Shifting Dutch priorities : changes in discourse, financing and decision-making for international development in the Netherlands

(1)

(2)

1

Master in International Development Studies

Graduate School of Social Sciences Faculty of Social and Behavioural Sciences

Universiteit van Amsterdam

Master’s Thesis

SHIFTING DUTCH PRIORITIES: CHANGES IN DISCOURSE, FINANCING AND DECISION-MAKING FOR INTERNATIONAL DEVELOPMENT IN THE NETHERLANDS

Gabriel Ramirez Acevedo Student number: 11181435 gabrielram@hotmail.com

10 January 2017

Supervisor: Dr. Courtney L. Vegelin. Master’s Director IDS, University of Amsterdam Second reader: Dr. Yves van Leynseele Master’s lecturer, University of Amsterdam Word count: 24,944

(3)

2 Abstract

Dutch development policy experienced changes in its funding and formulation in the period between 2007 and 2016, shaping priorities for international development and directly affecting developing countries in the process. This research examines the discursive principles in Dutch development policy and the decision-making processes involved in selecting certain development priorities to be funded by the Ministry of Foreign Affairs from the Netherlands. The study builds on the critical discourse analysis proposed by Maarten Hajer for exploring discourse in policies. Based on the question of “How are discourse, policy and financing from the government of the Netherlands shaping decision-making priorities for Dutch international development?” the research draws examples mainly from the Food Security spearhead priority of the Dutch development cooperation. Methodologically, this study combines a qualitative discourse analysis of policy documents and interviews to key respondents of the policy cycle, with a quantitative analysis of the budget and expenditure data available from the International Aid Transparency Initiative (IATI) registry. The main findings from the research suggest that in 2010 an emerging trade discourse contended with the established aid development discourse, which led to the creation of the “aid and trade” agenda by the Dutch government in 2013, in an effort to balance both positions under the pressure of budgetary cuts. Guided by the “aid and trade” policy, decision-makers make use of different discursive principles and of available funds from centralized and decentralized budgets, to determine what is deemed by them as the “best” option to be prioritized. However, the choices for implementation that are reflected in the expenditures until 2016 do not necessarily seem to reflect strong trend changes towards trade oriented priorities, with the exception of food security policy. At a theoretical level, this research may contribute to the debate of aid effectiveness and the relationship between politics and development policy. Empirically, it might prove useful for researchers and policy makers in understanding how development policies are made in the Netherlands, as well as for recipient countries and organizations to be able to have further insights into the processes that lead to certain development priorities.

(4)

3

4

5.3 Who makes the decisions in Dutch development policy? 40

Closing Remarks 44

Chapter 6. Policy-making: the “aid and trade” and food security policies 45

6.1 The good and the bad of the “Aid and trade agenda” 45

6.2 Drawing examples from food security policy 51

Closing Remarks 55

Chapter 7. Following the money 56

7.1 Analyzing the IATI data 56

Closing Remarks 63

Chapter 8. Final discussion and answers to research questions 64

Chapter 9. Looking Ahead 66

9.1 Personal Reflection 66

9.2 Future Research 67

9.3 Policy Recommendations 68

Appendices 69

Appendix A: Anonymized list of respondents 69

Appendix B: List of documents and policy analyzed discursively 70

Appendix C: Operationalization of Major Concepts 71

Appendix D: 2011 Dutch food security policy intervention logic 72

(6)

5 List of Figures

Figure 1: Map of the Netherlands ______________________________________________ 11 Figure 2: Policy Cycle________________________________________________________ 18 Figure 3: Conceptual Scheme _________________________________________________ 21 Figure 4: Dutch ODA and ODA by all OECD/DAC countries between 1960 and 2014 in percentage of GNI __________________________________________________________ 29 Figure 5: “Food Security” labeled disbursements 2007-2016 ________________________ 57 Figure 6: Reported “Food Aid/Food Security” disbursements 2007-2016 _______________ 57 Figure 7: Yearly disbursements by policy priority 2007-2016 for the Food Security priority 58 Figure 8: Aid for Trade disbursements 2007-2016 _________________________________ 59 Figure 9: Comparison between “Aid for Trade” and “Not Aid for Trade” marked disbursements 2007-2016 ___________________________________________________ 60 Figure 10: “Aid for Trade” marked disbursements for the “Food Security” budgeted policy priority 2007-2016 _________________________________________________________ 60 Figure 11: Comparison of the “Aid for Trade” marked disbursements for all budgeted policy priority lines 2007-2016 _____________________________________________________ 61 Figure 12: Number of “Aid for Trade” marked projects under the “Food Security” budgeted policy priority running per year from 2007 to 2016 ________________________________ 61 Figure 13: MASP budgets: Total, Food Security and Private Sector development for 2014-2017 ____________________________________________________________________ 62 Figure 14: Country disbursement IATI data from the policy priority marked as food security in the budget ______________________________________________________________ 62

List of Tables

Table 1: Perceived attitudes of political parties towards development ________________ 39 Table 2: Approaches, priorities and principles used in the 2014 MASPs ________________ 42 Table 3: Respondents’ perceptions of “Aid” and “Trade” ___________________________ 47 Table 4: Respondents’ perceptions of the “Aid and trade” approach __________________ 48 Table 5: Analysis of selected Choices in the 2014 Food Security policy _________________ 53 Table 6: Disbursements reported to IATI by policy priority (in € millions) _______________ 59

(7)

6 Glossary and Abbreviations

AIV: Advisory Council on International Affairs from the Netherlands (Adviesraad Internationale Vraagstukken)

CDA: Dutch Cristian Democratic Appeal party (Christen-Democratisch Appèl) CRS: Creditor Reporting System from the OECD

CSO: Civil Society Organizations

CSV: Comma Separated Value file extension. CU: Dutch Christian Union party (ChristenUnie)

FAO: Food and Agriculture Organization of the United Nations

FDOV: Facility for Sustainable Entrepreneurship and Food Security (Faciliteit Duurzaam Ondernemen en Voedselzekerheid)

FMO: Netherlands Development Finance Company (Nederlandse Financierings-Maatschappij voor Ontwikkelingslanden)

GDP: Gross Domestic Product

IATI: International Aid Transparency Initiative

IOB: Policy and Operations Evaluation Department from the Netherlands (Inspectie Ontwikkelingssamenwerking en Beleidsevaluatie)

IPGs: International Public Goods MASP: Multi-Annual Strategic Plans MCDM: Multiple Criteria Decision Making MDGs: Millennium Development Goals

MFA: Ministry of Foreign Affairs of the Netherlands NGOs: Non-Governmental Organizations

ODA: Official Development Assistance

OECD: Organization for Economic Co-operation and Development PvdA: Dutch Labor Party (Partij van de Arbeid)

PVV: Dutch Party for Freedom (Partij Voor de Vrijheid)

RVO: Netherlands Enterprise Agency (Rijksdienst voor Ondernemend Nederland) SDGs: Sustainable Development Goals

SMEs: Small and Medium Enterprises

SP: Dutch Socialist Party (Socialistische Partij) SRHR: Sexual and Reproductive Health and Rights UN: United Nations

VVD: Dutch People's Party for Freedom and Democracy (Volkspartij voor Vrijheid en Democratie)

WRR: Scientific Council for Government Policy from the Netherlands (Wetenschappelijke Raad voor het Regeringsbeleid)

WTO: World Trade Organization

Spelling note: This document is written using American English spelling, but the British spelling is kept for quotes from policy documents and academic papers.

Cover photograph: Binnenhof in the Hague. The political center of the Netherlands. Personal archive, 2016.

(8)

7 Foreword

I would like to thank many people who were involved in one way or another in making this thesis possible. First and foremost, I would like to thank my supervisor Dr. Courtney L. Vegelin who did not only tirelessly guided me through ups and downs of the research, but also gave me the moral support I needed to push through. Thanks for all the patience and for not losing the faith on us.

Additionally, I would like to show my gratitude to the respondents, who took time from their busy schedules to help me unveil the world of Dutch policy and politics. Without their insights and suggestions this thesis would be empty. I hope that I was able to present their thoughts faithfully.

I also want to acknowledge my fellow classmates, teachers and friends, in which I’ve found an academic and emotional support network while being abroad. I am glad that the opportunity of being part of the University of Amsterdam allowed me to meet and work with such extraordinary people.

I would like to thank very specially my parents, my sister and my cousin for the loving care, support and encouragement with which they have always been by my side. I feel incredibly thankful for being part of this family.

Finally, and most importantly, I want to thank my wife, Claudia. Without her determination, unconditional love, thoughtful advise and everlasting support, I would not have been able to make it this far. Thanks for embarking with me in this crazy adventure across the ocean and beyond.

(9)

8

Chapter 1. Introduction

During the first half of 2016, news about a human rights activist murdered in Honduras drew attention to the world concerning a hydroelectric plant that was being built as a development project near indigenous territories. The project was partially financed by the Netherlands Development Finance Company (FMO), a commercial entity in charge of channeling public and private funds for development projects around the world (FMO 2016). Concerns about the principles and safeguards that were supposed to be in place led to scrutiny of the decision-making processes behind the project, even at parliamentary levels. This particular case sparked my interest in gaining further knowledge about the interactions between discourse, policy, finance and decision-making, and their impacts for international development.

By researching development in developing countries only, as is traditionally done, much is lost in learning about how developed countries’ policies might shape what is actually happening in developing countries. The international impact of a political and electoral decision in the Netherlands, particularly in developing countries, as well as the processes involved in the making of said decision, matter and deserve to be studied. An average citizen in the Netherlands, or any developed country, will probably not reflect on how her or his decisions in politics will change the life of a small-holder farmer in Ethiopia, but the aggregated outcome of these decisions influence millions of people around the globe. It is in the light of these challenges that I undertook the current research by observing the links between changing discourses, policy and funding in a Western donor country, to ask greater questions about the impact of development cooperation in the field. More specifically, the Dutch government, which has been historically considered a frontrunner in Official Development Assistance (ODA), has apparently seen changes and cutbacks in its international development policy since 2010 (Spitz, Muskens, and Ewijk 2013). I intend to research the discourse changes behind those shifts, and compare them to the financial information available on project funds for development projects.

With this background in mind, I address the research question of how are discourse, policy

and financing from the Dutch government shaping decision-making priorities for international development programs? In this thesis, the reader will find, first, the relevance

and gaps in literature, the research questions and the research location. This will be followed by a chapter presenting a theoretical framework that includes definitions pertaining Development Discourse, Finance for Development, Decision-Making for Development and Policy Cycles. Afterward, a chapter with the methodological approach will be presented, followed by a general overview of the context in which the research is embedded. Then, I will present a discussion of the findings divided in three chapters, structured to connect them to the research questions. Subsequently, a chapter with the final analysis and the closing discussion will be presented, ending with a chapter looking into the possible future implications of this research.

(10)

9

1.1 Relevance of the research and gaps in literature

Given the influence of the Netherlands in the global development arena, understanding the forces behind the decision-making process is relevant not only for policy-makers and those who study Dutch development but also but for other donor countries and for a large population of vulnerable people in developing countries. Providing further insight into the relationships between discourse, policy and funding in decision-making for development from the Netherlands may prove to be useful for reflecting on the impacts of current development approaches, and for providing opportunities for improvement.

Additionally, the post-2015 agenda for development, condensed in the Sustainable Development Goals (SDGs) (2016) from the United Nations (UN), delineates the development priorities to be followed, posing additional challenges to the international development architecture. This research is deemed relevant in this international context to understand how governments from developed countries manage to introduce changes as a response to global commitments.

Starting from the policy agenda, the discursive guidelines of development are interpreted by decision-makers that have to determine how principles should be implemented in specific programs and projects around the world. The form in which discourse is understood from the viewpoint of agenda-setters, policy-makers, decision-makers and implementers will have different outcomes in the field, and will have different impacts that may or may not relate to their original intentions.

In this vein, and taking a closer look at the Netherlands in particular, it is found that little academic work has been done around the “aid and trade” agenda from the Dutch government. Given that Dutch policy is seen as a frontrunner in development and aid (OECD 2013), whose approaches are followed by other countries, it appears relevant to study the causes and effects of changes in policy. This research touches upon these changes and would like to contribute to the debate on it. Moreover, this research also expands on the report on “The Dutch and development cooperation” (Spitz, Muskens, and Ewijk 2013). The report commissioned by NCDO1 carried out a series of interviews and historical analysis on the evolution of the Dutch development policy, before the current “aid and trade” agenda was published in 2013.

For reasons that I will further discuss in Chapter 3, examples from the Dutch food security policy will be used throughout the thesis, not as a theoretical concept, to provide illustrations for the research. At the time of writing, the Policy and Operations Evaluation Department (IOB) was preparing a detailed impact evaluation of four projects implemented from 2012 to 2015 as part of the Food Security policy in Ethiopia, Rwanda, Bangladesh and Uganda (MFA and IOB 2013, 21). The terms of reference for this study gave me a better idea about how the projects are designed and implemented to fit the policy, while looking for some of the assumptions behind policy implementation (see Appendix D). The research I undertook had a significantly smaller scope –and fewer resources in staff, finances and time to carry it out– but also looked at how the principles and assumptions are connected at different levels of the policy processes.

1

(11)

10

I also found a gap in literature when comparing discourse and funding for development in the Netherlands. Looking at the transparency mechanisms available, I found in the International Aid Transparency Initiative (IATI) a great window to study the actual flow of Dutch financial resources for development. IATI is an international initiative that promotes transparency and open aid through reporting of development projects and financial resources. IATI will be further explained in sections 2.2.1 and 3.3.

The use of a mixed method approach allowed the analysis of what is expressed in policy in correspondence to the development initiatives that are being implemented and funded by governments. Shortly after this research is finished, a study will be published with a similar approach to the one that I undertook, by looking at how changes in the gender discourse in Canada are reflected in project funding, and using IATI data to present those changes (Swiss and Barry Forthcoming). Furthermore, the use of data reported to the IATI has not been widely used in giving an evidence-based approach to how the financial resources flow in terms of development projects. A paper from Szabo et al. (2016) calls for attention to the need to generate evidence-based accountability to achieve the SDGs, and reviews different options among which IATI appears as a robust tool to track ODA funds. In the same line, an article from Keijzer (2016) uses the example of IATI reported capacity-building projects, to draw conclusions on the impacts of short- and long-term accountability in development.

1.2 Research Questions

Main research question:

 How are discourse, policy and financing from the government of the Netherlands shaping decision-making priorities for Dutch international development?

Subquestions:

 What kind of discursive principles are perceived as guiding Dutch development interventions?

 How is Dutch development policy influenced in its formulation?

 How does decision-making happen in selecting international development priorities from the Netherlands?

(12)

11

1.3 Research Location

The research took place in the Netherlands, particularly in the cities of Amsterdam, The Hague and Wageningen (see Figure 1). Most of the Dutch governmental institutions, financial organizations, non-governmental organization (international and national NGOs) and Dutch knowledge institutes are located in these decision-making centers and constitute hubs where development stakeholders interact (ProDemos 2013). A detailed list of respondents is presented in Appendix A.

Figure 1: Map of the Netherlands

(13)

12

Chapter 2. Theoretical Framework

In order to address the research questions, it is necessary to identify the main concepts used to analyze the phenomena and position them in the field of development. To begin, the concept of development discourse is used to identify the priorities and principles that are found in policy and utilized by decision-makers. Subsequently, the concept of finance for

development allows the exploration of relationships between resource allocation and the

priorities funded in development. Then, the concept of decision-making for development is used to identify how this process is carried out by people combining discourse, policy and finance to determine the development priorities to be implemented by other stakeholders. Finally, the concepts related to policy cycles are reviewed to orient the research when observing the political and policy changes.

2.1 Development Discourse

As mentioned before, the concept of development discourse is intended to be used in this research as a lens through which the ideas, values and beliefs expressed by key informants are analyzed. These ideas were in turn classified to determine the principles utilized discursively in development. In order to determine what will be understood as development discourse, the concept will be disaggregated in a review of the approaches that have been considered by scholars, first on the concept of discourse, and then on the concept of

development.

For the concept of discourse, we can find extensive literature with definitions that depend on the field of study from which they are drawn. I will build from three definitions of discourse from academics studying phenomena in the development field.

First, Arturo Escobar defines discourse as “the process through which social reality comes into being -if it is the articulation of knowledge and power, of the visible and the expressible-”(1995, 39). To further articulate this view, the definition given by Phillips and Hardy “as an interrelated set of texts, and the practices of their production, dissemination, and reception, that brings an object into being” (2002, 3), narrows the idea to “texts”, which can be oral or written. This view, in line with Escobar’s definition, contemplates discourse as a process to give meaning to reality. Finally, the definition provided by Maarten Hajer, explains discourse as “an ensemble of ideas, concepts, and categories through which meaning is given to social and physical phenomena, and which is produced and reproduced through an identifiable set of practices” (2006, 67).

Development as discourse

As the scope of this research is limited to development in discourse, policy, finance and decision-making, development will be treated as a discourse on its own (Escobar 1995, 10). Development can be understood as a metaphor to show the essential growth of living things, which is applied to the economics of human communities. This necessary evolution is seen as benign and worth pursuing (Esteva 1992).

Andrea Cornwall further elaborates in describing development and its terminology as a series of “buzzwords” that are used to express a vague idea of improvement through action. For the author, these buzzwords are used as a way to gain admission to groups and

(14)

13

organizations, as well as accessing funds. In this sense, she adds that the discourse around development conveys ambiguous ideas that are useful for bringing actors together, as each implementing actor will give particular meaning to these ideas (Cornwall 2010).

Finally, in line with the idea of “development” as a buzzword, Gilbert Rist points out to a lack of real meaning on its own and to how it has become a way to justify all sorts of actions, as long as they are portrayed as the improvement of a current condition. One of the characteristics that the author gives to this term is that it came to be a logical and desired consequence, and therefore an uncontested concept in the policy world (Rist 2010).

Discourse and policy

Hajer’s approach to discourse and its study proves useful when applying the methodology of critical discourse analysis to policy and decision making (Hajer 2006, 68), moreover as it arose from the need of studying the discourse in the particular case of Netherlands. Building on the position of Chouliaraki and Fairclough (1999, 16) on how critical discourse analysis can be understood as both a method and theory, I will be using the same approach for this research (see section 3.3).

Hajer brings to light that, in terms of discourse, the supposition that people communicate ideas in an unequivocal manner is incorrect because each part of the communication process interprets concepts differently. However, the author mentions that this notion allows people to create coalitions of common understanding, what he calls

discourse-coalitions (Hajer 2006, 69–70). Discourse-discourse-coalitions are ideas that are shared by a group of

people in order to give meaning to their particular situation and time. These coalitions are formed by certain metaphors which embody values, beliefs and principles used by the group to represent reality. The metaphors are then enacted and formalized through practices, which entail norms and models to social interaction (Hajer 2006, 70). An example of this can be found in the concept of sustainable development, which can have different meanings and materializations when coming from the private sector side or from the environmental end (Gupta and Thompson 2010).

Additionally, Hajer connects discourse with policy by mentioning two steps through which discourse-coalitions evolve and become dominant. The first stage is when a particular set of principles start being used by the group to portray the world, which is defined as discourse

structuration. Then, when the discourse is materialized through organizational practices, the

author conceptualizes this step as discourse institutionalization (Hajer 2006, 70). The distinction between the aforementioned stages relates to the aim of this research, as

structuration will define the principles used in policy, financing and decision making (see

Chapter 5), while institutionalization will help to analyze the practices and implementation (see Chapter 6).

On this subject, Chouliaraki and Fairclough mention that discourse also concerns "networks of practices [that] are held in place by social relations of power, and shifting articulations of practices within and across networks are linked to the shifting dynamics of power and struggles over power" (Chouliaraki and Fairclough 1999, 24). The changes in discourse and practices are linked to changes in power dynamics. In development, they will shape the ways in which the practice of development is carried out. However, in this research I will not be focusing particularly on the power dynamics of development but this provides a broader framework in which to reflect upon the final conclusions.

(15)

14

The academics that have been mentioned in this review build upon the ideas of Michel Foucault on discourse. In brief, for Foucault, discourse “can be defined as the group of statements that belong to a single system of formation” (Foucault 1872, 107). However, Sally Hewitt gives an overview on how the Foucauldian discourse analysis approach can be applied in researching public policy. She highlights the relationship of discourse with the concepts of power and legitimacy of the government. Although this approach aims at understanding policy-making processes, discourse is partially put into practice through public policy. The dynamics of power are then reflected in the changing discourses, forging the organizations in charge of creating and applying policy (Hewitt 2009).

In policy research, Hajer and Versteeg emphasize the importance of discourse in development, as it creates a particular lens through which problems are selected to create policies, but also in selecting the answers for the problems identified (Hajer and Versteeg 2005, 179). The manner in which discourse permeates problem delimitation and its solutions will also be explained within the definition of decision-making for development. As will be seen in Chapter 5, the particular concepts that arise as ideas guiding Dutch development policy will be presented. These concepts, found recurrently throughout the research appear as important discourse metaphors. In this sense, the principles expressed in the policy, as well as by decision-makers and stakeholders are used in the analysis phase.

2.2 Finance for Development

Financing for development is a term that has been used to group collective efforts to mobilize funds for development actions around the world (UN 2003; UN 2009; UN 2015). This has been brought to wider attention by the UN by organizing three global conferences on “Financing for Development”, to achieve consensus and commitments on the funds required for development in 2002, 2008 and 2015, while having a strong position towards achieving transparency.

Although the concept of ensuring the financial means for development has gained attention recently, it is not a new idea. During the sixties, Jan Tinbergen, a Dutch economist in charge of the United Nations Committee on Development Planning, proposed that the UN establish a target of 0.75% of the Gross Domestic Product (GDP) to be devoted to ODA (UN 1968, para. 7; OECD 2002). In 1970, the then so-called “economically advanced countries” committed at least 0.7% of their gross national income to official development assistance during the UN general assembly (UN 1970, para. 43). This objective, however, has not been met by the majority of higher income countries.

To ensure that financing was mainstreamed in the international agenda, in 2000 the UN introduced the Millennium Development Goals (MDGs) (UN 2006) which included the need for a greater commitment from developed countries to fund –certain– development initiatives (Adam et al. 2015). After achieving certain contested results (Fukuda-Parr, Greenstein, and Stewart 2013), the MDGs were followed by the Sustainable Development Goals (SDGs) in 2015 (UN 2016). These goals include a broad set of principles regarding development cooperation and their relation to how to achieve them (Stevens and Kanie 2016).

(16)

15

The idea of financing for development is intertwined with the concept of the international

aid architecture. This term can be defined as “the subset of international financial

architecture rules and institutions affecting aid flows, including the way in which (bilateral and multilateral) aid is being allocated, the mix of official aid and debt flows, the use of official debt reduction and the accompanying policy requirements” (Cassimon, Claessens, and Campenhout 2007, 4). Increasingly, reporting has played a role in the aid architecture, by providing access to information for donors and recipients of aid, while being integrated into the planning of development agencies worldwide (Linders 2013). Among the reporting initiatives, the Creditor Reporting System (CRS) and the International Aid Transparency Initiative (IATI) have developed reporting instruments for government agencies and development organizations (OECD 2016a; IATI 2016a). The relevance of these reporting initiatives will be further analyzed in section 2.2.1 and in Chapter 7.

It is through the aid architecture that development funds allegedly flow from developed countries to developing countries. This has historically included the ODA provided bilaterally or through multilateral organizations but has evolved over time to include new forms of financing. According to the World Bank, the aid architecture has grown increasingly complex due to the diversity of actors involved, and due to earmarking the development flows (World Bank 2008).

The aforementioned concept of earmarked funding in development is understood as the process in which certain resources are allocated to specific themes, regions, countries or organizations (Michaelowa, Reinsberg, and Schneider 2014). Being able to identify the trends of earmarked funds in the aid architecture, allows studying how financial resources are being distributed, as well as looking at what is being prioritized financially. To achieve this, having access to transparent information is required in order to observe the actual allocation of resources in general. This topic will be analyzed in the following subsection, as it is a key issue for this research.

2.2.1 Transparency and accountability

Transparency and accountability are increasingly becoming guiding principles for policy and decision-making (see section 5.1). Having reliable information on budgets and expenditures2 for aid and development allows citizens to hold their governments accountable. It also allows recipient countries and organizations to plan ahead and to have greater control over budgets and impacts (Publish What You Fund 2016).

Jonathan Fox explains that the concepts of transparency and accountability do not necessarily follow each other. In his view, there are two types of transparency (opaque/clear) and two types of accountability (soft/hard). For transparency, the difference between the two categories lies in the reliability of the information that an institution shares about them and how easy it is for the public to access it. According to the author, achieving clear transparency does not necessarily lead to accountability, as this term relates

2

For the Dutch development case, the main difference between budget and expenditure is that budgets are the financial resources assigned beforehand to certain priorities, which are constructed by the Ministry and approved by parliament, while expenditures are the actual funds disbursed for the programs and activities developed (School of Data 2013).

(17)

16

to the power to demand answers from the institutions (soft accountability). Finally, hard

accountability is achieved when transparent information exists and institutions respond to

the demands, but also when institutions can be punished for their actions (Fox 2007). When talking specifically about opaque and clear transparency in budgets, Fox remarks that "the first approach would focus more on revealing the details of public-sector contracts, for instance, and assuring that funds were spent as intended. The second strategy, in contrast, would focus more on the impacts of public spending: how agencies actually used the funds, and to what effect " (Fox 2007, 667). This distinction is important when trying to critically assess the actual levels of transparency and accountability that governments give their citizens.

A global example of the current trends in transparency can be found in the International Aid Transparency Initiative (IATI). This tool was presented in 2008 at the third High-Level Forum on Aid Effectiveness in Accra for countries, NGOs, philanthropic and multilateral organizations to publish and access financial information on aid and development (IATI 2016b), with a special focus on accountability for the MDGs. The initiative was later promoted in 2011 at the fourth High-Level Forum on Aid Effectiveness in Busan as a mean to achieve “transparent and responsible co-operation” (OECD 2011). By 2015, members of this initiative comprised approximately eighty percent of the financial flows for development (Pamment 2016, 142). This particular instrument has been recognized and encouraged by the UN through its Addis Ababa Action Agenda (UN 2015). Importantly, the Netherlands adopted IATI as the official reporting standard for all development operations that are financed by the Ministry of Foreign Affairs, as well as for its partners (MFA 2013a; MFA 2016a; OpenAid 2016). This allows having access to the transactions given to organizations implementing development projects.

The sums of financial resources, however, are not able to explain by themselves the interests behind them (i.e. altruism, geopolitical, commercial, strategic, etc.), which is why the concept of financing for development will be analyzed in combination with the abovementioned concept of development discourse and with decision-making for

development, explained subsequently.

2.3 Decision-Making for Development

As I intend to study the decision-making priorities regarding development, the sense in which the concept of decision-making is comprehended needs some elaboration. I approach this section by looking at decision-making theoretically as an explanation for the particular choices made, and not as a method or a prescription on how to carry out a decision-making process.

In a still widely used definition, decision-making can be understood as “the process of selecting a possible course of action from all the available alternatives” (Hwang, Masud, and Paidy 1979, 1). In this sense, it can be seen as a way in which an “expert”, using available information and experience to overcome uncertainties, will select a particular option which is considered as the best to achieve the optimal solution scenario (Hallegatte et al. 2012, 13).

(18)

17

Hwang et al. describe the “Multiple Criteria Decision Making” (MCDM) method, used to make decisions when facing multiple inputs and outcomes (as opposed to having a binary set) (Hwang, Masud, and Paidy 1979). When explaining MCDM methods, the authors identify three major elements that characterize this model: “(1) a set of criteria of judgement; (2) a set of decision variables; and (3) a process of comparing the alternatives” (Hwang, Masud, and Paidy 1979, 6).

Ballestero and Romero express MCDM in a similar line. To the authors, decision-making is first constrained by limitations in the resources available, which will determine the decision variables to be used in the process. Then, these variables will create a limited number of options that realistically meet the restrictions, which is called by the authors as the “feasible set”. Following this first phase, the set of possible solutions is then filtered by the decision-maker according to the objectives and goals that wish to be obtained, which is defined as applying certain criteria to the set. Lastly, what is seen as the most suitable solution is chosen by the decision-maker. For the authors, it is important to evidence the difference between the first phase, in which the constraints and variables are seen as externally determined, and the second phase, in which the choices of the decision-maker come into place through the introduction of criteria and preferences (Ballestero and Romero 1998, 1). In order to operationalize the concept, the components of the decision-making criteria can be disaggregated. First, the objective of the criteria is determined by deciding what is seen as an improvement of a situation; this assumes that, according to certain values and beliefs of the decision-maker, there is a problem and it can be solved. Secondly, the goal is defined by setting a desired level of improvement that wants to be achieved, aligned with the objective. Lastly, the trade-offs determine what compromises are made in order to make a decision taking into account the aforementioned elements, as well as what is deemed by the decision-maker as the best option (Ballestero and Romero 1998).

Under this process, "decision-makers choose the ‘best’ among all possibilities available. However, the concept of 'best' is ambiguous and in many situations involve more than one criterion of choice" (Ballestero and Romero 1998, 2). This situation opens the door to multiple interpretations of what is understood as a priority or a suitable solution for a problem. It is, therefore, better analyzed through a qualitative approach, as it is intended for this research.

(19)

18

2.4 Policy Cycles

As this research aims partly at studying development policy, the theoretical model of a

policy cycle can be used to comprehend its processes and interactions. The policy cycle

encloses the different stages that policy undergoes. Concretely, policy cycle is understood as the processes of “agenda-setting, policy formulation, decision-making, implementation, and evaluation” (Jann and Wegrich 2007, 43) (see Figure 2). Nevertheless, these categories are not strictly delimited and may overlap with one another.

Figure 2: Policy Cycle

Source: Author based on Jann and Wegrich 2007.

To further dissect the concept, the agenda setting process includes the identification of a social problem in which the state can provide solutions, which are then incorporated into the government’s agenda (Jann and Wegrich 2007, 45). The agenda as such can be conceived as "a list of subjects or problems to which governmental officials and those close to them are paying serious attention" (Kingdon 1993, 40).

The phase of policy formulation happens when the issues existing in the agenda are taken up by government officials to create solutions (Jann and Wegrich 2007, 48). The

decision-making phase is intertwined with the formulation phase, as it is the moment in which, one

solution is negotiated and chosen from a set of solutions pertaining policy (Jann and Wegrich 2007, 49). As discussed in section 2.3, the last mentioned process includes the design of certain objectives, goals and trade-offs. On this subject, Michiel S. de Vries, when talking about policy-making in the Netherlands, states that “policymakers, like all other people, have to make choices under conditions of economic scarcity. The means available for adequate policies are always less than those required. […] It can be assumed that it is therefore necessary to prioritize. The obvious consequence of prioritization is neglect of those things not prioritized” (Vries 2005, 578).

(20)

19

Implementation, the fourth stage in the policy cycle, describes the process of adapting the

policy to practice. This phase is not bounded to the participations of just public officials, as it can involve other stakeholders3 and agencies selected, who carry out the policy (Newig and Koontz 2014). In the process of translating what was formulated into concrete actions, it is possible that the implementing actors interpret differently what was intended by the officials creating the policy (Jann and Wegrich 2007, 51), which is one of the main issues addressed in the current research.

The last part of the policy cycle is the evaluation process, in which the outcomes and results of the implementation are reviewed. This stage aims at determining if the policy had the desired impact on the problems that they were designed to address. Although this phase is described as the last stage in the cycle, evaluation is carried out in parallel to implementation and provides feedback to the agenda-setting process (Jann and Wegrich 2007; Newig and Koontz 2014)

In assessing the larger environment in which the policy cycle is embedded, the concept of the politics cycle appears as a major influence in how the different stages of policy are connected to politics. As articulated by Meadow and Steurer, the politics cycle in western democratic states is “determined by the periodic conduct of general elections which populate representative institutions, underpin the formation of governments, and help to define political priorities. […] Within an election period, the politics cycle involves the regular sessions of the legislature and meetings of the executive, the continuous formation of government priorities, and the annual budget process through which state expenditure is authorized” (Meadowcroft and Steurer 2013, 14).

The politics cycle constantly influences the policy cycle in all of its stages. This accounts not only for the regular interaction between politicians and policy-makers but also for how certain policies are viewed by the electorate (Meadowcroft and Steurer 2013, 15). Beyond being only connected to the agenda-setting phase in the policy cycle, politics will steer all the phases of the policy process. In this sense, the policy cycle is receiving constant feedback from the politics cycle, while also acting as a mean to achieve certain goals in the politics cycle (Meadowcroft and Steurer 2013, 16).

Closing Remarks

Having the aforementioned theoretical notions as a conceptual point of departure allowed me to explore the research questions of this study. First, “development discourse” gives a framework through which government policy can be understood theoretically and methodologically as a discourse. Second, “finance for development” allows looking at the financial flows to understand what gets supported or excluded from policy implementation. Third, “decision-making for development” gives a framework to understand the choices made by decision-makers. Finally, the concept of “policy cycle” structures the unit of analysis in the different phases where the policy is found.

3_{For this research, stakeholders are defined as the interested parties involved in the policy cycle such as}

agenda-setters, policy-makers, decision-makers, implementers and evaluators, represented by officers from the Ministry of Foreign Affairs of the Netherlands, Dutch embassies, organizations and people close to them.

(21)

20

The epistemological processes I chose for this research were selected to make a study of development policy, not necessarily on their outcomes, but of the assumptions on which they are based. The first goal of this chapter was to scrutinize development and development cooperation, and making the ontological claim that they are a discourse. By doing this, I try to analyze and highlight the premises involved in Dutch development policy formulation (Pauly 2016). The second goal of this chapter was to evidence the overlapping between theory and methodology when analyzing discourse and policy.

(22)

21

Chapter 3. Methodological approach

3.1 Conceptual Scheme

The main concepts explained in Chapter 2 are interrelated and allow exploring the questions posed in this research. As portrayed in Figure 3 below, the Dutch government uses development discourse and is also in charge of creating policy and providing the financial means for its implementation. Policy itself influences and is influenced by the principles in discourse and by the provision of funds by the government. Then, discourse is used in decision-making, which is guided by what policy dictates and by the limitations from available funds. Finally, development priorities are selected from the decision-making process, and are then transformed into practices implemented by stakeholders, who can sometimes persuade decision-makers and have interactions with the government based on their mutual interests4.

Figure 3: Conceptual Scheme

4

The way in which these concepts and interactions were operationalized specifically for the research can be found in Appendix C: Operationalization of Major Concepts.

(23)

22

3.2 Methodology

Methodologically, I undertook this research from a critical realism approach. According to critical realists, phenomena can be understood ontologically as a natural reality that can exist beyond human interaction (Bryman 2012). However, the form in which the researcher gains knowledge of these phenomena is through a critical constructivist epistemology, in the sense that meaning given to reality depends on social constructions, such as language and context (Maxwell 2012). In this sense, critical realism allows for an understanding of events and processes that exist in the real world, but that are shaped by social practices and institutions that allow for representations of such realities (Fairclough 2005, 922), and in a particular time (Fleetwood 2005).

As such, I used critical discourse analysis for this research, which builds on the tradition of critical realism (Fairclough 2005; Bryman 2012) as a way to identify the ideas used as principles and practices in the different stances regarding selection and funding of development priorities. Critical discourse analysis aims at studying the relationship between discursive and non-discursive components of reality, and their meanings (Fairclough 2005, 924). In this regard, critical discourse analysis proves useful in the analysis of how discursive components relate to changes in the decision-making processes and in discourse itself (Bryman 2012, 537).

Changes to original research proposal

The original research proposal was adjusted during the fieldwork phase. The most considerable modification occurred to the unit of analysis, which was first defined as the processes of decision-making at organization levels, and was subsequently changed to the processes of decision-making in the policy cycle. This choice was made as a consequence of the unexpected limited access to implementers and stakeholders, in comparison to the responses from individuals involved in policy.

As foreseen in the proposal, the original topic was narrowed down according to the critical discourse analysis. As the main interviews began to show a particular emphasis on Dutch food security policy, this research topic was chosen in order to draw examples from a policy in particular. Food security is one of the four policy spearheads of the Dutch government for development, making it an appropriate representation of the development policy processes in the Netherlands. It is important to note, however, that the choice of having a thematic line does not imply that the research is a case study of food security, as it still focused in the discourse analysis of development policy.

The operationalization of concepts, dimensions, variables and indicators that were set for the research proposal, were also reviewed during the field phase. For the concept of development discourse, the variables were changed to focus on the policy cycle. This revision was necessary after a higher rate of responses from people involved in the policy level of decision-making was noticed. The concept of financing for development also had some of the variables changed. The variable of “external funding” was removed from the budget dimension due to the unreliable and insufficient reporting of data concerning private funding of development projects.

Finally, in the expenditure dimension, the “financed organizations” variable was removed after considering that one organization could be in charge of running different types of

(24)

23

projects. Instead, the focus was shifted towards “financed sectors”. Sectors are understood as the categories under which projects are aggregated and aim to contribute to a bigger goal in the recipient’s structure, particularly for the IATI data utilized in this research (OECD 2016b).

3.3 Research Methods

The critical realist approach utilized in this study allows the use of mixed methods to assess the research questions. Qualitative methods were used to address the interviews and texts regarding discourse, policy and decision-making, while quantitative analysis was applied for the analysis of financial information on budgets and expenditures.

Furthermore, as mentioned in section 2.1, critical discourse analysis can be understood as both a method and theory (Chouliaraki and Fairclough 1999, 16). I used a particular approach to critical discourse analysis that was specifically designed for the study of development-related policy discourses in the Netherlands. This method stems from the following steps described by Maarten Hajer: 1) Undertaking a general document overview; 2) interviewing key guiding informants; 3) doing an in-depth analysis of documents; 4) focus-interviewing key respondents from the politics and policy cycle; 5) identifying context and importance of the responses; 6) analyzing the positionality of the actors involved in the exchange of arguments; 7) identifying the key moments that created shifts in positions; 8) identifying the practices arising from the discourse; 9) interpreting the discourses in place, their practices and their interactions (Hajer 2006, 73).

For the qualitative part of the research, the data was collected from semi-structured interviews with key actors and from documents related to policy and project evaluation. The interviews were conducted during the months of July through October 2016, with fifteen decision-makers and key stakeholders involved in different stages of the Dutch development policy cycle. The interviews focused on the understanding of values, beliefs, perceptions and ideas involved in prioritizing and selecting development priorities, as well as the perceived influences from external interactions. A list of the interviewees and documents can be found in Appendices A and B correspondingly.

For the quantitative analysis of financial data, descriptive statistics were used as a method to address the shifts and trends in financing for development. The financial information on budgets and expenditures was obtained from publicly accessible sources such as the Ministry of Foreign Affairs (MFA 2016b; MFA 2016c), OpenAid Netherlands (OpenAid 2016) and the International Aid Transparency Initiative (IATI 2016c), among others. The data collected was then analyzed by year, sector, recipient country, implementing organization, project, commitment and disbursement to evidence the changes and tendencies in funding development projects. This quantitative analysis was used to compare with the qualitative results from the discourse analysis.

After exploring different options for the analysis of the IATI information files, the choice to use both Microsoft Excel and Tableau Public was made as these software programs facilitated the process of aggregating and visualizing more than 5000 projects funded by the Ministry of Foreign Affairs of the Netherlands. The main IATI Comma Separated Value (CSV)

(25)

24

file from the Ministry of Foreign Affairs was downloaded and converted to Microsoft Excel. The resulting file was checked for errors, blank or double values and was then formatted according to the type of data available. Afterward, the file was uploaded to Tableau Public software, where the different interactions between dimensions and measures could be presented and analyzed. The decision to manage the IATI information in this manner and not through official Dutch IATI visualization websites (e.g. OpenAid), was made after considering that these websites did not allow large scale comparisons between different years and types of projects, nor filtering using all the categories available by the IATI standard.

At a later stage, the CSV file from the Ministry was linked and merged with the CSV file containing the yearly budget information from the Ministry of Foreign Affairs, which uses the same activity identification number for projects reported in the budget and to IATI. This process allowed contrasting the budgetary information with the actual disbursements, along with identifying the activities that received finance but were not initially presented in the budget.

3.4 Unit of Analysis and Sampling

The main unit of analysis for this research is the development policy cycle processes (Baxter and Jack 2008). Here, the processes and relationships between phases of the policy cycle were identified and analyzed for the development policies. In order to understand these phenomena, the unit of analysis was researched at the macro- level of analysis (Buechler 1993). The analysis was centered on the processes that are present in Dutch development policy and discourse, emphasizing the decision-making process and priorities involved in the selection of certain policies and priorities.

The respondents were selected using purposive sampling to correspond with the different stages of the policy cycle. Respondents from each phase of the cycle were chosen and interviewed to obtain a broader idea of how development policy is decided upon (Bryman 2012; Merriam 2002).

The analyzed documents were selected according to their relevance in development policy of the Netherlands. The main development policy document, A World to Gain: A New

Agenda for Aid, Trade and Investment (MFA 2013a), was analyzed along with many of its

supporting documents. The Coalition Agreements for the 2010 and 2012 governments,

‘Freedom and Responsibility’ (Rutte and Verhagen 2010) and 'Building Bridges' (Rutte and

Samsom 2012) were selected as part of the main government policy documents for each cabinet. Additionally, the documents Our common concern: Investing in development in a

changing world (MFA 2007); Letter to the House of Representatives Outlining Development Cooperation Policy (MFA 2010) and Letter to the House of Representatives Presenting the Spearheads of Development Cooperation Policy (MFA 2011a) were chosen to be studied in

the design of the development cooperation policies. As the report from the WRR Less

Pretension, More Ambition: Development Policy in Times of Globalization (Lieshout, Went,

(26)

25

Specifically for the purpose of drawing examples from food security policy, the following policy letters were used in the research: Kamerbrief Uitwerking Voedselzekerheidsbeleid (MFA 2011c); Letter of November 2014 from the Minister for Foreign Trade and

Development Cooperation and the Minister for Agriculture to the President of the House of Representatives on the Netherlands’ Contribution to Global Food Security (MFA 2014n); and Kamerbrief over resultaten Voedselzekerheid (MFA 2016d).

3.5 Data Analysis

The interviews with decision-makers and key stakeholders were recorded and stored electronically, ensuring anonymity during the whole process, including transcription. Texts from policy, evaluations and other relevant written information were saved in digital formats to allow their use with software tools.

AtlasTi was used to code and categorize information, as a strategy to find distinctive concepts in interviews and policy texts (Bryman 2012, 568). The results from this method informed the inductive process to perform the qualitative analysis required for addressing the research questions.

The financial information taken from the yearly budgets, expenditures and thematic funding assignments reported by the Dutch government to IATI was structured and analyzed using Microsoft Excel and Tableau Public software. This strategy allows the use of extensive data from budgets and public expenditures and its analysis through time. As a result from the quantitative data analysis, I generated a website allowing the public to review, interact and download the data, available in the https://public.tableau.com/profile/gabriel.ramirez#!/

website, under the name “IATI Dutch Food Aid 2007 - 2016”.

3.6 Scope and Limitations

The scope of my research is limited to the Netherlands and to stakeholders participating in Dutch public funding for development. This includes working with policy advisors, makers, officers and evaluators involved in development, as well as organizations implementing projects from the Netherlands, among others. In terms of a time frame, the analysis was delimited by the changes in Dutch public financing for development after 2010, year when a change was perceived in the Dutch policy focus, and since the public availability of financial information through IATI is only reliable after 2007.

Although studying project implementation would have shed a light on the larger consequences of discourse, policy and finance, this research focused on the policy cycle level due to time and resource restrictions. The scope was then limited to the decisions happening in the policy cycle and before project implementation.

It is important to note that the current research does not aim at being generalizable for all development policies, as it is bounded to a specific context and time. It can, however, inspire similar research approaches related to development discourse and funding.

(27)

26

3.6.1 Methodological Reflection

In the limitations that I had identified before initiating the interview phase, the most pressing one was the possibility being unable to access government officials willing to provide responses about their decisions on public funds. In practice, however, access to policy makers and public officials was uncomplicated and, despite a certain lack of response, a considerable number of officials responded positively to the invitation and gave in-depth answers to the questions proposed. The limitation of finding available respondents was related to the time of the year selected for the research, as the majority of the interviewees were not available during the summer holidays in July and August, extending in consequence the period of conducting interviews.

Additionally, the Netherlands is recognized as a country in which transparency and accountability for public outcomes are protected by law, and where reliable information is easily accessible. This particular situation indicated the Dutch system to be an appropriate setting for finding answers to the proposed research questions.

A limitation regarding the quality of this research is related to the veracity of the information given by public officials. This may arise from the officials’ fear of the being publicly judged for their actions, and the possibility of being scrutinized for previously made decisions. Therefore, I addressed this issue by communicating and engaging in confidentiality practices to protect the respondents. I also emphasized the academic aim of the research, diverging from being identified as a journalist or auditor judging their work. The result in building trust with the respondents was satisfactory and resulted in open responses about their opinions even when they could contradict the official stances.

Along the same line, the quantitative data gathered for the research may pose certain validity questions, as it was obtained from an external source (the IATI registry). The financial information regarding budgets and expenditures allocated for development projects is reported directly by the Dutch Government to the registry, and it is not possible to audit the veracity of the information as an independent researcher. However, the Dutch Government has adhered voluntarily to the IATI initiative and has pledged to provide transparent and accurate information available to the public.

To further assess the quality of the qualitative component of the research, I will use the concepts of credibility, transferability, dependability, confirmability and authenticity (Bryman 2012, 390–393) as quality criteria.

Credibility: On the one hand, the main technique used to achieve the credibility criteria with the obtained data was triangulation. This technique involved using more than a single method to study the social phenomenon (Bryman 2012). In this research, the answers from the respondents are confronted with the financial data available for Dutch food security development projects. It was also useful to have the Ministry of Foreign Affairs as the entity reporting to the IATI registry, since the category label of “food security” for each project and budget line should match the policy lines in the same theme. Even though the research did not include participant observation of the actual decision-making processes or a long-term ethnographic research, the possibility of doing the research in the same context, culture and politics in which I was immersed for nine months, gave an advantage in communicating with the interviewees.

(28)

27

Transferability: The criterion of transferability relates to the possibility of replicating the current research, which will largely depend on the cultural reference points of the respondents and the data available to contrast it. However, given publicly available financial data and access to policy and decision makers, parts of this research and its methods could be replicated for Dutch and foreign contexts.

Dependability: The transparency of the research is reflected on the dependability criterion. This was assessed by explicitly discussing the reasons behind the changes made to the initial approaches. This includes the change from a broader set of topics to be studied, to the specific theme of food security, as well as the unexpected disposition of government officials to respond to the interviews.

Confirmability: The confirmability criterion allowed me to constantly reflect on the possible interference that my personal background could represent in understanding the answers given by the respondents. As a researcher in the qualitative field, it is important to note that being a foreigner seemed to open more doors and invite deeper reflections by the interviewees.

Authenticity: Finally, regarding the authenticity criteria, I intended to represent as faithfully as possible the different perspectives provided by the respondents. Additionally, I wanted to present the opportunity for the Dutch society to have a deeper understanding on how decision-making processes happen in the Netherlands, while providing empowerment through knowledge about said processes to recipients.

In terms of language barriers, the interviews were conducted in English and most of the official documents were found in the same language, which helped in the communication processes without the need of a translator. When required, the few documents that were solely available in Dutch were translated with the help of bilingual native speakers and web based translators. The Dutch government publicly provides a translated copy of most of its policy, budget and reports, allowing me to assess whether the same concepts were used across the respondents and the documents.

3.7 Ethical Considerations

Ethics encompass the totality of the research. All stages of the study, from the preparation, selection of respondents, data collection, transcription, analysis, interpretation and disclosure were accompanied by ethical procedures and practices (Guillemin and Gillam 2004, 263; Miller et al. 2012, 2).

Continuous reflexivity is crucial for the type of research that was carried out. Assessing the ethical implications for the respondents and the received information was a constant throughout the research. Additionally, during the whole study, but particularly in the qualitative part, I was involved as a researcher myself analyzing phenomena and giving sense to the collected information, meaning that I had to assess myself with the same inquiry as the rest of the respondents (Guillemin and Gillam 2004, 274).

The principles of doing no harm and protection of privacy (Bryman 2012, 135) were present throughout the research. It is important that all respondents remain unharmed during and

(29)

28

after the research, especially due to condemnation or targeting as a consequence of providing information. This, along with confidentiality and privacy protection, entailed having especial care while conducting all efforts to provide anonymity to the respondents’ personal information and securing data during its analysis.

It was also of crucial importance to obtain informed consent from the respondents about participation in the research in conjunction with communication of the academic purpose of the research (Bryman 2012, 138). It is to note that during the research I did not come across information concerning illegal activities, circumstances under which the relevant authorities in the Netherlands would have been properly notified.

Finally, as this study investigates public funds and their destination, special attention has been given to the treatment of financial data, as well as to the implications of disclosing the outcomes of the study. It is of my particular concern the consequences that this research may have in the future towards assessing development cooperation in the Netherlands, as well as the impact that this may have for recipient countries. However, as a researcher I have taken a responsible approach in the management of information and in providing evidence-based conclusions.

Shifting Dutch priorities : changes in discourse, financing and decision-making for international development in the Netherlands

1

2

Abstract

3

Table of Contents

4

5

List of Figures

List of Tables

6

Glossary and Abbreviations

7

Foreword

8

Chapter 1.

Introduction

9

10

11

12

Chapter 2.

Theoretical Framework

13

14

15

16

17

18

19

20

21

Chapter 3.

Methodological approach

22

23

24

25

26

27

28