The effects of financial literacy and subjective norms on saving behavior

Article history: Received: May 30, 2020 Received in revised format: May 3


Introduction
The Harrod-Domar model stated that the economic growth of a country is determined by three factors, namely saving ratio, the marginal efficiency of capital, and capital. High capital of savings will affect the level of investment (Harrod, 1939). Therefore, saving becomes an important part of macroeconomic policy. The economy of a country is running well or not, which can be seen from the level of savings and investment. From the micro side, saving is an activity that is encouraged to become a behavior of human life that makes it undoubtedly a positive activity to deal with family economic problems that are unexpected. An act according to Theory Planned Behavior (TPB) developed by Ajzen (1991) begins with an intention, namely the saving intention. However, the habit of saving must begin through education that can generate awareness so that awareness and desire to save arise. Saving money is a way to control one's financial use for daily needs. The funds saved will be used for future needs. Gross Domestic Saving to gross domestic product (GDP) rate of Indonesia in 2018 was 33,07%. This data is lower than Cina (46.5%), Singapore (54.5%), and Brunei (56, 3%) (World Bank, 2029). The data shows that Indonesian people's understanding of financial literacy is still low. This understanding is important as a person's first step to saving and saving behavior (Chen & Volpe, 1998). This level shows that saving habits are still not good. BPS (Indonesia Statistics Central Bureau) reports the poverty rate in Indonesia until September 2019 was 25.95 million people (9.82 percent) spread in urban communities (7.02%) and rural areas (13.20%). The biggest contribution to poverty occurred in the staple food sector (73.48 percent). This data shows that the habit of saving becomes a big problem for the people of Indonesia since it has not been able to meet basic needs. (BPS, 2020). This low awareness of saving was realized by the government, so the government issued Presidential Regulation No.82 of 2016 concerning national financial inclusion strategies to increase public awareness of saving behavior (Presidential decree, 2016). This phenomenon proves that Indonesian people have a wasteful life behavior. They prefer to consume than save.
This study is based on the Theory of Planned Behavior (TPB) (Ajina, 2019). TPB stated behavior that is driven by intention, which is associated with saving behavior (Ajzen, 1991). The intention is built from three components: attitudes, subjective norms, and perceptions of belief in the existence of control if one acts (control beliefs) (Arifin, 2018). Applications of TPB related to the study of saving behavior have been widely studied which states that saving behavior is influenced by three factors namely financial literacy and attitude which are then related to literacy financial (Nidar & Bestari, 2012;Sondari & Sudarsono, 2015;Widyastutiet at el., 2016;Satsios & Hadjidakis, 2018). There are studies associated with social influences such as the role of parents and friends to subjective norms, saving interest, and attitude Widyastuti et al., 2016). Jamal et al. (2015) and Mahdzan and Tabiani (2013) saw financial literacy influenced Saving Intention and attitude towards saving, expressed Subjective norms influence saving attitudes.
From previous studies, there was a research gap regarding the relationship between Financial Literacy and Subjective Norms to saving behavior. The gap is in the form of a moderating variable which later became a model in this study. The purpose of this study is to re-explore the determinants of variables by building a model that explains (a) the direct effect of Financial Literacy, Subjective Norms, Saving Intention, and Attitude towards Saving on Saving behavior, (b) the indirect effect of Financial Literacy on Saving behavior through the mediating variable Saving Intention and Attitude towards Saving, and (c) The indirect effect of Subjective Norm on Saving behavior through the Attitude towards Saving moderation variable. The model was built using Theory Planned Behavior. To test the hypotheses, we use a sample of workers who work in the DKI Jakarta area. The reason for selecting these samples is since they have sufficient information on various savings products and can be considered representative of workers in Indonesia.

Theoretical Review and Hypotheses Development
Saving money is a residual income after being reduced by current consumption for a certain period (Browning & Lusardi, 1996). Savings' behavior includes two actions that coincide with the act of saving with the perception of future needs, and savings behavior is carried out to face risks if unexpected problems occur that require large funds (Satsios & Hadjidakis, 2018;Warneryd, 1989). Saving behavior is manifested in various actions based on intentions because of the future views, such as unexpected expenses, forcing to one person to set aside the income before being used for consumption, making estimates of future needs, avoiding expenses for things that are not important, and saving regularly (Ismail at al., 2013), also, in response to actions from the environment or people to obtain satisfaction, needs, and desires (Calderone at al., 2018). The intention to save is an individual's desire and desire to invest or save funds in financial institutions to meet future needs (Mahdzan & Tabiani, 2013). This desire becomes a force that can trigger individuals to pay attention to activities without coercion (Henager & Mauldin, 2015). The magnitude of this power depends on (a) Saving for a purpose, (b) Saving for risk, and (c) Saving barrier, a strong intention to save is realized in the form of the behavior of setting aside income for the unexpected, achieving goals in the future, willing to take risks due to profit opportunities, reducing consumption spending (Widyastuti et al., 2016). The research result proved that Saving Intention has a positive and direct effect on saving behavior (Rodermund, 2012;Satsios & Hadjidakis, 2018;Kisaka, 2014). Based on theoretical studies and relevant research results, hypotheses about the relationship between Saving Intention are built into the Intention of saving behavior.
Hypothesis 1: Saving Intention has a positive effect on saving behavior.
Attitude is a reaction and evaluation in the form of a person's positive or negative feelings towards an object which gives favorable or unfavorable evaluation results of behavior (Ajzen, 199;Berkowitz, 1972). Attitude consists of three components, namely affective, cognitive, and conative. The affective component in the form of individual emotional feelings and the cognitive component is in the form of an individual's belief or perception of an object (Berkowitz, 1972). The cognitive component in the form of a tendency to act by the attitudes of individuals (Secord & Backman, 1965). Attitude towards saving is a feeling and positive or negative response made by someone to saving activities. Individuals who are positive about saving activities will find it easier to do so, and vice versa. The positive attitude of saving includes giving a sense of security, providing benefits, and trends that exist in society (Widyastuti et al., 2016). Furnham and Tankel, (2015) found evidence that there is a positive relationship between saving attitude and saving behavior save even more.

Hypothesis 2: Attitude Towards Saving affects saving behavior.
Financial Literacy is knowledge in managing and solving financial problems (Chen & Volpe, 1998), shaping the behavior and mindset that give an effect on the financial situation of individuals to be better (Raju, Lonial, & Mongial, W, 1995). Behavior that is formed in the form of awareness to manage finances in the future to be able to meet their needs. For this reason, the research tried to increase the knowledge relating to the functions of loans for money, savings, banking services, pension knowledge, and others (Falahati et al., 2012). Individual finances become better because individuals care about the accuracy of transactions, always check financial statements, prepare retirement periods, and make financial asset choices more effective (Magendans, 2014). Good knowledge about financial assets, which shape, mindset, and awareness will encourage individuals to have high Saving behavior, even Financial Literacy is an important driving factor to form saving behavior . The result proved that Financial Literacy has a positive relationship with saving behavior (Jamal et al., 2015;, but the results of other studies proved not to have a significant relationship (Widyastuti et al., 2016;Sabri et al., 2019).

Hypothesis 3: Financial Literacy has a positive effect on saving behavior.
Financial Literacy is a process to improve knowledge and skills. This ability will be an individual's strength in managing finances and raising awareness (Sanjeewa & Hongbing, 2019) of the high importance of being able to meet financial needs in the future (Falahati et al., 2012). The process builds an attitude which then forms intentions (Ajzen, 1991a). The better a person's financial literacy is, the motivation to save is stronger. The result of research conducted on high school students proved that Financial Literacy has a positive effect on Saving Intention ( (Lestari et al., 2017). But in other research financial literacy was not significant to saving intention (Widyastuti et al., 2016).
Hypothesis 4: Financial Literacy has a positive effect on saving Intention.
Based on the theoretical framework that explained the partial direct relationship between Financial Literacy and Saving behavior (Hypothesis 3), Financial Literacy with Saving Intention (Hypothesis 4), and Saving Intention with saving behavior (Hypothesis 1), it should be assumed that Saving Intention mediates the relationship between Financial Literacy with Saving behavior. Thus, one hypothesis is built for this framework.
Hypothesis 5: Saving Intention moderates the relationship between financial literacy and saving behavior.
Individuals who have financial literacy in the form of knowledge and techniques to manage finances will be able to control financial use better. Some of the income will be saved for future financial needs (Zait & Bertea, 2014). This ability is an attitude dimension in TPB theory (Ajzen, 1991) due to financial literacy that is not only related to knowledge and abilities, it is also part of one's cognitive abilities (Supanantaroek et al., 2016). The attitudes related to finance (attitude toward saving) are a person's psychological characteristics related to personal financial problems. Openness to financial information, awareness to manage finances, not impulsive in consumption, and awareness to prepare finances in the future, will be an increasingly better outlook which is embodied in the form of individual attitudes to save. So, it can be stated that financial literacy has a positive influence on attitudes toward saving (attitude towards saving) (Gutter & Copur, 2011;Samadi, 2018;Salem & Chaichi, 2018). Little research has been conducted to examine the relationship between financial literacy and attitude towards saving. Zahriyan (2012) found evidence that financial literacy has a positive effect on attitude Towards Saving.
Hypothesis 6: Financial Literacy has a positive effect on attitude Towards Saving.
Theories that explain the partial relationship between financial literacy and saving behavior (Magendans, 2014), Financial Literacy with Attitude towards Saving (Gutter, 2008), and Attitude towards Saving on Saving behavior (Widyastuti et al., 2016), build a theoretical framework model, namely Attitude towards Saving moderates the relationship between Financial Literacy and Saving behavior.
Hypothesis 7: Attitude towards saving moderates the relationship between financial literacy and saving behavior.
Subjective norms in TPB are guided by social values that encourage someone to do or not to do an action. If the value allows you to do something and has a positive impact on individuals, then the individual will be motivated to follow that social value (Fiksenbaum et al., 2017). Awareness of the importance of setting aside his income for future needs so the person will set aside the income to save. If awareness is supported by the benefits provided by saving, then building a character in the form of saving habits (Sabri & Juen, 2014). The results of previous studies proved that positive subjective norms affect saving behavior (Widyastuti et al., 2016).
Hypothesis 8: Subjective norm influences saving behavior. Jamal et al., (2015) and  explained that subjective norms originating from the environment, such as friends, family had a great influence on one's attitude. An impulsive environment will shape a person into consumptive. As a result, individual attitudes in saving become low, and vice versa. Kisaka, (2014) and Rodermund, (2012) proved that subjective norms have a positive effect on attitudes toward saving (attitude toward saving). Existing attitudes will result in different ways of saving depending on the individual environment that shapes them.
Hypothesis 9: Subjective norms have a positive effect on attitudes toward saving (attitude toward saving). Kisaka, (2014) explained that subjective norms form beliefs about the value of the environment that allows or is prohibited from doing something. The environment is like friends, family, and couples, if the environment has a high impulsive environment, then individuals will tend to have a higher consumptive nature. This will form a low saving behavior. Satsios and Hadjidakis (2018) explained the subjective norm will form an attitude toward saving. (Gerhard et al., 2018) explained that attitude toward saving will form saving behavior. Based on the theories that explain the relationship between subjective norms with attitudes and saving behavior, and attitude with saving behavior, it can be said that attitude mediates the relationship between subjective norms with saving behavior.
Hypothesis 10: Attitude towards saving mediates the relationship between subjective norms and saving behavior.

Research method
The subjects in this study were workers under the age of 35 working in the DKI Jakarta Region. The reason for using samples of workers whose domicile in Jakarta is that they have sufficient access to information because they are in capital country. In addition, they are young workers who generally have good financial literacy. The research objects that build the research model in this study are financial literacy, subjective norm, saving intention, attitude toward saving and saving behavior. Saving intention and attitude toward saving become a moderator variable that explained the relationship between financial literacy with saving behavior. The research model also places attitude toward saving as the moderating variable of the relationship between Subjective Norms and saving behavior. Primary data is used to test the model. Data obtained by distributing questionnaires. The questionnaire was distributed online using the Google document application. Data samples were young employees (under 35 years old), each indicator item was measured using a scale. The range of scales ranges from 1-10, from the worst scale to the best scale. The result obtained as many as 469 data samples. This research examined the model by having the test of outer and inner models. Outer model is used to test validity and reliability while Inner model is implemented to examine the theory of the model, goodness of fit and the coefficient of determination and to verify the data used PLS based on SEM. PLS is used since it can analyze latent variables and the data which are not normally distributed . The variable in this study is the latent variable and it is measured using indicators. Each indicator is measured on a scale of 1-10. Scale 1 for respondents who answered poorly, and scale 10 for respondents who gave the best response. The variables and indicators used in this study are presented in Table 2. The research sample is workers working in the Capital City of Jakarta area. The data collected were 469 samples. The number of samples is based on demographic criteria such as gender, age group, and place of work, education, type of work, income, and status of ownership of residential homes. The sample is dominated by age groups that have just entered the world of work (47%) with 61% education is Bachelor (S-1), living relatively scattered in Five Areas of Jakarta. Most of the participants in our survey (96%) were single. Based on income, the highest amount was 38% above Rp10, 000.000. A description, complete description of the respondents is presented in Table 1. Reliability Tests are carried out using three criteria, namely Cronbach's Alpha, rho_A, and Composite Reliability. All three of these criteria use the same criteria. One variable is declared reliable if its value> 0.7 . Test result with these three criteria conclude that the model variables are declared reliable. This result is presented in Table 4. The purpose of Inner Model is to show three things, hypothesis testing, test the accuracy of the model (Goodness of Fit), and the coefficient of determination. The result of the statistical hypothesis test is presented in Table 5. In Table 5 it is proved that the direct relationship between the two variables is significantly positive, except for the relationship between subjective norms and saving behavior. This evidence supports hypotheses 1, 2, 3, 4, 6, and 9 but rejects Hypothesis 8. Indirect relationship proved that the Saving Intention and Attitude towards Saving variables are moderating variables between Financial Literacy and Saving behavior (Hypotheses 5 and 7). Attitude towards Saving variable moderates the relationship of subjective norm with saving behavior (Hypothesis 10). Test the model using the NFI Test: NFI values range from 1-10. The value indicated that the closer to 1, the less fit the model, and vice versa, the closer to 1, the more fit the model . NFI Test Result = 0.689. This proves the model is relatively fit, since it is close enough to 1.0. The coefficient of determination test is to measure how much the endogenous/ dependent variable is explained by the endogenous/ independent variable. The test determines the value of R Square . The result of the analysis of the coefficient of determination are presented in Table 6.

Discussion
Financial Literacy has a relationship, both directly and indirectly, with saving behavior. Indirect effects are mediated through saving intention and attitude towards saving. The direct relationship between the variable's financial literacy, saving intention and attitude towards saving has a positive relationship with saving behavior. This proves that financial literacy, saving intention, and attitude towards saving has a very important role to encourage saving behavior. The positive relationship between Financial Literacy and saving behavior will be encouraged by increasing the ability to control financial statements since this indicator provides the largest contribution from the Financial Literacy variable. Saving Intention mediates the relationship between Financial Literacy with Saving behavior proves that the Saving Intention variable will strengthen the relationship of financial literacy with saving behavior. Saving intention is a strong determination to save (Jamal et al., 2015). The magnitude of this power depends on (a) Saving for a purpose, (b) Saving for risk, and (c) Saving barrier (Widyastuti et al., 2016). Therefore, saving intention needs to be increased by raising awareness that future circumstances run the risk of large and unforeseen expenses. From the external side, it is very important to create conditions, namely the ease of saving on various assets and easy to control (Saving Barrier). Progress in the field of financial technology (fintech) is important because fin-tech provides these two things, namely the ease of choosing investment assets and the ease to control them. Attitude towards saving mediates the relationship between Financial Literacy and saving behavior proving that the Attitude towards Saving variable will further strengthen this relationship. The attitude that needs to be encouraged is a positive attitude about the importance of saving (Berkowitz, 1972). A positive attitude can be enhanced by the knowledge that risk is something that can be controlled and cannot be avoided (cognitive factors) and financial needs in the future are certain (Secord & Backman, 1964).
The Attitude towards saving variable mediates the relationship between Subjective Norms and saving behavior. But the Subjective Norm has no direct relationship with saving behavior. This proves that the existence of Attitude towards saving has a very important position. Positive subjective norms, such as encouragement from friends and family, cannot encourage a person to have saving behavior. Subjective Norm will encourage saving behavior if the individual has the behavior in the form of a positive response to saving activities. A positive response will grow if a sense of security is created when saving. Safety (ATS-1) is the main determining factor because it gives the highest weight of the attitude towards saving indicator. This finding proves that subjective norms are not an important consideration for investors in developing savings behavior. This finding is contrary to the results of the study Kisaka, (2014) and Satsios and Hadjidakis, (2018) but it is still within the framework of the TPB which explains that subjective norm is a dimension of intention (Ajzen, 1991).

Implications and Limitation
This research has an important impact, both directly and indirectly. In explaining saving behavior. This study implies that Financial Literacy and Subjective Norm are positive factors that will encourage saving behavior for the better. To improve Financial Literacy and Subjective Norms, Saving Intention and Attitude towards saving needs to be grown on each individual. This can be done through the ease of saving, the ease of saving to control their savings, and the awareness of the importance of preparing finances in the future. Furthermore, the existence of fin-tech can be two of three factors, namely the ease of saving and controlling savings. Fin-tech in Indonesia has not been able to provide security for the culprit. First, this study focuses on the workforce of formal workers who live and work in the Province of DKI Jakarta, so this sample does not represent the actual condition of the population of workers in Indonesia Therefore the findings in this study cannot be generalized and applied as a whole of workers in Indonesia. Most workers in Indonesia work in the informal sector. For better research results that depict workers in Indonesia, then the number and variation of samples need to be expanded. Second, the focus of this research is attributed to the role of financial literacy and subjective norm in influencing saving behavior, many other variables are influencing saving behavior. Future studies could include other variables in the model. Including these additional variables can shed further light on financial literacy and the subjective norm on saving behavior. Third, the extended saving behavior model could also be tested in specific products of securities (e.g. bond, common stock, insurance, etc.). Fourth, in this study using the TPB theory approach, for future research TPB can be combined with other management theories, such as innovation theory, Dynamic Capability Theory, etc.