Poverty, Economic Growth, Deprivation, and Water: The Cases of Cambodia and Vietnam

Transcription

1 Olli Varis Poverty, Economic Growth, Deprivation, and Water: The Cases of Cambodia and Vietnam Poverty reduction decorates all development agendas, but the complexity of the poverty issue is too often hidden behind simplistic indicators and development goals. Here, a closer look is taken at the concepts of deprivation and vulnerability as outcomes of poverty. Deprivation leads typically to social exclusion and marginalization; such groups are particularly weak in getting themselves out of poverty by self-help, and economic growth does not trickle down to these people. When looking at the connections between poverty reduction and economic growth, special emphasis should be put on the differences between modern and more traditional sectors: development of the modern sector should not marginalize and exclude those dependent on more traditional livelihoods. Two case studies The Tonle Sap area, Cambodia, and the Mekong Delta, Vietnam reveal that investment in education, empowerment of small-scale entrepreneurship and other means of microeconomic environment, along with good governance, infrastructure, and income distribution can ensure that economic growth includes the poorer echelons of society. POVERTY REDUCTION AND ECONOMIC DEVELOPMENT Should the priority of water-resources policy in the Lower Mekong River Basin be environmental conservation, maximized economic growth, or reduction of poverty as much as possible? The needs are huge in all of these aspects. Cautious scrutiny and compromise-building is needed because these three interlinked facets of development very easily collide in conditions such as those of the Mekong Basin. Here, the relation between economic growth and poverty reduction in the Mekong context is scrutinized by elaborating different concepts and definitions of poverty and comparing them to the conditions in the Mekong countries. Two case studies, the Tonle Sap area, Cambodia, and the Mekong Delta, Vietnam, are presented. Poverty and Wealth In macroeconomics, the basic premise of poverty alleviation is economic growth. Without growth, there is not more to distribute to the poor than before. Whereas this is basically correct, it tells only a part of the story. In too many low-income countries, the growth of the Gross National Income (GNI) means a widening income gap and polarization of the society as well as uprooting of uneducated people to urban shantytowns. Poverty reduction has found its way onto almost all national and international development agendas, including the key agendas related to the development of the Lower Mekong Basin (1). Roughly one-fifth of mankind is typically classified as being poor, but what actually is meant by poverty? Intuitively, within a familiar context, the concept of poverty should not be very unclear. In reality, however, the issue is far more multifaceted and complicated. Accordingly, the contemporary poverty-related literature is rich and highly inconsistent in approaches to analysis, alleviation, and definitions of poverty. The concept varies across scientific disciplines as well as cultures and has changed markedly over time (2, 3). Is wealth relative or absolute? Are some groups of people rich because some others are poor? At least the former ones have much power over the latter ones. Because the question of poverty is very difficult and many-sided, the answer is not as easy as the somewhat provocative previous statement might suggest. It is partly true, but poverty has definite dimension, which is more absolute, related to the capability of an individual or community to meet their basic personal needs, and to make their life better. There is not a single country where all people are equally wealthy. There have been several attempts to reach such a society, but so far, all have failed. Uneven wealth distribution is a reality everywhere, both inside and among countries. One generation ago, the richest 20% of the mankind were around 30 times richer than the poorest 20%, but now they are over 50 times richer (4). This fact has many reasons that raise fundamental moral questions, particularly at the global level. In many Asian countries, the celebrated Kuznets (5, 6) theory seems to be valid. It says that the income disparities grow with economic development, until a middle-income level is reached, and decreases thereafter (Fig. 1). The Mekong Basin countries have a wider gap between the rich and the poor, but it is not as wide as in many other Southeast Asian countries. Cambodia s inequality is somewhat higher than in the other countries within a comparable GNI level. If the theory of Kuznets turns out to be correct in Southeast Asia in the coming decades, there is a considerable risk that the social inequalities will grow in countries such as Laos, Cambodia, and Vietnam. Cornia et al. (7) analyzed the development of inequality in 73 nations throughout the world and concluded that in two-thirds of them, inequalities are growing, and inequalities are decreasing in only one out of six countries. As a positive result of this analysis, it is possible to conclude that given the stable and strong economic progress in China, South Asia, and Southeast Asia, and the fairly modestly growing inequalities, these regions Figure 1. Richest to poorest ratios: Ratios of the income levels of the richest 20% and the poorest 20% in selected Asian countries. These ratios are also related to the GNI per capita (4). Ó Royal Swedish Academy of Sciences

2 Figure 2. Macroeconomic view of poverty reduction: Macroeconomists emphasize economic growth and redistributive effects in poverty reduction but tend to forget the crucial role of local activities and empowerment. have a reasonable capacity to reduce their poverty. They are in this respect perhaps better off than most developing parts of the world. In economic terms, the practical prerequisite to poverty alleviation is economic growth (Fig. 1). Without growth, there is not more to distribute to the poor than before. The hot perennial theme for discussion is the dilemma between economic growth and income distribution within an economy. Redistributive policies were taken as self-evident in the west over decades, yet the boom of liberalistic thoughts in the 1990s, and particularly in the early 2000s, has vitalized the myth of income- and wealth distribution neutral growth (7, 8). The liberalistic trickle-down argument postulates that economic growth is the most efficient way to reduce poverty since the increasing wealth trickles down through the whole society and benefits also the poor. Redistribution of wealth and income distorts the economy, and growth declines. Growth is typically assumed to be distribution-neutral, meaning that growth benefits all equally. Others claim that this is far too slow an approach to poverty eradication. This view has a strong support from the Kuznets theory and innumerable other empirical analyses (c.f. Figs. 1 and 2). Wealth and income should be redistributed to the poorer echelons of the society for many reasons: Political, social, and economic stability are increased, fertility rate tends to go down along with better schooling possibilities and other social conditions that follow from redistributive policies and, above all, development of local markets (9). The Mekong region has seen a sharp split between communist, centrally planned societies and market-oriented economies in the past few decades. Thailand has followed the latter model, while the other countries have had periods with the dominance of the former one. China had loosened its communist policies already by the early 1980s under Deng Xiaoping s famous policy that allowed people to become rich. In Cambodia, the end of the communist era came in the early 1990s with the reestablishment of the Kingdom of Cambodia. Vietnam s Doi Moi policy has been somewhat equivalent to the Chinese approach, but it was initiated one decade later. In Vietnam, recent times have witnessed a strong liberalization of markets and the private sector. At the same time, social services have not totally been able to keep up with the economic development, and income gaps have been growing steadily. Yet, the rapid economic development, particularly in Thailand, Vietnam, and China, has allowed the rapid reduction of income poverty (1, 4). Poverty: Indicators and Beyond Since poverty has both an absolute and a relative dimension, its definitions include both these concepts. Another axis to the diverging definitions comes from the tradition of economics as science: the split to macro- and micro-oriented views and weightings is evident in many analytical approaches and definitions throughout economics. Many development agencies are in favor of using income indices such as one or two dollars a day. If an individual has a daily income less than such a threshold, he is classified as poor. The power and the weakness of this indicator come from its extreme simplicity. It is a straightforward tool in comparative studies but provides only very limited information about the individual s ability and preferences to access basic goods and services. Another class of poverty indicators includes those that measure the ability of humans to meet their basic needs. Besides an array of diet-related indicators, examples of these social indicators include population without sustainable access to improved domestic water source, children under weight for age, probability at birth of not surviving to age 40, and adult illiteracy rate. In fact, the United Nations Development Programme uses a combination of these four indicators as its widely used Human Poverty Index (10). Maxwell (3) provided a comprehensive summary of commonly used poverty indicators. Their suitability was analyzed within the context of the Lower Mekong Basin (11), and results show that, whereas numerical indicators such as those just mentioned are illustrative in international, macrolevel comparisons, the poverty-related matters are far more many-sided at the level of rural, peri-urban, and urban livelihoods. The vulnerability and sensitivity of those livelihoods to external factors, such as natural calamities (such as floods, droughts, and disasters) and socioeconomic issues (such as market instability, unclear land tenure, etc.), as well as internal factors, such as low education level, poor social cohesion, too one-sided income structure, and so forth, are clearly related to the poverty problem. Accordingly, I went back to the philosophy and approach of Amartya Sen, Nobel Laureate of Economics of 1998 (12), who extensively studied the interconnections among welfare, poverty, and famines. He has developed approaches to study the poverty question with a higher resolution than frequently used poverty indices. His concept of poverty is: There are good reasons for seeing poverty as a deprivation of basic capabilities, rather than merely as low income. Deprivation of elementary capabilities can be reflected in Catching fish from the Mekong River during dry season, Vientiane, Laos. (Photo: M. Keskinen) 226 Ó Royal Swedish Academy of Sciences 2008

3 Figure 3. The livelihoods of peri-urban Phnom Penh require more money in their daily operations than rural livelihoods. The income poverty classifications thus consider the former to be less poor than the latter. In terms of deprivation, the implication is reverse. (Photos: O. Varis) premature mortality (especially of children), persistent morbidity, widespread illiteracy, and other failures. Deprivation typically leads to social exclusion and marginalization. Such groups are particularly weak in getting themselves out of poverty by self-help, and economic growth does not trickle down to these people. Poverty in Traditional and Modern Sectors An important but largely neglected distinction must be made between people who live in more or less traditional manner, often on subsistence farming, fishery, etc., and such marginalized people as the landless and urban slum dwellers, even though economic indicators (such as USD 1 2 a day) do not make much of a difference between them (Fig. 3). In the case of the Mekong Basin, it is striking indeed to realize how strong the contrasts are between the traditional livelihoods relying mainly on subsistence livelihood activities and the emerging modern economic sector, on which most development projects tend to focus. Table 1 summarizes some of the key features of these two sectors. The subsistence sector is often called the traditional sector. In between the modern and the traditional populations, there is the urban informal sector; this includes illegal dwellers, shantytown inhabitants, and so forth, who, in a way, are living at the borders of the formal economy on subsistence level and are dominantly marginalized and deprived. In many cases, the expansion of the modern sector has introduced massive changes to more traditional livelihoods, i.e., those that make their living out of common pool resources. One problem that is related to this the challenge is: how do we ensure that the increased aspirations for the modern sector do not lead to marginalization and exclusion of the people from more traditional sectors. In Sen s (12) philosophy, local economic activities should be developed so as to allow the people to have control and ownership over the productive resources and the local markets, as well as other institutions. These local markets should connect to urban markets so that growing urban wealth would benefit rural population, and the wealth would trickle down. In Cambodia, for instance, such a situation is still very far away. The country has a violent recent history, its economy is still exceptionally dependent on foreign aid, and rural land tenure and other governance systems remain unclear and discontinuous. Although the economy has been growing, the disparities are widening, particularly between rural and urban areas. In terms of the Tonle Sap area, natural reserves are one way to conserve lake ecosystems and the nature of the lake basin as a whole. In this context, interesting questions arise in the relation to the traditional sector and natural reserves: Should we see the traditional sector as a part of the ecosystem to be conserved? Do we encounter human rights questions? How should we handle the poverty reduction issue in such situations? If conservation efforts hinder local economic activities, the countereffect may be the erosion of loyalty of those and related groups and subsequent expansion of illegal exploitation activities, at least in cases in which the governance system is weak, as in Cambodia. The contrast between economic development and environmental degradation tends to be characterized very one-dimensionally; the negative impact of social deprivation on the environment is a massive problem in most parts of the world (13), which is too often disregarded, and it can be tackled by social and economic policies rather than solely with environmental policies. Table 1. Subsistence and modern sectors: some attributes. Subsistence (traditional sector) Uses no or very little money People supply themselves with basic commodities, such as food, water, fuelwood, etc. Institutions are primarily customary, religious, etc. People are living within the nature Example activities: family farms/fishery/forestry for village-level supply Modern sector Is driven by money Nature is used as a resource of tradable goods that are primarily valued after trade Institutions are primarily set up by government, etc. People are using nature as a resource base and are living out of nature Example activities: cash-crop farming, commercial fishery, fish farming, industry, hydropower generation, urban water supply plants Ó Royal Swedish Academy of Sciences

4 Figure 4. WUP-FIN policy model: Results of the sensitivity analysis of the three development goals analyzed (poverty reduction, economic growth, and environmental sustainability) with regard to the eleven sector policies (14). POLICIES AGAINST POVERTY AND DEPRIVATION: TWO CASE STUDIES It becomes clear from the previous discussion that in the water sector, we need far more elaborate and tailored analytical tools than those that exist today in order to tackle policies related to poverty reduction. The policy analyses based on so-called WUP-FIN Policy Models and developed for the Tonle Sap area (14) and the Mekong Delta of Vietnam (Varis et al., in prep.) offer one possible approach to this kind of more elaborate analysis. WUP-FIN was a complementary project to the Mekong River Commission Water Utilization Programme. It is funded by the Development Cooperation Department, Ministry for Foreign Aaffairs, Finland. It was active from June 2002 to October WUP-FIN is an acronym in which the first part comes from the Water Utilization Programme and FIN comes from Finland. The results indicate that a compromise policy is possible, and it is radically more balanced and acceptable than any of the policies that target only one of the three goals at a time (18). Interestingly, some sector policies would be crucial for both the economy and poverty reduction, but not all. The huge shortcomings in education and institutions are the most obvious ones that most strongly block meeting these two goals. Community development (including also the local production Case 1: Tonle Sap, Cambodia The surroundings of the Cambodian Great Lake, the Tonle Sap, are among the poorest parts of the Mekong Basin. It is a unique and exceptionally rich wetland-lacustrine ecosystem, and it has a population of around 1.2 million in its vicinity. The chiefly subsistence-level livelihoods are profoundly dependent on the lake. The Tonle Sap area is the poorest part of Cambodia, with an average income of USD 0.4 per capita. In this analysis, the poverty concept was, however, elaborated to follow closely Sen s (12) concept. The main concern was to analyze the vulnerability of the livelihoods to changes in the environmental conditions and in the socioeconomy and, thus, their vulnerability to deprivation. The starting point was the Mekong Agreement of 1995, particularly, its vision statement. It specifies three development goals for the basin: economic growth, poverty reduction, and environmental sustainability. The possible combinations of sector policies (Fig. 4) for achieving these often conflicting goals were analyzed systematically using a probabilistic, Bayesian network model (15). Four policy scenarios were constructed, one promoting each of the three development goals separately and one integrated, compromise scenario. Besides these three development goals, the impacts on the vulnerability of livelihoods in four rural zones were observed, based on altitude, and the urban areas were analyzed as a fifth zone. 228 Living in water: the poorest settlements are the most vulnerable to any change and deterioration of the aquatic system. Poor urban housing in Phnom Penh, Cambodia. (Photo: O. Varis) Ó Royal Swedish Academy of Sciences 2008

5 Table 2. Results of the four development scenarios with respect to the three development goals: economic growth, poverty reduction, and environmental sustainability. Goal/Scenario Economic growth scenario Conservation scenario Poverty reduction scenario Integrated scenario Economic growth Environmental sustainability Poverty reduction Clearly better average than in the two other basic scenarios. High uncertainty. Lowest average and relatively high uncertainty. Lowest uncertainty and not much lower average than in the Economic growth scenario. Environmental problems worsen with all scenarios. The Poverty reduction scenario is slightly better than the Environmental sustainability scenario, and the Economic growth scenario is clearly inferior to these two. The Integrated scenario produces very similar results compared to the Environmental sustainability scenario. Somewhat better than the Conservation scenario. Somewhat weaker than the other scenarios. Slightly better than the two other basic scenarios, but the difference is small. Ranking Almost equal in average to the best of the basic scenarios (Economic growth) but much lower uncertainty. Still somewhat better than the Poverty reduction scenario: less uncertain and better average. systems and markets), in turn, has a very important function in poverty reduction, although not as immediate macroeconomic implications. Large-scale fisheries would be economically beneficial but counterproductive to poverty reduction. At the same time, with every scenario and sector policy, uncertainties related to their impacts remain very high and must therefore be appreciated. Institutional situation (including the weak formal institutions and distorted informal institutions) appeared to be a key factor with respect to all three development goals. Institutional development is strongly anchored to the dilemma of local traditions (violated by hostilities) and modern governance institutions (still overly weak). A sensitivity analysis (Fig. 4) provides important information about ways in which the scenario could be improved with respect to each of the three development goals. Investment in rural development, education, and public health, and small- and medium-scale fisheries would yield the best results in poverty reduction. Urban development and roads, in turn, would boost economic growth, and the latter would also be instrumental in poverty reduction. Strengthening formal institutions would enhance all development goals. The biggest surprise was that the policies included in the model appear to be relatively toothless in relation to environmental problems, particularly if these are defined as environmental sustainability, as in the Mekong Agreement of This is obviously due to the following reasons. First, the concept of environmental sustainability is not easily conceptualized concretely enough in order to be treated analytically. Some instances emphasize biodiversity conservation (19), while many others relate it to ecosystem management, pollution control, and waste management. In our model, it is a weighted combination of all these issues (14). Second, since the majority of the local population dwells in villages and makes their living from the lake or the floodplain in a fairly direct way, the environmental issues are very closely bound to social issues. Social developments therefore are tightly bound to environmental impacts, and, typically, what happens in model simulations is that improvements in social conditions tend to introduce both positive and negative environmental impacts, which partly cancel each other. The situation has different results with a more efficient governance system. Environmental deterioration, in turn, leads very easily to social deprivation. It must also be noted that the mighty floods of the Mekong, which raise the water level up to 10 m, govern the Tonle Sap system, and, subsequently, the surface area of the lake grows fourfold. The sediments and other mass flows are dominated by these monsoon floods. There are no strong handles in the national sector policies to allow the control of these issues. The Tonle Sap system is particularly sensitive to transboundary environmental impacts, i.e., effects of developments in Laos, Thailand, Myanmar, and China. Case 2: Mekong Delta, Vietnam The approach in this analysis of deprivation and poverty, and related policies to overcome these problems, was quite similar in the case of the Mekong Delta in Vietnam as in the previous case. The delta is very different, however, from the Tonle Sap area; it is densely populated and economically far more developed. Broadening income distribution, social exclusion, and disparities among different livelihoods are widespread problems on the delta, and perhaps more so as the Vietnamese economy is going through a hefty modernization process. Again, four different scenarios were used: Maximized economic growth, environmental conservation, or poverty reduction separately, and a combination of these. A Bayesian network model (15) was used. Three geographical regions of the delta were analyzed separately, namely its upper, central, and lower parts, which all differ substantially from one another socially, economically, and environmentally. Table 2 summarizes the main outcomes of the four scenarios. Whereas implementation of economic growth policies, environmental conservation policies, as well as social policies was all extremely important, still more important was the realization that we must find feasible compromises in which all of these three, sometimes conflicting and distant, goals can be achieved. The sensitivity analysis plot in Figure 5 clearly shows some of the important differences in the outcomes of those policies. A general feature seems to be that it is far easier to find policy combinations with the model and better results are produced with respect to each development goal in the central delta than in the two other zones. This is valid particularly for economic development, but also for the other goals. For the lower delta, in contrast, it appears most difficult to coordinate the three development goals. There are more confrontations than in the other zones. Economic development is prone to enhance polarization and poverty problems. The sharp contrast between commercial aquaculture development and small-scale farmers is growing in importance, and the latter are subjected to marginalization due to loss of livelihoods. A balance between economy and environment is difficult to achieve, particularly in the case of mangroves. Increasing salinity and consequent water-scarcity problems are obvious due to the growing volume of saltwater aquaculture. In all zones, environmental sustainability seems to be very difficult to maintain, even at the present level. Obviously, more stringent and explicit environmental policies would be necessary than those currently implemented on the delta. Most of the needed approaches call for economic growth and strong Ó Royal Swedish Academy of Sciences

6 Figure 5. Results of the sensitivity analysis. The plot shows the sensitivity of the goals to changes in sector policies. The further the line is away from the zero line (the circle denoted with 0.0%), the more impact the policy has on the respective development goal. The impact is negative if the line goes inside the circle, otherwise it is positive. institutional measures. Again, environmental sustainability was not easy to conceptualize. In our model, it is a combination of various issues and problems related to the environment and natural resources. The results related to the upstream impacts can be summarized in the following way. Although the Mekong is known for its massive floods, the delta is relatively resistant to them, at least if compared to other parts of the Lower Mekong Basin. This is in a large part due to the extensive physical infrastructure that has been constructed on the delta during the past several decades. Floods remain a problem anyhow, and the contemporary policy has been characterized by the phrase living with floods, which means adaptation of the economy and housing to floods. The dry-season low flows constitute another problem to the livelihoods on the delta. The further the saltwater intrudes inland, the lower are the flows. The major concerns in both respects are changes in water-quantity regime upstream, including dam construction, land-use changes, and land-cover changes, such as changing forest area. The most critical factors in this regard are changing flood pattern and, particularly, sharpening of the flood peak, as well as lowering of the dry-season low flow. In terms of water quality, changes in sediment transport as well as growing concentrations of nutrients, pathogens, and toxic substances are major issues. The model simulations indicate that environmental sustainability is more sensitive to the upstream effects than economic development or poverty reduction. Low flows and floods are particularly important, and among water-quality issues, the sediment regime appears to be the most influential. CONCLUSIONS Poverty reduction is one of three key development goals within the Lower Mekong Basin. In economic terms, poverty is most often measured with an income indicator, such as USD 1 a day: people below such an income level are classified as poor, and those above it are not considered as poor. It is astonishing how widely this concept is still being used by leading aid agencies despite its well-recognized problems as well as the availability of more elaborate alternatives, one of which was discussed here. An income indicator applies well to people within the modern sector but not to people who are living in the traditional sector (Table 1). Such an indicator may give an increased income value to an absolutely impoverished individual who was before a respected and prosperous farmer or fisher but was not using much money, but is now living in a shantytown of a big city because his livelihood was destroyed by the modern sector that took over or destroyed that part of the lake basin that traditionally (customarily if not legally) belonged to his tribe or family or village. How should we understand and develop the poverty concept to be more valid in coping with the differences between the traditional and the modern sector? Sen (12), among others, provided plausible alternatives, but their way to policy level would require further operationalization of the concepts as well as the appreciation of the complex character of the poverty issue it is not only an income issue but an outcome of social deprivation. Economic growth is a prerequisite of poverty reduction there is little doubt about that. China and Vietnam are among the many countries that have promoted socialist policies that 230 Ó Royal Swedish Academy of Sciences 2008

7 have not allowed much spread in the income distribution, but both recognized some decades ago that this policy did not lead to economic or social development. However, economic growth is by far an insufficient condition for poverty reduction. In fact, the relative poverty (i.e., income disparity) tends to grow in low-income countries as their economy develops, as was detected by Kuznets (5) decades ago. However, the growth of poverty must not take place investment in education, empowerment of small-scale entrepreneurship and other means of microeconomic environment, good governance, infrastructure, and so forth, together with income distribution, can allow the economic growth to also include the poorer echelons of society. Additional social policies are needed to work against the exclusion of the rural landless and underemployed people as well as the urban slum dwellers. We see that these are the key points necessary to develop the Lower Mekong Basin into a prosperous region that still would enjoy a rich natural environment. The opposite scenario a deprived, overly polarized, and poverty-driven society with degraded natural resources is also a real possibility. The Bayesian policy models proved to be a useful tool in analyzing impacts and uncertainties of different management options and finding compromise solutions between them. Thus, the policy models help to diagnose problems and find consistent views on the roles of different policies to tackle those problems. 9. Dagdeviren, H., van der Hoeven, R. and Weeks, J Redistribution does matter: growth and redistribution for poverty reduction. In: Growth, Inequality, and Poverty Prospects for Pro-poor Economic Development. Shorrocks, A. and van der Hoeven, R. (eds). Oxford University Press, Oxford, pp United Nations Development Programme publishes annually its Human Development Report, which includes several international comparisons using an array of human development oriented indices. For more information, see When ranking the peninsular Southeast Asian countries with respect to almost any of the poverty indicators, the result is more or less the same: Singapore comes by far first (with least poverty incidence), Malaysia and Thailand follow, and Vietnam comes next, well ahead of Myanmar, Lao PDR, and Cambodia. The last ones are among the leastdeveloped Asian countries in these comparisons. It is clear that the limits of such analyses are tested when tailoring poverty reduction policies at national or local levels. 12. Sen, A Development as Freedom. Oxford University Press, New Delhi, 396 pp. 13. Palo, M. and Mery, G. (eds) Sustainable Forestry Challenges for Developing Countries. Kluwer Academic Publishers, Dordrecht, Netherlands, 384 pp. 14. Varis, O. and Keskinen, M Policy analysis for the Tonle Sap Lake, Cambodia A Bayesian network model approach. Int. J. Wat. Res. Dev. 22, The Bayesian network methodology is based on the systematic analysis of causal interconnections in complex environmental-social-economic systems. The objective is to assess risks to various components of the environmental and social system under concern as consequences of different policy strategies. For more information, see 16 and Varis, O A belief network approach to optimization and parameter estimation: application to resource and environmental management. Artif. Intel. 101, Varis, O. and Lahtela, V Integrated water resources management along the Senegal River Introducing an analytical framework. Int. J. Wat. Res. Dev. 18, The plot shows the sensitivity of the goals to changes in sector policies. The further the line is away from the zero line (the polygon denoted with 0.0%), the more impact the policy has on the respective development goal. The impact is negative if the line goes inside the polygon, otherwise it is positive. 19. GMS Biodiversity Conservation Corridors Initiative. Greater Mekong Subregion (GMS) Core Environment Program, Manila, 32 pp. 20. The author deeply thanks Marko Keskinen, Mira Käko nen, and Matti Kummu, as well as the other members of the MRCS/WUP-FIN team, particularly Juha Sarkkula, Jorma Koponen, Markku Virtanen, Seppo Hellsten, Ulla Heinonen, and the Cambodian and Vietnamese co-workers and trainees. Also, the fluent cooperation with the Mekong River Commission and Finnish Environment Institute is appreciated. The Water Resources Laboratory staff from Helsinki University of Technology, and particularly Pertti Vakkilainen, is equally acknowledged. This work has received funding from the Academy of Finland Project References and Notes 1. Ward, K. and Rowcroft, P Economic development and water resource demands in the Lower Mekong Basin. In: The MRC Basin Development Plans Monographs. MRC Development Series. Mekong River Commission, Vientiane, pp Nafziger, E.W The Economics of Developing Countries (3rd ed). Prentice-Hall, Upper Saddle River, NJ. 3. Maxwell, S The Meaning and Measurement of Poverty. Overseas Development Institute, London. 4. World Bank World Development Indicators World Bank, Washington DC, CD-ROM. 5. Kuznets, S Economic growth and income inequality. Amer. Econ. Rev. 45, Kuznets, S Quantitative aspects of the economic growth of nations. Econ. Dev. Cult. Change 11, Cornia, G.A. (ed) Inequality, Growth, and Poverty in an Era of Liberalization and Globalization. Oxford University Press, Oxford, 438 pp. 8. Shorrocks, A. and van der Hoeven, R. (eds) Growth, Inequality, and Poverty Prospects for Pro-poor Economic Development. Oxford University Press, Oxford, 283 pp. Olli Varis is a senior researcher at the Water Resources Laboratory of Helsinki University of Technology, and he has broad and interdisciplinary experience in water, environment, and development research and consultation. He is a frequently used expert by various international organizations and the author of over 300 scientific papers. His address: Water and Development Research Group, Helsinki University of Technology, PO Box 5200, FIN Espoo, Finland. olli.varis@tkk.fi Ó Royal Swedish Academy of Sciences