Bank–Firm Relationships Do Perceptions Vary by Gender (pages 837–858)
Bank–Firm
& Relationships: Do
Perceptions Vary by Gender?
Patrick Saparito Amanda Elam Candida Brush
This study examines how small-business owners’/managers’ perceptions about their banking relationships are influenced by the gender of both the small-business owner/ manager and the bank manager. This study draws from social network theory and status expectations state theory to test how gender influences key perceptions about the bank–firm relationship. Using 696 matched firm owner/manager–bank manager pairs, our results show that male–male pairs of business owner/managers and bankers had the highest levels of trust, satisfaction with credit access, and bank knowledge, while female–female pairs had the lowest levels for each measure; with mixed pairs in the middle on all accounts.
Introduction
The number of female business owners is rising rapidly and many are creating substantial businesses. While most privately held businesses in the United States are majority-owned by men (62.5%), women continue to start businesses at three to four times the rate of men (U.S. Bureau of the Census, 2007). In the United States, there are more than 10.1 million women-owned firms comprising 40% of all privately held firms, employing more than 13 million people, and generating $1.9 trillion in sales as of 2008 (Center for Women’s Business Research, 2009). Commercial banks are the primary source of funding to small firms, dwarfing other debt markets and venture capital financ- ing (Berger & Udell, 1998). Further, despite many programs to foster loans to small businesses, constricted credit access continues to plague small firms (Coleman & Robb, 2009). We propose that issues surrounding bank–firm relationships are particularly impor- tant for female small-business owners who rely more heavily on commercial and personal debt, or personal savings and investments, with only an extremely small percentage drawing on private equity (Brush, Carter, Gatewood, Greene, & Hart, 2004; Coleman, 2000; Coleman & Robb; Watson, Newby, & Mahuka, 2009).
Please send correspondence to: Patrick Saparito, tel.: 610-660-1157; e-mail: [email protected], to Amanda Elam at [email protected], and to Candida Brush at [email protected].
July, 2013 837
Research on gender differences in the access to and use of credit for new and small firms is mixed. When controlling for various factors including industry and growth, women-owned firms are no more likely to be turned down for credit or to receive credit with less favorable terms (Orser, Riding, & Manley, 2006). Controlling for key industry and business factors, Robb and Wolken (2005) found no differences in bank lending practices, approval rates, or credit terms between men-owned and women-owned busi- nesses. Other research shows that while women place more importance on relational factors and advice in their interaction with banks, no gender differences exist with respect to the terms and conditions of bank financing, or the provision of service (McKechnie, Ennew, & Read, 1998). However, several studies challenge these findings with evidence that women are charged higher interest rates and need greater collateral to meet financing terms (Coleman, 2000), suggesting that bank managers assess loan applications and interact with female business owners in a discriminatory fashion (Buttner & Rosen, 1992; Carter, Shaw, Lam, & Wilson, 2007; Fay & Williams, 1993). In summary, the evidence that there are gender differences in the provision of credit is equivocal. However, what we do learn from these studies is that women tend to cluster into certain industries, have less industry experience, and smaller and slower growing firms than men (Coleman; McKechnie et al.; Orser et al.; Robb & Wolken).
Researchers generally agree that, compared with men, women have different expec- tations and perceptions for their businesses and banking relationships. Recent studies indicate that not only do female small-business owners start ventures with fewer resources, but they also have expectations for slower growth (Cliff, 1998) and are less familiar with credit sources (Carter, Brush, Greene, Gatewood, & Hart, 2003). Lower expectations for growth and a lack of familiarity with credit sources can cause borrowers to incur psychic costs associated with the loan application process (Kon & Storey, 2003). As such, women may be systematically more likely to be discouraged borrowers who choose not to enter into the credit market because of perceptions that their applications have a high probability of being rejected (Kon & Storey). This phenomenon may explain, in part, evidence from the Center for Women’s Business Research (CWBR) that shows that women are more likely than men to choose financial products based on positive experiences or relationships with lenders (CWBR, 2009). Hence, even though studies are mixed on whether or not female business owners are equally or less likely than male business owners to obtain bank financing, female business owners are still generally less satisfied with both the business-related and interpersonal aspects of their banking rela- tionships than male owners (Fabowale, Orser, & Riding, 1995). As such, important questions arise as to whether female business owners view their relationships with banks differently than do male owners (Coleman & Carsky, 1996), and whether the bank officer’s gender influences these perceptions.
In this study, we explore how the gender of the small-business owner/manager and of the bank manager responsible for the account influences various perceptions about the banking relationship, including trust, the bank’s knowledge of the firm, satisfaction with credit, and the likelihood of switching banks. We contrast key propositions from two theoretical perspectives to examine this relationship: social networking theory and status expectations state theory. Social networking theory suggests that gender homophily will lead to better perceptions of banking relationships between similar pairs in a given institutional setting (Brashears, 2008; Kim & Aldrich, 2005; McPherson, Smith-Lovin, & Cook, 2001). In contrast, status expectations state theory implies that, in the context of male-typed occupations, better perceptions of banking relationships will be found among men, not because of homophily, but because social ideals attribute greater competence to
838 ENTREPRENEURSHIP THEORY and PRACTICE 838 ENTREPRENEURSHIP THEORY and PRACTICE
Key Perceptions About the Bank–Firm Relationships
The perceptions of both bank officer and small-business owner/manager within bank– firm relationships play a pivotal role in the outcomes of bank–firm relationships (Haines, Riding, & Thomas, 1991; Holland, 1994; Saparito, Chen, & Sapienza, 2004; Uzzi & Lancaster, 2003). Guided by such perceptions, business owners choose to share informa- tion or expand business with banks and, in turn, banks choose to extend credit and other business services to small-business owners (Carter et al., 2007; Cole, Goldberg, & White, 2004; Holland; Kon & Storey, 2003; Saparito et al., 2004; Uzzi & Lancaster).
Research on gender’s influence on bank lending decisions has focused primarily on the perceptions of lending officers. Studies dealing with these key perceptions have identified the existence and use of subjective decision-making criteria that adversely affect female business owners’ access to credit (Carter et al., 2007; Cole et al., 2004; Fay & Williams, 1993). Despite loan officer training models that emphasize more objective, quantifiable criteria, such as capital assets, collateral, capacity, and conditions, research indicates that the character of the applicant stands out as a key basis for decision making among bank lending officers (Carter et al.). Use of such qualitative criteria may be especially common in smaller banks (i.e., under $1 billion in assets) as loan officers in these banks are more likely to consider their “personal interactions with and assessments of loan applicants” when making loan decisions (Cole et al., 2004, p. 249). Additionally,
a recent focus group study found distinct differences between the processes that male and female lending officers use in their decision making (Carter et al.). Male lending officers tend to rely more on gut instinct, rapport, and a diffuse sense of the overall process than female lending officers, who tend to focus on formal lending models that include specific lending terms, quality of the business plan, and the presence of brokers. Together, these findings have decided implications for demand-side arguments on the limited access to credit experienced by female business owners, especially with key structural controls concerning business characteristics and industry differences in place, but offer little insight into the supply-side arguments on the self-selection of female business owners out of the process of applying for credit (Carter et al.).
In this section, we introduce several factors of particular importance to small-business owner/managers and bank officers. We draw heavily from the research of Haines and colleagues (e.g., Haines et al., 1991; Riding, Haines, & Thomas, 1994) whose early work helped bring to light three of the four perceptual variables upon which we focus. The four key perceptions in our model are: trust between firm and bank (Saparito et al., 2004; Uzzi, 1999; Uzzi & Lancaster, 2003), the firm’s satisfaction with access to credit (Dunkelberg, Scott, & Cox, 1984; Ennew & Binks, 1993, 1997; Haines et al.; Riding et al.; Uzzi), a bank’s knowledge about the firm (Haines et al.; Uzzi), and small business proclivity to shop for alternate financial institutions (Haines et al.; Riding et al.; Saparito et al.).
The formation of these perceptions is complex and is based upon significant interac- tions between the small-business owner and bank manager (Holland, 1994; Uzzi, 1999). Although finance scholars have focused on direct factors such as how a firm’s size, age, or industry influences the interest rates on loans (e.g., Petersen & Rajan, 1994), manage- ment scholars have focused on how social embeddedness, measured by the length and breadth of the bank–firm relationship, influences perceptual factors such as trust and the likelihood of switching to an alternative bank (e.g., Saparito et al., 2004; Uzzi).
July, 2013 839
Trust
Trust is defined as the intention of one party to accept vulnerability based upon positive expectations of the intentions or behavior of another party (Rousseau, Sitkin, Burt, & Camerer, 1998). The role of trust in investor–small firm relationships has received increasing attention (e.g., Saparito et al., 2004; Uzzi, 1999; Uzzi & Lancaster, 2003; Yli-Renko, Autio, & Sapienza, 2001). For example, research finds that trust-filled rela- tionships (i.e., reciprocal trust) is positively associated with greater knowledge transfer between investors and emerging firms (Uzzi & Lancaster; Yli-Renko et al.). Trust facili- tates a more timely, relevant, and fine-grained flow of knowledge by providing expecta- tions that sensitive knowledge will not be misappropriated (Nahapiet & Ghoshal, 1998; Rowley, Behrens, & Krackhardt, 2000). Moreover, trust provides connections between parties that facilitate strong self-enforcing norms, expectations of reciprocity, and long- term commitment to relationships (Uzzi). Indeed, Saparito and colleagues found that a small firm’s trust in the bank reduced the firm’s likelihood of switching to an alternative bank and Berger and Udell (2002) found that a bank manager’s trust in the small firm is positively associated with a small firm’s access to credit. Thus, economic transactions embedded in trust-filled relationships “reduce fears of misappropriation because transac- tors anticipate that others will not voluntarily engage in opportunistic behaviors” (Uzzi, p. 484). Past research has studied trust from the perspective of the bank (e.g., Berger & Udell, 2002; Uzzi), from the perspective of the small firm (e.g., Saparito et al.), and the reciprocal trust within investor–firm relationships (Uzzi & Lancaster; Yli-Renko et al.). This research focuses on how gender influences various perceptions, including trust, from the perspective of the small-firm owner/manager.
Satisfaction With Access to Credit
Satisfaction with access to credit refers to a small-firm owner’s/manager’s satisfaction with the amount of credit generally made available by the bank relative to the size of the loan request; the terms of attaining credit (e.g., collateral or equity investment require- ments, etc.); and the interest rate (Ennew & Binks, 1993, 1997; Haines et al., 1991; Riding et al., 1994). Adequate access to financial resources is essential to new and small firm growth, and barriers to credit access can: confine and avert entrepreneurial efforts (Churchill & Lewis, 1986; Kirzner, 1982); result in operating difficulties or bankruptcy (Coleman, 2000; Laitinen, 1991); and, ultimately, affect the overall economic health of the business (Berger & Udell, 1998). As such, satisfaction with access to credit is critically important to how small-business owners and managers make sense of the relationship with their bank (Dunkelberg et al., 1984; Riding et al.).
Bank Knowledge About the Firm
A bank’s knowledge about a firm is a primary factor influencing small-firm access to credit (Cole et al., 2004; Diamond, 1991; Petersen & Rajan, 1994; Uzzi & Lancaster, 2003). Firms within the small-cap debt finance markets rarely have rated debt and often do not have reliable or certified financial statements (Uzzi & Lancaster). Consequently, knowledge about these firms’ situations is generally not publicly available and is unevenly dispersed (Petersen & Rajan; Uzzi & Lancaster). Efficient debt markets depend upon a sufficient understanding by investors (e.g., banks) to make informed investment decisions (Diamond; Uzzi & Lancaster). Insufficient knowledge and understanding surrounding a loan application increases the lender’s perceived risk of the loan, and lenders may decide
840 ENTREPRENEURSHIP THEORY and PRACTICE 840 ENTREPRENEURSHIP THEORY and PRACTICE
While all scholars agree that knowledge plays a pivotal role in bank–firm relation- ships, empirical work in this area has measured knowledge in various ways. Petersen and Rajan (1994) did not directly measure knowledge, but used the length and breadth of the bank–firm relationship as a proxy for knowledge. Their assumption was that longer and broader bank–firm relationships would allow the bank to accumulate greater knowledge about the small firm. Alternatively, Uzzi and Lancaster’s (2003) qualitative study mea- sured knowledge through a series of interviews with bank loan officers. A third approach to measuring knowledge has been to ask small-firm owners/managers their perceptions about the bank’s knowledge of the firm’s business and marketplace (Binks, Ennew, & Reed, 1992; Ennew & Binks, 1993; Haines et al., 1991). Since this study explores how gender influences the small-firm owner’s/manager’s perceptions about the bank–firm relationship, we measure small-firm owners’/managers’ perceptions about the bank’s knowledge of the firm’s business and marketplace.
Likelihood of Switching to an Alternative Bank
A small firm’s likelihood of switching to an alternative bank refers to the small-firm owner’s/manager’s current and future perceptions regarding the potential to shop for and switch to an alternative financial institution (Riding et al., 1994). Customers’ shopping activity has received considerable scholarly interest because a durable bank–customer relationship positively impacts a bank’s profitability (Berlin & Mester, 1998). This prof- itability arises because banks avoid the cost of replacing customers who have shopped and switched to another bank. Additionally, customers who are committed to a particular bank may be willing to pay a small premium in interest rates and fees, and long-term banking relationships facilitate the achievement of economies of scope through selling additional financial services and products (Berlin & Mester). In short, many activities that banks undertake are designed to maintain and expand their economic relationships with existing customers.
Perceptions about inter-firm relationships are influenced by many factors, and gender is one of these important factors (Ely, 1995). To our knowledge, no research to date has considered the importance of both the lending officer’s and small-business owner’s gender simultaneously and how these together influence the business owner’s perception of the banking relationship. Additionally, while various studies have exam- ined different subsets of these four key perceptions of banking relationships, we believe this is the first study to examine all four of these perceptions in the same study. We explore these key perceptions in the next section and propose competing sets of hypoth- eses (homophily versus status expectations) for each key perception identified as central to the bank–firm relationship.
Gender and Perceptions of the Bank–Firm Relationship
Although research appears to demonstrate that female- and male-owned firms are not significantly different in terms of credit access (Orser et al., 2006; Robb & Wolken, 2005; Watson et al., 2009), female business owners have less positive perceptions about many factors surrounding their business relationships compared with male owners (Fabowale
July, 2013 841 July, 2013 841
a reflection of discrimination or other factors coloring the experience for female business owners. In light of these findings, an exploration of the influence of gender on these central perceptual attributes of the bank–firm relationship is warranted.
Homophily
In this study, we draw on two theories to investigate the influence of gender on small-business owner–banker relationships. First, we consider social networking theory, particularly the concept of homophily. As one of the most robust findings in social network research, homophily has been defined as “the tendency for demographically similar people to associate with each other” (Brashears, 2008, p. 400). In other words, similarity breeds connection and results in social ties that not only confer important advantages, or resources, but also results in stronger ties that are more likely to endure over time (McPherson et al., 2001). Of course, while all social networks tend toward homogeneity, not all relationships are the same and some individuals are sought out more often than others (Kim & Aldrich, 2005). Indeed, research on gender and homophily suggests that male entrepreneurs accumulate better social capital than female entrepreneurs by developing larger professional networks with more diverse, powerful, and ultimately resourceful ties (Aldrich, Elam, & Reese, 1997; Brashears; Elam, 2008; Ibarra, 1992).
Such findings on the importance of homophily in individual-level interactions have powerful implications for the role that gender plays in connections made between small- business owners and bankers. If men have stronger professional networks, then, in a strictly structural sense, homophily produces certain professional advantages for men relative to women. However, there is more to the question of how homophily produces gender advantage or disadvantage than simply structural effects. Social status (i.e., the legitimacy or valuation of an individual based on ascribed status characteristics) also tends to confer different rewards, or resources, on women compared with men (Burt, 1998; Renzulli, Aldrich, & Moody, 2000). In this sense, homophily in professional relationships produces additional disadvantages—status disadvantages—for women compared with men. On this point, Ibarra (1992) found that men tend to form stronger homophilous ties across multiple networks, whereas women tend to adopt stronger and denser homophilous ties in personal networks and more instrumental and diverse ties in their professional networks. In a later study, Ibarra (1997) found that high advancement potential female
842 ENTREPRENEURSHIP THEORY and PRACTICE 842 ENTREPRENEURSHIP THEORY and PRACTICE
Importantly, whether homophily confers advantages or disadvantages on a given individual depends on the activities in question. In a study of business discussion net- works, Renzulli et al. (2000) found that the gender homophily effect in business network strength is explained largely by the proportion of kin in the business owners’ network, placing business owners with a large proportion of demanding kin ties in their social networks at a critical disadvantage. Homophily effects represent key structural differences in the networks of female business owners that put them at a clear disad- vantage in terms of accessing professional resources, compared with male business owners. However, in the realm of caring work and kin-keeping activities, the structural effect of homophily likely confers important advantages when it comes to caring for family members, especially children. In this study, our focus is on the structural impact of homophilous ties between business owners and bankers and how these pairings affect perceptions of the banking relationship.
Status Expectations State Theory
Next we consider status expectations state theory, in particular the concept of cul- turally defined expectations of competence. Status expectations states refers to the cultural beliefs organized along the lines of social status differences, like gender, that set individual expectations about how the self or others will perform at a given task (Foschi, 2000; Ridgeway & Correll, 2004). An important distinction in this theory is the concept of salience—when gender is salient in the context of a particular task situation, cultural beliefs about gender function as rules of the game (Ridgeway & Correll). In effect, when gender is salient, double standards are applied in prejudgments of com- petence (Foschi).
Research on the double standards of competence applied in professional contexts indicates that women face considerable disadvantages in both prejudgments of compe- tence and assessments of performance (Foschi, 2000; Ridgeway & Correll, 2004). In laboratory studies modeling the hiring process, researchers have consistently found that men tend to be rated or selected for hire more often than women, notwithstanding evidence of higher qualifications for women (Foschi; Ridgeway & Correll). Moreover, studies have found that women tend to favor evidence of qualification and ability over gender status in appraisals of professional competence, as compared with men (Foschi; Foschi, Lai, & Sigerson, 1994).
Status expectations state theory is an important, but underutilized, body of theory in gender studies (Chafetz, 1997). Gender scholars have argued that the persistent cultural beliefs linking gender with task-specific abilities constitute a gender system—i.e., an institutionalized system of relations organized around distinctions between two genders, resulting in different roles, identities, expectations of competence, performance assess- ments, and, consequently, the distribution of resources and rewards (Ridgeway & Correll, 2004). Cultural beliefs linking gender and expectations of competence are perpetuated in large part through confirmatory experience of individual actors. Ridgeway and Smith- Lovin (1999) argue that while research on peers with equal status and power show few gender differences in behavior, most gender interactions do not occur on an equal footing. In most professional interactions, men hold higher status positions and women lower status positions, leading to confirmation of existing beliefs.
July, 2013 843
Competing Hypotheses From Homophily and Status Expectations
The distinction between homophily and status effects has attracted little attention in entrepreneurship studies (Elam, 2008). One exception is Ruef, Aldrich, and Carter (2003) who found that homophily and ecological constraints work together to produce minority isolation among entrepreneurial founding teams. Typically, however, the application of the concept of homophily in social networking theory takes for granted the status differ- ences between men and women in the context of new/small-business ownership. In contrast, status expectations state theory makes explicit the possibility that status differ- ences resulting from gender-linked expectations of competence will vary with the task set as well as with the gender of the individual studied.
For this study, we develop four sets of hypotheses, comparing predictions of homoph- ily versus status expectations state effects on key aspects of small-business owner–banker relationships. We argue that both homophily and gender-linked status expectations of competence provide a basis for key perceptions in banking relationships. However, as theoretical constructs, each concept constitutes a distinct social process and must be considered separately.
Social networking theory posits that trust and positive perceptions are governed by the homophily mechanism (McPherson et al., 2001). In other words, individuals will experi- ence the highest levels of trust in relationships, or pairings, with similar others. As such, the homophily argument predicts that same gender pairings (male–male and female– female pairs) will show the highest levels of positive perceptions, followed by mixed-pair relationships with lower levels of trust and less positive perceptions.
In contrast, status expectations state theory argues that positive perceptions will most likely occur in relationships, or pairings, where the fit between the diffuse status charac- teristics of the person being judged (e.g., male or female) and the task set, or role, at hand (banker or small-business owner) follows widely held cultural ideals about gender and appropriate task sets (Foschi, 2000; Ridgeway & Correll, 2004). Both banking and business ownership are typically stereotyped as highly masculine endeavors (Ahl, 2004; Bird & Brush, 2003). Under the assumption that status expectations serve as the under- lying mechanism by which gender exerts an influence on the bank–firm relationship, the rankings should look quite different from those predicted by the homophily argument. In fact, the status expectations state theory argument predicts that for bank–manager/ business–owner relationships, respectively, we expect male–male pairings to have the most positive perceptions, followed by mixed pairings (i.e., female–male/male–female) and, finally by female–female pairings.
As a consequence, the competing hypotheses for these two theoretical perspectives appear as follows:
Homophily Status Expectations State Theory
Hypothesis 1a: Trust in the bank will be higher for male–male and Hypothesis 1b: Trust in the bank will be highest for male–male female–female pairs compared with mixed pairs.
pairs, followed by mixed pairs, and female–female pairs. Hypothesis 2a: Satisfaction with credit access will be higher for
Hypothesis 2b: Satisfaction with credit access will be highest for male–male and female–female pairs compared with mixed pairs.
male–male pairs, followed by mixed pairs, and female–female pairs.
Hypothesis 3a: Perceptions of the bank’s knowledge about the Hypothesis 3b: Perceptions of the bank’s knowledge about the firm firm will be higher for male–male and female–female pairs
will be highest for male–male pairs, followed by mixed pairs, and compared with mixed pairs.
female–female pairs.
Hypothesis 4a: Likelihood of switching banks will be lower for Hypothesis 4b: Likelihood of switching banks will be lowest for male–male and female–female pairs compared with mixed pairs.
male–male pairs, followed by mixed pairs, and female–female pairs.
844 ENTREPRENEURSHIP THEORY and PRACTICE
Research Methodology
We employed a matched-pair sample design of small-firm owners/managers and the bank managers who had responsibility for each of their firm’s accounts. The completion of the three-stage data collection resulted in 696 matched pairs of firm owners/managers– bank account managers. All dependent variables were collected directly from small- business owners/managers. This study design is appropriate given that the purpose of this study was to examine how gender influences firm owner’s/manager’s perceptions of the bank–firm relationship. Data regarding an individual’s gender and age were self-reported by both the small-firm owners/managers and the bank managers. Control variables were collected from the: small-firm owners/managers (i.e., firm size, firm age, growth rate, and business sector; the bank–firm relationship’s age; and whether the bank was the firm’s primary financial institution); bank (i.e., bank size) and the United States Federal Reserve (i.e., bank market competitiveness).
Participating Banks
In the first phase of data collection, we approached 286 banks in Connecticut, Missouri, New Jersey, and Pennsylvania. This geographic frame includes a good mix of rural, suburban, and urban areas providing for potentially more generalizable results. Since large banks are generally less involved with small business clients (Cole et al., 2004; Petersen & Rajan, 1994), we limited variation associated with bank size by targeting moderately sized banks with total assets of $100 million to $10 billion. Twenty-two banks agreed to have surveys distributed to both their managers and their small and mid-sized commercial customers and bank managers. This represented a 7.67% participation rate. As we explain more fully below, we asked banks to provide a client list, a lending officer list, and permission to survey both groups. Therefore, we recognized that the survey methodology to attain a matched response of small-firm owners/managers–bank manag- ers was extensive and we were not surprised at this response rate. Participation rates
between states did not differ significantly ( c 2 = 4.55, not significant [n.s.]). The median total assets for banks that participated was $248.5 million versus $237.2 million for banks that did not, a difference that is not statistically significant (F = 0.95, n.s.).
Sample of Small-Business Owners/Managers
Each bank compiled a list of small businesses that had both deposit and lending relationships with the bank. Small businesses were defined in terms of having a credit limit of $1 million, a criterion consistent with that adopted by the U.S. Small Business Administration. We randomly distributed 7,298 surveys to this sample list of bank clients and 1,093 surveys were returned, representing a 14.98% response. This response rate is consistent with similar studies collecting survey data from small businesses about their bank–firm relationships (e.g., Binks & Ennew, 1997; Haines et al., 1991; Petty & Upton, 1997). We asked that the survey be completed by either the business owner/operator or the person primarily responsible for interacting with the bank if there were multiple owners/ operators. In the case of multiple owners/operators of both genders, we used the gender of the owner/operator who was primarily responsible for interacting with the bank. Since surveys were anonymous, it was not possible to calculate differences between respondents and nonrespondents. However, late respondents are considered similar to nonrespondents (Churchill, 1991), and t-tests comparing early versus late respondents found no significant differences for any firm variables.
July, 2013 845
Sample of Bank Managers
In the final phase of data collection, each responding small-business owner/manager identified the bank officer primarily responsible for the company’s account, and each participating bank notified bank employees to expect a survey from the research program. Two hundred sixty-three bank managers were identified and sent surveys. Cover letters indicated that all responses would be anonymous and confidential and surveys were directly returned to the research office in an enclosed and stamped envelope. Managers returned 217 surveys, representing an 82.51% response rate.
Statistical Procedures
We used multivariate analysis of covariance (MANCOVA) to test for differences in levels of trust, satisfaction with credit, perceptions of a bank’s knowledge about the firm, and likelihood to switch to an alternative bank between female–female, mixed gender, and male–male small-business owner/manager–bank manager pairs. In order to better isolate and identify differences in our dependent variables due to gender, we were careful to control for factors that are known to influence bank–small firm relationships and the perceptions of these relationships.
Measures
Trust
We measured trust using Saparito et al.’s (2004) 4-item index measure. Using a 7-point scale (where 1 = very rarely true and 7 = very often true), small-firm customers rated the following items: (1) We feel that the bank would act in a fashion consistent with what we would recommend without prior discussion with us; (2) We can freely share concerns and problems about our company and know that they will respond construc- tively; (3) We can freely share concerns and problems about our company and know that they will be interested in listening; and (4) We share common business values with the bank (alpha = .89).
Satisfaction with Credit Access
We measured satisfaction with credit access by a 5-item scale. Items were created and adapted from measures used in large national investigations of small business banking in the United States, the United Kingdom, and Canada (e.g., Dunkelberg et al., 1984; Ennew & Binks, 1997; Haines et al., 1991). The measure was pilot tested on a sample of small businesses that were clients of a small business development center in a Northeastern U.S. state. Using a 7-point scale (where 1 = very dissatisfied and 7 = very satisfied), small-firm customers rated the following items: (1) The credit amount that the bank generally makes available when our company makes loan requests; (2) The bank’s security and collateral requirements for obtaining a loan; (3) The bank’s requirements for personal/ company equity invested in the business prior to granting a loan; (4) The bank’s financial reporting requirements for granting a loan; and (5) Interest rates charged on loans (alpha = .91).
We believe that it is important to point out that, in this study, we are measuring satisfaction with access to credit and not actual access to credit. We believe this is
846 ENTREPRENEURSHIP THEORY and PRACTICE 846 ENTREPRENEURSHIP THEORY and PRACTICE
Bank’s Knowledge of the Firm
Drawing upon the literature and previous measures (Binks et al., 1992; Haines et al., 1991), we created and adapted four items to measure a bank’s knowledge of the firm. This measure was pilot tested along with the measure for customer satisfaction with credit. Using a 7-point scale (where 1 = very dissatisfied and 7 = very satisfied), small-firm customers rated: (1) The bank’s knowledge of our business; (2) The bank’s knowledge of the local market/community; (3) The bank’s anticipation of our credit and other financial needs; and (4) The bank’s anticipation of our financial needs other than credit (alpha = .92).
Likelihood of Switching to an Alternative Bank
We measured likelihood of switching to an alternative bank using Saparito et al.’s (2004) 5-item measure. Using a 7-item scale (where 1 = very unlikely and 7 = very likely), small-firm customers rated the extent to which they were likely within the next year to: (1) switch to an alternative bank to service their borrowing needs; (2) switch to an alternative bank for checking and other deposit accounts; (3) move accounts to banks with slightly more attractive interest rates; (4) move accounts to banks with slightly more attractive fees; and (5) shop for banks with more attractive fees and interest rates (alpha = .92).
Control Variables
We controlled for several factors considered to influence the nature of bank–firm relationships (Petersen & Rajan, 1994; Uzzi & Lancaster, 2003).
Bank Market Competitiveness
Less competitive banking markets are associated with greater credit constraint prob- lems (Berlin & Mester, 1998; Uzzi, 1999). Bank concentration has been shown to affect the access to and interest rates of loans (Petersen & Rajan, 1994; Uzzi). Additionally, less competitive banking markets also limits the ability of firms to switch to alternative banking institutions (Saparito et al., 2004). The U.S. Federal Reserve Bank measures the concentration of bank deposits across geographic banking markets via a bank concentra- tion index. Higher index numbers mean higher concentrations of deposits among banking institutions with fewer branches and alternatives. This is interpreted to mean less compe- tition within a given banking market (Petersen & Rajan). We used this more precise measure of bank competition within local markets instead of using more course-grained binary variables for the state in which a bank is located.
Bank Size
Generally, larger banks are less involved with small-business lending than smaller banks (Berlin & Mester, 1998; Cole et al., 2004; Petersen & Rajan, 1994). Additionally,
July, 2013 847 July, 2013 847
Firm Size
The size of a borrowing firm may affect numerous facets of the bank–firm relationship including: bargaining power; importance to the bank; and access to credit (Petersen & Rajan, 1994; Riding et al., 1994; Wathne, Biong, & Heide, 2001). We used number of employees and sales revenues as size measures; the former may have an impact on indirect business for the bank, and the latter is more directly connected with the potential business of the specific customer firm with the bank.
Firm Growth Rate
Firms were resistant to reporting specific profitability figures. However, we were able to collect a measure of a firm’s growth rate. The growth rate of sales may signal the health of a firm as well as the potential future business for the bank (Berger & Udell, 1998; Berlin & Mester, 1998). Using a 7-point scale (where 1 = decreasing rapidly and 7 = increasing rapidly), small-firm customers rated the growth rate of their firm’s sales over the past 5 years (less if the firm was younger).
Firm Age
Firm age is positively associated with the likelihood of firm survival and it may be relatively more difficult for young firms to obtain alternative lines of credit at the same or better costs (Petersen & Rajan, 1994). Firm age was measured as the number of years the firm had been in operation.
Business Sector
A firm’s business sector can influence its borrowing needs, ability to provide col- lateral, and numerous other factors associated with its ability to attain loans (Berger & Udell, 1998; Berlin & Mester, 1998; Petersen & Rajan, 1994; Saparito et al., 2004). Consistent with the seminal work of Petersen and Rajan, we used the broad business sectors for controlling for these potential affects. We collected this data by providing small-firm business owners a list of broad business sector category names (e.g., Retail, Wholesale, Manufacturing, etc.) and asked the small-firm owner/manager to indicate which business sector best described the firm. Business sector was coded using dummy variables.
Firm–Bank Relationship Age
Uzzi (1999) suggested that firm–bank relationship age is an important as a proxy for the embeddedness of the relationship. Longer relationships allow both the bank and firm time to get to know one another, develop social ties, and exchange information (Berlin & Mester, 1998; Petersen & Rajan, 1994; Uzzi). Thus, the age of the firm–bank relationship can directly influence a bank’s knowledge about the firm and the firm’s access to credit
848 ENTREPRENEURSHIP THEORY and PRACTICE
(Berlin & Mester; Petersen & Rajan; Uzzi). We measured the bank–firm relationship age as the number of years the firm had a business account with the bank.
Primary Bank
Whether or not a bank is the firm’s primary banking institution is also an important factor influencing: the amount of information exchanged between the bank and the firm, the firm’s credit access, and the likelihood of the firm switching to an alternative banking institution (Berlin & Mester, 1998; Petersen & Rajan, 1994; Saparito et al., 2004; Uzzi, 1999). Therefore, we measured whether the bank was the firm’s primary financial insti- tution using a dummy variable (1 = if the bank was the firm’s primary bank and 0 = oth- erwise).
Small-Firm Owner/Manager Age and Bank Manager Age
The business owner’s/manager’s age is often used as a proxy for human capital (experience) and positively related to performance (Bates, 1990). We collected the age of the small-firm owner/manager and bank manger directly from each respondent in the respective surveys by asking them to identify what category best described their age group (i.e., 21–30, 31–40, 41–50, 51–60, 61–70, more than 70). We coded this using a 7-point scale (1 = 21–30 to 7 = more than 70).
Results
Responding bank managers and small-firm owner/managers were 47.1% and 27.6% female, respectively, with a median age range of 31–40 and 41–50, respectively. This age breakdown is consistent with the breakdown of small-business owners in the United States, whereby the bulk are between the ages of 45–54 (U.S. Bureau of the Census, 2007). Additionally, responding small-firm owners/managers within various business sectors ranged from 16.1% female in wholesale to 32% female in retail.
The correlations, means, and standard deviations appear in Table 1. As expected, trust, bank knowledge about a firm, and satisfaction with credit access all show large and highly significant positive correlations with one another (each at p < .01). Also as expected, trust, bank knowledge about a firm, and satisfaction with credit access show large and highly significant negative correlations with the likelihood of switching to an alternative bank (each at p < .01).
Because each of these variables is highly correlated with one another, it is appropriate to simultaneously test for differences using MANCOVA instead of separate ANCOVA analyses for each of the four dependent variables (Hair, Anderson, Tatham, & Black, 1995). Our hypotheses suggest two different orderings for the gender pairings—the homophily argument predicts that homophilous pairs (male–male and female–female) will produce the most favorable scores on test factors, while status expectations state theory suggests that male–male pairs will show best scores, followed by mixed gender pairs and female–female pairs. Statistical results support the assertion that gender pairings do influence the dependent variables (Wilks lambda l = .97, F 8, 1286 = 2.66, p < .01). Looking more closely, we see in Table 2 that female–female pairs show the lowest levels of small-firm owners’ trust in the bank, mixed pairs an intermediate level of small-firm owners’ trust in the bank, and male–male small-firm owners the greatest trust in the bank
July, 2013 849
Table 1 Correlations, Means, and Standard Deviations for Variables in the Study
1. BCI 2. Bank size
3. Firm employees
4. Firm sales
5. Firm growth
6. Firm age
7. Relationship age
8. Primary bank
9. Firm owner age
10. Bank manager age
12. Bank knowledge
.69** 14. Likelihood of switching
13. Customer satisfaction
3.37 2.93 21.57 21.02 21.61 14.29 Standard deviation
BCI, bank concentration index. PRA
CTICE
Table 2 Differences in Means by Entrepreneur–Bank Manager Gender †
Entrepreneur–bank
Bank knowledge Likelihood manger gender
Satisfaction with
Trust
credit access
of firm of switching
Female–female ‡ (n = 87)
20.68 (5.28) 11.24 (7.47) Male–female or female–male ‡ ‡ (n = 305)
20.98 (5.31) 11.27 (6.72) Male–male (n = 304)
21.95 (5.34) 11.41 (6.63) Total (n = 696)
‡ The standard deviation is provided in parentheses beside each cell mean. Means are adjusted for covariates.
(F = 5.67, p < .001). This pattern of results suggests that the status expectations state theory argument for differences in trust (hypothesis 1b) is supported, while the homophily argument for differences in trust (hypothesis 1a) is not supported.
Hypothesis 2a proposed a homophily argument for differences in satisfaction with credit access, while hypothesis 2b proposed that differences in levels of satisfaction with credit access would vary according to status expectations state theory. As shown in Table 2, female–female pairs show the lowest levels of satisfaction with credit access, mixed pairs an intermediate level of satisfaction with credit access, and male–male small-firm owners the most satisfaction with credit access (F = 4.65, p < .01). Again, this pattern of results suggests that the status expectations state theory argument for differ- ences in satisfaction with credit access (hypothesis 2b) is supported, while the homophily argument for differences in satisfaction with credit access (hypothesis 2a) is not supported.
Hypothesis 3a proposed a homophily argument for differences in levels of small-firm owners’ perceptions concerning a bank’s knowledge of the firm, while hypothesis 3b proposed that differences in levels of a small-firm owners’ perceptions about a bank’s knowledge about the firm would vary according status expectations state theory. Again, the results in Table 2 show that female–female pairs have the lowest levels of small-firm owners’ perceptions about a bank’s knowledge, mixed pairs an intermediate level of small-firm owners’ perceptions about a bank’s knowledge, and male–male small-firm owners the highest small-firm owners’ perceptions about a bank’s knowledge (F = 3.49, p < .05). Again, this pattern of results suggests that the status expectations state theory argument for a bank’s knowledge about the firm (hypothesis 3b) is supported, while the homophily argument concerning the bank’s knowledge of the firm (hypothesis 3a) is not supported.
Finally, hypothesis 4a proposed a homophily argument for differences in a small-firm owners’ likelihood of switching to an alternative bank, while hypothesis 4b proposed that differences in a small-firm owner’s likelihood of switching to an alternative bank would vary according to the status expectations state theory. As shown in Table 2, the differences between means for the likelihood of switching banks are not significant (F = 0.10, n.s.). Thus, neither the homophily argument (hypothesis 4a) nor the status expectations state theory argument (hypothesis 4b) are supported.
July, 2013 851
Discussion
We proposed a competing set of hypotheses drawing from homophily (hypotheses 1a, 2a, 3a, and 4a) and status expectations state theories (hypotheses 1b, 2b, 3b, and 4b). The homophily perspective proposes that the more we have in common, the more positive perceptions and expectations we have about the other party (Brashears, 2008). Connec- tions, so the argument goes, are more readily formed with similar others, perhaps leading to (or because of ) better communications. In contrast, the status expectations state perspective argues that, in the absence of better information, perceptions are based on mental models that support prejudgments of competence relative to a specific task (Foschi, 2000; Ridgeway & Correll, 2004).
Our findings suggest that perceptions about gender status and gender-appropriate work roles influence small-firm owner/manager perceptions about the bank–firm relation- ship and, as such, our findings were generally supportive of status expectations state theory. More specifically, our proposed hypotheses regarding a small-firm owner’s/ manager’s trust in the bank (hypothesis 1b), satisfaction with credit access (hypothesis 2b), and perception of the bank’s knowledge of the firm (hypothesis 3b) were supported. However, the status expectations state theory argument for differences in the likelihood of switching to an alternative bank (hypothesis 4b) was not supported. Our findings did not support any of the hypotheses based on the homophily argument.
While not all of our status expectations state theory hypotheses were supported, we did not expect a perfect fit. These findings advance our understanding about the way homophily and status expectations may shape the nature of bank–firm relationships. The homophily perspective argues that dealing with similar others produces predictable out- comes. Status expectations state theory argues that similarity is not sufficient because perceptions of competence of a predefined task will predict assessments of trustworthiness and fair treatment. In other words, when one party in a dyad does not conform to traditional occupational and gender expectations, the nature of the negotiations and interactions can be influenced as one or the other seeks to conform to gender stereotypes for the job at hand (Nelson, Maxfield, & Kolb, 2009). When parties are working to conform to certain status expectations, they may perceive the other person to be less trustworthy and/or to be less satisfied with the nature of the relationship. Our findings support earlier work by Ibarra (1992), who found homophily and status connections were positively correlated for men in advertising but negatively correlated for women. She argued that social networks play a role in influencing these results. It follows that if female business owners have few women in their financial networks, they would have certain expectations and perceived stereotypes about the likely competence of female bank managers. The extent to which gender stereotypes influence key perceptions of trust and satisfaction of the parties in banking relationships is a logical extension of this research.