Living amid energy efficiency and energy consciousness : an examination of the discrepancy between stated and revealed sustainable practice preferences of generation Y

(1)

Living amid energy efficiency and

energy consciousness

An examination of the discrepancy between stated and

revealed sustainable practice preferences of generation Y

(2)

(3)

3

Tiemen Frans Frederik Koch 10202471

tiemenkoch@gmail.com

June 2016

Master in Urban Geography (Human Geography) Graduate School of Social Sciences

Department of Human Geography, Urban Planning, and International Development Studies University of Amsterdam

Supervisors: Prof. Dr. D. Manting & Dr. W.R. Boterman Second assessor: Dr. W.P.C. van Gent

Number of words: ca. 32.000

(4)

(5)

5

Preface

Plenty of interesting and challenging topics have been examined during my period as a graduate student in Human Geography at the University of Amsterdam. These subjects stretch from living with difference in the Schilderswijk, one of the most segregated neighbourhoods in the Netherlands, to the impact of sea level rise on urban Miami. A completely different subject is addressed in this master thesis, especially when comparing to my bachelor’s thesis about neighbourhood participation in IJburg, Amsterdam. However, all these topics share one common aspect, the urban or spatial context.

The main topic of this master thesis is the sustainable practices of young adults. The connection between energy, society, and space has not yet been studied widely, but got my attention during my period as an exchange student at the University of Manchester. To exemplify, fuel or energy poverty, energy security, and how to enhance the energy efficiency of residential buildings are aspects that experience increased attention in studies in the United Kingdom. For case of the Netherlands, this topic is increasingly relevant because of the possibilities in residential buildings to reduce the energy usage that is necessary to achieve the sustainable goals by 2050. The focus on young adults is based on the assumption that they are the most sustainable conscious yet and are the ones to upgrade the housing stock.

I was privileged to write this final assignment in combination with an internship at the Netherlands Environmental Assessment Agency (PBL). During my stay I could not only get access to rich and relevant data, but also learn a lot about the organisation, the role and importance of such advanced institute, and in particular, the value of a human geographer in both research and society. I would like to thank my colleagues at PBL for always being willing to help and our many interesting conversations. Furthermore, I would like to thank my supervisors Dorien Manting and Willem Boterman for their support and feedback during the process of writing this master thesis. I have learnt a lot from our inspiring meetings and the structural feedback they have provided me. Finally, I would like to thank Wouter van Gent for being my second assessor.

Tiemen Koch

(6)

(7)

7

Summary

Generation Y, those roughly born between 1980 and 1995, is seen as the most sustainable generation yet. Nevertheless, it remains unclear whether this consciousness about the environment sticks with consuming fair trade and eco-friendly products, or if it also relates to enhancing the energy efficiency of the residential building by doing energy saving investments, e.g. solar panels, highly efficient water boiler. The expectation is that young adults have the intention to do such investments based on the norms and values of the postmodern society, but due to uncertainties related to their phase in life do not set these intentions to action. To test these assumptions, a combined data set has been analysed (i.e. WoON 2012 & Energy Module 2012) of which the findings are highlighted by qualitative and spatial research elements. The main conclusion of this master thesis is that young adults have more often the intention to do energy saving investments, but this age effect is not significant when it comes to having done such investments. In predicting the likelihood of having taken energy saving measures, the residential characteristics are of main importance. The tentative explanation of the findings is that an intention to change the residential location, which is believed young adults have more often, is not beneficial for doing long-term profitable energy saving investments.

Keywords: Energy Efficiency, Energy Saving Investments, Generation Y, Housing Pathways, Residential Buildings, Stated and Revealed Preferences, Sustainable Practices, Young Adults.

(8)

8

Samenvatting

Generatie Y wordt gezien als de meest duurzame generatie tot dusver. Het gaat hier om de mensen die ongeveer tussen 1980 en 1995 zijn geboren. Het is echter nog niet aangetoond of het veronderstelde milieubewustzijn zich alleen vertaald in het consumeren van duurzaam geproduceerde producten, of dat deze jongvolwassenen ook harde maatregelen nemen zoals het investeren in energiebesparende maatregelen in hun woonhuis. Dit zijn bijvoorbeeld investeringen in zonnepanelen, het verbeteren van de isolatie en het installeren van een hoogrendementsketel. De verwachting is dat de jongvolwassenen gebaseerd op de huidige waarden en normen omtrent duurzaam gedrag wel de intentie hebben om te investeren, maar door onzekerheden in hun huidige levensfase deze intentie niet waarmaken. Deze assumptie is getest door het kwantitatief analyseren van een gecombineerd databestand bestaande uit de WoON 2012 en Energie Module 2012 onderzoeken. De resultaten worden geïllustreerd met voorbeelden uit elementen van kwalitatief en ruimtelijk onderzoek. De hoofdconclusie is dat jongvolwassenen aanzienlijk vaker de intentie hebben, maar dat dit leeftijdseffect niet significant waarneembaar is voor de daadwerkelijke realisatie van energiebesparende investeringen. Aan de hand van logistische regressieanalyses is getracht het hebben van een intentie om te investeren en het daadwerkelijk uitvoeren van een investering te voorspellen. Uit de analyses blijkt dat residentiële kenmerken, zoals het bouwjaar van de woning en het type woning doorslaggevend zijn. Een voorzichtige verklaring voor de bevindingen is dat een verhuisintentie een negatieve impact heeft op het realiseren van energiebesparende maatregelen die pas op langere termijn rendement opleveren. Op basis van de bestaande theorieën en bevindingen van dit onderzoek kan worden gesteld dat jongvolwassenen mobieler zijn op de woningmarkt en daarom minder vaak hun intenties tot uitvoering brengen.

Sleutelwoorden: Energie efficiëntie, Energiebesparende investeringen, Generatie Y, Woontrajecten, Residentiële gebouwen, Intenties en daadwerkelijk gedrag, Duurzaam gedrag, Jongvolwassenen.

(9)

9

1. Introduction

In recent popular literature generation Y is described as the ‘most sustainability-conscious’ generation yet (Rayapura, 2014). Young adults are more supportive of stricter environmental laws and are more likely to attribute global warming to human activity and therefore favour environmentally friendly policies such as green energy development (Rayapura, 2014). Rayapura (2014) also states that ‘Gen Yers’ are most likely to pay more for responsible made products and want to work for companies that care about their environmental impact. Rogers (2013) wrote that generation Y is showing signs of a major shift in green attitudes and behaviour, and considers the young adults as ‘born green’ because they grew up in a society where eco-consciousness was becoming a norm. Moreover, Hume (2010) argues that generation Y, those roughly born between 1980 and 1995, is seen as the cohort that represents the future of our society. They are the future consumers, workers and innovators. According to Heist (2014), generation Y desires quality and authenticity which makes the young adults more willing to maintain, alter, repair, and create items. This new lifestyle has the potential to impact all sorts of consumer products because young adults want to hold onto their investments for the long-term (Heist, 2014). Repairing your own mobile phone, growing your own vegetables and the maintenance, repair and modification of your own clothes or even transport vehicle are examples of sustainable practices of generation Y.

The aforementioned examples of the supposed sustainability of generation Y are further elaborated upon in a book edited by Barendregt and Jaffe (2014). They use the term eco-chic for the combination of lifestyle politics, environmentalism, spirituality, beauty and health, combined with a call to return to simple living. Barendregt and Jaffe (2014) use the term eco-chic for the identity kit which emerged in the middle class, the newly rich and the traditional elites to combine taste and style with care for personal wellness and the environment. Being part of the eco-chic is a desire to signal exclusive taste and prestige and points to the gentrification of sustainability. An example can be for instance the positive connotations with investing in solar panels in opposition to investing in thermal insulation of the walls which cannot be seen from outside (Inspectrum, 2013). To some extent solar panels on someone’s roof function as a status symbol to show off how modern and environmental aware the people are (Inspectrum, 2013).

(12)

12

The notion that generation Y can be considered as ‘the most sustainability-conscious generation yet’ can be questioned as investments in energy saving measurements in the home might only be the practices of young adults who possess economic capital which is required to set intentions to action. When it comes to more expensive energy saving measures which mostly take more time and effort to realise, it can be expected there is a discrepancy between the intention to do such investments and the actual realisation. The latter can be further explained by using the terms stated and revealed preferences (e.g. Adamowicz et al., 1994; De Groot et al., 2008). In following of recent popular literature such as the articles by Rayapura (2014) and Heist (2014), it can be expected that young adults more often have the intention or stated preferences when it comes to investments which enhance the energy efficiency of the home. On the other hand, it remains unclear to what extent these stated preferences translate into revealed preferences. In order to find out more about the gap between the intention to invest or having done investments with regard to the energy efficiency of houses in which young adults live, the following problem definition is formulated.

To what extent do stated and revealed preferences correspond in sustainable practices of young adults and how can this concurrence be explained?

It might be the case that the actual sustainable practices of generation Y remain on the level of buying fair trade coffee or making use of public transport more often, but not much is written about stated and revealed sustainable investments of young adults who are willing to put in some work. Additionally, it is interesting to investigate the topic from a housing career perspective. Is generation Y doing sustainable investments in the home or just consuming sustainable products as a status symbol? The moment of buying a house is possibly a trigger for doing energy saving investments because then there is time and space to address such aspects.

The societal relevance of research on the topic of sustainable practices of generation Y is twofold. Firstly, if sustainable practices are only a desire to signal exclusive taste and prestige it might lead to an increasing gap between rich and poor. When the middle class can afford to invest in for example thermal insulation and solar panels and improve their house, it leaves less wealthy households with lower quality houses. It also refers to the process of ‘othering’ and the abjection of ‘other people’ who might not be able to afford sustainable living practices (Sibley, 1995). For instance, a hypothesis can be that investment and

(13)

13

divestment of certain groups can increase the polarisation on the housing market. Secondly, Hajer (PBL Leefomgeving, 2014) states in a presentation about the state of the Dutch living environment the need to upgrade the current housing stock instead of building new houses. As generation Y is seen as the future consumers, workers and innovators, it is interesting for various institutions to have a better understanding of the actual sustainable practices of the young adults of today (Hume, 2010). They might be the ones to achieve the need to make the current housing stock more sustainable. Moreover, the existing housing stock is one of the largest potential sectors to reduce the use of energy. Making the buildings more energy efficient is crucial to achieve the goal of reducing the CO2-emissions with 80 to 95% by 2050 (PBL, 2014). This is part of the goal of the Dutch national energy agreement for sustainable growth which strives for an ‘energy neutral’ built environment by 2050 (PBL, 2014). Nevertheless, it can be expected that high mobility of young adults, uncertainty on the housing market, and the lack of economic capital are not in favour of creating a more energy efficient housing stock and the achievement of sustainable goals.

The scientific relevance of studying sustainable practices of generation Y can be found in the contribution to the debate concerning changing lifestyles and housing preferences. In 2002, Clapham coined a new approach with the focus on housing pathways based on social constructionism which offers a way of integrating identity and lifestyle choice into housing analysis. Research on sustainable practices in the homes can therefore be seen as a derivative of the current scientific interest in the residential trajectories of young adults. The way in how young adults manoeuvre in the field of housing can be related to their supposing sustainable practices as living in the informal housing sector makes it difficult to request any subsidies.

The main theories and literature on the housing pathways of young adults and energy saving investments in residential buildings will be further elaborated upon in the next chapter which includes a theoretical framework of the wider scientific debate on the topic. Furthermore, this master thesis consists of a chapter concerning the research design and methods which includes the research question and sub-questions, definitions of the main concepts, and a conceptual model. The data collection and the limitations of the research will also be touched upon. The second part consists of the results and analysis including the answers to the sub-questions. Finally, in the conclusion the research question will be answered and followed by a reflection and recommendations for further research. The reference list and appendices are attached at the end of this master thesis.

(14)

14

2. Theoretical framework

This framework provides an overview of the relevant academic debate in which the master thesis is embedded. Firstly, a short introductory section follows which addresses the notion of a generation to get a better understanding of what generation Y actually means. Secondly, the theory of planned behaviour and the method of investigating stated and revealed preferences are highlighted which are used by academics studying the field of housing. Thirdly, literature concerning housing pathways and sustainable practices of young adults will be covered. And finally, a summary of the literature with regard to energy saving investments in residential buildings is provided.

2.1 Introduction: Studying a generation

According to Hume (2010), generation Y, those roughly born between 1980 and 1995, is the cohort that represents the future of our society. To study a specific generation or societal phenomena, researchers have been using positivist approaches in the past, but the dominant paradigms formulated in a societal context are very different from that of today. Globalisation characterised by the privatisation of the housing market, and also individualism which relates to the search for a sense of identity and self-fulfilment through lifestyle choices are part of the current postmodern society (Clapham, 2002; Hume, 2010). In line with Clapham (2002), Manting (2013) therefore emphasises the importance of studies and analysis through a ‘life course perspective’ which incorporates other aspects in life such as receiving a diploma and finding a job. The focus on individuals or households is important because spatial changes are the sum of the developments at the lowest aggregation levels (Manting, 2013).

So what defines a generation as the group to study from a life course perspective? According to Mannheim (1952), shared experience along biological or historical lines are possible but not the determining boundaries of a generation. Boundaries of generational experience may be socially and culturally influenced (Mannheim, 1952). Results of a research conducted by Grenier (2007) show generational tensions by the assertion that generations develop their own linguistic models which contribute to misunderstanding between age cohorts. Macky et al. (2008) theorise differences between generations to the occurrence of major influences in the environment during the ages in which early human socialisation occurs. These influences have an impact on the development of personality, values, beliefs and expectations which are stable once formed into adulthood. Thus, experiences and shifts in

(15)

15

the socio-cultural environment over time are critical in the socialisation years (Macky et al., 2008). Generation Y, those now roughly aged between 20 and 35 years, are likely to have their years of socialisation behind them which makes it an interesting case to study.

2.2 Bridge: from theory to theme

The theory of planned behaviour by Ajzen (1991) which is used to investigate the discrepancy between intentions and actual behaviour, is one of the key theories to address as it is widely used to get a better understanding of why intentions lead or do not lead to action. The latter theory is an extension to the theory of reasoned action and includes the exogenous variable of perceived behavioural control. When people have little control over the execution of an intention because of a lack of resources, their intentions to result in actual behaviour may be low even if their attitude is favourable (Madden et al., 1992). For instance, it is possible that students have the intention to reside in Amsterdam, but remain longer at the parental home because they lack economic, social, cultural or symbolic capital (Hochstenbach & Boterman, 2014; in following of Bourdieu, 1989). The discrepancy between intentions and behaviour can be explained by the underlying factor of triggers in the labour, demographic, or occupational cases. Union dissolution, union formation, childbirth, and losing a job are significant examples of triggers for actual moving or staying put behaviour (De Groot et al., 2011). Another way to examine the discrepancy between intentions and actual behaviour is the stated and revealed preference method. De Groot et al. (2008) for example, use a longitudinal research design to investigate to what extent stated housing preferences have turned into revealed housing preferences of Dutch households. More often a cross-sectional research design is used to compare people who are willing to move and other people who have moved. This type of research costs less time and money in comparison to a longitudinal research design for which the same people has to be followed over time. A cross-sectional research design will be used for the master thesis as it is the best option given the available time.

Recent studies have shown the increasingly complex pathways and various triggers young adults (i.e. generation Y) experience in their transition to adulthood (Heath, 2008; Lennartz et al., 2015). The suggestion is that young adults experience more struggles in housing markets which have taken a turn for the worse since the global financial crisis started in 2008 (McKee, 2012; Clapham et al., 2014; Lennartz et al., 2015). Additionally, Smith and Sage (2014) show for example that high mobility occurs during youth phases and migration as an important element in the transition from youth to adulthood. Therefore, it is interesting to further explore the literature about young people’s housing preferences and trajectories.

(16)

16

2.3 Housing pathways and preferences of young adults

Clapham et al. (2014) state that the general context will impact differently on young people with different resources and preferences, meaning there could be substantial variations in the pathways young people take over time. Clapham (2002) coined a new approach with the focus on housing pathways based on social constructionism which offers a way of integrating identity and lifestyle choice into housing analysis. This notion is based on the idea that housing is increasingly viewed by households as a means to achieve personal fulfilment rather than an end in itself (Clapham, 2002). Ford et al. (2002) for instance conceptualised the transition of young people into adulthood through five pathways, but also mentioned the ability of participation in housing markets which is determined by non-housing attributes. The latter is further investigated by Hochstenbach and Boterman (2014) for the case of young people in Amsterdam with a reduction to three pathways; a linear, a progressive chaotic, and a reproductive chaotic housing pathway. The combination of a pathway approach and the sociology of Bourdieu and De Certeau showed that class differences and inequalities between ‘outsiders’ and ‘insiders’ of young people are reproduced on the Amsterdam housing market (Hochstenbach & Boterman, 2014). Moreover, the ‘outsiders’ coming from outside the Amsterdam region with little knowledge of the local housing market and minor economic and other forms of capital are more likely to follow a reproductive pathway in the informal or temporary housing sectors. A reproductive housing pathway can be characterised by frequently undesired forced moves. However, most young people follow a progressive housing pathway by utilising social and cultural capital in the search for housing in the affordable private-rental sector (Hochstenbach & Boterman, 2014).

High levels of demand make the commodified housing stock relatively expensive (Hochstenbach & Boterman, 2015). Therefore, young adults in major cities are forced to follow more complex pathways in their early housing careers. In the case of Amsterdam there has been a great shift of younger households into more expensive rental housing. This can also be found in other European cities (Boterman et al., 2013; Lennartz et al., 2015). Homeownership is strongly in decline for young adults which even led to the label of ‘Generation Rent’ for young adults in the UK (Boterman et al., 2013; Lennartz et al., 2015). Although buying a house is more difficult than before the financial crisis, homeownership has become a key element of a successful lifestyle package (McKee, 2012).

(17)

17

2.4 Sustainable lifestyles of young adults

To elaborate on the variety and changing of lifestyles of young adults, Rogers (2013) wrote that generation Y is showing signs of a major shift towards green attitudes and behaviour. Rogers (2013) considers generation Y as ‘born green’ because they grew up in a society where eco-consciousness was becoming a norm. Hume (2010) undertook a generational approach to contribute to the sustainable practices debate and defines sustainability in a three dimensional approach consisting of social, economic and ecological dimensions which are mainly related to the production processes of companies. In contrast to Rogers (2012), Hume (2010) concludes that generation Y is considered socially, economically and environmentally conscious, a clear pattern of contradiction exists between what they know and what they practice in regard to sustainability efforts. The combination of lifestyle politics, environmentalism, spirituality, beauty and health, combined with a call to return to simple living is summarised in the term eco-chic in a book edited by Barendregt & Jaffe (2014). The term connects to the fields of ethical, sustainable, and elite consumption. Barendregt and Jaffe (2014) use the term eco-chic for the identity kit which emerged in the middle class, the newly rich and the traditional elites to combine taste and style with care for personal wellness and the environment. Being part of the eco-chic is a desire to signal exclusive taste and prestige and points to the gentrification of sustainability. According to Barendregt and Jaffe (2014) the most fruitful approach to green consumption is one that focuses on shared imaginaries, communal norms, and collective practices, as well as on civic responsibilities. These are all aspects that could have been of influence during the socialisation process of the young adults who are now part of generation Y.

The recent thesis by Hofman (2014) and the report with the results of the analysis on the WoON energy module by Tigchelaar and Leidelmeijer (2013) can be highlighted with regard to sustainable practices of households in the Netherlands. For instance, behaviour appears to have a great impact on the energy usage of households (Tigchelaar and Leidelmeijer, 2013; Hofman, 2014). Awareness about energy usage has a positive effect on sustainable behaviour and residents appear to live more sustainable in less efficient homes, whereas people in more energy efficient homes live less sustainable. Therefore, the theoretical difference between energy usage for different energy labels of houses is smaller in reality (Tigchelaar & Leidelmeijer, 2013). This is in line with the notion of the existence of a rebound-effect (Haas, 1997). Efficiency and behaviour may influence each other, because a higher efficiency leads to a lower energy usage and hence may change consumer behaviour with respect to choosing energy usage levels (Haas, 1997). Tigchelaar and Leidelmeijer

(18)

18

(2013) also found a mismatch between people who want to invest and where they live. People who are willing to invest, belong to the high income middle age group who have recently moved, whereas people who will not invest, live already in efficient houses and can be characterised as old people with a lower income who stayed put for a long time. In addition, homeowners stepwise invest in their houses to make them more energy efficient, where on the other hand people in the rental sector are difficult to stimulate to take measures (Vringer et al., 2014). Moreover, Hofman (2014) relates sustainable behaviour to the size of the house, income, size of the household and environmental awareness, but also the urbanisation rate and neighbourhood involvement seem to influence the probability of using less energy. Research in Australia conducted by Waitt et al. (2012) shows the importance of socio-cultural contexts in differentiating sustainable capabilities of the household. Women, suburban-detached households and lower income segments of the population are ultimately doing most of the work of being sustainable (Waitt et al., 2012). It should become clear that how households act is a different matter from what they know about sustainability. Willingness to act may be constrained not only by cultural and conventional limits, but also to what counts as acceptable consumption, practice and specific material, technological and economic restrictions and possibilities (Waitt et al., 2012). Sustainable practices of a household are understood as a resourceful process of juggling, trade-off and manipulation of daily routines, purchasing practices, and use of building, garden and domestic technologies (Waitt et al., 2012).

2.5 Energy saving investments in residential buildings

Besides studies on the sustainable aspects of certain lifestyles, also some research has been conducted on the main topic of this master thesis, namely energy saving investments in residential buildings. The residential buildings incur an important share of the national energy consumption. The improvement of energy efficiency in the building sector can have an important impact on the total energy consumption of a country and a considerable contribution in attaining the emissions objectives for a sustainable development (Banfi et al., 2008). Banfi et al. (2008) investigated the willingness to pay for energy saving measures in Switzerland’s residential buildings. The main results suggest that the benefits of the energy saving measures are significantly valued by the consumers. However, some barriers have been observed such as legal, structural and socio-economic, but the observed underinvestment is mainly due to a lack of information regarding the enhancement of the energy efficiency. Banfi et al. (2008) state that the government can reduce some of the barriers by supporting communication and information and granting additional financial resources to homeowners

(19)

19

who want to invest to overcome financial barriers. The research of Tommerup and Svendsen (2006) can add to these findings because according to them no essential barrier exists to realise a major part of the big energy savings potential. The major barrier for people not to energy-renovate their buildings seems to be the lack of knowledge and interest. The market for energy savings does not react rationally on traditional market forces, such as energy price level, which means that legislation and control are necessary and important tools for changing the behaviour of homeowners (Tommerup & Svendsen, 2006).

According to Poortinga et al. (2003) most studies on household energy usage had the focus on social and psychological factors influencing the acceptability of the energy saving measures with a lesser focus on the influence of physical characteristics. Poortinga et al. (2003) found that technical improvements were preferred over behavioural measures and especially shifts in consumption. Also, home energy saving measures were more acceptable than transport energy saving measures. Interestingly, the respondents aged between 20 and 39 years old thought that technical improvements were relatively more acceptable than respondents aged 65 years and older, and families and couples found technical measures relatively more acceptable than singles. Although the younger cohort is apparently more willing to invest in energy saving measures, Hamza & Gilroy (2011) argue that the decisions made by the baby boomer cohort may be fundamental to the success or failure of not only national but global climate change policies. The predictions for the baby boomer cohort are for a high spending generation looking to invest in their leisure time whether in the home or on holidays (Hamza & Gilroy, 2011).

Bell and Lowe (2009) investigated the modernisation of low rise housing in the UK. The authors state that energy savings technologies can be successfully applied as a project, and that it is well within the capacity of existing technology and construction practice, but the major problem that the UK and many other countries are facing is the transfer of such improvements into the housing stock as a whole. Bell and Lowe (2009) state that it will require policies which encourage and enable millions of individuals in owner occupied dwellings to make an investment in for example improvements of space heating, solar panels, insulation, and more efficient water boilers. Thus, the authors suggest that despite the opportunities, homeowners remain reluctant when it comes to energy saving investments in residential buildings. The research of Zundel & Stiess (2011) show similar findings. They state that homeowners can save considerable amount of energy and money if they adjust their homes according to the latest energy efficiency standards. The technical measures have proven to be profitable in relatively short payback times, if only additional costs are

(20)

20

considered. However, homeowners do not follow this type of rationality. Based on research conducted in Germany, Zundel and Stiess (2011) argue that the decision to invest in energy saving measures is shaped by both economic and non-economic goals and motives. Energy saving measures might lower the energy bill and secure the value of the house, but they must also correspond to the needs such as comfort, convenience or belonging in order to be realised. The combination of various needs controls the behaviour of homeowners. The energy-saving measures must correspond to the combination of wants and needs to have the chance to be realised (Zundel & Stiess, 2011). Judson & Maller (2014) made use of ethnographic case studies to investigate the extent to which low-energy and other environmental concerns are important in the process of housing renovations. Both current daily routines and those anticipated in the future have an impact on retrofitting practices which make the house more energy efficient. All their ethnographic stories show that energy efficiency is regarded of lesser importance than the accommodation of the household practices (Judson & Maller, 2014). Investments in enhancing the energy efficiency of the residential building therefore occur mainly in combination with renovations that involve a change in interior space such as adding bedrooms and/or bathrooms. Lastly, Nair et al. (2010) found that the majority of Swedish homeowners considered it important to reduce their household energy usage and adopted some measures. However, most of them undertook non-investment measures such as switching of lights. Besides the income of the household, the age of the building was found to influence the choice to invest in energy saving measures. Although, homeowners with a very old building were also more likely to take energy saving measures as a part of a major renovation (Nair et al., 2010).

The assumptions concerning the sustainable lifestyles of young adults are the main motivation to study the subject of energy saving measures in residential buildings in this master thesis. Not much have yet been written on the crossroads of the expected sustainability of young adults and energy savings investments in residential buildings. Therefore, in this master thesis the stated and revealed preferences are investigated with regard to energy saving measures in the home. This is done in the context of the housing pathways in order to find out how this impacts the decision to enhance the energy efficiency of the home.

(21)

21

3. Methodology

In this chapter the problem definition from the introduction is focused into a research question and three sub-questions. These questions are further elaborated upon in the section with the hypotheses. This is followed by sections on the operationalisation of the concepts, the research design and the data collection. This chapter ends with the possible limitations of this research and ethical aspects which should be taken into account.

3.1 Research question

The master thesis is set up to investigate the stated and revealed sustainable practice preferences of generation Y. Therefore, the following research question is formulated:

To what extent do stated and revealed preferences correspond in sustainable practices of generation Y in the Netherlands and how can this concurrence be explained?

Based on the theoretical framework the following three sub-questions are formulated:

1. What are the stated sustainable practices of generation Y concerning their homes and to what extent can this be placed in a generational perspective?

2. To what extent is there a discrepancy between stated and revealed sustainable practice preferences of generation Y?

3. How can the stated and revealed sustainable practice preferences be related to the housing pathways of young adults?

(22)

22

3.2 Definitions

The main concepts which are mentioned in the research question and sub-questions can be defined as follows:

Generation Y: Those roughly born between 1980 and 1995 who have been influenced by increasing processes of globalisation and individualism during their years of socialisation and which is seen as the most sustainability-conscious generation yet.

Sustainable practices: The resourceful process of juggling, trade-off and manipulation of daily routines, purchasing practices, and use of building, garden and domestic technologies of the household with the aim to reduce the ecological footprint. In the master thesis the main focus will be on domestic energy saving investments for example in thermal insulation, solar panels and a more efficient water boiler.

Housing pathways: Patterns of interaction concerning house and home, over time and space which are composed of individual housing steps to which the households in question managed to gain access. This approach offers a way of integrating concerns about identity and lifestyle choice into housing analysis in the postmodern search for self-fulfilment.

Stated preferences: Intentions to achieve a certain practice which depends on the perceived necessity of the act. Adaptive preference formation plays an important role in adjusting the intention before actual behaviour is accomplished.

Revealed preferences: Actual behaviour as a result of intentions which are set to action. Underlying triggers play an important role for intentions to result in action.

(23)

23

3.3 Conceptual model

The below depicted model shows a schematic overview of the research question.

The discrepancy between stated and revealed sustainable practices is the central theme in this master thesis. The stated and the revealed sustainable practices are investigated by introducing three types of characteristics which are expected to influence both aspects. Possibly, a certain type of characteristics is a stronger predictor for the intention to invest, whereas this can be a different or the same type of characteristics for the revealed sustainable practices. Finally, the discrepancy between the stated and revealed preferences is further investigated by focussing on the length of residence and the residential mobility. This is in order to place both the stated as the revealed sustainable practice preferences in the context of housing pathways. Then it can be investigated whether high residential mobility has an impact on the translation of the intention to invest in energy saving measures into actual investments. A recent change in residential location might function as a trigger to invest in energy saving measures. How the various types of characteristics are constructed and how the context of housing pathways is framed by the length of residence and residential mobility follows in the section with the data analysis and results.

3.4 Hypotheses

The majority of the literature which was used for the theoretical framework is based on research conducted in the United Kingdom and The Netherlands, i.e. an Anglo-Saxon context. This is important for the transferability of findings and setting up hypotheses for this master thesis. The latter consists of three sub-questions which together should provide an answer to the main research question. In this paragraph the sub-questions are further explained including possible and expected outcomes based on the literature.

Stated sustainable practices Revealed sustainable practices

-Length of residence -Residential mobility -Personal characteristics

-Household characteristics -Residential characteristics

In the context of housing pathways

-Personal characteristics -Household characteristics -Residential characteristics

(24)

24

The first sub-question seeks to provide statistical evidence of the stated sustainable practices of young adults. This is done by analysing data which include questions about the intention to invest in the house of young adults. The expectation is that young adults do have the intention to enhance the energy efficiency of their home, but it is also possible that only a specific age category within generation Y, e.g. those aged 30-35 years old and possesses the necessary economical capital, has this intention (Tigchelaar & Leidelmeijer, 2013). It can also be hypothesised in following of Nair et al. (2010) that mostly homeowners have the intention to invest in energy saving measures, although most of them eventually remain on the level of switching off the lights when not around. Also, personal and residential characteristics might be the most important predictors for having the intention as Nair et al. (2010) found that income and the age of the building influence the decision to invest.

The second sub-question seeks to investigate the gap between stated and revealed sustainable practices of generation Y. This has also been done by taking a quantitative approach. After investigating only the stated preferences, including the revealed preferences makes it possible to see who have actually invested in their home. It might be difficult to investigate this in a longitudinal way, but a cross-sectional design with a retrospective question is used; did you invest in energy saving measures the past five years? The hypothesis for this sub-question is that the gap between intentions to invest and actual energy saving investments is significant as it might primarily be the practices of those who can afford it (i.e. eco-chic in Barendregt & Jaffe, 2014). Furthermore, in following of Bell and Lowe (2009) and Zundel and Stiess (2011) homeowners might be reluctant when it comes to investments in energy saving measures because the decision is not only shaped by economic, but also by non-economic goals and motives. Nevertheless, homeowners are likely to have more possibilities to make energy saving modifications to their home than tenants.

The third sub-question includes two aspects of the housing pathways of young adults in order to find out if a change in residential location or length of residence has an impact on the sustainable practice preferences of young adults. This can be investigated by incorporating moving intentions or recent moving behaviour as an explanatory variable for the sustainable practice intentions of young adults. Although arriving in a new place might be a trigger for sustainable investments (Tigchelaar & Leidelmeijer, 2013), it can be expected that young adults who are very mobile in their early life stages (Smith & Sage, 2014), do not really have the intention to invest in enhancing the energy efficiency of their home. It is also possible that young adults who invest in their home are in later stages in life and decided to settle down for

(25)

25

a longer period of time (Tigchelaar & Leidelmeijer, 2013). Thus, a longer period of residence could increase the likelihood of having invested in energy saving measures.

3.6 Data collection and research design

This research primarily consists of quantitative methods, but also a qualitative and a spatial element are introduced which is in line with the idea that the positivist approaches from the past should be expanded with the studies from for example a ‘life course perspective’ or a ‘generational perspective’ (Clapham, 2002; Hume, 2010; Manting, 2013). Although, the basis of this study can be found in the quantitative part, a small qualitative/spatial part is used to make the findings more tangible. Therefore, this research has a cross-sectional design, but also minor characteristics of an intensive analysis of a single case-study, e.g. the Netherlands. In a cross-sectional research design, a body of quantitative data in connection with two or more variables is examined to detect patterns of association (Bryman, 2008, p.44). The research design for the three research questions is a cross-sectional design because young adults in the Netherlands will be investigated by analysing several variables which have been measured at one moment in time.

The three sub-questions are answered by making use of secondary data which means that existing data has been reworked and analysed. Existing data sets derived from ‘Woon Onderzoek Nederland (WoON) have been used. The main focus is on the energy module 2012 which is a derivative of the main research of WoON 2012. The respondents from WoON 2012 were approached to perform in one of the sequels of which the energy module is one. A specific section follows in chapter 4 about the statistical information of the Dutch housing market which is obtained every three years. The paragraphs on the used data set include information about the survey set-up, response rate, and possibilities for investigating the topic of sustainable practices by making use of these data sets in particular. Furthermore, two interviews were held with two couples who belong to generation Y to make the quantitative findings more tangible. These interviews took approximately 45 minutes at the houses of the respondents. The interviews were guided by a semi-structured item list which can be found in the appendix. Additionally, three maps have been created to incorporate a basic spatial analysis of the stated and revealed sustainable practice preferences in the Netherlands. More details about the procedures for the interviews and maps follow in the relevant chapters.

(26)

26

3.6.1 Units of analysis

The young adults who can be comprised to generation Y in the Netherlands are the units of analysis in this research. Because of continuously comparing to all other respondents, also other households will be investigated. Therefore, the units of analysis can be better described as the households that participated in the WoON 2012 and Energy Module 2012 researches. As the research should be representative for the population of the Netherlands, every inhabitant of the Netherlands is in potential a unit of analysis for this master thesis. The research is based on two levels of aggregation. The first is the age cohort level, level n. The second is level n-1, the level of the households. The age cohort level consists of every respondent who can be placed in a specific generation of approximately fifteen years. The two households that have been interviewed were selected on specific characteristics such as being homeowners or tenants in order to be of interest for this research. The households are thus found through purposive sampling (Bryman, 2008, p.414). The goal of purposive sampling is to sample participants in a strategic way, so that those sampled are relevant to the research questions that are being posed. However, purposive sampling does not allow the researcher to generalise to a population. This is not a problem as the interviews only provide some quotes to make the quantitative findings more tangible. The criteria for selecting the two respondents are thus derived from the findings of the quantitative analyses.

3.6.2 Operationalisation

The abstract terms in the research questions are not directly researchable indicators. It is necessary to translate the concepts into measurable variables. Only then it is possible to analyse the primary and secondary data. In this paragraph the conceptual model is processed into dependent and independent variables. These variables will become measurable by pointing out indicators (Bryman, 2008, p.144). The indicators are for example specific questions from the WoON 2012 survey. Variable ‘X’ is possibly influencing variable ‘Y’. The conceptual model depicts the relationship between stated and revealed sustainable practices and the variables or context which can be influencing this relationship. In this case the stated preferences are variable X and the revealed preferences variable Y. Nevertheless, these variables should become more concrete for this research. It is chosen to provide an overview of the variables in the second part of this master thesis because not for all three sub-questions the same variables are used. In each of the chapters the variables are explained to the fullest. This paragraph provides only a short introduction to what will follow and how the overall operationalisation of the concepts looks like.

(27)

27

For the first two sub-questions, indicators are mainly derived from the Energy Module 2012. The dependent variable which should refer to the sustainable practices is concerned with energy saving investments in the home. A possible indicator is for example; do you consider energy saving investments in the next two years? (question 52). Explanatory variables are sought in personal, household, and residential characteristics. These are for example; level of education, income of the household, year of construction of the house, and living in a rural or urban environment. The second research question seeks to quantify the discrepancy between stated and revealed sustainable investments. Therefore, based on descriptive statistics the difference between the young adults who did invest and who are willing to invest will be analysed. The main focus will be on the second dependent variable, namely the revealed practices. Indicators for this variable are for example; did you invest in the past five years in thermal insulation and/or double-glazing? (question 43), did you change the boiler in the past five years? (question 44), or did you apply any other energy saving measurements in the past five years? (question 45). In order to answer the second research question these indicators have been combined because not a comparable direct indicator as for the first research question could be derived from the questionnaire. The previously used model with the explanatory variables can be used again, whereas only a different dependent variable had to be constructed. To answer the third research question, explanatory variables such as willingness to move and a recent change of residential location are introduced. Combined with the length of residence, these variables relate to the housing pathways approach and possibly influence the intention to invest and having invested in the residential building.

3.6.3 Limitations and side notes

Most research questions cannot be answered fully because of limitations. Therefore, it is necessary to make these limitations and the consequences for the research questions explicit. It was an obvious choice to make use of quantitative methods for investigating the sustainable investments of households because of the existence of relevant data sets. The sample size and quality of research commissioned by the Dutch ministry of the interior and kingdom relations would not be reached in the time period the master thesis had to be written. Although, the WoON data and the more specific Energy Module contain data which can be used for research on the topic of sustainable practices, it still has some limitations. This research seeks to test hypotheses based on existing theories, but there are some constrains that come with making use of existing data.

(28)

28

Firstly, the data sets contain of a large number of respondents and questions, i.e. variables. The large number of respondents makes it possible to make a selection of only the respondents who were born in a specific year in order to focus on generation Y. However, because only cross-sectional data of 2012 is used, it is not directly possible to investigate a generation effect. If the latter wants to be achieved, data related to different moments in time should be researched. Investigating an age or a generation effect is further touched upon in chapter 4. In addition, the large amount of variables makes it difficult to pick the ones that are most suitable as indicators for this research. This was for instance the case for the construction of the variable which measures the revealed sustainable practice preferences as not a directly comparable indicator to the stated preferences could be derived from the questionnaire. Furthermore, all percentages that follow in this master thesis have one decimal number. Here a side note must be given because based on the data it is not possible to present such precise results for all of the analysis. In order to be consistent, it is chosen to present the percentages in the same way.

Secondly, because of using an already conducted survey it is not possible to influence the process of data gathering. There are no doubts about the quality of the research as it was conducted by professionals, but the focus of the surveys is not completely the same as the topic of this master thesis. The demarcation of the indicators for various variables depends on the questions in the surveys. Although a self-made survey also has its limitations, it likely has more indicators to research the topic. In this case the advantages of using an existing data set outweigh the limitations.

Lastly, some disadvantages occur when only relying on statistical evidence. It might be possible that certain mechanisms or triggers are not touched upon. Therefore, some qualitative/spatial analysis is performed in order to answer the third research question. Because of the balance between the methods is very uneven, there are limitations considering the validity, transferability, and reliability of the findings. It must be stated that the qualitative/spatial approach in the third research question only provides an exploration and illustration of the influence of the housing pathways of young adults on the sustainable practices. Still, the experience of gathering data and observing the statistical findings in a real setting has advantages for the master thesis as it leads to recommendations for further research.

(29)

29

3.6.4 Ethical aspects

As previously mentioned, the three sub-questions will be answered by making use of an already conducted research. This left no options for influencing the size and the selection of the sample. It might still be the question whether the sample is representative for everyone who belongs to generation Y in the Netherlands. The response analysis in the next chapter will elaborate more on this aspect. Because the surveys were set up and conducted on behalf of the Dutch government, it can be assumed that the data gathering was performed alongside the ethical guidelines.

However, because two semi-structured interviews were conducted with young adults, it is necessary to discuss some ethical aspects. When the respondents are asked about their triggers for not investing in energy saving investments in the home, it might be the case that they give socially desirable answers. Some people find it difficult to admit they don’t have the money for such investments. If a respondent does give a strong opinion and for example says that the landlord does not want to collaborate to realise such sustainable upgrades to the house, there are possibilities of harm to participants (Bryman, 2008, p.118). The respondent can suffer from negative consequences when they are cited in the master thesis. It is possible to make the quotes anonymous, but this has negative consequences for the reliability of the research. The respondents have been asked before the interviews if they can be cited in the master thesis. By asking further questions it can be tried to avoid social desirable answers. The interviews are semi-structured in order to get an answer to the most important questions, but this also leaves room for independent stories of respondents.

(30)

30

Part II: Data analysis and presentation of results

4. Stated sustainable practice preferences of young adults

Several characteristics can be of influence on the stated sustainable practice preferences of generation Y. These characteristics can be divided into personal, household, and residential variables of which the construction will be further elaborated upon in this chapter. In following of the theoretical framework, some hypothesis or expected outcomes have been formulated. For instance, it be assumed that many households do have the intention to do energy saving investments in their homes, but it is also possible that only a specific age category within generation Y, e.g. those aged 26-35 years old and/or those who possess the necessary economical capital, have this intention.

This chapter provides descriptive statistics and regression analysis which contribute to a better understanding of the stated sustainable practice preferences of generation Y. Besides, the latter analysis is complemented with a paragraph that focuses on the age effect of the age cohort that relates to generation Y. These quantitative analyses combined, provide an answer to the first sub-question:

What are the stated sustainable practice preferences of generation Y concerning their homes and to what extent can this be placed in a generational perspective?

The used data sets for the quantitative part of the master thesis are addressed in the following paragraph. It must be made clear that these data sets are freely accessible after registration at the Data Archiving and Networked Services (DANS) of the Royal Academy of Science (KNAW). The data files can be processed in the computer program SPSS Statistics, developed by IBM. Therefore, this program is also used for this master thesis to analyse the data. As this master thesis is written in combination with an internship at the Netherlands Environmental Assessment Agency, the data was provided through an internal database which includes some explanatory documents next to the particular data sets. These documents include for example, the complete questionnaire and a comprehensive fieldwork report provided by Inspectrum, one of the companies that collaborated with the Dutch government in order to successfully conduct the research.

(31)

31

4.1 Introduction: Analysis of survey data

Before jumping into the various variables and analysis, it is necessary to elaborate on the used data and preparations for the analysis. In a previous chapter it was mentioned that the WoON 2012 data will be used in combination with the Energy Module 2012 for the quantitative analysis. Every three or four years the Dutch government and affiliated companies conduct research on the composition of households, living expenses, housing needs and other aspects of the living environment in the Netherlands (BZK, 2013). The research consists of a survey which is held under people who are eighteen years and older and live in the Netherlands. Approximately 70.000 respondents participated in the research between September 2011 and May 2012 (BZK, 2013). The data was processed in a way that individual respondents cannot be traced in order to make it possible for research institutes and students to perform further analysis.

Although WoON 2012 is the main research, it is not the primary data set which is used for this master thesis. The focus of the master thesis is on the sustainable practice preferences of young adults which is more in line with the research conducted in the Energy Module 2012, a follow-up module of the WoON 2012. The Energy Module was held in 2006 for the first time and for the second time in 2012. As it is part of the WoON researches it is possible to enrich the data sets with each other. For this occasion, it was chosen to add the data of the WoON 2012 to the data of the Energy Module 2012. This could be done by joining the data based on a common indicator, respectively the individually assigned number to each of the respondents. The Energy Module 2012 mainly focuses on the energetic aspects of a house, but also on technical and human factors concerning the energy consumption of a household (BZK, 2012; De Jong, 2013). The research company Inspectrum, as well as other companies such as Telescore and Covadis, collaborated with the Dutch ministry of the interior and kingdom relations to conduct a survey and house audits to get a better understanding of how buildings and inhabitants perform on the terrain of energy consumption.

4.2 Sample & response

In a comprehensive document provided by Inspectrum, all choices and difficulties concerning the fieldwork for the research have been addressed (De Jong, 2013). The desired amount of respondents was approximately five thousand which included a completely filled in survey and a house audit (De Jong, 2013). In the sampling process, Inspectrum tried to reach respondents from different housing categories, as well as respondents from different income classes and different household compositions. The sample size for the Energy Module 2012

(32)

32

was based on forty thousand respondents who were previously approached for the WoON 2012 research. In total, data of 4790 audits and questionnaires were collected in 2012. The combined data sets offer a unique and detailed insight in the fundamentals of energy consumption of residential buildings in the Netherlands (Tigchelaar & Leidelmeijer, 2013). Furthermore, the respondents were asked in the questionnaire about doing investments in energy saving investments, but also about previous investments in energy saving measures such as installing double-glazing and insulating the walls. The aspects of willingness to investment and previous investments is of high value for this master thesis.

The survey consisted of eighty-five questions and took approximately twenty-five till forty minutes to complete (De Jong, 2013). The non-response was tried to keep as low as possible by approaching the respondents in various ways. They could for example complete the survey at the door, over the phone or by email. Nevertheless, some respondents did not finish the questionnaire or did not wish to participate in the research. This was mainly because the questionnaire and audit would take too long, whereas others were not home, left a ‘fake number’, or did not reply to their email (De Jong, 2013). According to the fieldwork report, 33,9% of the sample completed both the questionnaire and the audit, and 59% of the sample did not cooperate (De Jong, 2013). Although the completion rate of only one third seems a bit low, it is still a lot more respondents than could be contacted within the available time for the master thesis.

Moreover, the fieldwork was conducted by well-trained interviewers and inspectors under supervision of Inspectrum. The pilot version was executed by ABF research which also checked the structure of the questionnaire. All in all, the data collection and processing was performed by well-known companies and professionals. Therefore, it can be assumed that the data is of sufficient quality and forms a good fit for answering the research question.

In spite of assuming the data is of high quality, it is necessary to take a look at the response as some groups or categories might be overrepresented, whereas others might be underrepresented. The latter should be kept in mind in the concluding section as some findings cannot be generalised. In the fieldwork report it is stated that the response requirements were not reached. This was primarily due to a lack of response from respondents living in the private rental section (De Jong, 2013). Unfortunately, the fieldwork report does not further elaborate on the reasons why difficulties arose in reaching this particular group.

Besides what is stated in the fieldwork report, analysing frequency distributions of the data set show some details which are more relevant for this research. The frequency tables can

(33)

33

be found in the appendix (tables 4.1 & 4.2). The tables show the distribution of the categories for both the whole data set as well as for a selection of only the respondents who belong to the age group of generation Y. The particular variables and categories are addressed in the following paragraphs. For now, a few variables show an equal distribution of its categories when considering the whole data set. Gender, level of education, type of household, and to a lesser extent income of the household have an equal or normal distribution of respondents over its categories. Of course the process of recoding and combining categories is of influence on the distributions, but for some variables the distributions are obviously skew which can be related to the response or non-response rates. For example, native Dutch respondents are strongly overrepresented with 87,5%, whereas only 8,1% of the respondents are Western immigrants, and 4,4% are of non-Western origin. Also, people who are 56 years and older are overrepresented in the data with 49,9%. Overrepresentation of certain categories also occur in the type of housing (single family, 69,3%) and tenure situation (owner-occupied, 58,6%). Moreover, most of the respondents live in the larger cities of the Netherlands (52,3%), whereas 10,3 per cent of the respondents live in rural areas.

Most of the analysis in this chapter is based on a selection of those who belong to the age category of 19 to 35 years old. This selection holds 14,5% of the respondents and is thus only a small part of the whole data set. Not to forget the overrepresentation of elderly people. For this selection, the categories for type of housing and type of household are equally distributed, but the level of education is not anymore (e.g. 10% with a low educational level). Furthermore, ethnicity again shows a very uneven distribution as 84,1% of the respondents is native Dutch. To highlight a final aspect with regard to response rate, 82,5% of all respondents think energy saving behaviour is important. Based on the selection this percentage is a little lower (73,5%). Because this variable is derived from the Energy Module 2012, this might not come as a surprise. It can be expected that people who care about energy saving behaviour are more willing to participate in a research such as the Energy Module 2012.

4.3 Preparation for analysis

Dependent and independent variables should be chosen in order to perform quantitative analysis. Based on the scale levels of these variables, certain methods can be used to calculate for example the correlation between variables and to predict the contribution of the independent variables in explaining the dependent variable. This paragraph elaborates on the choices concerning the various variables and its indicators.

Living amid energy efficiency and energy consciousness : an examination of the discrepancy between stated and revealed sustainable practice preferences of generation Y