E NV I R O N M E N T 2454 r k 3 D E P A R T M E N T ^'~P A P E R S PAPER NO. 67 ENVIRONMENTAL MANAGEMENT. SERIES The Evolution of Environmental Assessment in the World Bank: from "Approval" to Results .Robert Goodland Jean-Roger Merciei January 1999 Environmentahi, and Sociallv Sustainable Develooment the World Bank ESSD The Evolution of Environmental Assessment in the World Bank: from "Approval" to Results January 1999 Papers in this series are not formal publications of the World Bank. They are circulated to encourage thought and discussion. The use and citation of this paper should take this into account. The views expressed are those of the authors and should not be attributed to the World Bank. Copies are available from the Environment Anchor, The World Bank, Room MC-5-105. Contents ABSTRACr V AcKNOWLEDGMENTS Vii INTRODUCrlON ix Chapter 1 Persuading the Proponent of the Needfor EA 1 Chapter 2 Predicting Adverse Impacts 3 EA Timing 4 EA Budget 5 Fielding Appropriate Disciplines 5 Agreeing on Each Measure in the EMP 6 Chapter 3 Securing Implementation Budget 9 Decentralization 11 Privatization 11 Chapter 4 Implementive Capacity 13 Chapter 5 Participation 17 Social Assessment 18 Chapter 6 Monitoring, Evaluation, and Audit 21 Chapter 7 Analysis of Alternatives 25 Sectoral EA 26 Environmental Management Series The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results Chapter 8 Conclusions 29 Other Conclusions 29 Guide tofurther sources of information 31 Boxes 1 Running Definition of EA ix 2 Internalizing Negative Environmental Externalities 4 3 The Need to Rank Environmental Impacts 5 4 Case Example Burkina Faso-Ouagadougou Water Supply Project 5 5 Case Example: Thailand's Pak Mun Hydro 8 6 India: Shrimp and Fish Culture Project 10 7 Implementation Impeded by Inflation and Wet Seasons 11 8 "Clean Island in Dirty Ocean" 17 9 The Evolution of Public Participation: Warning, through Consultation and Participation to Partnership -The Hydro Case 23 10 Case Example: Mali-The Selingue Dam 23 11 Laos: Nam Theun Hydro - Analysis of Alternatives 26 12 Future Directions of EA 30 Figures 1 The Environmental Management Process 1 2 Cost of EA relative to total project cost 10 3 The Old "Approval" EA System 14 4 New Style EA Emphasis: From Assessment to Implementation 15 5 The Three Main Environmental Units 16 6 Examples of Poverty/Environmental Linkages 22 Tables 1 Historical evolution of transparency and participation: Broadening the constituency of EA design teams-The case of Hydro 7 2 Components of EA Report 18 3 Environmental and Social Assessments 19 iv Environment Department Papers Abstract This paper suggests that the quality of weak or absent. Third, most environrnental Environmental Assessment (EA) of a management priorities lack adequate project -the "EA Report" -is necessary, but financial resources for their implementation. insufficient for successful EA. There are three We offer the case that the preventive and main constraints to successful EA in mitigatory measures, referred to as the EMP, developing countries nowadays. First is lack is the most important part of the EA package. of political will to implement environmental It should contain adequate provisions for safeguards. Second, the institutional capacity budget and capacity strengthening before the to implement environmental management is project can be accepted. Environmental Management Series V Acknowledgments Warm appreciation to our EA colleagues, comments on earlier drafts. Reinoud Post of Glen Morgan, Sergio Margulis, Serigne Omar the Netherlands EIA Commnission kindly Fye, and especially Tom Walton, for commented on the draft. Environmental Management Series vii Introduction The term "environmental impact assessment" successfully. The purpose of this paper is to (EA) was coined with the idea that the foster agreement that mitigation budget and assessment itself of the potential capacity strengthening are essential pre- environmental impacts was the goal. The conditions for all EAs of projects seeking emphasis was on assessment. Once we World Bank support. achieved the then difficult task of predicting an impact, mitigation would follow This paper makes the case that the three somehow. How naive we were in the 1970s! currently weakest links in the whole EA Fortunately, EA (defined below) has evolved. process are first, lack of political will to And so has the scope of environmental issues implement environmental safeguards and taken into account: from smoke pollution, make hard choices between short term and nature conservation, to environmentally development and longer term environmental sustainable development (ESD). The management. The second weak link is assessment of environmental impacts remains management th s end nk is a core element, but now we see that it is only seuri dgetnto imlent vtat a first step. The science of EA has evolved The third constraint is ensuring that such that assessment of impacts is becoming institutional capacity is strong enough to standard. I But once the impact has been implement mitigatory measures effectively. assessed, we need to design a set of In sum, the three major constraints are environmental management--mitigating and political support, money and trained people. monitoring -measures. That step also is This builds on the results of all EA reviews in progressing and not insuperable. The next the Bank (World Bank, 1993, 1996, a 1996b, two steps are the more difficult; namely, 1997, 1998). Ability to assess impacts and securing (a) the budget and (b) the design mitigation are no longer the main professionals to implement the mitigation constraints. Box 1 Running Definition of EA 1. Environmental assessment is an integral part of economic development. EA seeks to promote the sustainability of economic development. EA is basket of tools to improve environmental management and decision-making. 2. EA is a process that produces results. The main result is a set of safeguards against adverse impacts, first preventive -by adjustments during design; second by implementation of mitigatory measures; and third by monitoring and adaptive management during construction, operation, and thereafter. 3. Primarily, EA is designed to conserve or improve environmental services. Secondarily, EA reduces and compensates for the adverse environmental impacts of a proposed project. EA reduces costs by preventing damage, especially that harning the poor, and by reducing delays. Environmental Management Series ix The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results Endnote 1. As there remains disagreement on which impacts are "significant," impacts should be ranked in some order of significance. This fosters agreement over the significance threshold. Where simple ranking will not suffice, then valuation of impacts and comparative risk assessment should be fostered (vide infra). x Environment Department Papers Persuading the Proponent of the Need for EA The difficulty of persuading the project If the proponent of a major infrastructure proponent of the need for EA used to be one project does not see the need for a thorough of the first constraints to successful EA. EA, there are grounds for great concern. Nowadays, as a rule, EA is becoming Proponents lack of acknowledgement that mainstreamed, so persuasion should no environmental needs can be significant is longer be necessary. Unfortunately, often the major barrier to improvement. That persuading the proponent still needs forceful means either the proponent is naive and has attention in some cases. International and not experienced projects of this type before, most national EA standards require EA of all or the proponent has had experience with EA relevant projects from the earliest design and does not want to repeat it. Either way, as phase. If, in residual cases the national soon as proponent reluctance is detected, a legislation does not so require, international rethink is called for. Ownership is essential in standards should be followed if international EA if implementation is to succeed. Weak assistance is to be secured. However, there 'buy in' to the EA process suggests the are no grounds for complacency. The World borrower has little ownership and still Bank (1996) found that ". .there has been regards the EA as a product hurdle to limited progress in achieving ownership for overcome in order to secure the loan. In such EA work among borrowers," and there are cases it would be a waste to prepare an EA some cases ". . .which exhibit negligible that will not be implemented. ownership. . ." Comprehensive national legislation or Figure 1. The Environmental Management Process policies are normally essential for successful EA. Today, if a country lacks an Ideal Sequence Current Status effective national EA policy, that should send a strong signal of systemic weakness, Strategic Rare and should raise a red flag in any proposed trtegieEARare project in that country. In such a case, linking approval of the project to the adoption of national legislation -as a condition of appraisal for example -would Project-level EA Routine greatly increase the chances of project success, and would accelerate national environmental progress thereafter. If this is the first project that relies upon such legislation, then institution strengthening Environmental Post Audit Increasing is that much more urgent. It should be started as soon as possible after project Environmental Management Series The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results identification and continue for as long as commitment should automatically translate needed after project preparation. Experience into provisions for adequate budget and shows that project-specific guidelines and capacity strengthening. Accordingly, it is procedures are not as effective as country- profoundly unsettling if a project proponent level measures. questions the need for an EA. When it comes to the crunch proponents may 'accept' to get If implementation of environmental or social an EA done if they want international mitigation is a provincial rather than a federal finance. But 'acceptance' is not enough. It matter, that too is cause for concern. A shows lack of political will, which suggests national federal-level policy statement or there will be problems in effective mitigation legislation-should normally be considered a after the loan has been signed or after the prerequisite for major EAs. Experience shows project has been built. Of course, it is fine to that fostering project-level policies or remind proponents of the importance of EA, standards is too weak; provincial-level but if they are skeptical, or if it is the first standards often also are inadequate. major EA with which they have been Unfortunately, policies have to be top-down involved, then more attention will be needed. from the national level and adopted- modified or not -by provinces, The related case is where an EA is completed, municipalities, and projects. Proponents and the hurdle for financing is overcome, the contractors may claim that they will comply project documentation looks good, but with all relevant national laws or policies, thereafter the EA is not taken seriously. The even when such laws or policies are EA is a paper report which languishes inadequate. It is the duty of the financier to unimplemented. This is especially dangerous ensure that laws and policies are adequate at if the EA concluded that the currently the national level to protect investments and designed project should not go ahead. This ensure environmental prudence and due point is amplified in 'Implementation' diligence. (Section 4) of this paper. Implementation is fostered by having specific line items of an EA without political commitment is implementation budget integrated into the inadequate, risky, and wasteful. Political project construction budget. In addition, staff commitment should be expressed through are trained specifically to implement and use national legislation, official policies and the implementation budget. Reporting resource allocation. Political commitment is requirements, milestones and spot checks all the most important of the three main support implementation and help ensure that constraints to successful EA. Strong political the EA is implemented. 2 Environmnent Department Papers 2 Predicting Adverse Impacts Until recently, prediction of adverse impacts should read EAs that predict acid deposition was the bulk of what was known as EA. on rice paddies downwind from a new coal- Before then, environmental impacts or fired thermal plant will depress rice yields. externalities were not predicted and their costs fell across all society, especially on the However, governments are not as affluent poor. Internalization of external costs is the they once were, and the scope of governance fundamental goal of the entire environmental has been decreasing over the past 30 years. movement, and the proximate goal of EA. Decentralization proceeds apace, and Formerly, the details of project design were privatization of large sectors of the economy, scrutinized by EA professionals, and the formerly the sole purview of the government, likely environmental impacts or implications is sweeping the world. EA has to keep up of the design were predicted or assessed. with these powerful trends. Intersectoral That is why the process was called cooperation between ministries formerly was environmental impact assessment. Most of easier when central governments were strong EA effort went to predicting adverse impacts. and almost the only stakeholders. Yet Once an impact had been predicted, it was development projects and human assumed that the proponent would populations are increasing. In addition, the automatically desire to prevent it, or the private sector is often the project proponent, appropriate government agency would rather than the government. When these address the problem as soon as it was trends were discerned, the EA process started assessed. That was once the cutting edge of to emphasize mitigatory measures. These EA. Later, by the mid-1980s, the science of EA became formalized in the mid-1980s in eA.olater, byth the acdtuals thedsctien of the "Mitigation Plans". The emphasis began to evolved so that the actual prediction of the shift away from the prediction of impacts, impacts was no longer the major problem. ,, , , , , . . ~~~~and towards what to do about them. What to do about the assessed impacts now is becoming emphasized A new, low cost requirement should be that the impacts should be roughly ranked (Box). A crucial role for EAis to foster inter-sectoral Equal weight for major and trivial impacts coordination. The EA should promote suggests lack of judgment. If a major impact coordination between different line agencies is omitted and minor ones get much to prevent or minimize impacts. For example, attention, that too is a warning that the as soon as the Ministry of Health reads an EA process is flawed. The bigger impacts need which predicts an increase in malaria due to the most attention; the lesser impacts need an irrigation dam proposed for a new region less attention. of the country, the Ministry extends or strengthens its malaria control program in Induced Impacts: Induced or secondary that region. The Ministry of Agriculture impacts frequently can exceed direct or Environmental Management Series 3 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results primary impacts. For example, the Box 2 construction road to Nepal's Arun hydro Internalizing Negative would have been by far the biggest impact of Environmental Externalities this 43 ha. reservoir. Construction camp impact and that of the work force can exceed Mitigation plans deal mainly with 'traditional' en- the impacts of the civil works. vironmental costs and often err in underestimat- ing such costs. EA Timing * Resettlement costs almost always exceed initial estimates substantially (World Bank 1994), and Formerly, the biggest need in the EA process in most cases the oustees are not as well off af- was to assess impacts early on, either during ter the project. Does 'development' demand that the design process or during engineering oustees be, promptly but modestly better off immediately after their move? "No worse off" feasibility. Even now, in significant cases, the implies stagnation at best. EA begins after the project is already * Pneumoconiosis, silicosis, 'black-lung' disease, designed, so progress is still essential with SOx, and NOx have only fairly recently been regard to the timing of the EA. Where the included in C/B of coal-thermal projects. project is designed and only then is the EA * CO2costs are still normally externalized in coal- called in are becoming a rarity. That is thermal projects. Following Kyoto's December team calleo m are Decommg a rar1ty. 1 nat 1S 1998 UN Protocol, the World Bank Group is now success. The EA team now generally starts starting to calculate the GHG implications of all work simultaneously with the project design relevant projects. The use of US$20/tonne of C team. The design and the EA teams work emitted is becoming the norm (Goodland & El together for the whole period of project Serafy 1998), although likely to burgeon shortly. preparation, often two years or more. Many * Minimata disease victims (from industrial mer- EAs require one year of data as a miinimum to cury effluents discharged to the ocean) took c.25 years to be compensated. Australian Aborigi- accommodate seasonal phenomena, such as nes waited over 40 years to be compensated for animal migrations, and this is often damage from UK's nuclear testing on the impossible or overlooked. By far the most Woomera range in the 1950s. effective and inexpensive means of * The benefits of downstream replenishment of environmental improvement are those design soil fertility by annual flooding is still normally changes made for environmental reasons. At excluded from reservoir projects. changes made for environmental reasons. At s In cases of conflict, WTO/GATT normally pro- the end of the design period, the project has tect free trade, rather than internalizing an en- already become much better - with fewer and vironmental concern smaller impacts that need mitigating. Source: Daly & Goodland, 199. Experience shows that this is the way forward for peak effectiveness at lowest cost. This mean that the important measure of If an EA team is called in when the project is moving the dam site upstream had been being designed or worse, has already been foreclosed. That is a recipe for controversy designed, that should be a red flag to all and waste. concerned. While it still may be useful to carry out an EA on an already designed In addition, we must keep emphasizing that project, much of the opportunity of costless EA is not only an activity of project or low cost mitigation may have been preparation, although that is where it starts. foreclosed by then. For example, in the case The focus of EA must now become of the controversial Sardar Sarovar dam on implementation of the Environmental India's Narmada river, the dam's foundations Management Plan (EMP) which includes (a) had already been constructed before the mitigatory measures, (b) monitoring program World Bank was invited to join the project. budget, and (c) capacity building. 4 Environment Department Papers Predicting Adverse Impacts Box 3 forth. Obtaining soft money for the EA The Need to Rank Environmental Impacts perpetuates the myth that EA is an additional imposition. Soft money for EA represents a In the first draft EA-subsequently rectified-of subsidy for EA and should cease. EA should a recent $4 bn. oil pipeline project, the Greenhouse be mainstreamed by now. Gas (GHG) implications of the construction, bull- dozers, trucks, pumps etc. were all carefully cal- culated. However, the GHG implications of min- Fieldng Appropriate Disciplines ing 38,000 cu. mts. of oil per day were omitted. This is sound analysis. Sound analysis means This is an example of emphasizing the less im- portant, and not dealing with the significant. It fielding (a) appropriate disciplines, (b) with also shows how the EA goalposts keep moving. enough experience, (c) at the right times, (d) Until recently, proponents would not dream of for sufficient duration, and (e) with adequate calculating how much GHG they would emit, not resources. The Figure below shows the even from coal-fired projects, although this has evoution The dsgn team, sh o n the been 'best practice' at least in US EPA since 1991 evolution of the design team, based on the (Montgomery et al. 1991; US EPA 1992). Now, example of hydroproject planning. Some since the December 1997 Kyoto Protocol and since decades ago it might have been appropriate publication of the Bank's OD 10.04, an increasing to expect a single environmental generalist or number of projects calculate GHG emissions. Such 'environmental enineer' to undertake the calculations inform decision-makers of this exter-o nal cost, the first step towards the goal of inter- whole EA. No longer. If a river is involved, an nalization. The valuation of environmental costs aquatic biologist or fish biologist should now and benefits has progressed greatly in recent be required (not a fisheries or fish culturist). years. In addition, such calculations used in the 'analysis of alternatives' would tend to promote If resettlement IS sovolved, then an hydro and other renewables, and to demote coal anthropologist or'sociologist experienced in and other fossil-fuelled alternatives. This should resettlement and land use planners will be become standard methodology where practicable. essential. Biodiversity specialists are needed Environmental costs and benefits should be in- if wildlands or other habitat may be affected, cluded in the overall financial and economic evalu- and so on. Now it is generally accepted that a ation of projects as a matter of routine. ajo in.rNowrit e generall need a .major infrastructure project will need a minimum of half a dozen specialists on the EA Budget EA team for a couple of years. EA is a normal part of project preparation. This means the budget for the EA team is part Box 4 of normal project preparation costs. When EA Case Example was new and unknown, development Burkina Faso -Ouagadougou Water institutions tried to help developing countries Supply Project by securing additional extra 'soft' funds to The EA and the engineering feasibility were con- finance the EA in a 'fostering infant industry' ducted in parallel. Consultants prepared the EA approach. While this may have been with the design team. Then Burkinabe Govern- appropriate at the time, it has led to a belief in ment prepared the follow-on Environmental Man- agement Plan. Cooperation between environmen- some quarters that the project proponent will tal professionals, engineers, and economists be given extra funds for the EA. This should improved the design. In particular the width of be resisted. As EA is a routine part of normal the water supply dam's spillway was narrowed. project preparation, it should be financed This reduced the inundation area by 25%, de- project ~~~~~~~~~~~creased involuntary resettlement significantly, from the same budget as the rest of project and diminished other impacts. preparation, engineering feasibility, and Environmental Management Series 5 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results EA scoping can still be done fast and with needed for each rmitigatory measure. only a couple of experienced EA Financing has to be commensurate with need. professionals. Scoping should specify the It is the EMP that should be carefully referred disciplines and their duration needed for to in the legal contracts and budget. Loan each, as well as the necessary interaction with covenants should always -and usually do - all other specialists on the team. For example, specify, as a minimum, that the EMP shall be "one primatologist will be needed for 6 implemented on an agreed schedule, by an months in two periods of three months each agreed budget, by agreed on disciplines. for the wet and dry season or the breeding or Difficulties arise where EMP implementation migrating season, beginning at the same time costs are from counterpart budgets which as the engineering feasibility team starts materialize late or not at all. The Bank is loath work." to resort to remedies on a project because of environmental non-performance. Continuity of staff is important. Continuity of the mandatory Panel of Environmental and It is becoming common to separate the Social Experts also is important. Ideally, the assessment of impacts from mitigation. One same independent external panelists should recent project had one set of volumes for the work together, once or twice a year for a EA sensu stricto, and another set of EMP couple of weeks each time, for five or even volumes several months later. This is ten years, from project identification becoming counter-productive and onwards. Environmental managers working inefficient ..... In fact one can propose the with the scientific EA teams are becoming contrary case to consider the EMP as the more important. The management specialist main document and for relegating the converts the scientific recommendations into prediction of impacts to an annex to be operational programs and ensures they are consulted only for readers wanting details of implementable. why a mitigatory measure is necessary. An intermediate solution would be to insert the Agreeing on Each Measure in the EMP mitigatory measures after each ranked The Environmental Management Plan (EMP) impact prediction. is the single most important part of the EA. 'There is nothing more difficult to predict The package.of mitigatory measures normally placed after the EA itself has now than the future." Even the best designed been integrated into this comprehensive projects cannot plan for each adjustment Environmental Management Plan. The EMP which is made during project budgets for three priorities: (a) mitigation, (b) implementation. In that sense, the EMP's monitoring, and- (c) capacity strengthening. package of safeguard mitigatory measures The EA findings -the identification of should be flexible so that it can cope with impacts - must be followed by the more changes in project implementation. However, important next step, namely the design of this degree of flexibility should be mnitigatory measures for each impact, accompanied by a significant effort on designed as a safeguard against negative monitoring of implementation. Flexibility social and environmental impacts. Each also means contingency budgets are essential mitigatory measure must be accompanied by to finance emergencies or unforeseen an analysis of its associated costs and priorities. benefits. Some mitigations save money to the proponent. The EMP contains a budget for its Several projects have been known to force own implementation. Cost estimates are involuntary resettlement (only two 6 Environment Departrnent Papers Predicting Adverse Impacts Table 1. Historical evolution of transparency and participation: Broadening the constituency of EA' design teams-The case of Hydro Design Team Approximate Era2 Engineers 1930s Engineers + Economists Mid- 1940s Engineers + Economists + then add-on EIS3 to end of complete design Late 1970s Engineers + Economists + Environmentalists & Sociologists4 Late 1980s Engineers + Economists + Environment & Soc. + Affected People Early 1990s Engineers + Economists + Environment & Soc. + Affected People + NGOs Mid-I1990s Engineers + Economists + Environment & Soc. + Aff. People + NGOs + Fully Early 2000s? informed Public "Acceptance"5 I The World Bank's mandatory environmental assessment procedures are outlined in the three-volume "Environmental Assessment Sourcebook" (World Bank, 1991). 2 These approximate dates hold more for industrial nations than for developing ones, although meaningful consultations with affected people or their advocates and local NGOs, and the involvement of environmentalists in project design are now mandatory for all World Bank-assisted projects. 3 EIS are "Environmental Impact Statements." They were added on to the end of a completely designed project - a certain recipe for confrontation and waste. They did not last even one decade. 4 "Environmentalists and sociologists" should routinely include public health specialists, management specialists, anthropologists, community planners, etc. The case can be made that social impact assessment and health impact assessment should be formalized alongside EA. 5 For all new major national projects, all stakeholders - affected people, taxpayers and civil society - must be fully informed at the outset, participate in the decision-making process, and broadly agree with the results. While a national referendum or plebiscite is not necessary for such projects, some measure of acceptance and broad "no objection" will increasingly be necessary. households in the case of the Ghana Urban elements, based on experience and Transport Project; tens of thousands in the commonsense judgment. case of Zaire's Transmission line) that were not "on the books" during project The duration of the monitoring program preparation. This is when appropriate should also be tailored to the nature of the capacity (see below) and appropriate impact to monitor. In the case of the Lesotho monitoring mechanisms are needed. Highlands Water Project (LWHP) in the mid- 1990s, there was the need to monitor progress A good set of monitoring measures contains a in the net income of the resettled people, a comprehensive list of parameters to be requirement that is in line with the imperative measured and a procedure to feed this expressed in the Involuntary Resettlement information back into the hands of the Policy (OD 4.30). Given the difficult land relevant decision makers in the project situation and the poverty of the population to management. The list of parameters, begin with, an income monitoring program comprehensive as it should be, should also be over a period as long as 10 years was designed limited to significant, measurable, achievable and is being implemented. Environmental Management Series 7 The Evolution of Environmnental Assessment in the World Bank - From "Approval" to Results Box 5 Case Example: Thailand's Pak Mun Hydro In 1992 this hydroproject was designed to meet 136MW of peak load between 1800-2000 hrs daily by 1995. The EA team saw that involuntary resettlement numbers were large and the proponent's social record had been mixed in recent years. Affected people tallied 242 families during the design; eventually 2500 families were compensated, many for losses of fish. In addition, the EA team predicted that the reservoir would inundate part of a protected biodiversity reserve. The EA team recommended to the design team that the dam height should be lowered c.10 m., and, that the dam site should be relocated 1.5 km up- stream in order to reduce impacts. These are ac- ceptable mitigatory measures. Those two measures decreased involuntary resettlement, the biggest impact of the project, from 20,000 to 5,000 oustees. 8 Environment Department Papers 35 Securing Implementation Budget Good design is not enough. There are still mitigatory measure, schedule and personnel massive gaps between EMP recommen- for implementation. This should be part of dations and their implementation. total project costs, along with every other Implementation of the mitigation measures cost, and should be included in the legal has to be systematic; it has to follow covenants and loan agreements. established procedures and guidelines. The budget has to be fully integrated into total The standard clauses in the loan agreements project costs before loan signing. A separate should specify that all the recommendations budget for the Environmental Management and the entire EMP shall be implemented on Plan is a warning signal. Having to seek soft an agreed scheduled, using the agreed on money for the EA, or for its EMP, also is a budget, and by the now strengthened clear sign of weak political will. It is no institutions. Payment for performance works longer a favor to seek soft money for EA or effectively in OECD countries where implementation. Fostering an infant industry implementive capacity is adequate. by means of special financing, trust funds or Sometimes this flexibility is preferred over a whatever may have been appropriate until line item budget. The cost of the EA (usually the early 1980s, but no longer. Now it is a less than $500,000 for big infrastructure) is sign of weakness or resistance. If EA is not always small in comparison with the cost of accepted as a fully standard part of normal the project itself, which commonly exceeds project preparation costs, that is a warning of $200 Million (See graph). While cross- future problems. Adequate implementation conditionality is frequent for many aspects of budget is a signal of ownership, commitment performance, it is rare or absent for and buy-in, environmental performance and such as implementation of the EMP. Where construction races ahead while the EMP Environmental performance hinges on the staggers along, such cross-conditionality is timely implementation of the mitigatory called for. That it is rare shows that measures. The implementation schedule environment is still not taken as seriously as should be an integral part of the budget. The other project components. The best example extent to which the implementation schedule is where resettlement is delayed until the is being met should be the focus of reservoir starts to fill. Then resettlement is supervision and monitoring. Budgets have to rushed and oustees are hurt. be time-bound. Much implementation is time sensitive and cannot be left until later in the On the Bank's side, our supervision of project cycle. Implementation must include implementation of the mitigatory measures monitorable indicators, benchmarks, or remains weak. World Bank (1996) found our performance criteria. The critical elements supervision 'lax'. A full 16% of those projects here are clear budget categories for each rated as environmentally 'satisfactory' by the Environmental Management Series 9 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results supervision mission in fact had major problems or worse. Experience shows that a Box 6 professional environmentalist is required for India: Shrimp and Fish Culture Project routine supervision of the environmental This Category B project (IDA $85Million) became mitigation program for all Category A effective in 1992. Most of the project financed projects at least annually. shrimp culture in brackish water in West Bengal, Orissa and Andhra Pradesh. The EA specified the The rule here is that the EMP is the most mitigatory measures; the loan covenant mandated that they should be carried out. But due to delays important part of the EA. An EA can in no in contracting the monitoring, it was not carried circumstances be considered at all adequate out until discovered by the ninth supervision until the EMP is ready (together with budget, mission, which was the first with an environmen- schedule and adequate capacity to tal specialist on it. (World Bank, 1988.) For lack implement). In addition, the institutional of appropriate expertise, eight successive super- vision missions rated the environmental compo- capacity - the professionals expected to nent as "satisfactory" when in fact EMPs had not implement the mitigation -must be clearly even been prepared for the environmental com- identified (see below). If they cannot be so ponent! identified, a stringent institutional capacity strengthening effort is overdue. Effective implementation of EMP means two Grudging compliance or passive acceptance prerequisites: First, clear, firm and integrated of environmental precautions before loan budgets are needed to pay for the signing must be interpreted as a serious implementation. Second, implementation warning that political will is inadequate. needs adequate human capacity. Budgets Bank leverage declines sharply after should be the easier of the two. It should be appraisal and loan signing. If a measure mandatory that the budget to implement the cannot be agreed on in advance, then its EMP be required as an integral part of the budget will evaporate and implementation EMP first, and integrated into total project will be token at best. If the EMP becomes the costs before appraisal. main document, the assessment of impacts can be relegated to annexes which may be Environmental budgets differ from normal consulted by anyone needing to know why a project budgets in that environmental mitigation is proposed, or in order to see if measures and needs often extend well the proposed mitigation is appropriate. beyond construction and into the operation Figure 2. Cost of EA relative to total project cost EA cost vs Project cost 600 -.500 -4-Adjusted 3> 5000 ' / ; ' J j ' 1 | j Basic Data d) 200 IL 100 0 200 400 600 800 1000 EA (k US$) 10 Environment Department Papers Securing Implementation Budget phase of projects. Some environmental Decentralization works the other way too. impacts may become apparent some years Many provincial projects with substantial after the project has started to operate. Then impacts may never come to the attention of decommissioning, rehabilitation and the federal environmental ministry. If the restoration costs, depending on the post hoc project is large or covers more than a single environmental audit, are needed possibly project, it usually would be reviewed by the decades after operation begins. Budget for federal agency. The role of the financing priorities very late in the project cycle (such institution in this case is to agree on as decommissioning, rehabilitation) is implementive responsibility early on, analyze probably best provided for in escrow its capacity, strengthen it where necessary, accounts or performance bonds. But whatever and ensure that the implementation budget the mechanism, audit and decommissioning flows unimpeded and promptly to the agency costs need to be addressed during project tasked with implementation. design, and not left for years later. The most egregious case is when drainage costs of Privatization irrigation projects are needed a few decades after operation when water logging and Now that the private sector is investing in salination has begun to depress crop yields. large infrastructure projects, once the sole purview of the public sector, regulation needs Decentralization to be retained by government or devolved onto a nongovernmental regulatory agency EMP implementation and c ay are (for example, river basin authority), and importantly linked in another way. Th folwdbnheriaedvlpr project design and EA process are often followed by the private developer. copee byafdrlaenywihte Regulatory capacity should match private completed by a federal agency which then dee. ernrneitoifatutr , . . ., ~~~~~~developers entrance into infrastructure assigns implementation to a provincial agency following the start of construction. projects. The Provincial agencies may not be clearly Most of the environment consists of public informed or may not understand that they goods, common property, and open access have to implement the EMP. Or they may resources. It is the government's overriding know, but lack trained capacity or fail to receive their share of the EMP implementation budget from the federal Box 7 agency receiving the loan. Decentralization of implementation Iox 7 implementation needs to be addressed early Iedea bInf on during preparation of the EA and assessment of implementive capacity. Federal A project in an Amazonian country with rampant agencies sometime overlook the need to inflation in the late 1980s contained an implemen- transfer implementation budget to the tation budget that was so slow arriving from the central government to the distant project area that provnmcial agency newly responsible for it became reduced to a tiny fraction in purchasing implementation. Possibly the federal agency power. The deflated budget often arrived with the deems implementation small - something onset of the severe wet season which annually that can be carried out under the provincial halts all field work in any event. The purchases current pre-project budget. The project unit (health clinic building materials) may have arrived responsible for implementation needs with the onset of the rains and deteriorated by the adequate authority, and should not have to time the dry season began some months later. It aequate ah th yh . h took the finance institution almost five years to rely on provincial authorities which may have detect this situation and correct it. little incentive to support the EMP. detect_this_situation_and_correct_it Environmental Management Series 11 The Evolution of Environmental Assessment in the World Bank - From 'Approval" to Results responsibility to pursue the public good. Government is the only institution capable of As the private sector is now often more agile taxing, and often the main one involved in and better financed, leading private sector regulating. The Government's regulatory corporations need to be able to undertake EA, function deals to a great extent with how strengthen EA implementive capacity, and common property may be accessed or used, fully meet or exceed government standards. by whom, and under what conditions. Corporations are not interested in the Analysis of Alternatives. They have done On the other hand, the private sector seeks their own profitability assessment. They have private profit, their own self-interest, and that concluded they want to invest in a certain of their shareholders: Much private sector specific project, not in an alternative to that activity exploits common property. The more project. They are prepared to accept the risks costs that are externalized, the more involved. This is an added argument to profitable is the private sector. Government's promote the use of Sectoral EAs before the function is to balance private sector's private sector becomes involved. As the Bank profitability with public welfare. EA started often recommnends privatization as an off as a governmental tool to internalize element of Structural Adjustment Operations hitherto external costs; But now privatization (a full 40% of bank lending in FY 1997), they is sweeping the globe, EA must adapt. should also promote SEAs beforehand in Government's regulatory function must be order to foster least cost sequences and adequate before privatization takes place, adequate government regulation and especially in mandating the internalization of capacity thereafter. environmental impact costs, and the monitoring of environmental standards thereafter. 12 Environment Department Papers 4 Implementive Capacity Lack of implementive capacity has arguably it had been appraised and approved by the become the biggest constraint to effective Board, citing, inter alia, lack of implementive EAs, although institutional capacity capacity. The proposed environmental and strengthening has been mandated in Bank- social mitigation measures were satisfactory, assisted projects since 1991. During project but implementation of these measures, in preparation, we should be hosting national or addition to increased power rates "would sectoral workshops to foster consensus, have imposed requirements which the Bank financing a systematic series of intensive in- now judges to be beyond what Nepal could country courses, strengthening the best realistically have achieved at present." The national university to offer EA courses, and seven years of preparation of this project sending trainees to regional environmental could have been used to strengthen Nepal's training centers. Sending candidates for 12 implementive capacity. That is the new month master degrees in environmental challenge for EA professionals: How to science also is effective. Donors should strengthen EA capacity before a loan is coordinate to strengthen environmental signed? institutions. Bank policy mandates that the environmental unit in the implementive Overall, implementive capacity is more ministry should be strengthened, as well as important than implementation budget. the environmental ministry. Much can be Budget without capacity will not achieve achieved in capacity strengthening by implementation. Capacity strengthening fostering seminars between the takes a lot longer than securing budget. environmental units of various ministries and Strong institutions demand adequate budget. provincial works departments, as Historically, consensus building among the appropriate, several times a year. For various constituencies has rarely been a example in Laos, it has been very clear for a priority of EA. Local capacity is often decade or more that exporting overwhelmed merely by the EA team's work. hydroelectricity will be the main source of World Bank (1996) found that local EA foreign exchange. This means EA capacity is weak. As soon as the WBG gets professionals of several sorts and levels will serious about a project with a priori major undoubtedly be needed whichever specific environmental impacts, we should start EA hydro goes ahead. Yet, partly because of training, combined with other kinds of Thailand's 1997-1998 economic crisis, capacity strengthening. institutional strengthening has barely started. Capacity building is understood to mean a The new President of the World Bank Group combination of training, institutional took office on 1 June 1995. By August he had strengthening and networking. Isolated canceled Nepal's $lbn. Arun hydro long after individuals or organizations are doomed to Envirornmental Management Series 13 The Evolution of Environmental Assessmnent in the World Bank - From "Approval" to Results failure in this day and age. The start of During project preparation, we should be capacity strengthening is "needs analysis" or hosting national or sectoral workshops to "institutional analysis." This should become a foster consensus, financing a systematic routine part of the EA process and occur series of intensive in-country courses, early on in the preparation phase of the strengthening the best national university to early on in the preparation phase of the offer EA courses, and sending trainees to project cycle. As EA preparation and EMP regional environmental training centers. implementation are often the responsibilities Sending candidates for 12 month master of quite different institutions, the institutional degrees in environmental science also is analysis should identify who will be doing effective. Donors should coordinate to what, how the budget is expected to arrive strengthen environmental institutions. Bank policy mandates that the environmental unit from the central n, adthe in the implementive ministry should be extent to which the implementive agencies - strengthened, as well as the environental which may be provincial or municipal, rather ministry. Much can be achieved in capacity than federal or line agencies -are adequate to strengthening by fostering seminars between their future task. the environmental units of various ministries Figure 3. The Old "Approval" EA System The emphasis was on production of the EA report itself, not on implementation of the mitigatory measures. Phase of Project Cycle Phase of the EA Process Identification Screening Scoping Preparation EA REPORT hazy ~~~~~~~~~~weak implementatio Recommendations institutional budget st~~~~~~~~~~~~rengthening Construction and Operation Some Implementation 14 Environment Department Papers Implementive Capacity Figure 4. New Style EA Emphasis: From Assessment to Implementation Phase of Project Cycle Phase of the EA Process IdentHlaidon Categorization & Scoping Training & Capacity 1| Strengthening P--Ject Prepration Impact Assessment Design of EMP Appra Iii @. V 7EMP Implementationl Schedule Ngodtatlons Loan Covenant & Legal Agreements Operation Suparvlon ~ - ~ Monitoring & Evaluation IL * Post hoc Completion Reviews RestoratiornRehabilitatiornDecommissioning and provincial works departments, as Where to send people is less of an issue than appropriate, several times a year. For getting money to pay for capacity example in Laos, it has been very clear for a strengthening before we are committed to the decade or more that exporting project. Creative thinking will be needed. hydroelectricity will be the main source of proj ative thik ng wilibenee ed foreign exchange. This means EA uInnovation loans, Trust Funds, undisbursed professionals of several sorts and levels will funds from a previous project, LILs, Adaptive undoubtedly be needed whichever specific Lending, all may be useful sources of finance. hydro goes ahead. Yet, partly because of Thailand's 1997-1998 economic crisis, Effective executing organizations are institutional strengthening has barely started. essential for successful implementation of the Environrmental Management Series 15 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results Figure 5. The Three Main Environmental Units Financing Agency Capaciy Strengthening i Environmental Ministry Environmental Unit Capacity strengthening Regulations,monitoring Proponent's Environmental Unit mitigation. This is often the weakest link in unacceptable. The top dozen or so impacts the whole EA chain from analysis to capture most of the mitigation necessary. implementation. We must expect to sponsor This is not to say that all impacts may not intensive EA training courses for all major need mitigation, but that the most effort projects. Such capacity strengthening should should be allocated where it mitigates the ideally take place during project preparation. most. This is difficult as we are not good at financing training until we are sure the An important part of implementive capacity project will go ahead with us. This must is coordination between three entities. First is change. the environmental unit of the proponent or implementive ministry. Second is Based on experience, we now see that coordination with the central environmental effective "Implementation of the EMP" is the ministry or agency. Third is with the MFI or key to successful EA. The EMP is the single Bank's environrment staff. These three groups most important document. The impacts to be should agree on category of EA, screening, mitigated should be ranked, with much scoping and all other phases of the EA, attention to the major impacts, and much less especially implementation of the mitigation. attention to lesser impacts. The ranking of The allocation of institutional responsibilities impacts depends on professional,judgment, between the three entities should be clearly and is important. The specific ranking agreed on and specified at the outset and matters less than ranking categories. It is fine emphasized at appraisal. The responsibilities to categorize 700 impacts (for example, the $3 and budget allocation between the bn., 1040 km Chad Cameroon Oil Pipeline implementive environmental unit and that of Project) into three or so groups -important, the central governments environmental units medium and less important. But an EMP also need clarity. allocating equal attention to 700 impacts is 16 Envirorunent Department Papers 5 Participation World Bank policy is clear: EAs must be Ironically, colleagues interpreting the policy prepared ensuring maximum participation of took those two examples as the only two all stakeholders, especially affected peoples occasions for participation. And many and NGOs. Participation is a very powerful projects fail to ensure participation even at mechanism to foster realistic analysis, those two points. In any policy revision we effective mitigation and conscientious should clarify that those two occasions are a implementation. Participation is highly cost rock bottom minimum. Participation should effective. It even is a means to foster political be a continuous process from identification will. When we were drafting the EA policy though completion and beyond. The and the EA Sourcebook in 1990, we added expectations agreed on must be explicit and two examples of where participation would keyed into project time markers. This be especially effective, namely after the EA encourages both proponent and task category had been assigned, and as soon as manager to stick to implementation the first draft EA was substantially complete. agreements. The details of how participation Box 8 "Clean Island in Dirty Ocean'" The perennial dilemma of Bank-assisted development in general applies also to EA. The Bank is a project oriented institution. We seek to help by financing one project to higher standards than non-Bank assisted projects. The learning experience, so the story runs, helps the sector by example. When we have financed several successful projects in a sector, we may evolve our support to a sectoral project. The evolution from " specific project to a sectoral operation, and from there to macroeconomic support and structural adjustment is one of balance to be perennially adjusted. We may finance a project to higher environmental standards than others in that region or sector over which we have no control. But is that the most cost-effective way of improv- ing environmental quality? The financial increment in the Bank-assisted project might be more effectively invested in neighboring projects to raise their environmental performance somewhat, rather than solely in- vested in very high standards in only a single project. Brazil's Carajas Iron Ore project is a case in point. The borrower achieved very high environmental standards inside their mining concession. Immediately outside their concession illegal artisanal gold mining (using mer- cury), and deforestation for charcoal manufacture led to the classic clean island surrounded by a sea of appall- ing environmental decline. Would it have been more rational to accept slightly lower standards for the Bank- assisted iron ore project, while allocating the money saved to raising the environmental controls of the gold mining and charcoal industries? The answer is to get the initial vision right. Then the investment allocations would follow. Is the vision to supply more iron ore to the market or to conserve the regional environment for longer term benefits? Another example is the coal-fired thermal electricity industries of China or India. Should the EA of the next coal-fired coal plant insist on best available technology for the single new project, or seek marginally to improve the surrounding coal projects? Environmental Management Series 17 The Evolution of Enviromnental Assessmnent in the World Bank - From "Approval" to Results changed project design should be made Close cooperation between the EA and SA explicit and emphasized. teams is essential to ensure that all impacts, such as human disease exacerbated by the Social Assessment project, are properly addressed. The complementarity between EA and SA is Participation also is important as a means to apparent where EA usually has more integrate and build synergize social impact experience with disease vector breeding assessment and the strictly environmental/ habitat, while SA is strong on the biophysical side of EA. Formerly, EA was management of public and preventive health. As preventable diseases are so clearly shorthand for EIA and SIA. In the last year or concentrated among the world's poorest so SIA has come into its own in the World people, EA and SA must join forces to help Bank (Francis and Jacobs 1998). This is redress this inequity. The poor in cities also, warmly welcomed: Let a thousand flowers suffer disproportionately from environmental bloom. SA, which includes SIA, now has its damage such as unsafe sanitation, spoiled own department, staff, procedures, food from lack of refrigeration, toxic fumes guidelines, manuals, training etc. However, from inappropriate fuels and ventilation, city guidelines, manuals, training etc. However, air pollution from industry and vehicles. In the fundamental links between the two have addition, as the impacts of climate change are become more important than when they were appearing, the poor suffer most from weather more firmly linked. and disease. Table 2. Components of EA Report Component World Bank EBRD IDB AsDB AfDB Executive Yes Yes Yes Yes Yes Summary Policy, Legal, Yes Yes Yes Depends Yes Inst. Framework Project Description Yes Yes Yes Yes Yes Baseline Data Yes Yes Yes Yes Yes Environmental Yes Yes Yes Yes Yes Impact Analysis Cost-Benefit No No Yes Yes No Analysis Analysis of Yes Yes Yes Yes Yes Alternatives Mitigation Plan Yes Yes Yes Yes No Institution Building Yes Yes Yes Yes No Environmental Yes Yes Yes Yes Yes Monitoring Plan Consultation Yes Yes Yes Yes Yes Note: The Mitigation Plan, or rather, its implementation, has now become the most important element of the EA process, along with implementation budget and implementive capacity strengthening. Source: World Bank, 1996. 18 Environrnent Department Papers Participation Table 3. Environmental and Social Assessments Environmental Assessment Social Assessment Biophysical Changes Human Impact(s) Pollution Involuntary Resettlement Species Extinctions Gender Impacts Human Health/Disease Vulnerable Ethnic Minorities Greenhouse Gas Emissions Violence / Child Labor Deforestation Community Cohesion Habitat Loss Cultural Property Sanitation Employment and income generation Some important topics need to be closely to improve development through preparation coordinated between SA and EA. For of NEAPs or national environmental priority- example, fish protein resources of the project- setting exercises. But these are not frequent affected communities need to be assessed and (every few years or so), have not yet reached mitigated jointly. The risk of SIA separating full effectiveness, and are not project specific. from EA at the procedural levels may often Participation in NEAP preparation is be greater than the benefits. Now that they important and needs to be emphasized. The are separate stand-alone disciplines, we results should always be used during the should foster SA and EA done jointly or in start up of the next project-specific EA, but parallel by two linked teams working closely will never substitute for participation at the together in time and space. Both have project level. common stakeholders, and institutional needs assessment is often almost fully Participation, especially when it starts from overlapped. Participation involves almost the scoping phase, offers a powerful exactly the same stakeholders in EA and SA. opportunity for involvement of civil society. The benefits of retaining SA/ EA links are Systematic stakeholder analysis is not always clear, and to some paramount. The area of done to ensure all groups and interests have linkages between newly separated EA and SA opportunities. That argues strongly for is a topic for more attention in the near starting EA scoping and participation in the future, and is not amplified here. media, especially newspapers. Whole page inserts in newspapers can outline the scoping Formerly, it was civil society -the grassroots and the topics on which inputs from civil communities and affected people -that society are needed. Such inserts can provide raised attention to potential impacts. But now diagrams of current project proposals and EA has become institutionalized. EA is sufficient information to evoke meaningful normally done by professionals often in less discussion. Inserts also can provide sources of accessible institutions. Civil society's former further information and details of how role has thereby been reduced. Therefore, comments can be entertained. Such media participation also has become a valuable approaches vastly enhance subsequent public opportunity to reaffirm linkages with discussions. grassroots. Civil society has an opportunity Environmental Management Series 19 6 Monitoring, Evaluation, and Audit All EAs should have built-in independent the country or municipality improve effluent monitoring and evaluation mechanisms as quality to EU standards? The Bank's main part of the EMP section. The EMP should means of fostering compliance is through specify monitoring and audit in advance. This training as part of the project. Compliance is preferably set out in an agreed also is fostered by legal agreements on the Environmental Monitoring and Audit design and operation of the project itself, program to ensure the EMP actually gets inclusion of appropriate technology to meet implemented. Monitoring Plans need: (a) standards (for example, electrostatic agreed-on standards based on ambient or precipitators), and systematic reporting. The prevailing environmental quality, (b) reporting is a combination of self-reporting quantitative measurement, or agreed-on by the client and less frequent audits by an benchmarks, (c) systematic reporting by independent entity, which could be the appropriate specialists, and (d) agreed-on government's environmental agency or incentives. The World Bank's Industrial others. These are frequent during Pollution and Abatement Handbook (1997) construction and less frequent during specifies details in such cases. Where operation. In addition, Bank staff spot check quantitative indicators are less appropriate on site from time to time. (for example, municipal solid waste, biological impacts, green-end issues), In more advanced countries and for more specification of the measure and training may advanced polluting industries, incentives for be adequate. Contracts for standards need to compliance include posting performance be precisely written (that is, time-weighted bonds, escrow accounts, fractions of the averages, frequency, cumulative impacts, income stream (such as, electricity sales) methodologies, equipment, relations to allocated to environmental priorities, or for ambient quality, location of monitoring: at environmental performance. We find it the site or near the receivers of the impact, at difficult and often counterproductive to sampling stations, and so forth). withhold payments, to suspend disbursements, or to prosecute, especially in The Bank does not issue precise effluent BOT or BOOT arrangements. A mid-term standards for municipal waste waters as review is highly desirable for major these should be case-specific and should be infrastructure projects as a minimum. tailored and agreed on during the EA process. Often the question is resolved by Monitoring and audits are also excellent ways what a municipality can afford financially. to generate information and knowledge that For example, in Eastern European countries: should be plugged into EA work. EA is should they commit to achieving the EU basically a predictive, risk management Wastewater Directive? Over what period can mechanism. The prediction of impacts in EA Environmental Management Series 21 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results Figure 6. Examples of Poverty/Environmental Linkages Social Impacts Linkages Environmental Impacts Land o normalylackhard,recisequanttativereordedromarojectta-scarcity, io hmn S_ _ Landlessness, \, los Rapid fragmentation t ensoughagon, i n j k ~~~population X Cultivto of/ of the complexityo sgrowth a marginal landse to savings, Hr Shortened d onsacuaicatmon, 2 Enveronment Depl ing D men Pape + -~~~trs fuelwood Pvrty \Reduced a gricultura ( oil and g productivit eromSi < J ~~~~~~~~~~~supply V dpeo Source: Goodland, 1994. normally lack h ard, precise, quantitative recorded from a project that was imple- data, because ecology is young, and because mented long enough ago, is invaluable of the complexity of impacts and knowledge. interlinkages. Hard data on actual impacts 22 Environmnent Department Papers Monitonng, Evaluation, and Audit Box 9 The Evolution of Public Participation: Warning, through Consultation and Participation to Partnership-The Hydro Case 1. Pre-1950s: WARNING One-way information flow: oustees and other affected were warned that they would be flooded or impacted in a few weeks or months time and had to get out of the way for the greater good of distant citizens. 2. 1960s: INFORMATION Primitive participation in resettlement site selection: Oustees were informed that they would be flooded out, and were asked where they would like to move to among a few sites selected by the proponent; compensa- tion often inadequate. 3. 1970s: CONSULTATION Participation in resettlement site selection: oustees were consulted about their impending move, and invited to assist in finding sites to which they would like to move. 4. 1980s MEANINGFUL CONSULTATION Resettlement participation evolves into consultation: Oustees are meaningfully consulted in advance and can influence dam height of position on the river; oustees views on mitigation of resettlement are addressed. 5. 1991: MANDATORY CONSULTATION World Bank's "EA Sourcebook" mandates meaningful consultation in all EAs; EA is unacceptable without such consultation. 6. 1990s: STAKEHOLDER CONSULTATION Resettlement consultation evolves into stakeholder consultation: Stakeholders views are sought on all im- pacts, not just involuntary resettlement. 7. 1992: PARTICIPATION World Bank's EA Policy mandates participation. 8. 1996: World Bank's "Participation Sourcebook" published 9. 1997: Social Assessment separates from EA in World Bank Box 10 Case Example: Mali -The Selingue Dam After fourteen years of operation, the government examined the impacts of the Selingue Dam on the physical and socioeconomic environment. The impacts ranged from hydraulic and energy aspects to ecological, health and economic aspects. This led to the following recomrnendations: * Improve the standard of living of the rural population by providing running water and electrification; * Upgrade agricultural development by proceeding with planned opening of agricultural land and providing technical support; * Encourage better water resources management and coordinating lake activities; * Improve fishing regulation; * Provide better monitoring systems; * Improve natural resources management around the lake; * Reinforce local health plans at the community level. Environmental Management Series 23 7 Analysis of Alternatives Most EAs still are applied at the project level. hydro dam, they too would not be popular. The EA starts when a project has already But the ambit is narrowing. If the coal been decided upon - 'identified' in Bank thermal EA made a compelling case for gas terms. Project-level EA fails to help in project instead of coal the EA team might just may selection. While there is still much flexibility get away with it. Hydro EAs now regularly in design and much scope for mitigation of recommend lowering dam height or shifting impacts, project-level EA is useless in the the dam upstream (see Box 5). selection of the project in the first place. That is a strong argument for promoting the use of The Analysis of Alternatives was added to sectoral or strategic EAs (Goodland and EA procedures rather recently because Tillman 1996), but SEA is not the focus of this proponents erroneously thought that the EA paper. The 'Analysis of Alternatives' should assess impacts of the proponents mechanism was added to project-level EA in design alone. In other words, proponents did an attempt to rectify this weakness, but it is not welcome the EA to propose fundamental often too little and too late. project design changes. Analysis of alternatives is unwelcome at the time of Project-level EA can achieve progress in two project preparation; it should become part of important aspects. First, EA improves the sector work leading to project identification. design of an already agreed on project. Second, EA mitigates residual impacts. An That archaic situation has improved in two EA team contracted to undertake an EA of a ways. First, now that EA begins as soon as new highway would be fired if they project design begins, the two influence each concluded that a rail would be preferable to other positively. But the EA is still the only the road; and a canal or fluvial transport place to show that the proponents design is preferable to both. Yet modal choice in the least cost means to meet whatever goals transport is precisely where the greatest the project seeks. Thus analysis of environmental, social and efficiency gains can alternatives has grown in importance, and its be made. Whether expected or not, that is importance does not sit easily in current where major environmental and social frameworks. Now the private sector has benefits can often be reaped. In the transport grown so enormously in importance, analysis sector, the EA is often interpreted narrowly. of alternatives needs to be overhauled. A good road EA should emphasize mass transit, non-motorized transport (NMT), and Clearly, the government should ensure pedestrian facilities. demand side management (DSM) is well in hand before permitting new generation Similarly, if the EA team of a coal-fired capacity. Next, the government should thermal generating plant recommended a compare the least cost generation method to Environmental Management Series 25 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results meet projected demand such as hydro, natural gas, coal etc. Assuming hydro is Box 11 selected as least cost, then its position on the Laos: Nam Theun Hydro- river and the height of the dam should be Analysis of Alternatives optimized by intenalizing envirounment, In Laos PDR's private sector Nam Theun II hy- social and conventional criteria. The private dro, for example, the government fostered a com- sector is not interested in, nor has any control mendable and comprehensive Analysis of Alter- over, much of that process. For example, natives to ensure that Nam Theun II was indeed whereas DSM may be beneficial for the the least cost priority. The A of A examined im- country, the private dam proponent has no portation of coal from various countries (Austra- experience in DSM, and cannot make money lia, Indonesia), import of LNG from Oman and elsewhere, expansion of domestic lignite thermal from it. The private sector wants to invest $ capacity in Thailand, and Thailand's DSM pro- 'x' in order to generate and sell 'y' Kwh at a gram -among other alternatives. Only then were firm price by a fixed date. The private sector various hydros on other rivers analyzed. At the accepts the risk that the energy will be end of the Analysis of Alternatives study, Nam needed by a certain date and will be salable at Theun's dam height and position on the river were a certain price. Strengthenin goverunment analyzed. How much of such necessary sectoral a certam orlce Streethem vwork can a private proponent be expected to capacity to review EAs and to ensure the undertake? implementation of their mitigatory measures is the main need that the private sector cannot meet. The difficulty now arises, who to minor sets of alternatives, including the should pay for the analysis of alternatives in 'no-project' option. This should all be the frequent case where the private sector minimized by recourse to Strategic EA. seeks to irivest in a specific project? Sectoral EA Analysis of Alternatives usefully occurs at Because Analysis of Alternatives is becoming different levels. The most immediate level unwieldy, it should be replaced with Sectoral addresses other configurations and some EA instead (See SEA in 'Other Conclusions'). technology of the design of the project in E must be SEA n ther Analysiof hand. Different dam heights in hydro, We must be careful not to drop Analysis of different alignments in pipelines or roads, Alternatives before Sectoral EAs have become addition to bicycle lanes and walkways to standard though. Sectoral EAs are more highways: these are often useful. The overall effective, for example, in modal transport level of Analysis of Alternatives overlaps choice than in the EA of a road. Generating with sectoral assessment (SEA), such as choice is more effective in a SEA than in the comparing rail and road alternatives. Analysis of Alternatives of a hydroproject for example. Sectoral EA is the first step towards The timing of Analysis of Alternatives is Strategic EA (Goodland and Tillman 1996; crucial. All too often, major decisions have Verheem 1992,1998, Partidario 1996 a,b). The been made and lots of effort has gone into World Bank acknowledges the power and engineering design of one alternative deemed need of SEA, but we have not historically by economists as the 'optimal choice' years in been leaders in it. The most successful SEA advance. The last thing engineers and has been in Nepal's Power Sector operation managers want is to be forced back to the which ranked -socially and environmentally drawing board and start designing a different -57 hydro projects, DSM, import of alternative all over again. More often than electricity, and fossil-fueled thermal not, Analysis of Alternatives has to limit itself electricity. 26 Environment Department Papers Analysis of Alternatives The record for developing countries has been especially in the power, industry, and less impressive than in industrial countries transport sectors. Coal India has just because financial and human resource completed a Sectoral EA for all its 33 coal constraints hamper adoption of SEA. Bilateral mine operations, under a World Bank and multilateral aid agencies have started to assisted project. The Bank's (1992) EA encourage EA further upstream in planning Sourcebook has new sections on Sectoral EA, for developing countries. In the World Bank, EA of International Treaties, and drafts for Sectoral EAs are becoming more common, Regional EAs under preparation. Environmental Management Series 27 8 Conclusions Unarguably, the EA process has become the 3. Strengthen Capacity to Implement: This major meeting point between development takes the longest so should be started decisions and environmental management. It during project preparation under a LIL or seems likely to remain so in the future. It is undisbursed funds, or elsewhere. the responsibility of EA managers to exploit Institutional analysis is needed to show this opportunity for mutuality with decision- which institutions should be strengthened makers. Now that decision-makers realize (federal, provincial, municipal, private that environmental management is essential sector, academia etc.). National and to sustainable development, the EA process regional EA seminars can serve the dual can expect more support in the future. purpose of scoping the EA and starting Environmental specialists have to show the EA training based on today's project. As win-win solutions and how risks, costs, financing agencies are reluctant to finance impacts and delays are always reduced by a component unless the overall project prudent environmental management. The EA has been signed, we will have to seek process needs to adapt to the rapidly flexible and innovative means to changing nature both of economic overcome this understandable impasse. development and of environmental priorities. The next step in this Work-in-Progress will be the REDEEM project of Jean-Roger The three major conclusions are: Mercier: Refining the Design of Environmental Management Plans. 1. Implementation of Mitigation: Implementation of the mitigatory Other Conclusions measures specified in the EMP section of Project-Level EA needs to be continued and the EA is by far the most important part strengthened. In particular it must influence of the entire EA process. The mitigatory project design. Analysis of alternatives and measures should be ranked and agreed fully costed implementation of the EA's upon, scheduled, disciplines identified mitigation plan during project construction and their duration specified. and operation must be made more effective. 2. Implementation Budget: Each mitigatory Regional EA and Cumulative EA processes measure, or at least all those measures at should continue to be strengthened, and the top of the rank, should have careful more frequently used. and individual cost estimates. The total environmental implementation budget Sectoral EA phase-in should be accelerated in should be a fully integrated part of the order to reduce the cost and increase the overall cost tables of the project. effectiveness of the benefits of project-level Environmental Management Series 29 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results EA. Sectoral EA is a fast but low-cost and unsustainble projects should be dropped powerful tool to help in project selection. SEA from further consideration. This implies that improves both economic C/B and least cost sustainability analysis should be done on the analysis. Only sustainable projects should be portfolio of potential projects before SEA addressed during Sectoral EA and begins. Box 12 Future Directions of EA These first three refer mainly to project-level EA; they are traditional but reactive. * First, strengthen project-level EA, and ensure it becomes a project design influencing tool, parallel to and equivalent with engineering feasibility, and never a post hoc justification or cosmetic for a previously de- signed project. * Second, EAs need to be used when a specific region (for example, water basin or province is slated for a number of development projects. This becomes a "Regional EA." * Third, Cumulative EA should be routinely and systematically applied when a currently proposed project will be added to existing projects in the general area, and specifically takes foreseeable projects into account. Synergy between proposed, existing, and foreseeable projects are addressed in CEA. The next two refer to Strategic EA and apply EA beyond the project level. * Fourth, EAs of entire sectors, such as the power sector or the transportation sector, need to be completed, preferably before selecting the next project in that sector. This is the role of Sectoral EA, a subset of Strategic EAs. Economic least-cost analysis is an effective phase to add SEA. * Fifth, Environmental sustainability needs to be fostered by many means, one of which is EA. A major intel- lectual effort is needed to see how to foster sustainability and the role of EA in that process. Both Strategic EA and Sectoral EA can be used to promote sustainability by selecting only those potential projects that are sustainable. * Sixth, EAs need to be used in policy and program formulation, such as in designing structural adjustment, or in policy-based lending. EAs also should be used in national priority-setting exercise, such as the national budgeting process and in national approaches to environmental sustainability. EA of policies, programs, national budgets, legislation, and international treaties is "Strategic EA" (SEA). 30 Environment Department Papers Guide to further sources of information Arquiaga, M., Canter L. and Nelson D. 1994. CEAA, 1997. Cumulative effects assessment: Integration of health impact Practitioners guide. Ottawa, Canadian considerations in EIS. Impact Assessment Environmental Assessment Agency 12: 175-197. 68++p Bakkes, J. A. 1994. An overview of CEQ 1997. Considering cumulative effects environmental indicators: state of the art under the National Environmental Policy Act. Washington DC., Council on and perspectives. Nairobi, UNEP, Environmental Quality 64++p. [Cambridge Univ; Bilthoven, Nat. Inst Publ. Health & Emvironmental Daly, H. E. and Goodland R 1994. Ecological- protection] EA Tech. Rept 94-01: 72 p. economic assessment of deregulation of international commerce under GATT. Barber, M. C. 1994. Environmental monitoring Population and Environment 15(5):395- and assessment program: Indicator 427; 15 (6): 477-503. development strategy. Washington DC., US EPA/620/r-94/022: 74 p. Doyle, D. and Sadler, B. 1996. Environmental assessment in Canada: frameworks, Bojo J and Ekbom A. 1998. Poverty and procedures and attributes of effectiveness. Ottawa, Canadian eashinvionment: work d inBprog . Environmental Assessment agency, 44p. Washington DC., World Bank.v Ekbom, A. and Bojo, J. 1997. Mainstreaming Borrini-Feyerabend, G. (ed.) 1997. Beyond environment in country assistance strategies. fences: seeking social sustainability in Washington DC., Africa Environment conservation. Gland, Switzerland. IUCN, group, Discussion Paper 1: 32 p. WWF. World Bank, TNC, USAID, WRI, CARE: vol. 1: 129 p. Elkin, T. J. and Smith, P. 1988. What is a good environmental impact statement? Canter, L. 1996. Environmental impact Reviewing screening reports from assessment. New York, McGraw-Hill 600 p. Canada's national parks. Journal of Environmental Management 26: 71-89. CEARC, 1985. Social Impact assessment: a El Serafy, E. 1993. Country macroeconomic research prospectus. Ottawa, Canadian work and natural resources. Washington Environmental Assessment Research DC., World Bank, Environmental Councill-85/SIA: 17 p. Working Paper 58: 50 p. Environmental Management Series 31 The Evolution of Environmental Assessment in the World Bank - From "Approval" to Results Erickson, P.A. 1994. A practical guide to Goodland, R. and Tillman, R. 1996. Strategic environmental impact assessment. San environmental assessment: Strengthening Diego, Academic Press 266 p. the EA process (1-34) in Goodland et al. EA in Africa. Washington, DC., World Finsterbusch, K. 1995. In praise of SIA. Bank 169 p. Impact Assessment 13: 229-252. Goodland R. and Edmundson V. 1994. Fox, J. and Brown, L.D. 1998. The struggle for Environmental assessment and accountability [of the World Bank development. Washington DC., World Group]. Boston MA., MIT Press 551+p. Bank 159 p. Francis, P. and Jacobs, S. 1998. Goodland, R. and El Serafy, S. 1998. Institutionalizing social analysis at the Internalization of carbon dioxide World Bank. Manchester UK, Institute of emissions costs. Ecological Economics Development Policy ms: 17 p. (December). Gilgan-Hunt, E. and Mercier, J-R. 1998. Goodland, R. 1997. The strategic Sectoral environmental assessments in environmental assessment family. ECI: Sub-Saharan Africa. Washington DC, The Magazine of the IEA and EARA. World Bank, Africa region 60 p. (September): 17-20. Gilpin, A. 1995. Environmental impact Goodland, R. 1998. The biophysical basis of assessment: cutting edge for the 21st environmental sustainability (14p.) in van century. Cambridge, Cambridge Univ. den Bergh, J.C.J.M. (ed.) Handbook of Environmental and Resource Economics. Glasson, J., Therivel, R. and Chadwick, A. London, Edward Elgar Publisher. 1994. Introduction to environmental impact assessment: principles, procedures Hyman, E.L. and Dearden, K. 1998. and prospects. London, Univ.College of Comprehensive impact assessment London Press 342 p. systems. World Development 26(2): 261- 276. Goodland, R. 1994. Environmental sus- tainability: imperative for peace (2-46) in James, D. 1994. The application of economic Graeger, N. and Smith, D. (eds.) Environ- techniques in enyironmental impact ment, Poverty, Conflict. Oslo, Inter- assessment. Dordrecht, Kluwer Academic national Peace Research Council: 125 p. 292 p. Goodland, R., Mercier, J-R. & Muntemba, S. Jubb, C. and Underhill, B. 1990. Environmental (eds.) 1996. Environment assessment in assessment: impact on major projects. Africa: a World Bank commitment. Canberra, Bureau of Industry Economics, Washington DC., The- World Bank, Research Report 35: 61 p. AFTES: 169 p. Lee, N. and Colley, R. 1992. Reviewing the Goodland, R. and Sadler, B. 1996. The quality of environmental statements. analysis of environmental sustainability: Manchester, University of Manchester, from concept to application. Int. J. EIA Center, Department of Planning and Sustainable Development. Landscape, Occasional Paper 24: 46 p. 32 Enviromnent Department Papers Guide to further sources of infonnation Lee, N. and Hughes, J. 1995. Strategic Sadler, B. (ed.) 1996. International study of environmental assessment legislation and the effectiveness of EA. EA in a changing procedures in the community. Brussels, world; Evaluating practice to improve European Comm. 2-vols. performance. Ottawa, CEAA and IAIA: 248 p. Lee, N., Walsh, F. and Reeder, G. 1994. Assessing the performance of the EA Sadler, B. 1998a. Sustainability analysis. process. Project Appraisal 9(3): 161-172. Environmental Assessment 6(1): 12-14. Mercier, J-R and Anjali, A. 1998. World Bank Sadler, B.1998 b. Aide memoir to Group's assistance to EA capacity sustainability analysis.. EA: the magazine building in Sub-Saharan Africa. of the IEA and EARA (March): 12-13. Washington DC., World Bank (in preparation). Sadler, B. and Boothroyd, P. 1994. Traditional ecological knowledge and environmental McCold, L and Holman J 1995. Cumulative impact assessment. Vancouver, impacts in environmental assessments: University of British Columbia, Center for How well are they considered? The Human Settlements 75 p. Environ. Profess. 17: 2-8. Sadler, B. and Verheem, R. 1996. Strategic Montgomery, J., Soloman, B.D., Smith, J.B. & environmental assessment: status Berish, C.W. 1991. Addressing global challenges and future directions. The climate ihngm NEPA rreviews.TThe ciaechange inNP eiw.TeHague, Min Housing Spatial Planning Environmental Professional 13: 114-124. ante Envionment and the Envirornment Morris, P. and Therivel, R. (eds.) 1995. Scholten, J. 1996. Reviewing environental Methods of environmental impact impact assessment reports. Canberra, assessment. London, UCL Press 378 p. Epa. OECD, 1994. Towards coherence in environmental assessment. Paris, DAC 3 Spaling, H. 1995. Cumulative effects vols. assessment. Impact Assessment 12: 231- 251. Partidario, M. R. 1996. Strategic environmental assessment: key issues Spaling, H. and Smit, B. 1993. Cumulative emerging from recent practice. Environ. environmental change: conceptual Impact Assess. Rev. 16 (1): 31-55. frameworks, evaluation approaches, and institutional perspectives. Environmental Partidario, M. R. 1996. SEA-regulations and Management 17: 587-600. guidelines worldwide (15-29) in Therivel, R. and Partidario, M. R. (eds.) The Suter, G. W. & Barnthouse, L.W. 1993. practice of strategic environmental Ecological risk assessment. Boca Raton, assessment. London, Earthscan 206 p. FL., Lewis Publ. 538 p. Roe, D., Dalal-Clayton, B., and Hughes, R. United Nations, 1996. Current policies, 1995. A directory of impact assessment strategies and aspects of environmental guidelines. London, IIED/WRI/IUCN/ impact assessment in a transboundary OECD/DAC: 84 p. context. New York, United Nations 74 p. Environmental Management Series 33 The Evolution of Envirormental Assessment in the World Bank - From "Approval" to Results United Nations. 1994. Trends in Wood, C., Barker, A., Jones, C. and Hughes, J. environmental assessment of energy 1996. Evaluation of the performance of projects. New York, United Nations the EIA process. Brussels, European DDSMS, Energy Branch 79 p. Commission DG XI (B2): 2 vols. UNEP, 1994. Sub-Regional workshop on Wood, C. and Djeddour, M. 1992. Strategic environmental impact assessment for environmental assessment: EA of policies, Commonwealth countries of Eastern and plans and programmes. Impact Assess. Southern Africa. Nairobi, UNEP, Bull. 10(1): 3-21. Envirounment and Economidcs Unit 70 p. World Bank, 1991. Environmental assessment US EPA, 1992. Climate change and sourcebook. Washington DC., World environmental assessment. Washington Bank, 3 vols. DC., US Govt., Environmental Protection Agency (Region IV): 149 p. World Bank, 1996. Participation sourcebook. Washington DC., World Vanclay, F. and Bronstein, D. (eds.). 1995. Bank 259 p. Envirounmental and social impact World Bank, 1993. Annual review of assessment. Chichester, J. Wiley 325 p. environental assessment: 1992. Van Pelt, M.J.F. 1993. Ecological sustainability Washington DC, World Bank, and project appraisal. Aldershot, Environment Department 66 p. Avebury Ashgate Publ. 280 p. World Bank, 1998. Involuntary Resettlement Verhaheyen, R.F., Nagels, K., and Coenan, M. in Development Projects, World Bank 1997haheyen, i R.F., Nagels,K. assessmentand MTechnical Paper Number 80, Washington 1997. Integration of risk assessment and DC, The World Bank 88 p. health impact assessment in environmental assessment. NATO World Bank, 1996. The impact of Environmental Impact Assessment; environmental assessment: the World Wilrijk, University of Antwerp & Bank's experience: second environmental Reykjavik National Planning Agency: 99 p. assessment review Washington DC, World Bank 146 p. Verheem, R. 1992. Environmental assessment at the strategic level. Project Appraisal World Bank, 1996. Effectiveness of EAs and 7(3): 150-156. National Environmental Action Plans: a process study. Washington, DC., The Wade, R. 1997. Greening the World Bank: the World Bank, OED (June) No. 15835: 158p. struggle over the environment, 1970-1995. (Ch. 13: 611-734) in Kapur, D., Lewis J. P. World Bank, 1997. The impact of and Webb, R. (eds.) The World Bank: its environmental assessment. Washington first half century. Washington DC, DC., Environmental Department, Brookings Institution Press, 2 vols. Technical Paper 363: 146 p. Wood, C. 1995. Environmental impact World Bank, 1997. Pollution prevention and assessment: a comparative review. New abatement handbook. Washington DC., York, Wiley 337 p. Environment Department 344 p. 34 Environrnent Department Papers Guide to furtier sources of information World Bank, 1998. India: Review of the effectiveness of environmental assessments in World Bank-assisted projects FY 1990-97. Washington DC., World Bank, South Asia Environment Unit 63 p. Environmental Management Series 35 tnvironment Department The Wtrld Bank 1iS1 8H Street, N.W Washington, D.C. 20433 202 473,3641 2024770565 FAX Printed on 100% post-conisumer recycled poper