Stakeholder perspectives and challenges to the institutionalization of strategic environmental assessment in Botswana

ABSTRACT

This research compares perspectives regarding strategic environmental assessment (SEA) and evaluates SEA implementation bottlenecks in Botswana through the purview of two stakeholder groups, namely environmental assessment practitioners (EAPs) and engineers. It also evaluates SEA regulation and practices in the country. The results highlight that stakeholders in Botswana concur on many SEA-related aspects such as the definition of the concept of SEA and criteria for evaluating SEA effectiveness. Compared to engineers, EAPs have a much more accurate understanding of the type of effects to be assessed during SEA. Challenges to the success of SEA include passive and ineffectual public participation; lack of data sharing; limited technical capacity regarding SEA and dated guidance; deeply entrenched preferences for distinct SEA approaches; low proponent’s willingness to pay for environmental costs including SEA; oppressive practitioner pay structures; and, commercialised environmental assessment consultancy. These manifest as conflict over purpose and approach to SEA, limited awareness regarding SEA, de-skilled planners and diverging practices. These factors are also symptomatic of higher institutional problems such as intense corruption and would suggest that SEA in Botswana is not yet fully institutionalised.

KEYWORDS:

• Strategic environmental assessment
• stakeholder perspectives
• plurality
• implementation bottlenecks
• Botswana

1. Introduction

Strategic environmental assessment (SEA) is an internationally accepted framework for mainstreaming environmental and broader sustainability considerations into policies, plans, and programs (PPPs; see, e.g. Pope et al. 2013, 2018; Fischer and Retief 2021; Matome 2023). Within the ample body of literature devoted to SEA, there is consensus that SEA is context-specific and must therefore be tailor-made to the particular decisional setting in which it is applied (Partidário 2021). In this context, distinct SEA approaches and procedures, institutional and legal arrangements, as well as varied goals and purposes for SEA are apparent in the professional literature (Noble and Nwanekezie 2017; Noble et al. 2019; Tshibangu and Montaño 2019; Retief et al. 2021). Moreover, stakeholders differ in their value expectations, problem framing, legible key issues, impact significance ratings, preferred options and desired level of participation (Bond et al. 2013, 2018; Walker et al. 2016; Cape et al. 2018; Pope et al. 2018; Matome and Mulale 2023). ‘This diversity in theoretical grounding, procedural design, scientific methods and stakeholder engagement has given rise to a pluralistic nature of EA (environmental assessment; encompassing both, SEA and environmental impact assessment – EIA)’ (Cape et al. 2018, p. 32).

More recently, evaluating SEA practices has been the subject of many contributions to the international literature (Geißler et al. 2019; Therivel 2019; Bond et al. 2022). As a result, there is a considerable, albeit non-cumulative body of literature reflecting on the critical factors of SEA (Zhang et al. 2020). Here, it has been found that contextual factors diverge between countries thus impeding the direct transfer of SEA evaluation criteria across jurisdictions (Bond et al. 2022). As a consequence, it is fundamental to understand the critical factors of SEA within the appropriate legal, institutional and cultural arrangements – facets which influence how SEA is understood and implemented (Peterson and Vahtrus 2019; Therivel and González 2019).

A decade ago, Zhang et al. (2013) found 266 notions of critical factors of SEA, with 76% of them being of a more general nature. In a more recent literature synthesis, Bond et al. (2022) identified 21 contextual factors influencing SEA. While SEA-related research is growing internationally, a majority of it reflects on practice in a selected number of countries, in particular, the UK, China, Canada and the Netherlands (Fischer and Onyango 2012; Zhang et al. 2020). Here, SEA practice in most low- and middle-income countries, of which Botswana is part, is rarely reported on in the international literature (Fischer and González 2021, p. 4).

This paper aims to contribute to filling this empirical knowledge gap by comparing stakeholder perspectives regarding SEA and evaluating SEA implementation bottlenecks in Botswana through the purview of three stakeholder groups, namely environmental assessment practitioners (EAPs), engineers and physical planners. It also reflects on SEA regulation and practices in the country. In Botswana, EAPs are required by law to undertake SEA and EIA, however, they have never been included in research evaluating either (see, e.g. Makaba and Munyati 2018; Tshibangu 2018; Matome 2023; Matome and Mulale 2023). Until the advent of a professional association and compulsory certification scheme for EAPs in 2011, their duties were performed predominantly by engineers (Makaba 2014). Presently, engineers and physical planners serve as specialist consultants during EAs, as officials in technical departments, or as officials charged with the duties of overseeing the design and implementation of initiatives subject to EA in both, private and public institutions (Makaba and Munyati 2018; Segosebe 2020; Matome 2023). Albeit in only a few studies, engineers and physical planners have been included in EA-related research in Botswana (Makaba and Munyati 2018; Matome 2023).

Generally speaking, learning from practice is pivotal in advancing EA practice (Fischer et al. 2009; McLauchlan and João 2019). Where EA-related research is limited to enable learning at a system level (as is the case in Botswana; see, e.g. Makaba and Munyati 2018; Matome 2023), it is imperative to explore practice through the purview of the street-level bureaucrats (e.g. EAPs, engineers and physical planners). Such an approach can nurture a deep and thorough understanding of a given EA system as these stakeholders possess an insight into both, past and current EA practice (Peterson and Vahtrus 2019; Clarke & Vu, 2021). According to Zhang et al. (2013, 2020), critical factors include both enabling and constraining factors, however, this research focuses specifically on challenges (constraining factors) as they represent areas for further improvement (Fischer 2010; Walker et al. 2016; Matome and Mulale 2023).

In Botswana, mandatory requirements for preparing SEAs when formulating PPPs were introduced in 2005 (Makaba and Munyati 2018). Still, SEA practice in the country is relatively budding, with estimated 20 SEAs having been conducted from 2000 to 2020 (Matome 2021). Furthermore, SEA in the country remains relatively unknown and under-researched, procedures are unregulated, and SEA practice relies upon guidelines borrowed from neighbouring South Africa (Makaba and Munyati 2018; Matome 2023; Matome and Mulale 2023).

Though not all at once, these contextual aspects are also reported across SEA systems in most developing nations. For instance, SEA in Brazil is not mandatory at any level of decision-making and practice is heavily dependent upon (exotic) guidance from multilateral development agencies and international best practice principles outlined in the professional literature (Tshibangu and Montaño 2019). Furthermore, while 103 SEAs¹ have been prepared in South Africa since 1995 (Retief et al. 2021), a total of 27 SEAs have been prepared in Thailand to date (Chanchitpricha et al., 2019, Chanchitpricha et al. 2021). In this context, Botswana is representative of developing nations in general, and lessons drawn accordingly reflect SEA practice in – what are often – under-researched developing nations. Here, lessons from this research are valuable for an international audience dealing particularly with the SEA theory, which demands a full account of experiences with SEA from all contexts to be valid in informing universal SEA practice (Fischer and Gazzola 2006).

2. Methodology

2.1. Research design

This research adopted a cross-sectional research design, which is defined simply as a ‘one-point-in-time investigation’ (Mosetlhi 2012, p. 74). In the international literature, cross-sectional designs are heavily criticized for failing to demonstrate trends over time and are thus considered inferior to longitudinal studies (Mosetlhi 2012; Sedgwick 2014). Nevertheless, they generally have high reliability and external validity, in particular, ‘due to the use of representative comparison samples which are therefore statistically representative of study populations’ (Mosetlhi 2012, p. 74). Furthermore, they are more effective when there is a shortage of time and financial resources, and when determining associations to inform hypothesis-building and further research (Mosetlhi 2012; Sedgwick 2014). Notwithstanding the traditional prejudice against cross-sectional research, the design was deemed appropriate for this study- to compare stakeholder perspectives regarding SEA and determine obstacles to the institutionalisation of SEA in Botswana.

2.2. Data collection and analysis

A semi-structured online questionnaire was designed and administered through the Microsoft forms® platform. The survey consisted of seven main questions: six related to the concept of SEA, the relevance of key issues and type of effects evaluated during SEA, the litmus test for effectiveness, timing and role of SEA in planning (see Table 1) whilst the 7^th question related to the challenges that respondents could attribute as impeding SEA practice in Botswana. Justification for the choice of issues evaluated is provided in the next section. Whilst the first six questions were closed and answerable with a rating on a 5-point Likert scale (options were strongly disagree (1), disagree (2), uncertain/neutral (3), agree (4) and strongly agree (5)), the last question was open-ended enabling respondents to feature explanations (e.g. relating to causes of the observed challenges). This enabled an in-depth and rich understanding of both perceptions and challenges constraining SEA practice. It is noteworthy to highlight that these questions were preceded by three other questions relating to the role of the stakeholder in the SEA process (i.e. whether they are EAPs, engineers or physical planners) as well as their EIA and SEA experience.

Table 1. Group statistics.

	What was/is your role in the Environmental Assessment (EA) process?	N	Mean	Std. deviation	Std. Error Mean
Strategic Environmental Assessment (SEA) is a system of incorporating environmental considerations into policies, plans and programmes.	Environmental Assessment Practitioner (EAP)	15	4.47	1.246	.322
	Engineer	15	3.93	.961	.248
SEA studies (In Botswana) discuss issues relevant for decision-making.	Environmental Assessment Practitioner (EAP)	15	4.33	1.113	.287
	Engineer	15	3.87	.743	.192
SEA studies must examine secondary, cumulative and synergistic effects of proposed policy, plan or programme (PPP) initiatives and their options.	Environmental Assessment Practitioner (EAP)	15	4.07	.961	.248
	Engineer	15	3.33	.976	.252
SEA is only worthwhile if it results in modifications to PPPs or if it structures the planning process.	Environmental Assessment Practitioner (EAP)	15	4.07	1.100	.284
	Engineer	15	3.87	.743	.192
Most SEAs (In Botswana) are conducted when plans are largely done, and there is little room for SEA to influence them.	Environmental Assessment Practitioner (EAP)	15	3.40	.828	.214
	Engineer	15	3.00	.926	.239
SEA in Botswana is only conducted as a legal obligation, not a decision-support tool	Environmental Assessment Practitioner (EAP)	15	2.87	.743	.192
	Engineer	15	3.13	1.187	.307

It is significant to note that calls to contribute to the survey were made through the stakeholders’ professional bodies. That said, the survey included EAPs registered and certified by the Botswana Environmental Assessment Practitioners Association (BEAPA) as per the 2011 Environmental Assessment Act (Department of Environmental Affairs 2011), engineers registered with the Engineers Registration Board Botswana (ERBB) per the 2009 Engineers Registration Act (Government of Botswana 2009), and physical planners registered with the Director of the Department of Town and Country Planning (DTCP) per the 2013 Town and country planning act (DTCP 2013). For engineers, calls were also made through the Association for Consulting Engineers Botswana (ACEB) and the Botswana Institute for Engineers (BIE). Due to the low turn-up, calls for participation were also made through telephone calls and email; and, extended to private consultancy firms and public institutions such as local planning authorities.

Generally speaking, low turn-up is associated with significant survey biasedness (Mosetlhi 2012). Arguably, the influence of low turn-up on survey biasedness is limited particularly since respondents to SEA research surveys generally respond more as members of a specified class (e.g. government regulators, industry, SEA consultants) rather than as individuals (see e.g. Cape et al. 2018; Therivel 2019). Here, the low turn-up could be attributed to the limited stakeholder awareness regarding SEA (see section 3.2) and the unwillingness of some individuals to support research (as elicited during telephone calls and/or from email responses to the invitation).

Despite the method of contact, participation was purely voluntary. The results were also acquired anonymously (i.e. no identifiers such as names and contact details were required or stored by the system). In this context, the survey was open for 4 months, i.e. from early December 2020 until the end of March 2021. During this period, only 45 responses were obtained. Of these, 15 responses were incomplete with the respondents having attempted, in particular, the three questions relating to their role, EIA and SEA experience. Importantly, the 15 respondents had no SEA experience, little if not no EIA experience and broadly represented the three stakeholder groups. The remaining 30 responses were proportionally divided between engineers and EAPs. No responses were obtained from physical planners; however, through email responses to the invitation and during follow-up telephone calls, they suggested key SEA implementation bottlenecks disguised as justification for their general inactivity in the research survey (see Sections 3.2.1 and 3.2.2). In this context, comparing stakeholder perspectives regarding SEA was undertaken based specifically on the responses from 15 EAPs and 15 engineers.

In sum, 30 responses were received from 5 people with ample SEA experience, 10 people with some SEA experience and 15 people with no SEA experience. For the EAPS, four people had no SEA experience, seven people had been involved between 1 and 5 SEAs, three people had engaged in 6–10 SEAs and one person had engaged in more than 11 SEAs. For the engineers, 11 people had no SEA experience, 3 people had been involved in 1–5 SEAs and 1 person was involved in 6–10 SEAs. Meanwhile, six engineers had no EIA experience, two EAPs and eight engineers had been involved in 1–5 EIAs, four EAPs served in 6–10 EIAs, and one engineer and nine EAPs had served in over 11 EIAs. Accordingly, some stakeholder perspectives regarding SEA are likely to have been theoretical or to have been strongly influenced by the comparatively high EIA experience suggesting that they may not be based on practical experience with SEA. In this context, there is considerable uncertainty in the results.

The data were analysed using two methods. First, the stakeholder perspectives regarding SEA based on the six close-ended questions were analysed through the Student T-test run on the IBM Statistical Package for Social Sciences (SPSS Statistics Version 27.0). The T-test compares the mean values of a continuous dependent variable based on two categories of the independent variable (i.e. comparison groups), assuming normality and random independent samples (Okeh 2009; De Winter and Dodou 2010; Milenovic 2011). Although scholars generally argue that a Mann Whitney-Wilcoxon (MWW) is suited for comparing mean value of ordinal dependent variables (Mccrum-Gardner 2008; Webb and Chaffer 2016), it has been proven that ‘for five-point Likert items, the t-test and MWW generally have similar (statistical) power’ and that the T-test is robust to modest deviations from the test assumptions (De Winter and Dodou 2010). Here, though the test does not require homoscedasticity in ordinal scores, it requires equal variance amongst ranks of the scores (De Winter and Dodou 2010; Milenovic 2011). In this context, the T-test was used to examine the extent to which EAPs and engineers differ in their average perception regarding SEA. More widely, the research tests the extent to which profession accounts for plurality in stakeholder perception regarding SEA and thus aims to contribute to the scanty empirical literature examining the concept of plurality in EA (Cape et al. 2018; Chanchitpricha et al. 2021). Second, the data from the open-ended question was analysed through thematic content analysis – by reducing it into themes or concepts depicting a collective expression (Mbaiwa 2018; Mbaiwa and Hambira 2019; Matome 2023).

2.3. Justification for the choice of issues evaluated

Questions used in the survey were designed by the author following issues identified in contemporary SEA research. For instance, the professional literature highlights the need for SEA to achieve the environmental policy integration objective by incorporating environmental considerations as well as (environmentally sound) modifications into PPPs to facilitate sustainable development (Therivel 2019; Therivel and González 2019; Bond et al. 2022). In this context, good scoping, and shorter and proportionate SEAs are key areas of discussion in the SEA community (Fischer 2023). Resultantly, SEAs need to be ‘fit-for-purpose’ and focus on the actual issues relevant for decision-making (McLauchlan and João 2019). This can facilitate resource-use efficiency, which has become an important driver for EA simplification efforts internationally (Bond et al. 2014; Fischer et al. 2023).

Similarly, SEA needs to evaluate, among other equally relevant issues, secondary, cumulative and synergistic effects of proposed initiatives for the full suite of impacts to be considered adequate (Noble et al. 2019; Partidario and Monteiro 2019). In this context, legal disputes over the type, extent and methods of (strategic) impact evaluation test the legitimacy and even quash SEAs and associated plans, for example, in the English courts (Therivel 2013). Meanwhile, the timing of SEA v the planning cycle and concerns over the poor substantive outcomes of SEA, which appear to be driven in part by the trend to conduct SEA as a legal obligation rather than a decision-support tool is topical in contemporary SEA research (Partidario and Monteiro 2019; Therivel 2019). Here, scholars suggest that SEA must structure/shape planning rather than fine-tune plans (Peterson and Vahtrus 2019; Therivel 2019).

As suggested by the SEA literature cited above, these issues are covered extensively in the international professional literature. However, limited consideration has been given to them in SEA research in Botswana. Where they have been covered, the evaluation is often based on few responses acquired within the context of single SEA case studies (Makaba and Munyati 2018; Matome 2023; Matome and Mulale 2023). Accordingly, the author designed the questions on these topical issues to afford broader learning at a SEA system level.

3. Results and discussions

3.1. Comparing stakeholder perspectives regarding SEA in BOTSWANA

The results of Levene’s test of equality of variances, depicted by the significance of F, shown in Column 2 of Table 2 demonstrate that all groups have equal variances. The decision rule in Levene’s test is to reject the null hypothesis if the significance of F is less than alpha, set at 0.05 in this case. The interpretation of the results of the T-test, therefore, follows the assumption of equality of variance (as acquired from Levene’s test; see Table 2). Moreover, the interpretations are based on the group means shown in Table 1, but cognisant to the fact that the means are merely labels and not actual measures. In a T-test, the null hypothesis is that there is no statistically significant difference in the means of the two groups evaluated (De Winter and Dodou 2010). That is, on average, EAPs and engineers are similar in their perspectives regarding SEA. In this context, the decision rule is to reject the null hypothesis when the significance of the T-test (shown in column 3 of Table 2) is less than the specified alpha (set at 0.05).

Table 2. Independent samples test.

		Levene’s test for Equality of Variances		t-test for Equality of Means
		F	Sig.	T	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	95% Confidence Interval of the Difference
									Lower	Upper
Strategic Environmental Assessment (SEA) is a system of incorporating environmental considerations into policies, plans and programmes.	Equal Variances Assumed	1.213	.280	1.313	28	.200	.533	.406	−.299	1.366
	Equal Variances not Assumed			1.313	26.305	.201	.533	.406	−.301	1.368
SEA studies (In Botswana) discuss issues relevant for decision-making.	Equal Variances Assumed	1.790	.192	1.351	28	.188	.467	.345	−.241	1.174
	Equal Variances not Assumed			1.351	24.418	.189	.467	.345	−.246	1.179
SEA studies (In Botswana) must examine secondary, cumulative and synergistic effects of proposed policy, plan or programme (PPP) initiatives and their options.	Equal Variances Assumed	.998	.326	2.074	28	.047	.733	.354	.009	1.458
	Equal Variances not Assumed			2.074	27.994	.047	.733	.354	.009	1.458
SEA is only worthwhile if it results in modifications to PPPs or if it structures the planning process.	Equal Variances Assumed	2.183	.151	.584	28	.564	.200	.343	−.502	.902
	Equal Variances not Assumed			.584	24.581	.565	.200	.343	−.506	.906
Most SEAs (In Botswana) are conducted when plans are largely done, and there is little room for SEA to influence them.	Equal Variances Assumed	.019	.890	1.247	28	.233	.400	.321	−.257	1.057
	Equal Variances not Assumed			1.247	27.659	.233	.400	.321	−.257	1.057
SEA in Botswana is only conducted as a legal obligation, not a decision-support tool	Equal Variances Assumed	1.270	.269	−.737	28	.467	−.267	.362	−1.007	.474
	Equal Variances not Assumed			−.737	23.512	.468	−.267	.362	−1.014	.481

The results indicate that stakeholders generally agree that ‘SEA is a system of incorporating environmental² considerations into policies, plans and programmes’ (means are 4.47 for EAPs and 3.93 for engineers; see Table 1), and that stakeholders are congruent in this regard (0.2000 > 0.05; see Table 2). Whilst integrating environmental and sustainability considerations into PPPs is the universally accepted goal for SEA, scholars are divided on other aspects of SEA such as whether it should be a decision-informing or decision-making tool (Therivel 2019; Fischer and Retief 2021; Nwanekezie et al. 2021). Since this definition of SEA is an excerpt from the Botswana national SEA guidelines (see DEA 2009, p. 99), the results may very well reflect the strong effect of guidance materials in shaping how SEA is understood in a given SEA system. In this context, keeping SEA guidance materials up-to-date and accurate is imperative especially since they serve as reference materials for many stakeholders including practitioners and proponents (Annandale et al. 2021).

Likewise, EAPs and engineers concur that ‘SEAs (in Botswana) discuss issues relevant for decision-making’ (means are 4.33 for EAPs and 3.87 for engineers; see Table 1). In Botswana, the issues discussed in any EA are a product of terms of reference (TOR) set by the competent authority, the EAP, and the interested and affected parties during a scoping exercise (Department of Environmental Affairs 2011). The TOR is approved if the competent authority is confident that it may adequately help in guiding the appraisal and subsequent decision-making. In this context, it is assumed that the relevance of the issues discussed in SEA for decision-making is tacitly embedded, which explains the degree of confidence exhibited by stakeholders. Key informant interviews with planners in Botswana have also yielded similar conclusions (Matome 2023), suggesting that SEAs could become effective planning tools provided they are considered in decision-making.

However, stakeholders contrast significantly in their views regarding the type of effects to be assessed by SEAs (0.047- 0.05; see Table 2). In this context, engineers are more uncertain (mean = 3.33) whereas EAPs are convinced that SEAs must evaluate the secondary, cumulative and synergistic effects of proposed PPP initiatives and their options (mean = 4.07; see Table 1). Given that evaluating secondary and cumulative effects of PPPs constitutes part of the methodology components for a good quality SEA (Retief et al. 2008; Fischer 2010; Matome and Mulale 2023), it is suggested that, compared to engineers, EAPs have a high degree of ownership of the methodology aspects of SEA. The results therefore support the observation that the understanding of SEA in Botswana is limited among many stakeholders (Makaba and Munyati 2018; Matome 2023). As a consequence, it is imperative to improve the current degree of ownership of SEA exhibited by stakeholders to make them familiar with the techniques and methods of SEA to facilitate informed participation (Walker et al. 2016).

Nevertheless, the EAPs and engineers are inseparable in their conviction that ‘SEA is only worthwhile if it results in modifications to PPPs or informs planning’ (0.564>alpha-0.05; see Table 2). In contrast, 59% of respondents in the UK disagreed with the statement: ‘The appraisal is only worthwhile if it changes the plan/policy’ (Thérivel and Minas 2002, p. 84). While it is accepted that SEA can have other effects such as ensuring participation, transparency and accountability in decision-making as well as improving the environmental stewardship of stakeholders (Bina et al. 2011; Partidario and Monteiro 2019; Peterson and Vahtrus 2019), the absolute measure of the value of SEA (in a decision audit) should be judged concerning whether it and the extent to which it structures or informs decision-making, in particular, if it integrates environmental concerns into PPPs (Fischer and Retief 2021). Here, evaluations of SEA could then demonstrate whether or not SEA is contributing to the achievement of sustainable outcomes, which could help determine whether the benefits of SEA are worth the costs (Therivel and González 2019, p. 182). However, this represents a short-term perspective, meaning that stakeholders must become more accepting of the transformative potential of SEA – which can also strengthen transformations towards sustainability at a system level (Bina et al. 2011).

Based on Table 1, neither EAPs nor engineers are certain of whether SEAs in Botswana are conducted when plans are largely prepared leaving little room for SEA to influence them or not (mean = 3.40 for EAPs and 3.00 for engineers). Furthermore, there is no statistically significant difference in the views of EAPs and engineers (0.223 > 0.05; see Table 2). Whilst SEAs can be conducted as ex-post (i.e. prepared during PPP audit) assessment procedures, it has been found that the effectiveness of SEA in influencing decision-making tends to be higher when SEAs are conducted during early stages of planning, in particular, during plan conceptualisation and before PPP approval (Chanchitpricha et al., 2019; Therivel 2019). In this context, SEAs must serve as clear evidence for environmentally sound decision-making (Fischer and Retief 2021; Partidário 2021). Worryingly, EAPs (who are also the SEA consultants) are uncertain of the timing of SEAs in relation to planning cycles, which is part of the decisional context that must be evaluated and clearly understood as part of any SEA (Nwanekezie et al. 2021; Partidário 2021). As a result, it is suggested that, in Botswana, the limited influence of SEAs in decision-making (Makaba and Munyati 2018; Matome 2023) probably connotes that SEAs are conducted in isolation as technical consultancy exercises far removed from the planning and decision-making context.

Where stakeholders view SEA as being done in isolation of the broader decisional context, SEA would often be conducted as a legal obligation, ‘with planners treating SEA as a formality to be carried out rather than as a tool to inform their plan’ (Thérivel and González 2021, p. 102). In Botswana, however, EAPs and engineers are uncertain of whether institutions conduct SEAs as a legal obligation or as decision-support tools (means are 2.887 for EAPs and 3.13 for engineers; see Table 1). Also, Table 2 reveals that the sentiment is mutual to the two stakeholder groups (0.467 > 0.05). Given that some planners in Botswana are genuinely open to SEA as a mechanism for achieving sustainable planning outcomes (Matome 2023), SEA may still have a role in transforming Botswana’s planning culture towards sustainability albeit raising awareness regarding SEA and undertaking capacity building would be necessary (Makaba and Munyati 2018; Matome and Mulale 2023).

3.2. SEA implementation bottlenecks in BOTSWANA: a practitioner’s PERSPECTIVE

This section presents SEA implementation bottlenecks in Botswana as identified by survey respondents with SEA experience. In this context, they are based on practical experience rather than speculation. However, it is key to note that only the 11 EAPs (constituting 73% of the total number of respondents with SEA experience) shared the challenges they noted whilst conducting SEA, which makes the results biased towards the views of a single stakeholder group. Because the SEA-related challenges encountered by the technical departments, the local planning authorities, and the competent authority are included in recent research on SEA in Botswana (Makaba and Munyati 2018; Matome 2023), the results could be considered as expanding and deepening the available knowledge regarding SEA implementation bottlenecks in the country.

3.2.1. Limited awareness regarding SEA

According to most EAPs, ‘there are not many institutions that are familiar with the SEA process in Botswana’. Generally speaking, limited awareness regarding SEA was the most cited reason provided by most engineers, some EAPs, and almost all physical planners for not participating in the research survey. Interestingly, respondents indicated familiarity with EIA but not SEA – a similar picture was also obtained from the analysis of the EIA and SEA experience of survey respondents (see Section 2.2). In this context, one planner remarked as follows: ‘What is SEA? Is it similar to EIA? Because EIA I have heard of, but not SEA’.

Makaba and Munyati (2018) attribute the limited stakeholder awareness regarding SEA to the brief reference to SEA in the national EA legislation; however, it is suggested that the evolution of the physical planning legislation and practice may have also contributed to the disparities. For instance, the 1977 Town and Country Planning Act required EIA or similar procedures for significant land-use developments, with the result that a considerable number of major physical developments (e.g. the TransKhalahari highway project) were subjected to EIA since the 1980s (Mpotokwane and Keatimilwe 2003; Aniku 2011; Segosebe 2020). Reference to SEA was only made in the 2013 Town and Country Planning Act, which requires that development plans prepared for physical planning areas must be accompanied by approved SEAs (DTCP 2013). There are 18 physical planning areas in Botswana; however, only a few have completed and approved SEAs accompanying the respective development plans. These include the Molepolole planning area (Mosienyane & Partners International 2015), Selebi Phikwe planning area (Mosienyane & Partners International 2016) and Serowe planning area (Envirommetrix 2017). In this context, only a few planners (particularly those that have engaged extensively with development plans having SEA) had knowledge regarding SEA albeit limited (Matome 2023). As a result, it can be argued that in developing nations the limited exposure to SEA is one of the main drivers of limited awareness regarding SEA (Slunge and Tran 2014; Walker et al. 2016; Chanchitpricha et al., 2019; Tshibangu and Montaño 2019).

3.2.2. Passive and ineffective public participation

‘SEA should be stakeholder-driven, explicitly addressing the public’s inputs and concerns, ensuring access to relevant information of the PPP making process’ (Fischer and Gazzola 2006, p. 401). However, in practice, respondents highlight that planning and SEA are undertaken in silos, where their participation in SEA is only passive with most of the work being done by consultants. For instance, one physical planner commented: ‘I am not an environmentalist. While I may have been involved in some meetings regarding the plan you say has a SEA, my participation was limited to issues regarding the (current) spatial plan and problems connected to it’. Another planner added:

The meetings we attend are never about SEA studies. Rather, we are told about spatial plans and asked about planning issues we have noted (in the respective locality). At the end of the day, we are told the information will be used in a plan being prepared by the consultants. How the information is used thereafter, we are never really aware.

These results imply that since most of the work is done by consultants, government planners available in-house are generally spectators during development planning and SEA. Where consultancy is very strong, SEA practice tends to assume a technical-rational approach (Retief et al. 2008). The results of a consultancy-driven approach to SEA include the preparation of overly long and incomprehensible SEAs and de-skilled planners (Fischer et al. 2009; Jha-Thakur et al. 2009; Fischer 2010). As a result of the limited capacity regarding SEA in the Botswana public sector, the decision-makers rarely use SEA to inform their planning decisions (Makaba and Munyati 2018; Matome 2023).

Furthermore, the public meetings held during SEA in Botswana are often attended by an uninformed public (Makaba and Munyati 2018). As a result, one EAP observed that there were ‘poor responses to questionnaires by some key stakeholders’. Here, the capacity of the public to contribute meaningfully during SEA may be weakened limiting, more importantly, their role in decision-making and PPP formulation. As a consequence, a lack of ‘buy-in’ from the public and the public sector is possible, which can advance conflicts over the decisions (e.g. proposed projects, licensing and development guidelines) prepared during development plan-related SEA. So, the capacity of SEA to streamline project decision-making (which is one of the strong arguments that facilitated its wide adoption; see, e.g. Noble et al. 2013; Cape et al. 2018; Therivel and Gonzalez 2021; Gonzalez and Therivel 2022) may be lost. There is therefore an urgent need to improve stakeholder awareness regarding SEA to facilitate informed participation (Rega and Baldizzone 2014; Walker et al. 2016).

3.2.3. Institutional and technical limitations in coordinating SEA

In most developing nations, there is often a shortage of technical capacity to coordinate SEA (Slunge and Tran 2014; Hipondoka et al. 2016; Tshibangu and Montaño 2019). EAPs highlight a similar challenge in Botswana, stressing that ‘in most of the cases even the government officials working as Environmentalist do not have enough competency to direct the SEA process’ (see also Makaba and Munyati 2018). The limited institutional and technical capacity to coordinate SEA in the country has resulted in the approval of SEAs which performed poorly on critical components such as demonstrating effects of proposed initiatives on protected areas and cumulative effects assessment (Matome and Mulale 2023).

Furthermore, EAPs remark that ‘the competent national authority often requires EIA level detail in the strategic environmental management plan (SEMP) and monitoring which is not appropriate at the SEA level’. As a result, SEAs in Botswana do not satisfactorily explain how monitoring arrangements established during SEA can be tiered to project EIA to reduce duplication of efforts (Matome and Mulale 2023). This explains why monitoring sections are among the substandard components of SEA reports in Botswana – often lacking in terms of who and what of monitoring (Makaba and Munyati 2018).

These results confirm Matome and Mulale’s (2023) observation that the poor performance apparent in certain aspects of SEA practice in Botswana may be a result of an underdeveloped operational context. In this context, promoting the use of SEA across the world may be considered as the transfer of Western innovations to the resource-constrained and undercapacitated developing nations (Dalal-Clayton and Sadler 2004). But, ‘rarely has the introduction of (EA) instruments been accompanied by the level of capacity development necessary to prevent them from being manipulated to focus only on those elements with which the implementers are more familiar’ (Cashmore et al. 2009, p. 93). Since it is the knowledge of the stakeholder that comes to bear upon the SEA, the practice may be improved or ruined depending upon the sufficiency or accuracy of that knowledge (Bond et al. 2018; Chanchitpricha et al. 2021). As a consequence, it is suggested that quality control by independent SEA reviewers as suggested by, e.g. Fischer and Gazzola (2006) is imperative to a functional SEA system.

EAPs further observe that the evaluation of SEAs is often based on personal interpretations, which makes the review process inconsistent. For instance, one EAP said: ‘Institutionally no two DEA regional offices ask the same questions or review the SEA the same way. Recently I received over 5 pages of comments each on two separate but very similar SEA scoping reports from two different offices, No two comments were the same’. The lack of uniformity in the review of SEA is a prominent feature of SEA theory and practice reported, for instance, across distinct authorities implementing the European SEA Directive 2001/42/EC (EC European Commission 2009, 2017). While it may reflect the differences in the understanding of SEA amongst reviewers, inconsistency during scoping may explain why different environmental/sustainability issues receive unequal treatment during SEA (Therivel et al. 2009; Canter & Ross, 2014; De Montis 2014). This would also account for the uneven SEA report quality reported internationally (see, e.g. Bonde & Cherp, 2000; Fischer 2010).

As critically argued by some authors, one way of improving SEA practice is through active learning (Fischer et al. 2009; Jha-Thakur et al. 2009; Bond et al. 2013; McLauchlan & Joao, 2019). ‘This would require those who have carried out an SEA to think about how that SEA worked and what could be changed next time, and then applying that learning to the next round of plan-making and SEA’ (Thérivel and González 2021, p. 105). But ‘the skills required by planners, consultants and the community will differ from each other, and the range of skills required by each group may also be influenced by context’ (Jha-Thakur et al. 2009, p. 142). As a consequence, learning may become circumstantial. Here, active learning is ‘relatively simple for the consultants who carry out multiple SEAs, since it is in their interest to carry out high-quality and efficient SEAs, but may be harder for the planning team, particularly as it requires continuity of staff over time’ (Thérivel and González 2021, p. 105–106). If the evaluation of SEAs is based on personal and non-standardized criteria, learning from practice may be constrained since the knowledge obtained from previous encounters cannot be directly transferred to another. As a result, those engaged in the SEA process must become more open to new approaches to facilitate better learning (Bond et al. 2013; Walker et al. 2016).

3.2.4. Plurality in SEA approaches and familiarity with one

Internationally, SEA is typically implemented following discrete stages similar to those used in project EIA (Annandale et al. 2021; Nwanekezie et al. 2021). In Botswana, there are no mandatory approaches to undertaking SEA (Tshibangu 2018; Matome and Mulale 2023). In addition, SEA guidelines in the country stress that different approaches to SEA may be used depending on the peculiarities of each case; however, they do not elaborate on how SEA may be used in different contexts and instead elaborate on the EIA-like administrative procedures for SEA (Makaba and Munyati 2018; Matome 2023). As a result, respondents highlight that the competent national authority emphasises the use of the EIA-based SEA approach as a yardstick for a good quality SEA.

In the professional literature, it is widely accepted that SEA is a ‘multi-faceted and multi-dimensional assessment’ representing a ‘series of approaches operating along a spectrum from less to more strategic – from impact assessment-based to strategy-based – with each approach to SEA differentiated by the specific objectives of SEA application and the extent to which strategic principles are reflected in its design and implementation’ (Noble and Nwanekezie 2017). Here, ‘the “holy grail” is a situation where SEA is more closely integrated into the planning process – possibly to the point where there is no longer a differentiation between SEA and planning’ (Tetlow and Hanusch 2012, p. 17). Where EAPs utilize a strategy-based SEA, they note that ‘the reviewers were unable to differentiate between the plan and SEA process’ and would then demand a ‘SEA’, i.e. an impact assessment-based SEA. Accordingly, the use of an impact assessment-based SEA model is popular in Botswana (see, e.g. Mosienyane & Partners International 2015, 2016; Envirommetrix 2017), suggesting that the degree to which SEA is adaptive to the particular operational context is questionable.

Whilst it is acknowledged that different approaches to SEA are effective in different situations, it is also noted that consultants tend to prepare ‘frameworks for strategic decision-making’ rather than ‘an assessment per SEA’ (Retief et al. 2008, p. 504). Moreover, McLauchlan and João (2019) And Thérivel and González (2021) observe that stakeholders can learn about SEA and then apply that knowledge to create ‘cookie cutter’ SEAs or develop robust SEAs. That said, practitioners may probably be following the former logic to bypass the system, or reviewers may be failing to appreciate the strategy-based SEA because of too much familiarity with the EIA-based approach. Here, it is suggested that knowledge and learning are mainly undertaken by EAPs rather than regulators and that provided capacity building is done to nurture it, SEA may fail particularly in areas where a strategy-based approach is relevant.

3.2.5. Unavailability and inaccessibility of long-term environmental and baseline data

SEA is an evidence-based assessment procedure (Kørnøv and Thissen 2000; Fischer 2007; Partidário 2021). However, in Botswana, respondents decry the paucity of environmental and baseline data, emphasising that, during SEA, the” accessibility and availability of up-to-date statistical data was also a challenge”. According to previous research in Botswana, the availability or accessibility of environmental data is constrained by higher institutional problems such as privatisation of data (often to maintain power and for personal interests), the lack of data sharing policies, and deeply entrenched sectoralism (Perkins et al. 2013; Molebatsi and Kalabamu 2018; Mbaiwa and Hambira 2019). In this context, SEAs in the country often have significant baseline data shortages (Matome and Mulale 2023).

The availability of long-term statistical data is a key consideration during SEA particularly when attempting to use extrapolative methods such as trend analysis and modelling (Fischer 2007; Gonzalez et al., 2021). In this context, it is unsurprising that the description of the baseline environment in most SEAs in Botswana is mainly limited to inventories (see, e.g. Mosienyane & Partners International 2015; Geoflux 2017). Furthermore, Matome and Mulale (2023) found out that SEAs in the country often include a poor consideration of health effects which, according to Chalabi (2020) are better predicted using extrapolative health risk assessments. Extrapolative risk assessment models often have high degrees of confidence, robustness and certainty but very high statistical data requirements (Williams et al. 2010; Stewart and Hursthouse 2018). In this context, it is unsurprising that impact prediction during both, EIA and SEA, is based predominantly on more qualitative and subjective techniques such as checklists and matrices (Fischer 2007; Souloutzoglou and Tasopoulou 2020), meaning that the certainty of predictions and the legitimacy of the assessments (and associated decisions) is contentious (Therivel 2013). Here, tiering, e.g. the alternatives selected using such unscrupulous predictions may become questionable. So, the capacity of SEA to streamline EIA (to advance resource-use efficiency; see, e.g. Noble et al. 2013; Therivel and Gonzalez 2021; Gonzalez and Therivel 2022) may be lost, with the result that duplication of efforts between SEA and EIA could become problematic.

3.2.6. Financial deficiency and proponent’s control over capital invested during SEA

In Botswana, proponents declare financial deficiencies as a major contributory factor to the poor integration of SEA results into PPPs (Makaba and Munyati 2018; Matome 2023). Equally, the respondents maintain that there is a shortage of funds invested during SEA. In particular, one EAP asserted that: ‘financial constraints are an issue because to access other places one needs money and also to carry some researches needs money’. As a result, SEAs in Botswana have inadequate fieldwork (Makaba and Munyati 2018) and critical baseline data shortages (Matome and Mulale 2023). In this context, SEA consultants risk making misinformed decisions, which may jeopardise the capacity of SEA to protect the environment.

Moreover, some aspects of SEA, e.g. archaeological impact assessment (AIA) tend to require ground truthing for accuracy (Wilmsen et al. 2009; Thebe and Sadr 2017). Even highly computerised remote sensing and geographic information systems (GIS), which are typically employed during spatial plan-related SEA (Fischer 2007; Souloutzoglou and Tasopoulou 2020), often require ground truthing for accuracy (Sebego and Arnberg 2002; Matlhodi et al. 2019). Consequently, there is a need to improve funds allocated to SEA to enable intense fieldwork, which is currently a weak point of Botswana’s SEA practice (Makaba and Munyati 2018).

Furthermore, EAPs highlight that some proponents tend not to mind data shortages, suggesting that for some proponents the value of SEAs may not necessarily be related to the comprehensive data they often contain as proposed by various authors (see, e.g. Monteiro & Partidario, 2012). Here, proponents are generally inclined towards rapid and expert-driven SEAs rather than comprehensive and detail-oriented SEAs. One EAP put the issue as follows: ’ … there is always a conflict between what the Client is willing to pay and what the authorities require or expect. Technically the authorities require extensive studies (as per the guidelines) even if not necessary’. Apart from highlighting plurality in terms of the preferred SEA approaches, the results also imply that the proponents and authorities have deeply entrenched preferences for certain SEA approaches. ‘Thus the potential for dissatisfaction can arise from stakeholders guiding and funding SEAs to focus on elements they value rather than those valued by other stakeholders’ (Cape et al. 2018, p. 40). In this context, the flexibility of SEA (which is critical to its integration with the decisional process and its overall effectiveness; see, e.g. Noble and Nwanekezie 2017; Partidário 2021) may be lost. Eventually, SEA may be conducted as nothing more than a box-ticking exercise to acquire environmental clearance (Matome 2023).

Notwithstanding the financial challenges observed, Botswana has a regulated EA consultancy fee structure, which remunerates high-ranking EAPs more (see Table 3). According to some EAPs, the underlying assumption was that the high-ranking EAPs could (as a consequence of their profound experience) conduct EAs more quickly than the less experienced EAPs and, therefore, both could still gain fair payments in the end. However, the assumption neglected the fact that project durations are set by proponents probably based on project timelines not the rank of the contracted EAPs. That said, the pay structure is oppressive to low-ranking EAPs who incur similar assessment requirements and costs as does the more experienced high-ranking EAPs. For instance, the heightened pay enables high-ranking EAPs, where appropriate, to purchase additional EAP support and cover more ground quickly; purchase additional specialised technical skills necessary for the conduct of the assessment but not accounted for in the proponent’s assessment budget; or easily afford costs associated with fieldwork. Therefore, they are more resource-empowered than the lower tier EAPs. The resource inequality could therefore deter lower-ranking EAPs from actively participating and profitably benefiting from EA consultancy. It may also make them more susceptible to producing poor-quality assessments primarily as a result of financial constraints.

Table 3. The tariff of professional fees in Botswana.

CATEGORY OF PRACTITIONER	MINIMUM DAILY RATE (P)	MAXIMUM DAILY RATE (P)
EAP	1,600.00	2,600.00
Senior EAP	2,600.00	3,500.00
Principal EAP	3,500.00	5,000.00

Source: Department of Environmental Affairs (2012).

Furthermore, EAPs insist that the fee structure (shown in Table 3) is ineffective as proponents set fixed budgets for EAs (within which EAPs must operate). Here, they argue that allowing proponents to price assessments is ineffectual as they limit their costs by investing very little towards EAs. According to Pearce and Turner (1990), the willingness to pay method is useful for pricing an item an individual already values significantly. In Botswana, several authors (Toteng 2002, 2004; Mmopelwa et al. 2017; Matome 2023) assert that environmental considerations are often an afterthought and a subordinate priority to socio-economic imperatives and are thus largely underbudgeted. In this context, even if the fee structure shown in Table 3 exists, it does very little to enable reasonable investments into EA practice as proponents have a low willingness to pay for environmental costs such as SEA.

This is further compounded by the fact that EA practice in Botswana is commercialised and profit-driven. Whilst a strong consultancy sector may facilitate increased SEA practice (Retief et al. 2008, 2021), it also has the potential to advance exploitation where practitioners ration resources to maximise returns. In this context, practitioners may decry limited funds only to heighten consultancy fees and their associated profits. As such, it may be more relevant to provide budget estimates for specified environmental assessments rather than professional fees. Here, proponents could then negotiate with their chosen practitioner but within reasonable budgets. Otherwise, and because there are no procedures to enable EAPs to request additional funding (where appropriate), practitioners may prepare minimalist assessments fit for procedural clearance but less instructive for environmental protection and informed decision-making. Ultimately, SEA’s decision-support function may be compromised resulting in further dissatisfaction and resistance against it.

3.2.7. Limitations in available legislation and national SEA guidance materials

Legal frameworks are core to the successful implementation of SEA: they define the reasons and legal basis underpinning SEA conduct, set the institutional frame within which an assessment occurs as well as requirements for procedural and substantive clearance (EC 2009). In Botswana, it has been suggested that the lack of SEA-specific legislation is a contributory factor to the low technical quality of SEA reports, the limited awareness regarding SEA, and the ‘little effect’ of SEA on planning (Makaba and Munyati 2018; Matome and Mulale 2023; Matome 2023). EAPs also suggest that the lack of mandatory procedural requirements for SEA militates against their effort to prepare quality SEA: ‘The process of conducting EIA is more clear and elaborate while SEA is just vague. Regulations should be more elaborate on SEAs as they also have an impact on policies and plans’. Authors including Margato and Sánchez (2014) and Retief et al. (2008) have also argued that where SEA-specific legislation is short or absent, practice tends to be divergent, SEA identity is plural and, more importantly, investigating SEA quality and effectiveness becomes methodologically and conceptually difficult. That said, the distinguishing qualities and purpose of SEA in Botswana are uncertain and the legitimacy of any one SEA can be contested.

Similar to some Asian countries including China (Annandale et al. 2021), Botswana’s national SEA guidelines were developed over a decade ago and under the title ‘General guidelines for conducting EIA and SEA studies under the 2005 EIA Act’ (Department of Environmental Affairs 2009). The 2005 EIA Act has been repealed by the 2011 Environmental Assessment Act – also amended by the 2020 Environmental Assessment (Amendment) Act (Matome 2023). Ideally, the use of the 2005 EIA Act (Department of Environmental Affairs 2005) and any supporting materials (e.g. regulations and guidelines) must also be decommissioned, but one EAP remarked: ‘The guidelines, although still used applied to the 2005 EIA Act not the present Act’. The SEA guidelines are therefore regarded as illegitimate and outdated. As a consequence, one EAP commented: ”The guidelines are sufficient in explaining the process of conducting SEA studies, however, there is a need for update and review”.

Surprisingly, some EAPs further claimed that ‘Botswana has no guidelines for the preparation of SEA’ and that what is termed national guidelines are actually “South African“SEA guidelines (see also Matome and Mulale 2023). It also emerged that the Botswana national SEA guidelines (Department of Environmental Affairs 2009) have never been sanctioned by the national parliament, meaning that their state as official documents is only assumed. Here, EAPs admitted a tendency to rely upon ‘more elaborate’ international guidance for preparing SEAs. The commonly cited ones include the UK, Canadian, Kenyan, ‘South African’ guidelines, and the European SEA Directive 2001/42/EC. Accordingly, there is diversity in the approaches taken by SEAs prepared in Botswana; some SEAs (e.g. Envirommetrix 2017; Geoflux 2017) are prepared per the Kenyan National Environmental Management Agency (NEMA) SEA guidelines (NEMA 2011) whilst others (e.g. Mosienyane & Partners International 2015, 2016) followed the South African approach after DEAT (2000). It is significant to highlight that some of the SEAs (Mosienyane & Partners 2016; Envirommetrix 2017; Geoflux 2017) were prepared under the jurisdiction of the Serowe DEA regional office, suggesting that there are not only variations in the coordination of SEA across DEA regional offices but also within regions and across time.

Although international SEA systems tend to share the procedure of assessment, the substantive requirements differ significantly (Tshibangu 2018; Annandale et al. 2021). Consequently, SEA practice in Botswana is divergent, with different substantive issues being discussed to varying extents (Matome and Mulale 2023). Furthermore, while the European SEA Directive takes an EIA-based SEA approach (Bond et al. 2013), the South African approach revolves around presenting opportunities and constraints that the environment places on a PPP and thus risks missing the impact assessment aspect of SEA for a ‘framework for planning’ (Retief et al. 2008, 2021). In this context, not only would SEA approaches, roles and purposes (and evaluation criteria used by reviewers) differ but there is a likelihood that SEA report quality may be uneven. Resultantly, EAPs decried the need to develop ‘home-brewed’ guidelines and identity for SEA, and are, in this context, aligned with the prospects of the national guidelines which aim to develop local guidance based on accumulated knowledge and experience (Department of Environmental Affairs 2009). Such an approach is more capable of developing a context-sensitive SEA, which is key for the overall success of SEA (Retief et al. 2021).

4. Conclusion and recommendations

This research analysed stakeholder perspectives regarding SEA and determined SEA implementation bottlenecks in Botswana through the purview of two classes of street-level bureaucrats, namely environmental assessment practitioners (EAPs) and engineers. It also evaluated SEA regulation and practices in the country. In this context, a limited number of responses were received from EAPs and engineers, particularly from those with extended SEA experience.

The results indicate that while the stake/professional role in SEA conduct cannot be used to explain general knowledge regarding the definition of the concept and purpose of SEA (for this is mutual to all), it is a defining feature for familiarity with deep aspects such as methodology components of SEA (e.g. type of effects to be assessed). In this context, EAPs demonstrated a high degree of ownership of the methodology components of SEA. Furthermore, stakeholders concur that ‘SEA is only worthwhile if it results in modifications to PPPs or informs planning’. Worryingly, physical planners in Botswana lack even basic knowledge pertaining to SEA, however, spatial plans generally drive financial and infrastructural investment planning, meaning that they also have a significant effect on regional ecological footprints (Molebatsi and Kalabamu 2018; Matome 2023). To achieve the United Nations’ sustainable development goals through an influential SEA, it is therefore critical to make physical planners aware of SEA and its potential role in green spatial planning (Thérivel and Minas 2002; Morrison-Saunders et al. 2019; Matome 2023).

Moreover, the results demonstrate that SEA implementation in Botswana is strangled by several problems. These include, among others, passive and ineffectual public participation, lack of data sharing, limited technical capacity regarding SEA and dated guidance, deeply entrenched preferences for distinct SEA approaches, low proponent’s willingness to pay for environmental costs including SEA, oppressive practitioner pay structures, and commercialised environmental assessment consultancy. These manifest as conflict over purpose and approach to SEA, limited awareness regarding SEA, de-skilled planners and diverging practices. Moreover, it also emerged that there is limited buy-in as various institutions are unaware of SEA, that government officials performing the functions of the competent authority are under-capacitated to support SEA, and that stakeholders (including EAPs/SEA consultants) have a weak understanding of the planning and decisional context which SEA aims to inform. That said, these factors are uncharacteristic given how SEA must operate and it remains fair to say SEA in Botswana is far from being effectively institutionalised.

Most of the constraints to the effective use of SEA identified within the action arena may at first glance seem easy to address. Additional training programs can fill knowledge and awareness gaps; formal legal procedures or guidelines can be revised and improved; additional budgetary resources for conducting SEA may be made available by development agencies or by developing a clear ‘cost norm’ for SEA, and so forth. (Slunge and Tran 2014, p. 59)

However, a careful evaluation of these factors indicates that they may only be symptomatic of higher institutional problems. For instance, corruption in Botswana is widespread and has intensified from petty forms (e.g. theft and bribery) to grand forms such as state capture and white-collar crimes involving established businessmen, senior civil servants and the ruling party elites (Sebudubudu 2014; Mogalakwe and Nyamnjoh 2017; Mbaiwa and Hambira 2019). It also extends to include the formulation of self-serving policies and limited regard for the rule of law (Mogalakwe and Nyamnjoh 2017; Molebatsi and Kalabamu 2018; Mbaiwa and Hambira 2019). Also, cultural characteristics such as systems of ‘familism’, loyalty to the extended ‘families’ wrought around institutions and individuals and deference to authority (rewarded through nepotism and/or advancement) are well cultured in Botswana (Sebudubudu 2014; Mogalakwe and Nyamnjoh 2017; Molebatsi and Kalabamu 2018; Mbaiwa and Hambira 2019). In this context, passive stakeholder participation in decision-making (and SEA) as well as limited data sharing enhance these ills (Molebatsi and Kalabamu 2018; Mbaiwa and Hambira 2019). Therefore, much resistance to an effective SEA - involving an open, participative and stakeholder-driven process - may be encountered.

Furthermore, patron–client relationships and the desire to avoid conflicts (and accompanying sanctions) are also common cultural norms in Botswana (Sebudubudu 2014; Mogalakwe and Nyamnjoh 2017; Mbaiwa and Hambira 2019), suggesting that there is a likelihood that SEA in Botswana could be prepared as per the interests of the dominant stakeholders especially that practice is consultancy and profit-driven. That said, SEA consultants would avoid an in-depth evaluation of the decisional context in attempts to not ‘rock the boat’ and preserve relationships. In this context, flexibly applying SEA may be counterproductive; stakeholders may manipulate SEA to meet their specifics and ‘ultimately, this would lead to cementing the secretive, nontransparent and closed nature of PPP making and SEA’ (Fischer and Gazzola 2006, p. 405). Consequently, successful SEA implementation in Botswana and other jurisdictions with similar contextual factors would therefore require a legislated and structured process. That said, the results of this research confirm the notion of many scholars that, to be effective, SEA implementation (particularly in developing nations) must be tailor-made and sensitive to the prevailing institutional, socio-political and cultural context (Fischer and Gazzola 2006; Walker et al. 2016; Nwanekezie et al. 2021; Thérivel and González 2021). Also, SEA must strive to achieve institutional reform and democratise decision-making; that is, SEAs need to achieve higher degrees of transformative learning (Walker et al. 2016) and incremental effectiveness (Bina et al. 2011).

Acknowledgment

The ABSA FG Mogae scholarship fund is greatly acknowledged for the MSc grant to the author which enabled the research underpinning this article. The research was conducted under the Government of Botswana’s national research permit reference number ENT 8/36/4/xLVIII(44). The author would also like to express sincere gratitude to all institutions (BEAPA, ERB, BIE, ACEB and DTCP) which took their precious time and resources to facilitate the research, as well as all informants who participated in the research survey.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. This figure includes 50 SEAs prepared from 1995 to 2003 and 53 environmental management frameworks (i.e. para-SEAs) prepared from 2007 to 2015(Retief et al. 2021).

2. In Botswana, the term ‘environment’ includes ‘the physical, ecological, archaeological, aesthetic, cultural, economic, institutional, human health and social aspects of the surroundings of a person’ (Department of Environmental Affairs 2011: Section 2). As a consequence, it is synonymous with the term sustainability as used by e.g. Therivel et al. (2009).