Social Equity in Credit Card Approval Analysis

Credit Card Eligibility: An Analysis

How do factors of social status affect credit card eligibilty?

Overview of Analysis

Enhance Transparency

we aim to provide applicants with insights to improve their chances and increase transparency in the credit system

Identify Financial Discrimination

Using statistical analysis, we highlight trends and patterns that suggest biases in the approval process

Credit Approval Systems for Inclusivity

We provide recommendations for policy changes that could lead to more inclusive credit approval practices

Analyzing Credit Approval through lenses of Marxist and Feminist

Reveals how systemic biases disadvantage women and economically marginalized groups.

In today's world, credit cards have become an integral part of everyday life, particularly in the USA where they are widely used for various transactions. Credit cards commonly offer attractive benefits such as access to lounges and travel credits. Nonetheless, the way banks approve credit card applications is often described as a great mystery. It is not uncommon for people with poor financial profiles to be approved for credit cards while those with better qualifications are denied. Utilizing a dataset from Kaggle called the “Credit Card Eligibility Data: Determining Factors” compiled by Rohit Sharma, this project aims to reveal some of the mysteries behind the credit card approval process. Our focus is on identifying the factors that financial institutions use in making decisions regarding credit card applications. To achieve this, we perform data preprocessing, exploratory data analysis, and model building, as well as refinement, to gain deeper insights into how credit cards are approved. By identifying the key variables that influence approval decisions, we hope to demystify the process and provide clarity to applicants. Moreover, this project intends to increase transparency in the credit card approval process through actionable recommendations for potential customers on how they can boost their chances of getting a credit card approved.

As we embarked on this analysis, our initial findings revealed significant trends that underscore the complexity of credit card approvals. The data highlights how certain demographic and socioeconomic factors, such as age, employment status, and residential stability, interplay with financial behaviors to influence approval rates. For example, younger applicants or those in less stable jobs faced higher rejection rates, suggesting that financial institutions may favor long-term financial stability over immediate income potential.

Additionally, residential stability emerged as a surprising factor, where individuals who frequently moved had lower approval rates, possibly due to perceived risks associated with financial responsibility. These trends illuminate not only the overt criteria used by banks but also the subtle and hidden factors that shape credit card decisions. By uncovering these layers, our project aims to demystify the approval process and provide clearer guidance on how individuals can navigate these systemic barriers.

Our initial exploratory data analysis reveals compelling trends across various demographic and socioeconomic attributes that correlate with credit card approval rates. The data displays that gender, car ownership, and education play significant roles in the likelihood of approval. For instance, the dataset shows that among applicants, males have a higher approval rate, with approximately 4,000 approvals compared to around 1,500 for females. This disparity suggests a potential gender bias in the credit card approval process that merits further investigation. Car ownership is also another strong indicator of approval likelihood. The graph indicates that individuals who own a car are more likely to be approved, with over 4,000 car owners receiving credit cards versus about 2,000 non-owners. This could imply that owning a car is perceived by financial institutions as a marker of financial stability, which may put applicants who cannot afford such assets at a disadvantage. These insights indicate systemic biases that could potentially exacerbate existing social inequalities. Historically marginalized groups, such as women and low-income individuals, are seemingly at a disadvantage when it comes to credit card approval. This trend points to deeper issues in the credit system where perceived metrics of financial stability might disproportionately affect certain groups negatively. This suggests the need for a critical reevaluation of the criteria used for assessing creditworthiness.

To further understand the implications of these biases, watch Cathy O'Neil's TED Talk titled "The era of blind faith in big data must end," where she explores the hidden dangers of algorithms and their impact on society.

The TED Talk "The era of blind faith in big data must end" by Cathy O'Neil, delivered in April 2017, explores the hidden dangers of algorithms in our society. With 1,685,177 views, O'Neil discusses how algorithms, which are often perceived as objective and scientific, can have secret, harmful impacts. She coins the term "weapons of math destruction" to describe algorithms that decide who gets loans, job interviews, and insurance, often reinforcing unfair biases without any system of appeal. O'Neil emphasizes that algorithms are opinions embedded in code and stresses the importance of understanding and questioning the hidden agendas behind these formulas to avoid blind faith in big data.

Building upon these insights from the visualizations, it becomes essential to delve deeper into the theoretical frameworks of Feminism and Marxism to understand the broader implications of these biases. Feminism advocating gender equality provides a perspective to examine how the financial industry may perpetuate gender discrimination. The disparity we have observed in credit card approvals between male and female underscores the challenges women face in achieving financial independence. In addition, historical context supports this view, as women were not allowed to have credit cards in their own names until the passage of the Equal Credit Opportunity Act in 1974 (Frankel 1). This legislation was a significant milestone in advancing women's financial rights yet the inequality might still persist, suggesting the importance of establishing policies and legislations in tackling societal issues. Viewing these disparities from a feminist lens allows the understanding that women were historically denied financial opportunities, creating a cumulative disadvantage that affects their credit histories.

Furthermore, analyzing these biases through the lens of intersectional feminism reveals that the problem is more complex for women of color, who often face compounded systemic oppression based on both gender and race. Intersectionality, a term compounded by Kimberlé Crenshaw, highlights how various forms of discrimination intersect. The Urban Institute found that white communities have the highest median credit score of 727, while Native American, Black, and Hispanic communities held a credit score in the low 600s. (The Urban Institute, n.d.). This disparity in credit scores appears to be a direct result of historical practices that have consistently undermined the financial opportunities available to people of color. This disproportion coupled with the discrepancies present for women in general creates a significant barrier to entry for women of color.

Another critical factor to consider is the historical context of credit systems and how they evolved to prioritize certain groups over others. The credit scoring system, which became widely adopted in the 1980s, was initially designed to streamline the process of evaluating creditworthiness. However, the metrics used, such as credit history, income stability, and asset ownership, inherently favored those who already had access to financial resources and stability. This system thus perpetuated a cycle where those who were already financially secure continued to have better access to credit, while those who were economically disadvantaged faced significant barriers. This historical context is crucial in understanding why current credit approval processes might still reflect these entrenched biases and why there is a need for reform.

From a Marxist perspective, which critiques the class inequalities inherent from capitalism, the financial barriers faced by low-income individuals can be seen as a manifestation of economic inequality. Credit card companies, by preferring approvals to applicants with assets like cars and higher incomes, reinforce class distinctions, making it more difficult for economically disadvantaged individuals to access credit. This exclusion can perpetuate a cycle of inequality since credit cards serve as essential tools for managing finances and building credit history. Ultimately, this happens to reinforce the exploitative nature of capitalist structure where the profit being made by large corporations drives policies that disadvantage marginalized societal groups. Viewing these discrepancies through a Marxist lens reveals how financial systems are designed to benefit the wealthy while systematically excluding the financially disadvantaged. Thus, addressing these biases is helpful in tackling the broader issues of gender and economic inequality in our society.

With the increasing usage of credit cards, the credit card oligopoly no longer has an eligibility policy that will go unscrutinized. The jarring gaps between the eligible and ineligible based on criterion such as income, car ownership and gender highlight an unfair mechanism created by wealthy individuals leading banks such as JP Morgan and Bank of America. While we need to conduct more careful analysis on the percentage difference brought by the available factors, introductory results show that even factors such as being married or having a better education can affect the chances of getting a credit card. This appears unfair to low-earning or single citizens who may otherwise be in good credit standing.

Classical Marxism states that the superstructure must change to fulfill the material needs of society, and in this post-pandemic era, credit cards are indeed a fundamental necessity. Gender based discrimination for eligibility can also no longer hold, as it fails to treat different genders as equals, an ideology which is patriarchal and disempowering to women and other genders. The data clearly indicates that men are more likely to be eligible for credit cards, which impedes women’s participation in the economy. In light of this, we argue that banks need to change their credit card eligibility policies to demonstrate a policy which embraces diversity, equity and inclusion. To foster a more inclusive financial landscape, it is imperative that banks revise their credit scoring models to incorporate a wider range of financial indicators of creditworthiness. By integrating alternative data sources that don’t simply box individuals into traditional credit metrics, banks can develop a more nuanced understanding of an individual’s financial behavior.

The potential changes in policy and practice could include considering alternative credit data such as utility payments, rent payments, and other forms of regular financial commitments. These metrics can provide a more comprehensive picture of an individual’s financial responsibility, especially for those who may not have extensive credit histories. Additionally, implementing bias training for those involved in the credit approval process and regularly auditing approval data for disparities can help mitigate unconscious biases and ensure fairer decision-making.

In conclusion, the analysis of credit card approval processes reveals significant biases that reflect broader societal inequalities. By applying feminist and Marxist lenses, we can better understand the structural issues at play and advocate for changes that promote greater equity in financial systems. This project not only aims to demystify the credit card approval process but also to highlight the need for systemic reform to ensure that access to credit is fair and inclusive for all individuals, regardless of their gender, race, or socioeconomic status.

Page updated

Report abuse