Banking in Transition: The Effect of FinTech on Incumbent Banks in Sweden Sara Kristoffersson and Jakob Vilhelmsson Supervisor: Johan Stennek Master’s thesis in Economics, 30 hec Spring 2025 Graduate School, School of Business, Economics and Law, University of Gothenburg, Sweden Abstract This thesis examines the impact of Financial Technology (FinTech) on the market power and financial performance of incumbent banks in the Swedish banking sector. This study utilizes a panel dataset covering six major banks in Sweden from 2010 to 2023 and applies fixed effects regression model to examine the impact of FinTech presence, proxied by venture capital investments, on market power and financial performance. Market power is measured by the Lerner index, and financial performance is measured through return on assets and return on equity. The results indicate that after the implementation of the PSD2 directive, increased FinTech presence is associated with a decrease in incumbent banks’ market power. Additionally, the results indicate that FinTech presence has a negative impact on the financial performance of incumbent banks. These findings suggest that FinTech represents an emerging competitive force, and that regulation may play a key role in shaping its future influence on the banking industry. Keywords: FinTech × Market Power × Financial Performance × PSD2 × Venture Capital Investments × Swedish Banking Sector Table of Contents 1. Introduction ....................................................................................................................... 1 1.1 Research questions ....................................................................................................... 2 2. Literature Review ............................................................................................................. 2 3. Theoretical Framework ................................................................................................... 6 3.1 Banking operations ....................................................................................................... 6 3.2 Regulatory Policy ......................................................................................................... 6 3.3 Disruptive Theory ........................................................................................................ 7 3.4 Innovation and Incentives ............................................................................................ 8 3.5 Market Power ............................................................................................................... 9 4. Data and Methodology ..................................................................................................... 9 4.1 Measurement of Market Power and Financial Performance Measures ................. 10 4.1.1 Estimation of Market Power ................................................................................... 10 4.1.2 Justification for Total Assets as Bank Output ......................................................... 13 4.2 Panel data analysis ....................................................................................................... 14 4.2.1 Model specifications ............................................................................................... 15 4.2.2 Variable Description ............................................................................................... 16 4.3 Data Sample and Time Frame .................................................................................... 20 4.3.1 Data Collection ........................................................................................................ 21 4.4 Robustness and Potential Endogeneity Concerns ..................................................... 22 4.5 Limitations .................................................................................................................... 23 5. Result ............................................................................................................................... 24 5.1 Descriptive Statistics .................................................................................................... 24 5.2 Regression Results ........................................................................................................ 26 5.2.1 Market Power Analysis ........................................................................................... 26 5.2.2 Financial Performance Analysis .............................................................................. 28 6. Analysis ........................................................................................................................... 30 6.1 Market Power ............................................................................................................... 30 6.2 Financial Performance Measures ............................................................................... 31 7. Conclusion ....................................................................................................................... 33 References ............................................................................................................................... 34 Appendix A ............................................................................................................................. 43 Appendix B .............................................................................................................................. 44 B1. Robustness Analysis Using Number of FinTech Launches ..................................... 44 B2. Instrumental Variable Analysis ................................................................................. 46 1. Introduction In recent years, the Swedish banking sector has attracted attention, mainly due to their profits under a period of economic uncertainty. These profits, coupled with the sector’s high degree of concentration around a few actors, have intensified the debate concerning the level of competition in the banking sector, and the market power held by incumbents (SVT Nyheter, 2024). Historically, Sweden has been at the forefront of the FinTech sector with a growing number of FinTech firms emerging and entering the market, leveraging new technologies to deliver financial services and products in new innovative ways (The National Board of Trade, 2024). This development raises important questions about how FinTech is influencing the competitive landscape of the Swedish banking sector and how incumbent banks adapt to these emerging challenges. FinTech represents a promising development in the financial sector, with the potential to enhance competition and offer more tailored, efficient, and cost-effective alternatives to customers in an otherwise concentrated market. However, given the financial sector’s critical role in maintaining economic stability, the rapid pace of digital innovation introduces new and complex risks that must be carefully managed (Finansinspektionen, 2020). In 2018, the Payment Service Directive 2 (PSD2), which aims to modernize Europe's payment services for the benefit of consumers and businesses, was implemented in Sweden. The directive also aims to level the playing field and contribute to a more integrated and efficient European payment market. PSD2 enabled data sharing from banks to third party providers, under the consent of the customer, which facilitated FinTech development. This thesis aims to explore how the presence of FinTech in Sweden has influenced the market power, measured by the Lerner index, within the Swedish banking sector during the period 2010 to 2023. Further, it seeks to analyze how FinTech presence have influenced the profitability of Swedish banks, measured through return on assets and return on equity. The thesis also considers the PSD2 directive to evaluate whether regulatory policies have impacted FinTech presence and altered its influence on incumbent banks. Our primary hypothesis is that the growing presence of FinTech has a significant effect on both the market power and financial performance of incumbent banks in Sweden, with the expectation that increased FinTech presence results in a reduction in bank’s market power and profitability. 1 1.1 Research questions This thesis aims to examine the evolution of FinTech and its implications for the Swedish banking sector by addressing the following research questions: 1. Does FinTech affect the market power of incumbent banks in the Swedish banking sector? 2. Does FinTech affect financial performance measures of incumbent banks in the Swedish banking sector? 2. Literature Review FinTech has become an influential actor in the global financial sector, offering alternatives to traditional banking services. The existing literature mostly focuses on larger or more established markets, while certain regions and sectors remain relatively unexplored, highlighting the need for research across different contexts. FinTech firms tend to focus on the most profitable areas of banking, for example lending to borrowers with limited market power, payment services, and financial advice. This raise concerns that traditional banks may be demoted to financial utilities, offering only essential banking functions while FinTech firms capture more lucrative market segments. However, incumbent banks are not passive in this transition towards a more digital and innovative banking and financial environment. Banks have begun to adapt by integrating FinTech-driven innovation into their own business to stay competitive (Navaretti et al., 2018). The Swedish Central Bank states that FinTech contributes to increased competition within financial markets (Riksbanken, 2022). FinTech companies typically operate with newer and more modern systems compared to traditional financial institutions, enabling them to deliver innovative solutions for financial services more efficiently. As a result, FinTech firms can offer the same or similar services as traditional actors but at a lower cost or with more streamlined and rapid processes. Additionally, the Financial Stability Board (2019) highlights that the competitive pressure from FinTech and its impact on market dynamics and profitability should be closely monitored by regulators and supervisors. Although banks have previously faced competitive pressure from other entities, the rise of FinTech is particularly concerning due to 2 the rapid technological and digital advancements made in the past few years. Furthermore, the accelerated evolution of business models and a greater focus on technology, among both new entrants and established firms, may introduce new forms of operational risk, especially when such technologies are deeply integrated into firms’ core functions. New technologies such as big data, cloud computing and machine learning have contributed to changes in how financial services are offered, and FinTech firms are especially effective at leveraging these technologies. This capability enables FinTech to provide faster, more efficient and automated services, providing FinTech firms with a comparative advantage. In contrast, incumbent banks often depend on more relationship-based information and older legacy systems, which can limit flexibility and speed. Nevertheless, traditional banks still retain advantages in the form of higher consumer trust and operational experience (Navaretti et al., 2018). Furthermore, Stulz (2019) and the OECD (2020) observe that although FinTech has reshaped the structure, provision and consumption of financial services, these firms have yet to establish a dominant market position. Navaretti et al. (2018) argue that regulation has been, and remains, a fundamental factor in the development of the FinTech sector, almost as important as technological innovation itself. A key regulatory challenge is the tradeoff between competition and financial stability. While efforts to enhance competition in financial markets might suggest a lighter regulatory approach compared to the governing of traditional financial services, the rapid expansion of FinTech also introduces concerns in terms of financial stability that must be carefully addressed. Several studies have explored how FinTech reshapes the market power and financial performance of incumbent banks. Table 1 outlines core findings of some of the most recent studies we find relevant for this thesis. 3 Table 1. Core Findings from Recent Studies Author Year Country FinTech Presence Market Power Financial Findings Performance - Aggregate index of FinTech company development - Technological U-shaped relationship innovation Interest Rate between FinTech and Qi et al. 2022 China development index Spread - the market power of - Business traditional banks innovation development index of FinTech FinTech lending has a Cross- negative significant Cuadros-Solas et al. 2024 Country FinTech Lending Lerner Index - effect on market power of traditional banks Market Power: No ROA, NIM, significant effect FinTech Assets to NII to total Financial Saklain 2024 Australia Bank Assets Lerner Index revenue and Measure: Negative Loan to Asset significant effect on NII ratio to total revenue and Loan to Asset ratio Number of FinTech NIM, ROA, Consistent negative firms registered with ROE and Yield impact of FinTech on all Phan et al. 2020 Indonesia - the Indonesian on Earning four bank profit FinTech Association Assets. measures Total capital ratio, non- performing loans to gross FinTech development Zhao et al. 2021 China Financial innovation - loans ratio, cost significantly affects to income ratio, various aspects of banks’ performance ROA and liquid assets to total deposits ratio Economic Negative relationship Yao and Song 2021 China FinTech application - capital of between FinTech and index market risk bank performance FinTech credit Nguyen et al. 2021 Cross- FinTech credit - ROA, ROE, negatively affects Country and the Z-score profitability The U-shaped relationship between between FinTech and the market power of traditional banks identified by Qi et al. (2022) indicate that FinTech companies may weaken the market power of traditional banks at first, however, in later stages traditional banks adapt and regain their 4 market power. Cuadros-Solas et al. (2024) not only find that FinTech lending reduces market power but also uncovers an institutional dimension: countries with weaker legal frameworks see more pronounced effects. To further assess the robustness of the findings, the authors conduct an instrumental variable analysis using the percentage of rural population in each specific country as an instrument for FinTech lending. The results are in line with the results from the main analysis. Moreover, an alternative measure of FinTech is employed: Investments in the FinTech sector per capita and these results are also in line with the main analysis. In contrast, Saklain (2024), finds no strong evidence of FinTech disrupting market power. This raises questions about country-specific factors, such as regulation or consumer preferences. Partial evidence suggests that certain accounting performances of banks might decline due to the growth of FinTech firms. In the profit-oriented studies, Phan et al. (2020) highlight an important heterogeneity: younger banks are more agile and better able to adapt to FinTech pressure, despite the overall negative profitability trend. This generational divide within the banking sector is worth noting for policy and strategic planning. Yao and Song (2021) offer a more risk-focused angle by linking FinTech activity to market risk capital. Their robust findings show not only a negative relationship between FinTech and performance but also that asset size moderates this effect. This points to a potential scalability issue: larger banks may face greater difficulty adjusting to FinTech-led changes. Zhao et al. (2021) conduct a detailed heterogeneity analysis, finding that large state-owned and policy banks are disproportionately affected by FinTech. Interestingly, while profitability, asset quality and earning power decline due to FinTech industry development, these banks show improved capital adequacy and management efficiency, suggesting a trade-off between stability and performance metrics. Taken together, the majority of the results from the studies described show a negative effect of FinTech on market power and different aspects of financial performance. 5 3. Theoretical Framework This section addresses economic theories that explain competition and innovation in the banking sector. By incorporating perspectives on market structure, technological disruption and strategic firm behavior the framework helps assess how FinTech impact incumbent banks. 3.1 Banking operations Banks operate as a middleman that channels funds from savers to borrowers. In this process banks facilitate indirect finance, where savers deposit funds with the bank, which are later allocated to borrowers including households, entrepreneurs and individuals. While individuals could engage in direct finance, lending directly to borrowers, banks offer the advantage of scale, risk management and information processing. Making them a more effective and accessible alternative. Banks generate profit from this core activity by earning the spread between the interest charged on loans and returns offered on deposits (Van Hoose, 2009). Banking operations can be viewed through different perspectives. One perspective proposed by Sealey and Lindley (1977) is the intermediation approach, which states that banks are primarily seen as transforming liabilities into earning assets. The authors state that the assets of the bank should be considered as outputs and deposits to be treated as financial inputs used in the intermediation process alongside inputs like capital and labor. Banks collect deposits and transform them into assets such as loans and investments, which are considered as outputs. The model emphasizes the transformational role of banks, converting deposits into long-term assets generating profit and value. Another perspective, provided by DeYoung and Rice (2004), highlights the growing importance of non-interest income, such as fee-based services, including transaction services, payment services, investments and wealth management. This development is partly explained by technological advancements. While interest income through traditional banking activities remains a key component in the profits of banks, the role of other income sources has expanded. These emerging revenue streams are becoming increasingly important to banks and their business models in a more digitalized world. 3.2 Regulatory Policy Swedish banks are subject to extensive regulation requirements at both the EU and national levels, including requirements for capital, liquidity and diversification in investments 6 (Riksbanken, 2024). Regulations related to capital requirements and licenses influences both market competition and structure as they can directly and indirectly affect the number of firms operating in the market by limiting their ability or incentive to compete (Swedish Agency for Economic and Regional Growth, 2017). Nonetheless, regulation plays a crucial role in the bank sector, ensuring stability, maintaining trust and protecting consumers. The regulatory framework has not been able to keep up with the rapid rise of FinTech, which could enable FinTech to avoid regulation. However, this rapid rise also entails that FinTech is at risk of being excluded from essential payment infrastructures. For instance, in 2022, FinTech companies were not yet granted access to Sweden’s central payment system (Riksbanken, 2022). In their annual FinTech report, The Swedish FinTech Association (2024) identifies that FinTech companies experience heavy regulatory burden. Additionally, the FinTech association claims that FinTech companies are not able to compete on equal terms in the financial market. They state that traditional actors have regulatory advantages that hinder the ability of new actors to enter the market, for example in the payment market and house mortgage market. This creates uncertainty and deters innovation and entry due to difficulties in the handling of complex requirements and regulations. The report emphasizes that creating a regulatory landscape which facilitates entry, while simultaneously ensuring effective regulation is important to promote market entry and promote competition in the financial sector. 3.3 Disruptive Theory The disruptive innovation theory, introduced by Christensen in 1997, states that new entrants could disrupt established incumbents by initially targeting overlooked market segments with simpler and more accessible products and services (Christensen et al., 2018). Disruptive innovations occur when new firms introduce simpler solutions at a lower cost that initially serve niche or underserved customer segments but eventually expand and challenge incumbents. Incumbents could fail to respond properly to disruptive innovations because of their reliance on already in-place technologies and business models. Embracing innovation might undermine its own profitable products, while ignoring it increases the risk of becoming obsolete when disruptors grow (Christensen, 1997). According to Christensen et al. (2015) disruptive innovations most often begin as small-scale experiments and tend to prioritize getting the business model just right, rather than focusing on 7 the product. When they succeed, they climb from the low end of the market to the mainstream and can erode the incumbent firms’ markets share and their profitability. This process can take time, and incumbents can get quite creative in the defense of their established franchises. This concept is relevant to this thesis, as FinTech companies often begin as small-scale initiatives offering more user-friendly and efficient alternatives to traditional banking services. These characteristics align closely with the theory of disruptive innovation. Moving on to a similar theory, Spencer and Kirchhoff (2006) present Schumpeter’s theory of creative destruction from 1942. The main argument of the theory is that all important changes in the economy are initiated by innovations that slowly work their way into the economy, replacing outdated products and processes. The theory states that this process occurs continuously, as innovation enters the market, it disrupts the old ways of doing things and creates new opportunities and industries. 3.4 Innovation and Incentives Aghion and Howitt (1992) present a theory of endogenous economic growth based on creative destruction, a process where new innovations replace older more obsolete ones. In their model, growth occurs when firms invest in research and development which generate vertical innovations. Vertical innovations are innovations that improve product quality and replace older ones. This process eliminates the monopoly rents of previous innovations, thus reducing the incentives for current innovation for incumbents, known as the creative destruction effect. This process mirror Schumpeter ideas of growth through cycles of innovation that simultaneously create and destroy value. The model also demonstrates that the expectation of future innovations can deter current investments by incumbents due to reduced expected monopoly rents and rising competition for resources. Who has the incentive to innovate differs between the views of Schumpeter and Aghion and Howitt (1992). Schumpeter argues that the primary incentive to innovate lies with the incumbent firms, particularly those holding a monopolistic or oligopolistic position. Their market power creates incentives to innovate and enables sustained R&D. Incumbents innovate to strengthen and maintain their position in the market. In contrast, Aghion and Howitt (1992) argue that the incentive to innovate lies with the smaller entrant under a regime of creative destruction. Innovation becomes a tool of market entry and a way to challenge incumbent firms 8 and temporarily earn profits. This business-stealing effect creates incentives for outside contenders to innovate. This view also accounts for intertemporal dynamics, recognizing that incentives to innovate today depend on expectations about future competition and technological progress. For incumbents these expectations can create disincentives to innovate as future innovation might erode their profits. Further, incumbents also face costs of replacing older capital and technologies, which can further discourage R&D investments. 3.5 Market Power According to Carlton and Perloff (2005) a firm has market power if it is profitably able to charge a price above that which would prevail under perfect competition, which is usually taken to be marginal cost. The market power of banks arises mainly due to asymmetric information, switching costs, barriers to entry and network externalities (Carletti et al., 2024). In terms of asymmetric information, banks’ acquisition of borrower’s creditworthiness in either screening or monitoring allows incumbents to exercise market power over outside banks, due to their information advantage (Carletti et al., 2024). The rise of new technology has decreased this information entry barrier and made it possible to decrease the market power exercised from asymmetric information. Another important source of bank market power is switching costs. When moving from one bank to another, customers incur costs of different kinds related to the move. As a result, consumers might stay with the same bank for a long time and enable banks to “lock-in” these consumers and exert market power. Finally, network effects are another source of market power for banks. These effects imply that the value of banks services increases as more customers use them. Network effects have gained more attention due to digital platforms and the rise of FinTech, which rely even more on network effects (Carletti et al., 2024). 4. Data and Methodology This section describes the method, and the data utilized to be able to investigate how the rise of FinTech influences the market power and financial performance of incumbent banks. 9 4.1 Measurement of Market Power and Financial Performance Measures This thesis utilizes the Lerner index to quantify market power over the period 2010 to 2023. The Lerner index compares the price of a good to its marginal cost. In conditions of perfect competition, price equals marginal cost, resulting in a Lerner index value of zero. The index ranges from 0 to 1, with values approaching 1 indicating a higher degree of market power, characterized by firms setting prices above marginal cost, thereby reducing output and consumer welfare. Conversely, values approaching 0 suggest a competitive market environment with limited market power, where increased price competition benefits consumers and enhances overall welfare for the consumer (Elzinga & Mills, 2011). In addition, this study investigates the impact of FinTech on banks’ financial performance. To assess this relationship, key financial performance measures, return on assets and return on equity, will be employed. Return on assets reflects the efficiency with which a company utilizes its assets to generate profits, while return on equity measures a firm's ability to produce earnings from shareholders’ equity. By employing both return on assets and return on equity in the analysis it enables a nuanced understanding of how FinTech influences bank performance. While return on assets captures operational adjustments and shifts in how banks deploy their assets in response to competitive pressure, return on equity reflects the broader impact on profitability and the shareholders’ value. Given that FinTech firms affect the financial market in many ways, for example through product innovation and customer-centric service models, analyzing both metrics offers a more complete picture of banks’ financial responses and strategic repositioning. 4.1.1 Estimation of Market Power The Lerner index is computed by dividing the difference between each bank’s price and marginal cost by its price, demonstrated in Equation 1. 𝑃 −𝑀𝐶 𝐿𝑒𝑟𝑛𝑒𝑟 𝐼𝑛𝑑𝑒𝑥 = !,# !,#!,# 𝑃 (1) !,# In order to estimate the Lerner index, the price is proxied by dividing total income by total assets, in accordance with Turk Ariss (2010), Fosu et al. (2017), Clark (2018) and Cuadros- 10 Solas et al. (2024). This ratio captures how much the bank earns per unit of asset, which aligns with the notion that banks use their assets to produce earnings. Given the frequent usage of these proxies in previous banking literature, this way of proxying unobservable data ensures consistency with previous literature. The chosen price proxy represents the average revenue per unit of assets and includes both interest and non-interest income, which is in line with the shift towards more service-based banking operations. The Lerner index is a frequently used measure to assess the market power in the banking industry (Huang et al., 2017). According to White (2015) the Lerner index has become the standard measure of market power among economists. Several studies in banking have utilized the Lerner index as a measure of market power (see e.g. Berger et al, 2008; Cuadros-Solas et al., 2024; Saklain, 2024; Leroy & Lucotte, 2017; Beck et al., 2013; Koetter et al., 2012; Weill, 2013). The advantage of using the Lerner index is that it is possible to compute a value for each entity in the sample, while other methods such as the Rosse-Panzar H-statistic only compute a value for the entire sample (Huang, 2017). This also enables a more dynamic analysis as changes over time can be analyzed, which suits the analysis of FinTech growth, and technological change in general, as it emerges over time. There are some limitations to the Lerner index. Shaffer and Spierdijk (2017) highlight the uncertainty associated with the various proxies for in- and output. The use of proxies is a necessity to be able to investigate the effect of FinTech on the market power of incumbent banks, which is discussed in Section 4.1.1.1 and 4.1.2. Another limitation is that the Lerner index does not distinguish between market power and pricing behavior driven by cost structures. Specifically, a firm may set prices above marginal cost not because of market power, but to cover substantial fixed costs. This is relevant in the modern, technology-driven economy, where many firms rely on large investments. As Lindenberg and Ross (1981) note, such fixed costs can necessitate pricing closer to average cost, and the presence of investments may also create barriers to entry, introducing challenges in the interpretation of the Lerner index. To compute the Lerner index, the Transcendental Logarithmic Cost Function will be utilized to estimate the marginal cost. This method is widely used in previous banking literature (see e.g. Weill, 2013; Turk Ariss, 2010; Clark, 2018; Cuadros-Solas et al., 2024; Fernández de Guevara, 2005). 11 4.1.1.1 Transcendental Logarithmic Cost Function The Transcendental Logarithmic (Translog) Cost Function has been widely used in previous banking research, particularly when estimating market power through marginal cost and the Lerner index. The Translog cost function extends the Cobb-Douglas model by including interaction and quadratic terms in the model specification which allows for non-constant returns to scale and variable substitution elasticities. The function is a second-order flexible form which allows for more accurate estimations of the cost function compared to first-order forms. Even though second-order forms contain more parameters to be estimated, which could be a potential challenge, it is generally preferred due to its flexibility. Equation 2 presents the Translog cost function employed in this study, for notational simplicity, time and entity subscripts are omitted. The function incorporates a single output represented by the bank’s total assets, and three input variables serving as proxies for labor, capital and deposits. This specification follows the approach adopted by Beck et al. (2013), Berger et al. (2008) and Weill (2011). 𝑇𝐶 1 𝑝𝐿 𝑝𝐷 ln 0𝑝𝐶3 = 𝛽$ + 𝛽%ln (y) + 2𝛽&ln (𝑦) & + 𝛽' ln 0𝑝𝐶3 + 𝛽( ln 0𝑝𝐶3 𝑝𝐿 𝑝𝐷 𝑝𝐿 𝑝𝐷 (2) + 𝛽) ln 0𝑝𝐶3 ln 0 3 + 𝛽* ln (y)𝑝𝐶 ln 0𝑝𝐶3 + 𝛽+ ln (y) ln 0𝑝𝐶3 + 𝜀 In this expression, TC denotes total operating cost, and y represents the total assets for each bank. The input prices are non-observable and are therefore proxied in accordance with previous literature (e.g., Weill, 2013; Leroy & Lucotte, 2017; Turk Ariss, 2010; Berger et al, 2008; Beck et al., 2013). Price of labor (pL) is proxied by personnel costs1 over total assets, price of capital (pC) is proxied by operating expenses over fixed assets, and price of deposits (pD) are the interest expenses over total deposits. The model is normalized by the price of 1 Personnel cost is defined as salaries, wages, bonuses, commissions, changes in reserve for future stock option expense, and other employee benefit costs. Includes any expenses related to employment or retirement benefits, whether paid or deferred, recognized during the period. 12 capital (pC) to maintain linear homogeneity in input prices, as recommended in previous literature (e.g., Shaffer & Spierdijk, 2020; Shamsur & Weill, 2019; Clark et al., 2018). This normalization ensures that a proportional change in all input prices leads to an equivalent proportional change in total costs, holding output constant (Koetter et al., 2012). Based on the Translog cost function the marginal cost is calculated as follows: 𝑇𝐶 𝑝𝐿 𝑝𝐷 𝑀𝐶 = 𝑦 (𝛽% + 𝛽&𝑙𝑛𝑦 + 𝛽* ln 0𝑝𝐶3 + 𝛽+ ln 0𝑝𝐶3 ) (3) The input elasticities are estimated through regressions analysis, capturing the sensitivity of total operating costs to changes in input prices. Specifically, we extract the coefficients associated with total assets, the squared term of total assets, and the interaction term between total assets and each input price. These elasticities form the basis for calculating the marginal cost, consistent with methodologies adopted in prior studies. 4.1.2 Justification for Total Assets as Bank Output Defining bank output is inherently difficult due to the nature of banking services. Unlike traditional industries, banks do not produce a specific unit of output, rather, they provide a range of intermediation services. Moreover, modern banks have diversified beyond traditional interest-earning activities, increasingly generating revenue from fee-based services such as transfers and transactions and other financial products such as wealth management. As competition in the banking sector intensifies, particularly due to the rise of FinTech, banks must adapt by expanding their income sources. Angelini and Cetorelli (2003) argue that a product specific approach would fail to consider the ability of banks to act strategically in various markets, products and services within the finance sector, and therefore potentially bias the estimation of market power. Since banks, in recent years, rely more on service income and fee- based income, these components should be incorporated in the analysis of market power, supporting the total assets approach. The total assets approach offers a more comprehensive representation of the bank’s operations. Under this framework, one unit of output corresponds to one monetary unit of assets managed, and marginal cost reflects the incremental cost associated with expanding the balance sheet. 13 Alternative proxies for bank output, such as interest-earning asset (e.g., Clerides et al., 2015; Wang et al., 2020), were considered for this thesis. Even though the results were generally consistent with the total assets approach, interest earning assets categories are often less clearly defined in financial statements and may require imprecise estimations, increasing the likelihood of measurement error. Given these considerations, this study employs the total assets approach as the primary measure of bank output. This choice ensures alignment with established literature and provides a comprehensive and reliable representation of banking activity in an era marked by FinTech- driven transformation. This method is consistent with prior research, including Cuadros-Solas et al. (2024), Saklain (2024), Weill (2013), Leroy and Lucotte (2017), Beck et al. (2013), Berger et al. (2008) and Hainz et al. (2012). 4.2 Panel data analysis To examine if FinTech presence has an impact on the market power and financial performance in the Swedish banking sector, a panel data analysis with bank fixed effects to control for unobserved differences between banks is employed. The panel data approach is suitable since we examine the impact of FinTech growth over time among several banks in the Swedish banking sector, providing us with a panel data set for six banks over 14 years. The use of fixed effects in our analysis is based on the reasoning that we want to control for bank-specific characteristics. Different banks have different traits such as management styles, risk appetite, historical market position and brand strength, that are unlikely to change significantly over the short run but affect profitability and market power. In contrast, a random effect model assumes that unobserved bank-specific characteristics are uncorrelated with the explanatory variables, which in this case is unlikely, as FinTech presence is likely related to underlying bank strategies and structures. Fixed effects panel data regression analysis relies on several key assumptions. First, the independent variable, in this case, venture capital investments in the Swedish FinTech sector, must be uncorrelated with unobserved factors influencing bank performance. This means that there should be no omitted variable bias, ensuring that all relevant variables affecting market power and financial performance measures are included. Such variables could be those that 14 represent the concentration in the banking sector, the financial market conditions and technology adoption of banks. Another assumption is that there should be no perfect multicollinearity among the explanatory variables. Finally, to account for potential serial correlation within banks over time, clustered standard errors are employed. This addresses the fact that data within the same bank over time are not independent. By clustering at the bank level, the estimated results become more robust (Stock & Watson, 2020). 4.2.1 Model specifications The baseline model, Equation 4, examines the impact of FinTech presence on market power. This model is expanded in Equation 5, to account for the effect of the PSD2 directive, and is further extended in Equation 6 to include the interaction between FinTech presence and the PSD2 directive. All models incorporate bank-specific control variables (Bank) and macroeconomic control variables (Macroeconomic). Additionally, the market power analysis includes net interest margin (NIM) to control for the profitability of banks. The variables are elaborated upon in Section 4.2.2. 𝑀𝑎𝑟𝑘𝑒𝑡 𝑃𝑜𝑤𝑒𝑟!,# = 𝛽$ + 𝛽%𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# +𝛾%𝐵𝑎𝑛𝑘!,# + 𝛾&𝑀𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐# + 𝛾'𝑁𝐼𝑀!,# + 𝜀 (4) !,# 𝑀𝑎𝑟𝑘𝑒𝑡 𝑃𝑜𝑤𝑒𝑟!,# = 𝛽$ + 𝛽%𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# + 𝛿%𝑃𝑆𝐷2# +𝛾 𝐵𝑎𝑛𝑘 + 𝛾 𝑀𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 + 𝛾 𝑁𝐼𝑀 + 𝜀 (5) % !,# & # ' !,# !,# 𝑀𝑎𝑟𝑘𝑒𝑡 𝑃𝑜𝑤𝑒𝑟!,# = 𝛽$ + 𝛽%𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# + 𝛽&𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# × 𝑃𝑆𝐷2# + 𝛿%𝑃𝑆𝐷2# + 𝛾%𝐵𝑎𝑛𝑘!,# + 𝛾&𝑀𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 (6) # + 𝛾'𝑁𝐼𝑀!,# + 𝜀!,# For the financial performance analysis, the model specification follows the same structure as in the market power analysis and includes the same set of bank-specific and macroeconomic control variables. The primary differences lie in the choice of dependent variables, where the financial performance indicators return on assets or return on equity are used, and the inclusion of lagged values of these performance measures to capture the dynamic effects of prior financial outcomes. Equation 7 serves as the baseline model, while Equation 8 incorporates the effect of the PSD2 directive, and Equation 9 adds an interaction term between FinTech presence and the PSD2 directive. 15 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒!,# = 𝛽$ + 𝛽%𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# +𝛾%𝐵𝑎𝑛𝑘!,# + 𝛾&𝑀𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐# + 𝛾'𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 (7) !,#)% + 𝜀!,# 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒!,# = 𝛽$ + 𝛽%𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# + 𝛿%𝑃𝑆𝐷2# +𝛾%𝐵𝑎𝑛𝑘!,# + 𝛾&𝑀𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐# + 𝛾'𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 + 𝜀 (8) !,#)% !,# 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒!,# = 𝛽$ + 𝛽%𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# + 𝛽&𝐹𝑖𝑛𝑇𝑒𝑐ℎ 𝑃𝑟𝑒𝑠𝑒𝑛𝑐𝑒# × 𝑃𝑆𝐷2# + 𝛿%𝑃𝑆𝐷2# + 𝛾%𝐵𝑎𝑛𝑘!,# + 𝛾&𝑀𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐# + 𝛾'𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒!,#)% + 𝜀 (9) !,# 4.2.2 Variable Description This section provides a description of the variables utilized in the study. It begins by presenting the dependent variables, capturing the market power and financial performance of incumbent banks, followed by the explanatory variables measuring FinTech presence, and concludes with the control variables included to account for bank-specific and macroeconomic conditions. 4.2.2.1 Dependent variables To investigate the impact of FinTech firms on incumbent banks, this thesis utilizes three dependent variables: Lerner index, return on assets and return on equity. Lerner Index To assess market power, this study employs the Lerner index. Its application in the context of FinTech’s impact on the banking sector aligns with recent studies by Saklain (2024) and Cuadros-Solas et al. (2024). Furthermore, the Lerner index has been extensively used in earlier banking literature, including studies by Berger et al. (2008), Beck et al. (2013), and Weill (2013), confirming its relevance and robustness in measuring market power within the banking industry. Return on Assets (ROA) Return on assets is used as one of the financial performance measures in this study. Return on assets has been manually calculated by dividing net income by total assets. The metric reflects the profit generated per unit of assets and has been widely adopted in previous research 16 examining the effect of FinTech on bank performance, including studies by Nguyen et al. (2021) and Phan et al. (2020). Return on Equity (ROE) Return on equity serves as the second measure of bank performance. Return on equity has been manually calculated by dividing net income by total equity. The metric provides insights into how effectively a bank uses investors capital to drive earning growth and is frequently used in banking literature, including studies by Menicucci and Paolucci (2016), Athanasoglou et al. (2008) and Phan et al. (2020). 4.2.2.2 Explanatory Variable To evaluate the effect of FinTech presence on incumbents in the Swedish the banking sector, this study uses venture capital investments in the FinTech sector as the primary explanatory variable. To assess the robustness of the results, the number of FinTech launches per year is also considered. Venture Capital Investments in FinTech Venture capital investments are included in this study as the main explanatory variable. Venture capital investments play a crucial role in the development and expansion of FinTech firms. Including venture capital investments helps capture the financial mechanism that supports FinTech development and enables it to challenge incumbent banks. Zhao et al. (2021), emphasize the importance of financing metrics, such as registered capital, number of financing events and total financing amount in measuring FinTech maturity. Similarly, Cuadros-Solas et al. (2024) use the total annual investments in FinTech as a proxy for FinTech development. FinTech Launches The number of FinTech launches per year is included as a secondary explanatory variable, to assess the robustness of our findings. The variable reflects the expansion of the FinTech sector in Sweden and the increasing competition it brings to incumbent banks. An increase in FinTech startups signals growing competition for incumbent banks, particularly in areas such as digital payments, peer-to-peer lending, and alternative credit solutions. According to Phan et al. (2020) 17 and Zhao et al. (2021) the growth of FinTech increases market competition, influences banking efficiency and impacts profitability. 4.2.2.3 Control variables To ensure a more accurate estimation of FinTech’s impact on incumbent banks, this study includes control variables that account for bank-specific and macroeconomic variables potentially influencing market power and financial performance of incumbents in the Swedish banking sector. Revised Payment Services Directive (PSD2) The Revised Payment Services Directive (PSD2), introduced by the European Union and implemented in Sweden in 2018, represents a regulatory shift aimed at increasing competition, innovation, and security in the financial sector (ECB, 2018). This regulatory change has enabled the entry of new actors into the payment and digital banking space, weakening the dominance of incumbent banks in these areas. The PSD2 variable is constructed as a dummy variable, taking the value 0 for years prior to 2018 and 1 for 2018 and onwards, reflecting the implementation of the PSD2 directive. Bank Specific Variables Total Deposits Bank size is a standard control variable in the literature, as larger institutions typically benefit from economies of scale and greater customer reach. This study uses the natural logarithm of total deposits as a proxy for size. Larger banks often attract more deposits due to their branch network, brand recognition and established customer base. Deposits are also a core funding source, reflecting both financial stability and customer confidence. Loan Loss Provisions to Average Gross Loans The loan loss provision to average gross loans ratio is included to control for credit risk exposure. This variable captures how much a bank sets aside to cover potential loan defaults. Higher provisions suggest increased risk, which generally has a negative impact on profitability. Including this variable follows prior research by Phan et al. (2020) and Athanasoglou et al. (2008). 18 Liquidity Ratio The liquid assets to total assets ratio measure a bank’s liquidity position. Higher liquidity is typically associated with improved financial stability and performance. Including this variable follows prior research such as Nguyen et al. (2021), Cuadros-Solas et al. (2024) and Saklain (2024). Deposit Growth In addition to overall deposit levels, deposit growth is included to capture dynamic changes in market share. This variable reflects whether a bank is expanding or contracting in terms of its funding base, which may signal shifts in competitiveness and consumer trust. This variable follows prior research by Phan et al. (2020). Macroeconomic variables GDP Growth Gross Domestic Product (GDP) growth is used to account for general economic conditions and business cycle effects. Higher GDP growth is typically associated with better loan performance and increased banking activity, positively influencing bank profitability and market power. This approach follows Phan et al. (2020), Yoon et al. (2023) and Yao and Song (2021). GDP growth refers to the annual change in Gross Domestic Product at market prices compared to the previous year. Inflation Inflation, measured through Consumer Price Index (CPI), is included as a control variable to account for macroeconomic conditions that influence the banking environment, in accordance with Phan et al. (2020), Cuadros-Solas et al. (2024) and Saklain (2024). Changes in the general price level can impact both the cost structure and revenue generating capacity of banks. Higher inflation may allow banks to increase interest rates on loans faster than on deposits, potentially boosting profit margins. Therefore, inflation is expected to have a positive relationship with bank performance. Since profitability can enhance a bank’s ability to exert market power, through pricing strategies, expansion or product differentiation, it follows that inflation is also expected to be positively associated with market power. The Consumer Price Index refers to the change compared to the previous year, reported as annual averages. 19 Additional Control Variables Net Interest Margin (NIM) To account for changes in profitability, the net interest margin is included as a control variable only in the market power analysis. The net interest margin reflects the profitability of banks, and controlling for it helps to isolate the effect of FinTech on market power by accounting for underlying shifts in bank profit. The inclusion of a profit measure as a control variable follows prior research by Cuadros-Solas et al. (2024) and Saklain (2024). Lagged Financial Performance Variables Financial performance in banks often exhibits persistence over time, where past performance can influence current outcomes. To account for potential autocorrelation in bank performance, a one-year lag of the financial performance measures, return on assets and return on equity, is included in the performance analysis. This approach enhances the robustness of the estimation regarding the impact of FinTech presence and is consistent with previous research by Athanasoglou et al. (2008) and Phan et al. (2020). 4.3 Data Sample and Time Frame Our sample consists of six Swedish banks: Handelsbanken, Skandinaviska Enskilda Banken (SEB), Nordea, Swedbank, SBAB Bank and Länsförsäkringar Bank. The first four are the largest banks in the Swedish banking sector, collectively covering 64 percent of the Swedish banking market, based on deposits and loans to Swedish households and non-financial corporations (Finance Sweden, 2024). The latter two, SBAB Bank and Länsförsäkringar Bank, have emerged as increasingly influential actors within the Swedish banking sector with their increasing market shares in areas like mortgage lending, credit growth and digital banking services (Copenhagen Economics, 2023). Including SBAB Bank and Länsförsäkringar Bank in the sample offers a more comprehensive view of how both established and rising financial institutions respond to financial innovation. Furthermore, their inclusion enhances the policy relevance of the study, reflecting broader market dynamics and extending the coverage to 76 percent of the Swedish banking market. The period 2010 to 2023 is characterized by the rapid rise of FinTech and the increasing adoption of digital banking solutions, such as payments and lending. The period witnessed increasing digitalization and the transition towards a more cashless society. Additionally, key 20 regulatory changes, such as the PSD2 directive implemented in 2018, redefined data privacy standards affecting both banking and FinTech operations. The selected timeframe excludes the 2008-2009 financial crisis. This allows for a clearer analysis of the post-crisis transformation, where innovation and growth in sectors like FinTech could occur without the constraints and disruptions caused by the crisis. Finally, the period 2010 to 2023 captures the emergence, expansion and regulatory evolution of the Swedish FinTech sector, which makes an analysis of its impact on the Swedish banking sector suitable. 4.3.1 Data Collection Data on FinTech activity, the amount of venture capital investments and the number of FinTech launches in Sweden, were collected from the Sweden Tech Ecosystem (2025a; 2025b) for the period 2010 to 2023. The Sweden Tech Ecosystem is an open-access database maintained by Dealroom.co in collaboration with Business Sweden, the Swedish Institute, the Swedish Agency for Economic and Regional Growth and Vinnova (Vinnova, 2024). The Sweden Tech Ecosystem (2025c) defines FinTech as the intersection between finance and technology, further subdivided into eight sub-industries, as outlined in Table A1 in Appendix A. Reported venture capital investments figures were originally denominated in euros. To align with the monetary unit of this study (millions of Swedish kronor, SEK), all annual investment values were converted to SEK using the average annual exchange rate (EUR/SEK) for each year respectively. Exchange rate data was obtained from Riksbanken’s “Search interest rates and exchange rates” service (Riksbanken, n.d.). Financial and bank-specific data were retrieved from the S&P Capital IQ database. The database includes comprehensive financial information on Handelsbanken, Skandinaviska Enskilda Banken, Nordea, Swedbank, SBAB Bank and Länsförsäkringar Bank. For each bank, the four standardized datasets: Financial Highlights, Balance Sheet, Income Statement and Asset Quality Detail, were collected. The macroeconomic variables, consumer price index and gross domestic product were retrieved from Statistics Sweden (2025a; 2025b). Due to a single missing value in the Capital IQ dataset for Handelsbanken, additional disclosure was sourced from the bank’s annual report to ensure data completeness and consistency. The 21 missing value was not directly reported in Capital IQ but could be manually calculated using finical figures disclosed in the annual report.2 To validate the reliability of the Capital IQ data, values from non-missing years were cross checked against corresponding figures in the annual report, showing a high degree of consistency, and no discrepancies were found. 4.4 Robustness and Potential Endogeneity Concerns Due to the possibility that market power and financial performance might affect the amount of venture capital invested in the FinTech sector, the analyses may suffer from reversed causality. The endogeneity concerns are based on the reasoning that the development of the banking sector will affect the willingness to invest in the FinTech sector. To address the endogeneity concerns, two robustness analyses are conducted. The first robustness test will be carried out by replacing the dependent variable, venture capital investments in the FinTech sector with the number of FinTech launches per year. The number of FinTech launches indicates actual operational presence and the number of FinTech firms challenging incumbent banks. The robustness analysis is conducted in Appendix B. In the second robustness analysis we conduct an instrumental variable analysis, utilizing a Two- Stage Least Squares approach (TSLS). An instrument variable must affect the venture capital invested in the FinTech sector while not being directly tied to banks’ profits or market power, except through its influence on venture capital. Our instrument variable, the frequency of internet use in Sweden, collected from Eurostat (2024), needs to fulfill two assumptions in order to be a valid instrument. The frequency of internet use needs to be correlated with venture capital invested in the FinTech sector. Secondly, the instrument should not be correlated with unobserved factors that affect banks’ performance (Stock & Watson, 2020). Instrument variables measuring different kinds of internet coverage or access have been used in previous literature (e.g. Xu et al., 2023; Demir et al., 2020). The choice of frequency of internet use is further elaborated upon in Appendix B. 2 The missing value concerned liquid assets for Handelsbanken in 2021 and was due to an omission in the Capital IQ dataset. All necessary figures were available in Handelsbanken’s Annual and Sustainability Report 2022, which also included retrospective data for 2021. The value was manually calculated based on this information. 22 4.5 Limitations This study is based on group financial data of the selected banks. Not all banks provide geographically segmented financial information, such as annual reports specific to their Swedish operations, making it difficult to conduct a consistent and robust country-specific analysis. Using only Swedish-segmented data would have limited the analysis considerably, both in terms of sample size and explanatory power. While group-level data may not capture the full detail of FinTech presence to Swedish operations, it still provides valuable insights and an indication of broader trends. Länsförsäkringar Bank and SBAB Bank operate exclusively in Sweden, however, the four universal banks have a broader international presence. To estimate the Swedish share of their operations, we used the ratio of employees based in Sweden to the total number of employees in 2023, obtained from Finance Sweden (2024). This approach indicates that Swedbank, Handelsbanken and SEB conduct more than 50% of their operations in Sweden3, whereas Nordea operates primarily outside of Sweden, with only about 20% of its workforce based in Sweden. To account for this disparity, we re-ran the analysis excluding Nordea and found that the overall results remained unchanged. Previous banking literature assessing profitability and market power commonly utilize a general method of moments (GMM) approach in the econometric analysis, based on its ability to handle potential endogeneity concerns (see e.g. Athanasoglou et al., 2008; Yao and Song, 2021; Phan et al., 2020). However, the GMM approach performs better in large samples and datasets. The number of entities, in this case banks, is restricted to six, N = 6. The implications of the GMM estimator in small samples are relatively unexplored, and small samples may prevent the use of the full set of instrument variables available (Soto, 2009; Blundell & Bond, 1998). This implies that to utilize the GMM framework in small samples, the number of instruments must be reduced and could potentially introduce pitfalls. The concern of the number of instruments is also recognized by Hayakawa (2007). Blundell & Bond (1998) discuss finite sample bias in the GMM model, in a smaller time frame the GMM could be biased because it relies on instrumental variables that may be weak or imprecise in smaller samples. Therefore, a fixed effects approach will be utilized in accordance with Cuadros-Solas et al. (2024), Menicucci and 3 Share of employees in Sweden: Swedbank: 55%, Handelsbanken: 61%, SEB: 52%, Nordea: 22%. 23 Paolucci (2016) and Saklain (2024). Cuadros-Solas et al. (2024) state that their main findings do not change when applying a GMM framework, compared to the fixed effects approach. 5. Result The following section presents the result of the empirical analysis, beginning with descriptive statistics that summarize the key variables in this study. This is followed by the main regression results, where the impact of venture capital investments in FinTech on incumbent banks’ market power and financial performance is examined and interpreted. 5.1 Descriptive Statistics Table 2 presents descriptive statistics over the variables utilized in this study. The main explanatory variable, venture capital investments, shows a wide range of the investments made in the FinTech sector from 2010 to 2023. The venture capital investments increased significantly between the years 2018 to 2023 and reached their peak in 2021. The dependent variables, return on assets (ROA) and return on equity (ROE), provide insights into the profitability of the banks included in this study. Return on assets has remained relatively stable over the observed period, while return on equity has experienced a greater variation. The mean value of the Lerner index signals that the Swedish banking sector is characterized by moderate market power but not monopoly-level dominance. Figure 1 illustrates the average evolution of the Lerner index. From the figure, we can tell that the mean price in the observed period has been significantly higher than the average marginal cost, suggesting that incumbent banks maintained a stable level of market power over the observed period. Table 2. Descriptive statistics n Mean sd Min Max Range Lerner Index 84 0,43 0,10 0,14 0,66 0,53 ROA 84 0,53 0,22 0,10 1,20 1,09 ROE 84 10,51 3,01 4,07 17,19 13,12 Deposit Growth 84 11,80 25,59 -26,26 215,36 241,62 Loan Loss Ratio 84 0,07 0,08 -0,16 0,32 0,48 Liquidity Ratio 84 23,00 8,68 8,75 47,14 38,40 Log of Deposits 84 13,04 1,38 8,71 14,70 5,99 Net Interest Margin 84 1,01 0,25 0,50 1,94 1,44 Venture Capital Investments 14 5 009,82 6 477,61 101,14 20 289,72 20 188,58 24 Number of FinTech Launches 14 31,43 14,11 10,00 56,00 46,00 GDP Growth 14 4,56 2,46 -2,25 8,84 9,09 CPI 14 2,19 2,71 -0,18 8,55 8,73 PSD2 14 0,43 0,51 0,00 1,00 1,00 Note: Venture Capital Investments is presented in millions of SEK. ROA, ROE, Deposit Growth, Loan Loss Ratio, Liquidity Ratio, Net Interest Margin, GDP Growth and CPI are denoted in percentages. Figure 1. Overview of Mean Lerner Index, Price and Marginal Cost, 2010 - 2023 2 1,8 1,6 1,4 1,2 1 0,8 0,6 0,4 0,2 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Mean Lerner Index Mean Price Mean Marginal Cost Note: Mean price and mean marginal cost are presented in ten thousand SEK. For each variable, the mean is first calculated by computing annual values for each bank and then averaging these values across all banks and years in the data sample to provide an overall overview. The evolution of the Lerner index for each individual bank is illustrated in Figure 2. Across the observed period, SBAB Bank demonstrates the highest Lerner index values among the banks, signaling relatively strong market power. SBAB Bank’s high Lerner index, despite its smaller size, could be explained by its low marginal cost, potentially driven by digital operations and niche focus and not necessarily higher pricing or dominant market position. The Lerner index can overestimate market power in cases when marginal costs are low or price levels are stable or regulated, as is often the case in mortgages, a segment where SBAB Bank has primarily positioned itself. Nordea experienced a significant decrease between 2015 and 2019, before gradually converging to the levels of the majority of the other banks. Overall, the data indicates a general upward trend in market power for most banks between 2010 and 2016, followed by stabilization and a converging trend, apart from Länsförsäkringar 25 Bank. The variation in Lerner index levels across banks and over time suggests differing competitive positions and pricing behaviors in the Swedish banking sector. Figure 2. Lerner Index by Bank, 2010 – 2023 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Handelsbanken Nordea SEB Swedbank Länsförsäkringar SBAB 5.2 Regression Results This section first presents the findings from the market power analysis, followed by the results of the financial performance analysis. 5.2.1 Market Power Analysis In the market power analysis venture capital investments in the FinTech sector show no significant effect on incumbent banks’ market power in Model 1 and 2, presented in Table 3. This suggests that venture capital investments in the FinTech sector did not have a standalone effect on incumbent banks’ market power, even after accounting for the effects of PSD2. Model 3 indicates that the period prior to PSD2, venture capital investments had a positive significant effect on market power. However, the negative and significant interaction term indicates that this relationship changes after the implementation of PSD2, with venture capital investments instead being associated with a decrease in incumbent banks market power, holding everything else constant. These results are further supported by the robustness analysis, reported in Appendix B. 26 Lerner Index Table 3. Regression Result Lerner Index Dependent variable Lerner Index Model 1 Model 2 Model 3 (1) (2) (3) Venture Capital -0.00000 0.00000 0.00002** (0.00000) (0.00000) (0.00001) Venture Capital x PSD2 -0.00002** (0.00001) PSD2 ✓ ✓ Bank-Specific Controls ✓ ✓ ✓ Macroeconomic Controls ✓ ✓ ✓ Net Interest Margin ✓ ✓ ✓ Observations 84 84 84 R2 0.599 0.612 0.623 Adjusted R2 0.524 0.534 0.540 13.048*** 12.115*** 11.242*** F Statistic (df = 8; 70) (df = 9; 69) (df = 10; 68) Note: Venture Capital is denoted in millions of SEK. All estimates include bank fixed effects. Standard errors are clustered by bank. *p<0.1; **p<0.05; ***p<0.01. 27 5.2.2 Financial Performance Analysis The results of the analysis of how venture capital investments in the FinTech sector affect incumbent banks’ return on assets and return on equity are displayed in Table 4. Venture capital investments have a modest significant negative effect on the return on assets and return on equity in Models 1, 2, 4 and 5, holding all else constant. The inclusion of PSD2 did not significantly alter the effect of FinTech on incumbent banks’ return on assets nor on return on equity, as observed in Models 2 and 5. The interaction term between venture capital and PSD2 does not show a significant effect on return on assets nor on return on equity. This indicates that there is no significant difference in the effect of venture capital investments on the financial performance measures before and after the implementation of PSD2. 28 Table 4. Regression Results for Return on Assets and Return on Equity Dependent variable ROA ROE Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 (1) (2) (3) (4) (5) (6) Venture Capital -0.00001*** -0.00001*** -0.00001 -0.0001*** -0.0001*** -0.0003 (0.00000) (0.00000) (0.00001) (0.00004) (0.00004) (0.0003) Venture Capital x PSD2 0.00000 0.0002 (0.00001) (0.0003) PSD2 ✓ ✓ ✓ ✓ Bank-Specific ✓ ✓ ✓ ✓ ✓ ✓ Controls Macroeconomic ✓ ✓ ✓ ✓ ✓ ✓ Controls Lagged Financial ✓ ✓ ✓ ✓ ✓ ✓ Performance Controls Observations 84 84 84 84 84 84 R2 0.570 0.571 0.571 0.501 0.502 0.503 Adjusted R2 0.491 0.484 0.477 0.409 0.401 0.394 F Statistic 11.616*** 10.203*** 9.058*** 8.792*** 7.731*** 6.889*** (df = 8; 70) (df = 9; 69) (df = 10; 68) (df = 8; 70) (df = 9; 69) (df = 10; 68) Note: Venture Capital is denoted in millions of SEK. All estimates include bank fixed effects. Standard errors are clustered by bank. *p<0.1; **p<0.05; ***p<0.01. The results are to be interpreted carefully due to possible endogeneity concerns caused by simultaneity between incumbent banks’ market power and financial performance measures, and venture capital investments. Due to the potential endogeneity concerns, robustness tests have been conducted and are presented in Appendix B. 29 6. Analysis The findings suggest that venture capital investments in the FinTech sector have a modest negative significant effect on the financial performance of incumbent banks, consistent with our expectations and prior research. Regarding market power, the results indicate that venture capital investments reduce market power of incumbents when interacted with the PSD2 directive. 6.1 Market Power Swedish banks are characterized by their strong and stable market positions and are well equipped to adopt to the evolving banking landscape by incorporating or acquiring technologies and systems from FinTech competitors. The initial positive association between venture capital invested in the FinTech sector, prior to the PSD2 directive, and market power could possibly be explained by strategic responses by the incumbent banks, as discussed by Navaretti et al. (2018). The emergence of FinTech does not show a major effect in the observed period, which is in accordance with Stulz (2019) and the OECD (2020), who argue that banks are not passive actors and that FinTech firms have not yet established themselves as strong competitors to incumbent banks. However, FinTech may pose a larger competitive threat in the future depending on policies and regulations. In a market characterized by a few large actors, like the Swedish banking sector, regulations promoting competition and facilitating the entry of smaller firms are important. Our results suggest that the PSD2 directive has played a part in enhancing the competitive ability of FinTech, by enabling data-sharing and third-party access. The results further suggest that the directive enabled FinTech to assert competitive pressure on incumbents, decreasing their market power. Moreover, the results are line with theories proposed by Christensen (1997), and Aghion and Howitt (1992) which states that new technologies disrupt older ones by using simpler solutions for consumers and targeting underserved segments with new cost-effective solutions. Incumbents could be negatively affected by this due to their reliance on already in- place technologies and business models. The results from this study indicate that FinTech innovations alone might not be sufficient to impact the market power of incumbents. Furthermore, the findings suggests that that the emerging force of FinTech needs pairing with 30 suitable regulation and policies that can support their development within the Swedish financial sector. Our findings regarding the impact of FinTech on market power of incumbent banks are supported by the robustness analysis conducted in Appendix B, which includes an alternative measure of FinTech presence and an instrumental variable analysis. These additional analyses suggest that the effects of FinTech presence are supported over different specifications and proxies. 6.2 Financial Performance Measures Moving to the financial performance analysis, our results demonstrate that FinTech presence has a different effect on the financial performance measures return on assets and return on equity, compared to market power. The results indicate that venture capital investments have a standalone negative significant effect on return on assets and return on equity. This could be due to the rapid expansion by well-established new technologies that amplify the disruptive effect on incumbents. Additionally, venture capital investments in the FinTech sector also signal beliefs of potential future disruption by FinTech, hence impacting future market expectations, which potentially could explain the negative outcomes. The negative effect could also be explained by the intensified competition from FinTech activity that could imply pressure on incumbents through lost income, increased expenditures and tighter margins. This pressure might lead to reduced financial performance measures as consumers are more drawn to more agile and user-friendly FinTech solutions. Collectively, these dynamics suggest that, even in its early stages, FinTech can have a disruptive effect on the financial performance of incumbent banks. The decline in the financial performance measures, despite the positive effect on market power, could potentially be explained by the incumbent banks’ initially strong market position. With their established positions, incumbents have the ability to withstand short-term shocks while still retaining their pricing power. Moreover, the new technologies introduced by FinTech may force banks to accelerate their digital transformation and adapt to the technological shift (Navaretti et al., 2018). This strategic response may inflate short-term costs, potentially 31 reducing profitability, as measured by return on assets and return on equity, during periods of FinTech driven disruption. The implementation of PSD2 did not show any significant changes on FinTech’s effect on the financial performance measures nor did the interaction between venture capital investments and the implementation of PSD2. Our findings on the standalone effect of FinTech presence, measured through venture capital investments, are consistent with prior studies (see Phan et al., 2020; Zhao et al., 2021; Yao & Song, 2021; Nguyen et al., 2021). This suggests that the influence of FinTech presence in Sweden aligns with the patterns observed in the countries analyzed in the referenced studies. However, the results from the robustness analyses in Appendix B contradict the negative significant effect. The different proxies of FinTech presence show similar effects in the market power analysis, and opposite effects in the financial performance analysis. This underscores that different proxies capture different aspects of FinTech presence. Venture capital investments represent future market expectations and disruptive capabilities of FinTech, which could induce short-term costs and create uncertainty, which could have a negative effect on short-term financial performance. The number of FinTech launches on the other hand serves as a general indicator of market activity but may not reflect the same level of competitive intensity as venture capital investments. While a higher number of FinTech launches suggest growing interest in the sector, not all new launches possess the resources or fundings to challenge incumbents. In contrast, FinTech firms backed by venture capital often have a greater capacity to develop better technologies and grow faster and are therefore more capable of challenging incumbents. These proxies capture different dimensions of FinTech presence and could explain the variation in the results. As previously noted, the results should be interpreted with caution. Nonetheless, the results provide an indication of how FinTech presence impacts the market power and the financial performance measures for incumbents in the Swedish banking sector. 32 7. Conclusion The emergence of FinTech represents a transformative force within the Swedish banking sector, reshaping how financial services are delivered. Although the full extent of FinTech’s impact on incumbent banks remain difficult to quantify, this study contributes to the growing body of literature by examining how FinTech presence affects market power and financial performance in the Swedish banking industry. The findings suggest that FinTech has the potential to modestly erode the market power of incumbents, indicating a shift in the competitive dynamics of the sector. While the effects observed in this study are relatively limited, they may intensify over time as FinTech firms mature and regulatory frameworks continue to evolve. These results, supported by the robustness analyses, underscore the importance of fostering and enabling a regulatory environment, one that balances innovation and competition with the need to maintain financial stability and a level playing field. In contrast, the effect of FinTech on bank profitability, measured through return on assets and return on equity, appears less pronoun. Given that the robustness tests do not fully support the findings of the main analysis, the conclusions related to financial performance should be interpreted with caution. We presume that FinTech will continue to evolve and become more increasingly embedded in the Swedish society, partly driven by ongoing digitalization and demographic shifts toward a younger generation that is more inclined to adopt digital financial services. In response, incumbent banks may seek to protect their market position through strategies such as advocating for stricter regulatory measures or acquiring FinTech firms and their technologies. These developments highlight the need for a regulatory framework that not only facilitates FinTech presence and enhances competition but also safeguards against anticompetitive practices by established financial institutions. For future research we recommend narrowing the focus to specific FinTech segments, for example the sub-industries presented in Table A1, to better assess the differences in their competitive impact. Further, the analysis could be extended and include financial stability, to examine whether FinTech presence has any implications on the stability of the Swedish 33 financial system. Another interesting aspect could be to explore the implications of using different proxies for FinTech presence, as our findings suggest that these proxies capture distinct aspects of FinTech activity. An analysis of different measures in different contexts could enhance the understanding of various proxies of FinTech. References Aghion, P. and Howitt, P. (1992). A Model of Growth Through Creative Destruction. Econometrica, 60(2), pp.323–351. doi: https://doi.org/10.2307/2951599. Angelini, P. and Cetorelli, N. (2003). The Effects of Regulatory Reform on Competition in the Banking Industry. Journal of Money, Credit and Banking, 35(5), 663–684. Available at: http://www.jstor.org/stable/3649822 Athanasoglou, P.P., Brissimis, S.N. and Delis, M.D. (2008). Bank-specific, industry-specific and macroeconomic determinants of bank profitability. Journal of International Financial Markets, Institutions and Money, 18(2), pp.121–136. doi: https://doi.org/10.1016/j.intfin.2006.07.001. Beck, T., De Jonghe, O. and Schepens, G. (2013). Bank Competition and stability: Cross- country Heterogeneity. Journal of Financial Intermediation, 22(2), pp.218–244. doi: https://doi.org/10.1016/j.jfi.2012.07.001. Berger, A.N., Klapper, L.F. and Turk-Ariss, R. (2008). Bank Competition and Financial Stability. Journal of Financial Services Research, 35(2), pp.99–118. doi: https://doi.org/10.1007/s10693-008-0050-7. Blundell, R. and Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87(1), pp.115–143. doi: https://doi.org/10.1016/S0304- 4076(98)00009-8. Carletti, E., Leonello, A. and Marquez, R.S. (2024). Market Power in Banking. SSRN Electronic Journal. doi: https://doi.org/10.2139/ssrn.4691168. Carlton, D.W. and Perloff, J.M. (2005). Modern Industrial Organization. 4th ed. England Pearson Education. 34 Christensen, C.M. (1997). The innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. Boston, Mass.: Harvard Business School Press. Christensen, C.M., McDonald, R., Altman, E.J. and Palmer, J.E. (2018). Disruptive Innovation: An Intellectual History and Directions for Future Research. Journal of Management Studies, 55(7), pp.1043–1078. doi: https://doi.org/10.1111/joms.12349. Christensen, C.M., Raynor, M. and McDonald, R. (2015). What Is Distruptive Innovation. Harvard Business Review, 93(12), pp.44–53. Available at: https://hbr.org/2015/12/what-is- disruptive-innovation. Clark, E., Radić, N. and Sharipova, A. (2018). Bank competition and stability in the CIS markets. Journal of International Financial Markets, Institutions and Money, 54, pp.190–203. doi: https://doi.org/10.1016/j.intfin.2017.12.005. Clerides, S., Delis, M.D. and Kokas, S. (2015). A New Data Set on Competition in National Banking Markets. Financial Markets, Institutions & Instruments, 24(2-3), pp.267–311. doi: https://doi.org/10.1111/fmii.12030. Copenhagen Economics (2023). Konkurrensen i den svenska banksektorn. Available at: https://www.financesweden.se/fakta-och-rapporter/rapport-om- bankkonkurrensen/konkurrensen-i-den-svenska-banksektorn-2023/ [Accessed 2 April 2025]. Cuadros-Solas, P.J., Cubillas, E., Salvador, C. and Nuria Suárez (2024). Digital disruptors at the gate. Does FinTech lending affect bank market power and stability? Journal of International Financial Markets, Institutions and Money, 92, pp.101964–101964. doi: https://doi.org/10.1016/j.intfin.2024.101964. Demir, A., Pesqué-Cela, V., Altunbas, Y. and Murinde, V. (2020). Fintech, financial inclusion and income inequality: a quantile regression approach. The European Journal of Finance, 28(1), pp.1–22. doi: https://doi.org/10.1080/1351847x.2020.1772335. DeYoung, R. and Rice, T. (2004). Noninterest income and financial performance at U.S. commercial banks. Economic Perspectives, 28(4), pp.31–47. Federal Reserve Bank of Chicago. Available at: https://www.chicagofed.org/publications/economic- perspectives/2004/4qtr2004-part3-deyoung-rice [Accessed 15 May 2025]. 35 Elzinga, K.G. and Mills, D.E. (2011). The Lerner Index of Monopoly Power: Origins and Uses. American Economic Review, 101(3), pp.558–564. doi: https://doi.org/10.1257/aer.101.3.558. European Central Bank (2018). The revised Payment Services Directive (PSD2). European Central Bank. Available at: https://www.ecb.europa.eu/press/intro/mip- online/2018/html/1803_revisedpsd.en.html [Accessed 14 March 2025]. Eurostat (2024). Individuals - frequency of internet use [Dataset]. Available at: https://ec.europa.eu/eurostat/databrowser/view/isoc_ci_ifp_fu__custom_16781409/default/tab le?lang=en [Accessed 21 March 2025]. Fernández de Guevara, J., Maudos, J. and Pérez, F. (2005). Market Power in European Banking Sectors. Journal of Financial Services Research, 27(2), pp.109–137. doi: https://doi.org/10.1007/s10693-005-6665-z. Finance Sweden (2024). Bank- och finansstatistik 2023. Available at: https://www.financesweden.se/fakta-och-rapporter/statistik-och-rapporter/bank-och- finansstatistik/ [Accessed 19 May 2025]. Finansinspektionen, (2020). Erik Thedéen: FinTech and cyber threats – what impact do they have on financial stability? Finansinspektionen. Available at: https://www.fi.se/en/published/presentations/2020/erik-thedeen-fintech-and-cyber-threats-- what-impact-do-they-have-on-financial-stability/ [Accessed 16 May 2025]. Fosu, S., Ntim, C.G., Coffie, W. and Murinde, V. (2017). Bank opacity and risk-taking: Evidence from analysts’ forecasts. Journal of Financial Stability, 33, pp.81–95. doi: https://doi.org/10.1016/j.jfs.2017.10.009 Hainz, C., Weill, L. and Godlewski, C.J. (2012). Bank Competition and Collateral: Theory and Evidence. Journal of Financial Services Research, 44(2), pp.131–148. doi: https://doi.org/10.1007/s10693-012-0141-3. Hayakawa, K. (2007). Small sample bias properties of the system GMM estimator in dynamic panel data models. Economics Letters, 95(1), pp.32–38. doi: https://doi.org/10.1016/j.econlet.2006.09.011. 36 Huang, T.-H., Chiang, D.-L. and Chao, S.-W. (2017). A new approach to jointly estimating the Lerner index and cost efficiency for multi-output banks under a stochastic meta-frontier framework. The Quarterly Review of Economics and Finance, 65, pp.212–226. doi: https://doi.org/10.1016/j.qref.2016.09.004. Koetter, M., Kolari, J.W. and Spierdijk, L. (2012). Enjoying the Quiet Life under Deregulation? Evidence from Adjusted Lerner Indices for U.S. Banks. Review of Economics and Statistics, 94(2), pp.462–480. doi: https://doi.org/10.1162/rest_a_00155. Leroy, A. and Lucotte, Y. (2017). Is there a competition-stability trade-off in European banking? Journal of International Financial Markets, Institutions and Money, 46, pp.199– 215. doi: https://doi.org/10.1016/j.intfin.2016.08.009. Lindenberg, E.B. and Ross, S.A. (1981). Tobin’s q Ratio and Industrial Organization. The Journal of Business, 54(1), pp.1–32. doi: https://doi.org/10.2307/2352631. Md Sohel Saklain (2024). FinTech, systemic risk and bank market power – Australian perspective. International Review of Financial Analysis, 95 (Part A), pp.103351–103351. doi: https://doi.org/10.1016/j.irfa.2024.103351. Menicucci, E. and Paolucci, G. (2016). The determinants of bank profitability: empirical evidence from European banking sector. Journal of Financial Reporting and Accounting, 14(1), pp.86–115. doi: https://doi.org/10.1108/jfra-05-2015-0060. Navaretti, G.B., Calzolari, G., Mansilla-Fernandez, J.M. and Pozzolo, A.F. (2018). Fintech and Banking. Friends or Foes? SSRN Electronic Journal, 2 (2017). doi: https://doi.org/10.2139/ssrn.3099337. Nguyen, L., Tran, S. and Ho, T. (2021). Fintech credit, Bank Regulations and Bank performance: a cross-country Analysis. Asia-Pacific Journal of Business Administration, 14(4), pp.445–466. doi: https://doi.org/10.1108/apjba-05-2021-0196. OECD, (2020). Digital disruption in banking and its impact on competition. OECD Roundtables on Competition Policy Papers, No. 243. Paris: OECD Publishing. Available at: https://doi.org/10.1787/b8d8fcb1-en [Accessed 19 February 2025]. 37 Phan, D.H.B., Narayan, P.K., Rahman, R.E. and Hutabarat, A.R. (2020). Do financial technology firms influence bank performance? Pacific-Basin Finance Journal, 62, p.101210. doi: https://doi.org/10.1016/j.pacfin.2019.101210. Qi, H., Yang, K. and Wang, W. (2022). Does Fintech Change the Market Power of Traditional Banks in China? Journal of Business Economics and Management, 23(5), pp.1– 24. doi: https://doi.org/10.3846/jbem.2022.17184. Riksbanken (2022). What is fintech? Available at: https://www.riksbank.se/en-gb/press-and- published/publications/staff-memos/an-overview-of-fintech-and-cryptoassets/what-is-fintech/ [Accessed 10 February 2025]. Riksbanken (2024). Den svenska finansmarknaden. Available at: https://www.riksbank.se/sv/press-och-publicerat/publikationer/den-svenska- finansmarknaden/den-svenska-finansmarknaden-2024/fakta--regelverk-satter-ramar-och- spelregler-for-det-finansiella-systemet/ [Accessed 6 May 2025]. Riksbanken (n.d.) Search interest rates and exchange rates. Available at: https://www.riksbank.se/en-gb/statistics/interest-rates-and-exchange-rates/search-interest- rates-and-exchange-rates/ [Accessed 22 Apr. 2025]. Sealey, C.W. and Lindley, J.T. (1977). Inputs, outputs, and a Theory of Production and Cost at Depository Financial Institutions. The Journal of Finance, 32(4), pp.1251–1266. doi: https://doi.org/10.1111/j.1540-6261.1977.tb03324.x. Shaffer, S. and Spierdijk, L. (2017) Market power: competition among measures. Bikker, A.J. and Spierdijk, L. (eds). Handbook of Competition in Banking and Finance, p.11–26. Available at: https://www.elgaronline.com/edcollchap/edcoll/9781785363290/9781785363290.00007.xml Shaffer, S. and Spierdijk, L. (2020). Measuring multi-product Banks’ Market Power Using the Lerner Index. Journal of Banking & Finance, 117, p.105859. doi: https://doi.org/10.1016/j.jbankfin.2020.105859. Shamshur, A. and Weill, L. (2019). Does Bank Efficiency Influence the Cost of credit? Journal of Banking & Finance, 105, pp.62–73. doi: https://doi.org/10.1016/j.jbankfin.2019.05.002. 38 Soon Suk Yoon, Lee, H. and Oh, I. (2023). Differential Impact of Fintech and GDP on Bank Performance: Global Evidence. Journal of risk and financial management, 16(7), pp.304–304. doi: https://doi.org/10.3390/jrfm16070304. Soto, M (2009): System GMM Estimation with a Small Sample, UFAE and IAE Working Paper, no. 780.09, Unitat de Fonaments de l'Analisi Economica (UAB) and Institut d'Analisi Economica (CSIC), http://pareto.uab.es/wp/2009/78009.pdf. Spencer, A.S. and Kirchhoff, B.A. (2006). Schumpeter and new technology based firms: Towards a framework for how NTBFs cause creative destruction. International Entrepreneurship and Management Journal, 2(2), pp.145–156. doi: https://doi.org/10.1007/s11365-006-8681-3. S&P Global Market Intelligence (2025a). Länsförsäkringar Bank AB: Public company profile. Available through: S&P Capital IQ Pro [Accessed 17 Feb. 2025] S&P Global Market Intelligence (2025b). Nordea Bank Abp: Public company profile. Available through: S&P Capital IQ Pro [Accessed 17 Feb. 2025]. S&P Global Market Intelligence (2025c). SBAB Bank AB: Public company profile. Available through: S&P Capital IQ Pro [Accessed 17 Feb. 2025]. S&P Global Market Intelligence (2025d). Skandinaviska Enskilda Banken AB: Public company profile. Available through: S&P Capital IQ Pro [Accessed 17 Feb. 2025]. S&P Global Market Intelligence (2025e). Svenska Handelsbanken AB: Public company profile. Available through: S&P Capital IQ Pro [Accessed 17 Feb. 2025]. S&P Global Market Intelligence (2025f). Swedbank AB: Public company profile. Available through: S&P Capital IQ Pro [Accessed 17 Feb. 2025]. Statistics Sweden. (2025a). Consumer Price Index (CPI), Fixed Index numbers, total annual average, 1980=100. Year 1980 – 2024 [Dataset]. Available at: https://www.statistikdatabasen.scb.se/sq/162645 [Accessed 22 Apr. 2025]. Statistics Sweden. (2025b). Institutional non-financial sector accounts (ESA2010), current prices, SEK million. Year 1950 – 2024 [Dataset]. Available at: https://www.statistikdatabasen.scb.se/sq/162643 [Accessed 22 Apr. 2025]. 39 Stock, J.H. and Watson, M.W., 2020. Introduction to econometrics. 4th ed. Harlow: Pearson Education. Stulz, R.M. (2019). FinTech, BigTech, and the Future of Banks. Journal of Applied Corporate Finance, 31(4), pp.86–97. doi: https://doi.org/10.1111/jacf.12378. Svenska Handelsbanken AB. (2023). Annual and Sustainability Report 2022. Available at: https://www.handelsbanken.com/sv/investor-relations/rapporter-och-presentationer [Accessed 7 Mar. 2025]. SVT Nyheter, (2024). Finansministern om bankernas vinster: Vill se större konkurrens. Available at: https://www.svt.se/nyheter/inrikes/finansministern-om-bankernas-vinster-vill- se-storre-konkurrens [Accessed 16 May 2025]. Sweden Tech Ecosystem (2025a). Investment heatmap. Available at: https://sweden.dealroom.co/ [Accessed 16 Apr. 2025) Sweden Tech Ecosystem (2025b). Companies. Available at: https://sweden.dealroom.co/companies/f/founding_or_hq_slug_locations/anyof_sweden [Accessed 12 Feb. 2025]. Sweden Tech Ecosystem (2025c). Industries and sub-industries. Available at: https://sweden.dealroom.co/ecosystem/guides [Accessed 12 Feb. 2025]. Swedish Agency for Economic and Regional Growth (2017). Regulation and competition – A literature review. Available at: https://tillvaxtverket.se/tillvaxtverket/publikationer/arkiveradepublikationer/publikationer201 7/regulationandcompetition.1266.html [Accessed 6 May 2025]. Swedish FinTech Association (2024). Fintechrapporten 2024. Available at: https://www.swefintech.se/rapporter [Accessed 14 May 2025]. The Financial Stability Board (2019). FinTech and Market Structure in Financial services: Market Developments and Potential Financial Stability Implications. Available at: https://www.fsb.org/2019/02/fintech-and-market-structure-in-financial-services-market- developments-and-potential-financial-stability-implications/ [Accessed 26 Feb. 2025]. 40 The National Board of Trade (2024). The Swedish Market for Financial Technology. https://www.kommerskollegium.se/en/analyses-and-seminars/publications/market-studies/the- swedish-market-for-financial-technology/ [Accessed 14 May 2025]. Turk Ariss, R. (2010). On the Implications of Market Power in banking: Evidence from Developing Countries. Journal of Banking & Finance, 34(4), pp.765–775. doi: https://doi.org/10.1016/j.jbankfin.2009.09.004. Van Hoose, D. (2009). The Industrial Organization of Banking. 1st ed. Springer Science & Business Media. Vinnova (2024). Swedish startups attract investments. Available at: https://www.vinnova.se/en/news/2024/03/swedish-startups-attract-investments/ [Accessed 19 Feb. 2025]. Wang, X., Han, L. and Huang, X. (2020). Bank market power and SME finance: Firm-bank evidence from European countries. Journal of International Financial Markets, Institutions and Money, 64, p.101162. doi: https://doi.org/10.1016/j.intfin.2019.101162. Weill, L. (2011). Do Islamic Banks Have Greater Market Power? Comparative Economic Studies, 53(2), pp.291–306. doi: https://doi.org/10.1057/ces.2011.1 Weill, L. (2013). Bank Competition in the EU: How Has It evolved? Journal of International Financial Markets, Institutions and Money, 26, pp.100–112. doi: https://doi.org/10.1016/j.intfin.2013.05.005. White, L.J. (2015). Monopoly and Dominant Firms: Antitrust Economics and Policy Approaches. Blair, R.D. and Sokol, D.D. (eds). The Oxford Handbook of International Antitrust Economics, Volume 1. Oxford Academic. doi: https://doi.org/10.1093/oxfordhb/9780199859191.013.0014 Xu, Y., Liu, Y., Lee, H., Baire, S., Nakonieczny, J. and Zhao, X. (2023). Fintech Development and Firm Technological Innovation Efficiency: Empirical Findings in China. IEEE Transactions on Engineering Management, 71, pp.1–11. doi: https://doi.org/10.1109/tem.2023.3239499. 41 Yao, T. and Song, L. (2021). Examining the differences in the impact of Fintech on the economic capital of commercial banks’ market risk: evidence from a panel system GMM analysis. Applied Economics, 53(23), pp.2647–2660. doi: https://doi.org/10.1080/00036846.2020.1864275. Zhao, J., Li, X., Yu, C.-H., Chen, S. and Lee, C.-C. (2021). Riding the FinTech Innovation Wave: FinTech, Patents and Bank Performance. Journal of International Money and Finance, 122, p.102552. doi: https://doi.org/10.1016/j.jimonfin.2021.102552 42 Appendix A Table A1 shows the different sub-categories of FinTech, as well as a description of their operations. Table A1: Sub-categories of FinTech FinTech Sub-industry Description Payments Startups develop solutions to improve the way financial transactions are settled, and how money is transferred between two parties. Banking Startups develop solutions, and/or digitize the activities, services and products of traditional banks. Crypto and Defi Startups developing solutions for the use and exchange cryptocurrencies or financial startups using cryptocurrencies as a core feature in their business. Wealth Management Startups develop solutions assisting in investment decision-making or providing a way to invest in assets, stocks, securities and other assets. Mortgages & Lending Startups develop solutions enabling digital lending (loans, lending platforms), providing online mortgage brokerage services, providing finance for individuals and businesses. Insurance Insurtech is the intersection between insurance and technology. It included startups providing insurance services with digital-first and innovative models, or helping insurers, agents and brokers increase the efficiency of their processes Financial Management Solutions such as software and algorithms help companies and consumers Solutions better manage their financial operations and processes. E.g.: accounting software, billing software. RegTech Solutions to comply with regulatory requirements in financial services, from customer identification (KYC), anti-money laundering, fraud detection and compliance & reporting. Source: Sweden Tech Ecosystem (2025c) 43 Appendix B To assess the robustness of our findings, we conducted two additional analyses. The first analysis replaces the explanatory variable with the Number of FinTech launches per year. The second analysis employs an instrumental variable approach. B1. Robustness Analysis Using Number of FinTech Launches The results from the analysis, presented in Table B1, partly supports the main analysis and the impact of FinTech on incumbent banks market power in Sweden. Using the number of FinTech launches, instead of venture capital investments, we observe positive significant effects of the number of FinTech launches in Model 1 and Model 2, whereas the main analysis showed insignificant effects, both negative and positive. The inclusion of the interaction variable in Model 3 strengthens the results from the main analysis, showing both higher significance and larger effects. Overall, the number of FinTech launches is an indicator of sectoral change, accounting for new entrants, while venture capital show investments in already existing firms or new start-ups entering the sector. Both measures signal competitive pressure in different ways, the number of FinTech launches indicate new actors in the sector, while venture capital reflects investor confidence in the potential of FinTech. 44 Table B1. Robustness Analysis Result for Lerner Index Dependent variable Lerner Index Model 1 Model 2 Model 3 (1) (2) (3) FinTech Launches 0.003*** 0.003*** 0.003*** (0.001) (0.001) (0.001) FinTech Launches x PSD2 -0.002*** (0.001) PSD2 ✓ ✓ Bank-Specific Controls ✓ ✓ ✓ Macroeconomic Controls ✓ ✓ ✓ Net Interest Margin ✓ ✓ ✓ Observations 84 84 84 R2 0.678 0.678 0.693 Adjusted R2 0.618 0.612 0.625 18.385*** 16.120*** 15.321*** F Statistic (df = 8; 70) (df = 9; 69) (df = 10; 68) Note: All estimates include bank fixed effects. Standard errors are clustered by bank. *p<0.1; **p<0.05; ***p<0.01. The robustness analysis based on the financial performance measures, return on assets and return on equity, yields result that differ from that of the main analysis. For return on assets, using the number of FinTech launches indicates a reversed effect compared to the main analysis, which used venture capital investments. Additionally, the effect of FinTech presence on return on equity becomes statistically insignificant, suggesting weaker support for the return on equity results under the alternative model specification. 45 Table B2. Robustness Analysis Results for Return on Assets and Return on Equity Dependent variable ROA ROE Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 (1) (2) (3) (4) (5) (6) FinTech Launches 0.002* 0.001 0.002* 0.027 0.006 -0.010 (0.001) (0.001) (0.001) (0.023) (0.022) (0.024) FinTech Launches x PSD2 -0.001 0.048 (0.001) (0.037) PSD2 ✓ ✓ ✓ ✓ Bank-Specific Controls ✓ ✓ ✓ ✓ ✓ ✓ Macroeconomic Controls ✓ ✓ ✓ ✓ ✓ ✓ Lagged Profitability Controls ✓ ✓ ✓ ✓ ✓ ✓ Observations 84 84 84 84 84 84 R2 0.517 0.526 0.527 0.413 0.447 0.456 Adjusted R2 0.428 0.430 0.422 0.304 0.335 0.336 9.380*** 8.500*** 7.563*** 6.164*** 6.192*** 5.704*** F Statistic (df = 8; 70) (df = 9; 69) (df = 10; 68) (df = 8; 70) (df = 9; 69) (df = 10; 68) Note: All estimates include bank fixed effects. Standard errors are clustered by bank. *p<0.1; **p<0.05; ***p<0.01. B2. Instrumental Variable Analysis To mitigate the issue of endogeneity caused by simultaneity between venture capital investments in the FinTech sector and market power and the financial performance measurements, an instrumental variable analysis is conducted. In the first stage we model the venture capital investments in the FinTech sector using the instrument variable frequency of internet use in Sweden, which provides predicted values of the venture capital invested in the FinTech sector. These predicted values are then gathered and replace the original venture capital variable from the original model. This procedure results in the TSLS estimator. 46 Using the frequency of internet use in Sweden as an instrument for venture capital investments captures the technological infrastructure necessary for FinTech firms, and the consumer readiness for FinTech solutions, since the solutions rely on digital financial services which are app- or platform-based. The frequency of internet use in Sweden is consistently high during the period of interest, ranging from 76 to 95 percent, indicating that digital connectivity is deeply integrated into everyday life. As a result, variations within this already high range are unlikely to exert a direct or substantial influence on incumbent bank’s market power or financial performance. This supports the validity of internet frequency as an instrument variable, as it is more plausibly related to FinTech development than to the outcomes of incumbent banking institutions. The results from the Two-Stage Least Squares analysis reinforce the results from the main analyses. Table B3 presents the outcomes using the frequency of internet use in Sweden as an instrument for venture capital investments in the FinTech sector, with the Lerner index as the dependent variable. The findings in Model 3 are consistent with those of the main analysis. Moreover, the instrument added significance to Model 2. 47 Table B3. Instrument Regression Results for Lerner Index Dependent Variable Lerner Index Model 1 Model 2 Model 3 (1) (2) (3) Predicted Venture Capital 0.000004 0.00001*** 0.00005* (0.000002) (0.000002) (0.000026) Predicted Venture Capital x PSD2 -0.00006* (0.000027) PSD2 ✓ ✓ Bank Specific Controls ✓ ✓ ✓ Macroeconomic Controls ✓ ✓ ✓ Net Interest Margin ✓ ✓ ✓ Observations 84 84 84 R2 0.66661 0.60683 0.69196 Within R2 0.54116 0.45890 0.57605 Note: Venture Capital is denoted in millions of SEK. All estimates include bank fixed effects. Standard errors are clustered by bank. *p<0.1; **p<0.05; ***p<0.01. For the financial performance measures, presented in Table B4, the instrument variable reduces the significance for all models. We observe that the signs and sizes of the coefficients are consistent with those of the main analyses, indicating a broader, but not statistically significant, negative trend. 48 Table B4. Instrument Regression Results for Financial Performance Measures Dependent Variable ROA ROE Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 (1) (2) (3) (4) (5) (6) Predicted Venture Capital -0.00000 -0.00000 -0.00001 -0.000187 -0.000205 -0.00057 (0.00000) (0.00000) (0.000017) (0.000099) (0.000143) (0.000613) Predicted Venture Capital 0.00001 0.00045 x PSD2 (0.000016) (0.000614) PSD2 ✓ ✓ ✓ ✓ Bank Specific Controls ✓ ✓ ✓ ✓ ✓ ✓ Macroeconomic Controls ✓ ✓ ✓ ✓ ✓ ✓ Lagged Financial ✓ ✓ ✓ ✓ ✓ ✓ Performance Controls Observations 84 84 84 84 84 84 R2 0.82457 0.82148 0.81863 0.73545 0.73246 0.73803 Within R2 0.56195 0.55424 0.54711 0.49397 0.48825 0.49891 Note: Venture Capital is denoted in millions of SEK. All estimates include bank fixed effects. Standard errors are clustered by bank. *p<0.1; **p<0.05; ***p<0.01. 49