From Financial Literacy to Investment Intention: The Sequential Roles of Risk Perception and Trust

Abstract

The relationship between financial literacy and capital market participation remains a central focus of both theoretical and empirical research in behavioral finance. However, existing research has predominantly relied on direct-effect, mediation, or moderation frameworks, thereby offering only a partial understanding of how individuals make investment decisions under uncertainty. To address this limitation, this study develops a sequential cognitive–affective framework by integrating the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH). Within this framework, investment intention is conceptualized as the outcome of cognitive evaluations and affective responses, with financial literacy influencing these processes by shaping perceived risk and institutional trust. Utilizing a multistage sampling strategy, survey data were collected from 449 individual investors and analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results indicate that financial literacy is positively associated with investment intention, both directly and indirectly through a sequential mediation pathway. Specifically, higher financial literacy is associated with lower perceived risk, which subsequently strengthens trust in financial institutions and ultimately increases investment intention. These findings suggest that financial literacy functions not only as a cognitive resource but also as a psychological mechanism that influences how individuals interpret and respond to financial uncertainty. By validating a sequential cognition–affect pathway, this study provides a more comprehensive behavioral explanation for the inconsistent findings reported in prior research. The findings further suggest that financial literacy initiatives designed to address risk perceptions and institutional trust may be more effective in promoting capital market participation than programs focused solely on information provision.

1. Introduction

Capital markets play a fundamental role in modern economic systems by facilitating the efficient allocation of capital, strengthening corporate governance, and supporting sustainable economic growth (Beck et al., 2000; Levine & Zervos, 1998; Rajan & Zingales, 1998). Well-functioning capital markets also support effective price discovery and risk sharing, helping ensure that asset prices reflect both firm-level fundamentals and broader macroeconomic conditions (Beck & Levine, 2004; Levine, 2002; Fama, 1970). Despite these important functions, retail investor participation in capital markets remains limited in many countries (Arrondel et al., 2015; van Rooij et al., 2011). This challenge is especially evident in emerging economies, where the rapid expansion of digital financial services and supporting infrastructure has not been accompanied by a corresponding increase in retail investor participation (Demirguc-Kunt et al., 2022; Pazarbasioglu et al., 2020).

In advanced economies, higher levels of capital market participation are often supported by institutional mechanisms such as automatic enrolment policies, employer-sponsored retirement savings programs, and comprehensive financial inclusion initiatives (OECD, 2023; Guiso & Sodini, 2013; Choi et al., 2004; Madrian & Shea, 2001). In contrast, emerging economies generally experience lower participation rates and wider disparities in market access (Raut, 2020; Bongini et al., 2018). In Indonesia, for example, retail investors account for less than 5% of the total population despite substantial progress in financial digitalization and market accessibility (IDX, 2023). Moreover, investor participation remains concentrated among younger, urban, and highly educated individuals (OJK, 2023). This pattern is consistent with prior studies that highlight persistent disparities in capital market participation across socioeconomic groups (Arrondel et al., 2015; van Rooij et al., 2011).

Historically, limited participation in capital markets has been attributed primarily to structural and institutional barriers, including high transaction costs, limited access to financial products, regulatory constraints, and weak investor protection mechanisms (Guiso et al., 2003; Vissing-Jorgensen, 2002; Haliassos & Bertaut, 1995). While these barriers remain important, recent developments in behavioral finance have underscored the growing role of psychological factors in shaping how individuals interpret and respond to financial uncertainty (Hirshleifer, 2020; Barberis et al., 2018; Guiso & Sodini, 2013). Within this perspective, financial literacy, perceived risk, and institutional trust have emerged as important determinants of investment behavior, influencing how individuals evaluate investment opportunities, process uncertainty, and engage with financial markets (Georgarakos & Pasini, 2011; van Rooij et al., 2011; Guiso et al., 2008).

Financial literacy has been widely associated with greater market participation, enhanced portfolio diversification, and more effective long-term financial planning (Bucher-Koenen et al., 2017; Lusardi & Mitchell, 2014; van Rooij et al., 2011). Nevertheless, empirical findings remain mixed regarding the magnitude and consistency of this relationship. While some studies report a strong positive association between financial literacy and market participation, others suggest that this relationship depends on institutional, contextual, and psychological factors, particularly in emerging economies (Stolper & Walter, 2017; Fernandes et al., 2014). These divergent findings suggest that financial literacy serves not only as an informational resource but also as a behavioral mechanism that shapes how individuals interpret and respond to financial uncertainty.

Among the psychological factors associated with investment behavior, perceived risk and institutional trust are particularly important. Perceived risk reflects an individual’s subjective appraisal of uncertainty and the potential for financial loss, which may subsequently evoke affective responses (E. U. Weber et al., 2002; Loewenstein et al., 2001; Slovic, 1987). Consequently, higher levels of perceived risk are generally associated with lower participation in capital markets, especially in environments characterized by information asymmetry and market volatility (M. Weber et al., 2013; Guiso et al., 2003; Vissing-Jorgensen, 2002). Meanwhile, prior studies indicate that institutional trust helps mitigate perceived complexity and uncertainty, thereby fostering participation even in the presence of incomplete information (Georgarakos & Pasini, 2011; Guiso et al., 2008). Previous studies suggest that institutional trust may facilitate investment participation even when individuals face informational constraints, whereas low levels of trust can reduce the extent to which financial knowledge translates into investment behavior (Balloch et al., 2015; Christelis et al., 2010; Guiso et al., 2008).

Despite growing scholarly attention, the literature remains fragmented in two important respects. First, a substantial body of literature investigates financial literacy, perceived risk, and trust as separate or parallel determinants of investment behavior, frequently presuming their effects function independently and neglecting possible interrelationships among these variables (Mandić et al., 2026; Stolper & Walter, 2017; Lusardi & Mitchell, 2014). Second, studies investigating mediation mechanisms frequently treat perceived risk and trust as independent pathways rather than as interconnected components of a sequential cognition–affect process (Balloch et al., 2015; Georgarakos & Pasini, 2011; Guiso et al., 2008). Consequently, how cognitive appraisals of uncertainty shape subsequent affective responses during investment decision-making remains insufficiently understood.

Behavioral decision-making theories provide a useful foundation for addressing these limitations. The Theory of Planned Behavior (TPB) emphasizes the roles of cognitive evaluation and perceived behavioral control in forming behavioral intentions (Ajzen, 1991). Accordingly, TPB has been widely applied in financial research to explain how individuals form investment intentions under uncertain conditions (Raut, 2020; East, 1993). Complementing this perspective, the Risk-as-Feelings Hypothesis (RFH) posits that affective responses, including fear, anxiety, and trust, may influence decision-making either concurrently with or prior to deliberate cognitive processing (Loewenstein et al., 2001). Recent studies in behavioral finance further indicate that investment decisions are shaped by the interaction between cognitive and affective processes rather than by purely rational analysis alone (Slovic et al., 2004; Finucane et al., 2000), with recent evidence documenting how emotions and cognitive framing jointly influence financial risk-taking and investment choices (Brooks et al., 2023; Cantarella et al., 2023).

To address these theoretical and empirical gaps, this study develops and empirically tests a sequential cognition–affect framework in which financial literacy influences investment intention through perceived risk and trust. The proposed model suggests that higher levels of financial literacy diminish perceived risk, thereby reinforcing trust in financial institutions and ultimately augmenting investment intent (Lusardi & Mitchell, 2014; van Rooij et al., 2011; Loewenstein et al., 2001). Unlike previous studies that treat perceived risk and trust as independent or parallel mechanisms (Balloch et al., 2015; Georgarakos & Pasini, 2011; Guiso et al., 2008), this study argues that cognitive evaluations of uncertainty precede and inform the formation of trust. The underlying premise is that when individuals perceive financial uncertainty as more understandable and manageable, they are more likely to develop trust in financial institutions and market systems (Slovic et al., 2004; Loewenstein et al., 2001). By conceptualizing investment intention as the outcome of a structured cognition–affect process, this study seeks to clarify the mechanisms by which financial literacy shapes it.

Accordingly, this study examines the following research question: To what extent does financial literacy influence investment intention through a sequential process involving perceived risk and trust? The empirical analysis focuses on Indonesia, which is currently experiencing rapid digital transformation in the financial sector while continuing to exhibit substantial disparities in capital market participation and institutional trust (Abdurrahman, 2025; IDX, 2023; OJK, 2023; Sahay et al., 2020). This context is particularly suitable for examining how cognitive understanding and affective evaluations interact to shape investment intention under conditions of uncertainty.

This study makes three principal contributions to the behavioral finance literature. First, this study integrates cognitive and affective perspectives within a sequential explanatory framework of investment intention, thereby addressing the fragmentation that arises when the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH) are examined separately. Second, it advances understanding of financial literacy by reconceptualizing perceived risk as a subjective appraisal of uncertainty that influences subsequent affective responses, rather than solely as an objective assessment of potential loss. Third, it contributes to the trust literature by conceptualizing trust as an affective response shaped by prior cognitive appraisals of uncertainty rather than as an independent determinant of investment behavior. Taken together, these contributions provide a more comprehensive behavioral explanation of investment intention and offer practical insights for policies aimed at enhancing capital market participation.

The remainder of this paper is organized as follows. Section 2 reviews the relevant literature and develops the theoretical framework and research hypotheses. Section 3 describes the research methodology, including data collection procedures, construct measurement, and analytical methods. Section 4 presents and discusses the empirical findings. Finally, Section 5 summarizes the main findings, discusses the theoretical and practical implications, and outlines directions for future research.

2. Literature Review and Hypotheses Development

2.1. Theoretical Lens

This study is grounded in two complementary theoretical perspectives: the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH). Together, these theories provide a framework for examining how cognitive evaluations and affective responses jointly shape investment intention under conditions of uncertainty.

According to TPB, behavioral intentions are influenced by attitudes, subjective norms, and perceived behavioral control (Ajzen, 1991). In financial decision-making, perceived behavioral control is particularly important because investment activities require access to relevant information as well as confidence in navigating uncertain and volatile markets (Raut, 2020; East, 1993). Financial literacy strengthens perceived behavioral control by equipping individuals with the knowledge and skills needed to evaluate financial products, interpret market information, and manage uncertainty (Lusardi & Mitchell, 2014; van Rooij et al., 2011). However, while TPB explains deliberate and reasoned behavior effectively, it provides limited insight into the emotional responses individuals may experience when facing financial uncertainty.

The RFH addresses this limitation by proposing that responses to risk involve both cognitive appraisals and emotional reactions (Loewenstein et al., 2001). Within this framework, emotions such as fear, anxiety, and reassurance can shape decisions alongside—or even before—conscious reasoning (Slovic et al., 2004; Loewenstein et al., 2001). In financial settings, higher risk perception often generates negative emotions that discourage market participation (Balloch et al., 2015; Slovic et al., 2004; Haliassos & Bertaut, 1995). Within such an environment, trust functions as a positive affective response that helps reduce perceived complexity and emotional discomfort (Balloch et al., 2015; Georgarakos & Pasini, 2011; Guiso et al., 2008).

Based on this integrated perspective, investment intention emerges through a sequential process involving both cognitive and affective mechanisms. Within this framework, financial literacy influences how individuals appraise uncertainty, which subsequently shapes perceived risk and the development of institutional trust. This integrated perspective provides the theoretical foundation for examining how cognitive understanding and affective responses jointly influence investment intention.

2.2. Financial Literacy and Intention to Invest

Financial literacy is a critical determinant of individual financial behavior because it enhances the ability to understand, evaluate, and apply financial knowledge in decision-making processes (Lusardi & Mitchell, 2014; Huston, 2010; Noctor et al., 1992). This construct includes not only fundamental concepts such as compound interest, inflation, and risk diversification, but also a broader understanding of financial instruments, market mechanisms, and regulatory frameworks (Cossa et al., 2022; van Rooij et al., 2011; Huston, 2010). Financial literacy is therefore important in shaping how individuals interpret and respond to the complexities and uncertainties associated with financial decision-making (Kaiser et al., 2022; Fernandes et al., 2014; Hastings et al., 2013).

A substantial body of empirical research has shown that a higher level of financial literacy is associated with greater participation in capital markets, improved portfolio diversification, and more effective long-term financial planning (Kaiser et al., 2022; Bucher-Koenen et al., 2017; Dimmock et al., 2016; van Rooij et al., 2011). Nevertheless, the literature has focused predominantly on direct behavioral outcomes, providing comparatively limited insight into the psychological processes through which financial knowledge influences investment intention (Stolper & Walter, 2017; Fernandes et al., 2014).

Within TPB, the relationship between financial literacy and investment intention can be understood through perceived behavioral control (Raut, 2020; Ajzen, 1991). Perceived behavioral control refers to the extent to which individuals believe they possess the capability to act effectively under uncertain conditions. In this context, financial literacy strengthens perceived behavioral control by enhancing analytical skills and increasing confidence in managing complex financial situations (Kaiser et al., 2022; Lusardi & Mitchell, 2014; van Rooij et al., 2011).

Previous studies distinguish between objective and subjective financial literacy, with the latter often showing a stronger relationship with perceived behavioral control and investment confidence (Bellofatto et al., 2018; Allgood & Walstad, 2016; Xia et al., 2014). Individuals who perceive themselves as financially competent are more likely to view uncertainty as manageable and engage in financial decision-making, even under uncertain conditions (Allgood & Walstad, 2016; Fernandes et al., 2014; Bandura, 1997).

From a behavioral finance perspective, financial literacy reduces perceived barriers to market participation by enabling individuals to view financial decision-making as a process characterized by manageable uncertainty rather than overwhelming ambiguity (Lusardi & Mitchell, 2014; van Rooij et al., 2011; Heath & Tversky, 1991). As individuals become more capable of interpreting financial information and evaluating investment opportunities, their willingness to participate in capital market activities increases, even amid market volatility (Kaiser et al., 2022; Dimmock et al., 2016; Heath & Tversky, 1991). Taken together, these theoretical and empirical arguments suggest that financial literacy is positively associated with investment intention.

H1. 

Financial literacy is positively associated with investment intention.

2.3. Financial Literacy, Perceived Risk, and Investment Intention

Investment decisions are inherently characterized by uncertainty, making perceived risk an important determinant of how individuals interpret and respond to investment opportunities (E. U. Weber et al., 2002; Slovic, 1987). Perceived risk refers to an individual’s subjective appraisal of the likelihood and potential magnitude of financial loss (Sitkin & Weingart, 1995; Slovic, 1987; Bauer, 1960). Unlike objective risk, which is based on statistical probabilities, perceived risk reflects the subjective and psychological dimensions of uncertainty (E. U. Weber et al., 2002).

A substantial body of research suggests that higher levels of perceived risk are associated with lower investment intention and reduced participation in financial markets (M. Weber et al., 2013; Guiso et al., 2003; Vissing-Jorgensen, 2002). Individuals tend to become more cautious when potential losses are perceived as substantial or difficult to control (Loewenstein et al., 2001; Sitkin & Weingart, 1995; Kahneman & Tversky, 1979). Within the TPB framework, a higher level of risk perception can weaken perceived behavioral control by increasing uncertainty and reducing the perceived manageability of investment decisions, thereby decreasing investment intention (Raut, 2020; East, 1993; Ajzen, 1991).

Financial literacy plays an important role in shaping how individuals evaluate investment risk (Dimmock et al., 2016; Lusardi & Mitchell, 2014). In this context, highly literate individuals are generally better able to interpret market information, understand diversification strategies, and assess market volatility within the context of long-term investment decisions (Katnic et al., 2024; Kaiser et al., 2022; Stolper & Walter, 2017; van Rooij et al., 2011). As a result, they are more likely to perceive investment uncertainty as manageable and less threatening, which may reduce their subjective risk perceptions (Aren & Zengin, 2016).

Collectively, prior studies suggest that financially literate individuals are less likely to overestimate investment risk and are more likely to perceive uncertainty as understandable and manageable rather than inherently threatening (Aren & Zengin, 2016; Heath & Tversky, 1991). This perspective denotes that financial literacy contributes to a more balanced cognitive appraisal of uncertainty (Slovic et al., 2004; Loewenstein et al., 2001). Accordingly, perceived risk is expected to serve as a crucial mechanism through which financial literacy influences investment intent.

H2. 

Perceived risk mediates the association between financial literacy and investment intention.

2.4. Financial Literacy, Trust, and Investment Intention

Meanwhile, trust represents a critical psychological determinant of financial decision-making, particularly in environments characterized by complexity, uncertainty, and information asymmetry (Balloch et al., 2015; Georgarakos & Pasini, 2011; Guiso et al., 2008). In this study, trust is defined as an individual’s confidence in the reliability, integrity, and competence of financial institutions and market systems (Guiso et al., 2008; Fukuyama, 1996; Mayer et al., 1995).

From an affective perspective, trust reduces the psychological discomfort associated with making financial decisions under uncertainty (Slovic et al., 2004; Loewenstein et al., 2001). Prior studies show that individuals who have confidence in financial institutions and regulatory frameworks are more likely to invest, even when information is incomplete or control over outcomes is limited (Georgarakos & Pasini, 2011; Guiso et al., 2008). Empirical evidence further indicates that a higher level of trust is associated with greater market participation and a stronger willingness to allocate resources to financial assets (Kaustia et al., 2023; Balloch et al., 2015).

Within RFH, trust functions as a positive affective heuristic that enables individuals to act despite incomplete information and ongoing uncertainty (Slovic et al., 2002, 2004; Loewenstein et al., 2001). In this context, individuals may employ trust as a guiding mechanism when processing all relevant information proves difficult or costly, rather than relying solely on detailed cognitive evaluations (Georgarakos & Pasini, 2011; Guiso et al., 2008). Financial literacy can strengthen institutional trust by improving individuals’ understanding of investor protections, disclosure requirements, and regulatory oversight (Balloch et al., 2015; Kersting et al., 2015). Consequently, financially literate individuals are better able to assess the credibility of financial institutions and are less likely to develop distrust due to misinformation or uncertainty (Ng et al., 2016; Christelis et al., 2010; Tu & Bulte, 2010). Briefly, these arguments suggest that trust functions as an important affective mechanism through which financial literacy influences investment intention.

H3. 

Trust mediates the association between financial literacy and investment intention.

2.5. Serial Mediation of Perceived Risk and Trust

Although both perceived risk and trust influence investment intention, behavioral decision-making theories suggest that cognitive and affective processes are interdependent and often unfold sequentially (Loewenstein et al., 2001). Cognitive evaluations of uncertainty typically precede affective responses, implying that trust is partly shaped by prior cognitive appraisals of uncertainty (Slovic et al., 2004; Loewenstein et al., 2001; Sitkin & Weingart, 1995; Smith & Ellsworth, 1985).

Previous studies indicate that perceived risk and trust are closely interconnected. A higher level of uncertainty may undermine confidence within financial systems, whereas lower perceived risk fosters conducive conditions to trust establishment (Guiso et al., 2008; Siegrist et al., 2005; Das & Teng, 2004) When individuals perceive financial environments as less threatening and more understandable, their willingness to trust financial institutions and market mechanisms increases, thereby strengthening institutional trust (Georgarakos & Pasini, 2011; Heath & Tversky, 1991).

This sequential dynamic is consistent with the theoretical propositions of both the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH). Within the TPB framework, lower perceived risk enhances perceived behavioral control by making investment decisions appear more manageable (Ajzen, 1991). From the RFH perspective, reduced uncertainty fosters positive affective responses, such as trust, which subsequently encourage investment behavior even when some uncertainty remains (Loewenstein et al., 2001).

Empirical evidence further suggests that trust is less likely to develop in environments characterized by elevated volatility and perceived risk, highlighting the interdependence of cognitive and affective processes in decision-making (Kaustia et al., 2023; Kumar et al., 2024; Balloch et al., 2015). Taken together, these arguments suggest that financial literacy influences investment intention through a sequential mechanism through which cognitive appraisal of risk precedes and shapes the development of affective trust.

H4. 

Financial literacy is associated with investment intention through a sequential mediation pathway involving perceived risk and trust.

2.6. Conceptual Framework

By integrating the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH), this study proposes a conceptual framework that explains the mechanisms through which financial literacy influences investment intention. Rather than treating cognitive and affective mechanisms as independent and parallel processes, the proposed framework emphasizes their sequential and interdependent roles in investment decision-making under uncertainty. The framework incorporates four principal pathways:

• Direct pathway (H1), which represents the direct association between financial literacy and investment intention.
• Cognitive mediation pathway (H2), which represents the indirect association between financial literacy and investment intention through perceived risk.
• Affective mediation pathway (H3), which represents the indirect association between financial literacy and investment intention through institutional trust.
• Sequential cognition–affect pathway (H4), which represents the sequential mediation mechanism linking financial literacy, perceived risk, institutional trust, and investment intention. This pathway is based on the premise that lower perceived risk provides a cognitive foundation for developing stronger affective trust, which in turn enhances investment intention.

By examining these pathways simultaneously, the framework captures the interdependence of cognitive evaluation and affective response in financial decision-making. The conceptual framework and corresponding hypotheses are presented in Figure 1.

3. Methodology

This study adopts a deductive research design to examine the cognitive and affective mechanisms underlying investment intention in capital markets. The proposed conceptual framework is grounded in the Theory of Planned Behavior (TPB) (Ajzen, 1991) and the Risk-as-Feelings Hypothesis (RFH) (Loewenstein et al., 2001), which collectively provide the theoretical basis for linking financial literacy, perceived risk, trust, and investment intention.

Drawing on these perspectives, the study conceptualizes investment intention as the outcome of both cognitive evaluations and affective responses. Specifically, TPB provides the rationale for examining the role of cognitive appraisal and perceived behavioral control, whereas RFH highlights the importance of affective responses in decision-making under uncertainty. Integrating these perspectives enables a more comprehensive understanding of how financial literacy influences investment intention through the sequential cognition–affect process proposed in this study.

3.1. Instrument Measurement

Investment intention is defined as an individual’s willingness to participate in capital market activities and to allocate financial resources to achieve long-term financial objectives (Raut, 2020; East, 1993; Ajzen, 1991). Consistent with prior studies, investment intention is conceptualized as a multidimensional behavioral construct encompassing market participation, investment decision-making, and long-term financial orientation (Indrawati et al., 2025; Raut, 2020; Mayfield et al., 2008). In this study, investment intention is measured using seven items adapted from existing literature, and assessed on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Financial literacy refers to an individual’s capacity to understand and evaluate capital market products, investment mechanisms, and financial information (Lusardi & Mitchell, 2014; Huston, 2010). In this study, financial literacy is measured using six items adapted from van Rooij et al. (2011) and Sivaramakrishnan et al. (2017) on a five-point Likert scale. The construct comprises two dimensions: (1) familiarity with financial products and (2) familiarity with capital markets.

Perceived risk denotes an individual’s subjective evaluation of the likelihood of financial loss and uncertainty associated with investment activities (Sitkin & Weingart, 1995; Slovic, 1987; Bauer, 1960). Following Trang and Tho (2017), Munnukka et al. (2016), and Bhukya and Singh (2015), perceived risk is measured using six items representing three dimensions encompassing financial, security, and social risks. Although measured through multiple dimensions, perceived risk is modeled as a higher-order construct representing overall uncertainty associated with investment decisions.

Trust is defined as an investor’s confidence in the reliability, integrity, and competence of financial products and financial institutions (Guiso et al., 2008; Mayer et al., 1995; Fukuyama, 1996). Following Kaustia et al. (2023) and Nguyen et al. (2016), trust is measured using six items on a five-point Likert scale covering two dimensions: (1) confidence in financial products and (2) confidence in financial institutions.

All variables are specified as reflective constructs whereby observed indicators are assumed to reflect the underlying latent variables. Accordingly, changes in the latent constructs are expected to be consistently manifested across their respective indicators, consistent with established measurement theories and recommended practices in PLS-SEM.

Table 1. Operationalized Constructs and Measurement Items.

Variable	Dimension	Items	Source
Investment Intention (Y)	Continuous Investment Intention	1. I intend to invest regularly in capital markets. 2. I intend to continue investing in capital markets.	Raut (2020)
	Strategic Decision-Making	3. I carefully evaluate investment alternatives before making investment decisions. 4. I consider both short-term and long-term consequences when making investment decisions.	Indrawati et al. (2025)
	Long-term Investment Orientation	5. I intend to allocate a portion of my income to capital market investments. 6. I intend to maintain a diversified portfolio consisting of various capital market assets (e.g., stocks, bonds, and mutual funds). 7. I intend to invest consistently in long-term capital market assets.	Mayfield et al. (2008)
Financial Literacy (X)	Product Familiarity	8. I have a good understanding of investment products available in capital markets. 9. Stocks generally exhibit a higher price fluctuation than do other investment instruments over time. 10. Mutual funds allow investors to diversify investments across multiple financial assets simultaneously.	van Rooij et al. (2011)
	Capital Market Familiarity	11. I understand how capital markets operate. 12. I possess sufficient knowledge about capital market activities and mechanisms. 13. I understand the relationship between investment risk and expected return in capital markets	Sivaramakrishnan et al. (2017)
Perceived risk (Z1)	Financial Risk	14. I am concerned about the possibility of losing money when investing in capital markets. 15. Investing in capital markets involves financial uncertainty that may negatively affect my investment outcomes.	Bhukya and Singh (2015)
	Security Risk	16. I believe that investing in capital markets involves substantial financial risk. 17. I am concerned about fraudulent activities and institutional misconducts in capital markets.	Munnukka et al. (2016)
	Social Risk	18. Investing in capital markets may negatively influence how others perceive me. 19. I worry that unsuccessful investment decisions may affect my social image among friends or relatives.	Trang and Tho (2017)
Trust (Z2)	Trust in Financial Products	20. I have confidence in financial products offered in capital markets. 21. I trust the reputation of listed companies in capital markets. 22. I believe that capital market investment products are generally reliable for investment purposes.	Nguyen et al. (2016)
	Trust in Financial Institutions	23. I have a high level of trust in financial institutions operating within capital markets. 24. I have confidence in the regulators overseeing capital markets. 25. I believe that the government will protect investors participating in capital markets.	Kaustia et al. (2023)

summarizes the operational definitions, measurement items, and literature sources for each construct.

3.2. Sample Selection and Data Collection

The target population of this study comprises retail investors in the Indonesian capital markets. Given the large population of approximately 11.69 million registered investor accounts and their uneven geographical distribution (KSEI, 2023), a multistage sampling strategy is employed to improve the geographic representation of respondents across the principal regions of Indonesia.

Previous studies have shown that income level and financial capacity significantly affect participation in risky financial assets. Individuals with higher incomes generally face fewer liquidity constraints, and are therefore more likely to participate in capital market activities (Guiso et al., 2003; van Rooij et al., 2011; Vissing-Jorgensen, 2002; Haliassos & Bertaut, 1995). Accordingly, the first stage of the sampling process established a minimum monthly income threshold of IDR10 million. This criterion is intended to minimize the influence of liquidity constraints on investment decisions, and ensure that respondents possess sufficient financial capacity to participate in capital market activities (Vissing-Jorgensen, 2002; Haliassos & Bertaut, 1995). The selected threshold is broadly consistent with prior evidence indicating that Indonesian households with monthly incomes in the range of IDR 8–10 million generally experience fewer liquidity constraints and possess greater discretionary capacity to allocate resources to financial assets, equivalent to approximately three to six times the regional minimum wage (Tresnatri et al., 2025; Wardhani & Iramani, 2023).

In the second stage, the minimum sample size required for evaluating the proposed serial mediation model was determined. An initial estimate is calculated using the Slovin formula:

$$n={\displaystyle \frac{N}{1+N{e}^2}}$$

(1)

where n is the required sample size, N is the population size, and e is the margin of error. Given that the number of registered investor accounts is 11,689,647 (KSEI, 2023), the Slovin formula with a 5% margin of error yields a minimum sample size of approximately 400 respondents. This calculation provides an initial estimate of the sample size required to obtain broad population coverage.

To further ensure methodological rigor, the minimum sample size is also evaluated against the requirements of Partial Least Squares–Structural Equation Modeling (PLS-SEM). According to the 20-time rule, the minimum sample size should exceed 20 times the number of indicators associated with the most complex construct or the largest number of structural paths directed toward a latent variable (Hair et al., 2021; Kock & Hadaya, 2018). Since the investment intention construct contains seven indicators, the recommended minimum sample size is 140 observations. Accordingly, the minimum sample size requirement ranges from 140 observations based on the PLS-SEM rule to approximately 400 observations based on the population coverage consideration.

In the third stage, a quota sampling method was implemented to determine the geographic distribution of respondents. Given the unequal distribution of investors across Indonesia, this approach was intended to improve the proportional representation across major regions of Indonesia. As shown in

Table 2. Sample Size Determination and Regional Sample Distribution. This table shows the recommended sample size for each region, based on the distribution of respondents obtained through quota sampling. Investor size (%) indicates the geographic distribution of investors, based on the KSEI’s (2023) investor distribution statistics. Meanwhile, the sample size indicates the number of respondents in each region, based on KSEI investor distribution statistics, with a minimum sample size of 400 calculated using the Slovin formula.

Region	Representative Province	Investor Size, %	Sample Size
Java	DKI Jakarta	68.50%	274
	West Java
	Special Region of Yogyakarta
Sumatra	West Sumatera	16.75%	67
	North Sumatera
Kalimantan	East Kalimantan	5.25%	21
Sulawesi	North Sulawesi	4.75%	19
Bali and Nusa Tenggara	Bali	3.50%	14
Maluku and Papua	Maluku	1.25%	5
Total		100%	400

, the quotas were established based on KSEI investor distribution statistics (KSEI, 2023): Java accounts for 68.50% of the total investor population, followed by Sumatera (16.73%), Kalimantan (5.27%), Sulawesi (4.77%), Bali (3.50%), and Maluku and Papua (1.11%). Accordingly, regional quotas were established in proportion to these percentages using the minimum sample size of 400 respondents as the allocation baseline.

Primary data for this study were collected through an online survey administered from 1 February 2025 to 31 March 2025. Participants were recruited voluntarily via an online questionnaire distributed in collaboration with several investor communities and financial platforms, including the Indopremier Investment Forum (IPOT Buzz), the Phillip Securities Indonesia Investor Forum (Smart Trader), and the IDX Regional Development Information System (RDIS). To ensure informed and voluntary participation, the questionnaire included a comprehensive consent form outlining the study’s objectives and ethical considerations.

To safeguard respondent privacy and confidentiality, the questionnaire did not collect personally identifiable information, such as names, phone numbers, or email addresses. Instead, only demographic data—including age, gender, education, occupation, annual income, and area of residence—were collected for analytical and sampling purposes. This demographic information facilitated respondent profiling and allocation in accordance with the predetermined geographic quotas. Prior to completing the main questionnaire, participants were screened using preliminary questions to verify active participation in capital market investments and compliance with the minimum income threshold.

To achieve the minimum sample size recommended by the Slovin formula (400 observations), a total of 600 questionnaires were distributed to account for potential non-responses, incomplete submissions, and invalid responses, while ensuring adequate representation across regional quotas. Of the distributed questionnaires, 482 were returned, resulting in a gross response rate of approximately 81%. Following screening and validation, 449 complete and valid responses were retained for analysis. This final sample size exceeds the minimum requirements specified by both the PLS-SEM approach (140) and the Slovin formula (400), thereby providing adequate statistical power for serial mediation analysis and the broad geographic coverage of Indonesia’s registered investor population.

Table 3. Demographic Characteristics of Respondents.

Demographic Characteristic	Frequency	Percentage	Demographic Characteristic	Frequency	Percentage
Age			Occupation
≤28	78	17	Full-Time Trader	75	16.7
29–44	196	44	Entrepreneur/Self-employed	87	19.4
≥45	175	39	Educator/Academic Professional	49	10.9
Gender			Private Sector Employee	108	24.1
Male	283	63	State-owned Enterprise Employee/Civil Servant	94	20.9
Female	166	37	Professional (e.g., Lawyer, Consultant, etc.)	36	8
Levels of Education			Domicile
Doctoral Degree	27	6	Java	317	70.6
Master’s Degree	91	20.3	Sumatera	70	15.6
Bachelor’s Degree	253	56.3	Kalimantan	25	5.6
Associate Degree or Lower	78	17.4	Sulawesi	18	4
Annual Income			Bali and Nusa Tenggara	16	3.6
IDR120-180 Million	311	69	Maluku and Papua	3	0.7
>IDR180 Million	138	31

presents the demographic characteristics of the final sample. Respondents generally possess high levels of education, adequate income, and substantial geographic diversity, thereby rendering the sample suitable for analyzing investment behavior in the Indonesian capital markets. The actual distribution of respondents closely corresponds to the regional quotas estimated using the Slovin formula, maintaining a precision range within 5% with only minor variations across regions. Consistent with prevailing investor concentration patterns, Java accounts for the largest proportion of the sample. Minor discrepancies between projected and actual quotas are attributable to differences in voluntary participation rates and survey accessibility. Overall, the final sample demonstrates broad geographic coverage and closely reflects the regional distribution of investors reported by KSEI (2023).

3.3. Data Analysis Technique

This study employs Partial Least Squares–Structural Equation Modeling (PLS-SEM) as the primary analytical technique. PLS-SEM is particularly appropriate for examining complex behavioral models involving multiple latent constructs, higher-order constructs, and indirect relationships (Hair et al., 2021; Sarstedt et al., 2017). Given the prediction-oriented nature of the study and its emphasis on testing sequential mediation relationships, PLS-SEM is considered more suitable than covariance-based structural equation modeling (Hair et al., 2019, 2021).

PLS-SEM offers several methodological advantages for our research. First, it is robust to violations of multivariate normality and produces reliable estimates with a medium-sized sample (Hair et al., 2021). Second, it enables the simultaneous assessment of measurement properties and structural relationships among latent variables (Hair et al., 2021; Sarstedt et al., 2017). Third, it is particularly effective for evaluating serial mediation models, which are central to the proposed cognition–affect framework (Hair et al., 2021; Nitzl et al., 2016).

The analysis is conducted in two stages. The first stage involves evaluating the measurement model through the assessments of reliability and validity. The second stage involves testing the structural model, including direct, indirect, and sequential mediation relationships specified in the proposed framework. However, given the cross-sectional nature of the data, the findings should be interpreted as evidence of structural associations within the theoretical model rather than as definitive evidence of causality (Maxwell & Cole, 2007).

3.4. Common Method Bias

Given that this study relies on self-reported survey data collected from a single source, common method bias (CMB) represents a potential methodological concern. Following the recommendations by Podsakoff et al. (2003), several procedural remedies are implemented to mitigate this risk. Respondents were assured of anonymity and confidentiality to reduce social desirability bias, and all measurement items were adapted from validated prior studies and randomized to minimize the likelihood that participants could infer the study’s hypotheses.

To statistically assess the presence of CMB, a full collinearity test is conducted using inner-model Variance Inflation Factor (VIF) values within the PLS-SEM framework. According to Kock (2015), a VIF value below 3.3 indicates that common method bias is unlikely to threaten the model validity. Our results show that all inner VIF values range from 1.000 to 2.038, well below the recommended threshold. Therefore, common method bias is unlikely to pose a significant threat to the validity of either the measurement model or the structural model.

4. Findings and Discussion

4.1. Measurement Model Assessment

This study operationalizes four reflective latent constructs: (1) financial literacy, (2) perceived risk, (3) trust, and (4) investment intention. The measurement model is assessed using established criteria for reliability, convergent validity, discriminant validity, and multicollinearity. The results of construct-level reliability and validity assessments are presented in

Table 4. Construct Validity and Reliability.

Construct	Item	Outer Loading	Cronbach’s Alpha	ρA	Composite Reliability	AVE	VIF
Investment Intention (Y)	Y1	0.904	0.959	0.960	0.966	0.804	4.161
	Y2	0.898					4.021
	Y3	0.894					3.735
	Y4	0.890					3.808
	Y5	0.901					4.079
	Y6	0.902					4.261
	Y7	0.886					3.85
Financial Literacy (X)	X1	0.900	0.905	0.905	0.960	0.800	3.785
	X2	0.890					3.721
	X3	0.897					3.802
	X4	0.905					3.874
	X5	0.888					3.592
	X6	0.886					3.514
Perceived Risk (Z1)	Z11	0.905	0.951	0.952	0.961	0.804	3.941
	Z12	0.902					3.917
	Z13	0.892					3.538
	Z14	0.885					3.352
	Z15	0.897					3.791
	Z16	0.899					3.846
Trust (Z2)	Z21	0.902	0.952	0.952	0.961	0.805	3.967
	Z22	0.890					3.52
	Z23	0.904					3.903
	Z24	0.902					3.823
	Z25	0.896					3.802
	Z26	0.889					3.682

.

As exhibited in Table 4, all indicator loadings exceed the recommended threshold of 0.70 (Hair et al., 2021), indicating satisfactory indicator reliability and supporting convergent validity. Internal consistency reliability is assessed using Cronbach’s alpha, rho_A (ρA), and composite reliability. The results show that all constructs exceed the recommended threshold of 0.70 (Dijkstra & Henseler, 2015; Nunnally & Bernstein, 1994). In addition, composite reliability values approach or exceed 0.90, confirming strong internal consistency reliability across all constructs (Hair et al., 2021).

Furthermore, convergent validity is evaluated using the Average Variance Extracted (AVE). Since all constructs exhibit AVE values above the recommended threshold of 0.50, our results indicate that the latent constructs explain a substantial proportion of the variances in their respective indicators (Sekaran & Bougie, 2016). These findings confirm satisfactory convergent validity. Meanwhile, potential multicollinearity is tested using the Variance Inflation Factor (VIF). All VIF values remain below the conservative threshold of 5.0 (Hair et al., 2021), with the highest observed value being 4.261. Therefore, multicollinearity is unlikely to pose a significant threat to the stability of parameter estimates.

Table 5. Discriminant Validity.

Panel A. Discriminant Validity based on Heterotrait–Monotrait Ratio of Correlations Criteria.
	X	Y	Z1	Z2
Financial Literacy (X)
Intention to Invest (Y)	0.611
Perceived Risk (Z1)	0.513	0.617
Trust (Z2)	0.628	0.678	0.664
Panel B. Discriminant Validity Based on Fornell–Larcker Criterion
	X	Y	Z1	Z2
Financial Literacy (X)	0.894
Intention to Invest (Y)	0.585	0.896
Perceived Risk (Z1)	−0.488	−0.590	0.897
Trust (Z2)	0.598	0.649	−0.632	0.897

reports the discriminant validity assessment using the Heterotrait–Monotrait Ratio (HTMT) and the Fornell–Larcker criterion. As shown in Panel A, all HTMT values are below the recommended threshold of 0.90, indicating satisfactory discriminant validity among the latent constructs (Henseler et al., 2015). Panel B shows that the square root of AVE for each construct exceeds its correlations with other constructs (Fornell & Larcker, 1981), providing additional evidence of discriminant validity. Collectively, these findings indicate that the latent constructs are empirically distinct and that the measurement model satisfies established psychometric standards, thus providing a reliable foundation for subsequent evaluation of the structural model.

4.2. Model Fit and Predictive Assessment

Upon the validation of the measurement model, the structural model is evaluated with respect to model fit, explanatory power, and predictive relevance. The Standardized Root Mean Square Residual (SRMR) value of 0.031 is well below the recommended thresholds of 0.08 and 0.10, indicating a close correspondence between observed and model-implied correlations (Henseler et al., 2014; Hu & Bentler, 1999). In addition, the Normed Fit Index (NFI) value of 0.933 exceeds the recommended threshold of 0.90, providing additional evidence of satisfactory model fit (Hair et al., 2021; Henseler et al., 2016).

Furthermore, the model’s out-of-sample predictive capability is assessed using the PLSpredict procedure (Ringle et al., 2015). As reported in

Table 6. Out-of-Sample Prediction. This table reports predictive capability classifications based on the PLSPredict assessment criteria proposed by Shmueli et al. (2019).

Construct	Indicator	PLS		LM	PLS-LM	Predictive Capability
		RMSE	Q2 Predict	RMSE	RMSE
Intention to Invest	Y1	1.027	0.272	1.031	−0.004	Moderate
	Y2	1.093	0.259	1.102	−0.009
	Y3	1.032	0.314	1.029	0.003
	Y4	1.004	0.241	1.011	−0.007
	Y5	1.062	0.285	1.066	−0.004
	Y6	1.002	0.292	1.008	−0.006
	Y7	1.015	0.231	1.014	0.001
Perceived Risk	Z11	1.023	0.173	1.029	−0.006	High
	Z12	1.045	0.17	1.051	−0.006
	Z13	0.988	0.165	0.995	−0.007
	Z14	0.988	0.208	0.994	−0.006
	Z15	0.962	0.197	0.973	−0.011
	Z16	0.945	0.212	0.953	−0.008
Trust	Z21	0.992	0.26	0.998	−0.006	High
	Z22	1.002	0.281	1.012	−0.01
	Z23	1.01	0.313	1.02	−0.01
	Z24	1.08	0.285	1.088	−0.008
	Z25	1.008	0.295	1.018	−0.01
	Z26	1.038	0.263	1.043	−0.005

, all endogenous constructs exhibit positive Q² predict values. These results indicate meaningful predictive relevance of the model. Furthermore, the comparisons between the Root Mean Square Error (RMSE) values of the PLS-SEM model and the linear regression benchmarks reveal that the PLS model performs comparably to or better than linear regression across several indicators. Following the guidelines of Shmueli et al. (2019), these results suggest moderate-to-high predictive performance.

Construct-level predictive relevance is further evaluated using the cross-validated redundancy measure (Q²). Investment intention (Q² = 0.412) and institutional trust (Q² = 0.406) demonstrate substantial predictive relevance, whereas perceived risk exhibits a lower Q² value (0.190), indicating moderate predictive relevance (Hair et al., 2019). These findings suggest that perceived risk may be affected by additional factors beyond the current framework, including individual psychological characteristics, prior investment experiences, and broader environmental uncertainty. This interpretation is consistent with behavioral finance research, which conceptualizes risk perception as a multidimensional and context-dependent construct (Mandić et al., 2026).

The explanatory power of the model is assessed using the coefficient of determination (R²). The R² values for investment intention (0.518) and institutional trust (0.509) indicate substantial explanatory power, while that for perceived risk (0.238) demonstrates moderate explanatory power (Cohen, 1992). These findings further suggest that perceived risk is affected by behavioral and situational factors beyond financial literacy alone.

Overall, the structural model demonstrates satisfactory fit, meaningful predictive relevance, and substantial explanatory power. Taken together, these findings support the adequacy of the proposed sequential cognition–affect framework for explaining retail investment intention in the Indonesian capital markets.

4.3. Structural Model Evaluation and Hypothesis Tests

The structural model is evaluated using a bootstrapping procedure with 10,000 resamples to assess the significance of conjectured relationships, as suggested by Streukens and Leroi-Werelds (2016).

Table 7. Path Coefficient Results. This table reports the estimated path coefficients obtained from the PLS-SEM analysis. Panel A presents the direct effects of the structural model, capturing the relationships among financial literacy, perceived risk, trust, and investment intention. Panel B reports the specific indirect effects, examining the mediating roles of perceived risk and trust, including the serial mediation effect. Panel C presents the total effect of financial literacy on investment intention, reflecting the combined direct and indirect effects. X denotes financial literacy, Z1 denotes perceived risk, Z2 denotes trust, and Y denotes investment intention. Statistical significances are assessed using 10,000 bootstrapping procedures. Reported values include path coefficients, bootstrapped standard errors, and t-statistics. ***, **, and * indicate significances at the 1%, 5%, and 10% levels, respectively.

	Path	Coefficient	Standard Error	T-Statistics	p-Values	Decision
Panel A. Structural Model (Direct Effect)
Hypothesis 1	X → Y	0.264	0.05	5.243	0.000 ***	Supported
	X → Z1	−0.488	0.044	11.017	0.000 ***
	X → Z2	0.379	0.051	7.471	0.000 ***
	Z1 → Z2	−0.447	0.05	8.983	0.000 ***
	Z1 → Y	−0.251	0.052	4.843	0.000 ***
	Z2 → Y	0.332	0.058	5.677	0.000 ***
Panel B. Hypothesis Testing (Mediation)
Hypothesis 2	X → Z1 → Y	0.123	0.029	4.277	0.000 ***	Supported
Hypothesis 3	X → Z2 → Y	0.126	0.03	4.235	0.000 ***	Supported
Hypothesis 4	X → Z1 → Z2 → Y	0.072	0.017	4.272	0.000 ***	Supported
Panel C. Total Effect
	X → Y	0.585	0.039	15.005	0.000 ***

reports the estimated direct, indirect, and total effects, while Figure 2 illustrates the structural relationships among operationalized constructs. Consistent with Hypothesis 1, financial literacy exhibits a positive and significant association with investment intention (β = 0.264, p < 0.001). This finding suggests that individuals with higher levels of financial literacy are more inclined to participate in capital market activities. From the TPB perspective, this relationship implies that individual confidence increases as knowledge grows. Hence, financial literacy enhances perceived behavioral control by increasing individual confidence in managing complex financial environments.

The results further reveal that financial literacy is negatively associated with perceived risk (β = −0.488, p < 0.001) and positively associated with institutional trust (β = 0.379, p < 0.001). These findings suggest that individuals with greater financial knowledge are more likely to perceive market uncertainty as understandable and manageable. Meanwhile, these capabilities concurrently foster increased trust in financial institutions, as they better comprehend how financial markets operate. In addition, perceived risk is negatively associated with institutional trust (β = −0.447, p < 0.001) and investment intention (β = −0.251, p < 0.001), indicating that higher risk perception reduces both trust and the propensity to invest. Conversely, institutional trust is positively associated with investment intention (β = 0.332, p < 0.001), highlighting the pivotal role of confidence in financial institutions in fostering investment participation.

Collectively, these findings reveal two complementary mechanisms underlying investment intention: (1) a cognitive pathway involving the evaluation of market uncertainty and (2) an affective pathway involving institutional trust. These results support the integration of the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH) by demonstrating that cognitive appraisal and affective response jointly shape investment behavior under uncertainty.

The mediation analysis further identifies several distinct indirect pathways through which financial literacy influences investment intention. Specifically, financial literacy exhibits a significant indirect effect through perceived risk (β = 0.123, p < 0.001), supporting H2. Additionally, financial literacy yields a significant indirect effect through institutional trust (β = 0.126, p < 0.001), supporting H3. More importantly, the sequential mediation pathway involving both perceived risk and institutional trust is also statistically significant (β = 0.072, p < 0.001), thereby substantiating H4. These findings indicate that financial literacy affects investment intention through a structured cognition–affect sequence, in which lower perceived risk provides the cognitive foundation for the emergence of stronger affective trust. This represents the central theoretical contribution of our study, which demonstrates that the cognitive appraisal of uncertainty precedes and shapes affective trust in the investment decision-making process.

Finally, the total effect of financial literacy on investment intention is positive and significant (β = 0.585, p < 0.001), exceeding the magnitude of the direct effect alone (β = 0.264, p < 0.001). This indicates that the influence of financial literacy extends beyond the direct provision of financial knowledge, and operates through both cognitive and affective mechanisms. Taken together, the results reinforce the perspective that financial literacy functions not only as an informational resource but also as an important behavioral factor shaping how individuals interpret and respond to uncertainty in financial decision-making (Kaiser et al., 2022; Fernandes et al., 2014; Lusardi & Mitchell, 2014).

4.4. Discussion

4.4.1. Financial Literacy as a Cognitive Foundation of Investment Intention

Our findings reveal a positive and significant association between financial literacy and investment intention, thus supporting Hypothesis 1 (β = 0.264, p < 0.001). Within the TPB framework, this relationship reflects the role of financial literacy in strengthening perceived behavioral control, which refers to individual confidence in his/her ability to act effectively under uncertainty (Ajzen, 1991). Accordingly, financial literacy operates as an important cognitive resource that enables individuals to process financial information, evaluate investment alternatives, and make decisions in complex market environments (Che Hassan et al., 2023; Kaiser et al., 2022; Lusardi & Mitchell, 2014).

Capital markets are characterized by information asymmetry, uncertainty, and volatility, all of which may discourage the participation of less-informed investors (Guiso et al., 2003; Haliassos & Bertaut, 1995). In this context, financial literacy helps mitigate these barriers by improving individual ability to interpret financial information, assess investment opportunities, and understand market mechanisms (Kaiser et al., 2022; Lusardi & Mitchell, 2014; Hastings et al., 2013). This interpretation is consistent with prior studies that demonstrate that financial knowledge reduces the cognitive costs associated with market participation, and enhances individual confidence in financial decision-making (Stolper & Walter, 2017; Grohmann et al., 2015; van Rooij et al., 2011).

Our results are also consistent with recent empirical evidence across different financial contexts. A more recent study reports that financial literacy significantly increases the intention to invest in Islamic financial products among Pakistani investors (Naqvi et al., 2025). Similarly, Irfan et al. (2025) find that financial literacy positively affects investment intention through perceived behavioral control among Indonesian retail investors. Meanwhile, Eaw et al. (2024) demonstrate that financial literacy indirectly promotes green investment intention through attitudinal mechanisms. Collectively, these studies corroborate the notion that financial literacy serves not merely as an informational asset but as a broader psychological resource that supports financial participation.

More importantly, our study extends the existing literature by demonstrating that the effect of financial literacy cannot be fully understood solely through direct-effect models. Our findings show that although the direct association between financial literacy and investment intention is statistically significant, the total effect (β = 0.585) is substantially larger than the direct effect (β = 0.264). Such findings imply that financial literacy exerts much of its influence through additional psychological mechanisms. This helps reconcile the long-standing debate regarding the effectiveness of financial literacy in shaping financial behavior. Whereas some studies have questioned the behavioral impacts of financial education when examined in isolation (Fernandes et al., 2014), our results suggest that financial literacy becomes considerably more influential when its indirect effects are taken into account in how individuals interpret and respond to uncertainty. Accordingly, financial literacy should be viewed not only as a source of financial knowledge but also as a foundational cognitive capability that initiates broader behavioral mechanisms that lead to investment participation.

4.4.2. Financial Literacy and the Cognitive Recalibration of Perceived Risk

In addition to its direct association with investment intention, financial literacy exhibits a negative and significant relation with perceived risk, thereby supporting Hypothesis 2 (β = −0.488, p < 0.001). Moreover, perceived risk partially mediates the relationship between financial literacy and investment intention (β = 0.123, p < 0.001). These findings are consistent with both the TPB (Ajzen, 1991) and the RFH (Loewenstein et al., 2001), in which perceived risk operates primarily as a cognitive appraisal of uncertainty that subsequently affects behavioral intention. In this regard, the RFH complements the TPB by explaining how emotional response to uncertainty emerges from and interacts with cognitive appraisal in the formation of investment intention.

From a cognitive perspective, our results provide empirical support for the Competence Hypothesis (Heath & Tversky, 1991), which posits that subjective risk perception declines as domain-specific competence increases. In this sense, individuals with higher levels of financial literacy are better able to interpret market dynamics, understand diversification strategies, and distinguish temporary fluctuations from fundamental market trends (Kaiser et al., 2022; Lusardi & Mitchell, 2014; van Rooij et al., 2011). Consequently, uncertainty becomes more understandable and manageable rather than inherently threatening or ambiguous (Loewenstein et al., 2001; Heath & Tversky, 1991). This interpretation is consistent with prior evidence indicating that financial knowledge is associated with a more balanced evaluation of investment risk and a more informed response to market uncertainty (Kaiser et al., 2022; Stolper & Walter, 2017; Aren & Zengin, 2016).

Our research findings are also consistent with recent studies highlighting the mediating role of perceived risk in an investment context. In this regard, prior research reports that perceived risk mediates the relationship between product knowledge and investment intention (Hati et al., 2020). Meanwhile, several other studies find that perceived risk mediates crucial behavioral antecedents of online trading intention (Raut & Kumar, 2024, 2023). In a similar spirit, another study demonstrates that perceived risk functions as a key serial mediator linking investor behavior to investment performance (Subedi et al., 2025). Collectively, these findings corroborate the argument that perceived risk is an important psychological mechanism by which financial knowledge affects investment behavior.

More importantly, our findings offer a more nuanced understanding of the relationship between financial literacy and risk-taking behavior. Instead of encouraging indiscriminate risk-taking or speculative actions, financial literacy appears to facilitate a recalibration of subjective uncertainty. In this regard, financial knowledge enables individuals to align their risk assessments more closely with observable market conditions, thus supporting more informed and disciplined investment decisions. This interpretation is consistent with previous studies that associate financial literacy with more disciplined investment behavior (Kaiser et al., 2022; Stolper & Walter, 2017; van Rooij et al., 2011). This also reconciles conflicting findings in the literature. While financial literacy may increase market participation and trading activity (Glaser & Weber, 2007; Barber & Odean, 2000), our evidence suggests that its primary behavioral function is not to encourage greater risk-taking per se, but to promote a more accurate assessment of uncertainty and more informed investment decisions (Kaiser et al., 2022; Aren & Zengin, 2016; Lusardi & Mitchell, 2014).

4.4.3. Trust as an Affective Mechanism Translating Knowledge into Action

Whereas perceived risk primarily reflects cognitive evaluation, institutional trust functions as an affective mechanism that translates cognitive understanding into behavioral intention. The empirical findings indicate that financial literacy is positively associated with institutional trust (β = 0.379, p < 0.001) while trust is positively associated with investment intention (β = 0.332, p < 0.001). Subsequently, the indirect effect of financial literacy on investment intention through trust (β = 0.126, p < 0.001) provides support for Hypothesis 3, and highlights the important role of affective process in shaping financial decision-making.

From the RFH perspective, trust can be interpreted as a positive affective heuristic that reduces emotional discomfort under uncertain conditions (Slovic et al., 2004; Loewenstein et al., 2001). In institutional environments characterized by information asymmetry and regulatory complexity, trust serves as a psychological mechanism that facilitates actions despite incomplete information. To retail investors, confidence in financial institutions, regulatory bodies, and market infrastructure provides affective reassurance that complements cognitive analysis (Balloch et al., 2015; Georgarakos & Pasini, 2011; Guiso et al., 2008).

Our findings correspond with recent empirical evidence emphasizing the growing importance of institutional and technological trust within contemporary financial systems. A more recent study underscores that trust in financial institutions significantly strengthens the intention to participate in stock markets during macroeconomic uncertainty periods (Adil et al., 2023). Similarly, Tien et al. (2026) report that confidence in technology and automated financial systems is a key factor influencing investor willingness to adopt robo-advisory services. Taken together, those findings suggest that trust has become increasingly important within financial ecosystems, characterized by rapid digitalization, technological complexity, and evolving institutional structure.

More importantly, our results suggest that trust extends beyond rational expectations of institutional compliance, and encompasses a broader sense of psychological safety, whereby individuals feel sufficiently protected and supported to engage with uncertain financial environments. Consequently, highly literate individuals may still be reluctant to participate in capital markets if their trust toward financial systems remains weak. We highlight that investment intention is shaped not only by cognitive competence but also by affective assurance. Accordingly, knowledge alone may be insufficient to stimulate market participation unless it is accompanied by sufficient trust in the institutions that govern and support financial transactions.

4.4.4. The Cognition–Affect Sequence in Investment Decision-Making

The principal contribution of this study lies in providing empirical evidence for a sequential cognition–affect mechanism underlying the relationships among financial literacy, perceived risk, institutional trust, and investment intention. Consistent with Hypothesis 4 (β = 0.072, p < 0.001), this study’s findings document that financial literacy is related to lower perceived risk, which subsequently facilitates stronger institutional trust and ultimately enhances investment intention.

This sequential framework advances behavioral finance theory in several important respects. First, it demonstrates that affective trust is not independent of cognitive evaluation; instead, trust is shaped by prior cognitive appraisals of uncertainty. In this context, a higher level of perceived risk constrains the development of trust, consistent with risk-centered theories of trust (Guiso et al., 2008; Siegrist et al., 2005; Das & Teng, 2004). Meanwhile, trust is more likely to emerge when financial environments are perceived as understandable and manageable (Kersting et al., 2015; Heath & Tversky, 1991).

Second, our findings help explain why initiatives designed to strengthen investor confidence may be less effective when implemented independently of financial education programs. Institutional interventions focused exclusively on increasing public confidence are likely to have limited behavioral impacts if an individual lacks the cognitive capability necessary to evaluate investment-related uncertainty. This evidence suggests that financial literacy and institutional trust operate as complementary rather than substitutive mechanisms in promoting investment participation.

Third, our study extends mediation-based research on behavioral finance by demonstrating that cognitive and affective mechanisms operate in an interdependent sequence rather than in isolated or parallel processes. While previous studies typically examined perceived risk or trust as separate mediators (Raut & Kumar, 2024; Adil et al., 2023; Hati et al., 2020), we integrate these constructs within a unified cognition–affect framework. Consequently, this study contributes to the behavioral finance literature by illustrating that investment intention emerges through an interconnected process. We suggest that the cognitive evaluation of uncertainty shapes affective trust, which in turn influences behavioral intention.

Overall, our findings suggest that retail capital market participation reflects an interdependent psychological process in which cognitive understanding reduces perceived uncertainty, lower uncertainty facilitates trust formation, and trust subsequently strengthens investment intention. More broadly, the findings indicate that investment intention emerges not solely from financial knowledge or institutional confidence in isolation, but from the dynamic interaction between cognitive understanding and affective reassurance. This framework provides a theoretically grounded foundation for future behavioral finance research, and offers practical insights to policymakers seeking to promote broader and more inclusive participation in capital markets.

5. Conclusions and Future Research Agenda

5.1. Conclusions

This study examines the behavioral mechanisms underlying the relationship between financial literacy and investment intention in Indonesian capital markets. By integrating the Theory of Planned Behavior (TPB) and the Risk-as-Feelings Hypothesis (RFH), this study develops and examines a sequential cognition–affect framework to explain how financial knowledge affects individual willingness to participate in capital market activities amid uncertainty.

Our findings indicate that financial literacy is a significant determinant of investment intention, both directly and indirectly. More importantly, the results demonstrate that financial literacy affects investment intention through a sequential mechanism in which a higher level of financial literacy reduces perceived risk, lower perceived risk strengthens institutional trust, and stronger trust subsequently enhances investment intention. These findings suggest that financial literacy operates not only as a source of financial knowledge but also as a psychological mechanism that shapes how individuals interpret and respond to uncertainty.

This study contributes to the behavioral finance literature by addressing persistent inconsistency concerning the relationship between financial literacy and capital market participation. Rather than conceptualize financial literacy as an isolated informational resource, we demonstrate that its effect emerges through the interaction between cognitive and affective processes. Hence, this study extends existing literature by showing that perceived risk and trust are interconnected instead of independent mechanisms in the formation of investment intention.

Overall, we suggest that investment participation is not driven solely by financial knowledge or access to information. Instead, participation reflects a broader psychological process in which individuals first evaluate uncertainty and subsequently develop the confidence necessary to act. By highlighting this cognition–affect sequence, this study provides a more comprehensive behavioral explanation of investment intention and offers a foundation for future research examining the interplay between cognitive and affective factors in financial decision-making.

5.2. Theoretical Implications

This study contributes to behavioral finance and financial decision-making literature by advancing a sequential cognition–affect perspective of investment intention. While prior research has frequently examined cognitive and affective factors as independent or parallel determinants of financial behavior, the present findings demonstrate that these mechanisms operate in an interconnected and sequential manner. Specifically, financial literacy affects investment intention through a cognition–affect process in a sequential pathway. In this regard, perceived risk functions as a cognitive appraisal of uncertainty, while institutional trust emerges as an affective response that facilitates action under conditions of incomplete information.

First, this study extends the Theory of Planned Behavior (TPB) by demonstrating that the influence of financial literacy on investment intention cannot be fully understood through direct effects alone. Instead, perceived risk and institutional trust serve as sequential intervening mechanisms that underlie the relationship between financial literacy and behavioral intention. This enriches the TPB framework by highlighting how the cognitive evaluation of uncertainty interacts with subsequent affective responses in shaping investment-related decisions.

Second, the findings provide empirical support for the Risk-as-Feelings Hypothesis (RFH) within the context of an emerging capital market. The results indicate that risk perception is not merely an objective assessment of loss probability, but a subjective cognitive appraisal shaped by financial knowledge and prior understanding. Financial literacy enables individuals to interpret market uncertainty more effectively, thereby affecting both cognitive evaluations and subsequent affective reactions toward investment activities.

Third, the evidence supports the sequential mediation pathway among financial literacy, perceived risk, institutional trust, and investment intention. This finding contributes to the broader behavioral finance literature by clarifying how trust is psychologically formed. Rather than functioning as an isolated determinant of financial behavior, institutional trust is more likely to emerge from prior cognitive appraisals of uncertainty. This perspective provides a more integrated explanation of how confidence in financial institutions develops and influences investment decisions in capital markets.

More broadly, the findings suggest that investment intention cannot be adequately explained by the informational or rational framework alone. Instead, an investment decision emerges from the dynamic interaction between cognitive understanding and affective reassurance under uncertainty. By conceptualizing financial decision-making as a sequential cognition–affect process, this study provides a more comprehensive behavioral explanation of investment intention and offers a theoretical foundation for future research examining the interplay between cognitive and affective mechanisms in financial behavior.

5.3. Policy Implications

This study’s findings carry important implications for policymakers, particularly the Indonesian Financial Services Authority (OJK) and the Indonesia Stock Exchange (IDX), in designing initiatives aimed at expanding retail participation in capital markets. While existing financial literacy programs primarily emphasize technical knowledge, product familiarity, and numerical competence, the present findings suggest that such approaches may be insufficient when implemented in isolation. Financial literacy affects investment intention not only by increasing knowledge, but also by shaping how individuals perceive and respond to uncertainty.

Accordingly, future financial education initiatives should adopt a more comprehensive behavioral orientation. Beyond explaining investment products and market mechanisms, educational programs should be designed to reduce subjective perceptions of uncertainty and strengthen confidence in financial institutions. This may be achieved by increasing public awareness of investor protection systems, regulatory safeguards, dispute-resolution mechanisms, disclosure requirements, and market oversight frameworks. Such efforts can help individuals perceive investment-related risks as more understandable and manageable, thereby facilitating the development of institutional trust.

The findings also highlight the importance of aligning financial literacy policies with broader institutional trust-building initiatives. Since trust emerges from prior cognitive assessments of uncertainty, policies aimed solely at increasing public confidence may be less effective if individuals lack the knowledge needed to evaluate financial risks and opportunities. Similarly, financial education programs may fail to translate into greater market participation if confidence in financial institutions remains weak. Consequently, financial literacy and institutional trust should be viewed as complementary policy objectives rather than independent interventions.

More broadly, the results suggest that efforts to enhance financial inclusion should incorporate behavioral and psychological considerations alongside traditional informational approaches. Expanding retail participation in capital markets requires not only increasing access to financial knowledge but also fostering an environment in which individuals feel sufficiently informed, protected, and confident to engage with financial markets. By addressing both the cognitive and affective dimensions of investment decision-making, policymakers may be better positioned to promote broader, more sustainable, and inclusive participation in capital market activities.

5.4. Managerial and Industry Implications

The findings provide important implications to financial institutions, brokerage firms, investment managers, and digital investment platforms seeking to expand retail investor participation. The results indicate that institutional trust serves as a critical mechanism by which financial literacy translates into investment intention. Consequently, among individuals who already possess a basic level of financial knowledge, reluctance to participate in capital market activities is more likely to stem from concerns regarding uncertainty, institutional credibility, and platform reliability than from informational deficiencies alone.

Accordingly, investor engagement strategies should extend beyond the provision of financial information and focus on reducing perceived uncertainty while strengthening institutional trust. In addition to communicating investment opportunities and expected returns, financial institutions should emphasize transparency, regulatory compliance, investor protection mechanisms, cybersecurity safeguards, disclosure practices, and long-term institutional credibility. These initiatives can help investors perceive financial environments as more understandable, predictable, and trustworthy.

The findings further suggest that trust-building should not be viewed solely as a branding or reputation-management activity. Instead, trust is a strategic behavioral asset that enables investors to act under uncertainty. As such, organizations should integrate educational initiatives, transparent communication practices, and trust-building measures into a unified investor engagement strategy. This approach is likely to be more effective than relying exclusively on promotional messages that emphasize short-term financial performance.

These implications are particularly relevant within increasingly digitalized investment ecosystems characterized by online trading applications, mobile investment platforms, robo-advisory services, and AI-assisted financial technology. Financial interactions become increasingly mediated by digital systems, so technological trust, platform reliability, data security, and institutional credibility are anticipated to play an increasingly significant role in maintaining retail investor engagement. Firms that successfully combine financial education with trust-enhancing practices are more likely positioned to attract, retain, and engage retail investors in the long run.

5.5. Limitations and Future Research Agenda

Although this study provides both theoretical and empirical contributions, several limitations should be acknowledged, offering opportunities for future research. First, the sampling strategy is restricted to individuals with monthly incomes exceeding IDR10 million to ensure that respondents possess sufficient financial capacity to participate in capital market activities. While this approach enhances internal validity by minimizing the effect of liquidity constraints, it may limit the generalizability of the findings to a broader investor population. This caveat is particularly relevant since financial literacy, perceived risk, and institutional trust may operate differently among individuals facing stronger financial constraints and lower levels of financial inclusion. Future research should therefore examine whether the cognition–affect mechanism identified in this study is consistent across different income groups and socioeconomic strata.

Second, the cross-sectional design limits the ability to draw definitive causal inferences among the variables examined. The observed relationships are consistent with the proposed theoretical framework; however, experimental design, longitudinal or panel-based studies would enable researchers to inquire into how changes in financial literacy, perceived risk, and institutional trust affect investment intention over time. Such approaches would provide deeper insights into the dynamic evolution of cognitive and affective mechanisms in financial decision-making.

Third, this study relies on self-reported measures of investment intention rather than observed investment behavior. Although behavioral intention is a central construct in the Theory of Planned Behavior (TPB), prior research has consistently documented an intention–behavior gap. Therefore, future studies may benefit from employing experimental designs, direct observation of investment activities, or transaction-level data to determine whether the cognition–affect mechanism identified in this study is reflected in actual market participation and investment behavior.

Fourth, future research could extend the proposed framework to emerging digital financial environments, including robo-advisory services, AI-assisted investment platforms, social trading systems, and broader fintech-based investment ecosystems. In such contexts, investor decision-making is increasingly influenced by algorithmic processes and digital interactions. Consequently, constructs such as technological trust, algorithmic transparency, and digital financial literacy are likely to become increasingly important determinants of investment behavior. Future studies may explore whether technological trust serves as an affective mechanism analogous to institutional trust in a digital investment environment.

In summary, the limitations identified in this study provide several promising directions for future research. Future studies should examine whether the proposed cognition–affect mechanism operates similarly across socioeconomic groups, investigate its dynamic evolution using longitudinal research designs, and evaluate its influence on actual investment behavior rather than on behavioral intention alone. Furthermore, digitalization in financial services presents opportunities to extend the framework by incorporating constructs such as technological trust, algorithmic transparency, and digital financial literacy. Collectively, these research directions provide opportunities to further refine and extend the proposed cognition–affect framework across diverse populations, research settings, and financial environments, thereby advancing understanding of how cognitive and affective mechanisms jointly shape financial behavior under uncertainty.