Examining the Impact of Financial Literacy, Financial Self-Control, and Demographic Determinants on Individual Financial Performance and Behavior: An Insight from the Lebanese Crisis Period

Abstract

This study investigated the effects of financial literacy, financial self-control, and demographic determinants on individual financial performance and behavior during the Lebanese crisis period between 2019 and 2021. To the best of our knowledge, this may be the first study that compares the determinants of financial behavior for different generations, genders, marital statuses, and education and income levels. To do so, we conducted a comprehensive survey of 328 individuals and performed a logistic regression analysis. The empirical results show that an individual’s financial performance and behavior are positively affected by financial literacy, financial self-control, and demographic factors, in particular education and income levels. In addition, when we focused on the demographic factors, the results reveal that having good financial literacy increases the likelihood of an individual’s financial performance and behavior, in particular for Generations X and Z, males and females, single and married people, low- and high-educated people, and low- and high-income individuals. However, having good financial self-control only increases the likelihood of an individual’s financial performance and behavior at highly educated levels. The results are robust and come from various performed methodologies, and the results have important policy implications. The policies should be focused on enhancing an individual’s financial behavior and helping young adults acquire skills in self-control. Policies could also motivate local financial institutions to offer a variety of financial products and investment opportunities, targeting low-income and low-educated individuals, by providing subsidized funds with parallel mandatory financial studies.

1. Introduction

A more sustainable economy and society are the results of financial education, which encourages more ethical financial behavior and more ethical financial product usage. Not all financial behavior and performance existing on the market today display sustainability-related traits. In this regard, financial literacy plays a key part in a shift of strategy, one from the supply side via the consolidation of sustainable finances and one from the demand side through more sustainable consumer behavior [1,2]. Due to the significance of financial education in making ethical financial and economic decisions that will enhance future well-being, both academics and those in charge of economic policy are becoming more and more interested in the subject [3].

In the literature, several studies have investigated the determinants of financial behavior. Along with demographic control factors, such as age, gender, and marital status, the findings of several studies [4,5,6] show that financial literacy and financial self-control have a significant role in explaining an individual’s financial performance and behavior. People with a sound financial understanding are more likely to adopt wise financial practices that enhance their quality of life [7]. Financial responsibility, such as the amount and quality of savings, informed expenditure decisions, and informed investment decisions are positively correlated with education and financial knowledge [8]. Moreover, having high self-control when it comes to money and spending could help individuals make better financial decisions [4].

What about the determinants of an individual’s financial performance and behavior in Lebanon? Since the fall of 2019, Lebanon has been facing a severe financial, economic, and political crisis. The Lebanese crisis has evolved into one of the top three most severe crises since the mid-1990s, according to the World Bank: Lebanese Economic Update in October 2021 [9]. Throughout the crisis, Lebanon has been facing a severe drop in GDP, a banking liquidity problem, unofficial capital controls, a severe devaluation of the national currency, hyperinflation, and the COVID-19 crisis that impacted the whole world. Consequently, the entire Lebanese population has been facing economic, financial, and political uncertainty. The country has also been witnessing a wave of immigration among its Millennials and Generation Z (those born between 1994 and 2012). The impact of the abovementioned crisis has taken its toll differently on various Lebanese income levels [10]; the middle class has been most affected, and the Lebanese were a victim of a banking system Ponzi scheme due to their lack of financial understanding and risk to return assessment [11].

On the other hand, Generations Y (those born between 1989 and 1994) and Z (those born between 1994 and 2012) are the future of our society and economy. Their behavior, evolution, and future financial performance will impact the future and sustainability of our economy and well-being. Those individuals represent the consumers and savers of today and the investors of tomorrow. The postcrisis Lebanese economy will hence heavily depend on the behavior of this segment of the population, compared to Generation X (those born between 1960 and 1989), whether residing in Lebanon or abroad. On the other hand, Millennials and Zoomers, the generations born after 1989, may tend to make more uncontrolled financial decisions than their predecessors considering that they grew up in a relatively prosperous, stable, and open Lebanon compared to the Baby Boomer generation. After the postwar and postfinancial crisis of the 1980s, Lebanese Millennials and Zoomers heard of a crisis but never experienced it until 2019.

Moreover, Lebanon is facing not only a financial and political crisis but also countless setbacks, which prompted us to explore the origin of all decisions, more particularly individual financial decisions. To foresee the future, one should become interested in the behavior of young generations and how they perceive and prepare for the future, especially during a crisis. Hence, considering these facts, Lebanon could be an interesting case study, and the results could have a significant message for policymakers and scholars.

This study has significant contributions. First, this may be the first study aiming to explore the determinants of financial performance and behavior in Lebanon. In fact, to the best of our knowledge, limited studies have focused on exploring the determinants of financial behavior among Lebanese people. Second, there may be few studies (See the studies by [12,13]) on financial behavior, but they do not focus on the determinants of financial behavior especially by considering the demographic characteristics of individuals during the 2019–2021 financial crisis period. Third, despite prior studies, this study performed various approaches, such as logistic regression, generalized linear regression, and multinominal logistic regression to conduct this study. Lastly, the studies on Generations Y and Z (born after 1989) around the world have been limited to a descriptive analysis, as they have recently entered the labor market and thus entered national survey samples (See the study by [4]). Few financial behavior studies have been conducted on Millennials in the USA and other parts of the world (See the studies by [4,14,15,16,17,18]). To bridge this gap, the current study extends the frontier of knowledge by investigating the determinants of financial behavior among a set of demographic categories in the developing context of Lebanon, especially during the crisis of 2019–2021.

This study aimed to shed light on the following questions:

(1)

How are financial literacy, financial self-control, and demographic factors impacting an individual’s financial performance and behavior in Lebanon during the crisis period?

(2)

How are demographic characteristics of individuals impacting the determinants of an individual’s financial performance and behavior?

To answer the above questions, we conducted a comprehensive survey of 328 individuals and performed a logistic regression analysis. We also classified the individuals based on their age, gender, marital status, number of dependents, education levels, and income levels to specifically explore the role of the demographic factors. We then verified the robustness of the results using generalized linear regression and multinominal logistic regression.

The findings demonstrate that among all generations, financial literacy and self-control had a positive effect on enhancing an individual’s financial behavior. When we focused on the impact of demographic variables, the findings show that having a basic understanding of finance increased the likelihood of good financial behavior for members of Generations X and Z, both sexes, single and married people, low and high earners, and low- and high-educated people. For highly educated persons, financial self-control increased the chance of sound financial behavior and performance; however, it was less significant when compared to other demographics.

This result implies that good financial conduct and performance are favorably correlated with sustainability and long-term well-being. Therefore, governments need to pay closer attention to how younger generations behave financially. By incorporating financial literacy into school curricula and assisting children and teenagers in developing self-control abilities, the policies should work to promote good financial conduct. By providing subsidized funds, policies could also encourage local financial institutions to target low-income and undereducated people with a range of financial products and investment opportunities, while providing them with obligatory financial education as one of the financing conditions. The sustainability of the beneficiaries’ income and their financial performance in the future will both be improved by these programs.

The remaining parts of the paper are organized as follows: The Section 2 reviews the literature with a focus on theoretical, conceptual, and empirical discussions. The Section 3 outlines our research methodology and data. The Section 4 presents the research results and a discussion of key findings. The Section 5 outlines the contributions and implications for practice and research.

2. Literature Review

This section reviews theoretical, conceptual, and empirical literature to provide a strong foundation for the current study, as well as the need to properly situate the study within the body of existing knowledge in the field of financial behavior and its importance on financial performance and economic welfare and stability.

2.1. Theoretical Framework

This study adopted the social cognitive theory (SCT), which is one of the most widely used behavioral theories for explaining changes in the financial behavior of individuals and organizations [19]. It was developed in the field of psychology by Wood and Bandura [20] purposely to better understand how people learn within their social environment, including determinants of individual and group behaviors. In other words, the SCT postulates that behavioral change is strongly influenced by environmental factors, personal factors, and behavioral attributes and that each of these factors may affect or be affected by other factors [19]. Furthermore, Ozmete and Hira [19] opined that the central tenet of the SCT is self-efficacy, which refers to the belief in one’s ability and capability to perform and exhibit certain behavior that is perceived to be induced by positive incentives (positive expectations or rewards for performing the behavior). Self-efficacy is important in all learning situations because watching others perform a behavior influences the observers’ perceptions of their ability to perform the same behavior (self-efficacy), including shaping the expected outcomes and serving as strategies for effective performance [21].

However, self-efficacy declines when negative expectations outweigh positive expectations. Exhibiting specific behavior or performing certain actions comes with immediate and long-term outcomes or consequences. Self-efficacy is not a monolithic factor that is static; rather, it is progressive and can be enriched and improved through different stages of learning, such as providing clear instructions, skill development/training, and mentoring/coaching, to help model desired behavior, including financial risk reduction [19,22]. Therefore, it is useful and rewarding to understand how people and groups interact with their environment to manage their behavioral change based on stage elements and stage-specific attributes [23,24]. Because the SCT is a learning theory, it is useful in this study to provide a theoretical foundation for understanding the role of financial literacy in influencing the financial behavior of generational categories during periods of crisis. Therefore, self-efficacy is important in behavioral change analysis because it determines a person’s behavioral change after weighing the benefits (positive expectations) against the costs (negative expectations) within a given context or situation.

2.2. Financial Behavior Determinants and Hypothesis Development

Financial behavior determinants are many and diverse. These determinants are levelers, influencers, and motivators of financial decisions. In the financial and behavioral literature, the financial behavior of individuals is defined as how they decide on day-to-day financial matters, such as spending, saving, and wise investing [4,5,6]. Additionally, financial behavior connotes any aspect of human behavior that relates to and has connections with the management of money and financial resources [7].

Considering its importance, financial behavior is measured by evaluating individuals’ credit reports, credit scores, health insurance and amounts, life insurance and amounts, types of investments made by these individuals, and their financial retirement plans [5]. A financially well-behaved individual is never late on a bill payment, never bounces a check, and pays his credit card in full and on time. Individuals described as well-behaved financially also balance their checkbooks weekly, review their budget, and estimate their income and expenditures monthly [6]. In addition, being financially well-behaved means that the individual has savings and adequate investments for his needs and is risk averse [25]. Moreover, financially well-behaved individuals save more for their retirement [25] and have a diversified investment portfolio and adequate returns [26]. They also have less debt [27] and lower borrowing costs [28]. Subsequently, financially well-behaved individuals are ready to meet future uncertainty and financial shocks [29], such as the Lebanese 2019 financial crisis. Therefore, financial behavior is a result of many factors that dictate whether an individual is well-behaved financially or poorly behaved financially. Those factors cover mainly three evaluation areas, individuals’ current account financial management, investment decisions, and future financial management. Good financial behavior is defined as individuals being able to manage correctly all three financial areas.

Good financial behavior is determined by many factors. We will focus here on two of them: financial literacy and financial self-control, in addition to the demographic control variables, such as age, gender, and marital status. Several scholars mentioned that financial behavior differs among generations and generational categories. Therefore, the Millennials—the generation born between 1989 and 1994—compared to older generations, may show reckless financial performance and behavior [30] as well as financial overconfidence, especially among those who lack financial literacy [4]. In addition, financial performance and behavior depend on the financial environment of individuals.

2.2.1. Financial Literacy

Financial literacy has a critical role in determining an individual’s financial behavior. Good financial education and good financial comprehension of individuals will dictate their good financial decisions. Financial literacy plays a significant role in influencing one’s financial behavior, which helps people make better financial decisions in general and during financial crises in particular [6,31,32]. Financial literacy also means that those who comprehend financial matters are more likely to make better financial decisions and take adequate action regarding their future financial necessities [6,31]. The Gray, Montagnoli, and Moro [8] UK 2020 study revealed that education positively correlates with good financial behavior, such as savings in quality and amount [8]. The works by Bhushan and Medury [33], Andarsari and Ningtyas [34], and Rai, Dua and Yadav [35] showed that financial literacy has a positive impact on the financial behavior of women. Consequently, scholars [8,31,32] have agreed on the fact that good financial education and knowledge lead to better financial decision making, and that is true in all countries, cultures, and circumstances.

Good financial knowledge encourages people to form positive financial behaviors that would improve their quality of life [7]. Additionally, ref. [36] explained that changing the way people think financially, including their beliefs, attitudes, and goals, would likely change their financial behavior and financial capability. The empirical study of Grable and Joo [37] found that comparatively younger persons living in rented houses and who experienced other financial stressors exhibited poorer financial behaviors and hence required financial literacy or financial help-seeking. Furthermore, Ratten [38] found that people’s entrepreneurial inclination and financial learning tendency strongly influence and determine how they respond to marketing and e-finance (e.g., mobile banking). Hence, based on the above discussions and the facts from the extant literature, we expected to find a positive relationship between financial literacy and an individual’s financial behavior.

H1. 

Financial literacy has a positive impact on the individual’s financial behavior during the crisis period.

2.2.2. Financial Self-Control

An individual’s level of self-control, defined as “restraint exercised over one’s impulses, desires, or emotions” [4], has a considerable impact on his financial behavior. Self-control is another determinant of financial behavior, and individuals having good self-control toward spending and money could lead to making better financial decisions.

Individuals are considered well self-controlled if they control their spending behavior and if they have a good credit history. High self-control levels are linked to spending less and saving more [39], while low self-control levels lead to impulsive spending [40]. Thus, self-control is linked to financial behavior. Generally, financial self-control is measured in the literature by the Credit Card Use Scale containing 12 items to assess credit card behavior, such as always abiding by card limits, having a hard time repaying the credit card, impulsive shopping with card payments, and withdrawing cash from the credit card [41]. Self-control is also measured by the Consumer Spending Self-control Scale, which is the self-imposed standard reflecting the ability of individuals to monitor spending, focus on long-term financial goals, consider the consequences of purchases, etc. [42]. The self-control variable for individuals could therefore be a combination of self-restrictions and professional evaluations conducted by creditors.

Financial behavior and performance during a crisis depend on the individual’s financial literacy and self-control. It is also worth noting that financial well-being is better among individuals who have a higher level of financial education and knowledge [5,6,31,32,43] and among those who exercise self-control over their financial decisions [4,39].

Hence, based on the above discussions and the facts from the extant literature, we expected to find a positive relationship between financial self-control and an individual’s financial behavior.

H2. 

Financial self-control has a positive impact on the individual’s financial behavior during the crisis period.

3. Data and Methodology

3.1. Data Description

In this study, the data were collected by a survey broadcasted online in English and Arabic. The study adopted the stratified random sampling method and targeted two main communities: the USEK community, which includes employees, faculty members, and students, and the Lebanese University community, which includes employees, faculty members, and students. Both communities cover the three targeted generations of all income levels, both genders, and different education levels and are spread all over Lebanon. Questions were a collection of Likert scale and multiple choice qualitative questions covering basic financial knowledge, self-control, and performance and behavior of individuals.

Table 1. Sample distribution.

YOB	GEN	FE	MA	SI	MA	LE	HE	LI	HI	AVG Age	AVG No. of Dependents
1960–1989	X	88	81	49	120	16	153	15	154	43.53	1.59
1989–1994	Y	34	31	37	28	1.00	64	11	54	29.46	0.55
1994–2003	Z	49	45	90	4	11	83	55	39	20.98	0.25
	Total	171	157	176	152	28	300	81	247	34.28	1.00

Note: YOB is the year of birth; GEN is generation; FE is female; MA is male; SI is single; MA is married; LE is low educated; HE is highly educated; LI is low income; and HI is high income.

presents the distribution of 328 individuals who participated in this survey. As shown, 171 were females, and 157 were males. A total of 169 individuals were born between 1960 and 1989 and belonged to Generation X, with an average age of 43.53. Sixty-five were born between 1989 and 1994 and belonged to Generation Y, with an average age of 29.46. Ninety-four were born between 1994 and 2003 and belonged to Generation Z, with an average age of 20.98. A total of 152 were married; 300 were highly educated, and 247 had a high income. More than one person was typically financially dependent on Generation X members, compared to less than one for Generation Y and Generation Z.

3.2. Variable Selection

3.2.1. Dependent Variables

The dependent variable used to construct Equation (1) was financial behavior. We measured financial performance and behavior using a set of 9 questions on whether the individual sets a monthly personal budget, reviews the budget at the end of the month, plans the budget to reach his financial objectives, pays his bills on time, pays his loan installments on time, saves money monthly, has an emergency fund, has a pension fund, and receives credit when needed at the end of the month. These questions reflect whether individuals make good financial decisions. Responses were recorded on a Likert scale from 1 to 5. The number 1 reflects that the individual displays financial behavior in question 0% of the time, and 5 reflects that the individual displays financial behavior in question 100% of the time. The construction of the financial behavior variable is summarized in

Table 2. Sample questions for measuring financial performance and behavior variable.

Financial Behavior Evaluation Questions	Measurements
Question 1: I set a monthly personal budget. At the beginning of every month, I calculate the expected revenue and the expected expenditures.	1 reflects poor financial behavior 5 reflects good financial behavior
Question 2: I review and assess revenue and expenditures at the end of every month and analyze whether the budget was met or not.
Question 3: I plan my budget to reach my financial objectives (save or invest a target amount of money).
Question 4: I pay my bills (phone, electricity, rent …) on time.
Question 5: I pay my loan installment on time.
Question 6: I place a certain amount of money aside as savings every month.
Question 7: I have an emergency fund other than my savings.
Question 8: I have a pension fund other than my savings.
Question 9: I get a credit or a loan when I need money at the end of the month.	1 reflects good financial behavior 5 reflects poor financial behavior (Reverse coded)
Financial behavior binary score	Any score above or equal to 3.5 reflects good financial behavior and is assigned number 1, and any score below 3.5 reflects poor financial behavior and is assigned number 0

. This study used the median grade benchmark to evaluate the performance and behavior of individuals. Thus, the individuals had a bad financial performance if the median score was below 3.5 and vice versa.

3.2.2. Independent Variables

Financial literacy was measured using 5 multiple choice questions showing knowledge regarding the main financial concepts: inflation, risk, diversification, interest calculation, and liquidity. Responses for each question were converted into binary results, and each respondent had a score over 5, showing his level of financial knowledge. The details are shown in

Table 3. Multiple choice questions evaluating financial literacy.

Financial Literacy Evaluation Questions	Answer Choices	Answer Treatment
Question 1: Imagine that the interest rate on your savings account was 1% per year and inflation was 2% per year. After 1 year, would you be able to buy:	A.1. More than today with the money in this account A.2. Exactly the same as today with the money in this account A.3. Less than today with the money in this account A.4. Don’t know A.5. Refuse to answer	1 for the correct answer “A.3.” and 0 for the incorrect answers.
Question 2: When an investor spreads his money among different assets, does the risk of losing a lot of money:	A.1. Increase A.2. Decrease A.3. Stay the same A.4. Don’t know A.5. Refuse to answer	1 for the correct answer “A.2.” and 0 for the incorrect answers.
Question 3: Suppose you had $100 in a savings account and the interest rate is 20% per year and you never withdraw money or interest payments. After 5 years, how much would you have in this account in total?	A.1. More than $200 A.2. Exactly $200 A.3. Less than $200 A.4. Don’t know A.5. Refuse to answer	1 for the correct answer “A.1.” and 0 for the incorrect answers.
Question 4: Which of the following is considered a riskier behavior?	A.1. Investing your money in a company with a return rate of 15% A.2. Investing your money in a company with a return rate of 5% A.3. Don’t know A.4. Refuse to answer	1 for the correct answer “A.1.” and 0 for the incorrect answers.
Question 5: Which of the following is considered to be an emergency fund?	A.1. The 100,000 USD land you own A.2. The 10,000 USD savings account A.3. The 10,000 USD saved in your house vault A.4. The retirement account A.5. Don’t know A.6. Refuse to answer	1 for the correct answer “A.3.” and 0 for the incorrect answers.
Financial literacy variable		Score over 5

below.

In addition, inancial self-control was measured using a set of 5 questions presented in

Table 4. Sample questions for measuring financial self-control variable.

Financial Self-Control Evaluation Questions	Answers	Answer Treatment	Measurements
Question 1: “You got a bonus of 5000 USD cash this month. Will you book a holiday tomorrow that costs 2500 USD?”	“No” or “Maybe” suggest the individual is not an impulsive buyer.	1 for answers “No” and “Maybe” and 0 for the answer “Yes”.	The binary score of Holiday budget self-control value: Low self-control is indicated by 0, and high self-control is indicated by 1.
Question 2: “I tend to live for today without thinking about the future”. Question 3: “I prefer spending money now to saving it for the future”. Question 4: “Money is there to be spent”.	Likert scale scores: 1 means the individual strongly disagrees with the statement; 5 means the individual strongly agrees with the statement.	Median of the 3 Likert scale scores: Less than 3 reflects a strong attitude toward calculated spending and savings. At least 3 reflects a poor attitude toward calculated spending and savings.	The binary score of Attitude toward spending and savings value: A strong attitude toward calculated spending and savings is indicated by 1, and a poor attitude toward calculated spending and savings is indicated by 0.
Question 5: “Do you exercise any self-control strategies to avoid temptations?”	0 for answers “No” and “maybe” and 1 for the answer “Yes”.	0 reflects poor self-awareness, and 1 reflects high self-awareness.	The binary score of Strategies of financial self-control: Low engagement in strategies of self-control is indicated by 0, and high engagement in Strategies of self-control is indicated by 1.
Financial self-control variable		Median Holiday budget value, Attitude toward spending value, and Strategies of self-control value.	The binary score of financial self-control: Low self-control is indicated by 0, and high self-control is indicated by 1.

that reflected the degree of self-control respondents have. The answers to the five questions reflected whether the individual was an impulsive spender, what his attitude toward spending and saving was, and whether he believed he was exercising self-control strategies. Responses to this set of questions were transformed into binary results, with 0 reflecting poor self-control and 1 reflecting high self-control. The detailed construction of these variables is presented in Table 4.

Furthermore, following the literature, we included the demographic control variables, such as age, gender, marital status, number of dependents, education level, and monthly income level [44]. Age was a continuous independent variable. Gender was assigned 1 for females and 0 for males. Marital status was also assigned 1 for married and 0 for single people. The variable number of financially dependent individuals the respondent (number of children, spouses, elderly people, or others that may rely on the person financially and for whom the respondent is financially responsible) has was left unchanged as a discrete variable because the logistic regression allowed discrete variables as independent variables. The higher-level education variable was transformed into two categories: high education level for university and graduate respondents and low education level for high school and primary school respondents. A high education level was assigned 1, and a low education level was assigned 0. The monthly income variable was transformed into two categories: high income for respondents with a monthly income over LBP 10,000,000 and low income for respondents with a monthly income lower than LBP 10,000,000. The high income level was assigned 1, and the low-income level was assigned 0.

3.3. Models and Methodology

3.3.1. Conceptual Model

Following the prior studies [5,43,44], we constructed the conceptual model below to examine the relationship among demographics, financial literacy, financial self-control, and financial performance and behavior. The conceptual model of financial behavior had three fundamental parts: financial literacy, financial self-control, and other social–personal determinants. As explained above, financial behavior was based on six essentials, namely budgeting, self-financial evaluation, financial allocation, financial organization, saving and borrowing, and budget planning. Financial literacy was also the mix of awareness, information, skills, attitudes, and behaviors required to make wise financial decisions and, ultimately, attain individual financial well-being. As for self-control, it was based on three main concepts: attitude toward spending, strategies adopted to avoid temptations, and holiday budgeting. Gender, age, marital status, number of financial dependents, education level, and income level represented the demographic determinants of the financial behavior concept.

Equation (1) shows the practical linear relationship between the dependent and independent variables.

$$\mathrm{Financial}\mathrm{behavior}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1\mathrm{Financial}\mathrm{literacy}+{\mathsf{\alpha }}_2\mathrm{Financial}\mathrm{self}-\mathrm{control}+{\mathsf{\alpha }}_3{\displaystyle \sum }\mathrm{CV}+\mathsf{\epsilon }$$

(1)

where financial behavior is the dependent variable. Financial literacy and financial self-control, in addition to other control variables (CV), were considered independent variables. ε is an independent error term.

3.3.2. Methodology

In examining the effect of independent variables on financial behavior, this study performed logistic regression because the dependent variable financial behavior was binary in type [45,46,47,48]. In addition, the independent variables were mixed in type, with financial literacy going from 1 to 5, while financial self-control was binary in type. Age was a continuous variable. Gender and marital status were also binary variables. Education levels and income levels were translated into 0 and 1 variables. Our collected data were reduced to binary, over 5 scores, or continuous data. For further model result verifications, the generalized linear models and multinominal logistic regression were also used because both tests can be used when the dependent variable is binary in type.

Even though logistic regression was perhaps the most prevalent type of regression used for binary variables in the research literature, there are various other tests, such as the generalized linear model and the multinomial logistic regression that can be used. When values on the dependent variable represented unordered categories, multinomial logistic regression was typically utilized. When there was no linear relationship among the variables and the response variable was discrete or categorical rather than continuous, the generalized linear model was used. Both cases applied to our study [47,49,50,51].

4. Empirical Results

4.1. Univariate Analysis

As discussed, the dependent and independent variables included in the study were binary in form, which meant they only held two values, a minimum value of 0 and a maximum value of 1. The exceptions were financial literacy, age, and the number of financial dependents, which had Likert scale, continuous, and discrete forms, respectively.

As indicated in

Table 5. Descriptive statistics.

Variables	Sample Size	MIN	MAX	Mean	Std. Dev.	Skewness	Kurtosis
Financial behavior	328.00	0.00	1.00	0.75	0.43	−1.14	−0.70
Financial literacy	328.00	0.00	5.00	2.78	1.60	−0.21	−1.14
Self-control	328.00	0.00	1.00	0.56	0.49	−0.25	−1.94
Generation	328.00	0.00	1.00	0.44	0.49	0.23	−1.95
Age	328.00	18.00	61.00	34.28	11.87	0.56	−0.50
Gender	328.00	0.00	1.00	0.52	0.50	−0.08	−2.01
Marital status	328.00	0.00	1.00	0.46	0.49	0.14	−1.99
No. of dependents	328.00	0.00	5.00	1.00	1.24	0.98	0.12
Education level	328.00	0.00	1.00	0.91	0.28	−2.98	6.93
Income level	328.00	0.00	1.00	0.26	0.43	1.10	−0.78

, the dependent variable financial behavior had an average of 0.75, which meant that the respondents had good financial behavior rather than not, but with a standard deviation of 0.43, which meant values were relatively distant from the mean, so respondents either had good basic financial knowledge or bad basic financial knowledge.

As for the independent variable financial literacy, which represented a score of the level of financial knowledge, it ranged from 0 (no basic financial knowledge) to 5 (perfect basic financial knowledge). The average grade of financial knowledge of the 328 respondents was 2.78 over 5, which meant that the respondents in our sample had mostly bad basic financial knowledge rather than good basic financial knowledge given that 3.5/5 was the benchmark of good basic financial knowledge. Moreover, the financial self-control independent variable had an average of 0.56, which meant the respondents had moderately good self-control rather than not, with a standard deviation of 0.49, which meant the values were relatively distant from the mean, so the respondents either had good self-control or bad self-control.

The ages of the respondents ranged between 18 and 61, while the average age of the respondents was 34.58. The independent variable number of financially dependent individuals had a minimum value of 0, which meant the respondent was not financially responsible for anyone, while the maximum value was 5, which meant the respondent was financially responsible for five other individuals.

The three continuous independent variables of financial literacy, age, and number of dependents had a skewness of −0.21, −0.50, and 0.98, respectively, and a moderate kurtosis of −1.14, −0.50, and 0.12, respectively, which meant that the distributions of the variables were close to normal.

4.2. Logistic Regression Results

4.2.1. Pre-Estimation Tests

Before empirically testing the model, we tested the variables and model reliability checks.

Table 6. Variable reliability test.

	Reliability Statistics
Variables	Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	No. of Items
Financial behavior	0.72	0.71	8
Financial self-control Attitude toward spending	0.61	0.61	3
Financial self-control	0.25	0.29	3

and

Table 7. Model and goodness-of-fit tests.

Tests	Reliability Statistics
	Chi-Square	Degree of Freedom	Significance p-Value
Omnibus likelihood tests of model coefficients	32.73	9	0.00
Hosmer and Lemeshow goodness-of-fit test	5.08	8	0.74

show the test results. The reliability of the financial behavior measure was tested using the Cronbach’s alpha reliability test. The standardized value was 0.719, indicating that the measure was reliable.

In addition, Cronbach’s alpha was used to test the reliability of the financial self-control measure at two stages. The first stage constructing the attitude toward spending value was created by three answers to three Likert scale questions. As shown, the Cronbach’s alpha reliability test standardized value was 0.619, meaning that the measure was moderately reliable.

The second stage constructed the final financial literacy measure by calculating the median of three values: holiday budget, attitude toward spending, and strategies of self-control. Table 6 shows that the Cronbach’s alpha reliability test standardized value was 0.293, implying that the measure was not reliable; this could be because the values of the tested data were not Likert scaled. So, this result could be ignored moving forward.

The financial literacy variable was a score resulting from a set of questions, where the answers reflected the knowledge of the individuals and were not Likert scaled; thus, the reliability test did not apply.

Furthermore, as presented in Table 7, the likelihood ratio of the Chi-square test showed that the model was significant and fit. The Chi-square of the Omnibus tests was 32.730, which was significant at 1%. The Hosmer and Lemeshow test also showed that the Chi-square was 5.087 and insignificant, indicating that the model had a good fit for the data.

4.2.2. Results and Discussion

Table 8. Financial behavior model logistic regression estimates.

Independent Variables	Coefficients	Standard Errors	Odds Ratios
Financial literacy	0.30 *	0.08	1.35
Self-control (1)	0.51 **	0.28	1.67
Gender (1)	0.15	0.29	1.16
Marital status (1)	0.32	0.37	1.37
No. of dependents	0.10	0.13	1.11
Education level (1)	−0.98 **	0.57	0.37
Income level (1)	0.83 **	0.38	2.30
Age	−0.01	0.01	0.99
Generation	−0.17	0.27	0.83
Constant	0.64	0.77	1.91

Note: For Self-control, “good self-control” indicated by (1); for Gender, “female” indicated by (1); for Marital status, “married” indicated by (1); for Education level, “high education level” indicated by (1); for Income level, “high income level” indicated by (1) in the estimate tables. Significance: * = 1%; ** = 5%.

shows the logistic regression estimation results for all generations (X, Y, and Z). The results show that the coefficient of financial literacy was significant at a 1% significance level, which supported the findings of prior studies (e.g., Asaad 5]; Andarsari and Ningtyas [34]; Rai, Dua and Yadav [35]; and Aristei and Gallo [52]). This positive coefficient (0.306) of financial literacy implied that the financial behavior of Lebanese people was dependent on their financial knowledge and there was an increasing likelihood of good behavior if the individual had high financial literacy. In other words, having good financial behavior was dependent on whether the individual was knowledgeable of and educated on inflation, risk, diversification, interest rate calculation, and liquidity concepts. The higher than one odds ratio (1.359) also indicated that the likelihood of the individuals having good financial behavior will increase if an individual has high financial literacy.

So, consistent with prior studies [5,34,35,52], financial literacy has a significant impact on the financial behavior of individuals.

In addition, the coefficient of financial self-control was significant at a 5% significance level, supporting [30]. The positive coefficient (0.514) of self-control showed that self-control was positively related to the financial behavior of individuals. In other words, the financial behavior of Lebanese was dependent on their self-control. Good financial behavior was dependent on whether the individual was an impulsive spender, had a good or poor attitude toward calculated spending and savings, and whether he believed he should exercise self-control strategies or not control his behavior. The higher than one odds ratio (1.671) also indicated that the probability of the individual having good financial behavior will increase if he had high financial self-control. In conclusion, good financial self-control has a positive impact on good financial behavior, a result similar to prior studies by Rey-Ares, Fernández-López, Castro-González and Rodeiro-Pazos [4], and Lissitsa and Kol [30].

Focusing on the demographic variables, Table 8 shows the financial behavior was insignificantly impacted by an individual’s gender. In other words, being female or male did not affect financial behavior in Lebanon. Similarly, marital status and the number of dependents had no significant impact on financial behavior. Both results are consistent with the prior studies [4,15,35], which found that the gender, marital status, and the number of dependents factors were irrelevant to the financial behavior of individuals in the US and Spain. In contrast, the study by Dewi, Febrian, Effendi, and Anwar [16] found that women in general and married women, in particular, had better financial behavior than men, whether married or single.

Likewise, the coefficient of the education level was significant at a 5% significance level, implying that the financial behavior of Lebanese people was dependent on their education level. In other words, the negative coefficient (−0.985) of the education level control variable showed that the education level was negatively related to the financial behavior of individuals; this could be because lower-educated people may be more prudent with how they spend, especially during a crisis, and may be making better financial provisions than highly educated people since their expected income is lower than highly educated individuals. This result contradicts the study by Gray, Montagnoli, and Moro [8], which found that a higher education level led to better financial behavior. Additionally, the results show that the coefficient of income level was significant at a 5% significance level, implying that the financial behavior of Lebanese people was dependent on their income level. The positive coefficient (0.835) of the income level control variable showed that the income level was positively related to the financial behavior of individuals and was consistent with the work by West and Friedline [14]. The higher than one odds ratio (2.305) also indicated that the likelihood of individuals having good financial behavior will increase if they have a high income.

Lastly, the results show that the age and generation of individuals had an insignificant effect on financial behavior. These results contradict the results of studies (e.g., Kim and Lee [53]), which found that age had a significant effect on financial decisions. Overall, the results confirm the hypotheses and indicate that financial behavior significantly depends on financial knowledge and self-control. Notably, the results confirm the effect of education and income levels on an individual’s financial behavior in Lebanon.

4.2.3. Further Analysis: Do the Demographics Matter?

In this section, we specifically examined how demographics impact an individual’s financial performance and behavior using the logistic regression approach.

Table A1. Financial behavior determinants under the various individual characteristics.

Independent Variables	Generation									Gender
	Generation X			Generation Z			Generation Y			Male						Female
	CO	SE	OR	CO	SE	OR	CO	SE	OR	CO		SE		OR		CO	SE	OR
Financial literacy	0.33 *	0.13	1.39	0.39 **	0.17	1.48	0.12	0.21	1.12	0.25 ***		0.13		1.29		0.36 *	0.12	1.44
Self-control (1)	0.59	0.43	1.8	0.33	0.49	1.39	0.75	0.76	2.11	0.65		0.41		1.92		0.48	0.39	1.62
Gender (1)	−0.14	0.44	0.86	0.68	0.54	1.97	0.18	0.77	1.2	---		---		---		---	---	---
Marital status (1)	0.37	0.53	1.46	−2.33 ***	1.29	0.09	1.03	0.86	2.8	−0.26		0.60		0.76		0.76	0.50	2.14
No. of dependents	0.04	0.19	1.04	0.39	0.36	1.48	−0.07	0.39	0.93	0.12		0.21		1.13		0.08	0.18	1.08
Education level (1)	2.09 ***	1.1	1.12	−0.71	0.81	0.49	−19.58	40,193	0	−0.45		0.72		0.63		−2.32 **	1.12	0.09
Income level (1)	0.51	0.48	1.67	0.1	0.98	1.11	1.90 ***	1.13	6.68	0.60		0.50		1.82		1.25 *	0.66	3.50
Age	−0.03	0.02	0.96	0.01	0.11	1.01	−0.10	0.24	0.90	0.03		0.02		1.03		−0.03 *	0.01	0.96
Generation	---	---	---	---	---	---	---	---	---	−0.15		0.40		0.85		−0.08	0.37	0.91
Constant	3.15 ***	1.59	1.47	−0.38	1.23	0.68	1.48	1.35	1.56	0.03		0.02		0.43		2.76 **	1.31	1.84
H and L test	9.61			16.2			6.89			13.23						8.76
Independent Variables	Marital Status						Education Level						Income Level
	Single			Married			High Education			Low Education			Low Income			High Income
	CO	SE	OR	CO	SE	OR	CO	SE	OR	CO	SE	OR	CO	SE	OR	CO	SE	OR
Financial literacy	0.33 *	0.11	1.39	0.27 ***	0.14	1.32	0.27 *	0.09	1.31	0.89 ***	0.07	1.32	0.41 *	0.10	1.51	−0.31 **	0.29	0.73
Self-control (1)	0.51	0.35	1.66	0.61	0.47	1.85	0.65 **	0.28	1.76	0.86	1.25	1.82	0.37	0.31	1.45	1.81	0.81	6.16
Gender (1)	0.04	0.37	1.05	0.35	0.51	1.42	0.03	0.30	1.03	0.74	1.32	1.41	0.13	0.33	1.14	0.73	0.79	2.08
Marital status (1)	---	---	---	---	---	---	0.46	0.38	1.59	0.73	1.08	1.23	0.06	0.41	1.06	1.44	0.97	4.25
No. of dependents	0.21	0.21	1.23	0.04	0.20	1.04	0.08	0.14	1.09	0.64	1.12	0.92	0.19	0.15	1.22	−0.24	0.42	0.78
Education level (1)	−0.99	0.82	0.37	−1.19	0.84	0.30	---	---	---	---	---	---	1.03 ***	0.58	0.35	−1.11	1.44	0.42
Income level (1)	0.36	0.53	1.43	1.19 **	0.57	3.28	0.82 **	0.38	2.27	1.18	1.23	0.36	---	---	---	---	---	---
Age	0.01	0.01	1.01	−0.02	0.02	0.97	−0.01	0.14	0.98	0.29	0.55	1.01	−0.01	0.14	0.99	−0.09	0.05	0.99
Generation	−0.09	0.36	0.91	−0.50	0.45	0.61	−0.15	0.27	0.86	−0.67	0.80	0.81	−0.28	0.29	0.75	0.37	0.83	1.46
Constant	0.31	1.01	1.37	2.05	1.49	7.80	0.02	0.58	0.97	−1.02	1.22	0.92	0.55	0.80	1.73	1.26	1.11	0.35
H and L test	12.45			4.29			9.45			9.12			13.52			4.09

Note: For Self-control, “good self-control” indicated by (1); for Gender, “female” indicated by (1); for Marital status, “married” indicated by (1); for Education level, “high education level” indicated by (1); for Income level, “high income level” indicated by (1) in the estimate tables. H and L is Hosmer and Lemeshow test. CO is coefficients; SE is standard errors; OR is the odds ratio. Significance: * = 1%; ** = 5%; and *** = 10%.

in the Appendix A shows the results.

The empirical results reveal that financial literacy significantly impacted the financial behavior of Generations X and Z, but the financial behavior of generation Y was only affected by income level. Furthermore, the results imply that financial literacy significantly impacted the financial behavior of both males and females, but female financial behavior was also affected by education levels, income levels, and age.

Likewise,

Table 9. Robustness test.

Independent Variables	General Linear Model			Multinominal Logistic Regression
	Coefficients	Standard Errors	Odds Ratios	Coefficients	Standard Errors	Odds Ratios
Financial literacy	−0.31 *	0.08	0.73	−0.31 *	0.08	0.74
Self-control (1)	−0.51 ***	0.28	0.59	−0.51 **	0.28	0.59
Gender (1)	−0.15	0.29	0.85	−0.15	0.29	0.86
Marital status (1)	−0.32	0.37	0.72	−0.32	0.37	0.73
No. of dependents	−0.11	0.13	0.89	−0.11	0.13	0.90
Education level (1)	0.98 ***	0.57	2.67	0.98 ***	0.57	2.70
Income level (1)	−0.83 **	0.38	0.43	−0.83 **	0.38	0.43
Age	0.01	0.01	1.01	0.01	0.01	1.01
Generation	0.17	0.27	1.19	0.17	0.27	1.20
Constant	−0.64	0.76	0.52	−0.64	0.77	0.00
Omnibus test	32.73 *
Likelihood ratio test	---			32.73 *
Pearson’s test	---			319.49
Deviation test	---			321.71

Note: For Self-control, “good self-control” indicated by (1); for Gender, “female” indicated by (1); for Marital status, “married” indicated by (1); for Education level, “high education level” indicated by (1); for Income level, “high income level” indicated by (1) in the estimate tables. Significance: * = 1%; ** = 5%; and *** = 10%.

highlights that financial literacy significantly affected the financial behavior of both single and married individuals, but the financial behavior of married individuals was also affected by income levels. As for the education levels, the financial behavior of both high- and low-educated individuals was affected by financial literacy. However, self-control and income levels also affected the financial behavior of the highly educated individuals.

5. Robustness Check

The present study implemented the robustness analyses by performing various methodologies including the general linear model and multinomial logistic regression to control the reliability of the estimation results. Table 9 shows the results. It should be noted that the reference value for the predictors was “0” in the generalized linear model, which meant that the “negative” coefficient indicated a negative relation between the dependent variable “good financial behavior” and the predictor reference as 0 (poor financial literacy, low self-control, and so on). As shown, the result is similar to that generated using binary logistic regression. The generalized linear model results indicate that Lebanese individuals with low income, poor financial knowledge, and low self-control were less likely to have good financial behavior. In addition, generation, age, gender, marital status, and the number of dependents did not have an impact on the likelihood of Lebanese having good financial behavior. Consistently, the multinomial logistic regression showed that the predictors of low financial literacy, low self-control, and low-income level had a significant effect on the financial behavior of Lebanese. Applying the goodness-of-fit tests by Omnibus, Likelihood ratio, Pearson, and Deviation verified the validity and significance of the models.

6. Conclusions

For the past three years, Lebanon has been suffering the most severe economic, financial, political, and social crises. The Lebanese people have never been more uncertain about their economic and financial future than they are today. The future of Lebanon is dependent on its young generation’s performance and behavior and decision making. In particular, and within our field of interest, our study focused on financial behavior and decision making. To the best of our knowledge, there is a significant gap in the literature aiming to explore the financial behavior determinants, particularly by considering the demographic factors during the Lebanese crisis period. Therefore, this study fills this gap and investigated this relationship to open a new debate in the literature. To do so, this study conducted a comprehensive survey of 328 Lebanese individuals by asking a collection of Likert scale and multiple choice qualitative questions covering basic financial knowledge, self-control, and behavior.

The results show that financial performance and behavior were significantly improved by basic financial knowledge and financial self-control among all generations. In other words, the results show that having good financial behavior was dependent on good financial literacy and high financial self-control. Focusing on the effect of demographic factors, the results provide evidence that good basic financial knowledge improved the likelihood of financial behavior for Generations X and Z, for males and females, for single and married people, for low- and high-educated individuals, and for low- and high-income individuals. Financial self-control improved the likelihood of financial performance and behavior for highly educated individuals, while it was not that important when comparing other demographics.

Our results have important policy implications. Good financial behavior promotes the well-being and sustainability of the economy. As our research has enough evidence on the importance of good financial performance and behavior and its relation to financial literacy, it should urge policymakers to pay more attention to the financial behavior of new generations. The policies should aim to enhance good financial behavior by integrating financial knowledge into school curricula and helping children and young adults acquire skills in self-control. Policies should also work on improving literacy in general, because education leads to a better understanding of things and thus better behavior overall. In addition, for better financial behavior, companies and organizations should emphasize improving financial literacy as part of their human resources development plan, because it has been proven that good financial literacy improves individual financial behavior and performance and their well-being. Finally, government and the central bank could also motivate local financial institutions to offer a variety of financial products and investment opportunities targeting low-income and low-educated individuals, by proving subsidized funds. Those financial opportunities should be packaged with a financial literacy program for the beneficiaries. At the same time, these programs will enhance the sustainability of the income and the future financial performance of the beneficiaries.

For further studies, it is worth investigating how COVID-19, governance, and economic and political instabilities impact an individual’s financial behavior (e.g., [40,48]). In addition, further studies should be focused on examining the financial literacy–financial behavior relationship for other countries to provide a more comprehensive representation of good financial behavior determinants, which prepares people for setting a sustainable framework in the future.