**1. Introduction**

Although it only entered the literature five years ago, FinTech has been studied a lot. It refers to companies and Finance 4.0 that create financial technologies at the highest level. Globally, FinTech is being implemented rapidly in human life in recent years.

Of recent FinTech studies, some focus on all aspects of the issue in general (e.g., Arner et al. 2016; Zalan and Toufaily 2017; Dospinescu et al. 2021), while others examine morespecific aspects. These include studies related to banks and traditional financial institutions (Kotarba 2016; Buchak et al. 2018; Hu et al. 2019), venture capital, cryptocurrencies, and blockchain (Kaplan and Lerner 2016; Ante et al. 2018; Gozman and Willcocks 2019; Kim et al. 2018; Ji and Tia 2021; Mora et al. 2021), insurance (Yan et al. 2018b; Stoeckli et al. 2018), and asset management (Rogowski 2017; Dugast and Foucault 2018). While each study adds an important perspective on the subject, a bibliometric analysis can provide a broader perspective and assessment than has been the case for studies thus far.

A network analysis carried out through bibliometric analysis defines new areas and information on the subject more strongly. It can also identify research groups and researchers to show how various areas of thought have emerged. Finally, it can identify leading and influential researchers in these research groups, identify different and new issues addressed by these influential researchers, and identify areas of study related to these new issues.

**Citation:** Tepe, Gencay, Umut Burak Geyikci, and Fatih Mehmet Sancak. 2022. FinTech Companies: A Bibliometric Analysis. *International Journal of Financial Studies* 10: 2. https://doi.org/10.3390/ijfs 10010002

Academic Editor: Sabri Boubaker

Received: 18 October 2021 Accepted: 22 December 2021 Published: 28 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

This study provides a detailed and comprehensive analysis by identifying researchers and publications with high influence in this pool, starting with 401 studies focusing on future-oriented FinTech applications. Various performance indicators were calculated for the bibliometric analysis. The formulization of the methods used was encoded by the authors on an R-based basis using the R Studio 1.2 program. The data processed through the program were then handled with Gephi 0.8.2 and VOSviewer 1.6.11 programs for visualization and mapping purposes to obtain the final outputs.

FinTech, which is short for financial technology, has spread rapidly worldwide, although its importance varies from country to country depending on the level of economic development and market structure (Berkmen et al. 2019). The concept, which originated in the early 1990s, currently refers to a rapidly evolving process in financial services (Arner et al. 2017; Hochstein 2015). FinTech describes companies offering financial services using modern creative technologies that *"attract customers with products and services that are more user-friendly, efficient, transparent and automated than those currently available"* (Dorfleitner et al. 2017, p. 5). FinTech firms cannot be defined within legal parameters because they operate in different business lines and models, and a wide range of industries, from crowdfunding to credit providers, cryptocurrencies to angel-investment networks.

FinTech has developed through three basic stages (Arner et al. 2017). The first phase resulted from surplus production and technological innovations brought about by the industrial revolution with the use of the first simple abacuses. After the mid-1800s, the invention of the telegraph (Nicoletti 2017), and telegraph communication and intensive trade between countries, enabled financial transactions to be made on a global scale using technology (Standage 2013). From 1866 to 1967, the financial services industry was heavily connected with technology but remained largely analog. This period is called FinTech 1.0.

Developments in digital technology between 1967 and 2008, known as FinTech 2.0, enabled financial-services technologies to switch from analog to digital and become globalized. For example, Barclay's Bank was the first to introduce automatic teller machines (ATMs) in 1967 (Nicoletti 2017), while electronic payment systems significantly changed the financial structure, making money transfers between banks and countries quite easy. In 1975, the Basel Committee was established to reduce the risks of interbank money transfers, with new rules to regulate relations between international banks (History of the Basel Committee 2019).

In terms of what consumers expect from a bank, e-banking/m-banking, the possibility of performing ATM cash transactions, customer service, ease of use, and the volume of information on the card have become very important (Dospinescu et al. 2019). On the other hand, rapidly developing financial integration that connected global markets marked a new era, FinTech 3.0, after 2008, especially with the introduction of the Internet, when Wells Fargo presented the first internet-banking experience, and startups and technology companies began offering financial products and services directly to businesses and consumers (Arner et al. 2016). These FinTechs considerably damaged the profitability of the banking sector (Zalan and Toufaily 2017).

While we currently still live in FinTech 3.0, a future upgrade to FinTech 4.0 will happen. Arner et al. (2017) even claim that FinTech 4.0 arrived in 2018, thanks to applications such as the Internet of things, big data, artificial intelligence, and cloud computing.

Figure 1 shows how FinTech is segmented into four fundamental sectors (Dorfleitner et al. 2017). Financing is the segment that provides funding for individuals or organizations through crowdfunding and credit and factoring. Crowdfunding usually involves raising small amounts of money from large numbers of people via the Internet or social media. The most important feature is the setting of the deadline. If the target amount cannot be reached within the specified period, the operation is canceled (Lee and Kim 2015). Credit and factoring are the processes by which FinTech firms provide financing to individuals or companies cheaply and quickly by automating transactions in collaboration with banks. The second main sector, asset management, includes services such as social trading, Roboadvice, personal financial management (PFM), and investing and banking. The third sector, payment, refers to national and international payment transactions. These include virtual payment methods, such as cryptocurrencies and blockchains, which are used as alternatives to conventional monetary transactions. Finally, other FinTechs cannot be classified within the first three traditional banking functions. These include insurance, search engines, comparison sites, technology, IT, and infrastructure.

**Figure 1.** FinTech Segmentations. **Source:** Dorfleitner et al. (2017), cited by Al-Ajlouni and Al-Hakim (2018).

The term bibliometric was first used in the Journal of Documentation (Fairthorne 1969). Bibliometrics (sometimes called scientometrics) involves quantitative analysis, which is the main tool of science. It provides a statistical analysis of data, such as how frequently journal articles are cited. Comparisons can be made across countries and research branches. While bibliometric analyses are becoming increasingly popular, their novelty means that there are no studies yet that directly investigate FinTech. Only two (Milian et al. 2019; Wu 2017) have used the word FinTech in their titles. However, they did not exclusively focus on it. Instead, their analyses drew on the following segments: payments, deposit, and landing, insurance, capital raising, investment management, and market provisioning. A few bibliometric studies have focused on individual segments within FinTech, such as crowdfunding (Martínez-Climent et al. 2018; Blasco-Carreras et al. 2015; Blažun Vošner et al. 2017), payments (Karafiloski and Mishev 2017; Cao et al. 2017; Dabbagh et al. 2019; Zheng et al. 2018; Merediz-Solà and Bariviera 2019; Liu 2016), asset management (Yan et al. 2018a), and other financing functions (Kumari and Sharma 2017; Cancino et al. 2017).

#### **2. Research Method**

The basic aim of a bibliometric analysis is to collect previous literature and related topics on the research subject to form objective findings that can be tested and replicated. It aims to both categorize previous studies and offer a rigorous methodological examination of the research results. To show that the study adds new information to the literature, the results should be defined in accordance with the research questions.

#### *Research Questions*

In the present study, FinTech-related publications and researchers were subjected to structural categorical analysis. Following Milian et al. (2019), the following two basic research questions with three sub-questions and two others, a total of seven, were addressed.

RQ1. How has the literature developed between 2015 and 2021?

RQ1.1. What are the most influential studies and authors?

RQ1.2. What are the main studies in FinTech?

RQ1.3. What are the distributions and impacts of publications over time?

To respond to RQ1, it is necessary to group the important studies, identify the relationships between them, and categorize them within the framework of current studies. This leads to the second question:

RQ2. What are the important topics in the FinTech literature?

In responding to this question, Lotka's Law and Bradford's Law, which are classics in bibliometric analysis, were assessed for their compatibility with the data.

RQ3. Are the results compatible with Lotka's Law? RQ4. Are the results compatible with Bradford's Law?

#### **3. Sampling and Methodology**

Bibliometric analysis was used to determine the scope of the scientific FinTech literature. The bibliometric analysis used in this research is a very detailed and comprehensive analysis technique in this field.

#### *Bibliometric Analysis*

The bibliometric analysis identifies the most prolific countries and universities, and the most influential authors, studies, and journals. The FinTech and bibliometric analysis dataset for the study was taken from Scopus. While Web of Science was also scanned, it was excluded from the evaluation because it provided considerably fewer studies than Scopus. The bibliometric analysis technique aimed to reveal the evolution of the FinTech research literature in terms of RQ1.1, RQ1.2, and RQ1.3, specifically the most influential articles, authors, and topics.

Many factors can be examined in bibliometric analysis. However, the analysis to be performed must be suitable for the purpose. The present study followed the method proposed by Cadavid Higuita et al. (2012), Albort-Morant and Ribeiro-Soriano (2016), and Martínez-Climent et al. (2018). In this method, the indicators are divided into three types: quantity, quality, and structural indicators (Martínez-Climent et al. 2018). (1) Quantity indicators contain numerical data for the area to be analyzed. (2) Quality indicators show the academic impact of publications. (3) Structural indicators reveal the relationships between publications.

Social network analysis is used for measuring both quality and structural indicators. In social network analysis, the network consists of nodes connected through networks (Wasserman and Faust 1994). This determines the centrality of each author by the number of connections they make with other members of the network. Centrality has three main principles: degree, closeness, and betweenness (Freeman 1979, cited by Milian et al. 2019). Centrality degree indicates how many co-publications an author has. Betweenness measures the number of times a node captures the shortest route between two other nodes, and thus shows the binding role that the author plays among other authors. Farness is the sum of the shortest distance of one node from other nodes while proximity is the opposite of farness. The greater the degree, the less the total distance from one node to all other nodes (Milian et al. 2019). Authors with high proximity reach new information faster and spread their ideas more quickly.

RQ2 addresses the issue of which topics FinTech researchers focus on while Lotka's Law (Lotka 1926), which measures authors' scientific productivity, was addressed through RQ3. According to Lotka's Law, the number of authors contributing to the literature with n number of studies is 1/n<sup>2</sup> of the number of authors contributing to the literature with a single study.

RQ4 assesses Bradford's Law (Bradford [1929] 1985), which determines the distribution of references to journals. According to this law, a bibliographic study on any subject will show that there is a small core group of journals that publishes a third of all articles in this field. A second, larger group of journals publishes the next third while the biggest group of journals publishes the remainder.

#### **4. Findings**

This section presents the results, periods, publications, authors, and other information of the analyses.

A total of 636 publications were scanned in Scopus for academic papers on FinTech (including journal articles, conference papers, books, and book chapters). Figure 2 specifies the number of publications found by years.

**Figure 2.** Publishing trend in FinTech. Note(s): This figure represents the publication trend of academic papers on FinTech between 2015 and 2021. The data were retrieved from the Scopus database using the keyword "FinTech".

Figure 2 shows that FinTech has grown geometrically since 2015 when it first emerged as a concept. The papers were written by 1445 different authors from 387 different journals and books. In the scanned sources, the average citations per document were 7.52, the number of documents per author was 0.44 while the collaboration index was 2.75.

Table 1 shows which ten universities had the most affiliations of FinTech authors. Universities in Asian countries have contributed the most (6 of the top 10). Table 1 also shows total production (TP), total citations (TC), and citations per publication (CPP).



Note(s): This table was created with a dataset from Scopus via Excel.

Figure 3 shows which 10 institutions sponsored the most FinTech papers.

**Figure 3.** Top 10 funding sponsors of documents. Note(s): This figure represents the 10 institutes that sponsored the most academic articles on FinTech between 2015 and 2021. The data were taken from the Scopus database using the keyword "FinTech".

Figure 4 shows the geographical locations of all contributing countries, with the number of publications decreasing from dark to light blue, while grey indicates no contribution. China, the USA, and the UK have the highest contributions.

**Figure 4.** Geographical locations of contributing countries. Note(s): This figure was created with a dataset from Scopus via R Studio.

Figure 5 shows the country collaboration map. UK authors have 59 joint publications with authors in other countries, including 7 with Chinese authors, 6 with Australian authors, and the remaining 46 collaborations with authors in 25 different countries. The UK is followed by the US with 54 collaborations, China with 52, Australia with 43, and Singapore with 18.

Table 2 lists the 10 most productive journals. The journals that are not in the field of Finance and Entrepreneurship were excluded from the analysis results. At the top, Sustainability Switzerland has the most publications on FinTech, and a TC value of 96, whereas the second-ranked, Lecture Notes in Computer Science including subseries, has a TC value of 40. Despite only ranking tenth, Industrial Management and Data Systems has the highest CPP value at 33.2, and Financial Innovation has the second-highest CPP value at 15.28.

**Figure 5.** Country collaboration map. Note(s): This figure was created with a dataset from Scopus via R Studio.

**Table 2.** Most-productive journals.


Note(s): This table was created with a dataset from Scopus via Excel.

As the most productive country, China has 87 publications, 745 citations, and 8.56 citations per publication. Six of the top ten are Asian countries, two are European, while Australia represents Oceania. The most productive countries all have h-index values of 4 or above (Table 3).



Note(s): This table was created with a dataset from Scopus via Excel.

#### *4.1. Author Influence*

Table 4 shows which authors have been most prolific. Authors with four or more publications between 2015 and 2021 are listed in Table 4. Rabbani, with the most publications, has a TC score of 55. Tan, who is in second place, has a TC score of 105. More than half of Rabbani and Khan's citations are self-citations. Muthukannan has the lowest CPP score, which indicates a weak correlation between that author's large number of publications and their impact factor.

**Table 4.** The top 10 contributing authors' number of published articles and self-citations.


Note(s): This table was created with a dataset from Scopus via Excel.

Table 5 shows which authors received the most citations by year. Table 4 above shows citations based on total publications, whereas Table 5 shows the most citations received by one study. The 10 authors with the most citations had 1342 citations in total, with an average of 134.2 citations each.


Note(s): This table was created with a dataset from Scopus via Excel.

Figure 6 shows the frequency of citations of individual articles. The size of each node indicates the number of citations. The most-cited authors were Gomber et al. (2018), Lee and Shin (2018), Gomber et al. (2017), Gabor and Brooks (2017), Buchak et al. (2018), Gai et al. (2018), Schueffel (2016), Leong et al. (2017), Haddad and Hornuf (2019), and Shim and Shin (2016).

**Figure 6.** Frequency of citations of publications (Fractionalization). Note(s): This figure was created with a dataset from Scopus via VOSviewer.
