Financial Literacy, Self-Efficacy, and Risky Credit Behavior among Kredivo Users in Jakarta Area

Financial Literacy, Self-Efficacy, and Risky Credit Behavior among Kredivo Users in Jakarta Area

Putri Renalita Sutra Tanjung

Universitas Mercu Buana, Jakarta, Indonesia

DOI: https://doi.org/10.51244/IJRSI.2025.120600129

Received: 10 June 2025; Accepted: 17 June 2025; Published: 15 July 2025

ABSTRACT

The use of Buy Now Pay Later (BNPL) services such as Kredivo is increasing, especially in the Jakarta area. Many users take advantage of the convenience of this digital credit without adequate financial understanding, potentially leading to risky credit behavior. This study aims to analyze the effect of financial literacy and self-efficacy on risky credit behavior in Kredivo users in Jakarta. This quantitative research uses a survey method with a questionnaire distributed to 150 respondents who are active Kredivo users during 2024. The research instrument includes three scales: financial literacy, financial self-efficacy, and risky credit behavior. Validity and reliability tests were carried out on questionnaire items before data collection. The data analysis technique used descriptive statistics and multiple linear regression with a significance level of 5%. The results showed that financial literacy and self-efficacy had a significant effect on reducing risky credit behavior. Partially, financial literacy is significantly negatively related to risky credit behavior, as well as financial self-efficacy shows a significant negative influence on risky credit behavior. The two independent variables simultaneously explain a meaningful proportion of the variation in risky credit behavior in Kredivo users. These findings indicate that good financial knowledge and confidence in managing finances play an important role in preventing risky credit behavior. The practical implication is that increasing financial literacy and user self-efficacy needs to be the focus of education to reduce the risk of problematic credit in using paylater services.

Keywords: financial literacy; self-efficacy; risky credit behavior; paylater; Kredivo

INTRODUCTION

Buy Now Pay Later (BNPL) or paylater services have become a new trend in digital payment methods in Indonesia. Through this service, consumers can purchase goods instantly and postpone payment for up to 30 days or installments within a certain period of time. One of the leading BNPL providers in Indonesia is Kredivo, which offers instant credit convenience for e-commerce customers. The popularity of paylater services continues to rise; a Populix survey (2023) shows that around 45% of respondents have used paylater, where Kredivo is recognized by 38% of respondents as one of the popular paylater services in Indonesia. Especially in Jakarta – the center of the economy and urban lifestyle – the use of Kredivo is growing rapidly along with the rise of online shopping and the need for practical consumer credit.

Despite offering convenience, uncontrolled use of paylater can pose a financial risk to users. Many users, especially the younger generation, are not fully aware of the negative impact behind the convenience of this digital credit facility. The level of financial literacy among Indonesians is still relatively lower than the level of financial inclusion. This means that many people have become users of financial products (including paylater) but do not have adequate financial knowledge and understanding. The Financial Services Authority (OJK) in a national survey in 2024 noted that the financial literacy index of the Indonesian population had only reached around 65.4%, still lagging behind the financial inclusion index which was more than 75%. This gap indicates that there are users of financial products who transact without sufficient knowledge, making them vulnerable to unwise financial behavior.

One of the financial behaviors of concern in the context of paylater services is risky credit behavior. Risky credit behavior refers to actions in the use of credit that can increase the likelihood of default or financial problems in the future. Examples include habitually delaying bill payments, taking out credit beyond one’s ability to pay, or using credit facilities for impulsive consumer purchases. An indication of this increased risk can be seen in the latest OJK data, which shows a surge in the value of paylater loans along with an increase in the level of bad debts. By the end of 2024, the total national paylater debt was recorded at IDR 20.5 trillion (up 48% compared to the previous year), accompanied by an increase in the paylater Non-Performing Financing (NPF) ratio from 2.76% in October 2024 to 2.92% in November 2024. This fact illustrates the potential for debt trap if the use of paylater is not managed wisely.

Two important factors that are thought to influence how users’ credit behavior is formed are financial literacy and self-efficacy in a financial context. Financial literacy is generally understood as a person’s knowledge, skills and beliefs that influence their attitudes and behaviors in making quality financial decisions. With good financial literacy, individuals are expected to be able to understand financial information (such as interest, penalties and credit risk) and make rational decisions on debt. Conversely, low financial literacy is associated with less prudent credit decisions and an increased likelihood of non-performing loans. Limbu (2017) found that higher credit card knowledge is negatively associated with credit card mismanagement behavior (misuse), meaning that the more knowledgeable a person is about how credit works, the less likely they are to misuse credit. In the context of paylater, adequate financial literacy should help users understand the consequences of this short-term debt, thus avoiding risky behaviors such as defaulting on payments or spending beyond their means.

The second factor, financial self-efficacy, also plays an important role. The concept of self-efficacy is derived from Bandura’s (1977) social cognitive theory, which defines self-efficacy as an individual’s belief in his or her ability to organize and execute the actions necessary to achieve certain results. In the financial domain, financial self-efficacy is defined as a person’s belief in his or her ability to manage finances and achieve desired financial goals. People with high financial self-efficacy are confident that they can manage their spending, pay their obligations on time, and resist the temptation of excessive credit. Strong self-efficacy has been associated with positive financial behaviors, such as healthy saving habits and debt management, which in turn improve subjective financial well-being (Dare et al., 2022). In contrast, those who lack confidence in their financial capabilities may be tempted to use credit without careful calculation, or feel overwhelmed and delay debt repayment.

Various previous studies have examined the relationship between financial literacy and self-efficacy with financial behavior, including debt behavior. However, the findings are still mixed. Liu and Zhang (2021) in a study of college students using online credit in China found that financial literacy has a significant negative effect on risky credit behavior – the higher the literacy, the lower the propensity for risky credit behavior. The study also showed that self-efficacy partially mediated the relationship, confirming the important role of financial confidence in amplifying the impact of literacy.

In Indonesia, Constansje, Kurniasari, and Abubakar (2023) studied Generation Z paylater users and found that financial literacy was significantly negatively related to risky credit behavior, while self- efficacy did not show a significant direct effect. This suggests that basic financial knowledge is instrumental in suppressing risky credit behavior in the younger generation, while self-efficacy alone without knowledge may not be enough to change behavior. In contrast, Ernawati, Yusnita, and Wibawa (2024) in a survey of students using paylater in Tasikmalaya found that self-efficacy has a significant influence in reducing risky credit behavior, while financial literacy has no significant effect. Although the literacy of the respondents is good on average, the knowledge does not seem to be directly reflected in behavior, in contrast to self-efficacy which contributes significantly: students with high financial confidence tend to be more cautious in debt. Meanwhile, research by Pratama et al. (2024) involving university students in Purwokerto reported unexpected results: financial literacy and consumptive lifestyle had a positive effect on risky credit behavior, while self-efficacy had a negative effect. This result indicates that in certain contexts, students who are more financially savvy and lifestyle-oriented may be more active in utilizing credit (thus appearing to increase risk), although self-efficacy can curb such risky behavior. This difference in findings suggests that the effect of financial literacy and self-efficacy on credit behavior is not simple and may be influenced by sample characteristics as well as other situational factors (e.g. financial stress, self-control, or lifestyle).

Based on the above background, further research is needed to understand the relationship between financial literacy, self-efficacy, and risky credit behavior, especially in users of paylater services such as Kredivo in big cities like Jakarta. The research questions raised are: Do financial literacy and self- efficacy have a significant effect on risky credit behavior in Kredivo users in the Jakarta area?

The purpose of this study is to examine the effect of financial literacy and self-efficacy on risky credit behavior among Kredivo users in Jakarta in 2024-2025. By understanding the factors that influence risky credit behavior, it is hoped that the results of this study can provide input for stakeholders – from fintech companies providing paylater, financial regulators, to educational institutions – in formulating strategies to improve financial literacy and appropriate interventions to minimize the risk of default and unhealthy debt behavior among paylater users.

LITERATURE REVIEW

Financial Literacy.

Financial literacy is generally defined as a combination of knowledge, skills and beliefs that influence a person’s attitude and behavior in making decisions and managing finances in order to achieve financial well-being. Financial literacy indicators include an understanding of basic financial concepts (such as interest, inflation, time value of money), financial planning skills, knowledge of financial products and services, and skills in using financial products wisely. A person with high financial literacy is expected to be able to compare credit options, understand loan terms and fees, and realize the risks of over-indebtedness. Financial literature shows that good literacy is closely related to positive financial behavior. For example, Lusardi and Mitchell’s (2014) multi- country study found that highly literate individuals tend to be more active in saving for retirement and less likely to fall into consumptive debt. In contrast, low financial literacy is often associated with errors in financial decision-making, such as uncontrolled use of credit cards and investing in high-risk schemes. Limbu (2017) specifically examined college students who use credit cards and found that a higher level of credit card knowledge has a significant effect on reducing credit card misuse behavior. In the context of paylater services, financial literacy plays a key role so that users can understand the mechanism of the service (e.g. installment interest or late fees) so that they can avoid risky credit behavior such as delinquent payments or continuing to increase debt without planning.

Financial Self-Efficacy.

The concept of self-efficacy was first introduced by Bandura (1977), which refers to an individual’s belief in his or her ability to organize and perform certain actions necessary to achieve desired outcomes. Financial self-efficacy is the application of this concept in the financial domain, namely the extent to which a person believes he can manage his finances and achieve the financial goals that have been set (for example, saving a certain amount of money, paying off debt, or investing). Forbes and Kara (2010) state that financial self-efficacy reflects an individual’s confidence in his or her ability to take the financial actions necessary to achieve certain financial well-being or targets. Indicators of financial self-efficacy can be seen from a person’s beliefs in aspects such as: the ability to prepare a budget and stick to it, the discipline to pay bills on time, the ability to deal with financial emergencies, and feeling able to resist the temptation of excessive consumption despite credit facilities. Research shows that high self-efficacy correlates with healthy financial behavior.

Hilgert et al. (2003) found that individuals with high self-efficacy scores were more consistent in implementing good financial practices (such as budgeting and paying bills on time) than those with low self-efficacy. In the context of credit, those who are confident are usually more proactive in setting payment schedules and controlling the amount of debt, while those who are less confident may be easily anxious or procrastinate, increasing the risk of non-performing loans. Dare et al. (2022) also revealed that financial self-efficacy contributes to financial well-being through the mediation of positive financial behaviors – meaning that people with high self-efficacy tend to make beneficial financial decisions, which in turn leads to better financial conditions.

Risky Credit Behavior.

Risky credit behavior can be defined as any action or habit in the use of credit that has the potential to increase the risk of financial loss or default in the future. In this study, risky credit behaviors in paylater users include, among others: the habit of delaying or missing bill payment deadlines, taking or using credit limits beyond repayment capacity, having multiple loans/credits at once, and impulse buying with paylater beyond need or financial capability. These behaviors are considered “risky” because they can lead to debt accumulation, increased interest costs and penalties, deteriorating credit scores, and serious financial difficulties such as default. Studies on risky credit behavior often refer to the concept of credit misuse, such as with credit cards (Pirog & Roberts, 2007) or student loans. Factors driving risky credit behavior include lack of financial literacy, consumptive attitude, weak self-control, social group pressure, and low risk perception of debt. In paylater services, its easily accessible characteristics, instant credit limits, and attractive promotions can encourage impulsive shopping behavior and increase the risk of users falling into over-indebtedness. Rachmawati and Agustina (2021) in their research mention the phenomenon of “digging holes” experienced by some paylater users: taking out new loans to pay old installments, which is an indication of high-risk credit behavior.

The Effect of Financial Literacy and Self-Efficacy on Risky Credit Behavior.

Previous theories and empirical findings generally assume that high financial literacy and self-efficacy will reduce risky credit behavior. Financially savvy individuals tend to be more aware of the consequences of debt, so they are more careful in using credit and managing their loans. Similarly, individuals who have high financial self-efficacy feel that they can control their financial situation, so they are more disciplined in debt repayment and are not easily provoked to add debt outside the plan. The results of Liu and Zhang’s (2021) study support this assumption, where students’ financial literacy significantly reduces the tendency of risky credit behavior on online loans. The study also showed that self-efficacy acts as a partial mediator, which means that financial literacy also increases self-efficacy, and then the self-efficacy helps suppress risky behavior. In other words, financial knowledge makes a person more confident in managing credit, and the combination of the two results in wiser credit behavior.

However, research in Indonesia shows mixed results. Constansje et al. (2023) found that higher financial literacy significantly reduced the risky credit behavior of Gen Z paylater users, but self-efficacy had no significant effect. This may be because although the younger generation is confident in using financial technology, it is their basic financial knowledge that ultimately determines whether or not they are wise in taking on debt. In contrast, Ernawati et al. (2024) found that only self-efficacy significantly reduced the risky credit behavior of students who use paylater, while financial literacy was not significant. Allegedly, age and maturity factors influence; in college students, self-confidence may play a greater role in controlling behavior (i.e., they know the consequences but what determines is the ability to restrain and regulate actions, which is reflected in self-efficacy). While their average “good” literacy may not be heterogeneous or not sufficiently implemented in real behavior.

Pratama et al. (2024) added another perspective: students with higher financial literacy actually indicated more active risky behavior (perhaps because the group felt they knew better how to use credit or were driven by lifestyle), but self-efficacy still played a role in suppressing the risk. This finding reminds us that the relationship between literacy and credit behavior can be influenced by other variables such as attitudes, social norms, or motivation to use credit.

Seeing the differences in the results of previous studies, the hypotheses in this study are formulated as follows: (1) Financial literacy negatively affects risky credit behavior in Kredivo users (the higher the literacy, the lower the risky behavior); (2) Self-efficacy negatively affects risky credit behavior (the higher the self-efficacy, the lower the risky behavior). These two variables are also expected to simultaneously have a significant effect, meaning that together they contribute to explaining the variability in risky credit behavior of paylater users. The conceptual framework of this study can be described as the relationship of two independent variables – financial literacy (X1) and self-efficacy (X2) – to one dependent variable, namely risky credit behavior (Y).

RESEARCH METHODOLOGY

Design and Approach:

This study uses a quantitative approach with a survey method. The research design is explanatory to test the causal relationship between the independent variables (financial literacy and self-efficacy) and the dependent variable (risky credit behavior). data were collected using a closed questionnaire distributed online to the target respondents.

Population and Sample:

The population in this study are all Kredivo service users in the DKI Jakarta area. Given the very large population and not individually recorded, the study used a non-probability sampling technique with a purposive sampling approach. The inclusion criteria for respondents are: at least 18 years old, domiciled or active in Jakarta, and have used the Kredivo credit limit for shopping in the last 6 months. The sample collected amounted to 150 respondents, which is considered to meet the minimum limit for multiple linear regression analysis (rule of thumb: at least 50 + 8m, with m = number of independent variables, according to Green, 1991). Sample characteristics in summary: 54% of respondents were female and 46% were male. The age range of respondents was 19-40 years, with the majority (70%) aged 25-34 years. In terms of occupation, 60% are private employees, 20% are self-employed, 15% are students, and the rest are other professions.The average monthly expenditure of respondents is in the Rp5-10 million category. Most respondents (82%) have more than one paylater account other than Kredivo (e.g. Shopee PayLater or Akulaku), indicating a tendency to use multiple digital credit sources.

Research Instruments:

The instrument used was a closed questionnaire with a 5-point Likert scale (1 = Strongly disagree to 5 = Strongly agree) to measure the three constructs of the research variables. The questionnaire was prepared referring to the operational definition of each variable as well as adaptations from previous research instruments so that its content validity was guaranteed. The following are the details of the instrument along with examples of statement items:

Financial Literacy

Measured by 5 statements that assess respondents’ knowledge of basic financial concepts related to credit and credit management skills. Respondents are asked to rate how much they agree with statements that reflect their financial understanding. Sample item:

“I understand the concept of interest and late fees in paylater credit.”

“I know how to calculate monthly installments based on the loan amount, tenor, and interest rate.”

“I understand the impact of late payments on my credit score and the penalties I have to pay.”

“I have sufficient knowledge about the terms and conditions of use of the Kredivo service.”

“I understand the risks of going into debt beyond my ability to pay.”

Financial Self-Efficacy

Measured by 5 statements that measure respondents’ level of confidence in managing aspects of their personal finances, specifically related to the use of credit and debt control. Respondents rate the extent to which they feel capable of performing good financial management behaviors. Sample item:

“I am confident that I can pay my Kredivo bill on time every month.”

“I am confident that I can control myself not to use credit beyond my financial means.”

“I was able to develop a monthly budget that included installment obligations and consistently stick to it.”

“I feel like I can turn down a discount or a paylater offer if it’s going to put me in debt beyond my plan.”

“I am confident that I will be able to pay off all the loans I have on schedule.”

Risky Credit Behavior

Measured by 6 statements that reflect the frequency or tendency of respondents to engage in risky behavior in the use of paylater credit. The higher the score on this scale indicates the higher the risky credit behavior. Respondents rate how often or how applicable the following statements are to their situation. Sample item:

“I often pay Kredivo bills beyond the specified due date.”

“My Kredivo credit limit is often used close to or up to the maximum limit.”

“I use the paylater facility to purchase items that are beyond my financial means if I have to pay cash.”

“I have more than one paylater or online loan account to fulfill my shopping desires.”

“I was often tempted by paylater promos and bought things I didn’t really need.”

“There was a time when I couldn’t pay my paylater bill on time and it went into arrears the following month.”

Each statement item is compiled in Indonesian that is easy to understand and relevant to the situation of using Kredivo. Before mass distribution, a try-out of the instrument was carried out on 30 respondents who had similar characteristics to the research sample. This aims to ensure the instrument is valid and reliable. The item validity test was carried out using the Pearson Product-Moment correlation technique between the item score and the total variable score. The validity criteria use α = 0.05 (two-way) and a cut-off correlation value of > 0.3 as a minimum requirement for items to be considered valid (Azwar, 2012). The reliability test was conducted by calculating the Cronbach’s Alpha coefficient for each construct, with an α value ≥ 0.70 considered satisfactory (Nunnally, 1994).

The trial results show that all statement items have a grain-total score correlation above 0.30 (p < 0.01) so that all items are declared valid. The Cronbach’s Alpha reliability coefficient for each variable is: financial literacy = 0.82; self-efficacy = 0.85; and risky credit behavior = 0.79. Thus, the research instrument is declared reliable, which means that the inter-item internal consistency in each scale is well maintained.

Data Collection Procedure

The main data collection was conducted in November-December 2024. The questionnaire was distributed through an online form (Google Forms) whose links were shared in Kredivo user communities (e.g. social media groups) and financial-related online forums. Respondents filled out the questionnaire anonymously and voluntarily. At the beginning of the questionnaire, the purpose of the research is explained , a brief definition of each variable (so that respondents understand the context of the statement), and an affirmation that the data will be kept confidential and only used for academic purposes. Respondents were also asked to ensure that answers were given according to personal conditions and opinions. Of the 172 responses received, 150 were deemed complete and suitable for analysis (22 responses had to be eliminated because they did not meet the criteria: e.g. not domiciled in Jakarta or inconsistent entries).

Data Analysis Technique

The collected data was analyzed using SPSS version 26.0 software. The stages of analysis include: (1) Descriptive statistics to obtain an overview of the variables, including the minimum, maximum, mean, standard deviation, and categorization of variable scores. Score categorization is done by comparing the empirical mean with a rating scale (range 1-5) to interpret the level of financial literacy, self-efficacy, and risky behavior of respondents (e.g., mean ≥ 4.0 is categorized as “high”; 2.5-3.9 “medium/adequate”; < 2.5 “low”). (2) Classical assumption test on the data, including residual normality test (Kolmogorov-Smirnov), linearity test, and multicollinearity test among independent variables (with Variance Inflation Factor). (3) Multiple linear regression analysis to test the effect of financial literacy (X1) and self-efficacy (X2) on risky credit behavior (Y). The regression model tested:

Y=a+b1X1+b2X2+eY = a + b_1 X_1 + b_2 X_2 + e

with Y = risky credit behavior score, X1X_1 = financial literacy score, X2X_2 = self-efficacy score, aa = constant, b1,b2b_1, b_2 = regression coefficient, and ee = residual error. The test includes simultaneous significance test (F-test) to see whether X1X_1 and X2X_2 together have a significant effect on Y, and partial significance test (t-test) to see the effect of each variable X1X_1 and X2X_2 individually. The significance level used is α = 0.05 (5%). The coefficient of determination (R-square) is also calculated to determine the percentage contribution of the independent variable in explaining the dependent variable. The results of the analysis are presented in tabular form with narrative interpretation.

RESULTS AND DISCUSSION

Descriptive Analysis Results

At the initial stage, descriptive analysis was conducted to understand the characteristics of the research data, particularly the scores for the variables of financial literacy, self-efficacy, and risky credit behavior. Table 1 presents a summary of the descriptive statistics of the three variables, including the mean, standard deviation, minimum-maximum scores, and score trend categories.

Table 1. Descriptive Statistics of Research Variables (N = 150)

Variables	Min Score	Max Score	Mean	Std. Dev.	Category
Financial Literacy	2,80	5,00	3,87	0,52	High
Financial Self-Efficacy	2,60	5,00	3,95	0,58	High
Risky Credit Behavior	1,67	4,50	3,10	0,65	Fair/Medium

Source: Primary Research Data, 2024.

From Table 1 it can be explained that:

Financial Literacy

Respondents’ financial literacy scores were in the range of 2.80 to 5.00 with an average of 3.87 (SD = 0.52). This average is in the high category, which indicates that in general, Kredivo users in Jakarta who were sampled had a fairly good level of financial understanding and knowledge. They generally understand basic concepts such as loan interest, late fees, and debt risk. This can be caused by the characteristics of the sample, the majority of whom have a minimum education of S1 and have wider access to financial information. Nonetheless, there is still variation (min 2.80 indicates that there are a small number of respondents with relatively low literacy).

Financial Self-Efficacy

Self-efficacy scores ranged from 2.60 to 5.00 with a mean of 3.95 (SD = 0.58). This average is also high, which means that respondents are relatively confident in their ability to manage their finances, especially in terms of debt repayment and controlling credit use. In other words, many users feel confident that they can pay their bills on time and control their debt. This high level of self-efficacy may be influenced by respondents’ experience in using digital credit; for example, perhaps most have never defaulted on a payment and thus maintain their confidence. However, the standard deviation of 0.58 indicates a wide variation in self-efficacy among respondents.

Risky Credit Behavior

The risky behavior score has a range of 1.67 to 4.50 with an average of 3.10 (SD = 0.65). This average is in the moderate category, indicating that the credit behavior of Kredivo users tends to be at a moderate level. This means that respondents sometimes take risky actions in using credit, but not too often or extreme. Some respondents admitted to paying late or maximizing credit limits, but did not do so consistently. This moderate category can be interpreted as a cautionary signal: there is still unwise behavior, but not yet at a chronic level. The presence of respondents with scores as low as 1.67 reflects a very cautious user group (almost never engaging in risky behavior), while the maximum score of 4.50 indicates that there is also a segment that engages in risky credit-related behavior quite frequently.

In general, the descriptive results show that while respondents’ financial literacy and self-efficacy are on average high, risky credit behavior is not entirely low. This indicates that there may be other situational or psychological factors that sometimes make users continue to engage in risky behavior despite knowing it is unwise and feeling able to control it. For example, the temptation of promotions and ease of transactions may once or twice “trump” the knowledge and initial intention to be disciplined. This descriptive finding is in line with reports of the phenomenon that the younger generation, despite being technologically literate and well-educated, can occasionally fall into consumptive debt behavior due to environmental influences or momentary desires (Constansje et al., 2023).

Before conducting hypothesis testing, classical assumption testing has been carried out. The normality test shows that the residual distribution of the regression model is not significantly different from the normal distribution (Kolmogorov-Smirnov p> 0.05), so the normality assumption is met. The linearity test shows a linear relationship between each independent variable and the dependent variable. The multicollinearity test showed that the VIF values for financial literacy and self-efficacy were 1.21 (both the same), below the threshold of 10, and the tolerance was around 0.83, so it was concluded that there was no serious multicollinearity problem; literacy and self- efficacy were indeed positively correlated with each other (r = 0.45, p < 0.01) but this correlation was moderate and would not interfere with the regression analysis.

Regression Test Results: The Effect of Financial Literacy and Self-Efficacy on Risky Credit Behavior

To answer the research questions, multiple linear regression analysis was used. The F test is conducted to test the effect of the two independent variables together, while the t test is to see the partial effect of each variable. The results of the regression analysis are summarized in Table 2 and Table 3 below.

Table 2. F Test Results (Simultaneous Effect)

Model	R	R2	Adjusted R2	F	Sig. (F)
Literacy & Efficacy -> Risk Behavior	0,558	0,311	0,302	32,8	0,000 (< 0,001)

Source: Processed Data, 2024.

Table 2 shows that the coefficient of determination (R-square) is 0.311, which means that about 31.1% of the variation in risky credit behavior can be explained jointly by financial literacy and self- efficacy. The rest (68.9%) is influenced by other factors outside this model. The R-square value of 0.311 for individual behavior research is not bad, considering that financial behavior is usually influenced by many external factors (income, lifestyle, promotion, etc.). The F-test yielded F_count = 32.8 with significance p < 0.001. This indicates that the regression model incorporating financial literacy and self-efficacy is statistically significant in influencing risky credit behavior. Thus, it can be concluded that financial literacy and self-efficacy simultaneously have a significant effect on risky credit behavior. This finding answers the first research objective, namely that the two independent variables together have a role in changing the credit behavior of Kredivo users. Furthermore, the regression test results per variable (t-test) are presented in Table 3.

Table 3. Multiple Linear Regression Test Results (Partial Effect)

Independent Variable	Coefficient B	Coefficient β	t	Sig. (p)	Description
Constant (a)	5,214	–	–	–	(intercept value)
Financial Literacy (X1)	-0,228	-0,210	-2,45	0,016	Significant effect
Independent Variable	Coefficient B	Coefficient β	t	Sig. (p)	Description
Self-efficacy (X2)	-0,417	-0,378	-4,64	0,000	Significant effect

Notes: B = unstandardized regression coefficient; β = standardized regression coefficient.

Source: Data processed, 2024.

The multiple linear regression equation based on the output above can be written as:

Y^=5.214 – 0.228X1 – 0.417X2\hat{Y} = 5.214 \;-\; 0.228 X_1 \;-\; 0.417 X_2

with X1X_1 = financial literacy, X2X_2 = self-efficacy, and Ŷ = risky credit behavior (predicted score).

Interpretation of coefficients: the constant value of 5.214 indicates the predicted value of risky credit behavior (on a scale score of 1-5) if literacy and self-efficacy are 0. Conceptually, this constant is less meaningful because a score of 0 is outside the range of the scale used, but mathematically it shows the cut-off point of the regression model.

The coefficient B of financial literacy = -0.228 means that every increase of 1 unit of financial literacy score will decrease the score of risky credit behavior by 0.228, assuming other variables are constant. Similarly, the B coefficient of self-efficacy = -0.417 indicates that every increase of 1 unit of self-efficacy score decreases the risky behavior score by 0.417, ceteris paribus. The negative sign on both coefficients is in accordance with the proposed hypothesis, which is that literacy and self- efficacy have a negative impact (reverse direction) on risky credit behavior. From the partial test (Table 3) obtained:

Effect of Financial Literacy: The value of t_count = -2.45 with p = 0.016 (<0.05). This means that financial literacy has a significant negative effect on risky credit behavior. The first hypothesis (H1) which states that financial literacy has a significant negative effect is accepted. In other words, the higher the level of financial literacy of Kredivo users, the lower their tendency for risky credit behavior. This effect of literacy in the model appears significant but not very strong (β = -0.210), indicating that the direct contribution of literacy in reducing risky behavior is around 21% in standard metrics. Although the contribution is smaller than self-efficacy, literacy still has a meaningful impact. This result is consistent with several previous findings that emphasize the importance of literacy in preventing problematic debt behavior. Adequate financial knowledge allows users to understand the consequences of taking on debt, making them more cautious. For example, respondents who understand the calculation of interest and penalties tend to think twice before delaying payments, because they are aware of the additional costs. This finding contradicts the results of the study by Ernawati et al. (2024) who did not find a significant effect of literacy on Tasikmalaya students, but in line with the studies of Liu & Zhang (2021) and Constansje et al. (2023) which involved a more general population. This may be because the sample of this study (adult users in Jakarta) utilized their financial knowledge more effectively in practice than the specific student sample.

Effect of Self-Efficacy: The value of t_count = -4.64 with p = 0.000 (<0.001). This shows financial self-efficacy has a highly significant negative effect on risky credit behavior. The second hypothesis (H2) is strongly supported. The higher a person’s confidence in managing their finances, the lower their tendency to engage in risky credit behavior. Self-efficacy has a coefficient of β = -0.378, which means that this factor contributes relatively more than literacy in reducing risky behavior. It can be said that self-efficacy is the dominant factor in this model. This finding is consistent with the social-cognitive theory that self-efficacy influences action: users who believe they can control their finances are more likely to be disciplined in paying bills and limiting debt. This result is in line with the study of Ernawati et al. (2024) which found self-efficacy as the main determinant of student credit behavior, also supporting the findings of Pratama et al. (2024) that self-efficacy has a negative (reducing) effect on risky credit behavior. However, in contrast to the research of Constansje et al. (2023) where self-efficacy was not significant when examined in Gen Z. The difference may be due to the different differences in self-efficacy. The difference could be due to the different characteristics of the sample; in the working adult population (not just students), self-efficacy may be formed from life experiences and daily financial responsibilities so that it has a real influence on behavior. While in younger groups, self-efficacy may not always be accompanied by financial decision-making maturity.

It can be concluded from the regression results that both independent variables play an important role in the context of paylater use. High financial literacy helps users understand debt risks and how to avoid them, while high self-efficacy makes users more disciplined and able to execute their good intentions in managing credit. The two complement each other: knowledge without confidence may not be implemented, while confidence without knowledge can be dangerous (confident but missteps). Therefore, strategies to reduce risky credit behavior should include improving both financial literacy and self-efficacy.

DISCUSSION

The results of this study provide several important insights. First, the proven negative effect of financial literacy on risky credit behavior reaffirms the importance of financial education for users of digital credit services. Despite the many conveniences offered by fintechs such as Kredivo, users with better literacy tend to be more vigilant and responsible in utilizing them. They understand that credit facilities are not “free money” and are aware of the consequences of delaying payments or excessive borrowing. This finding echoes the statement that knowledge is power: financial knowledge gives individuals the power to say “no” to self-defeating financial decisions. In practical terms, this means that financial literacy programs promoted by OJK, banks, and fintechs must continue to be expanded and include material related to the healthy use of digital credit. For example, users need to be educated on calculating their personal debt-to-income ratio, recognizing the danger signs of over- leverage, and effective debt repayment strategies. This research supports the policy that improving people’s financial literacy can be one way to prevent an individual debt crisis, especially in the era of rampant digital lending.

Second, self-efficacy proved to be the more dominant factor influencing risky credit behavior. This shows that the psychological aspects of self-confidence and self-control are no less important than the cognitive aspects (knowledge). Users who have high self-control and are confident in their abilities appear to be better able to resist the temptation of momentary consumption financed by debt. They are also more alert to manage payments before they are due because they are confident that they can manage their salary/income to fulfill these obligations. This finding is in line with the behavioral finance perspective, which states that financial decision-making is influenced not only by rationality (knowledge) but also by psychological factors including confidence, discipline and habits. Interestingly, in some case studies, too high self-efficacy without literacy can be detrimental (e.g. individuals are overconfident and underestimate risks). In this study, however, it seems that the respondents’ literacy level is quite good, so the high self-efficacy functions positively as expected.

The practical implication is that financial education programs may need to include soft skills training components such as self-control, financial stress management, and forming a confident yet realistic mindset in managing finances. For example, personal financial mentoring or coaching can help improve users’ financial self-efficacy, in addition to theoretical knowledge.

Third, the contribution of the two variables in the model explains about 31% of the variance in behavior. There are still almost 69% of other influential factors that were not included in this study. Some potential factors include: income and economic conditions (people with economic difficulties may be forced to engage in risky behavior), personal attitudes and values (e.g. materialism or impulsive tendencies), social influences (peer pressure to consume), and platform features (large credit limits, ease of minimum payments, etc., may encourage risky behavior). Constansje et al. (2023) for example tested the role of financial stress as a mediator but in their case it was not significant, although logically financial stress can exacerbate behavior (stressed people can dig holes to close holes). Research by Pratama et al. (2024) included lifestyle variables and found a significant effect, meaning that a consumptive lifestyle can clearly increase risky credit behavior. Therefore, for a more comprehensive picture, future research can add variables such as financial stress, self-control, attitudes towards debt, and demographics to see their influence.

When compared to the research of Ernawati et al. (2024) among Tasikmalaya students, the results of this study differ on the effect of financial literacy. In that study, literacy was not significant, perhaps because the variation in literacy among respondents was low (many accounting/management students were already similarly educated) so the effect was not apparent, or it could be because student behavior was more influenced by lifestyle and peer factors than knowledge. Whereas in Jakarta, respondents have more heterogeneous backgrounds, so literacy appears as a differentiating factor. While the similarity of the results is that they both show the importance of self-efficacy. This reinforces the assumption that anywhere, training financial self-efficacy – for example by providing mastery experience, modeling, or support – can be the key to changing financial behavior.

The results of this study practically provide a message to paylater service providers (such as Kredivo) and regulators: increasing user literacy is not enough, it is also necessary to increase their financial self-efficacy. For example, the paylater application can feature educational features and reminders that not only provide information (knowledge) but also encourage user action (thus building habits and confidence, for example, budgeting features or payment simulations). OJK and industry players can collaborate to organize debt management workshops for young consumers, which include practical tips to control spending, set payment schedules, and case studies of successful debt repayment, to inspire and increase self-efficacy.

No less important, the role of family and environment also influences shaping one’s literacy and self- efficacy early on (Shim et al., 2010 mentioned the concept of financial socialization). T herefore, financial literacy education should start from school age and involve parents, so that the nextgeneration grows up with a strong foundation of financial knowledge and confidence.

RESEARCH LIMITATIONS

As a note, this study has several limitations. First, in terms of the survey design using a self-report questionnaire, there is the potential for social bias that might influence respondents’ answers. For example, some respondents may give idealized answers (over-reporting literacy or self-efficacy) or under-report negative behaviors because they are embarrassed to admit. While anonymous surveys attempt to minimize this, biases may still exist. Second, the sample scope is limited to Kredivo users in Jakarta and surrounding areas, so generalizations to the national paylater population should be cautious. The characteristics of urban Jakarta (higher income, easier access to information) may differ from other regions. Third, this study is cross-sectional (one-time data collection), so it cannot confirm the cause-and-effect relationship absolutely. Although theory supports the causal direction of literacy/efficacy -> behavior, it is possible that debt behavior also influences knowledge and confidence from experience (two-way relationship). Future longitudinal studies may provide more insight into these dynamics.

Finally, the risky credit behavior variables measured are still based on respondents’ perceptions. In the future, it would be better to add objective data, such as transaction history or respondents’ credit scores, to strengthen the findings.

CONCLUSIONS AND SUGGESTIONS

Conclusion: This study concludes that financial literacy and financial self-efficacy have a significant effect on risky credit behavior in Kredivo users in Jakarta. Higher financial literacy tends to reduce the tendency of risky credit behavior, meaning that users who are more financially savvy are less likely to commit actions such as defaulting on payments or taking on excessive debt. Similarly, higher self-efficacy is associated with reduced risky credit behavior; users who believe in their ability to manage money prove to be more disciplined and cautious in taking on debt. Together, these two factors explain about 31% of the variation in risky behavior, suggesting a significant role. Between the two, self-efficacy emerged as a more dominant factor than financial literacy. This emphasizes the importance of psychological aspects in addition to knowledge in determining individual financial behavior.

These results underscore that efforts to prevent risky credit behaviors – such as the bad habit of delaying payments or impulsive borrowing – need to focus on improving financial literacy as well as strengthening the self-efficacy of users of digital financial services. Financial knowledge provides the foundation for individuals to understand the consequences of each debt decision, while high self- efficacy ensures individuals have the commitment and confidence to act on that knowledge.

Suggestions: Based on the findings of the study, some suggestions can be proposed:

For Fintech Service Providers (eg Kredivo): It is recommended to improve educational programs and application features that support user financial literacy and discipline. For example, providing interactive educational modules on credit management, installment calculators, or proactive alerts when users approach their credit limit. In addition, gamification features that provide rewards for users who pay on time or do not overuse credit can be considered, thus encouraging self-efficacy and good financial habits. Kredivo and similar platforms can also regularly send practical financial tips content via email/notifications, because financial literacy must be continuously honed. For Regulators and Government: OJK and related agencies need to continue expanding the reach of the National Financial Literacy program, with a focus on the segment of young fintech users in urban areas who are prone to consumptive debt. Educational content should be relevant to the current situation (e.g. “smart use of paylater”, “calculating installment capacity”, etc.). In addition to knowledge, programs to increase financial self-efficacy are also important. For example, through personal financial planning training, financial decision- making simulations, or testimonials of role models who have successfully gotten out of debt, it is hoped that people will not only know but also be confident in managing their finances. Certain regulations may require paylater providers to display risk information clearly before users transact (similar to warning), so that users are always reminded and educated when using their services.

For the Community/Users: Each individual credit user should proactively improve their financial literacy, for example by utilizing OJK’s education portal (Sikapi Uangmu), attending financial seminars/workshops, or discussing with financial planners. Also improve self- efficacy by starting from small steps that work (e.g. the discipline of paying one bill on time for several months will build confidence to manage other bills). Users also need to set personal limits on debt, such as a 30% rule of maximum income for installments, and strictly adhere to these limits – this fosters self-control.

For Future Research: It is recommended to conduct a follow-up study by including additional variables such as self-control, income level, financial stress, or marketing promotion factors, to see the comprehensive influence on risky credit behavior. Qualitative research can also be explored, for example through in-depth interviews with problematic paylater users, to understand their motives and challenges in managing credit. In addition, longitudinal studies would be useful to see how literacy and self-efficacy affect behavior over time, as well as the impact of specific educational interventions. With the synergy between increasing literacy and self-efficacy, it is hoped that users of digital credit services such as Kredivo can optimally enjoy the benefits of easy credit without getting entangled in harmful debt behavior.

BIBLIOGRAPHY

1. Anestia, (2022, June 3). Paylater is increasingly favored by consumers for online shopping.
2. Bandura, (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
3. Constansje, , Kurniasari, F., & Abubakar, A. (2023). The Effects of Financial Literacy, Self-Efficacy, and Financial Stress on Risky Credit Behavior of Generation Z: Evidence from Pay Later Users.Journal of Entrepreneurship, Business and Economics, 11(1), 180-210.
4. DailySocial.id. (online article). Accessed from: https://dailysocial.id/post/paylater-makin-diminati- konsumen-untuk-belanja-online
5. Dare, E., Yeresyan, I., Carreno, A., & Grimes, M. (2022). How executive functioning and financial self-efficacy predict subjective financial well-being via positive financial behaviors. Journal of Family and Economic Issues, 44(2), 232-248.
6. Ernawati, , Yusnita, R. T., & Wibawa, G. R. (2024). The Effect of Financial Literacy and Financial Self- Efficacy on Risky Credit Behavior in the Use of Paylater (Survey on Students in Tasikmalaya City). Lokawati: Journal of Management Research and Research Innovation, 2(6), 26-40.
7. Financial Services Authority (OJK). (2016). Indonesia National Financial Literacy Strategy (Revisit 2017). Jakarta: OJK (Financial literacy is the knowledge, skills, and beliefs that influence attitudes and ..).
8. Financial Services Authority (OJK). (2023). Financial OJK Education Portal. (accessed January 22, 2024 from https://www.ojk.go.id/…).
9. Financial Services Authority (OJK). (2025). Press Release of National Survey Results on Financial Literacy and Inclusion 2025. Jakarta: OJK, May 2, 2025.
10. Forbes, , & Kara, S. (2010). Defining financial self-efficacy: Financial self-efficacy refers to one’s belief in his or her ability to achieve financial goals. (Definition cited in Chen et al., 2021, Frontiers in Psychology).
11. Limbu, B. (2017). Credit card knowledge, social motivation, and credit card misuse among college students: Examining the information-motivation-behavioral skills model. International Journal of Bank Marketing, 35(5), 842-856.
12. Liu, , & Zhang, H. (2021). Financial literacy, self-efficacy and risky credit behavior among college students: Evidence from online consumer credit. Journal of Behavioral and Experimental Finance, 32, 100569.
13. Populix. (2023). 8 Most Popular Paylater Services in Indonesia, Shopee Paylater Champion. Databoks.Katadata.co.id, October 25, 2023. (Populix survey results on paylater brand awareness).
14. Pratama, C., Anatasya, P. S., Selavi, A., Aseh, M., Ainunnisa, N. S., & Innayah, M. N. (2024). Self- Efficacy, Financial Literacy, and Lifestyle Towards Risky Credit Behavior. In I. Zutiasari & D. T. Kurniawan (Eds.), Proceedings of the 4th International Conference on Business, Accounting, and Economics (ICBAE 2024, Purwokerto) (pp. 123-132). EAI/Atlantis Press.
15. Risman, A. (2024). The Behavioral Finance of MSME: Digital Finance, Managerial Biases, Financial Literacy. Dinasti International Journal of Economics, Finance & Accounting (DIJEFA), 5(2).
16. Risman, Asep and Ali, Anees Janee and Soelton, Mochamad and Siswanti, Indra (2022): The behavioral finance of MSMEs in the advancement of financial inclusion and financial technology (fintech). Published in: The Indonesian Accounting Review , Vol. 1, No. 13 (8 February 2023): pp. 91-101.
17. Sari, (2021). The Effect of Paylater Use on Impulse Buying Behavior of E-Commerce Users in Indonesia. Journal of Business and Investment Research, 7(1), 44-57.
18. Sugiyono, (2018). Quantitative, Qualitative, and R&D Research Methods. Bandung: Alfabeta.