EJBF_13-1_2023-92-112

1. Introduction

In today’s business world, small and medium-sized enterprises (SMEs) constantly face challenges and opportunities in a dynamic and competitive environment. A crucial aspect that determines the success of these organizations is their capacity to manage their financial resources efficiently, particularly their working capital (Kayani et al., 2019). Mann (1918) defines working capital as money required to perform the existing operations of an entity. This is also known as the net working capital, i.e., the amount required to keep the business running. While Schaal and Haley (1991) consider working capital management as the management of current assets and current liabilities. This is an essential component when optimizing the cash conversion cycle and for financial management in firms as it allows their operations to function properly and provides the liquidity needed to face market challenges. Working capital in SMEs becomes particularly relevant because of the characteristics and limitation of these firms, such as the scale of operations, limited access to finance, and sensitivity to market fluctuations. There are specific implications of individual versus cooperative working capital management in a supply chain. In this regard, Huang et al. (2022) take up the work of Hofmann and Kotzab (2010), who recommend that managers establish long-term collaborative relationships, particularly in cooperative working capital management, in order to finance the weaker members and improve overall performance instead of benefiting just some of the participants.

Working capital management has been key throughout modern history to provide liquidity in supply chains (SCs), by means of financial collaboration tools within the framework of Supply Chain Finance (SCF), and the financial solutions that have stemmed from this such as Factoring (Gelsomino et al., 2016b; Gomm, 2010; Liebl et al., 2016). Since the mid-20th century, SCF has stood out as an important support medium for commercial and financial activities throughout a swathe of industrial sectors at an international level. SCF has attracted attention by being an innovative model that allows businesses, particularly SMEs—which are mainly family run (D’Angelo et al., 2016)—to convert their credit sales into cash sales without incurring additional liabilities. SCF seeks to optimize financial flows on an inter-firm level by using solutions implemented by financial intermediaries (Camerinelli, 2009) or technology providers (Lamoureux & Evans, 2011) in order to align financial flows with product and information flows within the SC (Gelsomino et al., 2016a; Wuttke et al., 2013b).

This collaboration between firms is particularly important for SMEs, because, within the supply chain framework, they present bottlenecks in financing (Gereffi & Fernández-Stark, 2018). Cooperation networks and mechanisms for guaranteeing the operational flow of finance have a major influence on growth and innovation (Alos-Simo et al., 2022; Araya-Castillo, 2022; Moon, 2022), especially in developing countries (Ayyagari et al., 2015). In this setting, it is important to emphasize the significance of commercial credit, which is used by firms of various sizes and for a variety of financial reasons (Abdulsaleh & Worthington, 2013).

SCF has been characterized by two-digit growth rates in many countries, an increase in the number of firms providing SCF solutions worldwide, and a swift adoption of electronic data exchange (Bickers, 2021). As a result of the health and economic crisis caused by the Covid-19 pandemic, more recently exacerbated by the crisis in Russia and Ukraine, SCs are being put to the test (Kilpatrick, 2022), leaving firms in serious financial difficulties. According to Bickers (2021), the short-term impact of the pandemic has been a 6.5% decrease in the volume of SCF operations through Factoring in 2020, which was the second drop this century, being preceded by that of the 2009 financial crisis. One of the major questions in the post Covid-19 era is how the end of financial support measures from governments will affect SCF markets (Bickers, 2021). Currently, the most vulnerable firms are trying to increase trade credit from their suppliers in order to complement other forms of finance, while organizations less affected by this credit crisis are taking on the role of liquidity providers, accepting longer payment terms. These effects contribute to the need for solutions and programs that optimize working capital, of which SCF is one of the most important. Thus, in this scenario of uncertainty, efficient finance management in SCs through profound changes in the near future takes on greater significance (Ishida, 2020; Kumar & Mishra, 2020; Pérez-Elizundia et al., 2020).

It should be borne in mind that SCs are dominated by multinational firms that take up the first levels of the chain. However, in recent years there have been increasingly more SMEs joining SCs, although their negotiating power is still limited (Lampón et al., 2018, 2022; Rodríguez-De la Fuente & Lampón, 2020). In terms of financing, this asymmetrical power structure often results in the imposition of long payment terms and transfer of the financing costs to suppliers located on the SCs’ lower levels (Lind et al., 2012). Furthermore, access to financing continues to be a major obstacle facing SMEs (Jansen et al., 2018). All these factors combined can affect the whole SC, which can lead to distribution problems for the leading buyers by exposing them to manufacturing or delivery interruptions (Brandenburg, 2016). SCF is the means used on many occasions to solve such financing problems, focusing on SMEs (Ali et al., 2020; Wuttke et al., 2013b). In this context, this work shines a light on the role of SMEs in SCF as they make up two thirds of the firms worldwide, and of these, family firms represent around 85% (D’Angelo et al., 2016).

By means of a literature review and bibliometric and content analysis on SCF, this work aims to identify the trends in scientific research on SCF in the area of SMEs, most of which are family run, and to introduce the theoretical approaches traditionally used on SFC, its most closely related topic areas, and future lines of research based on the emerging trends. The following research question (RQ) is therefore proposed:

RQ: What are the trends in scientific research on SCF found from 1970 to 2021 and what implications do they have for SMEs?

In answering this question, we will expose the existing limitation derived from fragmentation in the literature and semantics that prevails in this field of knowledge, and which creates a great deal of conceptual confusion. Therefore, we contribute to the debate on creating a holistic conceptual framework based on a consensus of universally accepted theories and terms. This is all done to improve understanding of SCF at both theoretical and practical levels and allow commercial stakeholders and regulatory authorities for SCs to make the most of opportunities and guide their policies.

Although there are previous bibliometric studies dealing with SCF analysis, they show limitations in search criteria as they do not consider the fragmentation that exists in the literature and semantics in this field of study, which was pointed out by Gelsomino et al. (2016a), Xu et al. (2020), and Parida et al. (2020). At the same time, none of the prior studies performs a bibliometric analysis on SCF centered on SMEs as this one does. The relevance of our work at a theoretical level consists in performing bibliometric analysis through science mapping with more robust criteria than those proposed by the previous literature. Analysis using these criteria made it possible, on the one hand, to identify the driving topics surrounding SCF (Supply Chain Management, Delayed Payments, and Sustainability), and the most relevant articles, journals, and authors in subjects linked to SCF (Reverse Factoring, Trade Credit, Working Capital Management, Risk Management, and SME Financing) in addition to the driving topics. On the other hand, this analysis made it possible to confirm the fragmentation and pave the way towards the construction of a comprehensive conceptual framework for universally accepted SCF for future research. Furthermore, this study takes up the most important questions identified by other authors in SCF and identifies aspects such as innovation, collaboration, and resilience as key factors in the development of SCF with the technological support of FinTechs, the adoption of Blockchain, and Artificial Intelligence to promote the sustainable performance of SCs.

The implications of this work for family firms lies in the advantages derived from their belonging to SCs, as well as the topic trends identified in SCF for decision making. The literature recognizes SMEs—the vast majority of which are family firms—as one of the main drivers of economic and social development worldwide. They play a crucial role in generating jobs and wealth, and their size allows them to adapt quickly to changes in the market and consumer demands (Hernández-Linares & Arias-Abelaira, 2022; Saavedra & Tapia, 2014). However, SMEs face major challenges to access sources of finance, which limits their growth and competitiveness. It is thus relevant to belong to SCs in order to access improvement flows that favor their growth and development. In this sense, the inclusion of SMEs in SCs can facilitate access to finance by having the support of a large corporation that governs the SC and have the benefit of guaranteed long-term contracts (Romero-Luna, 2009). Additionally, SMEs obtain further benefits by being integrated in the SCs, such as access to other markets, greater demand and price stability, improved administration and planning practices, the creation of a credit history, and access to training and technical assistance (Inegi, 2018). Furthermore, this study is of great interest for stakeholders, including banks, SMEs, commercial partners, employers, educational institutions, and public administrations, as it provides them with relevant information on current trends in SCF. This makes it possible to identify commercial opportunities and guide policies. These contributions take on particular relevance in the complex environment SCs are facing in the post Covid-19 era and in the new geopolitical context characterized by the conflict between Russia and Ukraine.

To respond to the research question, the article is organized as follows. Section 2 presents the literature review on theoretical approaches to SCF and the bibliometric analysis. Section 3 outlines the methodology and offers information on the methods used to select the data and analysis techniques. Sections 4 and 5 show the results and the discussion, respectively, based on analysis of science mapping. Finally, section 6 summarizes the conclusions, limitations, and future lines for research.

2. Literature Review

2.1. Theoretical approaches in supply chain finance

Together with the expansion of the SCF market over the last two decades, interest in this field of knowledge has been growing among academics. The number of scientific articles focusing on SCF has increased significantly. In the literature, there are many attempts to develop generalized views of SCF, both from a finance-oriented perspective (More & Basu, 2013) and a Supply Chain perspective (Wuttke et al., 2013a). However, these works are relatively scarce, and tackle their analysis without providing a comprehensive framework for study. Moreover, they still do not bridge the gap between both perspectives, that is, they do not offer either a holistic reference framework (Gelsomino et al., 2016a; Xu et al., 2020), or consensus on the terms used by the scientific community. With this high degree of fragmentation in mind, it is imperative that the volume of quality research available is organized in a unified literary corpus for future research (Gelsomino et al., 2016b; Parida et al., 2021).

Historically, SCF has been studied from three different approaches, ranging from the general to the specific (Liebl et al., 2016). (1) Financial Supply Chain Management, centered on the supplier/buyer relationship, particularly the cashflow that accompanies the physical supply of products (Thangam, 2012; Wuttke et al., 2013b). (2) The optimization of working capital and liquidity of the commercial parties in SCs (Supply Chain Financing), which includes financing before and after product dispatch (Meijer & Bruijn, 2013; Moon, 2022; More & Basu, 2013). (3) Financing for suppliers under Reverse Factoring schemes (Gelsomino et al., 2016a; Gomm, 2010; Huang et al., 2022; Klapper, 2006; Liebl et al., 2016; More & Basu, 2013; Tanrisever et al., 2015; Pérez-Elizundia et al., 2020; Wuttke et al., 2013b), which involve the assignment of collection rights to financial intermediaries based on the credit strength of a large buyer (Supplier Finance).

The topics surrounding SCF consider financial collaboration (Jin et al., 2019; Lampón et al., 2021; Pérez-Elizundia et al., 2023; Wandfluh et al., 2016), particularly trust and transparency (Dello Iacono et al., 2015; Liebl et al., 2016), and the search for a reasonable balance between payment terms and interest rates for the parties involved (Lampón et al., 2021; Shuzhen et al., 2014). Another instrument that is closely linked to SCF is Trade Credit, which has also been used and researched for many years (Cotler, 2015; Gelsomino et al., 2016a y 2016b; Klapper, 2006; Rodríguez-Rodríguez, 2008; Wuttke et al., 2013b). In addition to these dominant instruments (Reverse Factoring and Trade Credit), there are other emerging instruments for SCF, such as inventory finance (Buzacott & Zhang, 2004; Chen & Cai, 2011; Hofmann & Locker, 2009; Modansky & Massimino, 2011; Yan & Sun, 2013). We found no evidence of studies specializing in SCF in SMEs, or central considerations regarding their size, age, or ownership regime, although there is evidence of a growth in general study of family firms as a unit of analysis (Araya-Castillo et al., 2022).

2.2. Finance in supply chains and bibliometric studies

This section covers several bibliometric studies on SCF, from both a theoretical and methodological perspective. Xu et al. (2018) identified four major directions for research in all the studies on SCF: (1) Deterioration of the inventory model in the trade credit policy framework; (2) Inventory decisions with trade credit policy in more complex situations; (3) Interaction between replenishment decisions and deferred payment strategies in SCs; and (4) Functions of the finance service in SCs.

Liu et al. (2015) explored a new vision of the SCF field in China, the leading country in terms of scientific production on SCF. Parida et al. (2021) proposed a research framework based on the finance perspectives of the buyer and of the SCs, where the main contribution was identification of operational coverage strategies that lead towards sustainable SCF in line with the Sustainable Development Goals, as well as some of the emerging research areas in which SCF is still at a fledgling stage of application.

Tseng et al. (2021) proposed a bibliometric analysis from a regional geographical perspective in order to illustrate a general concept of sustainable SCF, which also revealed certain key indicators for continued improvement. Minh (2022) identified that technology development is a research trend for SCF, together with FinTech topic areas, such as data analysis, digital data storage, artificial intelligence, and Blockchain, among others.

Huang et al. (2022) highlighted that, in the context of constantly evolving SCF, existing literature reviews are limited due to the lack of integration of recent findings from 2010 to 2021. Prompted by this limitation, they attempt to fill this gap by researching new achievements that have been reported in the current literature.

In sum, all these works provide interesting ideas and trends in the SCF area. However, all of them warn of the limitations in the search criteria as they do not consider the literary and semantic fragmentation that prevails in this field of study. Table 1 summarizes the main bibliometric studies on SCF, the most relevant aspects, and the types of analysis and technique used.

Table 1. Main bibliometric studies on SCF

Work	Relevant aspects	Type of analysis and bibliometric technique
Huang et al. (2022)	Exhaustive study of all aspects of SCF, which warns of the lack of integration of SCF findings in the last decade and the need to update the SCF framework.	Exhaustive systematic literature review, combined with descriptive and content analysis in four stages: 1. Question formulation. 2. Document location. 3. Choice and assessment of materials. 4. Analysis and synthesis of contents.
Liu et al. (2015)	New view of China—leader in scientific production on SCF.	Analysis of contents in leading articles on China, comprising four steps: 1. Journal selection. 2. Article identification. 3. Category classification. 5. Topic assessment.
Minh (2022)	Study that identifies China as the leading country in SCF research and science and technology as the trend for research in this field.	Systematic literature review, network maps of international affiliation, and keywords, supported by the VOSviewer visualization tool.
Parida et al. (2021)	Sustainable SCF with a finance perspective	1. Exhaustive and systematic literature review. 2. Networks analysis and cluster analysis, citation and co-citation analysis supported by VOSviewer, CiteScape and Tableau & CitNetExplorer visualization tools. 3. Triangulation method to identify research perspectives.
Tseng et al. (2021)	Sustainable SCF with regional perspective.	1. Bibliometric analysis based on big data in the literature. 2. Keyword co-occurrence analysis supported by VOSviewer. 3. Hybrid analysis (quantitative and qualitative) of fuzzy methods for tackling uncertainty of the context, based on: i) Fuzzy Delphi method. ii) Entropy weight method. iii) Fuzzy decision-making trial and evaluation laboratory.
Xu et al. (2018)	Interrelation between inventory models and trade credit policies.	Exhaustive systematic literature review combined with bibliometric analysis, network analysis (citation and co-citation), and content analysis, supported by BibExcel analysis tool.

Source: Elaboration by the authors

3. Material and Method

This paper uses bibliometric analysis to define the theoretical structure of SCF based on performance and science mapping analyses, and co-citation and contents analyses with the support of SciMAT software. Bibliometric analysis is defined as the part of Scientometrics that uses mathematical and statistical methods to analyze the scientific activities in a research field (Callon et al., 1991; Cobo et al., 2011), and allows perspectives to be opened up on the contents and structure of that field (Mazandarani & Royo-Vela, 2022). The SciMAT software (Cobo et al., 2012) is used for its flexibility when selecting measures for obtaining and visualizing bibliometric networks, which is why it has become a popular tool among leading authors in bibliometric analysis (Aparicio et al., 2019; Arteche-Bueno et al., 2019; Cobo et al., 2014; López-Herrera et al., 2012; Mazandarani & Royo-Vela, 2022; Paule-Vianez et al., 2020). A performance analysis was developed using various basic bibliometric indicators (number of documents published, number of citations received, etc.) and the h-index. For a better interpretation of the results and categorization of the issues detected, both Strategic Diagrams and Thematic Networks were used. The process of applying science mapping analysis to define the conceptual structure of SCF consisted of seven steps (see Figure 1).

Figure 1. Workflow of science mapping analysis

Source: Cobo et al. (2012)

First, the various sources and databases for data gathering were identified, which, in our case, was Web of Science (Step 1). Usually, data extracted from bibliographical sources contain errors, which means data pre-processing must be performed, including filtering out duplicated or misspelt keywords (Step 2). A bibliometric network is then constructed using one unit of analysis, which, in our case, was the keyword. In this work, a cooccurrence relationship of terms has been used (Step 3), which is produced when two terms appear simultaneously in at least the title or the abstract of a specific article such that the more articles the terms coincide in, the stronger those terms are considered as being related to each other (Callon et al., 1991; Cobo et al., 2012; Coulter et al., 1998).

To obtain significant information about the area through the analysis, the bibliometric network obtained must be normalized (Step 4), in order to relativize the relationships between two units of analysis (Cobo et al., 2011), attaching more importance to those units with a low frequency and a high frequency of cooccurrence compared to those units with a high frequency and a low frequency of cooccurrence (Cobo et al., 2012). The measure for normalization used in this work is the eij equivalence index (Callon et al., 1991), expressed by the equation eij = cij2 / cicj, where cij represents the number of documents in which two words i and j cooccur, while cicj represents the number of documents in which each word i and j appears. In this way, if two words always appear together, the index will be equal to one, and if they never appear together in any document, the index is equal to zero.

Once the network is normalized, the topics can be extracted by using clustering techniques to divide the set of elements into different subsets (Step 5), whose nodes are closely linked to each other (clusters), and scarcely linked to the rest (Callon et al., 1991; Cobo et al., 2012; Mazandarani & Royo-Vela, 2022). This work uses the clustering algorithm based on Simple Centers, which has the advantage of assigning the word most related to the rest as the name of the topic (Cobo et al., 2012). To assess the position of each item in the area, we present the different clusters in what is called the Strategic Diagram (Callon et al., 1991), which allows us to place topics according to their centrality (X axis) and density (Y axis). Centrality measures the degree of interaction of a topic with the others, as well as its importance in the overall development of the scientific field (Cobo et al., 2012). Density measures the degree of cohesion for a topic, i.e., the internal force of the different links of the nodes within a cluster and can be understood as the topic’s measure of development. (Cobo et al., 2012). In this context, the literature classifies the topics in four categories in the Strategic Diagram (Callon et al., 1991; Cobo et al., 2011; Coulter et al., 1998):

— Driving topics: located in the upper right quadrant and showing strong centrality and density. They are well developed and important to the scientific field.

— Specialized topics: topics that are highly developed and specialized but isolated from the rest are located in the upper left quadrant. These topics show low centrality but high density and are marginally important to the scientific field.

— Emerging or declining topics: in the lower left quadrant and showing low centrality and density. These are little developed and marginal.

— Basic or overarching topics: located in the lower right quadrant, showing strong centrality but low density. These are interconnected with the other topics but scarcely developed.

In a bibliometric network, each unit of analysis can have a group of documents associated with it, which are used for the performance analysis (Cobo et al., 2012). This study has used bibliometric indicators for quality based on the number of documents associated with a topic, citations obtained, and the h-index (Cobo et al., 2012) as a later measure of performance (Step 6). Once the science mapping analysis is finished, the results and the maps must be interpreted using the experience and knowledge of the authors in order to discover and extract useful information that can be applied in decision making (Step 7).

We performed a longitudinal analysis covering 1970 to 2021 (full period), divided into three subperiods: 1970 to 2000 (SP-1), 2001 to 2018 (SP-2), and 2019 to 2021 (SP-3). Although there are some works on aspects related to SCF from the beginning of the 20th century, they are few and far between. It was towards the 1970s when formal research into SCF began to emerge (Huang et al., 2022; Xu et al., 2018), and so 1970 was chosen at the start point for this study. The year 2000 was chosen as the first cut-off point because publications on SCF, which had been practically non-existent in the 20th century, began to become relevant as the new century began. This was when the first formal definition of this field of study emerged(Xu et al., 2018). Likewise, 2018 is the second cut-off point because the first years of this century saw relatively scant scientific production on SCF. It should be noted that, in order to detect the topics of a discipline correctly, the defined subperiods need to be more or less in line with the number of documents (Cobo et al., 2012; López-Herrera et al., 2012). For a longitudinal analysis, division into sub-periods all of the same number of years does not make sense, bearing in mind the limited number of articles published in the last century and during the first decade of this one. Additionally, this criterion is justified by previous studies on science mapping analysis (Arteche-Bueno et al., 2019; Cobo et al., 2011, 2012).

This study uses the Web of Science (WoS) database, property of Clarivate Analytics, which enjoys recognition for works of this type (Arteche-Bueno et al., 2019; Paule-Vianez et al., 2020; Van Raan, 2014). WoS allows access to current and retrospective information in the highest impact research journals since 1900 and has the most complete quality coverage of all the scientific disciplines, including finance. The search only included articles indexed in the SSCI and ESCI databases, thus ensuring the high quality of the literature reviewed. One limitation that we must warn of is that the databases with international recognition prioritize the English language and omit contributions which are more local in nature and that identify particularities of the object of study.

The selection of search chains and keywords bore in mind that the SCF topic area does not have a unified concept and is seen differently by authors depending on their research approach (Gelsomino et al., 2016a; Minh, 2022; Pérez-Elizundia et al., 2020). Moreover, SCF is made up of the terms “Supply Chain” and “Finance” and is interdisciplinary, straddling logistics, SC management, and finance within SCs (Hofmann & Johnson, 2016; Liebl et al., 2016). To ensure that both terms are completely captured by the keywords, we chose two search criteria according to both previous bibliographical reviews on similar subjects (Minh, 2022; Xu et al., 2018), and the professional and research experience of the authors in the field of SCF and Global Value Chains within business management and the economy.

The first search criterion contains terms that characterize the SCF topic area introduced at the beginning of this section, such as Factoring, Inventory Finance, and Trade Credit, in line with the criteria chosen in the bibliometric study by Xu et al. (2018). The second search criterion contains terms related with Finance and which are in turn associated with Supply Chain, with the following search string: (Terms associated with SCF) OR (Terms associated with Finance AND Terms associated with Supply Chain) (Table 2).

The search was performed by topic using the Topic (TS) command in WoS, which includes terms from the research topic area located in the title, in the abstract, and in the keywords. It should be clarified that we opted to use the operator NEAR/x, an operator modality of AND. NEAR/x was used to find records that appear in the same document, but where the terms joined by the operator are within a particular number of words from each other. Thus, the operator NEAR/8 was used (Table 2). The search TS = (financ* NEAR/8 “supply chain*”) (i) finds records that contain financ* and supply chain* that are at most 8 words from each other, e.g., “It is critical for automobile industry supply chains to find seamless financing to operate”, and (ii) excludes records that contain, for example, the term “financial accounting” at the beginning of an article’s abstract and the term “supply chain” at the end. Despite the possibility of some records being excluded from the search, the number of results obtained allowed us to identify the main topic area of SCF dealt with in the literature. In order to guarantee that the NEAR/8 operator captured the records associated with SCF and disregarded those that are not, a visual review was performed on the titles and abstracts of a random sample of 50% of the records obtained. Later, with the help of SciMAT, it was possible to filter the records base even further.

Table 2. Search strings and key terms

Search criteria	Key terms	Search strings
Search Criteria 1	SCF related terms	“supply chain financ” OR “supplier financ” OR “receivables financ” OR “invoice discount” OR “invoice financ” OR “inventory financ” OR “financial supply chain management” OR “reverse factoring” OR “factoring financ” OR “financ* supply chain” OR “financ the supply chain” OR “channel financ” OR “value chain financ” OR (“factoring” AND “financ” AND “supply chain*”) OR (“factoring” AND (“SME” OR “SMEs”))
	OR
Search Criteria 2	Financing related terms	financ* OR “buyer* financ” OR “seller financ” OR “working capital” OR “financ risk” OR “credit risk” OR “commercial credit” OR “trade credit” OR “trade-credit” OR “bank credit” OR “bank-credit” OR “credit” OR “lend” OR “early payment” OR “delay* payment” OR “delay in payment” OR “payment delay” OR “deferred payment” OR “payment* deferral” OR “advanc* payment” OR “capital constrain” OR capital-constrain* OR “financ* constrain” OR financ-constrain* OR “cash conversion cycle*”
	NEAR/8
	Supply Chain related terms	“supply chain” OR “value chain” OR “production network*”

Source: Elaboration by the authors from the advanced search engine of WoS

4. Analysis and Results

This article contributes to the identification of leading, new, and emerging topics that currently characterize SCF based on more robust search criteria than those used previously in the literature. Indeed, Parida et al. (2021) warned that no other study on SCF had used an exhaustive combination of keywords as search criteria in the literature review as they did not consider all the financial and operational aspects. This work seeks to contribute towards correcting those previous shortcomings.

4.1. Historical evolution and performance analysis of the scientific production

The trends in the evolution and development of SCF have been very uneven throughout its history, characterized by an incipient interest in the mid-20th century, followed by accelerated growth in the last two decades. It can be seen that during the 20th century, no significant development was recorded in this area, with the first article being published in 1912 in the Journal of Accountancy followed by very sparse production until the 1980s, when it began to pick up marginally to reach a maximum in 1999 of five articles (Figure 2). The citations shown for articles published in the last century are practically nil. As this field of knowledge began to emerge, it can be observed that the SCF scientific community tends to use few keywords in its publications (1 or 2 words). The dearth of information in this subperiod prevents an analysis of word cooccurrence or scientific mapping being performed.

In the 21st century, scientific production in the field shows a compound annual growth rate (CAGR) of 34% from 2001 to 2021, and accelerated growth from 2015 (39%). 81.2% of all production on SCF is concentrated in the last eight years (2014-2021), while citations grow at an accelerated pace from the turn of the century, showing a maximum of 3,548 in 2018. An upward trend is identified from 2001 to 2018 (blue circles). This analysis reveals growing interest and development in the topic area, showing that SCF represents an important area for research in the future. It should be noted, however, that the more recent the publication date, the lower the number of citations due to the shorter time that articles have been available to the scientific community since first appearing.

Figure 2. Number of SCF studies on Web of Science database

Source: Elaboration by the authors based on WoS output

After obtaining the evolution data described above, the keywords were classified. The prepared file from WoS was imported into SciMAT to carry out the necessary corrections, categorizations, and analysis. Thus, duplicated keywords in articles were omitted in a filtering process to improve the accuracy and quality of the results (Cobo et al., 2012). The result was a total of 4,567 keywords that appeared in at least one document, of which 50 were concentrated in 34% of the total appearance frequency for keywords over the whole 1970-2021 period.

Using the word cloud technique, the ten most used keywords in the SCF field are Trade Credit, Supply Chain Finance, Supply Chain Management, Supply Chain, Management, Impact, Model, Performance, Policy, and Risk (Figure 3). It is worth noting that the first four keywords in this list coincide with those identified by Minh (2022), a leading author in SCF Bibliometry.

Analyzing the authors with the greatest scientific production in line with SCF (Table 3) reveals that the author with the most articles is Xiangfeng Chen, with 16 (left-hand section of the table). The most cited authors are Norris Krueger, Michael Reilly and Alan Carsrud with 2,116 citations from an article called “Competing models of entrepreneurial intentions” on incentives for entrepreneurship (right-hand section). However, it should be noted that although Table 3 shows the authors with the greatest production in topics related to SCF, Minh (2022) and Parida et al. (2021) identify the leading authors in SCF as a main research line, among whom are Hans-Christian Phol, Moritz Gomm, Luca Mattia Gelsomino, Alessandro Perego, Angela Tumino, Nina Yan, Xinhan Xu, David Wuttke, Dileep More, Preetam Basu, Federico Caniato, and John Liebl.

Figure 3. Top keywords in SCF field

Source: Elaboration by the authors from both “Words manager” module provided by SciMAT and www.nubedepalabras.es

Table 3. Performance by authors

Authors with most articles			Authors with most citations
Authors	Documents	Cites	Authors	Documents	Cites
Chen, Xiangfeng	16	784	Krueger, Norris F.	1	2,116
Bi, Gongbing	11	134	Reilly, Michael D.	1	2,116
Hofmann, Erik	10	384	Carsrud, Alan L.	1	2,116
Yan, Nina	10	223	Boons, Frank	1	923
Yan, Honglin	10	107	Ludeke, Florian	1	923

Source: Elaboration by the authors from the “Authors manager” and “Documents manager” modules provided by SciMAT

The five journals with most articles published include 17.5% of all articles (Table 4). They prioritize subjects of industrial organization, administration, and sustainability of SCs as a response to growing concern and increased awareness surrounding environmental protection (Zhao, 2018). That is, topics related directly to SCs, which is why these are the most productive journals on questions of SCF. This centers the subject within the area of logistics and production process administration.

Table 4. Most productive journals in the area

Journal	Launch date	Documents	%
International Journal of Production Economics	1991	89	6.2
Sustainability	2009	59	4.1
Computers & Industrial Engineering	1976	43	3.0
International Journal of Production Research	1961	32	2.2
Journal of Cleaner Production	2003	29	2.0

Source: Elaboration by the authors from the module “Journals manager” of SciMAT

4.2. Science mapping analysis

Science mapping analysis is a technique that aims to follow a scientific area over time in order to understand its structure, development, and its main participants (Noyons et al., 1999). The different types of information that can be used are known as units of analysis. The unit of analysis chosen for this study is “keyword”.

To perform an analysis using SciMAT, both a Strategic Diagram and Thematic Network are constructed. These make it possible to identify the main topics within a scientific field, showing their conceptual and cognitive aspects (Cobo, 2012). This analysis included both the whole 1970-2021 period and the 2001-18 (SP-2) and 2019-21 (SP-3) subperiods, shown in Figures 4, 5, and 6, respectively. The volume of the spheres is proportional to the number of documents associated with each topic in both the Strategic Diagram and the Thematic Network, and, in the latter, the thickness of the link between two spheres is proportional to the equivalence index, i.e., the degree of co-occurrence of two keywords in the same document.

1970-2021 period

Based on the Strategic Diagram (Figure 4a) and quantitative metrics (Table 5) for the entire 1970-2021 period, it is notable that the SCF topic area has a low centrality and density and is subordinated to SCM (driving topic) as a field of knowledge of greater breadth, centered on supply chain logistics and processes administration management. Regarding the metrics for the number of documents, later citations, and impact (h-index), it can be observed that : (1) Supply Chain Management is the main driving topic, standing out in relation to all others in terms of number of documents, citations, and h-index, followed by Delayed Payments, Coordination, and Sustainability; (2) The specialized topics show that SCF has positive results in the three criteria and Data Management scores low in them; (3) The basic and overarching topics, Products and Optimization, were found in fewer documents, obtained fewer citations, and had a lower subsequent impact; (4) The emerging or declining topics Global Value Chains, Determinants, and Systems showed poor performance in all three criteria; (5) The topics with high centrality and number of received citations show strong correlation.

From the thematic network (Figure 4b), a close relationship can be observed that exists between one of the topics emerging in the Strategic Diagram and the following subtopics:

(1) Delay with Trade Credit, Inventory Management, Payments Management, and Economics; (2) Supply Chain Management with Financial Management, Performance, and Risk Management; (3) Coordination with Credit Management, Capital Constraint, and Equilibrium Analysis; (4) Supply Chain Finance with Factoring, Working Capital Management, Banking, and SME; (5) Determinants with Collaboration, Competitive Advantage, and Supplier Financing; and (6) Global Value Chains with Globalization, Governance, and Shareholder Value.

2001-18 (SP-2) and 2019-21 (SP-3) subperiods

An analysis of the Strategic Diagrams for SP-2 (Figures 5a and 6a, and Table 6) reveals that the most relevant driving topics in terms of quality indicators start with Supply Chain Management, which is a long way ahead of Delayed Payments, Corporate Management, and Supply Chain. The driving topics for SP-3 are Trade Credit, Corporate Management, and Management, a long way ahead of Pricing Strategy. It is worth highlighting that the topic of SCF no longer appears in the strategic diagrams of any subperiod but in the thematic networks as a marginal topic associated with Trade Credit (2019-21). Likewise, the topic Corporate Management is also notable not only for prevailing during both subperiods but also because it becomes significantly strengthened in the second, increasing from 71 to 216 documents (Table 6). It can also be seen that, despite the major relevance of the Supply Chain Management topic in SP-2 as a driving topic, it does not appear in SP-3. However, it does represent an important topic associated with the thematic network for Corporate Management.

Figure 4. Strategic diagram and thematic network of the SCF field 1970-2021

a) Strategic diagram