RESILIENCE PATHWAYS

The European Journal of Development Research https://doi.org/10.1057/s41287-023-00605-w

ORIGINAL ARTICLE

Resilience Pathways of Informal Settlements in Nairobi: Stasis, Decline, Adaptation, and Transformation

Jan Fransen1 · Beatrice Hati2,3 · Naomi van Stapele4 · Samuel Kiriro5 · Rosebella Nyumba3

Accepted: 10 September 2023© European Association of Development Research and Training Institutes (EADI) 2024

Abstract

This study investigates resilience pathways of informal settlements, and their diverse impacts through community-based participatory research in Mathare and Koro- gocho, Nairobi. Results reveal that resilience pathways were relatively stable in the 1970s and 1980s, declined in the 1990s, and changed towards more adaptive approaches in the 2000s. Each period accommodates a dynamic mix of maladaptive, adaptive, and transformative resilience initiatives. Adaptive resilience initiatives are sustainable and adopt frugal practice to cope with resource constraints, without harming others. Transformative resilience initiatives are rare, as they require a sig- ni cant shift in societal norms and institutions. The study emphasizes the need to recognize the dynamics and variety of resilience initiatives in informal settlements. The study therefore recommends (1) recognizing and supporting adaptive and trans- formative resilience initiatives; (2) recognizing the dynamics and variety of resil- ience initiatives of informal settlements; and (3) addressing inhibiting institutional settings of informal settlements.

Keywords Resilience · Vulnerability · Frugality · Mathare · Korogocho · Kenya

Résumé

Cette étude examine les moyens de résilience à long terme des établissements in- formels, et leurs divers impacts, à travers une recherche participative chez les com- munautés à Mathare et Korogocho, Nairobi. Les résultats révèlent que les moyens de résilience étaient relativement stables dans les années 1970 et 1980, qu’ils ont di- minué dans les années 1990, et qu’ils ont changé vers des approches plus adaptatives dans les années 2000. Chaque période accueille un mélange dynamique d’initiatives de résilience inadaptés, adaptés, et transformatives. Les initiatives de résilience adap- tés sont durables et adoptent une pratique frugale pour faire face aux contraintes de ressources, sans nuire à autrui. Les initiatives de résilience transformatives sont rares, car elles nécessitent un changement signi catif des normes et institutions sociétales. L’étude souligne la nécessité de reconnaître les dynamiques et la variété des initia-

Extended author information available on the last page of the article

Vol.:(0123456789)

tives de résilience dans les établissements informels. L’étude recommande donc (1) de reconnaître et de soutenir les initiatives de résilience adaptative et transforma- tives; (2) de reconnaître les dynamiques et la variété des initiatives de résilience des établissements informels; et (3) d’aborder les cadres institutionnels inhibiteurs qui sous-tendent les vulnérabilités dans les établissements informels.

Resumen

Este estudio investiga los caminos de resiliencia a largo plazo de los asentamientos informales, y sus diversos impactos, a través de la investigación participativa basada en las comunidades de Mathare y Korogocho, Nairobi. Los resultados revelan que los caminos de resiliencia eran relativamente estables en las décadas de 1970 y 1980, dis- minuyeron en la década de 1990, y cambiaron hacia enfoques más adaptativos en la década de 2000. Cada período acomoda una mezcla dinámica de iniciativas de resil- iencia inadaptadas, adaptadas y transformadoras. Las iniciativas de resiliencia adap- tadas son sostenibles y adoptan prácticas frugales para hacer frente a las limitaciones de recursos, sin perjudicar a otros. Las iniciativas de resiliencia transformadora son raras, ya que requieren un cambio signi cativo en las normas e instituciones sociales. El estudio enfatiza la necesidad de reconocer la dinámica y variedad de iniciativas de resiliencia en los asentamientos informales. Por lo tanto, el estudio recomienda (1) reconocer y apoyar las iniciativas de resiliencia adaptada y transformadora; (2) reconocer la dinámica y variedad de iniciativas de resiliencia de los asentamientos informales; y (3) abordar los entornos institucionales inhibidores que subyacen a las vulnerabilidades en los asentamientos informales.

Introduction

The resilience of marginalized communities is high on the scholarly, social, and political agenda following the accentuated vulnerability of approximately 1.2 bil- lion urban poor residents housed in informal settlements (Mahendra et al. 2021). The co-occurrence of the global pandemic Covid-19 with a wide range of shocks, including the Ukraine war, persistent droughts, economic crises, and socio- political instability, intricately interwoven with ongoing deprivations, engenders informal settlements as ‘global hotspots of vulnerability’ (UN-Habitat 2020, p. 2). The capacities to cope in informal settlements are limited, yet (state) support is insu cient (Rivero-Villar and Medrano 2021). Urban poor communities strive to survive under constraints, but hybrid shocks have pushed millions of informal settlers below the poverty line, especially in Sub-Saharan Africa, threatening the progress made in poverty alleviation over decades (Yonzan et al. 2022). Resil- ience in this context draws attention. In 2014, the IPCC already warned that major disasters are likely to strike in informal settlements (Revi et al. 2014), as these settlements are vulnerable due to substandard and often illegal housing, infra- structure, services, and employment combined with limited governance capacity to reduce vulnerabilities (Seeliger and Turok 2014; Satterthwaite et al. 2020).

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

Urban resilience is a widely applied concept, yet it remains eminently fuzzy and contested in current policy and academic debates (De Carli 2016, p. 775). A popular application of the concept de nes resilience as the ‘capacity of a city (individuals, communities, institutions, businesses, and systems) to survive, adapt, and thrive no matter what kinds of chronic stresses or acute shocks they experience’ (Rockefel- ler Foundation and ARUP 2019). In a neo-liberal frame, informal settlers are por- trayed as incrementally building resilience as they confront precarity in their daily struggle for survival. Resilience evolves as communities gradually learn and prac- tice to survive, while concomitantly hazards are being eliminated so that they will not be required to deal with recurring risks (Kaika 2017). However, such a perspec- tive may over-glorify agency and aestheticize vulnerability, poverty, and su ering of the urban poor without necessarily exploring risks and government responsibility (Banks et al. 2020). The question is not only how vulnerable communities cope but also why they should cope with chronic stresses and acute shocks in the rst place: what institutional systems cause these vulnerabilities? A political-economic fram- ing brings to light that many vulnerabilities originate outside vulnerable communi- ties and require a systemic multilevel perspective with an eye for power imbalances (Kaika 2017).

A second contestation inherent in the resilience de nition is the interpretation and measurement of ‘transformation’. The concept of transformation is integral to resilience, emphasizing the ability of communities to recover, adapt, and progress towards a structurally improved state (Satterthwaite and Dodman 2013). However, the coping mechanisms of urban vulnerable communities are often insu cient to trigger transformation. They relieve acute su ering in the short term within limited time and space and may even “physically divert risk... to other members of the com- munity or, at best, function as temporal solutions that will have to be repeated in the future” (Rivero-Villar and Medrano 2021, p. 5). The application of a neo-liberal approach of resilience to these complex social systems without understanding their variety and power imbalances may hinder positive transformation (Pelling et al. 2012; Meerow & Newell 2019). Shocks and persistent deprivations point towards the need to transform informal settlements, but the transformation mechanisms within the context of informality and resource constraints remain unclear.

Thirdly, a challenge also lies in de ning ‘chronic stresses and acute shocks’. Resilience literature is concerned with disasters such as ooding, and drought linked to long-term stressors such as climate change. While these disasters do indeed cause havoc and occur more regularly and with increased intensity, our respondents o er a more socialized perspective. They de ne resilience as an everyday hustle to sur- vive in an uncertain environment. Their main chronic stresses are income loss and precarious livelihoods, unsafety, and illness. These everyday stressors are likely to increase in number and intensity due to the increased occurrence and intensity of disasters. By combining these notions, we contextualize the de nition of resilience to informal settlements as an everyday hustle to survive in a deeply uncertain envi- ronment, embedded in contextual (historical) practices and broader institutional frameworks.

This paper explores the long-term processes that emerge out of the systemic inter- action between institutional frameworks, vulnerabilities, and the everyday hustle

within informal settlements. Our analytical approach to resilience focuses on how the continual struggle for survival against multifaceted livelihood, health, and safety challenges is earned over time and space, also in the broader context of disruptive shocks such as gang entry and ousting or Covid-19. Resilience of informal settle- ments is not static: it changes for the good or worse in complex, systemic processes. This paper explains and compares resilience pathways of informal settlements over the last 50 years, de ned as space- and time-speci c processes within which insti- tutional contexts, vulnerabilities, and daily resilience initiatives of multiple actors co-evolve. This leads to our main research question: How do institutional contexts, vulnerabilities, and resilience initiatives interact to shape the resilience pathways of informal settlements in Nairobi, Kenya?

We trace the resilience pathways of two large informal settlements in Nairobi: Korogocho and Mathare. In both settlements, we nd that resilience pathways were relatively static in the 1970s and 1980s, declined in the 1990s and early 2000s, and transformed from 2006 onwards, leading to adaptive resilience pathways. We criti- cally explore these pathways and their relationship to the heterogeneous variety of often contradicting resilience initiatives within. Following Elmqvist et al. (2019) and Folke (2006), we do so by assessing if the initiatives are adaptive, maladaptive, or transformative. To appreciate resilience initiatives in a resource-scarce environ- ment, we turn to the concept of frugality, de ned as a creative and robust x using limited resources (Radjou et al. 2012).

The paper is structured as follows. “Theory” section brings together literature on resilience, informal settlements, and frugality. It discusses resilience pathways of an informal settlement, followed by a conceptual framework. “Methods” section subse- quently o ers the research methods, and “Findings” section mentions the ndings followed by debate, conclusions, and recommendations.

Theory

Resilience Pathways of and in Informal Settlements

Resilience pathways occur at multiple levels, which interact with each other over di erent spatial and temporal scales (DeWeijer, 2013 in Seeliger and Turok 2014). They follow path-dependent processes, whereby the past steers the direction of development (Folke 2006). In informal settlements, substandard built environments and institutional contexts are likely to condition resilience pathways. Understanding the resilience of informal settlements thus requires a historical, spatial, and multi- level perspective.

Exploring the resilience pathways of informal settlements is challenging, as it aims to zoom in on a meso-level phenomenon in uenced by macro-level institutional frameworks and encompassing heterogeneous subcommunities, households, resi- dents, and rms. We apply the concept of “general resilience”, de ned as the ability of a system and its subsystems to cope with shocks and stresses (Folke et al. 2010; Nykvist and Heland 2014). We treat general resilience as the resilience of informal settlements, acknowledging that they are open systems in which vulnerabilities and

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

resilience initiatives are conditioned by institutional contexts. We thereby perceive the general resilience of informal settlements as the meso-level meeting point of institutional contexts, vulnerabilities, and resilience initiatives. General resilience is subsequently operationalized for the core resilience challenges that our respondents have highlighted: livelihoods, with a focus on income and nance, health, and safety.

Whereas general resilience o ers a modal picture at the level of informal settle- ments, “speci ed resilience” describes how any part of the system or any actor in the informal settlement is impacted by and responds to shocks (Folke et al. 2010; Nykvist and Heland 2014). As informal settlements are heterogeneous, general resilience a ects subcommunities, households, and individual residents in di erent ways. In this paper, we describe how the general resilience of informal settlements changes over time and indicate how these changes impact the speci ed resilience of subcommunities and households in informal settlements.

Unpacking Resilience Pathways

To better understand the resilience pathways of informal settlements, we explore their institutional context, vulnerabilities, and resilience initiatives.

Informal settlements are often perceived as existing outside the formal planning and governance structures of cities, while they are deeply embedded in the urban fabric and socioeconomic systems. Informal settlements are both o cially recog- nized and regulated by city authorities, while also being neglected and stigmatized. For example, Nairobi County recognizes Mathare and Korogocho through speci c funds allocated for local development projects and engages with police, chiefs, vil- lage elders, and resident groups in matters of security. This recognition, although partial and contingent, demonstrates that informal settlements are not disconnected from the city, contrary to what terms like “non-city” may imply (Rojas Symmes et al. 2023). Instead, informal settlements are intertwined with, and in many ways, constitutive of not only the physical but also the social, economic, and political infrastructure of the city (Suhartini & Jones 2019). In fact, residents of informal set- tlements often navigate complex legal and administrative processes to access basic services and rights. In most cases, residents self-organize to engage with formal urban institutions to secure essential services and rights. As a result, many residents have developed innovative solutions to access basic services (Blomkvist et al. 2020). This leads to the emergence of complex governance arrangements centred around basic provisions, where actors within informal settlements become part of broader urban governance institutions, functioning as social, economic, and political nodes within the city.

Vulnerabilities in informal settlements are related to their demographic charac- teristics, spatial dynamics, and resource constraints. In neo-liberal urban landscapes, informal settlements are often located in high-risk zones or far away from employ- ment opportunities and services. Demographically, they are likely to house vulner- able and discriminated groups. Resource constraints, manifested in substandard infrastructure, precarious housing, and low incomes, make informal settlers more susceptible to shocks (Seeliger and Turok 2014; Satterthwaite et al. 2020).

Within the institutional and vulnerability context, informal settlers engage in make-do resilience initiatives, which involve everyday adaptations in immediate response to shocks and stresses (Jabeen 2019). These resilience initiatives are often precarious, temporary, frugal, and con icting. As they build on each other, informal settlements resemble dynamic makeshift environments (Pieterse 2008; Vasudevan 2015).

Informal settlers’ collective actions venture into developing shock-proof “hard” infrastructure or “soft” infrastructure, strengthening the capacity of people and their fragmented networks (Simone 2008). These actions may reduce vulnerabilities, such as through solid waste collection. However, collective action rarely challenges the broader institutional setting (Adger et al. 2009). Other actors, including govern- ments, donors, and rms, may also o er shock-reducing infrastructure, services, and support (Fransen et al. 2022). The governance arrangements that emerge when external actors partner with collective actions can lead to adaptive or transformative resilience pathways but can also result in participation fatigue, mistrust, elite cap- ture, or bureaucracy (Brandsen 2016; Fransen et al. 2022).

The multitude of resilience initiatives a ect resilience pathways in di erent ways. Following Folke (2006) and Elmqvist et al. (2019), we question whether these ini- tiatives are (mal)adaptive and/or transformative. Transformation requires changes in institutional contexts, which often involve more powerful actors such as govern- ments. Maladaptive initiatives increase vulnerabilities for others or in the long term. For example, if a government widens a road, it may improve the resilience of some urbanites but harm others if it leads to their displacement. Elmqvist et al. (2019) argue that self-organized initiatives may enhance speci ed resilience in the short run but overlook the resilience of the wider community or the longer term. A vulnerable household, for instance, may adapt to power cuts by illegally connecting to electric- ity, which increases the risk of res in the neighbourhood.

To identify whether resilience initiatives in a resource-scarce context are (mal) adaptive or transformative, we turn to the concept of frugality, which involves the innovative and minimal use of resources and intuitive processes to identify low-cost options by removing non-essential functions (Tiwari et al. 2014). The slogan “fail cheap, fail fast, fail often” (Radjou et al. 2012) represents the temporal, experimen- tal, and makeshift nature of frugal resilience initiatives. Frugal resilience initiatives o er a ordable, basic functionality with reduced total cost and easy access (Tiwari et al. 2014). Frugality may indicate adaptive or transformative capacity if it provides an a ordable and robust quick x to shocks and disturbances (Bahadur and Doczi 2016). Therefore, the adaptability of resilience initiatives is determined by who it bene ts or impairs, its likely long-term e ects, and its frugality (a ordability and quality).

Conceptual Framework

In line with theory, we conceptualize that resilience pathways of informal settle- ments are likely to be in uenced by their institutional context, vulnerabilities, and resilience initiatives (Fig. 1). The combination of (mal)adaptive and transformative

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

Fig. 1 Resilience pathways of informal settlements. Source Authors

resilience initiatives are formed by and form the institutional context and vulner- abilities of informal settlement. These interactions are expected to result in constant dynamics and may push resilience pathways in di erent directions. Resilience path- ways depict trends in the general resilience of informal settlements, which, in turn, impact the speci ed resilience of actors within informal settlements in various ways.

Methods

We conduct an exploratory qualitative, community-based participatory inquiry within the framework of a multiple case study design (Yin 2009). The study focuses on two large informal settlements, Mathare and Korogocho, which have a his- tory of over ve decades and are located in close proximity to each other (Fig. 2). Mathare, established in the 1920s, has not bene ted from formal redevelopment and only experienced a failed upgrading attempt in 2011–2012 under the phase 1 of the Kenya Informal Settlement Improvement Programme. On the other hand, Korogocho, established in the 1960s, has been a bene ciary of a rudimentary for- mal upgrading program. Recognizing the heterogeneity within the settlements, we select two neighbourhoods (Mitaa) from each settlement, anticipating that their diversity would be representative of the overall population variation in a minimal sense (Seawright and Gerring 2008, p. 297). In Mathare, we focus on Village 2 and Mathare 3C, which represent di erent socio-economic statuses and housing typolo- gies (shacks and concrete high rises) among the 13 villages in Mathare. In Korogo- cho, we examine two villages: Grogon, which has partially been upgraded, and K1

J. Fransen et al.

Fig.2 Map on Mathare and Kporogocho. Source Authors. Percentages are based on perceptions of a small sample and are therefore indicative. Sample size in Korogocho:12 in 1970s/1980s; 36 in 1990s/2000s. Sample size in Mathare: 17 in 1970s/1980s; 48 in 1990s/2000s

(Korogocho A), which has not been upgraded. While upgrading e orts are spatially bound, the changes a ect residents across all eight villages, although in di erent ways. Data from the selected Mitaa are aggregated and discussed at the settlement level, while leveraging the variety to add nuance to the resilience pathways.

Empirically, we employed a constructivist approach to examine the coevolution of institutional and vulnerability contexts and resilience initiatives. These constructs are operationalized in Table 1. Analytically, we applied the rounds model (Teisman 2000) to identify three distinct rounds with di erent con gurations of institutional and vulnerability contexts, resilience initiatives, and resilience outcomes: (1) stasis in the 1970s and 1980s, (2) declining resilience in the 1990s and early 2000s, and (3) adaptive pathways from around 2008 to 2020. These rounds are demarcated by decisive events, such as the rise and fall of powerful gangs as in uential actors. We are aware that a “one round, one pathway” approach is overly simplistic for complex and heterogeneous informal settlements. It represents general resilience, calculated as the modal score re ecting perceived improvements or declines in health, safety, and livelihood conditions within each round. While each round represents a distinct resilience pathway, we acknowledge that resilience outcomes vary and encapsulate di erent manifestations. This variety is partially described. Moreover, the bounda- ries between rounds are not rigid, as changes in pathways occur over several years.

In the study, we conduct individual storytelling sessions with 17 elderly inhabitants aged 66–80 years, with the youngest participant being 16 years old at the beginning of the rst round (1970s). These sessions comprehensively

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

Table1 Operationalization Variable

Institutional context Vulnerability context

Adaptability of resilience initiatives

Resilience outcomes

Source Authors

Sub-variable

Governance

Resources

Demographics Spatial setting

Actor Frugality

Adaptability General resilience

Indicators

Perceived role of/ coordination among actors Perceived resources

Density of settlement Locational risksDistance to work and services

Actor Quality A ordability

Sustainable use of resources

E ect on general resilience (harming others)

Perceived change in resilience (overall, health, safety, livelihoods)

describe and compare resilience, institutional context, and vulnerability from the 1970s to 2020. Semi-structured household interviews are conducted with 41 respondents aged 23–51 years, providing data primarily on the second round and partially on the third round. Additionally, we interview 29 key informants (such as village elders, community health volunteers, CBO representatives, chiefs, a businessman, and a religious leader) to enrich the historical narratives and household interviews and triangulate data. In total, 87 respondents are selected in a strati ed sample which aims for diversity in age, gender, education, and marital status (Table 2).

The data are analysed with the participation of community researchers in ve steps. First, we transcribe the data and developed coding schemes. Second, we apply a code book to group data for each settlement and round, incorporating the tacit knowledge of the community researchers. Third, we assess the perceived changes in health, safety, and livelihood for each respondent in each round and assign values, resulting in perceived general resilience outcomes for each set- tlement as modal scores. The factors across the three rounds are then con g- ured, leading to the grouping of pathways. Fourth, the participatory analysis is cross-checked by re-coding and regrouping the data using Atlas.ti software, and triangulating data with over 100 secondary reports, 23 of which are referenced in this article. By triangulating data we aim to overcome the recall bias. Finally, the re ned ndings are presented to residents of Mathare and Korogocho in data validation workshops conducted at the settlement level. This iterative process and triangulation increased internal validity, enhance the capacity of community researchers, and facilitate knowledge transfer to the communities. However, the external validity of the ndings is limited due to the context-speci c nature of the two settlements. Additionally, due to the small sample size within selected neighbourhoods, the study may not capture the full range of speci ed resilience and should therefore be considered exploratory.

J. Fransen et al.

Table2 SampleAge of all respondents

Gender

Education

Marital status of household interviews (41)

Total Korogocho Mathare

18–34 35–54 55–64 >65 30 33 7 17 34% 38% 8% 20% 18 12 2 8 45% 30% 5% 20% 12 21 5 9 25% 45% 11% 19%

Female Male 39 48 45% 55% 17 23 42% 58% 22 25 47% 53%

Primary Secondary Tertiary NA 9 21 14 43 10% 24% 16% 49% 4 14 3 19 10% 35% 8% 48% 5 7 11 24 11% 15% 23% 51%

Married Single Widowed Separated 21 16 2 251% 39% 5% 5%10 7 2 1

Source AuthorsAt least one di erently abled person was interviewed in each settlement

50% 35% 10% 5% 11 9 0 1 52% 43% 0% 5%

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

Fig. 3 Trends in general resilience of Mathare and Korogocho. Source Authors Findings

This section chronologically describes the resilience pathways of both settlements. In summary, respondents perceive that resilience pathways were relatively in sta- sis in the 1970s/1980s (which was a few years after the country’s independence in 1963), after which they declined in the 1990s which is labelled by other scholars ‘the lost decade’ (Rono 2002), and then increased again (Fig. 3). This trend appears to be most pronounced for perceived safety and least for perceived livelihood. Health resilience hardly reduced in Korogocho in the 1990s but increased sharply in the new millennium.

Mathare

Statis (Up To 1980s)

Mathare emerged in the 1920s (Chiuri 1978) and experienced growth in the 1960s as a bushy and sparsely populated village (M8, M14, M18). During the 1970s and 1980s, the settlement became denser as more people migrated to cities post-inde- pendence. Re ecting on those times, a 57-year-old resident (M62) narrated, “In those times, people were su ering a lot, but life was simpler”. People had limited resources, facing poverty, illiteracy, and weak employment skills (M14). However, the area had abundant land, although not fertile (M12), and the river provided clean water, while trees o ered fruits. The Mathare Quarry also o ered employment opportunities, particularly for the construction of the nearby Kenya Air force Moi Base (M8; Médard 2010; Andvig and Barasa 2014). Most houses were constructed using wood, mud, iron sheets, nylon papers, and/or cardboard boxes. Two elderly women referred to their makeshift houses in the 1970s/1980s as “homes”, but in subsequent rounds, they used the term “slums” (M64, M66).

Local vendors provided basic services, while many households resorted to reliev- ing themselves in the bush and sourcing water from the stream (M66). Some resi- dents established schools, and others advocated for services from the Nairobi City Council, forming the basis for community organizing connected to governance structures in the city through various arrangements. The government provided basic

infrastructure and services, which were perceived to be e cient but geographically distant (M10). Waste was collected weekly from designated solid waste collection points (M8), politicians facilitated access to housing, employment, and infrastruc- ture, and local chiefs facilitated community dialogues. Public toilets were numbered and used as landmarks for businesses and crime (M19).

During this period, most respondents perceived the level of resilience as average, serving as a benchmark (Fig. 3). The relative safety of Mathare was often associated by village elders (M71, M70) with the prevalence of non-tribal politics.

Peace was also very good, there was no tribal divisions, people lived together as one family. Division politics come later in the late 1990s when the multiple political parties was introduced. (M71).

Respondents in Mathare 3C, however, cite widespread petty crimes. An elderly resident stated that “one couldn’t even hold anything that could be seen sella- ble (1980s). One would be safe if he/she is known” (M68). This notion perpetu- ates to-date, “Mathare’s safety depends on you being a residence or a visitor” (M102). Drug abuse and alcoholism were also rampant (M14). Youth groups such as KANU youth wingers provided security in collaboration with area chiefs and were deployed to quell resistance to local administration. Other youth groups such as Bondeni and Kiamutisya had a more tense relationship with the local administration and police (M9; Van Stapele 2020).

Most respondents agree that health risks posed the largest challenge, as health services were hard to access. When people were ill, they “made use of herbalists, the well-known lady was Nyauyoma who assisted most people” (M9). Occasion- ally, local government would collaborate with local faith-based groups to provide health services, while the church brought in foreign doctors (M8). During a wide- spread cholera outbreak in 1987, the government responded fast: the area was immediately fumigated to avoid further spreading (M8).

The ability to cope with income loss was considered average (Fig. 3). Many households survived harsh periods by combining urban farming (Van Stapele 2014), sex work, re-using materials and brewing Chang’aa—illicit beer (M12; M14; M68):

Sex work is the oldest profession in Mathare. Most sex workers were single women, transferring the same to their daughters. (...) Young men indulged in criminal activities. (...) Chang’aa is the cash crop for Mathare residents, young men were the ones in charge of brewing while women sold the brew. (M68)

These livelihood activities enabled most residents to survive, but not to thrive. M72 noted that most households could a ord three meals a day. Community lead- ers o ered nancial support to some youth to alleviate poverty and counter crime (M70), while collective action (M8), including microcredit and merry-go-rounds (M9, M11) enabled some to smoothen income loss. However, other families were destitute. Some mothers picked potatoes and bananas that had been thrown away in neighbouring areas (M13). At the same time, households close to politics or with an entrepreneurial spirit ourished:

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

I came here with my husband when the lands were almost bare and it used to be a quarry, existing houses were countable. We are the ones who sold timber to people who were building at that time... I was the third person to set up water tap... then started selling water to the members of the community. (M66)

Decline (1990s and Early 2000s)

On average, respondents argue that resilience declined for most households in the 1990s and early 2000s (Fig. 3). Many households became more vulnerable for two reasons. First, as the settlement became denser, opportunities for urban agri- culture diminished, while competition for scarce resources intensi ed. Second, the role of the government underwent a change, becoming more political, while gangs gained greater control over services and public spaces (M7). This shift in the roles of government and gangs is believed to have started when a structural adjustment programme was announced in 1986 and intensi ed during politically instigated vio- lence in the advent of multiparty democracy in 1992, which created more opportuni- ties for gangs backed by local politicians. Starting from 1998, Mungiki groups grad- ually took control of public transport routes, illegal electricity distribution, public toilets, and the housing markets across most of Mathare, leaving only a few strong- holds for other groups. Residents began referring to them as “the local government” (Rasmussen 2010).

Respondents generally perceive a sharp decline in safety (Fig. 3). Most agree that safety services became politicized and fragmented. While some police o cers sup- ported community safety initiatives, others colluded with gangs and corrupted local leaders (M84, M106). Respondents’ perceptions on police killings of suspected criminals are strongly divided, ranging from support to condemning the police o c- ers as “killer cops” (M7, M8, M21, M84).

Due to political instigation and ensuing turf wars, Mungiki increasingly employed fear tactics and started demanding higher fees for local services, including security and electricity (M7, M69). Unsafety in Mathare was also associated with a pro- longed upgrading con ict in Mathare 4A, which had repercussions across house- holds in Mathare. The dispute between residents and upgrading partners revolved around the allocation of new houses and plots, as well as rent payments (Kigochie 2001). This con ict remains unresolved to this day.

Most respondents perceive that health resilience declined slightly as well (Fig. 3). The environment became less healthy as garbage collection by the city council ceased in 1992, public toilets were neglected, and violence increased. The occur- rence of diseases such as cholera, HIV, and tuberculosis rose, while major hospitals were located outside the settlement. Although health services were fairly cheap, they remained una ordable for many residents. Health risks escalated due to maladaptive resilience initiatives, including the brewing of chang’aa (illicit alcohol) with formal- dehyde (M73), unsafe electricity connections known as “sambaza” that caused res (M18, M19), makeshift water connections called “spaghetti” that exposed water to contaminants (Fransen et al. 2023), and the greasing of stagnant water to prevent mosquito breeding, which resulted in car oil seeping into drinking water.

Livelihoods were perceived to remain relatively stable over time, with continued access to jobs in the nearby city centre and higher-income areas. Respondents noted gendered livelihood roles, with women more likely to work as house helps in Indian estates, engage in sex work, sell chang’aa or vegetables, run food kiosks, or operate other small businesses (M67, M64). Some women were able to accumulate assets and purchase property (M62). On the other hand, men were more likely to engage in activities such as distilling chang’aa, working as casual labourers or security guards, or participating in criminal activities, including gangs (M62, M68; Van Stapele 2015). In certain households, particularly in Mathare 3C, connections with gangs resulted in increased wealth, while others struggled to a ord even one meal a day (M16, M72). Many resilience initiatives were perceived as maladaptive, including theft and the use of formaldehyde powder from mortuaries to accelerate and scale up the production of chang’aa (M17).

As the government withdrew from some basic services, more community-based organizations (CBOs) and non-governmental organizations (NGOs) began operat- ing in the area from the mid-1990s onwards, including the Mathare Youth Sports Association, Red Cross, and Médecins Sans Frontières. Some of these organiza- tions collaborated with local leaders (M9, M10). Additionally, private security com- panies emerged, primarily protecting businesses in the nearby settlement of East- leigh. However, they also provided employment opportunities as security guards (M10). Therefore, while general resilience declined, there were also some positive adaptations observed. By 2006, the institutional context had become intricate, with Mungiki exerting power, an increasing presence of CBOs and NGOs, and shifting roles of the government.

Self‐organization (2006 Onwards)

From November 2006 to February 2008, the police engaged in a battle with the Mungiki to regain control over informal settlements (Van Stapele 2015). During this violent period, the police recorded 108 extrajudicial killings in Mathare (Ombuor and Bearak 2020). Many residents no longer accepted the authority of the Mungiki, and some joined the ght, prompted by rising living costs after a tax hike on the local alcohol industry in November 2006 (Van Stapele 2015). Gangs also joined the con ict, attempting to seize pro table ventures and/or supporting politicians. The violence reached its peak during the post-election violence of 2007/2008, which e ectively ended the reign of the Mungiki groups in most villages (Bennett et al., 2015). In 2007, the government initiated “the peace project” to engage the youth in rebuilding peace, leading some individuals to reform from criminal activities and violence (M130). In 2013, the entrance of CBOs was further facilitated by the regu- larization of group registration under the Public Bene t Organizations Act of 2013.

Residents became suspicious of groups attempting to seize control, fuelled by fear of further violence (KNCHR 2008b), thereby creating opportunities for gov- ernmental agencies, smaller gangs, CBOs, NGOs, and private rms to ll the void. These various actors assumed the responsibilities previously held by the Mungiki groups, including garbage collection, security provision, rental housing, and dis- pute resolution. The government also increased its investments in local development

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

projects, although often driven by political motives. For instance, according to M7 and M14, political leaders were credited for the construction of infrastructure such as Mau Mau Road and public toilets. However, coordination among these multiple initiatives remained weak.

Most respondents perceived a gradual increase in resilience for the majority of households. Perceived safety improved but remained problematic (Fig. 3). Female respondents in Mathare 3C shared accounts of harassment and threats by gang members, particularly targeting single and divorced women (M75). Some families engaged with gang members socially in an attempt to ensure the safety of women (M90). The police provided security during the day but expressed fear of conduct- ing night patrols due to the presence of gangs (M93, M75, M72). Mathare also wit- nessed an increase in police killings of suspected gang members (M13; KNCHR 2008a; Van Stapele 2020). On one hand, gangs intimidated residents and domi- nated lucrative illegal businesses, including drug dealing. On the other hand, they o ered protection to informal settlers against police brutality and external crimi- nals (Mutahi 2011). Meanwhile, the police became increasingly intertwined with criminal gangs (M10), while the local government facilitated community policing through the Nyumba Kumi initiative. Security providers, suspects, and the police were at odds with each other but were also suspected of collusion, creating a uctu- ating, complex, and tense mix of interests. As M10 explained, “In some instances, you’ll see police patrol in your area, and after they have passed, the thieves take over... It makes you wonder if they are working together”.

Many respondents perceived an improvement in health resilience. They expe- rienced fewer diseases than before; for example, malaria stopped spreading dur- ing the COVID-19 pandemic, and tuberculosis became rare and curable. Better hygiene practices, increased awareness, and social measures not only fought against COVID-19 but also had a positive impact on other diseases (M17). Various adap- tive resilience initiatives were noted during the pandemic, such as government distribution of masks, cash transfers, installation of boreholes, and distribution of water and food. Community health volunteers provided support to local communi- ties, and water kiosks introduced digital payment systems under the auspices of the Nairobi City Water and Sewerage Company (M18). Furthermore, the government, CBOs, and NGOs constructed public toilets (M7, M14), private hospitals and clinics were opened in proximity with better equipment than before (M8), and the National Health Insurance Fund became available.

However, not all households experienced equal improvements in health, as the costs of water, toilets, and health insurance were perceived to be high (M7, M8, M18). For those who could not a ord these services, the government, NGOs, and CBOs o ered them at subsidized rates or even for free (M21). However, these ser- vices were often geographically distant (M8, M17, M18, M27). Solid waste man- agement was considered inadequate, as garbage collectors often disposed of waste in the river and did not educate households on proper waste management (M8).

Livelihood resilience is perceived to have slightly improved, although there are signi cant di erences between households. By 2019, the population of Mathare had increased to over 200,000 residents (KNBS 2019) and in its wake compe- tition for jobs, services, and infrastructure intensi ed (M11). However, new

income-generating initiatives emerged, such as subletting rooms, mask making, dig- ital work, or cleaning public space for money (M12, M22). Some households expe- rienced improved employment opportunities and gained access to better services (M8, M11). They also utilized digital information and e-banking services (M40), enjoyed better nutrition (M64), found stable tenancy closer to main roads (M7), and were able to save money (M74).

On the other hand, many people still resorted to working in Chang’aa dens (ille- gal alcohol brewing), drug-related activities, or engaging in sex work (M17). Most households remained poor (Give Directly 2021; M50, M64, M67). The poverty sit- uation was perceived to have worsened during the COVID-19 lockdown (M7). In response, households reduced their expenditure, engaged in “side hustles”, reused materials, sought income through government initiatives like “Kazi kwa Vijana” (Work for Youth), received nancial support from organizations like Give Directly, and/or accessed microcredit (M18). As the population continues to grow, reaching close to 250,000 people in 2022 (SDI-K, 2022), further changes in livelihoods are expected.

Korogocho

Stasis (1980s)

Korogocho, which means “scattered” or “broken”, is an informal settlement located at the foot of Nairobi’s land ll, approximately 11 km from Nairobi City Centre. In the early 1980s, the settlement consisted of scattered houses, with available land for urban agriculture and two clean rivers used for drinking and irrigation (K1, K3, K12). Most households relied on the land ll and urban agriculture for their liveli- hoods, as other job opportunities and services were relatively far away. The gov- ernment had limited involvement in the settlement, with key facilities such as the chief’s o ce, public o ce, police station, and public health centre located outside of Korogocho (K1, K10). Additionally, electricity was lacking in the settlement (K7, K9). The city council provided water taps along main roads, but limited water pres- sure led to water shortages and high prices (K2, K12). As a result, residents had to walk long distances in search of clean water. Over time, the settlement became more densely populated, particularly with garbage collectors working at the land ll. Within a decade, it grew to become the fourth largest informal settlement in Nairobi, with a population estimated between 120,000 and 150,000 people.

Respondents perceive the level of resilience during this period as somewhat lower compared to the present (Fig. 3). Particularly, health resilience was considered low, partly due to the distance to health services and a lack of awareness about ill- nesses (K12). Residents were vulnerable to various diseases, including malaria and tuberculosis, and there were cases of sexually transmitted infections attributed to the increasing incidents of rape against older women by young men (K10).

The perception of safety resilience varied widely (Fig. 3). Some respondents reported that crime was prevalent in Korogocho, with frequent reports of murdered individuals and bodies being dumped in nearby quarries (K12). However, K10

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

recalls that in the 1980s, the crime rate was relatively low. Youth groups such as the KANU Youth wingers provided security services (K13).

Most respondents perceived livelihood resilience during that period as more straightforward compared to the present, with fewer opportunities but also fewer risks and competition. According to K12, “We took three meals a day, but nowadays you skip meals or have tea without milk because you don’t have money”. K15 notes that residents were able to survive by supplementing their income from waste col- lection with urban agriculture and informal trade along the main road.

Decline (1990s)

In the 1990s, density increased, and local politicians began to establish their in u- ence through gangs after the realization of multiparty democracy (see also the sec- tion on Mathare). Since the government did not have a police station or o cial government representation, there was no formal oversight on the role of gangs. For security, residents relied on neighbouring settlements’ o ces, such as the Ruaraka police station and the Mathare Administrative Police camp (K9). At the same time, the land ll provided a constant ow of work, although it was unhealthy, unsafe, and casual.

Changes in the institutional and vulnerability context were associated with a perceived decline in resilience by most respondents, particularly in terms of safety (Fig. 3). Korogocho was considered the most unsafe settlement in Nairobi during the 1990s (K6), which had ripple e ects on the city. According to K9, “youths from age 15 and above had started going to the city centre and other estates to rob peo- ple. Some had already bought weapons”. Local gangs with ties to local politicians acted as the de facto local government. The police or fellow residents killed sus- pected criminals for engaging in crime, resulting in frequent burials (K9). In the early 1990s, the KANU Youth Wingers continued to provide security:

Security was provided by ‘KANU youths’ in collaboration with the area chief. They did not charge for security but charged anyone building houses or busi- ness kiosks. The main challenge was that their children were also involved in crime, thus making it very di cult to be e ective (K9).

When Mungiki groups entered Korogocho at the turn of the millennium (Anderson 2002), the settlement already had high levels of crime. Therefore, some respondents considered the Mungiki as an asset, despite their violent nature:

They did not charge the community but charged public vehicles operating on routes. (...) The community sought their intervention in dispute resolution where they had set up an o ce. According to me, they were a good alternative security (K12).

Many respondents perceived that health resilience remained troublesome (Fig. 3), primarily due to an unhealthy living and working environment, as well as distant health services. Tuberculosis and HIV/AIDS were the main illnesses they were grappling with. “Some people believed that diseases like HIV/AIDS came with the

whites” (K14) or were a result of witchcraft (K15). Increased unsafety caused by gangs was somewhat counterbalanced by a more active role of NGOs, CBOs, and the local Catholic Church, which started youth empowerment programmes (Achola and Ndung’u 2006). Livelihood resilience also remained troublesome according to most respondents (Fig. 3). Many households survived by working on the land ll, engaging in drug dealing, and sex work. Over time, Mungiki and other politically aligned groups became more violent and criminal. K1 expresses the dual role of gangs:

They (Mungiki) o ered income-generating activities for the youths but also promoted crime (...). This has led to many youths being killed by the police.

Upgrading (2000s)

In the new millennium, Korogocho experienced a signi cant increase in popula- tion, becoming a sprawling settlement hosting over 150,000 residents (Pamoja Trust 2001). It emerged as the fourth largest informal settlement in Nairobi. From 2003 to 2006, Mungiki groups gradually lost power in tense and violent con icts involv- ing the police, other gangs, and residents, although some groups remained active at the land ll even after the post-election violence of 2007/2008 (Bennett et al. 2015). During this period, Father Alex, an Italian priest from the Catholic church in Korogocho, negotiated an upgrading programme with the Italian Government under the Jubilee global campaign on debt cancellation (UN Habitat 2010). As part of the agreement, the Italian government cancelled a portion of Kenya’s debt in exchange for the reallocation of funds to the upgrading program. The Korogocho Slum Upgrading Programme (KSUP), with a budget of Ksh210 million (approxi- mately 1.6 million euros), commenced in 2008 by conducting a renumeration of the settlement. The programme was governed by a Residents Committee (RC), com- prising 6 elected representatives from each of the eight villages in Korogocho, the area councillor and the area chief. It was further organized into sub-committees that incorporated di erent interest groups, such as women, youth, and elders. The RC was responsible for community dialogues and communication, while the Catholic church acted as the trustee (K1). The physical infrastructure development included sanitation facilities, a road bridge, streets, street lighting, water supply, housing, and a police station (King’ori, 2014).

However, con icts arose during the implementation of KSUP, particularly between landlords (often also village elders) and tenants. Many landowners opposed the programme out of fear of losing their income. KSUP proposed sharing plot own- ership among four houses with tarmacked throughways, which would reduce tenancy and facilitate police surveillance of illegal industries (UN Habitat 2012). Tensions escalated during the construction of the ring road, resulting in the displacement of residents who were on the list for a house (Burugu 2015). The housing construction itself was also marred by con icts. KSUP faced an impasse between older residents (who were renumerated in 2008) and residents who moved in after 2008 but also claimed a plot and housing. Consequently, implementation remains incomplete due to con icting interests (Burugu 2015; King’ori, 2014). The upgrading programme

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

a ected all residents, even those not directly involved, providing opportunities for some while causing resistance from others. The upgrading scheme became a politi- cal tool through which some local groups connected with urban institutions, includ- ing the county government and political parties, while others remained isolated.

Despite these challenges, most respondents perceived a signi cant improvement in resilience (Fig. 3; UN Habitat 2012). Safety risks were reduced due to the pres- ence of a new police station within the area (K2). Additionally, the advent of mobile phones in the 2000s contributed to a drop in crime rates as it became easier to report incidents and contact security agents (K9). Government programmes such as com- munity policing and youth employment initiatives also enhanced security (K10; K19). However, while safety improved for many, crime and unsafety remained major concerns for some households. Vertigans et al. (2021) also observed an increase in crime rates during Covid-19 restrictions.

Perceived health resilience improved according to most respondents, attributed to the establishment of new health centres and mobile clinics (K17), sanitation facili- ties (K15), and the activation of community health volunteers (K2). “Nowadays, there are many organizations like Umande Trust, Pamoja Trust, and Action Aid that have established public toilets and sanitation blocks”, narrated K1. Awareness of diseases was also perceived to have improved. Especially during Covid-19, residents made e orts to keep their environment clean through community garbage collection (K1, K15), hand washing, mask-wearing, and maintaining social distance. These measures were also believed to have reduced the number of water-borne diseases like cholera. However, health risks and overall development varied between house- holds and neighbourhoods. In recent years, some schools and health clinics have been established in Korogocho A. Additionally, there has been investment in infra- structure, such as roads and water pipes, leading to a more development compared to Korogocho B, largely driven by collective e orts from residents. However, respond- ents noted that development in Korogocho A was still slow, and the settlement remains largely informal (K15, 16, 17, 19). Moreover, access to water improved on average, and Grogon is perceived to have a better water supply than any of the other villages (K40).

KSUP supported adaptive resilience initiatives by CBOs, which improved health and livelihood resilience. For example, a youth group initiated a plastic light bulb project in 2010, using bleach and soda bottles to provide light to houses without electricity, thus reducing costs (Kiti 2012). Several CBOs collaborated with an NGO to construct six biogas toilets and community kitchens between 2008 and 2010, o ering sanitation and a ordable gas to surrounding households. However, maladaptive initiatives such as illegal electricity connections continued to persist. “Nowadays, a lot of people have acquired electricity through illegal means, and less than 30 people have it through legal means” (K2). While illegal connections provide local jobs and are more cost-e ective (K22), they also increase the risk of re.

According to the perception of most respondents, livelihoods on average remained at a similar level, despite temporary labour-intensive and community-based con- struction projects (King’ori, 2014), microcredit schemes, the reuse of materials, and the emergence of CBOs from Mathare, such as the social justice movement, as well as national programmes like the youth fund and youth employment programmes.

However, corruption was reported to hinder the e ectiveness of government pro- grammes. Most available jobs were casual, and with Covid-19 restrictions, many struggled to nd employment.

A lot of people in the area are casual laborers who depend on jobs such as laundry, home management, among others in areas such as Eastleigh, and with the pandemic, people are not allowing others into their homes due to fear of spreading the virus (K4).

K20 and K22 noted that poverty remained prevalent, reinforced by uncertain, unsafe, and unhealthy employment at the land ll, limited job opportunities, extrajudicial killings, and drug abuse.

Debate

The previous sections have applied the concept of resilience pathways to informal settlements, highlighting long-term trends of general resilience. We nd that resil- ience pathways of informal settlements in uence and shape resilience dynamics within these communities, with varying impacts across di erent spatial and tempo- ral contexts. While resilience pathways describe resilience at the meso-level, they do not negate the potential for individual improvements within a declining path- way or suggest universal adaptation within an adaptive pathway. The generic trends observed at the meso-level cannot fully capture the internal diversity and institu- tional complexities within informal settlements. Rather, these trends serve to visual- ize long-term meso-level patterns that are dynamically linked to higher and lower levels. It is important for meso-level interventions, such as upgrading programmes, to recognize and consider the time- and space-speci c resilience pathways of infor- mal settlements. This understanding can inform more e ective and contextually rel- evant interventions.

Figure 4 illustrates the perceived resilience pathways of the two settlements. In the 1970s and 1980s, a state of stasis is observed, where resilience gradually changes and varies among the two settlements and local actors while remaining at a relatively simi- lar overall level. Resilience is perceived to have declined in the 1990s, characterized by

J. Fransen et al.

Stasis

Decline

Transforma on

Adap ve: self-organised

Transforma on

Mathare

Korogocho

Adap ve: coordinated

1970’s and ‘80s 1990’s

2000+

Fig. 4 Resilience pathways in two informal settlements in Nairobi. Source Authors

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

the withdrawal of government support and the gradual takeover by gangs. Additionally, higher population densities reduced opportunities for urban agriculture. This decline in resilience aligns with nationwide economic recessions that followed the “lost decade” of 1980–1990 (Rono 2002, p. 82). Similar declining pathways have been reported in the literature, attributed to factors such as violence (Bennett et al. 2015), natural disas- ters (Haokip 2018), and misguided policies (Folke 2006).

Both settlements entered a distinct period of transformation from around 2005 to 2008, triggered by violent con icts between gangs, residents, corrupt police guards, and other actors, as the government sought to regain control. These ndings support the notion that resilience pathways undergo changes at speci c moments when the existing con guration is no longer sustainable and the institutional context shifts (Folke 2006; Seeliger and Turok 2014).

Following the transformation, both settlements appear to have entered a period of adaptive resilience. These ndings con rm that adaptive resilience pathways emerge when actors learn from uncertainty and gradually adjust and improve resilience initia- tives. However, inhibiting institutional contexts are unlikely to be e ectively addressed (Adger et al. 2009; Folke 2006). In both settlements, changes in governance assem- blages are observed, with new actors emerging and interacting with broader govern- ance structures in various ways. This results in a wide variety of governance assem- blages, often lacking e ective coordination.

The theory section introduced the concept of frugality, suggesting that adaptable resilience initiatives in informal settlements tend to be frugal. The ndings indicate that adaptability and frugality of resilience initiatives are often associated. Most resilience initiatives are characterized by their low-cost and basic functionality. When external actors initiate initiatives that are not frugal, such as road tarmacking and widening, they may con ict with local resilience initiatives and interests. We con rm that maladaptive resilience initiatives lack sustainability (Elmqvist et al. 2019) and the robustness associ- ated with frugality (Tiwari et al. 2014), such as for illegal electricity connections and improper waste disposal.

On the other hand, we nd that adaptive resilience initiatives o er sustainable, low-cost, and robust quick xes. Examples include plastic bulb lights, community health volunteers providing basic health services, microcredit programmes, merry-go- arounds, and community service delivery. The fumigation campaign during the cholera outbreak serves as an example of an adaptive resilience initiative that may not necessar- ily be frugal. The upgrading programme in Korogocho potentially represents a trans- formative resilience initiative, although its potential has not been fully realized (Burugu 2015; King’ori 2014). Therefore, the ndings con rm that resilience activities alone are unlikely to transform institutional and vulnerability contexts, but they can enable informal settlers to cope with uncertainty (Bahadur and Doczi 2016).

Conclusions and Recommendations

The application of resilience literature to the context of informal settlements is com- plex and multifaceted. In our study, we de ne resilience as the everyday struggle to survive in an uncertain environment, shaped by settlement practices and broader

institutional frameworks. Resilient communities actively confront multiple chal- lenges and continuously learn to cope with disruptive shocks and disasters. Our nd- ings highlight distinct variations in perceived resilience over time and space, both within and between informal settlements. Speci cally, we identify periods of sta- sis, decline, transformation, and adaptation, each characterized by di erent dynam- ics between the institutional contexts, vulnerabilities, and resilience initiatives. The shifting responsibilities and roles of government and gangs strongly in uence resil- ience pathways in Mathare and Korogocho.

We also observe that actors adopt both maladaptive and adaptive resilience initia- tives. Hereby adaptability is often associated with frugality, encompassing low-cost and robust quick xes. Maladaptive resilience initiatives, although providing short- term survival, tend to be fragile, unsustainable, and may perpetuate poverty traps (Fransen et al. 2023). In contrast, adaptive resilience initiatives are frugal, sustain- able, and do not impose additional risks on others. Transformative resilience initia- tives are relatively rare, and the upgrading programme in Korogocho, while aiming for transformation, remains contested. Our study con rms that resilience initiatives, including frugal ones, are unlikely to fundamentally alter inhibiting broader institu- tional frameworks (Adger et al., 209; Bahadur and Doczi 2016).

Based on our conclusions, we provide several policy recommendations. We acknowledge and support community initiatives such as the Mathare Spatial Plan- ning Area Research Collaborative (MSPARC), which advocates for settlement upgrading and partnerships with local government, universities, and other actors. Recognizing that contemporary resilience pathways often fall short in addressing institutional contexts and vulnerabilities, we encourage a focus on “soft” infrastruc- tures, including employment services, policing, microcredit, education, and health services. These areas have the potential to signi cantly impact resilience with lower initial investments. Furthermore, we recommend stronger coordination among local government, CBOs, NGOs, and other local actors to support adaptive and trans- formative resilience initiatives while preventing maladaptive ones. Visualizing and supporting local resilience initiatives that are frugal, sustainable, and do not harm others is crucial.

The main contribution of our study lies in the conceptualization of resilience pathways in informal settlements. We suggest future research delving deeper and broader into this topic. Case study research can map the variety of speci ed resil- ience, analyse the interplay between di erent levels of resilience pathways, and explore the triggers of transformative resilience pathways. Comparative research across informal settlements can further investigate resilience pathways and general- ize our ndings. Lastly, we advocate for community-based research as it provides in- depth insights into the local context, empowers community researchers, and ensures data remain with the community.

Acknowledgements We acknowledge funding support from the Vital Cities and Citizens initiative of Erasmus University and the International Centre of Frugal Innovations (ICFI). This study would not have been possible without the partnership with Ghetto Foundation and ICFI. We thank the reviewers, mem- bers of EADI 2021, Peter Knorringa, and Joop de Wit for feedback on the paper.

Data availability Data is available upon request.

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

Declarations

Con ict of interest On behalf of all authors, the corresponding author states that there is no con ict of interest.

References

Achola, H.O., and J.M. Ndung’u. 2006. Koch life: Community sports in the slum. Nairobi: Paulines Pub- lications Africa.

Adger, W., S. Dessai, M. Goulden, M. Hulme, I. Lorenzoni, D. Nelson, L. Naess, J. Wolf, and A. Wre- ford. 2009. Are there social limits to adaptation to climate change? Climatic Change 93: 335–354.

Anderson, D. 2002. Vigilantes, violence and the politics of public order in Kenya. African A airs 101 (405): 531–555.

Andvig, J.C., and T. Barasa. 2014. A political economy of slum spaces: Mathare Valley. NUPI Working Paper 838. Oslo: Norwegian Institute of International A airs.

Bahadur, A., and J. Doczi. 2016. Unlocking resilience through autonomous innovation. London: Overseas Development Institute.

Bennett, R., G. Chepngeno-Langat, M. Evandrou, and J. Falkingham. 2015. Resilience in the face of post- election violence in Kenya: The mediating role of social networks on wellbeing among older people in the Korogocho informal settlement, Nairobi. Social Science & Medicine 128: 159–167.

Blomkvist, P., D. Nilsson, B. Juma, and L. Sitoki. 2020. Bridging the critical interface: Ambidextrous innovation for water provision in Nairobi’s informal settlements. Technology in Society 60: 10122.

Brandsen, T. 2016. Governments and self-organization: A Hedgehog’s dilemma. In Critical re ections on interactive governance. Self-organization and participation in public governance, ed. J. Edelenbos and I. van Meerkerk, 337–351. Cheltenham: Edward Elgar Publishers.

Burugu, D. 2015. Con ict in slum upgrading programme. A case study of Korogocho slum in Nairobi. The East African Review 151 (1): 10–17.

Chiuri, W. 1978. Mathare valley: A squatter settlement in Nairobi: A case study. Nairobi: Unicef. Elmqvist, T., E. Andersson, N. Frantzeskaki, T. McPhearson, P. Olsson, O. Gafney, and C. Folke. 2019. Sustainability and resilience for transformation in the urban century. Nature Sustainability 2 (4):

267–273.Folke, C. 2006. Resilience: The emergence of a perspective for social–ecological systems analyses.

Global Environmental Change 16: 253–267.Folke, C., S.R. Carpenter, B. Walker, M. Sche er, T. Chapin, and J. Rockström. 2010. Resilience think-

ing: Integrating resilience, adaptability and transformability. Ecology and Society 15 (4): 20. Fransen, J., B. Hati, R. Nyumba, and E. van Tuijl. 2023. Community vitality and frugal practices in infor-

mal settlements in Nairobi: Towards a typology. Cities 134 (January): 104179.Fransen, J., D.O. Peralta, F. Vanelli, J. Edelenbos, and B.C. Olvera. 2022. The emergence of urban com- munity resilience initiatives during the COVID-19 pandemic: An international exploratory study.

The European Journal of Development Research 34 (1): 432–454.Give Directly. 2021. GD Live Newsfeed. https://live.givedirectly.org/. Accessed 30 Nov 2021.Haokip, T. 2018. Role of CBOs in resilience building: Good practices and challenges. In Development

and disaster management, ed. A. Singh, M. Punia, N. Haran, and T. Singh, 281–299. Singapore:

Palgrave Macmillan.Jabeen, H. 2019. Gendered space and climate resilience in informal settlements in Khulna City. Bangla-

desh. Environment and Urbanization 31 (1): 115–138.Kaika, M. 2017. “Don’t call me resilient again!”: The New Urban Agenda as immunology... or... what

happens when communities refuse to be vaccinated with ‘smart cities’ and indicators. Environment

and Urbanization 29 (1): 89–102.Kigochie, P. W. (2001) Squatter rehabilitation projects that support home-based enterprises create jobs

and housing. The Case of Mathare 4A, Nairobi. Cities, Vol. 18, No. 4: 223–233.King’ori, A.N. 2014. In uence of community participation in completion of development projects: a case

of Korogocho slums, Nairobi County. Doctoral dissertation, Kenya.Kiti. 2012. Water Light Project—Korogocho slums. https://www.youtube.com/watch?v=OfaO9d_O7Mg.

Accessed 16 Dec 2022.

KNBS. 2019. Kenya National Bureau of Statistics. 2019 Kenya Population and Housing Census, vol. II. Nairobi: KNBS.

KNCHR. 2008a. ‘The Cry of Blood’ Report on extra-judicial killings and disappearances. Nairobi: KNCHR.

KNCHR. 2008b. On the brink of the precipice: A human rights account of Kenya’s Post 2007 elections violence. Nairobi: KNCHR.

Mahendra, A., R. King, J. Du, A. Dasgupta, V.A. Beard, A. Kallergis, and K. Schalch. 2021. Seven trans- formations for more equitable and sustainable cities. Washington DC: World Resource Institute. https://doi.org/10.46830/wrirpt.19.00124.

Médard, C. 2010. City planning in Nairobi: The stakes, the people, the sidetracking. In Nairobi today: The paradox of a fragmented city, ed. Hélène. Charton-Bigot and Deyssi Rodriguez-Torres, 25–60. Nairobi: IFRA.

Meerow, S., and J.P. Newell. 2019. Urban resilience for whom, what, when, where, and why? Urban Geography 40 (3): 309–329.

Mutahi, P. 2011. Between Illegality and Legality. (In)security, crime and gangs in Nairobi informal settle- ments. SA Crime Quarterly 37: 11–18.

Nykvist, B., and J. Von Heland. 2014. Social-ecological memory as a source of general and speci ed resilience. Ecology and Society 19 (2): 47.

Pamoja Trust. 2001. Korogocho informal settlements enumeration report, July, 2001. Nairobi: Pamoja Trust. https://www.globalhungerindex.org/case-studies/2009-kenya.html#fn-03-ref.

Pelling, M., D. Manuel-Navarrete, and M. Redclift. 2012. Climate change and the crisis of capitalism: A chance to reclaim, self, society and nature. Hoboken: Taylor and Francis. https://doi.org/10.4324/ 9780203146118.

Pieterse, E. 2008. City futures: Confronting the crisis of urban development. London: SAGE.Radjou, N., J. Prabhu, and S. Ahuja. 2012. Jugaad innovation: Think frugal, be exible, generate break-

through growth. London: Wiley.Rasmussen, J. 2010. Mungiki as youth movement: Revolution, gender and generational politics in Nai-

robi, Kenya. Young 18 (3): 301–319.Revi, A., D.E. Satterthwaite, F. Aragón-Durand, J. Corfee-Morlot, R.B.R. Kiunsi, M. Pelling, D.C. Rob-

erts, and W. Solecki. 2014. Urban areas. In Climate change 2014: Impacts, adaptation, and vulner- ability. Part A: Global and sectoral aspects. Contribution of working group II to the fth assessment report of the Intergovernmental Panel on Climate Change, ed. C.B. Field, V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White, 535–612. Cam- bridge: Cambridge University Press.

Rockefeller Foundation and ARUP. 2019. City resilience index. Understanding and measuring city resilience. https://www.arup.com/perspectives/publications/research/section/city-resilience-index. Accessed 8 Apr 2020.

Rojas Symmes, L., A. Cortés Salinas, and D. Moreno. 2023. Santiago, the non-city? Destruction, crea- tion, and precariousness of verticalized space. In Urbicide: The death of the city, 865–890. Cham: Springer.

Rono, J.K. 2002. The impact of the structural adjustment programmes on Kenyan society. Journal of Social Development in Africa 17 (1): 81–98.

Satterthwaite, D., D. Archer, S. Colenbrander, D. Dodman, J. Hardoy, D. Mitlin, and S. Patel. 2020. Building resilience to climate change in informal settlements. One Earth 2 (2): 143–156.

Satterthwaite, D., and D. Dodman. 2013. Towards resilience and transformation for cities within a nite planet. Environment and Urbanization 25 (2): 291–298.

SDI-K. 2022. Mathare Settlements: Pro ling Data Findings: ‘Making the invisible visible’. https://static1. squarespace.com/static/58d4504db8a79b27eb388c91/t/63849962a69e574c786e9a9c/1669634487 063/Mathare+Pro ling+Report_2022+Final.pdf.

Seawright, J., and J. Gerring. 2008. Case selection techniques in case study research. Political Research Quarterly 61 (2): 294–308. https://doi.org/10.1177/1065912907313077.

Seeliger, L., and I. Turok. 2014. Averting a downward spiral: Building resilience in informal urban settle- ments through adaptive governance. Environment and Urbanization 26 (1): 184–199.

Simone, A. 2008. People as infrastructure. In Johannesburg: The elusive metropolis, ed. S. Nuttall and A. Mbembe, 68–90. New York: Duke University Press.

Suhartini, N., and P. Jones. 2019. Urban governance and informal settlements. Cham: Springer.

J. Fransen et al.

Resilience Pathways of Informal Settlements in Nairobi: Stasis,...

Teisman, G.R. 2000. Models for research into decision-making processes phases, streams and decision- making rounds. Public Administration 78 (4): 937–956.

Tiwari, R., K. Kalogerakis, and C. Herstatt. 2014. Frugal innovation and analogies: some propositions for product development in emerging economies. Working paper. Technologie und Innovationsmanage- ment, Technische Universität Hamburg, Hamburg.

UN Habitat. 2010. Korogocho situation analysis. A snapshot. Nairobi: UN Habitat. https://habnet.unhab itat.org/sites/default/ les/documents/Final%20Report%20Korogocho%20S%20A%20a%20snapshot. pdf. Accessed 16 Dec 2022.

UN Habitat. 2012. Korogocho streetscapes: Documenting the role and potentials of streets in citywide slum upgrading. Nairobi: UN Habitat.

UN Habitat. 2020. RISE-UP: Resilient settlements for the urban poor, 1–4. Nairobi: UN Habitat. https:// unhabitat.org/programme/rise-up-resilient-settlements-for-the-urban-poor.

Van Stapele, N. 2014. Intersubjectivity, self-re exivity and agency: Narrating about ‘self’ and ‘other’ in feminist research. Women’s Studies International Forum 43: 13–21.

Van Stapele, N. 2015. Respectable ‘illegality’: Gangs, masculinities and belonging in a Nairobi ghetto. Doctoral dissertation, Universiteit van Amsterdam [Host].

Van Stapele, N. 2020. Police killings and the vicissitudes of borders and bounding orders in Mathare, Nairobi. Environment and Planning d: Society and Space 38 (3): 417–435.

Vasudevan, A. 2015. The makeshift city: Towards a global geography of squatting. Progress in Human Geography 39 (3): 338–359.

Vertigans, S., N. Mueller-Hirth, and F. Okinda. 2021. Resilience in a Kenyan informal settlement during the COVID-19 pandemic. Acta Academica 53 (2): 122–142.

Yin, R.K. 2009. Case study research: Design and methods, vol. 5. London: SAGE.Yonzan, N., A. Cojocaru, C. Lakner, D.G. Mahler, and A. Narayan. 2022. The impact of COVID-19 on

poverty and inequality: Evidence from phone surveys. Washington: The World Bank Blogs. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps

and institutional a liations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Authors and A liations

Jan Fransen1 · Beatrice Hati2,3 · Naomi van Stapele4 · Samuel Kiriro5 · Rosebella Nyumba3

* Jan Fransen fransen@ihs.nl

Beatrice Hati gitundu@iss.nl

Naomi van Stapele N.vanStapele@hhs.nl

Samuel Kiriro management@ghettofoundationkenya.org

Rosebella Nyumba rnyumba@c a.network

1Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3062PA Rotterdam, The Netherlands

Institute for Housing and Urban Development Studies (IHS), Vital Cities and Citizens (VCC),

2The Netherlands

International Institute of Social Studies (ISS), Erasmus University Rotterdam, Den Haag,

3Floor, Ring Road Westland’s Lane, Nairobi, Kenya

International Centre for Frugal Innovation (ICFI), Kenya Hub, Centenary House, Block B, 1

4of Applied Sciences, Den Haag, The Netherlands

5

Present Address: Centre of Expertise ‘Global and Inclusive Learning’, The Hague University Ghetto Foundation, Juja Road, Macharia Building, Nairobi, Kenya

J. Fransen et al.