FinTech Monopoly and Systemic Risk: Evidence From China

Abstract

With the rapid development of FinTech, it is of great significance to gain comprehensive insights into its potential risks. This paper focuses on the financial risks brought by the FinTech monopoly. We take the listed FinTech companies in China as samples to build the FinTech monopoly index and the systematic risk indicator, and the effect of the monopoly of Fintech companies on systematic risk is next analyzed using the fixed effect model. The results indicate that the FinTech monopoly will aggravate the systemic risk. Subsequently, the heterogeneous factors for the above nexus are investigated based on the micro characteristics of FinTech companies and the macro environment. On this basis, the mechanism of the above effect is also analyzed, and it is shown that the FinTech monopoly impacts systemic risk mainly via inducing excessive consumption, increasing banks’ credit risk, and inhibiting the development of FinTech.

Introduction

FinTech has been leaping forward with the boom of artificial intelligence, big data, cloud computing, blockchain, data mining, and other technologies over the years. Numerous studies have suggested that FinTech innovation penetrates a wide variety of financial fields (e.g., retail finance, wholesale payment, investment management, insurance, credit granting, and equity financing; Feyen et al., 2021; Financial Stability Board, 2017). Among a considerable number of promoters of FinTech innovation, technology-based companies represented by Internet enterprises also significantly boost the innovation and transformation of traditional financial services, besides the active layout of conventional financial institutions (Tsai & Peng, 2017).

Nevertheless, while promoting the growth of the global financial industry and facilitating investment and consumption, the FinTech led by technology companies also poses threats to the country’s systemic financial risk management (Wang et al., 2021). On the one hand, the business innovation and development of FinTech companies (e.g., securitization derivatives) result in more diversified financial risk sources, more covert risk transmission conditions, as well as more complex risk contagion paths (Murinde et al., 2021). On the other hand, since technologies mature progressively, numerous FinTech companies tend to further enter other businesses with platform advantages to develop more diverse and comprehensive business varieties and gradually dominate the market for new monopolies (Edelman, 2015). However, the financial business of the above monopolistic technology companies usually takes up a large market share on a huge transaction scale with extensive market coverage and cross-industry operation. Under ineffective operation, it is vulnerable to risk exposure and even severe risk contagion (Lu et al., 2023). Moreover, the risk prevention and control system of some monopoly platform institutions is incomplete in the development of varying qualifications, thus the potential danger of systemic financial risks is further aggravated (Wen et al., 2023).

Based on the above analysis, a research question is raised as to whether the monopoly of FinTech corporations will induce systemic risk. If so, what are the potential channels? Whether the above correlation is heterogeneous in different companies and macro environments? Studies have analyzed the systemic risks caused by banking monopoly, as well as the financial risks caused by the development of FinTech. However, there has been no research yet investigating the relation between FinTech monopoly and systemic risk. This paper aims to fill this gap in the literature.

We take China’s listed FinTech companies as samples to examine the above questions through empirical analysis. China is selected because of its world-leading FinTech development in Internet giants and FinTech startups (EY Global Ltd., 2022). China’s FinTech showed its momentum in 2013, when third-party payment companies typified by Alibaba extended their businesses into Internet finance, including online lending, online payment, and online equity-based crowdfunding that is dependent on e-commerce platforms that is Taobao (R. Li, 2018). By externality of platform economy, some technology companies gradually take the lead in the financial field with diversified business models and convenient business processes, covering almost all scenarios in retail finance (e.g., payment, banking, securities, insurance, and funds). Through years of development, China’s FinTech enterprises have shown a significant monopoly power in the financial market. Taking the payment business as an example, iResearch shows that Alipay and Wechatpay which are the financial business of Alibaba and Tencent respectively have a combined market share of 94% in 2022, with the former accounting for 55.1% and the latter accounting for 38.9%. Accordingly, the development of China’s FinTech monopoly is representative, and the research conclusions in China may provide a reference for other nations.

Under these circumstances, in this study, we first build the FinTech monopoly power indicator based on the annual financial data of Chinese listed FinTech companies and establish the synthesized systemic risk index by considering China’s macro-economic conditions from the perspectives of conventional financial institutions, emerging FinTech companies, capital markets, money markets, and real estate markets. Subsequently, empirical analysis is conducted from Q1 2014 to Q4 2020 based on the measures of FinTech monopoly, systemic risk, and other control variables to examine the impact of FinTech monopoly on systemic risk. Finally, we conduct the mechanism and heterogeneous analysis on the above relation.

The contributions of this study are presented as follows. First, it extends the strand of research examining the correlation between financial monopoly and financial risks (Beck et al., 2013; Goetz, 2018; Phan et al., 2019). The existing research on this issue has primarily focused on the risks triggered by the monopoly of conventional financial institutions (e.g., banks), whereas no empirical research has paid attention to FinTech companies, which as a group is a new and emerging financial business form. Numerous technological corporations in China presently have shifted from their main business to financial activities, thus progressively eroding the market share of conventional financial institutions and even taking the lead. Thus, it is imperative to assess the economic consequences of monopoly behaviors of these FinTech companies. Moreover, there has been very limited research addressing the correlation between FinTech monopoly and financial risks through both theoretical and empirical analysis. On that basis, this study identifies the significant effect of FinTech monopoly on financial risks and its working mechanism with empirical analysis, thus supplementing the research on the correlation between financial monopoly and financial risks.

Second, we enrich and extend the research on the connotation and factors of systemic risk. The existing research related to the coming source of systemic risk is mainly focusing on conventional financial institutions (Abduraimova, 2022; Aleksiejuk et al., 2002; Allen & Gale, 2000). However, the globally booming FinTech development in recent years makes financial business not only restricted to conventional sectors but involves technology-driven companies that could easily dominate the digital market. Thus, the connotation and source of a country’s systemic risk from the perspective of FinTech will be changed accordingly. In this study, synthesized systemic risk indicators are constructed by considering emerging FinTech enterprises and the effect of dominant FinTech on systemic risk is analyzed, thus extending and enriching the research on the determinants of systemic risk. Third, besides quantifying the effect of FinTech monopoly on systemic risk, this study further conducts heterogeneity and mechanism analysis on the above nexus, yielding some new results and findings, extending previous studies relating to financial development and financial risk.

The research conclusions take on a critical policy significance. On the one hand, the business monopoly of technology companies has become a phenomenal problem worldwide, such that the accurate evaluation of potential monopoly risks attributed to FinTechs may help regulators formulate more targeted anti-monopoly policies. On the other hand, it is imperative to boost the sustainable development of FinTech and inclusive finance in different nations by grasping the current situation of FinTech platform business and evaluating the effects and mechanism of platform monopoly on systemic risks. Only by in-depth analysis of the monopoly risks can the development of FinTech sector and the overall national strategy mutually reinforce each other, such that sustained and steady progress can be achieved through industry self-discipline and government regulation.

The rest of this study is organized as follows. Section “Literature Review” presents related literature. Section “Theoretical Analysis and Hypothesis Development” conducts the theoretical analysis. Section “Research Design” introduces the empirical design. In Section “Empirical Analysis,” the empirical results are analyzed, including benchmark regression and robustness tests. Further discussions are involved in Section “Further Discussions,” including heterogeneity analysis and mechanism tests. In Section “Conclusions,” the conclusions, policy recommendations, and research limitations are drawn.

Literature Review

The literature related to this paper mainly includes three aspects, the risk effect of financial monopoly, the risk effect of FinTech development, and the measurement of systemic risk.

Studies Focusing on the Nexus Between Financial Monopoly and Financial Risks

The existing studies on the relationship between financial monopoly and its financial risks mainly focus on the banking industry and there are two opposed views, one is the “concentration-stability” theory, and the other is the “concentration-fragile” theory (Beck et al., 2013; Boyd & De Nicolò, 2005; Goetz, 2018; Phan et al., 2019).

Studies have shown that the mechanism of improving bank monopoly power is conducive to risk prevention mainly in the following aspects. First, increasing bank concentration is likely to improve the “franchise value” of banks, and thus banks may actively reduce the risk of bankruptcy (Keeley, 1990). Second, the enhancement of bank monopoly position is conducive to reducing the risks caused by information asymmetry. As a special department, banks’ information production and analysis capabilities determine their operational efficiency and stability. As banking concentration deepens, more information can be collected, helping banks improve their internal credit analysis models through big data and thus improve loan quality (Petersen & Rajan, 1994). Third, bank concentration is advantageous to promote bank prudential behavior. Representative studies such as Allen and Gale (2000) find that when the number of banks increases to a certain extent, the optimal default risk value reaches the maximum, and the risk level of banks is the highest.

The view that the increase in bank concentration is likely to cause financial risks is mainly from the perspective that monopoly banks may require higher interest rates for loans and thus higher default rates (Borauzima & Muller, 2023). Based on Allen and Gale (2000) model, Boyd et al. (2006) pioneered the theory of negative correlation between bank concentration and bank stability. This theory entails that the traditional franchise value theory only considers the relationship between the competition in the deposit market and the stability of banks. If the loan market is added to the model, the impact mechanism of competition on stability will change. Specifically, if the bank’s investment in enterprises cannot be monitored at a low cost, the bank has greater loan pricing power in the uncompetitive loan market, and the enterprise bearing the interest burden itself has changed its bankruptcy risk due to the increase in financial costs. Later, scholars called the above mechanism risk transfer effect (Beck et al., 2006, 2013; Schaeck & Cihák, 2007).

Some scholars adopt market networking theory to investigate the impact of bank structure on financial risk contagion (Abduraimova, 2022). Allen and Gale (2000) propose the analysis of systemic risk contagion through the bank network structure, providing a micro theoretical basis for the study of this problem. Research has shown that bank monopoly mainly affects system risk through two aspects. The first channel is via the ability to carry the risks of financial institutions. When the bank is larger in scale, the capital is more sufficient, and the liquid assets are more abundant, the ability to withstand risks is stronger (Aleksiejuk et al., 2002). The second channel is through affecting the scale of interbank business. When banks become more monopolistic, the scale of interbank business may be smaller, and the intensity of financial risk transmission between departments may be reduced (Iori et al., 2006). Therefore, the more monopolized the banking system, the larger the scale of a single bank, the more difficult it is to infect the system risk.

Studies Focusing on the Nexus Between FinTech Development and Financial Risks

The effects of FinTech on banks’ operations are well-studied while research concentrating on banks’ risks is still in-developed (Lee & Shin, 2018). Some scholars claim that the traditional financial institutions face new risks connected with the emergence of innovative FinTech startups in qualitative analysis (Anagnostopoulos, 2018; Drasch et al., 2018). Based on theoretical analysis, some studies have explored the effect of FinTech on banks’ risks using empirical method from the perspective of macro FinTech innovation and development (Ng & kwok, 2017; Wang et al., 2021; Zhao et al., 2022). A few articles construct banks’ FinTech innovation indicator by using web crawler technology and word frequency analysis and find that bank FinTech significantly reduces credit risk in Chinese commercial banks (Cheng & Qu, 2020; C. Li et al., 2022).

Meanwhile, a growing number of literature examining the impact of FinTech on macro risks as the Financial Stability Board (2017) claims that FinTech activities could intensify risk contagion and asset volatility in the financial system, thereby undermining financial stability. Hasan et al. (2023) point out that because of the multiple interconnections, the risks inherent to the FinTech institutions could spill over to traditional financial institutions, possibly causing systemic risk. J. Li et al. (2020) find that FinTech institutions’ risk spillover to financial institutions positively correlates with financial institutions’ increase in systemic risk. However, some studies have shown that FinTech companies may not cause financial risks, such as Chaudhry et al. (2022) find that the tail risk of technology firms is higher than that of financial firms, whereas they are less likely to be in distress conditional upon a shock from the system.

Studies on the Measurement of Systemic Risk and Financial Monopoly

According to the International Monetary Fund, systemic risk refers to the failure of some financial institutions, which impedes the related function of the overall financial market and causes severe damage to the real economy and potential risks in the financial sector (IMF, 2016). Systemic risk can originate from risk events (e.g., credit risk, sharp fluctuations of asset prices, or insufficient market liquidity) in the dimensions of space and time (Cincinelli et al., 2022). From the spatial dimension, systemic risk is contagious within the financial system, as well as between the financial system and the real economy; from the dimension of time, it is a continuous variable accumulating, spreading, bursting, or resolving over time (Garcia & Rambaud, 2023).

After the global financial crisis in 2008, some scholars extracted information from microdata to measure systemic risk. The first type of model is based on CCA (Contingent Claims Analysis), while the second type of model is derived from the VaR (Value at Risk) theory, including the CoVaR model, ΔCoVaR model, etc. (Adrian & Bmnnermeier, 2009; Karimalis & Nokimos, 2018). However, the conclusions drawn from the above models are highly dependent on the sensitivity of model assumptions and parameter selection, which may underestimate the real risks when applied to the real economy (Baumöhl et al., 2022).

The comprehensive index method is also capable of measuring systemic risk (X. Zhang et al., 2023). This method is applied to build a comprehensive index by selecting risk indicators by the degree of correlation between a wide variety of economic indicators and systematic risk. The International Monetary Fund and the World Bank have launched the “Financial Sector Assessment Plan” since 1999 and both of them measure the soundness of a country’s financial system using the above comprehensive index method based on a series of macro and micro indicators. The comprehensive index method outperforms CCA or CoVaR models in measuring systemic risk for several reasons (X. Zhang et al., 2023). First, it is relatively simple and does not require complex theoretical models and various assumptions, which is more applicable to developing countries with underdeveloped financial markets and limited financial data. Second, it is capable of flexibly selecting appropriate financial and economic indicators according to the financial development of different periods and countries, without considering the specific causes of the financial crisis.

Presently, indicators reflecting institutional market power are mainly divided into structural indicators and non-structural measures (Leon, 2015). Specifically, the structure index reflects the concentration of the industry structure, referring to the total market share of the top N enterprises in the relevant market of an industry, including the concentration ration (CR) and Herfindahl-Hirschman index (HHI). However, some scholars claim that the concentration ratio cannot effectively measure the degree of competition in the financial industry, because it requires defining members in the industry, which makes it difficult to measure competition from the perspective of potential entrants and non-bank financial institutions, and also it relies on the restrictive assumption that all industry members face the same degree of competition (Bushman et al., 2016).

Non-structural indicators are used to measure the market power of an enterprise by using price-to-cost responsiveness indexes. For example, Panzar and Rosse (1987) propose H-statistics using the sum of the elasticity of a bank’s total income to the price of each factor with the assumption that banks reach long-term equilibrium revenue. Another commonly used non-structural indicator is the Lerner Index, which measures the market power of a specific firm by relating price to marginal cost (Lerner, 1934). Based on it, Angelini and Cetorelli (2003) propose the efficiency-adjusted Lerner index, which constructs a more efficient index by predicting cost and efficiency.

In sum, existing studies on financial monopoly and its impacts are mainly focused on traditional financial institutions, for example, banks. This is because, for many countries, especially emerging economies, banks occupy a dominant position in the financial system, so their monopoly or competitive behavior is crucial for a country’s financial stability. However, the recent rapid development of financial technologies has changed the financial structure greatly, making FinTech corporations an indispensable part of the financial system. Therefore, the behavior of these firms will also have a significant impact on the financial system. Unfortunately, however, as far as we know, no empirical studies have considered the risk effect of financial monopoly from the perspective of emerging FinTech companies till now. Although some scholars have noticed the prosperous development of financial technologies and conducted research regarding the effect of FinTech development on financial risks, however, few are from the perspective of FinTech monopoly. In fact, many initial financial technologies originated from internet companies, and these internet giants can acquire customers quickly through their monopoly advantages in other fields and conduct financial business for them, further exacerbating their monopolistic behavior. Therefore, the risk issues caused by the monopoly of these FinTech companies are worth paying attention to.

This paper extends the existing literature in the following aspects. First, we aim to construct a FinTech monopoly index based on micro-FinTech companies, to quantify the FinTechs’ monopoly power, which previous literature ignores. Meanwhile, we apply a comprehensive index method to measure systemic risk as previous studies do, but different from them, we also consider emerging FinTech companies, in line with the current situation and tendency in the financial system. Moreover, we propose to empirically quantify the effect of FinTech monopoly on systemic risk, so as to extend the literature on the economic consequences of monopoly behavior derived from traditional financial institutions. Finally, the heterogeneity factors and influencing mechanisms of the above relation are also analyzed, further expanding the literature on financial monopoly and financial risk.

Theoretical Analysis and Hypothesis Development

We propose there are some channels through which FinTech monopoly induce systemic risk. To begin with, the monopoly of FinTech companies may increase systematic risk through financial contagion channels. Financial institutions are different from the real industry, and once problems arise, their risks are likely to quickly spread to the entire financial system. With the entry of large technology companies into the financial field, they have formed a certain monopoly position and to some extent have become systemically important financial institutions. Some major technology companies in China have a loan scale of over one trillion yuan and have a large number of individual users and SME users. Their business scope involves multiple financial fields, and monopolistic platform institutions occupy a dominant market position and gradually become highly influential financial institutions. In addition, individual users and SMEs usually have weak professional knowledge in wealth management, which makes them prone to blindly following the crowd and following the trend. In the case of information asymmetry, there is a high possibility of a “bank run” of financial technology companies, which may spread the risk to the entire financial system and trigger systemic financial risks.

Next, the monopoly of FinTech companies may lead to high default financial risk by inducing excessive debt consumption. The customers of FinTech platform companies are mainly the long tail group that banks and other financial institutions scorn to reach out. Although the development of FinTech corporations has promoted financial inclusion to a certain extent, it should also be noted that many platforms, especially large FinTech ones will guide customers to excessive consumption. These firms acquire customers through other non-financial businesses such as e-commerce, and with the support of financial technologies, they combine their traditional businesses with financial ones to further expand their operation scope. They take more risks to obtain profits and extend loans to individuals who prefer advanced consumption and have weaker repayment abilities, a group that conventional banks deny lending. This disguised encouragement of advanced consumption in business will accumulate huge risks.

In addition, FinTech monopoly may cause systemic risk by weakening the stability of traditional financial institutions. The participation of large technology companies in the financial business such as deposit, loan, and wealth management, will compete with traditional financial institutions, thus increasing the operating costs of these conventional firms, eroding their profits, and weakening their stability. Meanwhile, monopolistic FinTech companies have stronger customer stickiness due to their diversified business and customized services, which will reduce the loyalty of bank customers, increasing their costs of maintaining customer relationships, further posing potential risks to traditional financial institutions.

Last but not least, the monopoly of fintech platforms may exacerbate systemic risk by curbing financial innovation. The monopoly of FinTech platforms may lead to difficulties in innovation among peers and curb market competition. On the one hand, monopolistic FinTech firms, to maintain their advantageous position, use their first mover advantage to suppress competitors and maliciously acquire innovative enterprises, ultimately leading to the stagnation of the entire industry’s development. On the other hand, FinTech enterprises leverage their advantageous position in data and customer resources to gain a huge competitive advantage. Then, through the vicious competition of price subsidies, they quickly occupy the market, defeat competitors within the industry, and create a “winner take all” situation. This new form of monopoly makes the traditional system of maintaining market fairness no longer effective, further curbing the healthy development of the entire industry.

Based on the above analysis, we propose the following research hypotheses that need to be empirically tested.

• H: FinTech monopoly is significantly positive with national systemic risk.

Research Design

In this section, the data, variable definitions, and econometric specifications are illustrated.

Samples and Data Sources

We apply quarterly data from Q1 2014 to Q4 2020 to conduct the empirical analysis. Year 2014 is selected as the starting year because Alipay, a third-party payment platform of China’s leading Internet company Alibaba, launched its online personal finance product “Yu Ebao” in June 2013, marking the beginning of Internet finance in China.

All listed A-share FinTech companies surviving from Q1 of 2014 to Q4 of 2020 in China are initially selected from the FinTech stock plate of Wind Database, with 45 companies in total. It should be noted that our sample does not include Alibaba, Tencent, JD, Meituan, and other famous FinTech companies listed abroad or overseas. Although the monopoly behavior of these companies is more prominent, due to the differences in accounting standards, we cannot include both domestically listed companies and overseas listed companies. Therefore, this will underestimate the monopoly power of FinTech companies in China to some extent.

We first remove ST (special treatment) and ST* listed companies, with the former indicating firms that have suffered losses for two consecutive years and have been subject to special treatment, and the latter referring to firms that have suffered losses for three consecutive years and being subject to delisting risk warnings. Then we remove firms with missing values for relevant information (e.g., total assets, ROA, leverage ratio), and firms that are listed for less than 3 years in the stock market. Therefore, 32 individual FinTech companies are considered, and we have a maximum of 896 firm-quarter observations. To avoid the effect of extreme values on the regression results, 1% outliers are deleted on both ends and there remain about 700 observations. Finally, we exclude observations that are missing for 1 or 2 years, making it a balanced panel data with 613 observations in the end. Moreover, the share of total assets of our sample on average accounts for approximately 80.17% of all FinTech companies in China, ensuring the representative nature of the sample.

The financial data of FinTech enterprises involved in this study are selected from the Wind Database and firms’ annual reports. Furthermore, the macroeconomic data primarily originate from the website of the National Bureau of Statistics of China.

Definition of Variables

Explained Variable: Systemic Risk

We adopt a comprehensive index method to measure systemic risk for its simplicity, flexibility, and independence of parameter estimation. Amid the prosperity of FinTech, the national monitoring index of systemic risk should be able to reflect its development. It is because the boom of FinTech in China increases the complexity, relevance, and infectivity of the original financial system, thus directly or indirectly causing higher vulnerability of China’s financial system.

To this end, in addition to the present systemic risk sources resulting from conventional financial institutions (Rahman et al., 2021), capital market (Feng et al., 2023), monetary market (A. Zhang et al., 2020), and real estate market (Chiang & Chen, 2022), we also take risks coming from FinTech companies into consideration, making the comprehensive index of systemic financial risk measure (CISFR) constructed from the above five dimensions. Table 1 presents the connotation of specific financial indicators under the respective dimension.

Table 1.

Primary Indicators of Comprehensive Index of Systemic Financial Risk (CISFR).

Dimension	Indicator	Economic connotation
Financial institution	Ratio of savings deposits	Reflect the stability of banks’ capital sources and the risk-resistance capability of financial institutions.
	The difference between loan and deposit growth rates	Reflect the liquidity risk of banks.
	Fund balance of bank financial products	Reflect the stability of banks’ capital sources and the risk-resistance capability of financial institutions.
	Proportion of non-performing loans	Reflect the quality of bank loan assets.
Capital market	Total value of domestic listed companies	Reflect the prosperity of stock market. The more prosperous the market, the greater the potential risk.
	Growth rate of stock trading volume	Reflect the investment vitality of stock market. The more active the market, the greater the potential risk.
	A-share average P/E ratio	Reflect the valuation of stock market. The greater the deviation from the benchmark, the higher the risk.
	Margin between the US and China 10-year treasury yields	Reflect the margin between domestic and foreign investment yields, the higher the margin. The more foreign inflows into China, the greater the potential risk.
	Margin between 10-year and 1-year treasury yields	Reflect various risks in the economy. The higher the margin, the greater the risk.
Monetary market	1-year Shibor-7-day Shibor	Reflect the liquidity risk, and smaller margin are usually out of sharply rising short-term interest rates. The lower the margin, the greater the risk.
	Import and export volume year-on-year	Reflect the prosperity of the real economy. The better the real economy, the smaller the exchange rate risk.
	REER index	Reflect the fluctuations of RMB exchange rate against various currencies in the foreign exchange market. The higher the value, the lower the RMB depreciation risk.
Real estate market	Growth rate of real estate accomplished revenue	Reflect the investment vitality in the real estate market.
	Growth rate of commercial housing sales	Reflect the prosperity of the real estate market.
	Hundred cities housing price index year-on-year	Reflect the price level of the real estate market.
FinTech	P2P turnover month-on-month growth	Reflect the vitality of online lending. The faster the transaction volume growth, the higher the credit risk.
	P2P problem platforms in the current month	The more problem platforms, the higher the risk.
	Interest margin of financing income	The more fund shortage in the market. The higher the margin, the greater the liquidity risk.
	Interest margin of online lending interest rate	Reflect the margin between P2P market and treasury bond market. The higher the margin, the greater the credit risk.
	Third-party payment scale	The larger the third-party payment scale, the greater the operational risk.

The detailed procedures for composing the systemic risk indicator are as follows. First, standardize each basic indicator within the sample period through the extreme value method. The indicators of positive value are processed in line with Equation (1) and the negative value according to Equation (2).

$${\mathrm{Z}}_1=\frac{{\mathrm{X}}_{\mathrm{i}}-\min (\mathrm{X})}{\max (\mathrm{X})-\min (\mathrm{X})}$$

(1)

$${\mathrm{Z}}_2=\frac{\max (\mathrm{X})-{\mathrm{X}}_{\mathrm{i}}}{\max (\mathrm{X})-\min (\mathrm{X})}$$

(2)

Where X refers to the indicators in Table 1, min (X) and max (X) represent the minimum and maximum values of the indicators during the sample period separately.

The second step is to synthesize the category index. Each index of the five dimensions is synthesized with an equal weight method. Thirdly, we synthesize the composite index. The crux is to determine the weight of each category index. To minimize the deviation of subjective assignment weights, we adopt the commonly used CRITIC (Criteria Importance Though Intercriteria Correlation) assignment method, in which the weight is determined by the product of the standard deviation of each category index and the conflict of each index. To be more specific, the larger the standard deviation of an index, the more information it contains, the smaller its correlation coefficient with other indicators, and the greater the conflict or index independence and thus a larger weight. The weight ω under the CRITIC assignment method can be expressed as Equation (3):

$$\mathrm{\omega }={\mathrm{\omega }}_{\mathrm{i}}=\frac{{\mathrm{\sigma }}_{\mathrm{i}}(\sum (1-{\mathrm{r}}_{\mathrm{ij}}))}{\sum ^{{\mathrm{\sigma }}_{\mathrm{i}}}(\sum (1-{\mathrm{r}}_{\mathrm{ij}}))}$$

(3)

Where ${\mathrm{\omega }}_{\mathrm{i}}$ is the index weight of i-dimension, ${\mathrm{\sigma }}_{\mathrm{i}}$ means the standard deviation of the i-dimension index, ${\mathrm{r}}_{\mathrm{ij}}$ represents the correlation coefficient between the i-dimension index and the j-dimension index. The standard deviation and correlation coefficient are inserted into Equation (3) to calculate the weight vector, then the CISFR is generated accordingly.

Main Explanatory Variable: FinTech Monopoly Index

Following Angelini and Cetorelli (2003) and Huang et al. (2016), we use the efficiency-adjusted Lerner Index, denoted as LI to measure the monopoly degree of FinTech companies. The calculation equation is written as follows:

$$\mathrm{LI}=\frac{\frac{\mathrm{PTP}}{\mathrm{TO}}+\frac{\mathrm{PTC}}{\mathrm{TO}}-\mathrm{MC}}{\frac{\mathrm{PTP}}{\mathrm{TO}}+\frac{\mathrm{PTC}}{\mathrm{TO}}}=\frac{\mathrm{PTP}+\mathrm{PTC}-\mathrm{MC}*\mathrm{TO}}{\mathrm{PTP}+\mathrm{PTC}}$$

(4)

Where PTP and PTC represent the company’s forecast profit and total cost, respectively; MC denotes the marginal cost of the company; TO expresses the company’s total output. Consistent with Degl’Innocenti et al. (2019), MC, PTP, and PTC in Equation (4) are obtained with the following two-input-one-output translog cost (profit) function:

$$\begin{array}{c}\multicolumn{1}{c}{\mathrm{lnT}{\mathrm{C}}_{\mathrm{it}}(\mathrm{lnT}{\mathrm{P}}_{\mathrm{it}})={\mathrm{\beta }}_{\mathrm{i}}\ln {\mathrm{Y}}_{\mathrm{it}}+\sum _{\mathrm{k}=1}^2{\mathrm{\alpha }}_{\mathrm{k}}\ln {\mathrm{P}}_{\mathrm{Kit}}+{\mathrm{\phi }}_{\mathrm{i}}{\mathrm{lnY}}_{\mathrm{it}}^2}\\ \multicolumn{1}{c}{+\sum _{\mathrm{k}=1}^2\sum _{\mathrm{K}=1}^2\ln {\mathrm{P}}_{\mathrm{kit}}\ln {\mathrm{P}}_{\mathrm{Kit}}+\sum _{\mathrm{k}=1}^2{\mathrm{\gamma }}_{\mathrm{k}}\ln {\mathrm{Y}}_{\mathrm{it}}\ln {\mathrm{P}}_{\mathrm{kit}}}\\ \multicolumn{1}{c}{+\mathrm{\sigma }\ln {\mathrm{E}}_{\mathrm{i}}+{\mathrm{\mu }}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\mathrm{it}}}\end{array}$$

(5)

Where TC denotes the company’s total cost, which is expressed as the total business cost; TP is the company’s pre-tax profit. Y represents the total assets of the company for output; P₁ expresses the labor cost in the input index, represented as the ratio of personnel expenses (annual change value payable to employee salary + cash paid to employees) to the total number of employees; P₂ denotes the capital cost, which is the ratio of interest expense to the company’s interest-bearing liabilities. The obtained PTC, PTP, and MC are substituted in Equation (3) to get the LI after efficiency adjustment. The higher the LI, the greater the market power of FinTech companies, and the higher the degree of monopoly. The structural indicator HHI is also adopted to measure the FinTech monopoly power for robustness check.

Other Controlled Variable

Referring to related banking and corporate finance literature, we select a series of controlled variables from three aspects, including basic firm-level variables, macro market environment, and global economic conditions. The impacts of these variables on financial risks have been widely examined in studies (Luo et al., 2022; Pour et al., 2023; Pozo, 2023; Qin & Zhou, 2019; Wang & Luo, 2022; Yin & Lu, 2022).

At the micro level, the size of FinTech enterprises (SIZE) is expressed as the natural logarithm of the total assets at the end of the year; profitability (ROA) is denoted as the return on total assets; the leverage ratio (LEV) is denoted as the ratio of shareholders’ equity in total assets; liquid asset (LIA) is obtained by dividing liquid assets to the total assets.

The domestic macro environment factors considered in the paper are presented as follows. GDP growth rate (GDPgr), which is obtained according to the growth rate of actual GDP (2018 = 100); monetary policy (M₁gr), expressed as M₁ monetary growth; financial development (STOCK), expressed as the proportion of stock market value in GDP. Since the national financial market risk is highly correlated with the global economy, we consider the interest rate margin between China and the United States, China’s real effective exchange rate, and foreign direct investment as global variables, abbreviated as AJRE, REA, and FDI, respectively.

Descriptive Statistics

Table 2 lists the descriptive statistics for the main regression variables. The FinTech monopoly (LI) of the sample firms is distributed with a mean value of 0.893 and a standard deviation of 0.086. The values range between a minimum of 0.32 and a maximum of 1. The relatively high average value of LI indicates that FinTech companies are highly concentrated. The average value of CISFR is 0.626, ranging from 0.367 to 0.635, showing a certain extent of fluctuation in this indicator. The distribution of other control variables is also within a reasonable range. For example, SIZE lies between 10.694 and 16.216 with a standard deviation of 0.897; the average value of ROA is 0.021, ranging from −1.068 to 0.297.

Table 2.

Descriptive Statistical Results.

Variables	Obs	Mean	Std. dev.	Min	Max
CISFR	613	0.525	0.064	0.367	0.635
FISR	613	0.659	0.185	0.137	0.866
CMSR	613	0.419	0.122	0.213	0.749
MMSR	613	0.503	0.129	0.319	0.730
RESR	613	0.490	0.163	0.206	0.785
FTSR	613	0.363	0.049	0.256	0.480
LI	613	0.893	0.086	0.320	1.000
SIZE	613	12.538	0.897	10.694	16.216
ROA	613	0.021	0.067	−1.068	0.297
LEV	613	1.688	0.726	1.014	7.283
LIA	613	0.693	0.155	0.277	0.985
GDPgr	613	1.614	3.420	−9.703	11.918
M1gr	613	2.396	3.265	−3.724	10.024
STOCK	613	2.428	0.293	1.823	3.296
REA	613	123.791	3.243	119.2	130.33
FDI	613	1.276	0.146	1.029	1.605
AJRE	613	110.924	4.835	102.366	122.118

Model Specification

The estimation model is constructed as below to verify the hypothesis:

$$\begin{array}{c}\multicolumn{1}{c}{\mathrm{CISF}{\mathrm{R}}_{\mathrm{t}}={\gamma }_0+{\gamma }_1·\mathrm{L}{\mathrm{I}}_{\mathrm{it}}+{\gamma }_2·\mathrm{MicroContro}{\mathrm{l}}_{\mathrm{it}}}\\ \multicolumn{1}{c}{+{\gamma }_3·\mathrm{MacroContri}{\mathrm{l}}_{\mathrm{t}}+{\gamma }_4·\mathrm{GlobalContro}{\mathrm{l}}_{\mathrm{t}}}\\ \multicolumn{1}{c}{+{\mathrm{u}}_{\mathrm{i}}+{\mathrm{Y}}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\mathrm{t}}}\end{array}$$

(6)

Where the subscript i denotes company i; t is the period; CISFR and LI express systemic risk and the degree of monopoly, respectively; MicroControl represents a series of control variables including SIZE, ROA, LEV, and LIA; MacroControl is domestic macro controlled variable (e.g., GDPgr, M1gr, and STOCK); GlobalControl denotes global economic variables (e.g., REA, FDI, and AJRE). ${\mathrm{\mu }}_{\mathrm{i}}$ expresses the individual fixed effects, Y_t is the time-fixed effect, and ${\mathrm{\epsilon }}_{\mathrm{t}}$ denotes a random error.

The fixed effect model (FE) is adopted for estimation. First, it is because the fixed effect model allows for individual effects that are unchanged over time. Furthermore, the above model makes enterprises irrelevant to the explanatory variables over time. The estimated standard deviation is adjusted in heteroscedasticity and then clustered at the FinTech company level. We also run a VIF test to check for the multicollinearity problems before regression, finding that there are no problems of multicollinearity since the mean VIF is 1.42 which is less than 10.

Empirical Analysis

In this section, regression analysis is conducted based on model (6) to investigate the effect of FinTech monopoly on systemic risk in China. Furthermore, robustness tests are performed by using alternative dependent and independent variables and changing estimation methods.

Benchmark Regression

Column (1) of Table 3 lists the results when there are no control variables, while columns (2) to (4) list the results after the addition of the control variables.

Table 3.

Empirical Results of FinTech Monopoly on Systemic Risk.

	(1)	(2)	(3)	(4)
Variables	CISFR	CISFR	CISFR	CISFR
LI	0.502*** (5.25)	0.253** (2.46)	0.262** (2.50)	0.136** (2.32)
SIZE		0.074*** (4.06)	0.078*** (4.00)	0.029** (2.53)
LEV		0.016 (0.68)	0.015** (1.89)	0.008** (1.79)
LIA		−0.000 (−0.01)	−0.003 (−0.04)	0.010 (0.27)
ROA		−0.018 (−0.55)	−0.031 (−0.72)	0.001 (0.02)
GDPgr			−0.001** (−2.73)	−0.001*** (−7.05)
M1gr			0.003*** (6.81)	0.005*** (23.65)
STOCK			−0.002 (−0.30)	0.009*** (3.98)
REA				−0.001 (−1.31)
AJRE				−0.008*** (−18.12)
FDI				−0.058*** (−16.38)
Constant	0.075 (0.88)	−0.658*** (−3.34)	−0.708*** (−3.38)	1.040*** (5.78)
Year dummy	Yes	Yes	Yes	Yes
Firm fixed effect	Yes	Yes	Yes	Yes
Observations	613	613	613	613
R 2	.180	.303	.326	.661
Number of code	32	32	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

As depicted in column (4) of Table 3, the regression coefficient of monopoly index LI is significantly positive at 1%, indicating that an increase of 1% growth in FinTech monopoly is associated with an increase of 13.6% in the systemic risk. The above results reveal that the rapid growth of FinTech in China has increased the complexity, relevance, and infectivity of the original financial system while further aggregating the fragility of the national financial system, in line with the “competitive-stability theory” of banking literature (Abduraimova, 2022; Borauzima & Muller, 2023).

The monopoly of FinTech may exacerbate systemic risk by intensifying traditional credit risks, creating cross-sector risks, and producing new types of financial risks. To begin with, traditional financial institutions are easily affected by the risks of FinTech companies. When financial institutions continue to strengthen their cooperation with third-party payment, P2P lending, crowdfunding, and other institutions, the chain reaction caused by non-standard cooperation, violations, and imperfect supervision can easily lead to accountability for traditional financial institutions, leading to credit risk outbreaks. As FinTech firms become more monopolistic, such traditional credit risk increases more. In addition, the development of financial technologies has promoted the integration of financial institutions and FinTech corporations, making risks easily crossing among these participants. FinTech companies usually have deep cooperation with banks and other financial institutions, so their operational risks can be transmitted to financial institutions such as banks through their relevant financial accounts. When FinTech companies have a monopoly position in the market, the strength of the above correlation increases. Last but not least, the development of FinTech monopoly is accompanied by new risks such as data risk and technological risk. For one thing, data privacy is the most important issue in the field of FinTech. Although more data can help improve the efficiency of credit evaluation, excessive collection of customer data by large FinTech firms may infringe on customer privacy. Moreover, once these companies establish a dominant and monopolistic position in the data field and use customer personal information for credit evaluation, they can engage in price discrimination, thus affecting the fairness of credit. On another, due to the current lack of effective breakthroughs in security technology in FinTech, the technical deficiencies of FinTech itself and its dependence on information systems can lead to a decrease in FinTech security performance and an expansion of the scope of security issues.

Other controlled variables also take on certain implications as listed in column (4) of Table 3. At the micro level, the regression coefficient of SIZE is .029, significant at 5%, implying that the larger the size of a FinTech platform, the closer its connection between the financial industry, and the wider range and speed of risk contagion, making it more likely to trigger systemic risk (Qin & Zhou, 2019). At the macro level, the regression coefficient of GDPgr is significantly negative at 1%, thus revealing that the emergence of systemic risk is correlated with the economic cycle. Optimistic market sentiment boosts investment while maintaining sufficient liquidity in the market during economic prosperity. While during the economic recession, investment is significantly reduced and liquidity is tightened, the risk in the financial market spreads rapidly with the connection between financial institutions, thus aggravating systemic risk. The regression coefficient of M₁gr is 0.005 with significance at 1%, indicating that monetary policy affects systemic risk. Specifically, the expansionary monetary policy will bring greater liquidity with sufficient capital in the market, whereas excessive liquidity may eventually cause speculation and bubbles that can increase systemic risk (Iwanicz-Drozdowska & Rogowicz, 2022).

Furthermore, the coefficient of STOCK is significantly positive at 1%, suggesting that the continuous expansion of the financial market scale represented by the stock market will exacerbate the risk contagion and resonance effect between financial institutions (Kanga et al., 2023). As a result, the risk exposure of the macroeconomy is enhanced. Besides, the regression coefficient of AJRE is .008, significant at 1%, implying a positive correlation with systemic risk. AJRE serves as the barometer of the global economy and it has been the ideal haven in economic crisis for decades. The rising AJRE reflects the global economic fluctuations and may trigger an economic crisis, thus causing higher systemic risk, which is in line with Hasan et al. (2023). The regression coefficient of FDI is −.058, significant at 1%, suggesting a negative correlation with systemic risk. FDI indicates the national economic development to a certain extent. Higher FDI reveals their more optimistic attitude toward international capital on the economic growth of the country, which is characterized by a healthier economic structure and a lower possibility of systemic risk (Kellard et al., 2020).

Robustness Tests

First, the effects of FinTech monopoly on systemic risk at its five dimensions are studied, and the results are listed in columns (1) to (5) of Table 4, with the five dimensions, are FISR, CMSR, MMSR, RESR, and FTSR, representing systemic risk from traditional financial institutions, capital market, monetary market, real estate market, and FinTech market, respectively. The results indicate that the coefficients of LI are significantly positive when the systemic risk is replaced by FISR and FTSR, whereas the coefficients of LI on CMSR, MMSR, and RESR are not significant, but all show positive signs.

Table 4.

Alternative Measures of Systemic Risk.

	(1)	(2)	(3)	(4)	(5)
Variables	FISR	CMSR	MMSR	RESR	FTSR
LI	0.204** (2.22)	0.047 (0.65)	0.097 (1.08)	0.195 (0.80)	0.067** (2.20)
Constant	4.103*** (9.67)	0.913*** (5.25)	0.504** (2.67)	−2.539*** (−4.85)	0.141 (1.19)
Micro control	Yes	Yes	Yes	Yes	Yes
Macro control	Yes	Yes	Yes	Yes	Yes
Global control	Yes	Yes	Yes	Yes	Yes
Year dummy	Yes	Yes	Yes	Yes	Yes
Firm fixed effect	Yes	Yes	Yes	Yes	Yes
Observations	613	613	613	613	613
R 2	.779	.237	.421	.461	.562
Number of code	32	32	32	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

Second, we apply an alternative measure of monopoly power and display the results in column (1) of Table 5. The quadratic sum of the total assets owned by top five FinTech firms is calculated as HHI indicator. The regression result shows that the concentration of FinTech companies will aggravate the systemic risk, consistent with our benchmark regression, indicating the robustness of the research result.

Table 5.

Alternative Measure of Monopoly Power and Different Estimation Methods.

	(1)	(2)	(3)	(4)
Variables	CISFR	CISFR	CISFR	CISFR
HHI	0.532*** (48.52)
LI		0.111*** (4.94)	0.136*** (2.70)	0.095*** (3.04)
Constant	1.198*** (29.42)	1.592*** (19.23)	1.040*** (2.70)	1.457*** (11.55)
Micro control	Yes	Yes	Yes	Yes
Macro control	Yes	Yes	Yes	Yes
Global control	Yes	Yes	Yes	Yes
Year dummy	Yes	Yes	Yes	Yes
Firm fixed effect	Yes	Yes	Yes	Yes
Observations	613	613	613	613
R 2	.771			.716
Number of code	32	32	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

Third, we apply alternative estimation methods which are the random effects model, the Driscoll-Kraay model, and the Prais model, and demonstrate the results in columns (2) to (4) of Table 5, respectively. The results show that the coefficients for FinTech monopoly power are significantly positive with systemic risk at the 1% level for all models, further validating our previous conclusion that an increase in the monopoly power of FinTech companies significantly raises systemic risk.

Further Discussions

Further discussions including heterogenous and mechanisms analysis are conducted in this section.

Heterogenous Analysis

The casual relationship between FinTech monopoly and systemic risk may be affected by several factors (e.g., firm-specific ones and macro ones). To test the heterogenous inference, the interaction term is added to model (6) to explore the role of factors in the nexus of interest.

Table 6 lists the heterogenous results of the enterprise’s micro characteristics on the relation between FinTech monopoly and systemic risk. Columns (1) and (2) demonstrate the regression results of firm size and leverage, respectively. As depicted in column (1), the FinTech platform’s size presents a positive effect on the “monopoly-systemic” relation, that is, the monopoly behavior of large-scale companies affects systemic risk more significantly than firms with smaller size. The FinTech platforms with large sizes are “too big to fail,” thus increasing the likelihood of systemic risk. Furthermore, the firm’s leverage also has an effect in increasing the risk of platform monopoly as suggested in column (2), that is, the higher the LEV, the greater the effect of FinTech monopoly on systemic risk. High leverage implies a low level of capital, which affects the company’s profitability and lowers its risk-resistance capability. Therefore, high leverage may affect the company’s operation capacity which soon induces systemic risk once a problem occurs.

Table 6.

Heterogeneity Test of Enterprises Characteristic.

	(1)	(2)
Variables	CISFR	CISFR
LI	−0.207 (−1.05)	0.037 (0.95)
LI × SIZE	0.028** (2.15)
LI × LEV		0.030*** (2.90)
Constant	1.414*** (15.88)	0.891*** (6.07)
Micro control	Yes	Yes
Macro control	Yes	Yes
Global control	Yes	Yes
Year dummy	Yes	Yes
Firm fixed effect	Yes	Yes
Observations	613	613
R 2	.658	.645
Number of code	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

Table 7 lists the results of the macro environment including both domestic and global economic conditions on the nexus between FinTech monopoly and systemic risk. Columns (1) to (3) illustrate the heterogenous regression results of domestic conditions which are economic growth, monetary policy, and capital market development, respectively, while columns (4) and (5) list the results of global conditions. From the perspective of the domestic environment, the GDP growth rate negatively affects the risks of FinTech monopoly, while monetary policy and financial development show a positive effect. Specifically, the economic growth will reduce the business operation and capital management risks led by monopolistic FinTech platforms, and alleviate the potential systemic risk; expansionary monetary policy may further trigger speculation in FinTech companies through the “risk-taking channel of monetary policy”; the boosted financial development may generate greater bubbles, which could lead to the monopoly of FinTech companies with massive funds for speculation, thus increasing the possibility of systemic risk.

Table 7.

Heterogeneity Test of Macro Conditions.

	(1)	(2)	(3)	(4)	(5)
Variables	CISFR	CISFR	CISFR	CISFR	CISFR
LI	0.138** (2.35)	0.124** (2.11)	0.103 (1.69)	1.166*** (17.90)	0.224*** (3.87)
LI × GDPgr	−0.001*** (−6.87)
LI × M1gr		0.005*** (21.40)
LI × STOCK			0.013*** (5.02)
LI × AJRE				−0.009*** (−19.86)
LI × FDI					−0.067*** (−15.84)
Constant	1.039*** (5.77)	1.062*** (5.87)	1.073*** (6.05)	0.172 (1.24)	0.960*** (5.36)
Micro control	Yes	Yes	Yes	Yes	Yes
Macro control	Yes	Yes	Yes	Yes	Yes
Global control	Yes	Yes	Yes	Yes	Yes
Year dummy	Yes	Yes	Yes	Yes	Yes
Firm fixed effect	Yes	Yes	Yes	Yes	Yes
Observations	613	613	613	613	613
R 2	.661	.659	.662	.660	.662
Number of code	32	32	32	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

As to the macro global environment, the rising AJRE indicates the possible economic fluctuations worldwide. Considerable safe-haven funds for the Japanese yen increase the financial risks following platform monopoly and even trigger systemic risk. The larger the proportion of foreign direct investment in total GDP, the more optimistic foreign capital regards the national or regional economy, which will lower the risks of business operation and capital management brought by the monopoly of FinTech platforms, thus decreasing the possibility of systemic risk.

Exploring the Underlying Mechanisms

As discussed in the theoretical analysis, some possible channels are proposed. To be specific, at the consumer level, this study attempts to explore whether the monopoly of FinTech platforms increases the leverage ratio of households by triggering excessive consumption and premature consumption, thus resulting in systemic risk. At the financial institutions level, we attempt to test whether the FinTech monopoly impacts systemic risk by increasing the credit risk. At the FinTech innovation level, we aim to examine whether the monopoly will transfer systemic risk by inhibiting FinTech development. The intermediate effect model is applied as follows:

$$\begin{array}{c}\multicolumn{1}{c}{{\mathrm{M}}_{\mathrm{t}}={\mathrm{\alpha }}_0+{\mathrm{\alpha }}_1·\mathrm{L}{\mathrm{I}}_{\mathrm{it}}+{\mathrm{\alpha }}_2·\mathrm{MicroContro}{\mathrm{l}}_{\mathrm{it}}}\\ \multicolumn{1}{c}{+{\mathrm{\alpha }}_3·\mathrm{MacroContro}{\mathrm{l}}_{\mathrm{t}}+{\mathrm{\alpha }}_4·\mathrm{GlobalContro}{\mathrm{l}}_{\mathrm{t}}}\\ \multicolumn{1}{c}{+{\mathrm{u}}_{\mathrm{i}}+{\mathrm{Y}}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\mathrm{t}}}\end{array}$$

(7)

$$\begin{array}{c}\multicolumn{1}{c}{\mathrm{CISF}{\mathrm{R}}_{\mathrm{t}}={\gamma }_0+{\gamma }_1·\mathrm{L}{\mathrm{I}}_{\mathrm{it}}+{\gamma }_2·\mathrm{MicroContro}{\mathrm{l}}_{\mathrm{it}}}\\ \multicolumn{1}{c}{+{\gamma }_3·\mathrm{MacroContri}{\mathrm{l}}_{\mathrm{t}}+{\gamma }_4·\mathrm{GlobalContro}{\mathrm{l}}_{\mathrm{t}}}\\ \multicolumn{1}{c}{+{\mathrm{\alpha }}_5·{\mathrm{M}}_{\mathrm{t}}+{\mathrm{u}}_{\mathrm{i}}+{\mathrm{Y}}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\mathrm{t}}}\end{array}$$

(8)

Where M_t represents intermediary variables (e.g., household leverage ratio [HousholdLEV], commercial bank non-performing loan ratio [NPL], as well as FinTech development index [FinTech]). Tables 8 to 10 list the results of the intermediate effect regressions.

Table 8.

Impact Mechanism at the Consumer Level.

	(1)	(2)
Variables	HousholdLEV	CISFR
LI	22.860*** (3.06)	0.013 (0.70)
HousholdLEV		0.006*** (70.38)
Constant	27.626 (0.91)	0.926*** (27.12)
Micro control	Yes	Yes
Macro control	Yes	Yes
Global control	Yes	Yes
Year dummy	Yes	Yes
Firm fixed effect	Yes	Yes
Observations	613	613
R 2	.634	.821
Number of code	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

Table 9.

Impact Mechanism at the Financial Institution Level.

	(1)	(2)
Variables	NPL	CISFR
LI	0.382*** (3.02)	0.016** (1.89)
NPL		0.400*** (68.96)
Constant	2.184*** (5.33)	0.167*** (4.73)
Micro control	Yes	Yes
Macro control	Yes	Yes
Global control	Yes	Yes
Year dummy	Yes	Yes
Firm fixed effect	Yes	Yes
Observations	613	613
R 2	.692	.837
Number of code	32	32

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

Table 10.

Impact Mechanism at the FinTech Development Level.

	(1)	(2)
Variables	FinTech	CISFR
LI	−4.506*** (−3.89)	0.019* (1.63)
FinTech		−0.024*** (−27.91)
Constant	10.295* (1.86)	1.346*** (18.38)
Micro control	Yes	Yes
Macro control	Yes	Yes
Global control	Yes	Yes
Year dummy	Yes	Yes
Firm fixed effect	Yes	Yes
Observations	613	613
R 2	.628	.752
Number of code	31	31

Note. It is t value in parenthesis. *, **, and *** indicate significance at 10%, 5%, and 1% significance levels, respectively.

We list the channel results at the consumer level in Table 8. Column (1) lists the regression result of the FinTech monopoly on HousholdLEV, and column (2) presents the regression result after the intermediary variable is added. As depicted in these columns, the explanatory variables in column (1) are positively correlated with HousholdLEV, and those in column (2) are significantly positive, whereas the regression coefficient of our main explanatory variable is not prominent. The above results reveal that the intermediary variable fully plays its role, that is, the influence mechanism of the benchmark regression triggers the systemic risk of monopoly platform with rising HousholdLEV, consistent with the analysis that FinTech monopoly will induce excessive and premature consumption, raising the consumer leverage and cause systemic risk.

Table 9 lists the regression results of the mechanism at the financial institution level. Column (1) lists the regression results of the FinTech monopoly on the intermediary variable, while column (2) lists the regression results after the inclusion of the intermediary variable. As shown in the columns, the non-performing loan ratio of banks is positively correlated with the explanatory variables, and the regression coefficient of NPL in the model (2) reaches .4, while the LI indicator is still significantly positive. The above results suggest that the monopoly of FinTech corporations will incur the system risk through the credit risk channel of commercial banks. The result reveals that the FinTech monopoly will intensify the price competition between conventional financial institutions and FinTech firms, impede the profits of commercial banks, and pose potential external challenges to them. To maintain profits, banks may lower the asset selection criteria, thus increasing credit risk. In the absence of supervision, FinTech platforms can down-regulate the threshold for lending to attract more users. However, this down-regulation will further aggregate the spread of financial risks to bank credit, thus triggering systemic risk.

The regression results of the influencing mechanism at the FinTech development level are shown in Table 10. Referring to current studies, the competitive FinTech development index is established by Principal Component Analysis with the search results of the Baidu Index (Wang et al., 2021). Column (1) lists the regression results of the monopoly FinTech platform on the intermediary variable, and column (2) lists the regression result after the addition of the intermediary variable. It is indicated that the FinTech development index is negatively correlated with the FinTech monopoly, and the regression coefficient of the intermediary variable in the model (2) reaches −.024 at a 1% level of significance, with the LI index significantly positive which is in line with the benchmark. Therefore, it reveals that the monopoly of FinTech companies can spread systemic risk by inhibiting the competitive development of FinTech.

Conclusions

In this paper, we first construct the monopoly index of FinTech companies and FinTech-involved systemic risk indicator and then empirically examine the effect of FinTech monopoly on systemic risk in China. The empirical results indicate that the monopoly of FinTech companies will exacerbate systemic risk. Moreover, the effect is heterogeneous among several factors, including micro characteristics (e.g., size and leverage ratio of FinTech companies), domestic macro factors (e.g., economic growth, monetary policy, and financial market development), as well as international macro factors (e.g., exchange rate and foreign direct investment). We also find possible channels through which FinTech monopoly works on systemic risk, which are via inducing excessive consumption and pre-mature consumption, increasing banks’ credit risk, and inhibiting the development of FinTech.

Policy Implications

Some policy implications can be derived from the above results. Firstly, the regulatory system of FinTech companies in financial business shall be continually improved, making them regulated by the financial industry rules; meanwhile, the authority should take legal measures to ensure a fair competition environment, preventing FinTech enterprises from relying on their technological advantages to gain market dominance and avoiding risks resulting from FinTech monopoly as much as possible.

Secondly, while limiting the monopoly development of Fintech firms, the authorities should also encourage the competitive development of Fintech. Although the baseline results show that FinTech monopoly will incur systemic risk, mechanism analysis demonstrates that a prosperous FinTech market where all participants have a fair development chance can bring financial stability.

Thirdly, our findings also indicate that FinTech monopoly can trigger financial risks by increasing excessive consumption. Therefore, it is also necessary to introduce corresponding measures to establish a healthy consumer market, which can prevent the excessive consumption brought about by the application of financial technologies such as big data of the FinTech platforms.

Fourthly, it is shown that systemic risk can be induced by FinTech monopoly via the instability of the banking system. Hence, commercial banks should also actively strengthen their digital construction, to prevent the adverse effects of FinTech development on themselves and avoid becoming the channels for financial risk transmission.

Limitations

This study has some limitations. On one hand, due to the availability of data, the FinTech companies selected in this paper are only listed companies in China’s A-share market, and the top companies in the industry with larger trading volumes are not included since they are listed abroad. Therefore, the risk effects of FinTech monopoly may be underestimated. On the other hand, due to the more complex business of FinTech companies compared to traditional financial institutions, the efficiency-adjusted Lerner index calculated in the paper may also be slightly different from reality. In addition, considering data availability, we only consider FinTech enterprises when calculating the FinTech monopoly indicator and do not include traditional institutions in the analysis.

In the future, with the improvement of information disclosure, more FinTech enterprises should be included in the sample to further improve the results. Moreover, since we do not categorize the segmented fields such as payment, software sales, loan business, and so on of the FinTech companies, there may be deviations between the measurement results of market power and the actual situation. In future research, it is necessary to analyze the relationship between the market power of different business types of FinTech companies in the industry and systemic risk. Meanwhile, if more data can be obtained, it is meaningful to consider the impact of financial structures which include both traditional institutions and emerging FinTech companies on financial risk.