Personal Income Tax and Citizens’ Trust in Politics: A longitudinal study
using Microdata in the Netherlands.
Leiden University
Faculty of Governance and Global Affairs
Master of Science in Public Administration: Economics & Governance Master Thesis
Name: Zongqi Xu
Student Number: s2435489 Date: 18-08-2020
Supervisor: Assistant Professor. Dr. Max van Lent Second Reader: Professor. Dr. Pierre Koning
Foreward:
I would like to express my sincere gratitude to my supervisor Assistant Professor Dr. Max van Lent for his guidance to my thesis. I am sincerely appreciate his advice on the topic, support and guidance on my work throughout the period. His supervision is valuable to this thesis, and I am very grateful for his continuous and knowledgable guidance, especially in this serious time of global pandemic.
Abstract
The factors that affect citizens' trust in politics have been widely studied. In this research, I contribute to the literature by studying the effect of changes in the personal income tax on trust in government and institutions. Citizens' paying taxes to the country is not only a mandatory obligation but also an expectation on more public goods and better service from the government to improve their well-being. Citizens' perception of government service is a critical criterion for evaluating their political trust. Therefore, there is a causal effect between tax policy and citizens' trust in government. The chapter first provides an overview of the main theoretical propositions in the literature and also the counterarguments. The next chapter illustrates the design of this research, the use of LISS panel survey data in the Netherlands as the data source1; and regression discontinuity is the foremost approach
to study this topic. There are two main study questions of this thesis. First, do similar people (in terms of income, €500 close to the threshold) who pay a higher tax rate, decrease their trust in politics. Second, does a decrease in the income tax rate over the years decrease trust in politics? The following chapters focus on the results and discussions of the findings. The empirical results indicate that paying a higher marginal tax rate does not diminish citizens’ trust in government and institutions. Second, a higher tax rate on salaries does not reduce individuals’ trust in government and institutions.
1 In this paper I make use of data of the LISS (Longitudinal Internet Studies for the Social Sciences) panel
Table of Contents
Chapter 1: Introduction... 5
Chapter 2: Literature Review...7
2.1 The theoretical framework of Trust... 7
2.2 The concept of personal income taxation... 12
2.3 Interaction between trust and income tax (a causal mechanism)... 17
Chapter 3: Assumptions and Hypothesis:... 20
3.1 Key assumptions:... 20
3.2 Hypotheses... 21
Chapter 4: Data and Research Methods...22
4.1 Data (Personal Income and Political Trust)...22
4.2 Research Approach: Regression Discontinuity Design...25
Chapter 5: Empirical Results ... 29
Chapter 6: Discussion of the results and Implications...39
Chapter 7: Conclusion...42
Chapter 8: Limitations of the Research... 44
Chapter 9: Suggestions on Future Research... 45
Chapter 10: Appendix... 45
Chapter 1: Introduction
There is an ongoing debate as to what determines citizens' trust in the government. Political trust is a crucial indicator of the political legitimacy, and trust is essential for the success of a range of public policies, it is a measure that reflects the overall confidence in political capability and governance. The concept of trust often occurs at the interpersonal level to express people's confidence in each other or the quality of service they received. The concept becomes much more complicated on the political level, and the relationship between political performance and citizens' trust may not as straightforward as it shows. Maximizing citizens' utilities is a vital approach for the government to increase citizens' political support. Citizens may have diverse expectations from government. Political institutions hold the role of intermediating between government and citizens with specific expertise and responsibilities such as Education and Health Care.
Citizens evaluation of government is from an emotional response, a rational perspective of benefit-cost analysis, and behavioural evaluations. The difference in debates remains as some research suggest that people have specific attitudes and opinions about the various units of the government other than the government as a whole. Whereas other studies rebut the argument by arguing that all citizens cannot know about the civil servants work in the institutions and the functions of the institutions. The ability of government and the institutions are always crucial to the citizens, the satisfaction with the governments' work generally exhibits trust in government, whereas dissatisfaction with their work would lead to distrust or low trust in the government. I suggest that the performance of government and institutions are critical criteria the citizens use to evaluate the government. Tax policy is a crucial part of the governments' fiscal policy that determine the government's revenue, and it is a sustainable source of government revenue to sustain the social program and public investment to foster the social development and economic growth. The objective of the government's tax policy is to allocate the resources from one group to another to develop its plans that improve the well-being of the citizens. The increase in income taxes advances the capability of the government to improve the welfare of the citizens.
This thesis aims to study the effect of changes in income tax policy on citizens' trust in politics. Many studies investigate the factors that affect the citizens' political trust and the impact of tax policy on citizens' behaviours. However, there is little research to examine the correlation between these two subjects, and rarely have research analyze trust from tax perspective. The literature of tax policy suggests that the increase in tax policy leads to a rising capability of the government to improve the well-being of citizens. Nevertheless, citizens perceive the increase in taxes as a decline in their well-being, which decreases their trust in government. People pay taxes not only because it is necessary accountability to the country, but citizens also want more public goods and better services from the government to improve their quality of life.
The first study question is whether people paying slightly higher taxes compared with similar individuals decrease their trust in politics. It is because people pay taxes for better services and more public goods, but people around the tax thresholds often receive similar services. People on the right side of the threshold pay additional taxes on exceeding amount. They may want more on public goods for the extra taxes contributed to the government. Nevertheless, they receive the same goods compared with similar individuals pay fewer taxes. In this situation, do they drop their trust in government? The income tax rate is not fixed every year; it always increases or decreases over the years. The second study question is that when the income tax rate decreases over the years, do people increase their trust in politics? The first study question concerns when people unluckily in a higher tax bracket, while the second study question investigates changes in what other is paying. When income tax rates increases, ceteris paribus, the people pay more taxes to the government. Do they in this situation show decrease in their political trust. This paper can supplement the exists research on factors that affect the citizens' trust in government and the influence of tax policy on people's behaviours.
The primary research approach used in this thesis is Regression Discontinuity Design. It is a method commonly used in Economic analysis. Regression discontinuity is a method that estimates treatment effects and establishing causality in a non-experimental setting. Observations in a dataset are separated into a control group which below the discontinuity and a treatment group when the assigned variables are above the cutoff point. For the first research question, the cutoff points lie on the threshold of the taxes on citizens' gross pay. The treatment group is people who above
the threshold for about €500 pay higher taxes than the control group, which is people who below the threshold for about €500. The methods aim to study the causal effect of tax on citizens' trust in government. For the second research question, this thesis employs the ordinary least squares regression analysis. When holds inflation rate and GDP growth across years as control variables, the thesis examine the correlation between changes in income tax rates and citizens' political trust. The datasets that this thesis employs come from the LISS panel dataset. The LISS panel dataset records citizens' trust in politics from two directions, which are 'confidence' and 'satisfaction'.
However, the panel did not explain clearly the difference between these two criteria. I believe the ideas of confidence and satisfaction complement each other. Citizens satisfaction yields their confidence with the government. If citizens are not satisfied with the government, they cannot build confidence in the government. The opposite is also exact. The main findings rebut the hypotheses. For the first hypothesis that whether people paying slightly higher taxes compared to similar citizens decrease their political trust. Most empirical results from regression discontinuity design cannot support the hypothesis. For the second hypothesis, when holding inflation and GDP growth controlled, an increasing trend of personal income tax rate did not show decreasing trust in politics. However, both results reflect a positive and significant coefficient which yields an increase in citizens' trust in government.
Chapter 2: Literature Review
2.1 The theoretical framework of Trust.
Changes in people’s view on trust in government have raised the concerns of political scientists since the 1960s (Chanley, Rudolph & Rahn, 2000). Their political trust enhances the legitimacy of the government. Politicians often care about this subjective feedback from citizens, mainly to keep their offices during the elections. There are many initiatives government proposed to reconstruct the public trust in government. These plans are the strategies that the government used to improve the quality of their service, which aims to stimulate public support to the government.
There are two main degrees of looking at political trust in literature: trust as estimation or evaluation. Some look at the trust is an idea that citizens are confident with the government outputs even though they are not involved in the process. (see
Sztompka, 1996; Gamson, 1968; and Coleman, 1990). It is a belief that the government would act on citizens' interest and maximize their utilities, even the future is uncertain. Other scholars' opinion yields a different dimension. They analyze trust as an evaluation of the performance of the government. (see Easton, 1965; Bouckaert & van de Walle, 2001; Hibbling & Theiss-Morse, 1995; and Zhang & Kim, 2008). If citizens satisfied with the governments' outcome, they return their trust to the government. Whether studies believe trust is estimation or evaluation on government is not a different concept, they are different degrees but not contradict to one another. The standard criteria behind the two dimensions are that citizens estimate or evaluate whether the government increases its utilities or satisfy their interests. Citizens place their trust based on the service of the government and whether the service offered to improve their well-being. If citizens believe they receive better welfare service and they are satisfied with it, they review the government positively and more confident with the government. Citizens' distrust in government is because they are not satisfied with the service they receive from the government or they perceive a decline in their well-being.
2.1.1 Some studies look at trust as estimation on future
Some believe trust is a prediction on the future. It is an idea that other people will look after trusters' interest and will not harm trusters and their interest. This belief yields the uncertainty which results from the potential actions of the trustees, whether they will behave in the way benefit the people or hurt the interest of people. Trust is used to eliminate the uncertainty, and to guarantee the outcomes in uncertainty is beneficial to trusters. When people place trust to others, people become more confident in the trustees' ability to behave in people's interest, and this confidence eliminates their doubts on the possible results in uncertainty. Sztompka echoed the opinion and argued that trust is bet on the future contingent behaviours of the others. His definition includes seven factors which are efficiency, reliability, fairness, regularity, benevolence and responsibility, and these factors are often as the criteria people utilize to evaluate their political trust (Sztompka, 1996). Gamson believes that trust is the possibility the government would produce the preferred results even though left untended (Gamson, 1968). Coleman links trust as a subcategory of risk: it is an expectation of benefit or loss that indicates whether it could be trusted or not (Coleman, 1990). Therefore, citizens' expectation on government is a critical condition of trust. If the outputs government-produced just as citizens expected, citizens' return to the government would be high trust. On the contrary, if the
outcomes are not as citizens expected, citizens become less confident or even distrust the government.
2.1.2 Other studies look at trust as an evaluation of performance in the past The previous section demonstrates that trust to some scholars is an estimation that people made about the future. However, many researchers believe trust is an evaluation based on the actions, performance or the reputations of government in the past. Easton developed an idea of specific and diffuse support to explain citizens' intention behind their political trust. Specific support is the satisfaction with the members of the system with the perceived output and performance of the political authorities (Easton, 1965). Diffuse support is a reservoir that collects positive attitudes or goodwill (Easton, 1965). Both supports asserts the importance of the performance of government in the prior periods. He believed that specific and diffuse support are different phenomena, and they have different independent indicators and consequences for the function of a system, and these two concepts often get mixed (Easton, 1965). The negative performance and low output of government often influence the diffuse support in the long run, but if there are scandals or corruption, both specific and diffuse support decreases (Bouckaert & van de Walle, 2001). Diffuse support often under pressure of the low performance in the long term, the case of corruption or scandal can add another level of less support (Zhang & Kim, 2018) especially when there is a scandal or corruption. People sometimes forget about their reasons why evaluate the public officials negatively, but the sentiments of the negative evaluation remained to lead to permanent dissatisfaction with government. Easton's approach also criticized as a result-focused mechanism but ignored the process. The cognitive process based on the rationality is also crucial, as well as the evaluative assessment to the performance of government (Hibbling & Theiss-Morse, 1995). The process of governments' work also matters. Some suggest the process of how does the government generate a particular outcome should also consider the evaluating criteria.
2.1.3 Trust as estimation on future or evaluation of past does not differ from one another
trust. Scholars in research did not argue against each other. Based on their studies, trust as estimation on future or evaluation of the past does not differ from each other. They are not conflicted ideas. People can have a positive evaluation of the past, and it generates confidence with future. What is essential behind the different dimension is the criterion that citizens use to cogitate if they should trust the government or not. Citizens have a positive evaluation of the government would lead to their confidence in government in an uncertain future. Trust among people is referring as interpersonal trust, and when interpersonal trust aggregated at the collective level. It would transfer into the collective trust, which is the uniform expectation from people. The uniform expectation is the collective utility or collective interest that citizens with the government to maximize. Therefore, citizens' utility and interests are basic standards that citizens rely upon and use to decide whether government and institutions are trustworthy or not. Lewis and Weigert explain trust from three directions, emotional, cognitive and behavioural. The emotional perspective implies that trust often requires emotional investment, and when governments' performance is unexpected or hurts the citizens' interest, emotion becomes visible, and citizens lower their trust (Lewis & Weigert, 1985). Cognitive perspective of trust occurs when rationality is high, but emotionality is low, people evaluate their trust based on benefit-cost of rational analysis instead of emotional response (Lewis & Weigert, 1985). The behavioural perspective of trust illustrates that when the adverse effects of violating the expectations, people would act in a particular way which decreases their trust in government (Lewis & Weigert, 1985). Lewis and Weigert's three-dimensional mechanism further asserts the importance of citizens' interest. Citizens' interests are a critical factor for their trust in government. When citizens satisfy with their interests received, they have a positive evaluation of the government and institutions, and more confidence with the government in the future.
Political scientists often employ survey to investigate whether citizens trust or distrust the government and institutions. In the LISS panel dataset, they have two methods which are 'confidence' and 'satisfaction'. The survey use questions of "can you indicate, from 0 to 10, how much confidence do you have in government and institutions"; and "how satisfied are you with government and institutions operated in the Netherlands, from 0 to 10". Nevertheless, the panel did not offer further explanations on these questions and the difference between them. A simple word of trust merely explains the attitudes towards government action. The discussion is then built on the satisfaction in the government and the institutions and service they provided. The increasing complexity of social issues and the need for wisdom in
politics to deal with the social issues make the service delivery much harder, and the traditional certainty cannot sustain. It fell into the criteria that the public considers to trust or distrust the government. Trust and distrust are two different ideas. Factors that affect the trust may be different from the factors that lead to the distrust. The implicit direction of these two mechanisms may lead to expectations of the citizens hard to achieve.
2.1.4 Does trust in government differ from trust in other political entities?
Another debate is whether trust in government means to trust in all institutions. Many scholars believe citizens cannot distinguish between government and institutions. If citizens trust government, it implies citizens also trust the institutions (see Poister & Henry, 1994; Stipak, 1979; Elchardus, 1998; Muller & Jukam, 1977). However, the opinion attracts much criticism that argues that people can distinguish between government and institutions. Trust in government does not lead to trust in political institutions. Institutions are created with different functions and specializations to achieve the desired interests and expectations of people. When some scholars employ survey data to examine the citizens' political trust, they found the result of trust in government was the same as other institutions. Their research found it is because some institutions who have a significant impact such as parliament, or political leaders may view by the citizens as the government, and the term of government here is aggregated as one unit (Elchardus, 1998). Their conclusion suggests an idea of generalization (Poister & Henry, 1994; Stipak, 1979; Muller & Jukam, 1977). The idea of generalization implies that there is only one opinion that explains why people trust or not trust the government is because people did not make the distinctions between those institutions. In their findings, respondents often did not distinguish between private and public services. Thus, the conclusions would be that trust in government wrote by those respondents do not necessarily connect with the government, and the concept of government is aggregated as one amorphous idea (Elchardus, 1998).
However, other researches lead to different conclusions. Trust in government is not the same as trust in all institutions, because the performance of institutions is different. Institutions with symbolic role receive more trust than other institutions. Citizens also place more trust in those institutions they had connection than others. In Klingemann's argument, the generalization does not exist; they all depend on the actual performance
of the specific institutions; similarities in the outcome arise from similar performance (Klingemann, 1999). People's evaluations in government were often different from the evaluations on the specific government departments such as police, health (Steen, 1996). This is particularly obvious when people show distrust in government is hardly maintained when people blame the politicians more than the civil servants and remain confident in the government. People find it is easier to evaluate the specific institutions according to the criteria and to evaluate their performance. However, the evaluations for the political leaders need to combine many criteria such as economics, health, military, environment, and it is hard for the public to consider all the criterion.
Moreover, institutions held the symbolic role receive more trust from citizens than specific institutions. Citrin and Green found that respondents' confidence in the US white house is higher than the national executive council. It is because the White House represents the power, but the executive team is a negative connotation of bureaucracy (Citrin & Green, 1996). Plus, citizens trust in government differ from trust in other institutions is because they never had a close connection with the government. If the institutions are closely related to citizens, they can write their trust score based on experience with these institutions (Bouckaert & van de Walle, 2001). Whereas the government and some institutions are not closely linked to citizens, and citizens may not write a high trust (or low) score as the other institutions.
2.2 The concept of personal income taxation
Income, in terms of meaning, is initially defined as 'the incoming money'. It plays an essential role in Economics, not just monetary value but shows capital, labour force participation and the disparity between rich and poor. The government then proposes income tax on individual or entities with their respective income or profits. It is generally calculated as the tax rate multiplied by the taxable income, with different tax brackets settled by the government that allows people paying their taxes based on the income. Tax policy is a vital part of the government's fiscal policy that is an essential part of the governments' revenue. Tax is a sustainable source of government revenue to sustain the social program and public investment to foster social development and economic growth. The programs like infrastructure, education, health and welfare programs are essential for a country to achieve a functional, prosperous and orderly society. To start or strengthen these programs, the government needs funds which they increase the tax rate to raise revenue. However, it is a
challenge to the government not only about the tax rates but also the tax base which establish a bundle to a certain level of income group as to their tax payable; governments are accountable to design the tax compliance system which would not discourage people from paying their taxes.
The structure of personal income tax is an essential part of the government's tax policy. The design of tax policy determines the desired policy outcomes are achievable, and at the same time, stimulating the economic increase and reducing the cost of collecting tax. It means that a well-designed taxation policy would provide enough funds for the government to invest in social programs that can improve the quality of citizens' life, and the improved social programs can create a better environment for domestic economic activities and attracting foreign investment and trade activities, which increase the economic performance of the country (Creedy, 2009). Another opinion illustrates that the tax brackets which determined by the income level play as an essential role in redistribution policies; especially with the labour markets which influenced by the shaping of work incentives and the cost of labour (Bovenberg, 2003). The tax burdens determined by the different earnings level can help too, for example, determine the money paid to the pension system. The design of a tax system can also encourage the labour market; a successful design in tax structure can cut the unemployment rate if the design shifts the tax burden to unemployment. If taxable income increase, progressive taxes moderate wages; therefore, it reduces the unemployment rate (Bovenberg, 2003).
The income tax policy promotes redistribution while enable equity principle. The primary objective of taxation policy is to redistribute resources from one group to another group to achieve the desired development goals without jeopardizing the economic objectives. Even though the tax policy is based on transferring from one income group to another, the equity of transferring is enabled. There are two dimensions to study the equity of tax policy: horizontal and vertical dimensions. The horizontal equity presents when there is equal tax treatment of taxpayers who have the same ability to pay the taxes (Szarowska, 2014). Vertical equity enables a suitable correlation between the relative tax brackets paid by people with different capacities to pay taxes (Szarowska, 2014). Therefore, the taxation policy is equitable across people with the equal ability to pay taxes, and it also enables the proper relationship with people who have different capability to pay taxes. The objective of tax policy, on the one hand, create the platform for transferring resources for the redistributive
purpose, it ensures the equity of allocating resources from one group to another; on the other hand, the platform created can achieve the development goals without sacrifice the economic objectives. Despite the objective, the taxation policy also has some indicators for the policy to success. There is no formula to settle the proper dimensions of progressivity of taxation. The main factors that taxation depends are the degree of interpersonal solidarity in society or the degree of shirking by potential recipients of transfers. Income taxes are progressive taxes, which the burden increases as the income increases. The success and sustainability of tax compliance determined by the people's linkage, whether people accountable to society and whether the taxpayers comply with the tax policies.
Another function of tax policy can maximize utilities of citizens. Regarding the practical advice about the tax policy, Adam Smith's famous four maxims are often quoted by scholars. The four maxims are contribution based on the capacity to pay, convenience, certainty and efficiency (Smith, 1776). Among these four maxims which are often extended and clarified, the redistribution was not included when the scholars discuss the ability to pay taxes. The independently given revenue requirement then indicated the appropriate tax rate. In old times, the main change in the method to tax came from the later integration of public finance to the general area of welfare economics, which introduces the idea to achieve the maximum utility (Creedy, 2009). The development investigates the broad implications for progressivity of the minimization of total disutility from taxation-ignoring any possible benefits. With the criterion of minimizing total sacrifice, progression arises from a declining marginal utility, but with the equal absolute sacrifice, it depends on the apparent behaviour of the marginal utility of income.
Personal income tax policy has other advantages such as guidance, and reducing income inequity. But studies also propose criticism to tax policy. There are some basic principles about the taxation structure such as a difference occurs between legal and economic incidence, taxes can be shifted like the tax capitalization, enormous efficiency costs can arise even when taxes appear to have little impact on the behaviour. However, many economists propose the idea of 'rule of thumb', which is meant only as a guiding aim (Creedy, 2009). Few countries score well when applying the basic 'broad base-low rate' rule of thumb, which suggests, through the excess burden correlation with the tax rate, a preference for a relatively flat income tax rate structure, although this does not allow for redistribution objectives and associated
trade-offs, which may lead to modifications (Creedy, 2009). Plus, the top level of the marginal income tax rate (a higher rate on the high-income group) is typically introduced, which can reduce the income inequality. However, tax rate progression is not needed for progressivity, and the top marginal rate usually raises little funds (Szarowska, 2014). The argument also used in discussing the labour supply elasticity, which is low for higher-income people, it means the low sensitivity of the supply of labour to a change in the salary rate (Bovenberg, 2003). So the incentives effects of reducing the income inequality in a top income rate are negligible.
2.2.1 The personal income tax system in the Netherlands
The personal income tax policy in the Netherlands is regulated by the Wet inkomstenbelasting 2001, and the fiscal year is the same as the calendar year (from January to December) (de Boer, 2016). According to the legislation, income in the Netherlands is divided into three categories (boxes) for tax purposes. Firstly, income from work and homeownership, which is progressive taxes on wages, social security benefits and pensions (KPMG, 2019). It has different tax brackets with the specific marginal tax rate. Box two focus on the personal wealth of substantial interest, there is usually a flat rate of 25% on income from a substantial interest if the taxpayer, or with a partner own 5% of shares or options in the company (KPMG, 2019). Box three focus on the savings and investments, there is a flat tax rate calculated based on income from savings, investment, shares and a second home, generally a 30% on the taxable income from savings and investment (KPMG, 2019). The total tax that taxpayers are liable to the government is the sum in the three boxes and then minus the tax credit, which is a tax incentive that government provide to the taxpayers that allow them to subtract the amount of the credit taxpayers have accrued from the total tax they owe to the government.
Table 1: The Personal Income Tax Rate in the Netherlands - Box One -(2007 - 2019) Year Tax credit 01 Margina l rate 01 total rate with the premium contributi on 01 Threshold 02 Margina l rate 02 total rate with the premium contributi on 02 Threshold 03 Margina l rate 03 Threshold 04 Margina l rate 2007 2043 2.5 33.65 17319 10.25 41.40 31122 42 53064 52 2008 2074 2.45 33.6 17579 10.7 41.85 31589 42 53860 52 2009 2007 2.35 33.5 17878 10.85 42 32127 42 54776 52 2010 1987 2.3 33.45 18218 10.8 41.95 32738 42 54367 52 2011 1987 1.85 33 18628 10.8 41.95 33436 42 55694 52 2012 2033 1.95 33.1 18945 10.8 41.95 33863 42 56491 52 2013 2001 5.85 37 19645 10.85 42 33363 42 55991 52 2014 2103 5.1 36.25 19645 10.85 42 33363 42 56531 52 2015 2203 8.35 36.5 19822 13.85 42 33589 42 57585 52 2016 2242 8.4 36.55 19922 12.25 40.40 33715 40.4 66421 52 2017 2254 8.9 36.55 19982 13.15 40.80 33791 40.8 67072 52 2018 2265 8.9 36.55 20142 13.2 40.85 33994 40.85 68507 51.95 2019 2477 9 36.65 20384 10.45 38.10 34300 38.1 68507 51.75
(source: OECD statistics database, available athttps://stats.oecd.org/)
Note: For both two lower taxable income thresholds, there is a separate column integrates not only income taxes but also mandatory pension, social security and state-funded medical care payments.
2001. There are four income tax thresholds and tax credits which changes almost every year. In theory, the reform in the taxation system not only focus on the efficiency of resource allocation, but the equity of the tax system also matters in the previous section, where tax policy shows equity from horizontal and vertical dimensions. The Dutch tax system complies with the equity theory with that taxpayers in a similar circumstance with the same financial ability bear the same tax burden (the horizontal equity). In contrast, people with better economic capability in the better situation bear the more substantial tax burden as a proportion to their income (the vertical equity). Another hypothesis focus on whether the reform in the tax system can stimulate the labour market. In some studies, their empirical results support the hypothesis that the relationship between income tax policies and the labour market. More specifically, de Boer in the research shows the reform in the Dutch tax system, especially the recent reform on tax credits, can increase the labour supply (de Boer, 2016). Jacobs et al. in 2009 employed a general equilibrium model to investigate the labour supply effects of a flat tax rate in the Netherlands. Their empirical results indicate that a system with a relatively low flat tax rate of 38% can lead to 1% of increases in total labour supply, earners face lower marginal income tax rate increase their labour supply (Jacob et al. 2009). According to their researches, the Dutch taxation design reflects the equal value and can motivate the labour supply.
2.3 Interaction between trust and income tax (a causal mechanism).
There are many research studies the factors which affect citizens' trust in politics. These studies mainly focus on broad economic factors, citizens' education and some social phenomena such as corruption and crisis. There is also much research that analyzes the impact of income tax policy as a crucial part of the government's fiscal policy. Many existing studies and literature examine the change of income tax policy and its impact on citizens' behaviour, particularly on tax compliance behaviours. These studies also examine whether the tax policy is sufficient to increase tax revenue. I hardly find relevant studies that investigate the correlation between changes in income tax policies and taxpayers' trust in politics. However, it is meaningful for the policymaker to investigate whether there is a connection between these two concepts. It is because both policies are crucial to the government's revenue, legitimacy and democracy.of taxes citizens need to contribute to the country. Income tax also crucial to the government. As shown in the Graph 1, for example, in the Netherlands, if social security contribution included, the income tax revenue (box 1) accounted for more than half of the government's total tax revenue. Therefore, citizens' compliance with tax is crucial to the government. The tax system often adjusts annually, how do the taxpayers perceive these changes? Citizens' trust in government enhance the legitimacy of the government, and the changes in the government's tax policy would affect the citizens' confidence in politics. The origin of the ideology of political trust occurs from people' consideration as to whether the government would act on their benefits, and if the citizens left untended, would the government produce the preferred results that act on the interest of the people and citizens satisfied with the results. If the government's action does not match with the citizens' expectation or condition; the result would lead to lower confidence in the government. The changes in personal income tax would then lead to the citizens' concern on political trust because citizens expect the government's actions in improving welfare services.
(Source: OECD Database)
Change in government's income tax policy would lead to a decline in citizens' trust because citizens perceive the tax growth as a decrease in their well-being. In theory, personal income tax policy has an ambition of maximizing citizens' utilities. Citizens expect the government to produce more public goods and better service to maximize their utilities. If the government's action does not match with the citizens' expectation or condition; it would lead to lower confidence in the government. The changes in personal income tax would then lead to the citizens' concern on political trust because
citizens expect the government's actions in improving welfare services as the research outputs demonstrate that people have inverse attitudes towards the tax revenue contrary to the government. Citizens have a negative evaluation of government when tax revenue increase, but positive evaluation on government when the tax revenue decreases (Elena, 2015). The research indicates those countries with growth in general government's total tax and non-tax revenue, the situation perceived by their citizens as the decline in their well-being. In contrast, the decrease in government' tax and non-tax revenue perceived by the citizens as the increase in their well-being; and the insignificant fluctuations in tax and non-tax revenue perceived by the citizens as a neutral influence on their welfare (Elena, 2015).
The change in income tax policy would lead to different results between trust in government and trust in institutions. In the literature, the studies have different opinions about trust in politicians and institutions. Some scholars found that citizens often exhibit more trust in institutions who generate a symbolic output than specific institutions. The counterargument would be people focus more on the structures of the political system instead of the personality of political leaders. In theory, political leaders are often hard to evaluate because people need more criteria that combine many factors such as economy, education, environment and health. However, people find it is easier to evaluate the specific political institutions, for example, if people feel positive as to the well-beings they received in Education or health care, they play more trust in these institutions. Services that citizens receive would influence their trust in government and institutions, and the increase in taxes would generally mean more efforts and funds would allocate on the governments' service which should exhibit a better quality of citizens' well-being (Oishi, Schimmack, & Diener, 2012). In Elena's research, citizens' and governments' perception of the correlation between taxes and services show the opposite situation (Elena, 2015). If people dissatisfied with the well-beings they received, they would show lower trust in institutions. Despite the arguments of whether politicians or institution citizens focus more, the most crucial factor here is that which part of the government that citizens hold as accountable to their well-beings. For instance, if citizens believe the education department is responsible for education services. When they dissatisfied with education, they would distrust the education department while still trust the government and other institutions. Therefore, the change in income tax policy may generate different results with trust in government and institutions.
Chapter 3: Assumptions and Hypothesis:
3.1 Key assumptions:
Assumption 1: Humans care about the services they received.
Although there is evidence suggests that people are emotional and moral driven creatures. This thesis assumes that people are self-interested and care about their service and public goods received. Research supports this assumption and suggests that humans are always motivated by self-interest and selfishness (Cropanzano & Goldman, 2005). Individuals' profit can be a powerful motivation to drive their actions, and their objective is to maximize their utilities. Because people care about their interests and utilities, and the citizens consider the government's services and public goods they offer as their interests. If the interests of citizens get hurts, they lower their trust in the government.
Assumption 2: People are honest with their taxes
Even though there are some studies conclude that non-compliance behaviour often occurs when people pay taxes. People often try to avoid taxes and even manipulate their taxes pay to the government. In general, citizens can manipulate their amount of taxes need to pay for the wealth taxes. However, with the personal income tax of box 1, it is difficult for people to manipulate their wage taxes. Paying taxes to a country is a fundamental obligation to citizens in each country. In the Netherlands, there are both punitive measures, as well as tax credits and deductions as incentives to ensure that citizens fulfil their obligation of paying taxes.
Assumption 3: Participants' responses to the survey are truthful
There are many debates as to whether the survey data is reliable or not. Many researchers have little confidence with the accuracy of survey data. They believe that survey data is not reliable because people always hide their actual performance from data and lies. This often occurs from some online random survey. However, with the LISS panel data archive that I employ, the panel is based on a real probability sample of households drawn from a population that registered at Statistics Netherlands. People are paid for their contribution to answering questions. Both fixed response from registration and paid response to incentivize the respondents to fill in the survey honestly compared with other typical surveys. Besides, a discussion paper for Statistics Netherlands studied the most survey in the Netherlands. Their empirical results ensure the accuracy of data from the LISS panel with a p-value less than 0.05
compared to other survey data in the Netherlands (Bruggen, van den Brakel & Krosnick, 2016).
Assumption 4: Respondents have a criterion for each political component
Even though there is many studies show people cannot distinguish between government and other institutions (Poister & Henry, 1994; Stipak, 1979; Muller & Jukam, 1977). These studies suggest that trust in government is trust in all institutions. However, this thesis supports that people have a specific evaluation of each institution according to their performance. Citizens can distinguish between government and other institutions. Respondents' trust in government differs from trust in institutions. This assumption echos many other studies (Klingemann, 1999; Steen, 1996; Citrin & Green, 1996; and Bouckaert & van de Walle, 2001).
3.2 Hypotheses
People paying higher taxes compared with similar individuals around the tax threshold decreases their political trust. An increasing marginal tax rate leads to lower trust in government and other political institutions.
Citizens are uncertain about the public goods they receive. If citizens are dissatisfied with the public goods and service they paid, this will lead to a decline in trust. In literature, people are generally left tended in the process of government's service. Trust in government to citizens is a belief that the government would produce public goods and better service to improve their welfare. Citizens' tax is the primary source of government funds to produce public goods. When governments propose to increase taxes to finance public investments, they are asking citizens to accept a potential trade-off (Jacobs & Matthews, 2017). The trade-off is to request taxpayers sacrifice now, in return for the promise of more public goods and better services in the future. From the citizen's perspective, the difficulty is that the costs of that trade-off are defined: which is the amount of taxes paid to the government. The promised benefits from the government, on the other hand, are uncertain. Citizens cannot ensure whether the promised benefits or public good is to act on their interest. If the public goods do not satisfy citizens' need, it affects the citizens' evaluation on government. People pay taxes for public goods even if the public goods they receive may not as they expected; in worse scenario, they lose interests (Reeskens & Vandecasteele, 2016). Under this circumstance, citizens may feel disappointed at the government and decrease their political faith. The uncertainty of public goods and government services
affects citizens' trust in government.
Citizens pay more taxes to want better government service or public good to improve their well-being. If they pay more taxes compared with similar individuals (in terms of income) but receive the same public goods and services, this would also result in lower trust. The income tax rate sometimes increases every year, and people need to pay more taxes than before. Individuals pay taxes to wish the government could, in return brings benefits to citizens, better service and more public good to taxpayers. Nevertheless, the nature of public good is that individuals cannot get exactly their money's worth, no less and no more, of precisely the public goods each wants (Stretton & Orchard, 1994). Some people get more benefits than others, even if they pay fewer taxes. If taxpayers perceive the redistribution is unfair or the public good does not satisfy their needs, they decrease their support for the government. Citizens around the taxable income thresholds often, ceteris paribus, receive similar public goods. However, people with an income slightly higher than the threshold pay additional taxes. Citizens pay more taxes with the belief of receiving more public goods or improved well-being. Given the circumstance that people paying slightly higher taxes receive public goods does not differ from similar individuals who pay relatively lower taxes. Do they show a lower trust in politics? The hypothesis for further investigation in this thesis is that people pay more taxes compared with similar individuals (in terms of income) decrease their trust in politics. Another hypothesis is a higher marginal tax rate lead to lower trust given people receive similar benefits from the government around the thresholds.
The remaining parts of this research paper proceed as follows: it starts by introducing the design of this study, the survey data employed in this thesis and the research methods going to use in the research. Then I will report the empirical results from the use of this method, and the discussion of the results. After all, I will conclude that this research includes the limitations throughout the study and some recommendations on future research on this topic.
Chapter 4: Data and Research Methods
4.1 Data (Personal Income and Political Trust)
their taxes), as well as their trust in government and institutions are main variables to this topic. The LISS panel survey data in the Netherlands record many sources of citizens' income. This research only employs the total gross wage they earned for the whole year, which is the one citizens apply to box 1 to pay the taxes. Based on the respondents' income, and combine them with the personal income tax rate in table 1, which indicates the marginal tax rate of each respondent. The outcome variables are very similar among the studies. As shown in the literature, the most survey show a list of institutions to the respondents and ask how does the trust in the government range from the lowest to the highest. For the outcome variables, the survey data did not record people's trust in governments directly, and instead, they ask the respondents as to whether they are confident and satisfied with the government and the specific institutions from 0 to 10. It can reflect whether citizens trust or distrust the government. However, the LISS panel did not explain precisely the difference between confidence and satisfaction in the survey. The outcome variables also distributed into specific institutions, which may help to analyze whether there is a generalization process that a common factor behind the citizens' trust in government and institutions.
The outcome variables: The Government,
The LISS panel survey records people's trust in government directly, which can help to observe the changes in citizens' trust in government. As the conflicting opinions among the different scholars that some believe the government is an aggregated idea that citizens need many criteria to evaluate the government and evaluation of the specific institutions based on the experience. In contrast, the other scholars suggest the citizens do not often make distinctions on the government and the functional departments, and there is one common factor behind their political trust. This survey data make specific questions that asking how does citizens trust in government and the other institutions. It can help to examine whether there is one common idea behind citizens' political trust or citizens have different criteria to evaluate the different institutions.
The Legislature (Parliament & the legal system),
The first measure in the survey is how does people's trust in the legislative institution represented by the nation's Parliament. The respondents can choose from 1 (totally distrust) to 10 (totally trust). Higher numerical values indicate people's higher trust in the country's Parliament. The measure of people's attitudes towards the country's Parliament, on the one hand, distinguish the executive power from legislative function,
which was a vital distinction in the debate; on the other hand, a single measure on the country's Parliament can show people have different attitudes towards different function of the political system, or the generalization exist. When disagreement comes, the ability to resolve it is essential to every citizen. The existence of the legal system is a procedure that interprets and enforce the law, and the purpose of the legal system is to offer a systematic and estimable manner for overcoming the disagreement. The ability that a legal system can resolve the citizens' disagreement also matter to citizens' daily life. The consideration of how respondents trust in the country's legal system can reflect how do the respondents feel about the ability of country's judiciary department to resolve the citizens' disagreements from 0 (the lowest) to 10 (the highest).
Government Services (Police, Education System, health care),
The Police department is often representing the civil authority of the government. They were ordinarily responsible for the maintaining of social order, public safety, enforcing the law and preventing, detecting and investigating the criminal activities. Public trust in Police is essential not only it reflects whether the society is safe, legislations are expertly executed, and criminal activities that threaten citizens' interest and social order are effectively detected or investigated; but also public trust is essential to the Police. It can help the police department to strengthen its connection with the public which encourage more engagement from the citizens in security and policing activities because the public, sometimes have the critical information and their trust in Police can encourage higher productivity and effectiveness. The education system, often referring to public schooling, is essential to citizens' well-being. The purpose of the government's taxation is to allocate the resources, collected revenue to the government allows them to develop its education plans. For people concerned with the education system or people involved in the education system may very sensitive to the education plans of the government. A change in taxation policy may affect their confidence in the government. Similar to the education system, health care is critical to the citizens' welfare. People cared about the health care system may be influenced by the taxation policy.
The Political Institutions (Politicians & the Political Parties),
Politicians were referring to someone who actively engaged in politics, particularly those people who hold the offices. As it is shown in the literature, that politicians often viewed by the public as the government, the debates in this topic often distinguished by whether the generalization occurs when people respond to their trust in the government. Another assumption is that people can distinguish the political
system from each other. The method used by the survey asked how respondents trust in government in these seven separate indicators, the single answer for to what extent does the respondent's trust or distrust the politicians can show whether people's attitudes towards the government was solely affected by the view on politicians, how does public trust in politicians correlated to other indicators. Political parties are the organized institutions that consist of people who share the similar objective or have the similar objective, and they have the candidates for election, which is an attempt to get the candidates elected which then implement the political agenda of their parties. The literature also indicates how people see different political parties matter to their lives, for example, left-party and right-party have different attitudes towards the employment, there have different policies to work on the unemployment which brings different outcomes. The public trust in political parties can reflect how do the different outcomes affect the public, and how do the respondents respond to the difference.
4.2 Research Approach: Regression Discontinuity Design
The principal methodology in this thesis is regression discontinuity, which is a way of estimating treatment effects and establishing a causal effect in a non-experimental setting. It is a design that examines the causal effect of interventions. This method through assigning a cutoff or a threshold above or below where intervention is assigned by comparing observations lying closely on both sides of the discontinuity. Some believe the causal inferences drawn from regression discontinuity design are potentially more credible and transparent than other methods from a typical 'natural experiment' approach (Alghumgham, 2019). Regression discontinuity design assigns subjects to a control group and a treatment group, and the outcome is compared to estimate the treatment effect. The experiment is a rigorous way of establishing causality between variables and measuring treatment effects to examine the potential causal effect (Imbens & Lemieux, 2008).
Regression discontinuity is a method that estimates treatment effects and establishing causality between two subjects. Observations in a dataset are separated into a control group which below the discontinuity and a treatment group when the assigned variables are above the cutoff point. The theoretical justification indicates that one need not assume the regression discontinuity separate treatment variation that is "as good as randomized" (Alghumgham, 2019). Instead, such randomized variations are the consequence of agents' incapacity to control the assignment variable near the threshold (Imbens & Lemieux, 2008). RDD can be used in a wide variety of subjects
and context, covering a large number of crucial economic questions to evaluate policy effectiveness (Alghumgham, 2019). Regression discontinuity is useful for program evaluation, such as to investigate if tax incentives to promote business development, individual income tax policies to promote work and savings, and other social programs (Lee & Lemieux, 2010).
The environment where regression discontinuity method often used by researchers is that a threshold in a continuous variable (Alghumgham, 2019). Income, for instance, generates a substantial change in a policy variable (e.g. the income tax rate that taxpayers must pay) (Alghumgham, 2019). People cannot violate the threshold in RD. There are many data close to the cut-off point for the forcing, policy, and outcome variables (Barreca, Lindo & Waddell, 2015). Other factors that might affect the outcome do not change discontinuously at the threshold. People do not manipulate the forcing variable close to the threshold to make themselves eligible (Barreca, Lindo & Waddell, 2015).
Besides, regression discontinuity is quantifying the average treatment effect at the cutoff threshold, and it impacts on citizens' political trust and taxpayers' behaviour (Imbens & Lemieux, 2008). The average treatment effect for the treated (ATT) is the concept. It measures the treatment effect that the income tax rate influences (Alghumgham, 2019). The main assumptions are: firstly, the probability of treatment receipt must be discontinuous at the cutoff point (Alghumgham, 2019). In a sharp regression discontinuity, all on one side as the control group, and all on the other side as the treatment group. Secondly, no discontinuity in potential outcomes in the cutoff (the continuity restriction) (Alghumgham, 2019). That is, no alternative interpretation should also show a discontinuity at the cutoff. Otherwise, it would serve as a causal confound (Alghumgham, 2019).
Regression discontinuity design is a standard method used in economic analysis. This approach suits this thesis which investigate the treatment effect of higher taxes. The main objective of using a regression discontinuity approach in this thesis is to do a causal analysis which is to analyze people around the tax brackets and compare their political trust. Another purpose is how does a change of income tax rates in a particular group of people affect their political trust. Notably, the regressional correlation between changes in the income tax rate and citizens' trust in politics. The income tax policy is a complicated system that the tax thresholds change almost every year. The primary purpose of using regression discontinuity is to observe the citizens' political trust when comparing similar individuals who pay different tax rates. The method of doing regression discontinuity analysis is to split the sample up in workers based on their income.
To study the intervention of the income tax threshold and comparing the similar individuals around the tax threshold and their political trust. The cutoff point is the tax thresholds set by the government. The treatment group who affected by the policy is people with taxable income higher than the thresholds that pay additional taxes. The control group is people below the threshold that pay lower taxes. In addition to regression discontinuity analysis on people who pay higher taxes, the choice of income group also needs to consider the effects of change in the income tax rate on citizens' political trust. As shown in Graph 2 above, the income tax rate for the lower three tax brackets often changes over the past decade between 2007 and 2019. The second income tax rate reflects a fluctuation with an overall downward trend from 41.40% in 2007 decreased to 40.85% in 2019. The third income tax rate remains unchanged until 2015. From 2016 onwards, the third tax rate shows an increasing trend from 40.4% in 2016 to 40.85% in 2018.
The approach of regression discontinuity to this thesis is that for the second tax bracket from 2007 to 2018 and the third tax bracket from 2016 to 2018. Since the LISS panel dataset only records people's responses in the Netherlands. The sample does not have enough data to run regression when I chose €100 and €200. I decide to choose €500 as the amount close to the thresholds. That is, €500 less than the threshold and €500 more than the threshold as the interval of the sample. For the lower-income taxes, the treatment groups are people who slightly above the second tax threshold (with no more than €500) from 2007 to 2018, and the control group is people below the second tax threshold (with no less than €500) over the period. For the upper tax bracket, the treatment group is people above the third tax thresholds (within €500) between 2016 and 2018. The control group is people with annual income below the tax threshold, and the amount below is no more than €500.
The equation for Regression Discontinuity Design:
g
gov G G Post Post
C 1 2( ) ,
Where C is the index of the citizens' confidence with government, institutions or services; G denotes respondents' gross salary in the year for tax purposes; the post is a binary indicator which taking value 1 for those individuals with gross income receive the treatment (treatment group, people affected by the tax policies), and value of 0 is taking for the control group who are not influenced by the policy.
Another objective of this study is to analyze the relationship between changes in the income tax rate and citizens' trust in politics. Regression analysis can estimate the relationship between a dependent variable (outcome variables) and independent variables (predictors). A regressional approach can reflect the causal correlation between changes in income tax rates and citizens' trust in politics. Regression approach is to prove the changes in income tax rates are conditions of people' trust in politics. To increase the accuracy of the outcomes, some variables such as inflation and GDP growth that may affect the citizens' pay should include in the model as control variables. The first regression is to analyze if the decreasing income tax rate between 2007 and 2018 affect respondents' trust in politics. The second regression is to analyze if the increasing income tax rates from 2016 to 2018 influence the citizens' trust in politics.
The unit of analysis is the entity being analyzed in this study, which is the citizens in the Netherlands with their income that influenced by the taxation policy, and their trust in politics. How does the individual response to political trust when they are paying different taxes both from vertical and horizontal dimensions. The relevant population is, as the whole, how does income taxation policy affect dutch people's political trust based on the observation from the sample data chosen to explore the causal mechanism between taxation policy and political trust in the Netherlands in general. A common threat to this causal mechanism is the reverse causality, which is not X who affects Y, but Y influence X. It means that is not the taxation policy which affects citizens' trust in government, but the public trust in politics influence the government's change in income taxation policy.
Chapter 5: Empirical Results
Table 2: Results of lower tax bracket: Citizens’ Confidence: 2007 - 2018
Year government Parliament Legal System PoliticianPolitical PartiesPolice EducationHealth
2007 0.19 0.14 -0.59 -0.10 -0.96 -0.16 0.20 -0.05 (1.20) (1.24) (1.05) (1.11) (1.28) (1.19) (0.82) (0.94) 2008 -3.39 -3.17 -1.30* -6.07 -3.59 -3.84 -1.49 -3.33 (2.44) (2.46) (1.70) (2.27) (3.12) (2.41) (1.13) (2.11) 2009 -0.009 0.09 1.88 -0.39 -0.26 0.50 0.54 -1.20 (1.13) (1.28) (1.64) (1.52) (1.36) (1.46) (1.31) (0.42) 2010 1.42 1.87 -1.40 1.21 1.32 0.57 1.31 -0.65 (0.67) (0.68) (0.64) (0.62) (0.61) (0.55) (0.48) (0.54) 2011 -0.41 1.26 -0.59 0.16 0.24 0.17 0.85 -0.49 (0.52) (0.53) (0.55) (0.50) (0.49) (0.46) (0.41) (0.45) 2012 -2.29 -1.74 -1.06 -3.57 -1.79 -1.33 0.41 0.21 (0.66) (0.69) (0.65) (0.64) (0.64) (0.56) (0.41) (0.41) 2013 -0.56 -0.56 -0.41 0.79 0.78 -0.57 -0.12 -0.60 (0.73) (0.80) (0.69) (0.71) (0.71) (0.61) (0.56) (0.61) 2015 -0.56 -0.56 -0.41 0.79 0.78 -0.57 -0.12 -0.60 (0.69) (0.068) (0.66) (0.71) (0.68) (0.62) (0.48) (0.59) 2016 1.50 1.71 -0.74 3.01* 3.21* -0.05 -0.09 1.12 (0.74) (0.77) (0.76) (0.77) (0.81) (0.67) (0.68) (0.66) 2017 -7.55 -0.04 1.25 -1.29 -1.29 9.13 1.65 1.65 (0.99) (0.87) (1.09) (0.98) (0.98) (0.98) (0.65) (0.72) 2018 1.89 2.10 3.48** 1.25 1.31 2.53** 2.28** 1.45 (0.90) (1.00) (0.84) (0.88) (0.91) (0.70) (0.59) (0.68)
Table 3: Results of lower tax bracket: Citizens’ Satisfaction: 2007 - 2018
Year government Parliament LegalSystem Politician PoliticalParties Police Education Health
2007 -1.25 0.39 0.39 -0.27 -0.19 -0.46 -0.27 -0.81 (1.13) (1.79) (0.52) (0.47) (0.46) (0.91) (0.61) (0.87) 2008 -2.87 -3.16 -6.54 2.12 3.62 0.85 3.06* 2.41 (0.51) (0.57) (0.60) (0.50) (0.50) (0.49) (0.45) (0.46) 2009 -0.80 -1.88 1.31 -0.73 -1.27 -0.07 -0.36 -1.95* (1.30) (0.57) (0.54) (0.54) (0.53) (0.50) (0.43) (1.09) 2010 1.37 1.57 0.54 0.94 1.28 1.24 1.87 -0.59 (0.67) (0.73) (0.63) (0.66) (0.66) (0.55) (0.52) (0.54) 2011 0.76 1.57 0.64 -0.56 -0.36 -1.29 0.85 -0.94 (0.55) (0.59) (0.59) (0.52) (0.53) (0.48) (0.47) (0.46) 2012 -2.29 -2.55 -1.07 -2.92 -2.44 -0.63 0.39 0.73 (0.70) (0.73) (0.66) (0.65) (0.65) (0.55) (0.46) (0.46) 2013 1.14 0.73 -0.25 0.32 0.84 -0.35 -0.66 -0.84 (0.76) (0.82) (0.74) (0.74) (0.77) (0.67) (0.62) (0.67) 2015 1.58 0.96 -1.33 0.32 -0.55 -1.52 -0.84 1.33 (0.75) (0.76) (0.68) (0.74) (0.73) (0.64) (0.49) (0.62) 2016 0.93 0.18 0.02 3.56* 1.38 0.56 0.48 0.64 (0.78) (0.82) (0.84) (1.25) (0.79) (0.71) (0.66) (0.65) 2017 1.05 1.07 0.84 0.58 0.61 1.86 0.94 0.48 (1.04) (0.99) (1.15) (0.98) (0.97) (1.00) (0.61) (0.72) 2018 2.01 2.73 3.59* -0.34 -0.12 2.33* -1.02 -1.63 (1.47) (1.02) (1.86) (0.85) (0.88) (1.13) (0.72) (0.80)
*/**/*** indicates the significane level at 10%, 5% and 1% respectively Table 4: Results of upper tax bracket: Citizens’ Confidence: 2016 - 2018
Year government Parliament Legal System Politician Political Parties Police Education Health
2016 -0.46 -0.36 0.13 -0.44 -0.50 -0.05 -0.33 -0.35 (0.51) (0.51) (0.50) (0.52) (0.53) (0.41) (0.39) (0.38) 2017 -1.77* -1.74* -0.71 -1.15 -0.49 -1.14 0.70* -0.70 (0.90) (0.76) (1.16) (1.45) (1.23) (1.51) (0.39) (0.74) 2018 -0.05 -0.36 -0.13 0.01 -0.17 0.22 0.17 0.05 (0.43) (0.46) (0.45) (0.44) (0.47) (0.36) (0.30) (0.31)
Table 5: Results of upper tax bracket: Citizens’ Satisfaction: 2016 - 2018
Year government Parliament Legal System Politician Political Parties Police Education Health
2016 -0.62 -0.56 0.09 -0.40 -0.65 0.61 -0.34 -0.63 (0.57) (0.54) (0.54) (0.55) (0.54) (0.49) (0.41) (0.40) 2017 0.05 -0.26 -0.84 -0.98 -0.73 1.30 -0.92 -1.52 (1.40) (1.31) (0.95) (1.39) (1.24) (2.67) (0.67) (0.89) 2018 -0.10 0.09 0.11 -0.40 -0.25 0.38 -0.009 -0.06 (0.48) (0.53) (0.50) (0.51) (0.51) (0.42) (0.31) (0.34)
Table 6: Regressional results of tax rate: Citizens’ Confidence: 2007 - 2018 governme
nt Parliament LegalSystemPolitician PoliticalParties Police Education Health
Coefficient 0.18*** 0.15*** 0.17*** 0.24*** 0.24*** 0.13*** 0.22*** 0.16***
(0.03) (0.03) (0.03) (0.03) (0.03) (0.02) (0.02) (0.02)
Inflation Yes Yes Yes Yes Yes Yes Yes Yes
GDP
Growth Yes Yes Yes Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes Yes Yes Yes
Observatio
n 46,298 46,290 46,289 46,287 46,290 46,294 46,264 46,263 R-squared 0.0111 0.0086 0.0133 0.0082 0.0084 0.0139 0.0116 0.0054
Constant 90.37*** 53.03*** -71.07*** 91.72*** 93.48*** -98.82*** -92.01*** -46.17***
(6.51) (6.67) (6.84) (6.45) (6.37) (5.97) (5.18) (5.41) Table 7: Regressional results of tax rate: Citizens’ Satisfaction: 2007 - 2018
governme
nt Parliament LegalSystemPolitician PoliticalParties Police Education Health
Coefficient 0.14*** 0.11*** 0.12*** 0.19*** 0.22*** 0.09*** 0.18*** 0.13***
(0.03) (0.03) (0.03) (0.03) (0.03) (0.02) (0.02) (0.02)
Inflation Yes Yes Yes Yes Yes Yes Yes Yes
GDP
Growth Yes Yes Yes Yes Yes Yes Yes Yes
Observatio
n 46,218 46,214 46,210 46,212 46,213 46,217 46,199 46,201 R-squared 0.0101 0.0088 0.0127 0.0085 0.0083 0.0133 0.0086 0.0061
Constant 80.27*** 55.64*** -48.46*** 87.01*** 85.39*** -78.66*** -74.76*** -22.41***
(6.85) (7.04) (6.97) (6.59) (6.57) (6.15) (5.62) (5.76)
Table 8: Regressional results of tax rate: Citizens’ Confidence: 2016 - 2018
government Parliament LegalSystem Politician PoliticalParties Police Education Health
Coefficient 0.56*** 0.67*** 0.82*** 0.42*** 0.47*** 0.45*** 0.44*** 0.51*** (0.06) (0.06) (0.06) (0.06) (0.06) (0.05) (0.04) (0.05)
Inflation Yes Yes Yes Yes Yes Yes Yes Yes
GDP
Growth Yes Yes Yes Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes Yes Yes Yes
Observatio
n 6,959 6,959 6,959 6,959 6,959 6,959 6,959 6,959
R-squared 0.0146 0.0188 0.0247 0.0111 0.0132 0.0138 0.0126 0.0191
Constant 91.98 95.10 107.43 -47.20 -37.09 -86.43 56.07 -65.05
(29.84) (29.51) (32.59) (28.78) (28.61) (18.34) (26.81) (14.42) Table 9: Regressional results of tax rate: Citizens’ Satisfaction: 2016 - 2018
government Parliament LegalSystem Politician PoliticalParties Police Education Health
Coefficient 0.66*** 0.75*** 0.81*** 0.45*** 0.50*** 0.48*** 0.49*** 0.52*** (0.07) (0.07) (0.06) (0.06) (0.06) (0.06) (0.05) (0.05)
Inflation Yes Yes Yes Yes Yes Yes Yes Yes
GDP
Growth Yes Yes Yes Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes Yes Yes Yes
Observatio
*/**/*** indicates the significane level at 10%, 5% and 1% respectively
R-squared 0.0156 0.0189 0.0225 0.0110 0.0130 0.0129 0.0121 0.0182
Constant 83.38 61.96 77.20 64.00 14.23 -46.35 25.66 -38.54
The table above reports the results for both regression and regression discontinuity. Firstly, let us address the outcomes for the second income tax threshold. Both results with the use of taxpayers' confidence and taxpayers' satisfaction show no significant results with trust in government. Respondents' trust in the Legislature, however, shows a difference with the outcomes between the use of confidence and satisfaction. The results of trust in parliament show no significant outcomes. Trust in the legal system, show 1.30 unit of decrease with confidence in 2008. 3.48 unit and 3.59 unit of increase with confidence and satisfaction respectively. Trust in political institutions in 2016 show positive and significant increase. Their satisfaction with politicians shows a significant increase in 2016. Taxpayers' confidence and satisfaction with the government services show positive and significant outcomes with police and education. Whereas 1.95 unit of decrease with their satisfaction with health care in 2009. Overall, people paying higher taxes did not reflect a decline with their trust in government. However, they show a change with trust in institutions and services in different years.
The income tax rate for the second tax brackets experiences a fluctuation with an overall downward trend between 2007 and 2018. The table reports the regressional correlation between changes in income tax rates and taxpayers' trust in politics. With the control variable of inflation and GDP growth included. The results show an overall positive and significant coefficient by employing data with confidence and satisfaction. Coefficients of confidence are generally higher than the coefficients of citizens' satisfaction. Among the political components, trust in institutions yields the highest increase compared with other dependent variables with 0.24 unit of increase with taxpayers' confidence with politicians and political parties, 0.19 unit of increase of satisfaction with politicians and 0.22 unit of increase with their satisfaction with political parties. Trust in legislature yields the lowest increase among the other components. Respondents reveal 0.15 unit of increase with their confidence with the country's Parliament and 0.17 unit of increase of their confidence with the legal system. The outcomes of legislature confidence slightly higher than the taxpayers' satisfaction with 0.11 unit of increase with Parliament and 0.12 unit of increase with the legal system. Respondents' trust with Education service has a higher increase compared with other indicators of the government's services. Taxpayers overall increase of 0.22 unit of confidence and 0.18 unit of satisfaction with the country's Education service. Taxpayers' trust in government also significant and positive, with 0.18 unit of increase with their confidence and 0.14 unit of increase with their satisfaction.
