Life Satisfaction and Tax Morale in Azerbaijan: Mediating Role of Institutional Trust and Financial Satisfaction

: This paper examines the relationship between life satisfaction (measured as the self-reported satisfaction of each individual with their past life and goal achievements) and tax morale (measured as the likelihood of an individual’s intrinsic motivation to pay taxes). Using a large-scale survey dataset from Azerbaijan, it is documented that life satisfaction is positively associated with tax morale. Life satisfaction plays a signiﬁcant role in increasing tax compliance practices. It is also important to note that there is a positive mediating effect of life satisfaction on tax morale through ﬁnancial satisfaction and institutional trust. In line with our hypotheses, the results of a series of analyses remain robust to different models. These results imply that a higher level of life satisfaction may increase the proportion of individuals who report the highest tax morale in Azerbaijan. Our ﬁndings have policy implications for Azerbaijan and other governments aiming to alleviate high levels of tax evasion.


Introduction
The concept of tax evasion has received considerable attention since the 1970s, with seminal contributions by Ref [1][2][3]. According to Ref [1], tax compliance increases or decreases with the probability of an audit and the size of potential fines that may be levied on the evader. In this rational choice approach, the assumption that taxpayers decide whether to pay taxes depends upon the probability of detection and the size of the penalty.
These early theories of tax evasion suggest that individuals are utility maximizers who consider the chances of tax cheating as a risky decision. However, it should be noted that the seminal works were incapable of explaining all deterministic factors of tax evasion by using these rational choice approaches. Therefore, most recent studies on tax evasion have been developing new behavioral models to explain this complex issue (e.g., [4], for review). Additionally, recent studies on tax evasion reveal that tax evasion can cause a reduction in tax revenue, which may lead to reduced social capital [5,6]. To solve this puzzle of tax compliance, in the tax compliance literature, rapidly growing empirical insights have been reported for many developed, developing, and transition countries [7][8][9][10][11].
In the tax compliance literature, it has been shown that tax evasion is prevalent in transition countries [12]. This shows that many countries in the transition process are not successful in planning proper tax systems, structuring tax administration, or designing government structures that individuals can trust. As Ref. [12] (p. 357) notes, "It might be useful not to focus on tax evasion only but to go a step back and analyse tax morale, the intrinsic motivation to pay taxes." According to the tax compliance literature, there is a strong negative association between tax evasion and the degree of tax morale [12]. Contrary to tax evasion, tax morale measures the moral obligation based on the taxpayer's attitude towards tax payment. It is novel to analyze tax morale as the dependent variable and investigate the factors that shape the tax morale of Azerbaijani taxpayers. This paper aims to shed light on the effect of life satisfaction on tax morale by investigating the mediating role of institutional trust and financial satisfaction in Azerbaijan. Moreover, life satisfaction is becoming an increasingly widespread phenomenon in transition countries, and it can positively affect the degree of an individual's tax morale. Therefore, examining the effect of individuals' life satisfaction on tax morale, and also the economic consequences of tax morale, in Azerbaijan is critical for a government aiming to tackle tax evasion practices.
The present study responds to the recent call to conduct ". . . an empirical investigation of the effect [of a] country's level of social and environmental sustainability on tax evasion since tax compliance is a type of social contract between government and its citizens [13] (p. 1)." Based on social and environmental sustainability, different forms of sustainability levels such as the protection of the environment, the reduction of poverty, the improvement of quality of life, and the quality of infrastructure have been classified (e.g., [13]). Our paper responds to these calls to use survey data to evaluate the association between life satisfaction and tax morale, particularly observing the effect of quality of life on tax morale.
Using a sample of 2208 respondents, the level of tax morale is positively associated with the level of life satisfaction. When we control for the mediating effects of institutional trust and financial satisfaction, we again find a significant positive association among the main variables of interest. Our results have policy implications for governments that are attempting to increase the intrinsic willingness of their citizens to pay taxes.
The current study contributes to the tax compliance literature in two ways. First, although there are many papers that examine the country-level drivers of tax morale in developing and developed countries [14,15], evidence on South Caucasian countries, particularly Azerbaijan, is scarcer and more limited [16]. Second, the current study extends and complements previous studies on the relationship between life satisfaction and tax morale by testing the mediating effects of institutional trust and financial satisfaction. To the best of our knowledge, these mediating effects have not been investigated in the tax compliance literature to this extent.
In the following section, we briefly review the conventional rational economic theories and generate hypotheses. In Section 3, we present our research design. In Section 4, we show our empirical findings, and in the last section, we finish with some concluding remarks.

Life Satisfaction
As mentioned above, one of the essential variables for tax morale is life satisfaction. According to the empirical analysis of Ref. [17], there is a negative relationship between tax evasion and life satisfaction, based on research carried out in Central and Eastern European countries. In this sense, causality is running from tax evasion to life satisfaction. Also, this idea is supported by the notion that people's happiness level can affect their economic decisions [18]. Research conducted in Portugal states that individuals satisfied with their lifestyles tend to comply with taxes [19]. Another important research paper on this topic also verified the negative relationship between tax cheating and happiness. Higher scores in happiness notably increased the rejection of tax evasion [20]. Researchers observed a strong positive relationship between tax morality and life satisfaction in Macedonia [14]. Thus, we formulate our main hypothesis: Hypothesis 1 (H1). The level of direct influence that life satisfaction has on tax morale is positive.

The Mediating Effect of Trust on the Association between Life Satisfaction and Tax Morale
Trust is one of the main observed factors in relation to tax evasion. It is believed that when taxpayers feel that they have a good institution that makes an equitable exchange of fiscal revenues, this encourages them to comply with tax payments [13]. Observations show that in a country where the responsibility of and belief in the government is low, the rate of tax evasion is increased [20]. Ref. [21] points out that the lower the level of trust in government, the higher the level of tax evasion among countries. If taxpayers perceive the tax burden to be unfair and distrust the authorities, they will evade taxes [11,[22][23][24][25]. Ref. [19] found a significant negative association between institutional trust and tax morale in Portugal. Similarly, research conducted in Croatia observed that the main determinants of tax evasion are low tax morality and a feeling of unjustified behavior of governments regarding collected taxes [16]. Ref. [26] shows that the higher the belief in procedural fairness, the lower the rate of tax fraud, and the higher the trust in tax authorities.
Thus, we formulate the second hypothesis: Hypothesis 2 (H2). The level of mediating influence of life satisfaction on tax morale through institutional trust is positive.

The Mediating Effect of Financial Satisfaction on the Association between Life Satisfaction and Tax Morale
Regarding the financial satisfaction dimension, there is a widespread view that tax morale is affected by the financial situation of the taxpayers. According to previous studies Refs. [14,[27][28][29][30], higher satisfaction with one's financial situation is positively associated with tax morale. Refs. [10,12,22] emphasizes that financial dissatisfaction can create a feeling of trouble when tax should be paid, and there is an imbalance between real and desired financial situations. In countries with transition economies, the possibility of this imbalance is high because of low standards of living. People can feel that taxes are a barrier to reaching their goals, and thus reduced financial satisfaction decreases their tax compliance. As suggested by Ref. [31], the higher the financial satisfaction of people, the higher the share of people who state that tax evasion is wrong. In a similar vein, Ref. [29] focused on the major determinants of tax morale in Iran. In their study, the financial satisfaction of the individuals was built on a taxpayer's current income compared with others. Building on social psychology theories, their results show that people with low financial satisfaction tend to evade taxes more for reasons other than attaining a better financial position. Ref. [32] emphasizes that taxpayers who are evading taxes can use this evaded tax money for their own financial satisfaction. Ref. [27] provided similar findings, indicating that taxpayers with personal financial constraints are more inclined to practice tax evasion than taxpayers with fewer financial constraints. Ref. [7] also find that financial satisfaction and dissatisfaction affect tax morale in a certain way, meaning that while financial satisfaction leads to higher tax compliance, dissatisfaction works in the opposite direction, and the probability of one cheating on taxes increases. However, Refs. [33,34] found that taxpayers with high financial satisfaction will also tend to evade taxes. They assume that taxpayers favor their economic situation rather than personal status. Therefore, we can expect that the effect of life satisfaction on tax morale depends positively on the financial satisfaction of the individuals. Thus, we formulate the third hypothesis: Hypothesis 3 (H3). The level of mediating influence of life satisfaction on tax morale through financial satisfaction is positive.

Sample
This research employs cross-sectional data from Social Survey-2 by Ref. [35]. Data were collected from October through December 2018. The 2208 respondents, who filled out self-administrated questionnaires, include employees, students, homemakers, the unemployed, and retirees. Data collection strategies were followed to achieve optimum representativeness, face-to-face (paper-based) and online (using paid and unpaid social media facilities). The sample frame covers people over the age of 17 from all regions of the Republic, with relatively limited access to rural areas. The research aims to investigate the life satisfaction and perceived tax evasion association among all employees (public, private, and small business owners). Therefore, students, homemakers, the unemployed, and retired people are excluded to achieve the target sample size (n emp = 1317, n male = 718, n f emale = 599, Mean age = 32.44, SD age = 11.47). Among the respondents, 53% of the sample represent the capital city and primary developed region, Baku, while 11.2% of respondents are living in the Absheron region (the nearest area to Baku), and 35% are from other regions of the Republic. To estimate the direct and indirect effects of life satisfaction on tax morale, we ran parallel mediation analyses using the SPSS Process Macro developed by Ref. [36].

Dependent Variable
Tax Morale (hereafter TM) is measured by the likelihood of an employee's intrinsic motivation to pay taxes, based on the employee's reaction to a hypothetical five percentage point increase in the labor income tax rate. A related question in the survey measures how a respondent's work motivation will change after a hypothetical labor income tax rate increase according to a scale ranging from 1 = work signi f icantly less to 7 = work signi f icantly more, with an eighth option "would try to evade taxes." The eighth option is an indirect way of measuring TM for each respondent. TM is a dummy variable equal to 1 if the respondent chooses the eighth option ("would try to evade taxes") and 0 otherwise (any other change in work motivation).

Independent Variable
Life satisfaction (hereafter LS) denotes the self-reported satisfaction of each respondent with their lives and goal achievements in the past. To measure LS, we use the Satisfaction with Life Scale advanced by Refs. [37,38], which shows the outcome through 5 items

Mediator Variables
Institutional trust (hereafter IT) represents an individual's trust in public institutions (education, health, legal, and police). The survey follows an indirect 4-leg methodology. Questions measure people's trust in the representatives of each institute-teachers, doctors, judges and court staff, and law enforcement officers-according to a five-point Likert scale (from 1 = totally do not trust to 5 = totally trust). An individual's IT score equals the sum of responses to each item, ranging from 4 to 20. This approach helps to overcome the political fear of the individuals and enhance the reliability of responses. The reliability of the scale is confirmed (α = 0.836).
Financial Satisfaction (hereafter FS) indicates self-assessed sufficiency of people's income. A one-item scale is used to measure FS ("how satisfied are you with your income?") according to a seven-point Likert scale, ranging from 1 = strongly dissatis f ied to 7 = strongly satis f ied. More income satisfaction means a higher FS.

Control Variables
Age denotes the age of the respondent, ranging from 17 to 65. Gender is a dummy variable that equals 1 if the respondent is female, and 0 otherwise. Job experience (hereafter JobExp) denotes the total professional experience of an employee measured in years within the 0.1-51 range. School i , College i , Master i , and PhD i are educational dummies, while Bachelor i degree holders are left as the base group. Lastly, marital status dummies (Married i and Widowed i ) are added to the model specification. The base group is Single i .

Model Specification
Three models are estimated to perform parallel mediation analysis with two mediators: Here, Z k denotes a set of covariates to control for heterogeneity bias in the models. To get more reliable results, we include the covariates in the estimation process at different stages. The first stage does not include any covariates, while the second stage's model simultaneously covers age and gender variables. Meanwhile, job experience and dummy variables representing the educational attainment level of the respondent are added at the third stage. Finally, the models of stage 4 also include marital status dummies. The dependent variable in model 3 is a binary variable. Therefore, we apply the Binary Logistic Regression estimation method in this case.
Regarding the impact of LS on TM: δ 1 shows the level of the direct impact; α 1 * δ 2 equals the indirect impact through institutional trust; β 1 * δ 3 denote the indirect impact through financial satisfaction; We expect that α 1 > 0, β 1 > 0, δ 1 < 0, δ 2 < 0, and δ 3 < 0, and that all are statistically significant at 95% level of confidence. Table 1 presents essential descriptive statistics of all variables. In the sample, 45.5% are females and 54.5% are males. In terms of educational attainment, the sample composition is as follows: high school graduates-16.8%, college graduates-13.5%, bachelor's-45.7%, master's-18.5%, PhD degree holders-5.3%. Among the respondents, 42.1% are single, while 50.3% are married, and 4.2% are widowed/divorced. Tax morale was observed among 17.3% of respondents (n male = 149, n f emale = 68, Mean age = 30.97) who are prone to evade taxes in response to the hypothetical tax change.

Regression Analyses
Covariates are added in stages, from none in model 1 to all in model 4. Table 2 presents regression outputs for the final specification, model 4, which includes all control variables. However, we will discuss the level of direct and indirect impacts from LS to TM at all stages according to Table 3 below.
As expected, life satisfaction has a direct positive impact on tax morale, i.e., greater satisfaction with life diminishes an individual's likelihood of trying to avoid taxes. According to the estimation outputs, a positive causality also appears from institutional trust and financial satisfaction to tax morale. Therefore, the empirical findings present strong supportive evidence for the research hypotheses.
Regarding the primary determinants of tax morale among the covariates, age has a positive impact, while females are more compliant with tax laws, and they are less likely to evade taxes than males. Individuals with more job experience and who have a PhD are more likely to avoid taxes.     Although Table 2 is informative enough to support the research hypotheses, calculating direct and indirect effects requires an additional step-bootstrapping analysis. In this context, Table 3 is more informative about the level of direct and indirect impacts (through institutional trust and financial satisfaction) of life satisfaction on tax morale.
Following a procedure that adds covariates in stages does not reveal any significant variability of the research findings. Direct and indirect impacts are significant at a 95% confidence level in all models. A small exception is the insignificance of indirect impact via financial satisfaction in model 2, which recovers after the inclusion of more covariates.

Discussion and Conclusions
The effect of tax evasion on life satisfaction on a within-country and between-country basis has been increasingly attracting interest in the tax compliance literature (e.g., Refs. [17,39]). A neglected aspect in the related literature is the reverse causality of the association Ref. [40]. The earlier studies mainly tested how tax evasion (or tax morale) affects individual life satisfaction [14,15].
Building on previous studies, we hypothesized that life satisfaction is positively associated with tax morale. Additionally, a positive mediating effect of life satisfaction on tax morale through financial satisfaction and institutional trust was hypothesized.
Using large-scale survey data on tax morale, we obtain strong support for the association between life satisfaction and tax morale. While the former is characterized by the likelihood of an individual's intrinsic motivation to pay taxes, the latter corresponds to the self-reported satisfaction of each individual with their past life and goal achievements. The study contributes to the literature on understanding tax morale in Azerbaijan. Additionally, the mediating effects of financial satisfaction and institutional trust were examined in predicting the relationship between life satisfaction and tax morale. After controlling for several socio-economic and demographic variables, the results remain robust to different models.
Drawing on the work of Ref. [13], we assume individuals with less tax morale do not have incentives for social and environmental sustainability. There is a social contract between taxpayers and the government in which taxpayers demand high transparency and accountability from the government when using fiscal resources. If they feel that their government does not allocate these resources fairly, they might refuse to obey the authority [41]. Based on these assumptions, we expect that the level of sustainability is increased when there is a strong commitment to paying taxes through high life satisfaction as well as institutional trust and financial satisfaction. Suppose the individuals perceive that they are financially above the minimum subsistence level and have higher trust in the tax authority. In that case, they are more satisfied with their life in general and truthfully declare their income to the tax authority.
The present study also contributes to a growing research agenda on the importance of formal and informal institutions for tax morale [42]. While formal institutions correspond to laws and regulations, informal institutions are unwritten beliefs, values, and norms. We included the trust factor as a formal institution, and for informal institutions, we considered life and financial satisfaction. Our main results (Table 2) and additional bootstrapping analysis (Table 3) demonstrated significant positive direct and indirect influences of life satisfaction on tax morale, considering institutional trust and financial satisfaction. This means that increases in life satisfaction influence the intrinsic motivation to pay taxes. The evidence is straightforward and in the same line as previous studies [10,12,14,22]. Additionally, we find that tax morale is harmed when institutional trust [11,16,[21][22][23][24][25][26] and financial satisfaction [7,10,12,14,22,[27][28][29][30][31][32] are low. Finally, we find a significant relationship between life satisfaction and tax morale when we use our controls such as age, gender, job experience, and educational factors.
Accordingly, the findings can have policy implications for governments that plan to decrease the level of tax evasion. For instance, the government of Azerbaijan needs to give more consideration to the happiness of their residents when they are imposing taxes. Our evidence should allow government policymakers to get a better understanding of the relationship between individual life satisfaction and tax morale and to implement adequate social and economic strategies to control the level of tax evasion to diminish its harmful impacts on society. Additionally, the interactions with formal and informal institutions can shape the tax morale among Azerbaijani taxpayers. First, high trust in public authorities (formal institutions) might increase the level of tax morality among taxpayers. This means that high trust in public authorities implies more transparency and accountability undertaken by the government, which will result in high tax morale. Second, high life and financial satisfaction (informal institutions) might also increase the level of tax morality among taxpayers. This means high life and financial satisfaction give rise to the minimum standard of living among taxpayers, which again will result in high tax morale. Thus, scrutinizing the interaction effects between formal and informal institutions in understanding tax morale might lead to new significant behavioral insights.
The current study has several limitations. First, the sample is from one country, which may decrease the reliability of our study. Nevertheless, this is a common problem whenever empirical analyses are based on a single country. Second, it is possible that the dependent variables may not reflect the true value of tax morale. However, the use of indirect methods to measure this variable should lessen this concern.
Future research may explore the effects of both financial satisfaction and institutional trust over time, as it is possible that both can change over time. Second, extending the sample with cross-country analysis may also improve our understanding of the current topic. Finally, additional mediating factors could be added to the current methodology to better explain the aforementioned association between the main variables.