Abstract
This paper draws on a survey of the 32 environmental impact assessment (EIA) reports approved in Malawi after six years of mandatory EIA. With the aim of assessing the extent to which the requirements of the EIA process have been complied with and determining which EIA components influenced the apparent compliance, it is shown that the quality of the reports is poor, with most of the EIA requirements being unsatisfactorily adhered to. The components that contributed to the poor quality are analysis of alternatives, consultation and public participation, scoping and environmental audit. Substance problems that affect the adequacy of the EIAs, practice problems dealing with the administration of the EIA process, and contextual problems and deficiencies in the EIA law are specifically identified as the flaws responsible for the poor quality. It is thus recommended, for EIA quality to improve, that the flaws be addressed, and particularly the factors which are inducing them.
1. Introduction
Environmental impact assessment (EIA) is an instrument for aiding and improving the decision-making process for development planning as well as a tool for safeguarding and managing the environment (Bisset, Citation1995; Gilpin, Citation1995). The way EIA is applied internationally is that it is prescribed by some national or supranational legislation or policy that designates a competent authority-administered process of responsible and circumspect development undertaking, involving the prior preparation and submission of EIA reports about the likely environmental, social and health effects of proposed projects as a basis for the projects' approval. The essence of the legislation is to regulate the undertaking of EIA by setting minimum regulatory requirements to the various components of the EIA process (e.g. scoping). This means that EIA undertaking is supposed to ensure compliance with the requirements both to the letter and the spirit of the legislation.
In Malawi, EIA is provided for under the Environment Management Act (EMA) (Malawi Government, Citation1996), and it is administered by a central government agency, the Environmental Affairs Department (EAD). Six years after enactment of the EMA, the EAD had called for and approved about 32 EIAs. Much as that would signify a step towards fulfilling the EMA's goal of subjecting projects to environmental scrutiny, it is important to evaluate the EIAs in order to ascertain how efficient they have been. That is to say, has their undertaking been true to the letter and the spirit of the EMA? Or, more precisely, have the regulatory requirements of the EIA process been complied with? The principal aim behind such evaluation is to provide feedback to a process that is inherently prone to flaws, and to provide the much-needed opportunity to constantly improve EIA at the policy, procedural and technical levels (World Bank, Citation1997).
The purpose of this paper, which is drawn from a study undertaken by the author (Mhango, Citation2003), is to report the findings of an EIA report review survey of the above-mentioned 32 EIAs. The survey had the main objectives of assessing the extent to which the requirements of the EIA process have been complied with, and determining which EIA components influenced the apparent overall process compliance. But first, an overview of the EIA procedure in Malawi is appropriate.
2. EIA procedure in Malawi
The procedure commences with screening, following the proponent's submission of a project brief or a project submission document in the case of public sector developments. The EAD in liaison with the Technical Committee on the Environment (TCE), a body composed of members appointed for their objectivity, credibility and special knowledge in a particular field, screens the project against prescribed criteria set under a category list of EIA mandated projects, and also determines whether or not a full-fledged EIA should be undertaken (). (Originally, the list of mandated projects consisted of EIA-mandated projects and projects for which EIA may be required. However, owing to the lack of clear-cut distinctions between the two categories, a one-list category of EIA-mandated projects only has since been adopted.)
Figure 1: EIA procedure in Malawi
Where a fullyfledged EIA is ordered, the proponent undertakes scoping to identify possible impacts and alternatives and any key issues that may have to be considered in the EIA appraisal. In this phase, public consultation is first undertaken through public hearings at a place the EAD deems necessary for the purpose of assessing public opinion about the project and its consequences. The proceedings from scoping make up the substance of the terms of reference (TORs) on which the EIA appraisal is based. When the EAD approves the TORs, the proponent (usually through hired consultants) carries out the EIA appraisal and subsequently produces an EIA report, which covers project description, feasible and preferred alternatives, consultation and public participation (CPP), impacts identification, prediction, evaluation, interpretation and mitigation, and provides a comprehensive environmental management plan (EMP).
Once the report is submitted, the EAD initiates a review and refers the report to the TCE, which assesses the adequacy of the information supplied in the report against prescribed criteria and own knowledge for the purposes of making an informed decision about authorising or rejecting the project. The review process may also include public scrutiny of the EIA reports at a hearing again summoned by the EAD. Depending on the scope and complexity of the project, an independent review panel may also be retained to advise the TCE. The granting of a development licence conveys the authorisation of a project.
Finally, in almost all cases whether licensed or not, some form of follow-up has to be instituted when the project starts to be implemented, in one or more of the following forms: EMP implementation; environmental monitoring; and environmental audit. However, this is not explicit and has been left at the discretion of the EAD.
3. Methodological considerations
Bearing on the fact that the study from which this paper is drawn was an evaluation exercise, a discussion of the methodological considerations in EIA evaluations is appropriate. According to MacLaren Citation(1985), there are six possible dispositions that an evaluator of EIA can take, which are:
| • | checking the accuracy and authenticity of impact predictions. | ||||
| • | determining the effectiveness of mitigation measures. | ||||
| • | evaluating the effectiveness of monitoring and surveillance. | ||||
| • | examining the measures adopted to involve the public in EIA. | ||||
| • | assessing the response of EIAs to regulatory requirements. | ||||
| • | providing data to permit others to benefit in future projects of a similar nature. | ||||
Any one of these dispositions involves re-examining retrospectively accumulated EIA experience (Ortolano & Shepard, Citation1995; Sadler, Citation1988). This implies that the evaluator endeavours to expose how previous EIA appraisals have performed. Therefore, intrinsic to any EIA evaluation is the question of how one measures EIA performance.
Basing his list on MacLaren's work, Ortolano Citation(1993) identified seven approaches to measuring EIA performance involving the review of previously approved EIA reports. The reports are reviewed for:
| • | compliance with applicable regulatory requirements. | ||||
| • | completeness with respect to the agreed scope of the EIA work to be undertaken. | ||||
| • | adequacy of the methods with regard to guidelines, peer review, judicial review. | ||||
| • | influence on the attention given to environmental factors in decision making; | ||||
| • | influence on the decision of whether to approve, reject or modify the project; | ||||
| • | cost-effectiveness. | ||||
| • | contribution to sustainable development. | ||||
The methodology followed in this paper was based on the first premise: ‘why review EIA reports?’ The essence of EIA is to collect information about the probable consequences of proposed projects and to report them to decision makers prior to project implementation in order to facilitate sound and informed decision making. The mode through which the consequences are reported is EIA reports, and hence reviewing the reports is an important indicator of the performance of the EIA process.
The EIA Centre at the University of Manchester Citation(1999) defined reviewing an EIA report as the process of assessing the adequacy and quality of individual EIA reports. Whereas adequacy entails using the review as a means of quality control, to check how satisfactory each report actually is and to detect significant deficiencies in it, quality entails evaluating the report against a review framework in a structured and systematic manner for either decision making or research and monitoring purposes. Some of the renowned review frameworks in the world include the Lee & Colley environmental impact statement (EIS) quality review package (Lee et al., Citation1999), the European Commission's (EC) EIS review checklist (Commission of the European Communities, Citation2001), the International Association of Impact Assessment's (IAIA) EIA effectiveness checklist (Sadler, Citation1996) and the Devuyst EIA process review package (Devuyst, Citation1994). Even though some of these packages were designed for specific countries or regions, they have on occasions been used in other countries, for example the Lee & Colley package used in Ireland (Dancey & Lee, Citation1993) and in Portugal (Canelas, Citation1992). However, owing to variations in formal EIA regulations between countries, and differences in available resources and variable objectives/goals for undertaking EIA reviews, it is platitudinous to outline review checklists taking into consideration a particular EIA regime's appertaining minimum regulatory requirements, especially when one is to adapt principles from renowned review packages (Devuyst, Citation1994).
4. Methodology
All the concerns raised above were taken into account in the methodology used in this paper. More precisely, the methodology adapted principles from the EC and IAIA's packages on: (1) the framing of review topics and categories under which review criteria are presented in the form of either questions or presumptive sentences; (2) the inclusion of minimum regulatory requirements pertaining to the particular EIA regime and good EIA/EIS practice; and (3) the description of the assessment symbols to be used, in which case the review marked whether a particular criteria or requirement was present or not in the report in terms of the adequacy of information supplied supporting the requirement. (This stand was adopted in case the reports are just perfunctory paperwork gimmicks of EIA undertaking, especially bearing in mind the factual observation that if no supporting information or reasoning is given for assertions in an EIA report, then by definition of good decision-making, the assertions are invalid no matter what their foundation is; Lambert & Wood, Citation1990). This, to some extent, also addressed the question of the reliability of the methodology and its consistency with international standards and review methodologies.
4.1 Review of EIA reports
The EIA reports which were reviewed were selected from the EAD's inventory of projects whose EIAs had been approved at the time of the study (i.e. April 2002). In total, 32 EIA reports were reviewed (see ).
Table 1: EIA reports studied
The review was actually undertaken through evaluation checklists/matrices which were designed primarily to depict the minimum Malawian EIA regulatory requirements for the following EIA components: screening, scoping, EIA terms of reference (TORs), analysis of alternatives, impact analysis, consultation and public participation (CPP), monitoring and evaluation, and environmental audit as provided for either under the EMA or the guidelines for EIA (Malawi Government, Citation1997). (See sections 5.1.1–5.1.8 below and Appendix 1.)
The review derived ‘yes or no’ scores which were determined in terms of whether a particular requirement under each EIA component was satisfied or not. Thus, under each EIA component, a percentage compliance grade (C) was determined by proportioning the number of requirements each EIA satisfied:
where S was a single count of each satisfied requirement and n was the total number of considered requirements under that component. Then, for each EIA, an average compliance grade (x¯) was determined as a mean of all its component compliance grades:
4.2 Hypothesis testing
A hypothesis that ‘the quality of EIA is low’ was framed for the analysis. (This was based on the fact that the main objective of EIA in Malawi is facilitating project compliance with the requirements of EIA; Malawi Government, Citation1997.) Because quality was referred to with respect to EIA performance quantified as the compliance of the EIAs with EIA requirements, a determining point where an EIA report would be considered as being of low or high compliance status was appropriate. Therefore a 50 per cent cut-off point was devised. This implied that the null hypothesis was translated so that ‘the mean compliance grade of the EIAs was equal to 50’ (i.e. Ho: µ = 50). Thus the mean (X¯) of the average compliance grades for the EIAs was computed and tested against a test value of 50 at 5 per cent significance level:
where µ was 50; s was the standard deviation of the data; n was the sample size (32); and where x¯ was each EIA's average compliance grade.
4.3 Principal component analysis
Finally, in order to determine components of the EIA process that contributed to and/or influenced the apparent overall process compliance, a statistical technique called principal component analysis (PCA) was used. This was for the purpose of recommending measures for enhancing the compliance of the most influential EIA components as a basis for improving or strengthening the apparent overall EIA compliance. The PCA was intended to retain EIA component variables that brought out most of the variation in the compliance of the EIAs with EIA requirements by generating a correlation matrix (R) on the compliance data set:
where where k = 1,2…p; j = 1,2…p; and where the numerator was the covariance between any two k and j variables and the denominator was the variance of any of the k and j variables. A set of PCA rules for retaining the most influential components was then employed.
5. Findings and discussions
5.1 The quality of EIA
Results from the EIA report review are given in .
Table 2: Distribution of EIA compliance grades (%)
Using a one sample t-test, the analysis revealed an insignificant p-value of 0.000 (i.e. p < 0.05). Consequently, the null hypothesis was rejected, implying that the performance of the EIAs, as far as complying with minimum EIA regulatory requirements is concerned, was low or poor, with the mean compliance across the study population falling way below 50 per cent. In fact, 93.75 per cent of the EIAs had a below-50 per cent compliance grade. This poor compliance was deduced to smack of flaws in the general practice of EIA in the country. In order to give a more detailed picture of the flaws, an analysis was performed to compare the number of projects that failed to satisfy any of the minimum regulatory requirements under each EIA component against those that satisfied some requirements. (The worst and better-performed components were determined by a criterion of 50 per cent, where the worst component was categorised as one that had 50 per cent plus of the projects failing to satisfy any of its requirements (i.e. not compliant).) provides the results, and major findings with each component are discussed below.
Figure 2: Performance of the EIAs/Projects (N = 32) with respect to the EIA process
5.1.1 Screening
Screening was one of the better-performed EIA components, with none of the projects failing to satisfy any of its requirements. In fact, description of the project and the existing or affected environment were the best-documented screening requirements. Most of the projects also did manage to highlight possible effects that would arise and how they were likely to be dealt with. However, some requirements were not well documented, including: a brief account of the presumed benefits and detriments/costs of the project in terms of ecological as well as socio-economic aspects, which dissuades the decisionmaker from drawing informed conclusions about calling for an EIA for the project or not; consequential developments to the project and their likely effects, which make the consideration of cumulative effects even more difficult; consideration of other potential uses of the natural resources being exploited in the project; and the issue of the effect of the project on local communities (e.g. in terms of land acquisition, involuntary resettlement, etc.). For instance, despite being surrounded by four villages, the Mpatsanjoka Dambo project involved land coverage of 6,000 hectares, no consequences to the public were considered in the EIA. In this particular case, the failure to consider this aspect does not reflect what would be considered good EIA practice, and further such oversights undermine EIA's purported intention of promoting sustainability with respect to implications for social disruption.
Be that as it may, some EIAs attempted to document some of these issues well. For instance, the ecotourism project on Mulanje Mountain considered the effects of consequential developments such as handicrafts, bee-keeping and cane harvesting, while the Area 32 infrastructure project considered other potential uses of the land on which the project was to be commissioned, such as a proposed construction of a sports arena and a proposed forest reserve. The Area 32 project EIA went even further and handled the issue of land acquisition by proposing compensations to the Department of Forestry, which was originally entitled to the land.
Aside from these issues dealing with adequacy of the EIAs in terms of regulated requirements, there are some issues dealing with EIA administration aspects that were also observed. On a positive note, it was encouraging to notice – on the part of the EIA administrator – that besides relying on legislated lists of EIA-mandated projects; they exercise rational judgement with respect to the project's environment of implementation when recommending projects for EIA. A good case in point is the project involving the Victoria Hotel, whose room capacity did not warrant conducting an EIA, but which was nonetheless subjected to one on the basis of the hotel's proximity to a watercourse (the Mudi river) and some human settlements (Sunnyside and Blantyre police residential areas). On a negative note, it was observed that there is limited or skewed application of EIA in the country. Most of the projects so far appraised for EIA are from within Lilongwe where the EIA administrator is stationed. There is a high likelihood here that projects in other parts of the country go without EIA simply because they are not monitored or policed by the EAD. Examples of such projects would include the government's southern regional offices commissioned in 2000 in the city of Blantyre, and the Chichiri shopping mall complex commissioned in 2001, also in Blantyre. While this could be attributed to the EIA administrator having inadequate resources (human and financial) to police such projects for EIA application and enforcement, it is undoubtedly true that without an explicit link between development planning regulations and EIA (as is the case), such policing would indeed be onerous. In Malawi there are a number of laws that mandate the inception of all sorts of development endeavours by way of some government line authority giving consent – for example, the chief inspector under the Factories Act CAP 55.07; the town planning officer under the Planning (Sub-Division Control) Act CAP 59.04; and the commissioner under the Mines and Minerals Act CAP 61.01) – yet under the EMA, which was preceded by most of the development laws, the line authorities are required or expected to obtain further authorisation for their projects from the EAD (another government line authority). The problem with such a set-up is that it leaves room for conflict of authority, which is likely to lead to situations where the decisions for and/or of EIA are easily overlooked or the EIA procedure is used more as a reactive than a proactive tool. What is also problematic is the difference in philosophy of planning and EIA, especially as the latter is commonly perceived as a preventive means to either stop development altogether or modify it, which may conflict with planning authorities' interests or development agendas (Rivas et al., Citation1994). It is important to note that strategic environment assessment (a form of EIA for plans, programmes and policies) is yet to be instituted in Malawi.
5.1.2 Scoping
Scoping was the second worst-performed component, with 73 per cent of the EIAs failing to satisfy any of its requirements. In actual fact, only four (the Mangochi bridge, the ethanol distillery plant, the Kapichira hydroelectric scheme, and the 132 kV transmission line) of the 32 EIAs documented scoping exercises (through stakeholder meeting minutes and official correspondence documents), yet scoping is the phase that defines the EIA by determining its range of activity or application and content or substance. The rest of the EIAs mainly only recounted that stakeholders were consulted, without giving proof of the same, for example in terms of the consultees, the scoping plan, venues, avenues, days, etc. As a result, it is impossible to appreciate whether scoping is indeed undertaken in the course of summoned EIAs. Much as the EIA administrator could be held liable for not enhancing the documentation of purported scoping exercises, the undertaking of scoping is likely to be disregarded in the absence of an enforceable regulatory mechanism compelling developers to genuinely carry out and authenticate scoping consultations.
For those EIAs that documented scoping consultations, it was observed that alternatives and probable effect mitigation measures were discussed. However, the consultations were mainly engaged with government authorities or other private companies, in most cases without evidence of input from the public. For instance, for the ethanol plant EIA, scoping consultees only included personnel from a similar development initiative, the Illovo ethanol plant in Dwangwa. While it is beyond the scope of this paper to document why most of the EIAs seemed to have sidelined the public in their consultations, some other findings in the country have suggested that it may be because of preconceived notions that the public is either too illiterate to contribute productively to the consultations or too unconcerned to care about what is really going on (Mhango, Citation2003). Both these notions are ill-founded, because the public can contribute greatly by allowing the developer to have access to indigenous knowledge, which could later be to the developer's advantage, as the public is most likely to cooperate and be willing to be involved if their ideas are seriously considered and incorporated. Past experiences have shown that sustainability of development activities is promoted or frustrated by people's conception and perceptions (Lidskog, Citation1997; Shepherd & Bowler, Citation1997; World Bank, Citation1997).
Other poorly covered scoping requirements included: documentation of the team that carried out the scoping exercise and the methods through which that was done; and documentation of the key issues (contentions as well as concessions) that emerged in the scoping consultations. The documentation of the team is important because it guides the decision maker in appreciating that the appropriate expertise was consulted and retained, and the documentation of the key issues demonstrates that divergent views were entertained and then amicably resolved.
5.1.3 EIA terms of reference (TORs)
TORs in the Malawian EIA procedures are required in both the tender documents and the final EIA reports. The reason for the former is for bidding amongs EIA experts/consultants, and for the latter for ease of reference when making out whether the EIA appraisal did address or perform the tasks set out for the EIA and with what exactitude.
From the analysis, the component TORs was slightly poorly performed, with 34 per cent of the EIAs failing to satisfy any of its requirements. As a matter of fact, as many as 12 of the 32 projects under study did not include TORs in their EIA reports. This not only makes it difficult to appreciate whether the TORs were proposed or not, but also to appreciate how the reports were effectively reviewed by the appropriate decision makers. For most of the reports that included the TORs, it was disappointingly observed that the TORs were generally generic or merely copied the examples of TORs given in the guidelines for EIA (Malawi Government, Citation1997) word for word. This leads to missing out on the substance of a project and the subsequent EIA in terms of what specifically ought to be done, and also reinforces the fact that the pre-TOR phase, scoping, is rarely genuinely undertaken. Again encouraging, but only slightly, was the observation that most of the TORs complied with the requirement to document the EIA's areas/issues of address and the requirement to consult the public and propose mitigation measures. However, specifics in the TORs were massively excluded. Specifics such as:
| • | a description of the project, which sheds light for bidding consultants on the nature of the project for which the EIA work is required and hence what that work would entail. | ||||
| • | a consultation strategy (i.e. possible consultees), which again gives foresight to bidding consultants on the social fabric of the area in which the EIA work would be undertaken. | ||||
| • | a time frame for the EIA, which has much to do with being aware of time, and with job competence on the part of bidding consultants. | ||||
| • | finally, the requirement of a multidisciplinary EIA team (i.e. specific expertise required in relation to the nature of the project and its possible consequences), which assists the developer in making an informed decision as to which consultants have the capacity to perform the job. | ||||
The disregard of these specifics leads to the conclusion that consultants bid and are awarded jobs of whose pertinent challenges and impediments they are unaware, which is most likely to compromise the quality of the work and the product they will produce. It ought to be noted here that the purpose of the EIA TORs is as much about accountability in the implementation of EIAs as it is about the investor putting his/her money where it is worth putting. In other words, it is the investors' opportunity to choose, from the environmental consultancy industry, professionals capable of executing the job efficiently. Hence, the TORs ought to be looked at as screening criteria or benchmarks for experts, and so must be specifically framed and not merely duplicated. A few examples of EIA reports where specifics were highlighted in the TORs include the ones for the infrastructure project at Lengwe, Majete and Mwabvi game reserves, where a description of the project was included, and for the Bua irrigation project where a list of possible consultees was specified, as well as a 30-day period for the EIA.
5.1.4 Analysis of alternatives
The examination of alternatives for the projects was slightly poorly performed, with 44 per cent of the EIAs failing to satisfy any of its requirements. Indeed, 14 of the 32 EIAs did not consider any alternatives at all, yet exploring project alternatives is the more proactive side of EIA as it enhances project design by providing for design modifications as opposed to the more defensive, difficult and costly task of reducing the impacts of a given design (World Bank, Citation1997).
In the case of the few EIAs that did consider alternatives, it was encouraging that the alternatives mainly concerned possible sites, design and construction options for the projects, and also possible uses of the resources being exploited. For instance, the Rose Farm Greenhouse EIA considered alternatives in terms of possible sources of water for irrigation, and the Victoria Hotel and the Lilongwe Casino Hotel EIAs considered alternatives based on uses of the land on which the projects were to be implemented as allocated by the planning departments in the Blantyre and Lilongwe city assemblies, respectively.
However, one of the discouraging observations about analysis of alternatives was that the requirement to evaluate the ‘no action’ alternative was massively ignored. This suggests that the analyses are often undertaken to justify a decision that has already been taken to commission a development project, rather than to propose and adopt the most feasible and sensible course of action. In fact, the ‘no action’ alternative is like the opportunity to analyse economic investment endeavours from the contemporary perspective of dire necessity for the development with respect to present as well as future generations (i.e. inter- and intra-generation equity). Another discouraging observation was the exclusion of supporting data in the analyses, which reveals that analytical approaches are rarely used. In fact, the commonly used approach was that of environmental merits and demerits of the projects, which otherwise assists in justifying the preferred alternative, but undermines the analysis as the justification is not founded on a broad range of viewpoints (e.g. social, cultural and economic ones). It should be remembered that EIA is primarily supposed to counterbalance the biophysical and socioeconomic aspects of a proposed development endeavour. An example of an EIA where the analysis was on a broad range of aspects is the one for the Area 2 commercial industry project, where amenity, social development and urbanisation reasons were all considered. Examples of the few EIAs that attempted to use analytical economic methods are for the infrastructure project at Lengwe, Majete and Mwabvi game reserves, which used the cost score ranking method; for the Mangochi Bridge project, which used the approximate cost for construction and maintenance, construction period, structural characteristics, availability of local materials, and land acquisition and compensations options to select the preferred site; and for the Mulunguzi dam project, which used the cost-effectiveness, least cost approach to justify the preferred site.
5.1.5 Impact analysis
Impact analysis was better performed, with only about 5 per cent of the EIAs failing to satisfy any of its requirements. Because impact analysis is the nucleus of any EIA appraisal, it was analysed under four crucial subcomponents: impact identification, prediction, significance and mitigation. gives the results.
Figure 3: Performance of the EIAs/Projects (N = 32) with respect to impact analysis
The well-performed parts of impact analysis were impact identification and mitigation, with only 6 and 9 per cent of the EIAs, respectively, failing to satisfy any of the requirements. It was encouraging to note that the majority of the EIAs documented impacts, and the manner in which this was conducted was well founded, as the impacts were identified with respect to stages of the project maintaining the cause-and-effect strategy commonly employed through the potential impact matrix method. However, identification of impacts in pre- and post-construction stages of the projects was poorly attempted. This implies that EIA teams mostly concerned themselves with the consequences of putting up a project than with the consequences of the project once put up. This code of practice undermines the very concept of environmental sustainability for which EIA is earmarked because, as there is a reaction to every action, there is no justification whatsoever to assume that a project would have no impacts in post-construction phases. In fact, even though they may not be severe by contrast, post-construction consequences are the most vital as they are usually cumulative in nature over a period of time.
In addition, supporting data and scientific or analytical methods were rarely used in the identification. The methodologies commonly employed were expert judgements, matrices and checklists, logical deductive processes and literature reviews of projects of a similar nature. Indeed, the Area 3 flower plantation EIA based its impact identification and significance on similar projects undertaken in Zambia and Kenya. This, however, undermines the effectiveness of impact analysis, because even similar projects are bound to have differences because they are implemented in different circumstances and environments. The requirements for social, economic and cumulative impacts were also poorly addressed. Again, this could be attributed to the limited use of scientific and technical methods, and also the limited multidisciplinarity of EIA teams, which moreover do not dovetail with the nature of the projects. Further, most of the EIAs did not consider indicators or measurable parameters for the identified impacts. An example of an EIA that included indicators is for the Namiwawa hotel project, which had indicators and standards for ambient noise levels. These, however, were again considered only for the construction phase of the project.
It should be noted that the emphasis on the use of technical, scientific and analytical approaches is not meant to undermine other methods such as expert judgements, matrices, etc. (especially noting that the method for assessing and evaluating impacts depends on the availability of financial resources; the type of impacts and their potential significance implying that in some cases descriptive approaches would suffice, and in others quantitative scientific methods would be ideal), but rather to underscore the point that over-reliance on subjective and descriptive approaches is likely to lead to disputes (sometimes legal, which may not only delay the project but exert unnecessary financial burdens on the investor); for example, decision-makers contending that the EIA expert's judgement is incorrect or miscalculated. In other words, these approaches increase the possibility of interference or sheer prejudice and disputes. On the other hand, much as the discrediting of findings from analytical techniques is not entirely ruled out, with the analytical approach this is not so likely, unless the methods themselves are perceived as unreliable or wrongly used or, more likely, as having been wrongly interpreted.
Impact mitigation was well performed, although not all impacts identified were superseded by a proposed measure of mitigation. Also stakeholder/public input in identifying mitigation measures was not adequate, even though some EIAs attempted to address this issue; for example, the Kapichira hydro scheme project stated having had a two-day EIA workshop to facilitate stakeholder input. However, the workshop proceedings were not documented in the report, which makes substantiation of the same not easy. Mitigation measures were also well addressed in the EIA for the infrastructure project at Lengwe, Majete and Mwabvi game reserves owing to consideration of mitigation measures for dam construction and ground water extraction, which were identified as feasible alternatives to the provision of water (which was a component of the project).
The worst-performed parts of impact analysis were impact prediction and significance with 75 and 63 per cent of the EIAs, respectively, failing to satisfy any of the requirements. The majority of the EIAs did not document the magnitude, significance or severity of the identified impacts. This not only makes the authenticity of the impacts, in so far as whether they are really likely to occur, impossibly difficult, but also hampers the auditability or assessment of the effectiveness of the mitigation measures and makes it difficult to know how to prioritise the identified impacts (Dipper et al., Citation1998; Wood, Citation1999; Wood and Dipper, Citation2000). It is worth recalling that the analysis of impacts is, first and foremost, supposed to direct whatever efforts of mitigation are proposed to the critical issues of the project and provide guidance for prioritising them. Without the prediction and rationally justifying the significance of an impact one cannot, with certainty, make out which consequences of a project are critical and warrant a high degree of priority. Only in the one case of the micro-nutrient and health project EIA was impact significance determined through the United Nations Environmental Programme impact significance formula, which has validity as it employs the contemporary methods of environmental or resource economics, which are used to only a limited extent in EIAs even though they are largely advocated and preferred (World Bank, Citation1997).
Astonishingly, almost all of the EIAs did not document possible unmitigated impacts, implying that the proposed mitigation measures would address all consequences of the projects. This is impractical given that EIAs are to a large extent associated with uncertainties (Ashby, Citation1976; Hellstrom & Jacob, Citation1996), which means that the accurate anticipation of EIA outcomes is simply not possible (Devuyst, Citation1994). Further, most of the reports did not document the EIA teams, which made it impossibly difficult to appreciate whether the multidisciplinary approach, advocated in EIA, was indeed engaged to identify the divergent issues to do with the projects' consequences. Examples of the few EIAs that documented multidisciplinary experts are the Rose Farm greenhouse project, which had a chemist, a civil engineer and an environmentalist, and the Victoria Hotel project, which went a step further and documented the educational and professional credentials of the EIA team members.
5.1.6 Consultation and public participation
CPP was yet another slightly poorly performed component, with 38 per cent of the EIAs failing to satisfy any of its requirements. Almost 20 of the 32 EIAs did not document any public consultation and the subsequent results regarding public participation, yet it is the key to ensuring that the EIA addresses issues of relevance to local communities who are the primary beneficiaries and victims of any change a development project may bring about.
Most of the EIAs only mentioned that the public was consulted and that they participated, without specifying who the consultees were, or when and how they participated and were consulted. Besides this, possibly because of factors such as limited public consultation notification, awareness, knowledge, participation skills and the funding for CPP programmes (Mhango, Citation2003), the lack of comprehensive mandatory regulatory procedures for consulting the public in the Malawian EIA law is a major flaw. This is so especially considering the status quo of the extant provisions for CPP in the EMA, more precisely sections 25(3), 26(1)(a) and 52(1) and (2). Primarily, these provisions are so much concerned with cautioning about the right to access EIA documentation that they preclude the basic and most important precaution of empowering the public to take part in summoned EIAs. This means going beyond conferring on the public a right to be consulted and actually enabling them to play lead roles in identifying, designing, directing and implementing any development project that may have an impact on their immediate environment, and therefore their way of life. Second, the provisions are passive as they are only concerned with the accessing of information, which is not only a post-EIA activity, but can also be beneficial only if there is a vigilant public and a supportive media cognisant of their legally conferred rights and duties, which is not the case in Malawi. The problem with public vigilance is, among other things, compounded by the lack of legal standing in the EMA (see Nyasulu, Citation2000). Lastly, section 26 (1)(a) is not only silent about CPP in other stages of the EIA process (e.g. scoping and the EIA appraisal itself), but also makes CPP discretionary and lacks fundamental aspects about compelling genuine undertaking of CPP. Such fundamental aspects include defining whom to consult, when to consult, where to consult, how to consult, what to consult about and, most importantly, thorough documentation of what transpires in the consultations.
For the few EIAs that documented CPP, the consultees were predominantly government authorities and not the public or their representatives [e.g. chiefs or non-governmental organisations (NGOs)]. On the part of the NGOs, one of the contributing factors is their lack of enthusiasm, because the requirement for their involvement in EIAs in the Malawian EIA procedures is voluntary as opposed to being obligatory. Indeed, despite a large number of registered environmental affiliated NGOs (Malawi Government, Citation2002), the Wildlife and Environmental Society of Malawi is the only NGO that frequently participated in the EIAs reviewed in this study. It was also observed that consultations were predominantly engaged in the pre-EIA stages (e.g. project proposition, land acquisition for the project, and to a limited extent to scope issues of address) as opposed to during the EIA appraisal. In addition, most of the consultation methods employed involved the one-way communication mode of the use of the media and questionnaires. These usually translate more to efforts to inform than to involve the public.
5.1.7 Monitoring and evaluation
Monitoring and evaluation was assessed with respect to the requirements for an EMP. Encouragingly, only about 19 per cent of the EIAs failed to satisfy any of its requirements, and as many as 22 of the 32 EIAs managed to provided an EMP highlighting monitoring measures for the identified impacts, following the sensible impact-monitoring measure-responsible official cost approach. For instance, the Rose Farm greenhouse EMP proposed surface water and soil sample collection and analyses for the concentration of nitrogen, phosphorous and potassium in Lilongwe river to monitor activities from the project such as fertiliser application; the Namiwawa Hotel EMP proposed sample collections and analyses of liquid effluents in the Mudi River during project construction; the ethanol plant EMP proposed periodic tests for pH, temperature, alkalinity, etc. for the wastewater to be produced from the plant before its discharge into Shire River; and the Bunda Quarry EMP included impact indicators or parameters, collection points and permissible standards (national and international) to be used to monitor the activities from the project. This is an indication of follow-up measures being seriously considered in EIAs in the country. However, the vital question of whether the EMPs are eventually implemented, and implemented to the spirit of the letter, is worth investigating.
Be that as it may, a good number of the EMPs were weak on handling such issues as inclusion of impact monitoring indicators, and monitoring other components of the environment other than water. For instance, the EMP for the ecotourism project on Mulanje Mountain concentrated on monitoring water pollution in Likhubula River and neglected all other identified impacts. Further, some common misconceptions are prevalent in environmental monitoring. For example, instead of providing an EMP, EIAs of some projects were proposing its establishment, whereas some EMPs that were proposed failed to distinguish between mitigation and monitoring measures or between a monitoring measure and a monitoring activity. For instance, while treating industrial wastes to reduce toxicity before discharge into a watercourse may count as a mitigation measure for a water pollution impact, the periodic and spatial collection of water samples from the watercourse for toxicity analysis very much counts as a monitoring measure for the impact and the mitigation measure. Further, while proposing the collection and analysis of water samples may be a monitoring measure, specifying what to analyse (e.g. levels of biological oxygen demand) counts as a monitoring activity. All these issues need to be clearly spelt out and not confused if monitoring is to be effective.
On another note, the status quo of the extant provision for environmental monitoring (i.e. section 28) in the EMA leaves a great deal to be desired. The section provides only for the monitoring of projects that existed before the law was enacted, thereby excluding all post-EMA projects, which could have repercussions on how serious developers take the requirement for environmental monitoring, and also legal implications in case of conflicts between opposing factions in a dispute.
5.1.8 Environmental audit
Environmental audit was the worst-performed component, with 93 per cent of the EIAs failing to satisfy any of its requirements. In fact, only two (the Mangochi Bridge and the Mulunguzi Dam) of the 32 EIAs documented programmes of environmental audit, yet it is the component that guarantees the effectiveness of the EIA process by promoting feedback, and monitoring the ‘putting into action’ of the recommendations or mitigation/monitoring measures stipulated in EIA reports.
This state of affairs could be attributed to the fact that environmental audit in the Malawian EIA law is not required explicitly, and that it has been left at the discretion of the EIA administrator who, in effect, carries out an environmental audit on the basis of projects being a cause of public concern after public complaints are filed. However, with a public that is poorly abreast of their legal rights (as the case is in Malawi), it is very unlikely that EIA vigilance on projects would bear fruitful outcomes, and also that the impacts identified in EIAs and the mitigation measures proposed thereof would indeed be effective no matter how well articulated they are.
5.2 Components influencing EIA performance
gives the results from the PCA analysis.
Table 3: Extracted PCA components (Total Variance Explained)
Following Mardia et al. Citation(1979) and Jolliffe Citation(1973), who recommended ‘retaining components that contribute to 96 per cent of the total variation in the data’ and ‘retaining those components whose eigenvalues are above a certain threshold value of 0.70’, the first five components were extracted. Cattell Citation(1966) proposed producing a ‘scree graph’ of the eigenvalues against the components to determine where large eigenvalues cease and small ones begin (see ).
Figure 4: Scree graph for the EIA compliance data
Subsequently, the first six components were extracted. Finally, following Kaiser Citation(1958), who recommended ‘retaining components with eigenvalues less that the average of the total eigenvalues (which is 1 in the case of the PCA using a correlation matrix)’, the first two components were extracted. Hence, using the rule of thumb for a compromise of the extractions from the above rules (Mardia et al., Citation1979), the first five components were, appropriately, the ones that were retained as the principal components influencing the variation. This implied that the dimensionality in the data was reduced from eight to five variables. Consequently, in order to discover the variables, a correlation matrix was produced which was then transformed by rotation (see ) to make interpretation of themes of the influencing factors and hence overall EIA performance from the matrix more straightforward.
Table 4: PCA component matrix (Rotated Component Matrixa)
From , component 1 (explaining 35.934 per cent of the variation) was dominated by CPP, which involves seeking and considering views about the project from interested and affected people, and alternative analysis, which involves consideration of all feasible alternatives to the project, including those from interested and affected people. The theme coming emerging from this factor seemed to be the involvement and participation of all stakeholders in the project. Component 2 was dominated by screening, which involves describing the project, including its objectives, size, activities, etc. so that an informed decision as to whether an EIA should be summoned or not is made. This was the second most influencing factor (17.265 per cent of the variation), and the theme that was represented was that of the level and type of assessment the project requires or warrants. Component 3 was dominated by scoping, which involves specifying the scope and identifying key issues to be addressed in the EIA. The third most influential factor was therefore seen as the preliminary consideration of key issues to dominate the assessment of a particular project (11.995 per cent of the variation). Last but not least, component 4 was dominated by impact analysis, which involves identifying, predicting, evaluating and mitigating the effects and risks of the project. The factor represented here was the actual integration and coherence between impacts the project is likely to have and measures the project is to institute, in order to minimise or otherwise offset them (10.924 per cent of the variation). Finally, component 5 was dominated by environmental audit, which involves examination of data collected from the project for purposes of assessing compliance with the terms and conditions of the EIA, so as to learn from experience and improve EIA. This was interpreted as the evaluation of the project after the EIA is conducted and the project is implemented (i.e. follow-up), and it was the last most influential factor in the EIA compliance performance (9.661 per cent of the variation).
Of the eight EIA components, the ones influencing the overall compliance in order of magnitude were CPP and analysis of alternatives, screening, scoping, impact analysis and environmental audit. The components that contributed greatly to the overall variation (i.e. 57.590 per cent) were CPP, analysis of alternatives, scoping and environmental audit, which were also the ones in which most of the EIAs under study performed poorly or failed to comply with any of their requirements (section 5.1). On the other hand, screening and impact analysis that had few of the EIAs failing to comply with any of their requirements contributed to the least variation (i.e. 28.189 per cent). This means that the apparent low or poor process compliance was influenced predominantly by CPP, analysis of alternatives, scoping and environmental audit.
6. Conclusions and recommendations
This paper has revealed that the quality of EIA reports in Malawi is poor as a result of the unsatisfactory undertaking of most of the components of the EIA process with respect to compliance with their minimum regulatory requirements. In fact, apart from screening and impact analysis, the requirements of scoping, EIA TORs, analysis of alternatives, CPP, monitoring and audit are largely inadequately adhered to. This poor compliance or quality of EIA emanates from three major flaws: substance problems that affect the adequacy of EIAs with respect to regulated requirements; practice problems in dealing with the administration of the EIA process; and contextual problems and deficiencies in the EIA legal and/or regulatory mechanisms. Several attributes accrue to these three major flaws, including, respectively:
| • | limited capacity, especially in terms of knowledge and comprehension of the requirements of EIA (both regulated and technical) among EIA stakeholders in general and EIA practitioners and decision makers in particular, probably owing to inadequate training, experience and sensitisation and also commitment to good EIA practice. | ||||
| • | inadequate capacity (financial and human) within the administrator institution of the EIA process, which has limitation implications for the application, monitoring and enforcement of EIA; the passiveness of EIA watchdog institutions in policing projects that may require EIA and defending public interests in EIA consultations; the lack of synchrony between processes of EIA and planning. | ||||
| • | the lack of mandatory regulatory procedures for some crucial components of EIA such as CPP and scoping and the exiguous status of extant provisions for some components of the EIA process such as CPP, environmental monitoring and audit, which makes their implementation impossible to achieve. | ||||
These observations have implications not only for the quality of EIA reports, but also for the effectiveness of the entire EIA process. And, as the quality of a product reflects the effectiveness of the process through which it is produced, the EIA process can principally be improved by addressing the above-noted flaws, particularly by improving the factors that are inducing them. It is thus recommended that:
| • | comprehensive sensitisation of the requirements of EIA be promoted among all stakeholders, particularly the use of analytical techniques during the analysis of impacts and alternatives, and also that the EAD should critically review the analysis of alternatives, to ensure that the preferred alternative is justified only after comprehensive synthesis of all feasible alternatives to a project. | ||||
| • | the review process of EIAs be strengthened to ensure that the requirements of EIA are adequately complied with by improving capacity within authorities and EIA practitioners, and also that the EAD should consider adopting formal review mechanisms such as the EC's review checklist to ensure that the review is transparent, immune from possible interference and prejudice, and that a report is assessed with respect to its articulation and coverage of different aspects of EIA and that the decision to approve a project is based on overall assessment | ||||
| • | capacity within the EAD be strengthened so as to increase the ability to police projects for EIA application nationwide and monitor them, and particularly that environmental district officers who are stationed in each district in the country should be empowered to take up this role, which should not be a problem given the devolution of powers to districts that is taking place under the decentralisation process in the country. | ||||
| • | planning laws should be reformed in order to streamline them with the provisions and requirements of EIA. | ||||
| • | post-EIA research (e.g. evaluation of the accuracy of impact predictions, the effectiveness of EIA methodology, monitoring the implementation of EMPs) be promoted in order to strengthen EIA quality control in the country. | ||||
| • | EIAs thoroughly document scoping and CPP exercises to ensure that the assertions about them are authentic, and also that the EAD should consider adopting a mandatory consultation and participation procedure with a strengthened regulatory capacity (e.g. the already-set-up development committees in each village in the country could be revitalised to make certain that the public really gets to be involved). | ||||
| • | the legal requirements for environmental monitoring and audit be improved by revising the EMA to include the provision of environmental monitoring for post-EMA projects, and also that the EMA stresses the current environmental audit provision (which translates more to reactive environmental auditing) or distinguishes it from EIA auditing which, as part of the broad field of environmental auditing, is a proactive evaluative tool for any commissioned EIA (Dipper et al., Citation1998; Morrison-Saunders, Citation1996). | ||||
Acknowledgements
Gratitude is due to Mrs Meya Kalindekafe and Dr Eston Sambo of the University of Malawi for all their labours in supervising the research from which this paper is drawn, and Mr L. Kazembe and Dr J.J. Namangale also of the University of Malawi for their constructive advice on statistical options for analysing the data collected for the research. Dr Dimitri Devuyst of the Free University of Brussels, Professor Christopher Wood and James Walmsley of the EIA Centre of the University of Manchester are also to be thanked for supplying and recommending publications on EIA evaluation instruments. Special appreciation goes to the Eastern and Southern African Universities Research Programme (ESAURP) for supporting the research with a grant, and the Department of Environmental Affairs in the Ministry of Natural Resources and Environmental Affairs in Malawi for their support and for permitting access to the EIA reports reviewed for the research.
Additional information
Notes on contributors
Solani Dennis Mhango
Mchiji District Assembly, Private Bag 1, Mchinji, Malawi.References
- Ashby , D . 1976 . “ Background to environmental impact assessment ” . In Environmental impact assessment , Edited by: O'Riordan , T and Hey , R D . Westmead, UK : Saxon House .
- Bisset , R . 1995 . Environmental impact assessment: a guide to good practice , Nairobi, , nKenya : UNEP .
- Canelas , L . 1992 . “ Assessing the quality of Portuguese environmental impact statements ” . In EIA Newsletter No.7 , Manchester : EIA Centre, University of Manchester .
- Cattell , R B . 1966 . The scree test for the number of factors . Multivariate Behavioural Research , 1 : 245 – 76 .
- Commission of the European Communities . 2001 . Guidance on environmental impact assessment: EIS review , Luxembourg : Office for Official Publications of the European Communities .
- Dancey , R and Lee , N . 1993 . The quality of environmental impact statements in Ireland: a comparative analysis . Project Appraisal , 8 ( 1 ) : 31 – 6 .
- Devuyst , D . 1994 . Instruments for the evaluation of environmental impact assessment , Brussels : Vrije Universiteit . PhD Thesis
- Dipper , B , Jones , C and Wood , C . 1998 . Monitoring and post-auditing in environmental impact assessment: a review . Journal of Environmental Planning and Management , 41 ( 6 ) : 731 – 47 .
- Eia Centre . 1999 . “ Reviewing environmental impact statements ” . In EIA Leaflet No.11 , Manchester : EIA Centre, University of Manchester .
- Gilpin , A . 1995 . Environmental impact assessment: cutting edge for the twenty-first century , Cambridge : Cambridge University Press .
- Hellstrom , T and Jacob , M . 1996 . Uncertainty and values: the case of environmental impact assessment . Knowledge and Policy , 9 ( 1 ) : 70 – 84 .
- Jolliffe , I T . 1973 . Discarding variables in principal component analysis I: artificial data . Applied Statistics , 21 : 160 – 73 .
- Kaiser , H F . 1958 . The varimax criterion for analytic rotation in factor analysis . Psychometrika , 23 : 187 – 200 .
- Lambert , A J and Wood , C M . 1990 . UK implementation of the European Directive on EIA: spirit or letter? . Town Planning Review , 61 ( 3 ) : 247 – 62 .
- Lee , N , Colley , R , Bonde , J and Simpson , J . 1999 . Reviewing the quality of environmental statements and environmental appraisals , Manchester : EIA Centre, Department of Planning and Landscape, University of Manchester . Occasional Paper No. 55
- Lidskog , R . 1997 . From conflict to communication? Public participation and critical communication as a solution to sorting conflicts in planning for hazardous waste . Planning Practice and Research , 12 ( 3 ) : 239 – 349 .
- Maclaren , J W . 1985 . “ Setting the stage ” . In Audit and evaluation in environmental assessment and management: Canadian and international experience, Vol. 1 , Edited by: Sadler , B . Banff, , Canada : Environmental Canada and the Banff Centre School of Management .
- Malawi Government . 1996 . Environment Management Act , Lilongwe : Department of Environmental Affairs .
- Malawi Government . 1997 . Guidelines for environmental impact assessment , Lilongwe : Department of Environmental Affairs .
- Malawi Government . 2002 . State of the environment report 2002 , Lilongwe : Department of Environmental Affairs .
- Mardia , K V , Kent , J T and Bibby , J M . 1979 . Multivariate analysis , London : Academic Press .
- Mhango , S D . 2003 . Environmental impact assessment in Malawi: an ex post evaluation of the quality and efficacy of the process , University of Malawi . Master's thesis
- Morrison-Saunders , A . 1996 . Environmental impact assessment auditing to determine environmental management performance . Eco-management and Auditing , 3 : 21 – 5 .
- Nyasulu , D Z . 2000 . Locus standi in environmental protection: the (con)fusion of public law remedies and Malawi jurisprudence . University of Malawi Student Law Journal , 6 ( 1 ) : 140 – 52 .
- Ortolano , L and Shepherd , A . 1995 . Environmental impact assessment: challenges and opportunities . Impact Assessment , 13 ( 1 ) : 3 – 30 .
- Ortolano , L . 1993 . Controls on project proponents and environmental impact assessment effectiveness . Environment Professional , 15 : 352 – 63 .
- Rivas , V , Gonzalez , A , Fischer , D W and Cendrero , A . 1994 . An approach to environmental assessment within the planning process: northern Spanish experiences . Journal of Environmental Planning and Management , 37 ( 3 ) : 305 – 23 .
- Sadler , B . 1988 . “ The evaluation of assessment: post-EIS research and process development ” . In Environmental impact assessment: theory and practice , Edited by: Wathern , P . London : Unwin Hyman .
- Sadler , B . 1996 . Environmental assessment in a changing world: evaluating practice to improve performance , Canberra, , Australia : Australia EIA Network . Final report of the International study of the Effectiveness of Environmental Assessment
- Shepherd , A and Bowler , C . 1997 . Beyond the requirements: improving public participation in EIA . Journal of Environmental Planning and Management , 40 ( 6 ) : 725 – 38 .
- Wood , C and Dipper , B . 2000 . Auditing the assessment of the environmental impact of planning projects . Journal of Environmental Planning and Management , 43 ( 1 ) : 23 – 47 .
- Wood , G . 1999 . Post-development auditing of environmental impact assessment predictive techniques: a spatial analytical approach . Journal of Environmental Planning and Management , 42 ( 5 ) : 671 – 89 .
- World Bank . 1997 . “ The impact of environmental assessment: a review of World Bank experience ” . Washington, DC : The International Bank for Reconstruction and Development/World Bank . World Bank Technical Paper No. 363