Determinants of Tax Non- Compliance Behavior in Ethiopia: The Case of Hosanna Town Taxpayers

In the Ethiopian government the second five year Growth and Transformation Plan, it has been clearly stated that efforts will be geared towards promoting compliance and equipping tax collection institutions with adequate enforcement power which will further boost revenue mobilization at federal, regional, Zonal and woreds levels. The country’s tax mobilization was also the lowest among sub-Saharan African countries and thus, objectives if the study the identifying the factors that determine tax noncompliance behavior has been open for empirical investigation. There are seven determinates included in this study, to be researched upon and are categorized into non-economic factors and economic factors. Authors used a cross-sectional survey method of research design. The target population of the study was category “C” tax payers in purposefully selected three category taxpayers of Hosanna town, Ethiopia ,this group are considered hard to tax because the law doesn’t require them to declare their income or keep books of account.. A sample of 247 Category “C” tax payers was randomly taken from the selected cities. The number of questionnaires completed and returned was two hundred and twenty-eight (247), constituting 76% response rate. Both primary and secondary data were collected. Seven explanatory variables were used in the ordered logistic regression model. The p-value was used to test the significant at 5 % significance level of the study variables. The results showed that rate of tax, future tax costs, tax education/tax knowledge and Gross of sales are factors that positively significantly affect tax noncompliance behavior. However, Individual Tax Morale, report of income tax, and public quality governance were not significantly at 5 % significance level affect non-tax compliance. Therefore, researcher recommended, the outcomes of the study may inform policymakers about the determinants of non-tax compliance behavior of tax payers in the Hosanna town and helps to formulate better policy decisions related to nontax compliance behavior.

Economic factors are those variables that intrinsically found in affecting one's value, such as tax rate, income level and expectation on future tax cost that will be penalized if the non-compliant is discovered.
The level of public governance quality and tax morale of taxpayers, are those behavioral influences that fall under category of non-economic factors. Tax education may also need to be taken into the consideration as one of the noneconomic factors in perpetrating tax non-compliance.

ECONOMIC FACTORS H1: There is a positively corrected relationship between tax rate and level of tax noncompliance among Business income tax taxpayers in Ethiopia
This hypothesis is aimed to examine the relationship between tax rate and level of tax non-compliance among individual taxpayers in Malaysia. According to studies, high tax rates leads to higher level of tax non-compliance (Crane & Nourzad, 1987;Obid, 2004;Ho et al., 2006;Bayer, 2006;and Ahangar et al., 2011). An increase in tax rate definitely will reduce the usable income, taxpayers have to pay higher tax in relation to higher income, and therefore it can be more profitable if taxpayers choose to engage in tax non-compliance activities (Obid, 2004). H2: There is a negatively correlated with in level of tax non-compliance and income levels among business income taxpayers in Ethiopia. This hypothesis is developed to find out whether income level will affect tax noncompliance behavior among individual taxpayers. Insufficient fund is a strong stimulator in tax non-compliance (Ritsema et al., 2003). The higher the income level, the more ones opposed to tax non-compliance (Ross & McGee, 2011). This is because, if one has higher income, ones may have more respect to authority. Conversely, if one has lower income, then would have less respect to authority and eventually less averse to tax non-compliance. H3: There is a positively correlated relationship between future tax costs and level of tax non-compliance among individual taxpayers in Ethiopia. Based on economic model, it is a norm that, before taxpayers decided to engage in non-compliance activities, they will evaluate the benefits derived from noncompliance whether are greater than the fines or punishment when being caught later (Bărbuţă-Mişu, 2011). Accelerate in fines and increase in the frequency of tax audit will affect the tax compliance behavior.  unswervingly confirmed that future tax cost does positively influence the taxpayers' decision making on the tax non-compliance. This hypothesis is developed to find out the impact of future tax costs towards level of tax noncompliance among the business income taxpayers in Ethiopia.

NON-ECONOMICS FACTORS H2: There is a negatively correlated relationship between tax education and level of tax non-compliance among Business income taxpayers in Ethiopia.
Laws are complex and difficult to understand if without a proper learning platform, because of its jargons and terminologies (McCaffery & Barron, as cited in Bărbuţă-Mişu, 2011), which may lead to misinterpretation or misunderstanding in tax laws. With adequate tax knowledge, taxpayers are able to understand the tax system and tax policies more comprehensively (Kasipillai, as cited in Roshidi, et al, 2007), eventually increase their tax compliance behaviour and increase their awareness in fulfilling their duty as taxpayers (Kasipillai & Mustafa, 2000). This hypothesis is to find out whether tax education has significant impact on level of tax non-compliance among Business income tax taxpayers in Ethiopia. H3: There is a negatively correlated relationship between tax morale and level of tax non-compliance among Business income tax taxpayers in Ethiopia. Tax morale is an important factor in the study of tax compliance, as Alm and Torgler (2006) found that over the various factors influencing tax non-compliance, tax morale has occupied more than 20% of the total portion that can explain the tax non-compliance behavior. Thus, ceteris paribus, if tax morale is high, the tax non-compliance is expected to be low (Kirchgassner, 2010; Ahangar et al., 2011). This hypothesis is to find out the relationship between tax morale and tax noncompliance among the business income tax taxpayers in Ethiopia. H4: There is a negatively correlated significant relationship between public governance quality and level of tax non-compliance among Business income tax taxpayers in Ethiopia.
The public governance quality does matter in influencing tax compliance (Kim et al., 2006). This is because; the willingness of voluntary tax compliance is led by the trust in political leadership and a favorable government administration. Taxpayers will also take into consideration on how and what government spends on (Ho et al. The schematic presentation of the theoretical framework identified for this study purpose is presented below:

Statement of the Problem
Tax non-compliance is a growing international concern for tax authorities and public policy makers since it seriously threatens the capacity of governments to raise required public revenue. The problem is more serious in developing countries particularly in Sub-Saharan African countries (Cobham, 2005). Likewise, in Ethiopia it is evidenced by the current low tax-to-GDP ratio of 11% which is far lower than the average for developed tax systems (25-35%), developing countries (18-25%) and even of the Sub-Saharan average (16%). The main reason of this low revenue collection performance is due to tax non-compliance and due to poor tax administration (IMF, 2015). Tax audit is the common method used by tax authorities to detect tax evasion or tax non-compliance. There are various types of tax audits methods available and used by The Hosanna town Revenue authority branch (RABH). One of the methods is through tax field audit which is carried out at taxpayer's business premises to check and ensure that their recordings and reporting of income and expenses are adopting the correct tax and accounting treatments. A penalty will be imposed for incorrect tax return with discrepancy in tax liability a penalty of 5% of the amount of unpaid tax on the first day after the due date passed and an additional 2% of the amount of the tax that remains unpaid on the first day of each month thereafter (Misrak, 2014).  According to table 1 above the Hosanna town revenue authority branch office (HTRABO) revenues collected from different tax sources have been significantly decreasing and increasing in the past 4 years. This shows that there is a serious problem of revenue generation and the performance of tax collection accordingly the government suffers to finance the social and economic agendas of the Zone (town). Moreover, determinants of tax non-compliance in Hosanna were not planned and controlled category basis that is category A, B and C tax payer's basis. In order to generate sufficient tax revenue, increasing parallel to the government expenditure as well as to have sustainable economic development and to reduce the dependency of the region, there should be corrective actions on the gap of tax administration systems. Specifically, the determinant of tax noncompliance in Hosanna on the category 'C' taxpayers needs a special attention. To improve the tax revenue, the government should work with category 'C' taxpayers. Therefore, examining the determinants of tax noncompliance at tax offices and taxpayers helps to take corrective actions so as to improving the tax revenue. Thus, this study intended to examine determinants of tax non-compliance in Hosanna of category "C" taxpayers in Hosanna town.
As to the researchers' knowledge, factors that determine tax non-compliance behavior in Self-Assessment System is not well understood and studies have not been carried out in Hosanna town, Ethiopia and some of related studies also did not incorporate different determinants of tax non -compliance. In addition, while reviewing the above related literatures, they found inconclusive results on tax non-compliance determinants. Hence, the above research gap motivated researchers to examine the factors that influence non tax-compliance behavior of tax payers in SAS by taking evidence from SNNPRS, Ethiopia.

Objectives of the Study
The main objective of the study was to identify the factors affecting taxpayers towards tax non-compliance in Hosanna Town. Specifically, this study has the following specific objectives; 1. To study the relationship between economic factors and tax non-compliance.

2.
To examine the association between non -economics factors and tax non-compliance. 3. To find out the possible ways in combating tax non-compliance among Hosanna Town business taxpayers.

Hypotheses
In order to achieve the objectives of the study; the determinants of tax non-compliance and how these determinants relate with tax non-compliance, the following literature driven hypotheses have been tested: H1: Tax rate is positively correlated with tax non-compliance. H2: Income level (sale) is negatively correlated with tax non-compliance. H3: future tax costa is positively correlated with tax non-compliance. H4: Tax education/tax knowledge is negatively correlated with tax non-compliance. H5: Tax morale is negatively correlated with tax non-compliance. H6: Public governance quality is negatively correlated with tax non-compliance.

Research Design and Methodology 6.1. Method
In order to have a better understanding of the situation and gather pertinent data, the researchers used both primary and secondary sources of data. To gather relevant primary data a survey method with self-structured questionnaire was used. A survey method of data collection was employed. A structured questionnaire was distributed to 247 Category "C" taxpayers in three sub-cities of Hosanna, namely Sechi-duna, Addise Ketema and Gofare-meda. The study was carried out on Category "C" taxpaying business enterprises, whose annual turnover is not more than 500,000 birr. The motivation of considering Category "C" is that taxpayers in this group are considered hard to tax group because the law doesn't require them to declare their income or keep books of account. Therefore, it can be said that they are non-compliant with the tax system owing to the absence of documenting their inventories and disclosure of their earnings. The questionnaires were adopted and developed with some modification from previous similar studies such as 15 respondents to answer question in a simple way. The questionnaire was also prepared in both English language and Amharic language to help reduce the impact of Language barriers.

Target Population
The total population of this research is Category "C" tax payers (small and micro sizes taxpayers) in Hosanna cities in SNNPRS, Ethiopia. According to Income tax proclamation number 282/2006/amended by proclamation No.979/2016 Category "C" tax payer is a business having an annual turnover of Birr 500,000 (Five hundred thousand Birr) or less. The motivation of considering Category "C" tax payers is that taxpayers in this group are considered as they are don't fully engaged in self-assessment system because the law not require them to declare their income or keep books of account in Ethiopian tax system.

Model
Tax non-compliance is measured through four items: I would pay taxes if my income is higher, I fully declare my principal income, but not including my part-time income, If the job is paid in cash basis, then it is alright not to report it in tax return and I would understate income (employment income, rental income and etc.) if the amount is relatively small..
For instance, acceptability of tax evasion can be measured by: "I wouldn't feel bad if I don't pay tax" (1 = completely agree, 5 = completely disagree). The average score over the five items will be taken as an index for tax compliance. Based on this score, taxpayers were categorized into three levels of compliance: low, medium, high. Given the scaled ranking information of the dependent variable, ordered logistic estimation is applied. The ordered logistics have the following form: Yi * =β0+ X1iβ1+ X2iβ2+ X3iβ3………..+ Xkiβk+€i Where y * i is the dependent variable (levels of compliance); β is the vector of estimated parameters and x i is the vector of explanatory variables; € i is the error term, which is assumed to be normally distributed (zero mean and unit variance).
Accordingly, the probability of each tax compliance level (low-y1, medium-y2, and high-y3) will be computed as follows: Yi=1 if yi* ≤ u1 if the average value is less than equal to 2 Yi=2 if u1< yi* ≤ u2, if the average value is greater than two and less than equal to 3 Yi=3 if yi* > u2, if the average value is greater than 3

Results
Seven variables were tested using ordered logistic regression, namely tax rate, income level, future tax costs, report income tax, public governance quality, tax education and tax morale. (Income) -sale. All the variables in the regression model are relevant to the study since the VIF factors are all below the benchmark of 10. This indicates the absence of multi-collinarity in the model. The likelihood ratio chisquare of 269.91 with a p-value of 0.0000 tells us that the model as a whole is statistically significant.
The finding imply that the significant factors that affecting the tax non-compliance in Hosanna town at the time of this study include the income tax rate (positive), tax knowledge(tax education) (negative), future tax cost (positive) and Gross sales( positive ) refer to table 2.
The result also suggests that other variables such as reporting income tax, individual tax morale and public governance quality were are not significant affect corrected with tax non-compliance at the time of study.
 Accordingly, the result of this study, tax knowledge (tax education) (β =-2.652519) has negative and significant relationship with tax non-compliance. Hence, hypothesis H1 is accepted. This result provides evidence that low tax knowledge would decrease tax non-compliance. This result is also consistent with literature review issue with adequate tax knowledge, taxpayers are able to understand the tax system and tax policies more comprehensively (Kasipillai, as cited in Roshidi, et al, 2007), eventually increase their tax compliance behaviour and increase their awareness in fulfilling their duty as taxpayers (Kasipillai & Mustafa, 2000). As far as high tax knowledge of taxpayers could discourage tax non-compliance, it is advocated that the tax authority should increase their providing more tax knowledge to a larger number of tax payers helps to prevent tax evasion with the aim of increasing tax non-compliance, decreasing the tax gap and achieving the missions of Hosanna town revenue.  With regarding to income tax rates of, the result of this study shows positively and significant relationship between income tax rates (β=0.3207795) and tax non-compliance. Previous studies, for example (Crane & Nourzad, 1987;Obid, 2004;Ho et al., 2006;Bayer, 2006;and Ahangar et al., 2011) have found that a high tax rates leads to higher level of tax non-compliance. An increase in tax rate definitely will reduce the usable income, taxpayers have to pay higher tax in relation to higher income, and therefore it can be more profitable if taxpayers choose to engage in tax non-compliance activities (Obid, 2004). As far as high income tax rates could encourage tax non-compliance, it is advocated that revenue authority should not increase the income tax rates with the aim of decreasing, the tax gap and achieving the missions of Hosanna town revenue authority.  Regarding to future tax costs and tax non-compliance this study was found positively (β=0.3250051) and significant relationship. Thus, hypothesis is H3, is accepted. This result is also consistent with finding of Hai and See (2011) unswervingly confirmed that future tax cost does positively influence the taxpayers' decision making on the tax non-compliance.  The finding of this study with regard to income level of taxpayers (β =.748063) has positive and significant relationship with tax non-compliance. Thus, hypothesis H2 is rejected. For example pas study Jackson and Milliron (1986) found that income level has a mixed and unclear impact on compliance but progressive tax rates might encourage the higher income group to evade rather than the lower income group because their (higher income group) tax rates and taxable income are high, thus, making the tax liabilities much higher than lower income group. Loo (2006) found that high income earners in Malaysia are prone to evading tax while Torgler (2007) reported that lower income earners in Western Germany were less compliant. According to this study found that high income level in Hosanna town to evade tax liabilities rather than the lower income level because their (high income level) tax rates and taxable income are high, thus making the tax liabilities much higher than lower income group. In this study, other variables such as the public governance quality, reporting income tax and individual tax morale were not significantly correlated with tax non-compliance decisions, even though previous studies in other countries found significant associations. For example, the role of public governance quality is minimized the tax gap and decreasing tax non-compliance was found to be important as  Possible Ways in Combating Tax Non-compliance in Hosanna town According to the respondents are required to rate to what extent they think the given solutions would be most effectively in deterring and combating tax non-compliance all ten possible methods in combating tax noncompliance are ranked accordingly based on responses from participants. The top five of most selected methods are: A. Government should enhance the fairness within tax regime by reducing tax rate and tax pressure towards lower income earners, for example, by increasing income tax bands; B. Increase tax enforcement such as tax audit; and also, C. Government should put more efforts in on-going anti-corruption campaigns. D. Stiff punishment towards tax evader, like imprisonment, increase fine rate, and so on; E. Increase public governance quality by increasing transparency and provide information on how tax revenue is being utilized;

SUMMARY AND CONCLUDING REMARKS
The determinants of issues of tax non-compliance are large in number, but this study was focused on two factors such as Economics factors as well as non-economics factors. The primary objectives of this were examining the determinants of tax non-compliance in Hosanna town, Ethiopia. Based on a survey conducted in Hosanna town using 247 respondents, an attempt was made to explain effects of non-compliance determinates. According to Results of the Ordered logistic regression analysis (seeTable2) suggested that tax non-compliance is significantly influenced by tax knowledge (tax education), income tax rates, future tax costs and income level/sales. This study also evidenced that, other variables such as reporting income tax, individual tax morale and public governance quality, were not significant determinants of tax non-compliance behavior in Hosanna town "C" taxpayers.
The results of this study also provide implications In order to effectively combat the issue of tax noncompliance, there must be cooperation between three parties, which includes government, tax authority and taxpayers. A triangular relationship must be formed intimately. the government that specific insights should allow policy makers to gain a better understanding of the key variables that are significantly associated with tax non-compliance and enable them to implement suitable strategies to minimize potentially damaging factors, and should also allow them to improve a government's tax revenue collections.
Further, it is recommended that this kind of study should be conducted at the national level to gain a better understanding on non-compliance determinates in the country.

Direction for Further Research
There are some limitations of this study. Firstly, it considers only Category "C" tax payers in one selected Town in South, Ethiopia. Therefore, the findings of this study may be difficult to generalize about all tax payers in Ethiopia. Hence, this study can be improved if it is done at national level or by comparing tax non-compliance behavior in different business sectors by considering other tax categories. Further researchers can include other tax non-compliance determinant variables that were not included in this study. The factors which included in