Financial Discrimination: Consumer Perceptions and Reactions

Abstract

Access to traditional financial institutions plays a key role in enhancing positive financial outcomes. However, some consumers within the United States experience discrimination from these same institutions. In particular, discrimination based on race and gender has historically been tied to outcomes such as lower service quality and a lack of access to credit. While the previous literature has discussed some of the discriminatory practices that these groups have faced, there is a lack of research on how these groups respond to discrimination from financial institutions. Through a series of logistic regressions, the authors analyzed how race, ethnicity, and gender are related to reporting experiences of discrimination. The authors then explored how consumers react to discrimination by looking at five reported reactions. Primary results show that Black consumers were more likely than most other racial groups to experience financial discrimination. Additionally, women were less likely than men to report financial discrimination. Race was shown to be a significant factor in four of the five reactions to discrimination, while gender was a factor in two of the reactions. The findings further show that after experiencing financial discrimination, most individuals turned to non-traditional financial services as a direct result of the bias or racism.

1. Introduction

Discrimination is closely linked to bias and racism, as these concepts commonly manifest in the mistreatment of individuals or groups (American Psychological Association, n.d.). Bias, defined as “an inclination or predisposition for or against something” (American Psychological Association, 2018), shapes how individuals view others based on personal traits or group membership. These biases can influence attitudes, behaviors, and decisions. For some individuals, biases can produce a favorable outcome such as when more attractive candidates are given preferential treatment in hiring and promotion decisions (Marlowe et al., 1996). However, when biases are acted upon in a harmful way, they manifest as discrimination. This discriminatory behavior is most often subtle, like giving unequal opportunities or privilege (Ode et al., 2022; Ozturk & Berber, 2020; Wingfield & Chavez, 2020). Therefore, discrimination can be thought of as a product of biases, particularly when the biases lead to the unequal treatment of marginalized populations.

Racism, one of the most pervasive forms of discrimination, arises from bias against people based on their perceived race or ethnicity (American Psychological Association, n.d.). Most often, racism is expressed through negative emotional reactions, stereotypes, and discriminatory practices aimed at persons of color (Pager & Shepherd, 2008). Racism operates at both the individual level, such as in personal interactions or hiring practices, as well as at the institutional level, where structures and policies favor certain racial groups over others (Banaji et al., 2021; Elias & Paradies, 2021; Pager & Shepherd, 2008; Pittman, 2020). The cycle of bias, racism, and discrimination often work together to reinforce barriers and limit opportunities for individuals targeted by these prejudices. Recipients of bias, racism, and discrimination have even been known to experience physiological and psychological responses such as increased blood pressure and heart rate, stress, anxiety, and depression (Brandt et al., 2022; Crockett et al., 2003; Javed et al., 2022; Remedios & Snyder, 2015; Wang & Narcisse, 2025).

Access to financial products and services are instrumental in an individual’s financial resilience, stability, and well-being (Nandru et al., 2021; Sakyi-Nyarko et al., 2022; White et al., 2022). However, there are disparities in treatment and access to favorable financial products and services, particularly along racial, gender, and socioeconomic lines (Scott et al., 2024). Previous research has documented instances of bias and discrimination in lending, banking, and investment practices, possibly leading to long-term economic disadvantages and threats to financial well-being for marginalized communities (Cheles-McLean, 2024; Fernández-Olit et al., 2019; Li & Zhou, 2024; Scocca & Meunier, 2022; Scott et al., 2024; Wherry & Perry, 2021). It should, however, be noted that some research suggests that factors such as lower credit scores and higher leverage are more substantial factors in credit lending outcomes, specifically mortgage lending, as opposed to discrimination (Bhutta et al., 2025). Regardless, it is still important to consider how consumers react to discrimination as consumers may respond in a manner that is financially disadvantageous. Some reports suggest that discrimination may lead consumers to lose trust in financial institutions or seek out alternative financial services that are often predatory (Rice, 2022; San Francisco Office of Financial Empowerment, 2020). For marginalized communities, this can create another hurdle to long-term and intergenerational financial success. Despite the importance of access to traditional financial services and the possibility that discrimination may lead consumers to avoid utilizing such services, to the authors’ knowledge, there has been a minimal amount of research documenting the specific actions individuals have taken after experiencing bias, racism, or discrimination from a financial institution.

This study explores two main questions to examine factors associated with experiencing bias or racism when working with financial institutions and the subsequent responses of those affected. (1) What factors are associated with reporting experiences of financial bias or racism? (2) What factors are predictors of the different reactions taken after experiencing financial bias or racism? Specifically, we are seeking to understand the role race, ethnicity, and gender play in reporting experiences of financial bias or racism (referred to as “financial discrimination” throughout the paper) in an American context and the actions that people take thereafter. A secondary dataset, focusing on the financial and health experiences of low- to moderate-income individuals during the COVID-19 pandemic, was used. The sample included a variety of racial groups. Just over 50% of the respondents were White, but BIPOC (Black, Indigenous, and People of Color) individuals were intentionally oversampled.

The results of the study suggest that Black consumers were more likely than all other racial groups to experience financial discrimination, except those identifying as Native/Pacific Islander, a group which includes American Indian, Alaska Native, Native Hawaiian, and Other Pacific Islander. Women were less likely than men to report financial discrimination. The descriptive data show that most people who experienced financial bias or racism turned to non-traditional financial services as a result. Race was a significant factor in four of the five reactions to discrimination, and gender was a factor in two of the reactions. Being Hispanic was not significantly associated with any of the reactions.

The paper is organized as follows: Section 2 provides a review of the existing literature, Section 3 contains an overview of the conceptual framework, Section 4 includes details about the data, variables, and methodology, Section 5 reports descriptive statistics and regression results, Section 6 is a discussion of the results with implications of the findings, and Section 7 provides the conclusions.

2. Literature Review

Financial services play a key role in creating positive financial outcomes for consumers. The ability to utilize high-quality financial services has been linked to enhanced financial well-being (Hoang et al., 2023), a higher net worth (Ehrmann & Ampudia, 2017), and increased mental health (Aguila et al., 2016). These services cover a range of responsibilities meant to provide communities with access to credit, banking accounts, and other services. It was estimated in 2023 that 96% of U.S. households had some form of bank or credit union account (FDIC, 2024). However, despite the majority of U.S. households maintaining some form of interaction with traditional financial institutions, certain communities have historically faced discrimination by these same institutions.

Modern day discrimination in financial institutions is widely believed to be the result of centuries of unfair regulation (Banaji et al., 2021; Lynch et al., 2021; Rice, 2022). Policies such as Jim Crow laws and the Indian Removal Act placed restrictions on where certain racial/ethnic groups could own or lease property (K.-S. Park, 2021; Rice, 2022). Some states even utilized these policies in the form of Alien Land laws in an attempt to restrict specific racial/ethnic groups from being able to immigrate to the United States—particularly immigrants of either Chinese or Japanese descent. Often, even a policy that did not explicitly mention race or ethnicity would still result in a negative impact on marginalized individuals. In particular, as part of the Home Owners’ Loan Act of 1933, the Home Owners’ Loan Corporation (HOLC) was established. This organization created maps for residential areas across American cities and color-coded specific areas based on the perceived risk of mortgage lending (Hillier, 2005). Those that lived in the green, or “best”, areas were often White, affluent residents. Conversely, the red, or “worst” areas, were often populated by low-income or non-White residents (Hillier, 2005; New York City Department of Health and Mental Hygiene, 2021). While affluent, White residents congregated in the suburbs, non-White and low-income residents were shut out of these communities (Arce, 2021). Through this process, bankers and credit lenders became part of the system of discrimination.

Today, the effects of policies such as redlining still impact the relationship between racially and ethnically marginalized communities and traditional financial institutions. For many of these communities, not only are these institutions more difficult to physically access (Broady et al., 2021; Sakong & Zentefis, 2025), but the quality of the financial services offered in demographically diverse areas is often lower compared to areas with a predominantly White population (Begley & Purnanandam, 2021). Furthermore, racially marginalized consumers may have to pay more to access these services. A previous survey conducted by Bankrate found that White checking account holders are more likely to report that they do not pay any bank fees compared to Hispanic and Black checking account holders (Wisniewski, 2020). White account holders who do pay such fees report paying USD 5 per month, on average, in bank fees, whereas Hispanic and Black checking account holders report paying USD 16 and USD 12 per month, respectively (Wisniewski, 2020). This disparity in access is one explanation as to why; when compared to White and Asian households, Black and Latino households are more likely to be categorized as underbanked or unbanked (FDIC, 2024; Grable et al., 2025). The FDIC found that individuals who are considered underbanked often lack affordable and easy access to traditional financial institutions. In turn, these individuals are also more susceptible to relying on alternative financial services and products such as predatory payday loans (FDIC, 2024).

Even when interacting with traditional financial institutions, non-White customers, particularly those who are Black, are more likely to experience racial discrimination (e.g., Greenlee, 2023, 2025). In 2023, Mark Greenlee highlighted evidence from social media of incidences of racial discrimination in banking among Black consumers, some even involving celebrities, in his paper, “Banking While Black, Part I” (Greenlee, 2023). Some of these incidents were followed by formal litigation (Greenlee, 2025). Research from Bates and Robb (2013) found that minority-owned businesses (MBEs) receive smaller loans and have loan applications rejected more often when compared to White-owned businesses. Additionally, Bates and Robb (2013) uncovered that if a business was located in a minority-majority neighborhood, that was positively associated with receiving smaller business loans. When looking at mortgage lending, a study by Rugh et al. (2015) found that Black consumers pay 5% to 11% more in monthly payments when compared to similarly qualified White consumers. Moreover, Black and Hispanic consumers are specifically targeted for subprime loans (Botein, 2013). This finding was further reinforced when the U.S. Department of Housing and Urban Development (2009) found that homeowners living in high-income Black neighborhoods were more likely to have subprime refinancing compared to homeowners living in low-income White neighborhoods.

Racial discrimination in financial institutions also occurs in more subtle ways. In one paper by Scott et al. (2024), the authors conducted a series of three studies to measure certain metrics of service quality. In their first study, a matched-pair mystery shopping field experiment was conducted wherein 12 Black and 12 White participants visited banks throughout the Atlanta metropolitan area to track service outcomes. The authors found that not only were Black participants provided with inferior service outcomes compared to White participants, but that Black participants also reported inferior service processes. Simply put, not only were the products offered considered inferior, but the service quality offered by employees was also considered inferior. A qualitative study by Friedline et al. (2023) echoed this finding when the authors interviewed front-line bank workers. Part of the interview was focused on the bank policies that facilitated discrimination against minority bank customers. For example, one bank employee mentioned being scolded for allowing “non-wealthy” customers to use the bank bathroom, which the employee mentioned was code for “non-White”. Customers may also experience differences in service quality based solely on their name. Hanson et al. (2011) found that when looking at real estate agents and property ownership, consumers with names that are stereotypically associated with being Black were less likely to receive responses to inquiry emails compared to names that are stereotypically associated with being White. Furthermore, emailed responses to the individuals with stereotypically Black names took longer to receive, were shorter, and were less polite.

When looking at discrimination based on gender, policy within the United States had to be specifically implemented to reduce discriminatory practices that impacted women’s ability to utilize financial institutions. For example, the Equal Credit Opportunity Act of 1974 was partially implemented to ensure that women, regardless of marital status, had the right to open their own credit and banking accounts (Rose, 2023). Moreover, credit lenders could legally discriminate against women in credit lending, which could limit women’s ability to purchase property or start businesses (Gates, 1974). This was often performed by credit lenders as women were viewed as high credit risks (Mailliard & Anderson, 1987). This perception in and of itself stemmed from the assumption that many women did not have an independent source of income (Mailliard & Anderson, 1987).

An interesting difference between discrimination based on race/ethnicity and discrimination based on gender is that there does appear to be some mixed results in regards to credit lending. Generally speaking, the previous literature argues that women face more challenges with obtaining credit (Alesina et al., 2013; Aristei & Gallo, 2016; de Andrés et al., 2021). For example, Chundakkadan (2023) found that women were less likely than men to receive business loans during a contraction period, and that this effect weakens in countries with high gender equality. The author argues that this is a form of “credit rationing” that stems from societal notions that women are less capable than men. However, some of the literature argues that women are actually favored for credit lending. Authors such as Hewa-Wellalage et al. (2022) posit that credit lenders favor women for business loans compared to men because they are seen as being lower risk.

For both race/ethnicity and gender-based discrimination, there is limited research on how consumers react when facing such events. Some individuals will report their experiences. For example, Massey et al. (2016) developed a qualitative database of 220 statements taken from documents pertaining to fair lending lawsuits that occur as a result of reporting unfair practices. They found that 76% of the compiled statements suggested structural discrimination wherein the policies of the financial institution disadvantaged minority consumers. One such statement reads as follows, “Since loan offers made more money when they charged higher interest rates and fees to borrowers, there was a great financial incentive to put as many minority borrowers as possible into subprime loans and to charge these borrowers higher rates and fees (p. 21)”. Additionally, 11% of the statements indicated individual discrimination wherein minority consumers report discrimination by an individual employee as opposed to by the firm itself (Massey et al., 2016).

Other consumers, due to monetary barriers of entry, a lack of trust in such institutions, and fears of potential discrimination, will proactively avoid working with financial institutions (Ards & Myers, 2001; Creamer & Warren, 2024). Even those customers who do work with financial institutions can report a lack of loyalty to the institution (Scott et al., 2024), which again, may reflect low levels of trust. Instead, these consumers may seek alternative pathways to financial services such as investment in cryptocurrencies (Reuters, 2024) and the use of payday lenders (Campbell et al., 2012). However, the previous literature has not explicitly discerned whether this behavior is due, at least in part, to having experienced discrimination from financial institutions.

3. Conceptual Framework and Hypotheses

The integrative consumer vulnerability framework combines two perspectives that are useful for a multidimensional exploration of the factors associated with consumers’ adverse experiences with financial institutions (Commuri & Ekici, 2008; Lim & Letkiewicz, 2023). The first component is grounded in Andreasen’s (1975) seminal concept of the disadvantaged consumer, which suggests that demographic characteristics, such as race/ethnicity and gender, constitute vulnerability. Consumers in historically marginalized identities, such as Black consumers and women, are often in positions of disadvantage and vulnerability when interacting with financial institutions (Garrett & Toumanoff, 2010). The second component incorporates the notion of the vulnerable consumer, which suggests that everyone, regardless of demographics, has the potential to be vulnerable depending on their life circumstances (Baker et al., 2005; Commuri & Ekici, 2008). In other words, vulnerability is not confined to demographics but rather is situational and dynamic. In this view, vulnerability may stem from internal conditions such as diminished cognitive ability, emotional distress, or transitional life stages, as well as from external forces like discrimination or social exclusion. According to Hill and Sharma (2020), disadvantaged consumers are disadvantaged because they are deemed unequal or worse off than other consumers in a specified context, whereas vulnerable consumers are vulnerable not necessarily because of a particular characteristic relative to other consumers, but because they are subject to harm by interactions with institutions or individuals when seeking to access goods and services in the marketplace. The independent variables (demographics, financial knowledge, and life circumstances) represent the sources and indicators of both structural disadvantage and situational vulnerability.

Furthermore, Hill and Sharma (2020) add methods to identify consumer vulnerability and a comprehensive definition of consumer vulnerability. They state that consumer vulnerability can be identified by experience or observation. Experienced vulnerability is when consumers feel and identify that they are vulnerable. Observed vulnerability refers to situations where a third party detects the vulnerability, regardless of whether the consumer feels at risk. Building on Hill and Sharma’s (2020) typology, our framework distinguishes between experienced vulnerability (as perceived by the consumer) and observed vulnerability (as inferred from the context or third-party observation). The dependent variables in our study, whether consumers report experiencing bias or racism and how they respond behaviorally, are the outcomes of these types of vulnerability.

Hill and Sharma (2020) define consumer vulnerability as “a state in which consumers are subject to harm because their access to and control over resources are restricted in ways that significantly inhibit their ability to function in the marketplace” (p. 551), and they argue that consumer vulnerability is often influenced by power asymmetries between individuals and institutions. They posit that consumer vulnerability is a lack of consumer agency, where consumers’ ability to make decisions without harm can be impacted by personal, social, and institutional factors. Taken together, this integrative view of consumer vulnerability asserts that some consumers are more susceptible to exploitation and abuse because they fall into certain demographic groups, while other consumers’ vulnerability is precipitated by situational or institutional factors which leave them feeling powerless. This conceptual framework allows us to examine who experiences financial discrimination and how individuals react to these experiences. Based on the aforementioned conceptual framework and the relevant literature, the following hypotheses are proposed:

H₁: 

Consumers with demographic characteristics that are historically considered to be vulnerable (e.g., racial/ethnic minority status, women) are more likely to report experiencing bias or racism from financial institutions.

H₁a: 

When compared to Black consumers, non-Black consumers will be less likely to report experiencing bias or racism from financial institutions.

H₁b: 

When compared to women consumers, men consumers will be less likely to report experiencing bias or racism from financial institutions.

H₂: 

Race will be associated with the reactions related to experiencing bias or racism from financial institutions.

H₃: 

Ethnicity will be associated with the reactions related to experiencing bias or racism from financial institutions.

H₄: 

Gender will be associated with the reactions related to experiencing bias or racism from financial institutions.

4. Data and Methodology

4.1. Data

This study relied on secondary data which was originally collected in the United States from 17 November 2021, to 15 December 2021. A multidisciplinary and multi-institution team of ten researchers developed and pilot-tested the survey which was intended to capture the financial and health implications of the COVID-19 pandemic. The team that originally collected the data used a third-party survey company, Qualtrics, which recruited respondents and administered the survey online. Qualtrics was instructed to oversample by race and income. While about half of the 3598 survey participants were White, those identifying as Black, Indigenous, and People of Color (BIPOC) were oversampled. The survey included a variety of questions on aspects of financial behavior and decision-making topics such as financial capability, financial stress, economic resources and assets, and financial experiences.

In this study, observations with missing values on any of the variables used in the analysis were removed from the final sample. As a result, the sample size went from 3598 to 3290. This approach was deemed appropriate because of the relatively low proportion of missing data. No imputation procedures were employed.

4.2. Dependent Variables

The main dependent variable for this study is based on the following question: (1) Do you feel as if you have experienced bias or racism when working with a financial institution? The responses were coded as 1 if respondents chose “yes” and 0 if respondents chose “no”. If the participants answered “yes”, then the following question was asked: (2) As a direct result of the bias or racism experience you had, which, if any, of the following things happened? The possible responses included the following: (1) I lost trust in that financial services company; (2) I lost trust in the financial services industry; (3) I started using non-traditional financial services companies for certain financial transactions (e.g., payday loans, retail stores); (4) I switched financial services companies; (5) I told my friends and family about the experience; (6) I started using alternative investment tools like cryptocurrency to avoid dealing with financial services companies; and (7) I did not do any of the following actions. Respondents were only able to choose one of the seven responses. Each behavior was examined separately. Due to the limited sample size, the following responses were combined: (3) I started using non-traditional financial services companies for certain financial transactions (payday loans, retail stores) and (6) I started using alternative investment tools like cryptocurrency to avoid dealing with financial services companies. Response (7) was not examined. Therefore, five separate reactions served as the dependent variable for the remaining models.

4.3. Independent Variables

In this study, the independent variables include race, ethnicity, gender, age, educational attainment, marital status, employment status, income, home ownership, objective financial knowledge, and subjective financial knowledge.

In the survey, race was categorized as follows: (1) American Indian or Alaska Native; (2) Black or African American; (3) Asian; (4) Native Hawaiian or Other Pacific Islander; (5) White; (6) More than one race; (7) Other; (8) Refused to answer; (9) Do not know. Given the small number of observations for some groups, they were combined. Specifically, groups (1) and (4) were combined into a new group we called Native/Pacific Islander, and groups (6) and (7) were combined into a new group we called Multiracial/Other, for a total of five racial groups. Ethnicity, a separate variable from race, was categorized as “Hispanic or Latino” or “Not Hispanic or Latino”. If respondents answered, “refused to answer” and “don’t know”, the observations were coded as missing values.

Gender was coded as a dichotomous variable that took a value of 1 if respondents identified as a woman and 0 if respondents identified as a man. Due to the very small sample size, other genders or those who answered “prefer not to answer” were dropped. Age comprised six groups: 18–24 years old, 25–34 years old, 35–44 years old, 45–54 years old, 55–64 years old, and 65 years old or older. Educational attainment was categorized as follows: high school or less, some college or an associate degree, bachelor’s degree, and graduate degree. Marital status comprised three groups: married, widowed/divorced/separated, and single. Employment status included employed (full-time, part-time, and self-employed), student (full-time and part-time), unemployed, and other. Income included seven groups: less than USD 15,000, USD 15,000 to USD 24,999, USD 25,000 to USD 34,999, USD 35,000 to USD 49,999, USD 50,000 to USD 74,999, USD 75,000 to USD 99,999, and USD 100,000 to USD 150,000. Those who answered, “Don’t know” and “Prefer not to say” were coded as missing values. Homeownership was a dichotomous variable that took a value of 1 if respondents owned their home and 0 if respondents did not own their home.

Objective financial knowledge was measured by using the so-called Big Three financial literacy questions (Lusardi & Mitchell, 2011). For each question a respondent answered correctly, one point was given, resulting in a minimum possible score of 0 and a maximum possible score of 3. “Don’t know” responses were counted as incorrect and “refuse to answer” responses were coded as missing values. Subjective financial knowledge was determined by the following question: “How knowledgeable do you consider yourself when it comes to financial products and services?” The responses included (1) no knowledge of financial services and products; (2) basic knowledge of financial services and products; (3) good knowledge of financial services and products; and (4) very good knowledge of financial services and products.

4.4. Methodology

This study used logistic regression via maximum likelihood as follows:

Model 1:

Pr (Y = 1) = F (b₀ + b₁ independent variables)

Where Y represents the respondents who have experienced bias or racism when working with a financial institution.

Model 2:

Pr (Lost trust in that financial services company = 1) = F (b₀ + b₁ independent variables)

Pr (Lost trust in the financial services industry = 1) = F (b₀ + b₁ independent variables)

Pr (Used non-traditional services or alternative tools = 1) = F (b₀ + b₁ independent variables)

Pr (Switched financial services companies = 1) = F (b₀ + b₁ independent variables)

Pr (Told friends and family about the experience = 1) = F (b₀ + b₁ independent variables)

Regarding Model 2, as this study primarily explored the determinants of each reaction, separate logistic regression models were developed to examine the associations. To address the concern of possible selection bias, this study considered the Heckman selection model. Across all five outcomes in Model 2, the selection models yield a non-significant likelihood-ratio test and unstable rho estimates, indicating that selection bias is not present. Furthermore, when compared with the Heckman selection model, the logistic regression model offered a substantially better fit, and a lower AIC and BIC. Therefore, logistic regression is more appropriate in this study. To test multicollinearity, this study computed variance inflation factors (VIFs = 1.17) among all the explanatory variables, suggesting that multicollinearity was not a concern.

5. Results

5.1. Descriptive Results

Table 1. Descriptive statistics with comparison by race.

	Full Sample (N = 3290)		Black (N = 729)		White (N = 1714)		Asian (N = 314)		Multiracial/Other (N = 355)		Native/Pacific Islander (N = 178)
	Prop.	Std. Dev.	Prop.	Std. Dev.	Prop.	Std. Dev.	Prop.	Std. Dev.	Prop.	Std. Dev.	Prop.	Std. Dev.
Experienced bias	0.21	0.41	0.32	0.47	0.15	0.36	0.22	0.41	0.27	0.44	0.25	0.44
Hispanic	0.20	0.40	0.10	0.30	0.20	0.40	0.03	0.17	0.54	0.50	0.17	0.38
Women	0.50	0.50	0.49	0.50	0.50	0.50	0.44	0.50	0.57	0.50	0.48	0.50
Age (in years)
18–24	0.17	0.37	0.25	0.43	0.11	0.31	0.24	0.43	0.23	0.42	0.13	0.34
25–34	0.31	0.46	0.38	0.49	0.27	0.44	0.31	0.46	0.39	0.49	0.28	0.45
35–44	0.24	0.43	0.23	0.42	0.25	0.43	0.25	0.43	0.22	0.41	0.29	0.45
45–54	0.13	0.33	0.08	0.28	0.15	0.36	0.11	0.32	0.09	0.28	0.19	0.39
55–64	0.08	0.26	0.05	0.21	0.10	0.30	0.03	0.18	0.05	0.21	0.10	0.29
65+	0.08	0.27	0.02	0.13	0.12	0.33	0.06	0.23	0.04	0.19	0.02	0.13
Education
High school or less	0.33	0.47	0.39	0.49	0.33	0.47	0.23	0.42	0.34	0.47	0.35	0.48
Some college	0.40	0.49	0.41	0.49	0.39	0.49	0.30	0.46	0.42	0.49	0.49	0.50
Bachelor’s degree	0.19	0.39	0.15	0.36	0.19	0.39	0.34	0.47	0.17	0.38	0.12	0.32
Graduate degree	0.08	0.27	0.05	0.21	0.09	0.28	0.14	0.34	0.08	0.27	0.04	0.19
Marital Status
Married	0.43	0.50	0.28	0.45	0.51	0.50	0.43	0.50	0.40	0.49	0.38	0.49
Widowed/Divorced/Separated	0.14	0.35	0.08	0.27	0.18	0.39	0.06	0.24	0.11	0.31	0.17	0.38
Single	0.43	0.49	0.64	0.48	0.31	0.46	0.51	0.50	0.50	0.50	0.45	0.50
Employment Status
Employed	0.61	0.49	0.68	0.47	0.58	0.49	0.57	0.50	0.64	0.48	0.57	0.50
Student	0.07	0.25	0.08	0.28	0.04	0.20	0.14	0.35	0.08	0.28	0.07	0.26
Unemployed	0.24	0.43	0.21	0.41	0.26	0.44	0.23	0.42	0.20	0.40	0.24	0.43
Other	0.08	0.28	0.03	0.17	0.11	0.31	0.05	0.23	0.07	0.26	0.12	0.32
Annual Income
Less than USD 15,000	0.18	0.38	0.22	0.42	0.15	0.35	0.16	0.36	0.20	0.40	0.29	0.45
USD 15,000 to USD 24,999	0.15	0.36	0.16	0.37	0.15	0.36	0.12	0.33	0.17	0.38	0.17	0.38
USD 25,000 to USD 34,999	0.16	0.36	0.17	0.38	0.16	0.36	0.11	0.32	0.17	0.38	0.11	0.32
USD 35,000 to USD 49,999	0.20	0.40	0.18	0.38	0.17	0.38	0.31	0.46	0.23	0.42	0.30	0.46
USD 50,000 to USD 74,999	0.11	0.31	0.10	0.30	0.12	0.33	0.06	0.24	0.10	0.31	0.04	0.19
USD 75,000 to USD 99,999	0.10	0.30	0.09	0.29	0.12	0.32	0.12	0.33	0.06	0.24	0.05	0.22
USD 100,000 to USD 150,000	0.10	0.31	0.07	0.26	0.13	0.34	0.11	0.32	0.07	0.25	0.03	0.18
Homeowner	0.40	0.49	0.31	0.46	0.48	0.50	0.34	0.48	0.31	0.46	0.31	0.47
Objective financial knowledge	1.34	1.01	1.07	0.89	1.43	1.03	1.61	1.05	1.26	0.97	1.27	0.99
Subjective financial knowledge
No knowledge	0.11	0.31	0.09	0.29	0.11	0.31	0.08	0.27	0.12	0.33	0.19	0.39
Basic knowledge	0.49	0.50	0.43	0.50	0.49	0.50	0.54	0.50	0.51	0.50	0.49	0.50
Good knowledge	0.31	0.46	0.35	0.48	0.30	0.46	0.31	0.46	0.28	0.45	0.24	0.43
Very good knowledge	0.10	0.30	0.13	0.33	0.10	0.29	0.08	0.27	0.08	0.28	0.07	0.26

shows the descriptive statistics of the full sample by race. The majority of respondents were White (52%), followed by Black (22%), more than one race or other (11%), Asian (9%), and Native/Pacific Islander (5%). About 20% of respondents identified as Hispanic and 50% of respondents were women. About 21% of the total sample reported experiencing bias or racism when working with a financial institution. Most respondents were between 18 and 44 years old (72%). In addition, about 40% of respondents had some college and 27% had at least a bachelor’s degree. About 43% of respondents were married, and 43% of respondents were single. In the full sample, 61% of respondents were employed, and 24% of respondents were unemployed. A large percentage of respondents (69%) had incomes of USD 50,000 annually or less. About 40% of the respondents owned their home. Also, the respondents’ objective financial knowledge score averaged 1.34 out of 3. About 49% of respondents reported having basic financial knowledge of financial services and products and 41% reported good or very good knowledge.

Table 2. Descriptive comparison by gender.

	Women (N = 1640)		Men (N = 1650)
	Prop.	Std. Dev.	Prop.	Std. Dev.
Experienced bias	0.20	0.40	0.23	0.42
Hispanic	0.25	0.43	0.15	0.36
Race
Black	0.22	0.41	0.22	0.42
White	0.52	0.50	0.52	0.50
Asian	0.08	0.28	0.11	0.31
Multiracial/Other	0.12	0.33	0.09	0.29
Native/Pacific Islander	0.05	0.22	0.06	0.23
Age (in years)
18–24	0.19	0.39	0.14	0.35
25–34	0.32	0.47	0.30	0.46
35–44	0.24	0.42	0.25	0.44
45–54	0.14	0.35	0.12	0.32
55–64	0.07	0.26	0.08	0.27
65+	0.04	0.20	0.12	0.32
Education
High school or less	0.34	0.47	0.33	0.47
Some college	0.44	0.50	0.36	0.48
Bachelor’s degree	0.16	0.37	0.22	0.41
Graduate degree	0.06	0.24	0.10	0.30
Marital Status
Married	0.44	0.50	0.42	0.49
Widowed/Divorced/Separated	0.16	0.37	0.12	0.33
Single	0.40	0.49	0.46	0.50
Employment Status
Employed	0.56	0.50	0.65	0.48
Student	0.08	0.27	0.06	0.23
Unemployed	0.29	0.45	0.20	0.40
Other	0.08	0.26	0.09	0.29
Annual Income
Less than USD 15,000	0.20	0.40	0.15	0.36
USD 15,000 to USD 24,999	0.17	0.38	0.13	0.34
USD 25,000 to USD 34,999	0.16	0.37	0.15	0.35
USD 35,000 to USD 49,999	0.18	0.39	0.22	0.42
USD 50,000 to USD 74,999	0.11	0.31	0.10	0.30
USD 75,000 to USD 99,999	0.09	0.29	0.11	0.32
USD 100,000 to USD 150,000	0.07	0.26	0.13	0.34
Homeowner	0.37	0.48	0.44	0.50
Objective financial knowledge	1.13	0.90	1.55	1.06
Subjective financial knowledge
No knowledge	0.12	0.33	0.09	0.29
Basic knowledge	0.52	0.50	0.45	0.50
Good knowledge	0.28	0.45	0.34	0.47
Very good knowledge	0.08	0.27	0.11	0.32

shows the descriptive statistics of the sample by gender.

Table 3. Responses after experiencing financial bias or racism with comparison by race and gender.

	Full Sample (N = 701)		Black (N = 233)		White (N = 259)		Asian (N = 69)		Multiracial/Other (N = 95)		Native/Pacific Islander (N = 45)		Women (N = 324)		Men (N = 377)
	%	Std. Dev.	%	Std. Dev.	%	Std. Dev.	%	Std. Dev.	%	Std. Dev.	%	Std. Dev.	%	Std. Dev.	%	Std. Dev.
Lost trust in that financial services company (N = 127)	0.18	0.39	0.22	0.41	0.17	0.38	0.12	0.32	0.14	0.35	0.22	0.42	0.16	0.37	0.20	0.40
Lost trust in financial services industry (N = 120)	0.17	0.38	0.20	0.40	0.15	0.36	0.14	0.35	0.20	0.40	0.11	0.32	0.16	0.37	0.18	0.39
Used non-traditional services and alternative tools (N = 187)	0.27	0.44	0.24	0.43	0.31	0.46	0.32	0.47	0.19	0.39	0.27	0.45	0.25	0.43	0.28	0.45
Switched financial services companies (N = 105)	0.15	0.36	0.14	0.34	0.14	0.35	0.13	0.34	0.22	0.42	0.16	0.37	0.16	0.37	0.14	0.35
Told friends and family (N = 88)	0.13	0.33	0.12	0.33	0.11	0.31	0.16	0.37	0.15	0.37	0.13	0.34	0.16	0.37	0.10	0.30

shows descriptive statistics of responses after experiencing financial bias or racism. About 27% of the respondents who said they had experienced bias or racism while working with a financial institution reported that after the ordeal, they turned to non-traditional financial services companies for certain transactions (e.g., payday loans, retail stores, etc.) or started using alternative investment tools like cryptocurrency to avoid dealing with financial services companies. About 18% of the respondents reported that they lost trust in that financial services company, and 17% reported that they lost trust in the financial services industry altogether. About 15% of respondents reported switching financial services companies and 13% of respondents shared their negative experiences with their families and friends. Eleven percent of the respondents reported that they did not do any of the above following their financial discrimination experience.

5.2. Logistic Regression Results—Experiencing Bias or Racism When Working with a Financial Institution

Table 4. Logistic regression on experiencing bias or racism.

	Est. Coef.	SE	OR
Race (Ref. = Black)
White	−0.8671 ***	0.1140	0.4202
Asian	−0.4663 **	0.1669	0.6273
Multiracial/Other	−0.3586 *	0.1587	0.6987
Native/Pacific Islander	−0.2729	0.1987	0.7612
Women (Ref. = Men)	−0.2764 **	0.0928	0.7585
Hispanic (Ref. = Non-Hispanic)	0.2964 *	0.1163	1.3451
Age (in years) (Ref. = 18–24)
25–34	0.0161	0.1338	1.0162
35–44	−0.0869	0.1474	0.9168
45–54	−0.5459 **	0.1915	0.5793
55–64	−0.4615 *	0.2322	0.6304
65+	−1.8268 ***	0.3898	0.1609
Education (Ref. = High school or less)
Some college	−0.0266	0.1085	0.9737
Bachelor’s degree	0.1810	0.1399	1.1984
Graduate degree	0.5429 **	0.1837	1.7209
Marital Status (Ref. = Married)
Widowed/Divorced/Separated	−0.1078	0.1592	0.8978
Single	−0.2504 *	0.1085	0.7785
Employment Status (Ref. = Employed)
Student	−0.0637	0.1846	0.9382
Unemployed	−0.4484 ***	0.1254	0.6387
Other	−0.4920 *	0.2409	0.6114
Annual Income (Ref. = Less than USD 15,000)
USD 15,000 to USD 24,999	−0.0276	0.1562	0.9728
USD 25,000 to USD 34,999	0.0128	0.1561	1.0129
USD 35,000 to USD 49,999	−0.1978	0.1525	0.8205
USD 50,000 to USD 74,999	−0.3640	0.1887	0.6949
USD 75,000 to USD 99,999	−0.3915 *	0.1942	0.6761
USD 100,000 to USD 150,000	−0.4855 *	0.2050	0.6154
Homeowner	0.2455 *	0.1013	1.2783
Objective financial knowledge	−0.0771	0.0489	0.9258
Subjective financial knowledge (Ref. = No knowledge)
Basic knowledge	−0.1257	0.1541	0.8819
Good knowledge	−0.0491	0.1635	0.9521
Very good knowledge	−0.0250	0.1980	0.9753
Intercept	−0.1118	0.2454	0.8942
Model fit
Pseudo R-squared	0.0695
Concordance rate	68.07%
Hosmer–Lemeshow goodness-of-fit	χ2 = 4.59	p = 0.800

Notes: N = 3290; * p < 0.05, ** p < 0.01, *** p < 0.001.

shows the logistic regression model of experiencing bias or racism when working with a financial institution. Compared with Black respondents, White, Asian, and Multiracial/Other respondents were less likely to report experiencing bias or racism. White respondents, for example, had approximately 58% lower odds of experiencing bias or racism comparative to Black respondents. Women were less likely to report experiencing bias than men. The odds of those identifying as Hispanic that experienced bias or racism were nearly 35% higher than the odds for non-Hispanics. Compared with those aged 18–24 years, respondents older than 44 years were less likely to report experiencing bias when working with a financial institution. Respondents with graduate degrees were more likely to report experiencing bias or racism than those with a high school or less education. Single people were less likely than married people to report the same. The odds of experiencing bias were about 36% lower for the unemployed versus the employed. Compared to those with incomes less than USD 15,000, those with incomes over USD 75,000 per year were less likely to experience bias. Homeowners had about 28% higher odds of reporting experiencing bias or racism than those who do not own their homes.

The model’s R-squared was 0.0695, suggesting that the predictors explained a modest proportion of the outcome. The concordance rate was about 68%, indicating the modest predictive accuracy in identifying customers who are at a greater risk of perceived bias. In addition, the Hosmer–Lemeshow goodness-of-fit indicated no evidence of poor fit (χ² = 4.59, p = 0.800). Furthermore, the ROC curve yielded an area under the curve (AUC) of 0.681, showing that the model had fair discriminative power. In other words, if one customer who experienced bias was randomly selected and another was selected who did not, the model would assign a higher predicted probability of bias to the person who experienced bias.

5.3. Logistic Regression Results—Consumers’ Reactions After Experiencing Financial Discrimination

Reaction #1—Lost Trust in That Financial Services Company

Table 5. Logistic regression on responses after experiencing financial bias or racism.

	Lost Trust Financial Services Company			Lost Trust Financial Services Industry			Used Non-Traditional and Alternatives
	Est. Coef.	SE	OR	Est. Coef.	SE	OR	Est. Coef.	SE	OR
Race (Ref. = Black)
White	−1.1597 ***	0.2298	0.3136	−0.9125 ***	0.2394	0.4015	−0.2662	0.1972	0.7663
Asian	−1.1570 **	0.4026	0.3144	−0.8455 *	0.3730	0.4293	0.1774	0.2787	1.1941
Multiracial/Other	−0.9483 **	0.3470	0.3874	−0.3591	0.3138	0.6983	−0.2800	0.3019	0.7558
Native/Pacific Islander	−0.3657	0.3753	0.6937	−0.9009	0.4923	0.4062	0.1266	0.3446	1.1350
Women (Ref. = Men)	−0.3924 *	0.1965	0.6754	−0.3079	0.1990	0.7350	−0.4061 *	0.1623	0.6663
Hispanic (Ref. = non-Hispanic)	0.4186	0.2409	1.5199	0.1685	0.2499	1.1835	0.0543	0.2037	1.0558
Age (in years) (Ref. = 18–24)
25–34	0.7607 *	0.3296	2.1397	0.2178	0.2671	1.2433	−0.1464	0.2219	0.8638
35–44	0.5322	0.3531	1.7026	−0.2010	0.3209	0.8179	−0.2925	0.2475	0.7464
45–54	−0.0609	0.4434	0.9409	−0.4955	0.4458	0.6093	−0.4280	0.3187	0.6518
55–64	−0.1822	0.5243	0.8334	−0.0457	0.5148	0.9553	−1.4991	0.5596	0.2233
65+	−2.1021	1.0919	0.1222	−0.9781	0.8401	0.3760	-	-	-
Education (Ref. = High school or less)
Some college	0.0158	0.2309	1.0159	−0.1821	0.2348	0.8335	−0.2452	0.1891	0.7825
Bachelor’s degree	0.1782	0.2898	1.1951	0.3439	0.2824	1.4104	−0.2981	0.2513	0.7422
Graduate degree	0.4524	0.3701	1.5720	0.6249	0.3764	1.8681	0.4573	0.2873	1.5798
Marital Status (Ref. = Married)
Widowed/Divorced/Separated	0.4906	0.2968	1.6333	−1.0469 *	0.4903	0.3510	0.2326	0.2750	1.2619
Single	−0.1923	0.2258	0.8251	−0.0510	0.2243	0.9503	−0.1132	0.1907	0.8929
Employment Status (Ref. = Employed)
Student	−0.5501	0.5444	0.5769	−0.0281	0.3748	0.9723	−0.2087	0.3227	0.8116
Unemployed	0.1363	0.2494	1.1460	−0.9131 **	0.3109	0.4013	−0.4170	0.2310	0.6590
Other	−0.6526	0.6239	0.5207	−0.6339	0.5784	0.5305	−0.4213	0.4869	0.6562
Annual Income (Ref. = Less than USD 15,000)
USD 15,000 to USD 24,999	0.5404	0.3166	1.7167	0.2530	0.3376	1.2879	−0.4524	0.2878	0.6361
USD 25,000 to USD 34,999	−0.0221	0.3487	0.9781	0.3691	0.3294	1.4464	−0.3265	0.2775	0.7215
USD 35,000 to USD 49,999	−0.1274	0.3387	0.8804	0.1999	0.3317	1.2213	−0.3404	0.2603	0.7115
USD 50,000 to USD 74,999	−0.1067	0.3957	0.8988	−0.1070	0.4104	0.8985	−0.1561	0.3094	0.8555
USD 75,000 to USD 99,999	−0.2731	0.4214	0.7610	−1.0683 *	0.5448	0.3436	−0.1339	0.3178	0.8746
USD 100,000 to USD 150,000	−0.4616	0.4473	0.6303	−0.3393	0.4627	0.7123	−0.2842	0.3358	0.7526
Homeowner	0.4623 *	0.2063	1.5877	0.1445	0.2194	1.1555	0.3135	0.1743	1.3681
Objective financial knowledge	0.2946 **	0.1019	1.3426	−0.0082	0.1064	0.9919	−0.3332 ***	0.0877	0.7166
Subjective financial knowledge (Ref. = No knowledge)
Basic knowledge	−0.6081 *	0.3049	0.5444	−0.4271	0.2816	0.6524	0.6113	0.3346	1.8429
Good knowledge	−0.2488	0.3127	0.7797	−1.0224 **	0.3262	0.3597	0.6242	0.3479	1.8668
Very good knowledge	−0.1674	0.3806	0.8459	−1.2208 **	0.4489	0.2950	1.0986 **	0.3741	3.0000
Intercept	−3.0194 ***	0.5290	0.0488	−1.7894 ***	0.4884	0.1671	−2.0633 ***	0.4614	0.1270
Model fit
Pseudo R-squared	0.0850			0.0884			0.0557
Concordance rate	73.55%			73.79%			68.72%
Hosmer–Lemeshow goodness-of-fit	χ2 = 5.97	p = 0.6508		χ2 = 4.51	p = 0.8082		χ2 = 12.21	p = 0.1421
	Switched Financial Services Companies			Told Friends/Family
	Est. Coef.	SE	OR	Est. Coef.	SE	OR
Race (Ref. = Black)
White	−0.7222 **	0.2657	0.4857	−0.7727 **	0.2919	0.4618
Asian	−0.4979	0.3984	0.6078	−0.1039	0.3812	0.9013
Multiracial/Other	0.2860	0.3191	1.3311	−0.0894	0.3604	0.9145
Native/Pacific Islander	−0.0379	0.4385	0.9628	−0.1519	0.4733	0.8591
Women (Ref. = Men)	0.0237	0.2095	1.0240	0.2858	0.2310	1.3308
Hispanic (Ref. = non-Hispanic)	0.1206	0.2572	1.1282	0.4196	0.2772	1.5214
Age (in years) (Ref. = 18–24)
25–34	−0.0742	0.3045	0.9285	−0.2818	0.3224	0.7544
35–44	−0.1654	0.3383	0.8476	0.1283	0.3388	1.1368
45–54	−0.7350	0.4695	0.4795	−0.3534	0.4853	0.7023
55–64	−0.2793	0.5088	0.7563	0.2072	0.5327	1.2302
65+	−1.3912	0.8387	0.2488	−1.4750	1.1079	0.2288
Education (Ref. = High school or less)
Some college	0.1665	0.2564	1.1812	0.2497	0.2738	1.2837
Bachelor’s degree	0.3749	0.3054	1.4548	0.4311	0.3520	1.5389
Graduate degree	−0.1831	0.4839	0.8327	0.7455	0.4422	2.1075
Marital Status (Ref. = Married)
Widowed/Divorced/Separated	0.2880	0.3325	1.3337	−1.1850 *	0.5481	0.3058
Single	−0.1860	0.2476	0.8303	−0.2304	0.2629	0.7942
Employment Status (Ref. = Employed)
Student	0.1129	0.4043	1.1195	0.3862	0.3738	1.4714
Unemployed	−0.6132	0.3325	0.5416	−0.5105	0.3267	0.6002
Other	−0.4158	0.5683	0.6598	0.0658	0.5202	1.0680
Annual Income (Ref. = Less than USD 15,000)
USD 15,000 to USD 24,999	0.5430	0.4386	1.7212	−0.5870	0.3709	0.5560
USD 25,000 to USD 34,999	0.6932	0.4262	2.0001	−0.3686	0.3498	0.6917
USD 35,000 to USD 49,999	0.7291	0.4114	2.0733	−0.7871 *	0.3657	0.4552
USD 50,000 to USD 74,999	0.5464	0.4746	1.7271	−1.2867 *	0.5279	0.2762
USD 75,000 to USD 99,999	1.0680 *	0.4516	2.9095	−0.8961	0.4663	0.4081
USD 100,000 to USD 150,000	0.0725	0.5553	1.0752	−0.7727	0.4742	0.4618
Homeowner	0.2229	0.2246	1.2497	−0.0771	0.2564	0.9258
Objective financial knowledge	0.0523	0.1101	1.0537	−0.0663	0.1224	0.9358
Subjective financial knowledge (Ref. = No knowledge)
Basic knowledge	−0.0028	0.3978	0.9972	0.2353	0.4253	1.2653
Good knowledge	0.2606	0.4080	1.2978	0.6079	0.4377	1.8367
Very good knowledge	−0.0646	0.4993	0.9374	0.0924	0.5440	1.0968
Intercept	−3.7066 ***	0.6247	0.0246	−3.0009 ***	0.6071	0.0497
Model fit
Pseudo R-squared	0.0597			0.0646
Concordance rate	70.61%			71.63%
Hosmer–Lemeshow goodness-of-fit	χ2 = 10.32	p = 0.2436		χ2 = 5.69	p = 0.6815

Notes: N = 3290; * p < 0.05, ** p < 0.01, *** p < 0.001.

shows the separate logistic regression for the post-financial discrimination reactions. Respondents identifying as Asian, White, and Multiracial/Other were less likely than those identifying as Black to report losing trust in the financial services company where the bias or racism occurred. The odds for White consumers were about 69% lower than the odds for Black consumers. Women were less likely than men to lose trust in the financial services company. Those aged 25–34 years old were more likely than the 18–24 years old group to lose trust in the financial services company. The odds for homeowners were about 59% higher than the odds for those who did not own a home. Those with higher objective financial knowledge were more likely to lose trust in the financial services company. Compared with those with no subjective financial knowledge, those with a basic knowledge were less likely to lose trust in the financial services company.

Reaction #2—Lost Trust in the Financial Services Industry

Rather than losing trust in a specific company, some respondents reported losing trust in the financial services industry altogether. There were differences by race as shown in Table 5. Asian and White respondents were less likely than Black respondents to report losing trust in the financial services industry. The odds for those identifying as White were about 60% lower than the odds for those identifying as Black. Widowed, divorced, or separated respondents were less likely than married people to lose trust in the financial services industry. Compared to those who were employed, unemployed respondents were less likely to lose trust in the financial services industry after experiencing financial discrimination. In addition, compared to respondents with incomes of less than USD 15,000 per year, those with incomes between USD 75,000 and USD 99,999 were less likely to lose trust in the financial services industry. Furthermore, the odds that those with good subjective financial knowledge reported losing trust in the financial services industry were about 64% lower than those with no financial knowledge. Also, those with very good subjective financial knowledge were less likely to lose trust in the financial services industry when compared to those with no knowledge.

Reaction #3—Started Using Non-Traditional Services or Alternative Tools

As a result of experiencing bias or racism when working with a financial services institution, some respondents started using non-traditional financial services and products like payday loans, retail stores, and alternative investment tools like cryptocurrency. There were no significant differences by race or ethnicity. However, women were less likely than men to turn to these non-traditional services and alternative tools. Objective financial knowledge was negatively associated with responding to financial discrimination by turning to non-traditional and alternative financial sources. Those with very good subjective financial knowledge had three times the odds of reporting using non-traditional services or alternative tools than those with no financial knowledge.

Reaction #4—Switched Financial Services Companies

In response to experiencing bias or racism when working with a financial services institution, some respondents switched financial companies as shown in Table 5. The odds that White respondents reported carrying out this change were about 51% lower than the odds for Black respondents. Those with incomes between USD 75,000 and USD 100,000 were more likely to switch financial services companies than those with incomes less than USD 15,000.

Reaction #5—Told Friends and Family

Some respondents told friends and family about their experience. Compared with Black respondents, White respondents were less likely to share their discriminatory experiences with families or friends. Widowed, divorced, or separated respondents were less likely than married respondents to tell their friends and families about the experience. In addition, those with incomes between USD 35,000 and USD 75,000 were less likely to tell their friends and families than those with incomes less than USD 15,000.

5.4. Model Comparison

To ensure the validity of our results, we tested for potential selection bias using Heckman selection models (

Table 6. Model Comparison.

	Lost Trust Financial Services Company	Lost Trust Financial Services Industry	Used Non-Traditional and Alternatives	Switched Financial Services Companies	Told Friends/Family
Logistic Regression
AIC	1046.244	1001.130	1385.981	936.495	820.5894
BIC	1235.302	1190.187	1566.490	1125.553	1009.647
Selection Model
AIC	3893.878	3889.996	4048.04	3852.553	3792.682
BIC	4284.191	4274.21	4432.26	4236.768	4176.896
Likelihood-Ratio Test of independent equations	χ2(1) = 0.06, p = 0.80	χ2(1) = 1.10, p = 0.2952	χ2(1) = 0.00, p = 0.9794	χ2(1) = 0.06, p = 0.8037	χ2(1) = 0.17, p = 0.6792
Estimated ρ (rho) from Selection Model	−0.093	0.996	0.0352	0.999	0.995

). The logistic regression models demonstrated better fit with lower AIC and BIC values across all specifications, and likelihood ratio tests were non-significant (p > 0.05). Based on these model comparisons, the logistic regression approach was appropriate for our analysis.

6. Discussion

The purpose of this study was twofold. First, we sought to investigate which consumers were more likely to experience financial bias or racism (i.e., financial discrimination) when working with a financial institution. Second, we examined consumers’ reactions following their negative experiences and the factors associated with them. Our findings show that Black consumers were more likely than all other racial groups, except Native/Pacific Islander consumers, to experience financial discrimination. There was no significant difference found when comparing Black respondents to Native/Pacific Islander respondents. The finding supports Hypothesis 1a and is in line with previous research which shows that Black consumers are more likely than others to have adverse experiences when navigating the financial services landscape (Bates & Robb, 2013; Botein, 2013). Similarly, respondents who identified as Hispanic were more likely to report experiencing financial discrimination compared to non-Hispanic respondents. This finding is also in line with previous studies conducted in this area (Botein, 2013). Surprisingly, women were less likely than men to report financial discrimination and therefore, Hypothesis 1b is not supported. Previous research on this topic has generally found that women, not men, are more vulnerable to bias when working with financial institutions (Aristei & Gallo, 2016; Chundakkadan, 2023; de Andrés et al., 2021).

There are possible explanations for this surprising gender-based result. One possible explanation is that this may signal increased feelings of empowerment among women, which is in line with Crockett et al. (2003), who found an inverse relationship between feelings of empowerment and perceived bias in the marketplace. Another explanation may be intersectionality. Our data consisted of a race/ethnicity oversample. Although there are fewer cues provided in the literature about how women of color recognize and respond to various forms of bias and discrimination, the concept of intersectionality suggests that women of color in particular may normalize discrimination and lower their expectations of fair treatment (Chaney et al., 2021; Cole, 2020; Remedios & Snyder, 2015). Finally, another explanation could be rooted in financial socialization. Research shows that early in life, both gender and race have a role in shaping financial messaging and financial behaviors (Wiley, 2024). As there is evidence suggesting that parents discuss money with sons earlier than with daughters (LeBaron & Kelley, 2021), women’s expectations when in professional financial environments may be lower than men’s as well as their perceptions of bias.

Regarding the paper’s second objective, the findings show that most people in the sample turned to non-traditional financial services after experiencing financial discrimination. Further, various findings related to a respondent’s race were uncovered. When considering who loses trust in financial services companies as a result of financial discrimination, we found that Asian, White, and Multiracial/Other respondents were less likely than Black respondents to report this reaction. Similarly, Asian and White respondents were less likely than Black respondents to report that they lost trust in the financial services industry altogether after experiencing financial discrimination. White respondents were less likely than Black respondents to either switch financial companies or tell friends and family after experiencing financial bias or racism. There were no statistically significant racial differences on the responses of switching to non-traditional financial services and alternative investment tools. It is notable that being Hispanic was not statistically significantly associated with any of the five reactions reported. Therefore, Hypothesis 2 was partially supported while Hypothesis 3 was not supported.

Gender played a role in the results of two of the reactions: lost trust in that particular financial services company where the bias happened and switching to non-traditional services and using alternative investment tools. Specifically, women were less likely than men to lose trust in the financial services company with which they had the negative experience. Women were also less likely than men to turn to non-traditional services and alternative tools like payday loans, retail stores, and cryptocurrency after experiencing bias or racism. There is evidence, in general, to support that women are less likely than men to invest in cryptocurrency (Alonso et al., 2023) and take on payday loans (Caplan et al., 2017). Hypothesis 4 was partially supported.

7. Conclusions

7.1. Academic and Practical Contributions

This study adds new knowledge to the literature by highlighting some specific reactions that people have after experiencing discrimination using a unique sample. Consumers lost trust in financial companies and financial institutions, switched financial companies, turned to non-traditional alternative products and services, and reported their experiences to friends and family. The findings carry implications for financial institutions, consumers, policymakers, and consumer advocacy groups.

The racial disparities in reported experiences of bias and racism suggest a need for enhanced oversight and regulation that encourages fair treatment for all consumers. Financial institutions should adopt stronger anti-discrimination policies. The Civil Rights Act of 1991 amended the 1964 Act to strengthen employee protection against employment discrimination on the basis of race, color, religion, sex, national origin, age, or disability (U.S. Equal Employment Opportunity Commission, n.d.). Financial institutions can extend these protections to include the way their employees interact with consumers. Financial institutions should also review their policies, procedures, and technologies to make sure internal practices do not disadvantage certain consumers in their transactions with the institution. These actions may help institutions mitigate the risks of jeopardizing profits from lost business, liquidity from lost deposits, and being exposed legally due to employees’ discriminatory behaviors (Li & Zhou, 2024; Scott et al., 2024).

The decline in trust among individuals who experience bias poses a significant challenge for the financial industry. The tendency of individuals who experience bias and racism to seek non-traditional services or lose confidence in financial institutions and the financial industry highlights the potential economic consequences of such biases. As trust weakens, individuals may turn to non-traditional financial services, such as cryptocurrency and alternative investment platforms, which typically offer fewer protections for consumers (Barcellos & Zamarro, 2021; Federal Reserve Bank of Kansas City, 2010). This shift could exacerbate financial instability for marginalized groups. To address this, financial institutions, regulators, and consumer advocacy groups within the financial industry should make transparency, accountability, and outreach a priority to restore trust. For example, publicly disclosing consumer complaints about discrimination and unfair treatment has proved to be successful in improving the quality of customer service and reducing outflows of deposits associated with discrimination (Li & Zhou, 2024).

Experiencing bias and discrimination also emphasizes the importance of financial education initiatives. For example, individuals with more financial knowledge were less likely to lose trust in the financial services industry. The expansion of financial education programs could empower consumers to make more informed financial decisions and better navigate discriminatory mistreatment.

7.2. Conclusions

The purpose of this study was to investigate the factors associated with reporting experiences of financial bias or racism and the factors which are predictors of the reactions taken after consumers have experienced such an event. There was an explicit aim to examine the role that race, ethnicity, and gender play within an American context. The results suggested that no other racial group was more likely than Black consumers to report having experienced financial discrimination. In addition, race was a significant predictor in four of the five reactions and the most persistent difference in reactions was between Black and White consumers. Women, on the other hand, were less likely than men to report financial discrimination and gender was a factor in determining two of the reactions. For both reactions, women were less likely than men to respond in that way. Being Hispanic was not significantly associated with any of the reactions. These results highlight the persistent racial economic disparities experienced among consumers in the United States. There is a cost to discriminating against consumers—both to the consumers and to businesses. Future studies could investigate what these costs are and how they affect individuals and firms in the long term. In addition, it would be insightful to learn more about the companies at which consumers have these experiences such as the firm’s size, success metrics, geographical location, financial position, composition of leadership, and initiatives for diversity and inclusion. Moreover, future research should incorporate qualitative research design to understand more about individuals’ experiences with financial discrimination and longitudinal data to learn how reactions to discrimination might evolve over time.

7.3. Limitations

While this study makes a contribution to the literature by exploring financial discrimination and the reactions that follow, there are some limitations. First, this sample is composed primarily of those from low to moderate income levels. Second, this sample is relatively young, with less than 30% of the respondents being 45 years old or older. Therefore, the findings are likely not generalizable to other groups, particularly wealthier and older consumers. The survey did not include a weight variable which would have allowed for a correction of sampling biases. Finally, the data for the survey used in this research was collected during the time in which COVID-19 was active between November and December of 2021. There is evidence to suggest that incidences of bias, discrimination, or racism, and perceptions thereof, were heightened during the COVID-19 pandemic, especially among those with Asian heritage (Darling-Hammond et al., 2020; Koo et al., 2023; McGarity-Palmer et al., 2024; Roberto et al., 2020). Much research conducted on this topic has dealt with everyday discrimination and disparate pandemic-related health outcomes for people of color (e.g., V. T. Park et al., 2022). While it is unclear exactly how bias or racism were working within financial institutions during that time, it is imaginable that the traditional American mechanisms of discrimination were lessened in some instances and increased in others, depending on the circumstances. For example, some researchers posited that George Floyd’s murder at the hands of Minnesota police officers in 2020 and the social movements which followed for years after heightened the public’s attention and sympathy towards racial inequities (e.g., Primbs et al., 2024); as a result, they hypothesized that discrimination would be less prevalent for Black men and women during the early COVID-19 pandemic (Chavez et al., 2022). Other studies uncovered that Black entrepreneurs had improved loan (Garcia & Ortega, 2024) and crowdfunding outcomes following racial protests (Koh et al., 2023). On the other hand, given the same set of facts, consumers may have simultaneously experienced a decreased sense of trust and faith in institutions during this time.