Behavioural Drivers of Financial Well-Being Among Private School Teachers: Reconsidering the Role of Saving Habits

Abstract

This study examines the relationships among financial literacy, risk tolerance, future time perspective, saving habits, and financial well-being. This empirical model addresses limited evidence in research on teachers' context related to financial well-being. This study used a quantitative approach, employing a questionnaire administered to private school teachers in Jakarta. The data were collected from 160 non-certified private school teachers who were aged 23 to 55. Partial Least Squares - Structural Equation Modelling (PLS-SEM) were used to analysed the data. The findings show that future time perspective is the strongest predictor, with positive effects on both financial well-being and saving habits. Saving habits also enhance financial well-being, suggesting that routine saving behaviour plays an important role in strengthening teachers' perceived financial security. Financial literacy does not directly improve financial well-being, but it significantly promotes saving habits, indicating that financial knowledge becomes more meaningful when translated into disciplined financial behaviour. Risk tolerance shows no significant direct and indirect effect on financial well-being, implying that willingness to take financial risks may not necessarily support wellbeing through saving behaviour. Then, saving habits were not proven to be a mediator between financial literacy, risk tolerance, future time perspective, and financial well-being. This is because saving is a necessity in the context of urban workers in big cities, so financial behaviour is not always initiated by intention, but rather by conditions that require them to survive. This study highlights the importance of future-oriented thinking and consistent saving habits as key behavioural foundations of financial well-being.

KEYWORDS financial well-being; saving habits; financial literacy; risk tolerance; future time perspective; private school teachers; behavioural finance.

Introduction and Theoretical Background

The significance of financial well-being (FWB) as a strategic concern in behavioural finance research is increasing, given its direct links to productivity, stress, quality of daily life, and economic development (Adee et al., 2024). FWB is also a critical indicator of the standard of living, particularly amid complex financial challenges (Radianto et al., 2026). A comprehensive understanding of FWB is imperative for obtaining a more holistic perspective on economic resilience (Oberrauch et al., 2023). In response to this urgency, several studies have emerged, including Mawadah & Sumiati (2025), which discusses FWB among young adults, Setiawan & Iramani (2023) in the context of bank employees, Margasari et al. (2024) who discuss FWB among university students, Muat et al. (2025) in the context of Indonesian millennials, and Chauhan et al. (2025) in the employee context. Study focusing on teachers, particularly those employed by private foundations, remains scarce. Teachers contend not only with escalating occupational demands but also with the same financial obligations. Those conditions weigh on most working adults, such as children's education, housing costs, emergency reserves, and retirement planning. In a city like Jakarta, these pressures take on a more specific characteristic such as price volatility and daily transportation expenditures. These dynamics make Jakarta a relevant and underexplored context for FWB research. Addressing this gap would benefit both scholars and practitioners in designing financial support interventions in the teaching sector. Recent literature indicates that financial literacy (FL) underpins decision-making capabilities and influences FWB through behaviours such as saving, bill-paying discipline, and debt management (Lusardi & Streeter, 2023). This study highlights cognitive influences on financial behaviour. The most suitable paradigm for studying FL and FWB is the Theory of Planned Behaviour (TPB) (Kanth et al., 2026). Based on TPB, behaviour is ultimately driven by intention. The theory has demonstrated strong predictive validity across a range of behavioural contexts. Several definitions have emerged regarding FL, such as an individual's cognitive and informational capacity to understand financial concepts (Xiao & O’Neill, 2016). FL involves understanding and interpreting financial terms, evaluating financial circumstances, and making sound financial decisions (Paneda & Albay, 2025; Pradhan et al., 2021). Based on these definitions, FL is the cognitive ability to understand financial concepts and evaluate available financial alternatives. However, FL cannot account for all variations in FWB. To translate FL into behaviour, several other factors specifically risk tolerance (RT) and future time perspective (FTP) must be considered. Previous studies have focused on factors that shape RT, including Hemrajani et al. (2023) and Song et al. (2023). The process of converting knowledge into action is influenced by psychological characteristics that shape how individuals perceive uncertainty. RT significantly influences the selection of financial instruments, patterns of fund allocation, and the consistency of saving. It can be described as an individual's psychological preparedness to confront uncertainty and variability in financial results (Hemrajani et al., 2023). So, RT is person's tendency to engage in actions where the outcome is uncertain. As demonstrated by Owusu et al. (2023) RT is a construct influenced by psychological and cognitive factors. Studies on RT and its influence on financial decisions have expanded across various theoretical frameworks. This construct serves as a strong predictor of financial behaviour (Brayman et al., 2017). At the same time, extant studies have established the effect of RT on financial behaviour, such as retirement readiness or savings (Nguyen et al., 2022). The Prospect Theory has served as a foundational model for illustrating how people assess different possibilities based on their values and preferences. Prospect Theory suggests that individuals evaluate outcomes based on perceived advantages and losses. However, the profits and losses involved in this decision are uncertain (Fadli et al., 2024). So, individual tolerance for risk is equivalent to the attitude that shapes decision-making. Furthermore, financial behaviour also influenced by FTP. Strong future-time individuals prioritize long-term rewards and defer present expenditure for savings and financial planning (Oberrauch et al., 2023). According to Kooij et al. (2018),FTP is a key issue while planning an individual's future. It is vital for well-being and encompasses the ability to forecast, strategize, and foresee desired future outcomes. FTP refers to an individual's inclination to prioritize future outcomes and integrate long-term factors into current decision-making. Although it can be learnt and may evolve in response to various situations, FTP is not a fixed personality trait. Kempson et al. (2017) included FTP in their conceptual model of FWB. Future-oriented individuals tend to be less impulsive when choosing between having something now and later, have a better ability to delay gratification, and store financial

reserves such as savings, investments, and insurance for the future (Van Raaij et al., 2023). The above explanation shows that those who are more mindful of future circumstances will have better financial habits, such as diligent saving, investing, and careful financial planning. Within this framework, saving habit (SH) is conceptualised as a mediation mechanism that links FL, RT, and FTP to FWB. Harahap et al. (2022) previously demonstrated a chain mechanism in which FL influences saving behaviour through RT, which serves as a serial mediator in the context of long-term planning. The model and context proposed in this study are different from a number of previous studies, such as Mawadah & Sumiati (2025) which discusses FWB among young adults, Setiawan & Iramani (2023) in the context of bank employees, Margasari et al. (2024) who discuss FWB among university students, Muat et al. (2025) in the context of Indonesian millennials, Chauhan et al. (2025) in employee’s context, Hamdallah (2026) conducted research on financial satisfaction by examining the role of financial coping, Liu & Yang (2026) examined factors shaping well-being in Chinese teachers, then Laschowski & O’Reilly (2025) identified factors shaping financial well-being in the context of retired athletes in North America. However, research integrating FL, RT, and FTP and positioning SH as a behavioural mediator of FWB remains relatively limited, specifically in the teaching profession and in developing countries facing urban cost-of-living pressures. Teachers in Jakarta are theoretically informative because they constitute a professional group with formal educational backgrounds. This makes them an important context for FWB research. This study addresses that limitation by integrating FL, RT, FTP, SH, and FWB within a single behavioural finance model. The theoretical contribution extends beyond adding another occupational sample. The central question is whether SH functions as the behavioural pathway through which cognitive capability, risk preference, and future orientation jointly influence well-being. Drawing on the Theory of Planned Behaviour, FL is positioned as a capability-related antecedent that supports perceived behavioural control, while SH reflects the repeated behavioural patterns through which intention and capability are put into practice. Prospect Theory complements this by explaining how RT shapes responses to financial uncertainty, though such tolerance does not automatically improve well-being unless it is channelled into disciplined financial behaviour. The study, therefore, advances a specific theoretical argument that FWB is not simply a product of knowing more or accepting more risk, but depends on whether individuals develop stable saving routines, particularly among those with a stronger FTP. Three contributions to behavioural finance and FWB research follow from this framing. First, most behavioural finance studies treat FWB as the outcome of a single dominant factor that is financial knowledge, attitude, or income. This study departs from that position by arguing that the interaction of cognitive, psychological, and temporal dimensions shapes FWB. FL captures the capacity to evaluate and execute financial decisions. RT reflects how individuals respond psychologically to uncertainty. FTP captures the degree to which decisions are oriented toward long-term rather than immediate consequences. To connect these antecedents, the study draws on the Theory of Planned Behaviour. Second, the Jakarta private-teacher context yields theoretical insight precisely because of its tensions: teachers are formally educated, yet many face income constraints, household obligations, and urban cost pressures simultaneously. The study examines how FL, RT, and FTP affect the SH and FWB of private school teachers in Jakarta. This theoretical framework reframes FWB not as a static outcome of income levels, but rather as a dynamic consequence of sustainable financial behaviours (Netemeyer et al., 2017). Collectively, these constructs provide a comprehensive explanation of individual differences in the ability to develop consistent SH and achieve FWB, especially in the context of the teaching profession. This context is particularly significant because educators serve as strategic agents in human development. However, their role is frequently underappreciated in behaviour-based FWB mapping. From an academic perspective, the present study makes three notable contributions. Firstly, an integrated model of psychological, capability, and behavioural wellbeing is tested. Secondly, pathways to FWB are distinguished, particularly through the lens of SH. Thirdly, empirical evidence is presented to relevant occupational groups to inform the design of institution-based literacy and well-being programs.

Methods

Research Type This quantitative study used a survey with a standardized questionnaire administered online to private school teachers in Jakarta, with individual teachers serving as the unit of analysis. The use of an online survey was considered appropriate to reach respondents efficiently across different schools and to provide flexibility in participation. The measurement items were adapted from established prior studies covering FL, RT, FTP, SH, and FWB. Since several items were originally developed in English, they were translated into Bahasa Indonesia and adjusted to fit the local context while maintaining conceptual equivalence. Prior to the main survey, the instrument was pretested to assess clarity, wording, and respondent understanding, and minor revisions were made accordingly. In terms of ethics, respondents were informed about the purpose of the study, the voluntary of participation, and the confidentiality of responses before completing the questionnaire. Only those who provided consent proceeded to the survey. After data collection, the responses were screened to identify missing data and careless responses. Incomplete questionnaires, straightlining patterns, and responses with unusually short completion times were excluded to ensure data quality. The final dataset consisted of valid responses deemed suitable for further analysis. All items on the questionnaire were measured using a 5-point Likert scale. To minimize the potential for common method bias, respondents were informed that there were no right or wrong answers, confidential of the answer and the data. The questionnaire was structured by separating the predictor and outcome constructs into distinct sections, with clear and nondirective wording. Then, common method bias was assessed using the full collinearity VIF procedure recommended by Kock (2015) with a threshold of 3.3. The conceptual model's validity and reliability, as well as the relationships among the study's constructs, were assessed using statistical analysis. The constructs measured in this study are latent, so their measurement was carried out through indicators that reflect those constructs. FL was measured using four questions adapted from Harahap et al. (2022), RT was measured using five questions adapted from Hermansson & Jonsson (2021), SH was measured using six questions adapted from Alkhawaja & Albaity (2022), FTP was measured using six statements, and FWB using seven statements, both adopted from Van Raaij et al. (2023).To maintain content validity, the adaptation process followed three principles: preserving the conceptual meaning of the original items, limiting revisions to contextual wording changes, and evaluating clarity and equivalence during pretesting. As such, the adapted items remained theoretically consistent with the original scales while becoming more understandable for the target respondents.

Population and Sample The population of the study consisted of teachers employed by private educational foundations in Jakarta. Respondents were selected through purposive sampling, meaning each individual had to meet specific criteria before being included in the study. According to Campbell et al. (2020), this selection approach strengthens research rigour and produces more reliable outcomes. To be included, a respondent had to be an active teacher at a private foundation school in Jakarta, non-certified at the time of data collection, and between 23 and 55 years old. Data collection took place from October to November 2025. The initial pool of potential respondents was identified through a systematic outreach process, in which five private schools in Jakarta were contacted and asked to provide access to their teaching staff. School administrators from each institution were approached directly to obtain permission and facilitate the distribution of the questionnaire to eligible teachers. The total number of teachers identified across the five schools who met the inclusion criteria amounted to 210 teachers. Of that amount, 174 returned the questionnaire, and after data screening, 160 valid responses were retained for analysis, yielding an effective response rate of 76.19%. This approach was designed to obtain comprehensive and reliable data, following the guidelines of Hair et al. (2019), which suggests a minimum sample size of 10 times the number of arrows in the research model.

Data Analysis This quantitative study uses Partial Least Squares - Structural Equation Modelling (PLS-SEM) with SmartPLS 4 software. PLS-SEM enables a thorough assessment of complex theoretical models, including measurement validity and hypothesis testing (Hair et al., 2019). PLS-SEM was selected here because the study emphasizes variance explanation and the estimation of predictive relationships among constructs. Given the study’s focus on mediation, moderate sample size, and the practical objective of understanding the drivers of financial well-being, PLS-SEM provided a more suitable analytical framework. The instrument's validity and reliability were established using a measurement model, following the guidelines of Hair et al. (2019). AVE indicated internal consistency above 0.50, CR above 0.70, and Cronbach's alpha approaching 1. Then, the structural model was evaluated through path coefficients, tstatistics, p-values, confidence intervals, coefficient of determination (R²), and collinearity diagnostics. To assess multicollinearity, inner Variance Inflation Factor (VIF) values were examined for all predictor constructs; all values fell below the threshold of 3.3 as recommended by Hair et al. (2019), indicating the absence of serious collinearity concerns among the constructs in the structural model. Hypothesis testing was conducted using the bootstrapping procedure in SmartPLS with the following exact settings: 5,000 bootstrap subsamples, no sign changes option, and bias-corrected and accelerated (BCa) confidence intervals at the 95% confidence level. All significance tests were performed as two-tailed tests at the 5% significance level (t > 1.96; p < 0.05), consistent with the conservative inference standard recommended for PLS-SEM in scholarly research (Hair et al., 2019). For the mediation analysis, indirect effects were estimated through the bootstrapping procedure using the same 5,000 subsamples. An indirect effect was considered statistically significant if the 95% BCa confidence interval of the indirect effect entirely excluded zero, with no overlap between the lower and upper bounds.

Result and Discussion

Respondent Profile The following table 1 describes the basic characteristics of private school teachers in the sample. The profile of respondents reveals that the sample included 160 teachers from private schools, with a fairly even gender composition: 84 males (52.5%) and 76 females (47.5%). In terms of age, the majority were between 23 and 30 years old (35.6%), followed by respondents aged 45–50 years (15.0%) and 41–45 years (14.4%). The other age categories ranged from 11.3% to 12.5%. Most respondents have a bachelor's degree (61.9%), followed by a master's degree (30.6%) and a doctorate (7.5%). Most participants had 1–5 years of experience (34.4%), while 17.5% had 10–15 years and 15–20 years, and 13.8% had over 20 years. For family dependents, the most frequently reported categories were two dependents (29.4%) and three dependents (27.5%). Measurement Model The measurement method used in PLS-SEM ensures that each latent construct in this study is assessed accurately and reliably through its indicators. This study used a reflective measurement model, as illustrated in the table 2. In the initial stage of measurement model evaluation, several indicators were examined for retention based on outer loadings, inter-item correlations, and indicator VIF values. Indicators FTP2, FTP3, FTP4, SH3, SH4, FWB2, and FWB4 were removed because their outer loadings fell below the recommended threshold of 0.70 (Hair et al., 2019). The removal of these indicators did not narrow the conceptual coverage of their respective constructs. For the FTP construct, the deleted items (FTP2, FTP3, FTP4) captured overlapping facets of time preference already represented by the retained indicators (FTP1, FTP5, FTP6). For the SH construct, SH3 and SH4 addressed peripheral behavioural dimensions that were sufficiently represented by the retained indicators (SH1, SH2, and SH5). The indicators together maintain adequate coverage of saving habit regularity and consistency. For the FWB construct, FWB2 and FWB4 exhibited conceptual overlaps with retained indicators. Their removal did not compromise the theoretical representation of FWB across its affective and cognitive dimensions, as evidenced by the five retained indicators (FWB1, FWB3, FWB5, FWB6, FWB7). Regarding FL3, this indicator exhibited an inter-item correlation exceeding 0.85 with FL2, indicating substantial redundancy between the two items. High inter-item Table 2. Measurement Model Constructs Indicators Outer Loadings FL1 0.801 0.927 0.762 FL2 0.891 FL4 0.908 FL5 0.886 FL FTP1 0.723 FTP 0.801 0.573 FTP5 0.753 FTP6 0.794 RT1 0.906 0.946 0.816 RT2 0.911 RT3 0.925 RT4 0.869 RT SH1 0.755 SH 0.788 0.553 SH2 0.717 SH5 0.758 FWB1 0.731 0.856 0.543 FWB3 0.753 FWB5 0.722 FWB6 0.751 FWB7 0.726 Sources: survey data, 2025 FWB

correlation justified the removal of FL3. Its exclusion did not undermine the construct's theoretical domain, as the four retained indicators (FL1, FL2, FL4, FL5) collectively represent the full concept of FL. Following indicator deletion, outer loadings for the remaining items ranged from 0.717 to 0.925, indicating adequate contribution to their respective constructs. FL loadings ranged from 0.801 to 0.908, and RT produced the strongest loadings overall (0.869–0.925). Loadings for FTP Table 1. Respondent Profile Demographic Characteristic Frequency % Gender Male 84 52.5 Female 76 47.5 Age 23–30 years old 57 35.6 31–35 years old 20 12.5 36–40 years old 18 11.3 41–45 years old 23 14.4 45–50 years old 24 15.0 51–55 years old 18 11.3 Education Degree Bachelor 99 61.9 Master 49 30.6 Phd 12 7.5 Years of Work Experience 1-5 years 55 3.4 5-10 years 27 1.9 10-15 years 28 1.5 15-20 years 28 17.5 >20 years 22 13.8 Number of Family Dependents 1 family member 28 17.5 2 family members 47 29.4 3 family members 44 27.5 4 family members 21 13.1 5 family members 20 12.5 Sources: survey data, 2025 Composite Average Variance Reliability Extracted (0.723–0.794), SH (0.717–0.758), and FWB (0.722–0.753) each met the 0.70 threshold. Convergent validity was assessed through composite reliability (CR) and average variance extracted (AVE). CR values ranged from 0.788 to 0.946, confirming internal consistency across all constructs. RT returned the highest CR (0.946), followed by FL (0.927), FWB (0.856), FTP (0.801), and SH (0.788). AVE values ranged from 0.543 to 0.816, with all constructs surpassing the 0.50 criterion. Common method bias was evaluated using the full collinearity VIF procedure outlined by Kock (2015), in which diagonal elements from the inverse of the latent variable correlation matrix serve as VIF estimates. All constructs returned value well below the 3.3 threshold: FL = 2.460, RT = 2.407, SH = 1.397, FWB = 1.345, and FTP = 1.240. These results suggest that common method bias is unlikely to threaten the validity of the findings. Discriminant Validity

Evaluating discriminant validity involves determining whether a latent variable explains more variance in its related indicator variables than the variance explained by other constructs in the same model. The discriminant validity of the results, evaluated using the Fornell-Larcker Criterion. Based on the table 3, the model's discriminant validity is satisfied because the square root of the AVE (FL = 0.873; FTP = 0.757; FWB = 0.737; RT = 0.903; SH = 0.744) is larger than the correlations between related components. Additional evidence, HTMT, also supported the improved discriminant validity of the final measurement model. A construct is considered to have adequate discriminant validity when the HTMT value is below 0.90, while a more conservative threshold of 0.85 may also be applied (see table 4). Coefficient of Determination (R-Square) The coefficient of determination for this study is shown in Table 5. The R-squared values indicate that the model Figure 1. Structural Equation Mode

explains 25.2% of the variance in SH and 25.6% in FWB, with slightly lower adjusted R-squared values (SH = 0.238; FWB = 0.237). Structural Model Hypothesis testing in this study was conducted using a twotailed approach within the structural equation modelling framework. PLS-SEM analysis in this study uses a two-tailed hypothesis test with a significance of 5%. The significance of Table 3. Discriminant Validity using Fornell-Lacker Criterion FL FTP FWB RT SH FL 0.873 FTP 0.131 0.757 FWB 0.252 0.407 0.737 RT 0.754 0.060 0.262 0.903 SH 0.443 0.260 0.362 0.405 0.744 Sources: survey data, 2025 Table 4. Discriminant Validity using HTMT FL FTP FWB RT SH FL FTP 0.182 FWB 0.289 0.561 RT 0.826 0.099 0.303 SH 0.596 0.416 0.516 0.543 Sources: survey data, 2025 Table 5. Coefficient of Determination R-square R-square adjusted SH 0.252 0.238 FWB 0.256 0.237 Sources: survey data, 2025

the path coefficients was assessed through the bootstrapping procedure, which involves repeated sampling to generate empirical t-values and p-values. A path coefficient was considered statistically significant if the t-value exceeded 1.96and the p-value was below 0.05, corresponding to a 95% confidence level under a two-tailed test. Figure 1 presents the structural equation model illustrating the path coefficients among the research constructs, along with the R² values for each endogenous variable. Furthermore, the PLS-SEM analysis results are presented in Table 6, which includes the coefficients, t-statistics, pvalues, and confidence intervals for the hypothesized path. Among the three paths that predict SH, two emerged as statistically significant. FL showed a meaningful positive effect on SH (β = 0.277, t = 2.692, p = 0.007; CI [0.079, 0.478]), providing clear support to this hypothesis. FTP similarly returned a significant positive effect (β = 0.213, t = 3.107, p = 0.002; CI [0.077, 0.348]), reinforcing the notion that a longer-term orientation is associated with stronger saving behaviour. Meanwhile, RT fell short of the significance threshold (β = 0.184, t = 1.763, p = 0.078; CI [−0.023, 0.385]). Although the effect was positive in direction, the confidence interval crossed zero, so this hypothesis was therefore not supported. Furthermore, among the four paths directed at FWB, two were statistically significant. FTP is the strongest predictor of FWB (β = 0.344, t = 4.515, p < 0.001; CI [0.197, 0.495]), with a notably narrow confidence interval that fully excluded zero, pointing to a robust and consistent positive effect. SH also contributed meaningfully to FWB (β = 0.210, t = 2.308, p = 0.021; CI [0.025, 0.383]), suggesting that habitual saving behaviour translates into improved financial outcomes. By contrast, RT did not reach significance (β = 0.161, t = 1.342, p = 0.180; CI [−0.072, 0.398]), and direct effect of FL on FWB (β = −0.007, t = 0.066, p = 0.947; CI [−0.219, 0.203]). Indirect effect To assess mediation, indirect effects were estimated through bootstrapping with 5,000 subsamples using biascorrected accelerated confidence intervals. An indirect effect was considered significant when the 95% confidence interval. The results are presented in Table 7. The indirect effect of FL on FWB through SH was positive (β = 0.058, t = 1.557, p = 0.119; 95% BCa CI [0.003, 0.147]), though the p-value fell short of the 0.05 threshold. The lower bound of the confidence interval barely cleared zero at 0.003. Under the two-tailed criterion adopted in this study, the Table 6. Structural Equation Model Analysis (Direct Effect) Direct Effect Original sample (O)/ β Sample mean Standard deviation (STDEV) 2.50% 97.50 % FL → SH 0.277 0.276 0.103 2.692 0.007 0.079 0.478 RT → SH 0.184 0.187 0.104 1.763 0.078 -0.023 0.385 FTP → SH 0.213 0.218 0.069 3.107 0.002 0.077 0.348 FL → FWB -0.007 -0.008 0.108 0.066 0.947 -0.219 0.203 RT → FWB 0.161 0.163 0.120 1.342 0.180 -0.072 0.398 FTP → FWB 0.344 0.352 0.076 4.515 0.000 0.197 0.495 SH → FWB 0.210 0.210 0.091 2.308 0.021 0.025 0.383 Sources: survey data, 2025 (M) Table 7. Structural Equation Model Analysis (Indirect Effect) Indirect Effect Original sample (O)/ β Sample mean (M) Standard deviation (STDEV) 2.50% 97.50% FL → SH → FWB 0.058 0.06 0.037 1.557 0.119 0.003 0.147 RT→ SH → FWB 0.045 0.046 0.025 1.789 0.074 0.004 0.102 FTP → SH → FWB 0.039 0.038 0.028 1.391 0.164 -0.008 0.101 Sources: survey data, 2025

mediation hypothesis for this path is not supported. Then, the indirect effect of RT on FWB through SH was positive but did not reach significance (β = 0.045, t = 1.789, p = 0.074; 95% BCa CI [0.004, 0.102]). Although the confidence interval did not cross zero and the p-value still exceeded 0.05, leaving the indirect effect statistically non-significant under the criteria that applied. Finally, the indirect effect of FTP on FWB through SH was positive (β = 0.039, t = 1.391, p = 0.164; 95% BCa CI [−0.008, 0.101]). Since the confidence interval crossed zero, there is insufficient evidence to support mediation along this path. The significant effect of FL on SH suggests that individuals with stronger financial knowledge are more inclined to save. A clearer understanding of how money works appears to support more deliberate budgeting. This finding aligns with a previous study by Lusardi & Streeter (2023) that FL underpins decisionmaking capabilities and influences behaviours such as saving and bill-paying discipline. Then, individuals who take future needs are more inclined to set money aside rather than spend it now. Those with a longer time horizon tend to be better at deferring immediate consumption in favour of future security. This finding is supported by Van Raaij et al. (2023), who mentioned that saving behaviour is the intersection of cognitive capability and motivational orientation. Another study by Badrudin et al. (2025) also stated that those who plan for the future tend to have managed financial behaviour. The non-significant effect of RT on SH is theoretically coherent. RT is more closely associated with investmentrelated decisions, such as selecting asset classes and tolerating market volatility, than with saving behaviour. Saving is largely routine and low-risk and does not require individuals to evaluate uncertain outcomes. This result is supported by previous research by Alfando et al. (2025) among Indonesian millennials, which shows that RT did not significantly influence retirement saving behaviour. For FWB, the findings indicate that FTP was the strongest predictor. This result is important because it suggests that individuals who possess a stronger future orientation are more likely to experience better FWB. A future-oriented person may be more careful in spending, more committed to financial goals, and more prepared for unexpected financial needs. This result is supported by previous research by Hoffmann & Risse (2020), which mentioned that changes in FTP will lead to better economic welfare. Then, the rise in personal life expectancy will necessitate individuals adjusting their savings to secure their financial stability later in life (Ostrovsky-Berman & Litwin, 2019). Confidence Interval T statistics (|O/STDEV| P values ) Confidence Interval T statistics (|O/STDEV|) P values The finding that SH significantly influenced FWB also confirms the importance of saving behaviour as a practical financial behaviour that contributes to better financial outcomes. Individuals who set aside resources today are effectively managing future uncertainty, adjusting their savings patterns in anticipation of financial needs that lie ahead. FWB is less a product of circumstances alone and more a reflection of sustained, deliberate financial habits that accumulated over time. This result is supported by previous research by Ostrovsky-Berman & Litwin (2019),which mentioned that individuals adjust their savings to secure their financial stability later in life. Furthermore, FL didn’t have a significant effect on FWB. The result suggests that possessing financial knowledge does not automatically translate into improved financial outcomes unless that knowledge is actively applied through concrete behaviours. Understanding how budgeting or saving works, for instance, means little if it never shapes how a person manages their money. The results imply that financial wellbeing is not merely the result of financial knowledge. The results are supported by previous research by Sabri et al. (2024), which mentioned that FL was found to have no significant influence on FWB. In this study, SH was not proven to be a mediator between FL, RT, FTP, and FWB. This is because SH is a necessity in the context of urban workers in big cities like Jakarta. Financial behaviour is not always initiated by intention, but rather by conditions that require them to survive. Saving is considered a culture taught by parents, without learning the importance of understanding financial concepts. Therefore, this context explains why SH does not mediate FL and FWB. Then, SH does not mediate the relationship between RT and FWB because RT does not become a foundation for someone to save. Regarding the role of SH in the relationship between FTP and FWB, it suggests that individuals who possess a stronger future orientation are more likely to experience better FWB, without having to go through the process of saving. This is because future-oriented individuals directly perceive they have made significant strides in achieving their well-being. Van Raaij et al. (2023) found that future time perspective has both direct and indirect effects on financial well-being, but the indirect effect through financial behaviour is smaller than the direct effect. Limitations and Cautions This study has several limitations. First, the crosssectional design does not allow firm conclusions about causal ordering among FL, RT, FTP, SH, and FWB. Second, the use of purposive sampling and the focus on private school teachers in one metropolitan area limit the generalizability of the findings to other occupational groups, regions, or institutional settings. Finally, the R-square values for SH and FWB suggest that important explanatory factors remain outside the current model. References Adee, S., Lau, E., & Wong, S. (2024). Mediating factors in retirement savings and well-being: a focused study on Sarawak. Cogent Economics and Finance, 12(1). https://doi.org/10.1080/23322039.2024.2422217 Aidil Fadli, J., Rusmanto, T., Kurniawan, Y., & Hutagaol-Martowidjojo, Y. (2024). The Interplay of Financial Availability, Herding Behavior, and Cryptocurrency Investment Experience Moderated by Government Policy: A Study from Indonesia. Qubahan Academic Journal, 4(4), 509– 527. https://doi.org/10.48161/qaj.v4n4a1144 Alfando, K., Njo, A., & Yuliana, O. Y. (2025). Empowering Indonesian Millennials: The Role of Financial Literacy, Goal Clarity, and Risk Tolerance in Retirement Savings. Financial Planning Review, 8(3). https://doi.org/10.1002/cfp2.70013 Alkhawaja, S. O., & Albaity, M. (2022). Retirement saving behavior: evidence from UAE. Journal of Islamic Marketing, 13(2), 265–286. https://doi.org/10.1108/JIMA-03-2020-0072 Badrudin, R., Fahlevi, M., Dahlan, S. P., Dahlan, O. P., & Dandi, M. (2025). Financial stress and its determinants in Indonesia: Exploring the moderating effects of digital knowledge, age, and gender. Journal of

Recommendations for Future Research To increase explanatory power (R²), further studies could include income & income volatility, debt burden, financial events, school support, cost-of-living pressures, self-control, locus of control, financial stress, or financial socialisation. Then test whether the model paths differ by age, work experience, number of dependents, or education level, as needed, so multigroup analysis or measurement invariance can be conducted in subsequent studies.

Conclusion

The results of this study concluded that FL and FTP were proven to influence SH, while RT was not proven to influence it. Furthermore, FTP and SH were proven to influence FWB, while FL and RT did not influence it. In this study, SH was not proven to be a mediator between FL, RT, FTP, and FWB. This is because SH is a necessity in the context of urban workers in big cities like Jakarta, so financial behaviour is not always initiated by intention, but rather by conditions that require them to survive. These findings carry several limitations. The cross-sectional design precludes causal inference, and the purposive sample of non-certified private school teachers in Jakarta constrains generalisability beyond this context. Then, the modest Rsquared values suggest that the model does not capture the full picture. Other factors, such as income stability, debt burden, household financial pressure, financial stress, and institutional support, likely contribute to FWB. Then, longitudinal designs, broader occupational samples, and a wider set of contextual variables would strengthen future efforts to explain FWB more completely.

Author Contributions

Jul Aidil Fadli contributed to the conceptualization of the study, methodology development, data collection, data analysis, and preparation of the original manuscript draft. Tegar Satya Putra contributed to data curation, manuscript review, and editing. All authors have read and approved the final version of the manuscript.

Funding

We, the authors, are grateful for the internal research grant from Universitas Esa Unggul in 2025.

Acknowledgements

The authors express their gratitude to Linna Hidayanti for her significant support and help throughout the data collection phase of this study. Open Innovation: Technology, Market, and Complexity, 11(2). https://doi.org/10.1016/j.joitmc.2025.100528 Campbell, S., Greenwood, M., Prior, S., Shearer, T., Walkem, K., Young, S., Bywaters, D., & Walker, K. (2020). Purposive sampling: complex or simple? Research case examples. Journal of Research in Nursing, 25(8), 652–661. https://doi.org/10.1177/1744987120927206 Chauhan, B., Kumar, S., Kumar, S., Kumar, M., & Tripathi, V. V. R. (2025). Evaluating the effect of financial literacy on saving behavior and financial well-being among employees of financial institutions. Accounting Research Journal, 1–22. https://doi.org/10.1108/ARJ-03-2025-0069 Fred van Raaij, W., Riitsalu, L., & Põder, K. (2023). Direct and indirect effects of self-control and future time perspective on financial well-being. Journal of Economic Psychology, 99. https://doi.org/10.1016/j.joep.2023.102667 Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. In European Business Review (Vol. 31, Number 1, pp. 2–24). Emerald Group Publishing Ltd. https://doi.org/10.1108/EBR-11-2018-0203 Hamdallah, F. (2026). Unveiling the dynamics of financial coping behaviors and subjective well-being amid income instability: Evidence from public civil servants in the education sector in Palestine. Social Sciences & Humanities Open, 13, 102740. https://doi.org/10.1016/j.ssaho.2026.102740 Harahap, S., Thoyib, A., Sumiati, S., & Djazuli, A. (2022). The Impact of Financial Literacy on Retirement Planning with Serial Mediation of Financial Risk Tolerance and Saving Behavior: Evidence of Medium Entrepreneurs in Indonesia. International Journal of Financial Studies, 10(3). https://doi.org/10.3390/ijfs10030066 Hemrajani, P., Rajni, Khan, M., & Dhiman, R. (2023). Financial risk tolerance: A review and research agenda. European Management Journal, 41(6), 1119–1133. https://doi.org/10.1016/j.emj.2023.10.004 Hermansson, C., & Jonsson, S. (2021). The impact of financial literacy and financial interest on risk tolerance. Journal of Behavioral and Experimental Finance, 29. https://doi.org/10.1016/j.jbef.2020.100450 Hoffmann, A. O. I., & Risse, L. (2020). Do good things come in pairs? How personality traits help explain individuals’ simultaneous pursuit of a healthy lifestyle and financially responsible behavior. Journal of Consumer Affairs, 54(3), 1082–1120. https://doi.org/10.1111/joca.12317 Kanth, D., Sinha, A. R., & Mahato, J. (2026). Beyond urban boundaries: understanding financial literacy, financial well-being and financial behaviour of rural women entrepreneurs in India. International Journal of Social Economics, Vol. 53 No. 4, 654–669. https://doi.org/10.1108/IJSE Kempson, E., Finney, A., & Poppe, C. (2017). Financial Well-Being A Conceptual Model and Preliminary Analysis Final edition. www.hioa.no/sifo Kock, N. (2015). Common method bias in PLS-SEM: A full collinearity assessment approach. In International Journal of e-Collaboration (Vol. 11, Number 4). Kooij, D. T. A. M., Betts, M., Kanfer, R., & Rudolph, C. W. (2018). Supplemental Material for Future Time Perspective: A Systematic Review and MetaAnalysis. Journal of Applied Psychology. https://doi.org/10.1037/apl0000306.supp Laschowski, G., & O’Reilly, N. (2025). The financial well-being of retired professional athletes: debunking the myth about financial irresponsibility. Sport, Business and Management: An International Journal, 1–19. https://doi.org/10.1108/sbm-01-2025-0025 Liu, F., & Yang, Z. (2026). Effects of salaries on the well-being of university teachers in China. International Journal of Educational Development. https://doi.org/10.1016/j.ijedudev.2026.103537 Lusardi, A., & Streeter, J. L. (2023). Financial literacy and financial well-being: Evidence from the US. Journal of Financial Literacy and Wellbeing, 1(2), 169–198. https://doi.org/10.1017/flw.2023.13 Margasari, N., Murti Andhini, M., & Seetha Bandara, R. A. (2024). Student Financial Well-being: Personal Factors and Financial Behavior as Antecedents of Financial Well-being. Jurnal Economia, 20(2). https://doi.org/10.21831/economia.v20i1.58070 Mawadah, B., & Sumiati, S. (2025). Determinants of financial well-being in young adults’ quarter-life crisis. Jurnal Ekonomi Dan Bisnis, 28(Oktober), 423–446. Muat, S., Mahdzan, N. S., Sukor, M. E. A., Fachrurrozi, F., & Sari, N. (2025). The capability. International Journal of Consumer Studies, 40(6), 712–721. https://doi.org/10.1111/ijcs.12285 role of financial inclusion, financial literacy and digital payment adoption

in Indonesian millennials’ financial well-being. International Journal of Bank Marketing, 1–31. https://doi.org/10.1108/IJBM-01-2024-0022 Netemeyer, R., Warmath, D., Fernandes, D., & Lynch, J. (2017). How Am I Doing? Perceived Financial Well-Being, Its Potential Antecedents, and Its Relation to Overall Well-Being. Nguyen, L. T. M., Nguyen, P. T., Tran, Q. N. N., & Trinh, T. T. G. (2022). Why does subjective financial literacy hinder retirement saving? The mediating roles of risk tolerance and risk perception. Review of Behavioral Finance, 14(5), 627–645. https://doi.org/10.1108/RBF-03-2021-0036 Oberrauch, L., Kaiser, T., & Seeber, G. (2023). Measuring economic competence of youth with a short scale. Journal of Economic Psychology, 97. https://doi.org/10.1016/j.joep.2023.102633 Ostrovsky-Berman, E., & Litwin, H. (2019). Social Network and Financial Risk Tolerance Among Investors Nearing and During Retirement. Journal of Family and Economic Issues, 40(2), 237–249. https://doi.org/10.1007/s10834-018-9592-5 Owusu, G. M. Y., Korankye, G., Yankah, N. Y. M., & Agyekum Donkor, J. B. (2023). Financial risk tolerance and its determinants: The perspective of personnel from security services in Ghana. Borsa Istanbul Review, 23(4), 852–864. https://doi.org/10.1016/j.bir.2023.02.005 Paneda, M. G. P., & Albay, E. M. (2025). Financial literacy of secondary school teachers in the Department of Education–Division of La Union. International Journal of Evaluation and Research in Education, 14(4), 2521–2529. https://doi.org/10.11591/ijere.v14i4.32038 Pradhan, R. P., Arvin, M. B., Nair, M. S., Hall, J. H., & Bennett, S. E. (2021). Sustainable economic development in India: The dynamics between financial inclusion, ICT development, and economic growth. Technological Forecasting and Social Change, 169. https://doi.org/10.1016/j.techfore.2021.120758 Radianto, W. E. D., Sanjaya, E. L., Samosir, F., Salim, I. R., & Radianto, D. C. (2026). The role of psychological mechanisms in enhancing students’ financial wellbeing: An integration of financial literacy and the conservation of resources theory. Decision Science Letters, 15(1), 103– 114. https://doi.org/10.5267/j.dsl.2025.10.006 Sabri, M. F., Anthony, M., Law, S. H., Rahim, H. A., Burhan, N. A. S., & Ithnin, M. (2024). Impact of financial behaviour on financial well-being: evidence among young adults in Malaysia. Journal of Financial Services Marketing, 29(3), 788–807. https://doi.org/10.1057/s41264-023-00234-8 Setiawan, H., & Iramani, R. (2023). Financial Well-Being Model for Bank Employees: the Role of Financial Behavior as a Mediator. Journal of Economics, Business, & Accountancy Ventura, 26(2), 205–219. https://doi.org/10.14414/jebav.v26i2.3948 Shawn Brayman, by, Financial Planner John Grable, C. E., Paul Griffin, C., & Michael Finke, C. (2017). Assessing a Client’s Risk Profile: A Review of Solution Providers (Vol. 71, Number 1). Song, C. L., Pan, D., Ayub, A., & Cai, B. (2023). The Interplay Between Financial Literacy, Financial Risk Tolerance, and Financial Behaviour: The Moderator Effect of Emotional Intelligence. Psychology Research and Behavior Management, 16, 535–548. https://doi.org/10.2147/PRBM.S398450 Xiao, J. J., & O’Neill, B. (2016). Consumer financial education and financia

References

Wong, S. (2024). Mediating factors in retirement savings and well-being: a focused study on Sarawak. Cogent Economics and Finance, 12(1). https://doi.org/10.1080/23322039.2024.2422217 Aidil

Martowidjojo, Y. (2024). The Interplay of Financial Availability, Herding Behavior, and Cryptocurrency Investment Experience Moderated by Government Policy: A Study from Indonesia. Qubahan Academic Journal, 4(4), 509– 527. https://doi.org/10.48161/qaj.v4n4a1144

Yuliana, O. Y. (2025). Empowering Indonesian Millennials: The Role of Financial Literacy, Goal Clarity, and Risk Tolerance in Retirement Savings. Financial Planning Review, 8(3). https://doi.org/10.1002/cfp2.70013

Albaity, M. (2022). Retirement saving behavior: evidence from UAE. Journal of Islamic Marketing, 13(2), 265–286. https://doi.org/10.1108/JIMA-03-2020-0072

Dandi, M. (2025). Financial stress and its determinants in Indonesia: Exploring the moderating effects of digital knowledge, age, and gender. Journal of Recommendations for Future Research To increase explanatory power (R²), further studies could include income & income volatility, debt burden, financial events, school support, cost-of-living pressures, self-control, locus of control, financial stress, or financial socialisation. Then test whether the model paths differ by age, work experience, number of dependents, or education level, as needed, so multigroup analysis or measurement invariance can be conducted in subsequent studies. Conclusion The results of this study concluded that FL and FTP were proven to influence SH, while RT was not proven to influence it. Furthermore, FTP and SH were proven to influence FWB, while FL and RT did not influence it. In this study, SH was not proven to be a mediator between FL, RT, FTP, and FWB. This is because SH is a necessity in the context of urban workers in big cities like Jakarta, so financial behaviour is not always initiated by intention, but rather by conditions that require them to survive. These findings carry several limitations. The cross-sectional design precludes causal inference, and the purposive sample of non-certified private school teachers in Jakarta constrains generalisability beyond this context. Then, the modest Rsquared values suggest that the model does not capture the full picture. Other factors, such as income stability, debt burden, household financial pressure, financial stress, and institutional support, likely contribute to FWB. Then, longitudinal designs, broader occupational samples, and a wider set of contextual variables would strengthen future efforts to explain FWB more completely. Author contributions Jul Aidil Fadli contributed to the conceptualization of the study, methodology development, data collection, data analysis, and preparation of the original manuscript draft. Tegar Satya Putra contributed to data curation, manuscript review, and editing. All authors have read and approved the final version of the manuscript. Funding We, the authors, are grateful for the internal research grant from Universitas Esa Unggul in 2025. Acknowledgements The authors express their gratitude to Linna Hidayanti for her significant support and help throughout the data collection phase of this study. Open Innovation: Technology, Market, and Complexity, 11(2). https://doi.org/10.1016/j.joitmc.2025.100528

Walker, K. (2020). Purposive sampling: complex or simple? Research case examples. Journal of Research in Nursing, 25(8), 652–661. https://doi.org/10.1177/1744987120927206

Tripathi, V. V. R. (2025). Evaluating the effect of financial literacy on saving behavior and financial well-being among employees of financial institutions. Accounting Research

Journal, 1–22. https://doi.org/10.1108/ARJ-03-2025-0069 Fred van

Põder, K. (2023). Direct and indirect effects of self-control and future time perspective on financial well-being. Journal of Economic

Psychology, 99. https://doi.org/10.1016/j.joep.2023.102667

Ringle, C. M. (2019). When to use and how to report the results of PLS-SEM. In European Business Review (Vol. 31, Number 1, pp. 2–24). Emerald Group Publishing Ltd. https://doi.org/10.1108/EBR-11-2018-0203

Hamdallah, F. (2026). Unveiling the dynamics of financial coping behaviors and subjective well-being amid income instability: Evidence from public civil servants in the education sector in Palestine. Social

Open, 13, 102740. https://doi.org/10.1016/j.ssaho.2026.102740

Djazuli, A. (2022). The Impact of Financial Literacy on Retirement Planning with Serial Mediation of Financial Risk Tolerance and Saving Behavior: Evidence of Medium Entrepreneurs in Indonesia. International Journal of Financial Studies, 10(3). https://doi.org/10.3390/ijfs10030066

Dhiman, R. (2023). Financial risk tolerance: A review and research agenda. European Management Journal, 41(6), 1119–1133. https://doi.org/10.1016/j.emj.2023.10.004

Jonsson, S. (2021). The impact of financial literacy and financial interest on risk tolerance. Journal of Behavioral and Experimental

Finance, 29. https://doi.org/10.1016/j.jbef.2020.100450

Risse, L. (2020). Do good things come in pairs? How personality traits help explain individuals’ simultaneous pursuit of a healthy lifestyle and financially responsible behavior. Journal of Consumer Affairs, 54(3), 1082–1120. https://doi.org/10.1111/joca.12317

Mahato, J. (2026). Beyond urban boundaries: understanding financial literacy, financial well-being and financial behaviour of rural women entrepreneurs in India. International Journal of Social

Economics, Vol. 53 No. 4, 654–669. https://doi.org/10.1108/IJSE

Poppe, C. (2017). Financial Well-Being A Conceptual Model and Preliminary Analysis Final edition. www.hioa.no/sifo

Kock, N. (2015). Common method bias in PLS-SEM: A full collinearity assessment approach. In International Journal of e-Collaboration (Vol. 11, Number 4).

Rudolph, C. W. (2018). Supplemental Material for Future Time Perspective: A Systematic Review and MetaAnalysis. Journal of Applied Psychology. https://doi.org/10.1037/apl0000306.supp

Reilly, N. (2025). The financial well-being of retired professional athletes: debunking the myth about financial irresponsibility.

Sport, Business and Management: An International

Journal, 1–19. https://doi.org/10.1108/sbm-01-2025-0025

Yang, Z. (2026). Effects of salaries on the well-being of university teachers in China. International Journal of Educational Development. https://doi.org/10.1016/j.ijedudev.2026.103537

Streeter, J. L. (2023). Financial literacy and financial well-being: Evidence from the US. Journal of Financial Literacy and Wellbeing, 1(2), 169–198. https://doi.org/10.1017/flw.2023.13

Bandara, R. A. (2024). Student Financial Well-being: Personal Factors and Financial Behavior as Antecedents of Financial Well-being. Jurnal Economia, 20(2). https://doi.org/10.21831/economia.v20i1.58070

Sumiati, S. (2025). Determinants of financial well-being in young adults’ quarter-life crisis. Jurnal Ekonomi Dan Bisnis, 28(Oktober), 423–446.

Sari, N. (2025). The capability. International Journal of Consumer Studies, 40(6), 712–721. https://doi.org/10.1111/ijcs.12285 role of financial inclusion, financial literacy and digital payment adoption in Indonesian millennials’ financial well-being. International Journal of Bank

Marketing, 1–31. https://doi.org/10.1108/IJBM-01-2024-0022

Lynch, J. (2017). How Am I Doing? Perceived Financial

Antecedents, and Its Relation to Overall Well-Being.

Trinh, T. T. G. (2022). Why does subjective financial literacy hinder retirement saving? The mediating roles of risk tolerance and risk perception. Review of Behavioral Finance, 14(5), 627–645. https://doi.org/10.1108/RBF-03-2021-0036

Seeber, G. (2023). Measuring economic competence of youth with a short scale. Journal of Economic

Psychology, 97. https://doi.org/10.1016/j.joep.2023.102633

Litwin, H. (2019). Social Network and Financial Risk Tolerance Among Investors Nearing and During Retirement. Journal of Family and Economic Issues, 40(2), 237–249. https://doi.org/10.1007/s10834-018-9592-5

Donkor, J. B. (2023). Financial risk tolerance and its determinants: The perspective of personnel from security services in Ghana. Borsa Istanbul Review, 23(4), 852–864. https://doi.org/10.1016/j.bir.2023.02.005

Albay, E. M. (2025). Financial literacy of secondary school teachers in the Department of Education–Division of La Union. International Journal of Evaluation and Research in Education, 14(4), 2521–2529. https://doi.org/10.11591/ijere.v14i4.32038

Bennett, S. E. (2021). Sustainable economic development in India: The dynamics between financial inclusion, ICT development, and economic growth. Technological Forecasting and Social

Change, 169. https://doi.org/10.1016/j.techfore.2021.120758

Radianto, D. C. (2026). The role of psychological mechanisms in enhancing students’ financial wellbeing: An integration of financial literacy and the conservation of resources theory. Decision Science Letters, 15(1), 103– 114. https://doi.org/10.5267/j.dsl.2025.10.006

Ithnin, M. (2024). Impact of financial behaviour on financial well-being: evidence among young adults in Malaysia. Journal of Financial Services Marketing, 29(3), 788–807. https://doi.org/10.1057/s41264-023-00234-8

Iramani, R. (2023). Financial Well-Being Model for Bank Employees: the Role of Financial Behavior as a Mediator. Journal of Economics, Business, & Accountancy Ventura, 26(2), 205–219. https://doi.org/10.14414/jebav.v26i2.3948 Shawn

Brayman, by, Financial Planner John

Finke, C. (2017). Assessing a Client’s Risk Profile: A Review of Solution Providers (Vol. 71, Number 1).

Cai, B. (2023). The Interplay Between Financial Literacy, Financial Risk Tolerance, and Financial Behaviour: The Moderator Effect of Emotional Intelligence. Psychology Research and Behavior

Management, 16, 535–548. https://doi.org/10.2147/PRBM.S398450

Neill, B. (2016). Consumer financial education and financial