Geography of Happiness: A Spatial Analysis of Subjective Well-Being 3031198700, 9783031198700

Citation preview

Contributions to Regional Science

Eric Vaz Editor

Geography of Happiness A Spatial Analysis of Subjective Well-Being

Contributions to Regional Science

This book series offers an outlet for cutting-edge research on all areas of regional science. Contributions to Regional Science (CIR) welcomes theoretically sound and empirically robust monographs, edited volumes and handbooks from various disciplines and approaches on topics such as urban and regional economics, spatial statistics, spatial econometrics, geographical information systems, migration analysis, land use and urban development, urban and regional policy analysis, interindustry analysis, environmental and ecological analysis, and related fields. All books published in this series are peer-reviewed.

Eric Vaz Editor

Geography of Happiness A Spatial Analysis of Subjective Well-Being

Editor Eric Vaz Toronto Metropolitan University Toronto, ON, Canada

Contributions to Regional Science ISBN 978-3-031-19870-0 ISBN 978-3-031-19871-7 (eBook) https://doi.org/10.1007/978-3-031-19871-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Contents

Happiness Geography: Defining the Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . Eric Vaz

1

Regional Challenges A Spatial Analysis of the Instagram Hashtag #happy: An Assessment of Toronto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Eric Vaz The Subjective Well-Being in North Africa and the Impacts on Agriculture and Urban Land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Azzeddine Bellout, Eric Vaz, Antonia Bousbaine, and Christopher R. Bryant

11

33

Territorial Challenges Researching Quality of Life in Old Age: Some Conceptual and Methodological Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . José de São José

51

Peripheral Retail Expansion: Social Implications and Spatial Inequalities the Case of the Île-de-France Region . . . . . . . . . . . . . . . . . . . . . André Torre and Océane Peiffer-Smadja

73

Locational Challenges Sustainable Cities, Quality of Life, and Mobility-Related Happiness . . . . 103 Karima Kourtit, Peter Nijkamp, and Marina Toger

v

vi

Contents

Tourism, Climate Change and Well-Being: The Products’ Diversity as an Opportunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 André Samora-Arvela, Eric Vaz, Jorge Ferreira, and Thomas Panagopoulos Tourism, Senses and Well-Being . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Dora Agapito

Happiness Geography: Defining the Field Eric Vaz

1 Introduction Recent advances in spatial analysis have led to a growing interest in applying geocomputation methods to complex issues (Du et al., 2020), often defined over large datasets framing socioeconomic variables (Ekbia et al., 2015) and the natural environment (Bryan et al., 2011). Much of this work has been carried out to optimize decision-making and lead to a better quality of life, economic growth and social stability (Higgins et al., 2014; Rosu et al., 2015; Ahlfeldt et al., 2020). Thanks to the advances of computational power, spatial analysis has reached a new stage, where nonlinear modelling approaches combined with stochastic modelling allow for a better understanding of the geographical environment (Silva et al., 2020) and therefore foster more accurate decision-making (Carver, 2019). This decision-making has focused predominantly on sustainable development, leading to better life quality, where spatial information on the natural environment has brought a key role, particularly in ecology, to understand the present and offer a more sustainable future. From an anthropocentric perspective, wellbeing is the relation of humankind’s social, economic, and environmental stability, maximizing the opportunities for sustainable development while leading to better life quality. This holistic approach of wellbeing closely reminds a much deeper concept found in Aristotelian Ethics. From a positivistic posture, understanding what we define as happiness is subjective and explores individual wellbeing within space and time (Houlden et al., 2019; Kosanic & Petzold, 2020; Shekhar et al., 2019). It is thus highly uncertain, difficult to measure socially, and even harder to quantify and explain. Nevertheless, the advances in complex system science and the exponential growth of big data and the possibility of tracking individual data have led to a resounding interest E. Vaz (B) Department of Geography and Environmental Studies, Toronto Metropolitan University, Toronto, ON, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 E. Vaz (ed.), Geography of Happiness, Contributions to Regional Science, https://doi.org/10.1007/978-3-031-19871-7_1

1

2

E. Vaz

in the integration of complex systems to define a spatial understanding of happiness. The centre stage where the action of happiness plays is geography (Brereton, 2008; Stanca, 2010; Ballas, 2020; Mavruk et al., 2021). The key players are surrounded by the myriad of data sources that intertwine with computational methods that elaborate on defining the field of happiness geography. The choice of variables related to wellbeing is chosen in three fundamental dimensions: Social, economic and environmental, and adapted to the spatial dimension of administrative boundaries that lead to the geography of happiness (David et al., 2014). The combination and spatial dynamics of the patterns found in the analysis of spatial wellbeing, and mostly, the potential of harnessing individual data and ongoing spatial and temporal response rates, leads us a step closer to quantifying happiness at a regional scale (Morrison, 2021). Such a field may well be defined as a novel contribution to the existing ideations of subjective happiness, where a new kind of spatial rationale leads to optimizing the geographical and planning environment of future regions and cities (Wang & Wang, 2016).

2 On Happiness as a Consequence of Sustainable Development In detriment to economic growth, much of our landscape has changed dramatically over the last decades. With the unprecedented effects of urbanization and population increase, human beings have significantly jeopardized the natural environment, heritage landscapes, and the stability of itself as a keystone species. The increasing competition and economic growth have led to a neglect of the natural environment, and since Brundtland raised the need for strategies for common goals of sustainable development. However, there is no consensus on what this sustainable development entails, particularly given the complexity of the very notion of what could be discarded and what should be kept for future generations (O’Neill & Kahn, 2000). Scholars agree, however, that sustainable development is crucial for the very survival of our species and that humankind must find a solution to readjust itself to its surrounding environment (Garibaldi & Turner, 2004). One of the culprits of environmental and landscape degradation has been the excessive growth human being has generated (Vaz, 2016). From an economic standpoint, more significant asymmetries between social strata lead to loss of resources from a Schumpeterian perspective (Puscaciu et al., 2016). In other words, society has taught us that only a small nucleus of people remains and holds wealth, while the large majority of humankind lives in challenging misery. There is a clear pattern of poverty, crime, and wealth at the spatial level, clearly traceable to the root of hubs of economic growth, such as urban areas. Therefore, urbanization processes are a logical consequence of economic prosperity, and while cities grow, infrastructures spoil the already environmentally depleted land use, where natural resources become scarce. This sinks process has led to new techniques that enable better monitoring and understanding of the spatial dimension,

Happiness Geography: Defining the Field

3

supporting the planning of urban and rural areas, and defining the socioeconomic strata from a geographic perspective, leading to more sustainable and sounder environments. Wellbeing is closely related to this very concept, as it assumes social and economic stability and integration of awareness and harmony with the natural environment. The Aristotelian notion of happiness furthermore expands on the idea that happiness is an absolute objective for humankind, only reflected in the existence of a common goal. From a sustainable development perspective, this common goal may become the attempt to transmit to future generations our social, economic, and natural environment, without any negative externalities that create dramatic changes, but offer a stable system of growth to the generations to come.

3 Spatial Information for Complex Decision-Making With the recent technological development witnessed in the last decades, computational systems have increased dramatically their ability to cope with large amounts of data and offering more complex analysis of data sets. This has brought significant advances in geocomputation, where the assembly, editing, and manipulation of data allow at present much more elaborate computational tasks than a decade ago. A large volume of geostatistical and spatial analysis methodologies has arisen from the possibility of understanding complex interactions, leading to advances in nonlinear modelling approaches that follow out of the traditional body of knowledge of statistical toolsets mainly available until the nineties. These new methods embed the possibility to observe behaviours and predict motion over space and time. It is only natural that the traditionally applied visions such as von Neuman and the scholarly work found in genetics have become interesting paths for the advances in spatial analysis and observation of empirical outcomes. One of these methods links to cellular automata, which have primarily been used to prospecting spatial phenomena such as land-use change. The very nature of land-use change modelling utilizing stochastic processes enabled in cellular automata shows that combined methodologies of geographical analysis embed well into the stream of complex systems, converging to the advances in the field of geographic information science. The cellular automata itself derives from a straightforward concept. A single cell iterates its state given the fact that growth is the produce of its adjacency. The anachronism further explores this simple growth mechanism known as the cell iteration over time that the future state of the cell is represented by a transition probability, in other words, the mechanics on how plausible it is for the cell to grow adjacently, given the probability of any other cell becoming similar to its adjacent type. Adding different categories to this, the cells take action to form certain determinants of change over space and time. In all its simplicity, this model represents a compelling concept of nonlinearity that has been widely used for planning purposes in the field of spatial analysis. It is therefore intriguing that the usability of this method has not yet been applied to other fields of social science, and can only be a result of the little exploration still at hand of complex systems modelling in line with fields such as the

4

E. Vaz

study of happiness, where some geostatistical methods have been shared especially in the field of wellbeing. Happiness is not relative, and neither is it ambiguous. It has been shown that (a) people tend to be unhappy under adverse conditions such as poverty, war, and isolation, (b) improvement or deterioration of at least some conditions does affect happiness lastingly, (c) earlier hardship does not favour later happiness, (d) people are typically positive about their life rather than neutral. The confusion, as argued, is between content and happiness. And the role of wellbeing is not restrictive to offering a spatial determinism to the concept of happiness at a spatial level, where environmental, social, and economic factors equate systemic happiness possible to be quantified, assessed, and translated into a planning practice benefits decision-making and society. This very transversal nature of what happiness represents makes it more attractive to be evaluated from a spatially explicit perspective, extending the assessments at the country level as proposed by ___ into local and regional level assessments of relative happiness using nonlinear dynamics. This paper defines exactly this: A combined method to assess regional and local happiness using an intricate connection between Markov Transition Chains, cellular automata, and scenario forecasting. It is shown that happiness, as represented, tends to shrink in more urban areas, suggesting that variables such as natural environment do play a crucial role in local and regional happiness. In regions such as the Greater Toronto Area, where urban growth is an unavoidable reality, happiness has an innate tendency to shrink, showing that by 2020, where forecasting is built upon in terms of cellular automata, urban regions in the downtown Toronto core have less propensity of having higher values of geographically defined local and regional happiness.

4 Book Organization The geography of happiness has been of growing interest with the advance of geocomputation and big data analysis. Drawing from well-established works found in economics from the 1970s, such as Easterlin (1974) addressed the potential of measuring the traditionally complex task of measuring individual utility. This has allowed to shift gears in terms of the application of individual utility maximization. Such research opened the opportunity to address locational aspects of individuals, their surrounding environment, and the importance of measuring the empirical outcomes of happiness data not as a complex set of understanding individual satisfaction. As a result, the eudemonic conceptualization of happiness becomes more likely to gain grounds than previously explored through the conceptualization of subject wellbeing. With the advances of big data and novel quantitative techniques, this fosters a unique opportunity to explore the spatially explicit aspects of happiness and better combine geography as a field that explores actively issues related to happiness and subjective wellbeing. The book shows the complexity of happiness geography within the larger framework of public policies and regional decision-making. The diversity of fields explored in this contribution creates the different dimensions that pave a clearer decision

Happiness Geography: Defining the Field

5

towards happiness geography studies and the multiple dimensions that this field attains within regional and governmental policy structures. Embedded in this is the unique attribution of geography as a key driver towards the exploration of distinct geographical frameworks that enable regional decision-making and planning towards knowledge that better addresses quality of life issues. Illustrated by the contemplation of regional intelligence and explored within the scalable determinants of geographical analysis, the instruments and tools provided in this book abridge the utmost critical dimensions of regional challenges that may be mitigated within the Anthropocene. The reason that the Anthropocene is worth studying in the context of happiness geography is one of the availability of data. Chapter 1 introduces the book by Eric Vaz, defining the importance of happiness studies within geography. The concept of happiness geography is novel and deserves exploration through chapters that holistically emerge to define the field itself. Part I, entitled Landscape Challenges, brings empirical and theoretical frameworks illustrating the importance of regional level, governance, and public policy for subjective wellbeing and assessing the opportunity of happiness to be measured at a geographical scale. The pressing issues of environment and ecosystem and landscape services in times of pressing change within the Anthropocene. Starting with Chap. 2, the opportunity of big data and novel social network tools such as Instagram showcase the relation of individual data points that can be geographically explored to understand subjective wellbeing. It furthers the relevance of using novel approaches both for regional science and happiness studies in harnessing these datasets as a novel way to obtain large data that grounds an intrinsically subjective field. Eric Vaz showcases the distribution of happiness hashtags in the city of Toronto. A city has increasingly grown and faces challenges of land use and urbanization in the coming decades. Chapter 3 continues the discourse of cities, extending the investigation to the impacts of subjective wellbeing in the developing world. Azzeddine Bellout, Eric Vaz, Antonia Bousbaine, and Christopher Bryant establish a link between the perception of wellbeing and the surrounding urban landscape contextualizing within the impacts of agricultural land and sustainable agricultural solutions as key drivers for subjective wellbeing, anchored in the complex need for happiness. Both Chaps. 2 and 3 illustrate the utmost importance of understanding regional dynamics in the face of changing environments during the Anthropocene. Distilling on empirical examples, it becomes evident that regional analysis is of significant relevance to create happier communities. The role of sustainable development and regional land use becomes patent in both these studies. While regional studies concerning happiness are resoundingly crucial for public policies and governance, Part 2 of the book entitled Regional Challenges addresses the demographic perspective of happiness geography. In Chap. 4, José São José approaches age and quality of life, opening the narrative towards the importance of understanding the integration of age within the sustainable patterns of life quality and thus the larger dimension of happiness geography. While several regions worldwide are changing their demographics, it is vital to understand the geographical impact of this change and its impact on the issues of quality of life and public policy

6

E. Vaz

agendas. Advancing this issue, Chap. 5, investigates the impacts of spatial inequalities are measured within the framework of territorial governance by André Torre and Océane Peiffer Smadja. They offer an insightful perspective on spatial deprivation in Île-de-France Paris region. Similar to the findings in Part 1, the conclusions of Torre and Smadja are related to the impacts of social and spatial inequalities and its implications of land, assessing the direct effects of change over several decades of retail. Part 2 points out the importance of regional perception of land-use dynamics but furthers the issue within a socioeconomic context of regional and global change. Happiness indeed becomes a construct of society and a result of the proper articulation of land use and mostly the integration through governance structures of sustainable development brought by understanding intrinsic factors within society, such as age, and recognizing extrinsic factors such as surrounding environment and policymaking. The ability to predict happiness becomes clear. It is a combination of quality of life within a robust understanding of our own sustainability as species. Part 3, entitled Economic Challenges, furthers this debate abridging the local aspects and the immediate call for action of sustainability and the promotion of our own happiness and wellbeing. Chapter 6, by Karima Kourtit, Peter Nijkamp, and Marina Toger venture in the importance of cities as a source for liveability and quality of life. The several urban challenges at hand pose the importance of assessing local dynamics and the relevance of mobility in the integration of geography. The authors discuss the new urban world within its challenges and adapt and become resilient by heralding happiness at the local level. Chapter 7 by André Samora Arvela, Eric Vaz, Jorge Ferreira, and Thomas Panagopoulos offer insights on the issues of wellbeing within the context of tourism. Of particular interest is the degradation of coastal environments and the issues on the dynamics of local change. Within the Mediterranean region, the authors call for localized studies that use a proactive engagement to reassign the tourism industry to a functional governance structure that equates its people’s sustainability and wellbeing. Chapter 8 by Dora Agapito closes the book employing addressing the importance of senses in the tourism industry. We are reminded how global and local governance structures and the intrinsic experience of individuals within a larger framework of regional structures create the experience of happiness using hedonic and eudaimonic outcomes. The geography of happiness becomes distinct through addressing the individual experience in the larger context of global and regional issues. Our individual experience offers the opportunity to integrate policy and governance the unique opportunity to witness sustainable development and regional sustainability with attention to land use, a chance to frame happiness geography as a field. This book takes a multi-dimension approach, a stark contrast to the current literature tying geography to happiness studies in the past. This approach has, in recent years, gained some traction, as it has become clear that the factors to model happiness are several and, mostly, ubiquitously available depending on a spatial, temporal, and geographical scale that is being examined. Regional scientists play a significant role in shaping the field, especially given their ability to look within different scales. The methods in regional science are indeed the most suitable not only to model the complexity of happiness but further establish a narrative that paves the way for a

Happiness Geography: Defining the Field

7

consistent analysis of new tools and techniques that embody spatial decision-making and territorial performance. In line with the requirement of local governance, regional decision-making exposes a dialogue between what truly is happiness and how policymakers and stakeholders may integrate these aspects at regional scales of interaction. Gravitating towards governance structures, the geographical determinants that lead to happier regions promote liveability and the wishes of any democracy: justice, equity, and individual freedom. The opportunity to understand happiness as a pivotal player in regional science merges the role of data-rich environments of the future, where considering the aspects of environmental sustainability, the promotion of happiness stands over the idea of subjective wellbeing, which has been amply discussed in economics. The difference indeed seems to shape the integration of a less economical and more social and environmental vision of wellbeing, wherein one concludes that the complexity of happiness is a result of a growing interest of happiness within the Anthropocene. With the many challenges regional development faces in times of unprecedented change, happiness geography remains an undeterred pathway to a more sustainable and equitable future where public participation, governance, and social and economic systems are part of a more extensive system of regional development and growth.

References Ahlfeldt, G. M., Bald, F., Roth, D., & Seidel, T. (2020). Quality of life in a dynamic spatial model. Available at SSRN 3751857. Ballas, D. (2020). The economic geography of happiness. In Handbook of labor, human resources and population economics (pp. 1–24). Brereton, F., Clinch, J. P., & Ferreira, S. (2008). Happiness, geography and the environment. Ecological Economics, 65(2), 386–396. Bryan, B. A., Crossman, N. D., King, D., & Meyer, W. S. (2011). Landscape futures analysis: Assessing the impacts of environmental targets under alternative spatial policy options and future scenarios. Environmental Modelling & Software, 26(1), 83–91. Carver, S. (2019). Developing web-based GIS/MCE: Improving access to data and spatial decision support tools. In Spatial multicriteria decision making and analysis (pp. 49–76). Routledge. David, S. A., Boniwell, I., & Ayers, A. C. (Eds.). (2014). The Oxford handbook of happiness. Oxford University Press. Du, P., Bai, X., Tan, K., Xue, Z., Samat, A., Xia, J., et al. (2020). Advances of four machine learning methods for spatial data handling: A review. Journal of Geovisualization and Spatial Analysis, 4(1), 1–25. Easterlin, R.A. (1974). Does economic growth improve the human lot? In: P. A. David, & M. W. Reder (Eds.), Nations and households in economic growth: Essays in honor of Moses Abramovitz. Academic. Ekbia, H., Mattioli, M., Kouper, I., Arave, G., Ghazinejad, A., Bowman, T., et al. (2015). Big data, bigger dilemmas: A critical review. Journal of the Association for Information Science and Technology, 66(8), 1523–1545. Garibaldi, A., & Turner, N. (2004). Cultural keystone species: Implications for ecological conservation and restoration. Ecology and society, 9(3).

8

E. Vaz

Higgins, P., Campanera, J., & Nobajas, A. (2014). Quality of life and spatial inequality in London. European Urban and Regional Studies, 21(1), 42–59. Houlden, V., de Albuquerque, J. P., Weich, S., & Jarvis, S. (2019). A spatial analysis of proximate greenspace and mental wellbeing in London. Applied Geography, 109, 102036. Kosanic, A., & Petzold, J. (2020). A systematic review of cultural ecosystem services and human wellbeing. Ecosystem Services, 45, 101168. Mavruk, C., Kıral, E., & Kıral, G. (2021). Spatial effects over time-framed happiness. Journal of Happiness Studies, 22(2), 517–554. Morrison, P. S. (2021). Wellbeing and the region. In Handbook of regional science (pp. 779–798). O’Neill, R. V., & Kahn, J. R. (2000). Homo economus as a keystone species. BioScience, 50(4), 333–337. Puscaciu, V., Puscaciu, F. D., & Puscaciu, R. M. (2016). Symmetries and Asymmetries in the Sustainable Development of European Union versus Romania. Acta Universitatis Danubius. Œconomica, 12(2). Rosu, L., Corodescu, E., & Blageanu, A. (2015). Does geographical location matter? Assessing spatial patterns in perceived quality of life in European cities. European Journal of Geography, 6(2), 15–34. Shekhar, H., Schmidt, A. J., & Wehling, H. W. (2019). Exploring wellbeing in human settlements-A spatial planning perspective. Habitat International, 87, 66–74. Silva, E. A., Liu, L., Kwon, H. R., Niu, H., & Chen, Y. (2020). Hard and soft data integration in geocomputation: Mixed methods for data collection and processing in urban planning. In Handbook of planning support science. Stanca, L. (2010). The geography of economics and happiness: Spatial patterns in the effects of economic conditions on well-being. Social Indicators Research, 99(1), 115–133. Vaz, E. (2016). The future of landscapes and habitats: The regional science contribution to the understanding of geographical space. Habitat International, 51, 70–78. Wang, F., & Wang, D. (2016). Place, geographical context and subjective well-being: State of art and future directions. In Mobility, sociability and well-being of urban living (pp. 189–230).

Regional Challenges

A Spatial Analysis of the Instagram Hashtag #happy: An Assessment of Toronto Eric Vaz

1 Introduction 1.1 Place-Based Happiness In recent years, the role of geography concerning happiness and subjective wellbeing has gained increasing interest, more so after the recent pandemic and in light of growing environmental change (Qasim & Grimes, 2022; Zacher & Rudolph, 2021). However, within the field of geography, literature has predominantly focused on the interaction of economic growth and well-being indicators (Hagerty & Veenhoven, 2003; Tella et al., 2003) leading to a gap of literature that directly addresses the subject of place-based happiness studies, such as the divide between urban and rural regions (Burger et al., 2020). Many epidemiological studies that gather locational data have paved the way to understanding indicators of self-harm, depression better, aging, and environmental injustice (George, 2010; Helliwell et al., 2020; Lew et al., 2019). This is surprising, as those antipodes of happiness seem to cluster clearly over geographical space. Quantitative techniques thus can incrementally be used according to location determinants, where locational data may have an outstanding role in comprehending the symbiosis of geographical and environmental factors shaping happiness (Brereton et al., 2008). Available socioeconomic variables pave the way for successful analysis of the geographical understanding of happiness further (Ballas & Dorling, 2013; Conceição & Bandura, 2008). Knowledge driven by census data and particularly geodemographics allows to profile and contextualize at a neighborhood scale the impacts of drivers such as income, employment, and age, merged with environmental quality, sustainability, and ecology to shape a deeper understanding of how communities are intrinsically satisfied with their lives, social contexts, and neighborhoods (Pfeiffer & Cloutier, 2016). The accumulation of wealth in this sense is far from being E. Vaz (B) Department of Geography and Environmental Studies, Ryerson University, Toronto, ON, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 E. Vaz (ed.), Geography of Happiness, Contributions to Regional Science, https://doi.org/10.1007/978-3-031-19871-7_2

11

12

E. Vaz

a unique determinant for the complexity of happiness, and it is perfectly justifiable to look at endogenous factors that shape the hedonic understanding of happiness (Ferreira & Moro, 2010; Graham, 2005). This has been a trend in recent literature on subjective well-being and happiness. While the terminologies have intermittently been used, the contextual analysis largely explores these as drivers to conceptualize and illustrate the optimization of happiness within a confined regional space. In recent years, most studies have focused on leveraging comparing indicators between nations and thus deterministically justify how planning and public policy lead to a country-scale interpretation of happiness. However, these studies are intrinsically aspatial, as they do not consider the local and regional indicators but look only at the global level. While such studies harness an essential value for international policy, they present pictures of the limited scope for integration and interaction of happiness studies at a local level and do not offer place-based solutions where regional impacts can have a profound role in leading to consistent growth of happiness. As we shape our cities to become more resilient, efficient, and smarter, it is of utmost importance to use locational data to measure happiness within the city. Such empirical studies aid in better understanding the context of inequality and injustice as potential drivers find common associations of happiness. This is the role of public policy and public health while promoting a society with higher subjective well-being indicators that directly impact livability, health, and happiness. Harnessing this data is not an easy feature, as local-scale analyses hardly show subjective happiness. The eventual modifiable areal unit problem (MAUP) may lead to a critical misinterpretation of socioeconomic results throughout the geographical interpretation of happiness (Davern et al., 2017; Ye et al., 2020). Therefore, it is relevant that spatial data dealing with happiness is harnessed over a larger body of sensed data from social media, where information can become an unbiased resource to understand the determinants of spatial allocation of happiness.

1.2 The Role of Geospatial Technologies The advances of geospatial technologies have had an increasing role in the usability of hand-held devices in particular to measure younger social groups (Collins et al., 2009). Spatially explicit content has grown significantly since the incorporation of global position system units in smartphones due to cheaper assembly processes harvested by the power of the crowd (Elwood et al., 2012). As digital content and technological advance is becoming progressively spatially enabled, this allows for a growing number of studies to be performed sharing important findings in the fields of social sciences (Anselin, 1999). The contributors of spatially explicit content are growingly becoming the users themselves, helping thus directly in the fields of volunteered geographic information, spatial crowdsourcing, and location-enabled decision systems. This consolidates a fine-grained analysis regarding spatial clustering of previously non-identifiable milieus, such as the cultural industries within cities (Currid & Williams, 2010). It is the sense of collectivity and knowledge of social

A Spatial Analysis of the Instagram Hashtag #happy …

13

Fig. 1 Scientific contributions of Instagram registered on Scopus

media that are establishing the advances necessary to identify previously unidentifiable spatial patterns. Mitchell et al. (2013) in their interesting and ample study using Twitter data suggest the existence of such a geographically integrated landscape. The collective sense of available data provides an ample resource to frame happiness as a coherent study in the field of geography taking advantage of the power of expression and is thus strongly linked to emotion (Dodds & Danforth, 2010). This equates an abundant and ubiquitous interchange to advance in the fields of happiness studies, but also in the field of Geographic Information Science. With the advent of applications such as Twitter, large dataset repositories with spatially enabled content have become available. This has led to a growing number of research to better understand collective and often spatial patterns for sentiment analysis and textual mining. Social media sites generally do not allow visualizing or collecting data from their servers. Hence, Instagram is one of the few exceptions that allows real-time access to its existing archive. Furthermore, Instagram provides a desirable tool to directly assess subjective emotion through images. It is only natural that a reported increase of scientific work with its API has exponentially grown since its appearance in 2010 (Fig. 1). The intricate and complex nature of urban regions has led to several challenges that have nested interdisciplinary solutions. Urban regions, as such, may be defined as the interaction between social, economic, and environmental components, where socioeconomic balances are paramount for well-being and the growth of smarter regions. This requires an effort both from public participation as well as governance, leading to functional transportation, as well as efficient infrastructure that paves a way toward sustainable urban habitats. In this sense, spatio-temporal understanding of changes within cities, and the integration of social behaviors within urban cores, are paramount to assess the livability, well-being, and regional dynamics of urban cores. Harnessing social data however is a difficult challenge, as socioeconomic data is usually not time efficient to measure the impacts of current status quo. In a datacentric smart city however, such data is of significant importance to tackle regional questions that would otherwise be hard to analyze. One of these issues is subjective well-being and happiness.

14

E. Vaz

1.3 Instagram Data Available geotagged information from the crowd corresponds to an online model that has successfully expanded in recent years. This evolution occurred thanks to the advances in handheld mobile devices that, through technological enhancement, brought embedded sensors that originated smartphones. While the terminology “smartphone” was coined in the mid-nineties, it was in 2007 that Apple revolutionized the concept. Apple’s successful iteration of the iPod throughout the nineties leveraged the opportunity to create a ubiquitous device. A device with a unique design and concept stimulated user engagement through the integration of media and content. In this sense, the iPhone became the first commercial iteration of a handheld computational device. Where sensorial capacity meets ergonomics. The combination of phone, video, music, and browsing made it the ideal instrument to cater to the growing demand of the web. Its two major innovations relate to the ongoing optimization of high-performance display and the refined touch screen technology. Several of the current media and web contents are nowadays optimized to cater to mobile devices instead of computers. With the incremental addition of embedded sensor technology, the smartphone is nowadays a pièce de résistance that provides a holistic integration of societal communication. This technological evolution follows closely in line with the emergence of Web 2.0, also known as the participatory and social web. The opportunity cost of user-generated content in large part due to the individualized web experience brought by smartphones has led to a steady increase in a personalized social experience of the web. Data and content have through Web 2.0 (Kim et al., 2013) become widely participatory and open, enabling sharing, and content generation at an unprecedented pace. The interactive dimension of information sharing and retrieval (Acquisti & Gross, 2006), allied with the interoperability and collaboration of users, has forwarded networking to be dominantly present in social media in particular in areas where landscape component is relevant (Munar & Jacobsen, 2014). Crowd-sourced data leads to an aggregation of different flows of content from individuals and thus offers a harvest of social information which can be used for large data analysis (Vertanen & Kristensson, 2011). With the advances of handheld devices (Ko et al., 2013), integration of applications and software taking advantage from a number of technological incorporations (high-resolution cameras, Global Positioning System, sensors, and more efficient battery life) have further fostered users to articulate their mobile phones into an opportunity for crowdsourcing of spatial data (Arsanjani & Vaz, 2015). Instagram, as a widely popular mobile photo sharing and video sharing application, has now become large part of present social media services (Memarovic, et al., 2013). With the advance of handheld mobile devices, the service has grown to become an app with over 200 million monthly actives, 20 billion photos shared and 80 million average photos per day (consulted 30th September 2015: instagram.com/press/). Sharing of media content is intrinsically a phenomenon of positive creation of social ties where a photograph represents the ideal fit of an emotional media file sharing it with the community (John, 2013).

A Spatial Analysis of the Instagram Hashtag #happy …

15

Active users (milllion) 1200 1000 800 600 400 200 0 2013

2014

2015

2016

2017

2018

2019

Fig. 2 Growth of Instagram users

Further information was gathered from the Internet Archive’s Wayback Machine as to visualize and validate the snapshots and browse history of Instagram user press data. This led to a clearer understanding of the evolution of Instagram users (Fig. 2). While active users show a steady growth, it is the number of photos per day that are most surprising. In the recent years, the volume of 5 million photos shared per day has grown to a total of 80 million photos shared per day. This leads to the tremendous potential of Instagram as a crowd-sourced data source for further investigation on the social and collective behavior of happiness (Durahim & Co¸skun, 2015). In urban regions, such tools are paramount to further support the growing use of incorporation of large data. Further, the advantage of having a media contribution in an instant, as the name Instagram suggests, leads to an intuitive practice of offering an insight on an individual moment, sharing this with the world and with a panoply of currently over 40 billion photos shared (statistic retrieved from instagram.com/press, updated on 30th September 2015). These instant moments are then followed with a hashtag1 that expresses often the sentiment of that moment and describes it often with a given emotion. The spontaneous action of taking the snapshot and posting it online allows having a very clear emotional appreciation of the content, as well as infers state of emotion. It is in this action, that from a more rigorous scientific perspective, that Instagram can offer a unique filter for social knowledge, deriving a social interpretation of collective expressions. Instagram’s new API not only allows registering the photos that are posted, but presently includes geographic metadata indicating the location of the Instagram picture, enabling the representation at spatial 1

A hashtag is a word or an unspaced phrase prefixed with the number sign (“#”). Used as a metadata tag, it is a common practice in microblogging and social mediaing services such as Instagram, Twitter, and Faceboook.

16

E. Vaz

level. The exact location of the Instagram photo is provided by GPS or cellular triangulation, available in the smart phone, and registered for the snapshot in the exact moment it is taken. This allows thus to have a very accurate geographical interpolation of the Instagram feature, which is then promptly added with a hashtag describing the picture. The advantage of having this process embedded in just a fraction of seconds allows for an accurate latitude and longitude information that can then be interpreted for further spatial analysis and integrated in a Geographic Information System. As such, the spatial data that can be collected through Instagram is not only an abundant source of social information that can be crowd sourced, but further provides locational accuracy that allows to establish geographical correlations with the images and even emotions, as the pictures represent always the moment spatially and temporally, where they are taken, which I discuss further in this paper.

1.4 The Geography of Happiness 2 Study Area Canada has a very diversified and heterogeneous urban and environmental landscape, varying greatly at provincial level. One of the fastest growing regions in North America has been witnessed in the Golden Horseshoe located in southern Ontario. This region is expected to foster a population of over 11.5 million by 2031 (Hemson Consulting, 2013). Growing immigration has led to rapid suburbanization of urban metropolises including the Golden Horseshoe, such as the Greater Toronto Area (Vaz & Bowman, 2013). While the ministry of infrastructure of Ontario still juggles with certain socioeconomic aspects that require careful management, the Greater Toronto Area has led the Golden Horseshoe, given its landscape, urban and economic characteristics, to become the fourth largest economic hub North America, followed by Chicago (Vaz & Arsanjani, 2015). Growth in the retail and real estate sector has brought a unique venture of economic prosperity in the region, where a rising demand from a technologically savvy younger population is changing the interaction with urban areas and its growing economy. Southern Ontario has therefore become not only a cradle for economic prosperity, but also a potential beacon for innovation and technological advance, in particular in the urban regions throughout the Greater Toronto Area (Fig. 3). This is strongly linked with the current potential for fostering well-being and subjective perception of health, suggesting that land use transitions within the regions are an important aspect of environmental sustainability, but also continued well-being of the Greater Golden Horseshoe’s growing population (Vaz et al., 2015).

A Spatial Analysis of the Instagram Hashtag #happy …

17

Fig. 3 Land use in the Greater Golden Horseshoe, Canada

3 Data An API was built to search for the hashtag “#happy” on Instagram. The API called for the following fields: id, user.username, created_time, location.latitude, location.longitude, tag, subtag, type, and images.thumbanil.url. The results are exported as a csv file and imported into a Geographic Information System (GIS) (Fig. 4). Further data handling was carried out by means of importing the csv file as delimited text file and geocoding each line into a point feature. This allowed for spatial representation of the Instagram features related to the searched tag. A total of 674,839 Instagram data were geocoded at international level based on the available latitude

18

E. Vaz

Fig. 4 Hashtag search engine for keywords

and longitude coordinates. The Fig. 4 shows the distribution of collected Instagram data with the hashtag happy in Toronto. Further, to the total of 674,839 data points, a total of 7694 records were discarded due to missing coordinates. Contrary to twitter, where just a small fraction of georeferenced information is present, Instagram shows a much larger data source of georeferenced data. This enables the potential of harnessing these data sources for spatial analysis on the complexity of social behaviors. The Fig. 3 shows the number of Instagram vector points collected along the study area of the Golden Horseshoe, corresponding to a total of 2232 point vector features. While the pattern initially seems more clustered in the urban area as expected, there is a strong relation between usage of technology and population density per census tract. To correct this, and have a more integrative vision of the distribution of the data, an ancillary data treatment was conducted as follows: (i) Point data was aggregated within each census tract in the Greater Golden Horseshoe, (ii) an aggregated indicator was built based on h n / pct , where h n represents the agglomerated happiness count per census tract and pct represents the existing population per census tract, and, finally, (iii) the data was normalized accordingly by scaling the values between 0 and 1 in i −E min ei = Eemin , where E min corresponds to the minimum valuable for the variable E, −E max and E max is the maximum value for the variable E. . The following choropleth map of the correct distribution of Instagram incidences was thus obtained (Fig. 5). Further, land use data was used from the DMTI CanMap Route for 2012. As part of the DMTI RouteLogistics Dataset, the dataset pertains several important anthropogenic land use categories: commercial, government and institutional, open area, parks and recreational, residential, resource and industrial, and waterbody.

4 Methodology This paper offers an integrated spatial methodology, creating an explanatory model for the incidence of happiness landscape based on Instagram data. In this sense, spatial impendence is seen as an important aspect of the integration of the different spatial analytical components (Figs. 5 and 6).

A Spatial Analysis of the Instagram Hashtag #happy …

Fig. 5 Distribution of #happy density within the Greater Golden Horseshoe

19

20

E. Vaz

Fig. 6 Methodology

Once the Data API has conducted successfully the search for the hashtag, features were geocoded accordingly over geographical space. This allowed for a first assessment of understanding if there is spatial autocorrelation in the region at global level. The successful negation of the null hypothesis, i.e., there is no significant spatial autocorrelation, confirmed by the evaluation of the Global Moran’s I algorithm, allowed for an assessment on the spatial impendence of neighbors at local level. This was achieved by the calculation of the Local G i∗ Statistic. The resulting hotspots of particularly significant incidences of happy hashtags were then assessed within an explanatory model of a Geographically Weighted Regression (GWR) encompassing socioeconomic variables, as well as the integrated land use data for the Golden Horseshoe’s hotspot, and the resulting metrics regarding land use typologies.

4.1 The Chase for Global Spatial Dependence Tobler’s first law of Geography deems that objects closer to each other are more related than objects spatially apart and this holds true for land use transitions (Vaz, 2014). This fundamental assumption has led spatial analysis and geostatistics to

A Spatial Analysis of the Instagram Hashtag #happy …

21

consider the fundamental relation of objects in space, defining what is considered spatial autocorrelation. Spatial autocorrelation assists in understanding these patterns either at a global or local level, envisioning the repercussions of spatially explicit models at regional and local level (Rey & Montouri, 1999). The global spatial autocorrelation exerts a clearer understanding of the spatial patterns for the entire region level, that is, the total area considered in the spatial system, and is defined as follows by integration of Moran’s I: n I = S0

Σn

Σn

i= j

j=1

Σn

wi, j z i z j

2 i=1 z i

where zi corresponds to the deviation of an attribute for feature i from its mean (x i – X), wi, j is the spatial weight between features i and j, and n is the total number of features. S 0 is the aggregate of all the spatial weights, which is calculated as S0 =

n Σ n Σ

wi, j

i=1 j=1

The value of Moran’s I ranges between −1 to 1 and a value of 1 indicates that there is a positive spatial autocorrelation, while a value of −1 shows a negative spatial correlation, i.e., the features are negatively spatially autocorrelated. An approximation of Moran’s I to 0 corresponds to randomness, that is, no spatial autocorrelation is present.

4.2 The Chase for Local Spatial Dependence In the case of “happy” Instagram events, the data was generalized to census tract level, as to allow the generation of a weight matrix, a contiguity condition to allow the generation of the Global Moran’s I statistic. Furthermore, as explored by Erdogan (2009) a score method of a z-score was calculated for a significance level of 5%, this allowed inferring within the threshold of greater than 1.96 or less than −1.96, the generalized spatial autocorrelation in a test of non-randomness. These features were then assessed for local and inter-regional clustering of high and low clusters. This was carried out by local spatial autocorrelation techniques as described by Ord and Getis and allowed to define the spatially explicit hotspots that constituted the happiness landscape. Local spatial autocorrelation defines high and low clustering over the spatially explicit model, supporting the creation of spatial hotspots, a fundamental determinant to understand the spatial landscape of spatially explicit features. Positive or negative spatial autocorrelation is thus calculated as a metric in locational analysis, leading to a better understanding of randomness of spatial phenomena (Getis, 2008). The Getis-Ord G i∗ statistics was carried out by counting the number of Instagram features at census tract level for the Golden Horseshoe. Contrary to global spatial

22

E. Vaz

autocorrelation, the Getis-Ord G i∗ statistics gives a value for each census tract which indicates the degree of high or low value clusters, creating an interpretative spatial landscape of the Instagram features. The higher Getis-Ord G i∗ statistics, the more an indication of that spatial clustering among features is registered. The Getis-Ord G i∗ statistics is computed as follows: Σn j=1 wi, j − X j=1 wi, j /[ Σ (Σ )] n 2 n nj=1 wi,2 j − j=1 wi, j

Σn G i∗

= S

n−1

where x j is the attribute value for feature j, w i, j is the spatial weight between features i and j, and n is the total number of features.

4.3 Spatial Accounting of Land Use Within the Happiness Landscape Land use poses a tool of utmost importance for understanding socioeconomic relations to spatial and geographical phenomena. Topological interactions of land can be quantified and described by means of aggregation of spatially explicit land use data. The diversified dynamics of land use in anthropogenic land use types extends to a deeper understanding on a plethora of question linked to health, economic growth, and well-being indices. In this analysis, different land use types are considered and the number of Instagram #happy contributions are counted for each given land use type within the defined coldspots and hotspots. The land use data was assessed in the following categories: commercial, government and institutional, open area, parks and recreational, residential, resource and industrial, and water bodies. The features were combined by means of the land use category and accounted as to understand the general pattern of feature quantities for land use typologies, allowing further landscape metrics to assess the characteristics of the land use patterns within the hotspots and coldspots (Fig. 7). Although it deals with big data and given that the usage of collected crowd-sourced content is a recent endeavor, we must assume that some error in the interpretation of textual information may be present. As such, a wide array of research still remains untapped, until we have further advanced with computational power that allows for big data handling to assess such large amount of contributions. For now, we can however when integrating this data with geography ascertain a set of important characteristics: (i) Most of the contributions on Instagram appear on urban areas, (ii) the location of Instagram contribution matters, as it has a profound emotional charge exploring the moment where the snapshot is taken, and (iii) the Instagram picture is uploaded with a series of hashtags, that allow further exploration on the subjective interpretation of the contributor. All of these characteristics lead to a spatially explicit and coherent resource of emotional pool that deserves further attention, given its

A Spatial Analysis of the Instagram Hashtag #happy …

23

Fig. 7 Land use and landscape metrics component

authenticity, its number of users, and the importance of images in human perception of reality and state of mind suggesting a strong connection between land use and the #happy hashtag. The generated raster land use features for 2012 were converted to point features and counted for each individual land use category for the pertaining census tract. This was carried out for the hotspots and the coldspots. A consistent accounting of the amount of land use within each census tract was thus achieved allowing to calculate Kendall’s tau-b for the land use distribution per census tract for the hotspots and coldspots (Fig. 8). The relation between land use categories was tested by means of a Kendall’s tau-b rank pairwise correlation. Kendall’s tau-b is given by: (a, b)−!(a, b) τB = √ √ N1 × N2 where (a, b) is a number of concordant pairs, and N1 is a number of data pairs not tied in a target feature, and N2 is another number of data pairs not tied in a target feature. As such, Kendall’s tau-b shows a probability, taking values between −1 and + 1, manifesting a positive correlation when the ranks of both variables increase, and a negative correlation indicated by an increase of one pair variable and the decrease

24

E. Vaz

Fig. 8 Land use accounting methodology

of the other pair. The usability of Kendall’s tau-b rank instead of Spearman’s rank correlation seems adequate given its potential of direct interpretation of probabilities within the land use categories of concordant and discordant pairs (Conover, 1980). Finally, landscape metrics pertaining urban land use were conducted to measure urban land use in relation to the #happy Instagram features relating to (i) density and (ii) fragmentation. As urban density, the population density was related as a ratio of built-up land use with the total population per census tract. Whereas fragmentation was measured by the existence of open spaces within the built-up environment per census tract. This allowed to create a regional perspective of a comparative analytical framework of fundamental urban metrics and their relation to the contributions to #happy within the identified urban cores.

5 Results 5.1 The Golden Horseshoe’s #happy Spatially Explicit Landscape The 2231 Instagram contributions corresponding to the period of 23rd of June to the 1st of July, 2014, allowed for a thorough analysis of the spatial landscape of Instagram contributions with the hashtag happy. The features were visually assessed allowing a preliminary perception that the majority of images found during this period corresponded to selfies.2 The spatial landscape of the #happy further showed a very strong spatial autocorrelation at global level, through its interpretation of the Moran’s I coefficient, having obtained a z-score of 57.022. This suggests that the Instagram contributions are not random over space, but manifest a strong spatially explicit autocorrelation for the entire region. A further assessment is carried out 2

selfie (noun): an image that includes oneself (often with another person or as part of a group) and is taken by oneself using a digital camera especially for posting on social networks (in MerriamWebster (https://www.merriam-webster.com/dictionary/selfie, consulted on: Oct 17, 2022).

A Spatial Analysis of the Instagram Hashtag #happy …

25

Fig. 9 Identified hotspots and coldspots within the Greater Toronto Area

by detection of clusters that allow further exploration between the differences at census tract level. This was carried out by the Getis-Ord G i∗ statistic. This statistic compares the expected value of a variable across census tracts, with the entire values registered over the entire Golden Horseshoe. It forms a standardized z-score and associated P value as described in the methodology. The variable related to Instagram contributions was thus assessed and hotspots as well as coldspots were found by the association of positive significant values and negative significant values, respectively. The two hotspots were found in the downtown Toronto core as well as Mississauga, suggesting a much stronger positive relation at urban cores, at a significance level of P < 0.5. The figure below shows the positive and negative spatial associations around the Greater Toronto Area (Fig. 9).

5.2 Toronto’s #happy Land Use Distribution A fundamental difference is present within the distribution of hotspots and coldspots of the #happy landscape. While most of the land use categories are similar regarding the percentages of Commercial, Government and Institutional, Open Area, and water bodies, hotspots of contributors are found significantly in residential areas, while coldspots are found predominantly in industrial areas, although parks and recreational areas are more present (Fig. 10). The assessment of a pairwise comparison within the correlation of land use suggests that for the variables, (COM = Commercial, WA = Waterbody, RI =

26

E. Vaz

50.00% 45.00% 40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Commercial Government and Open Area Institutional

Parks and Recreational Hotspots

Residential

Resource and Industrial

Waterbody

Coldspots

Fig. 10 Land use distribution in hotspots and coldspots

Resource and Industrial, GO = Government and Institutional, PR = Parks and Recreational, RES = Residential, OA = Open Areas) show significantly different distributions per census tract aggregates. Land use categories for the hotspots express a more heterogeneous distribution for land use categories (Table 1), while the correlation coefficient is highest for coldspots predominantly, and almost exclusively, for open areas (Table 2). The additional analysis of correlation between Instagram features and urban density and urban fragmentation displays interesting results regarding the hotspots. The τB value for urban density was of τB = 0.001449, and the value for fragmentation with Kendall’s tau-b rank was of τB = 0.081988, that is, the value of Instagram #happy contributions was much stronger correlated to fragmented areas. The main contribution of these areas pertained the existence of open areas and more natural environments. Table 1 Kendall’s tau-b for Hotspots GO

PR

COM

COM 1.00

WA 0.14

RI 0.05

−0.52

−0.24

RES 0.33

OA 0.24

WA

0.14

1.00

0.71

−0.43

0.05

0.62

0.33

0.05

0.71

1.00

−0.14

0.33

0.33

0.24

GO

−0.52

−0.43

−0.14

1.00

0.33

−0.43

−0.14

PR

−0.24

0.05

0.33

0.33

1.00

0.24

0.52

RES

0.33

0.62

0.33

−0.43

0.24

1.00

0.71

OA

0.24

0.33

0.24

−0.14

0.52

0.71

1.00

RI

A Spatial Analysis of the Instagram Hashtag #happy …

27

Table 2 Kendall’s tau-b for Coldspots COM

COM

WA

1.00

0.12

RI 0.14

GO

PR

0.04

0.11

RES 0.02

OA 0.19

WA

0.12

1.00

0.12

0.02

0.23

0.08

0.24

RI

0.14

0.12

1.00

0.12

0.25

−0.07

0.27

GO

0.04

0.02

0.12

1.00

0.28

0.00

0.18

PR

0.11

0.23

0.25

0.28

1.00

0.14

0.33

RES

0.02

0.08

−0.07

0.00

0.14

1.00

0.33

OA

0.19

0.24

0.27

0.18

0.33

0.33

1.00

6 Discussion Several intriguing results have shaped important findings through the integration of spatial analysis and landscape metrics in regard to the distribution of the hashtag “happy.” A closer inspection of the hotspots related to residential land use and happiness hashtags remits to the importance of home and neighborhood concept as a key determinant for happiness. This is sharply juxtaposed to industrial land, where compounding pockets of coldspots were found, suggesting the opposite scenario. Such findings are similar to what has been found within the ideation of self-harm and health perception as studied in Vaz and others in the Greater Toronto Area (2020). Indeed, within the current body of literature of social sciences, there is a growing shift between happiness studies and subjective well-being. While psychological positivism adopts a holistic approach toward integrating different fields of expertise, it has been rather hard to converge the complexity of happiness studies within a larger framework of quantitative versus qualitative methods (Delle Fave et al., 2011). Quantitative studies fall primarily in the category of subjective well-being (Linley et al., 2009) and therefore do not adequately portray the complexity of happiness (Ruggeri et al., 2020). This is a result of several limitations such as (i) hard data does not allow for subjective observation of events and two distinct definitions must exist between objective and subjective well-being as a result (Choi et al., 2020), (ii) inconsistent distribution of population and demographic characteristics (Ahmadiani et al., 2020), and (iii) overlapping meanings of census results and lack of normalization of data structures. These are a few of the limitations at hand that leads to a lack of discourse between the disciplines in regard to happiness studies (Soukiazis & Ramos, 2016). Studies such as the Oxford Happiness Questionnaire rely on a multi-tier interpretation of happiness and thus an encapsulated vision of happiness within a sectionbased approach. The questionnaire finds that sequential orthogonal factor analyses of the OHQ identified a single higher-order factor, which suggests that the construct of well-being it measures is uni-dimensional (Hills & Argyle, 2002). However, the

28

E. Vaz

complexity of measuring happiness has gained further expression through the integration of larger data repositories utilizing social media such as Twitter (Jaidka et al., 2020). Social sciences and the integration of regional sciences could well abridge the field further by leveraging sociology, economics (Graham, 2005), and geography (Vaz, 2020), which have thus far in their tradition consistently used indicators in subjective well-being and city livability but have failed to approach the larger scope of happiness. Nevertheless, it is clear that happiness is a vital and encompassing construct of regional development that should intertwine sociology (Kroll, 2014), economics (Graham, 2005), geography (Mavruk et al., 2020; Mitchell et al., 2013), spirituality (Berejnoi et al., 2019), and psychology. Therefore, understanding happiness is of critical importance for the development of regional planning and an efficient driver to mitigate health risks. The issue of measuring happiness is tied intrinsically in the absence of systematic data that directly infers on “happiness.” Trials to measure happiness have been conducted thanks to the advances of smartphones that monitor, measure, and report happiness to some extent and at different dimensions (Lakens, 2013; Lathia et al., 2017). However, such studies fail to offer a ubiquitous interpretation within a larger geographical boundary, as they are dependent on survey response and therefore foster only a restrictive viewpoint. The usage of Instagram abridges this gap, given the ubiquitous nature of the data and the intrinsic subjective response of users.

7 Conclusions Although happiness is subjective, the existence of large data with clear emotional impendence may shed significant light on the future research possibilities of understanding subjective well-being (Diener, 1994). The usage of Instagram and other social media data sources may provide useful insights of big data in social science that should be further explored (Rae & Singleton, 2015). This paper has shown that perception of happiness is divided within a geographical specter of rural and urban lands (Requena, 2015), but also that urban land use types and its connectivity is intrinsically linked to perception of subjective well-being at local level. In these areas, the transitions of urban land and environmental pressures such as growing metropolitan regions must equate these large data sources as to provide better policies, particularly addressing the integration of subjective well-being in rural and agricultural regions, where regional sciences may have profound impacts (Torre, 2015). In line with the morphology of urban land and land use, this paper has explored the ubiquitous application of social data sources for capturing a quantitative representation of spatial information (Kent & Capello, 2013). In fact, land use does not only seem to show the existence of coldspots and hotspots for the subjective sentiment of happiness, but is further clearly expressed depending on the sustainability of regions by urbanization type. People are happier in regions with more open areas and where an equilibrium between the natural environment and urban land use is found. Industrial

A Spatial Analysis of the Instagram Hashtag #happy …

29

zones show least amount of happiness in the context of the Greater Golden Horseshoe. Further assessment on the fragmentation of urban land as well as compactness within census tracts shares new light on the definition of happiness within the city: Are denser urban environments happier? The results seem to point out an intriguing response. The regions with more fragmented urban land, where wider and greener areas are found, lead to more subjective well-being and relative sentiments of happiness. These happyscapes—landscapes of spatial identity of happiness at regional level—shed new light on the intriguing complexity of cities, but also an important and quantitative vision on how urban regions must incorporate greener spaces, instead of fostering a single urban hub without the integration of landscape and nature. Limitations There were two sets of limitations found throughout this paper. A first limitation entails the sometimes-sparse availability of Instagram data with geographical coordinates. Due to user privacy concerns, not all users chose to share their location. The second limitation is that since 2018, Instagram stopped allowing the possibility to download content from its API.

8 Competing Interests The author declares that he has no competing interests. Acknowledgements The author would like to thank the two reviewers for the great comments that significantly helped to enhance this manuscript.

References Acquisti, A., Gross, R., (2006). Imagined communities: Awareness, information sharing, and privacy on the Facebook, 4258, 36–58; Danezis, G., Golle, P. (Ed.). Lecture notes in computer science: privacy enhancing technologies. Springer: Dordrecht. Ahmadiani, M., Brereton, F., Ferreira, S., & Moro, M. (2020). Spatial variation in life satisfaction: A happiness puzzle. Edward Elgar Publishing. Anselin, L. (1999). The future of spatial analysis in the social sciences, geographic information sciences. A Journal of the Association of Chinese Professionals in Geographic Information Systems, 5(2), 67–76. Arsanjani, J. J., & Vaz, E. (2015). An assessment of a collaborative mapping approach for exploring land use patterns for several European metropolises. International Journal of Applied Earth Observation and Geoinformation, 35, 329–337. Ballas, D., & Dorling, D. (2013). The geography of happiness. The Oxford Handbook of Happiness, 54, 465–481. Berejnoi, E., Ulluwishewa, R., Cloutier, S., Gibbons, L., Puga, S., & Uthumange, A. (2019). Spirituality: The missing link of sustainability and happiness as a framework for holistic development. In Perspectives on community well-being (pp. 203–224). Springer, Cham.

30

E. Vaz

Brereton, F., Clinch, J. P., & Ferreira, S. (2008). Happiness, geography and the environment. Ecological Economics, 65(2), 386–396. Burger, M. J., Morrison, P. S., Hendriks, M., & Hoogerbrugge, M. M. (2020). Urban-rural happiness differentials across the world. World Happiness Report, 2020, 66–93. Choi, N., Kim, J., & Lee, S. J. (2020). The usefulness of intersubjective community wellbeing as a community development indicator: Evidence from comparing three approaches to measuring community wellbeing. International Journal of Community Well-Being, 1–20. Collins, A. L., Sarkisian, N., & Winner, E. (2009). Flow and happiness in later life: An investigation into the role of daily and weekly flow experiences. Conceição, P., & Bandura, R. (2008). Measuring subjective wellbeing: A summary review of the literature. United nations development programme (UNDP) development studies, working paper. Conover, 1980 Conover, W. J. (1980). Practical nonparametric statistics. New York. Currid, E., & Williams, S. (2010). The geography of buzz: Art, culture and the social milieu in Los Angeles and New York. Journal of economic geography, 10(3), 423-451. Davern, M., Gunn, L., Whitzman, C., Higgs, C., Giles-Corti, B., Simons, K., Badland, H., et al. (2017). Using spatial measures to test a conceptual model of social infrastructure that supports health and wellbeing. Cities & Health, 1(2), 194–209. Delle Fave, A., Brdar, I., Freire, T., Vella-Brodrick, D., & Wissing, M. P. (2011). The eudaimonic and hedonic components of happiness: Qualitative and quantitative findings. Social Indicators Research, 100(2), 185–207. Diener, E. (1994). Assessing subjective well-being: Progress and opportunities.Social indicators research, 31(2), 103–157. Dodds, P. S., & Danforth, C. M. (2010). Measuring the happiness of large-scale written expression: Songs blogs, and presidents. Journal of Happiness Studies, 11(4), 441–456. Durahim, A. O., & Co¸skun, M. (2015). # iamhappybecause: Gross national happiness through Twitter analysis and big data. Technological Forecasting and Social Change, 99, 92–105. Elwood, S., Goodchild, M. F., & Sui, D. (2012). Researching volunteered geographic information: Spatial data, geographic research, and new social practice. Annals of the Association of American Geographers, 102(3), 571–590. Erdogan, S. (2009). Explorative spatial analysis of traffic accident statistics and road mortality among the provinces of Turkey. Journal of safety research, 40(5), 341–351. Ferreira, S., & Moro, M. (2010). On the use of subjective well-being data for environmental valuation. Environmental and Resource Economics, 46(3), 249–273. George, L. K. (2010). Still happy after all these years: Research frontiers on subjective well-being in later life. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 65(3), 331–339. Getis, A. (2008). A history of the concept of spatial autocorrelation: A geographer’s perspective. Geographical analysis, 40(3), 297–309. Graham, C. (2005). The Economics of Happiness. World Economics, 6(3), 41–55. Hagerty, M. R., & Veenhoven, R. (2003). Wealth and happiness revisited–growing national income does go with greater happiness. Social Indicators Research, 64(1), 1–27. Helliwell, J. F., Gyarmati, D., Joyce, C., & Orpana, H. (2020). Building an epidemiology of happiness (No. w28095). National Bureau of Economic Research. Jaidka, K., Giorgi, S., Schwartz, H. A., Kern, M. L., Ungar, L. H., & Eichstaedt, J. C. (2020). Estimating geographic subjective well-being from Twitter: A comparison of dictionary and datadriven language methods. Proceedings of the National Academy of Sciences, 117(19), 10165– 10171. John, N. A. (2013). The social logics of sharing. Communication Review, 16(3), 113–131. Kent, J. D., & Capello, H. T., Jr. (2013). Spatial patterns and demographic indicators of effective social media content during theHorsethief Canyon fire of 2012. Cartography and Geographic Information Science, 40(2), 78–89.

A Spatial Analysis of the Instagram Hashtag #happy …

31

Kim, M., Park, S. O., Joo, W.-K., Choi, K., Jeong, Y.-S., & Kim, Y.-K. (2013). Tag based collaborative knowledge management system with crowdsourcing. Journal of Internet Technology, 14(5), 859–866. Ko, S. M., Chang, W. S., & Ji, Y. G. (2013). Usability principles for augmented reality applications in a smartphone environment. International Journal of Human-Computer Interaction, 29(8), 501–515. Kroll, C. (2014). Towards a sociology of happiness: The case of an age perspective on the social context of well-being. Sociological Research Online, 19(2), 1–18. Lakens, D. (2013). Using a smartphone to measure heart rate changes during relived happiness and anger. IEEE Transactions on Affective Computing, 4(2), 238–241. Lathia, N., Sandstrom, G. M., Mascolo, C., & Rentfrow, P. J. (2017). Happier people live more active lives: Using smartphones to link happiness and physical activity. PLoS ONE, 12(1), e0160589. Lew, B., Huen, J., Yu, P., Yuan, L., Wang, D. F., Ping, F., Jia, C. X., et al. (2019). Associations between depression, anxiety, stress, hopelessness, subjective well-being, coping styles and suicide in Chinese university students. PLoS ONE, 14(7), e0217372. Linley, P. A., Maltby, J., Wood, A. M., Osborne, G., & Hurling, R. (2009). Measuring happiness: The higher order factor structure of subjective and psychological well-being measures. Personality and Individual Differences, 47(8), 878–884. Mavruk, C., Kıral, E., & Kıral, G. (2020). Spatial effects over time-framed happiness. Journal of Happiness Studies, 1–38. Memarovic, N., Elhart, I., Michelotti, A., Rubegni, E., Langheinrich, M. (2013). Social mediated displays: Integrating networked public displays with social media. In UbiComp 2013 Adjunct— Adjunct Publication of the 2013 ACM Conference on Ubiquitous Computing (pp. 55–58). Mitchell, L., Frank, M. R., Harris, K. D., Dodds, P. S., & Danforth, C. M. (2013). The geography of happiness: Connecting twitter sentiment and expression, demographics, and objective characteristics of place. PLoS ONE, 8(5), e64417. Munar, A. M., & Jacobsen, J. K. S. (2014). Motivations for sharing tourism experiences through social media. Tourism Management, 43, 46–54. Pfeiffer, D., & Cloutier, S. (2016). Planning for happy neighborhoods. Journal of the American Planning Association, 82(3), 267–279. Qasim, M., & Grimes, A. (2022). Sustainability and wellbeing: The dynamic relationship between subjective wellbeing and sustainability indicators. Environment and Development Economics, 27(1), 1–19. Rae, A., & Singleton, A. (2015). Putting big data in its place: A regional studies and regional science perspective. Regional Studies, Regional Science, 2(1), 1–5. Requena, F. (2015). Rural–urban living and level of economic development as factors in subjective well-being. Social Indicators Research, 1–16. Rey, S. J., & Montouri, B. D. (1999). US regional income convergence: A spatial econometric perspective. Regional Studies, 33(2), 143–156. Ruggeri, K., Garcia-Garzon, E., Maguire, Á., Matz, S., & Huppert, F. A. (2020). Well-being is more than happiness and life satisfaction: A multidimensional analysis of 21 countries. Health and Quality of Life Outcomes, 18(1), 1–16. Soukiazis, E., & Ramos, S. (2016). The structure of subjective well-being and its determinants: A micro-data study for Portugal. Social Indicators Research, 126(3), 1375–1399. Tella, R. D., MacCulloch, R. J., & Oswald, A. J. (2003). The macroeconomics of happiness. Review of Economics and Statistics, 85(4), 809–827. Torre, A. (2015). New challenges for rural areas in a fast moving environment. European Planning Studies, 23(4), 641–649. Vaz, E. (2014). Managing urban coastal areas through landscape metrics: An assessment of Mumbai’s mangrove system. Ocean & Coastal Management, 98, 27–37. Vaz, E. (2020). Regional science, international encyclopaedia of human geography (pp. 357–367). Elsevier.

32

E. Vaz

Vaz, E., & Arsanjani, J. J. (2015). Predicting urban growth of the greater Toronto area-coupling a Markov cellular automata with document meta-analysis. Journal of Environmental Informatics, 25(2), 71–80. Vaz, E., & Bowman, L. (2013). An application for regional coastal erosion processes in urban areas: A case study of the Golden Horseshoe in Canada. Land, 2(4), 595–608. Vaz, E., Cusimano, M., & Hernandez, T. (2015). Land use perception of self-reported health: Exploratory analysis of anthropogenic land use phenotypes. Land Use Policy, 46, 232–240. Vaz, E., Shaker, R. R., Cusimano, M. D., Loures, L., & Arsanjani, J. J. (2020). Does land use and landscape contribute to self-harm? A sustainability cities framework. Data, 5(1), 9. Vertanen, K., Kristensson, P. O. (2011). The imagination of crowds: Conversational AAC language modeling using crowdsourcing and large data sources. In EMNLP 2011—Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference (pp. 700–711). Ye, L. I. U., Tong, X. I. A. O., Yuqi, L. I. U., Yingzhi, Q. I. U., Yi, L. I. U., & Zhigang, L. I. (2020). Impacts of urban built environments on residents’ subjective well-being: An analysis based on 15-minute walking distance. Progress in Geography, 39(8), 08001270. Zacher, H., & Rudolph, C. W. (2021). Individual differences and changes in subjective wellbeing during the early stages of the COVID-19 pandemic. American Psychologist, 76(1), 50.

The Subjective Well-Being in North Africa and the Impacts on Agriculture and Urban Land Azzeddine Bellout, Eric Vaz, Antonia Bousbaine, and Christopher R. Bryant

1 Introduction Since the late 1990s, there has been a renewed and growing interest in the measurement of subjective well-being (Morrison, 2007) by geographers. This issue has now become a legitimate issue in geography, as evidenced by the profusion of books and articles published in prestigious journals (Helliwell et al., 2012; Brulé & Veenhoven, 2015; Weiner, 2009; Brereton et al., 2008). This work has contributed to the birth of a new orientation in geography, which studies subjective well-being (SWB) under the name of the “Geography of Happiness.” Geography can have a major influence on happiness. Different geographic factors could influence the likelihood of whether a person feels satisfied or less satisfied with their life by influencing the perceptions of happiness. This is not only evident in Western countries, but in many other parts of the world including North Africa, indicating that the factors of happiness may not be so different from one country to another. Happiness can have a significant national economic impact in different countries, making it an important issue for policy makers and economists (Ott, 2012). For example, happiness can affect productivity as well as the way you spend your time A. Bellout (B) University Akli Mouhand Oulhadj-Bouira, Geography, Algeria e-mail: [email protected] E. Vaz Department of Geography and Environmental Studies, Ryerson University, Toronto, ON, Canada e-mail: [email protected] A. Bousbaine Department of Geography, University of Liege, Liege, Belgium C. R. Bryant University of Montreal, Quebec, Canada School of Environmental Design and Rural Development, University of Guelph, Guelph, ON, Canada © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 E. Vaz (ed.), Geography of Happiness, Contributions to Regional Science, https://doi.org/10.1007/978-3-031-19871-7_3

33

34

A. Bellout et al.

and money. There have been indications that happiness can be affected by the urban and rural divide, where more rural areas could provide more recreational opportunities and a greater sense of community engagement, thereby increasing happiness. However, a recent study in the United Kingdom, using a generalized log on regional and local data from urban and rural areas, showed that regional rather than localized effects are more likely to affect happiness. Looking at factors such as income, health, and recreation, those living in some parts of the world have reported being much more satisfied than others have been in rural or more urban areas. When comparing regional and urban effects, the regional effect had a greater influence on respondents, although those living in more rural areas were also generally happier than urban dwellers. This might suggest that lifestyles and region-specific factors that include social networks and sense of community could more strongly affect happiness scores. Well-being is recently becoming a focus in both social policies and academic studies. Among different dimensions of well-being, the hedonic view of wellbeing has drawn much scholarly attention, especially in economics and psychology. Although geography research is still in its infancy, many researchers have noted spatial differences in relation to subjective well-being and recognized the importance of place to SWB. (Fenglong & Donggen, 2015) promoted studies on the multifarious links between geographical context and SWB. These territorial specific dimensions of well-being particularly in relation to local and regional people’s experiences and values represent a major set of factors that can contribute to the support of different kinds that are associated with agricultural land and production. Furthermore, human values as expressed by farmers and their families as well as by local food consumers are often seen together in local food purchasing projects (Bousbaine, 2020; Bousbaine & Bryant, 2017a), as has been the case in Liège and its surrounding agricultural territories, with their several food governance projects.

1.1 Geographical Determinants of Subjective Well-Being Well-being has different meanings in different disciplines, leading to heterogeneous definitions. Previous studies focused more on personal characteristics such as personality and socioeconomics. In contrast, only a limited number of studies have examined the effects of geographical context on subjective well-being. As a consequence, while deep associations between regional factors and individual outcomes have been documented (e.g., Gerstorf et al., 2010), the precise geographic factors that lead to regional differences in subjective well-being have not been extensively examined (Diener et al., 1995). However, a growing body of research has recently begun to address this problem. As a result, recent studies have begun to examine the effects of urban physicality (both natural and built), social environments, and adjacent amenities on subjective well-being. Although most research on subjective well-being has looked at individual processes, researchers are beginning to examine how it works in larger geographic units of analysis, such as nations, states, counties, and cities (Diener et al., 2010;

The Subjective Well-Being in North Africa …

35

Lawless & Lucas, 2011; Lucas et al., 2014; Oswald & Wu, 2010; Plaut et al., 2002; Rentfrow, 2018). Interest in the geography of well-being rests, in part, on developing an understanding of how local and national culture, society, politics, and economy might inform subjective assessments of the quality of life and people’s lives. Studying these questions is useful to researchers in psychology and economics because they may reveal more distant causes of subjective well-being and is useful to policymakers who may use this information to guide policy decisions. To this end, researchers have begun to study geographic differences in subjective well-being using very large samples from around the world to measure individual differences in subjective wellbeing. The level of well-being in a place (i.e., nation, state, county, or city) is generally based on the average subjective well-being score from a sample of people who live in the same place. In this way, researchers are able to make comparisons of well-being between geographic regions. And the evidence indicates that well-being is not evenly distributed, but geographically clustered. The results of international studies consistently show significant differences in subjective well-being from one country to another. For example, the results consistently show that residents of Canada, Denmark, Switzerland, and the United States score near the top on measures of subjective well-being, while residents of Eastern European and Africa score near the bottom (Diener, 2000; Veenhoven, 1993). Within many countries, including the North African countries, there is considerable variation in the social, economic, and political landscape within the country. Thus, analyzing and monitoring regional diversity within countries is valuable in increasing understanding of the nature of subjective well-being and the factors that affect it.

1.2 Subjective Well-Being, Agriculture, and Happiness Many developing countries have been and continue to experience processes of “structural transformation.” During which, a large number of workers switched from agriculture to other sectors, and from self-employment to wage work. North African countries have been undergoing rapid structural transformation and economic development. The share of the labor force in agriculture in these countries fell by more than 50% in 2012 (Arab Organization for Agriculture Report, 2012). The percentage of self-employed workers has decreased from 83 to 65% over the same period, and these trends are expected to continue in the coming decades. These occupational changes play an important role in increasing productivity and incomes (e.g., McMillan & Rodrik, 2011) and can also have important effects on subjective well-being. Benz and Frey (2008) have shown that self-employment is associated with much higher job satisfaction than paid work in several countries and these authors attribute this effect to higher levels of “procedural utility” for the self-employed. The transition from transplantation to non-transplantation can also affect subjective well-being. Most people in rural areas have grown up on a farm and can feel more efficient in that environment.

36

A. Bellout et al.

This chapter first systematically introduces concepts, measures, and theories of subjective well-being and then provides an overview of current studies related to geographic context and well-being. This chapter also includes two main parts: First, there are some issues about the links between subjective well-being (the personal well-being of people) and the type of impacts on agriculture and production processes, particularly the process of food production that contributes significantly to the well-being of consumers, especially local consumers. Agri-food is one of the most important sectors for North African economies. Although its contribution to national production varies in the region, in 2014 it represented 9.5% of GDP in Tunisia, 13% in Algeria and Egypt, and 15.6% in Morocco. In 2015, the sector employed 21.7% of the total labor force in Egypt, 15% in Tunisia, and about 40% in Morocco (Arab Organization for Agriculture Report, 2012). Second, there is the impact of people’s well-being on how urban lands and potential urban lands are conceived, planned, and developed in North African countries by presenting specific examples of the role of citizens and their well-being in all its dimensions in developing urban agriculture and improving the welfare of other citizens on a large scale.

2 Methodology 2.1 Study Sites Commonly, North Africa designates the region stretching from the Atlantic shores of Morocco in the west to the Suez Canal and the Red Sea in the east and includes the countries of Morocco, Algeria, Tunisia, Libya, and Egypt. North Africa connects two important and disparate regions of the world: Europe and Africa (Fig. 1). The region is bordered on the north by the Mediterranean Sea and on the south by sub-Saharan Africa. However, it must be emphasized that threequarters of North Africa’s territory is desert and not as densely populated as the coastal regions of the northern Mediterranean.

2.2 Characteristics of the Population and Economy of North African Countries The population of the five selected North African countries represents 2.41% of the world’s total population (Table 1) with a high population density in the Northern Mediterranean coastal areas. These countries vary greatly concerning population size, ranging from 100 million for Egypt to as few as 6 million for neighboring Libya. Population growth rates in the region mostly exceed the world average, except in Tunisia (1.12%), and are much higher again in Libya (1.5%) and Egypt (2.0%).

The Subjective Well-Being in North Africa …

37

Fig. 1 Countries of North Africa. Source Authors’

This is reflected by the predominance of youth class < 65 years (Tiliouine, 2015). Obviously, the rapid population growth places some burdens on government budgets to meet necessities such as schools, improved health care, improved housing, and employment opportunities. This, in turn, may amplify social discontent, particularly among unemployed college graduates located in countries with the fastest population growth rates (Salehi-Isfahani, 2010). Figure 2 shows the percentage of sectors’ contribution to the GDP. The service sector, which grew 3.3% in 2016, contributed 45% of regional GDP growth. The industrial sector, which grew 3%, contributed 46% of regional GDP, with the largest contribution from Algeria and the smallest from Libya. Growth in services and industry demonstrates the region’s strong recovery from the punishing effects experienced in the aftermath of the Arab Spring when the two sectors made consecutive negative contributions to regional GDP growth in 2012 and 2013. In contrast, agriculture, which grew a meager 0.24% in 2016, contributed less than 9% to regional GDP growth. This is due to the fact that much of the agricultural land is located in arid and semi-arid agro-ecological conditions. Thus, agricultural production is dependent on climatic hazards and often depends on irregular and insufficient rainfall. As a result, the extension of cultivated areas has been limited and the pressure on land continues to increase due to population growth. For the whole region, therefore, each hectare cultivated must feed today twice as many inhabitants. Solving the problem of food insecurity has been a major component of development policies for all North African countries, without exception. The agricultural

% employees

Service sector

43

60.3%

38.4%

20.7%

8.4 H

42.4 H (18%)

54%

46%

73.1%

5.2%

1.95%

19%

40%

%agricultural workers

Algeria 2,381,741

21.6%

5.7 H

Cultivated area (million)

% employees

8.7 H (68%)

Agricultural area (million)

Industrial sector

Agricultural sector

Source Arab Organization for Agriculture Report, 2019

Characteristics economy

48.9%

Urban

72.2% 51.1%

> 65 years

Rural

6.1%

< 15 years

Age-dependent population

%Distribution

37 1.23%

Population (millions)

Characteristics population

Population growth rate

Morocco 710,850

Countries

Total area (Km2 )

53.4%

33.2%

13.4%

1.2 H

5 H (55%)

65%

35%

74.5%

7.5%

1.12%

11

163,610

Tunisia

Table 1 Some population and economy-related Indicators for North African countries in 2019 Libya

/

/

3%

1.5 H

3.6 H (9%)

/

/

74.8%

4.6%

1.5%

6

1,759,540

Egypt

38%

30%

32%

3.5 H

4.1 H (10%)

42.7%

57.3%

73.2%

4.5%

2.0%

100

1,010,408

Total

/

/

/

20 million H

64 million H

/

/

72%

28%

/

196 millions

4,266,609 Km2

38 A. Bellout et al.

The Subjective Well-Being in North Africa …

39

Fig. 2 Sector contribution to GDP growth in the North Africa region, 2008–2016. Source African Development Bank 2018, Report North Africa Economic Outlook, P9

and agri-food policies that have been followed have aimed at ensuring an adequate volume of food for the population and stable supplies on national markets, while guaranteeing affordable prices for the most vulnerable segments of the population (especially in the regions). These policies have been associated with many incentives for the personal well-being of individuals that can occur in agricultural and production processes, especially the food production process which greatly contributes to the well-being of consumers, especially local consumers.

2.3 Data Collection The scarcity of work of geographers on well-being is surprising after a brief review of the ways in which personal well-being has been exploited in various social and economic fields in different geographical areas. This chapter deals with the relationship between subjective well-being and its effects on agriculture and urban lands, which is often still seen as a simple causal relationship so that one can easily build models around them. However, there is a very complex set of evidence in which the empirical results appear, at first glance, very contradictory. These contradictory outcomes appear to be explained on an individual basis when one explores the importance of the social, political, and economic contexts in which such change occurs, but the perception of the relationship more generally remains rather complex. This therefore requires a comprehensive look at the historical generation of the debate on this subject. It also requires understanding the context of the broader policy debate in the late twentieth century to make its seemingly intractable elements more understandable. In terms of developing improved contexts for the development of agricultural territories that contribute to people’s well-being, there are substantial potential opportunities to improve people’s well-being related to the advantages of agricultural activities that can contribute to this well-being (Oswald and Wu (2009).

40

A. Bellout et al.

One tremendous opportunity is to develop pertinent practical research (i.e., action research) to encourage farmers on the one hand and non-farm citizens on the other hand to work together using an Action Research approach (Bousbaine and Bryant (2015, 2017b)). Action research has been used in many domains and when the people involved are committed to achieving constructive results, the results can be extraordinary. In addition, secondary data is relevant to our study using publicly available databases and previously published research. The reports included the statistics of official bodies in the agricultural and economic sector, as well as the reports of the Arab Organization and the report of the African Development Bank, including demographic and economic information for the countries of the study area.

3 Results and Discussion 3.1 Subjective Well-Being and Its Relationship to Sustainable Rural Development: From the classical economic perspective, sustainability means continuity and maximization of economic well-being for the longest possible period. As for measuring this well-being, it is mostly in accordance with rates of income and consumption, and this includes many obstacles to human well-being such as food, housing, transportation, health, education, and the environment. Some researchers in the field of geography in the agricultural aspect and urban agricultural landscapes, which are basically one of the most important indicators of personal well-being because individuals in societies have access to a safe and healthy food source in addition to work that improves the standard of living for a family and others. Achieving high income is no longer a sufficient indicator to reach the desired level of subjective well-being, so it is necessary to put into place many considerations related to that, such as the environment and the social characteristics of the agricultural family members. Subjective well-being is one of the most important issues facing the world today and is central to the development of social policy for rural areas. The well-being attitudes of individuals/social groups may differ not only between rural and urban areas but also between rural areas. The concept of well-being in the rural environment primarily seeks to understand the interactions between the diversity of factors so that we can better understand individual, family, and societal pathways. Figure 3 shows that the existing territorial differences among rural areas have an impact on their development trends. According to the idea that well-being research can influence rural development, there can be a detailed interconnection between these two issues (Vaznonien˙e, 2014). The figure also indicates how the well-being research levels are related to the main rural areas’ management and development approaches, as well as how these two approaches become a part of well-being research levels.

The Subjective Well-Being in North Africa …

41

Fig. 3 Interconnection of well-being research and rural development. Source Gintar˙e Vaznonien˙e (2014), Well-being research for rural development, research for rural development, V 2, P248

Subjective well-being in agricultural areas, whether in the Rural or on the Periphery of cities, includes many factors that fall into a hierarchical form as in Fig. 3. Human well-being includes three stages: Stage 1 relates to the standard of living and health and the provision of a safe environment, as this stage is considered the basis for achieving the personal well-being of individuals, where individuals live in a more organized and appropriate environment with regard to agricultural areas, transportation, energy, a safe environment free of pollution, and others. Stage 2 is a reflection of the first stage, which includes the transition to a green economy and the provision of healthy food and a clean environment; as for Stage 3, it means ensuring access to high levels of services related to the transport network, communications, and advanced technology in all aspects of life (Fig. 4). Since their independence, North African countries have implemented many strategies to revive rural areas, such as the 2006 Rural Renewal Strategy in Algeria, the

Fig. 4 Subjective well-being hierarchy. Source Authors’ elaboration

42

A. Bellout et al.

Rural Tourism Development Strategy in Morocco, and other strategies that have been applied in the rest of the countries with the aim of improving the policy system for the integrated development of urban and rural areas, as well as accelerating modernization of Agriculture and rural areas in these countries to enhance rural sustainability. Rural sustainability is important not only in North African countries but also in other developing countries. Many developing countries, including North African countries, face important challenges in their economic and social development. The issue is more important than resolving the contradiction between the people’s increasing need for a better life and the unbalanced and insufficient national development, especially urban and rural development. The primary goal of the Rural Revitalization Strategy is to improve the lives of rural people. As indicated by the Organization for Economic Cooperation and Development, there is a growing demand for the use of subjective well-being as a way to move beyond the classic income-based approach.

3.2 Spatial Differences in Subjective Well-Being of North African Countries The majority of studies conducted on SWB indicated that there is a similarity in the average level of SWB in different places once we control personal characteristics (Morrison, 2007). This argument is especially prevalent among economists, who usually assume that people can appear to have a strong desire to move into regions with a better quality of life when transportation costs are low and information about what it would be like to live elsewhere (Faggian et al., 2012). As a result, with the more desirable places becoming as a result crowded and expensive, an equilibrium will be reached and the average level of luxury should be the same in all regions. In support of this suggestion, (Ballas & Tranmer, 2012) failed to find a statistically significant geographical variance in happiness after controlling for socioeconomic and demographic variables. However, the above arguments should be interpreted with caution. Since the spatial and institutional barriers are negligible, adaptation to local factors can be very slow (Faggian & Royuela, 2010) and the spatial distribution of SWB must be uneven in the long run. Several pilot studies have also noted important differences in SWB between different regions. The majority of current studies are based on the country level (Berry & Okulicz-Kozaryn, 2011). The country in which a person lives is said to have inevitable consequences for his or her life through employment, quality of health care, and others. In support of this argument, scholars from various disciplines have revealed large variations and fundamental differences in the SWB reported across nature (Kalmijn & Veenhoven, 2005; Veenhoven, 1993). National differences are usually examined by comparing the overall score of SWB in different countries (Diener et al., 1995) with a fixed effects test when adjusting individual life satisfaction

The Subjective Well-Being in North Africa …

43

Fig. 5 Average happiness scores for African countries in 2014. Source Gallup World Poll data

(Aslam & Corrado, 2012). Figure 5 shows the average happiness scores for North African and African countries together. With Fig. 5, we notice that North African countries are ranked first in terms of Gallup’s classification of happiness, and Algeria and Libya come to the forefront of these countries, due to their high oil revenues because they are countries with great oil potential in Africa, in addition to the agricultural potential in Algeria, which makes this sector of great importance to the government and the Algerian people. As for Tunisia, Morocco, and Egypt, they depend on tourism in the first place and agriculture in the second place. Through the limited literature on this topic, it was concluded that happiness in less developed countries, including North African countries, was greater in urban settings, but that this difference between urban and rural areas tends to disappear with economic development (Veenhoven & Ehrhardt, 1995). Figure 6 shows how average levels of subjective well-being in North African countries have increased at different rates for those living in urban and rural areas. This figure draws a distinction

44

A. Bellout et al.

Fig. 6 Urban–rural differences in subjective well-being in North African Countries. Source African Development Bank 2018, Report North Africa Economic Outlook, P9

between the way subjective well-being changes with economic development in the very large metropolitan centers compared to the smaller cities and rural areas. Since income and economic opportunity in cities are higher in stage (A) of Fig. 6, they are accompanied by higher levels of happiness than in rural areas. When incomes rise and technology develops, and when transportation and digital infrastructure improve, access to rural areas becomes more diverse. This broad shift in the nature of work is ultimately reducing the happiness differences between urban and rural areas to the point where average happiness levels in rural areas, villages, and small towns approach and even exceed those in large cities. The majority of people in stage (B) of Fig. 6 choose to live in urban areas because they provide a higher quality of life in terms of employment opportunities and access to public facilities and services. While the benefits of city well-being in developing countries may outweigh the disadvantages of settlements outside the big city, this may not be the case for the majority of urban dwellers in developed countries. Many residents of the restructured rural areas of advanced economies are no longer dependent on agriculture, and the expansion of urban centers means that many find themselves living and working near urban centers and are able to “borrow” the positive effects of much larger cities while they are there, relatively isolated from negative influences. There may also be a selection of unhappy people in the cities and happy people in the countryside. Our analysis confirmed that in many parts of the world there were no differences between urban and rural areas in individual happiness: the driving forces were personal characteristics such as age, income, and marriage, with some variation according to the level of development. However, there were exceptions: in rapidly urbanizing Africa and Asia, subjective well-being was higher in large cities than

The Subjective Well-Being in North Africa …

45

elsewhere; in Northern and Western Europe, happiness was lowest in large cities and increased to a maximum in rural areas today. Through this, it can be said that developing countries follow the same path that European countries followed in the past years.

3.3 Analytical Model and Variables Our main purpose is to investigate the effect of the impact of subjective well-being on agriculture and urban land. In some cases, it is relevant to speculate that causality may also run in the other direction (the grey arrow from happiness to Subjective Well-Being and its influence on consumer agri-food demands in Fig. 5). Figure 7 presents the analytical model used to identify elements of variables that may affect subjective well-being. We distinguish between two types of variants. “Exogenous variables” are variables that may affect subjective well-being and influence consumer demands for agri-food and may also directly lead to happiness. On the other hand, “Intermediate variables” are likely to be influenced by subjective wellbeing and thus may mediate a causal effect from subjective well-being to happiness. Controlling external variables allows us to determine the overall and causal impact of agriculture on happiness, including indirect effects operating through income, exposure to risk, social networks, and so forth. Exogenous variables include Areas that are exploited agriculturally, Diversity of production characteristics, Food security versus food insecurity, and Unused land with potential for agricultural production. All other control variables are viewed as “intermediate.” Among these variables, we first include a measure of income, measured at the household level. While the positive correlation between income and happiness is a record result in individual happiness analyses, it has been strongly discussed whether this association is driven by absolute or relative income (e.g., Berry, 2009; Cummins, 2000; Easterlin, 1974). This suggests that the effect of happiness on self-employment may also be weak. However, to take into account the possibility of an inverse link from happiness to selfemployment, we carry out effective analysis of variables, where self-employment is run through community-level properties, which are external to the psychological characteristics of respondents.

4 Conclusions Subjective well-being is one of the most important issues facing the world today and is central to the development of social policy for rural areas. The well-being attitudes of individuals/social groups may differ not only between rural and urban areas but also between rural areas. The concept of well-being in the rural environment

46

A. Bellout et al.

Fig. 7 Analytical model. Source Authors’ elaboration

primarily seeks to understand the interactions between the diversity of factors so that we can better understand individual, family, and societal pathways. Revealing the interconnection between well-being research and rural development is indeed important not only in the scientific context, but also for local authorities or local government. It gives the direction of how to observe changes both in well-being research at the local level and rural development in whole. It is observed by Fig. 5 that North African countries are ranked first in terms of Gallup’s classification of happiness, and Algeria and Libya come to the forefront of these countries, due to their high oil revenues because they are countries with great oil potential in Africa, in addition to the agricultural potential in Algeria, which makes this sector of great importance to the government and the Algerian people. As for Tunisia, Morocco, and Egypt, they depend on tourism in the first place and agriculture in the second place. The benefit of well-being research for rural development is important in many ways. It enables knowledge of the well-being of rural people, reveals existing problems as well as positive changes, gives the opportunity to monitor how well-being is improved, and informs rural actors about their role in welfare research and their potential impact on rural development.

The Subjective Well-Being in North Africa …

47

References Akkari, A., & Bryant, C. R. (2016). The co-construction approach to developing adaptation strategies in the face of climate change and variability: A conceptual framework. Agricultural Research, 2016, 2–12. Arab Organization for Agricultural Development(AOAD). (2012). Arab Agricultural Statistics Yearbook (Khartoum: AOAD), http://www.aoad.org/Agricultural_Statistical_Book_Vol32.pdf. Accessed on March 17, 2016. Aslam, A., & Corrado, L. (2012). The geography of well-being. Journal of Economic Geography, 12(3), 627–649. Ballas, D., & Tranmer, M. (2012). Happy people or happy places? A multilevel modeling approach to the analysis of happiness and well-being. International Regional Science Review, 35(1), 70–102. Benz, M., & Frey, B. (2008). Being independent is a great thing: Subjective evaluations of selfemployment and hierarchy. Economica, 75(298), 362–383. https://doi.org/10.1111/j.1468-0335. 2007.00594.x. Berry, B. J. L., & Okulicz-Kozaryn, A. (2011). An urban-rural happiness gradient. Urban Geography, 32(6), 871–883. Berry, A. (2009). Improving measurement of Latin American inequality and poverty with an eye to equitable growth policy. IPD working paper series. Bousbaine, A. D., Bryant, C. R. (2017a). Les systèmes innovants alimentaires, cas d’étude : la Ceinture Aliment Terre de Liège (Translation: Innovative alternative agri-food systems: a case study of the Liege Food-Land Belt) (Vol. 2, 20pp). Belgeo: Revue belge de géographie, ISSN: 2294-9135. Bousbaine, A. D., Bryant, C. R. (2017b). Action research: An essential approach to the development of the agricultural field of the Mitidja Plain, North of Algeria, and comparisons with other territories in developed countries. Bousbaine, A. D., & Bryant, C. R. (2015). The integration of action research and traditional field research to provide sustainable solutions to maintaining periurban agriculture. Geographical Research, 2015, 1–11. Bousbaine, A. D. (2020). Ville et agriculture face à l’émergence des systèmes agro-alimentaires innovants. Etude de cas dans deux agglomérations wallonnes : Charleroi et de Liège (Translation : Cities and agriculture faced with the emergence of innovative agro-food systems : Charleroi and Liège), 2020, Pp. 484 (PhD thesis). Under the direction of Serge Schmitz (Ulg) and co-direction of Christopher Bryant (U de Montréal and U de Guelph, Canada). Brereton, F., Clinch, J. P., & Ferreira, S. (2008). Happiness, geography, and the environment. Ecological Economics, 65, 386–396. Brulé, G., & Veenhoven, R. (2015). Geography of happiness. Configurations of affective and cognitive appraisal of life across nations. International Journal of Happiness and Development, 2(May/June). Cummins, R. A. (2000). Personal income and subjective well-being: A review. Journal of Happiness Studies, 1(2), 133–158. Diener, E. (2000). Subjective well-being: The science of happiness, and a proposal for a national index. American Psychologist, 55, 34–43. Diener, E., Suh, E., Smith, H., & Shao, L. (1995). National differences in reported subjective well-being: Why do they occur? Social Indicators Research, 34(1), 7–32. Diener, E., Helliwell, J. F., & Kahneman, D. (Eds.). (2010). International differences in well-being. Oxford University Press. Easterlin, R. (1974). Does economic growth improve the human lot? Some empirical evidence. In P. A. David & M. W. Reder (Eds.), Nations and households in economic growth: Essays in honor of Moses Abramowitz. Academic Press. Faggian, A., & Royuela, V. (2010). Migration flows and quality of life in a metropolitan area: The case of Barcelona-Spain. Applied Research in Quality of Life, 5(3), 241–259.

48

A. Bellout et al.

Faggian, A., Olfert, M. R., & Partridge, M. D. (2012). Inferring regional well-being from individual revealed preferences: The ‘voting with your feet’ approach. Cambridge Journal of Regions, Economy and Society, 5(1), 163–180. Fenglong, W., & Donggen, W. (2015). Geography of urban life satisfaction: An empirical study of Beijing. Travel Behaviour and Society. Vol. 5, https://doi.org/10.1016/j.tbs.2015.10.001. Gerstorf, D., Ram, N., Goebe, J., Schupp, J., Lindenberger, U., & Wagner, G. G. (2010). Where people live and die makes a difference: Individual and geographic disparities in well-being progression at the end of life. Psychology and Aging, 25(3), 661–676. Helliwell, J. F., & Wang, S. (2012). The state of world happiness. Helliwell, J. F., Layard, R. & Sachs, J. (Eds.) (2012). New York, UN Sustainable Development Solutions Network. Helliwell, J. F. (2015). World happiness report. Kalmijn, W., & Veenhoven, R. (2005). Measuring inequality of happiness in nations: In search for proper statistics. Journal of Happiness Studies, 6(4), 357–396. Lawless, N. M., & Lucas, R. E. (2011). Predictors of regional well-being: A county-level analysis. Social Indicators Research, 101, 341–357. https://doi.org/10.1007/s11205-010-9667-7 Lucas, R. E., Cheung, F., & Lawless, N. M. (2014). Investigating the subjective well-being of U.S. regions. In P. J. Rentfrow (Ed.), Geographical psychology: Exploring the interaction of environment and behavior. Washington, DC: American Psychological Association. McMillan, M., & Rodrik, D. (2011). Globalization, structural change and productivity growth. NBER Working Paper No. w17143, Available at SSRN: https://ssrn.com/abstract=1866102. Morrison, P. (2007). Subjective well-being and the city. Social Policy Journal of New-Zealand, 31, 74–103. Oswald, A. & Wu, S. (2009). Well-being across America: evidence from a random sample of one million U.S. citizens. University of Warwick working paper. Accessed June 18, 2010. http:// www2.warwick.ac.uk/fac/soc/economics/staff/academic/oswald/oswaldwuwellbeing2009.pdf. Oswald, A., & Wu, S. (2010). Objective confirmation of subjective measures of human well-being: Evidence from the USA. Science, 327(5965), 576. Ott, J. C. (2012). An eye on happiness. Happiness as an additional goal for citizens and governments. Ridderprint BV. Plaut, V. C., Markus, H. R., & Lachman, M. E. (2002). Place matters: Consensual features and regional variation in American well-being and self. Journal of Personality and Social Psychology, 83, 160–184. https://doi.org/10.1037/0022-3514.83.1.160 Rentfrow, P. J. (2018). Geographical variation in subjective well-being. In E. Diener, S. Oishi, & L. Tay (Eds.), Handbook of well-being. Salt Lake City, UT: DEF Publishers. DOI: nobascholar.com. Salehi-Isfahani, D. (2010). Human development in the middle east and north Africa. Human Development Report Office (HDRO), United Nations Development Programme (UNDP). Seddik, H, & Bryant, C., R. (2012). Quelles dispositions prendre afin que la campagne Algérienne réapproprie ses fonctions ? International Journal of Advanced Studies and Research in Africa, 3(1), 135–156. Tiliouine, H. (2015). Quality of Life and Wellbeing in North Africa—Algeria, Egypt, Libya, Morocco and Tunisia, Chapter22 In book: Global Handbook of Quality of Life Publisher: Springer Netherlands Editors: Wolfgang Glatzer, pp. 507–530. Vaznonien˙e, G. (2014). Wellbeing research for rural development. Research for Rural Development, 2, 248. Veenhoven, R., & Ehrhardt, J. (1995). The cross-national pattern of happiness: Test of predictions implied in three theories of happiness. Social Indicators Research, 34, 33–68. Veenhoven, R. (1993). Happiness in nations: Subjective appreciation of life in 56 nations 1946– 1992. Risbo. Weiner, E. (2009). The geography of bliss, one grump’s search for the happiest places in the world. Grand Central Publishing, New York, ISBN: 9780446511070.

Territorial Challenges

Researching Quality of Life in Old Age: Some Conceptual and Methodological Principles José de São José

1 Introduction The thought about quality of life (QoL) started many centuries ago, being found in the writings of the philosopher Aristotle (Alkire, 2015). However, it was only from the middle of the twentieth century that QoL timidly entered the agendas of scientific research and public policy. This occurred mainly in the 60s, particularly in the USA, where the first large national surveys on various social indicators related to QoL and well-being were carried out (Ferriss, 2004). The development of this type of surveys and other similar studies became known as the “movement of social indicators” (Ferriss, 2004). This movement was the first major reaction against the assessment of the progress and development of the nations based exclusively on economic variables, such as the gross national product (GDP) (Fernández-Ballesteros & Santacreu, 2014; Rojas, 2014). Although the topic of QoL has entered the research and public policy agendas in the mid-twentieth century, it was only later that it gained prominence in these two agendas. In the research agenda, this prominence occurred from the 1980s onwards (Galloway, 2006), especially after the creation of various QoL measurement instruments, such as the WHOQOL (The WHOQOL Group, 1995) and the SF-36 (Stewart & Ware, 1992). In the public policy agenda, the prominence of QoL occurred only in the twenty-first century. Three events were crucial to draw the attention of policy makers to the relevance of QoL. In 2007, the European Commission, the European Parliament, the Club of Rome, the Organization for Economic Cooperation and Development, and the World Wildlife Fund, organized an international conference entitled “Beyond GDP,” whose objective was to discuss the development J. de São José (B) Faculdade de Economia, Universidade do Algarve, Faro, Portugal e-mail: [email protected] Centro Interdisciplinar de Ciências Sociais (CICS.NOVA), Faculdade de Ciências Sociais e Humanas (NOVA FCSH), Lisbon, Portugal © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 E. Vaz (ed.), Geography of Happiness, Contributions to Regional Science, https://doi.org/10.1007/978-3-031-19871-7_4

51

52

J. de São José

of alternative metrics to measure the progress and well-being of nations. Two years later, the Commission of the European Communities made a communication to the Council and the European Parliament entitled “GDP and beyond. Measuring progress in a changing world,” concluding that there is a need of complementing GDP with statistics covering other dimensions of well-being/QoL. Also in 2009, the report by the Commission on the Measurement of Economic Performance and Social Progress was published, coordinated by Joseph Stiglitz, Amartya Sen, and Jean-Paul Fitoussi. One of the main messages of this report refers to the need “(…) to shift emphasis from measuring economic production to measuring people’s well-being. And measures of well-being should be put in a context of sustainability” (Stiglitz et al., 2009: 12). These three events formed the second, and decisive reaction, against measuring QoL based exclusively (or predominantly) through economic variables and were a decisive contribution to the creation of instruments for measuring QoL both at transnational and national levels. The Eurostat QoL Index and the OECD Better Life Initiative are examples of transnational measures of QoL. In turn, the Canadian Index of Well-Being and the New Zealand Living Standards Framework are examples of national measures of QoL. Alongside the development of transnational and national QoL instruments, there has been, especially over the last decade, a truly impressive increase in research not only on QoL in general, but also on the QoL of certain populations, including the older population. Research on QoL in old age has been driven by several factors, among which the following stand out: (1) the growing concern of policy makers about the impacts of the population aging, especially on social care and health systems (Bowling et al., 2013; Walker & Mollenkopf, 2007); (2) the use of QoL as an end point in the evaluation of social interventions and public policies (Bowling & Stenner, 2011; Verdugo et al., 2005); (3) the famous dictum of the World Health Organization (WHO) “years have been added to life and now the challenge is to add life to years” (Bowling et al., 2013; Walker & Mollenkopf, 2007); and (4) the exponential growth of research on health-related QoL (HRQoL), which predominates in the field of health sciences (Walker & Mollenkopf, 2007). The impressive increase of research on QoL in old age has not been accompanied by a consolidation of theoretical/conceptual and methodological work. Currently, there is no consensus on what QoL in old age is and how to measure it. This makes the stock of knowledge on this research topic difficult to trace and the comprehensive understanding of it difficult to achieve. If QoL in old age is, as it appears to be, an important issue from a scientific and public policy standpoint, then it would be beneficial for the scientific community and policy makers to reach an agreement on some basic conceptual and methodological principles. This agreement would facilitate the comparison between studies and the consolation of empirical evidence. This chapter discusses the conceptual and methodological aspects of QoL in old age based on a review of the most relevant literature on these two aspects. In addition, as a corollary of this discussion, it proposes a set of conceptual and methodological principles to study QoL in old age. Section 1 addresses the theoretical/conceptual aspects of QoL in old age, namely definitions, theories/conceptual models, lay views, and neglected conceptual aspects. This section ends with a proposal of a set of

Researching Quality of Life in Old Age…

53

conceptual principles for studying QoL in old age. Section 2 includes a discussion of the major methodological limitations in this field of research, followed by a proposal of a set of methodological principles for studying QoL in old age. In Sect. 4, some concluding remarks are provided.

2 What is QoL in Old Age? 2.1 Definitions of QoL in Old Age As already stressed, the literature on QoL in old age is immense. A search in the Web of Science (all databases), carried out on October 20, 2022, using the keywords “quality of life” and “older people,” “later life,” “old age,” “elders,” “elderly people,” and “older adults,” in the field “topic,” produced 37.601 results. Hambleton and collaborators (2009: 3) suggest that the literature on QoL in general “can be described as a jungle: vast, dense, and difficult to penetrate.” The same can be said in relation to the literature on QoL in old age. In the midst of this immensity of literature, it is not possible to identify a widely accepted definition of QoL in old age (Brown et al., 2004; Galloway, 2006; Halvorsrud & Kalfoss, 2007; Walker & Mollenkopf, 2007). Table 1 presents some of the many definitions of QoL in old age, which exhibit dissimilarities between them. The lack of a consensual definition of QoL may be explained by several factors, including: (1) the extreme complexity of the concept, which makes it “very difficult to pin down scientifically” (Walker & Mollenkopf, 2007: 4); (2) the adoption of different disciplinary perspectives to make sense of it (Walker & Mollenkopf, 2007); and (3) the conceptual confusion that still exists between this concept and other concepts, such as well-being, life satisfaction, happiness, successful aging, and active aging (Fernández-Ballesteros, 2011). Although different, the definitions listed above reflect the idea that QoL denotes an assessment/evaluation. This is supported by Veenhoven (2000: 3) with respect to QoL in general, who alert us to the importance of establishing “what thing is evaluated by what standard.” With respect to the issue of “what thing” is evaluated, there is a good agreement among academics in the field of QoL in old age (and QoL in general) that the “thing” is constituted by several dimensions (Brown et al., 2004; Fernández-Ballesteros, 2011; Higgs et al., 2003; Walker & Mollenkopf, 2007). Therefore, QoL is made of several dimensions reflecting the multiple domains of human life. Nevertheless, most of the research on QoL in old age has reduced QoL to the health dimension due to the predominance of studies on HRQoL (Fernández-Ballesteros, 2011; Halvorsrud & Kalfoss, 2007). There are several questions that arise regarding the dimensions of QoL in old age: (1) Which dimensions to consider? (2) Who chooses the dimensions? (3) Should

54

J. de São José

Table 1 Definitions of QoL in old age Authors

Definitions

Higgs et al. (2003)

“QoL should be assessed as the degree that human needs are satisfied” (p. 243)

Sarvimaki and Stenback-Hult (2000) “A sense of well-being, of meaning, and of value or self-worth” (p. 1027) Walker and Mollenkopf (2007)

“QoL should be regarded as a dynamic, multifaceted, and complex concept, which must reflect the interaction of objective, subjective, macro, micro, positive, and negative influences. (…) QoL in old age is the outcome of the interactive combination of life course factors and immediate situational ones. (…) when it comes to comparisons between young people and older people, health and functional capacity achieve a much higher rating among the latter. (…) the sources of QoL in old age often differ between groups of older people. (…) subjective self-assessments of psychological well-being and health are more powerful than objective economic or sociodemographic factors in explaining variations in QoL ratings. (…)” (pp. 8–9)

Register and Herman (2006)

“QOL is a cumulative process that is generated through an ongoing series of specific connections and disconnections that result from interactions with the forces and processes people encounter in their daily life. (…) Quality of life is a dynamic personal perception that is enhanced by positive life connections and diminished by disconnections.” (pp. 340–341)

Lawton (1991)

“Multidimensional evaluation, by both intra-personal and social normative criteria, of the person environment system of an individual in time past, current, and anticipated” (p. 6)

dimensions be valued? (4) Who should value dimensions? (5) What is the standard for a “good” quality of life? (6) Who sets this standard? With regard to the first question (Which dimensions to consider?), the research on HRQoL has focused mainly on the health dimension, while the studies that use a broader conceptualization of QoL have considered multiple dimensions, but there is no consensus on the set of dimensions to be assessed. Three well-known instruments to measure QoL in old age, namely the CASP-19, the WHOQOL-OLD and the OPQOL, propose different sets of dimensions. The CASP-19 includes four domains, including control, autonomy, pleasure, and self-realization (Higgs et al., 2003; Hyde et al., 2003). In turn, the WHOQOL-OLD includes six facets, namely sensory abilities, autonomy, past, present and future activities, social participation, death and dying, and intimacy (Power et al., 2005). Lastly, OPQOL includes eight themes: life overall, health, social relationships and participation, independence, control over life, freedom, home and neighborhood, psychological and emotional well-being, financial circumstances, and religion/culture (Bowling & Stenner, 2011).

Researching Quality of Life in Old Age…

55

The lack of consensus around dimensions is not surprising, as the assessments of QoL in old age have been carried out from the viewpoint of different scientific disciplines. Nevertheless, this lack of consensus is not the main problem. The prevailing ambiguity with respect to the nature of the dimensions, as well as the calculation of total scores mixing dimensions of different nature, and the under-exploration of interrelationships between different dimensions are the main problems. With respect to the first problem, it is important to be aware that some dimensions are not constituents or outcomes of QoL, but rather determinants of QoL. As highlighted by Grewal and collaborators (2006) and Higgs and collaborators (2003), it is important to distinguish what constitutes QoL in old age from the factors that determinate it. Veenhoven (2000), referring to QoL in general, also emphasizes that it is important to distinguish opportunities for a good life (life chances) from the good life itself (life results). But this distinction has not always been made. For example, the WHOQOL-OLD includes dimensions that refer to determinants of QoL, such as sensory abilities (e.g., impairments to senses) and dimensions that refer to outcomes of QoL, such as autonomy (e.g., freedom to make own decisions), offering the possibility to calculate a total score of QoL that mixtures these two dimensions. Another important issue in relation to the nature of the dimensions of QoL is that some refer to aspects intrinsic to individuals, while others refer to aspects extrinsic to individuals. Veenhoven (2000) advocates that it is also important to distinguish inner qualities of life (e.g., physical health) from outer qualities of life (e.g., clean air). However, sometimes these different aspects are also not differentiated when measuring QoL. For example, the OPQOL includes the health dimension, which refers to aspects intrinsic to the individuals, and the home and neighborhood dimension, which refers to aspects extrinsic to individuals, offering the possibility of calculating a total score of QoL that mixtures these two dimensions. The last issue in relation to the nature of the dimensions has to do with the distinction between objective dimensions (e.g., level of schooling and living conditions) and subjective dimensions (e.g., life satisfaction and subjective well-being). Usually, objective dimensions (and indicators) refer to those things that are not dependent on our understanding of them. A car is independent of our understandings of it; thus, a car is objective. In turn, subjective dimensions (and indicators) have to do with evaluations (or appraisals/appreciations) of certain things. These evaluations are subjective in the sense that they exist only insofar as the subject performs them. In addition, it is important to underline that subjective evaluations may have a good match in reality or not. Examples of the first situation are negative subjective evaluations of living conditions that are objectively “bad” or positive subjective evaluations of living conditions that are objectively “good.” In turn, examples of the second situation are negative subjective evaluations of living conditions that are objectively “good” or positive subjective evaluations of living conditions that are objectively “bad.” These situations are relevant and should be carefully analyzed. The discussion around the objective and subjective dimensions of QoL is also carried out by the Capability Approach, initially developed by Sen (1985). This approach argues against a resource-based approach to measure well-being/QoL,

56

J. de São José

which is an approach of objective nature. The main reason is that resources are not “intrinsically valuable,” that is, they are means to achieve goals (they have an instrumental nature). As Alkire (2015: 6) emphasizes, “Well-being arguably depends not on the mere existence of resources but on what they enable people to do and be.” On the other hand, the Capability Approach also argues against a utility-based approach to measure well-being/QoL, which is an approach of subjective nature, as some activities are worthwhile in themselves rather than means to achieve happiness and joy (Alkire, 2015). But perhaps the most powerful argument against a utility-based approach is the phenomenon of “adaptive preferences” (adapting preferences to the circumstances). People facing with severe deprivation, but who exhibit acceptable levels of QoL due to the adaptation of their expectations and aspirations, can thus remain outside the radar of public policies (Alkire, 2015). In this vein, Alkire (2015: 2) argues that “well-being and its lack should be measured in the space of capabilities,” instead of in the space of resources (material, social, etc.) or utility (subjective well-being, happiness and life satisfaction). Capabilities refer to “a person’s real freedoms or opportunities to achieve functionings” (Robeyns, 2017: 39), being functionings “those beings and doings that constitute human life and are central to our understandings of ourselves as human beings” (p. 39). Therefore, QoL is reflected on “the freedom people have to enjoy valuable activities and states” (Alkire, 2015: 3). This means that QoL results, essentially, from the opportunities people have to lead the life they value (Stephens, 2017). Hence, in addition to objective and subjective dimensions, the Capability Approach adds the concept of capabilities, which result from the combination of objective aspects, intrinsic or extrinsic to individuals (e.g., functional capacity, built environment) and subjective aspects, such as evaluations of objective things, which can be intrinsic or extrinsic to individuals (e.g., self-perception of functional capacity, satisfaction with the built environment). The objective-subjective dichotomy with respect to well-being is addressed in a more complex way by Sayer (2011). He advocates that well-being (and ill-being) are not merely states of mind but rather “objective states of being which people strive to discover, achieve or create” (p. 134). He justifies this by calling our attention to the fact that we can usually provide reasons for how we feel, meaning that “our subjective feelings seem to be about things which are objective in the sense of independent of them” (p. 134). He also says that “If well-being were merely subjective, then our views on it would be infallible” (p. 134). Nevertheless, Sayer (2011) clarifies that this objectivist conception of well-being does not imply that there is only one way of achieving it, as there are different kinds of well-being. But at the same time, he argues that this does not imply that “what is good is simply relative to one’s point of view” (p. 135), given that different cultures offer different forms of flourishing and suffering. In this vein, Sayer (2011) suggests that well-being is “objective and plural, but not relative” (p. 134). In the field of intellectual disabilities, Goode (1994) also suggests that “the QoL of a person reflects the cultural heritage of the person and those who surround him or her” (p. 148). The World Health Organization (The WHOQOL Group, 1995) also mentions the relevance of culture and value systems in assessments of QoL. Therefore, QoL is imbued with culture and cannot be understood outside of

Researching Quality of Life in Old Age…

57

it, although some authors claim that there are aspects of QoL that are universal, that is, found in all cultures (e.g. Schalock & Verdugo, 2002). There is a good agreement in the scientific community in favor of the inclusion of objective and subjective dimensions in the conceptualization of QoL in old age (e.g., Brown et al., 2004; Fernández-Ballesteros, 2011; Walker & Mollenkopf, 2007). However, Fernández-Ballesteros (2011) concluded, in an extensive literature review, that QoL in old age has been conceptualized mainly as a subjective state, depending on the individual evaluations of certain domains of life. With respect to the under-exploration of the relationships between dimensions of QoL in old age, most of the studies, mainly quantitative studies that used scales of QoL, have not, in fact, explored these relationships. The interconnection between different domains of life emerged as one of the conclusions of the qualitative study conducted by Hendry and McVittie (2004) on the perspectives of older people on the QoL in old age. Indeed, human lives are not organized in separate domains, like silos. Such a view would be artificial. Later on, we return to the Hendry and McVittie’s study. As we will also see later, some theoretical/conceptual models of QoL in old age assume that the interrelationships between dimensions exist. In what concerns the second question regarding the dimensions of QoL (Who chooses the dimensions?), the general tendency has gone in the direction of combining the perspectives of experts/researchers with the perspectives of older people themselves in the identification of relevant dimensions. This happened in the development of the CASP-19 and the WHOQOL-OLD (the creation of the OPQOL was based exclusively on the perspectives of older people). In this regard, it is important to say that over the past few decades, research on lay conceptions of QoL in old age has increased considerably (this topic is addressed later). Taking into account the older people’s perspectives on QoL in old age is important not only in qualitative studies, but also in quantitative studies, particularly in the construction of instruments to measure QoL. This means that studying/assessing QoL in old age should combine, preferably, lay views with expert views, avoiding any kind of paternalistic approach. Regarding the third and fourth questions regarding the dimensions of QoL (Should dimensions be valued? Who should value dimensions?), the predominant trend in quantitative studies that use scales of QoL is to attribute the same importance/relevance to different dimensions, as it happens with the three instruments mentioned above (the CASP-19, the WHOQOL-OLD and the OPQOL). This trend is not found in most of the qualitative studies, especially in those studies aimed at capturing the lay views of older people about QoL in old age, which have been also concerned with exploring the importance that each dimension of QOL has for them. In this regard, Felce and Perry (1993) and Cummins (1997) suggest that all domains of QoL should be weighted by the level of importance they have to the individuals. The Capability Approach also supports this, advocating that well-being/QoL measures require “reasoned ‘consensus’ on weights or on a range of weights” (Alkire, 2015: 13). For a discussion of different methods to define weights across dimensions and indicators of well-being/QoL quantitative measures, please refer to Decancq and Lugo (2012) and Decancq & Neumann (2014).

58

J. de São José

In relation to the last two questions regarding the dimensions of QoL (What is the standard for a “good” quality of life? Who sets this standard?), there is a dominant line of thought in the literature in QoL in old age (and QoL in general) pointing out to the inexistence of a universal “gold standard,” because the understandings of what constitutes a life of quality varies from culture to culture. Even though QoL may result from the level of discrepancy between people’s expectations and needs and the satisfaction of them (Galloway, 2006), both expectations/needs and acceptable levels of meeting these expectations/needs are culturally shaped. This refers to the important issue mentioned earlier of studying/assessing QoL in the context of specific cultural systems. As emphasized by Sayer (2011), different cultures offer different forms of flourishing and suffering.

2.2 Theories and Conceptual Models of QoL in Old Age One of the first models of QoL in old age was proposed by Lawton (1983) in the field of Gerontology, “The quadripartite model of the good life.” In his early work he did not explicitly refer to QoL, but rather to the term “good life,” which he defined as follows: “The good life is a grandiose construct, presuming to account for all of life. Indeed, the implication is that the good life (and its polar opposite, the bad life) subsumes all that we define as legitimate personal and social goals. Its sectors together include every aspect of behavior, environment, and experience” (p. 349). From his viewpoint, good life includes four elements: (1) behavioral competence; (2) psychological well-being; (3) perceived quality of life; and (4) objective environment. Lawton defines behavioral competence as “the theoretical upper limit of capacity of the individual to function in the areas of biological health, sensation and perception, motor behavior, and cognition” (Lawton, 1982a: 38 in Lawton, 1983: 350). In turn, psychological well-being “is one’s subjective evaluation of the overall quality of one’s inner experience.” (Lawton, 1983: 350). The third element, perceived quality of life, is defined as “the set of evaluations that a person makes about each major domain of his or her life.” (Lawton, 1983: 52). Finally, the objective environment “lies outside the individual and is capable of being counted or rated consensually by observers other than the subjects, or measured in centimeters, grams, and seconds.” (Lawton, 1983: 352). Lawton (1983) offers empirical evidence of the relationships between these elements, but he suggests that an improvement in one element does not always lead to improvements in other elements. This quadripartite model includes three inner elements (behavioral competence, psychological well-being and perceived quality of life) and one outer element (objective environment). Two of these elements are objective (behavioral competence and objective environment) and the other two are subjective (psychological wellbeing and perceived quality of life). It also considers the interrelationships between elements. Nevertheless, it does not distinguish explicitly determinants of QoL from outcomes of QoL.

Researching Quality of Life in Old Age…

59

More recently, conceptual models of QoL in old age have been developed in the field of Gerontological Nursing. Sarvimärki & Stenbock-Hult (2000) proposed the model “Quality of life in old age described as a sense of well-being, meaning, and value.” This model emphasizes the need of distinguishing the conditions that contribute to QoL from the aspects that constitute QoL. It is argued that conditions such as personal health and economic resources are important determinants of QoL, but they are not part of the QoL. These authors distinguish conditions of intraindividual nature (e.g., health and functional capacity) from conditions of external nature (e.g., environment, housing conditions). The constituents of QoL identified by these authors are: “a sense of well-being, of meaning, and of value or self-worth” (p. 1027). The sense of well-being refers to the hedonistic conception of well-being, that is, to the maximization of pleasure and the minimization of suffering. In this vein, a life of quality implies a life where pleasure, satisfaction, and joy compensate for pain and suffering. In turn, the sense of meaning complements the sense of wellbeing and refers to the eudamonic conception of well-being, including the notions of life with meaning and life with purpose. Lastly, the sense of value or self-worth “has to do with the experience of oneself as a person of value or as a person involved in worthwhile activities” (p. 1027). According to these authors, QoL is the outcome from the interaction between external conditions and intra-individual conditions. Contrarily to Lawton’s model (1983), this model distinguishes explicitly determinants of QoL from outcomes of QoL. However, although it also distinguishes between inner and outer aspects and presupposes the interrelation between external conditions and intra-individual conditions, this model conceptualizes QoL exclusively as a subjective phenomenon. Also in the field of Gerontological Nursing, Register and Herman (2006) developed “The Register Theory of Generative Quality of Life for the Elderly (GQOLE).” This theory, classified by the authors as a middle range theory, defines QoL as a “generative process that remains active throughout life without temporal or spatial constraints.” (Register & Herman, 2006: 340). More specifically, QoL is a cumulative process, which is generated “through an ongoing series of specific connections and disconnections that result from interactions with the forces and processes people encounter in their daily life” (pp. 340–341). It is clarified that QoL is a dynamic personal perception, which is “enhanced by positive life connections and diminished by disconnections” (p. 341). In this vein, QoL is generated as older people experience connection with six interrelated forces or processes, which “involves the act of being” (p. 344): (1) metaphysically connected (self-esteem, self-determination, cognition, purpose, optimism, satisfaction with life); (2) spiritually connected (prayer, worship, fellowship, meaning); (3) biologically connected (functional capacity, physical comfort, health promotion, health maintenance); (4) connected to others (social support, interpersonal dynamics, cultural dynamics); (5) environmentally connected (socio-economic status, transport, assistance instruments, security, aesthetics); and (6) connected to society (personal social system and global social system). This theory does not distinguish explicitly determinants of QoL from outcomes of QoL. Although taking into account objective and subjective “forces” or “processes,”

60

J. de São José

it clarifies that QoL is a “personal perception.” The strengths of this theory are the emphasis on the dynamic nature of QoL (a cumulative process) and the inclusion of connections with others as one of the important “forces” or “processes.” Lastly, in the field of sociology, Higgs and collaborators (2003) proposed a “needs satisfaction model” of QoL in old age. These authors, inspired mainly by the Theory of Human Need of Doyal and Gough (1991) and by the work of Giddens on modernity (1991), adapted a “needs satisfaction” approach to support a new measure of QoL in early old age (the CASP-19). They emphasize that a conceptualization of QoL in old age must not lose sight of the post-materialist culture and the reflexive modernization process. In this vein, they argue that “the exercise of autonomy and choice is an intrinsic aspect of being human” (p. 244), emphasizing also “the crucial role that lifestyle and self-actualization have in articulating reflexivity and the narrative of self” (p. 244). Based on these basic assumptions, Higgs and collaborators (2003) defined four domains of need: (1) control; (2) autonomy; (3) pleasure; and (4) selfrealization. Control refers roughly to having control in life whereas autonomy refers to having freedom of choice. Pleasure and self-realization have to do with “the reflexive process of self-realization through activities that make them (older people) happy” (p. 245). These authors claimed that QoL should be measured “as the degree that human needs are satisfied” (p. 243). According to this conceptualization, the four domains of need correspond to the outcomes of QoL, which are different from the factors that determine it. Higgs and collaborators (2003) underline that there are strong correlations between these domains of need, confirming their belief that “quality of life is a holistic, unitary phenomenon” (p. 248). On the one hand, this model conceptualizes QoL in old age as constituted by domains of need. However, in the field of intellectual disabilities, Goode (1994) suggests that QoL not only refers to the satisfaction of needs but also to the “opportunities” to achieve certain life goals, which resonates with the Capability Approach. On the other hand, this model conceptualizes QoL as constituted by domains of subjective nature. This is in line with the claim made by Veenhoven (2000), who argues that the best and easiest way to measure/evaluate QoL is through subjective appreciations of life, but it goes against the general trend in the field of QoL in old age of combining subjective dimensions with objective dimensions. To conclude, it is important to underline that although there are several definitions and some theoretical/conceptual models of QoL in old age, several literature reviews of empirical research on this field reveal that only a minority of studies provides a definition or a theoretical/conceptual model of QoL in old age (e.g., Halvorsrud & Kalfoss, 2007; Vanleerberghe et al., 2017). This shows that this empirical research tends to be a theoretical and under-conceptualized, compromising its validity (Higgs et al., 2003).

Researching Quality of Life in Old Age…

61

2.3 Lay Views on QoL in Old Age The literature on the lay conceptualizations of QoL in old age (definitions expressed by older people themselves) has increased over the past few decades. One of the indicators of this increase is the existence of several reviews of this literature. For example, Brown and Flynn (2004) conducted a systematic review of the literature on lay views of QoL in old age, published between 1982 and 2004 (42 papers), having identified the following components of QoL: (1) family and other relationships/contact with others; (2) emotional well-being; (3) religion/spirituality; (4) independence/mobility/autonomy; (5) social/leisure activities; (6) finances/standard of living; (7) own health; and (8) health of others. A more recent systematic review of the qualitative literature on the older people’s views of QoL (non-institutionalized older people), which included 48 studies published up to November 28, 2018, aggregated the domains of QoL in the following categories: (1) autonomy (being able to manage on our own, retaining dignity and not feeling like a burden); (2) role and activity (spending time doing activities that bring a sense of value, joy and involvement); (3) health perception (feeling healthy and not limited by physical condition); (4) relationships (having close relationships which makes us feel supported and enable us to mean something for others; (5) attitude and adaptation (looking on the bright side of life); (6) emotional comfort (feeling at peace); (7) spirituality (feeling attached to and experiencing faith and selfdevelopment from beliefs, rituals and inner reflection); (8) home and neighborhood (feeling secure at home and living in a pleasant and accessible neighborhood), and (9) financial security (not feeling restricted by the financial situation) (van Leeuwen et al., 2019). As we can see, the domains identified by these two systematic reviews are similar. These lay domains of QoL in old age have also some similarities with those proposed by academics (Brown et al., 2004; Fernández-Ballesteros, 2011). Among the studies included in this last literature review, there are two that stand out for having gone beyond the mere identification of QoL domains. The first study was conducted by Hendry and McVittie (2004), based on semi-structured interviews with non-institutionalized older people. This study shows that the older people’s accounts of their experiences related with different aspects of QoL were made in an interlinked way, that is, they spoke about several domains of QoL and the interconnections between them. According to these authors, this demonstrates that the different aspects/domains of older people’s lives are “inextricably linked,” and that older people “do not segment their lives into component parts” (p. 971). In addition, they also found that “quality of life was regarded as relative to the experiences of other people” (p. 967) in two ways: (1) the older people’s QoL was dependent on the lives of significant others, mainly family members; and (2) the older people’s QoL was dependent on comparisons between their own lives and the lives of other people, particularly the lives of other older people. Hendry and McVittie (2004) also found that several domains of older people’s lives were assessed by themselves in an ambivalent way (assessed positively in certain things and negatively in other things).

62

J. de São José

Lastly, they concluded that QoL “can be seen to be a highly subjective phenomenon” (p. 970) in two ways: (1) QoL is a dynamic project/process under regular monitoring and adjustment; and (2) QoL is an inner experience that may or may not be communicated to others. Both ways indicate that QoL is manageable. Hendry and McVittie think that all these issues are not well captured by conventional quantitative measures of QoL. The importance of the QoL of other people for our own QoL is also stressed by Sayer (2011). This author argues that “people’s relation to the world is one of concern” (p. 1). This means that all human beings care about things and others that things and others matter to all of us. This is so because, according to Sayer, we are social beings (dependent on others and involved in social practices), sentient beings (we feel things and we can flourish or suffer), and evaluative beings (we evaluate things and people). Sayer goes further suggesting that our QoL or well-being depends on the evaluations we make about those things and people that matter to us: “Their lives (people’s lives) can go well or badly, and their sense of well-being depends at least in part on how these other things that they care about—significant others, practices, objects, political causes—are faring, and how others are treating them” (p. 1). Nevertheless, Sayer asserts that social sciences fail to recognize this and, therefore, fail to recognize “what is most important to people” (p. 2). Goode (1994) also suggests that “The QoL of an individual is intrinsically related to the QoL of other persons in his or her environment” (p. 148). The second study was carried out by Grewal and collaborators (2006), a qualitative study (using in-depth interviews) with non-institutionalized older people. This study identified six factors that, according to the interviewees, add quality to their lives: (1) activities/doing something; (2) home/surroundings; (3) family and other relationships; (4) health; (5) standard of living/wealth; and (6) faith/religion/spirituality. The respondents also identified the factors that detract from quality in their lives: (1) bereavement; (2) the provision of informal care; (3) poor health; (4) poor finances; and (5) poor surroundings. Finally, this study identified the attributes of QoL: (1) attachment (feelings of love, friendship, affection and companionship); (2) role (having a purpose in life, being useful); (3) enjoyment (feelings of satisfaction and joy); (4) security (emotional, financial and health security); and (5) control (being able to make decisions autonomously). Grewal and collaborators (2006) underline that these attributes are similar to the domains of need identified by Higgs and collaborators (2003). Inspired by the Capability Approach, they conceptualize these attributes as functionings, emphasizing however that “it is the capability of older people to achieve these functionings that appears to be of greatest importance, rather than necessarily, achievement of specific functions” (p. 1899). This shows that what is most important for older people from the point of view of QoL is not so much the achievement of certain valued functionings but the real opportunities they have to achieve them. Breheny and collaborators (2016) reached the same conclusion. Goode (1994) also stress the importance of opportunities with respect to QoL. Laslty, Edgerton (1996: 88 in Rapley, 2008: 222) claims that “We should continue every effort to ensure that

Researching Quality of Life in Old Age…

63

persons … have access to better housing, health care, recreational activities, dignified employment, and everything else that an enlightened society can provide for its citizens. But we must never forget that all a society should do is provide options….” This line of research on the older people’s views of QoL is important to get a closer picture of the aspects that really matter for them and the importance of these aspects to QoL.

2.4 Neglected Aspects in the Conceptualization of QoL in Old Age Despite the efforts that have been made to conceptualize QoL in old age, there are some conceptual aspects that have been overlooked. In addition to the lack of a clear distinction of the nature of the dimensions of QoL, the under-exploitation of the interrelationships between these, the non-consideration of the importance of the lives of others for our own QoL and the opportunities to achieve a life of quality (capability to achieve QoL), there are other aspects that have been neglected in the conceptualization of QoL in old age. One of the most striking neglected aspects of the conceptualization of QoL in old age is its experiential nature. Even qualitative studies that focus on the older people’s views of QoL in old age have under-explored this experiential aspect. QoL is not only evaluated but also, and above all, lived. QoL is not only in the older people’s minds, bodies and their external environments, but also in their real lives, that is, in their daily actions and interactions in particular contexts. In fact, a life of quality only exists as long as it is lived. As Sayer (2011) emphasizes, QoL refers to “objective states of being which people strive to discover, achieve or create” (p. 134). Moreover, it is important to recognize that evaluations, feelings (and actions) are “situational,” meaning that they are highly dependent on social situations. As Jerolmack and Khan (2014: 186) suggest, “what we say and what we do are strongly influenced by situational factors.” Therefore, it is also important to access, directly, the older people’s everyday lives as they unfold, their lived experiences. This is highlighted by Stiglitz and collaborators (2009: 12): “Such a system (of measuring QoL) should not just measure average levels of well-being within a given community, and how they change over time, but also document the diversity of peoples’ experiences and the linkages across various dimensions of people’s life.” (Stiglitz et al., 2009: 12). In this regard, Gubrium and Lynott, (1983: 34) underlined that “The ethnographic data show that for old people (though not implying only old people) the quality of life is articulated through the ongoing experiences of those who live it.” A recent ethnographic study on gardening and well-being/QoL in old age (Robbins & Seibel, 2019) revealed the importance that social relationships and different temporalities (of gardening activity, aging experiences, etc.) have for the well-being of older people, something that, according to the authors, it could hardly be captured through interviews or questionnaires. Hence, QoL of older people should not be conceptualized

64

J. de São José

merely as an evaluative exercise, displacing the everyday life through which life unfolds. The second neglected aspect is the dynamic nature of QoL (Schalock & Verdugo, 2002), which is explicitly recognized in the literature on QoL in general, but only implicitly recognized in the literature in QoL in old age. For example, QoL can change according to the stage of the life course we are in (Schalock & Verdugo, 2002). The cumulative effects of multiple (dis)advantages are another issue related with the dynamic and procedural nature of QoL (Stiglitz et al., 2009). Walker and Mollenkopf (2007: 8–9) also underline the role of the life course: “QoL in old age is the outcome of the interactive combination of life course factors and immediate situational ones.” This shows that QoL may change over time and is a cumulative process (Register & Herman, 2006). The third neglected aspect has to do with the uneven nature of QoL. Sayer (2011) emphasizes that well-being/QoL is “unevenly achieved” (p. 137), calling our attention to the importance of social inequalities and its impact on QoL. Carr (2019) also stresses that social inequalities undermine QoL of older people. In turn, the report authored by Stiglitz and collaborators (2009) recommends that the indicators of well-being/QoL should allow measuring inequalities between people, socioeconomic groups, gender, and generations. Nevertheless, with very few exceptions (e.g., Laditka et al., 2009; Weidekamp-Maicher & Naegele, 2007), there has been little concern for examining the extent to which QoL in old age has been marked by social inequalities (e.g., gender and other inequalities). The last neglected aspect is the multilevel nature of QoL. On the one hand, QoL in old age can be assessed at different levels of reality, namely macro (e.g., nations), meso (e.g., local communities, organizations), and micro (e.g., individuals) levels (Brown et al., 2004; Fernández-Ballesteros, 2011; Schalock & Verdugo, 2002; Walker & Mollenkopf, 2007). On the other hand, QoL at one level is shaped by QoL at other levels, although in general terms, the influence occurs from the macro level toward the meso and micro levels (Turner, 2005). Taking into account the multiple levels of reality also means recognizing that, at each level, there are social forces that impose limits and opportunities to achieve a life of quality. Schalock and Verdugo (2002) suggest that the individuals’ QoL is affected by the forces operating at these three levels of reality. As Gabriel and Bowling (2004: 689) point out, “Greater recognition is needed in quality of life research that the influential domains and variables are not only people’s own personal characteristics and circumstances, but also that there is a dynamic interplay between people and the surrounding social structures of a changing society.”

2.5 Conceptual Principles for Studying QoL in Old Age Schalock and Verdugo (2002) argues that agreeing on some conceptual and methodological principles regarding QoL is more important than trying to reach a consensual definition of QoL. In this vein, Van Hecke and collaborators (2018) derived several

Researching Quality of Life in Old Age…

65

conceptual and methodological principles from the QoL framework of Schalock and Verdugo (2002). A modified version of the conceptual principles proposed by Van Hecke et al. (2018) is provided below (later an adapted set of methodological principles are also provided). These principles are interdisciplinary in nature, as they are drawn from the previous discussion that has an interdisciplinary tone. As Walker and Mollenkopf (2007:3) advocate: “QoL should be approached from an interdisciplinary perspective, given that practically all aspects of life, which are of a different nature, can be important for its quality.” Hence, based on the previous discussion, the conceptual principles here proposed for studying QoL in old age are as follows: (1) (2) (3) (4) (5) (6)

(7) (8) (9) (10) (11) (12) (13) (14)

QoL in old age is constituted by multiple and interlinked dimensions. Some dimensions of QoL in old age are constituents of QoL while others are determinants of QoL. Some dimensions of QoL in old age are intrinsic to older people while others are extrinsic to them. Some dimensions of QoL in old age are objective while others are subjective. The dimensions of QoL in old age can be differently valued by older people. QoL in old age may depend not only on the achievement of valued functionings (valued doings and beings) but also on the capability to achieve these functionings. QoL in old age is a construct made up of lay and expert views. QoL in old age is imbued with culture. QoL in old age is dependent on the lives of others, the interaction with others and social structures at different levels of reality. QoL in old age is a cumulative process, it is shaped by what happened in the older people´s past lives. QoL in old age is a lived experience. QoL in old age is dynamic, it can change over time. QoL in old age is marked by social inequalities. QoL in old age is a multilevel phenomenon, as it can be found at macro, meso, and micro levels of reality.

3 How to Measure/Assess QoL in Old Age As we have seen, there is no consensus regarding the conceptualization of QoL in old age. The same applies to methodological aspects, as there is no widely agreement on how to measure/assess it (Brown et al., 2004; Walker & Mollenkopf, 2007). The major limitations in measuring/assessing QoL in old age are as follows: (1) predominance of HRQoL measures, quantitative measures centered mainly on the health dimension (Halvorsrud & Kalfoss, 2007; Higgs et al., 2003: 239; Vanleerberghe et al., 2017); (2) the existence of few QoL measurement instruments focused specifically on the older persons (Halvorsrud & Kalfoss, 2007; Vanleerberghe et al., 2017); and (3) predominance of self-report measures, mainly based on scales of QoL, which have

66

J. de São José

not only a subjectivist bias, as they are based on respondents’ perceptions (FernándezBallesteros, 2011), but also an individualistic bias, given that they neglect the social dimensions of QoL (Gubrium & Lynott, 1983; Stiglitz et al., 2009).

3.1 Methodological Principles for Studying QoL in Old Age The set of conceptual principles proposed in the previous section have methodological implications. Table 2 presents the methodological principles corresponding to each conceptual principle, taking into account the set of methodological principles that Van Hecke and collaborators (2018) derived from the QoL framework of Schalock and Verdugo (2002). The first methodological principle (To explore various dimensions of QoL and the interrelationships between them) is widely accepted in the literature on QoL in old age and QoL in general, requiring no further discussion. With respect to the second methodological principle (To include dimensions of different nature, while maintaining their distinctions), the first part (to include dimensions of different nature) is also widely accepted, but the same cannot be said in relation to the second part (while maintaining their distinctions). For example, while some quantitative instruments aggregate in a single index indicators of outcomes of QoL with indicators of determinants of QoL (e.g., the WHOQOL-OLD), there are academics that reject this (e.g., Veenhoven, 2000, 2014), arguing that it does not make sense to put in the same index pears and apples and that, in doing so, it is not possible to explore the interconnections between inner and outer dimensions and between life chances and life results (Veenhoven, 2000). The third methodological principle (To incorporate the older people’s views) has been increasingly accepted in the literature on QoL in old age. This is reflected in the consultation of older people that took place in the development of the most used quantitative instruments to measure QoL in old age (e.g., the WHOQOL-OLD and the CASP-19). The older people´s views have also been taken into account in qualitative studies, namely in relation to the identification and valuation of different dimensions of QoL in old age. However, capturing the perspectives of older people may not be possible in some cases (e.g., when the older persons suffer from severe dementia). In these cases, it will be necessary to use proxy-reports, which can be provided by other persons, such as family members (Claes et al., 2012). Regarding the fourth methodological principle (To be attentive to the social and cultural embeddedness of QoL at different levels of reality), it is important to underline three aspects. The first aspect refers to the importance of looking at older people in their social contexts. In the 1980s, Gubrium and Lynott (1983: 37) called attention to some of the problems with QoL measurement instruments, namely its individualistic character, which neglected the “social features of daily living” that are relevant for QoL in old age. They argued that these instruments produced an account of individuals as if they were displaced from their social interactions and contexts. This is still valid in relation to the existing measures of QoL in old age. The second aspect

Researching Quality of Life in Old Age…

67

Table 2 Conceptual and methodological principles for studying QoL in old age Conceptual principles

Methodological principles

(1) QoL in old age is constituted by multiple and interlinked dimensions

(1) To explore various dimensions of QoL and the interrelationships between them

(2) Some dimensions of QoL in old age are constituents of QoL while others are determinants of QoL

(2) To include dimensions of different nature, while maintaining their distinctions

(3) Some dimensions of QoL in old age are intrinsic to older people while others are extrinsic to them (4) Some dimensions of QoL in old age are objective while others are subjective (5) The dimensions of QoL in old age can be differently valued by older people

(3) To incorporate the older people’s views

(7) QoL in old age is a construct made up of lay and expert views (8) QoL in old age is imbued with culture (9) QoL in old age is dependent on the lives of others, the interaction with others and social structures at different levels of reality

(4) To be attentive to the social and cultural embeddedness of QoL at different levels of reality

(13) QoL in old age is marked by social inequalities (14) QoL in old age is a multilevel phenomenon, as it can be found at macro, meso and micro levels of reality (6) QoL in old age may depend not only on the (5) To explore the role of older people’s achievement of valued functionings (valued capabilities doings and beings) but also on the capability to achieve these functionings (10) QoL in old age is a cumulative process, it is shaped by what happened in the older people´s past lives

(6) To adopt a diachronic/life course perspective

(12) QoL in old age is dynamic, it can change over time (11) QoL in old age is a lived experience

(7) To complement subjective evaluations and objective indicators of QoL with observed lived experiences

refers to the roles of social and cultural structures, at different levels of the reality, as sources of inequalities. It is well-known that “Race, gender, and SES [socioeconomic status] profoundly shape how a person ages because they influence the obstacles (or benefits) in childhood, adolescence, and adulthood that set the course for how he or she grows old” (Carr, 2019: 6). The third aspect refers to the role played by the cultural context in assessing QoL. While some authors argue for the possibility of reaching consensus around the dimensions of QoL between different

68

J. de São José

cultures (e.g., WHOQOL GROUP, 1995), others are skeptical about this possibility (e.g., Diener & Suh, 1997). At this respect, some academics (e.g., Cummins, 2005; Schalock & Verdugo, 2002) argue that the core domains of QoL are the same for all people, although their valuation by individuals may vary according to the cultural systems. The fifth methodological principle (To explore the role of older people’s capabilities) is strictly related with the previous one, since capabilities result not only from resources that are intrinsic to the individual (e.g., functional capacity), but also from the social and cultural constraints at different levels of the social reality. For example, the ability to lead a life of quality depends not only on individual resources, but also on collective resources, such as the existence of a good transport system and social facilities in the neighborhood. The sixth methodological principle (To adopt a diachronic/life course perspective) is justified by the changing and procedural nature of QoL. The life course perspective presupposes the multidimensionality of life, the interconnection between our lives and the others’ lives and the embeddedness of our lives in the social and cultural structures. Hence, the life course perspective relates with several of the previous methodological principles. If these perspectives, especially the cumulative inequality perspectives, are useful to understand disparities in later life physical health and economic security (Carr, 2019), they will also be useful to understand QoL in old age. The last methodological principle (To complement subjective evaluations and objective indicators of QoL with observed lived experiences) aims to counterbalance the dominance of “account-based approaches” (Jerolmack & Khan, 2014) in the field of QoL in general and QoL in old age in particular. The “account-based approaches” are those that collect individuals’ accounts, through interviews or surveys, about different research topics. They are based on what research participants say. However, what people say does not always correspond to what they actually do, that is, verbal responses do not always match up with behavior/action (Jerolmack & Khan, 2014). Echoing Jerolmack and Khan (2014: 178), it is argued that “Because meaning and action are collectively negotiated and context-dependent, we contend that self-reports of attitudes and behaviors are of limited value in explaining what people actually do because they are overly individualistic and abstracted from lived experience.” In this vein, it is proposed that an account-based approach to study/assess QoL in old age should be complemented by a direct observation of the daily lives of older people. In other words, it is proposed that verbal responses (what they say) need to be situated in relation to observed everyday experiences (what they do). This requires an ethnographic approach. The combination of different methodological strategies may contribute to methodological sophistication, something that, according to FernándezBallesteros (2011), is needed in research on QoL in old age.

Researching Quality of Life in Old Age…

69

4 Concluding Remarks The research on QoL in old age is voluminous, but we do not know very well what we talk about when we talk about QoL in old age, because there is no universal agreement about its conceptual properties. We also do not know very well how we can study/assess QoL in old age, being, in great part, a reflection of the conceptual uncertainty. These conceptual and methodological uncertainties are obviously not beneficial for the maturation of this field of research. In order to overcome these uncertainties, this chapter presents a set of conceptual and methodological principles based on a review of relevant literature. These are interdisciplinary principles that are consistent with the interdisciplinary nature of QoL in old age. These principles are expected to constitute an effective contribution to the conceptual and methodological consolidation of the research field on QoL in old age.

References Alkire, S. (2015). Capability approach and well-being measurement for public policy. OPHI working paper 94. Oxford University. Bowling, A., & Stenner, P. (2011). Which measure of quality of life performs best in older age? A comparison of the OPQOL, CASP-19 and WHOQOL-OLD. Journal of Epidemiology and Community Health, 65(3), 273–280. https://doi.org/10.1136/jech.2009.087668 Bowling, A., Hankins, M., Windle, G., Bilotta, C., & Grant, R. (2013). A short measure of quality of life in older age: The performance of the brief Older People’s Quality of Life questionnaire (OPQOL-brief). Archives of Gerontology and Geriatrics, 56(1), 181–187. https://doi.org/ 10.1016/j.archger.2012.08.012 Breheny, M., Stephens, C., Henricksen, A., Stevenson, B., Carter, K., & Alpass, F. (2016). Measuring living standards of older people using Sen’s Capability Approach: Development and validation of the LSCAPE24 (Living Standards Capabilities for Elders) and LSCAPE-6. Ageing and Society, 36(02), 307–332. Brown, J., & Flynn, T. (2004). The components of quality of life nominated by older people: A systematic review of the literature. In: J. Brown, A. Bowling, T. Flynn (Eds.), Models of quality of life: A taxonomy, overview and systematic review of the literature (pp. 78–100). European Forum on Population Ageing Research/European Group on Quality of Life Extending quality of life in old age (EQUAL). Brown, J., Bowling, A., Flynn, T. (2004). Models of quality of life: a taxonomy, overview and systematic review of the literature. European Forum on Population Ageing Research/European Group on Quality of Life Extending quality of life in old age (EQUAL). Carr, D. (2019). Golden years? Social inequality in later life. Russell Sage Foundation. Claes, C., Vandevelde, S., Van Hove, G., Van Loon, J., Verschelden, G., & Schalock, R. L. (2012). Relationship between self-report and proxy ratings on assessed personal quality of life-related outcomes. Journal of Policy and Practice in Intellectual Disabilities, 9(3), 159–163. Cummins, R. A. (2005). Moving from the quality of life concept to a theory. Journal of Intellectual Disability Research: JIDR, 49(Pt 10), 699–706. https://doi.org/10.1111/j.1365-2788.2005.007 38.x Cummins, R. A. (1997). The comprehensive quality of life scale—Intellectual disability (5th ed.) (ComQoL-ID5): Manual. Deakin University School of Psychology. Decancq, K., & Lugo, M. A. (2012). Weights in multidimensional indices of well-being: An overview. Econometric Reviews, 32(1), 7–34.

70

J. de São José

Decancq, K., & Neumann, D. (2014). Does the choice of well-being measure matter empirically? An illustration with German data. In M. Adler, & M. Fleurbaey (Eds.), Oxford handbook of well-being and public policy. Diener, E., & Suh, E. (1997). Measuring quality of life: Economic, social and subjective indicators. Social Indicators Research, 40, 189–216. Felce, D., & Perry, J. (1993). Quality of life: A contribution to its definition and measurement. Mental Handicap in Wales—Applied Research Unit, University of Wales College of Medicine. Fernández-Ballesteros, R. (2011). Quality of life in old age: Problematic issues. Applied Research Quality Life, 6, 21–40. https://doi.org/10.1007/s11482-010-9110-x Fernández-Ballesteros, R., & Santacreu, M. (2014). Quality of life: Problematic issues. In A. C. Michalos (Ed.), Encyclopedia of quality of life and well-being research (pp. 5363–5371). Springer. Ferriss, A. L. (2004). The Quality of life concept in sociology. The American Sociologist, 35, 37–51. Gabriel, Z., & Bowling, A. (2004). Quality of life from the perspectives of older people. Ageing and Society, 24(5), 675–691. https://doi.org/10.1017/S0144686X03001582 Galloway, S. (2006). A literature review. In: Scottish Executive Social Research, Quality of Life and Well-being: Measuring the Benefits of Culture and Sport: Literature Review and Thinkpiece (pp. 4–97). Edinburgh: Scottish Executive Education Department. Goode, D. A. (1994). Quality of life for persons with disabilities: International perspectives and issues. Brookline Books. Grewal, I., Lewis, J., Flynn, T., Brown, J., Bond, J., & Coast, J. (2006). Developing attributes for a generic quality of life measure for older people: Preferences or capabilities? Social Science & Medicine, 62(8), 1891–1901. https://doi.org/10.1016/j.socscimed.2005.08.023 Gubrium, J. F., & Lynott, R. J. (1983). Rethinking life satisfaction. Human Organization, 42(1), 30–38. Halvorsrud, L., & Kalfoss, M. (2007). The conceptualization and measurement of quality of life in older adults: A review of empirical studies published during 1994–2006. European Journal of Ageing, 4(4), 229–246. https://doi.org/10.1007/s10433-007-0063-3 Hambleton, P., Keeling, S., & McKenzie, M. (2009). The jungle of quality of life: Mapping measures and meanings for elders. Australasian Journal on Ageing, 28(1), 3–6. https://doi.org/10.1111/j. 1741-6612.2008.00331.x Hendry, F., & McVittie, C. (2004). Is quality of life a healthy concept? Measuring and understanding life experiences of older people. Qualitative Health Research, 14(7), 961–975. Higgs, P., Hyde, M., Wiggins, R., & Blane, D. (2003). Researching quality of life in early old age: The importance of the sociological dimension. Social Policy & Administration, 37(3), 239–252. Hyde, M., Wiggins, R. D., Higgs, P., & Blane, D. B. (2003). A measure of quality of life in early old age: The theory, development and properties of a needs satisfaction model (CASP-19). Aging & Mental Health, 7(3), 186–194. Jerolmack, C., & Khan, S. (2014). Talk is cheap: Ethnography and the attitudinal fallacy. Sociological Methods & Research, 43(2), 178–209. https://doi.org/10.1177/0049124114523396 Laditka, S. B., Corwin, S. J., Laditka, J. N., Liu, R., Tseng, W., Wu, B., et al. (2009). Attitudes about aging well among a diverse group of older Americans: Implications for promoting cognitive health. The Gerontologist, 49(Suppl 1), S30–S39. https://doi.org/10.1093/geront/gnp084 Lawton, M. P. (1983). Environment and other determinants of well-being in older people. The Gerontologist, 23(4), 349–357. https://doi.org/10.1093/geront/23.4.349 Lawton, M. P. (1991). A multidimensional view of quality of life in frail elders. In J. E. Birren, J. E. Lubben, J. C. Rowe, & D. E. Deutchman (Eds.), Quality of life in the frail elderly (pp. 3–27). Academic Press. Power, M., Quinn, K., Schmidt, S., & WHOQOL-OLD Group. (2005). Development of the WHOQOL-old module. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation, 14(10), 2197–2214. https://doi.org/10.1007/s11 136-005-7380-9 Rapley, M. (2008). Quality of Life Research. A critical introduction. Sage.

Researching Quality of Life in Old Age…

71

Register, M. E., & Herman, J. (2006). A middle range theory for generative quality of life for the elderly. ANS. Advances in Nursing Science, 29(4), 340–350. https://doi.org/10.1097/00012272200610000-00007 Robbins, J. C., & Seibel, K. A. (2019). Temporal aspects of wellbeing in later life: Gardening among older African Americans in Detroit. Ageing and Society, 1–21. https://doi.org/10.1017/s01446 86x19000813 Robeyns, I. (2017). Wellbeing, freedom and social justice. The capability approach re-examined. Open Book Publishers. Rojas, M. (2014). Quality of life, conceptualization. In A. C. Michalos (Ed.), Encyclopedia of quality of life and well-being research (pp. 5360–5363). Springer. Sarvimaki, A., & Stenbock-Hult, B. (2000). Quality of life in old age described as a sense of well-being, meaning and value. Journal of Advanced Nursing, 32(4), 1025–1033. Sarvimäki, A., & Stenbock-Hult, B. (2000). Quality of life in old age described as a sense of wellbeing, meaning and value. Journal of Advanced Nursing, 32(4), 1025–1033. https://doi.org/10. 1046/j.1365-2648.2000.01568.x Sayer, A. (2011). Why things matter to people. Social science, values and ethical life. Cambridge University Press. Schalock, R. L., & Verdugo, M. A. (2002). Handbook on quality of life for human service practitioners. American Association on Mental Retardation. Sen, A. K. (1985). Commodities and capabilities. Elsevier. Stephens, C. (2017). From success to capability for healthy ageing: Shifting the lens to include all older people. Critical Public Health, 27(4), 490–498. https://doi.org/10.1080/09581596.2016. 1192583 Stewart, A. L., & Ware, J. E. (1992). Measuring functioning and well-being: The medical outcomes study approach. Duke University Press. Stiglitz, J., Sen, A., & Fitoussi, J. P. (2009). Report by the commission on the measurement of economic performance and social progress. Paris: Institut national de la statistique et des études économiques (INSEE). The WHOQOL Group. (1995). The World Health Organization quality of life assessment (WHOQOL): Position paper from the World Health Organization. Social Science and Medicine, 41(10), 1403–1409 Turner, J. H. (2005). A new approach for theoretically integrating micro and macro analysis. In C. Calhoun, C. Rojek, & B. Turner (Eds.), The Sage handbook of sociology (pp. 405–422). Sage. Van Hecke, N., Claes, C., Vanderplasschen, W., et al. (2018). Conceptualisation and measurement of quality of life based on Schalock and Verdugo’s model: A cross-disciplinary review of the literature. Social Indicators Research, 137, 335–351. https://doi.org/10.1007/s11205-017-1596-2 van Leeuwen, K. M., van Loon, M. S., van Nes, F. A., Bosmans, J. E., de Vet, H., Ket, J., Widdershoven, G., & Ostelo, R. (2019). What does quality of life mean to older adults? A thematic synthesis. PLoS ONE, 14(3), e0213263. https://doi.org/10.1371/journal.pone.0213263 Vanleerberghe, P., De Witte, N., Claes, C., Schalock, R. L., & Verté, D. (2017). The quality of life of older people aging in place: A literature review. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation, 26(11), 2899–2907. https://doi.org/10.1007/s11136-017-1651-0 Veenhoven, R. (2000). The four qualities of life. Journal of Happiness Studies, 1, 1–39. https://doi. org/10.1023/A:1010072010360 Veenhoven, R. (2014). Quality of life (QOL), an overview. In A. C. Michalos (Ed.), Encyclopedia of quality of life and well-being research (pp. 5265–5269). Springer. Verdugo, M. A., Schalock, R. L., Keith, K. D., & Stancliffe, R. J. (2005). Quality of life and its measurement: Important principles and guidelines. Journal of Intellectual Disability Research: JIDR, 49, 707–717. Walker, A. & Mollenkopf, H. (2007). International and multidisciplinary perspectives on quality of life in old age. In: H. Mollenkopf, & A. Walker (Eds.), Quality of life in old age. International and multi-disciplinary perspectives (pp. 3–13). Springer.

72

J. de São José

Weidekamp-Maicher, M., & Naegele, G. (2007). Economic resources and subjective well-being in old age. In H. Mollenkopf, & A. Walker (Eds.), Quality of life in old age. International and multi-disciplinary perspectives (pp.65–84). Springer.

Peripheral Retail Expansion: Social Implications and Spatial Inequalities the Case of the Île-de-France Region André Torre and Océane Peiffer-Smadja

1 Introduction Inequalities amongst territories raise concerns all over the world. There is increasing evidence that urban sprawl and spatial inequalities in terms of public facilities and/or other amenities provision lead to social inequalities and affect upward mobility (Ewing et al., 2016). As a matter of fact, the majority of the population living in and suburban areas of rural areas near to the metropoles does not fully benefit from the globalized, connected, dynamic world of the cities. There is evidence in France that, while inequalities in income per inhabitant between regions and inequalities between Départements have decreased since the 1960s, spatial dichotomy between metropolis and their suburbs has been growing continuously (Davezies, 2012). One of the major concerns for suburban and rural territories is the loss of vitality and viability they have experienced. Town centre activities, particularly retail, are a key component of vitality in rural communities as rural retail sales are often used as an indicator of attractiveness of rural communities. That is why declining retail sales is a growing concern for local policy makers (Artz & Stone, 2006). The arrival of large decentralized stores in these communities has led to an intensive competitive environment for small shops. Retail development has indeed known great changes in forms and localization since the 1970s. It has experienced decentralization and an ongoing size increase (Colla, 2003). Retail decentralization was first used by Berry to refer to American retail expansion outside the central business district. At that time, retail amenities A. Torre (B) University Paris-Saclay, INRAE, AgroParistech, 16, Rue Claude Bernard, 75231 Paris Cedex 05, France e-mail: [email protected] O. Peiffer-Smadja University Paris-Saclay, AgroParistech, 16, Rue Claude Bernard, 75231 Paris Cedex 05, France © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 E. Vaz (ed.), Geography of Happiness, Contributions to Regional Science, https://doi.org/10.1007/978-3-031-19871-7_5

73

74

A. Torre and O. Peiffer-Smadja

were still grouping in poles at the edge of the city and settling close to other economic activities. Later in 1986, Schiller described three waves of decentralization for retail in the UK. These were closely linked to the arrival of new retail patterns; supermarkets and hypermarkets in the 1970s involving food retailing only, then retail warehouses five to ten years later, and finally in the 1980s, retail parks and outlets, involving mainly clothing, comparison goods and some supporting services traditionally found in the town centres began to settle out of town (Schiller, 1988). Until recently, the focus on retail decentralization was much more about retail patterns and about how much these retail patterns were changing. Links between sprawl and retail patterns have been investigated by Dodds (2015). Retail expansion has been massive in the surroundings of large metropolis such as Paris. It has led to high retail densities even in some remote areas of the Paris region. The competition for small retail units in the town centre of these areas is heavy and it can lead to spatial and social inequalities. Since the 1970s, retail expansion in France and in other developed countries remains at a high level. Until 2015, France had the largest shopping centre market in Europe (Russia has broken France’s 43-year reign as Europe’s largest shopping centre market in 2015) and the highest numbers of hypermarkets and supermarkets in Europe. Pressure for land efficiency is particularly high in the Paris region, as it concentrates employment, population and great quality soils for agriculture as well. Moreover, it has experienced a massive retail sprawl, as it contains 42.9% of the total amount of French large stores (above 500 employees). This expansion has raised great concerns amongst central administrators, local elected personnel and citizen of this region. Nowadays, there is a tendency towards giant projects combining food retailing, clothing, comparison goods and attraction theme parks, such as Europa City planned on 80 hectares at 17 km from the centre of Paris, which includes an aquatic attraction theme park and a skiing installation. Spatial and social inequalities are deeply linked and the links between them have been part of the focus of regional inequality literature (Wei, 2015). In the Ile-deFrance region, discontent amongst populations living in the suburbs and in declining rural areas and a feeling of abandonment are often expressed in their vote (Giblin, 2012). In the last presidential elections, Paris intramuros was the place where extreme voters were the least numerous (less than 5% of the voters), but between 20 and 50 km away from the capital city, the extreme vote was the highest with between 20 and 40% (Alidières, 2012). Localization of the habitat is now for a majority of geographers one of the main predictors of political orientation, even more efficient than socio-economic distinctions (Lévy & Le Bras, 2012). This paper aimed to be part of the literature on regional inequality. Our focus is on retail development and how it can impact the social and spatial inequalities of the Île-de-France region. In Sect. 2, we give a literature of the environmental, social and economic impact of peripheral retail expansion. In Sect. 3, we introduce general comments on retail expansion in the Ile-de-France region and its local socioeconomic implications based on the financial analysts’ reports, regional and local retail studies and local newspapers. In Sect. 4, we use descriptive statistics to illustrate

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

75

retail sprawl and spatial inequalities in terms of retail provision on the 1975–2015 period in the region Ile-de-France. In Sect. 5, we statistically study the links between retail and social characteristics in the municipalities of the Ile-de-France region using the number of shops and retail units, the retail floorspace built and socioeconomic variables including medium yearly income, percentage of households with non-taxable income and unemployment rate. We finish with a few conclusions.

2 Literature Review: Retail Decentralization and Its Socio-economic Implications In this part, we provide a survey on the literature on social and economic implications of retail expansion and its consequences for spatial equality.

2.1 Employment and the Survival of Small Retail Units in the USA From the 1980s to the 2000s, several studies focused on the impact of out-of-town retail stores and the socio-economic implications of retail decentralization. In the USA, socio-economic implications of Walmart expansion have attracted a great deal of investigation. A scientific and political debate grew on the social and economic costs and benefits of the of the world’s biggest retail expansion. Most of research works conducted in the late 1980s or the early 1990s consisted in case study analyses, non-econometric analyses. From the 2000s, econometric analyses have started to tackle the issue. Our review of econometric studies is based on our own literature review, as well as Bonnano and Goetz’s (2012) and Basker’s reviews (2007). The results are summarized in Table 1. In the USA, we found twenty-six econometric studies published in peer-reviewed journals that were assessing the socio-economic implications of Walmart expansion on local and regional market structures. All papers deal with the impacts of the opening or the presence of one or several Walmart stores either at the county level or in a perimeter close to Walmart set for the need of the study. Four main impacts are addressed: sales of other stores; the number of other stores, employment and wages and benefits. All the eleven studies addressing the impacts of Walmart on existing stores sales concluded that Walmart stores create heavy competition with other existing stores and capture the market shares of their competitors (see Table 1). For example, Ailiwadi et al. (2010) proved that after Walmart entered the market, sales decreased for more than 65% of the mass merchandisers (median decrease of 40%), while smaller decreases in sales were observed in supermarkets (median decrease 17%) and drug stores (6%). In California, Chiou (2009) assessed that Walmart’s entry not only

2003–2010 San Francisco Bay Area

1990–2005 Mississippi’s 82 counties

Boarnet et al.

Artz and Stone (2006)

Cardiff-Hicks et al. 1996–2013 National retail companies

City of Dallas 1990–2003

Ingraham et al.

Difference-in-difference approach

Scenario-based econometric analysis

Linear regression models

Spatial econometric

1977–1999 1750 County-level econometric US counties with above 1500 jobs

County-level spatial econometric

Basker

Hicks and Wilburn 1988–2000 counties in West Virginia

30 stores in areas Linear regression model surrounding Dallas 1987–1994 (monthly)

Capps and Griffin

Methods

Period of study Zone of study

Investigators

+

+

Employment

Table 1 Socio-economic impacts of large decentralized stores in the USA: a literature review

−

+

0

Wages/Health benefitsa

−−

−−−

−−

Effects on existing stores’ salesb

(continued)

−−

+

Effects on small retail stores’ populationc

76 A. Torre and O. Peiffer-Smadja

1992–2000 All US counties

2001–2005 8 counties in Pennsylvania

1988–1997 2065 Three-stage competition model small and medium-sized counties

1988–2003 Maryland

All US states 2000

1977–1995

Iowa counties 1989–2003

Dube et al. (2007)

Hicks

Jia (2008)

Hicks

Sobel and Dean

Neumark et al.

Hicks (2009)

County-level econometric

Econometric function of the time and distance from the first Walmart

Linear regression model

County-level Linear regression models

State and county-level econometric

Hierarchical Bayesian econometric approach

November 1999-June 2001 small suburban towns

Singh et al.

Methods

Period of study Zone of study

Investigators

Table 1 (continued)

−−

−−

−−

Employment

−

++

0 except for new hires in the retail sector: +

−−

Wages/Health benefitsa

−−

Effects on existing stores’ salesb

0

0

(continued)

−−−

Effects on small retail stores’ populationc

Peripheral Retail Expansion: Social Implications and Spatial Inequalities … 77

Econometric study

central and southern California

1983–2004 counties in Florida

1990–2004 (quarterly) all counties with W

Seven Walmart Before-and-after-with-control-group entries, 90 stores analysis

Washington, D. C. area all big box retailers 1976–2005

1996–2000 Dallas–Fort Worth

2006,2007 and 2008

Chiou (2009)

Paruchuri et al.

Drewianka and Johnson

Ailiwadi et al. (2010)

Haltiwanger et al.

Cleary and Lopez (2011)

Davis et al. (2012)

Local econometric

Structural model

Linear regression models using spatial components

County level linear regressions

Negative binomial model + endogeneity control

Methods

Period of study Zone of study

Investigators

Table 1 (continued)

0

−−

++

Employment

0

Wages/Health benefitsa

−−

−

Mass merchandisers: − −− Supermarkets: -

−

Effects on existing stores’ salesb

(continued)

−−

−

0

−

Effects on small retail stores’ populationc

78 A. Torre and O. Peiffer-Smadja

Spatial econometric analysis

1989–2002 counties in Indiana

1994–2006 All USA sites close to Walmart

1991–2003 all retail unit births and deaths all US counties

Hicks et al.

Ellickson and Grieco

Ficano (2013)

−−−

Within 2-mile of Walmart:otherwise: 0

Employment

Wages/Health benefitsa

b Positive

means increase in wages/ negative means decrease in wages). means increase in existing stores’ sales/ negative means decrease in existing stores’ sales. c Positive means increase in the small stores’ number/negative means decrease in the small stores’ number.

a Positive

One county econometric study

1976–2008 122 Iowa towns and cities with less than 20,000 population

Artz and Stone

Methods

Period of study Zone of study

Investigators

Table 1 (continued)

Within 2-mile of Walmart:- Further away: 0

Effects on existing stores’ salesb

−−

No effect on independent stores, but on others: -

Rural towns: ++

Effects on small retail stores’ populationc

Peripheral Retail Expansion: Social Implications and Spatial Inequalities … 79

80

A. Torre and O. Peiffer-Smadja

affected market shares of stores more similar to the chain, such as Costco (−12%), Target (−10%) and Kmart (−8%), but also those of specialized stores (ranging from −16% to −6%). The impact on the number of stores nearby or at the county level is measured using either the entry and exit rates of retail stores, or the births and deaths of retail stores (see Table 1). Over thirteen studies we have found, three only proved Walmart to have a positive impact on the entry of other retail stores, arguing that Walmart created a friendly environment for new retail stores to settle nearby. Three studies found no impact and the remaining seven studies found a negative impact of Walmart on existing store population. Jia (2008) using a competition model on a sample of 2065 small- and medium-sized counties proved that “Walmart’s expansion alone explains 50–70% of net exit of small discount retailers between 1988 and 1997”. Ficano (2013) carried out a nationwide study including all US counties, all Walmart stores and looked at the impacts of the Walmart expansion at the county level on the retail unit births and deaths between 1991 and 2003. She concluded that after 15 months of a new Walmart store entry, between 4.4 and 14.2 existing retail units closed while at most 3.5 new retail units opened. Moreover, she showed that in rural communities, the impact of Walmart is stronger, as retail deaths went up from 7.6 to 12.2 after a Walmart store entry in the county. As far as employment is concerned, we found ten econometric studies addressing the impacts of Walmart on employment on the long run. Three studies found that a Walmart store entry caused a slight permanent increase in county-level employment. One study showed no impact (Davis et al., 2009); one demonstrated a slight decrease in employment within two miles of a new Walmart, but no significant impact further away. However, five showed that overall Walmart’s entry or presence caused heavy decrease in employment at the county level. For example, the presence of Walmart in a county resulted in a loss in countywide annual retail employment of between 248 and 408 workers in the Maryland state. In a nationwide study on the 1977–1995 period, Neumark (2008) showed that for each job created by Walmart, 1.4 jobs are estimated to be lost. To finish, it has to be noticed that the eight studies about wages and benefits lead to inconclusive results as three of them showed a positive impact, three others showed no impact and the three remaining studies, including two on nationwide data proved a negative impact (see Table 1 for references).

2.2 Socio-economic Implications of Retail Decentralization in the UK In the UK, there is also an ongoing debate about external costs and benefits of large store development. Wrigley et al. (2009) used linear regression models with spatial components to assess the impact of large decentralized stores on the entry and exit rates of eleven categories of small stores (butchers, delicatessen, etc.) located within

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

81

a 10 km circle of 303 supermarkets in England. They showed that in the prosperous regions of London and the South East, the entries into the small store sector have been accelerated. However, in the rest of the UK, the majority of the categories (7 out of 11) of small retail stores showed higher net exit rates in centres/high streets that have experienced supermarket openings. In Britain, the main concern for public authorities about retail decentralization was the impacts on viability and vitality of communities, particularly rural communities (Ravenscroft, 2000; Thomas et Bromley, 2002). Indeed, small rural towns retail stores failure is seen by communities as a loss of vitality and viability among the communities. The model of large decentralized stores has impacted the communities in various ways, in their way of living, in the choices they have to reach grocery stores. Moreover, it led to communities, particularly rural communities to rely more on the use of private cars, even for short journeys to the grocery stores (Ronse et al., 2015).

2.3 Retail Expansion, Social Capital and Poverty A few studies have been addressing the issue of the links between poverty rates and large decentralized stores. Schuetz et al. (2012) examined the relationship between neighbourhood income and retail density for several types of goods and services in 58 large U.S metropolitan areas. They regressed retail employment measures on residential income measures in three models using employment densities, employment growth rates and income changes over two time periods (1992–2000 and 2001– 2006). It allowed them to give insights on the evolution of retail patterns according to income changes. They proved that high-poverty neighbourhoods have a higher density of supermarket units, but lower employment density, smaller units and fewer chain supermarkets. Neighbourhoods that experienced income upgrading, relative to the metropolitan area, saw larger gains in retail employment, while high-poverty neighbourhoods in which poverty increases experience smaller employment gains (or larger losses). Goetz and Swaminatham (2004) assessed the impacts of Walmart’s presence at the county level on the poverty rate in 1987 and 1998. They found that poverty rate in counties with new or existing Walmart stores is 0.2 and 0.09 percentage points higher, respectively, than those in counties without Walmart. Fitzgerald and Wirtz (2008) found that poverty rate reductions between 1989 and 2004 in the federal district of Minneapolis were smaller in Walmart than non-Walmart counties. Metzer and Schuetz (2012) analysed how retail services vary across New York City neighbourhoods with regard to income and by racial composition. Lower income and minority neighbourhoods have fewer retail stores, smaller average units and a higher proportion of “unhealthy” restaurants. An interesting debate opposed two papers on the impacts of Walmart on social capital. Both studies used the same social capital indexes built on several variables, including county-level data, such as voter turnout, number of associations,

82

A. Torre and O. Peiffer-Smadja

census participation rate and individual-level data, such as number of times in the preceding year the respondent played a “social” sport (softball, volleyball, tennis, golf or bowling), played cards, attended a religious service, hosted or attended a dinner party, etc. Goetz and Rupasingha (2006) found that Walmart reduced social capital at the county-level. However, Carden et al. (2009) by replicating their results showed that they were inconsistent. They found Walmart entry is associated with a statistically significant increase in the number of associations and a slight decrease of the number of associations in rural areas. Walmart leads to a statistically significant reduction in voter turnout in all three models, although the magnitudes are modest (0.2–0.7%). Finally, there is an increasing number of studies about food deserts and their social consequences, as well about the effects of food stores accessibility on health (Beaulac et al., 2009). These studies are part of the literature on the links between spatial inequalities, with a focus on retail development and social inequalities. Our aim is now to provide some evidence on these linkages in the region Ile-de-France.

3 Retail Expansion, Controversies and Regulation in the Île-de-France Region The previous literature review focuses mostly on the economic implications of the arrival of the big box retailers on the market and part of it on the social impacts for the communities living next to large decentralized stores. In this part, we focus on the concerns about social, economic and environmental impacts coming from the administrative bodies, the residents, the local elected personnel and other institutional agencies related to retail development. Our aim is to provide some evidence on these linkages in the region Île-de-France, based on the study of local newspapers, institutional reports and local plans.

3.1 National Regulation: An Entry Barrier to Large Stores Since 1973 The first supermarket and hypermarket opened respectively in 1958 and 1963 in France. At that time, independent small retailers were represented nationally and raised voices against the competition brought by big box retail stores. In 1973, the central government took some restrictive measures towards retail development projects above 1000 m2 of selling space in cities with less than 40,000 inhabitants and above 1500 m2 in cities with more than 40,000 inhabitants. They made it compulsory for these stores to ask for an authorization of zoning boards before applying for a planning permission.

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

83

This entry barrier is still in place, even if it has been through a lot of criticism. The opponents to the regulation found that it was aimed at preventing competition and was an unjustified barrier for retail stores. The European commission put a lot of pressure on the French government to review this regulation as it went against the principles of a free trade. It succeeded in 2008 as the new regulation was much less restrictive towards retail development. It stipulated retail development could only be stopped if the boards found adverse effects on landscape and environment. Any economic criteria such as the impacts of the proposal on other retailers were no more to be considered. The proponents of a regulation have raised concerns about the composition of the regulatory boards, mainly composed of local elected personnel, whose interest was to see new stores settle on their territories to obtain short run tax benefits. Other measures were taken by the central government in order to protect small retail units. In 1989, a national fund was created using the profits coming from the taxation of large economic activities to preserve the mom and pop stores. Municipalities could ask for financial help to be able to resort to their pre-emption rights in order to avoid changing the use of the land.

3.2 Environmental Concerns and Regional Plans in the 2000s In 2003, a national regulation was put in place to encourage municipalities to group together in order to produce local plans at a more global scale than the municipal level. In 2004, the government required each of these plans to include a specific document for retail development, included in large-scale planning documents, named Schémas de Cohérence Territoriale (SCOT). In 2015, the Île-de-France region was covered by thirty-three of these documents. We analysed twelve produced in rural areas produced between 2008 and 2014. They all recommended to maintain, revitalize or develop the small retail units in the town centres, ten of them advise the municipalities to control, reduce or even block large store production. Five of them produced a set of criteria for large retail development that mostly involves limiting the development to the areas already urbanized. Four plans advise municipalities to produce perimeters to save the existing retail units in the town centres and use the national financial help to buy any retail unit that would go bankruptcy and prevent a change of use.

3.3 Local Attitudes Towards Retail Development Since devolution was put in place in 1982–1983, mayors have been in charge of land use planning: they produce local plans and grant or refuse the planning applications, in accordance with the larger SCOTs regulations. Moreover in 1993, local elected

84

A. Torre and O. Peiffer-Smadja

Fig. 1 Approval rates of hypermarkets and supermarkets in the French local zoning boards between 1989 and 2007

personnel was put in the great majority of the zoning boards, in charge of granting or refusing permission for the large retail development. For municipalities, particularly for rural ones, financial benefits from taxation, the prospect of creation of jobs and vitality for their communities were at stake. Moreover, in case of a refusal, there was a heavy risk for them that the large store would be accepted in a nearby municipality and that they would hence suffer from a loss of vitality in their own municipalities. So, retailers have benefited from competition between small municipalities (Colla, 2003). Consequently, in the years 1993–1997, the rate of approval for large stores increased drastically as shown in Fig. 1. In the 2000s, the main concern of rural municipalities laid in the bankruptcy of small-town centres retail units and the consequent loss of vitality of the communities. A few studies have been released with alarming figures concerning the loss of small retail units and the vacancy rates. Nationally, the vacancy rate in the town centres is between 8 to 9% and 12 to 14% in small rural cities (Senate, 2016). In 2014, one of the leading real estate consulting groups declared that retail nonprime markets (mainly rural) would « continue to experience high vacancy rates consequently raising questions about the future of retail wasteland» (Cushman et Wakefield, 2014). Mayors and Chambers for retail and industry have put into place numerous strategies to cope with the increasing loss of quality of life in rural communities, mostly the loss of infrastructures and proximity services We provide a list of these strategies in Table 2, which summarizes the concerns towards retail development we could find using local newspapers, institutional reports and local plans. Environmental issues include congestion, car dependency, land consumption, low-density development and harm caused to the character and appearance of the area. Social issues include

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

85

loss of vitality and viability of the rural communities, difficulties to manage the vacant spaces. And finally, economic issues include loss of jobs in local retail market and households’ expenses on oil consumption due to car dependency. Measures and actions are taken by the local municipalities include using compulsory purchase orders to avoid change of use of retail units in the town centres, allowing discounts for any good that would be bought in the town centres retail units. Several articles in national newspapers have mentioned an overproduction of retail stores relative to the expansion of retail business in Île-de-France (Les Echos, 2014; la Depeche, 2016; L’Obs, 2016; Le Monde, 2016). In 2015, a meeting at the French National Assembly was dedicated to the issue of loss of vitality and viability in rural areas. Large decentralized stores were accused of being “ever further away, ever bigger”. Several members of local elected personnel raised concerns about the expansion of large peripheral retail stores; notably they accused the retail stores of being the cause of poverty amongst local citizens (La Tribune, 2015; Les Echos, 2014).

4 Retail Decentralization and Spatial Inequalities: Statistical Study In this part, we aim at measuring and characterizing retail expansion and decentralization and providing some insights about consequences of this expansion on retail forms.

4.1 Data Description We use four databases on retail development in Île-de-France and one database on the social and economic characteristics of the Île-de-France municipalities from the French National Institute for Statistics (INSEE). We exclude the city of Paris from our analysis as the retail market within the core city is highly competitive and contains specific stores called Grands magasins. The first database, named Sitadel, contains the floorspace by development type (housing, office, retail, warehouse) built between 1975 and 2013 in each municipality. We use the retail floorspace built each year from 1975 to 2013 in the municipalities of the Île-de-France region. The second database is produced by the INSEE, and named Connaissance locale de l’Appareil productif. It is composed of 134,916 retail units in the Île-de-France region, and includes each retail unit in France.1 Per unit, it contains the number of employees in FTE, the opening date and the localization at the municipality level. 1

The access to this database is protected by the French National Committee for statistical secret.

86

A. Torre and O. Peiffer-Smadja

Table 2 Concerns towards retail development and actions taken by local government Concerns relative to large decentralized stores

Actions and measures taken

Environmental issues Congestion, traffic pollution, car dependency Low density development and land consumption Harm caused to the character and appearance of the area Social issues

Quality of life and viability of the town rural centres Loss of attractiveness for many rural communities Drug traffic around the vacant retail units, insecurity

Covering vacant shop windows Negotiating with retailers to settle in town centre Launching major development project Fund raising amongst the local population to save a local shop Buying the vacant retail units to avoid the change of use (Compulsory purchase orders)

Economic issues

Loss of jobs in local independent retailers Households expenses on oil consumption

Associations of local independent retailers (events, fidelity program…) Private consultants paid by the municipalities to advise local independent retailers on accountability and management matters Reconstruction of the road network in the town centres and modernization of the public spaces Discounts, lottery competition for any purchase in one of the local town centre shops

Environmental issues Congestion, traffic pollution, car dependency Low density development and land consumption Harm caused to the character and appearance of the area Social issues

Quality of life and viability of the town rural centres Loss of attractiveness for many rural communities Drug traffic around the vacant retail units, insecurity

Covering vacant shop windows Negotiating with retailers to settle in town centre Launching major development project Fund raising amongst the local population to save a local shop Buying the vacant retail units to avoid the change of use (Compulsory purchase orders) (continued)

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

87

Table 2 (continued)

Economic issues

Concerns relative to large decentralized stores

Actions and measures taken

Loss of jobs in local independent retailers Households expenses on oil consumption

Associations of local independent retailers (events, fidelity program…) Private consultants paid by the municipalities to advise local independent retailers on accountability and management matters Reconstruction of the road network in the town centres and modernization of the public spaces Discounts, lottery competition for any purchase in one of the local town centre shops

The third database we use is an inventory of all stores in each municipality in the Île-de-France region. It includes 25 categories: hypermarkets (above 2500 m2 of selling space), supermarkets (between 400 and 2499 m2 ), large building material shops (above 400 m2 ), medium-sized food retail shops (between 120 and 400 m2 ), general merchandise, bakeries, butchers, frozen food stores, fishmongers, clothing stores, home furnishing, furniture and equipment, sports stores, florists, bookshops, etc. Contrary to the other databases, it does not include all retail units, such as eating and drinking units, car-dealers, real estate agencies and other specialized units. The total number of stores is 36,042 in Île-de-France. Our fourth database is the inventory of all stores above 1000 m2 in Ile-de-France. It contains the postal address and the total selling space of each of these stores. This database is produced by the Atelier Parisien d’Urbanisme. The database we use on social and economic characteristics contains: the medium yearly income per households, the percentage of households with non-taxable income, the unemployment rate and the percentage of retired people between the total population at the municipality level. To distinguish urban, suburban and rural municipalities, we use the typology of the municipalities produced by INSEE. This typology has been elaborated using the evolution of population, housing and employment densities. The urban municipalities are located within a dense urban environment including an employment level above 10,000 jobs. The suburban municipalities have known a steep increase in housing units located around employment centres. The rural municipalities have a low residential density, low increase in population and are located further away from the employment centres. In the seven Départements of Île-de-France, according to this classification, there are 395 urban municipalities, 673 suburban municipalities and 212 rural municipalities. They respectively group 6,200,329 residents (64.2% of the population in the region), 2,953,986 residents (30.6% of the population) and

88

A. Torre and O. Peiffer-Smadja

503,566 residents (5.2% of the population), corresponding to a total of 14.8% of the whole population of France.

4.2 Retail Expansion Between 1975 and 2014: Equally Distributed? Between 1977 and 2012, more than 15 million sq.mt of retail floorspace were added in the Ile-de-France region, which corresponds to 20% of the current total of retail floorspace in France. In this part, we aim at studying how large decentralized stores expansion has been distributed and its implications for spatial equality. In Fig. 2, we observe that retail decentralization followed population decentralization between 1975 and 1999. Retail development took place where population had settled a few years earlier and remained closer to Paris than the variation of the population. Between 1975 and 1983, 71% of the retail floorspace built was located between 5 to 20 km from central Paris (including about 25% at 15 km from Paris and 32% at 20 km). From the middle of the 1980s, retail floorspace was more decentralized than in the 1983–1990 and 1991–1999 periods, about 22% of the retail floorspace was built at 30 km from Paris (to compare to 11% in the 1975–1982 period) and 15% at 35 km (to compare to 7% between 1975 and 1982). We notice that between 2000 and 2012, retail development has settled even further away from the centre of the metropolis than the population had settled. About 5% of the total floorspace built between 2000 and 2012 was located 70 km away from the centre of Paris, while only less than 3% of the total variation of the population at the same period settled that far from central Paris (Fig. 3). People were even more likely to take their cars to reach the retail facilities and this could explain why we found the

Fig. 2 Retail floorspace built and variation of the population according to the distance to central Paris between 1975 and 1999

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

89

Fig. 3 Retail floorspace built and variation of the population according to the distance to central Paris between 2000 and 2012

highest retail densities (number of stores and retail floorspace per 1000 residents) in the remotest areas of the Ile-de-France region.

4.3 Retail Facilities in 2014 4.3.1

Food Deserts in Ile-de-France?

In 2014, 14.13% of the Île-de-France population lived in one of the 487 municipalities with no shops at all, including 10.58% of the population having no food store. Respectively, 4,2% of large stores, such as hypermarkets, supermarkets, home, furniture and equipment and 4,0% of small independent food stores (bakeries, butchers, general merchandise and fishmongers) were located in rural municipalities. Moreover, about 12.3% of the rural municipalities grouped 86% of the total number of the 25 categories shops in the rural areas (Table 3). These numbers can be explained by the high rates of bankruptcy of small retail units in France and particularly in suburban and rural areas. In the food-retailing sector, the market shares of small retail units dropped from 66.7% in 1970 to 30.5% in 1996, hypermarkets’ market shares increased from 3.6% in 1970 to 36.8% in 1996. Finally, between 1966 and 1998, the numbers of independent stores such as bakeries, textile stores, convenience stores and butcher’s shops decrease by between 50 to 85% (Insee, 1998).

90

A. Torre and O. Peiffer-Smadja

Table 3 Proportion of the population with no shops or no food stores in its municipalities Urban

Suburban

Rural

Total

Population (% of urban population)

Population (% of suburban population)

Population (% of rural population

Municipalities (% of municipalities)

Population (% of population)

With no shops

176,269 (2.84%)

915,055 (30.96%)

273,249 (54.3%)

344 (26.9%)

1,364,873 (14.13%)

With no food stores

259,285 (4.18%)

1,465,843 (49.6%)

317,949 (63%) 487 (38%)

4.3.2

2,043,077 (10.58%)

Versus a Consumers’ Paradise

About 20,5% of the municipalities, gathering 74% of the total population in 2015, took in 90% of the retail floorspace built between 1975 and 2013. In Fig. 4, we observe that part of the retail growth has been concentrated in the Villes nouvelles. In the 1960s, land use planning was heavily centralized and the government took several measures in order to stop congestion in Paris, decentralize economic activities from the core city. The Villes nouvelles policy was part of these measures; they aimed at avoiding urban sprawl by concentrating the development. The first one was implemented in 1970 in the Ile-de-France region. Since then, a total of cinq Villes nouvelles were created: Evry, Cergy-Pontoise, Saint-Quentin-enYvelines, Marne-la-Vallée, Melun-Sénart, grouping 70 municipalities in total. Part of the policy was to give subsidies to any economic activity that would be launched in these cities. This policy has proved successful to concentrate economic activities in the Villes Nouvelles (Schearmur & Alvergne, 2003). As far as retail development is concerned, 16.6% of the total retail floorspace has been built between 1975 and 2013 in these areas. In 2013, they gathered 10% of the Île-de-France population, 9.4% of the total number of retail units, 14.4% of the total employment in retail sector and 16.9% of the total retail floorspace of stores above 1000 m2 . In Fig. 4, we can observe that some suburban and even rural municipalities, located in the remote part of the region, have welcomed a high level of retail floorspace, which is not in accordance with the distribution of the population. Retail densities in terms of number of retail units per 1000 residents in urban, suburban and rural municipalities have evolved from respectively 0.1, 0.1 and 0.13 stores per 1000 residents in 1975 to 14, 17.19 and 15.35 stores per 1000 residents. Rural municipalities have the highest densities in terms of number of stores per 1000 residents, but retail development does not seem to be equally distributed in these communities as a few rural municipalities concentrate all retail floorspace built between 1975 and 2013.

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

91

Villes Nouvelles

Retail floorspace in 1000 sq.mt [0,5] (5,17] (17,35] (35,59] (59,89] (89,129] (129,194] (194,274]

10 km

Fig. 4 Retail floorspace built between 1975 and 2013 in the region Ile-de-France

5 Retail Expansion and Socio-economic Characteristics of Municipalities In this part, we aim at studying the socio-economic implications of retail expansion and giving some insights about the links between social inequalities and retail development in the municipalities of the Île-de-France region.

5.1 Retail Employment and the 2008 Crisis: Unequal Consequences Given the Location of Stores Despite an ever-increasing number of retail stores openings, retail employment decreased in urban and rural areas between 2007 and 2013, as employment in retail experienced a 13.4% drop in rural areas and a 2.4% decrease in urban areas (Fig. 5). Suburban areas have not been affected as employment has increased by 1.9%. The 2008 crisis strongly hit rural areas. We found out that, amongst the 39 rural municipalities that have one or more retail units employing more 20 employees, 31 of them have lost on average 17% of their total employment in retail after the 2008 crisis. The municipalities that are characterized by high levels of large decentralized stores seem to be more vulnerable to economic crisis. In Fig. 6, we spatially define the link between the number of large stores in 2007 (defined as the number of retail units employing more than 200 persons) and

92

A. Torre and O. Peiffer-Smadja

Fig. 5 Retail employment in rural, suburban and urban municipalities in Île-de-France between 2003 and 2013

the evolution of the unemployment rate between 2007 and 2012 in suburban and rural areas only. We observed that it is more likely that there is an increase of the unemployment rate in areas in which the number of large stores is higher.

Fig. 6 Large retail stores in 2007 and the evolution of unemployment rate between 2007 and 2012 in the region Île-de-France

Peripheral Retail Expansion: Social Implications and Spatial Inequalities …

93

Fig. 7 Spatial representation: clusters of municipalities in 2014

5.2 Social Inequalities and Retail Development: The Clustering of Municipalities In order to study the links between socio-economic characteristics and retail development, we produce a multiple variable analysis and a hierarchical clustering analysis in the municipalities of the region. We use six quantitative variables at the municipal level: the number of shops, the number of retail units, the medium yearly income per households, the percentage of households with non-taxable income, the unemployment rate, the percentage of retired people and the typology of each municipality: urban, suburban or rural. We find that the two first dimensions of the multiple analysis account for 78.35% of inertia (dimension 1: 54.49% and dimension 2: 23.86%). The first axis shows that the number of shops or units is negatively correlated with the medium income and the percentage of retired people. It means that the higher the numbers of shops and units in a municipality are, the lower the medium income and the percentage of retired people are. On the contrary, the number of shops and units is positively correlated with the unemployment rate and the percentage of households with non-taxable income, meaning that the municipalities with a high number of retail units are the

94

A. Torre and O. Peiffer-Smadja

Table 4 Correlation between socio-economic variables and retail characteristics in the Île-deFrance municipalities Coefficients of correlation

Medium yearly income per householda

Percentage of households with non-taxable income

Unemployment rate

Number of shops

−0.1933728

0.3201983***

0.4388576***

Number of retail units

−0.2400724

0.3702274***

0.4901769***

a Signif.

codes: 0 ‘***’.

poorest. We used partial correlations testing with the Pearson method to check the correlations between number of shops or retail units and income, unemployment rate and percentage of households with non-taxable income correcting from the population. Tests were successful (p-value