Exploring the impacts of coal mining on host communities in Shanxi, China – using subjective data

Resources Policy 53 (2017) 125–134

Contents lists available at ScienceDirect

Resources Policy journal homepage: www.elsevier.com/locate/resourpol

Exploring the impacts of coal mining on host communities in Shanxi, China – using subjective data

MARK

⁎

Qian Lia,c, Natalie Stoecklb, , David Kingc, Emma Gyurisc a b c

Lingnan College, Sun Yat-sen University, Guangzhou 510275, PR China College of Business, Law & Governance, James Cook University, Townsville, QLD 4811, Australia College of Sciences and Engineering, James Cook University, Townsville, QLD 4811, Australia

A R T I C L E I N F O

A BS T RAC T

Keywords: Coal mining Subjective indicators Shanxi

The neglect of the welfare of host communities in the current mining practice is partly due to the lack of a defensible measurement of the impacts of coal mining on host communities. Subjective indicators, superior to the traditionally used objective indicators in terms of informing policy makers of public preference, are barely used in mining impact assessment. The objective of this study is to illustrate approaches to use subjective data to investigate the impacts of coal mining. It looks at how satisfied people are with multiple wellbeing factors, what matters most/least to people, and at their perception of the impact that coal mining has on these factors. Comparisons are made between location categories characterized with different intensities of coal mining. Two composite indices that blends responses to questions about satisfaction and importance/perceived impacts of coal mining are constructed to identify policy priorities. The general and pervasive message is that coal mining does not seem to improve subjective satisfaction with those wellbeing factors, instead, it has negative impacts on a wide range of wellbeing factors pertaining to the natural environment and the economy. This paper supplies references for public policy to improve local wellbeing, and demonstrates approaches to use subjective data.

1. Introduction The spectacular development of China over the last 30 years was facilitated by energy derived from vast quantities of fossil fuels, mostly coal, making China the World's largest producer and consumer of coal. China's stellar development has delivered economic and social benefits on a vast scale, lifting a vast amount people out of poverty as well as consolidating China's global position as one of the leading economic and political powers of our age. Balanced against the benefits of development, coal mining, processing and utilisation are demonstrably responsible for negative impacts on multiple scales ranging from the global to the local. Apart from irreversible worldwide climate change, coal mining also has severe negative environmental, social and economic impacts on regional scales. Coal mining-led development, as much development based on other mineral extracting industries, often induces conflict between the objectives of the mining enterprise, the needs of local or host communities and the policy goals of national, regional and local governments (Li et al., 2012). While struggling to address the many environmental impacts of coal mining, governments of many developing countries, including

⁎

China's, lack the ability or political will to effectively address the many other impacts of coal mining, such as those affecting social injustices and inequity (Morrice and Colagiuri, 2013). This lack is partially due to governments in developing country emphasising overall, national economic growth and poverty alleviation over the concerns of mining's negative impacts on local communities in mining areas (OECD, 2002). Apart from focusing on national development objectives, another reason for the apparent neglect of host communities’ welfare is the lack of information about the consequences of mining pertaining to certain societal goals and standards (Noronha, 2001) that rely on accountable and fair measurement of the impacts of coal mining on host communities. Indicators, that are most commonly used to measure wellbeing of communities, including those of local mining communities (from here on identified as host communities), refer to objective and subjective indicators. The relatively few empirical studies that have explored the relationship between objective and subjective measurements of wellbeing tend to return contradictory results – the 2 sets of indicators are not always consistent (Schneider, 1975; Emmons and Diener, 1985; Oswald and Wu, 2010). The discrepancies between objective and subjective indicators reinforce the need for the parallel development

Corresponding author. E-mail addresses: [email protected] (Q. Li), [email protected] (N. Stoeckl), [email protected] (D. King), [email protected] (E. Gyuris).

http://dx.doi.org/10.1016/j.resourpol.2017.03.012 Received 23 May 2016; Received in revised form 10 February 2017; Accepted 30 March 2017 0301-4207/ © 2017 Elsevier Ltd. All rights reserved.

Resources Policy 53 (2017) 125–134

Q. Li et al.

of both sets of indicators rather than detract from the value of either (Lee and Marans, 1980; Veenhoven, 2002). However, the measurements of progress or wellbeing are dominated by objective indicators (e.g. Li et al., 2012). These are relatively easily captured and can be readily aligned with government policy and public expenditures (Diener and Suh,1997; Abdallah et al.,2011; Rablen, 2012). Progress defined by sets of objective measures is assumed to indicate uniform improvements in community wellbeing over a large number of communities in different social, economic, cultural and environmental settings. However, objective indicators may be heavily influenced by the values of those who construct them (Rablen, 2012) and they may not necessarily reflect the wellbeing experienced by individuals living in diverse communities. In contrast, subjective indicators are superior indicators of public preference (Veenhoven, 2002), thus may serve better to measure the impact of coal mining on the welfare of host communities. Measures of subjective wellbeing “are generally found to have a high scientific standard in terms of internal consistency, reliability and validity and a high degree of stability over time” (Welsch, 2006, p. 803). Subjective wellbeing are now at the forefront of some academic analyses (Chen and Lin, 2014), and has been measured in in OECD countries. In China, criticism of the country's narrow pursuit of GDP growth and calling for officials to pay more attention to subjective wellbeing has been rising. Guangdong Province took the lead to incorporate subjective wellbeing as one of its policy goals (Tang, 2011). However, subejctive wellbeing has not been applied as the measure of mining's impact on host communities, although Noronha (2001) promoted the measurement of subjective satisfaction of individuals within mining regions. Acceptance and widespread application of subjective indicators to gauge mining's contribution (both positive and negative) to the wellbeing of people in host communities requires further demonstration of the value of this approach to policy makers. Thus, this study uses subjective indicators to explore the impacts of coal mining on host communities in a case-study area of Shanxi province, China. In particular we sought confirmation or otherwise if respondents living adjacent or near coal mines and associated activities bear most of the costs while experiencing limited benefits of coal mining. By doing so, it innovatively identifies policy priorities to efficiently improve the wellbeing of host communities by identifying factors which people are most dissatisfied with, which are ‘most’ important to the public and which are also the most affected by coal mining.

Table 1 Criteria for selecting sampling sites that represent different intensities of coal mining. Classification category Place with coal mining

Place close to coal mining

Place far from coal mining

Defined as

Coal mining, with or without associated activities/facilities (such as coal washing and coal transportation), present within 10 km of the administrative/residential area and present high-intensity exposure to impacts of coal mining. Coal mining, with or without associated activities/facilities, present within 10–20 km of the administrative residential area and present moderate-intensity exposure to impacts of coalmining. Coal mining, with or without associated activities/facilities, absent within 20 km of the administrative/residential area and presenting none or only light exposure to coal mining impacts.

(Moffat et al., 2014a; Moffat et al., 2014b; Zhang et al., 2015) compared attitudes in mining and non-mining regions and in (nonmining) metropolitan areas to investigate Australian/Chinese/Chilean attitudes/perception towards mining. Within the Shanxi Province case study area, sampling focused on three major areas: Shouzhou, Yangquan and Linfen. Exploration of these areas revealed that coal extraction was typically closely associated and often co-located with coal stockpiles, transportation routes and coal washing facilities, all of which, alone or in combination, would impact local residents’ lives. Therefore, this study considered their variously combined impacts as the aggregated impacts of coal mining. After reconnaissance, sampling of communities was stratified by 10 km increments in distance from coal mines/mining operations (Table 1). Both urban and rural areas were sampled because of the significant differences in lifestyles, livelihoods, public infrastructures and incomes between rural and urban areas of China (Lu and Chen, 2004; Kanbur and Zhang, 2005; Sicular et al., 2007). As the access to urban facilities or major transport routes might have effect on quality of life (Brereton et al., 2008; Arifwidodo and Perera, 2011), villages were selected at similar distances from urban areas to minimize this effect. In China, coal mining is not allowed in the urban areas, thus limiting the sampling to 5 categories of locations; rural area with coal mining (Rural With), rural area close to coal mining (Rural Close), urban area close to coal mining (Urban Close), urban area far from coal mining (Urban Far) and rural area far from coal mining (Rural Far). 2.2. Questionnaire design and development There is no standard or commonly agreed set of ‘domains’ or ‘factors’ about which researchers and organisations, who are interested in wellbeing, collect data. Davey and Rato (2012) encouraged the use of diverse and creative measures; different instruments can be chosen according to study aims, sampling approaches, socio-cultural contexts, etc. In the absence of previous research using subjective wellbeing measures to assess the impacts of coal mining on human wellbeing, this study had to be exploratory. The number of different factors that affect wellbeing are enormous and diverse. Cummins (1996) demonstrated that 68% of 173 different variables from the literature could be grouped into seven life “domains”, including: material wellbeing, health, productivity, intimacy, safety, community emotional wellbeing and spirituality. As such, each life domain can be considered as an aggregate of a number of components/factors (Cummins, 1996). Different individuals with different backgrounds, understandings and experiences may group factors representing certain life domain in different ways. Therefore, empirical researchers must seek to strike a balance between developing questions that will collect, with high reliability and repeatability, sufficient information about a variety of life domains and the need to keep questions to a manageable number (Cummins, 1996). Factors that the broader literature had identified as being impacted

2. Material and methods 2.1. Case study area and sampling design The case study area, Shanxi Province, holds 27% of China's coal resources, making it one of China's most important coal mining areas (China Energy Information Network, 2009). The coal industry is the main contributor to local GDP, accounting for 56.6% of GDP in Shanxi Province in 2012 (Editor of Land & Resource Herald, 2013). The premise underlying our sampling design was that the magnitude of various impacts of coal mining were related to the distance between places of residence and coal mines. Thus, comparing subjective indicators obtained in a range of communities that have exposure to different intensities of coal mining could thus further the understanding of the extent to which coal mining affects wellbeing in different contexts. Although the number of previous studies that have made this comparison is limited, they do offer some clues about how to differentiate locations according to exposure to different mining intensities (e.g. Kitula, 2006; Hendryx and Ahern, 2009; Zhang et al., 2015). For example, a survey conducted by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) of Australia 126

Resources Policy 53 (2017) 125–134

Q. Li et al.

Table 2 Number of useful surveys collected at various location categories.

Sample size Proportion of total (%)

Urban Close

Rural With

Rural Close

Urban Far

Rural Far

Urban areas

Rural areas

Total

90 16.6

184 33.9

179 33.0

30 5.5

59 10.9

120 22.1

422 78.0

542 100

village. Places where local residents gathered for social interaction were also visited for data collection. Trained personnel, fluent in the local language and dialect, administered surveys. The diversity of respondents was monitored and controlled by the investigators. The core interest of this study was to investigate the impacts of coal mining on wellbeing in regions with different intensities of coal mining. Thus survey deployment was strategized to focus on areas that were (i) within 10 km or (ii) between 10 and 20 km from mines or mine associated activities. Our sampling was also driven by the knowledge that social science research in China is somewhat limited in that most social data have been collected from urban samples, overlooking rural residents who account for 60% of the total population (Davey and Rato, 2012). Thus the largest samples were collected from rural coal mining areas (Rural With and Rural Close). Sample sizes for locations far from mining activities were relatively small, and the non-parametric analytical approaches used accounted for that. A total of 542 usable surveys were collected within the case study area (Table 2).

by coal mining (either directly or indirectly) or important to wellbeing were thus used to guide the development of the survey instrument. Eventually, this study included questions about 29 factors, which covered as many as wellbeing factors identified from the literature that contribute to wellbeing (see Li et al., 2017). Larson et al. (2014) used a 5-point scale to measure the satisfaction with and the importance of different values associated with the Great Barrier Reef in Australia. A 7-point bipolar scale ranking from strongly dissatisfied (score 1) to strongly satisfied (score 7) was used in this study to measure satisfaction with wellbeing factors. This is because wider 7–11 point) scales permit a relatively precise response than narrower 2–5 point) scales (Mackerron and Mourato, 2009; OECD, 2013). Meanwhile, bipolar scale anchors (running between two opposing constructs, i.e. from completely unhappy to completely happy) are less ambiguous than unipolar scales (i.e. reflecting a single construct running from low to high), making it more likely that all respondents interpret the scale in the same way (OECD, 2013). Additionally, weighting the importance of wellbeing factors can provide more insights to inform public policies than only using satisfaction measurements (Larson, 2010; Larson et al., 2013, 2014; Chen and Lin, 2014). Thus this study also included questions asking about the importance of wellbeing factors to identify what matters most to the host community and to blend the satisfaction and importance of wellbeing factors. To align with the scale used to measure satisfaction with wellbeing factors, a 7-point scale was used to assess importance, ranging from very important to very unimportant. Despite the associations between coal mining and socioeconomic disadvantages, health problems and environmental quality, it may be difficult to link wellbeing with the impacts of coal mining (Hendryx and Ahern, 2009). To reduce this uncertainty, Moffatt and Pless-Mulloli (2003) use local residents’ perception of coal mining's impacts on wellbeing factors. Thus, this study also sought information on perceived ‘impact’ of coal mining on wellbeing factors. To align with the scales used on those indicators, a 7-point scale was used to elicit the perceived impact of coal mining on each wellbeing factor, ranking from a strong negative impact to a strong positive impact. The survey was designed to be relevant to adult respondents of all ages. The survey questions were developed in English first and forward translated into Mandarin by the senior author, then a backward translation, was conducted by a translator unfamiliar the original survey questions. Accuracy of the translation was verified by a backward translation that was then compared with the original English version. The finalised survey instrument consisted of 55 questions, 29 of which were used in this study. These questions sought respondents to rate on a scale of 1–7, their satisfaction with, the importance of and their views of coal mining's impact on each of 29 factors.

2.4. Data analysis To visually explore the differences and similarities in satisfaction with and the importance and perceived impacts of coal mining between places with different exposure of coal mining, the mean scores pertaining to each of the 29 wellbeing factors, for each location category were calculated and displayed on radar charts. Not all researchers agree that it is appropriate to use Likert data in this way (since Likert responses, strictly speaking, provide only ordinal information), so these charts should be interpreted as providing only an overview of respondents’ subjective wellbeing and attitude towards coal mining. Analytical approaches that are appropriate for Likert data (the Kruskal-Wallis pairwise comparison) were then undertaken to determine if there were statistically significant differences in the distribution of responses to questions about each factor between the 5 different location categories. Factors that people consider to be extremely important to their overall quality of life, and with which people are very dissatisfied, should be the management priority for policy makers, and improvement in these factors has the potential to improve the quality of life of residents (Cummins and Nistico, 2002; Cummins, 2003). This ensures that policy makers concentrate on improving factors which matter ‘most’ and with which people are also most dissatisfied. To this end an Index of Dis-Satisfaction (DS) was calculated following Larson (2010) and Larson et al. (2013). This index has been successfully used to create an ‘action list’ of wellbeing factors to identify management and policy priorities in different contexts, such as in the Australian Tropical Rivers region (Larson et al., 2013) or along the coast of the Great Barrier Reef (Larson et al., 2013). This entailed inverting the satisfaction score (S) that each respondent, i, gave to each factor, k (yielding Sik.) to calculate a dissatisfaction score: DSik = 8 – Sik, and then multiplying that dissatisfaction score by the importance score that each respondent gave to the corresponding factor (Wik) to calculate the Index of Dis-Satisfaction (IDS) for each factor k:

2.3. Data collection Various survey approaches were considered. A mail survey is impracticable in China because many households have no mail box, especially rural places. Door-to-door household surveys are impracticable in urban areas, because most urban residents live in apartments and are too suspicious to open their door for strangers. Thus in urban areas, we selected respondents on the streets or in public squares, parks, and shops in different suburbs. In rural areas, a door-knock survey, covering as large an area as possible, was conducted in each

IDSik =

1 N

N

∑i =1 Wik · DSik

(1)

Where N is the number of respondents from each location category. A plot of DS against W was generated, to provide a visual aid in the interpretation of scores (Fig. 5). 127

Resources Policy 53 (2017) 125–134

Q. Li et al.

all location categories. The mean score relating to income was approximately 4 and again there were no significant difference between coal mining and non-coal mining areas, indicating that people in coal mining areas, were not more satisfied with their family income than those from non-coal mining areas. Notably, rural residents living with and close to coal mining were less satisfied with income disparity than rural residents living far from coal mining. People living in coal mining areas were more dissatisfied with air quality, water safety, inflation and price of necessities than those living in non-coal mining regions. Those were also the factors with which people living in coal mining areas were most dissatisfied among all the factors. In contrast, people living far from coal mines were most dissatisfied with wellbeing factors relating to inflation, real estate prices and education. There were few wellbeing factors that people living in coal mining areas were more satisfied with than their ‘non-coalmining’ counterparts. Most interesting of all, perhaps, was the fact that people living in rural areas far from coal mining seemed more satisfied with more wellbeing factors relating to air quality, water safety, the price of necessities, the quality of government and fairness of income and social capital (trust, help and honesty) than those living in any other category of residential area. Distance from coal mining seems to be a decisive factor that affects the satisfaction with air quality and water safety, as people living in places far from coal mining, irrespective of whether they were in rural or urban areas, were more satisfied with air quality and water safety than those living in the places ‘with’ or ‘close to’ coal mining. Those living in rural areas felt better able to trust, help and be honest with each other than those living in urban areas. This urbanrural ‘divide’ with respect to social capital is likely due to more active social networks, civic participation and cohesion in rural areas (Hofferth and Iceland, 1998; Ziersch et al., 2009).

It is reasonable to assume that the perceived positive/negative impacts of coal mining on any given wellbeing factor may affect satisfaction with that particular wellbeing factor. Combining responses to questions about (dis)satisfaction and perceived impacts thus allows identification of the magnitude of coal mining's impacts and satisfaction level with given wellbeing factors. This draws the attention of policy makers to factors that coal mining most negatively impacts and with which people are most dissatisfied. To achieve this, Larson's IDS was altered to develop The Index of Dis-satisfaction and Negative Impact (IDSNI). Here, responses to questions about perceived impacts (rather than importance) were combined with dissatisfaction scores. This entailed inverting the impact score (I) that each respondent, i, gave to each factor, k (yielding Iik.) to calculate a negative impact score: NIik = 8 – Iik, and then multiplying that dissatisfaction score. Where N is the number of respondents for each type of case-study area.

IDSNIik =

1 N

N

∑i =1 DSik · NIik

(2)

This index was only calculated for people who lived ‘with’ or ‘near’ coal, as subjective wellbeing pf people living far from coal mining were not affected by coal mining. People living far away from coal, had no actual experience of coal mining's impacts, and were thus only able to reveal the way in which they thought mining might impact various wellbeing factors if coal mining were ever to start near their place of residence. 3. Results 3.1. Satisfaction with wellbeing factors Fig. 1 displays the mean scores of satisfaction with factors of wellbeing by location category. People throughout the case study area were quite satisfied with their health and their relationships (mean scores around 6) and no statistically significant differences were revealed between location categories in these scores by the Kruskal-Wallis tests. Responses to questions about satisfaction with family income were also similar across

3.2. The importance of wellbeing factors Evident from Fig. 2, health and relationships were the most important factors right across all the location categories, irrespective

Fig. 1. Mean scores relating to satisfaction with wellbeing factors by location categories Notes: 1. Each ' ‘*’ Statistically significant difference between the distribution of response to questions about satisfaction and importance; 2. Scale ranks from “very dissatisfied” (1) to “very satisfied” (7), 4 indicates “neutral”.

128

Resources Policy 53 (2017) 125–134

Q. Li et al.

Fig. 2. Mean scores relating to perceived importance of factors to overall wellbeing by location category Notes: 1. Each ‘*’ indicates that one of the paired observations was significantly different out of all paired comparisons across the 5 location categories; 2. Scale ranks from “very unimportant” (1) to “very important” (7), 4 indicates “neutral”.

impact (black line in radar chart) than did people living further away from coal mines. Importantly, no obviously positive impacts were identified by respondents from any of the five location categories. The green line shows the perceptions of people living far away from coal mines, who thus have little experience of coal mining's impacts. Although factors which related to the environment (e.g. air quality, water safety, access to water), the economy (e.g. inflation and real estate price) and health (physical and mental health) were all thought to be negatively affected by coal mining in all five location categories (mean score below 4), people far from coal mining seemed to underestimate some negative impacts of coal mining (e.g. the impacts on the quality of government and fairness of income) while exaggerating the potential positive impacts (on income and income disparities). In contrast, residents living in coal mining areas did not report an obviously positive impact of coal mining on

of proximity to mining. In marked contrast to satisfaction scores, for most factors, mean importance scores vary little across different location categories. It is noteworthy that residents living in rural settings located within 20 km of coal mining (Rural With and Rural Close) attached significantly greater importance to factors relating to air quality, water safety, inflation and prices of necessities than respondents from other types of residential areas. People living in coal mining areas were also more dissatisfied with those factors than people living in non-coal mining areas. 3.3. Perceptions of the impacts of coal mining on wellbeing factors Fig. 3 shows that for almost all factors, people living in rural areas with coal mining considered coal mining to have a more negative

Fig. 3. Mean scores relating to perceived impact of coal mining on wellbeing factors by location category. Notes: 1. Each ‘*’ indicates that one of the paired observations was significantly different out of all paired comparisons across the 5 location categories; 2. Scale ranks from “strongly negative” (1) to “strongly positive” (7), 4 indicates “no impact”.

129

Resources Policy 53 (2017) 125–134

Q. Li et al.

Fig. 4. The relationship between mean scores relating to dissatisfaction and importance Notes: Scale of satisfaction ranks from “very satisfied” (1) to “dissatisfied” (7), 4 indicates ‘neutral’; Scale of importance ranks from “slightly important” (3) to “very important” (7), 4 indicates “neutral”. No factors have score of ‘importance’ below 3; The red square captures all items with which people are dissatisfied and which they think are important.

3.5. The relationship between satisfaction level and the impact of coal mining

income, and reported negative impacts on income disparity. This aligns with the observation from section 3.1 above: There was no statistically significant difference among location categories in peoples’ satisfaction with family income, but there was significant difference among locations in people's satisfaction with income disparity. The factors related to the natural environment (air quality and water safety), which were believed to be most negatively affected by coal mining, were also those where the gap between satisfaction scores in coal mining and non-coal mining areas was greatest (see Fig. 1).

Fig. 5 displays the mean dissatisfaction of each factor plotted against its mean negative impact score for the coal mining areas. Only coal mining areas were included in this analysis. Factors relating to the environment and the economy, appear in the upper right quadrant of the graphs, indicating both high levels of dissatisfaction, and ‘blame’ being attributed to coal mining. Coal mining was also perceived to have negative impacts on others factors, such as health, but people seemed quite satisfied with these factors, suggesting that they might not be as high a policy priority as other factors (also, fixing environmental issues, could help mitigate at least some health issues). Table 4 presents the IDSNI scores associated with each factor, for each type of case-study area. Dust, air quality, air cleanliness, water safety, real estate prices and inflation were clearly identified as factors with the highest levels of dissatisfaction, on which coal mining is perceived as having a strong negative impact. These factors emerge at the top 5 of IDSNI ranking and are very consistent across all the coal mining areas.

3.4. The relationship between satisfaction and importance Fig. 4 shows the mean importance score associated with each factor, plotted against the mean Dis-Satisfaction score, using the whole sample. Generally, importance and dissatisfaction are inversely related (R2=0.216): People are less dissatisfied (more satisfied) with those factors deemed more important. Factors, such as those relating to the natural environmental (air quality, dust, air cleanliness and water safety) and the economy (inflation and real estate prices), receive high dissatisfaction scores and high importance scores. Generally, there are no factors that receive very high/low dissatisfaction scores but very low importance scores, which is consistent with the argument of Friedlander (1965) and Trauer and Mackinnon (2001). Table 3 presents the Indices of Dis-Satisfaction (IDS), for each factor, by location category. Strikingly, wellbeing factors with high IDS are different between coal mining areas (Urban close, Rural With and Rural Close) and non-coal mining areas (Urban Far and Rural Far). In coal mining areas, wellbeing factors relating to the natural environment (especially dust in the air) and the economy (inflation and the price of real estate prices) have the highest IDS, while in non-coal mining areas, environment indicators have relatively low indices; instead, real estate prices, inflation, property safety and education system emerge as policy priorities (with the higher indices). Real estate prices and inflation are common issues in all the places, with a slightly higher values of IDS in coal mining areas – indicating a bigger magnitude of problems in coal mining areas. Interestingly, wellbeing factors with low IDS are shared among all location categories.

4. Discussion No similar study could be found in a mining context with which to compare attitudes of respondents from different types of study region towards coal mining. However, similar observations have been found in a tourism context. Termed the “Irridex model”, this theory “postulates that resident's attitudes towards tourism are euphoric in the early stages, progressing to apathy, irritation and, eventually, antagonism” (Diedrich and García-Buades, 2009, p. 519). Similar effect of coal mining are particularly obvious in rural area, where the economy is less developed and less diverse than that in urban areas, and the available options for livelihoods is commensurably more limited (Kanbur and Zhang, 2005; Sicular et al., 2007). Thus, the prospect for higher incomes from mining is very attractive. Some residents in villages far from coal mining said: “We will be lucky if we have coal mining in our 130

Resources Policy 53 (2017) 125–134

Q. Li et al.

Table 3 IDSa for individual factors, arranged from highest to lowest score for each type of study region. Urban Close

Rural With

Rural Close

Urban Far

Rural Far

Wellbeing factors

IDS

Wellbeing factors

IDS

Wellbeing factors

IDS

Wellbeing factors

IDS

Wellbeing factors

IDS

Dust Water safety Real estate price Air quality Government Air cleanliness Inflation Property safety Education opportunity Price of necessities Fairness of income Education quality Income disparity Personal safety Family income

31.022 29.922 29.189 29.178 28.967 28.800 28.500 27.800 27.400 25.933 25.500 24.800 23.800 23.667 22.562

Dust Air quality Air cleanliness Inflation Real estate price Water safety Government Fairness of income Price of necessities Family income Property safety Income disparity Personal safety Education opportunity Education quality

39.561 38.065 37.416 35.692 30.813 29.612 28.308 28.168 27.678 25.967 25.598 25.523 24.037 22.935 22.701

Dust Inflation Air quality Air cleanliness Real estate price Government Fairness of income Family income Price of necessities Property safety Income disparity Water safety Education quality Education opportunity Housing

34.966 32.765 31.852 31.832 30.228 29.584 27.819 27.040 26.034 25.792 25.507 25.074 23.905 22.470 21.383

Real estate price Government Inflation Property safety Education opportunity Dust Fairness of income Family income Education quality Income disparity Personal safety Water safety Honesty Housing Air quality

29.367 27.567 26.667 25.933 25.333 22.333 21.600 21.000 20.667 19.933 19.800 19.633 19.633 19.567 18.467

28.288 27.458 27.051 26.000 25.169 23.492 23.288 22.186 22.051 20.458 19.780 18.797 18.690 17.814 17.644

Honesty

22.433

Water supply

22.505

21.201

Price of necessities

18.100

Trust Help

20.989 20.733

Housing Honesty

20.780 18.121

20.309 20.242

Personal physical health Air cleanliness

17.900 17.700

Electricity supply Honesty

16.322 16.322

Housing

20.522

17.958

20.121

Help

17.033

Dust

15.610

Water supply

20.111

17.621

Honesty

19.255

Trust

16.700

Trust

14.119

Transportation & communication Participation in social activity Personal physical health

19.189

Participation in social activity Transportation & communication Trust

Participation in social activity Trust Transportation & communication Personal safety

Real estate price Water supply Property safety Inflation Education quality Family income Education opportunity Government Housing Fairness of income Water safety Price of necessities Income disparity Personal safety Participation in social activity Personal physical health

17.033

Water supply

18.289

16.367

Family physical health

14.069

17.178

Help

16.879

Personal physical health

18.262

Transportation & communication Family physical health

16.233

Help

13.508

16.811

Personal physical health

15.869

Help

17.732

14.900

Personal mental health

13.458

Personal mental health Electricity supply Family physical health

15.878 15.789 15.611

Personal mental health Family physical health Electricity supply

15.682 15.603 14.631

Personal mental health Family physical health Family mental health

16.678 16.436 15.523

Participation in social activity Electricity supply Personal mental health Water supply

14.700 14.067 12.400

13.458 12.678 12.475

Family mental health Personal relationship Family relationship

14.767 12.267 9.722

Family mental health Personal relationship Family relationship

14.472 10.664 10.402

Electricity supply Personal relationship Family relationship

14.765 11.195 10.322

Family mental health Family relationship Personal relationship

12.200 11.600 10.000

Air cleanliness Air quality Transportation & communication Family mental health Personal relationship Family relationship

a

16.322

12.328 10.610 9.610

IDS is calculated using Eq. (1) in section 2.4. Higher value of IDS indicates factors of greater important and dissatisfaction.

situation. For example, people in the village with illegal coal mines (although not common) were likely to have stronger negative attitudes towards coal mining then those living in villages with legal coal mines. This is because, ‘Illegal’ coal mines, often involved with corruption, were reported to generate conflicts between the coal mine operators, the local governments and host communities, and often cause stronger negative impacts on local communities. The uneven distribution of benefits and costs could be well observed at individual level: People whose family income derived from coal mining clearly benefited from coal mining. In a village with coal mining, a respondent was hesitant to express herself openly: “I don’t want to say anything bad about coal mining, because my whole family is counting on it. If coal mines are closed here, my husband will lose his job, and we cannot make a living”. In contrast, those who were working in other industries (e.g. agriculture) in the coal mining areas, at times experience damages from coal mining that directly impinge upon their livelihoods and thus affect their wellbeing. A family reliant on agriculture obviously resented coal mining: “Coal mining brings me nothing but misery. A lot of my sheep died after they ate the grass covered by coal dust. Some of them fell down into a crack [in the ground] caused by coal mining, now I don’t have many sheep left”. However, the data collected by our questionnaire, part of which is reported in this study, indicate that the majority of local residents were unable to get jobs or income from coal mining. The proportion of family income derived from coal mining ranged from 0% (Urban Far and Rural far) to 32% (Rural with). Sampling locations classified as

village. Life will be better because we could get money from that”. Contrary to expectations, rural residents are likely to be negatively impacted by coal mining once they have it, revealed by the strongest dissatisfaction with and perceived impacts of mining on environmental and economic factors by residents of rural areas with coal mining. Using subjective wellbeing data, this study reveals that host communities do not seem to better off from coal mining in any aspects of life – there was barely a wellbeing factor with which people living in coal mining areas were more satisfied than people living in non-coal mining areas; people living in coal mining areas were significantly more dissatisfied with factors relating to the natural environment and the economy than those living in non-coal mining areas. These results substantiate concerns that in Shanxi, as elsewhere (Davis and John, 2005), host communities bear a disproportionate burden in facilitating national and regional development goals. This uneven distribution of benefits and costs from coal mining has been demonstrated by numerous studies in other part of China (Dai et al., 2014), mining in general in China (Lei et al., 2013), and mining in other countries (e.g. Australia (Carrington et al., 2011) and Tanzania (Kitula, 2006). The benefit flows largely to the central government and elsewhere, while local or host communities get small proportion of benefit from mining and bear most of the environmental and other social costs associated with mining (Davis and Tilton, 2005). The analysis above relies on aggregated data and thus describes the ‘overall’ situation in each location category. This does not mean that each village (or each individual within each village) is in the same 131

Resources Policy 53 (2017) 125–134

Q. Li et al.

Fig. 5. The relationship between mean scores relating to dissatisfaction and perceived impact of coal mining Notes: Scale of satisfaction ranks from “very satisfied” (1) to “dissatisfied” (7), 4 indicates ‘neutral’; Scale of negative impact ranks from “slightly negative” (3) to “strongly positive” (7); 4 indicates “no impact”. No factors have score of ‘negative impact’ below 3; The red square captures all items with which people are dissatisfied and which they think coal mining have negative impacts on.

companies hire miners from other regions is because of a perception that local residents are more ‘resilient’ and thus potentially less easily controlled or manipulated. Local residents have more family and community support, than miners brought in from elsewhere, and in

Urban Close and Rural Close derived 15% and 16% of their family income from mining. Anecdotal information collected during discussions with respondents when collecting data, suggests that one of the reasons coal

Table 4 IDSNIa for individual factors, arranged from highest to lowest score for three of the five location categories (Urban Far and Rural Far were excluded from calculating the IDSNI). Urban Close

Rural With

Rural Close

Wellbeing factor

IDSNI

Wellbeing factor

IDSNI

Wellbeing factor

IDSNI

Dust Air quality Air cleanliness Real estate price Water safety Inflation Fairness of income Price of necessities Government Income disparity Education opportunity Property safety Education quality Honesty Water supply Trust Help Personal safety Housing Transportation and communication Participation in social activity Personal physical health Family income Family physical health Personal mental health Electricity supply Family mental health Personal relationship Family relationship

31.60 29.07 28.99 27.21 26.88 24.72 23.22 22.30 21.62 21.31 20.80 19.17 18.79 18.04 17.30 16.79 16.49 16.16 15.27 14.75 14.74 13.35 13.13 11.98 11.33 11.09 10.45 8.60 5.94

Dust Air quality Air cleanliness Inflation Water safety Real estate price Fairness of income Government Income disparity Price of necessities Personal safety Property safety Water supply Housing Education opportunity Education quality Family income Participation in social activity Honesty Trust Help Personal physical health Family physical health Personal mental health Transportation and communication Family mental health Electricity supply Personal relationship Family relationship

39.27 37.57 37.20 27.10 25.49 24.20 23.14 21.91 21.54 20.82 18.80 18.66 18.19 17.60 16.78 16.49 15.96 14.35 14.20 13.86 13.15 13.07 12.86 12.83 12.19 11.14 9.17 7.34 6.62

Dust Air cleanliness Air quality Inflation Real estate price Water safety Government Fairness of income Price of necessities Income disparity Property safety Family income Participation in social activity Education quality Housing Education opportunity Water supply Honesty Trust Personal safety Transportation and communication Personal physical health Help Family physical health Personal mental health Family mental health Electricity supply Personal relationship Family relationship

30.01 27.41 27.26 26.19 23.64 22.09 21.30 20.79 20.43 20.34 16.54 16.41 16.40 15.97 15.27 15.11 14.43 14.22 13.86 13.84 13.55 13.05 12.37 11.88 11.51 10.59 9.14 7.24 6.17

a

IDSNI is calculated using Eq. (2) in section 2.4. Higher value of IDSNI indicate factors with which people are more dissatisfied and coal mining has stronger negative impacts on.

132

Resources Policy 53 (2017) 125–134

Q. Li et al.

People also expressed most concern about the impacts of mining on the environment (air quality and water safety) and health, which is consistent across all the case-study regions. Compared to people living in coal mining regions, people living in non-coal mining regions seemed to underestimate the negative impacts of coal mining while exaggerating the potential positive impacts. Respondents from rural areas with coal mining recorded the strongest perceptions about the negative impacts of mining on most wellbeing factors. Those with little experience of coal mining seemed to underestimate its potential negative impacts, and overestimate its positive impacts (in particular, impacts on their household income and on income disparities). All indicators, in particular the composed Index of Dis-Satisfaction (IDS) and Index of Dis-Satisfaction and Negative impacts (IDSNI), suggest that both environmental issues (air quality and water safety) and economic issues (the inflation rate and the price of real estate prices) were of high priority in coal mining areas, and environmental issues were paramount. Additionally, coal mining did not generate obvious benefits as people usually expected, in terms of income and jobs opportunities. Thus more benefits need to be allocate to host communities. The analysis of subjective data directly reveals public preference and helps identify policy priorities to mitigate the negative impacts of coal mining and improve the wellbeing of host communities. It justifies the use of subjective data in mining contexts, and demonstrates new approaches that could be used in the mining impacts assessment framework to better consider the welfare of host communities and improve the sustainability of mining industries.

the event of accidents or other disputes, they tend to demand higher compensation and are able to hold out for longer against the companies during negotiations. As such local residents may be more ‘costly’ and/ or more ‘difficult’ to deal with than migrant workers brought in from other areas. The distance of the individuals’ dwelling from coal mines or coal transportation routes also mattered. Those living closer to coal mines were more likely to encounter higher level of coal dust, house damage caused by regular explosions involved in coal mining. Those living beside roads used for coal transportation were strongly exposed to noise of heavy trucks than those living further away. The remarks of a villager succinctly summarize the focus of this study: “I felt lucky that we don’t have coal mines in our village, so I still have my farmland, I have been doing odd jobs with an unsatisfying income in the city, but my farmland makes me feel secure. Rumour has it that coal has been found underground in my village. I would definitely be the first person opposed to coal mining here, because ordinary people like me could not benefit from coal mining, only those officers or rich people can. If they start mining here, the only consequences will be that we will lose farmland and suffer from all the environmental damage from coal mining. Even now, it already has impacts on us. In the farmland beside the special road for coal transportation, the corncobs are black inside. But there was no compensation for that”. There is also much more diversity at an individual level. In particular, people's satisfaction with various wellbeing factors, the importance which individuals attach to wellbeing factors, and their perceptions about the impacts of coal mining on those factors may depend upon a host of other factors, such as ‘personality’, sociodemographic characteristics (e.g. age (Brereton et al., 2008), gender (Arifwidodo and Perera, 2011) and genetic factors (Zidanšek, 2007)) and economic status (e.g. income (Ferrer-i-Carbonell, 2005)) of individuals. Future studies might seek to control sociodemographic characteristics. The diversity of perspectives clearly indicates that regional or even individual specific policies are desirable rather than a single “standard” policy towards all mining across all regions within a single country. Nevertheless, the data reveal several common issues associated with coal mining which might be the policy priorities to mitigate the negative impacts of coal mining and minimize the externalities exerted by coal mining on host communities.

Acknowledgments The authors thank the research assistants, hundreds of anonymous interviewees and anonymous reviewers for reviewing this paper. References Abdallah, S., Mahony, S., Marks, N., Michaelson, J., Seaford, C., Stoll, L., THOMPSON, S., 2011. Measuring our Progress: the power of well-being. The New Economics Foundation (nef). Arifwidodo, S.D., Perera, R., 2011. Quality of life and compact development policies in Bandung, Indonesia. Appl. Res. Qual. Life 6, 159–179. Brereton, F., Clinch, J.P., Ferreira, S., 2008. Happiness, geography and the environment. Ecol. Econ. 65, 386–396. Carrington, K., Hogg, R., Mcintosh, A., 2011. The resource boom's underbelly: criminological impacts of mining development. Aust. N.Z. J. Criminol. 44, 335–354. Chen, S.-K., Lin, S.S., 2014. The latent profiles of life domain importance and satisfaction in a quality of life scale. Social. Indic. Res. 116, 429–445. Cummins, R.A., 1996. The domains of life satisfaction: an attempt to order chaos. Social. Indic. Res. 38, 303–328. Cummins, R.A., 2003. Normative life satisfaction: measurement issues and a homeostatic model. Social. Indic. Res. 64, 225–256. Cummins, R.A., Nistico, H., 2002. Maintaining life satisfaction: the role of positive cognitive bias. J. Happiness Stud. 3, 37–69. Dai, G., Ulgiati, S., Zhang, Y., Yu, B., Kang, M., Jin, Y., Dong, X., Zhang, X., 2014. The false promises of coal exploitation: How mining affects herdsmen well-being in the grassland ecosystems of inner Mongolia. Energy Policy 67, 146–153. Davis, G.A., Tilton, J.E., 2005. The resource curse. Natural resources forum. Butterworths, London, 233–242. Davey, G., Rato, R., 2012. Subjective wellbeing in China: a review. J. Happiness Stud. 13, 333–346. Diedrich, A., García-Buades, E., 2009. Local perceptions of tourism as indicators of destination decline. Tour. Manag. 30, 512–521. Diener, E., Suh, E., 1997. Measuring quality of life: economic, social, and subjective indicators. Social. Indic. Res. 40, 189–216. Editor of Land & Resource Herald, 2013. Shanxi : Coal sales revenue exceeded one trillion yuan in 2012. Land & Resource Herald. Emmons, R.A., Diener, E., 1985. Factors predicting satisfaction judgments: a comparative examination. Social. Indic. Res. 16, 157–167. Ferrer-I-Carbonell, A., 2005. Income and well-being: an empirical analysis of the comparison income effect. J. Public Econ. 89, 997–1019. Larson, S., 2010. Can the concept of human wellbeing help identify regional policy priorities?. Doctor of Philosophy, James Cook University. Larson, S., Stoeckl, N., Farr, M., Esparon, M., 2014. The role the Great barrier reef plays in resident wellbeing and implications for its management. Ambio, 1–12. Larson, S., Stoeckl, N., Neil, B., Welters, R., 2013. Using resident perceptions of values associated with the Australian Tropical Rivers to identify policy and management

5. Conclusion This research aimed to investigate the impacts of coal mining on the wellbeing of host communities in Shanxi Province of China using subjective indicators. Our research used subjective wellbeing indicators for three reasons: Firstly, the discrepancies between objective and subjective indicators reinforce the need for the parallel development of both sets of indicators, while objective indicators currently are dominant and subjective indicators have rarely been used in the mining contexts. Secondly, it has been repeatedly demonstrated that objective indicators of quality of life and wellbeing do not adequately capture peoples’ daily experience (Sarracino and Bartolini, 2015). Thirdly, subjective indicators of wellbeing tend to be outside the narrow pursuit of GDP driven growth and concurrent objective measures of development with the underlying assumption that aggregate growth in GDP results in citizens’ increased wellbeing (Tang, 2011). We found that people in coal mining areas were strongly dissatisfied with and attached great importance to wellbeing factors relating to the natural environment (water safety and air quality) and the economy (the inflation rate and real estate prices). In contrast, people in noncoal mining areas were generally satisfied with factors relating to the environment and did not attach great importance to those indicators. 133

Resources Policy 53 (2017) 125–134

Q. Li et al.

Kanbur, R., Zhang, X., 2005. Fifty years of regional inequality in China: a journey through central planning, reform, and openness. Rev. Dev. Econ. 9, 87–106. Kitula, A., 2006. The environmental and socio-economic impacts of mining on local livelihoods in Tanzania: a case study of Geita District. J. Clean. Prod. 14, 405–414. Hendryx, M., Ahern, M.M., 2009. Mortality in Appalachian coal mining regions: the value of statistical life lost. Public Health Rep. 124, 541. Oecd, 2002. Coal and sustainable develoment-achieve balance in priorities. OECD, Paris. Oecd, 2013. OECD Guidelines on Measuring Subjective Well-being. OECD, Paris. Oswald, A.J., Wu, S., 2010. Objective confirmation of subjective measures of human wellbeing: evidence from the USA. Science 327, 576–579. Rablen, M.D., 2012. The promotion of local wellbeing: a primer for policymakers. Local Econ. 27, 297–314. Sarracino, F., Bartolini, S., 2015. The dark side of chinese growth: declining social capital and well-being in times of economic boom. World Dev. 74, 333–351. Schneider, M., 1975. The quality of life in large American cities: objective and subjective social indicators. Social. Indic. Res. 1, 495–509. Sicular, T., Ximing, Y., Gustafsson, B., Shi, L., 2007. The urban–rural income gap and inequality in China. Rev. Income Wealth 53, 93–126. Tang, H., 2011. How to make China Happy. China dialogue [online]. Available: 〈https:// www.chinadialogue.net/article/show/single/en/4116-How-to-make-China-happy〉 (accessed on 10 May 2016). Veenhoven, R., 2002. Why social policy needs subjective indicators. Social. Indic. Res. 58, 33–46. Welsch, H., 2006. Environment and happiness: valuation of air pollution using life satisfaction data. Ecol. Econ. 58, 801–813. Zhang, A., Moffat, K., Boughen, N., Wanf, J., Cui, L., Dai, Y., 2015. Chinese attitudes toward mining: citizen survey –2014 Results. CSIRO, Australia, (EP 151270). Ziersch, A.M., Baum, F., Darmawan, I., Kavanagh, A.M., Bentley, R.J., 2009. Social capital and health in rural and urban communities in South Australia. Aust. N.Z. J. Public Health 33, 7–16. Zidanšek, A., 2007. Sustainable development and happiness in nations. Energy 32, 891–897.

priorities. Ecol. Econ. 94, 9–18. Lee, T., Marans, R.W., 1980. Objective and subjective indicators: Effects of scale discordance on interrelationships. Social. Indic. Res. 8, 47–64. Lei, Y., Cui, N., Pan, D., 2013. Economic and social effects analysis of mineral development in China and policy implications. Resour. Policy 38, 448–457. Li, Z., Nieto, A., Zhao, Y., Cao, Z., Zhao, H., 2012. Assessment tools, prevailing issues and policy implications of mining community sustainability in China. Int. J. Min., Reclam. Environ. 26 (2), 148–162. Li, Q., Stoeckl, N., King, D., Gyuris, E., 2017. Using both objective and subjective indicators to investigate the impacts of coal mining on wellbeing of host communities: a case-study in Shanxi Province, China. Soc. Indic. Res., 1–27. http:// dx.doi.org/10.1007/s11205-017-1624-2. Lu, M., Chen, Z., 2004. Urbanization, urban-biased economic policies and urban-rural Inequality [J]. Econ. Res. J., 6. Mackerron, G., Mourato, S., 2009. Life satisfaction and air quality in London. Ecol. Econ. 68, 1441–1453. Moffatt, K., Boughen, N., Zhang, A., Lacey, J., Fleming, D., Uribe, K., 2014a. Chilean attitudes toward mining: citizen Survey – 2014 Results. CSIRO, Australia. EP, 147205. Moffatt, K., Zhang, A., Boughen, N., 2014b. Australian attitudes toward mining: citizen Survey – 2014 Results. CSIRO, Australia. EP, 147205. Moffatt, S., Pless-Mulloli, T., 2003. “It wasn’t the plague we expected.” Parents’ perceptions of the health and environmental impact of opencast coal mining. Social. Sci. Med. 57, 437–451. Morrice, E., Colagiuri, R., 2013. Coal mining, social injustice and health: a universal conflict of power and priorities. Health Place 19, 74–79. Noronha, L., 2001. Designing Tools to Track Health and Well-being in Mining Regions of India. Natural Resources Forum. Wiley Online Library, 53–65. China Energy Information Network, 2009. Basic facts of Coal resources in Shanxi [Online]. Available: 〈http://www.sxcoal.com/video/62.html〉 (accessed on 10 March 2016). Hofferth, S.L., Iceland, J., 1998. Social capital in rural and urban communities. Rural Sociol. 63, 574–598.

134