The Lindi Bay LNG project: a resource blessing or curse for the local community?

(1)

1

The Lindi Bay LNG project: a resource blessing or

curse for the local community?

A Tanzanian case study

Abstract

The main objective of this thesis is to test whether newly discovered natural resources in the form of an LNG project in Lindi, Tanzania turn out to be a resource blessing or a curse at the local level. The Quick Scan Model is used as a methodological tool to make an economic analysis and built three scenarios to create an economic outlook for the local community. Due to its important role at the local level, the quality of local government institutions is newly added to the model. Only through empowering the local government institutions, corruption and rent-seeking can be turned into a focus on entrepreneurship and development. Focus on the local level is therefore essential to turn the Lindi Bay LNG project into a resource blessing for the local community. By informing the local community with three possible scenarios this research aims to empower the local community with knowledge and create honest expectations. Key words: natural resource curse, local economic development, LNG, scenario’s, case study, Tanzania

Final version, 28 June 2017 University of Groningen 

Faculty of Economics and Business 

Master International Business & Management Student number: s2016605

Name: T.F.J. Weustink

(2)

2

1. Introduction

During a field visit related to the recently discovered natural resources in Tanzania, it hit the author that Tanzania was to a point where the Netherlands once was in 1959 when it first found vast reserves of gas in Groningen. The similarities are striking, gas fields are situated far away from central government, and expectations around the discovery are high. Though, when all the media in the Netherlands broadcasted on adverse effects for the local economy of Groningen, a disturbing thought hit the author: “if a developing country as the Netherlands can discover the true cost of natural resources 60 years after discovery, what will be the real impact of natural resources in a developing country as Tanzania?”

(4)

4 Investment Decision.” For the GoT, the LNG project is one of weighing different scenario’s, for example will the gas be sold to the global market in return for tax revenues, creating the ability to invest, or will the gas be used domestically to develop industry and expand energy access? Differing in the level of development the Lindi Bay LNG export project is set to cost about $30-35 billion, representing more than a third of the country’s GDP and constituting the largest investment ever in Tanzania.

Based on the natural resource literature, some countries have managed the resource curse in a very positive way e.g. Norway. Especially developing countries enjoy negative side-effects mainly. In countries like Nigeria, resources have turned the country into a rentier state where only the elite enjoys the power and where through lobbying the system is enforced. Although resource curses can occur in both developed and developing countries focus in this thesis is a developing country: Tanzania. Moreover, in the literature on the natural resource curse, most focus is on the national effects while natural resources are often found in local communities e.g. Groningen and Lindi, where the impact (either positive or negative) are perceived highest. What differentiates this research from previous research is the hypothesis, that central to the reason why a natural resource curse is incubated in particular in local communities is that of the poor institutional quality.

(5)

5 the research question is as follows: to what extent will the ‘Lindi Bay LNG export project' result in a resource blessing or a curse for the local community in the Lindi Bay area?

This research question has led to the development of the following sub-questions to answer in a systematic and conclusive manner.

I. What is the current economic situation in the Lindi Bay area before the construction of the LNG project?

II. What are the positive, negative and neutral, yet possible economic scenario’s in the Lindi Bay area after the construction of the LNG export project?

III. What uncertainties and challenges does each of these scenarios bring to the local community?

IV. What is the role of local government institutions in the current economic situation in Lindi and how is it affected by the construction of the LNG export project? To answer the main question, first of all, an economic scan of the local economy of Lindi has to be developed. Through interviews with local respondents in Lindi and by incorporating the structure of the Quick Scan Model (QSM) by Pennink and Mulder, a local level scan is developed. As mentioned, the importance of the quality of local government institutions is regarded as an important factor in the local economy; it is therefore incorporated into the QSM by this research to assess the economy more accurately. To easily visualize the result of the QSM, this research will make use of a Radar chart to quickly view the score of nine economic indicators.

(6)

6 an LNG plant in Lindi is measured in a positive, negative and neutral impact scenario. To visualize the implications in each scenario a table is developed in which the effects are summarized per level of impact.

(7)

7

2. Literature review: Natural resource curse or a blessing?

This section consists of three different parts, first of all, the natural resource curse in general and on a national level will be addressed. Followed by a discussion on the importance of governmental institutions in guiding the economy towards a resource curse or blessing. At last, the impact of natural resources at the local level, the focus of this research, is discussed.

2.1. The Natural Resource Curse

The phenomenon of the natural resource curse is based on a country’s discoveries of natural resources as oil, gas, and minerals. These findings are in first instance greeted with enthusiasm and thought of as being a blessing which contradictory turns out to be a burden due to adverse effects on the economy. There are many countries where natural resources have proven to be a boon for its community e.g. Norway, Botswana and 19th century USA.1 However, these are merely the exceptions that prove the rule. One of the first explanations of this phenomenon was named by the Economist in 1997 as the ‘Dutch Disease.’2 Following the discovery of gas in the northern province Groningen and in Dutch territorial waters, the national currency was driven up, affecting the international competitiveness of Dutch manufacturing. Not only production is impacted by currency appreciation, with the discovery of natural resources labor and capital are both driven towards the resource extraction business with the services industry experiencing similar undermining effects.3 In fact, the Dutch Disease exemplifies the resource curse in how excessive dependence on natural resources leads to economic regression in a country.

Over the years, scholars have discovered further adverse effects regarding the natural resource curse, adding to an extensively growing literature. On that matter, abundant resource countries are found to invest less in their educational system and human capital.4 Moreover,

1_{See Fubing Su et al. (2016)} 2_{See The Economist (1997)}

(8)

8 scholars argue, besides appreciation of currencies, the fact that capital in abundant resource countries is sucked away from innovative industries and moved to the extraction industry.5 The development of investment in technology and innovation that is stored away is a blind sighting effect of abundance in natural resources, consequently creating a weak position for countries in the long-term when resources are depleted. In effect, when the spillover effects of human capital are the result of employment in the traded sector through inducing endogenous growth in both traded and nontraded sectors, natural resource exports lower the jobs in the traded sector and hamper learning by doing and therefore stunting the economic growth.6_{In this regard,} natural resources lead to an appreciation of the real exchange rate making a country less competitive on three aspects, first by making export more expensive, secondly by reducing the learning and knowledge of its labor force and thirdly through deindustrialization by moving capital primarily to the resource extraction industry.

This step of turning away from learning whether on purpose or by accident, in its most basic form, leads up to a rentier state. The rentier-state is an economy that experiences a resource boom in which lobbying and rent-seeking are the little activities to accrue capital and where the productivity potential of the society declines.7 An example is Nigeria where this rent-seeking proposition was favorable for to elite in an extent that power and lucrative business was secured through granting permits and licenses to loyal followers.8_{A country, in} this case, becomes too severally dependent on the newly discovered resources. Subsequently, "resource dependence elicits corruption and rent seeking via protection, exclusive licenses to exploit and export resources by the political elite, oligarchs and their cronies to capture wealth and political power. Adverse effects of the natural resource curse can be summed up in this regard as the decreasing investments in essential socio-economic indicators while creating a

5_{See Luciano (1990)}

6_{See van der Ploeg (2011)} 7_{See Torvik (2002)}

(9)

9 rent-seeker state in which a ruling elite takes control.

2.2. The role of government institutions

Although some authors argue the weakening of institutions as the minor effect of resource abundance, others argue that natural resources, on the other hand, have had a positive impact on the quality of institutions. Different literature, moreover, suggests that natural resources are mostly neutral on the quality of a countries institution and continue their debate on the matter of measuring.9 However, to explain why particular countries experience natural resources as a blessing instead of a curse, it is best to reverse the sequence of this argument. The initial power and strength of institutions in a particular country can be the determinant of the positive or negative impact of natural resources on the economy. Accordingly, "There is little evidence of a curse where natural resources are combined with human capital and appropriately functioning institutions10.” Furthermore, scholars have suggested that the effect of resource abundance can be explained through the resources to GDP ratio and inequality in the non-resource private sector.11 The emphasis on economic indicators here is the effect of a cause, in this regard the quality of institutions. In the case of developing countries, weak institutions can lead to corruption and rent-seeking behavior, while in developed countries resources are allocated inefficiently, as occurred in the Netherlands.12_{Therefore, the quality of} institutions proves to offer the ability to explain why natural resources turn out a blessing rather than a burden, or vice versa.

A textbook example of the resource curse is Equatorial Guinee, similar to Tanzania a developing country in center Africa, that has discovered vast amounts of natural resources. With the discovery of oil in Equatorial Guinee three decades ago, the elite in this country has exploited the revenues for themselves. With no institutions that enforce transparency and

(10)

10 accountability, the elite only diverts the oil revenues to themselves and consolidate their power. Due to the replacement of taxes by unearned income through natural resources, the social contract between the ruling government and population is severed.13 Wealth gained from resources worsens the quality of institutions at the same time even further since it allows the government to resist modernization, pacify dissent and avoid accountability.14 In developed countries where citizens pay their taxes to the government in turn for public services, the public officials in their way are bound and accountable to their constituencies. Without this incentive, institutions in developing countries are failing their role because the accountability has shifted. Another rhetoric that explains the resource dilemma based on weak versus strong institutions is the difference in entrepreneurship. In the latter, institutions are strong, and entrepreneurship is encouraged translating into more productive activities than rent-seeking activities. While in the rentier state, with a dysfunctional legal system and low transparency, rent seeking has a higher return on investment leading to less productive entrepreneurs. The difference in resource curse or blessing in this sense can be therefore assessed on the quality of institutions, however, as both can experience a resource curse the effects are different.

2.3. Local impact of natural resources

The local economic impact that natural resources have is merely as significant as at the national level. In comparison to the extensive literature that focusses on the economic implications of resource windfalls at the national levels, only a few studies have evaluated the within potential country impact.15 As scholars suggest, the resource curse is sometimes only associated with national effects, while these evidently occur in local areas as well.16_{By this} literature, there are three substantial ways in which natural resource affect the local economy.

13_{See Diamond & Mosbacher (2013)} 14_{See Isham et al. (2005)}

(11)

11 First, the local economy can be impacted through direct channels as increases in local income and employment. Secondly, through spillovers (e.g. when infrastructure is laid out in the local community), and at last, government spending.17 Although not severally investigated, the results of the local impact research have given literature both negative and positive outcomes.

As contested previously, what impacts economic development at the national level is the quality of institutions. One factor is, “the decentralization of executive and fiscal powers to local administrative units to foster regional autonomy, improve service delivery and increase government accountability.18_{” In this regard, local government revenue is expected to be used} on local public goods in better ways as local authorities have a greater understanding of the local needs.19 Furthermore, local governments may have conflicting ideas in comparison to the central government, or they cannot handle some revenues. One case study performed research on the allocation of oil royalties in Italy and concluded there was little impact regarding economic growth and occupation. Profoundly more negative is a case in Brazil, where the corruption of mayors has driven a wedge between some fiscal transfers and local outcomes.20 Here, the money that was generated from oil was used for re-election purposes. According to literature, “the basic message here is that decentralization of revenue spending can work only if government capacity is simultaneously improved.21 Therefore, to grasp direct channel benefits, local government spending and regional spillovers the economy is dependent on the quality of institutions at the local level.

Direct channels are links to natural resources that are frequently mentioned for example economic indicators like employment and income growth. According to literature, the development of oil and gas fields can cause significant impact to a local economy, generating

17_{See Cust & Poelhekke (2015)}

18_Ibid.

19_{See Oates (1972)}

(12)

12 jobs and extracting capital from other regions.22_{Imagine that through the development of an} LNG project and the secondary operations that follow, local spending effects through local employment can go up. In the case of Peru, scholars have proven that the impact of natural resources was substantial in the regard that household consumption and literacy was higher, poverty rates lower for mining districts compared to non-producing regions.23 Moreover, continuing on the same line of direct channels that affect the local economy. In a developed country as Australia the gas industry has shown that localities where significant investments in the gas industry have been made, higher incomes were found in comparison to similar localities with no gas.24 More interestingly is the fact that these results are based on employment in the mining and the non-mining sector. The direct effect of natural resources on the local economy, therefore, is positive in direct employment and indirect employment.

Other effects on the local economy are more indirect as the case is with spillovers, mitigation, and infrastructure. In the regards of spillovers, findings suggest that an increase in resource sector employment can lead to an overall rise in total jobs and even with rising wages in manufacturing, jobs in manufacturing can rise parallel. 25 On the other hand, US landowners collect royalty payments personally, and resource sector revenues are spent locally.26 In other countries, these royalties are collected by the central government, and wage earners are not hired locally.27_{Moreover, in the case of mitigation, a positive is observed for existing residents} in mining villages. Even though the massive influx of people, employment was created with relatively low skilled intensity and poverty decreased.28 Scholars note, however, that in the case of fewer up- or downstream activities (e.g. local refineries), a larger share of the manufacturing sector might suffer from Dutch disease.29 Furthermore, the development of

22_Ibid.

23_{See Loayza, Mier y Teran & Rigolini (2013)} 24_{See Fleming & Measham (2015)}

25_{See Allcot & Keniston (2014)} 26_{See Cust & Poelhekke (2015)} 27_Ibid.

(13)

13 infrastructure also impacts local economic development positively. Scholars note that oil-financed general purpose infrastructure has a positive effect on local economic development.30 Others suggest that especially the benefits of unlocking the domestic reserves and exporting these via ports to the world market have more impact than trade with neighboring countries.31 These indirect effects give an extensive variety on developing the local economy through spill-over effects of the direct channels.

As mentioned in this paragraph, the discovery of natural resources and industry surrounding the extraction can lead to economic benefits. However, the discovery of natural resources can also result in financial fall-back and only result in wins for an elite group. As discussed extensively, the mediator factor between natural resources and economic benefits in the economy is the quality of institutions in place. In the first instance, the quality of institutions is based upon ensuring accountability and transparency in dividing the benefits of newly found natural resources among all stakeholders. Furthermore, highly qualitative institutions provide that during the time of legal extraction, systems are in place to overcome rent-seeking behavior. Institutions should create trade openness and enable investments in technology and learn to keep developing the economy, to avoid a resource curse. In the next paragraph, the operationalization of measuring the local economic impact will be explained, and the complete methodology will be discussed.

30_{See Michaels (2011)}

(14)

14

3. Methodology

To grasp the uncertainty that the construction of an LNG plant in Lindi brings, an analysis is needed that systematically defines every aspect of the economy and the impact of natural resources. Therefore, this research will make use of different tools to gather and analyze data. Primarily, the Quick Scan Model (QSM) is used to create a local economic scan of the Lindi Bay area, based on interviews with local respondents. Secondly, by following the scenario planning literature three scenarios are built upon this economic scan by using data from expert interviews, reports by consultancy firms, NGOs and data from the local respondent's interviews. In each scenario, an answer will be given to the sub-questions of this research. First, in this paragraph the methodology in general is addressed. Subsequently, the method of the QSM and the radar charts will be discussed. At last, the expert interviews are presented followed by the method of scenario writing. In both cases, the data collection for both tools is addressed extensively.

3.1. The value of a case study

This research is based on an in-depth case study; the purpose here is to secure theoretical validity rather than the more conventional statistical validity required of other research. Scholars note that case studies are valuable in situations where little existing knowledge is available.32 Moreover, a case study is a description of a situation which is sensitive to the context in which the research occurs. There are six factors important to a well-written case study research to create a valid and reliable research.33_{First, a theoretical} underpinning must be present; the literature on the natural resource curse. Secondly, a case study is a method in itself. However, it is combined with other concepts and procedures in this case interviews, the quick scan model, and scenario writing. Thirdly, case studies can be quantitative or qualitative; this research is considered a qualitative research where theory is

32_{See Harris and Ogbonna (2002)}

(15)

15 tested, and interviews are the tool for gathering data. Furthermore, this research is based on gathering data through interviews and the analysis of academic literature. Moreover, the unit of analysis has to be defined; the LNG project in the Lindi Bay area where the focus lies on the local economy and the effects of natural resources. To deliver a scientifically justified case study, one has to explain how and why one uses data collecting techniques such as interviews. For that matter this research believes that through doing interviews, this case study can address the risks that come with the LNG plant most stringent due to local expertise in the area that would otherwise never have been described without interviews, "we start with the experiencing person and try to share his or her subjective view. The task of the researcher is to objectively describe it with depth and detail.34Accordingly, for the QSM, 4-8 respondents are required to create a reliable dataset.35 At last, to enhance the validity and reliability of the findings in this research entailing a scientific approach, the focus is therefore on multiple sources of evidence from different stakeholders and experts in the local economy of Lindi and also from experts in Dar es Salaam. This case study is a valuable addition to existing theory and methodology, though, adapted and tested on new ground as will be explained in the next paragraph.

3.2. The quick scan model

This research wants to test the possible effects of an LNG plant on the local economy of Lindi and if the result is a resource curse or blessing. To do so, a methodological tool needs to be used to test locally how direct channels, government spending, and spillovers, as substantial ways for natural resources, affect the economy. To measure the economic context in a specific region, Pennink and Mulder have developed a set of dimensions related to economic development at the national, regional and local level to make a quick scan after a crisis e.g. the earthquake in Haiti or other humanitarian crises.36 The original QSM consists of

(16)

16 a large set of socio-economic indicators tested on three different levels; the national, regional and local level. These economic indicators and the operationalization of these can be found in Appendix I. For this research; the QSM is slightly altered to the specific case of Lindi and the LNG project, focusing solely on the local level to more adequately test the resource curse hypothesis at this stage.

The three substantial ways in which the local economy can benefit the natural resources related to the local economic indicators of the QSM. In order to test the effect for direct channels in the local economy three economic indicators of the QSM fit the profile, for spillover effects of natural resources five economic indicators, and in order to test government spending this research adds the economic indicator Quality of local institutions, see table 1 for an overview of the economic indicators and the divide. As the aim of this study differentiates from the previous use of the QSM, for this particular purpose it is altered in such a way to actually measure and test the effects of Natural resources on the local economy. In order to make use of the QSM in this particular case, this research has changed the method in the following ways: a clear and only focus on the local level, emphasis on the future aspect of the economy and the role of the LNG project, the additional indicator quality of local government institutions, and lastly the difference in respondents. In the following paragraph, the QSM is explained by the hand of discussing the changes made for this research.

Table 1 Measurements of Local impact and the economic indicators of the QSM

Direct channels Spill overs Government spending

Small businesses & entrepreneurship

State of transportation infrastructure

Quality of local government institutions

Employment rates Currency/basis of trade The absence/presence of poverty Energy efficiency

(17)

17 3.3 Alterations to the original model

Although the quick scan is most preferably used during the aftermath of a crisis, the focus of this research is not on ‘quick' as there has been no crisis to anticipate. In this case, there is a 30- to 35-billion-dollar event to happen that will not only impact Tanzania as a country but specifically the small town of Lindi where the project is destined to happen. As scholars already noted, the model consists of so many (socio) economic indicators that the model is suitable for measuring all types of economies.37 The ability of the model to make a scan of an economic situation, while simultaneously producing a look forward by anticipating an event makes the QSM a match for this research. The power of the scan is found in addressing the current situation by interviewing local correspondents who address the economy's weaknesses and strengths, thereby providing a closer look at where the bumps in the road ahead are.

The fact that the economic indicators seamlessly fit within the three substantial ways of local economic impact empower the possibility to draw conclusions on the impact of natural resources. Merely elaborating on the capacity of creating an economic scan, this research expands the QSM to accurately predict an economy's future. What differentiates this change specifically is that respondents have to imagine a scenario where the LNG project is constructed and in operation after they have created the QSM. What this research adds to the current QSM is the second level, the ability to create scenarios. This research, therefore, moves the position of an individual event from prior, to after the QSM. The additional feature of this study to interconnect the QSM structure with scenario planning makes it possible to create an economic scan projecting the future. In this case, the QSM functions as a tool to predict which economic indicators are weak or strong in the case of a particular event, based on the view of local experts.

(18)

18 The QSM in its original form is designed to balance the more stringent focus on the national level by focusing on three levels namely: the national, regional and local. Scholars note that with any humanitarian crisis, the local level is the most important.38 Local economies, as a result of globalization, are the markets that can create jobs and further stimulate the local economy.39 Therefore, instead of solely focusing on the national level, the quick scan model has included the regional and local level. For each of these levels, a list of economic dimensions and indicators has been developed, see Appendix I. However, as scholars note: "all the quick scans for all three levels can be performed separately if one should only need information on a single one of these levels."40 Because the large LNG project is designated to be constructed in Lindi where it will alter the economy and the fact that consultants and other researchers have analyzed the national impact, this research will only focus on the local level. Accordingly, the model should be tested on an economy that offers still many changes for growth.41 The choice for Tanzania, with the focus in this case on Lindi, still fits the picture of a developing country as it did two years ago when Mulder for the first time tested the model.

In the original QSM, eight socio-economic indicators are analyzed to create the full picture of a local economy. In developing the quality and strength of the model, this research adds the ‘quality of local government institutions' as an economic indicator at the local level. Moreover, in testing the quality of government institutions, the possible impact on government spending can also be addressed through the decentralization of power. Moreover, it has an indirect effect on other economic indicators as mentioned above in the literature, high quality of government institutions can perform as the engine of economic development by creating a legal system in which corruption and rent-seeking behavior are discouraged, inherently encouraging for example entrepreneurship. Local government institutions play a central role in

38_Ibid.

39_Ibid.

(19)

19 local communities.42_{This research will, by assessing the quality of institutions at the local level} regarding decentralization of power, accountability and transparency enable the possibility to view whether the government institutions are capable of coping with a future crisis or event with the size of the LNG project.

To create an economic scan of the local economy as described above it is important to find local stakeholders that have knowledge on the economic indicators. Through semi-structured interviews, the researcher first asks the stakeholders to give their professional opinion on each economic indicator by indicating a score between 1 and ten on each sub-question, with the possibility to elaborate further by adding a note. This approach is different from the approach by Mulder in that the possibility of the subjectivity of the writer is reduced. For each economic indicator, the sub-questions are added up to combine an average. To come to an overall score per economic indicator, the scores of all respondents are averaged. In the end, to arrive at a final score for the economy of Lindi, the mean of all economic indicators is calculated. Moreover, the local stakeholders should be able to actively participate by considering the possible challenges and benefits related to the possible event where an LNG plant is constructed in Lindi. Therefore a certain level of English is required. To overcome the problem of the subjectivity of stakeholders perception, more than one stakeholder should be interviewed, with the ideal amount of 4 to 8 respondents per indicator.43_{Based on the economic} indicators, the questionnaire as developed by Mulder adapted with the new measurement ‘quality of government institutions' and extended with six open questions. The questionnaire can be found in Appendix II.

42_{See Raco & Flint (2001)}

(20)

20 3.4. Radar Chart

Following the example study of Mulder, this thesis will use the visualization of a radar chart to present a clear overview of the obtained data, as nine different economic indicators require a clean and proper summary of the results. The presentation of data written in a table is critiqued as not creating the easy overview of data that a quick scan model would need.44 Therefore, these radar charts prove to be an outcome that anyone can quickly understand. Each axis in the radar chart represents one local economic indicator. The lower the valuation per indicator, the lower the score will be on the axis on the graph. As scholars note, the visualization in charts will easily make the comparison of situations comprehensible for all users.45_{With this easy visualization, the local community in Lindi can identify the situation} easily, and a brief explanation on the economic indicators would suffice to educate the community about the effects. More detail on the operationalization of these economic indicators can be found in Appendix I.

3.5. Scenario planning

This thesis uses scenario planning, which has the potential of applying to almost any context, situation or problem, and its evolving nature.46 “The most valuable advantage of creating and using scenarios is the recognition that uncertainty is a basic feature of organizational environments.”47_{Scenario Planning initially was used by large corporations and} governmental institutions, but it has since then developed rather rapidly and is now used also for the benefit of communities. Several scholars have used scenario planning in small communities, in Panama, for example, researchers have looked at how uncertainty about the future and changing demographic will change a small community dependent on trading lobsters

44_{See Mulder (2014)}

45_Ibid.

(21)

21 and coconuts.48_{On the other hand, researchers have used scenario planning to investigate the} prospective result of banning the hunt on polar bears in a community where hunting these animals is still a big part of their life.49 The implications of scenario planning can be described in these cases as twofold, first scenario planning creates space for discussion on options that were previously discounted, and secondly, scenario planning is useful in empowering communities for the development of adaptive governance.50_{On both these matters, the} community of Lindi can benefit by expanding their knowledge and future options while simultaneously empowering their position.

To use the community-based scenario planning, this research follows the methodological basis as used in the small community in Panama.51 The process of this scenario planning occurred in four stages: First, "definition of the purpose of the scenarios." In this research, that purpose is clear: ensure and empower local actors in the Lindi Bay area with knowledge and inform the locality on challenges and uncertainty. Secondly, "information about the system's structure and the primary drivers of change. In the case of Lindi, the major driver of change will be the construction of an LNG facility, to analyze the structure this research will take a look at the economic profile of the Lindi Bay area by using the Quick Scan Model and interviewing key stakeholders. Thirdly, "generation of the scenarios." Through forecasting, this research will establish several scenarios where the economic indicators of the QSM are used to structure the research. The scenarios are grounded and developed on the current economic situation in Lindi, and the uncertainties and challenges ahead, according to local stakeholders. At last, the implications of the scenarios and the use by decision makers will be drawn. In the final conclusions on these scenarios, this research will summarize the implications of the most likely scenario thereby answering the main question, while emphasizing the role in each of the

48_{See Rawluk & Godber (2011)}

(22)

22 local government institutions.

3.6. Impact Overview

To easily give a clear and easy overview to the local community, at the end of each scenario a radar chart is presented in which the impact and direction of the economic indicator is drawn by an arrow, thick or small differentiating the impact. Furthermore, a full summary of the scenario’s is visualized at the end of each scenario including an overview of the developments. Moreover, with basic + and – to summarize the results of the economic indicators, conclusions can be drawn quickly. By providing simultaneously an overview of the developments in the economy that lead towards these results, the local community can grasp the prospective futures without having to read the entire research. Following the impact overview for each scenario, in addition to the radar chart for the current situation, a conclusion is given on the impact of natural resources on the local economy for each scenario.

3.7. Collecting data in Tanzania

The data collection was executed in Tanzania, two weeks in Dar es Salaam and one week in the Lindi Bay area. Through semi-structured interviews, stakeholders from different groups are asked to give their view on the effect of the LNG plant on the local economy in Lindi. By using interviews as a technique to gather knowledge, one has to acknowledge the fact that subjectivity can sneak its way into the results of this research. However, “subjective interpretation occurs throughout the total research process, including the final step of reading the research report.52" The role of the researcher, in this case, is to reflect the social world it tries to describe as objective as possible by detailing the stakeholder's background to understand the results in a scientific manner. One major consequence of acknowledging the subjectivity of a case study is the need to make the subjectivity visible to the reader, which is therefore emphasized

(23)

23 3.8. Data collection for the Quick Scan Model

Critical to the result of this local economic scan was a broad range of respondents possible to overcome subjectivity. For the research to be as objective as possible the target of this research was to interview local respondents from Governmental, Business and NGO organizations. Secondly, an important factor is that the respondents can critically assess the current economy of Lindi and their capacity to imagine, discuss and create new scenarios based on the event of an LNG project, hence a moderate level of education and English is required. After the data collection in Lindi, the total number of respondents for the economic scan is 6 in total. Due to difference in sexes, organizations, backgrounds, and professions a diversified scope of answers is secured. Out of 6, four were male, and two were female. The final respondents work for four different type of organizations, three of them are local, 2 of them are foreigners living in Lindi for over a year, and one is a foreigner living in Lindi for more than four years. Due to the language barrier and complexity of the questionnaire, it was not

within the capacity of this research to interview national SME owners or entrepreneurs. Fortunately, two out of six respondents actively work with their NGO to improve the quality of SMEs and entrepreneurship in Lindi while another works for one of the only large SMEs in Lindi. For an overview see table 2, and for further details see Appendix IV.

3.9. Data Processing – Local level

(24)

24 With all interviews in Lindi finished, the consequent data is turned into transcripts that can be requested with the author. To analyze the data and create radar charts an overview of the answered scores is made, see Appendix V, moreover an overview of open questions and answers is created, see Appendix VI. Finally, to create an overall average score per economic indicator, all sub-questions are weighted equally to create one score per economic indicator. With all scores for each economic indicator per respondent calculated, the average of the six respondents is calculated per economic indicator, all weighing equally. These scores are finally used to create the local radar chart. Moreover, the summarized open questions are used in the scenario writing.

3.10. Data collection for scenario writing

To develop scenarios on the economic situation in Lindi, four different data sources are used. First, several reports have been written, ordered by the government, by consultants, government and educational institutions to develop knowledge on the LNG project. Secondly, through semi-structured interviews with eight respondents representing five different types of organizations in-depth knowledge is documented about the LNG project and the economic prospects of Lindi. Of the interviewees, only one was female thereby under-representing her sex. Furthermore, six out of eight where national Tanzanians, only two are foreigner working

(25)

25 overview see table 3, and for further details see Appendix IV. Thirdly, due to the non-responsiveness of government institutions, a conference on Energy was attended to gather more data from a governmental perspective and to get confirmation on certain hypotheses by experts. At last, after the questionnaires with the local respondents were completed, short semi-structured interviews followed with the six respondents to observe their view on the LNG project, the greatest challenges, and benefits and the role of the local government institutions.

3.11. Data processing – expert level

To process all the data, tables are used in which the answers are differentiated on four scenarios. In preparing the answers of the respondents in this way, the data can be easily used to develop the current situation and three scenarios. Individual comments as made by the respondents are employed in multiple scenarios as they can be interpreted as spanning across both. A limitation of dividing the respondent's answers in this way is the subjectivity of the scenario writer. However, by carefully documenting and storing the interviews and recordings plus by forcing the scenario writer to be extra aware of this limitation this research hopes to overcome this matter. Furthermore, the local respondents have all rated their top 3 economic indicators by the most challenged and benefited. By using a simple point system where the number 1 indicator receives three points and the number 3 only 1 point, this research has created a list ranking the nine indicators from top to bottom. In congruence with the comments of the respondents, these are used in the three scenarios to develop the argument.

4. Data Analysis

(26)

26 discussed based on the expert and local interviews in combination with several reports. These positive, negative and neutral scenarios will answer the second sub-question: what are the positive, negative and neutral, yet possible economic situation in the Lindi Bay area after the construction of the LNG export project? As part of the scenarios, the uncertainties and challenges are addressed by using the local respondent results to answer sub-question 3: what uncertainties and challenges do each of these scenarios bring to the local community? Moreover, in the current economic situation, the role of local government institutions is discussed as an extra factor to the QSM. Additionally, in the three scenarios, this factor is discussed more extensively to answer the last sub-question: what is the role of local institutions in the current economic situation in Lindi and how is it affected by the construction of the LNG export project? Moreover, concluding each scenario the impact regarding the resource curse will be discussed. At last, in the discussion, the implications in regards to the natural resource literature is discussed combined with the most likely scenario according to this research.

4.1. Economy of Tanzania

Since its independence in 1961, The United Republic of Tanzania managed to grow its economy throughout the 1970s. In 1978, however, the economic decline started to settle in under the socialist government of Julius Nyerere lasting until the early 1990s when liberalization was allowed by the Tanzanian authorities to stimulate greater private sector engagement.53 With new investments in the mining industry, positive growth in Tanzania has been achieved from 1996 onwards. In the period between 2007-2014, the country accumulated a growth trend of around 7% annually.54_{After this first wave of liberalization policies, the} country is now working towards the realization of its development targets, articulated in the Tanzania Development Vision 2025. The program, in Swahili, called the Mkukuta II strategy

53_{See Roe (2016)}

(27)

27 has as its main targets, to reduce income poverty, improve the quality of life and social well-being and to improve governance and accountability.55 The strategy is based on the Malaysian "Big, Fast Results Model" and focus is on implementing specific goals along with a stipulated delivery timeline, with the primary goal to move Tanzania to the level of the middle-income country. One of the six key goals is energy, as consumption of electricity is positively correlated to Gross National Income.56_{The objective is to lead the country to five-fold its} electricity consumption by 2035, where currently Dar es Salaam is the only region that benefits the national electricity grid properly.

For the country to use electricity, Tanzania is producing gas for over ten years now in the Songo Songo region, an area south of Dar es Salaam. More than 75% of the gas produced, 74 mmscfs/day, is pumped through the Marine Gas Pipeline directly to the capital to be used for power generation. More south, the French-Anglo consortium of Maurel et Prom, and Wentworth in 2015 found new gas reserves in the shallow water of Mnazi Bay. It is due to significant investments by Chinese investors that a 487 km Onshore Gas Pipeline was built in August 2015 from Mnazi Bay towards Dar es Salaam to provide the capital with an additional 100 mmscfs/d. Moreover, the Tanzanian Petroleum Development Corporation (TPDC) which is formed in 1969 has the role of overseeing oil and gas operations. Over the last decade, due to the more liberalization policies, there have been enormous exploration activities in the Indian Ocean as concessions were granted by TPDC. The most impressive discoveries have been offshore along the Southern coast of Tanzania. Companies like the BG Group, Statoil, ExxonMobil, and Orphir have in total discovered natural gas, multiplying the Tanzanian reserves from 10 trillion to 57.2 trillion cubic feet (tcf). To place the discovery in perspective, the Songo Songo and Mnazi Bay resources combined are estimated to be two tcf.

55_{See Peng & Poudineh (2016)}

(28)

28 4.2. The LNG project

The major gas fields found by the IOCs are located on the Southern coast of Tanzania. To exploit the major gas discoveries in this region, the government has issued an integrated project team involving the four major corporations to study the viability of an LNG plant that would enable the possibility to exploit the gap in global supply while also meeting domestic needs.57 The first phase of the project was finished last year namely the land acquisition in the Lindi Bay area. The responsible party for this acquisition is TPDC, the operational government party involved. Currently, the IOCs and the government are in the pre-feed phase.58 The two main issues discussed between the IOCs and the government, according to experts, are land related issues (e.g. land reclamation, development, etc.) and the commission negotiation (who receives what part of the pie). Another problem is related to the local content policies. Due to the complexity of negotiations and the little experience of ministries, it is tough for the government to communicate and negotiate with the IOCs as one team. The TPDC as a central organ has to communicate with the IOCs in congruence with eight other ministries that al have little experience with a project this size. The projects estimated size is around 30-35 billion dollar, with still many variables open for negotiation ranging from constructing a port, airport or compound for construction works to an industrial park

In the case that the LNG project would reach FID, the IOCs are the responsible party for retrieving the gas, with BG in the lead, while TPDC has responsibility for the land and social issues surrounding the LNG project site. The role of TPDC here is to develop a strategy for the local community and together with local government institutions implement these. On the same level, the central government has set out the target for local content policies to move the IOCs in developing the local community besides building an LNG facility solely. “Local content is a commonly used term referring to the participation of nationals and companies in

57_{See Peng & Poudineh (2016)}

(29)

29 a sector's activities and supply chain, but could also extend to infrastructure and community benefit.59” In this regard, the local content policy as drafted by the GoT is centralized on the promise of securing jobs for Tanzanians in the oil and gas supply chain, emphasizing short-term opportunities. Currently, the GoT and IOCs are in negotiation on the final local content policies to be implemented.

Over the same period after the discovery of the gas in Tanzania, even triple the amount of gas was found in neighboring country Mozambique. According to some experts, the quality of this gas and the reachability of it is better. The general trend currently in Eastern-Africa, therefore, is that IOCs focus more on the project in Mozambique where, due to large debt problems, the willingness of the government is larger than in Tanzania. However, Mozambique is according to experts more explosive in comparison to the stable political Tanzania. On the other hand, Tanzania is regarded as a country with a socialist background, meaning that the attitude towards multinationals is hesitant. Moreover, the country is fed with high expectations regarding the LNG project, and it is almost as id cities of gold are promised by Ministers in the national media. Especially in the south this creates a large gap between reality and believe. In the next paragraph, this research will explain the economic situation of Lindi.

4.3. The economic situation in the Lindi Bay area

The Lindi Bay region is home to around 224.000 households which accumulates to approximately 900.000 inhabitants. In the region, around 600.000 inhabitants are between 0 and 35 years old, and around 78% of the people live in the rural area while 200.000 people are considered to live in the urban areas. Specifically, and the focus of this research, there is only one large urbanized area in the Lindi region, namely Lindi town with around 50.000 people.

(30)

30 Expectations around the new LNG project are primarily high in Lindi due to the government which created the idea that it will deliver many jobs to the local community, high tax revenues and ensuring electricity availability for the country. To describe the economic situation of the local community in Lindi, a summary of the responses of the six respondents per indicator will be given as outlined in the research methodology. To answer the fourth sub-question, the first economic indicator on local institutions will be elaborated more than the other economic indicators. Following the summaries on the economic indicators for the local community of Lindi, a radar chart will be represented to view the economic situation more clearly.

4.4. Quality of local government institutions

(31)

31 by the government institutions with the local community and SMEs on opportunities regarding the LNG project. Moreover, corruption within the local government apparatus is perceived to be very high.

On the matter of accountability, a 5.2 is the weighted scored which is interestingly lower than the transparency measure. Although both the district council representatives and the NGO volunteer give high ratings, 5 out of 6 respondents perceive the institutions to be less accountable than they are transparent. Controlled by the central government, the local institutions have little say in things, though the areas in which they have control, they accordingly take a little accountability for their actions.

At last, regarding decentralization of legal and fiscal power, a 4.8 is scored implying a low level of authority for the local government institutions. According to the district council representatives, there is little decision making power for local public institutions, all cases are decided upon by the central government. Local laws are called Sharia Ngogo, decision power over these are decentralized. However, these laws can only be created and enforced if they don't violate the mother law. Out of the six respondents, five agree with this view on a centralized government system where the local institutions have to obey and enforce the law of Dar es Salaam.

This research needs to note that this indicator is limited to the extent that two out of six respondents have given heavily skewed answers in comparison to the other interviewees. Therefore, to conclude, the local government institutions in Lindi are currently sufficiently transparent in their actions, score negatively on their accountability and regarding decentralized power score very low.

(32)

32 Small businesses and entrepreneurship

The main economic components in Lindi according to experts are mainly cashew farming, fishing and a little bit of cassava farming, while there is no particular industry or tourism. Most of the farming in Lindi is primarily for supporting the own family. On this matter, most of the respondents stated that on average they believe 83 % of the businesses is small, including the small farmers that occasionally sell their crops in town. Many families have a small grocery, clothing or another shop where they sell their small amount of goods to their neighbors. The overall professionalism of these businesses is considered to be low. One of the NGO volunteers that actively worked with SMEs in trying to upscale professionalism stated that almost no small businesses are using financial books. Due to small family businesses and helping each other out with granting small jobs to one another, hidden unemployment among the Lindi Bay area is accordingly high.

The mean score on Small Businesses and Entrepreneurship is 4.7.

Employment rates

Employment rates are estimated to be very low, averaging around 30%, with most ‘official' jobs are in local government, banks and NGOs. Especially among young boy's unemployment is extremely high because at the end of high school an English exam is required. However, students don't receive English classes as there is a lack of English teachers. Unable to finish high school, opportunities for jobs in Lindi are limited. Moreover, the Tanzanian coast culture is attributed to causing the lack of interest in working. In the case of farming on medium to large scaled farms, workers from Zimbabwe are frequently employed as they are eager to work while the people in the Lindi Bay area are not.

(33)

33 State of transportation infrastructure

From Dar es Salaam a newly constructed road was finished around 1,5 years ago. The road was built by Chinese and was the result of the recently discovered gas in the southern part of Tanzania, now connecting Mtwara and Lindi in respectively 8 and 10 hours to Dar es Salaam. The road, however, is built directly on the sand beneath it and although the road is not heavily used, cement trucks from Mtwara to Dar es Salaam are slowly destroying the road already due to their heavy weight. Besides this main road, the amount and quality of roads in the city center and surroundings of Lindi are weak. There is no railway in and around Lindi where there is a small airport with potential; however, the airport in Mtwara is used more frequently by for example IOC employees to fly to Dar es Salaam.

The mean score on the state of infrastructure is 4.33.

The presence or absence of poverty

The poverty in the Lindi area is accordingly very high, in one respondent view, it is astonishing in what way the people can survive on so little. Using family or friends, they support each other with everything they have. The average income is very low if you imagine that the minimum wage is 4.000 shilling for suitable jobs, while the locals among each other if they need help harvesting pay each other 2.000 shilling. The average income a day can be assumed not to be higher for most people than 2.000 shilling a day as no one has regular day-to-day jobs and the local wage is only that. When it, however, comes to consumption, this is perceived by many of the respondents to be higher than income. This is related to the fact that most of the money earned each day is immediately spend on food or other goods perceived important at that time.

(34)

34 Currency/ basis of trade

The currency used in Lindi, as mentioned, is Tanzanian shilling. On the ground of whether the system in Lindi is more monetized or trade-based the respondents lean more to the latter. Accordingly, there is still a system in Lindi in which work is done in return for goods and food. On the regards of inflation in Lindi, as much of the vegetables for sale on local markets are imported from Dar es Salaam prices are already relatively high for those goods. However, due to speculation on the LNG project, prices have risen slightly according to one respondent. Moreover, the currency in Lindi is perceived to be relatively stable in the same line as at the national level.

The mean score on the currency/basis of trade is 4.75.

Energy efficiency

In Lindi due to the little industry around the area, pollution is considered to be very low. However, others mention that garbage on the edge of the roads and the beach and ocean are a greater problem. Moreover, many of the people cook at night with charcoal which is polluting too. In regards to the community's energy input to the GDP, this is very low. There is no energy input in Lindi.

The mean score on energy efficiency is 3.67.

Innovation capacity

(35)

35 though even fewer people take advantage of the opportunity. Moreover, many people invested after the LNG rumors started, now everything is empty, and uncertainty in Lindi discourages innovation.

The mean score of innovation capacity is 4.2.

Accessibility of basic goods

The access to basic goods in Lindi is different than in a large city like Dar es Salaam. For example, vegetables are imported from Dar es Salaam to Lindi though this does not mean that all locals can afford the vegetables. Especially clean drinking water is considered a scarce product, where bottles of water are available in grocery shops most people use other local water sources. Moreover, in the Lindi Bay area, sanitary facilitations are considered relatively accessible to all locals. However, the availability of gasoline (petrol) is deemed to be low, although there are a few petrol stations in Lindi, many people cannot afford to buy and use it. The same is considered for healthcare; there is a hospital run by two German doctors though not everybody can afford it and on top of that locals have high trust in local witch doctors. In all 224.000 households, only 30.000 have electricity to use for lightning and small household electrical machines. Around 100 families use natural gas as their main energy supply for lighting. The most used power source to retrieve lighting is kerosene for 128.000 households. For cooking the most widely used source is firewood, still used widely spread across the region by 190.000 households, almost 90%.

(36)

36 4.5. Radar chart

In the radar graph below in table 4, one can see the overall economic situation in the local economy of Lindi. When combining all economic indicators together the aggregated level in Lindi is 4.3. In the original QSM, without the quality of local government institutions, the score for the local level would have been 4.2, however, due to the high score on the newly added indicator, this is 0.1 point higher in this revised version. The first conclusion that can be drawn on the local economy’s status is that it is insufficient. According to scholars, “a value marking closer to the center implies conditions that are relatively worse: less business, more unemployment, more inflation, less infrastructure, fewer roads, more electricity cuts et cetera.60_{” In congruence with observing and interviewing people in Lindi, this radar chart is} accurate on its average score, implying a relatively worse economy. Secondly, significant is the outlier of the score on quality of local government institutions. Due to high scores by two out of six respondents, the new economic indicator is skewed and higher in comparison to the

other economic indicators. The subjectivity of at least one of these respondents has to be taken into

60_{Pennink & Mulder (2014)}

Table 5 Radar chart local economy of Lindi

(37)

37 consideration as he represented the District Council of Lindi. At last, the economic indicator absence or presence of poverty scored the lowest of all indicators with a 3.3, which implies that poverty is perceived very high by local stakeholders. The radar chart in table 4 is the basis for which the scenario's in the next paragraph will be developed.

4.6. Scenario’s on the economic situation of Lindi

(38)

38 4.6.1. Scenario one – Positive yet possible development of Lindi with the LNG project

In this first scenario, the LNG project has a positive effect on the local economy. Besides the LNG facility, an industrial park and port are constructed. Political pressure ensures that local institutions are developed to secure sustainable development in Lindi.

4.6.1.1. Developments in Lindi, Tanzania

The discovery of natural gas in both Mozambique and Tanzania has caused for fierce competition between both countries to supply their offshore gas, through LNG, on the world market. Tanzania, due to its socialist history has always had an inward-focused energy strategy to provide its constituencies with energy.61 Mozambique, on the other hand, has a more outward-oriented energy strategy based on fractionalized political elites susceptible to external pressure.62 After the civil war in Mozambique that lasted until 1992, violent tensions arose again in 2013 between the two most famous political parties RENAMO & FRELIMO, as militants of RENAMO started their guerrilla attacks once more.63_{This political turmoil has} escalated in 2018 and resulted in the drawback of international oil companies and the putting on hold of the LNG project in Mozambique. On the other hand, Tanzania under its president Magufuli experienced liberalization of economic policies.64 However, in the 2020 election, President Magufuli unexpectedly is defeated by opposition leader Lowassa, a former prime minister. It is the first time in the country's history that the CCM party of Magufuli is defeated, primarily driven by the opposition's success in the southeast, which "is founded on the belief that natural gas revenue will not benefit the local community, but will instead be diverted to Dar es Salaam."65

Although the opposition has a more inward focused strategy than Magufuli's CCM, Tanzania is considered to be the most politically stable in the region and will not explode like

61_{See van de Walle (2001)}

(39)

39 other countries around that area. Therefore, the IOCs focus on the LNG project in Lindi. From the late 2010s, the prices of oil and gas start slowly rising again due to political unrest in Mozambique and other parts of the world while still, the demand for LNG is high in Asia.66 With the continuing increase in prices of oil and gas in the early 2020s, the LNG project in Lindi shows greater economic viability for both the IOCs and the Government of Tanzania (GoT). Therefore, with President Lowassa fresh in office in 2020, the GoT closes a deal with the IOCs in which 20% of the natural gas over the lifespan of the project is used for domestic supply. By doing so accordingly, Lowassa delivers on his promise to his political base, while 80% will be liquefied to LNG and sold on the open market by the IOCs.

The divide is in accordance, with the Petroleum Act 2015 under the Obligation Domestic Supply, where the IOCs and the GoT are required to come to an agreement on the percentage of domestic supply.67 During the negotiations, where the high oil and gas prices boosted the negotiation position of the GoT, it is decided that in Lindi a new port will be built to receive large LNG vessels and the Lindi airport will be developed to serve air transport from and to Dar es Salaam. As 20% of the gas will be used for domestics use, a 2000-acre industrial park will be constructed close to the LNG facility offering space to all sorts of industry such as a fertilizer plant and a petrochemical factory. Moreover, to provide the domestic market with gas and electricity, a pipeline will be connected to the Mtwara-Dar es Salaam pipeline to deliver gas to the national grid and a second electrification plant is built in Lindi to secure electricity supply in the southern part of Tanzania. As projected, the construction of the LNG facility takes four years before the LNG goes online in 2024, other facilities are developed gradually through the years following.

During the building of the LNG plant in several phases, different kind of skilled workers, approximately 1000 at the time, are stationed in an offsite compound in Lindi and in

66_{BP (2016)}

(40)

40 the town itself. After the construction is finished, 400-800 primarily expats are stationed in Lindi to control and operate the facility. Moreover, construction workers from all over Tanzania are moving towards Lindi to build all the required plants, accommodation, and infrastructure to develop further the Industrial Economic Zone to which the LNG plant is central.

TPDC has been able to minimize expectations during the extended negotiation phase 2017-2020, to overcome a similar situation as in Mtwara in 2015.68 Through consulting and planning and partnering with the local authorities, TPDC has been able to minimize the risks within the local community by relocating locals by providing alternative social services like education. In cooperation with the IOCs, TPDC has ensured the local community to adapt to the changing environment. For Tanzania to develop itself to a middle-income country and reach the goals that it has projected in the 2025 vision, the only option has been to increase the domestic supply of energy. Moreover, the GoT in cooperation with the IOCs, has changed the local content policy focus from oil and gas to a broader market focus, including Agriculture, Tourism, and Logistics to support the industry in Lindi. By emphasizing these high potential sectors, including also the domestic gas-based industries, Tanzania is accordingly found to have a far greater probability of achieving its 2025 development aspirations.69 With the LNG facility and an Industrial Economic Zone in Lindi, this is secured, however, communication with the local institutions and community has been essential to overcome situations like in Nigeria.70 In the next paragraph, the effects of natural resources on the domestic economic indicators as described with the QSM will be discussed.

68_{BBC (2013)}

The Lindi Bay LNG project: a resource blessing or curse for the local community?