Text of the BES Lecture delivered at the Annual Conference of the British Ecological Society, Oxford, on 7 September 2006. The exposition relies on my book, Economics: A Very Short Introduction (forthcoming, 2007, in the Very Short Introduction Series of Oxford University Press, Oxford). For their comments on an earlier draft, I am most grateful to Kenneth Arrow and Paul Ehrlich. The author is the Frank Ramsey Professor of Economics at the University of Cambridge and Fellow of St John's College, Cambridge, UK.
Professor Sir Partha Dasgupta, University of Cambridge, and St John's College, Cambridge, UK (e-mail: firstname.lastname@example.org).
- 1In this lecture I first offer what one could call the World Bank view of the recent macroeconomic history of a number of countries in the poor and rich worlds. Secondly, I demonstrate how our view of macroeconomic history changes if Nature is included as a capital asset in production activities.
- 2I conclude that high population growth in the world's poorest regions (South Asia and sub-Saharan Africa) has been an obstacle to the achievement of sustainable economic development in those areas. It is believed that people in those regions are, on average, less wealthy now than they were 35 years ago.
- 3When population growth is taken into account, the accumulation of manufactured capital, knowledge and human capital (health and education) has not compensated for the degradation of natural capital in South Asia and sub-Saharan Africa and, in all probability, even in the United Kingdom and the United States.
- 4It is possible that China is an exception to the economic forces experienced in other areas of the world.
- 5Synthesis and applications. The conclusions drawn here are very tentative and there is much further work to be conducted in understanding how ecological concerns can be incorporated into economic theory. It is important, therefore, that growth economists, demographers, governments and international agencies take this approach.
Ecologists and economists have much to say to each other, yet they rarely converse. There are exceptions, of course, but typically ecologists do not read economics even when they come to deploy economic reasoning, and economists blissfully neglect ecology when studying the economic progress of communities and nations.
In order to illustrate how ecological truths can be introduced into economic reasoning I outline the ‘World Bank view’ of recent macroeconomic history in a number of countries in both the poor and rich worlds. The statistics are widely known (Table 1) but can be used to show how very different macroeconomic history begins to look if one includes Nature as a capital asset in production activities.
Table 1. Various economic and demographic metrics for rich and poor nations
|GDP per capita||$30 000||$2100|
|Human Development Index||High||Low|
|Annual population growth rate (%): 1966–2004||0·8||2·4|
|Annual growth rate of GDP per capita (%): 1966–2004||2·4||1·8|
|Total fertility rate (TFR)||1·8||3·7|
|Adult literacy (%)||> 95||58|
|(female literacy (%)||(> 95)||(48)|
|Index of government corruption||Low||High|
|Life expectancy at birth (years)||78||58|
|Under 5 mortality (per 1000)||7||120|
|Rural population (% of total population)||10||70|
|Agriculture's share in GDP (%)||5||25|
The underlying motivations behind this lecture are thus twofold: first, I argue that it is high time that growth economists, demographers, governments and international agencies took Nature seriously. Secondly, I will demonstrate that progress has been made in recent years, among small groups of ecologists and economists, in developing appropriate ways to study the progress of economies –whether the study is for descriptive or for prescriptive purposes.
In order to discuss economic progress or failure, we need a measuring rod. The one most commonly used today is ‘gross domestic product per person’ – GDP per capita. Economists may have invented the concept and may have also pointed to its many limitations; but the term is so ingrained in public consciousness that reference to ‘Economic growth’ is understood widely to mean growth in real GDP per capita; which is growth in GDP per capita, corrected for inflation or deflation.
A country's GDP is the value of all the final goods that are produced by its residents in a given year. It is a measure of an economy's total output. But when a commodity is produced and sold, the price paid for the purchase finds its way into someone's pocket. Therefore, GDP can be measured also by adding up everyone's incomes – wages, salaries, interests, profits and government income. GDP and national income are therefore two sides of the same coin.
Although GDP is often said to measure wealth, it does not do so. GDP is a flow (e.g. dollars per year), whereas wealth is a stock (dollars – at one point in time). As the concept of GDP was developed originally for market economies, the values imputed to the goods were market prices. But by a clever construction of notional prices (called ‘shadow prices’; see below), economists have adapted GDP even for economies where large volumes of transactions are undertaken in non-market institutions.
Adjusting for differences in the cost of living across the world, global income per head today is about $8000 per year. However, for most of humanity's past, people have been abysmally poor. The economic statistician Angus Maddison has estimated from the very fragmentary evidence that exists that, at the beginning of our Common Era (ce 0) the per capita income of the world was about $515 per year at today's prices (Maddison 2001). If Maddison's estimate is even approximately correct, it means that the average person 2000 years ago enjoyed not much more than a dollar a day, a figure deemed by the World Bank as the line below which a person is in extreme poverty. Maddison has also suggested that the distribution of income 2000 years ago was remarkably equal: almost everyone, everywhere, was very poor. The figures he has constructed tell us furthermore that average world income and the regional distribution of income per head were pretty much the same in ce 1000 as they had been 1000 years earlier. It would appear that regional disparities became significant only from the beginning of the 19th century: income per head in western Europe had by then become three times that in Africa. But world income per head was still only $755 per year at today's prices, having increased by less than 50% over an 1800-year period, amounting to an annual growth rate of under 0·02%. The figure is extremely low by contemporary standards: the annual growth rate of income per head has been 2% per year over the past four decades. [Note that if a numerical entity – for example, real GDP per person – grows (or declines) at the annual rate of g%, that entity doubles (or halves) approximately every 70 g−1 years. For example, GDP per capita would double every 35 years were it to grow at an annual rate of 2%; and halve every 140 years if it declined at an annual rate of 0·5%.]
Large regional disparities in income are also less than 200 years old. The ratio of the average incomes in the United States and Africa has risen from three at the beginning of the 19th century to more than 20 today – about $38 000 compared to $1850 per year. Real GDP per capita in the United States has grown 30 times in size in 200 years, implying that the average annual growth rate of income per person there has been about 1·7%. In sad contrast, income per capita in Ethiopia is about the same today as it was 200 years ago (a little over $700 per year).
When countries are ranked according to GDP per capita today, middle-income nations (China, Brazil, Venezuela and Argentina are prominent examples) are spread thinly between two extremes A large cluster of poor countries (in sub-Saharan Africa, the Indian subcontinent, South-east Asia, Melanesia and Central America) – with a total population of 2·3 billion – produces an average $2100 per year per head, while another, smaller, cluster of rich countries (Europe, North America, Australia and Japan) – with a total population of just less than 1 billion – enjoys an average annual income of $30 000 (Table 1). The world is polarized. Moreover, with the possible exception of India, there is little sign that the poor world will catch up with the rich world in the foreseeable future. During the past four decades, real per capita GDP has grown at an average annual rate of 2·4% in rich countries, but at only 1·8% in poor countries (Table 1). Worse, within the poor world sub-Saharan Africa has experienced a small decline in real GDP per capita during the past four decades.
In contrast to poor countries, agricultural output is a small fraction of national income in the rich world. The share of agriculture in GDP is about 25% in the poor world; less than 5% in rich countries. Less than 10% of the population in rich countries live in rural areas. In contrast, more than 70% of people in poor countries live in villages (Table 1). This suggests that people in poor countries work mainly in economies that draw their production inputs directly from Nature; they are ‘biomass-based’ economies. Ecology is of direct concern to the world's poor in a way that it is not to the world's rich.
Recently, the United Nations Development Programme (UNDP) has sought to extend the basis on which the standard of living is measured. It has done so by constructing a numerical index (the Human Development Index: HDI) that combines GDP per capita, life expectancy at birth and literacy. Again, leaving aside a few exceptions, HDI has been found to be low in poor countries and high in rich countries (Table 1).
Proximate causes behind differences between the rich and poor worlds
What enables people in the rich world to be so much richer than people in the poor world? It is clear that people in rich countries have better equipment and support: electric drills are more powerful than pick-axes; tractors are superior to ploughs; and modern medicine is vastly more effective than traditional cures. It can be argued, therefore, that the accumulation of physical capital (more accurately, manufactured capital) in the rich world has been a significant contributor to the high standard of living people enjoy there.
Others have noted that people in rich countries are far better educated, implying that they are able to make use of ideas to produce goods that are out of reach for people in countries where large numbers are illiterate. A crude index of education is the proportion of adults (people aged 15 years and above) who are literate, the figure for which today is over 95% in the rich world, but only 58% in the poor world (Table 1). Gender inequalities are considerably greater in the poor than in the rich world. The proportion of adult women who are literate in poor countries is 48%, whereas in the rich world the corresponding proportion is more or less the same as that for men (Table 1). Allied to education is health. Life expectancy at birth in rich countries is now 78 years, whereas it is about 58 years in poor countries. Some 120 children among every 1000 of those under 5 years of age die each year in the poor world; the corresponding figure for rich countries is seven (Table 1). Relatedly, clean water and good hygiene have greatly reduced mortality in rich countries. About a quarter of the population in the poor world suffer from malnutrition, whereas the corresponding figure in rich countries is negligible. As undernutrition and vulnerability to infections reinforce each other, poor nourishment and mortality are related. There is evidence that undernourishment in early childhood affects the development of cognitive faculties. Taken together, the average person in the rich world is capable of supplying work of far higher quality and for many more years than his counterpart in a poor country. Education and health go by the name ‘human capital’. Literature pioneered by Schultz (1974) and Becker (1983) reveals that the accumulation of human capital has been a significant factor behind the high standard of living people in the rich world enjoy today.
Many economists, however, regard the production of new ideas as the prime factor behind economic progress. They suggest that rich countries have become rich because people there have been successful in producing ideas not only for new products (printing press, steam engine, electricity, chemical products, computers), but also for cheaper ways of producing old products (transportation, mining). Of course, education and advances in science and technology combine as an economic force. Primary and secondary education alone cannot take a society that far today. A country where tertiary education is low would not have a population capable of scientific and technological advances or capable of working with the most advanced technology.
Related to this is an issue that has proved to be far more contentious than it should have been: population growth. Even unaided intuition suggests that if numbers grow quickly, the rate at which capital assets must increase would need to be high in order to maintain living standards. If the desire to accumulate physical and human capital is the same in two countries and if rising numbers do not reduce the cost of accumulating that capital, the country where population grows at a slower rate can be expected to enjoy a higher living standard in the long term. Since the mid-1960s, population in what is today the poor world has grown at an average annual rate of 2·4%, while the corresponding figure in today's rich world has been 0·8% (Table 1). This is a large difference. Statistical demographers now at last agree that, controlling for other factors, countries where population increase has been large in recent decades have experienced slow growth in real GDP per capita. High population growth in today's poor countries has also put enormous pressure on their ecology, creating further problems for rural people (Dasgupta 2001), a causality that was explored carefully by Ehrlich, Ehrlich & Holdren (1977).
A country's population growth is affected not only by net reproduction but by net immigration, and also age distribution. In order to isolate net reproduction, it is common practice to work with the fertility rate (more accurately, the total fertility rate, or TFR), which is the number of live births a woman expects to deliver over her life. Suppose parents desire to have a certain number of surviving children. Then the fertility rate should decline once the mortality rate among children under 5 years starts to decline. Demographers have puzzled why reductions in fertility rates in today's poor world have been slower than they had expected. The first known decline in fertility rates in north-western Europe (England and France especially) occurred in the 17th century, when it fell from about 7 to 4. The fertility rate in the rich world today is 1·8 (below 2·1, the figure at which population would stabilize in the long term), whereas it is 3·7 in the poor world (Table 1). Despite a significant decline in child mortality rates, the TFR in a number of countries in sub-Saharan Africa continues to be between 6 and 8. We should ask whether there have been countervailing forces at work to keep fertility rates high in that continent. We should ask, too, whether the resulting population growth has been a factor in its terrible economic performance over the past four decades.
There are clear implications of high fertility rates for women's conditions and their contribution to the economy. In sub-Saharan Africa, extended breastfeeding has been a traditional practice for controlling pregnancies. Among the Kung San nomads of the Kalahari desert, children have been known to be breast-fed until they are 4 years old. Even if we were to ignore such extreme cases, successful reproduction in Africa would involve 2 years of pregnancy and breastfeeding. This means that in societies where female life expectancy at birth is greater than 45 years and the fertility rate is 8, girls can expect to spend more than half their fecund life (e.g. 15–45 years) in pregnancy or nursing; and we have not allowed for unsuccessful pregnancies. Under these circumstances rural women in poor countries are unable to seek employment outside subsistence agriculture.
No economist has ever claimed that there is a single driving force behind economic growth. All would appear to agree that the accumulation of manufactured capital, human capital and the production, diffusion and use of new scientific and technological ideas go together, each contributing positively to the contributions of the others. In the contemporary world an accumulation of, for instance, manufactured capital goods raises real GDP, other things being equal. This enables societies to set aside more of their incomes for education and health, triggering a reduction in both fertility and child mortality. Education increases GDP further, other things being equal, while reduced fertility and child mortality typically lower population growth; which, taken together enable societies to set aside more of their incomes for the production of new ideas. This raises the productivity of manufactured capital, which in turn brings forth further accumulation of manufactured capital and so on, in a virtuous cycle of prosperity. The flipside of this is, of course, a vicious cycle of poverty. The polarization that separates the rich and poor worlds today is a manifestation of those two trajectories. Economists use the terms ‘virtuous’ and ‘vicious’ cycles to characterize polarization (a few of us refer to vicious cycles as ‘poverty traps’); mathematicians describe the poor and rich worlds as being in two different ‘basins of attraction’.
It is possible to discover the relative importance of the various factors responsible for economic growth. No doubt the answer is different in different places and in different periods of history; but five decades ago, Robert Solow devised a way to decompose recorded changes in an economy's real GDP into their measurable sources (Solow 1957). Suppose that over an interval of time a country's real GDP has increased. Solow, and subsequently others, showed how to attribute that growth to increases in labour force participation (population growth; increases in women's employment in paid labour), the accumulation of human skills and manufactured capital, improvements in the quality of machinery and equipment, and so on. Now suppose that when all the contributions made by these factors of production are added together, we find that the sum falls short of real GDP growth. The shortfall can be interpreted as an increase in the overall productivity of the economy's capital assets; by which we mean that more output can be produced now than earlier, even if the amounts of such factors of production as machines and equipment and skills had remained the same. This is a formal way of acknowledging that there has been a general rise in the efficiency with which goods are produced. Economists call that rise growth in ‘total factor productivity’.
The situation arises when people acquire knowledge and make use of it, or when people make better use of what they already know. Hence, economists often refer to growth in total factor productivity as ‘technological progress’; but there are other changes in an economy that could leave an imprint on total factor productivity, such as improvements in the workings of institutions. Growth in total factor productivity may be an ungainly way to convey an idea, but it does reflect the unexplained part of real GDP growth well and is a recognized term in the economics literature.
Since the Second World War, growth in total factor productivity in the rich world has been considerable. It has been estimated, for example, that during 1970–2000 the average annual rate of growth of total factor productivity in the United Kingdom was 0·7%. Economists have estimated that, in contrast, total factor productivity declined slightly in a number of countries in sub-Saharan Africa during that same period.
What do these figures mean? Take the case of the United Kingdom. The country's real GDP grew at an average annual rate of 2·4%, which means about 29% of that growth (i.e. 0·7/2·4) could be attributed to increases in total factor productivity. At a 2·4% growth rate, real GDP in 2000 was twice the real GDP in 1970. Nearly a third of that increase can be attributed to growth in total factor productivity. In contrast, the economies in sub-Saharan Africa, where total factor productivity declined during that period, became less efficient in their use of such factors of production as machines and equipment, skills and labour hours. It is hard to believe that people in those countries systematically forgot technical knowledge they had known in the past; so the decline in factor productivity must have been due to a deterioration in local institutions, precipitated by civil wars and bad governance.
These statistics raise a puzzle. Today's poor countries lie mainly in the tropics, whereas the rich countries are mainly in temperate zones. No doubt the tropics are a breeding ground for many diseases, but they also harbour vast quantities of natural resources (timber, minerals and conditions suitable for the production of spices, fibres, coffee and tea). During the past several centuries, the countries that are rich today have been importing those very resources and products. They accumulated machines, human capital, and also produced scientific and technological knowledge. Why did the poor world not take advantage of their resource endowments to enrich themselves in the same way?
Colonization is a possible answer. Historians have shown that, from the 16th century, European powers have extracted natural resources from the colonies –including cheap (slave) labour – but have mainly invested the proceeds domestically. Of course, one could ask why it is that the Europeans managed to colonize the tropics; why colonization did not take place the other way around? Diamond (1997) has scanned 11 000 years of settled communities to conclude that Eurasia enjoys geographical advantages over the rest of the world that were crucial to the rise of economic activity there. That said, many of the most prominent of those ex-colonies have now been politically independent for decades. During that time real income per head in the rich world has increased over and over again. With the exception of a few striking examples in South and South-East Asia, however, most of the ex-colonies have either remained poor or become poorer still. Why?
North & Thomas (1973) and Landes (1998), among other economic historians, have argued that the rich world is rich today because, over the centuries, it has devised institutions that have enabled people to improve their material conditions of life. This is a deeper explanation. It says that people in rich countries work with superior technologies, are healthier, live longer, are better educated and produce many more productive ideas because they have been able to get on with their lives in societies whose institutions permit – even encourage –the economy-wide accumulation of such factors as production of machines, transport facilities, health, skills, ideas and the fruits of those ideas. The accumulation of productive capital assets is only a proximate cause of prosperity’; the real cause is progressive institutions.
One can also ask how and why in the past people in today's rich countries were able to fashion their institutions in ways that enabled those proximate causes of prosperity to explode there. One can even ask whether it was the institutions or whether it was the enlightened policies of the rulers that were responsible for the explosion. But then, policies are not plucked from the air; they emerge from consultations and deliberations within institutions. Nor is it likely that a policy designed to bring prosperity to a country will actually work unless the institutions there are capable of implementing it.
These dilemmas are of enormous importance for today's poor countries. What institutions should they adopt and what policies should their governments be encouraged to follow? There is little point in embarking on grandiose projects (steel mills, petrochemical plants, land reform, public health programmes, free education) unless a country's institutions have the necessary checks and balances to limit corruption and wastage. This brings us back to our earlier question: how did those institutions that promoted economic growth in today's rich countries become established and flourish? Despite in-depth study by the world's leading economic historians, the matter remains unsettled. There are also theoretical difficulties in answering the question (Dasgupta 2006). In view of the difficulties, it is safest to regard institutions as the explanatory factor when we seek to understand why the rich and poor worlds differ so much in terms of the standard of living.
The effectiveness of an institution depends on the rules governing it and on whether its members obey the rules. The codes of conduct in the civil service of every country include honesty, but governments differ enormously as to its practice. Social scientists have constructed indices of corruption among public officials. One such index is based on the perception private firms have acquired, on the basis of their experience, of the bribes people have had to pay officials in order to conduct business. The index (see Table 1) – which is on a scale of 1 (highly corrupt) to 10 (highly clean) – is less than 3·5 for most poor countries (African countries and Eastern Europe are among the worst) and greater than 7 for most rich countries (Scandinavian countries are among the best). It used to be argued that bribery of public officials helps to raise national income because it lubricates economic transactions. It does so in a corrupt world: if you do not pay up, you do not get to conduct business. But corruption is not an inevitable evil. There are several poor countries where corruption is low. Having to pay bribes raises production costs, so less is produced. Citizens suffer, because the price they have to pay for products is that much higher.
Economists have speculated that government corruption is related to the delays people face in having the rule of law enforced. It is proposed that delays are a way of eliciting bribes to hasten legal processes. To enforce a contract takes 415 days in the poor world, as against 280 days in the rich world. It may be that corruption is also related to government ineffectiveness. To register a business takes 66 days in the poor world, 27 days in the rich world. In poor countries registering property takes 100 days on average, while in rich countries the figure is 50 days. Some economists have suggested that government officials in poor countries create lengthy queues (i.e. government ineffectiveness) so as to elicit bribes from applicants if they want to jump those queues (i.e. corruption).
How do government corruption, ineffectiveness and indifference to the rule of law translate into the kind of macroeconomic statistics we have been studying here? They leave their imprint on total factor productivity. Other things being equal, a country whose government is corrupt or ineffective, or where the rule of law is not respected, is a country whose total factor productivity is lower than that of a country whose government suffers from fewer of those defects. Some scholars call these intangible but quantifiable factors ‘social infrastructure’, others call them ‘social capital’.
Sustainable economic development
Economic growth is a good thing. It may not buy happiness, but it usually purchases a better quality of life. Table 1 showed that growth in real GDP per capita comes hand-in-hand with improvements in the way people are able to live. But can economies grow indefinitely, or are there limits to growth? To put the question in a more contemporary form, is growth in real GDP compatible with sustainable economic development?
The above question is several decades old. If discussions on it continue to be shrill, it is because two opposing empirical perspectives have shaped them. On one hand, if we look at specific examples of natural resources (fresh water, ocean fisheries, the atmosphere as a carbon sink; more generally, ecosystems), there is strong evidence that the rates at which they are currently being utilized are unsustainable. During the 20th century the world population grew by a factor of 4 to more than 6 billion; industrial output increased by a multiple of 40 and the use of energy by 16; methane-producing cattle populations grew in pace with the human population; fish catch increased by a multiple of 35; and carbon and sulphur dioxide emissions by a factor of 10. The application of nitrogen to the terrestrial environment from the use of fertilizers, fossil fuels and leguminous crops is now at least as great as that from all natural sources combined. Ecologists have estimated that 40% of the net energy created by terrestrial photosynthesis is currently being appropriated for human use. These figures put the scale of our presence on Earth into perspective and reveal that humanity has created an unprecedented disturbance in Nature in a brief period of a century or so.
On the other hand, it is often argued that just as earlier generations in the West invested in science and technology, education and machines and equipment so as to bequeath to the present generation the ability to achieve high income levels, the current generation is now in turn making investments that will ensure still higher living standards in the future. It has also been argued that the historical trend in the prices of marketed natural resources, such as minerals and ores, has been so flat that there is no reason for alarm. Economic growth has allowed more people to have access to potable water and enjoy better protection against water- and airborne diseases. The physical environment inside the home has improved beyond measure with economic growth: cooking in the Indian subcontinent continues to be a major cause of respiratory illnesses among women. Moreover, natural resources can be so shifted around today that dwindling resources in one place can be met by imports from another. Intellectuals and commentators use the term ‘globalization’ to imply that location per se does not matter. This optimistic view emphasizes the potential of capital accumulation and technological improvements to compensate for environmental degradation. It says that economic growth, even in the form and shape it has taken so far, is compatible with sustainable development. This probably explains why contemporary societies are obsessed with cultural survival and are on the whole dismissive of any suggestion that we need to find ways to survive ecologically.
Broadly speaking, environmental scientists and activists hold the former view, while economists and economic commentators maintain the latter. It is no doubt banal to say that our economies are built in and on Nature, but it is a remarkable fact that most contemporary accounts of long-term economic possibilities facing humanity do not include natural capital. Nature does not feature in accounts of the macroeconomic history of nations because it does not appear in official publications of the vital statistics of nations. The extraction of minerals and fossil fuels is included (but not depreciated) in modern national accounts, but with the exception of agricultural land, natural capital makes very little appearance. In economics textbooks, Nature's services appear, in passing, only to be side-stepped. That is evident in the literature on the theory and empirics of long-term economic growth and the economics of poverty.
natural capital: classification
Natural capital is of direct use in consumption (e.g. fisheries); of indirect use as inputs in production (e.g. oil and natural gas); or of use in both (e.g. air and water). The value of a resource is often derived from its usefulness either as a source of food, or as an essential actor in ecosystems (e.g. a keystone species); but there are resources whose value is aesthetic (e.g. places of scenic beauty), or intrinsic (e.g. primates, blue whales, sacred groves), or a combination of all three (i.e. biodiversity). The worth of a natural resource could be based on what is extracted from it (e.g. timber), or on its presence as a stock (e.g. forest cover), or on both (e.g. watersheds).
Interpreting natural capital in this inclusive way allows us to add ecosystems to our list of capital assets. The services they produce include maintaining a genetic library, preserving and regenerating soil, fixing nitrogen and carbon, recycling nutrients, controlling floods, filtering pollutants, assimilating waste, pollinating crops, operating the hydrological cycle, and maintaining the gaseous composition of the atmosphere. A number of these services have a global reach (the atmosphere), but many are localised (microwatersheds) (see Daily 1997).
Pollutants are the reverse of resources. In general terms, ‘resources’ are ‘goods’ (in many situations they are the sinks into which pollutants are discharged), while ‘pollutants’ (the degrader of resources) are ‘bads’. If, over a period of time, the discharge of pollutants into a sink exceeds the latter's assimilative capacity, the sink collapses. Pollution is thus the reverse of conservation. In what follows, the terms ‘natural resources’ and ‘environment’ are interchangeable.
In order to demonstrate that economics can be integrated into the environmental sciences, two topical issues are discussed. The first is the subject of an acrimonious debate between those who favour free trade and those who are opposed to it, on the grounds that it often hurts the poorest in the world's poorest regions. The second is the belief that because the economic effects of carbon dioxide emissions into the atmosphere are likely to be felt by a future generation or two, we need not do anything now about climate change.
Trade expansion and the environment
There should be little doubt today that, other things being equal, freeing trade enables economies to grow faster. A large body of empirical work testifies to that. There is some evidence too that the poor, as a group, also enjoy the fruits of faster growth. However, as the environmental consequences of economic growth are rarely assessed, the case for freeing trade remains unclear. If those consequences are detrimental to many of the poorest in society, there is room for discussion about the merits of freeing trade without, at the same time, taking precautionary measures. Here is an example of how trade expansion can hurt.
An easy way for governments in poor countries that are richly covered in forests to earn revenue is to issue timber concessions to private logging firms. Imagine that logging concessions are awarded for the upland forest of a watershed. Deforestation contributes to an increase in siltation and the risk of floods downstream. If the law recognizes the rights of those who are harmed, the logging firm would have to compensate downstream farmers and coastal fishermen. But there is a gulf between the law and the enforcement of law. When the cause of damage is miles away, when the timber concession has been awarded by the state, and when the victims are a scattered group of poor farmers and coastal fishermen, the issue of a negotiated outcome usually does not arise. It can even be that those who are harmed do not know the underlying cause of their deteriorating circumstances. If the logging firm is not required to compensate those suffering damage, the private cost of logging is less than the true cost of logging, the latter being the sum of the costs borne by the logging firm and all who are adversely affected. From the country's point of view, timber exports are underpriced because there is excessive deforestation upstream. There is an implicit subsidy on the export, paid for by people who are evicted from the forest and by people downstream. The subsidy is hidden from public scrutiny; but it amounts to a transfer of wealth from the exporting country to those that import the timber. Some of the poorest people in a poor country would be subsidizing the incomes of the average importer in a rich country.
Unfortunately, the magnitude of those subsidies has not been estimated. International organizations have the resources to undertake such studies; but, to the best of my knowledge, they have not done so. The example should not be used to argue against free trade, but it can be used to caution anyone who advocates free trade while ignoring its environmental impacts.
Discounting climate change
The second example concerns the emission of greenhouse gases and the global climate change it is inducing, the subject of continuing study by the International Panel on Climate Change (IPCC). Carbon concentration in the atmosphere today is above the highest level reached in the past 600 000 years (not including the concentration of methane, another greenhouse gas). If current trends in carbon emissions continue, its concentration is expected to reach 500 parts per million (p.p.m.), nearly twice the preindustrial level (i.e. ce 1800) by the middle of this century, and could reach as high a figure as 750 p.p.m. (which is nearly three times the preindustrial level) by 2100. A doubling of present-day carbon concentration is expected to give rise to an increase in the mean global atmospheric temperature by 3–7 °C. With a trebling of concentration, it could rise by 6–11 degrees. The scale and speed of such a change are so great that the potential economic costs of global climate change will, in all likelihood, be huge. Nevertheless, when in 2004 eight eminent economists were invited to Copenhagen to offer advice on how the world community could most usefully spend $50 billion over a 5-year period, they placed climate change at the bottom of their list of 10 alternatives.
Their reasoning was based on discounting future costs and benefits at a positive rate. Reducing global carbon emissions or investing in technologies for carbon sequestration would involve costs now, but the benefits from averting economic disruptions would be enjoyed only 50–100 years from now. Long-term interest rates on government bonds in the United States have been 3–5% per year. When economists evaluate public projects, they typically use the figure to discount future benefits and costs, regarding it as the ‘opportunity cost of capital’, the term being applied to the rate of interest that could be earned by investing in government bonds rather than in the project whose benefits and costs are being evaluated. At discount rates of 3–5%, however, consumption benefits in the distant future appear minute today. If one discounts at 4% per year, a dollar's worth of additional consumption benefits 100 years from now would be worth less than 3 cents today; which is another way of saying that as a price for giving up $1 worth of consumption today, one would demand that more than $30 worth of consumption benefits be made available 100 years from now. A number of economic models of climate change have shown that if an annual discount rate of, for instance, 4% is used, the costs (which are negative benefits) are greater than the sum of the discounted benefits from curbing net carbon emissions. Doing something about climate change now, the calculations imply, would be to throw money away in a comparatively bad project.
Should the global community discount future consumption benefits at a positive rate? There are two reasons why it may be reasonable to do so. First, a future benefit would be of less value than that same benefit today if the global community is impatient to enjoy the benefit now. Impatience is a reason for discounting future costs and benefits at a positive rate. Secondly, considerations of justice and equality demand that consumption per capita should be smoothed across the generations. So, if future generations are likely to be richer than us, there is a case for valuing an extra dollar's worth of their consumption less than an extra dollar's worth of our consumption, other things being equal. Rising consumption per capita provides a second justification for discounting future costs and benefits at a positive rate.
Philosophers have argued that societal impatience is ethically indefensible, because it favours policies that discriminate against future generations merely on the grounds that they are not present today. Once we accept their argument, we are left with only the second reason for discounting future costs and benefits. But if rising per capita consumption provides the global community with a reason for discounting future consumption benefits at a positive rate, declining per capita consumption would provide it with a reason for discounting future consumption benefits at a negative rate.
Economists use positive discount rates in their models of climate change because the models assume that global consumption per head will continue to grow over the next 150 years and more even if net emissions of greenhouse gases follow current trends; which is to assume that climate change poses no serious threat to the future. But an increase in the mean global temperature by 3–5 °C would take the biosphere into a climatic zone that has not been visited in millions of years on Earth. The possible consequences of such changes to our productive base are so huge that it is not to be alarmist to question forecasts of continual economic growth even after Earth enters that zone. If nothing substantial is done today to discover ways to sequester carbon and to find alternatives to fossil fuels as sources of energy, there is a sizeable chance that global consumption per head, suitably weighted across regions and income groups, will decline. This may occur, for example, due to a big increase in the frequency of extreme weather events, more severe droughts in the tropics, the emergence of new pathogens and degradation of vital ecosystems. A negative rate to discount future consumption benefits is then appropriate. Notice, however, that applying a negative rate amplifies benefits in the distant future when viewed from the present; it does not attenuate them.
Empirical evidence from societal and personal choices suggests that a society ought to discount future consumption benefits at a rate of about three times the percentage rate of change of consumption per capita. Imagine that carbon emissions follow their current trends (which is often called ‘business as usual’). Consider a scenario in which global consumption per capita increases at an annual rate of 0·5% for the next 50 years and declines at 1% per year for the following 100 years. Under that scenario the global community ought to discount future consumption benefits at 1·5% per year for the next 50 years (3 × 0·5) and at –3% for the subsequent 100 years (3 × –1). A simple calculation now shows that a dollar's worth of additional consumption 150 years from now is worth $9 of additional consumption today. To put it another way, the global community should be willing to forego $9 worth of additional consumption today for an extra dollar's worth of consumption benefits 150 years in the future. The calculation reverses the message that has been conveyed by economic models of climate change.
There should be little doubt that private investors would be using a positive rate to discount their personal earnings even under the above scenario. They would be doing so because the interest rate offered by commercial banks on deposits would most probably remain positive; but there is no contradiction here. Under ‘business as usual’, the atmosphere is an open access resource. As long as people are free to emit carbon dioxide, there will be a wedge between private rates of return on investment and the rates that the world community ought to use to discount collective costs and benefits. The former could be positive even while the latter is negative. That wedge is a reason for controlling carbon emissions into the atmosphere and bringing the two rates closer to each other; it is not a reason for claiming that the problem of global climate change should be shelved for the future.
gdp and the productive base
A famous 1987 report by an international commission (widely known as the Brundtland Commission Report) defined sustainable development as ‘... development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ (World Commission 1987). In this reckoning, sustainable development requires that relative to their populations each generation should bequeath to its successor at least as large a productive base as it had itself inherited. Note that the requirement is derived from a relatively weak notion of intergenerational justice. Sustainable development demands that future generations have no less of the means to meet their needs than we do ourselves; it demands nothing more. But how is a generation to judge whether it is leaving behind an adequate productive base for its successor?
It is easy to see why focusing on GDP is inadequate. An economy's productive base is its stock of capital assets and institutions. By capital assets we now mean not only manufactured capital, human capital and knowledge, but also natural capital. It is evident that an economy's productive base will shrink if its stock of capital assets depreciates and its institutions are not able to improve efficiency sufficiently to compensate for that depreciation. GDP is an acronym for gross domestic product. The word ‘gross’ means that GDP ignores the depreciation of capital assets. It is certainly possible for a country's productive base to grow while its GDP increases, which is no doubt a path of economic development we all would like to follow; but it is also possible for a country's productive base to shrink during a period when GDP grows (Table 2). The problem is that the shrinking would not be immediately obvious. If the productive base continues to shrink, economic growth will sooner or later stop and reverse sign. The standard of living will then decline, but this state of affairs would not have been anticipated. So, growth in GDP per head can encourage us to think that the economy is healthy when it is not. Similarly, it is possible for a country's HDI to increase even while its productive base shrinks (Table 2). This means that HDI, too, can mislead.
Table 2. The economic progress of nations. The first column consists of refinements of the Hamilton & Clemens (1999) estimates of average inclusive investment as a proportion of GDP, expressed as percentages. The second column gives the average annual population growth rate. The third column gives estimates of annual growth rates of total factor productivity, which we are interpreting here as the annual percentage rate of change in a combined index of knowledge and institutions. The figures in the first three columns have been used to estimate the annual percentage rate of change in the productive base per capita, given in the fourth column. The places in question are sub-Saharan Africa, Bangladesh, India, Nepal and Pakistan (all poor countries); China (a middle-income country); and the United Kingdom and United States (both rich countries)
market prices as signals of resource scarcity
One could counter that a fixation on GDP or HDI should not prevent anyone from scrutinizing prices. One could even argue that if natural resources really were becoming more scarce, their prices would have risen, and that would have signalled that all is not well. But if prices are to reveal scarcities, markets must function well. For many natural resources, markets not only do not function well, they do not even exist (‘missing markets’). In some cases they do not exist because relevant economic interactions take place over large distances, making the costs of negotiation too high (e.g. the effects of upland deforestation on downstream farming and fishing activities); in other cases they do not exist because the interactions are separated by large temporal distances (e.g. the effect of carbon emission on climate in the distant future, in a world where forward markets do not exist because future generations are not present today to negotiate with us). Then there are cases (the atmosphere, aquifers, the open seas) where the migratory nature of the resource prevents markets from existing – they are ‘open access’ resources; while in others, ill-specified or unprotected property rights prevent markets from being formed (e.g. mangroves and coral reefs), or cause them to function wrongly even when markets do form (e.g. those who are displaced by deforestation are not compensated). The side effects of human activities that are undertaken without mutual agreement are called ‘externalities’ by economists. Our dealings with Nature are full of externalities. The examples suggest that the externalities involving the environment are mainly negative, implying that the private costs of using natural resources are less than their social costs. Being underpriced, the environment is over-exploited. In such a situation the economy could enjoy growth in real GDP and improvements in HDI for a long spell even while its productive base shrinks. As proposals for estimating the social scarcity prices of natural resources remain contentious, economic accountants ignore them and governments remain wary of taxing their use.
the environment: is it a luxury or necessity?
It is not uncommon to regard the environment as a luxury good, as in the thoughts expressed in prominent newspapers, that ‘economic growth is good for the environment because countries need to put poverty behind them in order to care’ (Independent, 4 December 1999), or that ‘... trade improves the environment, because it raises incomes, and the richer people are, the more willing they are to devote resources to cleaning up their living space’ (Economist, 4 December, 1999); but in the poor world the environment is an essential factor of production. When wetlands, inland and coastal fisheries, woodlands, forests, ponds and grazing fields are damaged (owing to agricultural encroachment, nitrogen overload, urban extensions, the construction of large dams, resource usurpation by the state, and so on), it is the rural poor who suffer most. Frequently, there are no alternative sources of livelihood for them. In contrast, for rich eco-tourists or importers of primary products, there is something else, often somewhere else; which means that there are alternatives. Degradation of ecosystems is like the depreciation of roads, buildings and machinery, but with two large differences: (1) it is frequently irreversible (or at best the system takes a long time to recover) and (2) ecosystems can collapse abruptly, with little prior warning. Imagine what would happen to a city's inhabitants if the infrastructure connecting it to the outside world was to break down without notice. Vanishing water holes, deteriorating grazing fields, barren slopes and wasting mangroves are spatially confined instances of corresponding breakdowns among the rural poor in poor countries. Modern economics helps to explain how an ecological collapse, such as those that have been experienced in recent years in the Horn of Africa and the Darfur region of Sudan, can trigger rapid socio-economic decline (Dasgupta 2003).
Economic development is sustainable if, relative to its population, a society's productive base does not shrink. How can one tell whether economic development has been sustainable? We have noted that neither GDP nor HDI will tell us. So what index would be appropriate? A society's productive base is its institutions and capital assets. As we are interested in estimating the change in an economy's productive base over a period of time, we need to know how to combine the changes that take place in its capital stocks and in its institutions. Intuitively, it is clear that we have to do more than just keep a record of capital assets (so many additional pieces of machinery and equipment; so many more miles of roads; so many fewer square miles of forest cover; and so on). An economy's productive base declines if the decumulation of assets is not compensated by the accumulation of other assets and expands if the decumulation of assets is (more than) compensated by the accumulation of other assets. The ability of an asset to compensate for the decline in some other asset depends on technological knowledge (e.g. double-glazing can substitute for central heating up to a point, but only up to a point) and on the quantities of assets the economy happens to have in stock (e.g. the protection trees provide against soil erosion depends on the existing grass cover). Clearly, however, capital assets differ in their ability to compensate for one another. Those abilities are the values we would wish to impute to assets. We need to have estimates of those abilities. This is where an asset's social productivity becomes an item of interest; the net increase in social wellbeing that would be enjoyed if an additional unit of that asset were made available to the economy, other things being equal. Putting it another way, the social productivity of an asset is the capitalized value of the flow of services an extra unit of it would provide society. An asset's value is simply its quantity multiplied by its social productivity.
As we are trying to make operational sense of the concept of sustainable development, we must include in the term ‘social wellbeing’ not only the wellbeing of those who are present, but also of those who will be here in the future. There are ethical theories that go beyond a purely anthropocentric view of Nature, by insisting that certain aspects of Nature have intrinsic value. This concept of social well-being includes intrinsic values if required. However, an ethical theory on its own will not be enough to determine the social productivities of capital assets, because there would be nothing for the theory to act upon. We also need descriptions of states of affairs. To add a unit of a capital asset to an economy is to perturb that economy. In order to estimate the contribution of that additional unit to social wellbeing, we need a description of the state of affairs both before and after the addition has been made. In short, measuring the social productivities of capital assets involves both evaluation and description.
Imagine that we have adopted a conception of social wellbeing (e.g. by adding the wellbeings of all people) and that we have an economic scenario of the future in mind (e.g. business as usual). In principle, we can now estimate the social productivity of every capital asset. One can do that by estimating the contribution to social wellbeing (the evaluative part of the exercise) an additional unit of each capital asset would make, other things being equal (the descriptive part of the exercise). Economists call social productivities of capital assets their ‘shadow prices’, to distinguish them from prices that are observed in the market. Although shadow prices pertain to commodities generally, not only to capital assets, we focus here on capital assets.
Shadow prices reflect the social scarcities of capital assets. In the world as we know it, estimating shadow prices is a formidable problem. There are ethical values we hold that are probably impossible to commensurate when they come up against other values that we also hold. This does not mean that ethical values do not impose bounds on shadow prices; they do; which is why the language of shadow prices is essential if we wish to avoid making sombre pronouncements about sustainable development that amount to saying nothing. Most methods that are currently deployed to estimate the shadow prices of ecosystem services are crude, but deploying them is a great deal better than doing nothing to value them.
The value of an economy's stock of capital assets, measured in terms of their shadow prices, is its inclusive wealth. The term ‘inclusive’ serves to remind us not only that natural capital has been included on the list of assets, but also that externalities have been taken into account in valuing the assets. Inclusive wealth is the sum of the values of all capital assets. It is a number that could be expressed in international dollars, for example.
We can summarize by saying that an economy's inclusive wealth plus institutions constitute its productive base. If we now wish to determine whether a country's economic development has been sustainable over a period of time, we have to estimate the changes that took place over that period in its inclusive wealth and its institutions, relative to population, of course. Earlier, we noted that changes in knowledge and institutions over time are reflected in movements in total factor productivity. So we break up the procedure for estimating changes in an economy's productive base relative to population during any period of time into five stages, as follows.
First, estimate the value of changes in the amounts and compositions of manufactured capital, human capital and natural capital – which we will call ‘inclusive investment’. (If inclusive investment is found to be positive, we may conclude that manufactured capital, human capital, and natural capital, taken together, grew over the period.) Secondly, estimate the change in total factor productivity. Thirdly, transform the two figures in a way that enables us to calculate the effects of the two sets of changes on the productive base. Fourthly, combine the two resulting estimates into a single number that can be taken to reflect the change that took place in the economy's productive base. Fifthly, make a correction for demographic changes to arrive at an estimate for the change that took place in the economy's productive base relative to population.
These five steps can be applied to forecasts of the future. The procedure outlined here is essential for anyone who wants to know whether the economic pathways we are currently pursuing can be expected to lead to sustainable development.
has economic development in recent decades been sustainable?
Hamilton & Clemens (1999) have estimated inclusive investment in different countries during the past few decades. They added net investment in human capital to existing country-wide estimates of investment in manufactured capital, and then substracted disinvestments in natural capital from that sum (step 1 above). The authors used official estimates of net national saving as proxies for net investment in manufactured capital. For estimates of investment in human capital, they used expenditure on education as a proxy. To quantify disinvestments in natural capital, they considered net changes in the stocks of commercial forests, oil and minerals and the quality of the atmosphere in terms of its carbon dioxide content. Oil and minerals were valued at their market prices minus extraction costs. The shadow price of global carbon emission into the atmosphere is the damage caused by bringing about climate change. That damage was taken to be $20 per tonne, which is, in all probability, a serious underestimate. Forests were valued in terms of their market price minus logging costs. Contributions of forests to ecosystem functions were ignored.
The list of natural resources in Hamilton & Clemens (1999) is very incomplete. It does not include water resources, fisheries, air and water pollutants, soil and ecosystems. The notion of human capital they deployed is inadequate because health does not enter the calculations, and their estimates of shadow prices are extremely approximate. Nevertheless, one has to start somewhere, and theirs is a first pass at what is an enormously messy enterprise. In another study (Dasgupta 2001), the Hamilton–Clemens estimates of inclusive investment was used to determine whether economic development in some of the major countries and regions in the rich and poor worlds has been sustainable in recent decades. Here, I study figures published recently by a group of ecologists and economists (Arrow et al. 2004), who refined those earlier estimates. Table 2 reports data that are in turn a refinement of Arrow et al. 2004). These data form a crude beginning to the study of sustainable development; but they are a start, and they reflect progress.
As an example, during 1970–2000 inclusive investment as a proportion of GDP was 8·8% annually in Pakistan. Total factor productivity increased at an annual rate of 0·4%. As both figures are positive, we can conclude that Pakistan's productive base was larger in 2000 than it had been in 1970; but Pakistan's population grew at a high 2·7% rate annually. The fourth column in Table 2 shows that Pakistan's productive base per capita declined in consequence, at an annual rate of 0·7%, implying that in 2000 it was about 80% of what it was in 1970.
In contrast, consider the United States. Inclusive investment as a share of GDP was 8·9% per year, which is only slightly larger than Pakistan's figure. Growth in total factor productivity (an annual 0·2%) was even lower than Pakistan's; but the population grew only at 1·1% per year, meaning that the productive base per capita of the United States grew at an average annual rate of 1%. Economic development in the United States was sustainable during 1970–2000, while in Pakistan it was unsustainable.
Interestingly, if one had judged their economic performances in terms of growth in GDP per capita, one would have obtained a different picture. As the fifth column of Table 2 shows, Pakistan grew at a respectable 2·2% rate per year, while the United States grew at only 1·1% per year. If one now looks at the sixth column, one will find that the United Nations’ Human Development Index (HDI) for Pakistan improved during the period. Movements in HDI tell us nothing about sustainable development.
The striking message of Table 2, however, is that during 1970–2000 economic development in all the poor countries included in the data set was either unsustainable or barely sustainable. For sub-Saharan Africa this was no surprise. Its inclusive investment was negative, implying that the region disinvested in manufactured, human and natural capital, taken together, at 2·1% of GDP. Population grew at 2·7% per year and total factor productivity barely advanced (annual growth rate: 0·1%). Even without performing any calculation, we should suspect that the productive base per capita in sub-Saharan Africa declined. The table confirms that it did, at 2·9% annually. The fifth column of numbers shows that GDP per capita in sub-Saharan Africa remained fairly constant; but the region's HDI showed an improvement, confirming once again that studying movements in HDI enables us to say nothing about sustainable development.
Pakistan is the worst performer in the Indian subcontinent, but the remaining countries in the region were little better when judged in terms of sustainable development. Inclusive investment in each country (Bangladesh, India, and Nepal) was positive, as was growth in total factor productivity. The two together imply that the productive base expanded in each country. However, population growth was so high that the productive base per capita just about grew – at annual percentage rates of 0·1, 0·4 and 0·6, respectively. Even these figures are most likely to be overestimates. The list of items used by Hamilton & Clemens (1999) in order to estimate inclusive investment did not include soil erosion and urban pollution, both of which are thought by experts to be problematic in the Indian subcontinent. Moreover, the human desire to reduce risk, mentioned earlier, implies that downside risks of natural capital degradation ought to be given a higher weight than a corresponding chance that things will turn out to be better than expected. So, if we allow for risk aversion, estimates of inclusive investment would be lowered. One cannot help suspecting that economic development in the Indian subcontinent was unsustainable during 1970–2000 but this is not apparent from the figures for GDP per capita and HDI. The former grew in each country in the region and the latter improved.
Inclusive investment in China was 22·7% of GDP, a very large figure in the sample of countries in Table 2. Growth in total factor productivity was a high 3·6% annually. Population grew at a relatively low 1·4% annual rate. We should not be surprised that China's productive base per capita expanded, in fact, at 7·8% annually. Per capita GDP also grew at an annual rate of 7·8%, and HDI improved. In China, GDP per capita, HDI and the productive base per head moved parallel to one another.
There is little to comment upon with regard to the United Kingdom and the United States. Both are rich, mature economies. Inclusive investment during 1970–2000 was modest, but then population growth was also low. Growth in total factor productivity was low. Although the figures imply that the productive base per capita expanded in both countries we should be circumspect because, as noted earlier, Hamilton & Clemens (1999) costed carbon emissions at too low a rate. GDP per capita increased in both countries and HDI also improved.
The figures in Table 2 are rough estimates, but they show how accounting for natural capital can make a substantial difference to our conception of the development process. Table 2 deliberately includes conservative assumptions regarding the degradation of natural capital. For example, a price of $20 per tonne of carbon in the atmosphere is almost certainly a good deal below its true shadow price (or cost). If we were to take, instead, the shadow price to be the not unreasonable figure of $50 per tonne, all the countries in Table 2, with the exception of China, would show a decline in their productive base per capita during 1970–2000. The message is sobering: over the past three decades, sub-Saharan Africa (home to 750 million people today) has become poorer if judged in terms of its productive base per capita; and economic development in the Indian subcontinent (home to over 1·4 billion people today) and in the United Kingsom and United States was either unsustainable or just barely sustainable. That said, it would be wrong to conclude that people in poor countries should have invested more in their productive base by consuming less. The production and distribution of goods and services in the poor world are highly inefficient. It would be wrong to regard consumption and investment in the productive base there as competing for a fixed quantity of funds. Better institutions would enable people in the poor world to both consume more and invest more.
What have we learnt?
When contemporary economists neglect to include ecological concerns in their analysis, the fault lies not in modern economics, but in my tribe. In this lecture I have tried to show that ecological truths can be introduced quite simply into economic reasoning. I have reported work conducted by a small group of ecologists and economists who have developed the correct way to study economic progress – whether the study is for descriptive or for prescriptive purposes. The macroeconomic history of nations looks very different when Nature is included as a capital asset in production activities. The tentative conclusions I draw from the evidence from a sample of rich and poor nations are:
- 1High population growth in the world's poorest regions (South Asia and sub-Saharan Africa) has been an obstacle to the achievement of sustainable economic development there. There are good reasons for believing that people in those regions are on average less wealthy now than they were 35 years ago. Relatedly,
- 2When population growth is taken into account, the accumulation of manufactured capital, knowledge and human capital (health and education) has not compensated for the degradation of natural capital in South Asia and sub-Saharan Africa and, in all probability, even in the United Kingdom and United States.
- 3China is possibly an exception to (1) and (2).
The conclusions are very tentative and there is much further work to be undertaken in understanding economic change. It is high time that growth economists, demographers, governments and international agencies took Nature seriously.