FUNDING DEVELOPMENT: ETHIOPIA TRIES TO STRENGTHEN ITS TAX SYSTEM, 2007 - 2018 SYNOPSIS In its 2006 national vision to end poverty, Ethiopia set its sights on becoming a middle-income country by 2025. It was a hugely ambitious goal for a country that, at the time, was one of the poorest in the world. To support development objectives put on hold during a decade of political turbulence, including a costly border war with Eritrea that drained public coffers, the Ethiopian government sought to expand its resources by significantly boosting tax revenues. The new plan called for a sharp increase in the ratio of tax revenue to the size of the economy-and within four years. The government merged its separate customs and domestic tax offices into a single entity and restructured the new agency's operations along functional lines, increased salaries, adopted stringent anticorruption rules, implemented a modern information technology system, and launched public awareness campaigns. It was important that the new revenue authority worked to improve its coordination with the tax offices of subnational governments, which operated with substantial independence under the country's federal system. Although unproven charges of corruption against the Ethiopian Revenues and Customs Authority's long-serving director general in 2013 stalled progress, a new round of IT and legal reforms in 2016 helped increase tax collection significantly: to US$7.8 billion in 2017 from US$1.3 billion in 2006 (measured in constant 2010 US dollars). Nonetheless, revenue gains continued to lag behind economic growth. In 2018, under a new prime minister, the government began to take further steps to strengthen tax collection. Leon Schreiber drafted this case study based on interviews conducted in Addis Ababa, Ethiopia, in October 2018. Case published December 2018. INTRODUCTION Following a decade of political and economic turbulence, Ethiopian Prime Minister Meles Zenawi entered the 2000s determined to ignite economic development in his country, Africa's second-most populous nation. But financing the government's double-digit growth ambitions and planned infrastructure projects-best exemplified by the Grand Ethiopian Renaissance Dam, a US$4-billion hydroelectric project that would be the biggest in Africa and which the government intended to fund using only domestic resources-would require significantly more tax revenue. During the 1990s, Ethiopia ranked as the third-poorest country in the world, with annual per-capita gross domestic product of less than $200.1 More than half of the population lived on less than $1.25 per day.2 The country also had struggled to collect taxes equivalent to even 10% of its annual GDP, placing it well below sub-Saharan Africa's average of 18%.3 In 2002, in an effort to improve the ratio and increase total tax collection from a base of just US$1.1 billion, the government introduced a modernized legal framework. To stimulate investment, a law on direct taxes reduced the corporate tax rate to 30% from 35%, introduced deductions for the calculation of business profits, created generous incentives for both foreign and domestic investors, and called for the issuance of taxpayer identification numbers (TINs) to all businesses and individual taxpayers in urban areas.4 (At the time, technology constraints made it difficult to issue TINs to taxpayers in rural areas.) A second law-on indirect taxes-established a value-added tax (VAT) to replace the existing sales tax, reduced export taxes, created new excise duties on luxury imports, and introduced mandatory income tax withholding by employers.5 Unlike a sales tax, which applied only to final purchasers, the new VAT was levied at each stage of the supply chain. The effect spread the burden of taxation more broadly, increased the potential for revenue collection, and created a clear audit trail from producer to end consumer. As part of the reforms, the government also shifted to a self-assessment system that enabled taxpayers to complete their own declarations rather than having all payments calculated by tax officials. Four years after those changes, however, the country's tax-to-GDP ratio had actually declined to 10.7% from 11.8% in 2002 despite three years of GDP growth in excess of 10% annually. Fragmentation had caused confusion and delay. For example, the new VAT suffered from crippling coordination problems in reconciling information from imports-which were handled by the customs authority-and domestic sales, which were administered by the inland authority. Girma Gebretsadik, who headed the large taxpayer office at the inland revenue authority at the time, explained that the country's tax service was divided into three organizations: a federal inland revenue authority that dealt with domestic taxes and accounted for the bulk of Ethiopia's tax revenues, a customs authority in charge of collecting trade taxes, and a ministry of revenue that provided oversight. The government went back to the drawing board. Following the 2005 general election, when violence erupted after Meles's governing Ethiopian People's Revolutionary Democratic Front won reelection with a much-reduced majority, the government doubled down on its push for economic development. As part of that renewed push for faster revenue growth, in November 2007 the civil service ministry and the minister of revenue, Melaku Fenta, appointed two dozen of the country's foremost tax experts to review procedures and propose improvements in order to overhaul the country's underperforming tax administration system. Alongside Fenta, a senior public servant who had worked on intergovernmental coordination and had a degree in tax administration from the University of Canberra in Australia, the expert team also included Gebretsadik as well as Girma Tafesse, the inland revenue authority's head of research and planning. The reform team had clear ideas about what it wanted to do, but it would soon learn that the path forward would be rocky. Change would proceed only in fits and starts-and eventually find a new champion in 2018. THE CHALLENGE Decentralization posed a daunting contextual challenge because Ethiopia's 1995 constitution had introduced sweeping reforms to devolve authority away from the central government. The country's federal political system comprised nine ethnically based regions and two chartered cities: Addis Ababa and Dire Dawa. The constitution also specified which taxes the federal government would collect, which taxes the federal government would share with the regions, and which taxes fell under the exclusive purview of regional governments. In 2007, regional revenue authorities collected and kept for themselves 22% of all government tax revenues in Ethiopia. In the face of the complicated division of powers, the federal government had to find ways of coordinating and cooperating with the regional revenue authorities. But because each region was constitutionally empowered to create its own legal framework, rules differed from one region to another. Philippe Dadour, a Canadian IT consultant who worked on tax administration in Ethiopia from 2005, cited the example of taxes on khat (a plant-based stimulant and appetite suppressant): "Oromia had a special rate for the export of khat to other regions, while the Amhara region's khat tax did not even mention export." The federal government had to assume greater responsibility in order to encourage standardization and prevent taxpayers from being burdened by widely divergent policies and procedures across the different tax authorities. In the view of senior officials, however, the most immediate problem facing the tax system was not decentralization but the division of responsibility between the federal inland revenue authority and the customs authority. The fragmentation hindered efforts to issue a single taxpayer identification number to each taxpayer because the customs and domestic tax offices kept separate records. The lack of coordination also meant that taxpayers had to follow different procedures at different tax offices and made it difficult for officials to reconcile import VAT paid to customs and VAT refunds later claimed from the domestic tax office. The lack of a centralized digital database aggravated the information management and coordination problems. With the exception of the VAT, all tax and customs procedures were still manual and paper-based. To process and record transactions for the newly created VAT, in 2003 the domestic tax office had rolled out the Standard Integrated Government Tax Administration System (SIGTAS), an information system for the administration of taxes that was in use by many countries around the world. To overcome the information fragmentation and enable Ethiopia to make use of modern technology, tax officials had to urgently expand the use of SIGTAS to cover all tax types at both the federal and regional levels. Insufficient human resources capacity posed a third operational challenge. Tsegabirhan Abay, a researcher at Addis Ababa University, wrote that the tax agencies suffered from "low pay, low human resource development [and] low motivation" as well as "institutional weakness . . . in terms of [the] identification of taxpayers, assessment capacity, administration, law enforcement, and taxpayer culture. With such weak tax administrative capacity, tax evasion was the rule rather than the exception, and foregone tax revenues [were] huge."6 In a system of low capacity and low morale, corruption flourished, and it contributed to the problems with the VAT, such as weak coordination and incomplete data as well as the failure of TINs to immediately improve compliance by making it easier to track and monitor individual taxpayers. In Transparency International's 2005 Global Corruption Barometer, which surveyed 55,000 people across 69 low- and middle-income countries, respondents in Ethiopia ranked tax revenue at 3.8 and customs at 3.6 on a scale of 1 to 5, where the top number signaled "extremely corrupt."7 In an International Centre for Tax and Development working paper, Giulia Mascagni, a former adviser to the Ethiopian finance ministry, wrote that some officials resisted reforms out of fear that a more efficient system would undermine the "unofficial benefits that tax officials were able to extract from a mismanaged system."8 A system of generous tax exemptions, motivated partly by the government's desire to attract foreign investment, also led to substantial revenue losses at large companies, including multinationals. Although Ethiopia failed to keep accurate data, estimates showed that for 2005, 2006, and 2007, total tax revenue foregone through exemptions amounted to 3.7%, 3.5%, and 4.5% of GDP, respectively.9 Although different ministries and different institutions such as the Ethiopian Investment Commission granted incentives, no single organization was responsible for measuring the impact of such arrangements and keeping track of when incentives were set to expire. Such lack of information and control left the system vulnerable to abuse. (See text box 1.) Box 1. Tax incentives The Ethiopian Investment Commission listed a wide range of tax incentives that reduced or eliminated the requirement whereby businesses had to pay certain types of taxes. Both foreign and domestic investors qualified for tax breaks on customs duties if they opened or expanded enterprises in manufacturing; agriculture; agro-industries; the generation, supply, and transmission of electrical energy; IT; tourism; construction contracting; education and training; luxury hotels; specialized restaurants; architectural and engineering consultancies; technical testing and analysis; capital-goods leasing; and the importation of liquefied petroleum gas and bitumen. The reductions ranged from a 100% exemption on the payment of customs duties and other taxes levied on the import of capital goods such as plant, machinery and equipment, and construction materials to a 15% reduction on the total value of imported spare parts for capital goods exempt from customs duties. Investors that received customs duty exemptions were allowed to import capital goods duty free: indefinitely if their investments were in manufacturing and agriculture and for a period of five years if their investments were in other eligible areas. Investors in certain types of manufacturing, agribusiness, electricity supply and transmission, and IT were entitled to income tax exemptions ranging from one to nine years based on the activity and geographic locations of their investments. The Ethiopian Investment Commission also designated an additional 18 geographic areas that entitled investors to a further 30% income tax deduction for three consecutive years following the expiry of other income tax incentives. Finally, investors that exported at least 60% of their goods or services or that supplied their products to an exporter gained another income tax exemption for an additional two-year period. 1 This text box is based on information contained on the Ethiopian Investment Commission’s website, accessed November 14, 2018; https://web.archive.org/web/20150213081952/http://www.investethiopia.gov.et/images/pdf/incentives.pdf. For Fenta, another major problem "was the unwarranted interference by some of the high-ranking officials in the party and government. These officials acted as a go-betweens between businesspeople and the tax authority, and they always wanted to bend the system in favor of those in the business community."10 Fenta knew he would need substantial political clout to tackle abuse and evasion. "The [reform] option I [wanted] to sign up for was to implement real unpopular reforms [that would] be at odds with most in the power corners. But I also told [Prime Minister Meles] that [this] option would require his active political commitment and support to the cause," Fenta said in a July 2018 interview with the Ethiopian Reporter, a newspaper published in the capital, Addis Ababa.11 (See text box 2.) Box 2. Revenue sharing between the federal and regional governments In practice, fiscal decentralization meant federal authorities had no direct roles in key areas of tax collection. Subnational regions collected and kept all revenues from five sources: (1) income taxes from non-incorporated firms, (2) corporate income tax from enterprises owned by regional governments, (3) personal withholding income tax from employees unless they worked for the federal government or enterprises owned by the federal government, (4) agriculture and land-use taxes, and (5) VAT from individual, non-incorporated traders. As a result, in 2007 the regional revenue authorities collected and kept for themselves a substantial share—22%—of all government tax revenues in Ethiopia. And although the federal government collected all corporate income tax from formally incorporated businesses across the country, it shared 50% of corporate revenues with the regions those companies were registered in. The same principle applied to VAT paid by formal corporations, which the federal government collected. But in the case of VAT, the federal government shared only 30% of the revenue with the regions VAT-paying firms were registered in. Fenta's reforms also had to take into account Ethiopia's narrow base of potential taxpayers. According to a 2016 study published by the United Nations Development Programme, despite accelerating economic growth in the early 2000s, about 36% of economic activity still took place in the informal sector-beyond the view of government tax officials.12 In addition, according to World Bank data from 2007, 84% of Ethiopians lived in rural areas dominated by subsistence agriculture, where average incomes were low and yielded little revenue.13 As the government ramped up its economic development drive, the revenue authorities had to bring more informal earners into the tax net and had to redouble their efforts to help taxpayers understand and comply with their obligations. FRAMING A RESPONSE In late 2007, Fenta's team of tax experts began work to revamp tax administration in Ethiopia. The team built on the government's September 2006 adoption of the national Plan for Accelerated and Sustained Development to End Poverty. The plan drew on consultations with international partners, civil society, and nongovernmental organizations, and it set out the government's strategic economic development framework for 2006-10. The primary objectives were to reduce poverty in Ethiopia and to enable the country to meet United Nations Millennium Development Goals. The framework included an ambitious call for the government to increase the country's tax-to-GDP ratio to 17% by 2010 from 10.7% in 2006. It further stipulated that "vigorous efforts will be made to further improve tax collection, and to combat fiscal fraud."14 And it pledged to accelerate the pace of the tax reform program. The plan identified five key interventions to improve tax administration: (1) strengthening and reorganizing revenue collection institutions, including by providing adequate staffing and training; (2) implementing the TIN system throughout the country; (3) improving implementation of the presumptive tax system that applied to non-incorporated businesses; (4) developing and implementing an audit program to cover all taxes; and (5) expanding and improving VAT administration.15 Fenta's reform team began one step at a time. "One committee focused on domestic taxes, and the other on customs," said Tafesse, who headed the domestic tax committee. "In the first part of the study, we identified the core business processes that had to be in place for all tax types: assessment, collection, and auditing. Next, we identified the limitations affecting each phase, and we drew up proposals for ways to reform that." The reform team also spent a week discussing possibilities with tax officials in the United Kingdom. After more than a year of research, the team produced a set of proposals to reengineer tax administration. At the federal level, the team called for the merger of three separate revenue institutions into a single revenue authority-and for the director general of a reformed tax authority to be a member of the cabinet (as was the case with the previous minister of revenue, which was a position separate from the minister of finance). The team further proposed that the new authority be organized along functional lines, develop standardized and clear policies and procedures, and be exempt from certain public service rules so that the authority could determine salary scales and easily fire underperforming or corrupt employees. In a reflection of Prime Minister Meles's support for the reform effort, the cabinet approved the team's proposals in a matter of a few months. In July 2008, the federal parliament issued a law (known in Ethiopia as a proclamation) that created the Ethiopian Revenues and Customs Authority (ERCA). The legislation declared ERCA to be "an autonomous federal government agency" that "shall be accountable to the prime minister."16 The proclamation empowered the prime minister to appoint a director general as well as deputy directors general. Meles appointed Fenta, the incumbent minister of revenue, as the first director general of the new tax authority, and the government abolished the position of minister of revenue. As director general of ERCA, Fenta remained a member of the cabinet. Gebretsadik said there was good reason to make the director general a cabinet member. "It created a mechanism to get political support from the higher-ups. It also meant the director general was not accountable to the finance minister, because he was parallel to the finance minister," he said. However, the move to designate the head of ERCA as a member of the cabinet also raised the possibility that the revenue authority could become politicized. The founding law also created an advisory board of "competent professionals" to provide strategic and policy advice for ERCA, although ERCA's director general chaired the board and had the power to select board members.17 The advisory board had only limited powers, and Gebretsadik said the cabinet effectively served as the executive board for determining salary levels and regulatory changes. At the same time, ERCA's senior managers accepted the study team's proposal to reorganize the authority along functional lines. Whereas the federal inland revenue authority had previously used a combination of directorates dedicated to specific tax types (VAT, withholding tax, corporate tax) and whereas other directorates organized according to function (declarations, assessment, payment, audit), the new organizational structure eliminated separate offices for different taxes. Following ERCA's creation, both domestic and customs taxes would be handled by directorates dealing with specific functions across all tax types. (ERCA retained the large taxpayer office that had initially been created as part of the former federal inland revenue authority.) With a national strategic plan that prioritized improved tax administration, as well as a revamped organizational framework that assigned greater authority and autonomy to ERCA, Fenta and his team had a solid foundation on which to build. GETTING DOWN TO WORK To ensure that the new organizational structure translated into improved tax administration, ERCA had to simplify its tax procedures, enhance staff capacity, and combat corruption. Other urgent priorities included expanding the use of IT and raising public awareness to boost voluntary compliance. Finally, ERCA also had to find ways of collaborating and coordinating effectively with the country's 11 regional revenue authorities, which enjoyed significant autonomy in collecting nearly a quarter of all government revenues. Building capacity and weeding out corruption The creation of ERCA provided Fenta's team with an opportunity to enhance staff capacity. The law that created the tax authority stipulated that ERCA would be exempt from strict public service rules that made it difficult to dismiss underperforming or corrupt employees. At the same time, the agency also gained the authority to increase staff salaries so as to attract talented personnel and strengthen anticorruption measures. Gebretsadik, whom Fenta appointed as ERCA's first deputy director general, said the team got to work right away. Many of the skilled officials who had been involved in the process reengineering study, conducted in 2007, joined a new tax transformation directorate, which incorporated the study's recommendations into new manuals and guidelines that would streamline and standardize procedures. "We conducted surveys across the different branch offices to find areas where procedures were applied inconsistently," Tafesse said. "After gathering that information, we prepared manuals, sent circulars to each office outlining the standard procedures, and conducted training. We used a lot of 'explanation by example.'" The new tax authority had more autonomy over hiring decisions than did its predecessor agencies, and it launched a wholesale reevaluation of its existing staff, based on skills and integrity. Fenta assembled a human resources placement committee to screen and propose staff for ERCA. Based on a variety of factors, "good staff were retained, and those who were not ethically good were laid off. Hundreds of staff were not rehired," Gebretsadik said. The guidelines included performance reviews, and evaluators favored younger employees. But Gebretsadik stressed that "the main factor was integrity. Those suspected [of being] corrupt were not placed." The tax office implemented new rules to combat corruption across the tax process. ERCA created a dedicated directorate for ethics that reported directly to Fenta. The directorate introduced a code of ethics and provided all staff members with training and a copy of the code. "The directorate got real teeth, and its proposals and recommendations were taken seriously," Gebretsadik said. For staffers implicated in corruption, the code stipulated levels of punishment, including demotion, suspension, and outright firing. The cabinet also issued a special regulation that denied ERCA employees the right to appeal to a court if, upon recommendation by the ethics directorate, the director general fired them on suspicion of corruption. The measure was controversial among staff members, but when the issue came to the cabinet for discussion, Prime Minister Meles backed Fenta's power to fire employees as a necessary tool for cleaning up the tax service. Thanks to the 2008 law that created ERCA, the new tax office also had the power to create its own intelligence and investigations team, and the director general could appoint prosecutors. Legally, the justice ministry delegated prosecution powers to ERCA. The measure enabled the office to prosecute both officials and taxpayers implicated in fraud or corruption-without relying on external agencies. "ERCA sent new staff to the police college to attend investigation training for three months, and training on tax laws was provided by lawyers from the prosecutions directorate," said Gebretsadik. Next, the senior management team turned its attention to the recruitment of new staff. Making ERCA a more attractive place to work was an important step, and in order to compete with private-sector salaries-which were generally higher for skilled workers than public service wages-ERCA used its newfound autonomy to raise staff salaries. The new salary levels were higher than for the rest of the public service, but they remained lower than for comparable positions in the private sector. "Our recruitment strategy focused mainly on high achievers from the public universities," Gebretsadik said. The tax office built relationships with the country's universities and specified selection criteria for talented graduates. In the first two years after ERCA's founding, "we hired more than a thousand new staff-mainly university graduates as well as senior officers from other government departments," Gebretsadik said. The tax office also introduced mandatory induction training for all new hires by way of an agreement with Addis Ababa's Ethiopian Civil Service University, an institution created in 1995 to build capacity in the country's public service. Gebretsadik said the training courses covered the different tax laws and their administrative procedures and ranged from one month to three months for newly hired auditors. Senior tax officials helped design the curriculum to make sure the new procedures would get incorporated into the induction training-including for the SIGTAS software system that would become the IT backbone of the tax system-and the university provided accommodation and access to learning facilities and computers. Going digital Replacing the tax system's manual, paper-based information management system with a modern digital database was a pressing priority both to reduce the time required to pay taxes and to limit opportunities for corruption. Following the 2003 introduction of SIGTAS to handle corporate VAT, the government in 2006 decided to use the system for all other tax types and across all branch offices. Working with Dadour's consulting team, ERCA's IT directorate installed computers, internet connections, and software in all branch offices. Next, to create an integrated central database that covered the entire country and included taxes from all levels of government, ERCA's IT team in 2008 launched a project to install the new information platform, SIGTAS, in the offices of Ethiopia's 10 regional revenue authorities. As part of the project's feasibility study, Dadour's team visited four regional authorities to examine the different legal frameworks and procedures for the taxes under the exclusive jurisdiction of the regions. Dadour said he found significant differences among the laws and that those differences in turn "impacted operations and computerization." The level of organization and computerization varied across regions: some regions had already started to install computer systems, whereas others were completely manual. Dadour's team integrated into the new SIGTAS platform the old, legacy software still used in the regions. Alongside the rollout of IT infrastructure, Dadour's team worked to overcome difficulties with the issuance of taxpayer identification numbers, TINs. Ethiopia lacked a uniform national identity system: regional governments issued their own ID cards in their own regional languages. That made it difficult to base the issuance of TINs on identity documents-except for citizens who had passports. In response, the IT team proposed using biometrics and added a module to SIGTAS that enabled the tax authorities to capture fingerprints and link them to TINs. With the back-end systems in place, in 2010 ERCA built an electronic tax-filing module on SIGTAS and rolled out e-filing to selected large taxpayers on a pilot basis. It extended the program to all large taxpayers during the following tax season and in 2013 expanded it to cover medium-sized taxpayers in urban areas where power supply and internet connections were adequate. Although ERCA had hoped to launch a system for electronic payments alongside e-filing, that goal proved beyond reach at the time. "Ethiopia was not ready for electronic payments because the banking sector was not ready," Dadour said. In addition to improving the speed of collection and lowering the costs of collection, e-payment would have helped minimize interaction between tax officers and taxpayers to reduce the risk of petty corruption. Dadour said the delay was frustrating: "That really undermined the whole point [of electronic services], because there's little point in filing online if you still have to go to ERCA to pay." New information technology held promise for addressing a related problem too. For a VAT to work, the government had to have an accurate record of sales or purchases, and in a country where much commerce was informal, capturing such information was usually difficult. To help solve the problem, ERCA pressed ahead with an initiative to use electronic cash register machines to minimize VAT fraud by more accurately recording transactions across all sectors of the economy. The machines, known as tills in some countries, issued receipts to customers following each transaction. A 2007 legal directive required incorporated firms as well as unincorporated businesses with annual turnovers of more than 100,000 birr (equal to about US$12,500 at the time) to purchase the machines from approved suppliers-at a cost of US$260 to US$730 in 2013-and use them to record transactions.18 After a firm bought a machine, it had to present a receipt as proof at its nearest ERCA office. Local suppliers approved by ERCA could fix faulty machines, and while the repairs were taking place, a company could use an official receipt book to issue invoices.19 ERCA began enforcing the directive in some sectors beginning in 2008 and then extended the changes through a phased rollout. The tax office focused primarily on Addis Ababa, the country's economic heartland, from 2008 to 2011. The program dealt first with hospitality businesses such as hotels, bars, and restaurants and then incorporated other large taxpayers and businesses located in the capital city, gradually reaching the rest of the country.20 When connected to the internet, the machines sent electronic data on each transaction to ERCA's computer servers in real time. However, based partly on internet reliability problems, ERCA also required firms to submit printouts of all transactions on a monthly basis alongside monthly VAT declarations. Although the machines thus improved the tax office's ability to match transactions against VAT declarations, use of the machines did not make it substantially quicker or easier for taxpayers to comply with VAT procedures. A 2018 study conducted by researchers at the International Centre for Tax and Development noted, "Firms have generally complied with this process, as not doing so would put them at risk...Failure to use the machine can result in harsh monetary penalties and even imprisonment, with a serious risk of business closure."21 Raising public awareness Reaching out to taxpayers was also on the reform agenda. As part of its organizational restructuring in 2008, ERCA created a taxpayer service division at its headquarters as well as at each of its 20 regional branch offices (not to be confused with the offices of the 11 different regional revenue authorities, which operated independently). To answer taxpayer questions, the authority also created its own call center to provide information during business hours. But to win compliance and goodwill required a more proactive approach. As a first step to improving its public image, ERCA in 2008 introduced a service-standards charter that outlined what customers could expect when visiting an ERCA office. At first, most of the authority's messaging focused on informing taxpayers of their rights and responsibilities under the charter and how to comply with procedures involving the VAT, which was still a relatively new tax in the country. "Starting from the director general to the lowest officer, we were all engaged in this campaign," Tafesse said. The authority tailored its educational programs to different types of taxpayers. For businesses and high-income earners, officials organized seminars to explain the changes then in progress. "We built close relationships with the country's business chambers, and we'd regularly conduct training for them in hotels and at conferences," Tafesse said. The authority also created its own television and radio programs called Gebi Lelimat (Revenues for Development), which broadcast weekly on stations across the country. In addition to interviews with experts who explained tax obligations and procedures, the programs frequently ran features on big development projects such as the Grand Ethiopian Renaissance Dam, destined to become the largest dam in the Africa when complete, and the government's project to build a tramway in Addis Ababa to showcase the link between paying taxes and development. The Gebi Lelimat Gazeta, ERCA's dedicated newsletter, which aimed at finance professionals, further amplified the message. To keep accountants and other tax practitioners up-to-date, the newsletter focused principally on communicating regulatory and other procedural changes. Finally, during annual tax week events, federal officials hosted promotional events throughout the country. ERCA's tax week was based on a similar event in the United Kingdom and was first held in 2008. During the week, the government exhibited investment projects financed using tax money, presented prizes to loyal taxpayers, answered tax questions, and hosted panel discussions that were broadcast live on television. Not surprisingly, some of the educational programs were more effective than others. For instance, at the 2018 Ethiopian Tax Research Network meeting in Addis Ababa, attended by the network's tax scholars and experts, Ato Getachew of the Ethiopian Development Research Institute discussed evidence indicating that taxpayers regarded training seminars and workshops as well as the weekly radio programs as the most useful initiatives. In contrast, taxpayers rated the call center and tax week celebrations as the least useful communication methods.22 Getachew also questioned why the tax week events took place during the middle of the financial year instead of closer to the annual tax deadline. Coordinating with regions To make the new tax policies and administrative process work, ERCA had to closely coordinate with the regional revenue authorities. The regional authorities enjoyed autonomy in collecting 22% of Ethiopia's total tax revenue. Regional governments did not remit that money to the federal government, and they got additional revenue from taxes collected by ERCA. Plus, the country's laws specified that ERCA would collect all VAT from formally incorporated companies, as well as corporate income taxes, and then share 30% and 50% of the respective revenues with the region the taxpayer was registered in. A key step in the improvement of coordination was taken by the 2008 project to install SIGTAS in the offices of regional authorities in order to create one central database that stored information on all taxes across federal and regional levels. But it was obvious that the regions would require ongoing capacity building and support, including help in using SIGTAS. As a result, ERCA established a directorate dedicated to regional cooperation. "As at the federal level, the major challenges in the regions are capacity and manpower," said Tesfaye Mergai, director of ERCA's regional support team. "So our whole purpose is to cooperate rather than monitor, because [regional revenue authorities] are not accountable to us. They are accountable to the regional governments." The new directorate put together a team of officials who traveled the country to build relationships with the regions. "We visit [regional revenue authorities] and collect information on how they are working. If there are best practices, we transfer that to other regions," Mergai said. The directorate also offered training when warranted. For example, Mergai added, "if we identified weaknesses [in how the regions conducted their audits], we asked ERCA's audit experts to provide training for the regions." In turn, regional officials coached officials in the woredas (districts), the local government offices. The directorate also worked to harmonize tax policies. During his project to roll out SIGTAS to the regions in 2008, Dadour found that regional authorities sometimes enacted laws and regulations that created different rates or administrative procedures for the taxes under their control. But after ERCA formed, there were no new disparities. "It is mostly a function of the good coordination between the federal and regional levels, because no law explicitly says that things like tax rates should be harmonized," Mergai said. Abdis Hufira, a director in the Oromia regional revenue authority, agreed: "We apply the same strategy [as ERCA], and the relationship between the federal and region is good." Every six months, ERCA brought together representatives of all the subnational authorities-nine regions and two cities-for consultative meetings to identify discrepancies and challenges in policies among the authorities. In addition, ERCA directly involved the regional revenue authorities throughout the policy-making process. For example, when ERCA drew up a draft regulation, it communicated the text to the regions, which were invited to submit written comments. Each region then held a series of meetings with ERCA and the finance ministry's tax policy department to work out differences and generate consensus on the proposed policy. The close consultations and improved relationships between the federal and regional revenue authorities paid off. After ERCA issued a policy directive, regions used the federal policy as a template for developing their own versions. "The regions will draft similar laws with only very minor changes, but regarding the rates and the tax base, there is no disagreement," Mergai said. Hufira added, "We basically translate [the proclamation] into Oromo and customize it to fit our regional situation." Reaching microtaxpayers in Addis Ababa In addition to ERCA's efforts to coordinate effectively with all 11 regional revenue authorities, the federal tax office in 2010 began to play a more proactive role so as to ensure that microenterprises-informal businesses, often run by households, that were unincorporated and had turnovers of less than Br100,000 (equal to about US$7,400 in 2010)-were registered and paid their fair shares of taxes. A long-standing presumptive turnover tax, administered by the regional governments, was the main tool in use for promoting compliance by microenterprises. Under the presumptive tax, microenterprises did not have to keep detailed accounting books or submit self-assessments. Instead, the regional revenue authorities used a standard assessment method to estimate the turnovers of microenterprises and then calculated a lump sum owed based on the sector in which the business operated. In essence, the presumptive tax created a much simpler framework whereby small businesses paid a lump-sum turnover tax rather than a complicated, corporate income tax. The lump sum generally amounted to about 2% of an enterprise's annual turnover. Although the presumptive tax simplified procedures and made it easier for microtaxpayers to comply, the regional authorities sometimes lacked the ability to visit and accurately assess all of their jurisdictions' registered microenterprises. The problem was particularly acute in Addis Ababa, which accounted for the vast majority of the country's small businesses. One marketplace location in Addis Ababa, called Merkato, included the biggest open-air market in Africa. Covering 114 hectares with nearly 15,000 enterprises that consisted of 2,500 stalls, 1,500 service businesses, and 80 wholesalers that employed more than 13,000 people, Merkato sometimes received more than 200,000 shoppers and visitors daily.23 The Addis Ababa Revenue Authority served 340,000 taxpayers divided into four categories based on annual turnover: micro (less than Br500,000, or US$17,000), small (Br500,000 to Br5 million, or US$17,000 to US$178,000), medium (Br5 million to Br40 million, or US$178,000 to US$1.4 million), and large (more than Br40 million, or more than US$1.4 million). The authority had 116 district offices that served micro and small taxpayers, and four offices for medium and large taxpayers.24 To generate greater tax compliance in the country's strongest economic region, beginning in 2010 ERCA assumed administrative control of the Addis Ababa regional revenue authority. (Addis Ababa remained under ERCA administration until 2018.) ERCA claimed that the controversial move was permissible under Ethiopia's constitution.25 In 2011, ERCA designed a new assessment framework for estimating turnover and taxes due from microenterprises. (Under the presumptive tax's legal framework, the assessment criteria had to be updated every three years.) Teams of officials crisscrossed the city to visit business premises and calculate dues. In 2012, ERCA also opened two dedicated offices at Merkato. The offices divided the market into blocks, and officials visited each individual trader. Those with turnovers of less than Br100,000 (equal to about US$11,400 in 2012) were required to pay the presumptive tax, and officials formally registered for payment of corporate income tax those businesses with higher turnovers. "We introduced the block system based on the methodology used by the Tanzania Revenue Authority, which divided businesses into blocks to make sure that all taxpayers get identified," said Sebsbie Fekade, a senior ERCA researcher. "Initially, there were more than 5,000 businesses [at Merkato] without licenses. Now they are registered and have TINs, and we give them ongoing training," Fekade said. When ERCA relinquished control over Addis Ababa in early 2018, the regional revenue authority retained the two offices at Merkato and had the tools to track and enforce compliance on the parts of small taxpayers in the capital city. In cases when payments amounted to less than Br1,000 (US$35), taxpayers paid in cash at the Merkato offices. If payment was going to be more than Br1,000, taxpayers had to pay with a payment order at a bank branch. OVERCOMING OBSTACLES For five years after ERCA launched, unwavering political support by Prime Minister Meles helped propel tax reform forward. "At the political level, the [prime minister] played a big role with regard to implementing the reforms," Fenta said. "In fact, Meles raised the issue with the party's executive committee and gave clear directions that the authority is off-limits to any unbridled interference by officials. . . . We worked pretty much without interference [under Meles]."26 The hands-off policy ended in August 2012 when Meles died in office and was replaced by his deputy, Hailemariam Desalegn. Fenta told the Ethiopian Reporter, "Everybody started . . . to create interference."27 In May 2013, less than nine months after Meles's death, a team of police officers arrested Fenta at his office in the ERCA headquarters building in central Addis Ababa. Carrying a warrant citing corruption charges, the team searched his office and his home. "The team that was sent to arrest me was really excessive; it [looked] like they [were] apprehending a [terrorist]," Fenta told the Ethiopian Reporter.28 Fenta insisted that the charges were "trumped up and based on hearsay," and he alleged that the action was in response to his refusal to allow powerful people-including cabinet ministers-to evade customs duties on luxury imports.29 Fenta's arrest, along with the arrests of some of his deputies, sent shockwaves throughout ERCA and triggered an exodus of skilled officials. Fenta was ultimately released in 2018, having spent five years in jail awaiting the outcome of a trial that never concluded. The reform process ground to a halt. "When I saw [Fenta] handcuffed and in jail, it was very difficult," said Gebretsadik, who left ERCA a year after the incident. "This man had dedicated himself to this business. [Officials] started asking who would be next. It introduced a huge sense of insecurity," he added. Tafesse agreed that in the wake of the arrests, "the whole tax administration activity stagnated. It was very frustrating. We were a dedicated team, but now everyone just refrained from taking decisions." Amidst the upheaval and insecurity, ERCA did not renew its contract with the IT consultants that had helped install SIGTAS. With insufficient technical support and with officials weary of taking initiative lest they cross powerful people, the IT system rapidly decayed. In November 2015, Dadour's team returned to Ethiopia under the new Tax, Audit, and Transparency Programme-which was funded by the United Kingdom's Department for International Development-to implement emergency repairs to SIGTAS. By 2015, the most pressing problem involved the e-filing system, which the country's largest taxpayers used for filing their income taxes. The system was on the verge of collapse. "When you had 11 taxpayers logged in concurrently, the system would slow down, and when a 12th person logged in, it would crash," Dadour said. "In some cases, you would log in with your own name but end up in someone else's account." Another persistent problem was that a lack of management control also meant some officials entered dates on the IT system based on the Gregorian calendar, the most widely used civil calendar, whereas others used the Ethiopian calendar, in which dates fluctuated seven to eight years behind dates on the Gregorian calendar. What seemed like a small issue had "big ramifications in terms of the penalties calculated by the system for late filing," Dadour said. Still another problem was that too many officials were authorized to overrule the system. For example, SIGTAS automatically refused to issue a tax clearance certificate if a taxpayer had outstanding tax debt. But officials often issued a certificate nonetheless, undermining "a great tool to enforce compliance," Dadour said. Dadour said his team implemented a series of technical fixes from 2015 to 2018 and worked to clear the "mountain of bad data." With e-filing back on track, the team also built a new module for electronic payments that it rolled out on a pilot basis in September 2015 and then to all large taxpayers in 2016. The emergency IT upgrades, as well as the government's issuance of a new tax administration law in 2016 that attempted to further standardize procedures across the tax service, stabilized ERCA's operations. However, in contrast to ERCA's ambitious early plans to roll out electronic services to all taxpayers, both e-filing and e-payment were still available to only about 1,500 customers registered with the large-taxpayer office. Dadour also feared that ERCA would continue to view its digital database system as a purely IT project. He said the impact of the system's poor integration with tax procedures and management upheavals on SIGTAS's functionality after 2013 revealed that ERCA should approach IT "as a wider reform, where it becomes the backbone of the system." He added, "I believe ERCA has learned not to blame its problems on the IT system-and not to believe a new IT system would fix everything." The instability also undermined anticorruption efforts-a problem that could not be solved by technical fixes alone. In a 2015 enterprise survey conducted by the World Bank, one in six companies in Ethiopia said tax officials routinely expected gifts. In Transparency International's Global Corruption Barometer, most recently conducted in Ethiopia in 2013, 41% of respondents said they had to pay bribes to the revenue authority.30 ASSESSING RESULTS The World Bank's 2018 Doing Business index revealed the extent to which leadership instability and other troubles were hurting ERCA's efforts to bolster tax administration in Ethiopia. Although the average rate of taxation levied on companies declined to 38.6% from 43.6% in 2008, bringing it more closely into line with rates other countries levied, by 2018 Ethiopia ranked 133rd in the world-from 29th in 2008-in ease of paying taxes. Companies had to make 30 payments during a year compared with 29 in 2008, which cost a company in Addis Ababa 306 hours to complete compared with 52 hours in 2008.31 Still, the initial phase of reforms from 2007 to 2013 produced measurable gains. Total taxes collected nearly tripled to $3.8 billion from $1.3 billion-and reached $7.8 billion by 2017. And the share of total government revenues financed by taxes grew to 82% in 2016 from 48% in 2007.32 Even though tax revenues rose strongly in absolute terms, Ethiopia fell well short of its initial aim to increase the country's tax-to-GDP ratio to 17%. The ratio increased to 12.3% in 2013 from 10.7% in 2007 but stagnated at about 12.4% from 2013 to 2017 as the growth in revenue collection failed to keep up with the country's rapidly expanding economy. Government statistics showed the economy grew by more than 10% annually after 2010,33 raising per-capita GDP to $550 in 2017 from less than $200 during the early 2000s.34 Loose constraints on tax exemptions contributed to the relative underperformance on tax collection compared with economic growth. According to the World Bank's 2016 review of Ethiopia's public expenditure, "tax exemptions and privileges towards promoting domestic and foreign investment explain a great part of tax revenue forgone."35 Although data was hard come by, World Bank figures from 2010 showed that exemptions caused Ethiopia to lose 49% of the potential taxes it could have collected on trade. "This represents an average of 4.8% of GDP, which is a significant revenue forgone," the World Bank concluded.36 It was not until 2017 that the government took a decisive first step toward better managing tax incentives and exemptions. Whereas tax policy decisions previously were made by a small, underfunded unit in the finance ministry, the government in 2017 upgraded the unit to a fully-fledged tax policy directorate. One of the new directorate's first tasks was to conduct a cost-benefit analysis of tax loopholes. When the research was complete, Mulay Weldu, head of the new directorate, said that from 2019, "we're going to make tax incentives part of the annual budget by listing them as expenditure items." With better data in place, the directorate hoped to eventually design more effective criteria for issuing exemptions-and to better control their implementation. Over time, ERCA's organizational restructuring and investments in building capacity and improving service delivery produced mixed results. From 2007 to 2015, Public Expenditure and Financial Accountability-a public financial management assessment program operated by development partners-registered some improvement in its assessment of Ethiopia's tax system. The country's ranking on the transparency of taxpayer obligations and liabilities improved to A from B-, and the effectiveness of tax registration and assessment improved to B from C.37 According to the World Bank's Doing Business index, in 2006 a company in Addis Ababa had to make on average 20 payments, spend about 52 hours per year, and pay 43.6% of its gross profit to be tax compliant.38 By 2012, ERCA had registered some improvements, because the effective tax rate was down to 31.1% and companies had to make one fewer payment per year, although they had to spend much more time-198 hours a year-on taxes.39 REFLECTIONS Girma Tafesse, head of the directorate that coordinated the Ethiopian Revenues and Customs Authority's (ERCA's) 31 branches across Ethiopia, stressed that there was a direct link between the disruption caused by Director General Melaku Fenta's removal in 2013 and Ethiopia's failure to achieve broad and strong gains in reforming its system of tax administration. "It had a big impact, because trained tax officers left the organization, and the turnover was very high. Capacity diminished. And decision making also became slow, because no one [felt empowered] to decide promptly," Tafesse said. Although the original decision to retain the head of ERCA as a member of the cabinet was based on the desire to give the revenue authority sufficient political clout to undertake difficult reforms, the politicization of the organization ultimately contributed to the unraveling of the reform process. After the arrest of Fenta, who had headed the domestic tax office and then ERCA for a total of eight years, the authority had four different directors general from 2013 to 2018. ERCA's experience also demonstrated a weakness inherent in two aspects of the reform process that normally might be considered advantages: strong support at the top level of government and the concentration of authority in one highly skilled individual. In this case, both made the institution vulnerable to interference. Prime Minister Meles Zenawi was a firm believer in overhauling Ethiopia's system of tax administration, and the reform effort relied largely on the close working relationship between Meles and Fenta. Aside from providing political support, Meles's government empowered Fenta by giving him significant powers to fire officials suspected of corruption without the option to appeal, to appoint ERCA's own investigators and prosecutors, and to chair and select members of the tax advisory board. However, because the reform program depended so greatly on the relationship between Meles and Fenta, the effort stalled after Meles's death opened the door to Fenta's arrest in 2013 on charges that remained unproven more than five years later. The loss of Fenta-and the related political pressure on ERCA-led to the departure of skilled personnel, whose absence created gaping holes in the agency's plans. Aside from slowing decision making and creating an insecure work environment, the impact of the leadership disruption was felt most acutely in the area of IT. Philippe Dadour, a Canadian IT expert who intermittently worked with ERCA for more than a decade, compared the Ethiopian case to his experience in Rwanda. Both Ethiopia and Rwanda initially viewed the implementation of the Standard Integrated Government Tax Administration System as a project limited to the IT directorate. However, through strong and stable strategic leadership, "Rwanda learned and transformed. . . . [The Rwanda Revenue Authority] realized that IT has to be owned by the whole organization because it becomes the backbone of the entire system," Dadour said. In contrast, even though ERCA managed to prevent the collapse of its IT system in the mid 2010s, it failed to meet its original goals of taking advantage of IT, using data strategically, and expanding the use of electronic services to the majority of taxpayers. "I believe Ethiopia learned lessons from the past decade, and hopefully it will approach [future IT upgrades] as wider strategic reforms whereby it aligns tax processes with organizational structure" with the IT system, Dadour concluded. In addition to the lingering issue of lost revenues through unconstrained exemptions, as well as politicization and lack of stable leadership at the tax authority, the decentralized nature of Ethiopia's political structure posed a third challenge that clouded the outlook for further progress on reform. Although ERCA and the finance ministry had built an effective system for coordinating both policy and administrative decisions with the country's 11 regional revenue authorities, the deeper, unresolved issue of revenue sharing fueled resentment in some regions. Their first complaint was that the federal government shared only 30% of the VAT revenue it collected from corporate taxpayers with the regions the companies were registered in. That was in sharp contrast to the arrangement for corporate income taxes, wherein the regions got 50% of the revenues ERCA collected. "We don't know what the basis is for the 30-70 split, but actually, it should be 70-30 in favor of the regions," said Abdis Hufira, a director at the Oromia regional revenue authority. "Because the services that support the companies are delivered by regional governments, regional governments should get the majority of the [VAT] revenues." A second, more fundamental problem plagued the relationship between revenue authorities and raised the possibility of political resentment. For the corporate income tax, with its 50-50 split, the federal government decided which region was entitled to the share for subnational government based simply on where the company was registered. In practice, that meant most of the money flowed to the country's capital and economic hub, Addis Ababa, where most companies chose to be headquartered. Even if a company had all of its factories and carried out all of its operations in a region outside Addis Ababa, 50% of all corporate tax generated by the company would go to Addis Ababa, and the region hosting its operations would get nothing. The impact was even more pernicious in the case of VAT administered by regional revenue authorities on non-incorporated traders, many of whom bought items in markets like the Merkato in Addis Ababa and then sold them elsewhere in the country. Even though traders paid 15% VAT upon purchasing their inventories in Addis Ababa, the law stipulated that they were entitled to a VAT input credit-also known as a refund-from the region the trader was registered in. And because the regional revenue authorities operated independently, it would be up to a regional state or city administration where the trader was registered to reimburse the trader. Wollela Yesegat, a professor at the Addis Ababa University, explained that "this created a problem for the government, because the trader is usually sourcing inputs from a wealthier region but then claiming credit from a poorer regional state. The effect is to shift revenue from the poorer regional states to wealthier ones." And because Ethiopia's federal regions are constituted on the basis of ethnicity, that arrangement also could fuel resentment that poorer regions housing people from a particular ethnic background were subsidizing other, wealthier regions. Plus, the division of authority that made regions responsible for collecting the presumptive turnover tax from unincorporated businesses-whereas ERCA administered taxes on formally incorporated businesses-made it difficult to build a coherent national message on the importance of paying taxes. With the responsibility for improving compliance and building a social contract with taxpayers dispersed between the regions and ERCA, Ethiopia's efforts to raise public awareness risked becoming fragmented and incoherent. Despite the lingering challenges, the inauguration of 41-year-old Abiy Ahmed as prime minister in early 2018 produced hopeful signs that reforms could get back on track. In early 2018, Ahmed pardoned and released thousands of political prisoners, including Fenta, who still had charges pending after five years in jail. A few months later, in November 2018, Ahmed's government raised ERCA's status by making it the Ministry of Revenues. Turning the authority into a full ministry bolstered its political clout and potentially gave it greater power to coordinate with the regional revenue authorities. Ahmed appointed former mayor Adanech Abebe of the city of Adama as the new minister of revenues. The government also established a powerful and independent tax appellate commission to investigate tax grievances and appeals by businesses. The commission reported directly to the prime minister, rekindling the hope that improved tax collection was again a political priority on the part of Ethiopia's prime minister and that Ahmed's government would finally be able to complete the reforms launched more than a decade earlier. References 1 World Bank Data, GDP per capita (constant 2010 US$), accessed November 6, 2018; https://data.worldbank.org/indicator/NY.GDP.PCAP.KD?locations=ET. 2 World Bank, Ethiopia Poverty Assessment 2014, accessed November 6, 2018; https://openknowledge.worldbank.org/handle/10986/21323. 3 Haile Kibret and Roza Mamuye, "Performance and Prospects of Tax Collection in Ethiopia," United Nations Development Programme, 2016, accessed October 22, 2018; http://www.et.undp.org/content/dam/ethiopia/docs/Performance%20and%20prospects%20of%20tax%20colllection%20in%20ET.pdf. 4 Giulia Mascagni, "A Fiscal History of Ethiopia: Taxation and Aid Dependence 1960-2010," International Center for Tax and Development Working Paper 49, March 2016, p. 31, accessed October 22, 2018; https://opendocs.ids.ac.uk/opendocs/bitstream/handle/123456789/12777/ICTD_WP49.pdf. 5 Giulia Mascagni, "A Fiscal History of Ethiopia: Taxation and Aid Dependence 1960-2010," International Center for Tax and Development Working Paper 49, , March 2016, p. 31. 6 Tsegabirhan Weldegiorigis Abay, Domestic Resource Mobilization in Sub-Saharan Africa: The Case of Ethiopia, North-South Institute, 2010, p. 28, accessed October 24, 2018; https://assets.publishing.service.gov.uk/media/57a08b07ed915d622c000a61/60667_Full_Case_Study_Ethiopia.pdf. 7 Transparency International, Global Corruption Barometer 2005, accessed November 6, 2018; https://www.transparency.org/whatwedo/publication/gcb_2005. 8 Giulia Mascagni, "A Fiscal History of Ethiopia: Taxation and Aid Dependence 1960-2010," International Center for Tax and Development Working Paper 49, March 2016, p. 31. 9 Tsegabirhan Weldegiorigis Abay, Domestic Resource Mobilization in Sub-Saharan Africa: The Case of Ethiopia, North-South Institute, 2010, p. 29. 10 Tamiru Tsige, "From the Office to the Prison Cell: The Story of Melaku Fenta," Ethiopian Reporter, July 28, 2018, accessed October 24, 2018; https://www.thereporterethiopia.com/article/office-prison-cell. 11 Tamiru Tsige, "From the Office to the Prison Cell: The Story of Melaku Fenta." 12 Haile Kibret and Roza Mamuye, "Performance and Prospects of Tax Collection in Ethiopia," United Nations Development Programme, 2016, accessed October 22, 2018; http://www.et.undp.org/content/dam/ethiopia/docs/Performance%20and%20prospects%20of%20tax%20colllection%20in%20ET.pdf. 13 World Bank Data, Rural population (% of total population), accessed November 13, 2018; https://data.worldbank.org/indicator/SP.RUR.TOTL.ZS?locations=ET. 14 Ministry of Finance and Economic Development, A Plan for Accelerated and Sustained Development to End Poverty (PASDEP), p. 62, September 2006, October 25, 2018; http://siteresources.worldbank.org/INTETHIOPIA/Resources/PASDEP_Final_English.pdf. 15 Ministry of Finance and Economic Development, A Plan for Accelerated and Sustained Development to End Poverty (PASDEP), p. 62. 16 Federal Democratic Republic of Ethiopia, Proclamation No. 587/2008: A Proclamation to Provide for the Establishment of the Ethiopian Revenues and Customs Authority, accessed November 6, 2018; http://www.erca.gov.et/images/Documents/Proclamation/Custom/415.pdf. 17 Federal Democratic Republic of Ethiopia, Proclamation No. 587/2008: A Proclamation to Provide for the Establishment of the Ethiopian Revenues and Customs Authority. 18 Giulia Mascagni, Andualem T. Mengistu, and Firew B. Woldeyes, "Can ICTs Increase Tax? Experimental Evidence from Ethiopia," International Centre for Tax and Development Working Paper 82, p. 10, July 2018, accessed October 30, 2018; https://opendocs.ids.ac.uk/opendocs/bitstream/handle/123456789/13845/ICTD_WP82.pdf?sequence=1&isAllowed=y. 19 Giulia Mascagni, Andualem T. Mengistu, and Firew B. Woldeyes, "Can ICTs Increase Tax? Experimental Evidence from Ethiopia," International Centre for Tax and Development Working Paper 82, p. 10. 20 Giulia Mascagni, Andualem T. Mengistu, and Firew B. Woldeyes, "Can ICTs Increase Tax? Experimental Evidence from Ethiopia," International Centre for Tax and Development Working Paper 82, p. 10. 21 Giulia Mascagni, Andualem T. Mengistu, and Firew B. Woldeyes, "Can ICTs Increase Tax? Experimental Evidence from Ethiopia," International Centre for Tax and Development Working Paper 82, p. 11. 22 Ato Abis Getachew, findings presented at the Second Meeting of the Ethiopian Tax Research Network, Capital Hotel, Addis Ababa, October 8, 2018. 23 City Government of Addis Ababa, Merkato, 2012, accessed October 31, 2018; https://web.archive.org/web/20130331123604/http://www.addisababacity.gov.et/index.php/en/about-addis/shopping/merkato. 24 "Ethiopia: Authority to Open Large Taxpayer Office," AllAfrica, August 26, 2018, accessed November 13, 2018; https://allafrica.com/stories/201808280321.html. 25 Constitution of the Federal Democratic Republic of Ethiopia, article 43 (3), 1995, accessed November 14, 2018; http://www.wipo.int/edocs/lexdocs/laws/en/et/et007en.pdf. 26 Tamiru Tsige, "From the Office to the Prison Cell: The Story of Melaku Fenta," Ethiopian Reporter, July 28, 2018, accessed October 24, 2018; https://www.thereporterethiopia.com/article/office-prison-cell. 27 Tamiru Tsige, "From the Office to the Prison Cell: The Story of Melaku Fenta." 28 Tamiru Tsige, "From the Office to the Prison Cell: The Story of Melaku Fenta." 29 Tamiru Tsige, "From the Office to the Prison Cell: The Story of Melaku Fenta." 30 Transparency International, Global Corruption Barometer 2013, accessed November 14, 2018; https://www.transparency.org/gcb2013/country?country=ethiopia. 31 World Bank, Doing Business 2018, p. 160, accessed November 1, 2018; http://www.doingbusiness.org/content/dam/doingBusiness/media/Annual-Reports/English/DB2018-Full-Report.pdf. 32 Author's calculations based on UNICEF, Budget Brief: A Synopsis Analysis of 2006-2016 National Trends and the 2017/2018 Federal Budget, p. 15, accessed November 1, 2018; https://www.unicef.org/esaro/UNICEF_Ethiopia_--_2017_--_National_Budget_Brief.pdf. 33 World Bank Data, GDP growth (annual %), accessed November 1, 2018; https://data.worldbank.org/indicator/NY.GDP.PCAP.KD?locations=ET. 34 World Bank Data, GDP per capita (constant 2010 US$), accessed November 6, 2018; https://data.worldbank.org/indicator/NY.GDP.PCAP.KD?locations=ET. 35 World Bank, Ethiopia Public Expenditure Review, April 2016, p. 16, accessed November 1, 2018; https://www.tralac.org/images/docs/9735/ethiopia-public-expenditure-review-world-bank-april-2016.pdf. 36 World Bank, Ethiopia Public Expenditure Review, April 2016, p. 16, accessed November 1, 2018; https://www.tralac.org/images/docs/9735/ethiopia-public-expenditure-review-world-bank-april-2016.pdf. 37 Public Expenditure and Financial Accountability, Ethiopia 2007, accessed November 1, 2018; https://pefa.org/assessment/et-oct07-pfmpr-public-en and Public Expenditure and Financial Accountability, Ethiopia 2015, accessed November 1, 2018; https://pefa.org/assessments/ethiopia-2015. 38 World Bank, Doing Business 2006, p. 124, accessed November 1, 2018; http://www.doingbusiness.org/content/dam/doingBusiness/media/Annual-Reports/English/DB06-FullReport.pdf. 39 World Bank, Doing Business 2012, p. 95, accessed November 1, 2018; http://www.doingbusiness.org/content/dam/doingBusiness/media/Annual-Reports/English/DB12-FullReport.pdf. Innovations for Successful Societies makes its case studies and other publications available to all at no cost, under the guidelines of the Terms of Use listed below. The ISS Web repository is intended to serve as an idea bank, enabling practitioners and scholars to evaluate the pros and cons of different reform strategies and weigh the effects of context. ISS welcomes readers' feedback, including suggestions of additional topics and questions to be considered, corrections, and how case studies are being used: iss@princeton.edu. Terms of Use Before using any materials downloaded from the Innovations for Successful Societies website, users must read and accept the terms on which we make these items available. The terms constitute a legal agreement between any person who seeks to use information available at successfulsocieties.princeton.edu and Princeton University. In downloading or otherwise employing this information, users indicate that: a. They understand that the materials downloaded from the website are protected under United States Copyright Law (Title 17, United States Code). This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. b. They will use the material only for educational, scholarly, and other noncommercial purposes. c. They will not sell, transfer, assign, license, lease, or otherwise convey any portion of this information to any third party. Republication or display on a third party's website requires the express written permission of the Princeton University Innovations for Successful Societies program or the Princeton University Library. d. They understand that the quotes used in the case study reflect the interviewees' personal points of view. Although all efforts have been made to ensure the accuracy of the information collected, Princeton University does not warrant the accuracy, completeness, timeliness, or other characteristics of any material available online. e. They acknowledge that the content and/or format of the archive and the site may be revised, updated or otherwise modified from time to time. f. They accept that access to and use of the archive are at their own risk. They shall not hold Princeton University liable for any loss or damages resulting from the use of information in the archive. Princeton University assumes no liability for any errors or omissions with respect to the functioning of the archive. g. In all publications, presentations or other communications that incorporate or otherwise rely on information from this archive, they will acknowledge that such information was obtained through the Innovations for Successful Societies website. Our status (and that of any identified contributors) as the authors of material must always be acknowledged and a full credit given as follows: Author(s) or Editor(s) if listed, Full title, Year of publication, Innovations for Successful Societies, Princeton University, http://successfulsocieties.princeton.edu/ ISS is program of the Woodrow Wilson School of Public and International Affairs: successfulsocieties.princeton.edu. ISS invites readers to share feedback and information on how these cases are being used: iss@princeton.edu. (c) 2018, Trustees of Princeton University. This case study is made available under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Leon Schreiber Innovations for Successful Societies (c) 2018, Trustees of Princeton University Terms of use and citation format appear at the end of this document and at successfulsocieties.princeton.edu/about/terms-conditions. 4 6 (c) 2011, Trustees of Princeton University