Does SME’s financing decisions follow pecking order pattern? The role of financial literacy, risk preference, and home bias in SME financing decisions

Abstract

This paper aims to explore the role of financial literacy, risk preference, and home bias in Minangkabau ethnic SME financing decisions, and whether their financing pattern follow the pecking order theory. The theoretical framework model was developed to determine the impact of financial literacy, risk preference, and home bias on the financing decisions of SMEs. The study applied the SEM-PLS method in testing the model. Data collection was carried out through online and offline survey techniques with a sample of 623 SMEs. The results reveal that home bias and risk preference have a negative direct effect on financing decision. Home bias also has a direct positive influence on risk preference. Financial literacy can minimize home bias from negative to positive toward financing decision. However, financial literacy does not have a direct effect on the financing decisions and does not have a significant direct influence on risk preference. Most SMEs have a low level of financial literacy, a high tendency for home bias and a high-risk preference. Financial literacy has strong positive moderation effect on the influence of home bias toward financing decision. Home bias make SME financing decision does not follow pecking order pattern. As SMEs financial literacy level is still very low, thus it is important for financial institutions to educate SMEs by providing several trainings about financial matters. The findings suggest to improve the financial literacy of SMEs informal association groups in order to help them reducing their home bias tendency. This study seeks to explore the pecking order pattern in financing decisions of SMEs.

Keywords:

• financial literacy
• home bias
• risk preferences
• financing decision
• SMEs

1. Introduction

The financing needs of each type of company will be different, it is easier for companies that go public to get sources of financing, while micro and small enterprises (hereinafter referred to as SMEs) must rely on their own sources of funds, especially when opening a business. Several studies of Small and medium enterprises (SMEs) and financing related to the theory of capital structure, that SMEs follow the pecking order theory, financing needs will be met from internal financing first, if financing is still needed then it will be met from external financing. Delic et al. (2016) found the dominance of SMEs using their own funds and bank loans when starting a business or when financing company growth and development. Based on the age level found in research by Nguyen and Luu (2013) which determines access to informal, formal, or both sources of funds, young managers’ access to sources of funds from informal sources such as family and friends. The older they are, the more likely they will be able to access banks or equity. The findings of Sanyal and Mann (2010), Fourati and Affes (2013) show that startup entrepreneurs are more likely to have foreign debt in their capital structure. For entrepreneurs who have a lot of human capital, are less likely to have external debt and are more likely to be financed internally. Home-based activities tend to be financed by internal equity contributions. Highly educated entrepreneurs have more external debt and attract more investors.

Financial literacy is an important thing for SMEs managers. Chen and Volpe (1998), Mouna and Anis (2013) explained that individuals with low education and income make the wrong decisions in their finances. Financial literate individuals are more likely to plan for debt provision and retirement (Lusardi et al., 2010. Eniola and Entebang (2017) in their findings stated that financial awareness is important for financial performance, this implies that company owners and managers are aware of the differences in several financial products that are profitable for them, external financing will not be sought if it is difficult to access. The individual decision-making process is influenced by the risk preferences of individual investors. In financial behavior, it is assumed that individuals can become irrational in decision making. This is due to many influencing factors such as psychological, sociocultural, and environmental factors.; Aren and Zengin (2016) state that investors who have high financial literacy tend to take risks to invest in equities and portfolios. Zwaan et al. (2020) stated that women tend to have low levels of self-confidence and financial knowledge and are unable to take risks. Kumar et al. (2017) show that there is a significant relationship between financial literacy and financial behavior and financial decisions. Several empirical financial literacy and risk tolerance has been studied, especially about investment decisions or retirement plans.

Financial decisions made by individual investors are not only based on company values but driven by their emotions such as home bias. This home bias is the tendency of investors to choose investment in their home country compared to a combination of stocks from foreign companies. Vries et al. (2017) find that investors show a familiarity bias when choosing between different companies to invest in. Several previous studies have found reasons for controlling for home bias such as inflation hedging and overconfidence (Musa & Simonov, 2004; Ulwan, 2021). As far as what has been studied, financial literacy affects risk preferences, and home bias towards SMEs funding decisions has not been found. The research gap in this study will reveal whether financial literacy variables directly affect risk preference and house bias or moderate the effect of risk preference and house bias variables on Minangkabau ethnic SME financing decisions. The Minangkabau ethnic community is known as one of the communities that carried out the tradition of migrating. Migrating can increase knowledge, broaden horizons, and motivate oneself to seek a better life throughout the country. The entrepreneurial spirit for the Minangkabau community is also very thick with promoting honesty in trading. Hastuti et al. (2015), Sutanto and Nurrachman (2018) stated that the success of the Minangkabau ethnic entrepreneurship in the migration location is inseparable from their character as hardworking, confident, independent, consistency, ingenuity, flexible, economical, and brave enough to face challenges and struggling to rely on oneself in new places. Based on the above empirical study on the characteristics of the Minangkabau ethnicity, this study reveals whether financial literacy, risk preferences, and home bias influence Minangkabau ethnic SME financing decisions. The empirical results of financial literacy have an indirect effect. The effect of financial literacy is mediated by the variables of home bias and risk preference in financing decisions. The data show that the average level of financial literacy is low with a high home bias leading to bias in assessing risk. Home bias increases resulting in the misperception of risk, so that there is an error in judgment in taking more risks than they can bear. Consequently, financing decisions use external financing which has a higher risk.

2. Literature review

2.1. Pecking order theory of financing decisions

Traditional financing decisions are based on two theoretical frameworks, namely trade off and pecking order theory. Theoretical trade-off model, capital structure is an assumption of a trade-off between the tax advantages of using debt and agency costs. The use of debt will increase the value of the company only to a certain extent. Jensen and Meckling (1976) and Myers (1984) found a negative relationship between capital structure and firm value, meaning that the addition of debt is a sign that the company’s financial performance is not good. By considering bankruptcy costs and agency costs, companies should use debt funds to an optimal level or often called the trade-off theory. This trade off theory implies that managers will think in terms of tradeoffs between tax savings and bankruptcy costs in determining their capital structure. Luigi (1958) pecking order theory directs the decision to choose company financing according to the level of financing, this concept minimizes the need for external financing. The company’s financing will be met first by internal financing sources, and if this is not sufficient, it will be met by external financing sources. And if additional external financing is needed then they choose the source of financing that can minimize the additional costs of asymmetric information.

Small and medium enterprises (SMEs) sources of financing are limited to banking and non-banking financing. If it is related to the capital structure theory that SMEs adhere to the pecking order theory, the financing needs will be met from internal financing first, and if this is insufficient it will be met from external financing. Delic et al. (2016) SMEs are most dominant in their financing using their equity funds and bank loans when starting a business or financing company growth and development. Funds from family and friends (FFF), as well as factoring, are among the least used sources of financing. In developed financial markets, funds from family and friends are one of the most accessible sources of financing for startup companies, while bank loans are not accessible to companies at this stage of their life cycle. Loans from suppliers are also an important source of financing for the growth and development of companies, but this source of financing is insufficient legal protection for investors. In Entrepreneurial start-up activities are more likely to have some external debt in their capital structure if they have more tangible assets that function as collateral and also have a legal form in the amalgamation (Fourati & Affes, 2013; Karyadi & Rizki, 2018; Sanyal & Mann, 2010). For entrepreneurs who have more human capital in entrepreneurial activities, they are less likely to have external debt and are more likely to be financed internally. Home-based activities are more likely to be financed by internal equity contributions. More educated entrepreneurs have more external debt and attract more investors.

2.2. Planned Behavior Theory (TPB) of Financial Behavior

Based on Planned Behavior theory (TPB), a person’s intention toward a behavior is formed by three factors, namely attitude, subjective norm, and perceived behavior control which have an impact on a person’s intentions or actions (Ajzen, 1991). Financial decisions related to the phenomenon of financial behavior in line with the demands of business and academic world developments, began to be addressed by the presence of behavioral elements in the decision-making process. Irrational investor decision making can be caused by cognitive and psychological biases. Financial behavior began to be recognized by various parties, especially academics after Solvic (1969) explained the psychological aspects of investment and stockbroker activities, Tversky and Kahneman (1974) stated that assessments under conditions of uncertainty can produce heuristics or bias

Investors in investing do not only use the estimated prospects of investment instruments. Investor psychology factors also have an important role in investing. Behavioral finance tries to explain and improve understanding of investors’ reasoning patterns, including the emotional processes involved and the extent to which they affect the decision-making process. Behavioral finance seeks to explain what, why, and how finance and investment, from a human perspective (Ricciardi & Simon, 2000). Behavioral finance complements traditional financial theory by providing behavioral explanations for the irrational behavior of investors. Shefrin (2000) defines behavioral finance as how psychological phenomena influence financial behavior. Nofsinger (2001) defines behavioral finance as studying how humans actually behave in a financial environment. In particular, study how psychology influences financial decisions, companies and financial markets. Based on the description above, it is clearly stated that behavioral finance is an approach that explains how humans invest or handle finances that are influenced by psychological factors.

2.3. Financial literacy and financing decisions

Lusardi et al. (2010) defined financial literacy as financial knowledge and the ability to apply it (knowledge and ability). The OECD INFE defines financial literacy as “the combination of awareness, knowledge, skills, attitudes, and behaviors necessary to make financial decisions and ultimately achieve individual financial well-being.” Measuring levels of financial literacy is a growing research issue around the world, with levels and reach. Financial transactions on electronic media are changing so rapidly, so is our understanding of competencies in financial literacy.

Financial literacy research has been widely conducted with individual characteristics and financial decision making, such as Chen and Volpe (1998), finding that individuals with little knowledge tend to have wrong opinions and decisions in the areas of general knowledge, saving, borrowing, and investing. Asaad (2015) found that individuals with good knowledge and high self-confidence are more likely to make “good financial decisions.” When self-confidence is high and actual knowledge is low, individuals are more likely to behave in making high-risk financial decisions.

2.4. Home bias and financing decisions

Financial decisions made by individual investors are not only based on information but also driven by individual emotions. Financial behavior tends to experience emotional bias. Emotional bias research has been carried out, one of which is home bias. Home bias in financial decisions occurs where investors tend to choose investments in their home country compared to a combination of stocks from foreign companies. Home bias and familiarity bias occur when investors prefer to invest in assets, they are familiar with. Some investors tend to be more comfortable buying company shares in their country rather than diversifying risk through investing in various countries (Dervishaj, 2018; Ullah, 2018). Vries et al. (2017) found that investors show familiarity bias when choosing between different companies to invest in. Several previous studies have found reasons for maintaining home bias such as hedging inflation and overconfidence (Musa & Simonov, 2004). Empirical evidence reveals that individual characteristics have a significant relationship with home bias, in fact men tend to have a greater home bias than women (Karlsson & Nordén, 2007).

From a sociological perspective, culture is an important part of understanding individual behavior. Cultural factors also play a role in investment decisions, as investors personally and collectively adhere to maintaining personal relationships within the organizations or communities of which they are members (Ellison & Fudenberg, 1993). Some evidence from behavioral finance research and other sociology suggests that investor behavior is linked to an individual’s cultural origins. This shows that individual investment behavior can be predicted based on cultural characteristics.

2.5. Risk preference and financing decisions

Investors in investing certainly hope to benefit from the investment. But in investing there must be a risk, in the form of capital gains or capital losses. The level of risk borne by investors differs from one another, depending on the degree to which investors prefer risk. However, various factors can cause individuals to have different judgments about taking risks. On the other hand, risk attitudes in a more specific context provide a more robust measure of the appropriate context. Gender differences in willingness to take risks may be part of the explanation for differences in behavior. Dohmen et al. (2011)) found that women choose low risk in making decisions. Weber and Hsee, (1998) define risk preference as an individual’s tendency to choose risky options. Bodie et al. (2014) classify the willingness of investors to bear investment risk into 3 types: 1) types of investors who dare to take risks called risk takers or risk lovers or risk seekers. 2) the type of investor who is willing to bear a risk that is proportional to the return they will get or what is called moderate risk. 3) the type of investor who is afraid or reluctant to assume risk is called a risk averter.

Aren and Zengin (2016) state that risk perceptions and literacy levels affect individual investment preferences, investors who have high financial literacy tend to take risks to invest in equities and portfolios. Chen and Volpe (2002) stated that women tend to have low levels of self-confidence and cannot take risks. Barber and Odean (2001) identified single men as more risk-taking than women, whereas the findings of Zwaan et al. (2020) that women have lower self-assessment and financial knowledge. These findings underscore the importance of financial literacy for self-assessment planning.

Based on the literature the following research hypotheses and framework (Figure 1) can be made:

Figure 1. Theoretical framework.

H1. Financial literacy has significant effect on financing decisions

H2. Home bias has significant effect on financing decisions

H3. There is a significant effect of risk preference on financing decisions

H4. Financial literacy has significant effect on home bias

H5. There is a significant effect of financial literacy on risk preference

H6. Financial literacy moderates the effect of risk preference on financing decisions

H7. Financial literacy moderates the effect of home bias on financing decisions

H8. There is a significant effect of home bias on risk preference

H9. Home bias moderates the effect of risk preference on financing decisions

3. Research method

This research is a survey research with quantitative analysis. This research was conducted in several districts and cities in West Sumatra Province. The research subjects are entrepreneurs from the Minangkabau tribe, who are widely known for their entrepreneurial orientation and migration to various regions in Indonesia and Southeast Asian countries.

This study distributed questionnaires directly to respondents out through online and offline survey techniques as a data collection technique. A total of 1000 questionnaires were distributed to SME owners who were used as samples. The sampling technique used is using simple random sampling. Of these, 642 questionnaires were returned. In other words, the response rate obtained is 64.2 percent. Of these, there were several questionnaires with incomplete answers. The final calculation shows that a total of 623 questionnaires are feasible for further analysis.

In this study, there are several variables studied. The estimated dependent variable is Financing Decisions (FD), while the independent variables in this study are Financial Literacy (FL), Risk Preference (RP), and Home Bias (HB). The variable of Financial Literacy (FL) is defined as a combination of cognitive ability and investment in human capital related to understanding financial decisions (Lusardi and Mitchell, 2014). The variable was measured by four indicators including identifying financial information, analyzing information in financial context, evaluating financial issues and applying financial knowledge.

The variable of Risk Preference (RP) is defined as the attitude of Minangkabau ethnic SME owners in holding towards risks on their decision-making behavior (Wen et al., 2014). The variable was measured by using 3 items adopted from Zyphur et al. (2009). Moreover, the variable of Home Bias (HB) is operationally defined as the behavior of Minangkabau ethnic SME owners who tend to only choose to invest in the area where they live with reasons that are less rational. The variable was measured by using three items adopted from including Lütje and Menkhoff (2007) including institutional characteristics, informational characteristics, behavioral characteristics.

The variable of FD is defined as the decisions of Minangkabau ethnic SME owners in determining the proportion of equity and debt capital that must be owned in their company’s capital structure. This variable of financing decisions is estimated with three measurements adopted from Nguyen et al. (2022) including (1) investment decision, (2) working capital decision, and (3) funding decision.

The questionnaire used 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). To analyze the hypothesis, this study employed the SEM-PLS method in testing the model.

4. Results

The test results to determine outer loading show that the values obtained are acceptable for items measuring the variables financial decision, home bias, risk preference and financial literacy (Table 1). The results also show a Cronbach’s Alpha value of more than 0.7. This is supported by the Composite Reliability value which is higher than 0.7. Thus, all variables in this study are declared reliable. The Average Variance Extracted (AVE) value is also declared acceptable according to the standard value used.

Table 1. Outer loading, construct reliability and Average Variance Extracted (AVE)

Construct	Item	Outer Loading	Cronbach’s Alpha	rho_A	Composite Reliability	Average Variance Extracted (AVE)
Financial Decision	FD	1.000	1.000	1.000	1.000	1.000
Home Bias	HB1	0.768	0.617	0.646	0.794	0.565
	HB4	0.648
	HB5	0.827
	HB*FL	0.827	1.000	1.000	1.000	1.000
Risk Preference	RP1	0.598	0.412	0.568	0.750	0.610
	RP3	0.929
	RP*FL	1.059	1.000	1.000	1.000	1.000
	RP*HB	1.465	1.000	1.000	1.000	1.000
Financial Literacy	FL	1.000	1.000	1.000	1.000	1.000

Furthermore, validity testing through Discriminant Validity (Fornell-Larcker Criterion) shows the values accepted for all variables in this study. In general, the Fornell-Larcker criterion is to assess discriminant validity by comparing the square root of the AVE value with the correlation of latent variables. In particular, the square root of each AVE construct must be greater than the highest correlation with the other constructs. The results show that the values resulting from the discriminant validity analysis are accepted (Table 2). This means that all variables in this study are valid and can be continued in the next analysis.

Table 2. Discriminant validity (Fornell-Larcker Criterion)

	HB	HB*RP	FD	FL	FL*HB	FL*RP	RP
HB	0.751
HB*RP	−0.056	1.000
FD	−0.126	0.074	1.000
FL	0.276	0.051	0.030	1.000
FL*HB	0.148	0.141	0.111	0.327	1.000
FL*RP	0.071	0.225	0.072	0.132	0.425	1.000
RP	0.414	−0.056	−0.136	0.053	0.070	0.070	0.781

Heterotrait-Monotrait Ratio (HTMT) test shows the value of validity testing. HTMT is the recommended alternative method for assessing discriminant validity. This method uses a multitrait-multimethod matrix as the basis for measurement. The recommended measurement value of the Heterotrait-Monotrait Ratio (HTMT) in PLS is less than 0.85. The test results show that all variables have an accepted value <0.85 (Table 3).

Table 3. Heterotrait-Monotrait Ratio (HTMT)

	HB	HB*RP	FD	FL	FL*HB	FL*RP	RP
HB
HB*RP	0.074
FD	0.169	0.074
FL	0.343	0.051	0.030
FL*HB	0.182	0.141	0.111	0.327
FL*RP	0.091	0.225	0.072	0.132	0.425
RP	0.745	0.121	0.164	0.085	0.139	0.119

Furthermore, to test the assumption that there is a relationship between independent variables, multicollinearity testing is carried out. The statistical results show the value received (Table 4). More specifically, the recommended value for testing the absence of multicollinearity is VIF < 10. The results show the VIF value for all variables is < 10. Thus, there is no multicollinearity in the model in this study.

Table 4. Collinearity statistics & VIF values

Relationship	VIF
HB * LK	1.000
HB1	1.232
HB4	1.177
HB5	1.297
KP1	1.000
PR * HB	1.000
PR * LK	1.000
PR1	1.072
PR3	1.072
Score_LK (FL)	1.000

The next test is to determine how far the independent variable reflects the variability of the dependent variable. The results obtained are reflected by the coefficient of determination or R-square. The values obtained from the test show values ranging from 0.045 (FD), 0.076 (HB) to 0.173 (RP; Table 5).

Table 5. R square

Variable	R Square	R Square Adjusted
HB	0.076	0.075
FD	0.045	0.036
RP	0.176	0.173

Next, an f-square effect size (f²) analysis was performed. This analysis is to obtain the f-square value which is used to determine the effect of the predictor variable on the dependent variable. The test results as shown in Table 6 show values ranging from 0.001 to 0.082. This shows the varying effect of the predictor variable on the dependent variable.

Table 6. F-square

	HB	HB*RP	FD	FL	FL*HB	FL*RP	RP
HB			0.009				0.210
HB*RP			0.002
FD
FL	0.082		0.000				0.005
FL*HB			0.009
FL*RP			0.001
RP			0.009

Furthermore, the fitness test is carried out using several indices such as SRMR, d_ULS, d_G, Chi-Square and NFI. The test is carried out by comparing the value of the saturated model with the estimated model for each index. The results show the values received for all indexes. This shows that the model in this study is fit (Table 7).

Table 7. Model fit

	Saturated Model	Estimated Model
SRMR	0.096	0.096
d_ULS	0.259	0.259
d_G	0.078	0.078
Chi-Square	289.519	289.360
NFI	0.401	0.402

In this study, the analysis conducted to answer the hypothesis. The analysis was then performed to describe the results of the analysis of financial literacy, risk preferences, and home bias against financing decisions. The results are shown in Table 8.

Table 8. Hypotheses Testing

Hypo-thesis	Path	Std Beta	Std. Error	t-value	Bias	Confidence Interval		p	Decision
						2.5%	97.5%
H1	FL -> FD	−0.011	0.045	0.254	−0.001	−0.106	0.072	0.799	Rejected
H2	HB -> FD	−0.150	0.040	3.740	−0.003	−0.224	−0.073	0.000	Accepted
H3	RP -> FD	−0.100	0.047	2.150	0.001	−0.188	−0.010	0.032	Accepted
H4	FL -> HB	0.276	0.040	6.967	−0.001	0.195	0.347	0.000	Accepted
H5	FL -> RP	0.053	0.042	1.264	−0.003	−0.030	0.134	0.207	Rejected
H6	FL*RP -> FD	0.027	0.047	0.584	−0.001	−0.055	0.125	0.559	Rejected
H7	FL*HB -> FD	0.102	0.044	2.283	0.004	0.014	0.185	0.023	Accepted
H8	HB -> RP	0.433	0.043	10.002	0.005	0.336	0.508	0.000	Accepted
H9	HB*RP -> FD	0.027	0.027	0.981	0.005	−0.015	0.088	0.327	Rejected

FL = Financial Literacy; HB = Home Bias; RP = Risk Preference; FD = financing Decision

The results of the hypothesis (H1) are rejected. This means that financial literacy does not directly affect financing decision making based on the financing source of SMEs. This is in line with Xiao (2011), Zhu (2017) most high-tech SMEs will choose to use retained earnings (internal funds), thereby growing their business. Funds from family and friends and factoring are the least used sources of financing. In advanced financial markets, funds from family and friends are one of the most accessible sources of financing for startup companies, while bank loans are not accessible to companies at this stage of their life cycle.

Testing of hypothesis (H2) shows risk preference directly influences financing decisions with negative path coefficient. This means that an increase in risk preference, SMEs Minangkabau uses internal funds. It can be understood that SMEs in running a business are not afraid of failure and bounce back with the capital they have. This is not in line with the theory of someone who has a high-risk preference (risk lover), so choosing to fund will choose external financing. This is in line with Barbosa et al. (2007) showing that individuals with high-risk preferences have higher levels of entrepreneurial intention and opportunity identification, while individuals with low-risk preferences have higher relationship potential and tolerance. Likewise, Qureshi et al. (2012) proved a significant and positive relationship between risk aversion and investment decision making.

Statistical testing of the third hypothesis shows that there is an effect of home bias tendency with Minangkabau entrepreneur financing decisions. The negative path coefficient means that the higher the level of the tendency for SMEs Home bias to have an impact on financing decisions in determining financing sources internally. The home bias tendency relates to information obtained from people they know. The information they believe is very supportive of the trend of home bias in financing decision making. SMEs have a high level of solidarity between traders, have associations formed to strengthen ties and obtain various information. This is following the assumption that people who have a high tendency of home bias when investing tend to choose investment in their own country. Likewise, for financing, internal funds, or own capital are chosen. This finding is in line with Musa and Simonov (2004) and Raut et al. (2018) that carried out the impact of familiarity depending on the level of investor information. In other words, investors having low knowledge will be influenced by familiarity. If it takes too long, it will have an impact on investor behavior. This suggests that investment choices are driven by the availability of information and familiarity, which is a substitute for good information.

The results of testing the fourth hypothesis shows that financial literacy directly affects the tendency of the home bias of SMEs. The path coefficient is positive. This means that an increase in financial literacy will have an impact on the increase in their home bias tendency. There is a diversification concept that has been violated. The concept of diversification is that with higher financial knowledge, it can reduce the tendency of home bias. Financial literate people are more likely to invest in foreign assets, and will therefore benefit from international diversification. The results of this study are in line with Bransch (2020) that professional financial advice for household financial literacy cannot compensate for the reduced home bias in investment. Lack of diversification explicitly focuses on how to make individuals financially literate. There may be something that is not conveyed in providing financial knowledge for individuals. The results of this study indicate that most SMEs have low financial literacy, with a high level of home bias tendency. SMEs with a low level of financial literacy, although the old business experience and an increasingly mature age affect the increasing trend of home bias. This is in line with Koenen et al. (2017) stating that individual who is home biased is more likely to invest in companies they know. The finding is also consistent with Coval and Moskowitz (1999) showing that local investors have the greatest comparative advantage over informally small firms. For uninformed investors, who face a more severe selection when investing in securities, they have a relatively smaller proportion of informed investors.

The testing results of the fifth hypothesis show that financial literacy does not directly affect the risk preferences of SMEs in financing decisions. This study is also supported by Gustafsson and Omark (2015) found evidence shows that individuals who rely on their intuition rather than financial literacy when facing financial risk, tend to show a higher tolerance for financial risk. Mudzingiri et al. (2018) stated that students use the information available to them to make decisions. However, this result is not in line with Aren and Zengin (2016) which states that risk perceptions and literacy levels affect individual investment preferences, investors who have high financial literacy tend to dare to take risks to invest in equities and portfolios. Investors with a low level of financial literacy tend to be risk-averse and prefer to deposit money to invest. Guiso and Jappelli (2006) stated that investors who are too confident, do not accept the information provided by financial advisors, banks, or brokers. Investors have more confidence in the information they get themselves. Gaudecker (2013) most of the disadvantages of inadequate diversification are obtained by overly confident investors, who are not financially literate or consulting financial advisors.

Testing the sixth hypothesis showed the moderating effect of the variable financial literacy and risk preference in SME financing decisions is proven insignificant. It is expected that the existence of financial knowledge can reduce high-risk preferences so that the choice of financing decisions can be well-diversified. In the process of making financial decisions, whether investment or financing, a person will be faced with several factors such as risk, ambiguity, and also many choices. These factors can cause bias in financial decision making. Rationally, individuals avoid risk. The decision-making process is influenced by individual risk preferences. In financial behavior it is assumed that individuals can become irrational in making decisions, this is due to many factors such as psychological, sociocultural, and environmental factors. This is not in line with Shusha (2017) revealing that financial literacy plays a role in moderating the relationship between demographic characteristics and their propensity to take risks. Kumar et al. (2017) showed that financial literacy has a significant positive effect on financial behavior, while Aren and Zengin (2016) state that risk perceptions and literacy levels affect individual investment preferences, investors who have high financial literacy tend to dare to take risks to invest in equities and portfolios. Yao et al. (2011) found that investment knowledge and experience can influence differences in perceptions of financial risk.

The results of testing the moderating effect of the financial literacy variable on the home bias (H7) in making financing decisions based on sources of funds are proven to be significant, meaning that the financial literacy variable mediates the home bias in decision making on Minangkabau ethnic SMEs. This means that financial literacy strengthens the tendency of increasing home bias to have an impact on making financing decisions, especially in the use of internal or external sources of funds. Most of the SMEs have low financial literacy, with a high home bias tendency to have an impact on the selection of sources of business funds coming from external funds, both formal and non-formal. Home bias is a phenomenon that continues to occur in financial markets, if the high home bias tendency has an impact on wrong financial decision making, it is possible to reduce home bias and ambiguity by providing good information about financial decisions. Research results that support Dlugosch et al. (2014) observed that the preference for known assets in the country is very strong to be ambiguous. It is possible to reduce home bias and ambiguity by providing better information about the return on assets. This finding is in line with Barber and Odean (2008) that investment is driven by attention and responds to pure labels, and more importantly, home bias is deliberately maintained in the face of uncertainty. These results impact interdependence and indicate a strong relationship between behavioral bias factors.

The results of hypothesis 8 show that home bias directly affects risk preferences. A positive path coefficient means a high level of home bias tendency, an increase in impact, or the courage to make high-risk decisions. This is in line with Gaudecker (2013), Guiso and Jappelli (2006) stating that investors who are too confident, do not accept the information provided by financial advisors, banks, or brokers. Agarwal et al. (2016) found bias harms investors, women are more afraid to take risks. Vries et al. (2017) found that investors exhibit a familiarity bias when choosing between different companies to invest in. Several previous studies have found reasons for controlling for home bias such as inflation hedging (Musa & Simonov, 2004), and overconfidence (Barber & Odean, 2011).

In testing hypothesis 9, it shows that home bias as a moderating variable of risk preference for financing decision making is not proven. In making rational investors’ decisions, investors make financial decisions to maximize their risk-return tradeoff. Sometimes individual financial decision making is irrational, the behavioral finance literature reveals that many types of bias also color the individual financial decision-making process, these biases are globally grouped into 2 (two), namely cognitive bias and emotional bias (Pompian, 2006). Home bias is one of the emotional biases that also colors in making financing decisions. Rational behavior is also necessary to be financially successful and to overcome this tendency. This is consistent with Marchand (2012) states that investors do not always act rationally because of the cognitive and psychological mistakes they have to face. They are influenced by behavioral factors in influencing investors who make financial decisions. In these circumstances, more comprehensive and prudent decision-making is impossible. Biases and heuristics present an effective way of estimating correct decisions. The results that also support Muradoglu et al. (2009) show that economic activity and state openness are not always the only prerequisites for attracting foreign investors. Olakitan and Ayobami (2011) shows that locus of control and risk-taking behavior do not together predict entrepreneurial success. The results of this research also show that there is no significant difference in entrepreneurial success based on gender differences.

Based on the results of the analysis, several hypotheses have an insignificant direct effect but have a significant indirect effect. An indirect effect between the two variables can occur when a variable affects another variable through one or more latent variables. In this study, three hypotheses state indirectly significant (Table 9), which are additional empirical findings in this study, which will be discussed below.

Table 9. Total, indirect effect & specific indirect effect

Total Indirect Effects
	Original Sample (O)		Sample Mean (M)			Standard deviation (STDEV)		T Statistics (|O/STDEV|)		P Values
HB -> FD	−0.044		−0.044			0.021		2.075		0.038
HB -> RP
HB*RP -> FD
FL -> HB
FL -> FD	−0.035		−0.035			0.013		2.669		0.008
FL -> RP	0.119		0.120			0.021		5.604		0.000
FL*HB -> FD
FL*RP -> FD
RP -> FD
Specific Indirect Effects
		Original Sample (O)		Sample Mean (M)	Standard Deviation (STDEV)		T Statistics (|O/STDEV|)		P Values
FL -> HB -> FD		−0.029		−0.030	0.013		2.337		0.020
HB -> RP -> FD		−0.044		−0.044	0.021		2.075		0.038
FL -> HB -> RP -> FD		−0.012		−0.012	0.006		1.953		0.051
FL -> RP -> FD		0.007		0.007	0.006		1.186		0.236
FL -> HB -> RP		0.119		0.120	0.021		5.604		0.000

The results of the first hypothesis indicate that financial literacy indirectly affects financing decisions, meaning that there are variables that mediate financial literacy in Minangkabau SME financing decisions. Table 9 specific indirect effects, show the variables that mediate financial literacy, namely the home bias. This result is an additional finding that shows that financial literacy significantly influences financing decisions mediated by the home bias.

5. Conclusion

The results showed that financial literacy, risk preference, and home bias have a significant influence on financing decisions. However, financial literacy does not significantly influence financing decisions. Home bias and risk preference significantly influence financing decisions to set up a business using external financing, both formal (bank) and informal (relatives, friends, or colleagues/partners). Financial literacy directly affects the home bias tendency of SMEs. Increased financial literacy, but does not reflect the truth because it does not decrease their home bias tendency. Furthermore, financial literacy does not directly affect the risk preferences of SMEs in financing decisions.

The moderating effect of financial literacy on risk preference is rejected, meaning that the financial literacy variable does not moderate (strengthen/weaken) risk preferences for financing decision making. Meanwhile, the test for the moderation effect of financial literacy on home bias is accepted. This means that financial literacy strengthens influence. This means that an increase in financial literacy, the tendency for home bias to increase has an impact on financing decisions. SMEs are mostly with low financial literacy, home bias tends to be high and the choice of sources of business funds is external, both formal and non-formal. Home bias can directly affect the risk preferences of SMEs. The results confirmed home bias as a moderating variable does not provide empirical evidence to accept hypothesis. High home bias does not strengthen or weaken the level of risk preference in external financing decision making. These results reveal that SMEs in meeting their financing needs do not follow the pecking order theory pattern, this is possible because of the role of financial literacy, risk preference, and home bias in SMEs.

Financial literacy does not have a direct influence on financing decisions. The results of the study on the direct effect of financial literacy with a significant positive home bias. Risk preference influences financing decisions significantly negatively. The effect of financial literacy through the variables of home bias and risk preference on financing decisions, with a negative path coefficient. The data show that the average level of financial literacy is low, resulting in a tendency for increased home bias to lead to bias in assessing risk, as a result of which risk preferences are higher. The tendency for home bias to increase leads to misperceptions of risk so that there is a misjudgment of taking more risks than they can bear. The consequence is that the financing decision to use external financing is a higher risk. The results of this study provide evidence that SMEs in meeting their financing needs do not follow the pecking order theory pattern.

As limitation, there are the need for empirical examination regarding controlling for endogeneity. This is important as endogeneity problems most often arise from omitted variables that correlate with one or more independent variables and the dependent variables in the regression model as stated by previous research (e.g., Bijmolt et al., 2005; Rossi, 2014). Omitting such variables induces a correlation between the corresponding independent variables and the dependent variables’ error term (Wooldridge, 2010). That is, the independent variables then not only explain the dependent variable, but also the error in the model. Moreover, even though there have been several previous studies related to gender issues in financing decision, the research model did not involve gender issues.

In the future, this research could also be extended by investigating familiar company characteristics and to link familiarity with financial performance to identify potential arbitration opportunities. Researchers can also try to identify the main factors that encourage investors to feel familiar with the company. In this study, the respondents were only in one ethnicity, for future research to compare with other ethnicities in Indonesia and other countries. The results of this study will assist financial advisors or financial institutions in providing training for SMEs that they can easily absorb and practice directly. One way to reduce home bias, because it is related to emotions and different cultures require a persuasive approach in providing education and training that is easy for them to understand.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors received no direct funding for this research.