The flood resilient city : how multi-level governance could complement concrete in Bangkok

(1)

Verena Stefanie Peter

MSc Master Thesis December 2014

International Development Studies

T

HE

F

LOOD

R

ESILIENT

C

ITY

–

H

OW

M

ULTI

-L

EVEL

(2)

The Flood Resilient City – How Multi-Level Governance

could complement Concrete in Bangkok

Master of Science Thesis

International Development Studies

November 28, 2014

Verena Stefanie Peter

10638075

Verena.peter@student.uva.nl

Supervisor: Dr. B. Lambregts

Amsterdam Institute for Social Science Research

University of Amsterdam &

Division of Urban and Environmental Planning

Kasetsart University, Bangkok

Second Reader: Dr. J.M. Bavinck

Graduate School of Social Sciences

University of Amsterdam

(3)

Dedication

I would like to dedicate this Thesis to my twin sister, Mirjam Peter, who spent some time

in the hospital after a terrible traffic accident while the fieldwork for this research was

undertaken. I wish I could have visited and brought you flowers.

(4)

Acknowledgements

I would like to thank my supervisor, Dr. Bart Lambregts, for his constant support and the

contacts provided during the fieldwork period, as well as for his valuable feedback and

advice during the writing process. I would also like to thank my translator and guide

Wiradet Norasetworachai, for helping me with so much more than just the Thai

language, for showing me Bangkok, and providing incredible amounts of assistance with

adjusting to life in Bangkok, as well as with conducting this research. Furthermore I

would like to express my gratitude to Dr. Maarten Bavinck, who kindly agreed to second

read this thesis, and moreover through his academic work provided a part of the

theoretical framework to this research. Finally, I would like to thank all members of the

Fah Mai Community, the Pra Cha Rat Samakki Community and the Soi Rot Fai

Worrapong Charansanitwong 46 Community of Bangkok for sharing their time and

experiences with me, as I could not have done it without all of you, and I will honour

what I learned from each of you.

Thank You!

(5)

Abstract

For economic, cultural and social hubs, such as the mega-city Bangkok, floods are not merely a threat to individual livelihoods and human well-being, but to the overall development process of the city or region as well. How can large costal urban areas, like Bangkok, effectively protect themselves from increasing flood risks in the future? This interview-based fieldwork study researches how the combination of structural and rather intuitive non-structural measures can contribute to the creation of flood resilience in Bangkok.

In practice, the well organized combination of response strategies, resources and experiences across a city's institutions and inhabitants can be a smart, comparatively economic and quick response, altering urban flood resilience, in the face of increasing storm and flood threats triggered by global warming. In theory, this study will link together approaches of engineering resilience and socio-ecological resilience, by focusing on the governance-setting between engineered structural flood resilience strategies and rather intuitive individual and community responses to an environmental shock. The interdisciplinary nature of this combined approach adds substantially to the academic debate, which tends to have a one-sided focus on either structural or social response strategies, depending on the field of study.

The overall aim of the study is to investigate how community flood resilience strategies are integrated with larger structural flood resilience measures in urban Bangkok, and what this means for the success or failure of flood governance. For this holistic approach to flood management and governance, ordinary communities of Bangkok with different socio-economic backgrounds were included into the study, as well as central flood-related district-, metropolitan-, and royal institutions, and several independent experts.

This research finds that in the case of Bangkok, substantial gaps remain with regard to the integration of multi-scalar flood resilience strategies. Even though different kinds of flood response strategies are to be found at all levels, they are only poorly interlinked and supportive of each other. The result is a divided network of flood response, which costs time and resources. This situation is mainly caused by a lack of communication, cooperation and information flow, and a non- transparent chaotic division of tasks and responsibilities in case of crisis, both between citizens and institutions, as well as between institutions.

(6)

List of Acronyms

Acronym Term

BMA Bangkok Municipalitan Administration LDG Local District Government

USAID United States Agency for International Development LIC Low Income Communities

MIC Middle Income Communities

List of Figures, Tables and Photos

Figure 1

The Governability Approach

Figure 2

Conceptual Scheme

Figure 3

Location of Bangkok

Figure 4

Location of Bang Phlat Distirct

Figure 5

Bang Phlat, Google Earth

Figure 6

Flood Governance Network of Bangkok

Table 1

Operationalization of Major Concepts

Table 2

Distribution of Income in LICs during the 2011 Flood

Photo 1

Road to Samakki Community, 2011

Photo 2

Pra Cha Rat Samakki Community, 2014

Photo 3

Food Team of Fah Mai Community, 2011

Photo 4

Woman cooking for the Fah Mai Community, 2011

Photo 5

Fenced houses of Soi Rot Fai Worrapong Charansanitwong 46

Community

(9)

Photo 6/7

The path leading to Fah Mai Community, 2014

Photo 8/9

Drainage System in the border region of Bangkok Province and

Samut Prakan Province (red circle marks Bangkok)

Photo 10/11/12

Two of the five costal water gates where water is pumped into the

Thai Gulf

(10)

1

1.

Introduction

"Worldwide disasters associated with natural hazards severely undermine local and national development efforts to support livelihoods, stimulate economic growth, and achieve overall human well-being. [...] Floods affect more people than any other type of natural disaster and cause some of the largest economic, social, ecological, and insured losses" (Keating et al, 2014: 1).

1.1. Context

The current and future socio-economic impact of floods is expected to be enormous. Floods are not merely a threat to individual livelihoods and human well-being, but to the overall development processes of entire regions or cities as well. Obviously, holistic approaches to cope with and combat severe flooding are urgently needed, and in the face of climate change, more than ever. Like any complex multi-scalar issue, the issue of flooding needs to be addressed on various scales in order for flood prevention or coping to be successful. It is only logical that local and municipal efforts need to be combined, well aligned and coordinated to increase the resilience of mega-cities against flooding.

The underlying issue of this research is not just flooding; it is also climate change and the expected increase in intensity and frequency of floods, as well as the need for highly populated areas to adapt and build resilience to this threat. This research focuses on the resilience creation or enhancement through good governance of existing resilience activities within multi-scalar cities.

1.2. Flooding and Resilience

"Traditionally flood risk reduction was mainly delivered by structural solutions, such as dikes, levees, embankments etc" (Linnerooth-Bayer et al, 2014: 2). These measures are usually very cost-intensive and rigid in their ability to adapt to evolving flood threats, which means they are at risk of being insignificant if the flood is too high or approaches from an unexpected direction. Moreover, these measures focus on higher authorities like municipal governments and institutions while ignoring the large potential of local community adaptation capacities. The strengths of local measures is that these "measures usually do not require large investment in hard-engineered infrastructures, as typically do structural measures, but rely on a good understanding of flood hazards in complex socio-ecological contexts" (Linnerooth-Bayer et al, 2014: 2). In the recent past the attention has gradually shifted from hard structural strategies, to softer strategies of adaptation. The shift from structural to soft measures is also the shift from "only hydraulic and engineering [solutions to] also socio-economic and environment aspects" (Tingsanchali, 2011: 25), mainly driven by reoccurring failures of the structural measures. While structural measures are supposed to function like a fortress against floods, the soft measures introduce a new socio-ecological form of resistance against floods - resilience.

To protect coastal urban areas, like Bangkok, effectively from increasing flood risks, the art lies in combining structural and rather intuitive non-structured measures to create a holistically flood resilient city. In order to make long-term integrated approaches to flood management

(11)

2

successful the engagement of the communities at risk of flooding is critical. Building or strengthening

community flood resilience is an urgent challenge in urban regions threatened by flood all over the world. However, there is only fragmented knowledge of community resilience strategies, or how well they are integrated with larger structural flood resilience measures in urban settings. In urban settlements where people's everyday lives greatly depend on larger systems of organization, which they themselves have no direct influence on, higher authorities and networks play an important role in creating and enabling resilience.

1.3. Research Aim and Relevance

The primary aim of this research is to fill the gap of knowledge on urban flood resilience strategies and their inter-scalar integration and compatibility, using the case of Bangkok. Periodical floods are not new to the city of Bangkok and many of its residents. Bangkok, having a certain amount of experience with floods can possibly serve as a learning case for building up flood resilience measures on various levels for other cities that face increasing levels of flooding in the future. Flood-experienced mega-cities like Bangkok are potentially extremely relevant to the climate change adaptation debate. What experience do residents of Bangkok have with flooding? How do they cope with it? What experiences does the municipality of Bangkok have with flooding? What coping-strategies did they develop? What is the relationship between municipal and community flood resilience strategies? How can they support each other in successfully preparing for and overcoming times of severe flooding? These are all important questions when looking ahead to the increasing challenge of climate change adaptation in coastal zones or lower basins of large perennial rivers. Hence this research, even though it only deals with Bangkok, has greater relevance in the light of combining community and governmental efforts in a sort of loose governance system around the challenge of flooding. Flood resilience strategies at the municipal and local level in Bangkok ought to be based on decades of learning from experience. The process of learning is crucial in building up resilience strategies. Learning from existing examples like Bangkok can hasten this usually long process of learning and preparing for future floods.

The academic relevance of this thesis is the application of the governability framework, designed to assess any form of societal governance setting and so far only applied to the governance of fisheries. This research will apply a simplified form of this framework onto the multi-scalar governance and coordination of different resilience strategies and actions against flooding.

1.4. Research Question

Creating flood resilience in large urban settings cannot be the responsibility of one group alone. Building flood resilience requires the interaction of various actor groups along several levels of flood coordination and management - from the individual person managing the flood around his or her house and neighbourhood, to local governmental institutions coordinating the immediate flood response, to the larger scale metropolitan or national efforts to strategically plan measures and projects against flooding. This is why the central question of this research focuses on the interactions and integration of inhabitants and governmental bodies in a flood situation and the

(12)

3

creation of a loose management or even governance networks around the issue of flood. Hence the main research question is:

How are community flood resilience strategies integrated with larger structural flood resilience measures in urban Bangkok, and what does this imply for the success or failure of flood governance?

Sub-questions:

1. What are the non-structural flood resilience strategies of the residents of Bang Phlat ?

2. What are the structural flood resilience strategies of the local and municipal institutions of Bang Phlat and Bangkok?

3. To what extent is a multi-scalar integration between non-structural flood resilience strategies of Bang Phlat's Residents and structural flood resilience strategies of local and municipal institutions achieved, and what are main challenges in this integration process?

1.5. Outline of the Chapters

In the following chapter the theoretical framework is outlined. It rests on a conceptualization of Resilience, a conceptualization of network governance, and the so called governability framework. The third chapter starts by presenting the main and sub research questions, and moves on to the conceptual framework of this thesis, which combines the theoretical framework and concepts with the posed research questions. Afterwards it outlines the operationalisation of said concepts. Then chapter three continues with describing the research location and the research methodology, discussing the units of analysis, research methods used, the process of sampling, and the limitations and ethics of this research. Furthermore, it explains the empirical context, hence the issues at stake in the context of the research location, Bangkok. The fourth chapter focuses on the first sub-question, hence on the flood resilience strategies of inhabitants of Bang Phlat collected during the fieldwork. The fifth chapter answers the second sub-question, describing the flood resilience strategies applied by the Governmental Institutions. The third sub-question is addressed in chapter six, outlining the interconnections, and nature of such, between the resilience strategies of all levels found in the field. Chapter seven draws conclusions from all three sub-questions and aims to answer the main question posed in this research, by placing the findings into the theoretical framework, addressing eventual shortcomings of the research and providing recommendations for future studies.

(13)

4

2.

Theoretical Framework

The theoretical framework is a combination of the basics of the governability framework by Bavinck et al (2013) and theories on resilience and network governance. The governability approach is introduced by Maarten Bavinck, Ratana Chuenpagdee, Svein Jentoft, and Jan Kooiman in their joint work 'Governability of Fisheries and Aquaculture: Theory and Applications'. Usually the governability approach provides a framework to assess and evaluate the governing-capacity of a societal system, however in this research it will be used to analyze the resilience strategies against flooding on different governing levels - the community level and the institutional level of flood governance - as well as their multi-scalar integration into a larger governance network, and what this implies for the success or failure of flood management in Bangkok.

As the governability approach presents the larger embracing theory within this research, it is explained first. The governability framework is used in this research to coat the resilience theory and provides the larger societal image describing how the parties involved relate to each other and position themselves in the greater societal system. Secondly, a section on network governance provides further theory complementing the governability framework. Thirdly, resilience is addressed by providing a theory background on resilience, and specifically social resilience, which is relevant to this research. Several conceptualizations of resilience are presented in this chapter, ending in Tyler and Moench's (2012) framework for urban climate resilience which provides the conceptualization of social resilience used in the further course of this research. Lastly, the relationship between social resilience and governance within this research is explained.

Before starting with the explanation of the main concepts, one central question has to be asked: Why link resilience against flooding with governance at all? The answer is that good management, organization and coordination - hence governance - increase the likelihood of sustaining desirable pathways of development, especially in changing environments where the future remains more or less unpredictable and surprise is likely (Walker et al, 2004; Adger et al, 2005). Through good governance the status quo but also the future development path of a city/region become more resilient to damage or destruction.

2.1. The Governability Framework

According to Bavinck et al, a societal system is defined by three interrelated components: (1) the system-to-be-governed - i.e. the object of governance, (2) the governing system - the subject, and (3) the governing interactions between the two (Bavinck et al, 2013: 12). The idea is, that Governors, the governed, and their interactions all contribute to the quality of governance. Figure 1 shows and explains the governability approach in its original form.

(14)

5

Figure 1 - The Governability Approach, Bavinck et al. (2013)

The system-to-be-governed is the best way to start the journey through the framework, as it describes the complex arrangements naturally in place that any governing system should address. The system to be governed is a combination of a Natural System - e.g. a river, a forest, a desert, a lake - and a Human or Social System structured around the Natural System. The Natural System strongly influences the Social System in shape and characteristics as it builds the environmental base for its existence, and the Social System has a large influence on the Natural System as it makes use of it or forms it according to its needs. Both are really diverse in kind and their relationship is not necessarily harmonious. In contrast, especially in urban areas, factors like dense population, extensive use of all kind of natural resources, and pollution are aspects of the social system, that can stress or overexploit the natural system.

In order to coordinate and manage the co-existence of natural and social systems, any societal system will sooner or later develop some forms of coordination systems (a governing system) which are more or less successful in addressing the complexities, diversities, dynamics and scales in the system-to-be-governed. Hence how much a governing system actually manages to govern or influence a system to be governed does not only depend on the governing system, but also on the system that it attempts to govern.

As the governing system regulates the system-to-be-governed, and the system-to-be-governed responds to that – whereby a response can also mean the absence thereof –, a relation between the two forms. This relation is described by the term "governing interactions" (Bavinck et al, 2013) (see Figure 1) and the form and quality of such interactions will mirror central features and capacities of the governing system and the system-to-be-governed. For smooth governing, a convergence between values, norms, strategies, instruments etc. in the governing system and the system-to-be-governed is desirable. There are three basic so-called modes of governing interactions (Bavinck et al, 2013: 21). They are self-governance, which is "the capacity of social entities to govern themselves (...) [and can be] found in all societies" (Bavinck et al, 2013: 21), hierarchical governance, which "is the usual style in which governments interact with their citizens" (Bavinck et al, 2013: 22), and co-governance, which is the sharing of governance responsibilities of authorities and other organizations with other stakeholders (Bavinck et al, 2013: 23).

"Governance also involves the need for people involved in a system-to-be-governed to pro-act or re-act to activities by governing institutions [...] and thus make governing a process of

(15)

6

interaction" (Bavinck et al, 2013: 20). General advantages of citizens' participation in public affairs, such as flood prevention and management, are "positive effects such as development, education, learning, integration, improvement of results, better understanding and control, legitimacy, and accountability" (Bavinck et al, 2013: 20). While generally recognized disadvantages are participation as "inefficient, politically naïve, unrealistic, disruptive and dangerous. It frequently results in excess time consumption, costs, hostility, worse outcomes, loss of control and complacency" (Roberts, 2004, in Bavinck et al, 2013).

This research is interested in these governing interactions, between the system to be governed (Bang Phlat District in Bangkok) and the governing system (the LDG, BMA and higher authorities) with regard to flood coping and management strategies. The form of such interactions will mirror central features of integration and compatibility of flood resilience strategies on the national, municipal, local and even neighbourhood level in Bangkok. The governability framework allows to go past identifying the levels of flood resilience strategies, but judge whether the case of Bangkok is an example for a well-integrated flood governance system or not. Integration, compatibility, interconnectedness or interoperability (Bavinck et al, 2013) are usually desirable features when two or more levels of any form of organizational order face a common challenge. Hence, the governability framework relates the resilience strategies of the governing to those of the to-be-governed, and presents a tool for analysing the two parties' nature of governance.

2.2. Network Governance

The governability framework of Bavinck et al, is kept very broad and is meant to identify the structures and frames of any governance setting and its social and natural environment. To narrow its focus some additional theories on network governance need to be added at this point. The governing interactions are where theories of network governance are to be placed in the theoretical framework of this thesis. Network and Governance are two buzz words of our time, hence a definition of the two words that fits into this research is necessary. Governance has many definitions but for this research the concept is taken out of its usual political or economic sphere and used to describe a more social process, namely the managing process of a natural disaster by the people living in the affected urban area. Hereby governance is defined in its loosest form as "the negotiated interaction of a plurality of public, semi-public and private actors to [manage] ... complex, fragmented and multi-layered [problems]" (Sørensen and Torfing, 2005: 195) like the clash of a natural system through a natural disaster with its social system. In case of a flood in a large city, all actors involved into coping with this problem are connected on multiple levels and often in an overlapping manner, sometimes directly and sometimes indirectly, sometimes hierarchically and sometimes horizontally, sometimes loosely and sometimes formally - in other words they build a network of connections and interactions that rests on a multi-scalar integration of their actions. All together they determine the integration, compatibility, interconnectedness or interoperability of the governance network which in this case revolves around the issue of flooding. As flooding is usually a negative event for all parties involved, a common general goal characterizes this network, namely to avoid and manage the flood as well as possible. In practice, however, integration, compatibility, interconnectedness, interoperability, or a collective way to manage the natural disaster proves hard to be achieved. Key features that provide the basis for integration, compatibility, interconnectedness and interoperability are participation, communication and a clear or transparent division of

(16)

7

responsibilities or tasks. These features have to be present before one can even address higher goals of governance such as accountability, legitimacy or even democracy.

According to this definition, a governance network has to be neither necessarily democratic nor egalitarian or free of any suppression, domination or patronage, in order to qualify as one. On the contrary, governance networks can be characterized by a mix of ties and connections of different forms which are best described by Bavinck's et al hierarchical, co-governing and/or self-governing types of interactions.

As said, a governance network around the issue of flooding includes many people, more so in urban settings compared to rural settings, as simply more people are affected by the collective governing process of floods. One can roughly distinguish between the group of directly flood-affected people that engage in the governing process by being flood-affected by the problem of a flood and merely responding in some way or another to it, and the formally affected people that are not necessarily personally flooded but carry out any form of responsibility, duty or other task related to the flood and hence affect the governing system around flooding. The second group can be state-driven actors like governmental institutions and their sub-offices, the military and in the case of Thailand the royal flood initiative that carry the responsibility for official flood response and prevention, and non-state driven actors like NGOs, private donors, private electricity companies or transportation providers affected by the flood. This research will focus mainly on the first group and the state-driven actors of the second group which both built the units of analysis. The non-state driven actors are touched upon through their involvements with the two units of analysis but are not a unit of analysis by themselves. The reason for this are the time and resource constraints of this research, but also the still dominating belief that, compared to the private market, the state-driven actors carry the larger and main responsibility of reacting to a natural disaster threatening the well-being of their citizens.

2.3. Resilience

The Resilience Theory provides a certain focus in the governability framework, which by itself is a very broad assessment tool for any governance setting. One of its strengths is that it can be applied to analyze all sort of socio-ecological governance setups. But if it is not filled with an additional theory or topic that narrows the focus of governance assessment, it can easily remain too abstract or too holistic to identify individual rudimentary problems, and thereby will not indentify concrete solutions to these problems.

Resilience is a term originating from the field of medicine and ecology, only rather recently has it been placed in the social field of management and planning (Davoudi et al, 2012). There are many definitions of Resilience. Gunderson et al (1997) define resilience as something's/ someone's ability to reorganize itself/oneself. Holling et al (1995) define resilience as "the buffer capacity or the ability of a system to absorb perturbations, or the magnitude of disturbance that can be absorbed before a system changes its structure by changing the variables and processes that control behaviour." The most basic definition is probably to describe resilience simply as “the ability of a system to absorb disturbances and still retain its basic function and structure” (Walker and Salt, 2006: 1).

Throughout the academic literature the concept of resilience consists of various different dimensions, even though they sometimes converge, depending on the field and purpose of study.

(17)

8

Several authors identify "the human domain and the biophysical domain [as] interdependent" (Walker and Salt, 2006: 38) and consider the system as a whole as a subject of resilience (Adger, 2000; Tyler and Moench, 2012). In this case the primary dimensions of resilience become ecological resilience and social resilience (or actor resilience), which are often complemented by the notion of institutions, which "in this case are defined in the broadest sense to include habitualized behaviour and rules and norms that govern society [as well as] formal institutions" (Adger, 2000: 348). Sometimes institutions are seen as an extra dimension (Adger, 2000), or as a linkage element between the ecological and social dimension of resilience "condition[ing] the way that agents and systems interact to respond to climate stress" (Tyler and Moench, 2012: 315)

The focus of this research, which connects the concept of resilience with governance, is social resilience (or actor resilience) as social agents mainly shape the rules, norms, behaviour and formal or informal institutions that all together govern flood response. Including the ecological dimension exceeds the resources of this research. This is however not too dramatic, as the capacities and actions of social agents embody a main factor of creating environmental shock resilience, especially in urban settings, and are worth examining separately. As the ecological dimension of resilience will be neglected, so will be the natural system of the governability framework (see above) even though the broad ecological facts concerning flooding will be provided in the form of context in Chapter 3.

Social Resilience is one dimension of overall resilience, but is further conceptualized into different dimensions by different authors. Three possible dimensions of social resilience, drawing from Carpenter et al. (2001) and Folke et al. (2010) are adaptation in the face of disturbance, self-organization, and learning. These dimensions are still rather wide for a fieldwork study. Self-organization for example can depend on so many variables, such as a person's or group's resources, networks and availability of knowledge or information. Moreover this conceptualization is better suited to connect the concept of resilience to the sustainability debate , rather than the governance debate, as it stresses processes of social change rather than processes of management.

W. Neil Adger, who defines social resilience as "the ability of groups or communities to cope with external stresses and disturbances as a result of social, political, and environmental change" (Adger, 2000: 347), claims that "social resilience has economic, spatial, and social dimensions" (Adger, 2000: 349) referring to the socio-economic background behind any persons resilience-act, and stresses the connection between the social actor and his/her spatial surroundings. This conceptualization, with its strong connection to ecological resilience, does not allow the researcher to neglect the ecological dimension of resilience as intended in this research (for the reasons mentioned above).

Tyler and Moench (2012), in their framework for urban climate resilience, identify agent resilience as one of three main dimensions of resilience, next to system resilience and institutions which do or do not foster resilience. In the governability framework from Bavinck et al (2013), agents and institutions are not separated as strictly, which makes sense, as institutions are basically built and managed by agents and hence are an informal or formal manifestation of agent behaviour. In the governability framework, the governors can also be the governed, even united in the same person, which is more realistic than putting actors and rules/institutions into two different categories. Consequently, this research will use Tyler and

(18)

9

Moench's conceptualization of actors resilience, but attempts to use the same conceptualization on institutional actors instead of separating them from other social actors. Moreover, by separating actors from institutions and rules, the element of self-organization can be missed (Tyler and Moench, 2012: 320). However, self-organization is expected to be an important feature of flood resilience in large urban spaces, especially in cities were governmental jurisdiction and activity do not reach out to all places and citizens. The focus of Bavinck et al's framework lies at the governing interactions between the different dimensions (not at the dimensions themselves) and is hence a much more modern approach to the interplay of (eco)systems, social actors, and institutions. Hence, from here on forward the concept is referred to as social resilience, instead of actor resilience, as it includes social actors and social institutions.

What makes Tyler and Moench's work fit into the governance debate is their aim, which is to develop a "framework for planning practitioners ... to adjust policies, practices and plans in order to avoid negative impacts of climate change" (Tyler and Moench, 2012: 311/312).

Tyler and Moench's first dimension of social resilience is Responsiveness or the "ability to organize, or reorganize in a timely manner; ability to identify, anticipate, plan and prepare for a threat, disruptive event or organizational failure; and to respond quickly in its aftermath" (Tyler and Moench, 2012: 316). This dimension seems to overlap a bit with what other authors call adaptation or adaptive capacity (Carpenter et al, 2001; Folke et al, 2010; Scoones, 2009) as the act of response to change or shock is partially similar to the act of adaptation. They are however not entirely the same, as responsiveness includes a certain notion of anticipation and pre-impact planning followed by a quick response, which adaptation does not necessarily include. This makes Tyler and Moench's conceptualization of social resilience very applicable to rapidly occurring shocks like a flood. Factors influencing responsiveness are knowledge and skills that can help a person to anticipate and plan, a good degree of information and communication allowing a person to better judge a situation of shock and consequently plan his or her response to it, and a clear division of tasks and responsibilities especially if many people are involved. Consequently, good governance influences responsiveness.

Tyler and Moench's first dimension, responsiveness, is inter-linked with the second dimension, Resourcefulness or the "capacity to mobilize assets and resources for action. This includes the ability to access financial and other assets, including those of other agents and systems, through collaboration" (Tyler and Moench, 2012: 316). Also other authors see an element of dependency in responsiveness or adaptation. Scoones (2009) states that adaptive capacity depends very much on the resources one has at hand to master any new situation, meaning resources in the broadest sense including social capital, financial or economic capital, and natural or spatial capital. Note that the capacity to take action/respond using one's own or network resources, mentioned by Tyler and Moench (2012), suggests that there is a certain diversity, or at least flexibility of resources necessary to act. In line with this, Adger (2000) states, that functional diversity contributes to resilience as it allows to bridge the temporary breakdown of one or several functions of a (social) system through other functions of the same system, and thereby enables this system to maintain and possibly even re-establish itself in its old form – the tipping point of system failure is higher. Hence the concepts of diversity and flexibility connect resourcefulness and responsiveness with each other, as a response depends on the diversity and flexibility of resources.

(19)

10

Tyler and Moench's third dimension of social resilience is the Capacity to Learn, or the "ability to internalize past experiences, avoid repeated failures and innovate to improve performance. This includes the capacity to build and retain knowledge over time" (Tyler and Moench, 2012: 316). It is crucial to understand the origins and circumstances of people's vulnerability, to develop resilience strategies decreasing vulnerability. Understanding and reacting accordingly is a learning process which is crucial for the development of resilience, as resilience is rarely naturally given. Moreover, learning from previous shocks can increase the preparedness for future shocks, in the sense that the gained knowledge or skills can help to anticipate and respond better to the future threats. Hence the learning capacity closes the circle of resilience again by touching upon the first dimension, Responsiveness.

Through "building up communication networks and sharing knowledge" (Schmiedgen, 2013: 10), lessons learned can be distributed and hence serve a wider range of people. Community networks of shared information, knowledge, and mutual assistance are effective tools of self-organization and, though often neglected, are vital characteristics of "resilient cities, capable of preparing for and recovering from impacts" (Farinaci et al, 2010: 7). This learning process can be individual but, especially in urban environments, a shared or collective learning-from-experiences is desirable for the entire system to develop a larger scale of resilience than one person or group of persons could create by themselves relying on their relatively smaller amount of resources, networks, and experiences. In general "social resilience is defined at the community level rather than being a phenomenon pertaining to individuals" (Adger, 2000: 349). Hereby one person's or group's resilience strategies can benefit someone else beside themselves alone, or entirely benefit people other than the person(s) implementing the act of resilience. This means that the resilience creator does not necessarily have to be the resilience beneficiary. This also means that there can be free riders in the creation of urban resilience, who benefit from other people's strategies to cope with an external shock without preparing or acting themselves.

2.4. Governance and Resilience

This thesis puts the key focus not solely on resilience but also on governance and how good governance essentially helps to create or strengthen resilience. Therefore it is necessary to integrate a theory or framework that singles out governance as the key element of the success and failure of societal systems (governability framework). Resilience theory attempts to understand the difference between success and failure of a socio-ecological system in case of shock and if, as Bavinck et al (2013) rightfully observe, governance is the key to socio-ecological organization, then governance deserves a spotlight position when talking about social resilience efforts. Social resilience, more than other forms of resilience, is based on human interaction and organization, which puts governance at the heart of social resilience creation. The paradox of combining the governability framework with social resilience theory is that, while the governability framework starts out as the outside frame to position social resilience within a specific social governance set-up, governance at the same time becomes the core of social resilience creation. No other social resilience theory places that much importance on governance (Adger, 2000; Davoudi et al, 2012; Farinaci et al, 2010; Holling et al, 1995; Scoones, 2009; Schmiedgen, 2013; Tyler and Moench, 2012).

(20)

11

3. 

Research Methodology

This chapter presents the research questions, the conceptual scheme for this thesis, and the operationalisation of the major concepts that develop from the research questions in line with the theoretical framework. Furthermore, a section each is devoted to the research location, the units of analysis, all research methods used, the sampling methods, limitations, the ethics, and the empirical context of this research.

3.1. Research Question

This research aims at answering the following main research questions and three further sub-questions.

How are community flood resilience strategies integrated with larger structural flood resilience measures in urban Bangkok, and what does this mean for the success or failure of flood governance?

Sub-questions:

1. What are the non-structural flood resilience strategies of the residents of Bang Phlat ?

2. What are the structural flood resilience strategies of the local and municipal institutions of Bang Phlat and Bangkok?

The sub-questions 1 and 2 enable me to first of all distinguish between the different categories of actors which are divided by the governability approach into formally governing actors (institutions) and to-be-governed actors (Inhabitants). Along the same lines sub-questions 1 and 2 distinguish between structural and non-structural resilience strategies. Hence, the first two sub-questions enable me to distinguish between the different actors and their strategies within the governance of flood resilience nexus, before evaluating their level of integration.

The third sub-question then addresses the different types of interactions and issues within the integration process. This question hence focuses on the governing interactions between the actor categories. In this research, as within the governability approach, they are the central and final feature of successful or unsuccessful governance. Even if sub-question 1 and 2 come to positive conclusions about the existence of flood resilience strategies, a positive feedback on overall urban flood resilience depends very much on the answer to the third sub-question.

3. To what extent is a multi-scalar integration between non-structural flood resilience strategies of Bang Phlat's Residents and structural flood resilience strategies of local and municipal institutions achieved, and what are the main challenges in this integration process?

3.2. Conceptual Scheme

For the conceptual scheme, the most basic concepts of the governability framework are used to frame the situation of social governance structures and the threat of flooding in an accessible way. Through this framework, one can address flood resilience strategies within the

(21)

system-to-12

be-governed and the governing system, and address their governing interrelations. A distinction is made between official structural flood management strategies designed by government actors and official staff in the governing system, and non-structural flood-coping strategies developed by locals living in a district of Bangkok that is regularly flooded. The intention is to analyze what they are, if they match, interact with each other, are supportive of each other or are based on the same realities and, if they thus contribute to a comprehensive flood management or, if not why not. Resilience theory fails to adequately deal with the issue of governance therefore the governability framework provides a great addition in this research to look at resilience strategies in combination with human organizational interaction and governance.

Figure 2 shows how the original governability framework was adapted to suit this research. The governing system, the system to be governed, and the governing interactions were kept as categories and so were their basic positions to each other (see section 2.1.). Within each sub-system, the focus lies on the flood resilience strategies developed. The research sub-questions one and two are addressed in the upper and lower light-blue box, asking about existing flood resilience strategies at both levels, within the governing system and the system-to-be-governed. In the original governability framework, the gearwheels in the middle describe the possible interaction/governance modes (self-governance, hierarchical governance, and co-governance) of municipal and local flood resilience measures. These are very general and broad based features which are very useful for grouping and labelling certain interactions and provide a good high-level overview. However, for this research's purpose, they do not give enough information about the practicality and emergency functioning of the flood governance network. Hence, other features identified in chapter two on the theoretical background of this thesis replace these governance modes and are meant to analyse the integration, compatibility, interconnectedness, and interoperability of the flood governance network (in line with Bavinck et al, 2013). These features are basically created through participation, communication, and a clear or transparent division of responsibilities or tasks on all levels. They have to be present before one can even address higher goals of governance, such as accountability, legitimacy in the decision-making process, or even democracy. This doesn't mean these later features are not desirable as well. Throughout the fieldwork I was reminded that one has to start at a more basic level in Bangkok. Consequently, the middle section is meant to answer the third sub-question, namely to what extent is a multi-scalar integration between non-structural flood resilience strategies of Bang Phlat's residents and structural flood resilience strategies of local and municipal institutions achieved, and what are the main challenges to this integration process?

The following conceptual scheme, is a diagram of Bangkok's flood governance configuration. The main focus of interest is what sets of strategies exist at both levels and how their integration is, or is not, contributing to the success or failure of Bangkok's flood management. Hence how

are community flood resilience strategies integrated with larger structural flood resilience measures in urban Bangkok, and what does this mean for the success or failure of flood governance?

(22)

13

Figure 2 - Conceptual Scheme

3.3. Operationalisation of Main Concepts - Social Resilience and

Network Governance

Table 1 demonstrates how the main concepts, identified in the theoretical framework, will be broken down into dimensions, variables, and indicators to make them accessible to field research. The concept of social resilience is divided into three dimensions, which are split into various variables that will be analysed on the ground. The indicators show possible information that is useful to conduct this analysis of the variables, and thereby answer the research questions. The operationalisation of flood resilience is mainly based on the work of Tyler and Moench (2012). As presented in Chapter 2, their main dimensions of agent/social resilience, in this case in relation to flooding, are responsiveness, resourcefulness, and learning capacity. Responsiveness depends on the variables 'capacity to anticipate and plan' as well as the 'ability to respond to the unexpected'. Commonly, responsiveness is indicated by "the ability to [predict], organize, or reorganize in a timely manner, the ability to identify, anticipate, plan and prepare for a threat, disruptive event or organizational failure, and to respond quickly in its aftermath" (Tyler and Moench, 2012: 316). Resourcefulness can vary depending on one's 'capacity to take action', one's 'own resources' and one's 'network resources'. Commonly, resourcefulness is indicated by "the capacity to mobilize assets and resources for action, [which] includes the ability to access financial and other assets [of one's own] and those of other agents and systems through collaboration" (Tyler and Moench, 2012: 316). While, finally, the capacity to learn includes the variables 'experiences', 'failures', and 'practices' and is commonly indicated by "the ability to internalize past experiences, avoid repeated failures and innovate to improve performance. This includes the capacity to build and retain knowledge over time" (Tyler and Moench, 2012: 316).

(23)

14

The concept of network governance (see Table 1) is split into three dimensions, which in turn are split into several variables, and those again into various indicators. The operationalisation of the governing interactions is inspired by Bavinck et al (2013) in the sense that the end goals of comprehensive governing interactions should lead to the integration, compatibility, interconnectedness, and interoperability of the flood governance network. However due to experiences in the field, they had to be brought down to a more basic level of integration, namely participation, a starting point for compatibility and interconnectedness, namely communication, and a more basic feature of interoperability, namely the transparent division of responsibilities and tasks within the network.

Concept Dimensions Variables Indicators Social Flood

Resilience Responsiveness Capacity to anticipate and plan

How is flood risk estimated? Recognize danger, organize or reorganize in a timely manner Identify threat, anticipate, plan and prepare for threat, disruptive event or organizational failure Development of a strategy or plan

Respond quickly in aftermath of the crisis or shock

Respond to the

unexpected Prepared excess capacity or backup e.g. savings, insurance, design of house, boat etc. The flexibility to perform tasks under a wide range of conditions

Flexibility to rearrange assets to find new ways Safe failure: loss of key components does not lead to a catastrophic failure of related systems Resourcefulness Capacity to take

action Ability to properly organize assistance or self-help Ability to mobilize own assets and resources for action

Ability to make use of one's network resources Own Resources Financial or economic resources or assets e.g.

income, second residence outside flood area, savings...

Physical Resources e.g. body health, age... Natural Resources

Network

Resources Access and mobilize community resources through collaboration Access and mobilize resources of other local networks through collaboration

Access and use resources of municipal and national networks through collaboration Access and use resources of business network through collaboration

Capacity to Learn Experience Ability to internalize past experiences Interpret experiences

Failures Identify failures Avoid repetition

Practices Ability and willingness to change practices and innovate to improve performance

Build and retain knowledge over time

Concept 2 Dimensions Variables Indicators

Network

Governance Participation Self-initiatives General willingness to help yourself our others Amount of effort one is willing to summon up Amount of effort one is able to summon up Access to other Size and quality of own social network

(24)

15

parties of the

network Constrains or obstacles in contacting others (higher authorities or persons) Responsiveness of network partners Communication Willingness to

communicate Interest in the issue Share information with others Ask for information from others Quantity of

Contact Amount of contacts to other parties of the network Diversity of parties within network Quality of

Contact Mutual respect and interaction _{(Un)biased distribution of preventive and acute} information, assistance and services

(Non-)occurrence of corruption Transparent Division of Tasks and Responsibilities Availability of information on responsibilities and liabilities related to flooding

Who knows who does what in case of emergency?

Where to ask for which sort of help?

Early warnings and sufficient information to act Constant and reliable update of information on the flood situation and on what has been done already and by whom

Table 1 - Operationalization of Major Concepts

3.4. Research Location

This section explains why Bangkok is suited as a research location for studying flood resilience developments at the community, local and municipal level. It also introduces the district of Bang Phlat as the more specific research location.

3.4.1. Bangkok

Bangkok is the capital of Thailand and has reached the status of a mega-city. It is located at the eastern coastline of Thailand at the Gulf of Thailand (see Figure 3). "It annually releases tremendous amounts of greenhouse gases into the atmosphere, thus contributing to the acceleration of global warming and the other deleterious impacts that will be the outcome of that warming, such as flooding ..." (BMA, GLF, UNEP, 2009: 4). On the other hand, Bangkok is also very vulnerable to these outcomes due to its geographical characteristics - large parts of the city are barely above sea level. "Bangkok is naturally prone to flooding and,

Figure 3 - Location of Bangkok owing to the rapid urbanization in recent decades, many

long existing watercourses, such as canals, ditches and ponds, were filled in and replaced by roads, buildings and other structures, thus exacerbating the effects of heavy rains" (BMA, GLF, UNEP, 2009: 8). Both, climate change mitigation and adaptation are challenges Bangkok is facing. The exposure to flood risk, a growing population and its assets, and the added threat through a changing climate in Bangkok, is a dangerous combination for the maintenance of livelihoods and overall human-well being.

(25)

16

3.4.2. Bang Phlat District

Bang Phlat is one of the 50 city districts (Thai: Khet) of Bangkok, and located just west of the Chao Phraya river. Compared to other districts it is rather small. This neighbourhood has a high risk of flooding and hence qualifies for this research. It is located within the inner city of Bangkok (Figure 4; District number 25) and 40.000 households are registered at the Local District Office of Bang Phlat, and another estimated 10.000 exist without official registration (I42). According to Adul Yothasmutr, the current President of the Local District Council of Bang Phlat, this adds up to about 100.000 inhabitants which live on an area of 11,360 km2_{. In the}

North Bang Phlat forms a part of the province border between Bangkok Province and Nonthaburi Province seperated by a Klong (Thai for canal). The Chao Phraya river runs for 6 km along the entire east side of Bang Phlat (Figure 5), and the Somdet Phra Pin Klao Road, the Borom Ratcha Chonnani Road and the Bang Klong Noi Canal (I42) form its southern border. Bang Phlat is mainly a residence area of Bangkok. Neither the sky train system, nor the metro line, which are the two main mass transit systems of Bangkok beside the Bus system, connect to Bang Phlat (though a sky train track is currently under construction along the main street Charansanitwong).

Figure 4 - Location of Bang Phlat District Figure 5 - Bang Phlat, Goolge Earth

3.5. Units of Analysis

The research question implies two units of analysis, flood resilience strategies of normal residents and flood resilience strategies of governmental institutions. First local resilience strategies, in Bang Phlat were studied focusing on flood response, resourcefulness, and learning capacity. In former times Bang Phlat used to be agricultural land providing food and nutrition to the old centre of Bangkok. Nowadays it is part of the city and a residential area consisting of mainly lower income and middle income classes. Therefore three willing communities of Bang Phlat were included into this research. As lower income communities sill represent the majority of the district, two lower income communities are included into the research (Fah Mai Community and Pra Cha Rat Samakki Community). The other one is a middle income community (Soi Rot Fai Worrapong Charansanitwong 46 Community) to also represent the growing part of middle income inhabitants of the District. Hence the units of observation (residents of Bang Phlat) represent the targeted district in terms of income distribution. This is important as, among others, financial resources are an indicator for flood resilience. I do not intent to compare resilience strategies of lower and middle income respondents. Including both of them only

(26)

17

serves the representativeness of the research. Moreover all age groups above 18 years are represented among the respondents. However, the agreement with the community leaders was to note down gender, address, and composition of the household of each respondent only, as a compromise between creating reliability for this research, and respecting the privacy of the respondents. However, for further studies it would still be easy to trace back the communities and households which participated in this research, which provides the basic degree of reliability a fieldwork study needs.

In total 51 residents of Bang Phat (39 lower income community members, 12 middle income community members) helped to construct an image of what individual households and their close neighbourhoods do before, during and after a flood to cope with the extra added stress of a natural disaster. Only one member per household was allowed to respond to interview questions, to avoid receiving two times the same information, as households usually acted as one in preparing and coping with flooding. Through this large number of respondents the internal validity of this research is strengthened. The external validity is though somewhat weaker, as a sample of 51 people are neither enough to represent Bang Phlat, nor the entire city of Bangkok or densely populated flood areas in general.

Secondly, the governmental or official flood measures of the same district, Bang Phlat, were studied, also looking at the same criteria. Hence the second unit of observation are themunicipal flood institutionswith jurisdiction inBang Phlat. In total three levels of Institution are included into this research, the Bang Phlat Local District Office and Local Politicians, the Bangkok Metropolitan Administration and the Royal Flood Initiative who is responsible for flood related projects all over Thailand and part of the national ministry of Irrigation. As the research depended on the cooperation of said political bodies, I could not decide on my own who and how many people could be interview from each of the corresponding bodies (14 respondents in total). At the local level a group interview was conducted with the Assistance of the Local District's Director, the head of the public work department and one of their technicians, the head of the social development department, the local affairs officer, and a mechanic engineer employed by the LDG. On the one hand, the group interview had the advantage of receiving holistic answers to all interview questions, which often thematically overlapped between two or three departments. On the other hand the group interview setting allowed the local district officers to be perfectly aware of each others' statements and align them accordingly, leading an external observer to the assumption of problem-free unity amongst them.

On the metropolitan level, representatives of the department of drainage and sewerage and the city planning department were interviewed independently from each other. On the royal or national level two senior engineers (apart from each other), involved in planning and operating the larger flood projects, agreed to be interviewed. All official respondents interviewed answered in the name of their institution, which should guarantee a high degree of reliability of the information gathered. A remaining risk is, that another representative of the same institution would have given different responses.

On top of that three independent experts were interviewed. All three are professors at different universities of Bangkok, working in the field of community empowerment and flood adaptation, community participation and local governance, and community participation and upgrading. These interviews were practical to fill some theoretical knowledge gaps and develop contacts to further possible respondents.

(27)

18

3.6. Methods

This research is mainly based on qualitative data collection methods. Qualitative methods are adequate to portray the individual subjective experiences with flooding which this research is interested in. Findings are consequently tied to a certain point in time and a certain phenomena, namely resilience against flooding, hence they are very useful to draw conclusions on the current situation of the phenomena of flood resilience in Bangkok and more specifically in Bang Phlat, but cannot be used to make generalizations about other phenomena that link to resilience, beside flooding. The findings of this research are built on observations, semi-structured interviews, and focus groups. Moreover secondary sources are used to check or corroborate the findings of this research. The following section will describe how each method was used in more detail.

3.6.1. Observations

During my eleven weeks in the field, I conducted observations with regard to flood resilience as an outsider but also through participating in everyday city-life. Important observations made were integrated into the semi-structured interviews. The method of observation was rather limited to large scale flood related projects that are visible also when no flooding is visible, like dams, canals, etc. I could not observe actual behaviour during flood, as Bangkok was not flooded during my fieldwork period. The observations primarily served the purpose of confirming the existence of certain building project that I had been told about during interviews.

3.6.2. Semi-structured Interviews

To research flood coping strategies of the municipality of Bangkok within Bang Phlat, selected individual officials were asked to take part in semi-structured interviews. The interviews took the indicators for responsiveness, resourcefulness and learning capacity as starting points and several questions were asked to each dimension of the flood resilience concept. In addition to that questions about governance, information flow and communication were asked. The findings from the interviews showed what flood resilience strategies for Bang Phlat, and in general, exist on the different government level in terms of responsiveness, resourcefulness and learning capacity. Moreover they showed how interconnected each institution was with higher or lower levels, residents and other relevant actors, and what characterised these connections.

Residents of Bang Phlatwere also asked to participate in semi-structured interviews addressing their flood experiences and possible responsiveness, resourcefulness and learning capacity. Moreover the questions investigated if residents made use or profit from any local, municipal or national flood-coping strategies, how they interacted with the municipality if applicable, and how that effects their own strategies. A condition for participation is that they had personally experienced at least one flood. The findings of the interviews demonstrated how residents of Bang Phlat cope with flooding and how their strategies fit into the overall flood-governance system.

3.6.3. Focus Groups

The research also included two focus groups with middle income residents of Bang Phlat, with once five and once seven respondents.. This was the wish of the middle income community, as for them it was more practical to collectively meet on two weekends in a row rather than squeezing individual interviews into their busy time schedules. It is rather hard to find a fitting

The flood resilient city : how multi-level governance could complement concrete in Bangkok

Verena Stefanie Peter

MSc Master Thesis December 2014

International Development Studies

T

HE

F

LOOD

R

ESILIENT

C

ITY

–

H

OW

M

ULTI

-L

EVEL

The Flood Resilient City – How Multi-Level Governance

could complement Concrete in Bangkok

Master of Science Thesis

International Development Studies

November 28, 2014

Verena Stefanie Peter

10638075

Verena.peter@student.uva.nl

Supervisor: Dr. B. Lambregts

Amsterdam Institute for Social Science Research

University of Amsterdam &

Division of Urban and Environmental Planning

Kasetsart University, Bangkok

Second Reader: Dr. J.M. Bavinck

Graduate School of Social Sciences

University of Amsterdam

Dedication

I would like to dedicate this Thesis to my twin sister, Mirjam Peter, who spent some time

in the hospital after a terrible traffic accident while the fieldwork for this research was

undertaken. I wish I could have visited and brought you flowers.

Acknowledgements

I would like to thank my supervisor, Dr. Bart Lambregts, for his constant support and the

contacts provided during the fieldwork period, as well as for his valuable feedback and

advice during the writing process. I would also like to thank my translator and guide

Wiradet Norasetworachai, for helping me with so much more than just the Thai

language, for showing me Bangkok, and providing incredible amounts of assistance with

adjusting to life in Bangkok, as well as with conducting this research. Furthermore I

would like to express my gratitude to Dr. Maarten Bavinck, who kindly agreed to second

read this thesis, and moreover through his academic work provided a part of the

theoretical framework to this research. Finally, I would like to thank all members of the

Fah Mai Community, the Pra Cha Rat Samakki Community and the Soi Rot Fai

Worrapong Charansanitwong 46 Community of Bangkok for sharing their time and

experiences with me, as I could not have done it without all of you, and I will honour

what I learned from each of you.

Thank You!

Abstract

Table of Contents

List of Acronyms

List of Figures, Tables and Photos

Figure 1

The Governability Approach

Figure 2

Conceptual Scheme

Figure 3

Location of Bangkok

Figure 4

Location of Bang Phlat Distirct

Figure 5

Bang Phlat, Google Earth

Figure 6

Flood Governance Network of Bangkok

Table 1

Operationalization of Major Concepts

Table 2

Distribution of Income in LICs during the 2011 Flood

Photo 1

Road to Samakki Community, 2011

Photo 2

Pra Cha Rat Samakki Community, 2014

Photo 3

Food Team of Fah Mai Community, 2011

3. 