TAX COMPLIANCE COST BURDEN AND TAX PERCEPTIONS SURVEY IN ETHIOPIA MAY 2015 TAX COMPLIANCE COST BURDEN AND TAX PERCEPTIONS SURVEY IN ETHIOPIA MARCH 2016 Disclaimer The organizations (that is, IBRD, IFC, and MIGA), using their best efforts in the time available, have endeavored to provide high-quality services hereunder and have relied on information provided to them by a wide range of other sources. However, they do not make any representations or warranties regarding the completeness or accuracy of the information included in this report, or the results that would be achieved by following its recommendations. About the World Bank Group’s Trade and Competitiveness Global Practice (T&C GP) The Trade & Competitiveness Global Practice (T&C GP) provides policy advice and lending support to help client countries increase trade and investment, improve productivity and competitiveness at the national and industry levels, and create an inclusive, competitive private sector. The T&C GP works with governments to identify policies that promote growth, while helping identify and remove impediments to the smooth functioning of markets (such as gaps in coordination, undersupply of public goods, non-competitive market structures, and regulatory constraints). The Ethiopia Investment Climate Program, managed by the World Bank Group’s Trade and Competitiveness Global Practice, aims at streamlining and simplifying high priority regulatory practices and processes burdensome to the private sector and address investment climate issues that are holding back investment and productivity growth in Ethiopia. Acknowledgments The WBG would like to thank the Ethiopian Revenue and Customs Authority for their support in conducting the tax compliance cost burden and tax perceptions survey in Ethiopia. The survey was conducted by Department of Economics of the Addis Ababa University. The survey is conducted as part of the Tax Administration Component of the WBG Ethiopia Investment Climate Program. Mamo Mihretu, Program Manager at WBG, coordinated the overall work and program. This report is prepared based on the results of the survey. Contributors to the report include Wollela Abehodie Yesegat (Consultant), Denis Vorontsov (Consultant), Jacqueline Coolidge (Consultant), and Laurent Olivier Corthay (Senior Private Sector Specialist). We gratefully acknowledge the financial contributions made by Department of Foreign Affairs, Trade and Development (DFATD) of Canada, Sweden International Development Agency (SIDA), Department for International Development (DFID) and the Italian Cooperation. Contents Executive Summary vii Background xiii 1  Overview of Ethiopian Tax System and Recent Reform Initiatives 1 2  Objectives and Methods Adopted 5 1.1 Survey: Sampling Design 5 1.2 Survey: Instrument and Actual Conduct of the Surveys 7 3  Survey Results 9 3.1 Profile of Respondents 9 3.2 Tax Compliance Costs 14 3.3 Perception of Taxpayers to Aspects of the Tax System and the Business Environment 22 3.3.1 Views about Registering for Taxes 22 3.3.2 Perception of Sales Register Machine (SRM) Usage 26 3.3.3 Perception of VAT Registration and Refund 26 3.3.4 Perception Towards Tax Inspections, Penalties and Appeals 28 3.3.5 Perceptions of Tax Compliance and the Business Environment 31 3.3.6 Difficulties with the Tax System 35 3.3.7 Computer and Bank Account Usage and Other Issues 36 4  Conclusions and Recommendations 41 References 47 Annex 1: Tax Compliance Costs—International Comparisons for Ethiopia 49 Annex 2: Examples of Thresholds Applied to Reduce the Compliance Burden of Small Businesses 61   iii Figures Figure E1: Bookkeeping Practices of Businesses—Formal viii Figure E2: Bookkeeping Practices (percent)—Informal viii Figure E3: Businesses Involved in Outsourcing and In-House Activities for Bookkeeping and Tax Accounting (percent) viii Figure E4: Distribution of Full Outsourcing Cost of Tax Compliance by Tax Type (in percent) (N 5 171) ix Figure E5: Average Total Compliance Costs (in ETB) ix Figure E6: International Comparison of Tax Compliance Costs x Figure E7: Perception about Tax Inspections (N 5 1003) xi Figure 2.1: Sectoral Distribution of Informal Businesses (percent) (N 5 499) 7 Figure 3.1: Position of Respondents—Formal Businesses (in percent, N 5 1003) 9 Figure 3.2: Gender of Respondents—Formal (in percent, N 5 1003) 10 Figure 3.3: Gender of Respondents—Informal Businesses (N 5 499) 10 Figure 3.4: Gender of Owners—Formal (in percent, N 5 1003) 10 Figure 3.5: Period of Establishment 11 Figure 3.6: Working Premises of the Sample Firms (N 5 499) 11 Figure 3.7: Number of Employees—Informal (N 5 499) 12 Figure 3.8: Bookkeeping Practices of Businesses—Formal 12 Figure 3.9: Bookkeeping Practice of Businesses—Formal (by category) 13 Figure 3.10: Bookkeeping Practices (percent)—Informal 13 Figure 3.11: Bookkeeping Practices by Location (percent, yes)—Informal 13 Figure 3.12: Average Total Compliance Costs (in ETB) 15 Figure 3.13: Tax Compliance Cost as a Share of Turnover by Turnover Band 16 Figure 3.14: International Comparison of Tax Compliance Costs 17 Figure 3.15: Businesses Involved in Outsourcing and In-house Activities for Bookkeeping and Tax Accounting (percent) 18 Figure 3.16: Distribution of Full Outsourcing Cost between General Bookkeeping and Tax Compliance (in percent) (N 5 174) 19 Figure 3.17: Average Cost of Full Outsourcing for Businesses That Outsource (in ETB) (N 5 174) 19 Figure 3.18: Distribution of Full Outsourcing Cost of Tax Compliance by Tax Type (in percent) (N 5 171) 19 Figure 3.19: Average Cost of (full) Outsourcing for Businesses That Outsource by Tax Type in ETB (N 5 171) 19 Figure 3.20: Person-Days Spent on In-house Bookkeeping and Tax Compliance (N 5 627) 20 Figure 3.21: Distribution of Time Spent on In-house Tax Accounting by Tax Type (N 5 592) 20 iv  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.22: Person-Days Spent on Tax Compliance by Tax Activity (N 5 627) 20 Figure 3.23: In-house Compliance Cost of Time Spent on General Bookkeeping and Tax Accounting in ETB (N 5 537) 21 Figure 3.24: Opinion of Respondent Firms about the Biggest and Second Biggest Disadvantages of Registering for Taxes (N 5 1003) 23 Figure 3.25: Opinion of Respondent Firms about the Biggest and Second Biggest Disadvantages of Registering for Taxes (by tax payers’ category) (percent) 24 Figure 3.26: Reasons for Giving Up or Never Considered Nor Applied (N 5 498) 24 Figure 3.27: Most Important Factors That Encourage or Persuade Business to Consider Registering for Tax (N 5 499) 25 Figure 3.28: Perception on the Likelihood of Registering for Tax (N 5 499) 25 Figure 3.29: General Perception on First Biggest Disadvantages of Registering for Tax (N 5 499) 25 Figure 3.30: Biggest Advantages of Using Sales Register Machine (percent) (N 5 1003) 26 Figure 3.31: Biggest Disadvantages of a Sales Register Machine (percent) (N51003) 26 Figure 3.32: First and Second Biggest Advantages of Registering for VAT (percent) (N 5 1003) 27 Figure 3.33: First and Second Disadvantages of Registering for VAT (N 5 1003) 27 Figure 3.34: Eligibility for VAT Refund (percent) 27 Figure 3.35: Application for VAT Refunds (percent) 27 Figure 3.36: Reasons for Not Applying for the VAT Refund (N 5 82) 28 Figure 3.37: Time Taken for VAT Refund Process (N 5 75) 28 Figure 3.38: Perception about Tax Inspections (N 5 1002) 29 Figure 3.39 Inspections and Audits by Turnover 29 Figure 3.40: Number of Inspections and Audits and Time Spent Among Those Who Reported any Inspection 29 Figure 3.41: Whether or Not Appeals Were Filed (N 5 146) 30 Figure 3.42: Perception about Statements Related to Tax Appeals (N 5 68) 30 Figure 3.43: Opinions on Tax Compliance (N 5 1003) 31 Figure 3.44: Respondents Perception about Taxes—Formal (percent) 32 Figure 3.45: Perceptions about Taxes and Tax System—Informal (N 5 499) 33 Figure 3.46: Respondents Perception about Business Environment—Formal (percent, N 5 1003) 34 Figure 3.47: Perception on Elements of Business Environment (N 5 499)—Informal 34 Figure 3.48: Perceptions about Calculating and Filling in Tax Returns 35 Figure 3.49: Perceptions about Submission of Tax Returns and Payments 35 Figure 3.50: Perception about Filling Out and Submitting Tax Returns—Informal 36 Contents  v Figure 3.51: Percentage of Businesses Who Stated Availability of Information Source on Taxation 36 Figure 3.52: Firms’ Use of Computers (percent) 37 Figure 3.53: Firms’ Use of Internet 37 Figure 3.54: Firms’ Use of Bank Account 37 Figure 3.55: Use of Bank Account for Business Operations (by legal status) 38 Figure 3.56: Use of Bank Account (informal) (percent) (N 5 499) 38 Figure 3.57: Knowledge about Tax Information and Requirements by Location—Informal 39 Figure 3.58: Perception about Keeping Accounting Records and Books for the Businesses—Informal 39 Tables Table 2.1: Distribution of Sample Firms—Formal (by region and category) 6 Table 2.2: Sectoral Distribution of Businesses—Formal (by taxpayers’ category) 6 Table 2.3: Business Category by Turnover (percent) 6 Table 3.1: Gender of Respondents—Formal (by taxpayers’ category) (in percent) 10 Table 3.2: Number of Workers in Sampled Businesses—Formal (total and by categories) 11 Table 3.3: Types of Taxes Paid by Category and Ownership Type (percent) 14 Table3.4: Average Tax Compliance Costs (in ETB and as a share of turnover) by Taxpayers’ Category 16 Table 3.5: Tax Compliance Costs (in ETB and as a share of turnover) by Taxpayers’ Ownership Type and Sector 17 Table 3.6: Time Spent on Bookkeeping and Tax Accounting by Ownership Type and Turnover 21 Table 3.7: In-house Compliance Cost of Time Spent on General Bookkeeping and Tax Accounting in ETB (by ownership type and turnover) 22 Table 3.8: In-house Compliance Cost of Time Spent on General Bookkeeping and Tax Accounting in ETB by Category 22 Table 3.9: General Opinion on the Reasons for Not Applying for Tax Registration (among those who have considered but did not apply) (N 5 166) (percent)—Informal 23 Table 3.10: Reasons for Not Filing an Objection (N 5 53) (percent) 30 Table 3.11: Appeals Resolved and Still in Process (N 5 68) (percent) 30 Table 3.12: Percent of Total Annual Sales Declared for Tax Purposes by Turnover (among those reporting less than 100%) 31 Table 3.13: Percent of Actual Expenditure Overstated in Tax Returns by Turnover 31 Table 3.14: Business-Related Visits to Tax Offices—Formal 39 vi  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Executive Summary This study attempts to estimate tax compliance costs unlikely to register for tax in the next two years was and assess views of taxpayers on aspects of the tax about 26 percent. system in Ethiopia. The study uses evidence mainly About 63 percent informal business operators from a survey of both formal and informal businesses in claimed to have neither considered nor applied to Addis Ababa and four major cities (Adama, Hawassa, register for tax by the time of the survey, although Mekele and Bahir Dar) in the four largest regional it varies across locations. Some of the reasons for states. The survey covered 1003 formal businesses not applying for registration included perceptions (based on a representative sample of business taxpayers about the financial burden of being formal (including from ERCA database) and 499 informal businesses taxes, fees) (27 percent), and the concern that tax (using area-based sampling) in the above mentioned laws and regulations are not clear, fair or appropriate locations. Survey questionnaires were informed by the (15 percent). results of four focus group discussions conducted in Addis Ababa and Adama. About half of formal businesses in the survey were keeping all receipts and invoices in an organized Most of the firms in the sample are small firms: More manner: although there are substantial differences than 82 percent of the firms are small firms belonging across the three categories: almost all category A to category C and they are mainly sole-proprietorship taxpayers do this but only 40 percent of category C (93 percent) form of establishments. taxpayers. Overall, about 42 percent maintain full The largest proportion of businesses (formal and records of revenues and expenses (not necessarily using informal) operate in service and trade sector: about computer). A slightly larger percentage of businesses 51 percent and 40 percent of businesses (formal) were (about 52 percent) do bookkeeping for revenue but found to be engaged in the service and trade (both not necessarily for expenses. Only about 3 percent of wholesale and retail trade) sectors. Similarly, about informal businesses keep receipts, although about 87 percent of informal businesses that participated in 16–17 percent report that they keep records of sales the survey reported that their primary activity is services and/or purchases (Figures E1 and E2). or trade. A larger proportion of businesses submit business profit Most businesses that participated in the survey obtained tax return: close to 60 per cent of formal business tax identification number (TIN) and registered for VAT respondents reported that they submitted business (if required) within the period 2002–2004 EC. During profit tax return while 31 percent and 12 percent this period, more than 64 percent of businesses that were found to have submitted turnover tax and VAT participated in the survey obtained TIN; and about returns respectively. Withholding tax on payments and 60 percent of the VAT payers (that participated in the employment related taxes and contributions survey) registered for VAT. were primarily paid by category A and category B taxpayers. About half of informal business survey respondents reported that they are likely or very likely to register A large share of businesses in the survey perform for tax in the coming two years. The proportion of bookkeeping activities in-house: in particular, there respondents who said that they are unlikely or very is limited outsourcing of bookkeeping activities of Executive Summary    vii Figure E1: Bookkeeping Practices of Businesses—Formal Using a specialized accounting or bookkeeping software (N = 86) 66.4 33.6 Using a computer for purposes of bookkeeping (N = 1003) 8.6 91.4 Doing full records of revenues and expenses (not necessarily using) (N = 1003) 42.3 57.7 Bookkeeping for revenue (not necessarily for expenses) (N = 1003) 51.8 48.2 Keeping all receipts and invoices in an organized manner (N = 1003) 49.5 50.5 0 20 40 60 80 100 120 Yes No Figure E2: Bookkeeping Practices (percent)—Informal Using a specialized accounting or bookkeeping software 100 Using a computer for purposes of bookkeeping (N = 499) 100 Doing full financial accounting (not necessarily using a computer) 96.4 3.6 Keeping records of purchases (N = 499) 83 17 Keeping records of sales (N = 499) 84.4 15.6 Keeping all physical receipts and invoices (N = 499) 96.8 3.2 0 20 40 60 80 100 120 Yes No businesses, with about 85 percent of them doing it consider VAT as being the most complex and often completely in-house. requiring external support (Figure E3). Close to half of businesses that file VAT returns The cost of outsourcing depends on tax type: half of outsource their VAT compliance activities (completely the cost of outsourcing tax compliance activities is for or partially): about a third of businesses that file for tasks related to business profit tax (50 percent) which is business profit tax outsource profit tax activities followed by value added tax (20 percent) and turnover completely or partially (with similar figures for tax (18 percent) (Figure E4). withholding income tax on payments and employment Businesses that outsource business profit tax compliance related taxes and contributions). Only 15 percent of activities incurred an average of ETB 2,267 per business ToT payers outsource. This suggests that businesses Figure E3: Businesses Involved in Outsourcing and In-House Activities for Bookkeeping and Tax Accounting (percent) General bookkeeping (N = 761) 84.8 7.6 7.6 Business profit tax (N = 692) 68.4 12.5 19.1 Value added tax (N = 409) 50.5 24.2 25.3 Turnover tax (N = 267) 85.1 7.7 7.2 Withholding income tax on payments (N = 230) 69.2 21.4 9.4 (Employment related contributions (N = 345) 70.7 16.3 13.0 Other taxes (N = 99) 99.0 0.6, 0.4 0 20 40 60 80 100 120 Completely In-house Partially In-house Completely Outsourced viii  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure E4: Distribution of Full Outsourcing Cost of Total tax compliance cost, consisting of outsourcing Tax Compliance by Tax Type (in percent) (N 5 171) and in-house costs, is high in Ethiopia: in the year 8% 1% 2012/13, the average cost to a business for general 3% bookkeeping was found to be ETB 9,804 (USD 523.2). Business Profit Tax In the same year, average total tax compliance cost of a Value Added Tax business including costs of acquisition and maintenance 18% Turnover Tax 50% Withholding Income of software and hardware was ETB 7,609 (USD Tax on Payments 406) while average total tax compliance cost without Employment Related acquisition and maintenance costs was ETB 5,842 20% Contributions (USD 311.7) (Figure E5). Other Taxes Tax compliance costs varies with tax type: business profit tax takes the largest share of total tax compliance in the 2012/13 fiscal year, which is the highest for a costs (both internal and external costs), followed by tax type. Value added tax compliance is the second value added tax and turnover tax. The burden of highest in terms of average outsourcing cost by tax type compliance costs of business profit tax and value added amounting to ETB 913. tax shows that these two taxes do have complexities The number of person-days spent increases as turnover that resulted in taxpayers incurring high costs of increases: on average, a business spends about compliance in terms of staff time and professional fees 106 person-days of its employees’ time for in-house paid to external assistance. bookkeeping and tax compliance. General bookkeeping Tax compliance cost varies with the size of businesses: accounted for about 74 percent (78 person-days), while the average tax compliance cost of a business (with tax compliance accounted for 26 percent of the total acquisition and maintenance costs of software and time spent (i.e., about 28 person-days). Large businesses hardware) as a share of turnover is about 5.4 percent spend more time on bookkeeping and tax accounting while the share of tax compliance cost in total related activities than smaller businesses. turnover without acquisition and maintenance costs Time spent on in-house tax compliance activities varies is 4.7 percent. Small businesses face a relatively larger with tax types: half of the total time spent by a business tax compliance cost as a share of turnover compared on tax accounting goes to business profit tax while with large businesses. Tax compliance cost as a share of another 31 percent goes to turnover tax. Value added turnover tends to decrease as the turnover of businesses tax takes 8 percent of the time spent. Figure E5: Average Total Compliance Costs (in ETB) 12000 9804 10000 7609 8000 5842 6000 4000 2000 0 General Bookkeeping Tax Accounting (incl. acquisition Tax Accounting (excl. acquisition and maintenance cost) and maintenance cost) Executive Summary    ix Figure E6: International Comparison of Tax Compliance Costs 14 Tax Compliance Costs as a Percentage of Turnover 12 10 8 6 4 2 0 0–5 5–13 13–24 24–41 41–66 66–104 104–161 161–246 246–374 374–567 567–855 855–1287 1287–1936 1936–2909 2909–4369 4369–6558 6558–9843 Turnover in '000 USD South Africa Ukraine Uzbekistan Kenya Georgia Ethiopia Nepal increases suggesting that it is regressive. Similarly, The process of complying with taxation is perceived the relative tax compliance cost (excluding cost of to be more burdensome than the amount of tax itself: acquisition and maintenance) as a share of turnover about 54 percent of businesses agree to the statement is larger for category C (5.03 percent) compared with that the process of tax compliance is more burdensome category A (3.85 percent). In absolute terms, however, than the actual tax due. tax compliance cost for category A businesses is well Avoiding government retribution (formal and informal over 5 times the cost for category C businesses (both businesses) and better access to land and premises when tax accounting cost includes and excludes (informal) are perceived to be the biggest advantages of acquisition and maintenance costs). Overall, the total registering for taxes: relatively larger percentage of both tax compliance cost in Ethiopia is estimated to be about formal and informal businesses in the sample believes that ETB 5.8 billion (about USD 309.5 million)1 and ETB avoiding government retribution is the biggest advantage 7.5 billion (USD 400.5) (depending on whether tax of being a registered taxpayer (about 23 percent each). In compliance cost includes acquisition and maintenance addition about 22 percent of informal businesses perceive cost of software and hardware). This is between 4.5 and better access to land and premises as the first biggest 5.8 percent of Ethiopia’s total government revenue advantage of registering for tax. collected in 2012/13 and about 1 percent of its GDP during the same year. Businesses perceive high tax burden or high tax rates as the biggest disadvantage of registering for taxes: Tax compliance cost in Ethiopia is high in comparison businesses perceive higher tax burden or high tax rates to the countries considered in this study (Figure E6), (34 percent), complicated tax compliance procedures since most of the countries that have had such a survey (13 percent), harassment by government officials were the ones already known to have a problem with (8 percent), and frequent inspections (8 percent) as tax compliance costs. some of the disadvantages of registering for taxes. 1 Assuming average exchange rate for the fiscal year Businesses perceive some advantages of registering 2012/13, USD 1 5 ETB 18.59. for VAT: cash flow benefits and participating in x  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure E7: Perception about Tax Inspections (N 5 1003) Other reasons (N = 1002) 0.34 Inspections are used as an additional revenue collection mechanism for the government (N = 1002) 16.88 Because of someone’s interest to create obstacles 9.36 to others (e.g. to competitors) (N = 1002) Inspection officers are trying to justify their jobs 19.17 (N = 1002) Inspection officers are looking for 19.15 informal payments (N = 1002) Legitimate reasons, i.e., inspections are really 36.1 necessary to deter tax evasion (N = 1002) 0 5 10 15 20 25 30 35 40 government tenders have been considered as the biggest reasons. This is lower compared to the perception advantages of registering for VAT. on the same issue in other countries such as Georgia. About 19 percent of tax inspections and tax audits Businesses perceive some disadvantages of registering were also perceived to be happening because inspection for VAT. Among the various disadvantages for officers were looking for informal payments, and registering for VAT, higher prices of goods and services another 19 percent because it was in someone’s interest compared to those by non-VAT registered businesses to create obstacles to others (Figure E7). considered the first and second biggest disadvantages for registering for VAT (42 percent). Simplify the tax regime for micro and small enterprises: to reduce the tax compliance burden on smaller Businesses also perceive multiple advantages and businesses as is evident from this study and from the disadvantages of sales register machine (SRM) review of the standard assessment system and also usage: advantages include less opportunity for theft encourage business formalization simplification of the (20 percent), updated and easily available sales tax regime for this segment is worth considering. information (22 percent), and easy to comply with tax requirements (17 percent). In terms of disadvantages, Offer a fixed tax and turnover based tax to respectively difficulty of correcting errors (36 percent) and cost micro and small enterprises: in between poverty—level of the machine (19 percent) were perceived to be micro –enterprises paying a fixed tax and medium-large disadvantages of using SRM. businesses paying regular profit tax and VAT, it is often helpful to offer an intermediate regime for “small” Businesses face frequent inspections/audits, although businesses that can pay a turnover tax in lieu of both larger businesses experienced relatively more profit tax and VAT (with an option for any business to inspections, compared with smaller businesses: register for regular profit tax and VAT Voluntarily). In the survey indicates that around 31.7 percent of addition, adjusting business categorization threshold for respondents reported that their businesses had faced inflation is worth considering. some kind of inspection or audit by tax authorities in the last two tax years. About 48 percent and 46 percent Review the income tax rate structure and exempt of category A and B businesses experienced inspections income threshold: reviewing the income tax rate or audits. structure (along with the tax exempt threshold) in addition to simplification of the entire tax system is Businesses perceive only a relatively a small proportion important in reducing the burden of taxes and changing of inspections and audits as happening for legitimate the attitudes of taxpayers and also encouraging reasons: businesses perceived that only about 36 percent formalization of businesses. The income tax exempt of tax inspection and audits happened for legitimate threshold should be set a approximately the poverty Executive Summary    xi level, and adjusted regularly over time to account for Investigate the entire dispute resolution system: it inflation and other relevant factors. is important to investigate why a large number of appeal decisions were not in favor of the tax authority Adjust the VAT registration threshold periodically for and what causes them. It is in general important to inflation: the VAT registration threshold should be investigate the entire dispute resolution system and adjusted periodically for inflation and is usually set at address the issues in the tax system. the lower turnover level for a “medium” size business (taking into account also the capacity for the necessary Improving the risk based audit system and hold bookkeeping for VAT compliance. The current discussions with businesses: to mitigate the compliance threshold is quite low by international standards. burden (pertaining to audit) on businesses especially smaller ones and also increase taxpayers’ perception ERCA should try to expedite payment of refund to that tax audit is for legitimate reasons, it is important taxpayers to help businesses avoid unnecessary cash to consider improving the risk based audit system flow difficulties. In particular adopting of risk based in addition to developing the culture of holding refund processing and also accepting bank bonds as a discussions with taxpayers. security for VAT payment (risky businesses) could be considered as suggested at the Ethiopian Chamber of Exercise caution in imposing business profit tax based commerce and Sectoral Association’s forum. on bank deposits: it is important to exercise caution by the tax administration in imposing income tax on bank Reduce the frequency of VAT filing: the compliance deposits as that would increase taxpayers’ mistrust cost is high for VAT, it is recommended to reduce the on the tax authority and reduce the incentive to use frequency of VAT filing (e.g., make it quarterly), for banking in business operations. relatively smaller businesses. xii  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Background The extent of tax burden is one of the main factors that revenue as a share of GDP is expected to increase from the private sector considers when deciding to create, 11.3 percent in 2010/11 to 15 percent in 2014/15. move or expand its businesses as high tax burden is a Accordingly, the success in increasing tax revenues, disincentive to private sector development. It should be which is of critical importance for the success of other noted that both the burden of tax payments and of tax components of the GTP, depends for most part on the compliance, which includes the time and cost associated performance of the Ethiopian Revenue and Customs with preparing tax returns, filing, effecting payment Authority (ERCA) whose interaction with the taxpayers and interacting with tax authorities, are likely to deter plays a significant role. business formalization. Evidence from a number of The actual tax revenue as a share of GDP ( 13 surveys (Coolidge, 2012) indicates that the burden of percent) is low compared to that of the average for tax compliance can often be heavier than the actual Sub-Saharan African and low income countries. The tax payments, especially for small businesses (e.g., share of tax revenue in GDP hardly changed, suggesting equivalent to a turnover tax of 5% or more). that tax revenue (which should rise relative to GDP The Government of Ethiopia (GoE), in its five year due to progressive rate structure) has failed to grow as Growth and Transformation Plan (GTP) (2010/11– much as expected with the overall economic growth. 2014/15), indicates that the expansion and quality This situation in the revenue system might be due to of infrastructure investment and the development of such things as poor tax compliance behavior and weak small and micro enterprises (SMEs), which are the tax administration. In order to address issues of tax basis for the transition towards medium and large-scale compliance to some extent, reducing complexities in the manufacturing industries are, among others, crucial tax regime may be necessary. pillars for sustained growth. This implies two things: Complexity of tax regimes (e.g., multiple taxes, several first, enhancing the participation of the private sector in different bases, requirements for multiple filings per the development process; secondly, improving domestic year, etc.) increases tax compliance costs, especially in resource mobilization to finance massive investment developing countries. Findings from the Investment initiatives envisaged in the GTP (MoFED, 2010). Climate Department of the World Bank Group (WBG) The capacity of any government to mobilize domestic indicate the severity of tax compliance burden for resources is a crucial factor in the process of poverty businesses, especially for micro, small and medium reduction and its overall economic development. enterprises (IFC 2010). Cognizant of this, the GoE has given due emphasis This analytical report intends to assess tax compliance to the introduction of reforms to its revenues and costs and businesses’ perceptions towards taxation in expenditures management. This emphasis can be seen Ethiopia. The report is based on the findings of a survey from GoE’s five year GTP where the expected growth of businesses (both formal and informal businesses) of the revenues from domestic sources in general, and in Ethiopia as conducted by the Department of its tax revenue in particular, relies on the government’s Economics of the Addis Ababa University on behalf of plan to improve tax compliance, broaden the tax the Ethiopian Revenue and Customs Authority (ERCA) base and to improve the tax administration capacity and the International Finance Corporation (IFC)/ of revenue authorities. As indicated in the GTP, tax World Bank group. Given the diversity of the country, xiii the study covered four regional states apart from The report is organized in four sections. The first part Addis Ababa. The findings of the study are expected presents an overview of the Ethiopian tax system and to offer tax policy makers and tax administrators an recent reform initiatives; this is followed by a discussion opportunity to pinpoint specific problems to help them of research objectives and the methods employed reduce the cost of complying with tax policies and in section two. Section three presents results of the procedures, thus improving the revenue performance survey while the last section presents conclusions and and also the efficiency and business-friendliness of the recommendations. tax system. xiv  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia 1 Overview of Ethiopian Tax System and Recent Reform Initiatives The 1995 Constitution of the Federal Democratic • Taxes on incomes derived from large scale mining Republic of Ethiopia (FDRE) classifies taxation power and all petroleum and gas operations, and royalties into three: as those assigned exclusively to the federal on such operations government, regional states, and concurrently to Following the assignment of revenues, tax both regional and federal governments. As per the administration organs are ERCA and regional revenue constitution, regional states have the power to levy and authorities. ERCA is responsible for administration collect taxes from sources assigned to them. of revenues that belong exclusively to the federal Federal government revenues include: government and those concurrently owned by both. • Income tax on employees of the federal government Regional revenue authorities are entrusted with the and international organizations; responsibility of administering taxes assigned to them. • Income, profit, sales and excise taxes on enterprises The principal taxes that affect businesses in Ethiopia owned by the federal government; are income taxes, value added tax (VAT), turnover tax • Tax on income of air, rail and sea transport services; (TOT) and excise taxes. The subsequent discussions briefly present the basic features of each of these taxes. Regional states revenues include: Income tax proclamation No. 286/ 2002 (as amended) • Income tax on employees of the state and of private and council of Ministers regulations 78/ 2002 (as enterprises; amended) provide the legal basis for income taxation • Taxes on the income of private farmers and farmers in Ethiopia. According to this legislation, there are incorporated in cooperative associations; four schedules of income, namely: income from • Profit and sales taxes on individual traders carrying employment (schedule A), income from rental of out a business within their territory; buildings (schedule B), business profit (schedule C) and • Profit, sales, excise and personal income taxes on other income (schedule D). A further look at the income income of enterprises owned by the states; tax under schedule C (business profit tax) shows that the tax is imposed on profit obtained from an Joint federal and regional states revenues: entrepreneurial activity. It is chargeable at rates ranging • Profit, sales, excise and personal income taxes on from 10 to 35 percent if the taxpayer an individual enterprises they jointly establish; (unincorporated entity) and 30 percent on profits • Taxes on the profits of companies and on dividends earned by a body (an incorporated entity). The income due to shareholders; and tax legislation classifies businesses into three categories   1 as A, B, and C. Category A businesses are those that 2 percent2 of the payments for the purchase of goods have annual turnover of ETB 500,000 or above and all and services.3 In addition importers are required to incorporated entities; category B includes those that are pay 3 percent of the Cost, Insurance and Freight (CIF) not already classified as A and have annual turnover value of goods imported for commercial purposes as of more than ETB 100,000 but less than ETB500,000. a withholding income tax on imports. Withholding Category C taxpayers are those that are not already income tax on payments is required to be remitted included in categories A and B and have annual to the tax authority within 30 days from the end of turnover up to ETB 100,000. the month in which the income was withheld. Both withholding income tax on imports and purchases of For categories A and B the tax is assessed based on the goods and services are allowed to be offset against profit and loss statement prepared in accordance with business profit tax due from the supplier by the end of the Generally Accepted Accounting Principles (GAAP) the accounting year in which the tax was withheld. subject to the specific rules as provided in the income tax legislation. Further, as per the income tax legislation businesses are obliged to withhold employment income tax On the other hand, the tax from category C taxpayers is (schedule A income) from payments to employees and levied and collected in a separate regime—the “standard remit the amount to the concerned tax office in the next assignment” which uses a classification of businesses month the income is withheld. into more than 80 business sectors and 19 turnover bands. This regime is based on daily sales estimates, Schedule B, as noted above, deals with income from which the government carries out once in a few—years rental of buildings. The basic provisions with respect to time. For example, the Addis Ababa city administration tax rates, determination of taxable income etc are very carried out daily sales estimates for category C similar to schedule C income (for business profit). businesses twice since the overhaul of the income tax Proclamation 285/2002 and council of Ministers’ system in the year 2002. Daily sales are estimated by regulation 79/2002 (as amended) provide the legal basis a committee of tax officials by asking the concerned for the imposition and collection of VAT in Ethiopia. taxpayer questions and also observing the business Similarly, proclamation 308/2002 (as amended) location and the nature of the business. In practice, the governs the imposition of TOT. In terms of design tax is calculated using the annual turnover estimated VAT is imposed on the supply of goods and services based on the daily sales estimate and profit margin other than exempted supplies (such as bread and milk). specified in the regulation (as amended). Category C VAT is based on the invoice credit method in which taxpayers are expected to make an annual declaration taxpayers are given credit for the VAT paid on inputs of annual sales, but are not required to keep books of when it is supported by the relevant documents. The account. If category “C” taxpayer maintains book of tax is also based on the destination principle in that accounts acceptable to the Tax Authority, the taxpayer imports are taxed but not exports. VAT is chargeable may pay the tax on the basis of book of accounts. at a standard rate of 15 per cent on all taxable supplies Categories A and B are required to file tax returns and of goods and services other than those zero rated pay the tax due in respectively 4 months and 2 months (mainly exports) and exempted. VAT registration is from the end of their accounting year while category C required by businesses that have annual turnover of taxpayers pay the tax due in one month from the end ETB 500,000 and more. In addition to annual turnover of the accounting year. The latter may request to make as a basis for registration of taxpayers, Ethiopia uses their tax payments in installments in cases of cash flow difficulties. 2 If the recipient does not provide the tax identification number then the law requires 30 percent of the payment to The income tax legislation requires such agents as be withheld. incorporated entities, government organizations, and 3 The withholding income tax on payments threshold for Nongovernment organizations (NGOs) to withhold goods and services is respectively ETB 10,000 and ETB 500 per one transaction. 2  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia sector specific registration schemes, in that those In addition to business profit tax and VAT/TOT, engaged in such sectors as plastic and plastic products businesses engaged in the importation and or manufacturing, shoes manufacturing, computer and production of selected products are liable to excise accessories suppliers and gold smiths are required to taxes as provided in excise tax proclamation 307/2002. register for VAT regardless of their annual turnover.4 Excisable goods include textile and textile products, The accounting and reporting period for VAT is one vehicles, tobacco and tobacco products, drinks calendar month regardless of the size and capability (including all types of alcoholic and soft drinks), sugar of businesses. The VAT legislation allows refunds to and salt. The base for computation of excise tax is the be made to mainly exporters within two months from cost of production for locally produced goods and cost, the time refund application is filed. Non-exporting insurance and freight (CIF) value and custom duties of taxpayers are required to carry forward excess credits imported goods. The tax rate ranges from 10 percent to the next five accounting periods (five months); if to 100 percent. Generally excise tax is payable on there are still unused excess credits it is allowed to be imported goods at the time of clearing the goods from refunded within two months from the time of lodging customs area and on goods produced locally, not later application for refund. Very recently, the government than 30 days from the date of production. has introduced VAT withholding scheme where federal In addition to the above domestic taxes affecting government institutions are required to withhold the businesses, there are duties and taxes imposed on the VAT on their purchases and remit the amount to the tax import and export operations of businesses. Since this authority. report is about tax compliance costs and perceptions Businesses supplying taxable goods and services and towards taxation focusing on domestic taxes, we do not not required to register for VAT are expected to pay deal with these duties on international trade. TOT on their supplies of goods and services. TOT is considered as an equalization tax between VAT registered and unregistered businesses. The applicable Recent Tax Reform Initiatives rates are 2 percent on goods supplied domestically In an effort to streamline and simplify the tax system, and on contractors, grain mills, tractors and combine– the GoE adopted a series of tax reform measures harvesters; and 10 percent on all other services (e.g., focusing on reducing tax rates, removing unproductive barbers/hairdressers, car mechanic, etc.). Certain taxes, broadening the tax base and improving and services are exempt, including financial services, the modernizing revenue collection (MoFED, 2014). supply of electricity, water and kerosene, the provision The key reform measures taken in the early 2000 of transport; educational institutions, child-care). were overhaul of the income tax regime which was Estimation of sales for category C taxpayers follow the operational for over four decades (with no major same procedure as for income tax, but must also take changes), and introduction of VAT as a replacement into account an estimate of the split of income between to sales tax. Specifically, the introduction of taxpayer goods and services. identification number (TIN) along with finger prints and the presumptive tax regime (standard assessment) The accounting period for TOT differs among the for category C taxpayers could be mentioned, among different category of businesses. Category A businesses others. have accounting period of one month; in that taxpayers in this category are required to file and pay turnover tax The introduction of electronic sales register machine on a monthly basis. Category B businesses are required is also a very recent phenomenon in the Ethiopian to file and pay their TOT on a quarterly basis while tax system. The government has further launched category C businesses pay TOT annually. electronic filing system and established a call center. The Electronic Single Window Project (for a better trade 4 Voluntary VAT registration is allowed for those that facilitation) is the other initiative. transact at least 75% of their transactions with VAT registered businesses. Overview of Ethiopian Tax System and Recent Reform Initiatives   3 In respect of tax administration, recently the tax Very recently ERCA and Addis Ababa city government’s administration was organized as a separate and revenue agency have merged and ERCA is currently autonomous government body, which later resulted administering taxes on behalf of the Addis Ababa in the establishment of Ministry of Revenue (MoR) City Revenue Agency. To closely follow taxpayers in coordinating and supervising the three revenue Merkato, ERCA has opened tax offices in Merkato. agencies of the Federal Government, namely Federal Currently, ERCA administers domestic taxes in its Inland Revenue Authority (FIRA), Ethiopian Customs 10 branches offices (Large Taxpayers’ Office (LTO), Authority (ECA) and the National Lottery. Later West Addis Ababa Branch, East Addis Ababa Branch, the government reorganized and restructured the Mekelle, Adama, Hawassa, Bahir Dar, Jimma, Dere tax administration at federal level by merging FIRA, Dawa, Kombolcha). Similarly for customs duties ERCA ECA and MoR into one authority- ERCA. ERCA is has 11 branches throughout the country (Addis Ababa organized as an authority with direct accountability to Kaliti, Addis Ababa Airport, Adama, Mojo, Dere the Prime Minister. Dawa, Jigjiga, Moyalle, Kombolcha, Mekelle, Bahir Dar, and Millie customs branch offices). 4  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia 2 Objectives and Methods Adopted The main objective of this study was to estimate tax TIGRAY and Afar), Bahir Dar (AMAHARA and compliance costs for private businesses in Ethiopia and Benishangua), Adama (OROMIA, Dire Dawa, Harari assess business taxpayers’ perception on the tax system and Somali), and Hawassa (SNNPRS and Gambela). and the business environment at large. The valid sample size was 1003 for formal businesses. In order to achieve the above objective, the study The regional distribution of the sample firms is as employed a survey of businesses—both formal and follows: approximately 37 percent were from Addis informal businesses—operating in Ethiopia. The Ababa City Administration; 25 percent were from subsequent discussion presents the survey design in Oromia, Harari, and Somali regional states and Dire respect of sampling, instrument design and actual Dawa City Administration; about 19 percent were from conduct of the survey. Amhara and Benishangul-Gumuz regional states; about 11 percent were from Tigray and Afar regional states; and about 9 percent were from SNNP and Gambella 2.1 Survey: Sampling design regional states (Table 2.1). Ethiopia has nine Regional States and two City In terms of the legal status of firms in the sample, the Administrations (Addis Ababa and Dire Dawa). Each largest proportion (92.5 percent) of 1003 firms belongs region is divided into zones and each zone into Woredas. to sole-proprietorship status. Cooperatives and private Woredas are further divided into Kebeles, the lowest limited companies represent very small proportions, administrative units. In City Administrations, especially respectively, 4 and 3.3 percent of the total firms in the in Addis Ababa, the administrative division is slightly sample, mostly in category A. However, about 18% of different. Each city is divided into sub-cities and each the businesses in category B reported they were a PLC. sub-city into Woredas, the lowest administrative unit. The survey was conducted in four major regional states In-line with the sectoral configuration of firms in the and Addis Ababa. The total number of eligible business country, the largest proportion of the 1003 firms in the taxpayers in the sampling frame was 987,923 (formal sample (about 50 percent) operate in the service sector businesses). The eligible sampling frame was stratified followed by trade (both wholesale and retail trade) by region, sector, category (A, B and C) and ownership which represents about 41 percent. Firms that engage status and the sample was randomly selected. In each in the manufacturing sector are only about 8 percent of region, one major city or town was selected using urban the sample. As might be expected, manufacturing firms population size as a criterion. In total, five major cities are more likely to be found in category A (where they were covered: Addis Ababa, Mekelle (representing make up over 20% of the group) than in categories B   5 Table 2.1: Distribution of Sample Firms—Formal (by region and category) Category A Category B Category C Location N Percent N Percent N Percent Addis Ababa 57  65 61  67 251  30 Oromia, Dire Dawa, Harari, and Somali 18  20 17  18 219  27 Amhara and Benishangul-Gumuz  6   7  9  10 172  21 Tigray and Afar  4   4  3   3 103  12 SNNPR and Gambella  3   4  2   3  79  10 Total 87 100 92 100 824 100 Table 2.2: Sectoral Distribution of Businesses—Formal (by taxpayers’ category) Category A Category B Category C (N 5 342) (N 5 253) (N 5 408) Manufacturing 20.7 9.8 6.3 Trade (wholesale and retail) 42.6 28.3 42.6 Other services 35.6 59.8 50.5 Others 1.1 2.1 0.6 Total 100 100 100 Table 2.3: Business Category by Turnover (percent) Category A Category B Category C   (N 5 342) (N 5 253) (N 5 408) Total (N 5 1003) Less than ETB 100,000 20.7 38.0 88.1 77.7 ETB100,001-ETB500,000 20.7 47.8 11.4 15.6 ETB500,001-ETB 10,000,000 49.4 14.1 0.5 6.0 Above ETB 10,000,000 9.2 0.0 0.0 0.8 Total 100.0 100.0 100.0 100.0 or C (where they make up 10% and 6% respectively) 100,000 consistent with the definition of category C (Table 2.2). businesses.5 The distribution of businesses by category and turnover In respect of informal businesses, the sampling band shows that about 40 percent of those in category technique adopted was area based sampling. The A were found to have turnover less than ETB 500,000. sampling frame consisted of the list/size of urban These businesses could be incorporated entities for they population by zone, the administrative level at which are categorized as A regardless of their annual turnover. the aggregation was made. From those in category B about 30 percent were found to have turnover less than ETB 100,000 (Table 2.3). This could be due to the fact that, for the year under 5 It is important to note that the difference between reported consideration, those businesses’ turnover was than the category and category of businesses as per ERCA’s records ETB 500,000. A majority of those in category C (about might be due to the fact that ERCA’s records mostly show the category of businesses at the time of registration for 90 percent) were found to have turnover less than ETB taxes. 6  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 2.1: Sectoral Distribution of Informal approximately 78 percent are from this survey site. Businesses (percent) (N 5 499) The remaining 22 percent of the firms is almost 3.2 evenly distributed across the other four survey areas: 10.6 approximately 6, 4, 6, and 5 percent, from respectively Mekelle, Bahir Dar, Adama, and Hawassa. The distribution of the sample firms in terms of their primary activities is similar to that of the formal businesses. That is, only very small percent of informal businesses that participated in the survey reported that 32.9 53.5 their primary activity is manufacturing, while about 87 percent of firms reported to operate activities that are classified as trade and other services (Figure 2.1). 2.2 Survey: Instrument and actual conduct Manufacturing (N = 16) of the surveys Trade (wholesale and rentals (N = 266) Other Services (N = 164) Based on the literature (IFC 2011), two structured Other (N = 53) survey questionnaires for both formal and informal business surveys were developed. The questionnaires The urban population by zone was prepared using the were further developed using data obtained through 2007/8 Population and Housing Census conducted by focus group discussions (FGDs) and on the outcomes the Central Statistical Agency (CSA). Before drawing of pilot testing (conducted with 40 respondents). the actual sample for the survey, a complete list of zones Questions in the instruments included those pertaining and Kebeles/Woredas (lowest administrative units) was to estimation of compliance costs (formal business prepared for Addis Ababa, Mekelle, Bahir Dar, Adama, survey questionnaire), perception of tax administration and Hawassa. The actual sample was selected using procedures and difficulties with the tax system. The probability sampling with probabilities proportional to reference period for all questions was the fiscal year urban population size of the Kebeles/Woredas located in 2005 E.C. (2012/2013). each city/town. Both surveys of formal and informal businesses were The valid sample size achieved for informal business conducted using face-to-face interview approach. The survey is 499 informal business operators. The largest survey was conducted between May 2014 and August proportion of the firms is sampled from Addis Ababa; 2014. Objectives and Methods Adopted    7 3 Survey Results This section presents the results of the survey 3.1 Profile of respondents (formal and informal businesses). Specifically, it Almost all the respondents to both formal and informal presents the findings of the study with respect to business surveys were owners of the businesses. Of tax compliance costs of formal businesses and the total formal business survey respondents, about perceptions of both formal and informal businesses 90 percent were owners followed by director/manger about aspects of tax compliance and the business (or deputy) representing 7 percent (Figure 3.1). And environment. even higher proportion of informal business survey Before analysis of the data began, it was necessary to respondents (98 percent) were owners of the businesses. generate sampling weights to obtain representative In terms of the taxpayers’ category (formal business data that could be generalized to the target survey), most of the owner respondents are in population. Frequency distributions were calculated category C. As might be expected for relatively larger and, as part of the analysis of the estimated firms, there is more diversity in categories A and B, with compliance costs, international comparisons with a “Director/Manager” answering for about a quarter the results of similar surveys were made. It should of category A businesses and 18 percent of category be noted that the percentages and averages reported B businesses. A chief accountant answered for about are weighted averages or percentages unless stated 7 percent of category A businesses. otherwise. Figure 3.1: Position of Respondents—Formal Businesses (in percent, N 5 1003) Others 0.7 Accountant 0.5 Chief Accountant/Financial Director 0.2 Director/Manger (or deputy) 6.6 Owner 91.9 0 10 20 30 40 50 60 70 80 90 100   9 Figure 3.2: Gender of Respondents—Formal Figure 3.3: Gender of Respondents—Informal (in percent, N 5 1003) Businesses (N 5 499) 70 60.5 27.5 60 50 39.5 40 72.5 30 20 10 0 Male Female Male Female Gender wise, most of formal business survey smaller businesses have relatively higher proportions of respondents are male representing 72.5 percent female ownership. while female respondents accounted for 27.5 percent In addition, in terms of the gender of owners of (Figure 3.2). Combining the figures for the position of businesses (formal) about 67 percent were found to respondents in the establishment with that of gender be owned by men while women owned businesses distribution indicates one of the stylized facts of were lightly more than a quarter of the establishments business firms in developing countries like Ethiopia, covered by the study (Figure 3.4). i.e., women are underrepresented both as owners of business firms and in positions of decision making in In the case of informal businesses, consistent with the business firms. gender of respondents (most of them were owners), the gender distribution indicates that about 60 percent The gender of respondents in the different tax payers’ of businesses in the sample are male-owned while categories shows opposite trends in the sense that the 39 percent are female-owned percentage of male respondents decreases from category A to C while that of female respondents’ increases from Formal business survey data reveals that most of category A to C. This seems to indicate that women are the businesses (more than 64 percent) in the sample relatively more engaged in smaller business (either as obtained Tax Identification Number (TIN) over the owners or managers); given the largest proportion of period 2002–2004 EC. In terms of the time period of the respondents in the sample are owners (Table 3.1). establishment, most of the informal businesses that participated in the survey (approximately 76 percent) Similarly in the informal business survey, the results were established during the period 1991–2000 showed male respondents dominance (Figure 3.3). (Figure 3.5). However, the gender gap is smaller for the informal businesses (approximately 60 and 40 percent for male and female respondents, respectively) compared to Figure 3.4: Gender of Owners—Formal (in percent, that of the sample formal businesses presented. Thus, N 5 1003) we see a continuation of the same trend, i.e., relatively 4.9 28 Table 3.1:  Gender of Respondents—Formal (by taxpayers’ category) (in percent) 67 Category A Category B Category C (N 5 342) (N 5 253) (N 5 408) Male 87 75 71 Female 13 25 29 Male Female Multiple Owners 10  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.5: Period of Establishment Figure 3.6: Working Premises of the Sample Firms 4.8 (N 5 499) 5.21 20.04 19.2 22.44 5.41 15.63 31.26 76.0 Before 1991 1991–2000 After 2000 At the same premises where you live At separate premises that you own At separate premises that you lease/rent in When informal business respondents were asked At separate premise/place neither owned nor rented in Your place of business is variable about their working premises, about 20 percent of Other (specify) them operate at their living place, 15% rent a separate working premise and only 5% of them own a separate working premise (Figure 3.6). More than 31 percent and working premises shows that overall the average of the firms reported that their working premises are number of workers in the formal businesses is about separate from their living place but neither own or 5 workers per establishment. Of this number, about rented in the place. Significant number of respondents 4 workers are full time workers (Table 3.2). As should (more than 22 percent) indicated that they do not be expected, category A businesses have higher numbers have a fixed business place—their places of work are of workers (about 26) than categories B (7 workers) variable. 6 and C (2 workers). In addition to the above examination of the results (in The data from the informal business survey reveal both surveys) with respect to number of employees that most of the firms in the sample are operated by 6 Table 3.2: Number of Workers in Sampled Businesses—Formal (total and by categories) Variable N Mean Median Std. Dev. Min Max Total number of workers 1003 5 3 23 1 1618 Full time workers 1003 4 2 18 0 878 Part time workers 1003 1 0 12 0 74 Total number of workers (category A) 342 26 8 73 1 1618 Full time workers (category A) 342 20 5 57 1 878 Part time workers (category A) 342 7 1 38 0 740 Total number of workers (category B) 253 7 3 8 1 40 Full time workers (category B) 253 5 2 5 0 25 Part time workers (category B) 253 2 0 5 0 33 Total number of workers (category C) 408 2 2 2 1 3006 Full time workers (category C) 408 2 1 1 0 67 Part time workers (category C) 408 1 0 2 0 300 6 This is just one large vegetable cooperative which employs 300 temporary workers and does not have permanent workers. Although this cooperative is self-reported to be under category C, it is not clear why this cooperative is in this category. Survey Results   11 Figure 3.7: Number of Employees—Informal (N 5 499) 0.4 4+ 2.4 Number of Workers 0.2 3 3.6 4.6 2.2 2 10.8 21.8 13.4 1 84.6 71.1 83.8 0 0.4 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 Share of Workers in Percent Part-time Full-time Total a single person (Figure 3.7). More than 70 percent of A businesses were found to have kept all receipts informal businesses that participated in the survey are and invoices in an organized manner. Similarly about operated by 1 person and approximately 85 percent of 80 percent of businesses in category A who use a the firms reported that the number of full-time workers computer for purposes of bookkeeping were found to be is just one. Furthermore, most of the firms (about using specialized accounting or bookkeeping software. 84 percent) do not have part-time workers. This is not On the other hand, only 40 percent of businesses surprising given that most informal firms are operated in category C were found to have kept receipts and by individual owners. invoices in an organized manner (Figure 3.9). Both formal and informal business survey respondents In the case of informal businesses almost all of them were asked about their bookkeeping practices. The reported that they did not maintain physical receipts results show that about half of the formal sector and invoices of transactions (Figure 3.10). Only respondents reported that they kept all receipts and 15.6 percent and 17 percent said that they kept records invoices of transactions in an organized manner, of sales and purchases, respectively. This shows that while 42 percent maintained full records of revenues the bookkeeping practice in the informal sector is very and expenses (not necessarily using computer) limited compared to that of the formal sector. (Figure 3.8). Only about 9 percent use a computer for In terms of regional variation, about 18.8 percent of purposes of bookkeeping; about two-third of those respondents in Adama and Mekelle maintained physical who use computers also use specialized accounting or receipts and invoices of transactions, compared with bookkeeping software. 15.1 percent in Addis Ababa and 13.6 percent in The bookkeeping practice of formal businesses by Bahir Dar (Figure 3.11). Keeping records of purchases category also reveals that about 98 percent of category Figure 3.8: Bookkeeping Practices of Businesses—Formal Using a specialized accounting or bookkeeping software (N = 86) 66.4 33.6 Using a computer for purposes of bookkeeping (N = 1003) 8.6 91.4 Doing full records of revenues and expenses (not necessarily using) (N = 1003) 42.3 57.7 Bookkeeping for revenue (not necessarily for expenses) (N = 1003) 51.8 48.2 Keeping all receipts and invoices in an organized manner (N = 1003) 49.5 50.5 0 20 40 60 80 100 120 Yes No 12  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.9: Bookkeeping Practice of Businesses—Formal (by category) (a) Bookkeeping Practices by Category (b) Using a Specialized Accounting or Bookkeeping Software (for those that use a computer) (N = 342) (N = 253) (N = 408) (N = 205) (N = 55) (N = 10) C No 59.6 54.9 65.8 97.2 No 65.2 C Yes 40.4 45.1 34.2 2.8 Yes 34.8 No 14.1, 18.5, 29.3, 75 No 22.7 B B Yes 85.9 81.5 70.7 25 Yes 77.3 No 2.3, 16.1, 11.5, 54 No 20 A A Yes 97.7 83.9 88.5 46 Yes 80 0 50 100 150 200 250 300 350 0 20 40 60 80 100 Keeping all receipts and invoices in an organized manner Bookkeeping for revenue (not necessarily for expenses) Doing full records of revenues and expenses (not necessarily using a computer) Using a computer for purposes of bookkeeping Figure 3.10: Bookkeeping Practices (percent)—Informal Using a specialized accounting or bookkeeping software 100 Using a computer for purposes of bookkeeping (N = 499) 100 Doing full financial accounting (not necessarily using a computer) 96.4 3.6 Keeping records of purchases (N = 499) 83 17 Keeping records of sales (N = 499) 84.4 15.6 Keeping all physical receipts and invoices (N = 499) 96.8 3.2 0 20 40 60 80 100 120 Yes No Figure 3.11: Bookkeeping Practices by Location (percent, yes)—Informal 80 60 15.6 40 25 3.1 3.1 13.6 21.9 21.7 17 20 0 Mekelle Bahir Dar Adama Hawassa Addis Ababa (N=32) (N=22) (N=32) (N=23) (N=390) Doing full financial accounting (not necessarily using a computer) Keeping records of purchases Keeping records of sales Keeping all physical receipts and invoices and or sales seems important compared with other tax returns which was the largest followed by turnover bookkeeping practices. tax (36 percent of formal businesses). Value added tax return submission was reported by about 12 percent Businesses were asked about the types of taxes for while the corresponding figure for employment related which they submitted tax returns in the last tax year tax and contributions was about 15 percent. Other (2012/13). About 61 percent submitted business profit Survey Results   13 Table 3.3: Types of Taxes Paid by Category and Ownership Type (percent) Category Ownership Type Individual/Sole Private Proprietorship Limited Cooperative   A B C Percent Company Society Others Business profit tax (N 5 740) 85 79 55 58 85 71 64 Value added tax (N 5 410) 77 37  3  9 70 32 56 Turnover tax (N 5 255) 14 37 32 31  6 43  8 Withholding income tax on 45 19  2  4 66 28 62 payments (N 5 228) Employment related 59 39  6 10 77 48 90 contributions (N 5 340) Other taxes (N 5 157) 10 14 24 22 30 14 13 tax returns submitted include withholding income tax submitted value added tax returns compared with on payments (8 percent) and “other taxes” (including sole proprietorships (9 percent). This is reversed for excise tax, building rental income tax, and property tax) turnover tax (6 percent) for private limited companies (24 percent). We also note that well over one-fifth of and 31 percent for sole proprietorships. We may also businesses (and over half of category C taxpayers) have note the lower percentage of businesses submitting not reported submitting tax returns mainly because they value added tax and turnover tax is due to exemptions are operating under the standard assessment whereby from payment of value added tax and turnover tax the tax they pay is fixed by the tax authority. Other for businesses such as providers of education, health businesses reported paying rental income tax (included and transport services as well as those that rent out under “other taxes”) (Table 3.3). residential premises. As expected a much larger percentage of businesses from category A (77 percent) submitted value added 3.2 Tax compliance costs tax returns compared with category C businesses (3 percent). On the other hand, a larger percentage of Sandford (1995) defined tax compliance costs as costs category B (37 Percent) and C (32 percent) businesses incurred by taxpayers in meeting the requirements laid submitted turnover tax returns compared with those in on them by the tax law and the revenue authorities; category A (14 percent). The proportion of businesses costs over and above the actual payment of tax; costs in a category that submitted returns for other taxes is which would disappear if the tax was abolished. much larger for category C (24 percent) compared with In this study tax compliance costs (TCCs) is estimated others, due to the relative significance of rental income using the following: tax for this group. TCCs 5  In house cost of time spent by employees We also note from Table 3.3 that, as expected, a 1 much larger proportion of private limited companies Outsourcing costs (85 percent) submitted profit tax returns compared 1 with sole proprietorships which have the lowest share Costs of software/ hardware/ dataware/ by ownership type (58 percent). Similarly a larger information percentage of private limited companies (70 percent) 14  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.12: Average Total Compliance Costs (in ETB) 12000 9804 10000 7609 8000 5842 6000 4000 2000 0 General Bookkeeping Tax Accounting (incl. acquisition Tax Accounting (excl. acquisition and maintenance cost) and maintenance cost) For the ease of respondents, the questionnaire asked 5 cost of in-house time spent for tax accounting 1 them first about the total amount of time spent on both cost of outsourcing for tax accounting 1 acquisition general bookkeeping and tax compliance tasks. and maintenance cost of software/hardware/dataware/ information In-house cost of time spent by employees 5 time spent by various individuals on bookkeeping and tax The average total cost of a business for general accounting tasks * salaries (salaries of relevant workers bookkeeping was found to be ETB 9,804 (USD 523.2)7 was asked in the survey); in the last tax year (2012/13). In the same year, average total tax compliance cost of a business including costs of Outsourcing costs 5 Outsourcing cost paid to outside acquisition and maintenance of software and hardware personnel for bookkeeping and tax accounting tasks; was ETB 7,609 (USD 406) while average total tax Cost of software/hardware/dataware/information 5 compliance cost without acquisition and maintenance Money spent by business on acquisition and costs was ETB 5,842 (USD 311.7) (Figure 3.12). maintenance of software, hardware, dataware and The relative cost of tax compliance was calculated as a information in the five year period before the survey percentage of turnover. The average tax compliance cost divided by five. of a business (with acquisition and maintenance costs of The overall cost is then divided into general software and hardware) as a share of turnover is about bookkeeping costs and tax accounting costs. In this 5.4 percent while the share of tax compliance cost in study general bookkeeping is defined to include all total turnover without acquisition and maintenance bookkeeping activities that will be undertaken even if costs is 4.7 percent (Table 3.4). In absolute terms, the business did not comply with tax requirements. All tax compliance cost for category A businesses is well other activities are considered under tax accounting. over 5 times the cost for category C businesses (both General bookkeeping cost 5 cost of in-house time when tax accounting cost includes and excludes spent for general bookkeeping + cost of outsourcing for acquisition and maintenance costs). However, relative general bookkeeping. tax compliance cost as a share of turnover is larger for category C compared with category A. Tax accounting cost 5 cost of in-house time spent for tax accounting 1 cost of outsourcing for tax accounting 7 Average exchange rate for the year 2012/13 was USD 1 5 Or ETB 18.59 (NBE annual report 2012/13) Survey Results   15 Table 3.4:  Average Tax Compliance Costs (in ETB and as a share of turnover) by Taxpayers’ Category Category A B C N 323 199 201 Tax accounting cost in ETB (excl. acquisition and maintenance cost) 17185 10068 2637 Tax accounting cost in ETB (incl. acquisition and maintenance cost) 21639 13162 3203 Tax accounting cost as share of turnover (%) (excl. acquisition and 3.85% 3.84% 5.03% maintenance cost) Tax accounting cost as share of turnover (%) (incl. acquisition and 4.70% 5.39% 5.51% maintenance cost) In addition, tax compliance cost as a share of turnover whether acquisition and maintenance cost of software tends to decrease as the turnover of businesses increases and hardware is excluded or not). The relative cost of suggesting that it is regressive (Figure 3.13). This is true tax compliance does not differ much across most of whether we include or exclude cost of acquisition and the ownership types with the exception of cooperative maintenance cost of software and hardware. Those with societies where the percentage is visibly lower than the turnover under ETB 100,000 faced compliance costs others. of almost 7 percent (for those who had tax hardware/ In terms of sector of businesses, compliance cost is the software); otherwise about 6 percent. lowest for ‘other services’ and the highest for ‘other Table 3.5 shows tax compliance costs by ownership sectors’ which largely includes agriculture. On the other type and sector of businesses. Sole proprietorships hand, the relative cost of tax compliance is the highest have the lowest tax compliance cost (ETB 4,046) while for trade and the lowest for ‘others’ the tax compliance cost of other ownership types is Figure 3.14 shows tax compliance cost as a share of much higher (for example, the cost for private limited turnover for different countries including Ethiopia. For companies is ETB 28,104 or ETB 35,793 depending on Figure 3.13: Tax Compliance Cost as a Share of Turnover by Turnover Band Tax Accounting Cost-turnover Ratio (%) 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 100,000 100,001 to 500,001 to Over or less 500,000 10,000,000 10,000,000 Turnover in Birr Tax accounting Tax accounting compliance cost as compliance cost as percent of turnover percent of turnover (excl. acquisition and (incl. acquisition and maintenance cost) maintenance cost) 16  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Table 3.5: Tax Compliance Costs (in ETB and as a share of turnover) by Taxpayers’ Ownership Type and Sector Ownership Type Sector Individual/ Private Sole Limited Cooperative Other Proprietorship Company Society Others Manufacturing Trade Services Others N 541 128 42 12 67 334 281 41 Tax accounting 4046 28104 11671 26267 9789 5448 5189 16611 cost in ETB (excl. acquisition and maintenance cost) Tax accounting 5279 35793 13645 36333 17673 6821 6661 21658 cost in ETB (incl. acquisition and maintenance cost) Tax accounting cost 4.77% 4.92% 3.52% 5.36% 3.18% 5.77% 4.07% 2.61% as share of turnover (excl. acquisition and maintenance cost) Tax accounting cost 5.45% 5.66% 4.04% 5.52% 4.40% 6.26% 4.74% 3.19% as share of turnover (incl. acquisition and maintenance cost) Figure 3.14: International Comparison of Tax Compliance Costs 14 Tax Compliance Costs as a Percentage of Turnover 12 10 8 6 4 2 0 0–5 5–13 13–24 24–41 41–66 66–104 104–161 161–246 246–374 374–567 567–855 855–1287 1287–1936 1936–2909 2909–4369 4369–6558 6558–9843 Turnover in '000 USD South Africa Ukraine Uzbekistan Kenya Georgia Ethiopia Nepal Survey Results   17 Figure 3.15: Businesses Involved in Outsourcing and In-house Activities for Bookkeeping and Tax Accounting (percent) General bookkeeping (N = 761) 84.8 7.6 7.6 Business profit tax (N = 692) 68.4 12.5 19.1 Value added tax (N = 409) 50.5 24.2 25.3 Turnover tax (N = 267) 85.1 7.7 7.2 Withholding income tax on payments (N = 230) 69.2 21.4 9.4 Employment related contributions (N = 345) 70.7 16.3 13.0 Other taxes (N = 99) 99.0 0.6, 0.4 0 20 40 60 80 100 120 Completely In-house Partially In-house Completely Outsourced this purpose we use the tax compliance costs measure Studies in other countries also show that the size of for Ethiopia that excludes costs of acquisition and compliance cost varies across countries.9 For instance, maintenance of software and hardware so that the the overall average tax compliance cost was about USD results are more comparable across countries. We may 1.48 billion or 1percent of GDP for Ukraine in 2007. also note that such comparisons have some limitations Similarly, for Indonesia, TCC as a share of GDP and as detailed definitions (such as legal requirements and revenue was about 0.19 percent and 3.2 percent in 2010, general practices of record-keeping and reporting) differ respectively (Susila and Pope, 2012). A similar study across countries. Generally tax compliance cost as a in Nepal indicates that the overall average compliance percentage of turnover for Ethiopia is higher compared cost of about 0.46 percent of GDP or 2.7 percent of to that of countries including Kenya and Nepal tax revenue in 2012 (SEDF, 2012). In Slovenia, total (Figure 3.14), (see Annex 1 for detailed informational compliance costs (for personal income tax) accounted comparisons). Tax compliance cost in Ethiopia is high for around 0.18 per cent of GDP and 2.8 percent of even in comparison to the countries considered, since personal income tax revenue (Klun, 2004), while the most of the countries that have had such a survey were figure for Croatia was 0.03 percent of GDP and 0.81 the ones already known to have a problem with tax percent of personal income tax (Blazic, 2004). Overall, compliance costs. there is considerable variation in terms of the size of tax compliance cost across countries depending on Based on the results of the survey, with about one their economic size and tax revenue. For instance, tax million businesses in Ethiopia, we estimate the total compliance cost as share of tax revenue was about tax compliance cost to be about ETB 5.8 billion (about 11.5 percent (or 1.1 percent of GDP) in Armenia USD 309.5 million8) and ETB 7.5 billion (USD 400.5) (Jrbshyan and Harutyunyan, 2006), while the figure for (depending on whether tax compliance cost includes Canada was 2.7 percent (Charron et al., 2008). acquisition and maintenance cost of software and hardware). This is between 4.5 and 5.8 percent of In order to estimate tax compliance costs precisely it is Ethiopia’s total government revenue collected in important first to examine the practice of using external 2012/13 and about 1 percent of its GDP during the assistance for tax compliance (a direct monetary same year. cost for a business) versus doing work in-house (i.e., staff time). As Figure 3.15 shows a large majority of 8 Assuming average exchange rate for the fiscal year 9 See Yesegat (2009) and Susila and Pope (2012) for detailed 2012/13, USD 1 5 ETB 18.59. review of literature on compliance cost. 18  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.16: Distribution of Full Outsourcing Cost Figure 3.18: Distribution of Full Outsourcing Cost between General Bookkeeping and Tax Compliance of Tax Compliance by Tax Type (in percent) (N 5 171) (in percent) (N 5 174) 1% 8% 3% Business Profit Tax Value Added Tax Cost of Turnover Tax Cost of 18% outsourcing for 50% outsourcing for Withholding Income general tax related Tax on Payments bookkeeping, tasks, 0.55 0.45 Employment Related 20% Contributions Other Taxes businesses do most or all of their bookkeeping and In terms of distribution of the external professionals’ tax accounting activities in-house. Only 15 percent of costs of tax compliance across tax types, half of this businesses outsource any of their general bookkeeping is for tasks related to business profit tax (50 percent) work, although about half outsource at least some followed by value added tax (20 percent) and turnover of their VAT compliance work, and about one third tax (18 percent) (Figure 3.18). outsource at least some of the compliance tasks for Businesses that outsource business profit tax compliance business profit tax. activities incurred an average of ETB 2,267 per business Looking at costs based on the practice of using external in the 2012/13 fiscal year, which is the highest for a assistance for tax compliance and general bookkeeping tax type. Value added tax compliance is the second activities reveals that over half of the total external highest in terms of average outsourcing cost by tax type professionals’ cost was incurred in tax compliance amounting to ETB 913 (Figure 3.19). activities. The internal costs of tax compliance measured in terms As Figure 3.16 shows 55 percent of the outsourcing cost of the monetary value of staff time spent for general goes to cover costs of tax compliance while 45 percent Figure 3.19: Average Cost of (full) Outsourcing for is for general bookkeeping activities. Average external Businesses That Outsource by Tax Type in ETB (complete outsourcing) cost in ETB and its distribution (N 5 171) between general bookkeeping and tax compliance is 2500 shown in Figure 3.17. 2267 Figure 3.17: Average Cost of Full Outsourcing for 2000 Businesses That Outsource (in ETB) (N 5 174) 1500 9000 8216 8000 1000 913 817 7000 6000 500 339 5000 4505 156 47 4000 3711 0 tio ed x x x t s 3000 en Ta Ta Ta xe bu at ns ym tri el Ta it d er on R of de 2000 Pa ov er C ent Pr Ad rn th on ss m O Tu e 1000 oy ne x lu Ta Va pl si Em Bu e 0 m co In Cost of Cost of Total Cost of g in Outsourcing for Outsourcing for Tax Outsourcing ld ho General Related Tasks ith Bookkeeping W Survey Results   19 Figure 3.20: Person-Days Spent on In-house Figure 3.21: Distribution of Time Spent on In- Bookkeeping and Tax Compliance (N 5 627) house Tax Accounting by Tax Type (N 5 592) 106.16 5% 4% 2% 77.99 28.17 31% 50% Bookkeeping Tax Accounting Bookkeeping and Tax Accounting bookkeeping and tax compliance related activities 8% reveals similar picture. In terms of in-house time about 74 percent of the time (78 person-days) was spent on Business Profit Tax general bookkeeping while the rest (about 28 person- Value Added Tax days) was spent on tax compliance tasks (Figure 3.20). Turnover Tax Withholding Income Tax on Payments Half of the total time spent by a business on tax Employment Related Contributions accounting goes to business profit tax while another Other Taxes 31 percent goes to turnover tax. Value added tax takes 8 percent of the time spent (Figure 3.21). Figure 3.22: Person-Days Spent on Tax From the discussions so far, it can be noted that Compliance by Tax Activity (N 5 627) business profit tax takes the largest share of total tax 10.92 compliance costs (both internal and external costs), 10.28 followed by value added tax and turnover tax. The burden of compliance costs of business profit tax and value added tax shows that these two taxes do 6.66 have complexities that resulted in taxpayers incurring high costs of compliance in terms of staff time and professional fees paid to external assistance. In addition, looking at the in- house tax compliance costs by tax activity indicates that on average a business spent about 11 person-days on filing activities during the last tax year which is the highest time spent by Pre-filing Filing Post-filing tax activity. Pre-filing took the second highest of the three activities (10 person-days) with post-filing taking by ownership type and turnover. By ownership type the smallest amount of time (about 7 person-days individuals or sole proprietorships have the lowest (Figure 3.22). compliance costs in terms of person-days spent on Table 3.6 presents person-days spent on bookkeeping bookkeeping and tax accounting. For example, sole and tax accounting as well as the share of worker types proprietorships spent about a quarter of the time that 20  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Table 3.6: Time Spent on Bookkeeping and Tax Accounting by Ownership Type and Turnover Ownership Type Turnover in ETB Private 100,001 Individual/Sole Limited Cooperative 100,000 to 500,001 to Over Proprietorship Company Society or Less 500,000 10,000,000 10,000,000 N 521 120 39 292 188 168 44 Person days spent on 96 271 151 92 97 229 522 bookkeeping and tax accounting Person days spent on 73 149 109 70 71 144 337 bookkeeping Person days spent on 23 123 42 22 26 84 185 tax accounting Share of person-days spent on tax accounting by worker type (percent) Bookkeepers and 14% 71% 57% 13% 20% 69% 83% accountants Owners and managers 84% 24% 43% 87% 76% 26% 12% Others 2% 6% 1% 0% 4% 5% 5% private limited companies spend on tax accounting. In (bookkeepers/accountants, managers/owners and other terms of share of person-days spent on tax accounting by workers) working on general bookkeeping and tax ownership type sole proprietorships rely on time of owners accounting related tasks. The daily equivalent of such and managers much more than the other groups which payments was then used to calculate the cost of person- is to be expected (Table 3.6). We note that the number of days spent for each category of workers. person-days spent increases as turnover increases. The total in-house cost of general bookkeeping was In terms of worker type used, the share of the total found to be ETB 9,382 (USD 500.6) while the cost of tax time spent increases with turnover for bookkeepers accounting (tax compliance) was ETB 3,906 (USD 208.4) and accountants showing that as businesses become for a business in the last tax year (Figure 3.23). larger they rely more and more on bookkeepers and accountants for their bookkeeping and tax accounting Figure 3.23: In-house Compliance Cost of Time activities. The share of time spent by owners and Spent on General Bookkeeping and Tax Accounting managers decreases as turnover increases which is in ETB (N 5 537) generally to be expected. The share of the ‘other’ 9381.77 worker category does not change much as turnover increases except those in the lowest turnover category where the share is zero (Table 3.6). In-house compliance cost of time spent on bookkeeping and tax accounting is calculated by converting time spent into money using salary of workers types 3906.23 involved. Salary of worker types is obtained from the survey. Businesses were asked to indicate the average gross remuneration (including salaries, bonuses, insurance and all other benefits) per month per person a similar business would pay. Respondents provided answers to each of the three groups of workers Cost of General Bookkeeping Cost of Tax Accounting Survey Results   21 Table 3.7:  In-house Compliance Cost of Time Spent on General Bookkeeping and Tax Accounting in ETB (by ownership type and turnover) Ownership Type Turnover in ETB Individual/ Private 100,001 Sole Limited Cooperative 100,000 to 500,001 to Over Proprietorship Company Society Others or Less 500,000 10,000,000 10,000,000 N 521 120 39 12 292 188 168 44 Cost of general 7856 32501 15226 33172 6618 9137 22007 74435 bookkeeping Cost of tax 2623 27686 6128 29453 2162 4170 15317 45235 accounting Table 3.8:  In-house Compliance Cost of Time at large. Specifically, the subsequent discussions focus Spent on General Bookkeeping and on views pertaining to registration for taxes, sales Tax Accounting in ETB by Category register machine usage, VAT registration and refund Category process, penalties, appeals, and inspections and audits, A B C and tax compliance. N 245 184 263 3.3.1 Views about Registering for Taxes In-house compliance cost of 24179 14749 6529 time spent on gen. bookkeeping As shown earlier, the largest share of formal business In-house compliance cost of 15176 9135 1954 survey respondents obtained their TINs during time spent on tax accounting 2002–2004 EC. Asked about the disadvantages of registering for taxes, a relatively large proportion of formal businesses indicated higher tax burden or high The relative importance of compliance cost of general tax rates (about 34 percent) as the biggest disadvantage. bookkeeping and tax accounting is different across Further, such factors as complicated tax compliance ownership type. In particular, the cost of tax accounting procedures (13 percent), high cost of raw materials for sole proprietorships is less than half of the cost of and inputs (9 percent), harassment by government general bookkeeping while the share of tax accounting officers (8 percent), and frequent inspections (8 percent) in total in-house compliance cost is much larger for also stood out as the other biggest disadvantages of other ownership types (Table 3.7). In terms of turnover registering for taxes (Figure 3.24) of businesses, as would be expected, the in-house As the perception of respondents about the biggest cost of compliance increases for those with larger disadvantages of registering for taxes shows, the turnover. This is also true by category where category different categories of taxpayers encounter different A businesses have a higher in-house compliance cost constraints. The only problems that seem to attract compared with category C (Table 3.8). a relatively significant proportion of firms as the biggest disadvantage of registering for taxes is higher 3.3 Perception of taxpayers tax burden/tax rates: 26 and 37 percent of firms to aspects of the tax from category B and C, respectively, identified this problem to be the biggest disadvantage (Table 3.9). system and the business However, when we aggregate over some of the environment disadvantages listed we see some similarity in terms of This section presents the views of respondents to two disadvantages: higher tax burden and/or tax rates aspects of the tax system and the business environment and complicated tax compliance procedures. These two 22  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.24: Opinion of Respondent Firms about the Biggest and Second Biggest Disadvantages of Registering for Taxes (N 5 1003) Others (please specify) 8 Reduced competitiveness 6 Complicated tax compliance procedures 13 Frequent inspections 8 Harassment by government officers 8 Higher cost of raw materials and inputs 9 Higher tax burden/tax rates 34 Higher cost of labor 1 Higher utility cost 5 Reduce attractiveness of buying from and selling to non-... 8 0 5 10 15 20 25 30 35 40 Percent Second Biggest Disadvantage Biggest Disadvantage Table 3.9:  General Opinion on the Reasons for Not registering their business for tax.10 About 63 percent Applying for Tax Registration (among claimed to have neither considered nor applied to those who have considered but did not register for taxes. Over 30 percent of respondents apply) (N 5 166 (percent)—Informal reported that they have considered applying for but Finding necessary information is too 9.5 never tried it. complicated and difficult Tax laws and regulations are not clear, fair or 14.9 Respondents who said they had considered but never appropriate tried to register were also asked to provide reasons for I don’t want to deal with the tyranny/corruption 0.6 not applying for tax registration, and about 26 percent of public o said that financial burden of being formal (including Lack of enforcement 0.6 taxes and fees) was the main reason for not applying Financial burden of being formal (including 26.2 for tax registration. About 15 percent of respondents taxes and fees) said that lack of clarity in tax laws and regulations as Lack of trust in government to use tax revenues 2.4 productively another reason for not applying for tax registration. I have no knowledge and/or capacity to deal 6.5 About 17 percent also believe that a business like theirs with tax related should not pay taxes (Table 3.9) A business like mine should not pay taxes 16.7 Those respondents who replied “neither considered Tax authorities do not treat all businesses equally 4.2 nor applied” were asked why they did not consider Other 6.0 to apply; and about a third of respondents replied Total 100 “No one will help me” for registering for tax with tax authorities. Similarly, about 23 percent and 20 percent disadvantages are indicated as the biggest disadvantages also indicated “it was too complicated” and “I had of registering for taxes by 30, 47, and 49 percent to wait too long” respectively as their reasons for not of the firms in categories A, B, and C, respectively considering tax registration(Figure 3.26). (Figure 3.25). Similarly, informal business survey respondents were 10 Only one percent of respondents have a TIN, but not filed asked about whether or not they have ever considered tax returns in the last three years. Survey Results   23 Figure 3.25: Opinion of Respondent Firms about the Biggest and Second Biggest Disadvantages of Registering for Taxes (by tax payers’ category) (percent) 5 4 3 15 11 15 16 12 8 10 Category C Second Biggest (N = 408) Biggest 8 4 1 37 8 8 9 12 6 8 9 9 4 12 7 11 10 17 10 11 Category B Second Biggest (N = 253) Biggest 8 4 2 26 9 9 4 21 7 10 15 8 7 6 10 12 10 20 6 7 Category A Second Biggest (N = 342) Biggest 12 8 2 14 16 10 4 16 9 10 0 20 40 60 80 100 120 Percent Reduced Attractiveness of Buying from and Selling to Non-registered Businesses Higher Utility Cost Higher Cost of Labor Higher Tax Burden/Tax Rates Higher Cost of Raw Materials and Inputs Harassment by Government Officers Frequent Inspections Complicated Tax Compliance Procedures Reduced Competitiveness Others Figure 3.26: Reasons for Giving Up or Never Respondents were also asked about factors that might Considered Nor Applied (N 5 498) encourage them to consider registering for taxes 35 and operate in the formal sector. A relatively higher 32.7 percentage (about 28 percent) of respondents showed 30 that reduction of tax rates as the most important factor 25 23.1 24.1 for registering for tax (Figure 3.27). 20.1 In addition, when respondents were asked about 20 whether or not they have any intentions of registering 15 for tax in the next two years. About half of respondents 10 had said that they are likely or very likely to register for tax in the coming two years. However, the percentage 5 of respondents who said that they are unlikely or very unlikely to register for tax in the next two years was 0 about 25 percent (Figure 3.28). It Was Too No One Will I Had to Wait Other Complicated Help Me Too Long 24  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.27: Most Important Factors That Encourage or Persuade Business to Consider Registering for Tax (N 5 499) Other (N = 4) 0.8 Nothing would encourage me to register (N = 52) 10.8 Less corruption in government (N = 7) 1.5 Better usage of taxes by the government/less corruption (N = 22) 4.6 More fairness from tax authorities officers (N = 30) 6.3 Reduction of tax rates (N = 132) 27.5 An awareness campaign by the tax authorities (N = 34) 7.1 An education campaign by the tax authorities (N = 34) 7.5 Existence of a more transparent tax system with better information (N = 36) 7.2 Decrease in penalties and fines (N = 30) 6.3 Simplifying procedures for tax registration and reporting (N = 35) 7.3 Setting a minimum threshold of income that is not taxable (N = 62) 12.9 0 5 10 15 20 25 30 Figure 3.28: Perception on the Likelihood of Registering for Tax (N 5 499) 120.0 100.0 80.0 40.9 34.8 58.1 48.7 49.6 68.8 60.0 27.3 47.8 22.6 40.0 23.9 12.5 32.3 20.0 31.8 18.7 17.4 28.7 26.5 9.7 0.0 Mekelle Bahir Dar Adama Hawassa Addis Ababa Total Very Unlikely + Unlikely Undecided Very Likely + Likely Figure 3.29: General Perception on First Biggest Disadvantages of Registering for Tax (N 5 499) 8.4 Other 19.8 4.2 Reduced competitiveness 8.0 9.8 Complicated tax compliance procedures 21.4 Frequent inspections 8.0 30.9 8.0 Harassment by government officers 25.1 Higher cost of raw materials and inputs 5.0 11.4 45.9 Higher tax burdon/tax rates 61.5 1.0 Higher cost of labor 2.6 5.8 Higher utility cost 10.8 Reduced attractiveness of buying from and 3.8 8.4 selling to non-registered businesses 0 10 20 30 40 50 60 70 First Biggest Disadvantage First and Second Biggest Disadvantage Informal business survey respondents were also asked (Figure 3.29). Respondents said that high tax burden, about the biggest disadvantages of registering for taxes frequent inspections, harassment by government officers and a larger proportion (46 percent) ranked higher and complicated tax compliance procedures as the first tax burden/tax rates as the first biggest disadvantage and second biggest disadvantages of registering for taxes. Survey Results   25 Figure 3.30: Biggest Advantages of Using Sales Register Machine (percent) (N 5 1003) Other (N = 36) 6.3 Able to get tax credit (VAT credit) (N = 22) 1.8 Increases business reputation (N = 22) 2.2 Saves time (N = 147) 14.3 Better sales and inventory control (N = 175) 16.2 Easy to comply with tax requirements (N = 171) 17.2 Updated and easily available sales information/data (N = 277) 22.3 Less opportunity for theft by my employees (N = 165) 19.6 0 5 10 15 20 25 Percent of Business 3.3.2 Perception of Sales Register being the “difficulty of correcting errors” (over Machine (SRM) Usage 35 percent) followed by “cost of buying the machine” Sales register machine is required to be used by (over 15 percent) (Figure 3.31). businesses in categories A and B. Survey results on SRM usage reveals that only about 80 percent of 3.3.3 Perception of VAT Registration respondents in category A was found to have SRM, and and Refund only just over half (52 percent) of category B businesses With regard to the advantages of registering for VAT, reported having one. Meanwhile, just over 4 percent of Cash flow benefits and opportunity for participating in category C businesses were found to have an SRM. government tenders were shown to be the first biggest advantages of registering for VAT as reported by Asked about the “advantages of using” SRM, about 24 percent and 15 percent of respondents respectively 22 percent cited “updated and easily available sales (Figure 3.32). Respondents also noted that registering information,” and almost 20 percent reported “less for VAT allowed them to trade with a larger number of opportunity for theft by employees” as biggest suppliers and customers (Figure 3.32). advantages of using SRM (Figure 3.30). Other advantages noted included that use of SRM makes it Regarding the disadvantage of registering for VAT, “easy to comply with tax requirements” and “better among others, higher price of goods and services sales and inventory control.” compared with non-VAT registered businesses were considered to be the first and second biggest However, respondents also reported several disadvantage of VAT (Figure 3.33). Additional first disadvantages of using SRM; the biggest disadvantage Figure 3.31: Biggest Disadvantages of a Sales Register Machine (percent) (N 5 1003) Other (N = 51) Bigger tax burden (N = 33) Limited maintenance service providers (N = 48) Requires training workers (N = 54) Difficulty to use other options when it is not working (N = 51) High maintenance cost (N = 86) Electricity/internet interruption (N = 101) Cost of buying the machine (N = 136) Total Difficulty of correcting errors (N = 443) 0 20 40 60 80 100 120 Percent of Businesses 26  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.32: First and Second Biggest Advantages of Registering for VAT (percent) (N 5 1003) Other 8.3 18.6 Easier to get loans 3.2 7.6 Able to get (VAT credit) 4.3 8.0 Increased business reputation 9.2 15.4 Able to buy from a larger number of suppliers 11.6 23.2 Able to sell to a larger number of customers 11.6 30.3 Able to get VAT refunds 13.1 25.2 Cash flow benefits: VAT payments are collected from customer 14.8 23.4 Be able to participate in government tenders 23.9 48.3 0 10 20 30 40 50 60 70 80 First Biggest Advantage (N = 1003) First and Second Biggest Advantage Figure 3.33: First and Second Disadvantages of Registering for VAT (N 5 1003) Other 12.7 27.8 Reduced attractiveness of buying raw materials and other 10.8 28.9 Inability to sell on the black market 8.7 18.3 Higher prices of goods and services 20.7 41.6 Costs of time-consuming accounting procedures 10.2 20.5 More harassment by tax authorities 11.0 20.2 More inspections/audits 16.7 29.7 Delay of VAT refund 9.2 12.8 0 10 20 30 40 50 60 70 First Biggest Disadvantage (N = 1003) First and Second Biggest Disadvantages Figure 3.34: Eligibility for VAT Refund (percent) Figure 3.35: Application for VAT Refunds (percent) 120 120 100 100 80 80 43.0 42.9 36.4 60.9 50.0 60 87.2 78.7 60 95.8 40 40 57.0 57.1 50.0 63.6 20 39.1 20 12.8 21.3 4.2 0 0 Total A B C Total A B C (N = 585) (N = 305) (N = 145) (N = 135) (N = 171) (N = 133) (N = 32) (N = 6) Yes No Yes No biggest disadvantages included more inspections/ majority (about 87 percent) reported that they were audits and harassment by tax authorities as reported not eligible for VAT refund in the 2012/13 fiscal year by 16.7 percent and 11 percent of respondents, (Figure 3.34). respectively. From among eligible for VAT refund, only about The study also assessed perceptions about the VAT 57 percent said they had applied for VAT refund in the refund system. Of the respondents who were filing VAT tax year under consideration (Figure 3.35). (less than 20 percent of the total), the overwhelming Survey Results   27 Figure 3.36: Reasons for Not Applying for the VAT Refund (N 5 82) Other Complicated formalities for refund, would not get the refund It might induce a tax audit in this business Complicated formalities for refund Refund would take longer time than offset We would not get the refund anyway It was not worth because the refund amount was too small 0 10 20 30 40 Percent of Respondents About 43 percent of eligible respondents did not apply be noted that VAT refund processing took more than for VAT refund in the last tax year. Although several 90 days for about 5 percent of the respondents. reasons were mentioned for not applying for the refund, the major ones included small refund amount 3.3.4 Perception Towards Tax (35.1 percent), belief they would not get the refund Inspections, Penalties and Appeals (15 percent), and complicated formalities for refund In this section we present views of taxpayers towards (15.3 percent) (Figure 3.36). Although the percentage of tax inspections, penalties and appeals. Concerning respondents who were in view of not getting the refund audit and inspection, on average only about 36 percent and complicated procedures is not big, these issues of them perceived to be happening for legitimate need to be addressed in order to maintain a healthy reasons (“inspections are really necessary to deter VAT system that has less impact on doing business evasion”), leaving the largest share to other reasons environment. such as perceptions that tax inspections and tax audits With regard to VAT refund process and period, happen as mechanisms of seeking informal payments respondents (that reported they had applied for VAT by tax officials (19 percent), officials inspect and audit refund) indicated that the average time taken by the taxpayers only to justify their jobs (19 percent), while tax office for processing refund was about 24.5 days 9 percent of tax inspections and tax audits happen (N5 22). About 80 percent, 85 percent and 95 percent because of someone’s interest to create obstacles to of respondents also reported that processing VAT others which would spoil the business environment. refunds took them up to 30 days, 60 days and 90 days A further 16 percent were perceived to be happening respectively (Figure 3.37). From Figure 3.37 it can as an additional revenue collection mechanism for the government (Figure 3.38). From the above discussion, it can be noted that only a small percentage of audits and Figure 3.37: Time Taken for VAT Refund Process inspections are perceived to be legitimate, which should (N 5 75) be further investigated to understand the prevalence of 100 this perception. 95.1 95 In comparison, perceptions among formal businesses in 90 Georgia were that more than 70 percent of inspections 85.6 were for “legitimate reasons” (see Annex 1). 85 80.9 In terms of the actual audit and inspection experience, 80 about 32 percent said their business had been inspected 75 or audited by tax authorities in the last two tax years (Figure 3.39). A significantly higher proportion 70 30 60 90 of relatively larger businesses (45.5 percent) faced 28  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.38: Perception about Tax Inspections (N 5 1002) Other reasons (N = 1002) 0.34 Inspections are used as an additional revenue collection mechanism for the government (N = 1002) 16.88 Because of someone’s interest to create obstacles 9.36 to others (e.g. to competitors) (N = 1002) Inspection officers are trying to justify their jobs 19.17 (N = 1002) Inspection officers are looking for 19.15 informal payments (N = 1002) Legitimate reasons, i.e., inspections are really 36.1 necessary to deter tax evasion (N = 1002) 0 5 10 15 20 25 30 35 40 Figure3.39: Inspections and Audits by Turnover 120 100 80 51.6 65.8 54.5 60 70.5 68.3 40 20 48.4 45.5 29.5 34.2 31.7 0 Less than ETB ETB 100,001- ETB 500,001- Above ETB Total (N = 1003) 100,001 (N = 501) ETB 500,000 ETB 10,000,000 10,000,000 (N = 224) (N = 234) (N = 44) Yes No inspections and audits compared to smaller businesses Figure 3.40: Number of Inspections and Audits and (29.5 percent which is generally considered an Time Spent Among Those Who Reported appropriate audit focus. Any Inspection 6 However, overall inspections are much lower in 4.5 4.3 4 3.4 countries that use risk based audit (RBA) for audit 2 selection. For example, as per FIAS (2008), small and 0 medium businesses (those with turnover under about Number of Times Calendar Days Person-Days $2 million) in South Africa11 in the year 2006 faced only about a 2 percent chance of inspection for Income Tax and Employees taxes and about 3 percent for VAT. Some spent about 3 person days during inspections and of the less populous provinces showed a slightly higher audits. It should be noted that the response on the likelihood of inspection (see Annex 1 for details). number of times (4.5 times on average) includes all types of inspections (desk audit, inspection of purchases With regard to the frequency of inspections, total days, and sales, SRM usage etc) and other visits of tax and total person days spent on inspections, those who officials. reported inspections said that their businesses were inspected between four to five times, each lasting for With respect to penalties, from all survey respondents about 4 days (Figure 3.40). The respondents said they (formal businesses) about 10 percent reported that they incurred penalties and or fines in the last two tax years. From respondents that incurred fines and or 11 South Africa is an example of a country with a well-functioning system of Risk Based Audit. penalties about half (47 percent) had filed an objection Survey Results   29 Figure 3.41: Whether or Not Appeals Were Filed Table 3.10: Reasons for Not Filing an Objection (N 5 146) (N 5 53) (percent) Audit outcomes, penalties and/or fines were fair 18.4 Appeals would cost too much 4.1 47.1 Appeal procedures take very long time 36.7 The advance payment to file appeal was too high 0.0 52.9 Amount of additional tax, penalties/fines was too 6.1 small We would not get a fair trial 34.7 It would induce additional tax audits in our 4.1 business Yes No Totals do not add up to 100 due to multiple coding nature of the question. (Figure 3.41). The appeals were higher for large taxpayers (about 67 percent), compared with just under Table 3.11: Appeals Resolved and Still in 47 percent for smaller businesses (Figure 3.41). Process (N 5 68) (percent) In respect of the reasons for not filing objections How many of these appeals were resolved to the 60.3 business’s satisfaction? against the decisions of tax officers a large percentage How many of these appeals were resolved not to 36.5 of respondents indicated the length of time appeal the business’s satisfaction procedures take and the feeling that fair trial would not How many of these appeals are still in process? 3.2 be obtained (36.7 percent and 34.7 percent respectively) as their reasons. Although it was not investigated in the The study further investigated the perception of survey, it might be because taxpayers lack confidence on respondents towards tax appeal related issues. While the quality of their self-assessment (Table 3.10). most respondents somewhat disagreed with the Table 3.11 shows the number of appeals resolved to statement that “Tax appeals process is quick and business’ satisfaction, number of appeals not resolved efficient”, most respondents said that “Tax officers to the business’ satisfaction and number of appeals do not ask for any informal payments during tax still in process. While about 60 percent of the appeals appeals”. The results with respect to efficiency of the were resolved to the business’ satisfaction, 37 percent appeal process appear to be consistent with the results of the appeals were not resolved to the business concerning the length of time the appeal procedures satisfaction. This sends the message that there is a need take as presented previously. These perceptions of to investigate further why a large number of appeal business taxpayers are likely to make the dispute decisions were not in favor of the tax authority and settlement system not being seen as a useful method what causes them. of resolving tax disputes and might force them to use Figure 3.42: Perception about Statements Related to Tax Appeals (N 5 68) Tax officers do not ask for any informal payments during tax appeals 2.97 Tax appeals process is fair and just 2.24 Tax appeals process is quick and efficient 2 Tax appeals process is simple and easy 2.46 1 2 3 4 5 Fully Disagree Fully Agree 30  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia an illegal means of resolving tax issues (like by bribing opinion, what percent of total annual sales do such officers and other mechanisms). It is, hence, important businesses in Ethiopia declare for tax purposes?”. Of to address the issues in the dispute resolution system. the total number of 991 respondents, about 7.1 percent believed that businesses like theirs would declare 3.3.5 Perceptions of Tax Compliance 100 percent of their total annual sales for tax purposes. and the Business Environment The remaining 92.9 percent believed that they would The study further assessed taxpayers’ perceptions declare, on average, about 53 percent of their total towards compliance and respondents were asked to rate annual sales for tax purposes. Generally, larger businesses general statements regarding the methods used by typical perceived better compliance than did smaller ones. businesses in Ethiopia to artificially reduce tax burden on In terms of overstatement in expenditures, of the total a scale of 1 to 5 where 1 means “Fully disagree” and 5 number of respondents, about 22 percent reported “do means “Fully agree.” As per the results, respondents view not know.” About 4 percent indicated that businesses like sales without invoices and use of fictitious invoices as the theirs overstated zero percent of actual expenditure in their main mechanisms used to evade taxes (Figure 3.43). tax returns. The remaining 74 percent of the respondents Table 3.12 provides average responses to the question believed they overstated, on average, about 38.4 percent of “having in mind businesses similar to yours, in your their actual expenditures in tax returns (Table 3.13). Figure 3.43: Opinions on Tax Compliance (N 5 1003) Businesses make unregistered sales (i.e., without sales 3.45 invoices) to artificially reduce tax burden (N = 1003) Businesses use fictitious/fake invoices to 3.1 artificially reduce tax burden (N = 1002) Businesses use fictitious/fake businesses to 2.95 artificially reduce tax burden (N = 1003) Businesses make payments of unofficial salaries to their 2.96 workers to artificially reduce tax burden (N = 1003) 1 2 3 4 5 Fully Disagree Fully Agree Table 3.12: Percent of Total Annual Sales Declared for Tax Purposes by Turnover (among those reporting less than 100%) Turnover (in ETB) Less than ETB 100,001 – ETB 500,001 – ETB Above ETB Total ETB 100,001 ETB 500,000 10,000,000 10,000,000 N 913 458 206 214 35 Percent of total annual sales 53.4 51.6 61.5 54.8 64.4 declared for tax purposes Table 3.13: Percent of Actual Expenditure Overstated in Tax Returns by Turnover Turnover (in Ethiopian ETB) Less than ETB 100,001 – ETB 500,001 – ETB Above ETB Total ETB 100,000 ETB 500,000 10,000,000 10,000,000 N 772 372 176 184 40 Percent of annual expenditures 38.4 38.7 36.3 37.3 53.1 overstated Survey Results   31 Figure 3.44: Respondents Perception about Taxes—Formal (percent) A business can carry out tax C (N = 408) 24.9 30.1 23.2 15.9 5.9 compliance process smoothly even without connections with B (N = 253) 29.1 17.5 23.5 26.1 3.9 tax officials A (N = 342) 30.9 24.3 18.6 17.9 8.3 C (N = 408) 10.7 28.7 27.0 27.7 5.9 Tax officials have integrity B (N = 253) 15.4 31.9 15.3 33.1 4.3 A (N = 342) 15.8 26.8 21.3 27.8 8.3 C (N = 408) 5.0 8.8 26.8 31.3 28.2 The use of sales register machines beneficiary to B (N = 253) 2.84.7 13.3 33.7 45.5 businesses which use them A (N = 342) 4.72.5 3.1 27.2 62.6 C (N = 408) 18.2 33.3 18.4 24.9 5.2 Tax officials are fair in their assessments B (N = 253) 23.7 31.3 18.3 22.2 4.5 A (N = 342) 22.9 23.5 23.1 19.3 11.2 C (N = 408) 4.6 8.5 10.8 38.6 37.5 Business would be willing to pay more tax if corruption B (N = 253) 8.0 6.2 15.1 30.4 40.3 is not prevalent A (N = 342) 4.2 6.2 8.2 36.3 45.1 C (N = 408) 15.8 24.5 22.4 28.2 9.2 It is justifiable if a business under reports its income B (N = 253) 20.2 21.3 28.4 18.3 11.8 in order to pay less tax A (N = 342) 26.9 18.7 13.9 25.4 15.1 C (N = 408) 8.1 26.0 19.7 37.2 9.0 Tax officials are competent and knowledgeable B (N = 253) 15.4 30.5 15.4 32.4 6.4 A (N = 342) 10.1 31.9 15.5 36.0 6.5 The process of complying C (N = 408) 12.0 24.8 18.1 32.4 12.8 with taxation is more burdensome than the B (N = 253) 7.4 21.0 17.4 30.8 23.4 amount of tax itself A (N = 342) 7.7 22.4 9.6 40.2 20.2 Taxes paid to the C (N = 408) 3.6 7.7 7.6 34.9 46.2 government are ultimately used for the benefit B (N = 253) 9.0 5.9 5.1 38.3 41.7 of the general public A (N = 342) 2.4 7.1 11.0 30.9 48.6 0.0 20.0 40.0 60.0 80.0 100.0 120.0 Fully Disagree Somewhat Disagree Neither Agree Nor Disagree Somewhat Agree Fully Agree Both of these results (misstatement of sales and changing the perceptions of taxpayers towards others’ expenditures) show how respondents perceive their compliance behavior. peers’ tax compliance behavior, which is expected to In addition, the perception of firms about various reflect their own compliance decision. As preliminary aspects of taxes is another item that the survey findings of a study by Yesegat and Fjelstad (2014) attempted to capture (Figure 3.44). The responses show revealed, there is a statistically significant relationship that almost about 80 percent of all categories of firms between taxpayers perception of others’ compliance either agree fully or somewhat, that the taxes paid to and their own attitudes towards compliance; the latter the government are ultimately used for the benefit of is expected to affect taxpayers actual compliance the general public, which is a positive indicator of tax behavior. It is hence important for the government to morale. Also, about 90 percent of firms in category A, work not only improving compliance behavior but also 32  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia 80 percent of those in category B, and 60 percent for the benefit of the general public and businesses. in category C perceive that the use of sales register Further, respondents seemed to agree that there are machine is beneficial to businesses which use them. many benefits for paying taxes. However, they also believe that businesses would be willing to pay more However, about 46 percent of firms in category A, taxes if corruption is not prevalent (Figure 3.45). 55 percent of those in category B, and 51 percent of those in category C disagree or somewhat disagree that In addition, considering the severity of some factors the assessment of tax officials is fair. Most businesses that are commonly mentioned as constraints faced by also tend to agree that “businesses would be willing businesses operating in developing countries, perception to pay more tax if corruption is not prevalent”, and of the business environment was assessed. Figure 3.46 disagree that “A business can carry out tax compliance shows a relatively larger proportion of formal process smoothly even without connections with tax businesses indicated that access and cost of finance is officials.” either a very severe problem or major problem affecting business performance. Another interesting finding is that about 60 percent, 54 percent and 44 percent in categories A, B and C Similar to formal businesses, among the statements respectively fully and somewhat agreed that the process presented to informal businesses, most respondents of complying with taxation is more burdensome than agreed to that access to finance (availability and cost) the amount of tax itself. and availability and cost of working premises pose moderate to major problem to the business environment Similar to formal businesses, informal businesses agreed (Figure 3.47). that taxes paid to the government are ultimately used Figure 3.45: Perceptions about Taxes and Tax System—Informal (N 5 499) All information about tax compliance is easily available 2.68 Every business should pay taxes 2.94 Tax rates for businesses like mine but registered for 3.17 tax are too high Keeping records for tax purposes is time-consuming 3.2 and expensive The process of complying with taxation is more burdensome 3.25 than the amount of tax itself Businesses registered for tax are often subjected 3.34 to audits and inspections Professionals bookkeepers, accounting and 3.41 tax advice are too expensive Tax procedures are too complicated 3.41 Most businesses do pay taxes 3.41 Tax authorities are unfair 3.41 Businesses are willing to pay more tax if corruption 3.77 is not prevalent There are many benefits for businesses that pay taxes 3.87 Taxes paid to the government are ultimately used for the 4.12 benefit of the general public and businesses 1 2 3 4 5 Fully Disagree Fully Agree Survey Results   33 Figure 3.46: Respondents Perception about Business Environment—Formal (percent, N 5 1003) Access to piped water 49.4 18.7 14.0 17.9 Access to tax related information 36.0 25.9 22.3 15.8 Internet system failure 84.9 6.7 3.3 5.1 High staff turnover 62.5 17.6 11.0 8.9 Access to foreign exchange 83.4 6.4 3.6 6.7 Frequent change in policy 47.3 21.5 15.3 15.9 Transport excluding infrastructure 47.0 23.6 13.5 15.9 Access to road infrastructure 50.1 21.4 12.2 16.3 Various inspections 37.7 29.4 18.5 14.5 Availability and cost of business premises 28.2 19.5 15.7 36.6 Quality of telecommunication 26.9 22.0 20.9 30.4 Tax administration 26.3 22.7 22.8 28.3 Tax rates 21.5 21.3 25.7 31.6 Institutional stability 52.1 23.9 14.5 9.6 Labour regulations 77.2 14.7 5.03.1 Access to skilled labour 66.1 16.7 7.3 9.9 Access and quality of electricity 27.8 19.5 17.5 35.3 Customs and trade regulations 48.4 24.7 15.2 11.6 Crime 55.7 23.7 12.8 7.7 Access to legal system 45.7 30.9 13.4 10.0 Corruption 46.8 21.3 15.9 15.9 Business licensing 50.0 24.7 15.4 9.9 Access and cost of finance 12.6 11.9 20.7 54.8 0.0 20.0 40.0 60.0 80.0 100.0 120.0 Not a Problem Minor Problem Moderate Problem Major + Very Severe Problem Figure 3.47: Perception on Elements of Business Environment (N 5 499)—Informal Access to finance (availability and cost) 1.2 Availability and cost of business premises 1.2 Business licensing and permits 1.6 Various inspections 1.6 Transportation issues excluding infrastructure 1.7 Quality of telecommunications services 1.8 Access to piped water 1.9 Access to legal system 1.9 Access to and quality of electricity 1.9 Corruption 2.1 Crime, theft and disorder 2.1 Access to road infrastructure 2.2 Tax administration 2.2 Customs and trade regulations 2.2 Tax rates 2.5 Institutional instability 2.6 Labour regulations 3.3 Access to skilled labour 3.7 1 2 3 4 5 Not a Problem Very Severe 34  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.48: Perceptions about Calculating and Filling in Tax Returns Employment related contributions (N = 340) 8.7 41.9 49.4 Withholding income tax on payments (N = 228) 4.6 45.2 50.3 Turnover Tax (ToT) (N = 260) 19.4 45.9 34.7 Value Added Tax (VAT) (N = 412) 14.9 50.2 34.9 Business Profit Tax (N = 737) 25.4 47.4 27.2 0 20 40 60 80 100 120 Difficult Manageable Easy 3.3.6 Difficulties with the Tax System The overwhelming majority of respondents considered Formal business respondents were asked about the submission of tax returns and payments as manageable difficulty of calculating and filling in tax returns and for all tax types. Among the different types of taxes, about 25 percent of those paying business profit taxes it was easier for firms to submit tax returns and pay reported that calculating and filling in tax returns withholding income tax on payments than other forms were difficult, while 47 percent viewed it manageable of taxes. (Figure 3.48). Similarly, close to half of respondents Informal businesses were also asked about their considered calculating and filling in tax returns to be perceptions about the difficulty of filling out and manageable for all types of taxes. However, 15 percent submitting tax returns. About 34 percent of respondents and 19 percent of the VAT and TOT payers respectively said that filling out and submitting tax returns as quite considered the process to be just easy. a problem or extremely problematic, with variations When respondents were asked about the level of across locations. About 41 percent and 30 percent of difficulty in submitting forms and paying taxes, respondents in Hawassa reported that this activity as about 31, 18 and 28 percent thought it was relatively quite a problem or extremely problematic, respectively difficult to submit tax forms and pay for business (Figure 3.50). profit tax, VAT and TOT, respectively (Figure 3.49). Figure 3.49: Perceptions about Submission of Tax Returns and Payments Employment related contributions (N = 339) 43.1 45.0 12.0 Withholding income tax on payments (N = 228) 47.0 41.1 11.9 Turnover Tax (ToT) (N = 261) 29.6 42.4 27.9 Value Added Tax (VAT) (N = 405) 33.1 49.3 17.5 Business Profit Tax (N = 739) 26.5 42.6 30.9 0 20 40 60 80 100 120 Easy Manageable Difficult Survey Results   35 Figure 3.50: Perception about Filling Out and Submitting Tax Returns—Informal Total (N = 199) 11 29 53.6 26.7 7.7 Addis Ababa (N = 390) 9.5 21.3 42.1 20.5 6.7 Hawassa (N = 23) 6.3 34.8 34.8 21.7 8.7 Adama (N = 32) 34.4 31.3 25 3.1 Bahir Dar (N = 22) 9.1 50 36.4 4.5 Mekelle (N = 32) 6.3 34.4 50 9.4 0 20 40 60 80 100.0 120 140 Percent of Respondents Not a Problem at All A Little Problematic Not Sure Whether it’s a Problem or Not Quote a Problem Extremely Problematic Businesses were asked to indicate which information Although the direct ERCA sources of information were sources on taxation were available to them. For not perceived as readily available to most respondents, those businesses that indicated the availability of an among those who could use them, they were considered information source, they were also asked whether they “preferred” sources by large majorities: about three used that information source. The extent of preference quarters of such businesses (for the Bulletin boards and of businesses was also examined for those information the website) and over 85 percent for the Circulars and sources they used. Figure 3.51 shows the percentage of Publications. businesses for which the information source is available for possible use in relation to taxation. Website of tax 3.3.7 Computer and Bank Account Usage authorities, e-mails and SMS/text message were the and Other Issues least frequently reported sources of information that The survey data show low penetration of computers in are available. The availability of each of these three the business firms in the country, (and given that the sources was mentioned only by about 7 percent of the survey tended to ignore businesses in smaller towns, the businesses or less. Television/cable, radio and family overall figures for businesses in the country are likely to and friends were the three most frequently mentioned be even lower). As shown in panel (a) of Figure 3.52, sources available to obtain information on taxation. Figure 3.51: Percentage of Businesses Who Stated Availability of Information Source on Taxation SMS/Text message (N =1003) 4.2 Community members/Peers in business (N = 1003) 47.8 Family and friends (N = 1003) 66.5 Specialist tax consultants/Professionals (N = 1003) 19.2 Workshops etc. organized by tax offices (N = 1003) 42.8 Newspapers (N = 1003) 38.9 Radio (N =1003) 71.5 Television/Cable (N = 1003) 78.0 e-mails (N = 1003) 2.7 Bulletin boards at tax offices (N = 1003) 45.4 Circulars (N = 1003) 19.5 Publications of tax offices (N = 1003) 25.7 Website of tax authorities (N = 1003) 7.2 36  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 3.52: Firms’ Use of Computers (percent) (a) Does the establishment use computers (b) Does the firm use computers for for business purposes? (N = 1003) accounting/tax purposes? (N = 289) 12 37 63 88 Yes No Yes No Figure 3.53: Firms’ Use of Internet of whether the firm uses internet for business purposes. 100.0 91.3 As reported in panel (a) of Figure 3.53, only about 80.0 9 percent of total firms in the sample reported that 60.0 54.5 45.5 40.0 they use internet for business purposes. Those who 20.0 8.7 responded that they use internet for business purposes 0.0 Use of Internet for Use of Internet were asked if they use it to obtain and process Business Purposes for Tax Accounting information for tax purposes. Only 45.5 percent of (N = 1003) Purposes (N = 198) these firms reported that they use internet for reasons associated with tax (see panel (b) of Figure 3.53). Yes No The majority of informal respondents do not use the largest proportion of firms in the sample (more computers including laptops for business purposes. than 88 percent) reported that they do not use They do not use internet either. Only 0.80 percent computers for business related activities and only about of the respondents used computers and internet for 12 percent responded favorably. Of the sample firms business purposes. that responded positively to the question of whether the The data on firms’ use of bank accounts also show firm uses computers for business purposes or not (i.e., that the number of firms that use bank accounts for less than 12 percent of the total sample), 63 percent their operations is low. As reported in panel (a) of reported that they use computers for accounting and/or Figure 3.54, only about 45 percent of the firms in the tax purposes while the remaining 37 percent reported sample reported that they use bank account for their that they do not use them for such purposes (see panel transactions. This number seems low given that the (b) of Figure 3.52). firms are formal firms and they are all situated in the Similar to that of computer use, only a very small biggest urban areas of the country. Of the sample firms number of firms responded positively to the question that responded using bank accounts in their operations, Figure 3.54: Firms’ Use of Bank Account (a) Does the establishment use bank accounts (b) Is the bank account specific to in its business operations? (N = 1003) the establishment? (N = 594) 27.9% 44.8% 55.2% 72.1% Yes No Yes No Survey Results   37 Figure 3.55: Use of Bank Account for Business Operations (by legal status) 100 100 80 79.4 75 57.8 60 42.2 40 20.6 25 20 0 Individual/Sole Private Limited Cooperative Society Others (N = 14) Proprietorship Company (N = 139) (N = 52) (N = 798) Yes No only about 28 percent of them indicated that the bank Figure 3.56: Use of Bank Account (informal) account is specific to the business establishment (see (percent) (N 5 499) panel (b) of Figure 3.54). This is not surprising since 16.4 most of the firms in the sample (about 93 percent of the total sample) are sole-proprietorships where the distinction between the transactions of the business and the owner are mixed. Use of bank account for business operations in terms of legal status shows that not surprisingly sole-proprietorship businesses turn out to be those 83.6 using bank account the least. As can be seen from Figure 3.55, about 58 percent of sole-proprietorship Yes No firms indicated that they do not use bank account for business operations. The proportion of firms that use sole-proprietorship, small in size, and make their bank account is larger in the Private Limited Company transactions by cash than through bank account. ownership structure. In general, the survey data show the limited use of In the case of informal businesses, about 16.4 percent of computer, internet and bank accounts by businesses respondents used a bank account for business activities. (both formal and informal). However, only 1.2 percent of them had a separate bank In addition to the above, when asked about whether or account for business operations, indicating that the not firms or any of their employees made any business- majority of respondents used a private bank account for related visits to tax offices, a large proportion of category business operations (Figure 3.56). A and half of B taxpayers made visits to tax offices. In line with the discussion above, on the use of bank However, about 41 percent of category C taxpayers account by the business firms in the sample, only reported such visits. Although businesses are required to about 26 percent of transaction turnover of the whole make business-related visits to tax offices, outsourcing firms in the sample is through bank. The magnitude tax-related functions could be a reason for some of these transactions are about 40 percent for firms in businesses, especially for category A and B taxpayers not category A, about 29 percent for firms in category B, visiting tax offices. On average, a business firm made and about 10 percent for category C firms. The results about 5 visits in the previous tax year (Table 3.14). are, again, consistent with the size and characteristic In addition, informal business survey respondents were features of the firms in the three tax payers’ categories. asked to rate general statements related to knowledge That is, as discussed earlier most of the firms are 38  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia In addition to the knowledge about tax laws and Table 3.14: Business-Related Visits to Tax Offices—Formal regulations informal business survey respondents were asked about various tasks of business operations Category such as keeping accounting records and books for   Total A B C the business. Only 11 percent reported that keeping N 1003 342 253 408 accounting records would not be a problem while the Yes 43.2 58.6 50.0 40.8 remaining (89 percent) indicated a little problematic, No 56.8 41.4 50.0 59.2 quite a problem, extremely problematic and not sure whether it is a problem or not (Figure 4.58). of tax regulations and requirements. About 33 percent Similar to the above picture, in all locations considered indicated that they were knowledgeable or very less than 10 percent of respondents indicated that knowledgeable about tax regulations and requirements keeping accounting records is not a problem at all. while the rest were either not knowledgeable, not sure Specifically, in Hawassa, there were not respondents or somewhat knowledgeable. The results by location who said “not a problem at all”. These results in also reveal that only 14 percent of respondents in relation to the perceptions of keeping accounting Bahir Dar, were found to be knowledgeable or very records and knowledge about tax laws indicate the need knowledgeable while the remaining 86 percent were for the tax administration, the government at large, to either not knowledgeable, not sure or somewhat work on awareness creation not only about tax laws knowledgeable (Figure 4.57). but also record keeping. Figure 3.57: Knowledge about Tax Information and Requirements by Location—Informal Total (N = 199) 16.8 35.7 14.8 26.7 6.0 Addis Ababa (N = 390) 19.0 35.1 15.4 24.4 6.2 Hawassa (N = 23) 4.3 56.5 39.1 Adama (N = 32) 3.1 46.9 15.6 31.3 3.1 Bahir Dar (N = 22) 31.8 27.3 27.3 13.6 Mekelle (N = 32) 3.1 21.9 9.4 50.0 15.6 0.0 20.0 40.0 60.0 80.0 100.0 120.0 Percent of Respondents Not Knowledgeable at All Somehow Knowledgeable Not Sure Whether Knowledgeable or Not Knowledgeable Very Knowledgeable Figure 3.58: Perception about Keeping Accounting Records and Books for the Businesses—Informal Total (N = 199) 14.0 23.2 37.7.8 19.8 5.2 Addis Ababa (N = 390) 14.6 22.8 38.2 19.2 5.1 Hawassa (N = 23) 30.4 30.4 30.4 8.7 Adama (N = 32) 15.6 28.1 31.3 21.9 3.1 Bahir Dar (N = 22) 18.2 4.5 36.4 27.3 13.6 Mekelle (N = 32) 12.5 31.3 43.8 12.5 0.0 20.0 40.0 60.0 80.0 100.0 120.0 Percent of Respondents Not a Problem at All A Little Problematic Not Sure Whether It’s a Problem or No Quite a Problem Extremely Problematic Survey Results   39 4 Conclusions and Recommendations Using the survey data the study attempted to estimate Such findings are not unusual internationally. On the and analyze tax compliance costs in Ethiopia for the other hand, it is often critical to take steps to reduce year 2012/13. The average tax compliance costs per the compliance burden on small businesses as much as business were estimated to be ETB 6,753 (USD 360) possible. When tax compliance costs are 5 percent of (including the acquisition and maintenance costs) and turnover, it is the equivalent of an additional 5 percent ETB 5,520 (USD 297) (excluding the acquisition and turnover tax (which, even assuming a generous profit maintenance costs). The total compliance costs of margin of 20 percent, would be equivalent to a taxes in Ethiopia in the year 2012/13 were estimated 25 percent profit tax on top of all other taxes being to be ETB 6.7 billion (about USD 360 million). paid by a small business). Such a burden reduces the This is between 4.5 and 6 percent of Ethiopia’s total competitiveness of businesses, especially, small domestic government revenue collected in 2012/13 (depending on businesses. whether tax compliance costs include acquisition and In addition, the study revealed the tax compliance maintenance cost of software and hardware) and about cost burden by tax type and by sector. As results 1 percent of its GDP during the same year. showed the compliance costs of business profit tax Examining tax compliance costs estimates as a share of is much higher than that of other types of taxes for turnover shows that it was an average of 5.3 percent in all business categories. Following business profit tax, the year under consideration. The tax compliance costs VAT and TOT compliance costs were found to be (in absolute terms) were found to be larger for category significant. This high burden of tax compliance costs “A” businesses than that of category “C” businesses. on relatively smaller businesses is likely to deter them The results presented earlier showed that the TCCs/ from fulfilling the requirements imposed on them. The turnover ratio for smaller businesses is high compared burden of compliance costs increases the cost of doing to that of larger businesses, providing evidence that business. Analyzing the burden of tax compliance costs tax compliance costs in Ethiopia are regressive. That by business sector also revealed that the trade sector is, small businesses bore disproportionately higher (both wholesale and retail trades) bore a heavier tax compliance costs burden compared to large businesses. compliance cost burden than others. Survey respondents (especially smaller ones) also In addition to the estimation of tax compliance costs perceived the burden of tax compliance as being higher for formal businesses, the study assessed taxpayers’ than the amount of tax payable.   41 perceptions on the advantages and disadvantages of and also the burden of tax audits and inspections registering for taxes and the business environment in necessitated the review of the system in light of the general (by both formal and informal businesses), as international experience and the reality in the country well as experiences and perceptions about VAT and and replacement of the regime by a more simplified and refund process, SRM usage, inspections and audits, more objective system similar to the practice in other penalties and appeals, and communications with ERCA developing countries. (for formal businesses). High tax rates/ burden and Many other countries have greatly simplified taxes complicated procedures were found to be the biggest for micro-enterprises, taking into account both their disadvantages of registering for taxes by both formal very low earnings and their relatively lack of capacity and informal businesses. High price of goods and for detailed bookkeeping. While the desire to improve services compared to VAT unregistered businesses was “fairness” (vertical/horizontal equity) is usually the perceived to be the biggest disadvantage of registering motivation for large matrices of sectors and business for VAT while “correction of errors” was found to be sizes, these systems sacrifice simplicity for the hope of the biggest difficulty in using SRM. fairness, but are usually ineffective in achieving fairness. Recommendations Many studies of micro-enterprises show that their sales In addition to the relatively high estimated compliance are highly variable from year to year and their profit costs, the perception of high tax burden/ tax rates margins range from substantial losses to quite high and complicated tax procedures were mentioned as profit margins, which also vary widely over the years “disadvantages of registering for taxes” by both formal and across locations. Thus many countries have opted and informal business survey respondents. Simplifying for much simpler fixed-tax systems (often administered the tax system and reducing the burden of tax at the regional or local level, as is already the case in compliance is critical for encouraging formalization of Ethiopia), along with encouragement to keep proper businesses and fostering a positive perception towards books (see IFC (2007), pp 60–72). Such a system could the entire tax system. offer a simple lump-sum (with very limited cells in a matrix, e.g., city versus town/village; services versus Specifically, in order to reduce the tax compliance trade and manufacturing) that would substitute for both burden on smaller businesses and also encourage income tax and VAT. However, ERCA would also have formalization, simplification of the tax regime for to take steps to mitigate abuse of the system (which micro and small enterprises is worth considering. As often becomes a problem, as seen in other countries) by can be seen from the tax compliance costs burden, under-declaring turnover. the perception of taxpayers and also from a review of the micro and small business taxation legal The trade sector (both wholesale and retail trades) framework, the regime is complex, time-consuming accounted for a large proportion of the business and costly. For category C taxpayers, there are over environment in Ethiopia. Heavier tax compliance costs 80 categories of sectors and nearly 20 turnover bands. burden on this segment implies that this sector is likely In the implementation of the legal framework, the tax to be deterred from fulfilling tax requirements. administration relies on daily sales estimates of each In respect of the disadvantages of registering for taxes and every taxpayer under that regime. The estimation both formal and informal businesses showed that high uses mostly such subjective criteria as location and tax burden/ tax rates as the biggest and the second nature of the business, the volume of transactions, biggest disadvantages of registering for taxes. Especially, expenditures for business and non-business purposes. in the case of smaller businesses this was found to be The practice of taxing micro businesses coupled with the biggest concern apart from complicated procedures. the tax compliance burden as revealed by the relatively Informal operators further indicated that financial high tax compliance costs of smaller businesses burden of being formal (including taxes and fees) and 42  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia lack of clarity of tax laws and regulations as major adjusting business categorization thresholds which have reasons for not applying for tax registration. Following been eroded by inflation is worth considering. this, informal businesses indicated that reduction in Steps should also be taken to simplify category A tax burden/tax rates as one of the factors that would taxation. For example, it might be helpful to eliminate encourage them to register for taxes and operate in sector selected registration requirements for VAT. It the formal sector. The very low income tax exempt might also be beneficial to reduce the number of goods threshold (ETB 1800 per annum, which is only about subject to excise (e.g., eliminating excise taxes for salt, US$100) might be one of the reasons for such a view sugar and soft-drinks). Another important measure especially by smaller businesses. Reviewing the tax should be to consider reviewing the withholding tax rate structure (along with the tax exempt threshold) (on payments and imports) against the profit tax of in addition to simplification of the entire tax system is suppliers, which tends to discourage use of domestic, important in reducing the burden of taxes and changing small business suppliers. the attitudes of taxpayers and also encouraging formalization of businesses. With regard to the disadvantages of registering for VAT, most formal businesses noted higher prices of The income tax exempt threshold should be set at goods and services compared to others (i.e., non-VAT approximately the poverty level, and adjusted regularly competitors) as the biggest disadvantage. Such a result over time to account for inflation and other relevant is consistent with what taxpayers usually raise as a factors. The VAT threshold also needs to be adjusted concern at various forums. Such a problem could be periodically for inflation, and is usually set at the lower because businesses that were required to register for turnover level for a “medium” size business (taking into VAT have not been registered; and/or after registration account also the capacity for the necessary bookkeeping if there are businesses that do not issue invoices, such for VAT compliance). The current threshold, at a complaint is likely to arise. In the case of the latter, about the equivalent of US$25,000 is quite low by the survey revealed sales without invoices as the most international standards. The fact that about half of common ways that businesses use in reducing their businesses that are registered for VAT are outsourcing tax liability. Under-declaration of sales is always a at least some of the compliance burden suggests that problem when most transactions are in cash. However, the complexity is more than many Ethiopian businesses as a growing proportion of transactions take place can currently handle on their own. The relatively high electronically or through the banking system (including compliance costs associated with VAT also suggest mobile-phone payments such as M-PESA in Kenya and that the burden is onerous for the smaller businesses. M-declaration in Rwanda), documentation of sales will By comparison, the VAT threshold in South Africa become increasingly more comprehensive. is currently about US$85,000; Kenya’s is about US$55,000 (see Appendix 2). Another way to help increase the culture of requesting for receipt by consumers could be useful. One In between poverty-level micro-enterprises paying a fixed instrument that could be considered in this regard is tax and medium-large businesses paying regular profit consumers’ receipt “award” as used in other countries tax and VAT, it is often helpful to offer an intermediate including Rwanda. regime for “small” businesses that can pay a turnover tax in lieu of both profit tax and VAT (with an option Furthermore, the study found the average time taken for any business to register for regular profit tax and by the tax office for processing VAT refund to be about VAT voluntarily). While some countries offer more than 24.5 days. Although this study doesn’t attempt to see one rate (e.g., a lower TOT rate for trade and a higher the nature of VAT refund requests, this result appears rate for services), the structure should be kept simple, to be applicable to those businesses whose refund and the revenue administration should offer training requests are allowed to be processed within a short to encourage proper bookkeeping. In addition to this, period of time. These businesses are mostly engaged Conclusions and Recommendations   43 in export; there are, however, others engaged in non- (2013) also noted correction of errors as one of the export business and have refund claims due to the difficulties in using SRM; this is because correction of existence of VAT excess credit (for over 5 months)12 errors is not automatic and needs some paper work and and or the operation of VAT withholding scheme. For repeated visit to the tax office. Further investigation of the such taxpayers, the average time to process VAT refund practice of correcting errors in using SRM and identifying is likely to be more than the above mentioned period. possible ways addressing this concern of taxpayers is As the data maintained by ERCA shows the number crucial. Further, it is important to create awareness among and volume of VAT refund requests is increasing; this taxpayers about the benefits of using SRM. is creating delays in processing refund requests and As reported earlier in the results section, the inefficiency is imposing burden on not only taxpayers but also in the appeal process and also the length of time the tax administration. In this context and given the the procedures take were considered to be issues in base is quite low, the above mentioned result has to be the dispute resolution system. These perceptions of examined with caution. business taxpayers are likely to make the dispute ERCA should try to expedite payment of refunds to settlement system not being seen as a useful method taxpayers to help businesses avoid unnecessary cash of resolving tax disputes and might force them to use flow difficulties. In particular, adoption of risk based an illegal means of resolving tax issues (like by bribing refund processing and also accepting bank bonds as officers and other mechanisms). In addition to the a security for VAT refund payment (risky businesses) above, examination of the views pertaining to appeal could be considered as suggested at Ethiopian Chamber decisions reveals that relatively a small proportion of of Commerce and Sectoral Association’s (ECCSA) appeal cases were resolved in favor of the tax authority. forum (ECCSA 2014). Although this result needs to be substantiated through examination of the actual appeal decisions, it sends the The compliance cost is high for VAT. It is recommend message that there is a need to investigate why a large to reduce the frequency of VAT filing (e.g., make it number of appeal decisions were not in favor of the quarterly), for relatively smaller businesses. tax authority and what causes them. It is, therefore, The other finding is that a substantial proportion of important to investigate the entire dispute resolution respondents did not apply for VAT refund; this was system and address the issues in the system. due to the view that, among others, refund amount The other interesting finding is that only 36 percent was small (35.1 percent), belief they would not get of inspections and audits were considered to be for the refund (15 percent), and complicated formalities legitimate reasons, leaving the largest share to other for refund (15.3 percent). Although the percentage of reasons. These other reasons include tax inspections respondents who were in view of not getting the refund and tax audits happen as mechanisms of seeking and complicated procedures is not big, these issues informal payments by tax officials, officials inspect and need to be addressed (through awareness creation and audit taxpayers only to justify their jobs, and someone’s making payments for genuine refund requests) in order interest to create obstacles to others. From the above to maintain a healthy VAT system that has less impact discussion, it can be noted that only a small percentage on business environment. of audits and inspections perceived to be legitimate. In terms of SRM usage, although respondents noted In terms of actual audit and inspections, about the existence of advantages, the difficulty of correcting 30 percent of smaller business showed that they errors was found to be the biggest disadvantage of using faced audit and inspections. This result evidences the the machine. Other studies such as Yesegat and Fjelstad burden on relatively smaller businesses due to tax authorities’ audit and inspections. With regard to the 12 Non-export businesses are required to carry over their VAT frequency of audits and inspections, the average was credits to the next five accounting period (five months) before applying for VAT refund. found to be 4.5 times; it should be noted that this 44  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia 4.5 average frequency of audits includes all types of In terms of bank account usage, as the results showed, inspections (desk audit, inspection of purchases and of those that used bank accounts only a small sales, SRM usage etc) and other visits of tax officials. proportion had bank account specific to the business The frequency of audits and inspections shows the establishment. This result evidences the limited use of extent of frequent contact between taxpayers and tax banking by businesses and also mixing personal and officials which might create room for corruption (which business accounts. Overall, the survey results suggest in turn might account for the fact that businesses about 75% of all transactions take place in cash, perceived about 20% of inspections to be motivated by including about 90% for category C taxpayers and “inspection officers looking for informal payments”). 60% for category A taxpayers. This signals the need to In addition, this result might have some link to the view exercise caution by the tax administration in imposing of respondents that only about 36% of the audits and income tax on bank deposits as that would increase inspection happen for “legitimate reasons.” taxpayers’ mistrust on the tax authority and reduce the incentive to use banking in business operations. The above mentioned issues pertaining to audit and inspection might be due to the fact that the tax Businesses typically visit tax offices, and this is more administration’s failure to discuss with businesses (to so for category A and B tax payers. Furthermore, the explain why the business has been selected for an audit in-house cost of compliance for filing taxes is higher and also the audit results to convince the taxpayer that than that of pre-filing and post-filing activities. It is, the audit finding is justified). Furthermore, the audit therefore, recommended that electronic filing (efiling) of compliance burden on relatively smaller businesses tax reports and payments, and mobile phone payments signals two things: the compliance burden on smaller/ should make it possible to considerably reduce the businesses, and the pressure in using the scarce tax time costs and the overall tax compliance costs for administration resources. Therefore, to address these businesses. It is known that the tax authority has issues in relation to audit and inspection, as part of launched using efiling for larger businesses (medium building a modern tax administration, improving the business to some extent) in Addis Ababa. Extending the risk based audit system is an important consideration in efiling system to those taxpayers who are able to use the addition to holding discussions with businesses about system is worth considering. Furthermore, for the efiling tax audits and their outcomes. system to be effective in reducing the compliance costs of taxpayers, it has to be supplemented by epayment or With regard to others’ compliance behavior, a large payment through the banking system. Payment using majority of businesses perceive the existence of mobile phone is also recommended to be considered as misstatement of both sales and expenditures for the one option in reducing the compliance costs burden, purpose of reducing taxes. Such a perception of others’ especially, on smaller businesses. noncompliance is expected to have impact on their own compliance decision. As preliminary findings The fact that 20 percent of category A and 49 percent of a study by Yesegat and Fjelstad (2014) revealed of category B tax payers do not use sales register there is a statistically significant relationship between machine calls for the need for awareness creation and taxpayers’ perception of others’ compliance and their education on the advantages of using of sales register own attitudes towards compliance; the latter is expected machine for business activities. to affect taxpayers actual compliance behavior, as Both for outsourcing and in-house tax accounting, the literature shows. It is, hence, important for the compliance cost is the highest for business profit tax. tax authority to work towards not only improving Turnover tax and value added tax also contribute compliance behavior but also highlighting positive a lot to compliance cost depending on the business evidence of good compliance behavior (e.g., annual category considered. We also find that the total award ceremonies in each region for “good taxpayers”). Conclusions and Recommendations   45 compliance cost as a share of turnover is generally viewed the task of bookkeeping as problematic. high for Ethiopia. Thus, total compliance cost could be These issues call for the government, in particular, tax reduced if a comprehensive capacity building measures administration’s efforts in creating awareness among are taken such as to identify factors that contribute informal traders. to high compliance cost in general and outsourcing in The fact that about half of respondents indicated particular. that they are likely or very likely to register for tax in As shown earlier in section 3, only a very small the coming two years and about 24 percent have not proportion informal business survey respondents decided means that more education and awareness indicated that they are very knowledgeable or creation is required for informal operators and in knowledgeable while the majority were not or were particular providing support and assistance is critical in unsure. In addition, the majority of respondents the formalization process. 46  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia References ´ H. (2004), Tax Compliance Costs of Small Blažic, MoFED (2013), Macroeconomic performance Business in Croatia. Institute of Public Finance. assessment of Ethiopia: achievements, challenges Coolidge, J. (2012),“Findings of tax compliance cost and the foundations for sustainable and inclusive surveys in developing countries,”eJournal of tax growth and transformation (1991/92–2011/12), research, 10(2): 250–287. Addis Ababa, Ethiopia. FDRE (1995), The Constitution of the Federal Sandford, C. (1995), Taxation Compliance Costs: Democratic Republic of Ethiopia of 1995. Measurement and Policy, Fiscal Publications, Bath. Articles 95–99, Negarit Gazeta, Year 1, No. 1. SEDF (2012), Nepal tax compliance cost survey FDRE (2002), Value Added Tax Proclamation, report-2012, International Finance Corporation Proclamation No. 285/2002, Negarit Gazeta, (IFC), World Bank Group, IFC Advisory Services in Year 8, No. 33. SouthAsia. FDRE (2005), Reorganization of the Executive Susila, B. and Pope, J. (2012), The magnitude and the Organs of the Federal Democratic Republic of features of tax compliance costs of large companies Ethiopia of 2001, Proclamation No. 256, Negarit in Indonesia”, Australian Tax Forum Volume 27 Gazeta, Year 8, No. 2 (repealed and replaced by No. 4. Proclamation No. 471/2005). Yesegat, W. (2009), Value added tax in Ethiopia: A FDRE (2008), Ethiopian Revenues and Customs study of operating costs and compliance, PhD Authority (ERCA) Establishment of 2008, Article 9, thesis, Faculty of Law of the University of New Proclamation No. 587, Negarit Gazeta, Year 14, South Wales, Australia. No. 44. Yesegat, W. and Fjelstad, Odd-Helge (2013), view of International Finance Corporation (IFC) (2007), paying taxes in Ethiopia: preliminary qualitative Designing a Tax System for Micro and Small results, a paper presented at the International Businesses: Guide for Practitioners. Center for tax and Development (ICTD) annual International Finance Corporation (IFC) (2010), Tax conference held from December 11–13, 2013, compliance cost surveys: using data to design Lome, Togo. targeted reforms, practice Note No. 8. Yesegat, W. and Fjelstad, Odd-Helge (2014), Views of International Finance Corporation (IFC) (2011), Tax paying taxes in Ethiopia: Preliminary survey results, Perception and Compliance Cost Surveys: A Tool paper presented at the ICTD’s annual conference for Tax Reform. held from Dec 07–09, Arusha, Tanzania. Klun, M. (2004), Compliance Costs for Personal Income Tax in a Transition Country: The Case of Slovenia. Fiscal Studies 25 (1).   47 Annex Tax Compliance Costs— 1 International Comparisons for Ethiopia13 1.  Introduction13 The core questions of interest are about the time and cost of basic tax compliance tasks, including primarily The Investment Climate Department of the World in-house staff time (in person-hours per year X gross Bank Group (WBG) has undertaken over a dozen “tax wage rates) plus (when relevant) costs for hiring compliance cost surveys” (TCCS) in developing and external tax preparers or advisors, tax forms, sales transition countries over the past several years, and registers, or other out-of-pocket expenses. We took care has amassed a wealth of empirical data documenting to separate the costs of tax compliance from those of the severity of the compliance burden for micro, small general bookkeeping, asking specifically about both. and medium enterprises, and perceptions about tax compliance from both formal and informal businesses. For example, in Armenia, the average business spent a total of about 400 person-hours per year on tax Although the various TCCS are difficult to compare, compliance, which was valued at just over $1000. In due to the fact that each was tailored to the particular other countries, the range has extended from less than conditions and priorities in each country, they allow 100 hours (or even 50 hours for sole proprietors using at least for direct comparison of the time required for simplified tax regimes) to well over 1000 hours (e.g., in specific tax compliance tasks in terms of person-hours Ukraine before recent reforms). Table below illustrates per year. the range for “medium” sized businesses, focusing on The findings of the surveys can not necessarily be taken time required for Income tax, VAT and payroll taxes in as typical for developing countries, as the majority of several of the countries in the database.14 client governments only requested a tax compliance Complexity of tax regimes (e.g., multiple taxes, several cost survey if there was a reason to believe that it was different bases, requirements for multiple filings per a problem for business taxpayers (with the exception year, etc.) appear to drive up tax compliance costs, as of South Africa, where we piloted the first such do requirements to submit detailed accounting records survey, and where tax compliance costs appear to be manageable for the majority of businesses). 14 Source for all countries other than Ethiopia is Coolidge (2012); for Ethiopia it is mean hours per year of staff time (for a business doing all its tax compliance work 13 This section is extracted from Coolidge, Jacqueline, in-house) for Business Profit Tax, VAT, Employment 2012: “Findings of Tax Compliance Cost Surveys in and other Withholding taxes for a business in the ETB Developing Countries”, eJournal of Taxation, Vol 10, No. 500,000—10 million turnover band. A “medium size” 2, Oct. 2012, pp 250-287. Available at: http://www.asb. business is defined following the Doing Business Paying unsw.edu.au/research/publications/ejournaloftaxresearch/ Taxes indicator, specifically a business with turnover equal Documents/paper3_v10n2_Coolidge.pdf to 1050 x per capita GNI (World Bank Atlas method).   49 Figure 1: Average Hours per Year for Tax Compliance (medium size business) Total Hours of Staff Timer per Year 2000 1870 1800 1600 1400 1200 1030 1000 800 699 649 672 632 600 508 400 200 105 0 Armenia Nepal Peru S. Africa Ukraine Vietnam Yemen Ethiopia along with tax returns (as in Peru and Armenia), usually doesn’t know how much time the business spent especially if tax accounting differs significantly from internally on such tasks. In countries where almost all financial accounting. Substantial changes to tax tax compliance work is done in-house (including Nepal legislation and regulations, even if they are designed to and Ethiopia), we would interview businesses directly; simplify the tax regime, require time for taxpayers to in countries where most such work is outsourced, learn about them. we would interview the tax preparers (e.g., Peru). In countries where there is a broad mix of the two Of course, there are several caveats regarding the practices (e.g., in South Africa) it was necessary to survey data. Many small business taxpayers find it interview both groups in order to get a comprehensive conceptually difficult to separate tax compliance picture of the issue. costs from general bookkeeping, as they tend to carry out bookkeeping only to the extent required for tax compliance. Even knowledgeable tax experts may 2.  Regressivity in Tax disagree with one another about where to draw the Compliance Costs line, known as the “disentanglement problem.”15 In fact, Sandford and others have pointed out that there A very common pattern seen across all countries are benefits to tax compliance to the extent that small where TCC have been studied is a striking degree of business management is improved by the financial regressivity: they are much heavier, as a percent of knowledge gained from tax compliance that wouldn’t turnover, for smaller businesses than for larger ones.17 have happened otherwise. In such an analysis, gross While medium and large businesses usually spent less tax compliance costs minus benefits of tax compliance than 1/10th of 1% of their turnover in TCC, small equal net tax compliance costs.16 This Annex focuses on businesses often face TCC of 5% or more of turnover, gross TCC. which can be compared to an extra tax burden.18 For businesses that outsource some or all of their tax Studies undertaken on TCC in developed countries compliance work, the owners are usually unaware of before the year 2000 typically show burdens of one the time spent by their accountant, while the accountant 17 See, for example, OECD (2012) 15 See Turner, Smith and Gurd (2008). 18 It should be noted that in addition to the problems of 16 See Sandford, Godwin and Hardwick (1989), Tran-Nam, the valuation of the time of small business-owners, such Evan, Walpole and Ritchie (2000); and Lignier (2006 and estimates can also be affected by under-reporting of 2009). turnover by business respondents. 50  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 2: South Africa Tax Compliance Cost SouthSurvey Africa Tax Compliance Cost Survey Compliance Burden for Preparation of Tax Returns as a Percent of Turnover (firms registered/not registered) for VAT; Mandatory at R300,000 6.0 5.0 % of Turnover 4.0 Firms Registered for VAT 3.0 2.0 1.0 Firms Not Registered for VAT 0.0 0.15 0.3 0.65 3.5 10 Turnover (in R million) Source: FIAS 2007. tenth of one percent of turnover or less for businesses The OECD provides a number of international over about $200,000. However, a more detailed study comparisons for thresholds for Income tax, presumptive in New Zealand documented TCC of up to 20% for tax and VAT. As seen from the table in Annex 1 below, businesses with turnover under $20,000 (Colmar many countries have increased their thresholds to Brunton, 2005; although most of these businesses reduce the administrative burden from their taxpayers. may have been start-ups with both low turnover and Setting a high threshold seems attractive given that the relatively high costs in “learning the ropes” of business contribution of small business to revenues collected tax compliance). One key question which could be is small,20 though for businesses under the threshold, addressed by regression analysis would be the relative they may be encouraged to underreport turnover to importance of the TCC burden of “start-ups” versus take advantage of reduced rates that come with the “smallness” in this pattern, given that most startups are lower thresholds and this would create distortion of also small. competitive neutrality between businesses.21 The WBG’s first TCCS was undertaken in 2006 in South Africa, where we estimated TCC of about 5% of 3.  Difference between turnover for businesses with turnover below $50,000 those who outsource tax and that were registered voluntarily for VAT. Those of similar size not registered for VAT had TCC/turnover of compliance work and those less than 3%.19 who do it in house Other countries showed even more extreme TCC As described above, countries differ substantially in burdens for small businesses, although some of the their habits of outsourcing tax compliance. While most differences may be exaggerated due to relatively higher of the former Soviet countries involved in TCCS seem estimates of the imputed wages of owner/managers or to prefer (or need) accountants on staff, those in Latin possibly due to under-reporting of turnover by survey America are almost opposite: in Peru, for example, tax respondents (see Figure 3 on next page). filings are legally required to be prepared by a certified accountant, most of whom service multiple clients (although large businesses mostly have them on staff). 19 Here it should be noted that businesses voluntarily registered for VAT were probably different from similar- sized businesses in other respects, such as numbers of 20 IFC (Engelschalk) “Designing a Tax System for Micro and employees and consequent implications for payroll and Small Businesses” page 47. other taxes, so the difference can not necessarily be 21 IFC (Engelschalk) “Designing a Tax System for Micro and ascribed to VAT alone. Small Businesses” page 46. Tax Compliance Costs—International Comparisons for Ethiopia   51 Figure 3: Tax Compliance Cost as Percentage of Turnover 14 Tax Compliance Costs as a Percentage of Turnover 12 10 8 6 4 2 0 0–5 5–13 13–24 24–41 41–66 66–104 104–161 161–246 246–374 374–567 567–855 855–1287 1287–1936 1936–2909 2909–4369 4369–6558 6558–9843 Turnover in '000 USD South Africa Ukraine Uzbekistan Kenya Georgia Ethiopia Nepal Source: Compilation from TCCS for countries cited. Figure 4: Outsourcing Practices among Small Business Taxpayers in South Africa, by Turnover Band Less than R300K 35 44 20 45 R300K to R1000K 31 56 13 41 R1000K to R6000K 43 48 9 33 R6000K to R14000K 45 49 6 30 Over R14 Million 49 42 9 26 Total 43 46 34 11 Completely Outsourcing Partially Outsourcing Not Outsourcing at All Outsourcing Index Source: Coolidge, et. al, 2009. In South Africa, the pattern is more mixed: some all their business profit tax compliance work in-house, businesses do all their tax compliance work in-house, over half of VAT payers doing it all in-house, 85% of some outsource it entirely, and some use a mix of in- TOT payers, and 70% of employment tax payers. (See house work and outsourcing. (See Figure 4). Figure 5). Ethiopian businesses do more of their tax compliance For the most part, small businesses in South Africa base work completely in-house, with over two-third doing their decisions on the availability of the necessary skills. 52  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 5: Outsourcing Practices among Business Taxpayers in Ethiopia, by Type of Tax General bookkeeping (N = 761) 84.8 7.6 7.6 Business profit tax (N = 692) 68.4 12.5 19.1 Value added tax (N = 409) 50.5 24.2 25.3 Turnover tax (N = 267) 85.1 7.7 7.2 Withholding income tax on payments (N = 230) 69.2 21.4 9.4 Employment related contributions (N = 345) 70.7 16.3 13.0 Other taxes (N = 99) 99.0 0.6, 0.4 0 20 40 60 80 100 120 Completely In-house Partially In-house Completely Outsourced Figure 6: Reasons for Decisions to Outsource or Not in South Africa 7.4 0.5 9.3 Tax is a specialist field 10.2 No time to do it internally Tax expertise of my current tax practitioner It is difficult to keep up- to-date with changes in 72.5 tax laws Others Among those who do most of their tax compliance 4.  Tax Inspections And Audits work in-house, most say they have sufficient expertise. Tax inspections and audits are usually a much more Among those who outsource, most say they do it important concern for businesses in developing because “tax is a specialist field” that they lack. Only and transition countries than they are in developed about 10% of those who do not outsource say it is countries, and add significantly to tax compliance costs. because it costs too much, although among the smallest Most revenue authorities in developing and transition businesses, 25% say that is their reason (Figure 6). countries do not have a well-functioning system of Examining the data in South Africa more deeply, it “risk based audit”, nor of reliable “self assessment” appeared that the mix of in-house and out-sourced on the part of taxpayers that they feel they can trust. tax compliance was significantly more expensive than Compounding the problem, many individual tax either doing all the work in-house or out-sourcing all officials in such countries are alleged to unofficially the work. Apparently, when part of the work is done “negotiate” with taxpayers to reduce their tax payments in-house (e.g., basic bookkeeping, VAT accounting in return for a bribe. The latter practice may reduce etc) and some tax compliance work is outsourced, the overall costs (tax payments plus bribes) to the taxpayer outside tax practitioner needs to review the in-house at the expense of the Treasury.23 work before he/she can proceed to the tax compliance work, in order to check for mistakes and correct them if 23 It is also possible that some tax officials threaten taxpayers necessary.22 with a higher tax bill than their legal liability; taxpayers may rather pay a bribe to the tax official to pay the “right” amount of tax than go through the uncertain (and usually 22 Coolidge, et. al, 2009. lengthy and expensive) process of trying to appeal. Tax Compliance Costs—International Comparisons for Ethiopia   53 Figure 7: Incidence of Inspections among Business Taxpayers in Ukraine 80% 76 73 70% 60 60% 50 48 48 50% 45 42 40 40% 37 36 35 35 32 32 29 30 31 30% 20% 10% 0% Less than UAH 300 000- UAH 1–5 UAH 5–35 Over UAH Ukraine UAH 300 000 1 million million million 35 million (average) Tax Authority Pension Fund Social Insurance Funds Source: Coolidge, et. al, 2009.Source: IFC (2009). A well-designed and well-implemented system of risk- Ethiopia’s figures look similar to Ukraine’s figures, based audit, alongside self-assessment of tax liabilities with just under one third of businesses experiencing by business taxpayers, should both improve the an inspection or audit within the previous two years efficiency of the tax regime and reduce opportunities (Figure 8). for bribery inherent in face-to-face contacts between tax In most countries where the World Bank Group has officials and taxpayers. It also reduces wasted time for done a TCCS, the revenue authorities state that their taxpayers and thereby reduces tax compliance costs. official figures show a much lower rate of inspection. In In Ukraine in 2007, for example, even small businesses the case of Ukraine, after discussions with the State Tax faced about a one-third chance of inspection not only Committee and focus groups of business accountants, by the tax authority but also by the pension fund and we concluded that the majority of “inspections” were in social insurance fund (see Figure 7). fact brief checks of cash registers. In the case of Kenya, Figure 8: Inspections and Audits among Business Taxpayers in Ethiopia 120 100 80 51.6 54.5 70.5 65.8 68.3 60 40 48.4 45.5 20 34.2 29.5 31.7 0 Less than ETB 100,001 ETB 100,001-ETB ETB 500,001-ETB Above ETB 10,000,000 Total (N = 1003) (N = 501) 500,000 (N = 224) 10,000,000 (N = 234) (N = 44) No Yes 54  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 9: Different Types of Tax Inspections in Georgia Desk tax audit 0.1 Planned field tax audit 0.1 Unplanned/ control field tax audit 0.3 Chronometrage 0 Controlling purchases 0.8 Visual inspection 0.5 Stock-taking/ inventory 0.1 Control of cash register usage 2.5 0 0.5 1 1.5 2 2.5 3 Source: IFC (2011) [Georgia]. officials from the Kenya Revenue Authority (KRA) said form of tax inspection (usage of cash registers is a that many of the short visits were likely “informational legal requirement). visits” by tax officials, which were not inspections at • chronometrage/invigilation: Tax inspectors monitor all. However, they also speculated that there may be sales (of a restaurant or a hotel) for a certain period some “unofficial visits” or even fraud—i.e., people of time, usually for at least one week. If sales during claiming to be tax inspectors who are not. this period exceed sales from previous period The TCCS in Georgia included a breakdown of types of as reported by business, then the business gets inspections (see Figure 9): penalized. Such penalties existed during the survey but have since been abolished. Definitions of some of the more technical terms in the chart above are as follows: By way of comparison, we have used the case of South Africa as an example of a country with a well- • control of cash register: Check that cash register is in functioning system of Risk Based Audit (RBA). Overall, compliance with set standards. small and medium businesses in South Africa in 2006 • visual inspection: A visual inspection of premises (those with turnover under about $2 million) faced without checking accounting documents, stock, etc. only about a 2% chance of inspection for Income Tax • controlling purchases: Tax inspectors buy goods and Employees taxes and about 3% for VAT. Some of from a store and check if salesperson registers the the less populous provinces showed a slightly higher transaction in cash register and gives them receipt. If likelihood of inspection (see Figure 10). not, business is penalized. This is the most frequent Figure 10: Incidence of Tax Inspections by Type of Tax and by Region—South Africa (2006) Income Tax VAT Employees Tax Mean by provinces 1.7 11 3 11 2.3 Mean by respondents 1.7 10 3.1 10 2.2 Limpopo 1.6 9 1.6 9 Western Cape 1.5 8 2.7 8 1.1 Eastern Cape 1.3 7 2.1 7 2.4 Kwazulu Natal 1.4 6 3.2 6 1.5 Gauteng 1.8 5 3.3 5 2 Mpumalanga 1.5 4 5.7 4 0.5 North West 1.9 3 4.4 3 1.7 Freestate 3 2 1.7 2 4.3 Northen Cape 0.9 1 2.2 1 6.2 0 0.5 1 1.5 2 2.5 3 3.5 0 1 2 3 4 5 6 0 2 4 6 8 Source: FIAS, (2008). Tax Compliance Costs—International Comparisons for Ethiopia   55 Figure 11: Perceptions of Tax Inspections in Georgia Someone creates obstacles to others Inspector officers justifying their jobs Informal payments Additional revenue collection for gov't Legitmate reasons 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 Percentages Figure 12: Perceptions of Tax Inspections in Ethiopia (percent of inspections) Other reasons (N = 1002) 0.34 Inspections are used as an additional revenue collection 16.88 mechanism for the government (N = 1002) Because of someone’s interest to create obstacles to 9.36 others (e.g. to competitors (N = 1002) Inspection officers are trying to justify their jobs 19.17 (N = 1002) Inspection officers are looking for informal payments 19.15 (N = 1002) Legitimate reasons, i.e. inspections are really 36.1 necessary to deter tax evasion (N = 1002) 0 5 10 15 20 25 30 35 40 Of course, businesses also understand the need for at (e.g., “tax officers are trying to justify their jobs” or least some inspections to catch and deter evasion. They “used as an additional revenue collection mechanism by perceive, however, that many inspections are motivated government”) or even corruption (“inspection officers by a need to achieve revenue targets, to extract bribes, are looking for informal payments”). See Figure 12. or in some cases to harass “opposition figures” or The World Bank’s Enterprise Surveys, carried out in competitors to “cronies” of politicians. almost all developing and transition countries, provides A very striking difference between Georgia and Ethiopia cross-country data on the perception of corruption in is evident in the perceptions of business taxpayers about the tax administration. Ethiopia compares quite well the “legitimacy” of most inspections. In Georgia, the within Sub-Saharan Africa, similar to South Africa and great majority of business taxpayer perceive that tax Namibia. See Figure 13.24 inspections are carried out primarily for legitimate reasons, e.g., to deter under-reporting of sales and 24 In order to compare the estimates above, it is important to multiply the “percent of inspections [where] income (see Figure 11). inspection officers are looking for informal payments” by the “percent of businesses experiencing an inspection By contrast, in Ethiopia, only a minority perceive tax in the previous two years” divided by 2: 19.2% * inspections as legitimate, and much higher percentages 31.7%/2 5 3%, which is well within the margin of error for the Enterprise Survey estimate for Ethiopia of “Percent than in Georgia perceive bureaucratic motivations of Firms expected to give gifts in meetings with tax officials” 5 3.8% 56  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Figure 13: Percent of Firms Expected to Give Gifts in Meetings with Tax Officials Sub-Saharan Africa 17.4 Angola 34.2 Botswana 8.4 Burkina Faso 6.7 Burundi 22.6 Cote Ivoire 13.6 Ethiopia 3.8 Ghana 11.4 Kenya 17.5 Malawi 11.4 Mozambique 9.8 Namibia 2.6 Nigeria 22.9 Rwanda 4.5 Senegal 18.7 South Africa 3.1 Tanzania 14.6 Uganda 14.3 Zambia 9 Zimbabwe 12.6 0 5 10 15 20 25 30 35 40 Source: http://www.enterprisesurveys.org/Data/ExploreTopics/corruption. Evasion is another critical problem in developing and common evasion tactic is the use of fake invoices transition countries, and although it is very difficult (Figure 15; but essentially the same as “overstatement to make cross-country comparisons, Ethiopia seems of costs”). However, it should be noted that the form to have similar problems to other countries, such of the question was different in the two countries. as Armenia. While businesses in both countries are In Armenia, the data are about the percent of most likely to evade taxes by under-declaring sales respondents mentioning a particular type of evasion revenue, the second most common evasion tactic in practice, while in Ethiopia, the data are about the Armenia is “payment of unofficial salary” to workers subjective responses about statements of the practices (Figure 14), while in Ethiopia, the second most among businesses. Figure 14: Perceptions of Tax Evasion in Armenia (percent of respondents who reported at least one method of tax evasion as “popular”)* Popular Methods of Tax Evasion in Armenia Declare only part of the revenue 52.4% Payment of unofficial salary 26.0% Overstate of costs 8% Use fiction firm 7.4% Fraudulent abuse of tax privileges 5.1% No way 1.5% *Among businesses who cited at least one method of evasion Tax Compliance Costs—International Comparisons for Ethiopia   57 Figure 15: Perceptions of Tax Evasion in Ethiopia (average ranking of agreement) Businesses make unregistered sales (i.e. without sales 3.45 invoices) to artificially reduce tax burden (N = 1003) Businesses use fictitious/ fake invoices to artificially reduce 3.1 tax burden (N = 1002) Businesses use fictitious/ fake businesses to artificially 2.95 reduce tax burden (N = 1003) Businesses make payments of unofficial salaries to their 2.96 workers to artificially reduce tax burden (N = 1003) 1 2 3 4 5 Fully disagree Fully agree Figure 16: Percent of Total Annual Sales Declared for Tax Purposes by Region in Ethiopia (excluding those who said 100% or “don’t know”) (N 5 921) Total 53.4 Addis Ababa (N = 458) 51.7 SNNPR and Gambella (N = 68) 62.9 Oromia, Dire Dawa, Harari, and Somali (N = 181) 63.6 Amhara and Benishangul-Gumuz (N = 124) 41.1 Tigray and Afar (N = 82) 48.6 0 10 20 30 40 50 60 70 More specifically, the survey interviews asked, regarding References “businesses like yours”, what proportion of income Colmar Brunton Social Research Agency, New Zealand, was typically reported officially for tax purposes. In the (2005) “Measuring the tax compliance costs of case of Armenia, the share of income reported for tax small and medium-sized businesses—a benchmark purposes was about two-thirds, while in Ethiopia it is survey” http://www.taxpolicy.ird.govt.nz/news/ archive.php?year52005&view5383 just over half (Figure 16). Coolidge, Jacqueline, 2012: “Findings of Tax Finally, it is important to repeat and stress the point Compliance Cost Surveys in Developing that international comparisons in tax compliance cost Countries”, eJournal of Taxation, Vol 10, No. 2, Oct. 2012, pp 250-287. Available at: http:// and perception surveys should be considered carefully, www.asb.unsw.edu.au/research/publications/ taking into account differences across countries in ejournaloftaxresearch/Documents/paper3_v10n2_ tax regimes (both laws and regulations), the wording Coolidge.pdf of questions, and cultural/national differences in the Coolidge, Jacqueline “Tax Compliance Cost Surveys: frankness of survey respondents. Using Data to Design Targeted Reforms” World 58  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Bank Group, Investment Climate In Practice Note Publications in association with The Institute for no. 8, Feb. 2010. Available at: http://www-wds Fiscal Studies. .worldbank.org/external/default/ Hasseldine, J; C T Sandford; Jeff Pope; Chris Evans WDSContentServer/WDSP/IB/2010/04/19/0003349 (2001) “Taxation Compliance Costs: A Festschrift 55_20100419043011/Rendered/PDF/540860BRI0T for Cedric Sandford,” St. Leonards, N.S.W.: axC10Box345636B01PUBLIC1.pdf Prospect Media Coolidge, J., Kisunko, G. and Ilic, D. (2009), “Small IFC (2009) The Costs of Tax Compliance in Businesses in South Africa: Who outsources tax Ukraine. Available at: http://www.ifc. compliance work and why” World Bank Policy org/ifcext/fias.nsf/AttachmentsByTitle/ Research Working Paper No. 4873. World PublicationCR_CostofTaxCompliance/$FILE/ Bank, Washington DC. Available at: http://econ. CostofTaxCompliance_Ukraine.pdf worldbank.org/external/default/main?searchTxt5 IFC (2011) IFC Tax Perception and Compliance Cost WPS4873&detailMenuPK564264748&docTY56 Surveys: A Tool for Tax Reform, World Bank 20265&menuPK564264748&pagePK564166018 Group, March 2011. Available at: https://www &piPK564165415&siteName5EXTDEC&theSite .wbginvestmentclimate.org/uploads/TPCCS_ PK5469372 Consolidated_Web.pdf European Commission (2004) European Tax Survey. IFC (2011) The Costs of Tax Compliance in Armenia. Working paper n 3/2004. Available at: https://www.wbginvestmentclimate. Evans, Chris (2003) “Studying the Studies: An overview org/advisory-services/regulatory-simplification/ of recent research into taxation operation costs,” business-regulation/business-operations/upload/ eJournal of Tax Research, Vol. 1, No. 1 Armenia-Report-Cost-of-tax-compliance.pdf FIAS (2007) “SME Taxation: A Toolkit for IFC (2011) “Georgia Tax Compliance Cost Survey Practitioners” 2010” Power-point presentation FIAS (2007) “South Africa: Tax Compliance Burden for OECD (2012) “Reducing the Tax Compliance Burden” Small Businesses: A Survey of Tax Practitioners,” Sandford, C. (1994), “International Comparisons of available at: http://www.ifc.org/ifcext/fias. Administrative and Compliance Costs of Taxation,” nsf/AttachmentsByTitle/South_Africa_Tax_ Australian Tax Forum Practitioners_Report/$FILE/SouthAfrica_FIAS_ USAID (2008), “Formal SMME Tax compliance Survey Tax_Practitioners_Report.pdf Report: Prepared for National Treasury Republic FIAS (2008) “Yemen Tax Cost of Compliance Survey” of South Africa”, available at: http://www.fias.net/ (unpublished manuscript) ifcext/fias.nsf/Content/Pubs_BusinessTaxation FIAS (2009) “Yemen: Designing a new SME regime: USAID (2008), “Informal SMME Tax Survey Report: Analysis of compliance cost survey data” Prepared for National Treasury Republic of South (unpublished manuscript) Africa”, available at: http://www.fias.net/ifcext/fias. Godwin, M. (1995) “The compliance costs of the nsf/Content/Pubs_BusinessTaxation United Kingdom tax system” In: Tax Compliance Costs—Measurement and Policy. Bath: Fiscal Tax Compliance Costs—International Comparisons for Ethiopia   59 Annex Examples of Thresholds Applied 2 to Reduce the Compliance Burden of Small Businesses Area of Tax Law Country Examples of Thresholds for Small Businesses 25 To pay Australia Quarterly payments where annual liabilities  $25,000 ($US 22,000) employee Canada26 From 2008, employers whose average monthly liability is  $C3,000 (previously  withholdings of $C1,000 will be eligible to remit source deductions quarterly instead of monthly. presumptive tax Germany Quarterly payments where prior year wages tax (PYWT)  3,000; annual payments tax where PYWT  800; otherwise monthly Ireland Quarterly payments where total PAYE and PRSI payments for year are  30,000 (from April 2006); otherwise monthly NZ Monthly payments where annual PAYE liability  $NZ100,000; otherwise twice monthly To make Australia Annually if not registered for GST and notional tax  $8,000 ($US 7,000); otherwise advance quarterly unless defined as large taxpayer. payments of Czech Rep. Quarterly payments where prior year liability was between CZK30,000–150,000 corporation (approx. $US 2–10,000); monthly for larger taxpayers Income tax Ireland New small companies with first year tax liability  150,000 not required to make advance payments; other small companies can make advance payments based on prior year liability. Slovak Rep. Quarterly payments where prior year liability was between SVK50,000–500,000 (approx. $US 3–30,000); monthly for larger taxpayers Canada Increase the threshold (from 2008) below which corporations are eligible to pay federal corporate taxes annually from $1000 to $C3000 z To register for Austria Must register if annual turnover  30,000 (approx. $US 42,500) VAT Canada Generally, must register if annual taxable sales 5 $C30,000) Ireland Must register if annual turnover  35,000 (services) or 70,000 (goods) 25 Japan Must register if annual turnover  JPY10 mill. (approx. $US 75,000) Singapore Must register if annual turnover  $S 1 mill. (approx. $US 690,000) Slovak 26  SKK 1.5 mill. (approx. $US87,000) Must register if annual turnover South Africa Must register if annual turnover  R1,000,000 (approx. $US85,000) UK Must register if annual turnover  £58,000 (approx. $US87,000) (continues) 25 The currency conversions shown in this table were calculated at various times during preparation of the 26 All the examples for Canada in Table 6 include legislative report. Given the recent volatility in currency markets, amendments that have been announced, but are not law as readers may wish to make their own just-in-time of October 30, 2007. They are expected to be passed into calculations. the law by the time this report is released.   61 Continued Area of Tax Law Country Examples of Thresholds for Small Businesses To file returns Australia Annual for voluntary registration; quarterly where turnover  $A2 million and pay VAT ($US 1.64 million); otherwise monthly. Austria Quarterly returns if prior year turnover  22,000; otherwise monthly. Canada Starting in 2008, annual return and quarterly payment if annual taxable sales  $C1.5 million (previously $C500,000). Annual payment if net tax  $C3,000 (previously $C1,500) Czech Rep Quarterly payments if annual turnover  CZK 10 mill. (approx. $US 65,000); otherwise monthly Germany Quarterly payments if prior year tax  6,136 (approx.$US8,000); annual payments if prior year tax  512; otherwise monthly Use of cash Australia Optional use where annual turnover  $A2 mill. (approx. $US 1.7 million) accounting Austria Optional use for professionals and other businesses if turnover of last or simplified two years  110,000 calculation basis Canada Optional use for prescribed traders where turnover  $C200,000. Also, optional for VAT liability simplified input credit calculation where annual sales  $C500,000 and annual determination purchases  $C2 million Germany Optional use where turnover  250,000 (approx. $US370,000) Ireland Optional use where turnover  1m (approx. $US1.35 mill.) South Africa Optional use where turnover  R2.5 mill. (approx. $US370,000) UK Optional use where turnover  £660,000 (approx. $US1.3 mill.) VAT invoices Austria Use of simplified invoices (threshold 150). Germany Use of simplified invoices (threshold increased from 100 to 150). To pay Australia Quarterly payments where annual liabilities  $ 25,000 ($US 22,000) employee Canada27 From 2008, employers whose average monthly liability is  $C3,000 (previously withholdings of $C1,000 will be eligible to remit source deductions quarterly instead of monthly. presumptive tax Germany Quarterly payments where prior year wages tax (PYWT)  3,000; annual payments tax where PYWT 800; otherwise monthly Ireland Quarterly payments where total PAYE and PRSI payments for year are  30,000 (from April 2006); otherwise monthly NZ Monthly payments where annual PAYE liability  $NZ100,000; otherwise twice monthly To make Australia Annually if not registered for GST and notional tax  $8,000 ($US 7,000); otherwise advance quarterly unless defined as large taxpayer. payments of Czech Rep. Quarterly payments where prior year liability was between CZK30,000–150,000 corporation (approx. $US 2–10,000); monthly for larger taxpayers Income tax Ireland New small companies with first year tax liability  150,000 not required to make advance payments; other small companies can make advance payments based on prior year liability. Slovak Rep. Quarterly payments where prior year liability was between SVK50,000–500,000 (approx. $US 3–30,000); monthly for larger taxpayers Canada Increase the threshold (from 2008) below which corporations are eligible to pay federal 27 corporate taxes annually from $1000 to $C3000 Turnover tax Germany Mandatory accounting threshold set at 0.5 mill. (Previously 0.35 mill). 27 All the examples for Canada in Table 6 include legislative amendments that have been announced, but are not law as of October 30, 2007. They are expected to be passed into the law by the time this report is released. 62  Tax Compliance Cost Burden and Tax Perceptions Survey in Ethiopia Area of Tax Law Country Examples of Thresholds for Small Businesses Employment tax USA Annual filing and payment (previously on a quarterly basis) where liability  $1,000 per year. Turnover tax Germany Mandatory accounting threshold set at 0.5 mill. (Previously 0.35 mill). Employment tax USA Annual filing and payment (previously on a quarterly basis) where liability  $1,000 per year. Sources: Revenue body web portals; IBFD; OECD Consumption Tax Trends (2006), Tax Administration in OECD Countries: Comparative Information Series (2006) Examples of Thresholds Applied to Reduce the Compliance Burden of Small Businesses   63