Potential for – and benefits from – local content in Swedish wind power projects
Marcus Edlund Daniel Eriksson
Master of Science Thesis Stockholm, Sweden 2013
2
Potential for – and benefits from – local content in Swedish wind power projects
Marcus Edlund Daniel Eriksson
Master of Science Thesis INDEK 2014:72 KTH Industrial Engineering and Management
Industrial Management SE-100 44 STOCKHOLM
3 Master of Science Thesis INDEK 2014:72
Potential for – and benefits from – local content in Swedish wind power projects
Marcus Edlund Daniel Eriksson
Approved
2014-06-04
Examiner
Niklas Arvidsson
Supervisor
David Bauner
Commissioner
Vattenfall AB
Contact person
Tobias Nylander Abstract
The construction of wind power is strongly associated with negative local externalities in terms of noise, shadows, visual impact and effects on local environment. To compensate for these negative effects, wind developers seek to find methods to create more local value. The purpose of this study has thus been to identify and evaluate potential methods to increase the local value creation from onshore wind power projects, mainly in the Swedish context.
Firstly, from the literature review and interviews, four different approaches to create local value has been identified. The four identified methods to create local value are (1) community funds, (2) local ownership, (3) modernization and (4) local content. Of these identified methods, local content is deemed to have most potential in creating local value.
The use of local content has generally been strong in the UK, why this study comprises a field study that reveal that British wind developers manage to appoint up to 20-30 % of the total capital expenditure to local companies. Compared to Sweden, the same number is as little as one percent. The explanation to this significant difference could be explained by “three I:s”,, namely, (1) Identification of local companies, (2) Information of opportunities and requirements and (3) Incentive creation for the main contractors.
For Swedish developers to overcome the problems associated with the three I:s, the study presents seven activities that are possible to implement directly into the development process of wind power. Together these activities create a comprehensive way of enhancing the possibility for local companies to be involved in the construction phase, and by doing so, increasing the local value creation.
Key words
wind power, local value creation, local content, local procurement
4
Exekutiv sammanfattning
Denna studie är genomförd som ett examensarbete på KTH inom institutionen för Industriell ekonomi och organisation. Studien vänder sig i första hand till vindkraftsutvecklare och i andra hand till kommuner, närboende och entreprenörer verksamma inom vindkraft.
Det övergripande syftet med studien har varit att identifiera och analysera hur vindkraftsutvecklare kan skapa mer lokalt värde och därigenom utveckla närområdet i samband med nya vindetableringar. Definitionen av lokalt har primärt varit inom kommungränsen för etableringen men hänsyn har även tagits till regionalt värdeskapande vilket inkluderar angränsande kommuner.
Metoder för att skapa lokalt värde från vindkraft
Studien visar på fyra olika metoder för att skapa lokalt värde, nämligen (1) ”bygdepeng”, (2) lokalt delägarskap, (3) modernisering och (4) lokalt innehåll i processen. Av dessa är användandet av ”bygdepeng” den klart mest använda metoden av Svenska utvecklare.
”Bygdepeng” innebär att utvecklaren avsätter en del av intäkterna (vanligen 0,1-1 procent) till närliggande föreningar eller byar. Lokalt delägarskap innebär att närboende och/eller kommun erbjuds möjlighet att investera i hela eller delar av vindkraftsturbiner och på så sätt få del av intäkterna. Modernisering definieras som aktiviteter i samband med uppförandet av vindparken som utvecklar bygden. Exempel på detta kan vara upprustning av vägar alternativt att möjliggöra för uppkoppling till den fiberkabel som ansluts till vindparken för fjärrstyrning.
Tanken är att man utnyttjar de resurser som används för konstruktionen vindparken för att åstadkomma positiva åtgärder för området i allmänhet. Sista värdeskapande metoden är involverandet av lokala företag och resurser i samband med vindkraftsetableringen vilket inom forskarvärlden benämns ”local content”.
Analysen av dessa fyra metoder visar att ”bygdepeng” och lokalt delägarskap inte optimala ur ett värdeskapande perspektiv. Problem som är associerade med ”bygdepeng” är bland annat svårigheter att definiera vilka som har rätt till ersättning samt hur fonden skall administreras.
Vidare kan ”bygdepengen” framstå som en muta, vilket vissa utvecklare framhåller som problematiskt. Lokalt delägarskap är en metod som visat sig framgångsrik i bland annat Danmark, där 20 % av vindkraftsparker enligt lag skall erbjudas till lokala aktörer. Svensk lagstiftning har tyvärr visat sig minde anpassad för lokala delägare, framför allt skattesystemet som är ofördelaktigt för små aktörer med liten egen förbrukning. Eftersom Sverige dessutom upplever historiskt låga elpriser och låga ersättningsnivåer på elcertifikat är marknadsrisken betydande för små aktörer och de potentiella vinsterna i dagsläget små. Summerat innebär detta att varken ”bygdepeng” eller lokalt delägarskap bidrar till signifikant värdeskapande lokalt och studien förordar därför andra lösningar.
Modernisering har visat sig kunna bringa värde till lokalbefolkning i ett antal projekt i Sverige. Både exempel på vägupprustningar och en möjlighet att installera fiberkabel finns identifierat i Sverige. Fördelar med modernisering är att det dels ger långvarigt värde i området och att det är i linje med utvecklarens normala verksamhet. En identifierad framgångsfaktor är ett nära samarbete mellan vindutvecklare och kommun, vilket bör leda till samfinansiering av moderniserande initiativ. Detta medför en reell upprustning som är klart
5 mer ekonomiskt än om kommunen hade finansierat själv. Ett förslag till kommuntjänstemän inom vindkraft är således att se över dessa möjligheter i samband med nya etableringar och agera proaktivt snarare än reaktivt. En nackdel med modernisering är att metoden är situationsberoende vilket innebär att det är långt ifrån alla projekt som innehåller områden som lämpar sig för denna typ av samarbete. Eftersom modernisering generellt inte är applicerbart har studien därför inte fokuserat på denna typ av värdeskapande, men noterar att metoden i sig är intressant och bör användas om möjligt i vindkraftsprojekt.
Involvering av lokala företag har störst potential för värdeskapande
Lokalt innehåll har i studien visat sig vara den metod med störst potential eftersom den ekonomiska potentialen för lokala företag är hög samt att eventuella förslag kan appliceras på alla projekt. Studien visar dock att lokalt innehåll inte används i stor utsträckning i Sverige för tillfället. Flera utvecklare har uppmärksammat frågan men ansatserna för att skapa lokalt innehåll förefaller vara slumpartad. En uppskattning av en utvecklare i Sverige visar att mindre än en procent av det totala värdet i vindkraftsprojekt tillfaller lokala aktörer. Om detta jämförs med en fältstudie från Storbritannien, vittnar det om en mycket längre framskriden process där 20-30 procent av totala investeringen tillfaller lokala aktörer.
Ett antal strukturella hinder försvårar involverandet av lokala företag
Undersökningen av den svenska vindkraftsmarknaden har påvisat ett antal strukturella hinder som försvårar involverandet av lokala företag. Vindutvecklare i Sverige använder generellt sig av en eller flera stora huvudentreprenörer av vindkraftsparken. Detta skapar ett beroendeförhållande eftersom dessa då blir ansvarig för den lokala leverantörskedjan.
Problemet som uppstår är att huvudentreprenören inte vet vilka resurser eller förmågor som finns lokalt utan använder sig av tidigare etablerade underleverantörer. Ett annat problem är också att lokala aktörer ofta saknar nödvändiga certifikat eller licenser för att vara anställningsbara som underleverantörer. Således behövs information om vad huvudentreprenörerna har för krav, men också vilka specifika aktiviteter som en vindkraftsbyggnation skapar.
En jämförelse mellan Storbritannien och Sverige visar att brittiska vindutvecklare mer strukturerat arbetar för att inkludera lokala förmågor i huvudentreprenörernas leverantörskedjor. Dessa skillnader har i denna rapport kategoriserats enligt tre I, vilket står för Identifikation, Information och Incitamentsskapande.
Framgångsfaktorer är ”de tre I:na” - Identifikation, Information och Incitament
Identifikation handlar om att i ett tidigt skede finna lokala företag och näringsidkare som på något sätt kan bidra i uppförandet av vindkraft. Detta görs med fördel tillsammans med kommunen som har kunskap om det lokala företagslivet. Framgången i identifieringen kan bero på kommunens involvering vilket innebär att aktiva kommuner kan öka sannolikheten för större lokalt innehåll. För att möjliggöra för lokala näringsidkare att anmäla intresse föreslås också att vindutvecklaren använder sig av formulär på sin hemsida syftande till att skapa en databas med alla identifierade företag och deras förmågor.
Information handlar om att berätta för de identifierade lokala företagen dels vad en vindkraftsetablering innebär men också vilka krav som ställs för att bli en anställningsbar
6 underleverantör. Studien har visat att många lokala företag inte har insikt i vilka verksamheter som behövs vid vindkraftsbyggnation. Därmed inser företagen inte att de faktiskt kan bidra och blir därför passiva i debatten om vindkraftsetableringen. Tiden har visat sig vara en nyckel eftersom informationen innebär att många små företag behöver nyanställa, gå ihop med andra lokala företag eller införskaffa nödvändiga certifikat för att bli anställningsbara.
Förslaget i studien blir således att så tidigt som möjligt, både via möten och informationsmaterial förklara för lokala företag var deras verksamhet passar in i den större konstruktionen. Spridandet av informationen bör vindutvecklaren ansvara för och innehållet bör till viss del komma från en huvudentreprenör.
Incitamentsskapande handlar om att påverka huvudentreprenören att, i de fall där det är möjligt, välja lokala alternativ istället för tidigare kontakter. Detta eftersom involverandet av lokala företag ställer högre krav på huvudentreprenörerna och ökar deras komplexitet. Studien föreslår tre sätt att skapa incitament för att välja lokala alternativ. Först måste huvudentreprenören få åtkomst till den identifierade databasen för att inse vilka lokala förmågor som existerar i området. Vidare bör vindutvecklaren specificera i kontrakten att de förväntar sig att huvudentreprenören arbetar för att maximera involverandet av lokala företag och att detta kommer att vara en grund i utvärderingen. Utöver detta bör en uppföljning göras för att få insikt i vilka aktörer som faktiskt kontrakteras, både för att visa på vikten av frågan men också för att vindutvecklaren skall få reda på hur mycket lokalt värde som skapas.
Föreslagen process bestående av sju specifika aktiviteter
Som ett sista steg i studien föreslås en process bestående av sju specifika aktiviterer som är möjliga att implementera direkt i vindutvecklarens verksamhet. Aktiviteterna utgör tillsammans ett förslag för hur utvecklare kan hantera identifiering och informering av lokala företag samt skapa incitament för huvudentreprenörer. Värt att notera är att både kommuner och huvudentreprenörer också har en roll i att lyckas maximera det lokala värdet som skapas av vindkraft i Sverige.
7
Acknowledgements
We would first of all like to express our gratefulness to Mr. David Bauner (PhD) of the Unit for Industrial Dynamics at the Department for Industrial Economics and Management of the Royal Institute of Technology in Stockholm. His valuable input and feedback, often provided at the break of dawn, has helped us to keep a holistic approach throughout the process, which has lifted the quality and coherence of the thesis to new heights.
Secondly, and equally important, we wish to thank Vattenfall AB in general and Peter Tornberg in particular for giving us the opportunity to perform this study within the organization. Furthermore, a special appreciation goes to our mentor and guiding star Mr. Tobias B.C. Nylander for his irreplaceable and persistent support, guidance and inspiration during the course of the study.
Lastly, and ever so important, we would like to express our deepest and most sincere thankfulness to our families and friends who have been there for us through thick and thin, sickness and health, and happiness and despair during our five years at the Royal Institute of Technology. Without you we would never have made it, and for that you have our eternal gratitude.
Stockholm, June 2014
Daniel Eriksson
Royal Institute of Technology
Marcus Edlund
Royal Institute of Technology
8
Table of Contents
1. Int roduction ... 11
1.1. Probl em definiti on ... 12
1.2. Definition of the term “local” ... 12
1.3. Purpose st at ement ... 13
1.4. Research questi ons ... 14
1.5. Delimi tat ions ... 14
1.6. Thes is outline ... 14
2. M ethodology ... 15
2.1. Research des i gn and dat a col lecti on methods ... 15
2.2. Limi tat ions ... 18
2.3. Refl ection regarding validit y and reli abili t y ... 19
2.4. Refl ection regarding generaliz abilit y and t rans ferabil it y ... 20
3. Theoreti cal fram ework ... 21
3.1. Wind power in a larger cont ext ... 21
3.2. Publ ic opi nions regarding wi nd power ... 22
3.3. Corporat e Com munit y Invol vement (CC I) ... 24
3.4. Local value creation from wind power ... 27
4. R esul ts and di scus sion ... 36
4.1. Di rect econom ic benefit s ... 36
4.2. Moderniz ation ... 39
4.3. Local cont ent ... 41
4.4. The three I:s - s uggest ed sol utions t o identi fi ed probl ems ... 50
4.5. Il lust ration of s uggest ed acti viti es t o impl em ent in process ... 54
5. Conclusi ons ... 57
5.1. Suggestions for further res earch ... 59
6. Bi bliograph y ... 60
9
Table of figures
Figure 1 – Conceptual visualization of the classification of local and regional areas (created by the authors) ... 13 Figure 2 – Conceptual model of the methodology encompassing the three methods and the literature review (created by the authors) ... 18 Figure 3 – Development of public attitudes towards wind power over the different project phases. Denotations of Devine-Wright (2005) without brackets and Wolsink (2007) in brackets. ... 23 Figure 4 – Conceptual model of the topic of Corporate Community Involvement, the two courses of actions and examples of their respective activities (created by the authors) ... 25 Figure 5 – Categorization of local value creation activities, building upon Brown, et al.
(2012), Rygg (2012) and Munday, et al. (2011) (created by the authors)... 28 Figure 6 – Illustration of the toll-gate process used in wind power projects by the case study company. In chronological order, however the length of the different phases does not represent their respective duration (created by the authors)... 42 Figure 7 – Illustration over the main process of wind power procurement showing the three sub-areas (created by the authors) ... 43 Figure 8 – Conceptual picture with the three I:s including actors and main purposes (created by the authors) ... 53 Figure 9 – Illustration of the suggested process including the seven proposed activities (created by the authors) ... 54
Table of tables
Table 1 – List of interviewed local stakeholders ... 17 Table 2 – Overview of the approved permits in the county of Jämtland showing that community funds are to perceive as a standardized approach for local value creation (Vindkraftscentrum, 2013) ... 36 Table 3 – An examination of the Swedish turbine market in order to evaluate the feasibility for local content ... 44 Table 4 – Summary of the three sub-processes and their part of CAPEX, potential for increased local content and comments ... 46 Table 5 – Illustration of the differences between Sweden and the UK using the presented three I:s ... 49 Table 6 – Overview of the suggested activities ... 58
10
Abbreviations
BoP Balance of Plants
CAPEX Capital expenditure
CCI Corporate Community Involvement
CPR Common Pool Resource
CSR Corporate Social Responsibility
FTE Full-time equivalent
LTS Large Technical System
NIMBY Not-In-My-Back-Yard
O&M Operations and maintenance
PIMBY Please-In-My-Back-Yard
SME Small-Medium Enterprises
TWh Terawatt hour
11
1. Introduction
The world’s energy consumption is constantly rising and the demand is expected to continue to increase by 1.6 % annually until 2030 (IEA, 2013a; Makridis, 2013). Simultaneously, despite the vast consensus regarding their negative impact on the climate (IPCC, 2011; Doran
& Zimmerman, 2009; Oreskes, 2004), fossil fuels continue to be the dominant source of energy (IEA, 2013b). In order to reduce the emissions of CO2 and other greenhouse gases into the atmosphere, an increased use of renewable energy technologies is essential (Panwar, et al., 2011). One of the most promising and cost-competitive alternatives to fossil fuel based energy available today is onshore wind (IEA, 2013a; Söderholm, et al., 2007), which has also been the most successful technology of the last decade (Haas, et al., 2011).
The development of wind power globally has surged in recent years and the total installed capacity rose from 31 gigawatts in 2002 to ca 283 gigawatts at the end of 2012. Today, wind power generated electricity constitutes more than 7 % of the electricity supply in Europe and wind power accounted for 26 % of the new capacity installed during 2012. (GWEC, 2013) In Sweden, where the proportion of renewables in the energy mix has been historically large due to the country’s good conditions for hydro power, the trend has been similar. Driven by a number of factors, including relatively high electricity prices and the introduction of the electricity certificate scheme in 2003, the wind power segment is constantly growing (Bergek
& Jacobsson, 2010; Michanek & Söderholm, 2006). During 2013, more than 300 wind turbines were installed in Sweden with a combined capacity of almost 800 megawatts, resulting in a total national capacity of 4,5 gigawatts (Swedish Wind Power, 2014; Swedish Energy Agency, 2013b). In total, Swedish wind turbines generated approximately 9.9 TWh of electricity during 2013, which represents about 6.6 % of the country’s total production (Swedish Energy Agency, 2014; SCB, 2014). This could be compared with hydro power (60.8 TWh, 37.5%), nuclear power (63.6 TWh, 39.1%) and thermal power (15.2 TWh, 9.4%) (SCB, 2014). However, the expansion of both hydro and nuclear power is currently deemed as politically problematic and in reality, wind power is the most preferred alternative today for electricity generation capacity expansion (Energimyndigheten, 2010).
In Sweden, about 90% of the territory is considered rural and 24 % of the population lives in such areas. It is considered a political challenge how to alleviate urbanization and increase the development in rural areas (Department of Agriculture, 2009). In this, wind power is by many perceived to be a promising way of stimulating rural development as wind farms typically are built in such areas (Phimister & Roberts, 2012; Söderholm, et al., 2007). Unfortunately, as the current standard within the industry overlooks this opportunity, in reality, the value that gets created rurally is considered insufficient (Vindval, 2012). Studies show that even though the GDP of the rural area increases, the effect on household economy is limited (Phimister &
Roberts, 2012). The debate on why this is, and more importantly, how to create more local value for the community is ongoing but no conclusive solution has been presented (Rygg, 2012; Munday, et al., 2011; Brown, et al., 2012).
12 1.1. Problem definition
The construction of a wind farm is strongly associated with negative local externalities in terms of noise, shadows, visual impact and effects on the local environment (Aitken, 2010b;
Meyerhoff, et al., 2010; Wolsink, 2007). As the local communities experience these drawbacks whilst the earnings are harvested by the wind developers and the global society due to the increase of renewables, the problem of social cost arises (Coase, 1960). Coase´s theorem concerns the situation where one part benefits from a certain decision at the expense of someone else’s prosperity. To compensate for such impacts, some of the earnings should be transferred to the afflicted side of the engagement (Coase, 1960). Currently within the wind industry, the most frequently used approach to achieve this is to offer direct economic compensation focused to the adjacent habitants, denoted ‘community funds’ or “financial support for local development” (Aitken, 2010a; Bell, et al., 2005; Munday, et al., 2011).
However, these community funds are problematic as it is difficult to define the boundaries of the affected community as well as to determine what the money should be used for (Bristow, et al., 2012; Gross, 2007). Experiences from previous wind power projects show that the money rarely is used to long-term development of the community, and community funds are thus not an effective way of creating consisting value (Brown, et al., 2012). From a wind developer’s perspective, it is also possible to consider the direct economic compensation as a bribe of local communities, as they are involved in the permit processes (Aitken, 2010a).
Furthermore, community funds are affecting the economic viability of the projects as a certain percentage is dedicated to communities. Hence, both communities and wind developers are searching for new ways of compensating for the externalities that are more beneficial for both parties (Munday, et al., 2011; Baxter, et al., 2013).
The necessity for a well-functioning compensation system is even more actualized in the Swedish context, as the necessary building permit can only be granted under the condition that the local municipal council gives its approval (Swedish Government Reports, 2009).
Studies show that the most common reason for a disapproval of planned wind farms is opposition from local stakeholders (Bergek, 2010; Söderholm, et al., 2007; Wüstenhagen, et al., 2007). Thus, developers need to create significant value for local communities, not only reallocate value because it is fair, but also because the local communities possess political power to stop prospective wind establishments. The core problem is thus how a wind power developer could maximize the local socio-economic benefits without compromising the projects’ economic attractiveness.
1.2. Definition of the term “local”
In order to bring some clarity to the discussion about local value creation, it is necessary to elaborate on the definition of localness. For our purposes, the term “local” primarily refers to the immediate vicinity of a proposed wind farm, in which the residents are affected during the construction and/or the operation of the farm. However, this is not limited to the geographical area where the population is directly affected visually, aurally or otherwise by the turbines.
Rather, it also includes the area where various stakeholders are affected by created employment opportunities, tax revenues, increased purchasing power of customers etc. Even
13 if these socio-economic aspects are far from always limited to the municipality where the wind farm is to be built, we argue that the municipal border represents an appropriate boundary for the “local area”. The underlying assumption is that the decision to approve the building of a wind farm is mainly influenced by the residents of the municipality in which the wind farm is to be built. Ultimately, the decision lies with municipal politicians, acting on behalf of local residents and industry. When a wind farm is to be built within two municipal territories, both of them should be considered to represent the local area, as both of them must give their consent.
Nonetheless, it may also be important to consider a larger reference area. In the “regional”
area, we include the surrounding municipalities and typically refer to the entire county and occasionally an even wider area. We deem a more distinct definition of the term unnecessary for our purposes and conclude that the regional area is where actors are affected in any way by the building of a wind farm. Regional stakeholders are likely primarily affected by the favorable socio-economic effects mentioned above. However, the significance of this for the approval process is not entirely clear and needs to be scrutinized. While some local stakeholders may favor general regional value creation, others might perceive benefits distributed to neighboring municipalities as unfair, since the negative aspects of the project does not affect these actors. A conceptual visualization of the classification of the local and regional areas is shown in Figure 1 below.
Figure 1 – Conceptual visualization of the classification of local and regional areas (created by the authors)
1.3. Purpose statement
The purpose of this thesis is thus to identify and evaluate potential methods to increase the local value creation from onshore wind power projects in Sweden. Specifically, the focus is to assess different strategic approaches for wind developers to implement that maximize the creation of local value whilst maintaining the cost competitiveness of the projects.
14 1.4. Research questions
In order to achieve the objective of this thesis, the following overall question needs to be answered:
How could Swedish wind developers maximize the local value creation deriving from wind power projects, without deteriorating the economic attractiveness of the investments?
In order to answer the overall question, three sub-questions are posed and sought to be answered:
1. What means of creating value for the local communities are currently used by developers in the wind industry?
2. How do the identified approaches affect the communities’ economy and how do they influence the cost competiveness of the project?
3. What aspects of the approaches affect the feasibility for implementation and need to be considered (e.g. legal issues or structural barriers in the industry)?
1.5. Delimitations
This paper will primarily be studying local value creation in Sweden and thus omit possible approaches not suitable for Swedish market conditions. Furthermore, as local value creation is highly dependent on the definition of local, the paper will solely use the definition of the term presented above. Consequently, even though increases in national value creation constitute a highly interesting area within the topic, such aspects will be neglected and the focus will be on the adjacent communities. The underlying rationale for excluding the national perspective is that the community has to be affected by the externalities in order to qualify for compensation, in accordance with Coase (1960).
1.6. Thesis outline
In the next chapter, the chosen methodology is presented. This section covers the approach and the selected methods to obtain the data needed for the analysis. In Chapter 3, a review of the literature conducted and presented in order to build the theoretical framework for the study. Public opinions regarding wind power development are explored as well as the field of corporate community involvement. Furthermore, the various tools for local value creation are examined before a more exhaustive elaboration on theories specifically concerning local content is conducted. The results and discussion of the study are presented in chapter 4, which is followed by the conclusions (chapter 5). The paper concludes with suggested areas for further research within the topic of local value creation, including issues that have been excluded from this study.
15
2. Methodology
The foundation of this thesis is a case study, conducted at one of Europe’s largest energy utilities, which is also one of the largest developers and operators of wind power in Europe.
This method was chosen in order to explore the phenomenon of local value creation in wind power in its natural setting and to enable a more in-depth understanding of the problem. The study could arguably be considered opportunistic as the availability of data from the studied company was unique (Collis & Hussey, 2009). However, the approach could somewhat be seen as a preparation phase for action research due to the implementation phase that will follow the study. Due to the prerequisites underpinning this study it is considered an inductive approach, moving from general to specific knowledge (Collis & Hussey, 2009).
The study was primarily based on qualitative data collection methods which were multifaceted, as Eisenhart (1989) suggested that greatest results are obtained by a combination of several qualitative methods. In seeking to achieve generalizability, the study comprised a large sample of interviews which were the base for the data collection. The interviews were then analyzed trough coding, in order to assess the benefits and find the parameters that could be viewed as drivers for local value creation. As the study was not based on a hypothesis, the nature of the outcomes was not clear at the research design stage. Therefore, the methodology is categorized as exploratory and highly iterative (Collis & Hussey, 2009).
2.1. Research design and data collection methods
The study was divided into four different parts, including the literature review that was conducted and iterated throughout the project. The purpose of the research design was to ensure a high validity and a high number of interviews were thus prioritized. Furthermore, as triangulation would increase the validity and reliability of the study, the research design also included a field study. The process could be described chronological as follows even though the process in reality should be viewed as an iterative one:
1. An assessment of current local value creation processes among Swedish wind developers was done, in order to understand the present situation. The data was collected through semi-structured interviews to enable elaboration and exploration of all possible processes.
Seven of the largest wind developers were chosen due to the sizes of their historical projects, of which five chose to participate.
2. Correspondingly, local stakeholders’ opinions towards both completed wind farms and the created local value were assessed. The data was collected through semi-structured interviews to allow the interviewee to elaborate on their answers. Examples of stakeholders were local politicians, nearby habitants, officials from the municipalities and industry experts from NGOs. The aim was to explore how the current approaches within the industry are perceived and compare this to the views of the wind developers. Also, identified possible process changes were subsequently discussed in order to assess how these would be received by local stakeholders.
3. The research design also encompassed a field study of a state-of-the-art project built by the case study company in the UK, in which local value creation through the use of local
16 content has been a high priority. The field study included a workshop to conduct interviews and collect opinions from experts in the field of local content as well as a thorough investigation of all documentation related to the project. Thus, the data collection was both qualitative and quantitative. This approach enabled in-depth identification of best practices within procurement as well as communication towards local communities and industry. The findings were then analyzed in order to identify specific activities aimed at increasing the use of local content and to evaluate the transferability of these activities to Swedish conditions.
2.1.1. Assessm ent of value cr eati on am ong Sw edish wind devel opers
The first step was to assess historical initiatives regarding local value creation in Sweden. To accomplish this, all of the major Swedish wind power developers were approached. After investigating the market in Sweden, seven developers were contacted. The selection criterion used was that the developer had constructed at least one wind farm in Sweden. Having constructed a wind farm, rather than merely acting as the operator of one, indicates experience of having to overcome local opposition and thereby experience of initiatives to local value creation. Of the seven contacted developers, five accepted to be interviewed and take part in the study, which meant a response rate of 71 %.
The data collection method was semi-structured telephone interviews to enable the interviewees to elaborate and answer freely within the topic. At least one interview for approximately one hour was conducted per developer, but in some cases there were one or more follow-up interviews. All answers were documented by an auditor while another person asked the questions. This approach enabled full focus both on the questionnaire with follow- up questions as well as on transcribing the answers.
As the study itself was conducted at a wind developer, more interviews were held with employees of the company in order to fully understand the process and complexity of a wind developer. A total of 15 interviews were held with employees from different parts of the organization.
2.1.2. Assessm ent of th e local stakeh older s ’ opinions
The second step was to assess and analyze how the local stakeholders perceived and valued the initiatives aimed at creating local value. A total of 10 semi-structured telephone interviews were conducted with politicians, local municipality officials and other relevant persons with knowledge about local value creation within the wind power industry. Another 3 interviews were conducted with nearby habitants or other persons directly affected by wind power developments, either face-to-face or by telephone, as seen in Table 1. The condition was that the interviewees had experienced a wind establishment in their immediate vicinity to be eligible for the study. This qualitative approach aimed to identify the personal opinions made semi-structured interviews appropriate as it gave the interviewee an opportunity to further elaborate beyond the structure of the prepared questions (Collis & Hussey, 2009).
This sample reflects the different opinions regarding what parameters is deemed most important concerning local value creation. The answers were coded in order to find common
17 themes as well as municipal-specific opinions. The interviews enabled the identification of what initiatives identified in the literature and the industry overview that is considered most effective in creating local value.
Table 1 – List of interviewed local stakeholders
Occupation/Organization Abbreviation Origin and comments
Politician
P1 Åsele, 1 large wind farm
P2 Bräcke, Pioneers in wind power in Sweden P3 Vaggeryd, Strong political opposition P4 Dorotea, Many ongoing projects Municipality official
MO1 Östersund, Strong local opposition MO2 Åre, Several rejected permit applications MO3 Sorselse/Storuman, 1 large wind farm Wind power associations WPA1 Hela Sverige ska leva
WPA2 Nätverket för vindbruk Local habitants
LH1 Åsele municipality
LH2 Sorsele municipality
LH3 Åre municipality
University professor Prof1 Chalmers, Gothenburg
2.1.3. Field study to ben ch m ark best-practi ce proces ses
The last part of the study was to identify current best-practice processes using a benchmark field study and analyzing what implications this would have on Swedish conditions. This was done by visiting an ongoing wind project being built in southern UK named Pen y Cymoedd, in which considerable focus has been laid on maximizing the local value creation by the use of local content in the process.
The field study comprised several different data collection methods. Firstly, semi-structured interviews were conducted by telephone with responsible employees from the developer to understand their process and considerations. Secondly, a visit to the UK and the head office of the developer was done. During one full day, a workshop was conducted in order to gather as much information as possible regarding the processes and the internal considerations made.
The final part was to go through all documentation regarding both communication and procurement practices to find initiatives that might be applicable to Swedish conditions. This multifaceted approach was chosen as it is suggested by Eisenhart (1989) to use interviews, workshops and documentation when collecting information in a case-study.
The aim of the methodology was to identify and assess the most suitable changes to Swedish development standards in order to increase the level of local content, and thereby the local value creation. This was done with consideration to both level of localness, economic impact and practical implications of the project. A conceptual illustration of the approach is shown in Figure 2, which displays how the three different methods together with the literature review lead to the final conclusions and recommendations.
18
Figure 2 – Conceptual model of the methodology encompassing the three methods and the literature review (created by the authors)
Finally, the identified conclusions and recommendations that are both possible and encouraged by local stakeholders will be applied on a new project at the case study company.
The result of the introduced changes will however not be included in the results, as the time frame of wind power projects are considerably longer than the scope of this project. The academic contribution will therefore be the suggested changes aimed at increasing the local value creation whilst maintaining the cost competitiveness of the projects. The practical result will hopefully be a more mutually beneficial process, leading to an increased feasibility for developers to obtain building permits and ultimately an expansion of renewable energy generation in Sweden.
2.2. Limitations
This subsection will cover a discussion regarding the scope of the paper and approach some of the shortcomings that may have affected the outcomes. The main areas that could be considered limitations are mainly time and resource constraints which for instance affected sample sizes. Below are the identified limitations of the research presented and discussed as well as their implications for the study:
Time constraints have limited the scope and result of the project as the project had a pre-set deadline. The quality and validity of the study would possibly have been
19 enhanced by (for instance) larger samples, a more thorough literature review or a higher number of benchmarking field studies. This limitation was minimized by a comprehensive plan and continuous hard work to maximize the output.
As mentioned above, sample sizes are a limitation as the validity and generalizability is affected by the quantity of empirical data on which the study is based. The samples could be considered small according to the population, except regarding the Swedish wind developers. Also, there were some difficulties in finding the right interviewees and the responses are to some extent a result of the interviewees’ availability and willingness to participate in the study. This limitation was minimized by careful sample selection, and making sure that each interviewee had sufficient time allocated for the interviews.
Respondents in the study may be affected by recent projects, and own personal biases might affects how they respond to the questions. The collected answers contain both outspoken opinions as well as unconscious subjectivity which limit the extent of both validity and generalizability. By using semi-structured interviews and follow-up questions, the aim has been to get the interviewees to further elaborate their answers and thereby trying to reduce any biases.
2.3. Reflection regarding validity and reliability
The validity in this study is considered high due to triangulation of three different data collection methods. The opinions of developers, local stakeholders and the experiences from the UK field study are considered to collectively exhaust the phenomenon of local content in wind power development. Naturally, the number of interviews will affect the validity of the study. As the study of the current practices among Swedish wind developers is based on interviews with a large share of the major wind developers in Sweden, the validity of these findings could be considered high. However, the number of interviewed local stakeholders is few compared to the total population of officials, politicians and local residents. This may reduce the study’s validity as another researcher could interview a whole set of new stakeholders that might look upon wind power differently. Wind power is an area exposed to personal biases and beliefs, and another sample of local stakeholders may therefore generate completely different answers. Also, as semi-structured interviews were conducted, it gave further room for off-topic discussions hence affecting the validity negatively.
Furthermore, if another country was selected for a field study, the outcome may be affected as the conditions and operations regarding local content could be very country specific.
Therefore, a higher validity could have been reached if this study included more field studies from other countries. However, such an approach would be too time consuming and this study was therefore limited to only one. Given this time constraint, we argue that choosing a best- practice project is an appropriate approach as the processes and initiatives incorporated in this project have been evaluated and verified in an actual setting. Therefore, it is believed that the quality and effectiveness of the identified activities to implement in the Swedish wind power process possess a high validity.
20 The reliability of the study could definitely be increased if the number of interviews were higher, especially regarding local stakeholders. More interviews would increase the number of opinions and enable this study to encompass a larger number of perspectives. Another factor affecting the reliability is the use of semi-structured interviews, which encourages the interviewee to elaborate further than the questions. If the interviews were complemented with a questionnaire survey, which actually was discussed as a method to assess the opinions of local stakeholders, the reliability may have been higher. Thus, the chosen method gives a lower reliability, but the weighting was to more accurate capture the phenomenon and opinions regarding local content which also is suggested by Collis & Hussey (2009). Hence, the argument underlying this choice is that a qualitative approach more truly captures the complex conditions of wind power even though a quantitative approach would be easier to recreate.
2.4. Reflection regarding generalizability and transferability
The generalizability of this study is considered high as it comprises both Swedish and the UK conditions, which increase the probability for national transferability. Even though the national permit processes differ somewhat, the comparison between the UK and Swedish communities show that they are alike in structure and capabilities. Therefore it is deemed likely that the suggested process is possible to transfer to other European countries. The criterion for transferability is a similar tender process, as the relationship between the main contractor, the developer and the local community is decisive for the suggested process to be applicable. Thus, the generalizability from one nation to another within wind power is deemed high and the recommended process should be possible to implement without any major changes.
The findings of this paper could also be used in other industries, as the problem definition is based on Coase´s (1960) theorem comprising externalities and the compensation for harvesting local resources. We argue that the three I:s framework presented in this paper also could be transferred to industries such as oil, mining, infrastructure etc. in order to increase the local value creation and contribute to rural development. The applicability of the conclusions in this paper however depends on to which extent the development requires main contractors. Yet, a brief overview of the similar industries indicates that the use of better information, identification and incentive creation could increase the local value creation.
Thus, when harvesting local resources with adjacent habitants, we deem it possible to use some of the findings in this study.
21
3. Theoretical framework
As presented in chapter 1, the topic of this paper is local value creation within the Swedish wind power industry. In the following chapter, previous research and current trends within the area are identified and elaborated on, deriving from an extensive review of the literature.
This assessment is then used in order to construct the theoretical framework underpinning the conducted study. Firstly, as wind power is a subset of a larger system, the chapter will commence with a system approach followed by a more detailed literature review regarding public opinions towards wind power, Corporate Community Involvement (CCI) and the topic of local value creation in general and local content in particular.
3.1. Wind pow er in a larger context
Wind power is one of many methods for electricity generation, which in research traditionally is regarded as a large technical system (LTS) (Markard & Truffer, 2006). The idea of LTS origins from Hughes (1987) and encompasses for instance roads, railway and the electrical grid. Modern life as we know it is dependent on these systems and in general, LTS is characterized by strong path dependency and high barriers for radical innovations (Hughes, 1987). Historically, the development of LTS has in many cases contributed to the development of the rural surroundings (Marwell & McQuarrie, 2013). In Sweden, this is proved in Karlström´s (1985) review of the Swedish industrialization. The findings presented were that the success of the rural development was dependent on how much the system and socio-economic factors were aligned. This could be applied to the present expansion of wind power as many studies indicate that this could be a way for rural areas to re-industrialize and develop (Zerriffi & Wilson, 2010; Phimister & Roberts, 2012; Slattery, et al., 2011).
In Sweden, the path dependency in electrical generation generally has been strong (Sandsmark & Tennbakk, 2010). Today, the electricity generation in Sweden consists mainly of nuclear and hydro power that together amount for 82 % of the produced electricity (Energimyndigheten, 2013). However, the expansion of both hydro and nuclear is currently considered as politically difficult and in reality, wind power is one of few realistic alternatives today (Swedish Government Office, 2014). Thus, if the Swedish generation system is to expand, a development of wind power is needed.
Unfortunately, however, a building permit for a wind farm can only be granted under the condition that the local municipal council gives its approval (Swedish Government Reports, 2009). This so called “municipal veto right” frequently prevents the realization of wind power projects and has been subject for criticism from politicians and trade organizations for its hampering effect on the wind power development in Sweden (Nordin, et al., 2013; Pettersson, et al., 2010; Swedish Wind Energy, 2010). Using Hughes (1987) denotations, this could be perceived as a reversed salient, as it hampers the incremental development of the LTS that is the Swedish electricity generation system. Municipalities’ disapproval of planned wind farms often originates in opposition from local stakeholders, primarily related to concerns about noise and visual impact of the landscape (Bergek, 2010; Söderholm, et al., 2007; Aitken, 2010b; Meyerhoff, et al., 2010). Gaining community acceptance, defined by Wüstenhagen et al (2007, p. 2685) as “the specific acceptance of siting decisions and renewable energy
22 projects by local stakeholders, particularly residents and local authorities”, is thus crucial for wind power developers in order to overcome the reversed salient. Therefore, the next chapter will discuss the nature of the public opinions and their effect on the development of the electrical LTS.
3.2. Public opinion s regarding w ind power
Wind power is often perceived as one of the most promising renewable energy sources in reducing the need for fossil fuel based electricity generation (IEA, 2013a; Söderholm, et al., 2007). At the same time, a rather simplistic view that the technology causes global and national advantages but local disadvantages is quite widespread (Rygg, 2012). According to several researchers, including Aitken (2010a), Bell et al., (2005), Wolsink (2007) and Ladenburg (2008), numerous polls and surveys show high public support for the wind power technology in general. However, the attitude on a local level towards building new onshore wind turbines and replacing old ones, so called repowering, is not always as positive (Wüstenhagen, et al., 2007). This opinion discrepancy is for example acknowledged by Krohn
& Damborg (1999) and Waldo (2012) as well as Söderholm et al. (2007), who conclude that despite the occurrence of local opposition in Sweden, experience implies that people generally demonstrate a positive attitude towards wind power. Similarly, Toke (2002) found that nearly 80 percent of the population in the UK supports wind power, but that only one quarter of the planned wind power installation is commissioned. In the German state of Brandenburg, more than 27,000 residents signed a petition in 2009 to ban the construction of new wind turbines closer than 1,500 meters to any residential area (Meyerhoff, et al., 2010). The consequence of such public behavior, Wolsink (2007) argues, is ultimately a reduced potential for the countries to reach the national renewable energy targets.
Interestingly enough, and somewhat contradictory, opposing arguments regarding wind power often concerns ecological issues and natural values (Waldo, 2012). This irony is acknowledged by Söderholm et al. (2007, p. 396), who notes that “While the general public, national and global energy policies, and indeed many scientists all point out wind power as particularly environmentally friendly, most of the objections to its expansion tend to have environmental origins”. Primarily, the visual impact on the landscape is argued to be the most important issue (Eltham, et al., 2008; Wolsink, 2007; Jobert, et al., 2007) and large farms with several wind mills are consequently perceived more negatively than few or solitary turbines (Devine-Wright, 2005; Thayer & Freeman, 1987). Another aspect influencing public opinions is noise from the rotor blades and the turbines (Waldo, 2012; Eltham, et al., 2008; Oerlemans, et al., 2007). The actual experienced noise disturbance from a wind farm has however been found to correlate more to the person’s opinion of the visual impact of the turbines (Pederson
& Persson Waye, 2004). Other commonly used arguments in these discussions are possible negative effects on the fauna and biodiversity, particularly bird life (Meyerhoff, et al., 2010;
Solli, 2010; Ladenburg, 2008), land and property values (Söderholm, et al., 2007) and the recreational attractiveness of the area (Frantál & Kunc, 2011; Brittan Jr., 2001).
Evidentially, wind power involves a number of different aspects to consider and planning a wind farm requires developers to cope with differing opinions and concerns of various stakeholders. It has been showed that public attitudes towards wind power on a local level
23 may differ between the different phases of a specific project. Both Wolsink (2007) and Devine-Wright (2005) found a V-shaped development of the level of local acceptance, as seen in Figure 3 below. As noted above, the attitudes towards wind power are high on a general level, when no local schemes have been presented. As can be seen in Figure 3, the level of acceptance is lowest during the construction phase, where concrete plans are presented and publicly debated (Wolsink, 2007). Eventually, when the project is finalized, people adopt a more positive view (Devine-Wright, 2005), and according to Wolsink (2007) sometimes even higher than pre-project levels.
Figure 3 – Development of public attitudes towards wind power over the different project phases. Denotations of Devine- Wright (2005) without brackets and Wolsink (2007) in brackets.
A frequently used explanation to local opposition towards nearby application of wind power is the so called Not-In-My-Back-Yard (NIMBY) syndrome (Burningham, 2000; Wolsink, 2000; van der Horst, 2007). The traditional view of the NIMBY-concept is that people generally support wind power, but are strongly opposed to the building of turbines in the vicinity of their homes (Wolsink, 2000). However, several researchers claim this attribution to selfish motives to be too simplistic (Aitken, 2010b; Wolsink, 2007; Jobert, et al., 2007). The assumption that such behavior could be derived from an attitude-behavior gap is argued by Bell et al. (2005) to be inaccurate. Similarly, Aitken (2010b) means that a dichotomous categorization of supporters and objectors are of little explanatory use and argues that both positive and negative opinions regarding wind power are legit and should be treated as valid arguments in every individual situation. The literature concerning ‘NIMBY-ism’ seems to be in agreement that few affected habitants will object wind power no matter what, and that it rather is a matter of information (Aitken, 2010b), participatory (Wolsink, 2000) and fairness (Wolsink, 2007; Jobert, et al., 2007).
3.2.1. Im plicati ons for win d devel opers
So, it appears that developers who encounter local resistance against a wind farm should not focus on refuting opposing arguments. Rather than assuming that opponents are misinformed
