Picture two people applying for a credit card. Both feel sure they understand how interest, minimum payments, and late fees work. But one of them actually knows the material, and the other only thinks they do. That gap between how much someone believes they know about money and how much they truly know has a name in behavioral finance: financial overconfidence. And a new study suggests it can quietly steer people toward missed payments and mounting balances.
The research, published in the Journal of Behavioral and Experimental Finance, examines how this confidence gap is linked to credit card delinquency, and whether that link plays out differently for men and women. Its central finding: financial overconfidence is associated with a higher likelihood of falling behind on credit card debt, but the connection appears weaker among women than among men.
The question behind the study
Credit card debt in the United States has climbed to record levels, and a large share of cardholders pay only the minimum each month, racking up compound interest and fees. Researchers have long looked at psychological reasons why people mismanage debt, including a tendency to underestimate how fast interest grows and a bias toward spending now rather than later.
Malvika Chhatwani of XLRI – Xavier School of Management in India focused on one particular psychological trait. She wanted to know whether people who overrate their own financial knowledge are more prone to credit card trouble, and whether gender changes that relationship. The title of her paper, “Girls will be girls?”, nods to a well-known earlier study titled “Boys will be boys,” which found that overconfident men traded stocks more aggressively and lost more money.
To understand the study, it helps to separate two kinds of financial literacy. Objective financial literacy is what a person actually knows, measured by a quiz. Subjective financial literacy is how knowledgeable a person feels. When someone’s feeling of competence runs ahead of their actual score, that mismatch is labeled overconfidence. When the reverse happens, it is called underconfidence.
How the study was built
Chhatwani drew on the 2021 National Financial Capability Study, a survey overseen by the Financial Industry Regulatory Authority, known as FINRA. After removing incomplete responses, her sample included 19,795 individuals across the United States, and the results were weighted to reflect the broader population.
Objective financial literacy came from five quiz questions covering topics like interest and risk. Subjective financial literacy came from a single question asking people to rate their overall financial knowledge on a scale from 1 to 7. Chhatwani then compared each person’s scores against the sample averages and sorted everyone into four groups: financially overconfident (low actual knowledge, high self-rating), financially underconfident (high actual knowledge, low self-rating), moderately high confidence (high on both), and moderately low confidence (low on both).
Credit card delinquency was measured with an index built from six behaviors, including not paying the balance in full, carrying a balance and being charged interest, paying only the minimum, getting hit with a late fee, exceeding a credit limit, and using a card for a cash advance. Each behavior added a point, so scores ranged from 0 (no problems) to 6 (all six).
Because that delinquency score is a ranked scale rather than a simple yes-or-no outcome, Chhatwani used a method called ordered logistic regression, which estimates how a given factor shifts the odds of landing in a higher delinquency category. She also accounted for age, education, marital status, ethnicity, number of children, income, and employment status.
What the analysis revealed
Compared with people in the moderately high confidence group, those in the overconfident group had roughly twice the odds of credit card delinquency. That result lines up with the idea that people who overestimate their financial skill may put off paying bills because they assume they can handle the debt.
A wider pattern also emerged. It was not only overconfidence that tracked with trouble. People with moderately low confidence had about 2.27 times the odds of delinquency, and the underconfident group had about 1.47 times the odds. Chhatwani interprets this as evidence that any mismatch or shortfall in financial confidence, not just overconfidence alone, is associated with worse credit card behavior. The people who fared best were those whose high confidence was backed by high actual knowledge.
The gender piece is where the study’s title comes into play. When Chhatwani added a comparison between men and women, women who were financially overconfident showed about 25.5 percent lower odds of delinquency than overconfident men. In plain terms, the harmful link between overconfidence and missed payments was stronger for men than for women.
One detail from the descriptive data adds nuance. In this sample, women actually reported slightly higher rates of financial overconfidence than men, which runs counter to the common finding that men tend to be more overconfident. Even so, that overconfidence appeared less damaging for women’s credit card behavior.
How the author explains the gender gap
To make sense of this, Chhatwani turns to gender role theory, which holds that men and women are socialized into different expectations. Financial decision-making has traditionally been framed as a “masculine” activity, with men cast as household breadwinners. The author argues that when men are overconfident in a domain already tied to their expected role, that confidence may push them toward riskier behavior. For women, whose traditional roles have excluded financial decision-making, she suggests that overconfidence may instead act as a helpful counterweight to lower self-assurance, softening rather than worsening outcomes. This is the author’s interpretation of the pattern, not a directly tested mechanism.
Chhatwani ran eight robustness checks to test whether the findings held up. These included a technique called propensity score matching to reduce the chance that overconfident people simply differed in background characteristics, adding controls for risk and time preferences, redefining the overconfidence measure, and examining each of the six credit card behaviors on its own. The main results stayed consistent across most of these tests, though the gender difference weakened or lost statistical significance in a couple of specifications, including one that pooled three survey years together.
Practical takeaways and caveats
Chhatwani suggests the findings point toward financial education that does more than teach facts. Because both overconfidence and underconfidence were linked to worse outcomes, she argues that programs should help people calibrate their confidence to match their actual knowledge. She also raises the possibility of gender-specific approaches, along with product design ideas such as personalized feedback on spending or behavioral nudges and reminders to prompt on-time payments.
Several limits deserve attention. The study is cross-sectional, meaning it captures a single snapshot in time and cannot establish that overconfidence causes delinquency. The relationships described are associations. The data also come from FINRA’s survey pool, which skews toward people who engage in investing, so the results may not extend neatly to everyone. And the survey relies on self-reported behavior, which can carry its own inaccuracies. As Chhatwani notes, longitudinal studies following the same people over time would be needed to sort out cause and effect.
