Saku Haiko Development of Scope 3 emissions of Finnish multinational manufacturing organizations A configurational perspective Vaasa 2025 School of Technology and Innovations Master’s thesis in Industrial Systems Analytics Master of Science in Technology 2 UNIVERSITY OF VAASA School of Technology and Innovations Author: Saku Haiko Title of the thesis: Development of Scope 3 emissions of Finnish multinational manu- facturing organizations: A configurational perspective Degree: Master of Science in Technology Discipline: Industrial Systems Analytics Supervisors: Jouni K Juntunen, Khuram Shahzad Year: 2025 Pages: 66 ABSTRACT: Climate change and sustainability have increasingly shaped the operational model of multina- tional enterprises (MNEs), particularly within the manufacturing sector. One of the major factors has been carbon dioxide (CO2) emissions, the monitoring and reduction of which have been sig- nificantly tightened. Most recently, the focus has been on indirect Scope 3 emissions, which originate in the upstream and downstream parts of a company’s value chain. Therefore, they account for the majority of the company’s total CO2 emissions. Despite their importance, Scope 3 emissions remain complicated and inconsistently reported. This study addresses this gap by examining the conditions under which Finnish multinational manufacturing companies have suc- ceeded or failed in reducing their Scope 3 emissions. This study investigates the development of Scope 3 emissions of Finnish multinational manufac- turing companies using a configurational method. This research focuses on how firm-level con- ditions such as financial slack resources (revenue and profitability), R&D intensity and interna- tionalization relate to Scope 3 emission reductions. The aim of the study is to identify pathways at the firm-level conditions that lead to the reduction of Scope 3 emissions. The focus is also on examining how the interaction between the selected conditions influences the change in emis- sions. The empirical phase is performed using the fsQCA analysis method, which enables the identifi- cation of multiple configurational causal pathways. Data is collected from 19 Finnish manufac- turing firms over a five-year period (2019–2023) through publicly available annual and sustain- ability reports, focusing on five key variables: revenue, adjusted profitability, R&D intensity, de- gree of internationalization, and changes in Scope 3 emissions. The study implements these con- ditions using fuzzy calibration and constructs truth tables to identify sufficient and necessary configurations for emission reduction. This study identifies configurations of conditions that lead to both the reduction and non-reduc- tion of Scope 3 emissions, highlighting the asymmetric and conjunctural nature of causal rela- tionships. The results indicate that no single condition is universally sufficient to explain reduc- tions in Scope 3 emissions. The results show the complexity of Scope 3 emissions and that the selected conditions were not able to consistently reflect which pathways lead to emission re- ductions. Instead, the results highlight the importance of the company’s environmental strategy, which was enabled by the selected conditions. Theoretical findings imply practical implications for firms seeking to align environmental performance with strategic decision-making. KEYWORDS: Scope 3 emissions, sustainability, configurational methods, manufacturing 3 VAASAN YLIOPISTO Tekniikan ja innovaatiojohtamisen yksikkö Tekijä: Saku Haiko Tutkielman nimi: Suomalaisten monikansallisten valmistusorganisaatioiden Scope 3 - päästöjen kehitys: Konfiguraationäkökulma Tutkinto: Diplomi-insinööri Oppiaine: Industrial Systems Analytics Työn ohjaajat: Jouni K Juntunen, Khuram Shahzad Valmistumisvuosi: 2025 Sivumäärä: 66 TIIVISTELMÄ: Ilmastonmuutos ja kestävyys ovat yhä enemmän muokanneet monikansallisten yritysten toi- mintamallia erityisesti valmistussektorilla. Yksi suurimmista tekijöistä on ollut hiilidioksidipääs- töt, joiden seurantaa ja vähentämistä on tiukennettu merkittävästi. Viime aikoina painopiste on ollut epäsuorissa Scope 3 -päästöissä, jotka ovat peräisin yrityksen arvoketjun ylä- ja loppu- päästä. Siksi ne muodostavat suurimman osan yrityksen hiilidioksidipäästöistä. Tärkeydestään huolimatta Scope 3 -päästöt ovat edelleen monimutkaisia ja niistä raportoidaan epäjohdonmu- kaisesti. Tämä tutkimus korjaa tätä aukkoa tutkimalla olosuhteita, joissa suomalaiset monikan- salliset tuotantoyritykset ovat onnistuneet tai epäonnistuneet vähentämään Scope 3 -päästö- jään. Tässä tutkimuksessa tarkastellaan Scope 3 -päästöjen kehitystä suomalaisissa monikansallisissa tuotantoyrityksissä soveltamalla konfiguraatiomenetelmää. Tutkimus keskittyy siihen, miten yri- tystason olosuhteet, kuten taloudelliset resurssit (liikevaihto ja kannattavuus), T&K-intensiteetti ja kansainvälistyminen liittyvät Scope 3 -päästövähennyksiin. Tutkimuksen tavoitteena on tun- nistaa yritystason olosuhteissa polkuja, jotka johtavat Scope 3 -päästöjen vähentämiseen. Pai- nopisteenä on myös se, kuinka valittujen olosuhteiden välinen vuorovaikutus vaikuttaa päästö- jen muutokseen. Empiirinen vaihe suoritetaan fsQCA-analyysimenetelmällä, jonka avulla voidaan tunnistaa useita konfiguraatiollisia kausaalipolkuja. Tietoja kerätään 19 suomalaisesta teollisuusyrityksestä vii- den vuoden ajalta (2019–2023) julkisesti saatavista vuosi- ja vastuullisuusraporteista, joissa kes- kitytään viiteen keskeiseen muuttujaan: liikevaihto, oikaistu kannattavuus, T&K-intensiteetti, kansainvälistymisaste ja muutokset Scope 3 -päästöissä. Tutkimus toteuttaa nämä olosuhteet käyttämällä sumeaa kalibrointia ja rakentaa totuustaulukoita, jotta voidaan tunnistaa riittävät ja tarpeelliset konfiguraatiot päästöjen vähentämiseksi. Tutkimuksessa tunnistetaan olosuhteiden konfiguraatioita, jotka johtavat sekä Scope 3 -päästö- jen vähenemiseen, että vähenemättömyyteen, korostaen kausaalisten suhteiden epäsymmet- ristä ja yhdistelmäluonnetta. Tulokset osoittavat, ettei mikään yksittäinen ehto ole yleisesti riit- tävä selittämään Scope 3 -päästöjen vähenemistä. Tulokset tuovat esiin Scope 3 -päästöjen mo- nimutkaisuuden ja sen, etteivät valitut ehdot pystyneet johdonmukaisesti kuvaamaan, mitkä po- lut johtavat päästövähennyksiin. Sen sijaan tulokset korostavat yrityksen ympäristöstrategian merkitystä, jonka valitut ehdot mahdollistivat. Teoreettiset havainnot sisältävät käytännön im- plikaatioita yrityksille, jotka pyrkivät yhdistämään ympäristösuorituskyvyn strategiseen päätök- sentekoon. Avainsanat: Scope 3 -päästöt, kestävyys, konfiguraatio menetelmät, teollisuus 4 Contents 1 Introduction 7 1.1 Research problem, objectives, and question 8 1.2 Structure 10 2 Literature review 12 2.1 Scope 3 emissions as a challenge for companies 12 2.2 Introduction to sustainability and internationalization 14 2.2.1 Relationship between internationalization and sustainability 19 2.3 Impact of climate change on multinational enterprises 20 2.3.1 Risks and opportunities for MNEs due to climate change 22 2.4 Sustainability-oriented innovation 24 2.4.1 SOI development stages 25 2.4.2 Research and development capability 27 2.4.3 Financial slack and sustainable development 28 2.5 Summary of the theory 29 3 Methodology 31 3.1 Qualitative comparative analysis 31 3.1.1 Characteristics of fuzzy-set QCA 32 3.1.2 Fuzzy sets and their membership scores 33 3.1.3 Calibration of raw data to fuzzy sets 33 3.1.4 Process of establishing a truth table 34 3.1.5 Minimization with Boolean algebra 35 4 Empirical study: Development of Scope 3 emissions 36 4.1 Data collection 36 4.2 Construction of variables 37 4.3 Data calibration 40 5 4.4 Analysis and results 43 4.4.1 Configurations leading to reduced Scope 3 emissions 47 4.4.2 Configurations leading to non-reduced Scope 3 emissions 49 5 Discussion 51 5.1 Theoretical contributions 51 5.2 Limitations 55 5.3 Future research 56 References 58 6 Figures Figure 1. Venn diagram of three dimensions of sustainability. 15 Figure 2. The dimensions of SOI. 25 Tables Table 1. Climate change-related risks. 23 Table 2. Collected data from 19 Finnish industrial manufacturing companies. 37 Table 3. Outcome and conditions. 40 Table 4. Qualitative thresholds of the outcome and conditions. 40 Table 5. Values of fuzzy set calibration. 42 Table 6. Necessity analysis for single conditions. 43 Table 7. Truth table for positive outcome. 45 Table 8. Configurations of reduced Scope 3 emissions. 46 Abbreviations CH4 Methane CO2 Carbon dioxide csQCA Crisp set qualitative comparative analysis EU European Union FSA Firm-specific advantages fsQCA Fuzzy set qualitative comparative analysis GHG Greenhouse gas ICT Information and communication technology MNE Multinational enterprise N2O Nitrous oxide PRI Proportional reduction in inconsistency QCA Qualitative comparative analysis R&D Research and development SOI Sustainable-oriented innovation 7 1 Introduction Climate change has progressed significantly on our planet, and its impacts have been substantial (WWF, 2019). WWF (2019) highlights the effects of climate change, including global warming, greenhouse gas emissions, extreme weather events, rising sea levels, and ecosystem changes, all of which are consequences of climate change. The average temperature of our climate has increased by 1.1 degrees compared to pre-industrial times (WWF, 2019). This is only 0.4 degrees short of the 1.5-degree limit agreed upon in the 2015 Paris Climate Agreement, which must not be exceeded to prevent global warm- ing (Finnish Ministry of the Environment, 2025). The Paris Climate Agreement also set a goal of reducing global greenhouse gas (GHG) emissions (Finnish Ministry of the Envi- ronment, 2025). To achieve this, businesses and various stakeholders are expected to provide innovations and solutions to reduce GHG emissions. These so-called GHGs causes global warming, which leads to harmful effects on the planet. But what exactly are these GHGs? The European Parliament (2023) clearly de- scribes the function of GHGs, explaining that these gases absorb heat from the Earth's surface and retain it in the atmosphere, preventing it from escaping into space and help- ing to maintain the Earth's temperature. Therefore, the increase in GHG resulting from human activities, such as fossil fuels, accelerates the greenhouse effect, thus warming the planet. According to the United States Environmental Protection Agency (2025), GHGs consist of carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and fluori- nated gases. Approximately 80 % of greenhouse gases consist of carbon dioxide, 11 % methane, 6 % nitrous oxide, and the remaining portion fluorinated gases (United States Environmental Protection Agency, 2025). Therefore, carbon dioxide is by far the most common GHG, which is why it is typically mentioned when discussing emissions. The manufacturing industry is one of the largest producers of emissions, and this has been acknowledged for a long time (Jing et al., 2019). Still, the emission levels have not reached the desired targets, and there is a need for necessary actions. The GHG emis- sions of the manufacturing industry consist of different parts of the value chain. 8 According to Jäppinen et al. (2014), 50% of the supply chain GHG emissions in the for- estry industry are from long-distance transportation, wood processing, and refinement operations. Taking this into account, material procurement is one of the largest contrib- utors to emissions in the manufacturing industry, and evidence from China shows that steel is one of the materials whose production generates a massive amount of emissions (Jing et al., 2019). Industrial emissions consist of both upstream and downstream activities, covering all stages from raw material procurement to the final product reaching the end user. The emissions resulting from these activities can be further divided into direct and indirect emissions, with indirect emissions referred to as Scope 3 emissions. These Scope 3 emis- sions cover the majority of a company’s emissions (Schulman et al., 2021). This is why it is important to understand Scope 3 emissions, what they consist of, and how they can be reduced. Internationalization is one of the factors that has increased emissions caused by companies’ value chains (Schulman et al., 2021). These emission challenges and societal pressures, such as regulations and laws, have driven companies toward more sustainable practices. According to the European Council (2024), the goal is to make Europe climate-neutral by 2050, meaning that net green- house gas emissions would be zero by then. To reduce emissions and achieve climate neutrality, industries need sustainable innovations and solutions. This requires invest- ment, capital effort, and knowledge of sustainable practices from companies and other stakeholders, which can be financially and time-consumingly challenging. 1.1 Research problem, objectives, and question Increasingly, sustainability, the maintenance of natural ecosystems, and the reduction of GHG are being highlighted in society and, consequently, in the business world. Compa- nies are aware of upcoming obligations and regulations related to the reduction of GHG emissions and generally support a more sustainable future. However, implementation is 9 still in an ongoing development phase, as GHG emissions are still far from optimal. Thus far, companies have primarily focused on reducing and monitoring emissions caused by their own operations (Scope 1) and emissions from the production of purchased energy (Scope 2) (Emborg et al., 2023). Scope 1 and 2 GHG emissions have been clearly tracked and identified, allowing companies to improve their operations to reduce these emis- sions over a long period of time. However, Scope 1 and 2 emissions typically account for only a small portion of a manu- facturing company’s total emissions, which is why indirect Scope 3 emissions come into play, usually covering at least over 75% of the company’s total emissions (Tian et al., 2025). Making the reduction of Scope 3 emissions an effective way to reduce a com- pany’s overall emissions. However, due to the challenging and complex nature of moni- toring Scope 3 emissions, they have been excluded until the GHG Protocol provided guidelines for tracking Scope 3 emissions and the EU introduced emission targets under various regulations. This is reflected in companies’ Scope 3 emissions, which have started to receive more attention in recent years due to increased transparency of sustainability. Thus, the research problem of this study can thus be considered the development of indirect Scope 3 emissions in manufacturing companies, which have been scarcely stud- ied so far. Therefore, it is interesting to identify the factors that influence the development of Scope 3 emissions when the goal is climate neutrality in Europe. This makes the research interesting, as there is clearly a need for it. This study examines the potentially different circumstances that affect Scope 3 emissions. Setting the first objective to determine un- der what circumstances internationalization, R&D investments or financial slack re- sources influence Scope 3 emissions. These factors are relevant to the research and its methodology, and therefore, the objective narrows the scope of the thesis and provides a clear guideline for achieving the research goal, such as identifying which resources are the most effective ways to enhance sustainability. Evidence has been found from Euro- pean manufacturing companies, indicating that R&D positively impacts emission 10 reductions (Apergis et al., 2013) and is therefore a relevant factor for this study. Accord- ing to Nguyen et al. (2019), high internationalization provides a better opportunity to make strategic decisions related to a company’s resources. These resources can be uti- lized in sustainability efforts that lead to a reduction in Scope 3 emissions. The second objective is to investigate how the interaction of a company’s conditions af- fects emissions from activities that the company indirectly influences within its value chain under Scope 3. This objective helps research explore potential patterns of defined conditions that indirectly affect the value chain. According to previous research, financial slack has influenced the sustainability expenditures of small and medium-sized enter- prises, allowing for further research into the impact of financial slack on sustainability in large companies (Boso et al., 2017). Both research objectives define and support the broader research question, which is as follows: • What configurational analysis of the manufacturing sector can reveal about path- ways to reduced Scope 3 emissions? The research question supports the research objectives of the study and aims to study and analyze the configurations of the manufacturing sector in order to identify pathways and strategies that help reduce Scope 3 emissions. This thesis has one clear research question that helps the reader understand the objectives and the process of the research. Potentially, it will be identified which factors have influenced changes in Scope 3 emis- sions within Finnish companies’ value chains during the observation period. 1.2 Structure The structure of this thesis consists of six chapters. The first chapter consists of an intro- duction to the researched topic. Chapter one covers background information, research problems, the research question, objectives, and the limitations of the study. Chapter two covers the literature review on the researched topic, which is essential for the study 11 and provides a solid foundation for understanding the analysis and solutions of the re- search. The theory in chapter two is based on the conditions and outcome selected in the empirical phase, which are essential for achieving the research objectives. Chapter three discusses the theory of the configurational analysis method used in the study, in- cluding its structure, characteristics, and applicability. This theory is later utilized in the empirical phase of the study. Chapter four is the actual empirical phase, during which the selected conditions and out- come are analyzed using a configurational approach. Chapter four thoroughly presents data collection, the selection of conditions and the outcome, as well as all phases of the analysis, which are crucial to consider in terms of the analysis's reliability. Towards the end of chapter four, the results of the analysis are discussed. In chapter five, the results of the research are discussed in general, considering the theory, and practical perspec- tives are presented. Chapter five also highlights the limitations of the results. The thesis concludes with chapter six, where possible perspectives for future research are dis- cussed, which could lead to new conclusions and solutions. 12 2 Literature review The purpose of this chapter is to introduce the theory used in this study for the reader and to provide a good basis for understanding the theory in terms of the importance and topicality of the study. Chapter two consists of four main theoretical sections, which are further divided into smaller subchapters to comprehensively cover the research objec- tive. First, the reader is provided with an understanding of Scope 3 emissions. The sec- ond part of the chapter discusses the theory of sustainability and introduces the per- spective of internationalization into the discussion. Next, the impact of climate change on multinational enterprises is discussed. Finally, sustainability-oriented innovation will be explained. Overall, the aim is to provide a solid foundation for empirical research and the conclusions drawn from the results. 2.1 Scope 3 emissions as a challenge for companies The GHG Protocol is an internationally used standard for measuring and reporting GHG emissions (Greenhouse Gas Protocol, 2025). The GHG Protocol defines emissions into three different categories, i.e., Scope 1,2 and 3. These represent the emissions generated from the different value chains of the company. Scope 1 emissions consist of the com- pany’s direct emissions, which come from its own operations, such as production (Greenhouse Gas Protocol, 2011). Therefore, it is easy for a company to track and iden- tify Scope 1 emissions. In contrast, Scope 2 and 3 emissions are indirect, with the differ- ence being that Scope 2 emissions come from the company’s own energy consumption, while all other indirect emissions fall under Scope 3 (Greenhouse Gas Protocol, 2011). Indirect emissions therefore consist of upstream and downstream activities. Scope 3 emissions play a major role in a company’s total emissions. According to previ- ous studies, Scope 3 emissions account for over 75% of a company’s total emissions, with sectors such as the food industry reaching over 80% and the retail sector up to 90% (Asif et al., 2022; Schulman et al., 2021). Considering the Scope 3 emissions caused by 13 upstream and downstream activities, the manufacturing sector also has very high emis- sions due to their long-value chain operations. The manufacturing industry is considered one of the largest contributors to GHG emissions (Hertwich & Wood, 2018; Sun et al., 2022). Therefore, it is important to understand which specific activities of a company contribute to GHG emissions. These Scope 3 emissions can be further categorized into different sections based on the company’s upstream and downstream activities they originate from (Greenhouse Gas Protocol, 2013). Upstream emissions are categorized from one to eight, and downstream emissions from nine to fifteen, as outlined below: Upstream Scope 3 emissions 1. Purchased goods and services 2. Capital goods 3. Fuel- and energy related activities 4. Upstream transportation and distribution 5. Waste generated in operations 6. Business travel 7. Employee commuting 8. Upstream leased assets Downstream Scope 3 emissions 9. Downstream transportation and distribution 10. Processing of sold products 11. Use of sold products 12. End of life treatment of sold products 13. Downstream leased assets 14. Franchises 15. Investments (Greenhouse Gas Protocol, 2013). 14 The list details all emissions arising from the value chain, allowing an analysis of how the distribution of Scope 3 emissions is divided. Schulman et al. (2021) highlight the magni- tude of emissions in categories 1, 9, 10, and 11 compared to other categories in the food industry sector. The purchased goods and sold products being the largest sources of emissions. Schmidt et al. (2022) emphasize the challenge of emissions reporting that companies face, particularly in categories 11 and 12, because it is difficult for companies to track these downstream activities themselves. In this case, the calculations of Scope 3 emissions have been inaccurate, leading to the situation where the best reporters have been the worst polluters (Schulman et al., 2021). 2.2 Introduction to sustainability and internationalization The words “sustainability” and “sustainable development” are certainly familiar to many today because of their central operating principles and global climate change (Niemi, 2024, p. 15). But what are the differences between them? Because the terms sustaina- bility and sustainable development can be easily mixed up. Therefore, it is crucial to un- derstand the differences between them. “Sustainability refers to actions that are able to use resources without depleting them from future generations” (Niemi, 2024, p. 18). Its modern term, sustainable development, was developed in 1987 in the Brundtland Re- port by the World Commission on Environment and Development (WCED), where sus- tainable development was described as an activity that “meets the need of the present without compromising the ability of future generations to meet their own needs” (Brock- ett & Rezaee, 2012, p. 4; Portney, 2015, p. 2-3). In this work, the term "sustainable de- velopment" is utilized, even though it shares similar characteristics with “sustainability”. The Brundtland report highlighted three dimensions of sustainability, which are eco- nomic, social equity and environment, and those dimensions establish the concept of sustainability (Portney, 2015, p. 6). The report also emphasizes the importance of all three dimensions and how they must be balanced without sacrificing the other (Portney, 2015, p. 6). A Venn diagram below in Figure 1 shows the balance of the dimensions of 15 sustainability, and this is what sustainable development aims for: all dimensions of sus- tainability must be considered for sustainability to be achieved. Figure 1. Venn diagram of three dimensions of sustainability (Portney, 2015, p. 7; Purvis et al., 2019, p. 682). The Venn diagram is the most well-known way to illustrate the balance of sustainable development (Purvis et al., 2019, p. 682). The strength of the Venn diagram lies in its simplicity and its ability to show what sustainability consists of and how it should be approached, which is why it was presented in the Brundtland Report in 1987 as a depic- tion of sustainability dimensions (Purvis et al., 2019). However, new ways to represent dimensions of sustainability have been developed, such as the pillar model, which closely resembles the Venn diagram but differs in that the dimensions are depicted as equal pillars (Purvis et al., 2019). The problem with these representations is that, in re- ality, the dimensions of sustainability are not equal, even though equality is the goal (Vanhala & Ristaniemi, 2022, p. 31) In reality, decisions and developments in the field of sustainability are predominantly made based on the requirements of nature and the environment, with social and eco- nomic requirements considered only afterward (Vanhala & Ristaniemi, 2022, p. 31). It can generally be said that regardless of which model we use, society is not yet sustaina- ble, and the goals of sustainable development must be genuinely considered with future generations in mind. For example, in many European countries such as Finland, many 16 people enjoy good social conditions, but at the same time, our consumption levels are not ecologically sustainable (Vanhala & Ristaniemi, 2022, p. 33). Therefore, it is im- portant to understand what the dimensions of sustainability entail in order to recognize the areas that require improvement. Environmental sustainability is one of the dimensions of sustainability which was previ- ously introduced. It can be said to be the most important dimension of sustainable de- velopment in terms of significance (Vanhala & Ristaniemi, 2022, p. 31). But what does ecological sustainability really consist of? The Finnish Ministry of the Environment (2023) describes environmental sustainability as “the fundamental requirement of sustainable development is the preservation of biodiversity and the functioning of ecosystems, as well as the long-term adaptation of human economic and material activities to the car- rying capacity of nature”. This means that humans must not endanger the life-supporting natural systems: the biosphere, atmosphere, waters, or soil and bedrock (United Nations, 2025). Additionally, according to the United Nations (2025), human-caused emissions must not exceed the resilience of nature. Human well-being and other dimensions of sustainability, social and economic, ultimately depend on environmental sustainability. Social sustainability is the second dimension of sustainability that must be considered with a view toward a better future. The goal of social sustainability is to ensure the preservation of conditions for well-being for future generations (Finnish Ministry of the Environment, 2023). To preserve future generations' well-being, we must sustainably ad- dress population growth, poverty, food and healthcare, gender inequality, and the pro- vision of education (Finnish Ministry of the Environment, 2023). The United Nations (2025) also highlights cultural sustainability in their sustainable development agenda, which can be linked to social sustainability. The third and final dimension of sustainability is economic sustainability. Economic sus- tainability, like the previous dimensions of sustainability, also plays a crucial role in en- suring a better future. The Finnish Ministry of the Environment (2023) describes 17 economic sustainability as “balanced growth that is not based on long-term indebted- ness or the depletion of resources”. The sustainable use of renewable natural resources, material efficiency, and the circular economy can promote economic sustainability (Busi- ness College Helsinki, 2019). A sustainable economy must be built on a solid foundation because it enables other aspects of sustainable development (Finnish Ministry of the Environment, 2023). Economic sustainability can be divided into three decision-making sectors: society, con- sumers, and companies (Business College Helsinki, 2019). According to Business College Helsinki (2019), society plays a significant role in the development of a sustainable econ- omy. The importance of this role stems from society's need to provide people with prod- ucts and services that are produced sustainably (Business College Helsinki, 2019). Thus, economic sustainability is achieved from society's perspective. Society's role is also to enact laws and regulations to promote a sustainable economy (Business College Helsinki, 2019). From the perspective of a sustainable economy, consumers can influence sustainability through their behaviour and choices, which have a significant impact on creating a sus- tainable future (Business College Helsinki, 2019). Consumers could choose sustainable products and utilize renewable energy sources, thereby supporting sustainable develop- ment. Ultimately, consumers play a crucial role because demand is creates the supply of sustainable products and services (Business College Helsinki, 2019). For this reason, companies must be able to offer products and services to consumers that are sustainably produced. Business College Finland (2019) emphasizes the importance of a company’s sustainable economy, where the company is responsible for complying with societal laws and regu- lations, sustainably producing products and services, and communicating its sustainable practices. Therefore, companies have a great responsibility to operate sustainably, de- pending on the industry, for sustainable development to be achieved. Overall, it is 18 important to note that when promoting a certain dimension of sustainability, it interacts simultaneously with another dimension of sustainability (Finnish Ministry of the Envi- ronment, 2023). This effectively highlights the usefulness of the Venn diagram in illus- trating how each dimension plays a significant role in sustainable development. As large companies seek opportunities for growth, they often need to expand their mar- kets, which frequently involves increasing exports and expanding operations into foreign markets (Nguyen et al., 2019). It is called internationalization, which is a very familiar term in the business world today. According to Nguyen et al. (2019), internationalization is a multi-phase process that also requires consideration of markets, resources, cultures, regulations, and many other factors. Internationalization has been studied in numerous scientific studies, and various theories have been developed to enable effective opera- tions in foreign markets (Korsakienė et al., 2021). The first theory of internationalization is considered the Uppsala model, which was de- veloped in the late 1970s. This model suggests that a company seeks to internationalize by acquiring and utilizing knowledge and expertise from its own international operations while simultaneously expanding its foreign activities (Dominguez & Mayrhofer, 2017). This theory has been considered very linear due to its straightforward development, where internationalization is seen as automatically evolving over time and therefore later, the Uppsala model has been updated to better align with the current global busi- ness environment, which, according to the latest research, has been found to be curvi- linear (Dominguez & Mayrhofer, 2017; Korsakienė et al., 2021). Another more modern internationalization theory is called the born global model, which operates in contrast to the Uppsala model. In this model, companies aim to expand in- ternationally without a particularly stable domestic business base. The key characteristic is to become international in a very aggressive manner (Nguyen et al., 2019). The rapid development of digitalization has driven the born global model forward, enabling smaller companies to become global (Dominguez & Mayrhofer, 2017; Nguyen et al., 19 2019). It can be concluded that large companies that have been established for a long time in the market have followed the traditional and developed Uppsala internationali- zation model, thus building a large global market before the advent of digitalization, when there was less need for aggressive internationalization. 2.2.1 Relationship between internationalization and sustainability Climate change is a major problem on our planet, and our market environment is also global, which inevitably drives sustainability challenges but at the same time also creates opportunities. Since the basics of internationalization and sustainability have been intro- duced, it is crucial from a research perspective to understand the connection between them, particularly how internationalization can advance sustainability. Internationalization constantly subjects a company to the societal pressures of the coun- try where it operates (Gómez-Bolaños et al., 2020). These pressures are caused by legis- lation, practices, and cultural differences, especially when it comes to a company ex- panding its operations from a developed country to a developing one (Gómez-Bolaños et al., 2020). Therefore, it forces companies to consider local sustainability issues in the target country to meet sustainability goals. Gómez-Bolaños et al. (2020), however, men- tion that companies relocating to developing countries can improve their environmental management and do not intentionally take advantage of the target country's looser reg- ulations to evade stricter sustainability regulations in their home country. Earlier studies have tried to show that companies relocate to places where they can pollute more, and this practice is referred to as the pollution haven hypothesis (Li & Zhou, 2017). However, this theory has been refuted by later studies, and companies have been able to improve sustainability practices in developing countries through their for- eign operations, which plays a crucial role in addressing global environmental issues (Gómez-Bolaños et al., 2020). Companies have replicated the strictly regulated practices of their home country in their foreign operations, which has improved their sustainability 20 efforts, maintained their trustworthiness, and thereby enhanced their competitiveness (Christmann & Taylor, 2001). There has not been a significant direct improvement in environmental performance be- tween internationalization and environmental sustainability in the energy sector, consid- ering that internationalization was expected to achieve emission reductions (Gómez-Bo- laños et al., 2020). However, the connection between these two terms has enabled com- panies to achieve significant improvements in environmental management, which has especially enhanced the transparency of large companies regarding their sustainability efforts (Gómez-Bolaños et al., 2020; Marano & Kostova, 2016). Therefore, it is interest- ing to see whether sustainable internationalization of large companies can lead to im- proved environmental performance in the industrial manufacturing sector. 2.3 Impact of climate change on multinational enterprises A multinational enterprise (MNE) is an enterprise whose operations and management are located in one country, known as the home country, but it also has operations in other countries (Eurostat, 2025). Thus, MNE has many value chains, involving numerous internal and external stakeholders, which require risk management and strategic deci- sion-making (Gasbarro et al., 2017; Kolk & Pinkse, 2008). This responsibility is signifi- cantly influenced by climate change and related regulations, which are driving enter- prises towards more sustainable operations and development. This theoretical chapter discusses how climate change can provide MNEs opportunities for sustainable innova- tion and firm-specific advantages (FSAs). Adarkwah and Malonæs (2022) describe the definition for FSAs as follows: company’s unique resources and capabilities that enable successful internation- alization and provide a competitive advantage. These advantages can turn out as intangible assets, capabilities gained through learning, or privileged relationships with external actors (Adarkwah & Malonæs, 2022). 21 Internal competitive advantages include technological expertise, reputation, organiza- tional competence, and operational efficiency (Lee & Rugman, 2012). With the help of FSAs, the company can achieve growth and profitability, and maintain its own market which, are vital features for the company’s financial condition. Lee and Rugman (2012) emphasize that FSAs are a key factor in terms of the competitiveness of MNE. Climate change is one of the factors driving companies to develop their operations to achieve competitive advantage and effective international business (Kolk & Pinkse, 2008). Ac- cording to Kolk and Pinkse (2008), developments related to climate change that drive competitive advantage are called “green” competitive advantages. Given the current strict regulations and climate change targets, green investments can be considered quite common in MNEs (Kolk & Pinkse, 2008; Marano & Kostova, 2016). Kolk and Pinkse (2008) highlight the importance of companies adapting their FSAs while considering the market changes caused by climate change. Therefore, MNEs must ana- lyze how they aim to adapt their FSAs to maintain their value. Climate change affects technological development, which can either destroy or strengthen a company’s capa- bilities (Kolk & Pinkse, 2008). Kolk and Pinkse (2008) describe these as a competence- enhancing change that enables the development of new technologies based on existing capabilities and a competence-destroying change that forces the company to develop entirely new capabilities to replace outdated technologies. Therefore, MNEs must adapt to these changes strategically in different ways. According to Lavie (2006), MNEs can follow three frameworks in their FSAs in order to adapt to technological developments related to climate change. 1. Capability evolution, where FSAs can be gradually adapted to allow existing ca- pabilities to be retained and modified to align with new technology. 2. Capability transformation, where core FSAs are preserved and applied with new operational methods, allowing the company to leverage both new and existing resources. 22 3. Capability substitution, in which old FSAs are replaced with completely new ca- pabilities (Lavie, 2006). The choice of the above-mentioned strategies is influenced by the industry where MNEs operate because technology changes may be more demanding in certain fields or whether capabilities need to be transformed across the entire value chain. However, Lavie (2006) emphasizes the utilization of the capability transformation framework for effective sustainability activities. Kolk and Pinkse (2008) summarize the biggest chal- lenge for MNEs as the ability to assess in advance how much climate change-driven tech- nological changes will impact their business. MNEs want to be as certain as possible that environmental investments have the poten- tial to improve operational efficiency and, therefore, enhance competitiveness. Namely, if MNEs invest in a rigid environmental investment with high regulatory or standard- related uncertainties, it may lead to failed investments, which is something that large multinational enterprises want to avoid at all costs (Kolk & Pinkse, 2008; Rugman & Verbeke, 1998). Therefore, MNEs face significant institutional pressure, requiring them to consider regulations both in their home country and in the countries where they op- erate. As a facilitating factor from the perspective of MNEs' decision-making, there are Europe's strict sustainability goals and regulations, such as the Green Deal and the EU Taxonomy, which provide clear guidelines for future investments and financial security for corporate investments, which facilitates the fight against climate change. 2.3.1 Risks and opportunities for MNEs due to climate change Climate change creates both challenges and opportunities for MNEs, which can affect the company either directly or indirectly (Gasbarro et al., 2017). Therefore, it is im- portant for the company to recognize the existing risks related to climate change and be able to turn them into opportunities through various actions. According to Gasbarro et al. (2017), recognizing these risks and opportunities is the first step toward developing 23 a climate strategy that aims to prevent the growth of GHGs. It has been observed that companies tend to delay their actions regarding climate change because it creates un- certainties, such as difficulties in integrating sustainable practices into their core opera- tions or a lack of knowledge about the necessary components for sustainable business practices (Melkonyan et al., 2024). Therefore, it is important to understand the risks as- sociated with climate change so that MNEs can implement sustainable practices aimed at reducing GHG emissions without jeopardizing their operations or profitability. Gasbarro et al. (2017) have aggregated the key climate change-related risks that slow down MNEs' adaptation to sustainable operations. The key climate change-related risks are listed in Table 1 below. Table 1. Climate change-related risks (Gasbarro et al., 2017). Risks Examples Regulatory Carbon taxes, emission reduction requirements (EU Taxonomy) Physical changes Changes in conditions which could affect opera- tions Product and technology innovation Adopting innovative technology under environ- mental uncertainty Changes in customer needs Changes in customer demands towards climate- friendly Reputation Use of products and processes with a negative im- pact on climate leading to a negative reputation Financial impacts Sustainability performance effect on investment decisions Operational efficiency Change in the consumption of the production method Table 1 shows that climate change poses a wide range of risks to MNEs from various aspects of their business operations, which can have either a direct or indirect impact on their activities. It is clear to say that companies experience significant pressure from climate change and the uncertainties that it causes (Melkonyan et al., 2024). On the other hand, the risks in Table 1 can also be seen as opportunities, such as reputation, where a company’s own positive impact on climate can directly enhance its own image. 24 In turn, a better reputation increases a company’s market value, enabling research and development investments towards sustainability innovations, allowing the company to enhance its sustainable operations (Gasbarro et al., 2017). It can be said that there are many risks, but they are interconnected and create a chain reaction of opportunities for companies to operate more sustainably. 2.4 Sustainability-oriented innovation As we understand, sustainable operations require innovations, and these innovations are needed across different areas of the company to achieve sustainable operation. Therefore, there is a concept called sustainability-oriented innovation (SOI) which fo- cuses on transforming a company’s overall operations and ideology to achieve social and environmental value in addition to economic value (Adams et al., 2016). SOI is needed for a company to cohesively create and maintain sustainable operations as an organiza- tion. The ignorance of analysing SOI barriers may lead to inefficient resource use related to sustainability (Ivanov, 2024). Therefore, this chapter focuses on the theory of SOI and how it can be used to influence and achieve sustainable operations within a company. The chapter also discusses the three development stages of SOI and what they require from a company. Adams et al. (2016) outline three dimensions of SOI, which help build an understanding of holistic sustainable operations. A holistic understanding of sustainability is important because, in the past, the major focus of sustainable innovation has been on technology and products. These three SOI dimensions compiled by Adams et al. (2016) are tech- nical/people, stand-alone/integrated, and insular/systematic. Understanding the dimen- sions of SOI is important when we later discuss the three stages of SOI development. Figure 2 below illustrates the dimensions of SOI and their respective purposes to achieve sustainable business. 25 Figure 2. The dimensions of SOI (Adams et al., 2016). According to Figure 2, technological innovations also require the contribution and exper- tise of the company’s people, and the ways of operation play a significant role in achiev- ing sustainable practices. Another perspective is a broad view of sustainable innovation, where the entire organization systematically participates in sustainable innovations and thinking, rather than just specific departments focusing on it (Adams et al., 2016). The middle line in Figure 2 represents that dimension. Adams et al. (2016) mention how SOI aims to spread sustainable thinking throughout the entire business operation. The pur- pose of the last dimension in Figure 2 is to create an understanding of SOI, where inte- gration of innovation leads to better sustainable practices with companies engaging with various external factors such as governments and investors rather than only operating internally (Adams et al., 2016). 2.4.1 SOI development stages Those dimensions and research related to the three-category classification, where inno- vations are sorted into three different categories, form the foundation for the develop- ment stages of SOI (Adams et al., 2016; Tukker & Butter, 2007). The three-category model includes system optimization, where, for example, a shift to cleaner technology occurs; individual innovations, such as changing production chains; and finally, large-scale changes to the entire system, influenced by societal needs such as regulations (Tukker & Butter, 2007). 26 The three development stages of SOI are: 1. Operational optimization, where small changes or improvements are made to improve efficiency. 2. Organizational transformation, where sustainability is integrated into the business strategy and company culture. 3. Systems building, where innovations are used to build broader, sustaina- bility-driven ecosystems (Adams et al., 2016). The goal of operational optimization is to achieve more with less by complying with reg- ulations and leveraging existing innovation capabilities. The process focuses on internal and incremental innovations, utilizing existing knowledge (Adams et al., 2016). On the other hand, new information can be obtained by acquiring external experts. In organiza- tional transformation, sustainability is integrated as a cultural and strategic norm into the company’s values whereby new values and functions are introduced (Adams et al., 2016). These functions are human-centred, significantly enhancing the integration of sustainable development into the company. According to Harsanto et al. (2024), opera- tional optimization is the most common approach to utilize SOI in manufacturing com- panies. According to Ivanov (2024), a strong external focus helps avoid unnecessary investments. Unnecessary investments are harmful to companies, especially when they are involved in sustainable innovations, which already involve significant risks and uncertainties. Ad- ams et al. (2016) emphasize the collaboration of key stakeholders and the integration of the SOI culture throughout the company. Organizational transformation can be de- scribed as doing good by doing new things. Lastly, systems building, where the purpose is to “doing good by doing new things with others”. Adams et al. (2016) describe the goal of the last activity of SOI as creating new 27 business models by leveraging external collaborations and new systemic solutions that support the established sustainable values. It is important to understand the complexity of the system, build trust, and identify means of change in order to address sustainable development challenges through SOI (Adams et al., 2016). On the other hand, the high- est level of SOI is the most difficult to utilize, and Harsanto et al. (2024) state in their study that it is the least used SOI approach in manufacturing companies. In conclusion, SOI can provide companies with a competitive advantage. Achieving this requires a systematic approach, organizational transformation, and active collaboration with stakeholders. Dangelico et al. (2017) particularly emphasize the integration of in- ternal and external resources to achieve sustainable innovation. By intentionally aligning their philosophies, values, products, processes, and practices to create social and envi- ronmental value alongside economic returns, companies can effectively integrate sus- tainability into their core operations (Harsanto et al., 2024). 2.4.2 Research and development capability R&D capability is a key factor for the continuous development aiming to create and im- prove new products, processes and services. R&D capabilities can be defined as re- sources that include technology, human expertise, strategic knowledge, and a company’s infrastructural competence (Kim & Choi, 2020). These capabilities are needed, especially internally, but also externally. External capabilities help save assets and accelerate the process while also being integrated with internal capabilities, driving R&D growth (Kim & Choi, 2020). In particular, it is crucial to leverage external capabilities in environmental innovation processes and closely integrate them with internal capabilities (Watson et al., 2018). According to Ruffoni et al. (2018), R&D investments have a positive impact on the in- crease of innovations and productivity. However, R&D activities require more than just financial investment; dynamic capabilities are considered one of the key focus areas in 28 R&D that provide desired outcomes. Aldabbas and Oberholzer (2023) describe dynamic capabilities as a company’s activity where internal and external capabilities are coordi- nated, learnt, and transformed to quickly respond to changing situations, such as sus- tainable development. Simply, they are processes that utilize the company’s resources. Dynamic capabilities have a significant impact on a company’s performance and com- petitive advantages (Aldabbas & Oberholzer, 2023). Therefore, it is important to under- stand how dynamic capabilities contribute to achieving competitive advantage and per- formance. From an R&D perspective, it is important to leverage the dynamic capabilities of learning and transformation. Learning capabilities promote the emergence of innovations, and particularly, the integration of internal and external knowledge facilitates R&D (Aldabbas & Oberholzer, 2023). Sustainability requires continuous new innovations, making learn- ing within the company essential. It is strategically important for competitive advantage that a company is able to produce and share knowledge internally so that new ways of thinking and learning spread throughout the organization (Roth, 2003). This leads us directly to the second dynamic capability, namely transformation, which allows R&D to become agile, enabling the company to quickly adapt to changing situa- tions. Transformation capabilities are needed to create efficient processes and to re- spond to the rapid changes in sustainable development (Aldabbas & Oberholzer, 2023; Watson et al., 2018). Thus, new sustainable and versatile products are achieved through R&D. However, this requires active collaboration with external factors due to the com- plexity of sustainability issues (Dangelico et al., 2017). 2.4.3 Financial slack and sustainable development To cover research and development costs and enable possible new investments, a com- pany must have availability of financial slack. This so-called financial slack refers to the company’s financial resources that are used for unexpected situations and investments 29 rather than immediate needs (Liang et al., 2023; Mahmood et al., 2023). These resources consist of the company’s internal funds, which could be utilized for sustainability inno- vations which include uncertainties. Due to internal resources, it is important that the company utilizes its funds properly (Liang et al., 2023). Previous studies have shown that the use of slack resources towards sustainability has had a positive impact on environ- mental performance (Modi & Cantor, 2021), which is important for the target of reducing Scope 3 emissions. On the other hand, sustainability actions involve many risks, requiring a significant number of slack resources. Boso et al. (2017) state that a company with a large financial slack can have a more se- cure impact on sustainability investments. Financial slack can be considered one of the most important resources a company can utilize. Financial slack arises from a company’s profitability, where the operating profit from their revenue, after taxes and other ex- penses, remains as cash reserves or liquidity (Boso et al., 2017). Therefore, a profitable company has the freedom to conduct research more freely, providing security against mistakes. Profitability plays a crucial role in a company’s competitiveness, enabling greater use of financial slack resources for sustainable investments (Hofer et al., 2012). 2.5 Summary of the theory To summarize theoretical literature, a summary of the theoretical components has been compiled to facilitate the understanding of the upcoming empirical part and to explain why the choices have been made. So far, sustainability and corporate internationaliza- tion have been examined, along with their interconnection and how climate change af- fects the operations and strategies of multinational enterprises. We conclude that sus- tainability has influenced the operations of international companies in many ways, in- cluding creating risks, societal pressures, and opportunities. It has also been observed that multinational companies can create competitiveness under the guise of sustainabil- ity and climate change (Kolk & Pinkse, 2008). Therefore, internationalization of compa- nies has been chosen as one of the factors for empirical study. 30 The study has also explored sustainability-oriented innovation, covered its development stages and focused on research and development capabilities as well as the impact of financial slack. The theory highlighted the importance of sustainability-oriented innova- tion and how extensively it impacts the entire company’s operations, ranging from tech- nology to external collaboration. The theoretical section provided insights into the im- portance of innovations and the significant role that financial resources play in the im- plementation of innovations. Thus, for the empirical study, three other factors were identified that are seen to influence sustainable operations. These three factors are the company’s revenue, the company’s adjusted operating profit, and input in research and development. 31 3 Methodology This chapter discusses the analysis method used in this study, called qualitative compar- ative analysis (QCA). The goal is to address the key principles, components, and effec- tiveness of the method in relation to the objectives of the study. First, the chapter dis- cusses the foundation of QCA and its applicability; after that, it moves on to examine its characteristics and strategy. The purpose of this chapter is to gain a clear understanding of the analysis method's theory and applicability, considering the objectives of this study. 3.1 Qualitative comparative analysis QCA is a methodological tool that utilizes characteristics of both qualitative and quanti- tative analysis. (Rihoux & Ragin, 2009; Schneider & Wagemann, 2012). This study em- phasizes a quantitative method that utilizes numerical data to understand causality and potential trends related to the research outcomes. QCA is characterized by a set-theo- retic approach, where the cases are assigned to a set based on how they meet the se- lected conditions, and the selection is being made using membership scores (Schneider & Wagemann, 2012). Membership scores indicate whether the cases belong entirely or partially to sets, which can be examined using multivalued, crisp or fuzzy sets in QCA (Rihoux & Ragin, 2009; Schneider & Wagemann, 2012). Thus, information is obtained on which combinations of conditions lead to a specific outcome. Crips set QCA (csQCA) and fuzzy set QCA (fsQCA) will be discussed later in this chapter. Greckhamer et al. (2018) emphasize the im- portance of conditions influencing the outcome and how they should be guided by the- ory and case knowledge for QCA to be high quality. QCA was originally designed for small cases, and even though considering small-case QCA analysis, the acquired theory can be evaluated and developed, allowing for a deeper exploration of the studied case (Di Paola et al., 2025). Therefore, QCA is effective with a 32 small case dataset because it can examine complex combinations of factors instead of the statistical dependence of individual variables. QCA is utilized in fields such as social and business sciences because they require a complex understanding of cause-and-ef- fect relationships. Thus, it works excellently for the strategic analysis of a company’s sus- tainable development because sustainability involves many complex factors that simul- taneously influence each other (Di Paola et al., 2025). A single factor rarely leads to an outcome; instead, a combination of multiple factors is required, which is called a conjunctural causation (Schneider & Wagemann, 2012). In addition to conjunctural causation, a key feature of QCA for analyzing complex condi- tions is asymmetry. This examines the relationship between causes and effects, meaning that certain factors leading to an outcome do not necessarily prevent the same outcome from occurring in their absence (Rihoux & Ragin, 2009; Schneider & Wagemann, 2012). Therefore, QCA is a useful method considering the complex research setting of this study. Studies related to QCA have become more common in innovation and business and man- agement research, especially with fsQCA, which identifies the most precise outcomes from case study research (Kraus et al., 2018). QCA has produced qualitative results, lead- ing to conclusions and opportunities for further research in innovation and technology research (Huang et al., 2016). Therefore, QCA has been chosen as the analysis method for this case-oriented research to generate reliable conclusions on sustainable innova- tion and practices while reducing Scope 3 emissions. Next, the structure of QCA, its com- ponents, and key considerations will be discussed to understand the choices made in the empirical part of this study. 3.1.1 Characteristics of fuzzy-set QCA In order to conduct fsQCA, data must be collected from the cases, showing the desired conditions and the outcome, which can be used to conduct the analysis. Rihoux & Ragin (2009) emphasize that the conditions must influence the chosen outcome for the 33 analysis to be convincing. Generally, it is recommended to keep the number of conditions between 3 and 5 to maintain the validity and reliability of the analysis (Schneider & Wagemann, 2012). Once the data of cases are collected related to the analysis and con- ditions and outcome are defined based on studied theory, the characteristics of fsQCA must be understood. fsQCA analysis consists of the following phases: 1. Collection of raw data on conditions and the outcome 2. Calibration of raw data to fuzzy sets 3. Process of establishing a truth table 4. Minimization with Boolean algebra 3.1.2 Fuzzy sets and their membership scores QCA most commonly uses either crisp or fuzzy sets to determine whether a member belongs to a set or not. Fuzzy sets differ from crisp sets in that the membership scores can range between 0 and 1, rather than being strictly 0 or 1 (Schneider & Wagemann, 2012). Thus, fuzzy sets can represent full membership (1) and full non-membership (0), and a crossover point (0.5) that determines which group the value belongs to. Therefore, there are partial memberships (0.9), for example, where a case belongs almost entirely to the set but not completely. This reveals a subset relationship, where subsets (X) and (Y) also belong to a larger set (Z). According to Scheider and Wagemann (2012), subset relationships allow fsQCA to reveal which factors together influence the outcome, even though on their own they may not have significant impact. 3.1.3 Calibration of raw data to fuzzy sets The collected raw data must be transformed into fuzzy sets, a process known as calibra- tion. Calibration is the first step of the actual QCA process, where the values of the vari- ables in the data are calibrated into membership scores as the fuzzy set (Rihoux & Ragin, 34 2009). In direct calibration, three threshold anchor values between 0 and 1 are chosen to represent full membership, full non-membership, and the crossover point for each condition and outcome by the researcher (Mello et al., 2021). The determination of cal- ibration must always utilize theoretical and conceptual knowledge (Kraus et al., 2018; Rihoux & Ragin, 2009). In defining threshold values, justifications must be provided, even though these definitions can be made rationally (Kraus et al., 2018). Calibration can also be performed indirectly, where the researcher must pre-group the membership values of cases based on raw data, allowing anchor points to be set for threshold values. (Schneider & Wagemann, 2012). According to Mello et al. (2021), indirect calibration is a less commonly used calibration method. 3.1.4 Process of establishing a truth table After calibration, the truth tables can be created. The truth table is a matrix, considered the most essential tool in QCA, where the columns represent conditions and the rows represent all possible combinations of conditions, which are referred to as causal factors (Rihoux & Ragin, 2009). In addition, the truth table shows the outcomes, where value (1) represents an achieved outcome and value (0) represents a non-achieved outcome, al- lowing an assessment of whether the outcomes are sufficient or not. Finally, the truth table also reveals values for coverage and consistency, which are important characteris- tics for the reliability of the analysis. The consistency reflects how well a combination of conditions produces the desired outcome. It is important that the consistency value is at least 0.75 for the results to be considered reliable (Chang & Cheng, 2014; Schneider & Wagemann, 2012). Coverage is another metric that must be considered in QCA, aiming to measure how common the condition combinations are among all cases where the outcome occurs (Schneider & Wagemann, 2012). 35 3.1.5 Minimization with Boolean algebra Finally, truth tables are minimized based on Boolean algebra, where unnecessary varia- ble combinations are eliminated, and the truth table is reduced to the simplest combi- nations to capture the most essential cause-effect relationships (Kraus et al., 2018; Ri- houx & Ragin, 2009). Minimization utilizes consistency and frequency threshold values. The frequency threshold determines which condition combinations are included in the analysis and which are not. For small sample sizes (10-50), the threshold value is set to 1 (Kraus et al., 2018; Rihoux & Ragin, 2009). Meaning that any combination that scores at least 1 is included in the analysis. After minimization, three different solutions can be obtained from the truth tables, and these are conservative, parsimonious, and interme- diate. In the conservative solution, logical remainder is not utilized, which makes the solution very complex and thus the analysis challenging (Álamos-Concha et al., 2022; Rihoux & Ragin, 2009). On the other hand, the parsimonious solution is the opposite of the conservative solution, where all possible remainders are used to simplify the solu- tions (Álamos-Concha et al., 2022). Lastly, the intermediate solution is an intermediate model of the conservative and parsimonious solution, where only credible remainders based on theoretical and empirical knowledge are utilized to achieve a good balance in the results (Schneider & Wagemann, 2012). 36 4 Empirical study: Development of Scope 3 emissions This chapter discusses the empirical phase of the study, where the different stages of the analysis are reviewed step by step. First, the raw data collection phase is reviewed, ex- plaining what data was collected and how it was gathered for the case study. After this, the chapter discusses and justifies the selection of variables, referred to as conditions and outcomes. After this, the chapter moves on to the QCA analysis, where the different stages of the analysis are discussed in detail, considering the collected cases and varia- bles. Finally, the last part of the chapter discusses the results using case examples. 4.1 Data collection The data for this study consisted of 19 Finnish manufacturing companies. The studied companies were multinational and well-known in Finland and have a significant impact on Finnish GHG emissions. Data were collected over a five-year period from 2019 to 2023 to ensure suitable fundamentals for QCA and a credible study. This period is important for the study, as the calculation and reporting of Scope 3 emissions for large Finnish manufacturing companies primarily started in 2019, allowing their development to be tracked up until 2023. The data collection process was straightforward, with key figures primarily gathered from companies’ annual and sustainability reports. The selection of companies was assisted by the Finnish company Upright, which gathers data on compa- nies and calculates their net impacts related to sustainability. Upright made it possible to filter Finnish manufacturing companies based on their revenue, which facilitated the data gathering. Annual and sustainability reports were also utilized after the analysis to support the interpretation of the results, providing practical and realistic evidence to back the findings and conclusions. The selection of companies was influenced by factors such as revenue, as it indicated whether the company had the necessary data to be included in the study. A high revenue indicated that their reports contained the necessary data on Scope 3 emissions. 37 Information on the company’s revenue, profitability, research and development invest- ment, foreign revenue, and Scope 3 emissions was collected from the companies’ re- ports. Table 2 was created from the collected data, summarizing the key values and fig- ures for future analysis, and Table 2 is presented below. Table 2. Collected data from 19 Finnish industrial manufacturing companies. Company Revenue Profitability R&D investment Foreign revenue Scope 3 Fiskars Group 1167.6 0.102 0.016 0.528 1.877 Gargotec 3783.8 0.078 0.027 0.523 -0.115 Glaston 193.6 0.037 0.041 0.491 -0.062 Huhtamäki 3784.8 0.089 0.007 0.607 -0.186 Kemira 2942.8 0.084 0.011 0.491 -0.473 Kone 10459 0.118 0.017 0.604 -0.148 Konecranes 3404.6 0.085 0.014 0.490 -0.224 Metso 4146.8 0.106 0.014 0.755 -0.162 Metsä Board 2065.6 0.134 0.004 0.366 0.736 Neste 18274.4 0.102 0.004 0.204 -0.182 Nokia 22907.6 0.065 0.190 0.723 -0.061 Nokian Tyres 1427.3 0.123 0.019 0.291 0.844 Outokumpu 7148 0.036 0.002 0.356 -0.210 Purmo 771.8 0.075 0.007 0.061 -0.381 Stora Enso 9969.6 0.105 0.013 0.303 -0.290 UPM 10162.4 0.135 0.030 0.384 0.064 Vaisala 455.2 0.116 0.129 0.654 2.142 Valmet 4365.6 0.089 0.020 0.593 -0.122 Wärtsilä 5281.8 0.072 0.038 0.677 0.179 4.2 Construction of variables From the collected data, conditions and the outcome were compiled for the configura- tional method. Theory supports using between three and five conditions in configura- tional analysis (Schneider & Wagemann, 2012). A total of four conditions were collected, 38 and based on them, one outcome was determined for the study. With the aim of study- ing the development of Scope 3 emissions and the factors influencing them, the outcome for the configurational analysis was chosen as the reduced Scope 3 emissions in Finnish manufacturing companies. The Scope 3 emissions data for the outcome was collected from sustainability and annual reports over a five-year period (2019-2023). The emission figures were collected from the reports as CO2 equivalent, which were then divided by the company’s size, i.e., its revenue. Finally, the difference in emissions between 2023 and 2019 was calculated, resulting in the 5-year change. The data collection was compli- cated by the companies’ methods of calculating Scope 3 emissions and the absence of Scope 3 emission categories from some calculations, which presents challenges and lim- itations for this study, and these can be addressed in future research. This is due to the voluntary nature of companies’ Scope 3 emissions reporting, as well as the challenges of tracking emissions due to the complexity of value chains, leading to varying reporting (Schulman et al., 2021). Based on the outcome, four conditions were selected as variables that support the the- ory, allowing the study of changes in Scope 3 emissions. These conditions are financial slack, which is broken down into the firm revenue and impact of profitability as two sep- arate conditions, capability in innovation, and internationality of company. The data for the first condition, high firm revenue, was collected from the companies’ annual reports. The company’s revenue was analyzed over a five-year period from 2019 to 2023. The average of these values was calculated and used as the first condition in the configura- tional analysis. The second condition was chosen to be the high profitability, which examined the com- pany’s adjusted operating profit %. Adjusted operating profit % was chosen to exclude one-time and unusual expenses that are not part of the company’s normal business op- erations. Therefore, a more accurate figure of the core operations was obtained. The adjusted operating profit % was also obtained from the companies’ annual reports for the period 2019-2023, and an average was calculated from these. The first two 39 conditions support financial slack, as these values create financial resources for the com- pany to invest in sustainability (Boso et al., 2017). Therefore, the first two conditions were selected for configurational analysis. The high R&D intensity was chosen as the third condition for the analysis. The purpose of this condition is to examine the company’s R&D expenditure, which indicates how much the company is willing to invest in product development. Previous studies have shown how R&D investments have a positive impact on reducing CO2 emissions (Baek & Lee, 2023; Paramati et al., 2020). Therefore, capability in innovation has been included as a condition in this study. Data on R&D investments was collected from the companies’ annual reports, and the average was calculated from 2019 to 2023. This was then ad- justed by the company’s average revenue to obtain an actual financial investment, taking the company’s size into account. This provided the R&D percentage of revenue, which is utilized in the analysis. As the fourth and final condition, the high internationalization was chosen. Previous studies have shown that internationalization and sustainable practices have a positive relationship with each other (Chiarvesio et al., 2015; Zhang et al., 2023). Therefore, in- ternationalization was chosen as a condition, as it is seen to have an impact on the out- come of the configurational analysis, and the theory supports this hypothesis. Data on the internationalization of companies was collected from annual reports from 2019 to 2023. The measurement assessed the ratio of foreign revenue to the company’s total revenue. Domestic revenue was defined as revenue from Europe, allowing the revenue from the remaining continents to be calculated as foreign revenue based on geograph- ical distribution. Thus, the amount of foreign revenue was calculated, and its ratio to the company’s total revenue was determined, from which the average was taken for the se- lected period. The selected conditions, outcome, their measurement methods, and data sources are summarized in Table 3, which helps in understanding the overall framework. 40 Table 3. Outcome and conditions. Conditions and outcome Measure Source Conditions High firm revenue Average of the company’s revenue from 2019 to 2023 Annual reports High profitability Average of adjusted operating profit (%) from 2019 to 2023 Annual reports High R&D intensity Average of R&D investment in proportion to revenue from 2019 to 2023 Annual reports High Internationalization The ratio of foreign revenue to total revenue (average from 2019 to 2023) Annual reports Outcome Reduced Scope 3 emissions The change in Scope 3 emissions from 2019 to 2023 relative to revenue Sustainability and annual reports 4.3 Data calibration Next, using the data, conditions, and the outcome, a calibration process was conducted to define threshold values, allowing numerical data to be converted into membership scores. Threshold values were established for each condition and outcome, and their crossover points were defined. Thus, the fuzzy-set technique was used in the calibration process. This required the use of direct calibration, where three qualitative thresholds were selected to determine the membership level in the fuzzy set (Rihoux & Ragin, 2009). The threshold values were divided into three categories: full non-membership (0), cross- over point (0.5), and full membership (1). Threshold values were assigned based on the data and theoretical support to represent the three aforementioned limits. The thresh- old values were selected as shown in Table 4. Table 4. Qualitative thresholds of the outcome and conditions. Conditions and outcome Threshold full non- membership Crossover point Threshold full membership High firm revenue 2000 4500 12000 High profitability 0.05 0.1 0.12 High R&D intensity 0.01 0.02 0.1 High Internationalization 0.25 0.5 0.7 Reduced Scope 3 emissions 2 0 -0.4 41 The threshold values for the high firm revenue were chosen directly, meaning that the determination of threshold values utilized empirical data and statistical methods to es- tablish robust threshold values. The threshold for full non-membership was set for 2000, while the threshold for full membership was set at 12000. Based on the data, the cross- over point of 4500 was selected between these values. For the high profitability thresh- old values, previous theory and empirical data were utilized in defining anchor points. 0.05 was chosen as the threshold for full non-membership, representing a company’s low operating profit %. On the other hand, 0.12 was chosen as the threshold for full membership, which was theoretically suitable for a high operating profit %. A crossover point of 0.1 was chosen between these anchor points, representing the transition be- tween good and poor operating profit. high R&D intensity, the crossover point was chosen to be 0.02, considering the full mem- bership threshold of 0.1 and the full non-membership threshold of 0.01. The selection of anchor points was based on raw data in collaboration with theory, which provided insights into the positioning of the points. The data required an understanding of the differences between manufacturing sectors, such as ICT and chemical and mineral indus- tries, where the ICT sector is known for its high R&D intensity, while the chemical and mineral sectors have a more moderate intensity (Kayal, 2016). The choices were also influenced by Finland’s goal to achieve a 4% R&D intensity (Deschryvere et al., 2021). The threshold values for the condition of high internationalization were based on empir- ical data and statistical insights. The threshold value for full non-membership was set at 0.25, and the threshold value for full membership was set at 0.7. The crossover point was set at 0.5. The dataset primarily consisted of highly international companies, so this factor had to be considered when selecting anchor points, combining empirical data on internationalization. Finally, the threshold values for the outcome variable, the reduced Scope 3 emissions, were determined. The full membership threshold was set at -0.4, as the goal is to reduce emissions, making the change predominantly negative. In turn, the full non-membership threshold was set at 2, and the crossover point was set at 0. This 42 theoretically guided crossover point divided the cases into those that have reduced Scope 3 emissions and those whose emissions have increased. The selection of anchor values was influenced by empirical data and theoretical knowledge, considering that companies aim for carbon neutrality and compliance with the Paris Agreement on cli- mate change (Finnish Ministry of the Environment, 2025). Thus, membership thresholds were established for all conditions and the outcome, enabling the calibration into fuzzy set values. The calibration was performed with the R programming language utilizing the QCA package, which produced the results shown in Table 5. Table 5. Values of fuzzy set calibration. Company High firm revenue High profitability High R&D intensity High Internationalization Reduced Scope 3 emissions Fiskars Group 0.019 0.580 0.217 0.603 0.059 Gargotec 0.301 0.211 0.567 0.582 0.699 Glaston 0.006 0.024 0.686 0.474 0.613 Huhtamäki 0.301 0.349 0.022 0.828 0.797 Kemira 0.138 0.285 0.059 0.472 0.970 Kone 0.912 0.930 0.321 0.821 0.749 Konecranes 0.216 0.292 0.144 0.471 0.838 Metso 0.397 0.720 0.143 0.977 0.768 Metsä Board 0.054 0.994 0.008 0.171 0.253 Neste 0.996 0.587 0.009 0.030 0.793 Nokia 0.999 0.115 0.998 0.964 0.610 Nokian Tyres 0.026 0.967 0.396 0.078 0.224 Outokumpu 0.739 0.022 0.006 0.154 0.824 Purmo 0.012 0.188 0.025 0.006 0.943 Stora Enso 0.895 0.670 0.118 0.089 0.894 UPM 0.902 0.994 0.587 0.203 0.476 Vaisala 0.008 0.913 0.982 0.906 0.041 Valmet 0.461 0.338 0.500 0.798 0.711 Wärtsilä 0.576 0.164 0.660 0.931 0.434 43 4.4 Analysis and results After calibrating the data into a fuzzy set, the actual analysis was initiated to obtain con- figurational results. First, it was examined whether there were any necessary single con- ditions, meaning that a certain condition must be present in a positive or negative out- come. For a single condition to be necessary for the outcome, its consistency should be above 0.9 (Ragin, 2009). The conditions were analyzed based on positive and negative outcomes. Table 6 presents the results of the necessity analysis, showing that the highest consistency was 0.857 for the positive outcome (reduced Scope 3 emissions). This indi- cates that the absence of capability in innovation is not entirely necessary. However, the coverage is high, meaning that the condition is present in many cases. All in all, no nec- essary conditions were found in the results of the necessary analysis. Table 6. Necessity analysis for single conditions. Condition Reduced Scope 3 emissions Non-Reduced Scope 3 emissions Consistency Coverage Consistency Coverage High firm revenue 0.567 0.834 0.488 0.447 ~High firm revenue 0.624 0.661 0.819 0.541 High profitability 0.493 0.617 0.769 0.601 ~High profitability 0.681 0.825 0.510 0.386 High R&D intensity 0.374 0.679 0.653 0.740 ~High R&D intensity 0.857 0.798 0.717 0.417 High internationalization 0.590 0.722 0.692 0.528 ~High internationalization 0.615 0.761 0.636 0.492 The next step was to examine whether there is any set of conditions that is necessary for a positive or negative outcome. The analysis was conducted using the logical "OR" and with the QCA package in R (programming language). The QCA package utilized the Super Subset method, which was used to identify the necessity of condition sets for the out- comes. Four sets of conditions were found for the positive outcome that could be con- sidered necessary. However, by utilizing both theoretical and empirical insights, it was concluded that the necessity of the sets of conditions had a trivial impact on the 44 outcomes (Goertz, 2006). On the other hand, no set of conditions was found to be nec- essary for the negative outcome. Next, a sufficiency analysis was performed, and truth tables were constructed to simplify and identify causal combinations to increase robustness. First, a positive truth table was created, identifying the combinations of conditions leading to the outcome. The con- sistency threshold was set at 0.8, which was used to filter the truth table to identify all combinations that are sufficient for the outcome. Theory supports the principle that the consistency threshold must be at least 0.75 (Schneider & Wagemann, 2012). Therefore, weak or so-called unreliable combinations were eliminated. From Table 7, two combina- tions resulted in an outcome of 0, as their consistency was too low compared to the threshold value. The frequency threshold was set to 1, which is common in small sample sizes (10-50), as it helps maintain a sufficiently large number of combinations in the analysis, making them significant. The final threshold set was the proportional reduction in inconsistency (PRI), with a value of 0.6. The goal of this was to eliminate inconsistency from the results. Due to the high consistency of values, the PRI threshold did not cause any changes to the positive truth table results, as can be seen in Table 7. However, low PRI values were considered, and observations were made based on them in the final configurations. The robustness of the results was also examined by utilizing the negative truth table, which assessed the asymmetry of the causal combinations. Checking the robustness of the analysis enhances the credibility of the analysis and removes randomness (Schneider & Wagemann, 2012). 45 Table 7. Truth table for positive outcome. High revenue High profitability High R&D intensity High foreign revenue Outcome n Consistency PRI Cases 1 0 0 0 1 1 1.000 1.000 Outokumpu 1 1 0 0 1 2 1.000 1.000 Neste, Stora Enso 1 1 0 1 1 1 1.000 1.000 Kone 1 1 1 0 1 1 0.937 0.274 UPM 0 0 1 1 1 2 0.929 0.745 Gargotec, Valmet 0 0 0 1 1 1 0.916 0.815 Huhtamäki 0 0 0 0 1 3 0.915 0.844 Kemira, Konecranes, Purmo 0 0 1 0 1 1 0.899 0.583 Glaston 1 0 1 1 1 2 0.854 0.503 Nokia, Wärtsilä 0 1 0 1 1 2 0.848 0.646 Fiskars Group, Metso 0 1 0 0 0 2 0.644 0.286 Metsä Board, Nokian Tyres 0 1 1 1 0 1 0.608 0.071 Vaisala Next, the truth tables were then minimized by using Boolean algebra, resulting in sim- plified solutions. The solutions were obtained in conservative, parsimonious, and inter- mediate forms. Using previous theory, the intermediate form was selected from the re- sults, which was used to obtain the final minimized configurations. Finally, four configu- rations from the intermediate solutions were identified as positive outcomes, leading to a reduction in Scope 3 emissions. These configurations were classified as S1-S4 according to Table 8. In turn, four configurations also emerged from the negative outcome inter- mediate solutions that did not lead to a reduction in Scope 3 emissions. These, in turn, were classified as N1-N4. The final configurations also utilized the parsimonious solution, which helped identify the core and peripheral causal conditions. These final configurations were visually marked in Table 8. The core causal conditions (present and absent) were determined from the parsimonious and intermediate solutions, meaning that the condition had to be found in both solutions. The present core causal conditions were marked with a large 46 dark circle, while the absent core causal conditions were marked with a large circle con- taining a cross inside. If a condition was found only in the intermediate solution, the condition was assigned peripheral. Peripheral causal conditions were marked in the same way, except that small circles were used. Meaning that the present causal condi- tion was marked with a small dark circle, and the absent causal condition with a small circle containing a cross inside. From Table 8, it can also be seen that there are areas with no markings, indicating that the presence of the condition is not significant for the con- figuration. Table 8. Configurations of reduced Scope 3 emissions. Condition S1 S2 S3 S4 N1 N2 N3 N4 Configurations leading to reduced Scope 3 emissions Configurations leading to non-reduced Scope 3 emissions High firm revenue High profitability High R&D intensity High interna- tionalization Strong cases Huhtamäki, Kemira, Konecranes, Purmo Neste, Outokumpu, Stora Enso Gargotec, Nokia, Valmet, Wärtsilä Fiskars Group, Kone, Metso Vaisala Metsä Board, Nokian Tyres UPM Glaston Consistency 0.89 0.90 0.82 0.88 0.97 0.84 0.94 0.86 Raw coverage 0.52 0.37 0.30 0.32 0.35 0.42 0.23 0.24 Unique coverage 0.18 0.20 0.05 0.09 0.15 0.23 0.08 0.06 Solution consistency: 0.84 Solution consistency: 0.84 Solution coverage: 0.89 Solution coverage: 0.72 Core causal condition (present) Peripheral causal condition (present) Core causal condition (absent) Peripheral causal condition (absent) 47 4.4.1 Configurations leading to reduced Scope 3 emissions Four configurations (S1-S4) were found to lead to a reduction in Scope 3 emissions. The consistency of the positive solutions was 0.84, and the coverage was 0.89. The results were therefore reliable due to the high consistency, and the high coverage indicated the similarity of the companies in the dataset, meaning that the obtained configurations covered a large portion of the selected companies (Rubison et al., 2019). S1 was the first configuration that led to the desired outcome, where low R&D intensity was considered a core condition and low firm revenue as a peripheral condition. It in- cluded four companies: Huhtamäki, Kemira, Konecranes and Purmo. These companies were able to achieve a reduction in Scope 3 emissions without making significant invest- ments in R&D or peripherally being large in terms of revenue. High profitability and in- ternationalization did not affect the outcome of this configuration, so they were left empty. Huhtamäki and Kemira are good examples of improved Scope 3 emissions reduc- tion with low R&D intensity, where sustainability is already at the core of their operations and mindset, making extensive R&D investments unnecessary to achieve emission re- ductions. As a packaging company, Huhtamäki has leveraged an efficient recycling sys- tem and collaboration with value chain partners, significantly reducing emissions from the end-of-life processing of sold products. Kemira, on the other hand, has focused on the sustainability of its suppliers of purchased goods and services, significantly reducing Scope 3 emissions. The second configuration that led to a reduction in Scope 3 emissions was S2, where high revenue was considered a core condition alongside peripheral low internationaliza- tion. High profitability and R&D intensity had no impact on the results of this configura- tion. Three companies fell under this configuration: Neste, Outokumpu and Stora Enso. All these companies primarily operate domestically and have significantly reduced their Scope 3 emissions. With high revenue, these companies have had financial slack re- sources to focus on sustainability, including Scope 3 emissions. At the same time, large companies are expected to make clear improvements and set targets for emission 48 reductions. Neste’s Scope 3 emissions are mainly composed of the end of supply chain, i.e., the use of sold products, and therefore, they have set a goal to reduce the emission intensity of sold products by 50%. Although high R&D intensity has not been required to reduce emissions in this configuration, Neste has successfully decreased the indirect emissions from sold products with its renewable products. The third configuration was S3, which led to a reduction in Scope 3 emissions. The posi- tive outcome was achieved with low profitability as a core condition and high R&D in- tensity as a peripheral condition. In the S3 configuration, high revenue and internation- alization conditions were ambivalent for the outcome. A total of four companies be- longed to this configuration. These companies were Gargotec, Nokia, Valmet and Wärt- silä. All the companies manufacture technically demanding products, which naturally re- sults in high R&D intensity, while their profitability remains at a low level. Manufacturers of demanding products naturally take emissions into account in their products and strive to minimize the emissions generated during use. A good example is Nokia, which has continuously developed the energy efficiency of its products and promoted the availa- bility of renewable energy among its customers to reduce the Scope 3 emissions gener- ated using sold products. The final configuration that led to a reduction in Scope 3 emissions was S4. This config- uration included three companies, which were Fiskars Group, Kone and Metso. The con- figuration had high internationalization and low R&D intensity as the core conditions, with high profitability as a peripheral condition, leading to a positive outcome. High rev- enue was considered insignificant for the outcome. Kone and Metso operated on a highly global and profitable scale and have gained a competitive advantage through sustainable FSAs, thereby also reducing Scope 3 emissions. High internationalization requires signif- icant resources for value chain activities, including collaboration with numerous stake- holders. Therefore, Kone and Metso have successfully achieved sustainable operations through their active and demanding upstream and downstream activities, which are re- flected in their positive emissions reduction results. Metso has set emission reduction 49 targets for its suppliers, enabling them to reduce emissions from purchased goods and services. On the other hand, Kone has achieved emission reductions through systematic communication with partners and by improving the energy efficiency of their products, which has reduced emissions from the use of sold products. 4.4.2 Configurations leading to non-reduced Scope 3 emissions Four different configurations were also found for the negative outcome, meaning solu- tions that do not lead to Scope 3 emission reductions, and these were labelled N1-N4. The solutions had a consistency value of 0.84 and a coverage value of 0.72, which was lower than the coverage of the positive solutions, while the consistency remained the same in both outcomes. Negative configurations included significantly fewer cases than the positive ones. N1 was the first configuration that did not lead to a reduction in Scope 3 emissions. In this configuration, the core conditions were high profitability and R&D intensity, while the peripheral conditions were low revenue and high internationalization. Vaisala was the only case that corresponded to the N1 configuration. As one of the smallest compa- nies, Vaisala’s high profitability, R&D intensity, and internationality have not been suffi- cient to reduce Scope 3 emissions. The result has been influenced by the evolving data tracking and collection of Scope 3 emissions, which have unfairly impacted Vaisala’s out- come. Especially when considering Vaisala's positive emission reduction targets, which they are actively pursuing to reduce Scope 3 emissions. The second configuration that failed to reduce Scope 3 emissions was N2, where the core conditions were low revenue, low internationalization, and high profitability. The low R&D intensity was considered a peripheral condition for the outcome. The configuration included two cases, which were Metsä Board and Nokian Tyres. As Nokian Tyres manu- factures tyres for vehicles, it is understandable that there are significant emissions re- sulting from the use of sold products. In 2023, approximately 90 % of Nokian Tyres’ Scope 50 3 emissions consisted of emissions from the use of sold products. Therefore, it is difficult as a tyre manufacturer to influence emissions related to the use of products, as the very nature of driving a vehicle inherently generates emissions. In the N3 configuration, the core conditions were high profitability, high R&D intensity, and low internationalization. In turn, the only peripheral condition was high turnover, and UPM was the only case company in N3 configuration. Considering the company Metsä Board from the previous N2 configuration and UPM from the N3 configuration, these are two forest industry companies that have not succeeded in reducing Scope 3 emissions. Both forestry companies have achieved high profitability, but their revenue has not grown significantly, and their Scope 3 emissions results have also been weak. UPM has also invested in R&D intensity, yet they have not achieved significant improve- ments in Scope 3 emissions. Finally, configuration N4 did not lead to a reduction in Scope 3 emissions. The core con- ditions of the N4 configuration were high R&D intensity and low internationalization. The peripheral conditions were low revenue and low profitability. Glaston was the only company included in this configuration. Glaston appeared in both positive and negative configurations, meaning it should be examined critically. From a theoretical perspective and based on previous examples, Glaston was placed only on the side of the negative outcome (N4). This is influenced by the absolute variability of emissions, which has been observed in Glaston’s operations and is expected in the future as well. Therefore, Glaston was placed on the negative side of the configurations, and the company can thus be considered to have failed in reducing its Scope 3 emissions. Although, with high R&D intensity and emission targets, Glaston will be able to reduce Scope 3 emissions arising from the value chain in the future. 51 5 Discussion The aim of this study was to examine the relationship between the conditions of financial slack, R&D investments, size of the company, and internationalization and answer the research question: What configurational analysis of the manufacturing sector can reveal about path- ways to reduced Scope 3 emissions? The aim was also to see the interaction of these factors and influence companies’ indi- rect value chain-related Scope 3 emissions. The configurational analysis yielded interest- ing results, further enriched by the diversity of the case companies. The fsQCA analysis used in this study effectively highlighted complex causal relationships, which are com- mon when examining the factors related to combinations. The analysis identified config- urations that led to reduced Scope 3 emissions and, conversely, those solutions that did not yield the desired outcome. This study contributes to the literature on corporate environmental management in four ways. Refines the resource-based view by showing how financial slack, R&D, and inter- nationalization interact in the transition to reduce Scope 3 emissions, offers practical insights for companies aiming to reduce indirect Scope 3 emissions across their value chains, provides new empirical evidence on the drivers and limitations of Scope 3 emis- sions in the manufacturing sector, and demonstrates the usefulness of fsQCA in captur- ing configurational causality. 5.1 Theoretical contributions Financial slack, specifically the conditions of a firm’s revenue sufficient use in this study, contributed to the reduction of Scope 3 emissions. The theory supports the argument that the size of a company’s revenue positively impacts its environmental management, 52 which in turn can influence vario