What drives small and medium enterprises to establish and terminate banking relationships?

Abstract

Purpose

This paper investigates the factors crucial for small and medium enterprises (SMEs) in establishing business relationships with banks in Pakistan.

Design/methodology/approach

To investigate how SMEs select banking relationships using criteria, such as decision factors, decision-makers, and decision processes, a comprehensive literature review was used to classify SMEs' decision factors for bank selection. A survey questionnaire was distributed to 200 SMEs, randomly selected from the Small and Medium Enterprise Development Authority database in Pakistan. Probit/Tobit model is estimated to explain the behavior of SMEs.

Findings

The results reveal that SMEs consider a bank's Reputation, Price, and Location essential while establishing bank relationships. SMEs tend to terminate relationships with banks when the Price and Location of the bank are considered essential factors in the relationship with the banks. Price and Location are necessary for SMEs to reduce banking relationships. The SMEs also tend to reduce if they get attractive offers, or the SMEs are recommended to make a banking relationship. This study also provides intuitions for bank policymakers to design policies to retain SME customers and attract new business relationships.

Practical implications

The research emphasizes the importance of competitive and transparent pricing strategies in designing products for SMEs. Banks must prioritize their Reputation and credibility to attract and retain relationships with SMEs.

Originality/value

The study attempts to provide evidence on the SME-Bank relationship focusing on the factors that are crucial for SMEs to decide while establishing business relationships with banks. Also, most of the related literature focuses on developed countries; this research adds to the literature on SMEs' behavior, particularly in a developing country's context.

Citation

Afridi, M.A. and Tahir, M. (2023), "What drives small and medium enterprises to establish and terminate banking relationships?", EconomiA, Vol. 24 No. 2, pp. 230-248. https://doi.org/10.1108/ECON-09-2022-0131

1. Introduction

The firm–bank relationship has remained a subject of immense discussion in the recent past (see Berger & Udell, 2006; Degryse, Kim, & Ongena, 2009; Agostino, Gagliardi, & Trivieri, 2012; Schwert, 2018) because of its nature and consequences for both the banks and firms (Cull, Haber, & Imai, 2011; Koch & MacDonald, 2010). There is, however, no consensus on a single definition of firm–bank relationship; that is, different studies have defined firm–bank relationship differently (see Diamond, 1984; Rajan, 1992; Ongena & Smith, 2000; Boot, 2000; Berger & Udell, 2002). In a nutshell, the relationship between firm–bank is contractual, based on quality, transparency and confidentiality of customers’ reports (Berger, 1999; Dong & Li, 2010). Similarly, banks have remained an essential source of funding and financial services for firms (see Cole, 1998; Cenni, Monferrà, Salotti, Sangiorgi, & Torluccio, 2015). Hence, the relationship between a firm and a bank is based on mutual benefits (Fischer, 1990; Petersen & Rajan, 1994).

What drives small and medium enterprises (we use SMEs and firms interchangeably) to establish and terminate banking relationships, what decision factors, in other words, motivate them to initiate and, later, terminate the relationship with their bank(s), who is responsible for taking such decisions and, lastly, what decision processes are followed by SMEs in their banking relationships are the research queries this study intends to answer in the context of Pakistan. The financial sector of Pakistan has remained bank-based despite various regulatory reforms undertaken by the State Bank of Pakistan (SBP), the country's central bank, to encourage and establish a market-based financial sector.

The SME [1] has played a profound role in developing both developed and developing economies around the globe. SMEs have proven their significance in creating job opportunities and income for low-income families, fostering economic growth and contributing to social stability in a developing country like Pakistan. The government of Pakistan and the SBP have played a vital role in developing the country's SME sector [2]. The contribution of SMEs to the country's GDP is 40%, and more than 80% of the total non-agricultural workforce has been employed by the SME sector (SMEDA) [3].

Our study is an application to the existing literature in three ways. Our study, so far we know, is the first one that directly answers the research queries raised above in a developing country like Pakistan. Secondly, a vast amount of literature on finance discusses the relationship between banks and their borrowing firms over different periods around the globe (see Agarwal & Elston, 2001; Yao & Ouyang, 2007; Berger, Klapper, Peria, & Zaidi, 2008; Huang, Huang, & You, 2015; Belaid, Boussaada, & Belguith, 2017; Nagano, 2017), there are, nevertheless, very few studies that have investigated the relationship between firms and their lending banks (Ongena, Tümer-Alkan & Vermeer, 2011; Saparito, Elam, & Brush, 2013; Canales & Nanda, 2012; Braggion & Ongena, 2013; Tang et al., 2017). We aim to fill this gap by exploring the relationship between SMEs and their banks in a developing country like Pakistan. Stated otherwise, we examine what factors motivate SMEs to establish and, later, terminate a relationship with their banks, who is responsible for making such decisions and what is the decision process, respectively. Thirdly, investigating the relationships is of profound significance, given the SMEs' pivotal role in the national economy in job creation, income flows to low-income families stimulating economic development and accomplishing inclusive economic growth. We examine the relationship between these variables of decision factors, decision-makers and the decision process of SMEs, the probability of ending their relationships and their propensity to retain a number of banking relationships. Therefore, we are confident and convinced that this is the first study investigating firms' relationship with their banks in Pakistan. For this purpose, we test our collected sample by employing Tobit and Probit estimation procedures. The survey-based study involves key decision-makers such as managers and chief financial officers of the respective SMEs.

The study, notably, has instructive suggestions for the banking sector of Pakistan, in general, and the Small and Medium Enterprises Development Authority, in particular. For instance, firms prioritize terminating relationships regarding the “Price” of the product. The study offers policy recommendations and future directions so that the SME sector of the country might be further researched, explored and strengthened. First, given its profound role in the economy, the pricing policies should be scrutinized and revised to preclude price fluctuations and prevent undesired monopolies of banks. In other words, the price impact of lending should be investigated, which will also help to identify credit-rationing in the lending relationships. Secondly, future research should identify loopholes in firm–bank relationships (such as information asymmetry and hold-up problems in which banks tend to exercise monopoly and extract higher rents for a loan contract). Finally, future research should also explore why firms establish new banking relationships (or switch to other banks in the financial sector).

The rest of the paper is structured as follows: The next section explains literature related to the firm–bank relationship; Section 3 briefly discusses the SME sector of Pakistan. Section 4 discusses the data sample, definitions of variables and the methodology employed; Section 5 results are presented; Section 6 of the study discusses results, and the last section concludes the study.

2. Literature review

The effects of firms' and market characteristics on banking relationships have been widely explored in banking literature (see, Ongena & Smith, 2000; Von Rheinbaben & Ruckes, 2004; Degryse et al., 2009; Yu, Sopranzetti, & Lee, 2012; Wang, Han, & Huang, 2020). On the other hand, various studies have explored the concentration of relationships in the banking market (for example, see Avom, Bangake, & Ndoya, 2022; Lu, Wu, & Liu, 2020; Guiso & Minetti, 2004; Bris & Welch, 2005; Guiso & Minetti, 2010; Ongena, Tümer-Alkan, & Westernhagen, 2012). Firms tend to retain concentrated relationships with their banks. They are more likely to default on their loans (Hart & Moore, 1998), which can lead them to take excessive risk (Dewatripont & Maskin, 1995) and, ultimately, to financial distress (Carmignani & Omiccioli, 2007). Therefore, it is appropriate to maintain less concentrated relationships to ensure against liquidity risk and credit risk (Elsas, Heinemann, & Tyrell, 2004).

On the duration of the relationship, the study of Berger and Udell (1995) shows that in the long-run relationship with banks, firms are advantaged to get more credit from their banks, besides paying lower rates on the borrowed funds. Recently, Chodorow-Reich, Darmouni, Luck and Plosser (2022) documented evidence that SMEs obtain shorter maturity credit lines than large firms. Contrarily, on investigating the Norwegian market for relationship duration by conducting duration analysis, the findings of Ongena and Smith (2001) indicate that long-term relations with banks tend to end more likely. The findings are supported by the study of Farinha and Santos (2002), indicating a positive relationship between the length of bank relationships and the probability that firms replace or add relationships with banks in the long run. At the same time, Angori (2020) investigated the role of firm–bank relationships and lending technologies in firms' access to credit and showed that firms' credit availability improves when banking relationships are tighter.

The Reputation of banks is essential for SMEs in various ways, such as their ability to access financing, build trust with stakeholders and leverage digital banking services. As such, SMEs should carefully consider the Reputation of banks when choosing financial institutions to work with, and banks should focus on building and maintaining their Reputation to attract and retain SME customers. In this regard, Jackowicz and Kozłowski (2019) studied the impact of social ties on SME managers' perceptions of their firm–bank relationship. Further, on the Reputation of banks, a study in Bahrain on college students reveals that, among other factors like car parking at banks' sites, the presence of ATMs, and the behavior of bank employees, the students consider the Reputation of the bank as an important factor in choosing of their bank (Almossawi, 2001). Besides other vital determinants, the Reputation of banks has a significant role in relationships with their firms (Kaur & Gupta, 2023).

Investigating the difference between the decision made by a group and an individual, the findings of different studies are mixed (see Mu, Van Riel, & Schouteten, 2022; Bone, Hey, Suckling, 1999). For instance, Barber, Heath, and Odean (2003) indicate that groups, as compared to individuals, make better decisions and that the decisions made by the group are more conservative (Adams & Ferreira, 2010). The findings, however, are opposite to the findings of Kocher and Sutter (2005), indicating that the decisions made by groups are not superior; however, groups, as compared to individuals, tend to learn swiftly. The study of Prather and Middleton (2002) suggests otherwise. For example, the study's findings indicate that team-managed and individually managed funds have no difference in outcome. The study supports the findings of Bone, Hey, and Suckling (1999), which suggest no difference between decisions made by a group or individual.

Although the above studies show that much has been contributed to banking literature, it does not mean that the findings of these studies apply to other nations of the world. The findings of one study significant in one country may be insignificant in another due to differences in the country's cultural, environmental, legal and investment environments. The decision factors and processes significantly impacting banking relationships may prove contrary in other countries. One developing country lacking such study is Pakistan, an emerging country in Asia. The availability of information on decision criteria, decision factors, decision process, and finally, managers and CFOs involved in making decisions for SMEs enables us to conduct a study and extend existing literature to the context of the SME sector of Pakistan.

3. SME sector in a glance – theoretical background

SMEs significantly alleviate poverty and financial inclusion goals. The SME sector of Pakistan contributes hugely to the national economy by enhancing economic growth, lifting the lower end of society from poverty and increasing export earnings. The government, in general, and the SBP, in collaboration with SMEDA (Small and Medium Enterprise Development Authority), in particular, put high emphasis on the development of the SME sector over the years. To actualize the goals and objectives, SBP, in close collaboration with SMEDA over the years, promulgated several reforms for the sector to accomplish the objective of greater financial inclusion and help alleviate poverty in the country. The sector was in its zenith in 2004–2006 when the financing shares as a percent of the total private sector grew to 17% (Rs. 400 billion) in 2006.

The global financial crisis, coupled with multiple domestic poor law and order situations and extreme power outages, dented the sector severely. The financing to the sector touched Rs. 284 billion, and shares as a percentage of the private sector reduced to 8% in 2013. Consequently, small business owners were reluctant to borrow from the banks due to looming uncertainties and high-interest rates, and, on the other hand, banks found SMEs to be less profitable and highly risky, thus reducing their shares of lending to the sector. To tackle the constraints faced by the SME sector, the SBP intervened with several reforms, by defining separately SMEs, revising the SME Prudential Regulations put emphasis on the provision of credit to women, set targets for the banks and approved Financial Institutions Act 2016 from the parliament, respectively.

The number of SMEs in the country, according to a recent report published by the SBP [4, 5] is 3.2 million, constituting around 90% of the country's business. Stated otherwise, the 3.2 million SMEs in the country contribute around 30% to GDP and 25% to the country's export earnings, thus legitimizing the utmost significance and potential of the sector. The financing to the sector, due to the facilitative role of SBP, expanded to around Rs. 401 billion, with SME borrowers reaching the figure of 177,000 [6]. The country's SME sector performance improved as a result of taking services of the UK's Department for International Development and improvement in domestic macroeconomic indicators, such as improvement in law-and-order situation as well as improvement in power supply to the sector.

Before sanctioning any SME loan, banks analyze borrowers using the 5C's tools and techniques. For instance, the character of the borrower, repayment capacity of the borrower, capital (debt/equity ratio) to be invested by the borrower, condition (of the market/industry/seasonality) and collateral (or hypothecated assets/stock) of the borrower, respectively, are set as prerequisites for loan acquisitions by the banks. The off-site inspection and on-site inspection by banks are after the approval of loans; when the business enters the operational phase, respective bank officials pay regular visits to the business site and value their business and collateral with two aims: first, to see whether the acquired loans are spent accordingly and, secondly, to value the collateral and business to safeguard the interests of the banks. In the off-site inspection, banks require businesses to produce annual/monthly/quarterly (depending on the period of the loan) reports duly audited by registered audit firm(s) (registered on the panel of SBP/banks).

On the contrary, how businesses examine their banks and how they choose, in other words, their banks while acquiring loans for their business ventures is a query yet to be explored. What motivates businesses to start and end their banking relationships, and what are the decision factors and processes employed by the businesses in their banking relationships are queries this study aims to answer. Therefore, this study intends to investigate the role of decision factors and decision processes in establishing a firm–bank relationship in a survey study. Businesses prioritize having lesser banking relationships when it comes to the Reputation of the banks. When Price is taken as a decision factor, it is revealed that businesses tend to terminate relationships with their banks and reduce banking services. The decision processes, furthermore, are significant in reducing banking services because banks prefer to switch relationships due to products offered and recommended by other banks.

4. Data

For its successful completion, the study utilizes primary data by surveying SMEs in the Punjab province of Pakistan. The study has chosen Punjab due to the high number of SMEs, the expected higher response rate and easy access to SMEs. The survey has a target list of 250 SMEs, of which 210 responded. Two hundred responses were useable for analysis purposes. The survey involved chief financial officers (CFOs) and managers of SMEs to get reliable information for analysis. In most cases, the CFO/or manager has filled out questionnaires, making the data more reliable and trustworthy for analysis. However, in the absence of the position of CFO in the targeted SME, the survey interviewed any member responsible for the accounts/finance department and decision-making.

4.1 Survey and variables definitions

The study adopts a questionnaire obtained from the study of Ongena, Tümer-Alkan, and Vermeer (2011). However, the adopted questionnaire has been modified accordingly due to variances in the business nature and business environment of Pakistan and the Czech Republic. The questionnaire consists of thirteen relevant questions. The dependent variables in the study are relationships and switching behavior, while the independent variables are decision factors, decision process, decision-maker and control variables. The control variables, such as the number of branches of respective banks, the number of employees in that bank and market shares, are obtained from annual reports of the respective banks. The definitions are given in Table 1.

4.1.1 Dependent variables

Relationships and Switching behaviors are two sets of dependent variables in the study. Relationships include “Largest shares” in concentrated banking services offered to SMEs by a particular bank. In most theoretical literature, the relationship between firms and their lending banks has remained to the extent of credit availability to the firms (Lehmann & Neuberger, 2001) and a relationship based on trust between firms and banks (Foglia, Laviola, & Reedtz, 1998). A strong relationship, in summary, between firms and their banks is vital because it reduces default rates (Fiordelisi, Monferrà, & Sampagnaro, 2014).

Switching behavior captures whether SMEs have stopped or reduced banking relationships. Switching occurs when firms in the financial sector have more options and substitutes available (Bernet & Denk, 2000). Switching to other banks, on the other hand, from existing ones is related to changes in lending rates because switching is common with troubled firms (Boot & Thakor, 1994). The Stop is a dummy variable equal to one if SME has stopped its relationship with the bank and equals zero if SME has not stopped its relationship with the bank. Reduced is a dummy variable equal to one if the firm has reduced banking services and equals zero if not. The “Largest shares” in descriptive statistic (see Table 2) has a mean value of 0.56, suggesting a more robust appetite for concentrated banking relationships by SMEs. The descriptive statistic is given in Table 2.

4.1.2 Decision factors

Decision factors are the factors firms consider essential in relationships with their banks. An index is used, from 1 to 4, with a higher score of 4 indicating firms find relationships with their banks more critical. Questions 8 and 12, respectively, constitute decision factors. The decision factors in our study include “Reputation” (Reputation of the bank), “Price” (Price the banks offer), and “Location” (Location of the bank), respectively. The mean value of “Reputation” (3.18), “Price” (3.22) and “Location” (3.15) suggest SMEs find these decision factors as necessary at the time of choosing their banks. The descriptive statistic is given in Table 2.

4.1.3 Decision-maker

The independent variable “Decision Maker” has been chosen to investigate whether or not “Decision Maker” matters in strengthening and stabilizing banking relationships. The decision maker can be CFO or board (or parent company). If the decision-making does not fall under the authority of the CFO, the decisions are made by the board to choose bank(s). Our study considers the two responses as equivalent. Furthermore, for simplification, we have created a dummy variable for decision-makers, which is equal to one if CFO takes the decision or equals zero if not. The identification of “Decision Maker” or type of “Decision Maker” and the issues raised in various studies (Adams, Almeida, & Ferreira, 2005; Rockenbach, Sadrieh, & Mathauschek, 2007) are closely related. The findings of these studies indicate that groups, compared to individuals, tend to make more conservative decisions. These studies that groups or team management make conservative decisions are justified by the study of Detragiache, Garella and Guiso (2000). The findings of Detragiache et al. (2000) indicate that teams make conservative decisions to diversify a firm's liquidity risk, thereby maintaining multiple banking relationships.

4.1.4 Decision process

The study classifies the “Decision Process” into two groups; the first is information variable, and the second is related to the timing of evaluation of bank relationships. The information variable group includes information used by SMEs at the time of selecting a bank. The variables related to information, interpreted as dummies, are tender (tender info), received offer (Offer info) and recommendation (Recommendation info). The second group of the Decision Process, evaluation timing, is about how often and when SMEs evaluate their banking relationship. For this purpose, questionnaires were presented to SMEs to make appropriate choices.

4.1.5 Control variables

Eight control variables are included in the study, which is then classified into three groups, namely “Relationship,” “Firm,” and “Bank Control.” Relationship control variables are proxies of the scope of bank relationships, that is, what sort of services are used by the firm? We include “Lending,” “Asset Management,” and “Commercial Finance” for the analysis. The second control variable group includes information about ownership size and nature. Finally, the third group involves additional relevant information about the bank, such as the number of branches the bank, number of employees and market shares. We have collected all required data from respective SMEs on their operation sites.

4.1.6 Graphical research design

5. Methodology and results

5.1 Relationships

This section of the study explores the determinants of “Relationships.” The empirical specification is as follows:

(1)Relationships =f (Decision Factors,Decision Process,Decision Maker,Control variables)

The above empirical model enables us to measure bank relationship concentration (or Largest Shares). Since the dependent variable(s) is binary for estimation, this study employs the probit model (Rao and Winter, 1978) to get the likelihood of occurrence and the marginal effects.

5.1.1 Largest share

We employ the Tobit estimation procedure to measure the role of decision factors on SMEs “Largest Shares' in banking services. The findings depicted in Table 3, reveal that among decision factors, Reputation, Price, and Location is significant in determining SMEs ‘Largest Shares’ in banking services. The empirical estimations indicate the coefficient of ‘Price’”in all specifications is significant at one percent and five percent, respectively, indicating that SMEs are very conservative when it comes to “Price” of the product the bank offers to SMEs. Price, in other words, decreases the concentration of banking services banks offer to SMEs. However, this suggests that SMEs diversify the liquidity risk by increasing the number of relationships with banks (Detragiache, Garella, & Guiso, 2000). Also, “Location” is positively significant, and the findings show its importance for SMEs as “location” positively determines the largest shares implying that SMEs prioritize branches nearest to them. SMEs consider Location an essential decision factor while establishing relationships with the nearest banks' branches. In that case, our findings are consistent with the study of Agarwal and Hauswald (2010). The study of Agarwal and Hauswald (2010) shows that borrowers tend to stop relationships with their banks that are farther away from them.

As for decision processes, both “Offer Information” and “Recommendation Information” tends to reduce the concentration of SMEs' relationship with the bank, implying that when SMEs get accessed by the new banks and are offered new products and services or if someone recommends the SMEs the product and services, the SMEs tend to engage in the relationship with new banks and diversify their relationship with banks while getting various products and services.

Furthermore, lending (negatively significant at one and five percent significance level) and commercial finances (negatively significant at one percent significance level) indicate that when using lending and commercial finances, the concentration decreases, implying diversification of liquidity risk by SMEs, hence proving the compromise effects as indicated by Adams and Ferreira (2010). It is understandable now that decision-makers do not play a vital role, although they determine banks. Simultaneously, the other control variables are also noticeable due to their relationship with the dependent variable, the largest shares. We find evidence that SMEs need to concentrate on relationships with their banks. In other words, they prioritize multiple banking relationships, findings consistent with the study of Detragiache et al. (2000).

5.2 Switching behavior

In this part, we explore what determinants, decision factors, makers or processes, induce SMEs to terminate relationships or reduce banking services with their banks. For this purpose, we simplify the survey questions into binary variables. “Stop” and “Reduce” are defined as dummies, with Stop indicating whether or not the SMEs have stopped relationships with their banks, and Reduce suggesting whether the SMEs have reduced banking services from the banks. In other words, “Stop” and “Reduce” as dependent variables suggest if there have been changes in relationships with banks over the recent past years. The relationship Control variable is not included because it includes those services that are being used by SMEs currently that are expected to increase. Separate results are presented in Table 4 (for stopping the relationship) and Table 5 (for reducing banking services).

5.2.1 Stopping a relationship with a bank

We estimate the probit model to explain the behavior of SMEs by using the decision factor. We transform the survey questions into binary variables for simplicity and understanding.

(2)Prob(Stop=1) =f (Decision Factors,Decision Maker,Decision Process,Control Variables)

The results are presented in Table 4. None of the determinants in Stopping relationships with banks matters for decision factors in the analysis. Only the insignificant bank branches negatively impact Stopping relationships, implying that the higher the number of branches of banks, the lesser the SMEs tend to stop relationships with banks.

The positively significant “Price” indicates that SMEs that stress the Price of services offered by banks are more likely to stop their relationship with their banks, showing the significance and importance of Price as a decision factor in stopping a relationship with banks. Simultaneously, we find that although SMEs prefer branches nearest to them, they can also stop their relationship with their banks if the Reputation of the banks is not satisfactory. Regarding the decision process, by emphasizing recommendations and offering information, the relationship with banks is likely to continue.

5.2.2 Reducing services from a bank

The exact mechanism is followed for this part of the analysis. This part of the analysis explores SMEs' behavior in reducing their relationships with banks. We only change the dependent variable from “Stop” to “Reduce” while all explanatory variables remain unchanged. Reduce, a dummy variable which equals one if SMEs have reduced their services from banks and equals zero if not. The results of probit estimation are presented in Table 5.

Prob(Reduce=1) =f (Decision Factors,Decision Maker,Decision Process,Control Variables)

The significant and positive relation of Price with Reduce indicates that SMEs considering the Price of banking services offered are likely to reduce banking services from their banks. On the other hand, SMEs that consider Reputation as an essential decision factor are less likely to reduce banking services.

Location is negatively significant at a ten percent significance level, indicating that SMEs considering Location as an essential decision factor are unlikely to reduce banking services. If we argue that when SMEs consider Location as an essential decision factor and at the same time are less likely to reduce banking services, our findings are in line with the study of Agarwal and Hauswald (2010).

We also find evidence related to the Decision Process, of reducing banking services by SMEs that use received offers and recommendations in their relationships with banks. In other words, an increase in Offer info and Recommendation info inclines SMEs to decrease banking services and switch to other banks. This proves that SMEs tend not to concentrate on their banking relationships and that they prefer to and can switch banking services as needed without primarily impacting the overall banking services they obtain from branches of the banks operating near them. The results indicate that CFO does not influence the dependent variable as a decision-maker. This is not surprising because CFO is only sometimes supposed to make decisions at a higher level in the firm. However, he can also be supposed to make decisions at a lower level of the SME instead, and the CFO is less likely to reduce services from the banks. So far firm control variable concerned with the estimation suggests the probability of reducing banking services from the banks. There is no effect of bank controls on the dependent variable Reduce.

6. Discussion

This study explores the firm–bank relationship; instead of banks choosing their firms, it investigates how SMEs choose their banks. A criterion is developed for the analysis purpose. The “Decision factors” criteria included the “Reputation” of the bank, the “Price” of the banking services and the “Location” of the bank(s). Similarly, the criteria of the “Decision Process” involved information variable group used by SMEs when selecting the bank(s) such as “Tender info,” “Received Offer” (Offer Info) and “Recommendation info.” The decision process also includes evaluation time, showing how often SMEs evaluate their banking relationship. CFO or manager is Decision Maker in the SMEs. Dummies have been created for the variables with a value of one indicating “yes” and a value of zero otherwise.

A total of eight variables are included in the study, which is then classified into three groups, namely “Relationship,” “Firm,” and “Bank Control.” Relationship control variables are proxies of the scope of bank relationships, that is, what sort of services are used by the firm? We include “Lending,” “Asset Management,” and “Commercial Finance” for the analysis. The second control variable group includes information about ownership size and nature. Finally, the third group involves additional relevant information about the bank, such as the number of branches, employees, and market shares. We have collected all required data from respective SMEs on their operation sites.

The results suggest that SMEs that emphasize “Reputation” as an essential decision factor in their relationships with banks have fewer and longer banking relationships. The study by Damberg, Schwaiger, and Ringle (2022) supports these results as SMEs in Pakistan are mainly family-owned and self-financed; therefore, they maintain fewer banking relationships. However, they maintain longer-term relationships with their banks due to developing business rapport and trust over time. Similarly, banks tend to retain their Reputation and trust with the SMEs in terms of offering quick and fast services, displaying leniency at the time of loan negotiation and clearance of SMEs' cheques at the earliest. In contrast, observing SMEs, what in theory is not being practiced. For instance, we have observed that payments were made on behalf of SMEs even before reaching the person of the respective SME to the bank. In other words, banks make payments on behalf of SME(s) to the supplier when the client (SME) calls them and even when the cheques are sent in the late hours.

We also find in the analysis that SMEs considering Price as an essential decision factor tend to stop banking relationships very soon. The outcomes we obtained align with the findings reported in the research conducted by Bakhtiari, Breunig, Magnani, and Zhang (2020) and Rosavina, Rahadi, Kitri, Nuraeni, and Mayangsari (2019). Price of the product or service offered by the banks makes SMEs' nature of retaining relationships volatile and elastic. This is possible because banks in the locality earn from “bundles of services”, offering one product or services cheaply and inexpensive and offering another product or service costly, thereby offsetting the impact which enables bank(s) to earn more from “bundle of services”. In other words, banks offer SMEs a bundle of services instead of one product or service at a time.

SMEs are inclined to retain relationships with nearby banks and are less likely to reduce their services from those banks. This finding is aligned with the results of a study by Cowling, Lee, and Ughetto (2020). Location, Reputation and favorable prices convince SMEs to establish and maintain relationships with their banks. The bank(s) nearest to the SMEs and offering the best and most timely services to the SMEs at competitive prices are defined as the best choice banks by the SMEs.

Due to ownership structure (families primarily own SMEs) and management of SME(s) by the CFO, it is found that stopping the relationship with banks or reducing services from the banks are primarily unlikely and less likely. CFO is optional to make every decision, but he is also likely to be responsible for lower-level decisions. We found evidence at the interview when CFO was responsible for decisions at the lower and middle levels of the SME, while the decisions at the higher level were taken by the owner of the SME(s). In some instances, the owner even used to interfere in middle-level decisions, which was supposed to be not his job to do.

Regarding the decision process, we find that by emphasizing recommendations and offering information, the relationship with banks is more likely to stop except in one (fifth) specification, which shows a negative impact on the dependent variable suggesting something otherwise. Similarly, the negatively significant values of “Offer info” and “Rinfo” indicate that they negatively determine largest shares. In other words, an increase in Offer info and Recommendation info prompts SMEs to decrease banking services and switch to other banks. This proves that SMEs tend not to concentrate on their banking relationships and prefer to switch banking services as and when needed without primarily impacting the overall banking services they obtain from branches of the banks operating near them. Similarly, we find evidence related to the Decision Process of reducing banking services by SMEs that received offers and recommendations in their relationships with banks.

Simultaneously the other control variables are also noticeable due to their relationship with the dependent variable, the largest shares. We find evidence that SMEs tend not to concentrate on relationships with their banks. In other words, they prioritize multiple banking relationships. Multiple banking relationships are beneficial for SMEs because it is the way to diversify their risk related to liquidity. Also, multiple banking relationships are established by SMEs to avoid creating an environment of monopoly for banks. Banks in the locality compete to gain maximum shares at competitive prices and with Reputation. This situation is a win–win for SMEs and banks because SMEs have multiple choices and many banks to establish relationships with or terminate relationships and switch to another one in case of any dissatisfaction, and banks have many SMEs to attract at competitive prices and by offering services to SMEs at their disposal.

7. Conclusion

In this study, we investigate the firm–bank relationship in Pakistan. A survey was conducted in the Punjab province of Pakistan – a hub of SMEs in the country, on detailing how SMEs establish and terminate relationships with their banks, what factors motivate them to initiate and end relationships with banks, who decides about relationships with banks – CFO or manager, and, finally, what is the decision process, respectively. For the successful completion of the survey and to obtain reliable information for analysis purposes, primary data were collected in the form of on-site interviews with CFOs and managers (in the case when a manager was responsible for dealing business affairs of the firm). Furthermore, we investigate determinants of the “Largest Shares” (in banking services) of the bank and “Switching” (the chances to stop the relationship with the bank) behavior of the firm as well as “Reducing” services from the banks.

The estimations and results in our study suggest that the number of bank relationships is lower when SMEs use a bank's Reputation as a decision factor in their relationship with the bank(s) and are not likely to reduce banking services. However, when the SMEs emphasize Price, they are more likely to stop relationships and reduce services with and from the banks. When SMEs prioritize CFO decisions and the Reputation of the banks, then the offers are ignored or, in other words, SMEs have lower needs for finances, so they have fewer relationships with banks. When Price is prioritized or preferred over Reputation of the banks, there is a higher probability that SMEs stop relationships.

In the case of decision-making by the CFO of the SMEs, if CFO is authorized to decide relationships with the banks, the number of relationships with banks is lower or fewer. When the received offer is used as a source of information, SMEs tend to have more banking relationships. Similarly, relationships with banks are higher when SMEs use lending and services relevant to commercial finances.

The study offers practical implications and some recommendations for policymakers. Given the profound role of banks in the economy, competitive and transparent pricing strategies are essential while designing products for SMEs to preclude price fluctuations and prevent undesired monopolies of banks. In other words, the price impact of lending should be investigated, which will also help to identify credit-rationing in the lending relationships. To attract and retain SMEs relationship, the banks must focus on and maintain their Reputation and credibility. The policies must be devised while incorporating the decision factors and process to provide SMEs with enduring banking services in Pakistan.

Table 1

Definition of the variables

	Variable Name	Description	Unit, values and measurement	Source
Relationships	Largest share	Concentration (largest share) of bank relationship	%	Survey Q-2a
Switching behavior	Stop	If the company has stopped a relationship with a bank in the recent past	1= yes, 0= no	Survey Q-5b
	Reduced	If the company has reduced the services from a bank in the recent past	1= yes, 0= no	Survey Q-5a
Decision factors	Important when choosing banks
			4=Most important/3=Rather important/2=Rather unimportant/1=Not important
	Reputation	Reputation of the bank	4, 3, 2 or 1	Survey Q-12
	Price	Price offered	4<<1, average of two responses	Survey Q-8 and Q-12
	Location	Location of the bank	4, 3, 2 or 1	Survey Q-8
	Speed	Speed in processing by the bank	4, 3, 2 or 1	Survey Q-8
	Negotiation	Ease of negotiation with the bank	4, 3, 2 or 1	Survey Q-8
	Consultancy	Consultancy by the bank	4<<1, average of two responses	Survey Q-8
	Corporate culture	Bank’s image suit corporate culture	4, 3, 2 or 1	Survey Q-8
	Services	Quality of bank’s services	4<<1, average of two responses	Survey Q-8 and Q-12
	Network	Network of the bank	4<<1, average of two responses	Survey Q-12
	Market dominancy	Market dominancy of the bank	4, 3, 2 or 1	Survey Q-12
	Relationship	Personal relationship with the bank	4, 3, 2 or 1	Survey Q-12
Decision maker	CFO	Who is responsible in choosing banks?	1= CFO decides 0= Parent company or the board decides	Survey Q-11b
Decision process	Information used when choosing banks
	Tender info	(Public) tendering	1= yes, 0= no	Survey Q-11a
	Offer info	Received/requested offers	1= yes, 0= no	Survey Q-11a
	Recommendation info	Recommendation	1= yes, 0= no	Survey Q-11a
Relationship controls		Bank services the company currently uses
(Scope)	Lending	Lending and structured finance services	1= yes, 0= no	Survey Q-6a
	Asset management	Asset management, custody/depository services and debt market instruments	1= yes, 0= no	Survey Q-6a
	Commercial finance	Export finance, factoring/commercial finance and leasing	1= yes, 0= no	Survey Q-6a
Firm controls	Ownership	Sole proprietorship, partnership etc.		Survey
	Firm size	No. of employees		Survey
Bank controls	Log bank branches	Log of “Number of bank branches” in Pakistan		Website of respective bank
	Log bank employees	Log of “Number of bank employees” in Pakistan		Website of respective bank
	Bank's market share	Advances to SME’s		Website of respective bank

Table 2

Descriptive statistics

Variable	Obs	Mean	Std. Dev	Min	Max
Largest share	200	0.56	0.21471	0.2	1
Stop	200	0.41	0.493068	0	1
Reduced	200	0.65	0.478167	0	1
Decision Factors
Reputation	200	3.18	0.707178	1	4
Price	200	3.22	0.697717	1	4
Location	200	3.15	0.787784	1	4
Speed	200	3.54	0.547631	2	4
Negotiation	200	3.28	0.784011	1	4
Consultancy	200	2.89	0.530188	1.5	4
Corporate culture	200	3.19	0.637126	1	4
Services	200	3.27	0.42473	1.5	4
Network	200	2.88	0.443936	1.5	4
Market Dominancy	200	3.24	0.645802	1	4
Relationship	200	3.2	0.66499	1	4
Decision maker
CFO/Board	200	0.39	0.488974	0	1
Decision process
Offer Info	200	0.79	0.40833	0	1
Recommendation info	200	0.76	0.428155	0	1
Relationship controls
Lending	200	0.78	0.411853	0	1
Asset Management	200	0.97	0.171015	0	1
Commercial Finance	200	0.4	0.491127	0	1
Firm controls
Employees	200	128.12	59.36376	20	240
Ownership	200	1.64	0.481205	1	2
Bank controls
No of bank branches	200	927.94	512.9367	85	1,540
No of employees	200	9027.23	3890.975	1,912	14,054
Bank market share	200	0.0798	0.045345	0.00707	0.151799

Source(s): Authors' own work

Table 3

Largest share

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Variables	Model 1	Model 2	Model 3	Model 4	Marginal effects 4	Model 5	Marginal effects 5
Reputation	0.292**			0.318***	0.281**	0.319***	0.281**
	(0.123)			(0.119)	(0.119)	(0.118)	(0.190)
Price	−0.0787***			−0.0555**	−0.0432**	−0.0555**	−0.0434
	(0.0257)			(0.0252)	(0.021)	(0.0252)	(0.021)
Location	0.0281*			0.0261*	0.0235*	0.0260*	0.0238*
	(0.0168)			(0.0144)	(0.012)	(0.015)	(0.013)
CFOBoard		0.0378		0.0397	0.0348	0.0403	0.0338
		(0.0325)		(0.0313)	(0.027)	(0.0324)	0.028()
Offer Info			−0.143***	−0.130***	−0.114***	−0.130***	−0.1142***
			(0.0387)	(0.0385)	(0.033)	(0.0385)	(0.033)
Recommend info			−0.135***	−0.113***	−0.111***	−0.113***	−0.1111***
			(0.0375)	(0.0374)	(0.032)	(0.0375)	(0.032)
Lending	−0.108**	−0.163***	−0.139***	−0.0935**	−0.074**	−0.0938**	−0.0728**
	(0.0432)	(0.0395)	(0.0378)	(0.0416)	(0.035)	(0.0418)	(0.036)
Asset mgt	0.274***	0.287***	0.294***	0.296***	0.277***	0.296***	0.1775***
	(0.0935)	(0.0954)	(0.0906)	(0.0892)	(0.078)	(0.0893)	(0.078)
C Finance	−0.132***	−0.152***	−0.116***	−0.111***	−0.094***	−0.111***	−0.0936***
	(0.0318)	(0.0323)	(0.0316)	(0.0312)	(0.027)	(0.0313)	(0.027)
Ownership						−0.00200	0.0023
						(0.0316)	(0.014)
Log Bank Branches	−0.0290*	−0.0343*	−0.0281*	−0.0299*	−0.028*	−0.0299*	−0.028*
	(0.0174)	(0.0179)	(0.0169)	(0.0168)	(0.014)	(0.0168)	(0.014)
Constant	0.711***	0.697***	0.847***	0.773***		0.776***
	(0.183)	(0.157)	(0.153)	(0.178)		(0.188)
Observations	200	200	200	200		200

Note(s): The Tobit estimation results with dependent variable “Largest Shares” are presented for five different models in the table. The first row of every explanatory variable presents estimation. The standard errors of the estimation are given in brackets with adjacent stars indicating significance level. The marginal effects for model 4 & 5 “Largest Shares” are given in column 5 and 7. Table 1 defines every variable

*** significant at 1%, ** significant at 5%, * significant at 10%

Source(s): Authors' own work

Table 4

Determinants of stopping relationship with a bank

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Variables	Model 1	Model 2	Model 3	Model 4	Marginal effects 4	Model 5	Marginal effects 5
Reputation	0.0252			0.00435	−0.0038	−0.0387	−0.0178
	(0.131)			(0.134)	(0.051)	(0.139)	(0.048)
Price	0.265**			0.296**	0.110**	0.255*	0.0901*
	(0.134)			(0.141)	(0.053)	(0.150)	(0.050)
Location	0.307***			0.315***	0.165***	0.192**	0.616***
	(0.118)			(0.118)	(0.045)	(0.126)	(0.141)
Offer Info		0.231		0.179	0.0726	−0.136	−0.0419
		(0.232)		(0.239)	(0.091)	(0.260)	(0.087)
Recommend Info		0.183		0.129	0.0524	−0.177	−0.0567
		(0.220)		(0.228)	(0.087)	(0.251)	(0.084)
CFO/Board			−0.216	−0.184	−0.0763	0.107	0.031
			(0.186)	(0.193)	(0.073)	(0.211)	(0.071)
Log Employees						1.084***	0.364***
						(0.205)	(0.053)
Log Bank Branches	−0.0274	−0.0208	−0.00275	−0.0159		0.00731
	(0.103)	(0.103)	(0.103)	(0.104)		(0.109)
Constant	−0.822	−0.413	−0.126	−0.844		−6.158***
	(0.885)	(0.725)	(0.675)	(0.907)		(1.392)
Observations	200	200	200	200		200

Note(s):The probit estimation results with dependent variable “Stop” are presented for five different models in the table. The first row of every explanatory variable presents estimation. Marginal Effects for Stopping Relationship with Banks for model 4 & 5 are given in column 5 & 7. The standard errors of the estimation are given in brackets. Table 1 defines every variable

*** significant at 1%, ** significant at 5%, * significant at 10%

Source(s): Authors' own work

Table 5

Reducing services from a bank

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Variables	Model 1	Model 2	Model 3	Model 4	Marginal effects 4	Model 5	Marginal effects 5
Reputation	−0.101			−0.134	−0.0426	−0.178	−0.052
	(0.138)			(0.145)	(0.047)	(0.149)	(0.044)
Price	0.478***			0.365**	0.012***	0.401***	0.012***
	(0.139)			(0.146)	(0.045)	(0.153)	(0.043)
Location	−0.229*			−0.219*	−0.072***	−0.109	−0.0318
	(0.124)			(0.125)	(0.040)	(0.133)	(0.039)
Offer Info		0.658***		0.544**	0.179**	0.299	0.0885
		(0.234)		(0.245)	(0.078)	(0.261)	(0.077)
Recommend Info		0.715***		0.552**	0.182**	0.307	0.091
		(0.224)		(0.234)	(0.074)	(0.247)	(0.073)
CFO/Board			−0.238	−0.197	−0.0636	0.0746	0.0243
			(0.188)	(0.206)	(0.068)	(0.224)	(0.067)
Log Employees						0.861***	0.26***
						(0.199)	(0.051)
Log Bank Branches	0.0971	0.102	0.124	0.114	0.037	0.134	0.04
	(0.107)	(0.105)	(0.104)	(0.110)	(0.036)	(0.114)	(0.034)
Constant	−0.731	−1.321*	−0.328	−1.160		−5.381***
	(0.925)	(0.747)	(0.681)	(0.958)		(1.405)
Observations	200	200	200	200		200

Note(s): The probit estimation results, with dependent variable “Reduce” are presented for five different models in the table. The first row of every explanatory variable presents estimation. Marginal Effects for Reducing Relationship with Banks for model 4 & 5 are given in column 5 & 7. The standard errors of the estimation are given in brackets. Table 1 defines every variable

*** significant at 1%, ** significant at 5%, * significant at 10%

Source(s): Authors' own work

Notes

1. SMEs are defined as a firm which does not employ more than 250 persons (Manufacturing) and 50 persons in (trade/services) (SBP). SMEDA defined SMEs as firms where employment size is up to 250, paid-up capital and annual sales are up to Rs. 25 million and Rs. 250 million, respectively.

2. Establishment of institutions like SMEDA (in October 1998) and SME Bank (in January 2002). Source: SMEDA and SME Bank

3. Source: Article on the state of SMEs in Pakistan published by SMEDA (Small and Medium Enterprises Development Authority)

4. Policy for Promotion of SMEs published in December 2017. For more details, visit www.sbp.org.pk

5. For more details about SME sector contribution and development, visit https://smeda.org/

6. Infrastructure, Housing, and Finance Department.

Conflict of interest/ethical statement: The authors declare no conflict of interest. The authors would also like to confirm that this paper is an outcome of author’s own work and was not published elsewhere neither is considered for publication elsewhere.