World Bank Reprint Series: Number 478 REP478 Michael M. Cemnea Understanding and Preventing Impoverishment from Displacement Reflections on the State of Knowledge F | t .s > X8fe, Reprinted with permission from Journal of Refugee Studies Vol. 8, No. 3 1995. Copyright © Oxford University Press, 1995. Michael M. Cernea Understanding and Preventing Impoverishment from Displacement Reflections on the State of Knowledge Journal of Refugee Studies Vol. 8, No. 3 1995 Understanding and Preventing Impoverishment from Displacement: Reflections on the State of Knowledge* MICHAEL M. CERNEA The World Bank, Washington The remarkable progress in social science research on resettlement during the last decade is defined by the author in terms of (a) knowledge acquisition-the addition of considerable in-depth and 'extensive' new knowledge; (b) significant shifts in research trends-from academic inquiry to operational research, from description to prescription, from writing ethnographies of past cases to crafting forward-looking policy frameworks; and (c) development and diversification of research models-particularly an evolution from the stress-centred model to the impoverishment/re-establishment centred model in analysing resettlement. The impoverishment risks model consists of eight recurrent and interlinked processes. It reveals how multifaceted impoverishment caused by displacement occurs via induced landlessness, joblessness, homelessness, marginalization, increased morbidity, food insecurity, loss of access to common property and social disarticulation. The conceptual model of impoverishment through displacement also contains, in essence, the model for the positive re-establishment of those displaced, which requires turming the impoverishment model on its head. The author analyses in detail the drop and the reversal in the income curve of resettlers during displacement and relocation, and points out the financial premises for income recovery. The two key priorities recommended for future resettlement research are: (a) research on re-establishment experiences, and (b) research on the economics of displacement and recovery. It is with commitment to active engagement in-rather than passive contemplation of-the development process that we inaugurate this conference. This is the first time that a large international meeting has been devoted fully to the subject of development-induced population displacement * Keynote Opening Address presented at the International Conference on Development Induced Displacement and Impovenshment, organized by the Umversity of Oxford and the Refugee Studies Programme in Oxford, England, January 3-7, 1995. This paper will appear in McDowell, C. (ed.) Resistng Impoverishment. Tackimg the Consequences of Development-Induced Displacement (Refugee Studies Programme Forced Migration Series) ** The views, findings, and interpretation contained in this study are those of the author and should not necessanly be attnbuted to the institutions with which he is associated. OOxford University Press 1995 246 Michael M. Cernea and resettlement. Indeed, such a meeting is long overdue, and we thank Oxford University and the Refugee Studies Programme for bringing together social researchers and development practitioners to reflect on this controversial topic. Convening this international conference sends a double message: that development-caused displacement is a matter of international concern, rather than only of local or national relevance; and that public awareness-explicitly, political and ethical awareness-of the seriousness and complexity of this issue has crossed a significant threshold. This landmark event should be consequential not through its implications for research alone but, more important, through its well-conceived recommendations for resettlement policy and actual practice'. Four paramount questions structure my discussion: -What is the state of the art in social science research concerning involuntary resettlement? -What is the essence of the displacement/resettlement cluster of issues? -How are social scientists addressing people's resistance to, or their participation in, resettlement? -And last, but equally important: What are the current priorities, both in research and at the operational level, for improving resettlement? Knowledge About Resettlement: The State of the Art Anthropologists and sociologists initiated research on forced population displacements before other disciplines and before such processes became a subject of public policy. Researchers such as David Butcher-whom we remember and miss with great sadness-and other British scholars including Brokensha (1965) and Chambers (1970), as well as Roy-Burman (1961) in India, and, especially, Scudder (1973), Colson (1971), and Gans (1968) in the US, have created an initial and rich 'bank of knowledge' on resettlement, to use Elizabeth Colson's phrase. Unfortunately, it is also true that policy makers have not exactly crowded as customers to this golden knowledge bank. An important change, vastly consequential in the long run, happened fifteen years ago, in 1980, when the World Bank became the first development agency to adopt an explicit policy concerning involuntary resettlement, a policy formulated by social scientists and grounded in social research (World Bank 1980; see for details Cernea 1986, 1988, 1993). As a result, social science research on involuntary resettlement has literally exploded, especially during the last ten years, practically carving out a topical 'subfield' in social science. Books and dissertations, studies and symposia, evaluation reports and policy documents, university syllabi, training courses, and briefs for parliamentary hearings or courts of law, have widened the intellectual endeavours focused on resettlement issues, as documented recently by an impressive bibliography (Guggenheim 1993). Preventing Impoverishment from Displacement 247 This surge in research and publications has been largely responsible for the intellectual construction of development-induced displacement as a social problem of worldwide dimensions. However, more than the academic research, it was the growing movement of resistance to displacement at the grassroots, supported by many NGOs, that has powerfully re-cast displacement as a political issue. How can we summarize the major advances during the last ten years in the social science research on resettlement? The following main characteristics of this progress, in my view, stand out. More Intensive Knowledge First, we have a more intensive knowledge about resettlement today than ever before; we can name it 'thick' knowledge, to paraphrase a widespread term in anthropology. We have made enormous cognitive gains in terms of depth, refinement, and detailed understanding of how displacement actually occurs, gains that have exponentially enriched the body of our knowledge on the subject. (We cannot yet say the same, however, about our knowledge of the re- establishment process, following displacement.) The main factor that accounts for this intensive knowledge is the unprecedented 'hands on' involvement during this decade of scores of social scientists in development programmes that involve displacement. Many development anthropologists have moved from the traditional role of participant observers in the process, to become participant actors, actors who use their expertise to find ways of avoiding or minimizing displacement. They also help on the ground to mitigate its adverse consequences, by improving the planning, monitoring, supervision, analysis, and evaluation of displacement and resettlement operations. This type of operational and action research is still an unheralded contribution. Little is known about it by the social science community at large. But from my vantage point, privileged to read hundreds of field reports, I stand witness to the good analytical quality, intricately detailed field assessments, and sharpened practical expertise reached by scores of resettlement social scientists. They are making a tangible contribution to protecting the welfare of people caught in forced displacement processes. More Extensive Knowledge Traditionally, the building blocks of knowledge in our area were extracted from research on dam displacements (mainly by anthropologists in developing countries) and urban displacements (mainly by sociologists in industrialized countries). What is relatively new during the last decade is the expansion of knowledge about displacements caused in other sectors, and affecting additional population groups, as well as of forced relocations unrelated to development projects. Such new sectors, involving new issues and a variety of 248 Michael M. Cernea types of resettlement, are: forestry (Turnbull 1987; Fernandes, Das and Rao 1989); mining, especially open pit mining and the construction of thermal energy plants (the literature on Singrauli is one example); the establishment of biosphere reserves and parks (Brandon and Wells 1992; West and Brechin 1991); the conversions in land use (Lane and Pretty 1990); transportation corridors; urban growth and environmental infrastructure in developing countries (Cernea 1989; Jellinek 1992; Davidson and assoc. 1993); politically mandated mass relocation (Clay 1988; Pankhurst 1992; Diece and Viezolli 1992; deWet 1993); and resettlement caused by structural adjustment reforms. Attention to the increased diversity of sectoral circumstances and the variety of previously unstudied types of displacements have broadened our empirical basis and the scope of our concepts and reasoning. Before moving to the third characteristic, I would like to emphasize that in both aspects discussed above a remarkable trend is the growth of resettlement research in developing countries, particularly in India, Brazil, China, Mexico, Argentina (e.g. Fernandes and Ganguly-Thukral 1989; Baboo 1992; Mathur 1994; Bartolome and Barabas 1990; Joshi 1991; Odidi Okidi 1992; Ribeiro 1994; Scott 1992; Huang 1984; Wangxiang 1993; for a detailed bibliography see Guggenheim 1993). From Description to Prescription Third, resettlement research has advanced from descriptions to prescriptions, and from academic analysis to operational research. Early resettlement 'writings ventured less into the territory of recommendations, despite their excellence in analysis. The more prescriptive posture of recent work (Bartolome, Cernea, Cook, Fernandes, Guggenheim, Partridge, Scudder, and many others) reflects the growing maturity of this domain, the increased self- ccnfidence of its research-informed practitioners, and the relevance of this type of' social knowledge to development's dilemmas. Development/applied anthropology and sociology are making some of their finest contributions ever in this area. Policy Frameworks Fourth, resettlement research has crossed the threshold from producing piecemeal ethnographies to crafting policy frameworks. Ethnographies deal with consummated cases, policies guide future processes. Case studies should remain a staple in our research menu, but there is great gain embodied in the ability to articulate credible policy proposals. Social science concepts and the involvement of professional social scientists have been decisive, not only in the fonrnulation of the World Bank's resettlement policy, but also of the policy of all 25 OECD countries' donor agencies, and of sectoral policies in Brazil, Colombia, and other countries. This is also true for the growing debate cunrently going on in India about the adoption of resettlement policies. We Preventing Impoverishment from Displacement 249 should not overstate the situation, however: the fact that social science can help formulate resettlement policies, and has done so in a few places, does not mean that this is currently happening routinely. Structural and political obstacles continue to impede the incorporation of social science research into policy and law. In scores of developing countries the problem of policy adoption is far from being resolved, or even being approached (Cernea, forthcoming 1995). Quantification Fifth, resettlement research has succeeded in generating quantified estimates at global level of the magnitudes of development-induced displacements. The most significant estimate, arrived at by our group at the World Bank, has emerged from collecting, verifying and combining a vast body of data. The estimate is still conservative, yet nothing less than stunning: we found that about ten million people annually enter the cycle of forced displacement and relocation in two 'sectors' alone-namely, dam construction, and urban/ transportation. That means that about 90-100 million people have been displaced during the decade to which I refer. Compare this with the numbers of refugees, long recognized as a major contemporary problem. Development- caused displacements, that seemed to be piecemeal occurrences and were estimated as totalling far less than the number of refugees worldwide (assessed by UNHCR at 15-20 million), have turned out to be a much larger process than all the world's annual new refugeeflows. Refugees and development-displacees, of course, are not 'numbers' that compete with each other, but are parallel global dramas, sometimes intertwined. But the shock value of numbers and quantifications is part of the social construction of involuntary resettlement as a problem, and increases public receptivity to the messages of resettlement research. Note, also, that the ten million figure annually, or 100 million people for the decade, are still partial figures: they do not include the populations displaced within what I called earlier the 'new sectors': displacements from forests and reserve parks; mining and thermal power plant displacements; and other similar situations. We simply have not yet generated reliable aggregate data for all sectors worldwide. Shifting the Key Emphasis Sixth, research on displacement has shifted its key emphasis from being stress- centred (the Scudder-Colson temporal model of resettlement phases, 1982) to being impoverishment-centred (the Cernea model, 1988, 1990). Through this shift, research conceptualizes and theorizes better the essence of displacement. The behaviour of displaced and resettled populations can be understood and explained more fully as a response to economic, cultural and social impoverishment than to 'stress'. By incorporating the shock/stress dimension into a comprehensive socio-economic impoverishment model, social science 250 Michael M. Cernea research can shift effectively from mainly reporting displacement traumas to predicting trends and prescribing actual remedies. It zeroes in on what must be the practical heart-of-the-matter in resettlement: preventing impoverishment and reconstructing livelihoods. I believe that these six characteristics capture the essential lines of progress in the state of our art. We can be collectively proud that we have achieved intensive and extensive knowledge expansion; that we have shifted from academic analysis to operational research, and from description to prediction and prescription; that we are moving up from case ethnographies to policy formulation; from investigating discrete cases to quantified assessments of worldwide trends; and that we have changed the focus from stress-centred to poverty-centred theory and operational analysis. However, research about some other important issues remains unacceptably vveak and insufficient. On these we can and must do better. I will highlight some of these neglected areas of inquiry later. fPeventing lmpoverishment: Eight Main Risks The essence of the displacement/resettlement nexus is the development-induced irnpoverishment of some population segments and the efforts to prevent and overcome it. Progress in knowledge notwithstanding, thorny problems and disastrous effects continue to plague displacement operations in many countries. One major result is impoverishment, another is increased resistance and political tension surrounding involuntary resettlement. During 1993-94, 1 had the responsibility of leading a Task Force mandated to review all 1986-1993 World Bank-financed projects involving involuntary population displacement. For comparative purposes, this review also included many projects not assisted by the Bank. Exactly a year ago at Oxford University, more than 120 researchers and NGO representatives engaged in a consultation about the mid-term findings of this review. By now, we have completed and published the study (World Bank 1994) and I will discuss some of its final findings, pertinent to preventing impoverishment caused by displacements. An important part of our study focused on how impoverishment happens. In poorly handled displacements, severe impoverishment and social dis- inlegration affect large numbers of people, and heavy costs extend well beyond the immediately affected populations; conversely, socially and economically responsible resettlement can prevent impoverishment and can exlend benefits to the regional economy and host populations. Ensuring that involuntary resettlement is avoided or minimized-and when unavoidable, is carried out without impoverishing the people displaced-is necessary on both economic and ethical grounds. Our study focused on Bank-assisted projects, rather than on the more numerous displacements occurring in the countless projects supported from domestic resources alone. However, since the present paper addresses Preventing Impoverishment from Displacement 251 resettlement issues as a worldwide problem, and is not a World Bank report, let me state bluntly that the worst consequences of forced displacement occur in domestic projects that are not guided by national or international policy norms. Our study found that impoverishment and brutal violations of basic human rights happen most frequently in programmes that are not subject to agreements on policy guidelines and to professional outside review, supervision, and evaluation. Such domestic projects account for the overwhelming majority-at least 95 per cent!-of the millions and millions of people forcibly displaced worldwide. This fact is the irrefutable argument for the adoption of national policies and legal frameworks for resettlement in all developing countries. How does impoverishment through displacement occur and how can it be prevented? I have examined abundant empirical evidence and compared the findings of numerous anthropological, sociological and geographical studies, in order to distinguish the basic processes that occur when people are forcibly displaced. What these comparisons revealed, beyond the enormous diversity of individual situations, are patterns of recurrent characteristics. Initially I defined seven characteristics or sub-processes (Cernea 1990), and subsequently identified one more, outlining a pattern of eight typical interconnected trends (Cernea 1993). Together, they constitute the model of the overall impoverishment process. This process captures not only the economic, but also the social and cultural impoverishment, reflecting the fact that displaced people lose natural capital, man-made capital, human capital, and social capital. The eight sub-processes that converge in impoverishment are not the only but rather the most important ones. In different locations, they occur with variable intensities. Very concisely, these are: 1. Landlessness. Expropriation of land removes the main foundation upon which people's productive systems, commercial activities and livelihoods are constructed. This is the principal form of de-capitalization and pauperization of displaced people, through loss of both physical and man- made capital. 2. Joblessness. Loss of employment especially affects urban people, but also occurs regularly in rural areas, displacing landless labourers and service workers, artisans and small businessmen. Creating new jobs is as difficult as finding empty lands, and resulting unemployment or underemployment lingers long after physical relocation. 3. Homelessness. Loss of housing and shelter is temporary for most displacees, but for some it remains a chronic condition. In a broader cultural sense, homelessness is also placelessness, loss of a group's cultural space and identity, or cultural impoverishment, as argued by Downing (1994, 1995) and by students of 'place attachment' (Low and Altman 1992). 4. Marginalization. Marginalization occurs when families lose economic power and slide downwards: middle income farm-households do not become landless, but become small landholders; small shopkeepers and 252 Michael M. Cernea craftsmen are downsized and slip below poverty thresholds. Relative marginalization often begins long before the actual displacement; for instance, when lands are condemned for future flooding and are implicitly devalued, new public and private infrastructural investments are prohibited, and the expansion of social services is undercut. 5. Increased Morbidity. Serious decreases in health levels result from the outbreak of relocation-related parasitic and vector-born diseases (malaria, schistosomiasis) and from increased stress and psychological traumas. Vulnerability to illness is increased, and unsafe water supply and waste systems tend to spread infectious diseases, diarrhoea, dysentery, etc. 6. Food Insecurity. Forced uprooting increases the risk that people will fall into chronic food insecurity, defined as calorie-protein intake levels below the minimum necessary for normal growth and work. Sudden drops in food crops availability and/or incomes are certain during physical relocation, and hunger or undernourishment tend to become lingering long-term effects. 7. Loss of Access to Common Property. For poor people, particularly for the landless and otherwise assetless, loss of access to non-individual, common property assets belonging to communities that are relocated (forested lands, water bodies, grazing lands, etc.) represents a cause of income and livelihood deterioration that is systematically overlooked and typically uncompensated in government schemes (with few exceptions, particularly China). 8. Social Dis-articulation. The dismantling of communities' social organization structures, the dispersion of informal and formal networks, associations, local societies, etc., is an expensive yet unquantified loss of social capital. Such 'elusive' dis-integration processes undermine livelihoods in ways uncounted and unrecognized by planners, and are among the most pervasive causes of enduring impoverishment and dis-empowerment. This conceptual construct of the impoverishment process not only has cognmtive value, but operational utility as well. First, it provides a risk model for planning, by synthesizing the lessons of many complex past processes and predicting almost certain outcomes if this warning model is ignored. These eight processes must be seen as a set of potential risks-high-probability risks-that will undoubtedly become real if unheeded, or can be avoided if anticipated and purposively counteracted. But like every risk forecast derived from past experience, this conceptual-predictive model can act as a 'self-destroying prophecy' (Merton 1962). Risk recognition and analysis are crucial for the argument that impoverishment through displacement can be avoided. Second, and most important, this conceptual construct of impoverishment through displacement contains in nuce the model for the positive re- establishment of those displaced. Indeed, if we reverse this model, it suggests Preventing Impoverishment from Displacement 253 precisely what needs to be done to restore, and whenever possible improve, the livelihoods and incomes of those displaced. Reversed in this way, the model offers the matrix for the subsequent process-the rehabilitation that must follow displacement: in other words, landlessness risks should be met through land-based planned re-establishment; homelessness-through sound shelter programmes; joblessness-through employment creation; inevitable disarticulation-through purposive community reconstruction and host- resettler integrative strategies, while deliberately enabling, so far as is feasible, those displaced to share in the specific benefits generated by the programme for which they had to relocate. For instance, enabling reservoir resettlers to relocate on the newly irrigated lands rather than around the reservoir; or providing electricity to relocated villages; or earmarking a share of the revenues resulting from selling energy; or providing access to affordable housing in newly constructed urban areas (see also Cernea 1988, 1991). The basic policy message embodied in the impoverishment/reconstruction model is that the intrinsic socio-economic risks must be brought under control through an encompassing strategy, and cannot be tamed through piecemeal random measures based solely on financial compensation for condemned assets. Such strategies, however, are seldom developed and even more seldom carried out. We must therefore ask: which factors transform these potential risks into sad reality? Despite variations between countries, I see several similar common causes of failure: 1. Policy. The vast majority of development-caused displacements worldwide are carried out through approaches that pursue expropriation and are not focused on income re-establishment. Expropriation laws provide for compensation for condemned assets, but they don't aim at, nor do they promise, re-establishment of prior levels (Shihata 1991, 1993; Paul 1988). Even worse, many decision makers and planners do not understand that approaches essentially limited to compensation are structurally insufficient and prone to failure. In practice, property acquisition procedures do not provide enough resources to enable displacees to purchase replacement lands and other assets, and frequently exclude or underpay significant numbers of affected people. Most governments use the might of the law and their institutional instruments to forcibly displace people, but do not sufficiently use their institutional instruments to facilitate 'land for land' alternatives. 2. Finances. The financial resources allocated for displaced people usually fall short, often by large amounts. Limited or absent political commitment, budget constraints, and inflexible allocation procedures shrink resources further. 3. Methodology. The current methodology for the economic and financial analyses of displacement/relocation costs tends to externalize these costs, rather than to internalize them within project budgets. This is not a minor 254 Michael M. Cernea issue. Some tend to dismissively belittle it, but it is one of the well kept 'secrets' of resettlement misplanning and failure. When resettlement costs are underassessed (whether deliberately or by error), displacees and hosts are forced to bear an undue share of the burden, which reduces their living standards. 4. Weak Institutions. The institutions in charge typically lack a policy mandate, commensurate organizational capacity, and professional social engineering skills. 5. Authoritarianism. Both the displaced and the host populations are not empowered to adequately participate in the planning and execution of the relocation especially in negotiating viable solutions. Effective legal mechanisms for negotiating and resolving grievances are often absent or subverted. The Resettlers' Income Curve To take this discussion a step further, let us examine and deconstruct the income curve of resettlers. This analysis will demonstrate why compensation for lost assets is alone incapable of restoring prior income levels, and why distinct resources should be invested for resettlers' 'take-off.' In this argument, I am drawing upon the Scudder-Colson (1982) analysis of the phases of resettlement and upon an environmental economics analysis by Pearce (1993). Resettlers' income curve during the displacement-relocation process consists essentially of four phases: -- a slow-growing or flat segment preceding displacement; --a sudden downward segment at dislocation; --a more-or-less flat segment during the adjustment/transition period at the new site; and --an upward segment once income restoration begins (see Figure 1). The curve NR (for No Resettlement) illustrates the potential path of income and asset accumulation without the project and where the income development curve of the area population is estimated conservatively at, say, one per cent per year or less. When a project displaces people, this income curve-assumed to be rising orLly slowly-is suddenly interrupted by displacement, at point D. For the aflected people this means an immediate decline in income and assets, represented by the downward segment D-D,, followed by a transition and adjustment phase, represented by the flat segment D1-A (for Adjustment), during which the income remains low, as economic activity is just being resumed at the new site during transition. If the resettlement package provided by the project works, a new development path is established from point A up,wards, represented as path R (for Rehabilitation). Preventing Impoverishmentfrom Displacement 255 Fgure 1 Resettlers' Income Curve During Displacement and Relocation NR E Di A t t+n t+x Time To achieve recovery at the same income level as without the project, development path R has to rise faster for a certain period to catch up to the point at which those displaced would have arrived spontaneously at time Bt + x. Path R must therefore be necessarily steeper than the NR 'normal' curve. Replacement of productive assets lost alone may, at best, create the material basis for starting on path RI, but additional investment resources for re-establishment and restoration are necessary for path BR to be steeper than path NR and catch up with it. Essentially, this indicates that if the pre-displacement community was experiencing even slowly rising living standards because of asset accumulation, the replacement of existing assets is not enough: the catching-up process involves some additional investment and asset creation. If path R continues to be steeper than NR after t + x, continuing the A-C line beyond the C (catching up) point would mean that the resettlers improve over their pre-displacement income. Figure I also suggests what would happen if there was only asset replacement. After the adjustment/transition period between D1-A (t to t+n on the time axis) development would recommence along path RI, but the resettled community would be worse off than it would have been without the project, even though it has resumed its development at the same rate as if the project had never occurred. The income forgone, lost because of dislocation, is represented in a simplified way by the white and shaded areas (DDIAC). This income loss is one of the hidden costs of the project, rarely taken into account in the analysis of project costs. Ideally, these losses are to be covered through compensations and grants for displacement hardship, taking into account the basic types of losses and risks that I described earlier. Most important, income loss compensation must be distinguished from asset loss compensation, which 256 Michael M. Cernea involves only the cost of replacing capital assets as they exist at time t and restoring them, or different income generating assets, at the new site. The Bank's resettlement policy requires, and various resettlement studies recommend, that compensation for assets lost must be at replacement cost, not at the 'market worth' of the lost assets. Compensation for assets and income lost, plus support for take-off, enables path R to be achieved: this is the ideal composition of the financial package for re-establishment. No externalization of any part of these losses to the people displaced must occur. This analysis also shows that the length of the adjustment transition period is a critical variable, on which the amount of income forgone (lost) by those resettled, and thus the real cost of the project as well, clearly depend. The shorter the transition, the less additional costs are caused by the project at large, and the sooner the relocated community gets back on its own feet. When the project is able to deliver a good resettlement package promptly to the affected people at dislocation time, the loss tends to be reduced (the white area DD1 C) and project costs are lower. Conversely, if land compensation is under- estimated, delivered slowly, and court grievances entail delays in payment, the transition is harder, more painful and additional costs (the full area of DD1AC) are incurred. An even better option is to prepare resettlement lands and 'packages' well in advance of dislocation: for instance, China's Shuikou Hydropower project, which I visited in 1994 for the fourth time, shows that establishment of orange orchards by the project is feasible several years beforehand, so they can be at, or close to, first harvest when handed over to resettlers. In this way, transitions can be shortened and income restored faster, with gains to resettlers and savings to the project. In a sociological perspective, the anatomy of re-establishment costs also highlights the importance of using other levers and resources, of a non- financial nature, to reduce costs and accelerate re-establishment. The process of income restoration is far from being just a cost-compensation transaction: it is a social readjustment process of considerable complexity that also depends upon non-economic factors such as people's forms of social organization, mobilization of their energy and innovativeness, participation, motivation, as well as upon the project's institutional efficiency. In summary, the above analysis leads to important conclusions. The economics of resettlement involves distinct cost categories critical for achieving project results without leaving people worse off; these expenses must be internalized in the overall project budget and financing. Yet current worldwide practice, with limited exceptions, does not use this kind of analysis, externalizes significant costs, and applies flawed methods in the economic and financial analysis for assessing risks and needed resources. C'onversely, when the real costs are fully recognized, important options open up ithat can help reduce them. The sociological design of resettlement plans should therefore go hand in hand with a sound economic and financial analysis of recovery costs and mechanisms for income restoration. Preventing Impoverishment from Displacement 257 Preventing impoverishment and restoring incomes and livelihoods are not intractable tasks; they are feasible, but very difficult, and require adequate resources, strong political and financial commitment from governments and development sponsors (including the private sector) and intense, strong participation by the populations affected and their organizations. Resistance and Participation The currently heightened international interest in resettlement derives in part from the increasingly organized and publicized resistance to forced displacement in many developing countries. Even if carried out adequately, involuntary resettlement causes social disarticulation. When executed with disregard to people's basic entitlements, it leads to additional social pathologies: avoidable hardship and pain, more losses, social protest, delays of project benefits, sharp political tensions. A Bank field mission reported, for instance, that in the Singrauli Thermal Project in India people opposed the bulldozers which were sent to force them out of their houses, and successfully blocked construction of a project building. Another field mission to the Subernarekha Irrigation Project reported strong local protests which have contributed to delaying the impoundment of the Chandil reservoir for two years. In Mexico, protests over the proposed San Juan Tetelcingo project (for which the Bank had previously declined financing) included highway blockades and a large demonstration in Mexico City. In Brazil, community activists formed the Regional Commission Against Large Dams (CRAB), which has since evolved into a national federation of people affected by displacement. In Indonesia, the resistance of many families from the Kendung Ombo reservoir has evolved from refusal to move, to court actions all the way to the country's Supreme Court. The anthropological literature, mostly silent about resistance until a few years ago, has started to analyse the growth, structure, patterns and outcomes of expanding social movements opposing displacement (Oliver- Smith 1991; Good 1992; West 1991 and others). There are deep causes at the roots of these social conflicts. While, clearly, there are projects where involuntary resettlement can and should be avoided entirely, equally clearly, there are many cases where infrastructural projects are indispensable and where developing societies must balance benefits such as safe water supplies, irrigation, efficient transportation systems, or urban growth with the costs and pains of resettlement. Many development programmes intrinsically embody or entail this kind of political and economic trade-offs and conflicts. These conflicts occur because long-term national or regional interests served by these programmes often cut across the interests of smaller groups, local communities, or some individuals. National interests and needs usually prevail. Conflicts emerge because the gains expected from these projects in the long term impose hard to bear losses in the short term. As one resettler in India summarized it, 'For their tomorrow, we are giving our today.' 258 Michael M. Cernea Development can never be completely free of conflicts. But it is unacceptable to unnecessarily exacerbate, instead of reconcile, such conflicts by absence of protective social policies. The recovery of resettlers' livelihoods is a matter of right and resources. Resource allocation is, in the last instance, a political matter, and not just an economic or financial one. As governmental agencies are using the weight of the state and the force of the law to impose expropriation and displacement, it is incumbent upon governments to also enable those displaced to get back on their feet and share in the benefits for which they are displaced. Opposition to displacement is not cost-free and painless to those who resist. Government officials or agencies who belittle opposition as the product of a handful of agitators or 'trouble making NGOs' are deceiving themselves, deceiving the public, and distorting the facts. As responsible social research has found, when large numbers of people engage in active resistance to resettlement they do so because they have no other effective avenues to defend, bargain for, and promote their interests. Resettlers' struggles have extracted big improvements in the 'packages' given to resettlers. The story of the Itaparica resettlement in Brazil is a case in point. The flames of resistance are often ignited not intrinsically by displacement's hardship itself, but because the policy vacuums and legal vacuums leave few alternatives to political struggle. In such vacuum, abuses and violations go unchecked and unpunished, and grievances have little standing in courts. National policy standards are needed in every country to institute norms of fairness, compelling to implementing agencies. Yet, the enactment of protective policies is resisted by powerful vested interests. We can often contemplate a socially schizophrenic situation: in the same country people displaced by the same government under two different projects are granted more entitlements and protection under a project co-financed by the World Bank and subject to the Bank's resettlement policy, than their neighbours in the next distinct displaced by a fully domestic project, to whom the same government denies identical compensation norms. Extending the application of the Bank's policy beyond Bank-assisted projects would be a major step forward. Recently, market-inspired advocacies increasingly plead the substitution of government agencies with the private (for profit) sector in financing energy investments that entail displacement. Such a trend is in many cases financially sound, but will not diminish the need for protective policies and legal frameworks; if anything, it would reinforce that need. Even more than public sector projects, market economies depend on clear operating rules that apply to all players. Another factor that exacerbates opposition is the fact that many government agencies decide in secrecy on displacement needs, amounts, procedures, terms, and deadlines , without the consultation and participation of those who have their most vital interests at stake. This often forces people to exercise participation by opposition and demonstrative resistance, rather than by bargaining and cooperation in finding the best possible solutions. Not Preventing Impoverishment from Displacement 259 providing information to affected people allows for abuse of entitlements, and ends up engendering hostility vis-d-vis project authorities. Is it paradoxical to advocate the 'participation' of people who are displaced in the projects that seek to involuntarily displace them? Strange as it may appear to some, encouraging and securing the active participation of people in solving the difficult problems of resettlement is essential, in fact even more important than participation in many other types of projects. In substance, there is no paradox: there is acute necessity, and mutual interest. Securing genuine participation is indispensable, because without the initiative and mobilization of the affected people's creativity and energies resettlement cannot succeed. Effective participation is apt to improve the recognition of people's needs, project design and implementation, and particularly the finding of better suited, acceptable solutions. Participation has prerequisites and takes several forms. The first pre- condition is to inform the affected populations in time about the need to resettle, about their entitlements, eligibility, options, due process, grievance mechanisms. Next, people and their organizations should be consulted and involved in finding or pondering alternative options and solutions to avoid or minimize displacement, or to identify suitable relocation sites. Further, many NGOs have proven themselves very effective in designing resettlement plans and realistic options, acceptable not only to the people but also to the government, and in mobilizing the energies of the resettlers for better implementation, self-help, self-organization, and monitoring of the planned relocation. Starting from 1994, a novel approach is mandated in this respect in World Bank-assisted projects: to improve participation in resettlement, the Bank requires its borrowers to make the resettlement plans available in draft to the public in the project area for review and comment before the Bank approves the funding for the new project. Recent trends in resistance movements is to go beyond project-focused claims and to bring up fundamental issues of resettlement macro-policy. An example comes from India, where a draft national resettlement policy was prepared recently by government agencies for formal adoption. What I find extraordinarily significant is that one Indian NGO has multiplied and sent this draft out to hundreds of various NGOs in all Indian states, to trigger a truly broad national discussion of policy principles for development-necessitated resettlement. One does not need to be a social scientist to predict the heightened awareness such a broad discussion will cultivate, the influence it may have over resettlement operations in many ongoing domestic and internationally-assisted projects, and the expanded opportunities for converting good practices and legitimate claims into firm national policy norms. Current Priorities in Resettlement Research Despite major progress in social science research on resettlement, which I discussed at the beginning, important variables of the resettlement process 260 Michael M. Cernea continue to be understudied and insufficiently known. Some personal suggestions about overarching research priorities are as follows. The first is to study in more depth the re-establishment segment of the displacement-resettlement continuum. Understandably, the dramas of sudden displacement have attracted more observers than the long, slow, arduous, unspectacular travails of rebuilding livelihoods. But if social research is to illuminate people's ways out of the displacement dramas, studying the successful adaptive strategies, the rebuilding of production systems, and the creation of new social organization patterns will eventually reduce the hardship and quicken the recovery of future resettlers. Longitudinal studies, comparative studies, short-term impact research stand to discover important responses to many unanswered questions about 'what works' and what does not when dismantled societies, kinship systems, and local cultures tend to reassemble, change, and function within new encapsulating environments. Along this line, I would propose a specific follow up to this present conference, a second international conference hosted by the University of Oxford and the Refugee Studies Programme, in one or two years' time, focused on researching how best to 'reconstruct the livelihoods' of development- displaced peoples and of refugees. Moving from the analysis of impoverishment mechanisms to primarily researching and synthesizing the practical experiences of resettlers' coping and adaptation strategies, and the various forms of financial and institutional support that governments can give, would be beneficial in practical ways. I anticipate important gains in knowledge for policy from the field research that the preparation of such an international conference, and its aftermath, would stimulate. No less important a priority, in my view, stands beyond the confines of sociology or anthropology: it is squarely in the court of our colleagues, the economists. Official misunderstanding or sheer ignorance about the complex economics of displacement and recovery are simply appalling in many agencies and countries. Many pitfalls in current practice can be traced to the sorry state of the economic research on resettlement and to the flawed prescriptions for economic and financial analysis, and for planning, in this domain, as I emrrphasized earlier. Perhaps anthropologists and sociologists have too exclusively colonized and monopolized resettlement. We must become more inviting of complementary perspectives. Further, important cognitive gains can be harvested from interdisciplinary and comparative analysis in resettlement. Empirical research in the last decade has advanced mainly in a two-pronged manner: as ethnographic-academic research and as operational action-oriented research. Based on both, more concept-building and new theoretical syntheses are now possible and needed. Unjustified insularity and clogged communication channels between neighbouring research fields, like ihe deep dichotomy between research on refugees and research on development displaced people to which I referred elsewhere (Cernea 1991), as well as the yet unbridged divide between disaster research and displacement research, hamper both knowledge and practice. Preventing Impoverishment from Displacement 261 Here lie other challenges to us for comparative, overarching syntheses. Evaluation and operational research on resettlement outcomes must be expanded, and not only as an initiative of individual researchers but as a responsibility of governments and institutions. My colleagues and I were gratified to see that our comprehensive evaluation study on policy and performance in the World Bank's resettlement portfolio (World Bank 1994) has been an instant best-seller and has become influential in both practice and research. To broaden the exchange of policy experiences, a similar multicountry review could be carried out, for instance, around OECD's resettlement guidelines, adopted in 1991. In fact, many bilateral donor agencies, such as GTZ, NORAD, ODA, CIDA, Caisse Centrale, OECF, etc., have long years of experience (that precedes the adoption of the OECD guidelines) with projects that entail involuntary resettlement. I am convinced that very useful lessons could be learned and exchanged through such comprehensive agency studies, lessons that are still insufficiently distilled and disseminated. And it may well be in the spirit of the World Summit on Social Development to propose that national governmental reviews on population displacement and resettlement become part of the post Summit work on improving social policies, side by side with poverty and other critical issues. The sober analysis of global development trends tells us that involuntary resettlement will not disappear. It is in fact likely to increase, due to urban growth, demographic trends, infrastructural investments, need to expand irrigation and food production, and also because of policy induced (as opposed to project induced) population displacements. Recognizing that a certain degree of displacement cannot be avoided does not mean, however, that it should be accepted as a God-ordained tragedy, with little more than a compassionate shrug of the shoulders. If resettlement must happen, then the issue is to change the national and international norms guiding it. Let me sum up: for this to become a reality, the basic steps are: first, the enactment of firm policies and legal frameworks; second, the allocation of commensurate resources that will allow internalization of costs; third, the empowerment of resettlers, enabling them to have voice and participation in the decisions and procedures regarding their relocation; and fourth, continuation of innovative social research. The urgency and goals of such research are inscribed in the very theme of our Conference: to prevent impoverishment and enable those displaced to share in the gains and not just in the pains of development. Grateful thanks are expressed to Dr. Ruth Cernea for her helpful comments in the preparation of this paper, and to Scott Guggenheim for signals from afar in the field. Gracie Ochieng processed it meticulously, beyond the call of duty. 1. A propitious circumstance enhanced the potential impact of this conference. Sixty days later, the first World Summit devoted to 'social development' convened in Copenhagen. One of the three central themes on the Summit's agenda was 'social integration.' Surely, involuntary displacement is a very painful type of 'social dis- integration'. Thus a proposal made in this Keynote Address was accepted by the 262 Michael M. Cernea Conference, which created a working group, chaired by Prof. Theodore Downing, charged to synthesize the Conference's policy relevant conclusions. The recommendations of the conference were formally submitted by the Refugee Studies Center to the January 1995 session of the UN Preparatory Committee of the Summit (held in New York) where they were discussed and supported by the representatives of several governments and national and international NGOs, with OXFAM playing a leading role in this process. From developing countries, some local NGOs sent their message on resettlement to the summit and documented it with empirical research data (Pandey 1995). As a result (and despite the advanced stage of the Summit's preparation), a statement on development-related displacement was introduced in Chapter II of the Summit's draft Programme of Action. It urges govermnents to avoid or minimize displacements whenever possible, and most importantly, to adopt 'policy and legal frameworks' for re-establishing those displaced. That paragraph was subsequently adopted by the Copenhagen Summit on March 12, 1995 as part of its formal documents, in the Section II devoted to the 'eradication of poverty' and the 'formulation of integrated action strategies' to that end. That section states: Governments are urged to integrate goals and targets for combating poverty into overall economic and social policies and planning at the local, national and, where appropriate, regional levels by: ... (d). Selecting, wherever possible, development schemes that do not displace local populations, and designing an appropriate policy and legal framework to compensate the displaced for their losses, to help them to re-establish their livelihoods and to promote their recovery from social and cultural disruption (Copenhagen Programme of Action adopted by the World Summit for Social Development, Copenhagen, 6-12 March, 1995). 2 For instance, enabling reservoir resettlers to relocate on the newly irrigated lands rather than around the reservoir, or providing electricity to relocated villages; or earmarking a share of the revenues resulting from selling energy; or providing access to affordable housing in newly constructed urban areas (see also Cernea 1988, 1991). The following list is an incomplete bibliography of works consulted recently or used in the preparation of this paper. ADAMS, W. M. (1993) 'Development's Deaf Ear: Downstream Users and Water Releases from the Bakalon Dam, Nigeria.' World Development 21(9):1405-1416. BABOO, B. (1992) Technology and Social Transformation. The Case of the Hirakud Multi-Purpose Dam Project mn Orissa. New Delhl, India: Concept Publishing Company. BARTOLOME, M. and BARABAS, A. (1990) La presa Cerro de Oro y el ingeniero El Gran Dios. Mexico DF: Instituto Nacional Indigemsta. B]ROKENSHA, D. and SCUDDER, T. (1968) 'Resettlement' in Rubin, N. and Warren, W. M. (eds.) Dams in Africa. An Interdisciplinary Study of Man-Made Lakes London: Frank Cass. BiLUTCHER, D. (1990) 'Review of the Treatment of Environmental Aspects of Bank Energy projects,' World Bank Industry and Energy Department, PRE Working Paper, March. CERNEA, M. M. (1988) Involuntary Resettlement in Development Projects Policy Guidelines In World Bank-Financed Projects World Bank Technical Paper No. 80. Washington, DC. - -(1988) 'Involuntary Resettlement and Development', Finance and Development 25(3):44-46 - (1989) 'Metropolitan Development and Compulsory Population Relocation. Policy Issues and Project Expenences', in Regional Development Dialogue 10(4) - -(1990) 'Poverty Risks from Population Displacement m Water Resource Projects,' Harvard University, HIID, DDP No.355, Cambndge, Mass. -- (1990) 'Internal Refugee Flows and Development-induced Population Displacement', Journal of Refugee Studies 3(4):320-339. Preventing Impoverishment from Displacement 263 --(1991) 'Involuntary Resettlement: Social Research, Pohcy, and Planning', in Cernea, M. M. ed. Putting People First: Sociological Variables in Rural Development. 2nd ed. New York, London: Oxford University Press. - -(1993) 'Anthropological and Sociological Research for Policy Development on Population Resettlement' in Cernea M. M. and Guggenheim, S. Anthropological Approaches to Resettlement: Policy, Practice, Theory, Boulder Col.: Westview Press. - (1995) 'Public Policy Responses to Forced Population Displacement', in Dreze, J., Samson, M. and Singh, S. eds. Displacement and Resettlement in the Narmada Valley. Forthcoming 1995. CHAMBERS, R. ed. (1970) The Volta Resettlement Experience. London: Pall Mall Press. CLARK, S., COLSON, E., LEE, J. and SCUDDER, T. (1994) 'Ten Thousand Tonga: A Longitudinal Anthropological Study from Southern Zambia, 1956-1991.' Draft, unpublished. CLAY, J. W. (1988) 'Villagization in Ethiopia', m Clay, J., Stemgraber, S. and Niggh, P. eds. The Spoils of Famine. Ethiopian Famine Policy and Peasant Agriculture. Cambndge, Massachusetts: Cultural Survival. COLSON, E. (1971) The Social Consequences of Resettlement: The Impact of the Kariba Resettlement Upon the Gwembe Tonga. Manchester: Manchester University Press. DE WET, C. (1993) 'A Spatial Analysis of Involuntary Community Relocation: A South Afncan Cast Study', in Cernea, M. M. and Guggenheim, S. E. eds. Anthropological Approaches to Resettlement: Policy, Practice, and Theory. Boulder, Colorado: Westview Press. DIECI, P and VIEZOLLI, C. eds. (1992) Resettlement and Rural Development in Ethiopia, Milano: Franco Angei. DOWNING, T. E. and KUSHNER, G. eds. (1988) 'Human Rights and Anthropology,' (Cultural Survival Report 24), Cambridge, MA: Cultural Survival Inc. DOWNING, T. E. (1994) 'Social Geometncs: A Theory of Social Displacement in Resettlement.' Paper presented at the International Congress of the Americanists. Stockholm/Uppsala, Sweden. - - (1995) 'Mitigating Social Impovenshment when People are Involuntarily Displaced,' (paper presented for the Oxford Conference on Development-Induced Displacement and Impoverishment). FERNANDES, W. and GANGULY THUKRAL, E. eds. (1989) Development, Displacement and Rehabilitation, New Delhi: Indian Social Institute. FERNANDES, W., DAS, J. C., RAO, S. (1989) 'Displacement and Rehabilitation. An Estimate of Extent and Prospects.' In Femnandes, W. and Ganguly Thukral, E. eds. Development, Dusplacement and Rehabilitation. New Delhi: Indian Social Institute. GANGULY THUKRAL, E. and SINGH, M. (1994) 'Dams and the displaced in India' m Mathur, H. M. ed. Development, Displacement and Resettlement. Focus on Asian Experwences, Delhi. Vikas Pubbshing House. GANS, H. J. (1968) People and Plans. Essays on Urban Problems, New York: Basic Books. GIBSON, D. (1993) 'The Politics of Involuntary Resettlement: World Bank supported Projects in Asia,' unpublished Ph.D. dissertation, Department of Political Science, Duke University. GOOD, C. (1992) 'Making the Struggle One Big One- Nahuatl Resistance to the San Juan Dam Mexico', paper presented at the Yale program on agranan studies. Yale University: New Haven. GUGGENHEIM, S. E. (1989) 'Development and the Dynamics of Displacemenft, in Fernandes, A. ed. Workshop on Rehabilitatton of Persons Displaced by Development, Bangalore: Myrada. - - (1994) Involuntary Resettlement: An Annotated Reference Bibliography For Development Research. Environment Woring Paper No.64, The World Bank. HUANG, Y. F. (1984) 'On the Reform of Resettlement Policies in China', Chinese Journal of Water Conservation, No.6. JOSHI, V. (1991) Rehabilitation-A Promise to Keep. A Case of Sardar Sarovar Project. Ahmedabad: Tax Publcation. LANE, C. and PRETTY, J. N. (1990) 'Displaced Pastoralists and Transferred Wheat Technology in Tanzania,' IIED, Sustainable Agriculture Pro,gramme, Gate Keeper Services, No. SA20. LAISSAILLY-JACOB, V. (1986) 'Un Exemple Ephenire de Planification du Developpement: I'AVB en Cote d'lvoire (1969-1980), Cahiers d'Etudes Africaines, 103, XXVI-3, pp.333-348. LOW, S. and ALTMAN, 1. eds. (1992) Place Attachment, New York: Plenum Press. MATHUR, H. ed. (1994) Development, DLsplacement and Resettlement. Focus on Asian Experiences. Delhi: Vikas Pubhshing House Ltd. MERTON, R. K. (1962) Social Theory and Social Structure. New York: Free Press: pp. 421-36. MORSE, B. and BERGER, T. (1992) Sardar Sarovar- Report of the Independent Review, Ottawa- Resource Futures International, Inc. 264 Michael M. Cernea NACHOWITZ, T. (1988) 'Repression in the Narmada Valley,' Cultural Survival Quarterly, 12(3):22-24. ODIDI OKIDI, C. (1992) Policy and Legal Framework in Development-Driven Involuntary Resettlement in African Countries, draft, Kenya: Moi University, School of Environment. OLIVER-SMlTH, A. (1991) 'Involuntary Resettlement, Resistance and Political Empowerment,' Journal of Refugee Studies 4(2):132-149. PANDEY, B. (1995) Development Induced Displacement. An Agenda for the Social Summit. Bhutaneshwas, Onssa: ISED. PANKHURST, A. (1992) Resettlement and Famine in Ethiopia. The Villagers' Experience. Manchester, New York: Manchester University Press. PARTRIDGE, W., BROWN, A. B. and NUGENT, J. B. (1982) 'The Papaloapan Dam and Resettlement Project: Human Ecology and Health Impacts,' in Hansen, A. and Ohver-Smith, A. (eds.). Involuntary Migration and Resettlement, Boulder, Colorado: Westview Press. PARTRIDGE, W. (1989) 'Population Displacement Due to Development Projects,' Journal of Refugee Studies, 2(3). PAUL, J. C. N. (1988) 'International Development Agencies, Human Rights and Human Development Projects', Denver Journal of International Law and Policy 17(l):67. PEARCE, D. (1993) 'The Economics of Involuntary Resettlement', consultant report to the Task Force for the Bankwide Resettlement review, manuscript, processed. RIBERIO, G. L. (1994) Transnational Capitalism and Hydropolitics In Argentina., The Yacyreit High Dam. University Press of Florida. JROY-BURMAN, B. K. (1961) Social Processes in the Industrialization of Rourkhela, New Delhi: Census of India. SALEM-MURDOCK, Ml (1989) Arabs and Nubians in New Halfa, University of Utah Press, Salt Lake City. SCOTT, R. P. (1992) 'Dams, Forced Resettlement and the Transformation of the Peasant Economy in the Sio Francisco River Valley, Brazil,' paper presented at the XVII International Congress of LASA, Los Angeles. SCUDDER, T. and COLSON, E. (1982). 'From Welfare to Development: A conceptual Framework for Analysis of Dislocated People' in Hansen, A. and Ohver-Smith, A., eds., Involuntary Migration and Resettlement, Boulder: Westview Press. SCUDDER, T. (1973) 'The Human Ecology of Big Projects: River Basin Development and Resettlement,' in Siegel B. ed., Annual Review of Anthropology, Palo Alto. -- - (1991) 'A Sociological Framework for the Analysis of New Land Settlements', in Cernea, M. M. ed. Putting People First. New York: Oxford University Press. S]HIHATA, 1. F. 1. (1991) 'Involuntary Resettlement m World Bank Financed Projects', in The World Bank in a Changing World, The Netherlands: Martinus Nijhoff Pubhshers, p.181. -- - (1993) 'Legal Aspects of Involuntary Population Resettlement', in Cernea, M. M. and Guggenheim, S. E. eds. Anthropological Approaches to Resettlement. Policy, Practice, and Theory. Boulder, Colorado. Westview Press. TIJRNBULL, C. M. (1987) The Mountain People. Touchstone Books. WALI, A. (1989) Kilowatts and Crisis: Hydroelectric Power and Social Dislocation in Eastern Panama. Boulder, CO: Westview Press. WEST, P, and BRECHIN, S. eds. (1991) Resident Peoples and National Parks. Tucson. Arizona: University of Arizona Press. WANGXIAiNG, C. (1993) 'Resettlement Associated with Hydro Projects in China' International Water Power and Dam Construction, 45(2):19-20. WODRLD BANK (1980) Operational Manual Statement 2.33. Social Issues Associated with Involuntary Resettlement in Bank-Financetd Projects. Washington, DC. ---(1990) Operational Directive 4.30: Involuntary Resettlement, Washington, DC, June. ---(1994) Resettlement and Development: The Bankwide Review of Projects Involving Involuntary Resettlement 1986-1993. The Environment Department. Washtngton, DC. THE WORLD BANK A partner in opening markets and strengthening economies - e ,. to improve the quality of life and expand prosperity FOR PEOPLE EVERYWHERE Headquarters 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. Telephone: (202) 477-1234 Facsimile: (202) 477-6391 Telex: MCI 64145 WORLDBANK UCI 248423 WORLDBANK Cable Address: INTBAFRAD WASHINGTONDC World Wlide Web: http://www.worldbank.org E-mail: books@worldbank.org European Office 66 avenue d'Iena 75116 Paris, France Telephone: (1) 40.69.30.00 Facsimile: (1) 40.69.30.66 Telex: 640651 7 okyo Office E; Iokusai 13uilding 1-1 Marunouchi 3-chome C 'hiyoda-ku, Tokyo 100, Japan elephone: (3) 3214-5001 F,csimile: (3) 3214-3657 T elex: 265838 The full range of World Bank publications, both free and for sale, is described in the annual Index of Publications, and of the continuing research program of the World Bank, in World Bank Research Program: Abstracts of Current Studies. The most recent edition of each is available without charge from: DISTRIBUTION UNIT, PUBLICATIONS OFFICE OF THE PUBLISHER THE WORLD BANK THE WORLD BANK 6 AVENUE d'IENA 1818 H STREET, N W. 75116, PARIS WASHNGTON, D.C. 20433 FRANCE U.S.A. ISSN 0253-2131