which the returns from investment in it can be appropriated by the investor; and so forth. The interplay of market and non‐market institutions is a crucial determinant of well‐being. The performance of one depends on the performances of the others. Nor are institutions given and fixed. They evolve. For example, Ensminger (1990) has narrated how common grazing lands among the Orma

Table 7.1. Property Rights to Natural Resources




Cultivated land, cattle, oil and minerals

Communal (common‐property)

Grazing land, threshing grounds, ponds, local forests, inland and coastal fisheries

Open access

Atmosphere, international waters


Commercial forests, oil and minerals

State as trustee

National parks, cultural monuments

(p.117) in north‐eastern Kenya became segmented into private land. She showed that the transformation took place with the consent of the elders of the tribe, and attributed this to changing transaction costs brought about by cheaper transportation and widening markets. The elders were, not surprisingly, from the stronger families, and it did not go unnoted by Ensminger that privatization accentuated inequalities.

Institutions can also be created or destroyed by design and legislation. What was once a relatively efficient institution for a resource base may cease being so when circumstances change. Nostalgia for communitarian solutions is an undependable emotion.

7.5. Global and Local Environmental Problems

The links between rural poverty and the state of the local resource base in poor countries offer a possible pathway by which poverty, resource degradation, and even high fertility feed upon one another in a synergistic manner.29 Recent experiences in sub‐Saharan Africa and the Indian sub‐continent are not inconsistent with this possibility. An erosion of the local resource base can make certain categories of people destitute even while the economy's gross national product (GNP) increases. The thought that entire populations can always be relied upon to make the shift from resource based, subsistence existence to a high‐income, industrial one is belied both by theory and by contemporary experiences.30

Rural poverty, resource degradation, and institutional failure pull in different directions and are together not unrelated to an intellectual tension between the concerns people share about global warming and acid rains, which sweep across regions, nations and continents, and about those matters (such as the decline in firewood or water sources) that are specific to the needs and concerns of the poor in as small a group as a village community. Environmental problems present themselves differently to different people. In part, this is a reflection of the tension just noted and is a source of misunderstanding of people's attitudes. Some identify environmental problems with poverty and unprecedented population growth in the South, while others identify them with wealth and unprecedented expenditure patterns in the North. (I am using the geographical terms in their current geo‐political sense.) Even though debates between (p.118) the two groups often become shrill, each vision is partly correct. There is no single environmental problem. Consequently, there is no single valuation or evaluation problem: rather, there is a large collection of them.31 Thus, growth in industrial wastes and resource use have come in tandem with increased economic activity. In the former Socialist block neither preventive nor curative measures kept pace with the production of waste. Moreover, the scale of the human enterprise, by virtue of both unprecedented increases in the size of the world's population and the extent of economic activity, has so extended our use of natural resources that humankind is today Earth's dominant species. During the twentieth century, world population grew by a factor of four to more than 6 billion, industrial output increased by a factor of 40, energy use by a factor of 16, methane‐producing cattle population grew in pace with human population, fish catch grew by a multiple of 35, and carbon and sulfur dioxide emissions by a factor of 10. The supply of nitrogen to the environment from the use of fertilizers and from burning fossil fuels is now of the same order of magnitude as biological nitrogen fixation. Vitousek et al. (1986) estimated that 40 per cent of the net energy created by terrestrial photosynthesis (i.e. net primary production of the biosphere) is currently being appropriated for human use. This is of course a rough‐and‐ready figure. Moreover, net terrestrial primary production isn't given and fixed: it depends in part on human activity. Nevertheless, the figure does put the scale of the human presence on Earth in perspective. The figures also give us an idea of the unprecedented disturbance to the natural environment that has been created by human activity in a short space of time.32

If the demonstration effect is a powerful force behind consumer demand (Section 3.4), a substantial portion of our use of the natural resource base should be seen as a collective waste. The waste occurs not because of inadequate property rights to the resource base, but because preferences are interdependent.33 In principle, the demonstration effect can give rise to any one of a number of patterns of resource use. For example, each household would desire high levels of consumption (despite the hard work involved in earning the incomes necessary for high levels of consumption) if other households were to attain high levels of consumption, but would desire moderate levels of consumption if other households were to attain moderate levels of consumption.34 It could even (p.119) be that people would be happier under the latter circumstances. Consumption taxes would be a policy implication.

The reference group giving rise to the demonstration effect can expand, because newspapers, radio, television (now also the internet) transmit information about other life‐styles.35 The media are then a vehicle by which conformism increasingly becomes based on the behaviour of a wider population than the local community: the reference group widens. This pathway helps explain recent trends in reproductive behaviour in a number of poor countries, where fertility rates displayed little to no trend in the past, but have declined in recent years.36 The presumption is that the poor are prone to imitating the rich, but the rich aren't prone to imitating the poor.

It would seem, then, that the enlargement of the reference group has two broad effects: a downward trend in fertility rates, and an upward trend in acquisitiveness and economic growth. Other things being the same, the former would reduce human expropriation of the products of nature, while the latter would increase it.

On the other hand, economic growth itself has brought with it improvements in the quality of a number of natural resources. The large‐scale availability of potable water and the increased protection of human populations against both water‐ and air‐borne diseases in industrial countries have in great measure come in the wake of growth in national income these countries have enjoyed over the past two hundred years or so. Moreover, the physical environment inside the home has improved beyond measure with economic growth. For example, cooking in South Asia continues to be a central route to respiratory illnesses among women. Such positive links between economic growth and environmental quality often go unnoted by environmentalists in the North. I would guess that this lacuna is yet another reflection of the fact that it is all too easy to overlook the enormous heterogeneity of the natural environment, ranging as it does from the atmosphere, oceans, and landscapes to water‐holes, grazing fields, and sources of fuelwood. Both this heterogeneity and the diversity of the human condition need to be kept in mind.

7.6. Technological Biases

In earlier chapters it was noted that estimates of socio‐economic indicators currently in use are biased because they don't incorporate changes in the natural environment. The price of natural resources on site is frequently zero, even (p.120) though they are scarce goods. Commercial rates of return on investments relying on such resources are higher than their social rates of return. Resource‐intensive projects appear better looking than they actually are. Over time, an entire sequence of resource‐intensive technologies is installed. Moreover, people learn by doing and learn by using, not only in installed technology, but also in research and development. The development and use of technology reflect processes that are path‐dependent. The conclusion is depressing: it may require a big push to move us away from the current profligacy in our use of natural resources.

These arguments imply that modern technologies are not always appropriate technologies, but instead are often unfriendly towards those who depend directly on their local resource base. This is likely to be especially true in poor countries, where environmental legislations are usually neither strong nor effectively enforced. The arguments help explain why the poorest in poor countries, when permitted, have been known to protest against the installation of modern technology. The transfer of technology from advanced countries can be inappropriate even when that same body of technology is appropriate in the country of origin. This is because the social scarcity of natural resources, especially local resources, varies from country to country.37 A project design that is socially profitable in one country may be socially unprofitable in another. This may be why environmental groups in poor countries not infrequently appear to be backward‐looking, trying to unearth traditional technologies for soil conservation, water management, forest protection, medical treatment, and so forth. To do so isn't to assume an anti‐science stance: it could be to infer that wrong prices can tilt the technological agenda in a wrong direction.38

The bias towards resource‐intensive technologies extends to the prior stage of research and development. When natural resources are underpriced, the incentives to develop technologies that would economize on their use are lower than what they should be. Often enough, once it is perceived that past choices have been damaging to the environment, cures are sought, whereas prevention would have been the better choice. Chichilnisky and Heal (1998), for example, in an ex post study of political choice over improving water quality, compared the costs of restoring the ecological functioning of the Catskill Watershed in New York State to the costs of replacing the natural water purification and filtration services the ecosystem had been known to provide in the past by building a water purification plant costing 8 billion US dollars. Their figures reflect the advantages of the choice that was in fact made: preservation over construction. Independent of the other service the Catskill watershed provides, and ignoring the annual running costs of 300 million US dollars for a filtration (p.121) plant, the capital costs alone showed a more than six‐fold advantage for investing in the ecosystem. Their investigation offers a rough estimate of the social worth of the watershed itself.

But customary habits of economic thinking are hard to overcome. Accounting for the natural environment, if it comes into the calculus at all, is an afterthought to the real business of ‘doing economics’. A recent issue of The Economist (25 September 1999) carried a 38‐page Survey of the World Economy in which the natural resource base made no appearance in the authors' assessment of what lies ahead. I doubt though that many readers will have noticed this. Even today the natural environment has not entered the common lexicon of economic reasoning. In Chapter 9 we will see that assessments of economic performance can be very misleading when the natural resource base is neglected in the calculus.


This chapter is based on Dasgupta and Mäler (1991, 1995) and Dasgupta (1993, 1996).

(1) See e.g. Pigou (1920), Lindahl (1958), Coase (1960), Kneese and Bower (1968), Arrow (1971), Meade (1973), Mäler (1974), and Baumol and Oates (1975). Sandmo (2000) contains a fine overview of environmental externalities. Optimal public instruments for the preservation of amenities and the control of pollution have been the focus of attention in the USA, influenced considerably by the tasks facing the US Environmental Protection Agency (EPA); see Cropper and Oates (1992). Morgenstern (1997) is a collection of studies on regulatory work undertaken at EPA. On the theory of transboundary pollution, see Mäler and de Zeeuw (1998), Folmer and von Mouche (2000), and Folmer and de Zeeuw (2000), and on transnational commons, see the essays in Dasgupta, Mäler, and Vercelli (1997).

(2) Dasgupta (1982b).

(3) Gordon (1954) was the first to analyse the implications of open access to a resource base. Scott (1955) is an original study on the effects of open access to fisheries, and Milliman (1956) is another on the effects on groundwater. Garrett Hardin's invention of the admirable metaphor, ‘the tragedy of the commons’ (Hardin, 1968), has done much to create public understanding of the problems that arise when property rights to resources are inadequate. On the impact of agriculture on groundwater, see Giacomelli, Giupponi, and Paniconi (2001).

(4) Dasgupta and Heal (1979: ch. 3) developed a formal, game‐theoretic model of reciprocal externalities. They showed that, even when there isn't open access, there can be a tragedy of the commons if those enjoying rights of use do not act with restraint.

(5) See Lind (1982), Arrow et al. (1996), and Portney and Weyant (1999).

(6) I am assuming here that there are no external disbenefits inflicted on others when fishermen fish. In Sect. 8.1 I discuss cases where there are external disbenefits associated with the exploitation of natural resources under open access.

(7) Strictly, they are ‘non‐convex’ processes. A process is said to be convex if, given any two time paths that are feasible under the process, all time paths that are weighted averages of the two (with positive weights) are also feasible. Non‐convex processes are therefore non‐linear. Koopmans (1957) is the classic reference on the subject.

(8) See also Ch. 12.

(9) This has been confirmed in the case of India by Agarwal and Narain (1989) and by Chopra, Kadekodi, and Murty (1990).

(10) See Agarwal and Narain (1989) and Chopra, Kadekodi, and Murty (1990).

(11) Jodha (1986, 1995) and Cavendish (1998, 1999).

(12) For a similar picture in the West African forest zone, see Falconer (1990).

(13) I provide a brief account of the idea of social capital in Chs. 9 and 12. For a more extensive account, see Dasgupta (2000a).

(14) See Feeny et al. (1990), Ostrom (1990), Somanathan (1991), Bromley et al. (1992), Acheson (1993), and Baland and Platteau (1996). The economic theory underlying communal management was developed in Dasgupta and Heal (1979: ch. 3). Ostrom (1990: ch. 3) ranges over a number of long‐enduring common‐property resources in Nepal, the youngest of which has been found to be 100 years old, the oldest more than 500 years old.

(15) McKean (1992) and Ostrom and Gardner (1993). See also Molians (1998), who has identified differences based on gender. In India caste also plays a role in segregating access to the commons. For example, people belonging to ‘scheduled castes’ are frequently not permitted to draw water from the wells in use by caste Hindus.

(16) Banerjee et al. (2001) have uncovered similar phenomenon in agro‐industrial cooperatives. Their theory is broader and doesn't rely on the bargaining model of Nash.

(17) B. Agarwal (1989) and Baland and Platteau (1996) provide examples.

(18) Bardhan (1996).

(19) Holden, Shiferaw, and Wik (1998).

(20) For estimates of annual rates of deforestation in sub‐Saharan Africa, see Cleaver and Schreiber (1994).

(21) Homer‐Dixon, Boutwell, and Rathjens (1993), Homer‐Dixon (1994), and Collier and Hoeffler (1998). In a study comprising 120 countries, Deacon (1994) has offered suggestive statistical evidence of a positive link between political instability and forest depletion. On poverty and degradation of the local environment, see B. Agarwal (1986, 1989), CSE (1990), Dasgupta (1993, 1995, 2000b,c), and Cleaver and Schreiber (1994).

(22) Lopez (1998) estimates that income at the village level has declined by some 15%.

(23) See Thomson, Feeny, and Oakerson (1986) and Baland and Platteau (1996).

(24) Reed (1996).

(25) Heath and Binswanger (1996) have extended their basic argument to a number of other countries.

(26) In a wider discussion of the conversion of forests into ranches in the Amazon basin, Schneider (1995) has shown that the construction of roads through the forests has also been a potent force. Other examples of policy‐induced environmental deterioration are the massive agricultural subsidies in the European Union. These are known to have encouraged agricultural practices harmful to aquatic ecosystems. In their estimate of world‐wide subsidies for the use of natural resources, both visible and hidden, Myers and Kent (2000) suggest an annual figure of 2 trillion US dollars, which is about 10% of the world's output of goods and services.

(27) See Bromley (1991) for a wide‐ranging discussion.

(28) Ostrom and Schlager (1996) contains an account of the conditions necessary for effective communitarian management.

(29) See Sect. 5.7.

(30) Cleaver and Schreiber (1994), Filmer and Pritchett (1996), and Aggarwal, Netanyahu, and Romano (2001), among others, offer suggestive evidence. In their study of the effects of stabilization and structural adjustment programmes in the Philippines, Cruz and Repetto (1992) offer evidence in support of a positive feedback between poverty, population growth, and soil degradation. These matters are further discussed in Chs. 9 (Table 9.2) and 12.

(31) Dasgupta and Mäler (1991, 1995), Dasgupta (1993), Reardon and Vosti (1995), and Vincent and Ali (1997) go into these distinctions in greater detail.

(32) This was the theme of a special symposium in Science, 1997, 277 (see especially the article by Vitousek et al.). See also McNeill (2000) for global statistics on changes in the magnitude of the perturbations that were made to the natural environment during the 20th century.

(33) Technically put, there are demand externalities.

(34) This is an example of multiple equilibria. We encountered the possibility of multiple equilibria in Sect. 4.4.

(35) Freedman (1995), Bongaarts and Watkins (1996), and Iyer (2000).

(36) Dasgupta (2000b,c).

(37) This is proved formally in the Appendix.

(38) Agarwal and Narain (1996) is an interesting recent study in this vein.