Analysis of Residents’ Livelihoods in Transformed Shantytowns: A Case Study of a Resource-Based City in China

Abstract

Research on shantytown transformation in the context of building sustainable human settlements has tended to concentrate on macro and objective factors. However, there is still room for expanding research on the livelihoods of residents in transformed poor communities of resource-based cities. This study is based on household survey data after the coal mine shantytown transformation in Datong, China, and uses the entropy method and Logit regression model to analyze the livelihood level and impact of livelihood capital on the choice of livelihood strategies of the residents after shantytown transformation. Based on the development characteristics of the Datong coal mining community, the traditional sustainable livelihood analysis framework is improved, constructing a livelihood capital evaluation system including natural capital, physical capital, financial capital, social capital, human capital and cultural capital. The research indicates that the livelihoods of the residents after shantytown transformation are in a state of low-value aggregation and differentiation, their livelihood strategies are still dependent on coal mining, financial capital and cultural capital have significant positive impacts on the residents’ livelihood strategies of choosing coal mine-related industries, and males and elderly people have a greater likelihood of choosing coal mine-related industries. This study provides micro-level explanations for the livelihood status and livelihood strategy choices of residents after shantytown transformation in a resource-based city and provides policy enlightenment for local governments on how to promote the sustainable development of residents in coal mining communities.

1. Introduction

Poverty is an intractable disease for human society, and urban poverty in developing countries is concentrated in slums [1]. Slum dwellers face harsh living conditions, fragile livelihoods, vulnerability to disease, and exclusion from formal governance structures [1,2]. The world endeavors continuously to improve the quality of life of slum dwellers. Statistics show that the lives of 320 million slum residents were improved from 2000 to 2014 [3]. Building sustainable human habitat communities has long been one of the major development challenges for human societies [4]. Improving housing is one of the main ways that countries build sustainable human communities. However, in many developing countries, while the housing conditions of poor people have been improved, the livelihoods of inhabitants may still be at risk, and concerns have been raised about how to prevent upgraded neighborhoods from declining into poor settlements again after a number of years [5]. Poverty alleviation is the ultimate goal of housing improvement and relocating residents [6]. The urban shantytown transformation project is an important part of China’s anti-poverty process.

In recent years, along with the process of energy conservation and emission reduction worldwide, just transition during economic transformation in resource-based regions has increasingly become a hotly debated topic in the international community [7,8,9,10,11,12]. Poor people in resource-based areas, who have long depended on local resources for survival, will suffer a severe impact on their livelihoods once they are confronted with resource depletion and a broader policy environment that discourages the exploitation of natural resources [12,13].

Hence, it is significant to assess the livelihoods of slum dwellers in resource-based areas after slum rehabilitation. In 2023, the UN-Habitat proposed a higher sustainable development goal that advocates inclusive, safe, resilient and sustainable cities and human settlements [4]. This means that transforming poor settlements is a long-term enterprise and that the sustainability of residents’ livelihoods is a primary issue for the sustainability of the community and the city.

The literature on the transformation of poor settlements indicates that transformation is an intervention to improve the living quality of poor residents [14] and that improving the electricity supply, roads, sanitation and other public services in settlements are the main components of shantytown transformation [15,16]. Specific approaches to shantytown transformation mainly include in situ upgrading and resettlement [17,18] and are conducted through full intervention by national governments, cooperation between international development agencies and governments, and the participation of civil society organizations [19,20,21]. The object of this study is the shantytown residents living in a resource-based city who have realized an improvement in their housing conditions by means of relocation under the complete intervention of the government.

Examining the relevant studies reveals that research on slum upgrading in developing countries has concentrated on assessing the course and outcomes of upgrading policies at the macro level [22,23,24,25,26], while only a few studies have explored the life economic impacts of slum upgrading on the affected residents at the micro-level [27,28,29]. Nevertheless, it has to be admitted that analyses at the macro-policy level may completely fail to capture the impact of specific slum-upgrading projects on the real lives of residents [30]. Furthermore, in recent years, although scholars have been emphasizing the livelihood problems of residents in resource-based areas undergoing economic transformation [10,11,12,31,32,33], and most of the current studies remain at the theoretical level, concrete empirical research is lacking.

Therefore, we examined the livelihoods of residents who live in transformed shantytowns in a representative resource-based city in China as an example. This study aimed to clear up two questions: [i] What is the livelihood status of residents resettled after shantytown transformation in the study area? [ii] How do changes in livelihood capital affect the choice of livelihood strategies of local residents in the community? In this study, a questionnaire survey was conducted in July 2023 to collect primary data, which focused on the current livelihood status of the residents of the study area almost ten years after the completion of the shantytown improvement project. The expanded sustainable livelihoods framework (SLF) was applied in the analysis. The entropy method was used to measure the value of the livelihood capital of the surveyed residents, and the Logit model was employed to analyze the impact of the livelihood capital possessed by the residents on their livelihood strategies.

The marginal contributions of this paper are mainly in the following areas. First, this study focuses on the level of residents’ livelihood capital and its impact on livelihood strategies after shantytown transformation in a resource-based city of China. On the one hand, it expands research on the impact of transforming urban poor settlements at the micro-level, and on the other hand, it enriches research on the livelihoods of residents in resource-based areas in the context of green and low-carbon development. Second, this study enriches the application of the sustainable livelihood theoretical framework. Adding cultural capital to the framework helps to understand the livelihood strategy choices of people living in coal mining resource-dependent areas in a more nuanced way. Third, this empirical study reveals, through a bottom–up approach, which livelihood capitals of residents in resettled poor neighborhoods should be attended to, and the policy implications of the study can guide low-income communities in resource-dependent cities to design more effective programs to help residents achieve a smooth transition and sustainable development.

The Section 2 provides a literature review targeting the perspectives of shantytowns and their transformation and concepts related to livelihood capital. Section 3 describes the study area, the survey data obtained, and methodology. Section 4 describes and analyzes the results. Section 5 provides a discussion, and Section 6 concludes with conclusions and policy implications.

2. Literature Review

2.1. Shantytown Transformation

Shantytown is a manifestation of slums. Researchers usually use the five types of housing deprivation proposed by UN-Habitat as a starting point for discussing slums, where any given dwelling or household lacking any one or more of [i] improved water, [ii] improved sanitation, [iii] safety and security, [iv] living space, and [v] durable housing structures can be recognized as a slum [34]. This work is likewise based on this definition in selecting the study area. However, in specific countries or regions, the historical causes of slums vary considerably.

The most dominant view in the academic community is that the impetus for the emergence of slums stems from the massive concentration of people brought about by industrialization and urbanization. It can be traced back to the 19th century in the United Kingdom, where slums and the ‘factory of the world’ emerged at the same time during the Industrial Revolution [35]. Martinez Veig [22] argued that political–economic processes were the root cause of poverty and inequality among slum populations. Behind the proliferation and dislocation of slums are systemic inequities regarding the distribution of the benefits of urbanization and the social reproduction of labor [21].

In China, the concept associated with slums is ‘shantytown’. China’s shantytowns are mainly manifested in large areas that were formed by the concentration and reproduction of urban industrial workers during the period of unsound housing systems in the last century [36]. This paper focuses on the residents of a resource-based city in China who have undergone shantytown transformation, formerly living in shanties next to coal mines, in areas where their housing conditions meet the UN-Habitat definition of a slum.

Urban shantytown transformation is one of the approaches to solving urban problems and an opportunity for the sustainable development of a city and its poor inhabitants. Many governments make slum transformation an important priority and adopt different approaches to improve the living conditions of slum dwellers [24]. The dominant international perception of slum governance has moved from clearance to upgrading, from exclusion and expulsion, to recognition and acceptance [14]. Particularly, the impact of post-World War II urban regeneration, led by the UK and the USA, was enormous, with large-scale reconstruction of housing and public infrastructure to safeguard the housing needs of the citizens and to restore order to cities [22,37]. Subsequently, countries such as Brazil and India have followed the example of developed countries [25]; however, the results have not met policy expectations [38]. In the case of Mumbai, India, for example, where the vision of slum clearance was repeatedly proposed in the late twentieth century, limited resources and political interference slowed down the progress of slum redevelopment programs [38]. Similarly, Rio de Janeiro’s slum clearance operations in the 1960s and 1970s ended in failure [39]. The critical scholarship that emerged after this period also demonstrated that mainstream thoughts such as slum clearance were non-humanitarian and ineffective [21]. As a result, the authorities had to shift their mindset on governance from eradication to upgrading. UN-Habitat and the Government of Kenya have partnered to upgrade Kibera in Nairobi, one of the largest slums in Africa, to improve housing infrastructure and raise incomes [17], but the results have been less than satisfactory [27]. Fernandez and Calas [27] point out that upgraded housing does not truly benefit slum dwellers because they cannot afford to pay the rent and the structure of the housing is not conducive to normal economic activity, which deprives them of their livelihood capital instead of accumulating it. It is worth noting that in the 1990s, slums were integrated into the formal city as ‘informal areas’, and governance thinking shifted from localized clean-ups to holistic urban integration [25,40]. However, even in doing so, there has been much controversy, and scholars argue that upgrading will only have limited improvement in the living conditions of the residents, strongly calling for more research on the impact of housing transformation on slum dwellers [27].

Studies on shantytown transformation have centered on policy evaluation and social participation in transformation programs [14,41,42], political operations [43], transformation models [44] and assessment of the impact of transformation, including livelihoods and social stability [20,28,45]. A study in Africa shows that how residents perceive the effects of shantytown transformation is influenced by economic, technological, and environmental factors, among which affordability and the sense of housing use are both important [20]. Borsuk observes that a shantytown transformation in Turkey is a process of deprivation accumulation which exacerbates the precariousness of survival for local women and makes their access to resources more limited [46]. It is known from Debnath et al.’s [28] study that the quality of housing in India’s slums after slum upgrading is low as well as poses health hazards to the residents, the higher electricity bills for the low-income groups living in the new houses is another form of exploitation. The situation is different in China, however, where research demonstrates that shantytown transformation significantly reduces the level of residential energy consumption and helps lift dwellers out of energy poverty [26]. We can learn that all these studies emphasize, either directly or laterally, that shantytown transformation programs should value the affected residents’ availability of physical, natural, human, financial, and social capital and reduce their vulnerability to risks to make their livelihoods sustainable.

2.2. Sustainable Livelihoods for Shantytown Residents

At present, with increasing knowledge, ‘low income’ is no longer the only indicator of poverty; poverty is the ‘capability’ to gain access to basic material survival opportunities [47]. The connotations of the sustainable livelihood capital framework fit aptly with this thought put forward by Amartya Sen. ‘Livelihoods’ is used extensively as a mature approach in various fields of academic research and investigation. The emergence of ‘sustainable livelihoods’ from an idea to a framework occurred in the 1990s, represented by a White Paper on World Poverty Elimination published by the UK Department for International Development (DFID) in 1997, and livelihoods became the focus of attention for many governments and NGOs [48]. The sustainable livelihoods framework (SLF) understands issues such as the resources people have and the impact on their livelihood strategies at a micro-scale, providing an action-oriented approach for governmental and nongovernmental organizations, emphasizing the holistic, humanistic, and local dimensions, which often aims to enable poor people to ‘use their initiative to achieve self-help’ [48]. The SLF is popular with researchers and policymakers and can serve to assess the consequences of development activities in regions on the livelihoods of poor people. Previously, the SLF has been mainly applied in studies related to rural farming households, whose livelihoods have been affected by various reasons, such as land policies [49], relocation and resettlement [50], and natural disasters [51]. But the SLF is gradually being used in urban contexts as well, and the sustainability and safety of urban dwellers also need to be valued, especially in cities in developing countries [29].

According to the framework developed by DFID, sustainable livelihoods can be quantified around five dimensions: natural capital, physical capital, financial capital, social capital, and human capital. As methods for assessing sustainable livelihoods are not harmonized across countries, researchers have adapted elements of this framework, which is highly inclusive and flexible, for practical application. Wu et al. [50] added cultural capital to the SLF in their study of farmers’ livelihoods in ethnic minority areas of Yunnan, China, where they argued that culture could also affect farmers’ production and consumption patterns. Information capital and psychological capital were included in the exploration of how climate change perceptions affect farmers’ ability for sustainable livelihoods by Guo et al. [52]. The SLF initially targeted rural areas across the globe, but with globalization and urbanization, it became geared towards urban use [53]. Continued development of specific areas of the city leads to sustainable urban development, which continues to improve the livability of the area; however, the direction of the impact of sustainable urban development on the level of economic development of the city is uncertain [54]. Therefore, in this study, it is meaningful to examine the livelihood levels and livelihood strategies of vulnerable groups after urban shantytown transformation.

There have been several studies that focus specifically on the livelihood vulnerability of disadvantaged groups, such as farmers, fisherfolk, coastal communities, pastoralists, and the urban poor who are prone to changes or risks to their livelihoods due to external factors, such as natural causes and political and economic environment [55,56,57,58]. It has been well researched and confirmed that among these identified vulnerable groups, smallholder farmers, pastoralists, fisherfolk and residents of coastal communities are highly dependent on natural resources and climate change for their livelihoods, in addition to unnatural factors such as policies, markets have a significant impact on their livelihoods [59,60,61,62]. In comparison, the livelihoods of urban slum dwellers are not just threatened by natural disasters but also by pressures related to land insecurity, spatial constraints, unemployment, crime, corruption and cultural exclusion [63]. Differing from the extent of attention given to farming households, the urban poor are often excluded from urban programs [64], and participants involved in responding to risk management tend to disregard vulnerable urban residents [62].

Previous studies have demonstrated that slum dwellers are more vulnerable and less resilient than other groups in the face of risks or disasters such as climate change, public health emergencies, food security, and socio-economic environments [2,65,66,67]. Akther’s research has shown that the livelihood status of slum dwellers drops to poverty level during periods of stress [58]. Ezeh et al. [2] conclude that there are neighborhood effects in slums that lead to more pronounced vulnerability to risk, and that neighborhood effects are categorized into four types, namely, the physical environment, social interactions, geographic factors and institutional factors. The increased risk of livelihood deprivation faced by slum dwellers is the consequence of the closely shared physical and social environment associated with the particular community space [2]. Overall, the livelihood situation of slum dwellers is characterized by a low capital base, weak livelihood strategies to overcome risks, and a high degree of informalization of livelihoods in terms of housing, employment, health and other basic amenities [29].

It was mentioned earlier that the transformation of urban shantytowns is an intervention that often involves resettlement [17,20]. The resettlement policies aimed at improving the housing conditions of the residents and preventing risks may be the best option for the residents [68]. In short, governments in developing countries remain convinced that transforming informal settlements is an important part of promoting sustainable urban development and improving the livelihoods of poor people [69]. Therefore, there is a need to focus on the livelihoods of people living in transformed urban poor settlements.

However, the literature on comprehensive assessments aimed at the livelihoods of shantytown residents after transformation in resource-based cities is still relatively scarce, especially in China. Due to the late start of shantytown reform in China, there are fewer studies on the livelihoods of these residents. Hence, in this study, we are concerned with residents of shantytowns whose livelihoods have been altered by resettlement while experiencing the risk of drastic changes in the social environment. The sustainable livelihoods of shantytown dwellers are defined as the enhancement of their livelihoods in risky environments created by a variety of factors, using the rights and resources available to them, including sufficient income, adequate space for housing that is affordable and safe, full access to public services, economic opportunities, etc. [4]. This means that it has three basic attributes, which are possessing capital (including physical, social, and natural capital), stable income sources, and having the ability to improve the level of livelihood (work experience, education, health, and psychological) [29]. In this study, we apply the improved SLF to analyze the livelihood of residents in China’s Datong coal mine shantytowns after transformation.

3. Study Area and Methodology

3.1. Study Area

The area selected for this study is Datong, Shanxi Province, China. Datong is extremely rich in coal resources and is the ‘coal capital’ of China. Large-scale coal mining created a concentration of shantytowns, which were built on the hills, covering a geographical area of about 7.32 million square meters with 230,000 residents. The local residents are all employees and family members of a coal mining group (see Figure 1), who built shanties on site adjacent to the coal mine to gather into living quarters.

In 2006, the Datong municipal government began to transform the coal mine shantytowns, which took eight years, and rebuilt a new community with an area of 9.57 square kilometers 10 km away from the city center to resettle the residents of the shantytowns in the mines, naming it ‘Hengan New District’ (see Figure 2). The Chinese word ‘Heng An’ means ‘safe forever’, which expresses people’s blessing for miners, as their work is extremely risky.

Before relocation, it was common for a family of six to live in a ten-square-meter shanty. After the transformation, the housing area of each household is kept at 45–80 square meters. The district can be considered a corporate community as the residents who work in the coal mining industry are in a state-owned conglomerate. As shown in Figure 3, it is a typical shantytown redevelopment community, where all the mine’s people are resettled to live in a completely new place, moving from shanties to concrete buildings.

3.2. Methodology

3.2.1. Data Acquisition

The data for this study were obtained from the questionnaire survey on livelihood capital of the residents of Hengan New District, which was conducted in July 2023 by the group. The questionnaire utilized the five-point Likert scale method to illustrate the indicators. Interviews and secondary sources as supplementary. The data collection was divided into two phases: the preliminary pre-survey and the formal survey. The team conducted a one-week pre-survey in June 2023 regarding the situation of the residents in Hengan New District, with the survey targeting some of the residents who were randomly visited. Preliminary knowledge of the overall overview of Hengan New District and the history of resettlement was established through the pre-survey, and the formal survey was begun after the questionnaire was revised and improved according to the actual situation. It is known from the pre-research and secondary sources that the residents surveyed belong to the whole relocation to the same district and the residents’ livelihoods are homogeneous, so the questionnaires were distributed by random sampling.

The formal survey was carried out through a field survey randomly distributing questionnaires to participants aged 18 years and above, who voluntarily received the questionnaires to fill them out. The survey distributed 250 questionnaires and collected 245 valid questionnaires, with a validity rate of 98%. The questionnaire consists of two main parts, the first of which is the demographic characteristics of the residents, including gender, age, marital status, family size, year of engagement in the workforce, household economic status, life satisfaction, and livelihood strategies. The second part contains the status of six types of livelihood capital with respect to the residents after their relocation: natural capital, human capital, physical capital, financial capital, social capital, and cultural capital. Statistics on the basic situation and income data of resettled families in the surveyed areas are as follows (see

Table 1. Characteristics of respondents.

Statistical Items	Variable Name and Assignment	Proportion
Gender	Female = 0	52.24%
	Male = 1	47.76%
Age	18–29 years old = 1	4.49%
	30–40 years old = 2	28.57%
	41–50 years old = 3	48.16%
	51–60 years old = 4	16.33%
	Over 61 years old = 5	2.45%
Household size	2 persons and fewer = 1	2.04%
	2 persons = 2	8.16%
	3 persons = 3	41.63%
	4 persons = 4	36.33%
	Over 5 persons = 5	11.84%
Level of education	Elementary and below = 1	11.43%
	Junior high school = 2	50.61%
	High school = 3	17.55%
	Technical/vocational high school = 4	6.12%
	Secondary/specialized/undergraduate = 5	13.88%
	Postgraduate = 6	0.41%
Working time	0–5 years = 1	37.96%
	6–10 years = 2	13.88%
	11–15 years = 3	18.78%
	16–20 years = 4	14.29%
	Over 21 years = 5	15.10%
Monthly per-capita household income	Less than RMB 1000 = 1	18.37%
	RMB 1000–1999 = 2	23.27%
	RMB 2000–2999 = 3	19.59%
	RMB 3000–3999 = 4	14.69%
	RMB 4000–4999 = 5	10.20%
	Over RMB 5000 = 6	13.88%

). All questionnaires were anonymous due to ethical requirements. Since the overall Cronbach’s alpha coefficient of the survey sample indicators is 0.739, the questionnaire questions passed the internal consistency test.

3.2.2. Improved Sustainable Livelihoods Framework

Culture as a capital can shape social class [70]. Cultural capital has the same nature of conversion to economic capital as human and social capital. According to Bourdieu [70], one of the connotations of cultural capital involves a person’s participation in and familiarity with a particular cultural activity in a society, and it can also be the ability to gain access to a certain cultural resource. A study regarding the immigrant earnings in the United States suggests that cultural capital does have the capacity to increase economic returns in the labor market [71]. In addition, culture is a major factor affecting the health and socio-economic integration of immigrants [72]. More importantly, existing studies on emotional attachment in coal mines have found that coal mining communities do have unique cultural and ‘regional memory’ [73,74,75].

Since this survey targets people living near coal mines, and the study area has a long history of ‘coal culture’, cultural capital is a factor that we have to take into account. For example, according to the field survey, we know that the winter solstice of the lunar calendar every year for the residents of the coal mine shantytowns is more solemn than China’s ‘Spring Festival’, all miners are on vacation this day, and the male miners collectively go to the temple to worship the ‘Yao Shen‘ (meaning the god of the coal mine), praying for peace in the coming year, so this day is also called ‘Yao Shen Festival’ (see Figure 4a). In addition, on the winter solstice and the Spring Festival, local residents used to ‘build booming fire’ with coal and ‘revolve around the booming fire’ to pray for good wishes, which are also a ritual that have survived to this day (see Figure 4b). The shantytowns in this area were created by ‘coal’, and the shantytown transformation was also caused by ‘coal’, so ‘coal’ is the link between the people and society. Where there is a history of coal mining, coal tends to maintain a consistent socio-economic identity in the perceptions of the local public, and it is a special existence for the people [9]. The Hengan New District is a typical resource-dependent community and the main responsibility for the shantytown transformation is borne by the state-owned enterprises in the area. Whether cultural capital has an impact on the livelihood strategies of dwellers is also a question of interest. Therefore, in conjunction with the field survey, this study adds cultural capital to SLF as a primary indicator to be measured. And we used ‘frequency of participation in festivals in their own and other communities’ and ‘knowledge of customs and traditions in their own community’ as indexes for evaluating the cultural capital of the surveyed residents.

3.2.3. Construction and Measurement of Livelihood Capital Evaluation Indicator System

This study constructed an improved livelihood capital evaluation system for Hengan New District residents based on the traditional SLF, with reference to the existing quantitative assessment results of livelihood capital and combining the production and life characteristics of the transformed shantytown residents (see Figure 5). The system is the basis for quantitative analysis of the livelihood capital of dwellers and facilitates further measurement of the stock of livelihood capital as well as livelihood strategies after resettlement.

Three indicators were used to measure human capital in this study: health status, level of education, and skills training. Education is recognized as the most fundamental dimension in measuring human capital, followed by studies showing that the health of the workforce and the skills index are also relatively important indicators of human capital [76,77]. In livelihood studies targeting farmers, land and ecosystems are important expressions of natural capital [78]. For residents of transformed urban coal mine shantytowns, natural capital is the ownership of their housing area and the ecological environment around the settlement where they live. Physical capital is a form of tangible capital, and the tools of production and living in a household, including consumer durables, are classified as physical capital [79]. This paper uses three indicators to portray physical capital: type of housing after resettlement, fixed assets, infrastructure, and public services. Financial capital has a pivotal role in the SLF, which can refer to regular cash receipts or access to different forms of credit lines [80]. We included annual household income and ease of access to borrowing in our measure for financial capital. Additionally, since shantytown transformation is a governmental act, the economic assistance received by the residents during relocation must also be taken into account.

Social capital exists in human relationships, where people can accumulate the resources they need in social networks, and developing individual social capital is an essential aspect for promoting sustainable livelihoods [81]. Therefore, participation in community organizations, connections with community residents, and relationships with acquaintances who hold certain resources were chosen to represent it. Cultural capital, a resource unique to the Hengan New District, was measured using the frequency of participation in festivals and celebrations in this community and elsewhere, as well as the degree of knowledge about the customs of the community.

The six dimensions of livelihood capital for this study were weighted, which in turn allowed for the calculation of a livelihood capital score for each resident surveyed to assess their sustainable livelihood level. The methods of quantitative empowerment of livelihood capital are mainly divided into subjective and mathematical empowerment methods [82]. In this paper, the entropy method is used to weight the livelihood capital indicators, which is a mathematical empowerment method and has been widely applied in the field of livelihood studies. In this paper, a comprehensive evaluation of the livelihood capital of residents after shantytown transformation, the process of calculating weights by the entropy method is as follows.

(1)

First of all, in order to eliminate the effect of the original data scale, this paper adopted the standardization of polar deviation to standardize the original data. The formula is as follows:

$${X^{\prime }}_{ij}=\frac{X_{ij}-min\left\{x_{ij},\cdots ,x_{nj}\right\}}{max\left\{x_{ij},\cdots ,x_{nj}\right\}-min\left\{x_{ij},\cdots ,x_{nj}\right\}}$$

(1)

${X^{\prime }}_{ij}$ is the value of the j-th index for the i-th sample ($i=\mathrm{1,2},\cdots ,n;j=\mathrm{1,2},\cdots ,m$). For convenience, the normalized data are still denoted as $X_{ij}$.

(2)

Then, the standardized livelihood capital indexes of individual residents were weighted. The formula for its calculation is:

$$W_j=\frac{1-e_j}{{\sum }_{i=1}^m\left(1-e_j\right)}$$

(2)

Among them, $e_j=-K{\sum }_{i=1}^m\left[\left({X^{\prime }}_{ij}/{\sum }_{i=1}^m{X^{\prime }}_{ij}\right)\times \ln \left({X^{\prime }}_{ij}/{\sum }_{i=1}^m{X^{\prime }}_{ij}\right)\right],$ $K=\frac{1}{\ln \left(m\right)}$.

In the above formula, $W_j$ is the weight of index j, $e_j$ denotes the information entropy of the j-th index, K is a constant, and m is the total number of samples evaluated.

(3)

Lastly, the livelihood capital values of all the surveyed residents were summed up based on the corresponding weights and sequentially calculated.

The comprehensive evaluation model of the level of sustainable livelihoods for residents is as follows:

$$C_i={\sum }_{j=1}^n\left(W_j\times X_{ij}\right)$$

(3)

In the formula, n is the total number of indexes, and $C_i$ denotes the composite evaluation value of the subject under study.

3.2.4. Model Construction

Livelihood capital can be regarded as a concentrated representation of the material wealth in households, and the resource endowment of residents is closely related to the choice of livelihood strategies. Proper strategies can enhance the sustainability of livelihood capital. Seeking and developing policies that create and support livelihoods are a priority for policy makers, and investment in all types of capital contributes to improving the livelihoods of poor people [83]. This emphasizes the important impact of livelihood capital on strategy choices. The result of descriptive statistical analysis shows that the livelihood strategies of residents after shantytown transformation, as measured by the main income sources of households, mainly consist of working in coal mines and related fields, and working in non-coal-mining industries, accounting for 38.37% and 61.63%, respectively. In this study, residents’ livelihood strategy is the dependent variable, with 0 indicating non-coal-mining industries and 1 indicating coal mining-related industries.

The Logit model was applied to explore the effects of livelihood capital and individual characteristics on the livelihood strategies of households in the Hengan New District. In Formula (4), i denotes the i-th resident, and P denotes the probability of choosing a coal mine-related industry. The explanatory variable X is the value of each livelihood capital (natural, physical, human, financial, social, and cultural capital) and personal characteristic, ${\beta }_0,{\beta }_1,\cdots {\beta }_{\rho }$ are proxy-estimated coefficients for the explanatory variables. The Logit model is applied in this study to measure the probability of occurrence of residents choosing coal mining related industries and non-coal-mining industries after changing the explanatory variables by 1 unit. The specific form of the Logit model is as follows:

$$\ln \left(\frac{P_i}{1-P_i}\right)=\alpha +{\beta }_0{x}_1+{\beta }_1{x}_2+\cdots +{\beta }_{\rho }{x}_{\rho }$$

(4)

4. Results

4.1. Evaluation of the Livelihood Status of Residents after Coal Mine Shantytown Transformation

Table 2. Description of livelihood capital indicator data for relocated residents.

Livelihood Capital	Indicator	Mean	Median	Standard Deviation
Human capital	Health status	4.16	4	0.94
	Level of education	2.62	2	1.21
	Skills training status	0.53	1	0.50
Natural capital	Ecological environment of the settlement	3.60	4	0.76
	Housing space ownership	21.15	20	5.89
Physical capital	Housing type	2.96	3	0.22
	Household fixed assets	0.41	0.44	0.15
	Infrastructure and public service conditions	10.68	11	2.12
Financial capital	Access to financial assistance	3.63	4	1.76
	Overall household income	2.22	2	0.97
	Access to credits	2.92	3	1.03
Social capital	Relative is public officials	0.16	0	0.37
	Involvement in community organizations	0.18	0	0.39
	Frequency of neighborhood contact	3.24	3	0.92
Cultural capital	Frequency of participation in festivals in the community	2.86	3	1.15
	Frequency of participation in festivals outside the community	2.31	2	1.14
	Degree of knowledge of community customs and traditions	1.73	2	0.62

shows the description of each livelihood capital indicator for dwellers. In terms of human capital, the overall health of respondents is good, but the average level of education is low between middle and high school. In terms of natural capital, the per-capita housing occupancy of residents after the shantytown reform is 21.15 m², but the standard deviation is relatively large, which indicates that the degree of divergence of the per-capita housing area is large. With regard to financial capital, the annual household income of residents is concentrated at RMB 30,000–50,000, which is still in the lower-middle range. Most residents, however, received financial assistance for housing from the government and corporation at the time of relocation. The situation of participation in community organizations is unsatisfactory, but since the shantytown transformation has adopted the centralized resettlement method, the frequency of interactions between neighbors is still high, and the social network structure is less disrupted. As for cultural capital, residents participate more frequently in holiday celebrations in their own community than they do outside of it.

According to the entropy method, the weights of each indicator of human capital, physical capital, financial capital, natural capital, social capital, and cultural capital of the Datong coal mine shantytown residents were calculated (see Figure 6), and then the six livelihood capital scores and the total livelihood capital score were derived based on the weights. The evaluation results of the sustainable development of the Hengan New District residents are as follows. The results of the sustainable livelihood assessment, as shown in Figure 6a, indicate that the livelihood capital of the Datong coal mine shantytown residents is generally at the level of low-value agglomeration. It is evident from Figure 6a and

Table 3. Descriptive statistics of livelihood capital score for residents of Hengan New District.

Livelihood Capital Type	Capital Value	Standard Deviation	Median	CV
Score	0.303	0.169	0.241	0.558
Human capital	0.0689	0.0432	0.094	0.627
Natural capital	0.0107	0.0029	0.0104	0.272
physical capital	0.0151	0.0037	0.0156	0.245
financial capital	0.0502	0.0248	0.0499	0.494
Social capital	0.0942	0.144	0.0119	1.533
Cultural capital	0.0634	0.0403	0.0603	0.635

Note: The coefficient of variation (CV) is the ratio of the standard deviation of the raw data to the mean. Using CV can be unaffected by differences in the indicator’s magnitude.

that most of the residents’ livelihood capital is centrally distributed around the mean level of 0.303. In addition, the livelihood capital of the residents has a tendency to be differentiated, with 2.45% of the residents having a livelihood capital value of less than 0.1 and 6.94% of the residents above 0.6.

There is obvious differentiation in the structure of livelihood capital among the relocated residents. Figure 6b shows the distribution structure of capital values for each category. From the mean values of the six livelihood capital scores, the order for the distribution of the six capitals among the residents is social capital > human capital > cultural capital > financial capital > physical capital > natural capital. Higher values for social and cultural capital reflect the richness of social networks, cultural resources, and interaction practices in the shantytown. This is a reflection of the fact that shantytown renovation has not destroyed the social network structure of the residents, who have a high degree of social integration. Human capital also performs well, mainly due to the good health status of the population in the study area, the rise of the tertiary sector following the shortening of the distance between the resettlement sites and the urban areas by the transformation, and the fact that the local coal mines are rich in resources and have a good industrial basis, which provides employment opportunities for the young and middle-aged workforce, so that population loss is not serious. Financial capital is in the middle of all capital, shantytown transformation is led by the government, the government and enterprise provide most of the housing funds for the residents, the bank provides loans for the residents, and the people have little difficulty in borrowing, but the overall income of the people is still low, which is inseparable from the general environment of social and economic development. Natural and physical capital account for a lower proportion of total livelihood capital compared to other capitals; nonetheless, for resettled residents, housing and community peripherals have been upgraded tremendously, and natural and physical capital are no longer critical elements in enhancing livelihood capacity.

The variation coefficient can characterize the degree of dispersion among the indicators and can be approximated to reflect the inequality of the residents’ livelihood capital in the study area (see Table 3). The relative divergence of the six livelihood capitals is as follows: social capital > cultural capital > human capital > financial capital > natural capital > physical capital. Although the values of natural and physical capital are relatively low among the six types of capital, they also have the smallest degree of variation, indicating that the overall inequality of natural and physical capital possessed by residents is low. This is because the government centrally builds resettlement housing for residents, with uniform construction standards and size, and residents enjoy the same public facilities and services.

4.2. Quantitative Analysis for the Impact of Livelihood Capital on Livelihood Strategies

Livelihood capital status influences the livelihood strategies of residents to meet their livelihood objectives. Among the 245 households in this survey, 94 households are engaged in coal mining-related industries and 151 households are in non-coal-mining industries, on the basis of which the logit regression results of livelihood capital and livelihood strategies were derived; the results are shown in

Table 4. Regression results of livelihood capital on livelihood strategies.

Variables	(1)	(2)
	Regression Coefficient	Regression Coefficients after Adding Gender and Age Variables
Human capital	−9.467 ***	−5.694
	(3.454)	(3.711)
Natural capital	−22.15	−3.606
	(56.26)	(58.49)
Physical capital	−45.35	−47.84
	(43.07)	(44.69)
Financial capital	25.88 ***	25.47 ***
	(7.170)	(7.554)
Social capital	−0.724	−0.0660
	(1.064)	(1.153)
Cultural capital	12.36 ***	12.31 ***
	(4.086)	(4.245)
Gender	−	1.007 ***
	−	(0.318)
Age	−	0.571 ***
	−	(0.194)
Observations	245	245
Pseudo R2	0.1478	0.2126

Note: *** indicates significance at the 1% statistical level. Standard errors are in parentheses.

and

Table 5. Regression results for indexes of livelihood capital on livelihood strategies.

Primary Indexes	Secondary Indexes	Tertiary Indexes as Variables	(1)	(2)
			Regression Coefficient	Regression Coefficients after Adding Gender and Age
Livelihoods capital	Human capital	Health status	−166.5 **	−68.00
			(73.89)	(82.72)
		Level of education	48.74 ***	62.48 ***
			(16.96)	(18.55)
		Skills training status	−11.01 ***	−8.190 *
			(3.934)	(4.230)
	Natural capital	Housing space ownership	−2.137	−8.230
			(115.3)	(120.5)
	Physical capital	Household fixed assets	−135.1 **	−141.8 **
			(59.63)	(60.82)
	Financial capital	Overall household income	40.49 ***	37.76 **
			(14.12)	(14.77)
		Access to credits	−45.75 *	−30.94
			(26.60)	(28.17)
	Social capital	Frequency of neighborhood contact	19.91	6.291
			(49.97)	(54.45)
	Cultural capital	Frequency of participation in festivals in the community	61.64 ***	53.06 **
			(20.86)	(21.80)
Individual characteristics	Gender	−	−	0.780 **
		−	−	(0.354)
	Age	−	−	0.771 ***
		−	−	(0.236)
Observations			245	245
Pseudo R2			0.2525	0.3066

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% statistical levels, respectively. Standard errors are in parentheses.

.

The regression results shown in Table 4 indicate that there is no significant correlation between whether or not residents of Hengan New District choose coal mine-related industries and the natural capital, physical capital, or social capital that they possess. This is mainly attributed to the fact that after the unified transformation by the government, the living environment and material conditions of the relocated residents are maintained in a relatively equal situation, with little difference between each household, so that the influence of natural, material and social capital on the choice of working in the coal mining-related industries or in the non-coal-mining industry is not significant. Human capital, financial capital and cultural capital are all significant at the 0.01 level in the willingness of residents to choose livelihood strategies.

As indicated in Table 4(1), with all else unchanged, human capital has a negative effect on the willingness to choose the coal mining industry. This suggests that as human capital increases, residents are less likely to choose to work in the coal mining industry. Of these, as shown in Table 5, the higher the level of education, the greater the likelihood of working in coal mining-related industries, while those who receive more skills training are more likely to choose other types of industries to earn a living. This is also verified according to the field survey; with the modernization of the coal mining industry, the coal mining industry has gradually increased the requirements for qualifications, and most of the elderly employees working in the coal mines have received education above the level of junior colleges. Livelihood strategies may diversify as people acquire more other skills. Although physical capital does not significantly affect residents’ livelihood choices in general, one of the indicators, household fixed assets, has a negative effect on residents’ choice to work in coal mining-related industries. Compared to human capital, both financial capital and cultural capital have significant positive effects on residents’ livelihood strategies. The higher the overall household income, the greater the likelihood that residents will choose a coal mining-related industry. Residents who attend traditional festivals and celebrations in their own community with greater frequency are more likely to choose to work in coal mining industries.

Specifically, after shantytown transformation, the role of cultural factors in the study area does not weaken with changes in material conditions and neighborhoods, but rather becomes an important reason to support the residents’ engagement in coal mining-related industries. This suggests that although the difficulty of governance in urban transformed shantytowns lies in the complexity brought about by social integration and social cohesion, the solution to the micro-level problems of shantytowns is still inseparable from community culture which is the core element. Addressing the governance of urban shantytowns also should focus on community cultural construction, maximizing the contribution of cultural capital to the choice of livelihood strategies for residents. Financial capital has an important position in the structure of livelihoods. It reflects the actual state of wealth of the residents following shantytown transformation. It is an indisputable fact that coal mines can bring enormous wealth to the local area, so people with more income are more likely to engage in coal mine-related businesses.

It is interesting that by adding gender and age to the regression model, we find that the effect of human capital becomes non-significant, while both gender and age have significant positive effects on residents’ engagement in coal mine-related industries (see Table 4(2) and Table 5(2)). Keeping all other conditions constant, males are more likely than females to be engaged in coal mine-related jobs, and the older the labor force, the more likely they are to be working in coal mine-related industries. Specific to the indicators of human capital, the changes in the effect of health status on livelihood strategies in coal mines is from significant to non-significant, as well as the reduction in the significant effect of skills, from significant at the 0.01 level to the 0.1 level. There was no change regarding the impact of education level on livelihood strategies. This suggests that gender and age as innate individual characteristics dilute the impact of health status on livelihood strategies. Whereas acquired education and skills have always been essential dimensions of human capital [77]. When gender and age are included in the model, the effects of financial capital and cultural capital on whether or not residents are engaged in coal mining-related industries do not change noticeably. Gender inequality is evident in the coal mining-related industries, which are not conducive to women’s employment due to the labor-intensive and risky nature of the industries. The positive effect of age on livelihood strategies also implies that the coal mining industry is not very attractive to the younger generation of the labor force, and compared to emerging industries, young people prefer the latter.

4.3. Robustness Test

In order to verify the robustness of the results of the above analyses, the model substitution method was used for validation first. A probit model and OLS estimation methods were employed to validate the impact of residents’ livelihood capital and personal characteristics on the choice of livelihood strategies. The regression results, as shown in

Table 6. Substitution model method: impact of livelihood capital and individual characteristics on strategies choice.

Variables	(1)	(2)
	Probit	OLS
Human capital	−3.2209	−1.0939
	(2.22)	(0.72)
Natural capital	−1.7796	−2.0092
	(34.26)	(11.50)
Physical capital	−27.7804	−8.7773
	(26.44)	(7.72)
Financial capital	14.9227 ***	4.7414 ***
	(4.43)	(1.36)
Social capital	0.0497	−0.0315
	(0.64)	(0.25)
Cultural capital	7.4335 ***	2.1620 ***
	(2.51)	(0.79)
Gender	0.6012 ***	0.1943 ***
	(0.19)	(0.06)
Age	0.3405 ***	0.0967 ***
	(0.11)	(0.03)
Observations	245	245
R2	0.2131	0.2505

Note: *** indicates significance at the 1% statistical level. Standard errors are in parentheses.

, show that the coefficients on the effects of financial capital, cultural capital, gender and age are all significant at the 1% statistical level.

Table 7. Substitution model method: indexes of livelihood capital and individual characteristics on strategies choice.

Primary Indexes	Secondary Indexes	Tertiary Indexes as Variables	(1)	(2)
			Probit	OLS
Livelihoods capital	Human capital	Health status	−40.7865	−10.3901
			(47.54)	(13.02)
		Level of education	37.1491 ***	10.7090 ***
			(10.79)	(3.01)
		Skills training status	−4.9129 **	−1.4129 **
			(2.46)	(0.68)
	Natural capital	Housing space ownership	−20.5998	−5.8224
			(48.63)	(13.65)
	Physical capital	Household fixed assets	−84.6628 **	−20.5881 **
			(35.43)	(9.07)
	Financial capital	Overall household income	22.4737 ***	6.5004 ***
			(8.70)	(2.40)
		Access to credits	−19.6795	−5.0057
			(16.42)	(4.38)
	Social capital	Frequency of neighborhood contact	4.7758	1.0375
			(31.61)	(8.38)
	Cultural capital	Frequency of participation in festivals in the community	30.6939 **	8.4070 **
			(12.66)	(3.44)
individual characteristics	Gender	-	0.4367 **	0.1322 **
		-	(0.21)	(0.06)
	Age	-	0.4520 ***	0.1141 ***
		-	(0.14)	(0.04)
Observations R2			245	245
			0.3076	0.3445

Note: ** and *** indicate significance at the 5%, and 1% statistical levels, respectively. Standard errors are in parentheses.

demonstrates that the direction and significance of the regression coefficients are stable for each type of livelihood capital tertiary indicator specifically. The results of the test are consistent with those of the Logit main regression, indicating that the empirical results are robust.

Next, the robustness of the results was verified by varying the sample size. Considering that residents over the age of 60 in China have already reached retirement age and are likely to no longer earn a living in the job market, this portion of the sample was excluded, and a final screening yielded a final sample of 239. Logit model and variables were kept invariant to verify the robustness of the results. The results with changed sample size are shown in

Table 8. Varying-sample-size method: impact of livelihood capital and individual characteristics on strategies choice.

Variables	Regression Coefficient	Variables	Regression Coefficient
Human capital	−5.613	Social capital	0.116
	(3.822)		(1.183)
Natural capital	8.326	Cultural capital	11.48 ***
	(59.39)		(4.273)
Physical capital	−50.89	Gender	0.981 ***
	(45.59)		(0.328)
Financial capital	28.96 ***	Age	0.756 ***
	(7.827)		(0.220)
Observations	239	Observations	239
Pseudo R2	0.2307	Pseudo R2	0.2307

Note: *** indicates significance at the 1% statistical level. Standard errors are in parentheses.

and

Table 9. Varying-sample-size method: indexes of livelihood capital and individual characteristics on strategies choice.

Primary Indexes	Secondary Indexes	Tertiary Indexes as Variables	Regression Coefficient
Livelihoods capital	Human capital	Health status	−101.5108
			(85.76)
		Level of education	62.4268 ***
			(18.93)
		Skills training status	−8.3081 *
			(4.40)
	Natural capital	Housing space ownership	−35.3261
			(85.76)
	Physical capital	Household fixed assets	−144.1328 **
			(61.76)
	Financial capital	Overall household income	38.8372 **
			(15.16)
		Access to credits	−27.3887
			(29.10)
	Social capital	Frequency of neighborhood contact	−16.4041
			(56.37)
	Cultural capital	Frequency of participation in festivals in the community	57.1323 **
			(23.57)
Individual characteristics	Gender	−	0.7615 **
		−	(0.37)
	Age	−	0.9629 ***
		−	(0.27)
Observations			239
Pseudo R2			0.3282

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% statistical levels, respectively. Standard errors are in parentheses.

, which show that the coefficients of the variables changed minimally, with the direction of the coefficients and the level of significance remaining the same. As a result, further validation shows that the empirical results are robust.

5. Discussion

This paper empirically analyzes the current state of livelihood capital of resettled residents and the logic of their livelihood strategy choices after shantytown transformation in a resource-based city in China by using the improved SLF.

The findings of this study can be summarized in two parts. First, on the basis of scientifically constructing the indicator system and determining the weights of the indicators, the research yields specific measurement results of residents’ livelihood capital. Based on the results, we can determine the livelihood status of the people living in the transformed shantytowns. The overall level of livelihood capital is relatively concentrated, but a small number of ‘livelihood disadvantaged groups’ and ‘livelihood advantaged groups’ emerge at the same time. This confirms that coal mining communities themselves are characterized by structural inequality [84]. Although residents were resettled to a new place to live, their social and cultural capitals are prominent in the livelihood capital structure. Some studies have shown that coal mining communities typically exhibit strong social and intergenerational ties to the industry and that residents in coal mining communities possess a special sense of identity and community solidarity that will last locally for a long time, notwithstanding the gradual depletion of mineral resources [73,85]. This study suggests that the government does not destroy the local social network structure while improving the physical conditions of housing for residents, and preserves the long-standing coal mining and enterprise culture, which constitute important reasons for the high scores in social capital and cultural capital. The availability of human capital is enhanced by local well-supported educational resources and location advantages after shantytown transformation. This finding is in accordance with Zanoni et al.’s study that slum-upgrading programs do contribute to increased school attendance [86]. Our findings show that residents also have higher financial capital scores, which contradicts the results of several slum-upgrading studies in Africa and India, which found that housing upgrading exacerbated the economic burdens of residents [27,37]. The results of the analysis in this paper reveal that financial compensation for housing resettlement and reduced difficulty in obtaining loans are the reasons for higher financial capital scores.

Second, the results of the analysis answer the question about how changes in residents’ livelihood capital affect livelihood strategies. The results of descriptive statistics show that residents’ livelihood strategies can be categorized into engaging in coal-related industries and non-coal industries, and regression analyses found that there are significant differences in the effects of various types of livelihood capital and individual characteristics on livelihood strategies.

The second issue is discussed in relation to the three key findings below. Firstly, the results of the survey and analysis in this paper show that residents of transformed coal mine shantytowns are still more likely to engage in coal mine-related work, and the employment status of randomly surveyed residents also shows that coal mine-related jobs occupy a large proportion of the total. This indicates that the livelihoods of the inhabitants of the surveyed areas are still very much reliant on coal mining resources. In contrast, a recent survey among residents of an Indian coal mining city showed that people prefer other types of industries to coal-related jobs, and the authors argue that renewable energy can create opportunities for employment in replacing non-renewable energy sources [8]. The study is positive about the livelihood impacts of reducing fossil fuels. The difference between these two results may be related to the stock of local coal resources and the development level of the clean energy industries.

Secondly, this paper added cultural capital to the SLF considering the features of the coal mining community and focused on the importance of culture in the choice of livelihood strategies of the residents, and the study found that those with higher values of financial and cultural capital are more likely to work in industries related to coal mining. This result verifies previous sociological research on coal mining communities. Previous studies have already revealed the meaning of mining culture for local people; for example, the ‘minescape’ has been regarded by scholars as a figurative tool with socio-cultural value beyond its purely physical nature [74], and the people involved in the extractive industry are not motivated by purely economic rationality; rather, people living in physically segregated coal mining communities have distinctive emotions, cultures, and values [74]. Coal is more than just a resource with economic value here; it acts as a link reinforcing the emotive nature of the connection between people and place in a ‘minescape’ [73].

Thirdly, the regression results show that gender and age have positive effects on whether or not to engage in coal mining-related industry strategies. One of the reasons why coal mining areas have long been characterized by high levels of poverty leading to the formation of shantytowns is that women’s participation in economic activities is severely limited, with the majority of females working as unpaid domestic servants and males earning income from the mining industry to support their families in the coal mining communities [87]. Female exclusion from mining jobs remains unchanged, and the employment environment for females has not improved significantly, despite the drastic changes in the living conditions of the inhabitants in the study area. In coal mining communities with a long history of mining, the coal mining industry gradually forms generational transitions [73], and it is common to see several generations of a family being miners during the coal prosperity period. Elderly residents are more likely to have a ‘generational sentiment’ [73] towards coal mining and ‘to be nostalgic for the heyday of the mining’ [9], while the younger generation, living in an era of decline in the coal industry and growth in new energy sources, are more likely to choose alternative livelihood strategies.

These findings respond to the findings of previous studies. As Turley et al.’s review study examining the impacts of shantytown transformation revealed, the effects of housing interventions on residential economic deprivation and unemployment have been mixed [69]. Although poor-friendly housing reconstruction programs implemented by the government and the physical environment in which people live have been improved, there have been no significant changes in their livelihood strategies [27].

The theoretical and practical significance of this study is reflected in the following aspects. On the one hand, it analyzes the impact of livelihood capital on livelihood strategies at the micro-level and improves the analysis framework of sustainable livelihoods for coal mining community residents by supplementing cultural capital, which draws attention to the uniqueness of coal mining communities in resource-based cities. On the other hand, this study is conducive for the relevant authorities to recognize that shantytown transformation in resource-based areas is not only about material and economic support, but also about the importance of building the maintenance of social network structures and the underlying culture in the community, and most importantly, creating an employment environment with diversified livelihood strategies, as well as making policies that comprehensively consider the different factors affecting livelihood strategies.

There are several limitations of this study that are expected to be further addressed in the future. Firstly, the results of this paper were analyzed in the context of a resource-dependent urban shantytown transformation scenario in China, and the uniqueness of China’s shantytown transformation governance may limit the applicability of the results of the analysis, but it does not preclude the same implications for cities with similar conditions in other countries. Secondly, using field survey data from 245 households in 2023, presenting the livelihood strategy choices of residents in the surveyed areas at the micro-level is relatively limited, and external factors including sectoral development, policy guidance, and other socio-economic conditions might also influence residents’ livelihood strategy choices. Future researchers could explore further evidence at the macro level. Finally, this paper only uses data from the 2023 field survey, and there are no continuous dynamic data to utilize, which can only reflect the current livelihood status of residents. Future work should be devoted to exploring more periods of data to reflect longitudinally and comprehensively the changes in residents’ livelihoods before and after shantytown transformation.

6. Conclusions and Implications

As an ‘Intervention’ of urban renewal in China, shantytown transformation aims to improve the living conditions of people in backward urban areas, and industrial and mining shantytowns in resource-based cities are the focus of the transformation. This study explores a neglected theme: the livelihoods of residents in resource-based cities after the transformation of poor settlements. This paper focuses on the residents in the Datong coal mine shantytowns in China after the completion of the transformation in 2013, utilizing field survey data for the year 2023 to evaluate the residents’ livelihood capital status based on the improved SLF and analyzing the impacts of livelihood capital on the residents’ livelihood strategies at the micro-level. The main contents and findings of the study are as follows.

(1)

This paper constructed a livelihood capital evaluation system for residents of shantytowns in China’s Datong coal mine after transformation. This evaluation system consists of 17 indexes, including 3 tertiary indexes for human capital, 2 tertiary indexes for natural capital, 3 tertiary indexes for physical capital, 3 tertiary indexes for financial capital, 3 tertiary indexes for social capital, and 3 tertiary indexes for cultural capital.

(2)

The entropy method was used to measure the livelihood capital score and evaluate the current level of livelihood capital of the residents. Overall, livelihood capital values are in a state of low-value aggregation and divergence. Specifically, the values of social capital, human capital, cultural capital, and financial capital are relatively prominent in the livelihood capital structure of the residents.

(3)

This study analyzed the effect of livelihood capital possessed by residents on the choice of livelihood strategies. The statistics show that the livelihood strategy of the surveyed residents is still dependent on coal mining-related industries. Logit regression analysis shows that financial capital and cultural capital have a significant positive contribution to residents’ choice of livelihood strategies in coal-related industries. Moreover, after adding gender and age variables, the results show that males and the older residents are more likely to choose coal-related industries. Even with the addition of gender and age variables, the positive effect of financial capital and cultural capital on the choice of livelihood strategy in coal mine related industries remains significant.

Combining the results of the study and the problems identified in the research, this paper proposes several policy recommendations to enhance the level of livelihood capital of the residents and to realize the sustainable development of the new community. The first involves giving full play to the cultural advantages of coal mining communities and strengthening the cultural capital of new communities to make them new points of economic growth. Secondly, the business environment should be optimized, and financial support for small and microenterprises and self-employed businessmen should be strengthened, so as to create channels for diversified employment. It is important to encourage and guide residents to optimize their financial capital allocation by diversified means. Thirdly, residents should be encouraged to increase their willingness to support and invest in their children’s education while improving the construction of local talent markets and promoting entrepreneurship and innovation, so that human capital can be effectively utilized. Last but not least, the fourth recommendation involves strengthening the cooperation between universities and the region, fully activating the scientific and technological innovation resources of the local coal mining industry, promoting the transformation of scientific and technological achievements regarding coal, and going to the market to drive employment. Additionally, the employment pattern of traditional coal mines should be further shifted, enabling more females to share the benefits brought by coal mining resources.