Personal relationships of rural small businesses with community banks in times of crisis

Purpose – Asmany businesses facedeconomicdisruptionduetothe Covid-19pandemicand soughtfinancial relief,existingbankrelationshipsbecamecriticaltogettingaloan.Thisstudyexaminesfactorsassociatedwith thedevelopmentofpersonalrelationshipsofruralsmallbusinesseswithcommunitybankrepresentatives. Design/methodology/approach – We applied a mixed-method approach. We employed descriptive statistics,principalfactoranalysisandlogisticregression fordataanalysis.Wedistributedanonlinesurvey to rural small businesses in five states in the United States. Key informant interviews with community bank representatives supplemented the survey results. Findings – Abusinessowner ’ strustinabankerwaspositivelyassociatedwiththeestablishmentofabusiness – bank relationship. However, an analysis of individual trust ’ s components revealed that the nature of trust is complex, and a failure of one or more components may lead to decreased trustworthiness in a banker. Small businesses that preferred personal communication with a bank were more inclined to relationship banking. Research limitations/implications – Due to the relatively small sample size and cross-sectional data, our results may not be conclusive but should be viewed as preliminary and as suggestions for future research. Bankers should be aware of the importance of trust for small business owners and of the actions that lead to increased trustworthiness. Originality/value – The study extends the existing knowledge on the business – bank relationship by focusingmainlyonsocial(insteadofeconomic)factorsassociatedwiththeestablishmentofthebusiness – bank relationship in times of crisis and high uncertainty.


Introduction
Rural small businesses play an essential role in the prosperity and revitalization of the rural economy and social well-being.Their pre-and post-disaster financial health represents a crucial aspect affecting a business's resilience and recovery.An indicator of financial health that can lead to long-term small business disaster recovery is access to financial resources (Hiramatsu and Marshall, 2018).Banks, particularly community banks, play a key role in providing rural small businesses with the financial resources needed during times of crisis.In fact, Marsh (2015:195) states that "community banks remain important to consumers who prefer the relationship-banking model" and "community banks are particularly important to small businesses, farmers, commercial real estate owners, and individuals."Relationship banking is specific for community banks in rural areas as they tend to be more involved in the local community, including the business community.These close social ties enable rural banks to obtain high quality and quantity of soft information about small businesses.This soft information may effectively help lenders make informed decisions about the riskiness and quality of the borrower and decrease the probability of loan default (Berger et al., 2014;DeYoung et al., 2019;Petach et al., 2021).
The strong interlinking of rural small businesses with their communities became even more critical during the Covid-19 pandemic.Businesses faced an unprecedented economic disruption due to the Covid-19 outbreak (SBA, 2020), and many small businesses sought financial relief through the Coronavirus Aid, Relief and Economic Security (CARES) Act's programs.The relief options included Paycheck Protection Program (PPP) forgivable loans to help keep a business' workforce employed, Economic Injury Disaster Loans (EIDL) to help businesses experiencing revenue loss, the Shuttered Venues Grant, the Restaurant Revitalization Fund and the U.S. Small Business Administration (SBA) debt relief.
In this article, we focus on PPP loans as an example of how a business-bank relationship increases access to financial resources.PPP required a relationship with a banking institution eligible to provide SBA loans.Not only did small businesses have to have a relationship with SBA loan-eligible banks, but they may have found the payroll expenditure difficult to calculate for their financial institutions, which, based on the PPP program fee structure, had little incentive to help them (Harrington, 2020).Harrington (2020) points out that businesses with existing bank relationships were more likely to access PPP funds quickly and efficiently.For example, local community banks were able to expedite small business loans at faster rates than large national banks such as Bank of America or Chase.Various studies connect relationship banking with increased borrowing activities and improved access to capital (e.g.Petersen and Rajan, 1994;Gill and Wilson, 2021).The interrelatedness of rural small businesses and community banks is an important aspect of achieving business and community prosperity.Thus, the objective of this article is to evaluate how trust in a community bank, satisfaction with a community bank's interaction, and preference for personal communication with a bank are associated with the development of the personal relationships of small business owners with community bank representatives.
To achieve our objective, we applied a mixed-method approach.We conducted an online survey in five states in the United States: Indiana, Michigan, Ohio, Kentucky and Illinois.We used principal factor analysis and logit regression analysis to estimate the social aspects of the business-bank relationship and the probability that small business owners would have a personal relationship with a community bank.We split these factors into three groupscharacteristics of a business-bank relationship, business characteristics and business owner characteristics.The qualitative component of our research included face-to-face key informant interviews with community bank representatives that supplemented the survey results.
The study contributes to the existing literature on business and community development, especially in times of crisis.In addition, the study adds to the existing knowledge around the factors that may influence the development of a personal relationship between small business owners and community bank representatives.We conclude that trust in a community bank, satisfaction with a community bank's interaction and preference for personal communication with a bank might assist in a business owner's decision to create a relationship with a community bank.Our results may facilitate efforts of rural small businesses, bankers and JSBED policymakers to develop relationships that provide the flexibility for adapting to a disaster such as the Covid-19 pandemic.
The article is structured as follows.In Section 2, we focus on a literature review and theoretical grounding regarding perceptions about a business-bank relationship.In Section 3, we introduce the study area, data collection and research methods used to identify factors associated with small business-bank relationships.In Section 4, we describe the variables included in the model to test our hypotheses.We present quantitative and qualitative results and discuss how the characteristics of a banking relationship and small businesses may be associated with the creation of a business-bank relationship in Section 5. We also acknowledge the study limitations in this section.Finally, in Section 6, we summarize our findings and highlight the business-related implications of banking relations in times of crisis.

Literature review
In the first part of the literature review, we focus on a brief description and comparison of two types of banking institutions -large and community banks.In the remaining sections, we concentrate on key perceptions of a business-bank relationship together with the characteristics of a business-bank relationship, small business characteristics and business owner demographics.

Large banks versus community banks
The primary purpose of financial organizations is to provide a wide variety of lending, investment and deposit products.Different financial institutions exist to serve a diverse clientele.Some institutions focus on providing accounts and services for the general public, while others serve clients with more specialized needs.
Large banks are defined as banks with at least $10bn in total deposits, while small banks are those with less than $10bn in total deposits (Wiersch et al., 2022).Community banks are defined by the Federal Deposit Insurance Corporation as those of limited size and geographic scope that provide traditional banking services to their local communities.In general, these banks obtain most of their core deposits locally and make most of their loans to local businesses (FDIC, 2012).
Community and large banks offer similar financial products and services, including bank accounts, credit and debit cards and personal and business loans.Still, there are several significant differences between these two financial institutions.Large banks are, in general, well-established banking institutions offering a diverse portfolio of financial products and services to individuals and different-sized businesses and institutions.They typically serve larger geographic areas and possess an extensive network of ATMs and physical branches nationwide and worldwide.Unlike large banks, community banks often offer a limited number of products and services.However, the Small Business Credit Survey conducted by the Federal Reserve Bank revealed that applicant firms were continuously more satisfied with their experience at small banks compared to large banks, finance companies or online lenders between 2018 and 2021 (Wiersch et al., 2022).
Community banks represent a crucial source of funding for small businesses as they offer them various loan products and services.They are more tied to the communities they serve and provide more personalized, relationship-based banking services requiring local knowledge, more frequent personal contact, individual analysis and continued administration (FDIC, 2020).
Community banks are successful providers in areas experiencing a population inflow and related local small business boom.Nevertheless, community banks can also meet the financial needs of less economically vibrant areas, including rural areas with population outflow and locations where employment is concentrated in small establishments dependent on community banks' financing (FDIC, 2020).The limited products and services can be outweighed by the faster local approval process compared to a more complicated loan approval committee in a larger bank.

Defining a business-bank relationship
A personal relationship with a bank is one of the most important relationships a business owner has.A business-bank relationship can be defined as a social relation (Fiske, 1992) where a business owner chooses to share information or expand activities with a bank.In return, a banker offers products and services designed to help the business owner run the business (Carter et al., 2007;Cole et al., 2004).This relationship represents more than repetitive actions and requires mutual respect and knowledge (Barnes, 2003).This statement is reinforced by Tyler and Stanley (1999), revealing that business bank customers often consider relationship banking and the delivery of bank services as the same thing.Banks also proactively build long-term relationships and improve their product portfolio to gain new customers.This service is called relationship marketing and underlines the importance of close social relationships between a lender and a borrower (Haubrich, 1989;Johnson et al., 1996).
Research into a business-bank relationship reveals that a stronger relationship is associated with better credit availability, more favorable collateral requirements, easier access to loans, more favorable interest rates and lower fees (Petersen and Rajan, 1994;Kano et al., 2011;Berger et al., 2014;Gill and Wilson, 2021).Bank managers can provide a business owner with contacts to potential suppliers, customers or funding resources (Dyer and Nobeoka, 2000).Rosenfeld (2014) finds that a strong relationship increases the likelihood of moderately financially distressed firms succeeding in getting a loan.
Developed relationship lending also overcomes information asymmetry and increases familiarity with a client (Perry and Coetzer, 2009).Lenders can use "soft information" about the borrower when enough hard information is unavailable to make informed decisions about a firm's potential riskiness (Berger and Udell, 2002;Brown and Zehnder, 2007).Developing bank connections as a social relationship can also be one of the options to relieve costly selfmonitoring, as many well-established businesses are open to revealing their "soft" information to a banker to minimize problems associated with credit rationing (Gill and Wilson, 2021).
In the following sections, we introduce factors that have been found to be associated with a business-bank relationship.We split these factors into two groupscharacteristics of a business-bank relationship (preferred way to communicate with a bank, trust and bank's interaction with a business owner) and business and business owner's characteristics (loan history, legal structure, family business, business activity, number of employees, years of experience, gender and educational attainment).
2.2.1 Business-bank relationship characteristics.Essentially, the effectiveness of a business-bank relationship depends on the degree of trust between all network participants, including competitors.Trust contributes to eliminating risk within the network.As such, it is a fundamental factor that defines the strength of the links and the long-term viability of the ties between the various parties.An extensive social dialogue between trading partners is important in creating high levels of trust (Ratnasingham, 1998;McGowan et al., 2001;Howcroft et al., 2007).In contrast, limited in-person interactions decrease the level of trust (Bhattacharya et al., 1998).In this context, Hansen (1995) describes a relationship among the members of the entrepreneur's network as a social structure and their interaction as a social process.Reciprocal trust between a banker and a business owner enables a bank to access soft business information (McCabe et al., 2003), reduces information asymmetry and secures the quality of the social bond (Ferrary, 2003), decreases the number of JSBED discouraged borrowers from applying for a loan (Tang et al., 2017) and reduces the chance of switching to an alternative bank (Saparito et al., 2013).Entrepreneurs who believe that a banker trusts them are more likely to behave in a reliable and honest way (Howorth and Moro, 2006).Mayer et al. (1995) refer to three distinct characteristics of a trustee's (a banker in our study) trustworthiness: ability, benevolence and integrity.The ability to accomplish the task consists of aspects such as skills and competencies that are often task and situation-specific.It means that attributes of the ability itself do not make a banker a trusted person.Benevolence means the extent to which a trustee is willing to improve the relationship between a bank and a business owner (a trustor) through positive actions, such as community involvement.Integrity extends competencies and relationship perception of a positive orientation towards the business owner, and it is especially important in the early stages of relationship development.In the small business-bank relationship, it can be perceived by the rather intrinsic nature of having the best interest of a business in mind, sharing common values or listening to a business owner's concerns and problems.
Different scholars studied satisfaction with the business-bank interaction and the bank's knowledge of a business (e.g.Haines et al., 1999;Madill et al., 2002).Saparito et al. (2009) show that trust and bank knowledge about a firm are largely and significantly positively correlated.Community banks especially receive high satisfaction ratings for their performance in meeting financial needs and maintaining strong banking relationships (Scott, 2004).If small business owners are dissatisfied with the advice and support provided by a bank, they tend to reduce their use of bank services or switch to another bank (Perry and Coetzer, 2009).
The Internet and various online platforms offer a new form of business-bank interaction.As a result, the social aspect of a personal relationship is partially or fully replaced with remote relationships and online trust (Gillen et al., 2000).Yet, bank clients can decide to utilize remote or personal interaction based on their individual attitudes toward technology.Customers who feel more comfortable using online technology may choose this option.Bank clients who prefer traditional personal relationships may eliminate or reject online banking (Thornton and White, 2001).Nevertheless, Gilbert and Choi (2003) claim that online banking needs to be considered an inevitable reinforcement of a business-bank relationship.However, the findings of Howcroft et al. (2007) indicate that small business owners are reluctant to accept Internet banking fully.The Internet also impacts relationship banking and the customer's satisfaction with a business-bank relationship because small business owners consider the Internet an essential component of the relationship (Howcroft et al., 2007).Later research shows that small businesses adjusted to Internet banking and enjoyed using this relatively new bank product (e.g. de la Torre et al., 2010).
The literature about relationship banking provides a wide variety of elements important for the development of a personal business-bank relationship.However, as shown in previous studies, interpersonal trust is considered a fundamental element of relationship banking and, in turn, is even more important in times of crisis and uncertainty.As the timing to make small business funding available was critical in 2020, PPP loans were distributed through banking institutions eligible to provide SBA loans.Due to the first-come, first-serve nature of the program, an existing personal relationship with an eligible bank became crucial for a successful PPP loan application for small business owners (Hubbard and Strain, 2020).Therefore, we asked the question: does trust as a complex construct contribute to establishing new or improving an existing business-bank relationship?We dived deeper into the trust components using a model by Mayer et al. (1995) that introduces three factors of perceived trustworthiness (ability, benevolence and integrity).Based on the reviewed literature, we tested trust in a banker as an aspect of establishing a personal business-bank relationship and tested the following hypotheses:

Business-bank relationships and crises
H1. Business owner's trust in a community bank is expected to have a positive association with the development of a personal business-bank relationship.
We tested decomposed trust with two additional hypotheses: H1a. Business owner's satisfaction (as a proxy for ability) with banker's interaction, knowledge of their business and local market/community are expected to have a positive association with the development of a personal business-bank relationship.
H1b. Business owner's perceived aggregated benevolence and integrity in a banker are expected to be positively related to developing a personal business-bank relationship.
Another crucial aspect of relationship banking is how a client interacts with a bank.Because a business owner can decide whether to utilize remote or personal interaction based on their preferences, it poses the question: What kind of interaction with a bank may evoke a business-bank relationship?Thus, we tested the following hypothesis: H2. In-person communication with a bank is expected to have a positive association with the development of a personal business-bank relationship.

Business and business owner characteristics.
Loan history represents another aspect of a business-bank relationship and the closeness of a borrower to a lender.Peterson and Rajan (1994) indicate that the smallest businesses have 95% of their loans from a single lender, compared to 76% of the largest businesses.Businesses tend to concentrate their borrowing from one source, but the concentration decreases as the size of a business increases.Many factors attributed to business heterogeneity impact the loan acceptance rate -financial history and performance, stage of business development, age of firm (Haines et al., 1999), friendly environment and loan manager behavior (Tang et al., 2017) and business and bank location (DeYoung et al., 2019).Thus, exploring whether a small business owner has a bank loan is an essential indicator of a potential strong business-bank relationship.
Research findings on gender differences in loan and credit accessibility and use vary based on data used, location, country and other aspects.As the number of female-owned businesses grows, research related to the characteristics of female entrepreneurs is of central interest.Several research studies show that female-owned businesses do not experience differences in bank lending practices, credit and loan acceptance rates and credit terms compared to male-owned businesses (Orser et al., 2006;McKechnie et al., 1998).Haines et al. (1999) do not support a gender bias as they find that male and female entrepreneurs do not differ substantially in terms of bank credit (such as interest rates or amount of collateral), financial characteristics and nonfinancial attributes.
However, other studies show evidence that female-owned businesses must pledge collateral more often (Calcagnini et al., 2015), are charged higher interest rates and have to provide greater collateral than male-owned businesses (Coleman and Robb, 2009).Bellucci et al. (2010) find the likelihood that female entrepreneurs need to provide collateral is 5.2% higher than for male borrowers, although they do not pay higher interest rates.As a result, women may become discouraged borrowers not entering the credit market.They tend to be less confident than men due to the perception their applications would be declined (Kon and Storey, 2003;Tang et al., 2017).Carter et al. (2007) suggest that bank managers act in a discriminatory manner while assessing female business owners' loan applications.
Women more often launch businesses in slower-growing service and retail industries.At the same time, men prevail in manufacturing, construction and faster-growing hightechnology sectors (Miskin and Rose, 1990).Retail and services are often considered relatively risky, which can be negatively reflected in bank loan accessibility and loan conditions (Haines JSBED et al., 1999).Female-owned businesses also tend to have less structured legal forms with a lower share of partnerships or limited liability companies that might prevent them from accessing bank loans (Belluci et al., 2010).Female-owned businesses are, on average, smaller and younger than male-owned businesses, which results in a shorter business-bank relationship (Haines et al., 1999;Calcagnini et al., 2015).
The age of a business owner or manager is often used as an alternative for business experience (Bates, 1990).Moreover, according to several studies (Dries and Swinnen, 2010;Jonson, 2002;McMillan and Woodruff, 1999), business experience positively affects access to credit.Gill and Wilson (2021) show that bank connections are positively correlated with owner education.
In general, family businesses often prefer to dispose of their own financial resources.Many of them also have family bank connections that contribute to their improved business performance (Gill and Wilson, 2021).Moreover, family-owned businesses might have a stronger relationship with their main bank (Berger et al., 2014).
The size of a borrowing firm may affect a business-bank relationship from different perspectives.Small businesses are less likely to deal with multiple banks (Perry and Coetzer, 2009).Small business lending is often based on informational asymmetries between lender and borrower that may impact the types of lenders and lending technologies used by a lender (Berger et al., 2014;DeYoung et al., 2019;Petersen and Rajan, 1994).Information asymmetries can be amplified as small businesses are less likely to be monitored by rating agencies (Petersen and Rajan, 1994), and their financial statements are unlikely to be compiled or audited by a professional accounting firm (Allee and Yohn, 2009).Larger-sized businesses have stronger negotiating power and are usually considered less risky than smaller businesses (Berger and Udell, 2002).
A business-bank relationship may differ according to the business and bank location.DeYoung et al. (2019, p. 100) suggest that "the advantages associated with relationship-based lending are likely to be more pronounced in rural places, where personal relationships are an integral part of the social fabric."Rural loans also have about an 11% lower probability of default (DeYoung et al., 2019).Larger distances, typical for rural areas, separate firms from lending branches and are considered a negative factor for credit access (Bellucci et al., 2010).From the bank's perspective, it is more challenging to assess borrowers located far from a bank (Hauswald and Marquez, 2006).At this moment, the role of local bank branches is even more important as they can collect soft information about their borrower and pass it to the bank's headquarters.This scenario saves transportation costs but also makes it difficult to maintain this soft information while passing it to its final place for consideration (Stein, 2002).
Existing personal relationships with community banks may enable easier access to capital.As relationship lending is an essential component of financial stability, particularly among rural, small and new businesses, the presence of community banks is a fundamental aspect of regional economic development.Petach et al. (2021) point out differences in the geographic dispersion of community banks and emphasize the high levels of community banking activities in rural Midwestern counties.Rupasingha et al. (2019) found that small businesses located in counties with more bank services are less likely to fail and that businesses in rural counties have a higher chance of survival than businesses in metro counties.

Materials and methods
To test our hypotheses, we applied a sequential mixed-method approach using quantitative and qualitative methods.We began with a quantitative survey focused on a sample of small business owners to collect data for our analysis.We continued with a qualitative method involving face-to-face key informant interviews that complemented the survey results (Creswell and Creswell, 2018).

Small business online survey
In the quantitative research component, we investigated the most critical factors associated with the personal relationships of business owners with community bank representatives.We targeted the survey towards small business owners in rural areas in five Midwestern states: Indiana, Michigan, Ohio, Kentucky and Illinois.Following the USDA-Economic Research Service, we define rural areas as nonmetro areas that include open countryside, rural towns (places with fewer than 2,500 inhabitants) and nonmetropolitan urban areas (urban areas with populations between 2,500 and 49,999 that are not part of metropolitan and larger labor market areas) (ERS, 2022).
We aimed to gather at least 200 valid responses to ensure our sample size was big enough for the logistic regression analysis.Although SBA defines small businesses as firms with fewer than 500 employees, according to U.S. Census Bureau data, only 18% of businesses had more than 20 employees, and 80% had no paid employees (SUSB, 2018).Therefore, in this study, we consider small businesses as businesses with up to 20 employees.
We designed an online survey using Qualtrics comprised of four sections, including questions regarding a business-bank relationship, loan experience, business demographics and personal characteristics.We describe a personal business-bank relationship as a relationship involving repeated contact between a business banker and the business owner.A business owner chooses to share information or expand activities with a bank.In turn, a banker offers products and services designed to help a business owner run the business.We applied two different strategies of respondent identification and survey distribution to guarantee the required sample size and to avoid the potential selection bias.We used random stratified sampling and panel data collection conducted by the Qualtrics XM company.
The first strategy was to conduct a survey only in Indiana.To avert the potential selection bias, we applied random stratified sampling to extract 3,000 emails using the D&B Hoovers database [1].We identified the number of emails in each of the ten Indiana Small Business Development Center's (ISBDC) regions based on a share of the state population in each region and estimated the number of companies per ISBDC region.As we primarily focused on rural businesses, we avoided large metropolitan areas like Indianapolis.Then, we assigned 25% of 3,000 businesses to manufacturing industries (NAICS 31-33) as manufacturing, in general, accounts for a quarter of jobs and state GDP in Indiana.Subsequently, we divided the remainder of the establishments into two groups, NAICS 21-48 (Mining, Construction, Trade, Transportation) and NAICS 51-81 (Services).We limited the employee size to a maximum of 20 employees.The majority of emails involved contacts of the business executives, owners or managers, and the email type was campaign verified.Wherever feasible, we also downloaded direct mailing addresses.We did not target farmers (as their activities and lenders are often too different from non-farm small businesses) and governmental organizations.Being aware of the potential non-response bias, we distributed the survey three times in 7-day intervals and replaced bounced-back emails, representing about 7% of distributed emails, using the same procedure described above.Despite our best efforts, the total number of valid responses was only 97.
As we did not gather enough valid responses, we decided to implement the second strategy and contracted Qualtrics XM to collect the remaining data for us [2].The company targeted rural businesses in Indiana and surrounding border states (Michigan, Ohio, Kentucky and Illinois) to collect 104 valid responses.We identified the sample as owners of small businesses with up to 20 employees located in rural areas, not farmers.Qualtrics XM distributed the online survey and collected responses in the spring of 2022 [3].In compliance with the Institutional Review Board's protocol, the data collected were strictly anonymous to ensure the confidentiality of responses.The final sample consists of 201 valid responses, with 64% of small businesses located in open countryside and rural towns and 36% in nonmetropolitan urban areas.JSBED

Interviews with community bank representatives
Qualitative topical interviews enabled us to explore how bank representatives perceived the personal business-bank relationship.We used the face-to-face key informant interviews with community bank representatives using open-ended questions to learn more about the motivation and implementation of a business-bank relationship in practice and supplement the survey results.We designed our interviews to achieve the credibility of our qualitative research through its transparency, consistency and communicability (Rubin and Rubin, 1995).In addition, the interviews were intended to increase the reliability of the previous analysis by including an open-ended question about maintaining existing and generating new relationships with small business owners.
Study collaborators from rural Hoosier Heartland ISBDC helped us identify community banks and create a list of potential key informants.We targeted the invitation list because we were addressing community bank representatives who could provide us with their observations and opinions on business-bank relationships from the perspective of community banks.We distributed interview invitations to community bank representativesprimarily bank directorsacross the Hoosier Heartland Region using email addresses from a generated list of potential key informants.Three directors of different community banks gave their consent to be included in the study.We did not connect their names with anything they said to protect the bankers' privacy.This precaution also prevents the possibility of linking the banker with their clientssmall business owners.
An experienced team member conducted the interviews in the interviewee's preferred location.All three interviewees were asked the same open-ended questions to guarantee a relevant level of consistency.Each interview lasted between 70 and 90 min, depending on the informant's volubility and information content.The interviewer followed the interview protocol and asked additional questions only to garner the necessary information.
The interviewer started the interview with an introduction and reading ground rules regarding confidentiality and voluntariness.After that, the interviewer asked open-ended questions focused on: (1) The interviewee's professional background and experience with a market.Can you tell me a bit about your work in general and your work and experience concerning a small business-bank relationship?What is your bank's geographic market for small business loans?
(2) Small business loans.What loan products does your bank offer to small businesses, and where can a small business apply for a loan?
(3) Maintaining existing and generating new relationships with small business owners.How does your bank maintain existing relationships with small businesses?How does your bank generate new relationships with small businesses?What are the needs and expectations of small business owners/managers and bank managers actively involved in the relationship?What is the preferred way to communicate with your established clients?
To ensure transparency and transcript availability, the interviewer recorded all interviews.The third-party vendor, rev.com, generated the transcripts of the interviews using artificial intelligence (a machine-generated transcript).We treated transcripts as memory aids bolstering the interviewer's notes to ensure an accurate representation of what was said, not by whom.We used the precise quotations of interviewees' firsthand experiences to give legitimacy to our arguments resulting from the quantitative analysis.

Logistic regression model
Logistic or probit regression analysis is often applied in business-bank relation studies to identify factors affecting a decision-making process related to a business-bank relationship Business-bank relationships and crises (Peterson and Rajan, 1994;Berger et al., 2014;Kano et al., 2011;Gill and Wilson, 2021).We employed binary logistic regression analysis to estimate the influence of the following variables on the existence of a business-bank relationship: trust in a community bank, satisfaction with a community bank's interaction and a preference for personal communication with a bank.Table 1 provides a summary of variable definitions.

Dependent variable Bankrelation
As a business owner, do you have a personal relationship with your banking institution for your business?
1 -A company has a personal relationship with a community bank 0 -Otherwise

Independent variables
Business-bank relationship ability1 Satisfaction with the bank's interaction (the ability to talk with the business owner or manager)

-Satisfied 0 -Otherwise ability2
The bank's knowledge of your business 1 -Satisfied 0 -Otherwise ability3 The bank's knowledge of the local market/community 1 -Satisfied 0 -Otherwise ability4 The bank's anticipation of a small business owner's financial needs other than credit/loan 1 -Satisfied 0 -Otherwise ben_int1 We trust that our bank has our best interests in mind when they give us recommendations 1 -Agree 0 -Otherwise ben_int2 We can freely share concerns and problems about our company and know that they will respond constructively The dependent variable bankrelation indicates the respondent's answer to the question, "As a business owner, do you have a personal relationship with a community bank for your business?"Fifty-four percent of respondents reported they have a personal relationship with a community bank.Forty-six percent of respondents indicated otherwise.The response "Otherwise" includes a personal relationship with a large/national bank or another type of banking institution (especially credit unions) and businesses that do not have a personal relationship with a bank.4.2 Independent variables 4.2.1 Trust and trust components.We tested hypotheses H1, H1a and H1bthe effect of trust in a community bank and its factors -by applying the three components of trust (ability, benevolence and integrity) proposed by Mayer et al. (1995).We identified seven aspects that determine diverse aspects of trustworthiness.We measured ability through questions about satisfaction with a community bank's interaction (Scarpi and Visentin, 2015;Saparito et al., 2013) using a 5-point Likert scale (where 1 5 not at all satisfied, 2 5 slightly satisfied, 3 5 moderately satisfied, 4 5 very satisfied, 5 5 extremely satisfied).Respondents rated the following satisfaction-related questions: (1) the bank's interaction with a business owner or manager; (2) the bank's knowledge of the client's business; (3) the bank's knowledge of the local market/community and (4) the bank's anticipation of a business owner's financial needs other than credit/loan.The survey did not include enough questions related to benevolence and integrity to analyze them separately, so we aggregated relevant questions into a 3-item scale variable ben_int.We measured benevolence/integrity using a 5-item index measure (Saparito et al., 2013;Howcroft et al., 2007) that we adjusted to the purpose of our study.Small business owners used a 5-point Likert scale (where 1 5 strongly disagree, 2 5 somewhat disagree, 3 5 neither agree nor disagree, 4 5 somewhat agree, 5 5 strongly agree) and rated the following statements: (1) we trust that our bank has our best interests in mind when they give us recommendations; (2) we can freely share concerns and problems about our company and know that they will respond constructively and (3) we can share common business values with the bank.The pairwise correlation coefficients were between 0.295 and 0.688, showing that ability and benevolence/integrity variables were significantly positively correlated (Table 2).
We employed the methodology used in previous lending relationship studies (Moro and Fink, 2013;Howorth and Moro, 2012), and tested how ability (a proxy for satisfaction) and benevolence/integrity-related variables might impact the willingness to develop a banking relationship.In contrast to their studies that operationalize trustworthiness and analyze ability, benevolence and integrity from the banker's perspective, our study looks at trustworthiness and its elements from the perspective of a small business owner.We used a factor extraction methodprincipal factor analysis (PFA) to reduce seven variables into smaller components and determine the number of factors to retain.We located those factors with eigenvalues greater than the average of the initial communalities (squared multiple correlations).The analysis showed that the average of these communalities was 0.518.The first two factors were associated with eigenvalues greater than 0.518 (3.337 and 0.794, respectively).We confirmed these results by a scree test.Although this unrotated solution provided us with relatively easily interpretable results, we used varimax rotation with the Horst normalization procedure to further polarize the loadings.Since all the loadings were above the threshold of 0.4, we set it along with the two-factor solution.As we planned to use these two factors in subsequent analysis, we tested the reliability of the summated scale (1)(2)(3)(4)(5) based on Cronbach's alpha coefficients (0.81 and 0.87, respectively) which showed the relatively high internal consistency of the items.

JSBED
The rotated factors showed that the first four items (ability) loaded to the first factor.The last three items, related to benevolence and integrity (ben_int), loaded to the second factor.Table 2 shows the two factors resulting from PFA.
4.2.2Bank communication.For our last hypothesis (H2), we tested the effect of preference for personal communication with a bank on the existence of a business-bank personal relationship including the variable bankcomm.This variable estimated a respondent's preferred way of communicating with a bank.The value of 1 indicated "personal" communication, and 0 was used for "hybrid or e-banking".Nearly 40% of respondents preferred personal communication with their bank.
4.2.3Control variables.The vector for business characteristics involved six variables.The variable bankloan examines the borrowing pattern of business owners.It represents a concentration of borrowing when on average, small businesses tend to have only one lender compared to large firms with about three lenders (Peterson and Rajan, 1994).It is a dummy variable taking the value of 1 if a business owner had a bank loan or line of credit and 0 if otherwise.Sixty percent of small businesses in our sample had a bank loan or line of credit.
The business legal form (legalform) is a dummy variable that takes the value of 1 if a business owner was a sole owner or sole proprietor and 0 if otherwise (partnership, limited liability company -LLC, corporation or subchapter corporation).Familyowned is a dummy variable that equals 1 if the owner identified their business as a family business and 0 if otherwise.Femaleowned is a dummy variable with the value of 1 if the business owner was a woman and 0 if otherwise.Women own 34% of businesses, and 62% are family-owned businesses.We included business size using the number of employees (employees) as a continuous variable.
Variable busactivity is a dummy variable describing the primary business activity.The variable consists of two aggregated categories: 1 5 Industry sectors (natural resources; mining, oil and gas extraction, construction, manufacturing, wholesale and retail trade), 0 5 Services (professional services, education, health care, entertainment and food services).Businesses in our sample are almost equally distributed between industry sectors and services.Small businesses are often concentrated in sectors that require fewer capital assets.Forty-three percent of the businesses in our sample fall within the services category, mainly professional and food services.
The variable for education is a categorical variable that equals 1 if a business owner's educational attainment was grade 12 or GED and less, 2 5 college 1 year to 3 years and 3 5 4year college graduate and higher.The length of a relationship between a small business owner and a community bank representative is an important dimension of a relationship.The business-bank relationship may be correlated with years of experience in the current business.Thus, we included years of business experience (experience) as a continuous variable.Summary statistics for the variables are shown in Table 3.

Results and discussion
We used logistic regression analysis to determine the effect of trust, satisfaction and communication preference on having a personal relationship with a community bank.To test our hypotheses, we ran two regression analyses.The first model (Model 1) includes a variable trust that consists of all three elements of trustworthiness (ability, benevolence and integrity).In the second model (Model 2), we decomposed trust and tested the significance of its two factors: ability (ability) and aggregated benevolence/integrity (ben_int).Both models contain a variable for preferred bank communication (bankcomm).
After culling for missing values, the final sample was reduced from 201 to 135 observations for both models.Using a link test, we tested the model specification as the logit regression (Pregibon, 1980;Turkey, 1949).The variable hatsq in both models was (1) trust  3. Summary statistics of variables JSBED insignificant (P>jzj equals 0.163 for Model 1 and 0.59 for Model 2).We did not identify any multicollinearity in the final sample.The lowest tolerance value of the variance inflation factor (VIF) (Chatterjee and Hadi, 2006) was 0.49 and the highest was 0.96 for Model 1, and 0.49 and 0.86 for Model 2. Finally, we employed Pearson's goodness-of-fit test to find other significant explanatory variables predicting RELATIONSHIP after running the models.The Pearson statistics were 0.63 in Model 1 and 0.47 for Model 2, which were not significant and indicated a good model fit.

Results
Table 4 provides the regression analysis results for the determinants of small business owners having a relationship with community banks.The analysis reveals that trust in a banker is statistically significant and positively associated with a business owner's willingness to develop a business-bank relationship and supports hypothesis H1.Ability, as a component of trust and a proxy for satisfaction, is also statistically significant and positively associated with relationship banking.This implies that business owners who are satisfied with the ability of a banker to talk with them to learn about their business, local market or community and anticipate their financial needs may be more likely to create a personal relationship with a bank and, thus, supports hypothesis H1a.Interestingly, aggregated benevolence/integrity (ben_int) as another trust component is not statistically significant.This suggests that the business-bank relationship may not be associated by a business owner's belief that a bank has their best interest in mind, that they can share their business concerns and that the bank has shared values.Thus, the analysis does not support hypothesis H1b.

Business-bank relationships and crises
The preferred way to communicate with a bank seemed to be a crucial aspect of having a business-bank relationship.Preferring personal communication (bankcomm) is positive and statistically significant.Therefore, business owners who prefer personal communication may be more likely to have a relationship with a community bank, which supports hypothesis H2.This implies that businesses favoring personal contact with a bank may be more willing to create a business-bank relationship than those choosing a fully remote connection via e-banking or a hybrid option.A hybrid option means that businesses usually use e-banking for transactions or routine financial needs and leave the personal interface for more complicated financial matters.Our results also suggest that business legal form, business owner's gender, business activity, number of employees, business experience and education are not statistically significant factors in the development of a personal relationship with a community bank.5.1.1Robustness check.Following Moro and Fink (2013), we tested the interdependency of trust and its components -ability and ben_intand other business-bank relationship-related variables.Zand (1972) argues that trust development is a nonlinear and rather circular process, making it difficult to untwist causes and effects.However, Mayer et al. (1995) show that trust is dependent on a number of internal and external factors, such as the characteristics of the business and business owner, but at the same time, trust is expected to be separated and different from the business owner and business characteristics.Thus, we used regression analysis with trust, ability and ben_int as dependent variables and regressed them against business and business owner characteristics.All three regressions are not significant.The outcomes indicate the results' robustness and differentiate trust, ability and ben_int from other independent variables.It means that trust and its components serve as crucial aspects of the personal business-bank relationship and extend the list of variables defining the banking relationship.

Discussion
Our study shows that trust, defined by all three factors of perceived trustworthiness (ability, benevolence and integrity), may be associated with the business's decision to establish a business-bank relationship.There is no generally accepted definition of trust, but the concept proposed by Mayer et al. (1995) is widely recognized.It defines trust as "the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party" (p.712).In line with this definition, our results reveal that the propensity to trust a banker may lead to a higher willingness to set up a relationship.Multifaceted trust includes various elements, for example, a banker's best interest in mind, sharing values and problems and a banker's interaction and knowledge of a business and community.This points to the complexity of trustworthiness.Our findings are consistent with Frankel (2008), who argues that trust is related to a rational belief that a banker tells the truth and keeps their promises.Uzzi (1999) claims that relational trust is associated with sharing values and objectives.This may be even more important in rural communities where business owners are more embedded in their communities from a business and family perspective (Amato et al., 2021).In fact, Guiso et al. (2004) find that trust-based financial operations are more common in small communities with more social capital.Our findings concur with Guiso et al. (2004) that business loans are financial transactions requiring trust, particularly between the rural small businesses we surveyed and community banks.
However, based on our findings, business owners might be aware of their potential vulnerability while sharing concerns and problems about their company with their bankers.This is consistent with a culture of secrecy, when the business can have an active role in the JSBED decision-making process of how much confidential information, they are willing to provide to a lender (Von Rheinbaben and Ruckes, 2004).This secrecy is in contrast with how much information the banker would like businesses to share with them.Banker 1 contributed to this point from the bank's perspective, I would say that if somebody's got an established relationship, it just makes sense that we've got more history and more of a comfort with how they operate and what their profitability is.We expect regular updates on what their financial position is, there again always looking for them sharing all the details of what the conditions are.
A business's decision on how much soft and hard information to share with a banker is a crucial aspect of a business-bank relationship as it has a strong implication on information asymmetry.Most survey respondents reported they feel they can freely share their business values, problems and concerns with a banker.However, feeling pressured to actually share their private information might impact an existing or future business-bank relationship.Pasiouras et al. (2021) find that a national/regional culture of secrecy affects the number of bank relationships.The culture of secrecy is out of our study's scope but would deserve future research.Our findings suggest that bankers may need to continuously invite business owners to share information about their business without business owners feeling it puts their business at risk if things are not positive.Business owners should feel that their banker is a partner and not just a supplier of credit.
Although different studies dispute the link between satisfaction and customer loyalty (Williams et al., 2011), we found that the satisfaction of business owners with the banks' interactions may positively affect a business-bank relationship.The ability of a banker to talk with their clients influences their clients' satisfaction and the extent of actual or potential relationships (Perry and Coetzer, 2009).On the other hand, community banks also strongly emphasize communication with their clients.As Banker 2 suggested, You [business owner/manager] need a banker that's going to listen to you and then from listening to you, start asking the questions to get you to understand what it is that you really need.

And added,
Because if I can get an understanding of whatever it is that you're involved with, if I can get an understanding for that, now I can help you.But if I'm not real clear on that, I may give you a solution that's totally off base, and that's not what you need.That's why I'm a firm believer in going to the customer as opposed to the customer coming here.
The banker statements coincide with the results of our model that businesses that prefer personal communication over remote communication are more likely to develop a personal business-bank relationship with a community bank.Our findings concur with Howcroft et al. (2007), who argue that small business owners prefer personal contact with their banks, especially in times of crisis.Bank managers play a crucial role in a social network by providing information about alternative financial resources and potential customers (Dyer and Nobeoka, 2000).To this point, Banker 2 claimed, I think that our role as a community banker is I need to really understand your [business owner/ manager's] business, and the only way I can do that is I have to stay in constant contact with you.Not every week but at least once a month or once a quarter at the very end.
Findings about providing a banker with sufficient information and its verifiability align with the results of other studies (Peterson and Rajan, 1994;Kano et al., 2011) and are particularly important during times of crisis.
Small business disaster loans increase the likelihood of business recovery from both natural disasters and the Covid-19 pandemic (Davlasheridze and Geylani, 2017;Deitch and Business-bank relationships and crises Corey, 2011;Hiramatsu and Marshall, 2018;Katare et al., 2021).Torres et al. (2019), in their study of small businesses after Hurricane Katrina, find that small businesses with links to community institutions such as banks were more likely to be resilient.Given that disaster loans have a significant effect on small business recovery, it is important for businesses to have pre-established relationships with a bank prior to moments of crisis.Banker 1 expressed the opinion on providing PPP loans to well-established customers as, . . .and there was some basic financial information that were required of those PPP loans, and I would share that it was much more convenient if someone was an existing client and we prepared a PPP loan application for them, and helped them secure it, than if it was not a client of ours.
However, a very small percentage of disaster-stricken small businesses apply for disaster loans (Runyan, 2006;Zhang et al., 2009).Katare et al. (2021) reveal that less than a third of small business owners in their study applied for PPP loans.Bureaucratic red tape and lack of experience in loan applications are some of the main reasons for the lower prevalence of SBA disaster loan applications among small business owners (Runyan, 2006).In line with findings of Runyan, the most frequent responses provided by the respondents about why they do not have a relationship with a bank include no need to have a bank, a bank having no interest in the business, difficulty with the loan application and a high probability of loan rejection.Thus, a banker's ability to establish trust and open paths of communication with the businesses they serve today may be the very thing that helps these businesses access the resources they might need to recover from future disasters.

Study limitations
As part of this research, we also acknowledge several limitations, mainly related to methodology, and offer suggestions for further research.This study provides insights into the social aspects associated with creating business-bank relationships.Our results imply that those relationships are not necessarily associated with personal and business characteristics but may be based on the business owner's trust and personal communication with a bank.However, it can be argued that the reverse causation scenario is highly likely, and the already existing formal or informal relationship with a bank positively influences trust and satisfaction.Handling endogeneity in the cross-sectional dataset is extremely difficult because all variables are measured simultaneously.Nevertheless, it is possible to prioritize variables and results based on the theoretical arguments and the theoretical plausibility of reciprocal effects (Lynch and Brown, 2011).The use of cross-sectional data is another limitation of this study.Unlike longitudinal studies, our dataset does not separate between a presumed cause and its possible effect.To address this drawback, follow-up interviews with interviewed community bankers and surveys targeting small businesses can determine changes in the behavior of bankers and small business owners during the time of crisis and after the event.
Likely common to many other voluntary online business surveys, an exposed weakness in this method of data collection is non-response bias.The reasons for the low number of responses might include ease of not responding and unwillingness to share personal and business data.Because the number of observations is not optimal, we backed our arguments with thorough theoretical reasoning and the qualitative research component -key informant interviews (Mehmetoglu and Jakobsen, 2017).In view of the limited sample size that we employ in our research, we do not consider the study findings as definitive.Rather, they should be viewed as preliminary and indicative, inviting further investigation.
Another limitation relates to common method bias caused by the possible common rater effect.This effect arises when the respondent (a rater) of dependent and independent variables is the same and leads to a tendency to respond to the survey questions in a JSBED consistent way (Podsakoff et al., 2012).We minimized this response bias by applying different measurement instruments.For example, our dependent variable is a binary variable, and trust and satisfaction-related questions were measured using a Likert scale format.In addition, the order of our survey questions does not suggest a causal link between dependent and independent variables to avoid pivoting respondents' answers to fit the assumed relationship.Notwithstanding its limitations, this study provides evidence to bolster the significance and value of an interdependent social relationship between rural small enterprises and their banks, particularly during periods of crisis and uncertainty.

Conclusion
We recognize that the financial health of rural small businesses is crucial for their prosperity and survival and, subsequently, for the social and economic development of communities.Thus, we underline the significance of relationship banking for small rural businesses in times of crisis and high uncertainty, such as the Covid-19 pandemic.We examine how trust in a community bank, satisfaction with a community bank's interaction and preferred way of communication with a bank may be associated with the willingness of business owners to establish a personal relationship with their bank representatives.
Our results contribute to the literature on the role of trust and its components in relationship banking.A business owner's trust in a banker may be positively associated with the establishment of a business-bank relationship.However, an analysis of individual trust's components reveals that the nature of trust is complex, and a failure of one or more components may lead to decreased trustworthiness of a trustee.These findings suggest that bankers in community banks should be aware of the importance of trust for small business owners and act in a way that leads to their increased trustworthiness.For example, higher satisfaction with bank services and community involvement may increase trust in a banker.In contrast, the feeling of being under pressure to share all business information may discourage a business owner and reduce the trustworthiness of a banker.
Preference for in-person interaction with a bank can be viewed as extended trust in a banker.A business owner who prefers in-person communication with a banker over a virtual option may feel more confident in sharing information and inclined to develop a relationship with a bank.The study outcomes propose that maintaining a business-bank relationship is not only important to recover from normative business cycle shocks but also from nonnormative shocks such as natural disasters and the Covid-19 pandemic.Recent disasters have made it clear that business owners need to build personal relationships with their banks to access the resources they might require during a disaster.