Reforming Public Financial Management in Somaliland Faculty Research Working Paper Series July 13, 2015 Yusuf Mohamed Ismail1 Yusuf Soraan MA of Tax and Custom Administration

Abstract. This study elaborates framework for understanding public financial management reform based on the four drivers of public sector reform: COPS—context, ownership, purpose and strategy. Public financial management is a core function of the state and its sovereignty and it is not an appropriate arena for foreign aid intervention—governments must fully own it, which was a key to success.

Keywords: public financial management reform, foreign aid, budget support, economic

development, Somaliland Author’s E-mails: [email protected] or [email protected] This Faculty Research Working Paper Series on the Present text and the author would welcome any comments.

1

Yusuf Soraan is deeply committed to revenue reform issues and has extensive knowledge of tax and custom administration

(Graduate Institute of Tax and Custom Administration in Ethiopian Civil Service University); He participate the inception phase of 6879: Somaliland strengthening revenue policy and administration; so he currently is serving as Municipal Finance advisor with Hargeisa Municipality, Somaliland.

Hopefully for students, but especially practitioners, this study starts to address this oversight and provides a new landscape for understanding and guiding PFM reforms.

2

Historical Context Somaliland covering an area of 137,000 km per square kilometer along the coast of the Gulf of Aden, and bordering Djibouti to the northwest, Ethiopia to the south and west and the Somali to the east, the population consists of nomads 55% and urban and rural dwellers. The annual population growth rate is 3.14%, the annual birth rate is 4.46% and the death rate is 1.32%. Somaliland tax system was first introduced by the British Protectorate administration in the early 20th century, the people of Somaliland Protectorate considered taxation an unjust imposition by the colonizing or Protectorate power and thus it is unclean and profane. In addition to that the business and trade activities were very limited; At that time the British Protectorate administration faced a constraint in taxing, and its budget was required subsidies from the British government, because the amount of revenue collected was not applicable to cover the expenditures of the administration. Following more than 75 years of Protectorate rule, Somaliland British Protectorate gained its independence on 26 June 1960. Just five days later, on 1 July, it joined with the former Italian Somalia to create the unitary state of the Republic of Somalia. This was intended to precede the Pan-African dream of that period, but produced a major catastrophe for the people of Somaliland. Despite early optimism associated with independence and nationhood processes of unification, integration and rule proved difficult. The formal administrative structures for the new state were put in place but, after nearly a decade of unified Somali nationhood married by increasing levels of state dysfunction, conflict and clannish, Somalia’s second president, Abdirashid Ali Shermarke, was assassinated on 15 October 1969. On the 21 October, General Mohammed Siyad Barre led a bloodless coup that brought Somalia under highly centralized military rule for the next two decades. The 1980s were marked by deterioration into full-scale civil war that pitted the Somali army against insurgent groups. In the formal regions of Somaliland British Protectorate opposition to the government was led by the Somali National Movement (SNM).This organization formed in 1981. The weight of insurgent group’s confrontation and war proved too much for Barre’s regime. The Somalia’s

3

central authority fell in 1991, leaving in its wake decimated social, political and economic infrastructure, particularly in the regions of the formal British Somaliland Protectorate. In 18 may 1991 Somaliland British Protectorate claimed back gain its independence and formed its own government and ratified a new constitution with broad public support; and also build a system of government of a multiparty electoral democracy featuring a bicameral parliament. The president, Members of the House of Representatives and Local Councils have all been chosen through peaceful and transparent elections that have been witnessed and confirmed by international observers. This allowed the country and its population to recover from the civil war; but regarding our failed union with Somalia, we are by no means the first African State to have entered into a voluntary union with another the subsequently withdrawn from that union intact; but there are several countries have all done likewise and have never been punished for it as Somaliland is being punished; but in spite of the fact that some progress has been made compared to the original point of departure.

4

REFORMING PUBLIC FINANCIAL MANAGEMENT IN SOMALILAND Designing an appropriate and workable tax reform is not an easy task in any country. To have such a reform accepted and then successfully implemented is much more difficult. Managing the whole process of tax reform from inception to conclusion is far from simple, not least because of the inherently political nature of any significant tax reform. There is no simple, magic answer to the many complexities facing those who undertake this task. Each country, and indeed each major reform, will, and must, take its own particular path. Any major tax reform is unique in the sense that it takes place in the unique circumstances of that place at that time. Public financial management reform has played a critical role contributing to economic stability, nation building, and poverty reduction, it also enables the government to perform effectively and deliver public goods and services to its citizens, but these issues apparently required committed leadership. This study elaborates a framework for understanding Somaliland ministry of finance; secondly the study to give readers a glimpse of some of the different ways in which tax reform can be approached and, to highlight particular aspect that may help illuminate and how it might be managed; thirdly although the substance of reform cannot be avoided, in keeping with the thrust of this study, the main emphasis here is public financial management reform based on the four drivers of public sector reform: COPS—context, ownership, purpose and strategy; However, the study argues that the government’s efforts in this regard are mostly ineffective and, worryingly, may have unintended consequences for reforming public financial management in Somaliland.

5

SOMALILAND MINISTRY OF FINANCE The Ministry of Finance Profile The mission of the Ministry of Finance is “raising sustainable revenue growth, proper taxation laws, and transparent public funds management”. This mission statement supports it’s broader vision of “developing Somaliland into a country free of poverty and economically sustainable”, which is in line with the overall vision of the Somaliland Government for “a stable, democratic government and prosperous country where the people enjoy a high quality of life.” The Ministry of Finance derives its mandate from the Constitution of Somaliland, which provides for proper budgetary and expenditure management of government financial resources, including:  Formulating financial and economic policies;  Developing and maintaining sound fiscal and monetary policies that facilitate socioeconomic development;  Regulating the financial sector, which is central to the development of the country and on which all other sectors depend for investment resources;  Managing revenues, expenditures and borrowings made by the government, ensuring that it mobilizes adequate resources to support government programs and activities. Consequently, the Ministry has the task of developing sound fiscal policies that ensure sustainable budget deficits;  Ensuring that government expenditure is within the revenue collected to reduce domestic borrowing, which tends to cause negative ripples in economic management;  Managing and co-coordinating bilateral and multilateral development financing and the associated technical assistance; and  Assisting and co-coordinating government ministries/departments in the preparation of the annual national budget. The legislative framework underpinning the Ministry of Finance’s activities is yet to be established and there are a number of laws that have yet to be enacted. These include: 

Public Financial Management & Accountability Bill



The National Audit Bill



The Public Procurement Bill



The Customs Bill



Somaliland Revenue Bill 6

According to the Ministry of Finance Personnel Section, the employee count for Permanent Officials includes 966 staff within the whole Ministry of Finance and 188 staff in the Hargeisa 2 Head Office. From a departmental perspective, the breakdown is as follows: Table 1: Staff Breakdown, Ministry of Finance, Hargeisa Head Office Department Department of Admin & Personnel

Number of Personnel 33

Department of Planning and Statistics

9

Department of Budgets

18

Internal Auditor

14

Accountant General

114 (Hargeisa: 72 & Regional: 42)3

Staff numbers in the Inland Revenue and Customs Departments are shown below. Currently the Customs Department is slightly larger than the Inland Revenue Department, but unfortunately it couldn’t be that. Table 2: Staff Breakdown, Inland Revenue and Customs Departments Department Inland Revenue Department

Number of Personnel 379

Inland Revenue Head Office & Hargeisa District Offices

248

Regional Offices

131

Customs Department

399

Customs Head Office Maroodi-Jeh Regional Office4 (including Hargeisa District Offices) Regional Offices5

63 83 253

2

The capital city of Somaliland Accountant General regional staff included in this section for reference 4 Maroodi-Jeh regional office includes offices within the larger region. It has 4 major offices: Hargeisa (53 staff including airport customs offices), Abaarso (21 staff), Faraweyne (4 staff) & Balligubadle (5 staff). 5 These offices are within the remaining five regions, for a total of 13 offices. 3

7

Minister of Finance holds a meeting each week with the Director General DG and his six directors (director of finance and administration; director of planning and statistics; director Customs; director of Inland Revenue; director of budget; director of Human resource).Topics of discussion are likely centre around a mix of political, executive and management affairs of the Ministry. The Ministry does not appear to have a well-defined management cadre outside the DG/Director team. The concept of ‘management’ does not appear to be well established for instances; management’s role in selection, training, orientation, performance management, discipline and setting work expectations (e.g. job descriptions) is either sporadic or non-existent. Appointment, transfer and disciplinary processes are centralized to a few people within line programmes, mostly at the executive level. Supervisory responsibilities are inconsistent, with many leaders having a very wide and ill-defined span of control. The management cadre is also undefined and titles and roles appear to be more concrete and understood starting at the director role, though even there are no written position descriptions defining responsibilities and qualifications. Revenue analysis: is currently constrained as most revenue data is collected and recorded manually, making information difficult to access, retrieve and organize and calling into question its accuracy. Although generally data recording systems are underdeveloped at all levels, remarkable progress has been achieved in some areas to explain the end of this paragraph; In the Department of Inland Revenue all data systems are manually created and maintained. There are no automated systems to perform functions such as registration, calculation of tax obligations, recording of tax payments and other basic functions. Since taxpayers are not assigned a tax identification number (TIN) the possibility of aggregating revenue data by taxpayer is virtually impossible at this stage, Similarly, in the Department of Customs all revenue data systems were manually created and maintained, including information on declaration of goods to be imported or exported, registration of goods, calculation of duties and tax obligations, recording of payments and other basic functions; Today, while the rest of the Customs offices still follow a manual process, the Port of Berbera has an operational information system that has been in place since November 2014. Although taxpayers do not have an assigned TIN, importers are issued a license number (Inception Report). The majority of government revenue is collected in the districts either at customs offices or central authority revenue collection offices. Both Customs and internal revenue collections and documentation are manual processes with little effective 8

supervision or control. Sums collected are deposited in district branches of the Central Bank of Somaliland and reports telephoned in to Hargeisa. At present there is no tax forecasting unit or dedicated functions in charge of preparing revenue forecasts at the ministry of finance. The Director of Planning and Statistics, based on his knowledge of historical revenue data and vast experience with the tax system, prepares the tax revenue estimates that are considered in the budget. There is no defined methodology for the calculation of future revenues. The set of macroeconomic data and other variables taken into account for estimation are not established institutionally or clearly defined in a set of operational guidelines and the main challenge the ministry of finance faces to develop capacity for revenue forecasting is the lack of availability of data. However, using recently systematized revenue and other economic data currently available (e.g. inflation estimates elaborated by the Ministry of National Planning and Development: the annual inflation-adjusted growth rate has been around 20 percent for the past five years 2010-2014), a revenue forecast model can be developed and institutionalized; currently the ministry of finance does not have a tax policy unit 6. Revenuerelated policy analysis and assessments of tax reforms are largely the responsibility of the Director of Planning and Statistics. Nonetheless, since 2012 important steps for developing the tax system have been taken, such as tax classification by type and code assignment based on the IMF GFS manual7 (Inception Report). Domestic taxes have a narrow tax base. Even though there is a wide range of domestic taxes being imposed, few goods and services are taxed. There are a number of “small” fees and specific taxes applied by the revenue department and municipalities, for which the collection costs for the tax administration plus the compliance costs to taxpayers could exceed revenue collected. Additionally, tax expenditure analysis is absent and the cost of exemptions, deductions, and concessions to the budget remains unknown; current policies regarding international trade taxes have several limitations. First, because the current tax base for import taxes is based on predetermined (not market) values, there are many items that are overvalued and many others that are undervalued. Second, because the exchange rate used by the Customs 6

Until 2012 there was a Fiscal Policy Unit at MoF that, among other objectives, was tasked with analyzing options for reforming Somaliland’s tax system 7 The Directorate of Planning and Statistics has an electronic copy of the 2014 IMF-GFS Manual (pre-publication draft) and is aware of the revenue classification system proposed in the Manual.

9

Department is significantly lower than the market exchange rate, revenue collections remain below potential. Third, there are additional charges levied alongside import duty, making the effective tax rate on imports substantially different from the statutory rate established in Customs legislation (Inception Report). Tax and Customs Administration: The Department of Inland Revenue at the Ministry of Finance is responsible for the administration of domestic taxes and non-fiscal revenues. Domestic taxes are governed by the Body of Laws dating back to 1966. The rates in the body of laws and some pertinent provisions have been changed over the years, but the legislation has not been updated accordingly and the changes are hard to track; whereas customs operations and taxes on international trade are managed by the Customs Department and governed by Customs Law No. 91/96 of 1966 and related regulations; although the fragmentation and incompleteness of the current legal framework hampers the establishment of a solid tax system (required to sustainably improving tax collections), The fragmented across many regulations and amendments to legislation and an exhaustive list is not readily available in the ministry of finance and there is no clear timeline for the passage of the draft tax and Customs laws currently awaiting government approval. The Inland Revenue Department is in charge of both direct and indirect domestic taxes. In terms of direct taxes, business income and rent income are taxed at 10% of net profit, while the employment income rate is 6%, of which 1% is stamp duty. The principal indirect tax is the sales tax which is imposed at 5% at retail level, at customs on imports and on locally manufactured goods at manufacturing level. There are miscellaneous taxes and fees that Inland Revenue collects in addition to the major taxes mentioned above, as indicted in figure 1. Most of these taxes are collected by customs at the point of entry, a practice that is common within emerging revenue agencies. Although collecting taxes at entry points is straightforward and limits leakages, it imposes a burden on taxpayers, who must pay income-related taxes upfront. Due to this disadvantage, as revenue administrations improve, such taxes are often replaced by domestic tax mobilization. In addition to taxes and fees, Inland Revenue collects a long list of non-fiscal revenue such as fees for passports and visas, penalties, traffic fines, etc. Almost all non-fiscal revenue paid to the central government is collected by the Inland Revenue and Customs Departments, but business registration service fees collected and used by the Ministry of 10

Commerce and Investment are a notable exception, and there may be others. Ideally, all central government revenue should be collected and accounted for by the central government revenue administration and only subsequently allocated to government departments through the central budgeting process (Inception Report). Figure 1: Other Miscellaneous Taxes and Fees 3% circulation tax, which is basically vehicle title transfer tax; 2. 3% registration tax for transfer of title on land when obtained from government by a person and on all contracts such as construction and other services; 3. 2.1% registration tax for transfer of title on land from one existing land owner to another person; 4. 2% administration tax imposed on invoices, submission of completed tax returns forms such as income tax, fees and value of imports (CIF value plus duty) at customs point; 5. 2.5% stamp duty on contract value plus the 3% registration fee 6. Road tax, which is fixed amount based on capacity of vehicle measured by horsepower or number of seats in the case of passenger vehicles, collected every six (6) months; 7. 2% sur-tax (Bariga) on value of imports collected at customs point; 8. 12.5% municipality tax on import duty plus sales tax collected at customs point; 9. 3% harbour tax on import value (CIF plus customs duty) imposed only at Berbera port; 10. Annual license fee on vehicles; 11. A one-time plate number fee; 12. Trade license fee. Sources of the inception report From a macroeconomic growth perspective sustained rapid growth of domestic revenue will be the most important progress toward fiscal sustainability, but in Somaliland as the ministry officials revealed domestic revenue are not large enough to cover total recurrent expenditures because the domestic revenue from the successful telecommunications and remittances industries didn’t features in the top ten sources of domestic revenue with the implication that many streets traders continuous to pay higher relative share of their earnings in taxation than Somaliland’s 11

largest companies; it can be understood that the total recurrent expenditures covered the capital investment financed by external aid channeled through the international agencies. In addition to such mainstream activities, compared globally, paying taxes in Hargeisa companies would spend on average 188 hours8 preparing tax returns and complying with other tax related requirements; No one likes taxes but taxes are necessary both to finance desired public spending in a non inflationary way and to ensure that the burden of paying for such spending is fairly distributed. Doing Business measures the time and cost (excluding tariffs) associated with exporting and importing a 20-foot container by sea transport, and the number of documents necessary to complete the transaction. The indicators cover documentation requirements and procedures at customs and other regulatory agencies as well as at the port. They also cover logistical aspects, including the time and cost of inland transport between company warehouses and traders’ ports; for instance Hargeisa is an inland city. The primary port used by the local traders is the port of Berbera, on the south side of the Gulf of Aden, about 160 kilometers northeast of Hargeisa. The port’s container-handling facilities present some insufficiencies. Due to limited space, container cargo handling and other trade-related operations take place in the same area. Infrastructure limitations together with operational inefficiencies cause major delays in the peak season of livestock exports, when the presence of a high number of animals in the terminal makes container handling operations hard to perform. At the same time, container ships with imports have to wait up to one week outside the port for vessels exporting livestock to leave before they can enter and unload. Compared globally to 183 economies measured by doing Business 2012, Hargeisa would rank 127 on the ease of trading across borders—ahead of Benin (129), and Sierra Leone (132), but behind Liberia (116) and the Republic of Yemen (118). An entrepreneur in Hargeisa needs to submit 7 documents9, wait 25 days, and spend $1,920 to import a 20- foot standardized container of cargo through the port of Berbera. To export through the same port, an entrepreneur needs to submit 6 documents10, wait 22 days, and spend $1,940. This is slower and more expensive than the case of Djibouti where importing takes 18 days and costs $911, while

8

36 hours are dedicated to corporate income taxes, 72 hours to sales taxes and 80 hours to labor-related taxes; this is below the average 318 hours spent in Sub-Saharan Africa economies. 9 The bill of lading, the packing list, the commercial invoice, the certificate of quality, the customs declaration form, the certificate of origin, and the boarding list. 10 the bill of lading, the packing list, the commercial invoice, the certificate of quality, the customs declaration form and the delivery order

12

exporting takes 18 days and costs $836. Compared to the average economy in Sub-Saharan Africa countries, the process is quicker and less expensive in Hargeisa. In fact in Sub-Saharan Africa, importing requires an average of 8 documents, takes 37 days and costs $2,502, while exporting requires 8 documents, takes 31 days and costs $1,960. This is because of infrastructure limitations. The road that connects Hargeisa to the port of Berbera is narrow and not well paved, which prevents large container trucks from using it. Hargeisa’s urban infrastructure does not allow for the free circulation of heavily laden or larger trucks. As a consequence, cargo for exports is typically manually loaded onto smaller trucks at the warehouse, taken to Berbera, and then manually loaded onto containers available at the port. Additionally, import containers are seldom released for transportation on land—or, if they are, it is against high deposits and the risk of detention charges. Because the shipping lines do not have empty-container depots near Hargeisa or any other inland destination, importers must either pay to return the empty container to port or unload the container in the port. This makes inland handling and transportation cumbersome, driving costs up.11 Local Governments: Local governments report their revenue collection to the Ministry of Interior (MoI). The Department for local government at the MoI receives data on paper and keeps in files but the department does not do any analysis or revenue forecasting. The relationship between the MoF and MoI in relation to tax revenue is not institutionalized. The Department of Planning and Statists of the MoF does not include local revenue in its analysis. It is important to note that at the tax information system of the Municipality of Hargeisa is relatively developed. In fact taxpayer registration, tax assessment and payment systems are much more developed than at the central level. The Municipality has two systems in place in all tax districts the Automated Information and Financial Management System (AIMS) and the Billing and Information Management System (BIMS), which have benefited the local governments and increased their overall revenues. There are more than 10,000 taxpayers registered in the system at the moment. These systems can be of significant benefit for the development of a national tax management system, not only for the local experience gained in developing and managing such systems, but also for the valuable taxpayer data already recorded and systematized which can be used to populate a national revenue IT system. 11

Doing Business in Hargeisa at www. Doing business.org/sub national

13

Support to revenue generation at the district level is provided by the UN Joint Programme on Local Governance (JPLG). Sub-national revenue derives from a mix of central government transfers, JPLG contributions and local taxes and fines, and has been steadily increasing. However there is a high dependency upon central government transfers and JPLG contributions, much attention needs to be given to strengthening local revenue collection capacity and efficiency of the 34 districts in Somaliland, local revenue collections are greatest in Berbera and Hargeisa and mostly derived from commodity, property and trade taxes. In Gabiley and Sheik, local revenue collections are minimal comprising only 8% and 13% of total revenue sources respectively. Gabiley’s local revenue is almost entirely generated by commodity taxes imposed upon Khat which is predominately imported at Kalabaydh close to the border with Ethiopia. In summary the ministry of finance in Somaliland, the majority of the private sector is small informal businesses which do not maintain good financial information on which taxes can be assessed meaning inland liabilities are presumed and negotiated rather than accurately calculated. The entire process is subject to delays, transcription errors and interference and lacks effective controls, oversight or monitoring of outstanding amounts. The process is subject to significant loses, under collection and incidences of fraud, in addition to that in terms of revenue policy, the main constraints are underdeveloped data recording systems and revenue collection analysis capacity, a lack of capacity for tax forecasting, and limited tax policy analysis functions, with no tax policy unit currently in place; In terms of revenue administration, there is an absence of clearly defined functions and performance management capacity in the Inland Revenue and Customs Departments; In terms of management and organizational reform some of the key constraints are an underdeveloped; Human Resource function at the Ministry of Finance, a lack of change management and leadership capacity, patronage-focused recruitment, and weak Human Resource and employee management.

14

APPROACHES TO TAX REFORM In a recent paper, Lledo, Schneider and Moore set out a useful categorization of some of the many approaches to tax reform found in the literature12. The four approaches they distinguish may be somewhat cavalierly categorized as follows: (a) The public economics approach. Found in any public economics text13, the focus of this approach, in essence, is on setting up a tax system that (in the purest version of the approach) maximizes social welfare, balancing efficiency and equity in accordance with society’s (assumed) objectives. Although this strongly normative approach dominates the economic literature, it has not had any greatly marked influence to date in the real world of tax reform. This approach usually stresses neutrality as the immediate objective of policy reform. (b) The macroeconomic approach. Although perhaps not the best label for it, the macroeconomic approach is also common ground among economists, especially those not specializing in public economics. This approach focuses less on the internal structure of the tax system and more on the impact of taxes on real aggregates, such as the growth rate and the level and distribution of income and wealth. Though hampered by the lack of much solid evidence in developing countries, beliefs about such matters (though seldom burdened by much consideration of either administrative or political concerns) have at times carried considerable weight in tax policy debates. This approach usually stresses the role of taxation in promoting economic growth. (c) The administrative approach. As this author has discussed at length elsewhere14, this approach emphasizes the interplay between what can be done and what should be done. Successful reforms, this approach emphasizes, invariably pay close attention to the administrative dimension, although they are not necessarily constrained by perceptions of existing administrative capacity. This approach often stresses simplicity. (d) The political approach. This approach recognizes both that there is no such thing as the “benevolent dictator” implicitly 12

Lledo, Victor, Aaron Schneider and Mick Moore, “Pro-poor Tax Reform in Latin America: A Critical Survey and Policy Recommendations”, Institute of Development Studies, University of Sussex, March 2003. 13 With special reference to developing countries, see Newbery, David and Nicholas Stern (eds.), The Theory of Taxation for Developing Countries (New York: Published for the World Bank by Oxford University Press, 1987); and Ahmad, Ehtisham and Nicholas Stern, The Theory and Practice of Tax Reform in Developing Countries (Cambridge: Cambridge University Press, 1991). 14 Bird, Richard M., “The Administrative Dimension of Tax Reform”, paper prepared for the World Bank workshop on Practical Issues of Tax Policy in Developing Countries, Washington, D.C., 28 April to 1 May 2003. 6. Hettich, Walter and Stanley Winer, “Blueprints and Pathways: The Shifting Foundations of Tax Reform”, 38 National Tax Journal 423-446 (December 1985).

15

assumed in the public economics approach and that taxation always reflects the clash and interplay of interests (and ideas) within an evolving institutional context. Some who take this approach see taxation as an integral part of the social contract – essentially a quite different normative approach (often called “fiscal exchange”) in which we, the people, agree (usually implicitly) to be taxed in such and such a way provided that you, the government, simultaneously agree to provide such and such a set of services. Others stress instead the role of taxation and especially tax reform as an exercise in political legitimacy, or as an instrument and indicator of state institutional capacity, or as an integral part of the democratization process. This approach often stresses especially the need for both equity (considering both sides of the budget) and transparency and accountability. One need not agree with all these arguments to share Hettich’s and Winer’s generally skeptical conclusion that, since none of these approaches satisfactorily factors in either administrative or political considerations, none is likely to offer much useful advice to governments that are attempting to reform real tax systems in real countries in real time. If theory seems to have disappointingly little to offer to those charged with instituting and implementing sustainable tax reform, where else can we look? Another approach is to begin by noting that the process of tax reform may, at one level, be thought of as being divided into three distinct stages. First, policies are formulated, then they are authorized legally, and finally they are implemented.15. The Indonesian example thus clearly demonstrates that even a low-income country can develop and introduce a major tax reform in a relatively short time, substantially improve the tax system as a result, and sustain these good results for many years. It can be done. But how was it done? Five aspects of the process in Indonesia may be singled out 16.In the first place, and importantly, the tax reform was clearly “owned” by the Indonesians. Although considerable use was made of foreign expertise (hired and paid for by the government itself), the reform was initiated and shaped by a strong Minister of Finance in close collaboration with the planning ministry. The key group of officials involved had substantial experience and continuity before, during, and after the reform period. Secondly, ample time (about two years) was devoted to preparing and evaluating policy options – which were, in condensed form, often presented to (and debated 15 16

See Bird, supra note 5 2004 International Bureau of Fiscal Documentation

16

before) the Minister himself – and then, importantly, to drafting the necessary legislation to implement the options selected. Thirdly, most revenue sources (except tariffs) were included in the reform, and attention was paid not only to tax structure issues but also to tax administration and compliance issues. The reform was intended to be, and largely was, unusually comprehensive in scope. Fourthly, it was originally intended that the entire reform would be presented as a package at the same time, in the expectation that this would be politically more acceptable than a series of measures over an extended period of time. As it turned out, however, the major components of the reform – income, sales and property tax reforms – were in fact introduced at different times. Finally, considerable investment was made both in training tax officials to run the new system and in upgrading the information component of tax administration. In an interesting paper on tax reform in Indonesia (and some other countries), Arnold Harberger concluded that the major lessons to be learned from such experiences was simply the need “... to pound home incessantly the importance of things we knew about all along: (a) clarity of conception in designing a reform, (b) professional-level attention to detail in converting that conception into laws, regulations, and procedures, and (c) administrative machinery for implementing the reform efficiently, fairly, and above all in the long run.” The Indonesian reform clearly scored very well on the first two of Harberger’s points – careful design and attention to detail. Despite the attention paid to the administrative issue, however, one of the major problems encountered with the tax reform turned out to be the general lack of support and enthusiasm for reform from the tax administration. It is very hard to put a new system in place if those who are supposed to make it work have no interest or incentive to do so. Indonesia offers yet another instance demonstrating that close attention must always be paid to the administrative dimension of tax reform.

17

COPS: THE DRIVERS OF PUBLIC SECTOR REFORM COPS drive public sector reform: context, ownership, purpose and strategy (Table 3). The principal driver of public sector reform in Ethiopia was the context of a new government using decentralization to improve service delivery as well as promote its legitimacy; contextual preconditions for reform were largely in place as decentralization defined the need, the government’s homegrown Civil Service Reform Program specified the help required, and urgency was paramount.17 Successful public sector reform is rare in Africa. Over twelve years, Ethiopia transformed its public financial management to international standards and now has the third best system in Africa that is managing the largest aid flows to the continent.18 The challenge for technical assistance was to craft a strategy that built upon these preconditions and established trust. Trust of external technical assistance was especially important, as the government was new and rightfully leery, if not suspicious, of foreigners in sensitive government agencies. Context explains the broad canvas of the reform and the specifics of reform of each financial system (e.g. how overnight devolution to districts made the budget planning reform extremely urgent for regions). Ownership of reform is often given lip service and confused with commitment, but is rarely unpacked in terms of the share holders and their shares (e.g. wholly or minority owned by government, donor/lenders, contractors) and the agents involved in the reform (the saints who support reform, the demons that don’t, and the wizards who design it and help the saints implement reform) (Peterson, 1998). The purpose of reform is often assumed, indeed not critically scrutinized, and not used to define a vision of where it is all going. Unfortunately, in Africa the vision of public financial management reform is one of attempting the summits of international best practice, rather than consolidating the basics of a firm financial plateau. The success of the Ethiopian reform was in large measure due to an appropriate strategy of reform focused on recognition and improvement of existing systems and judicious change.

17

Public sector reform requires four necessary, though not sufficient conditions: trust, need, help and urgency Peterson, 1996a. 18 The best public financial management in Africa is ranked as follows: first--South Africa (population 49.3 million); second--Mauritius (population 1.3 million); and, third—Ethiopia (population 85.2 million) (World Bank, 2010). In an assessment of the evolution of the public financial management systems of sixteen African countries from 2001 to 2007, which covers the core period of PFM reform in Ethiopia, only Ethiopia scored a nineteen percent improvement. Other countries that have been touted as the success stories in Africa scored lower: Burkino Faso (8.3%), Mozambique (4.8% with 9.5% mid-period decline), and Tanzania the oft-touted poster child (12.5%).

18

Table 3 The Drivers of Public Sector Reform Context

Macro-level: political, social, economic Mid-level: administrative structure, bureaucratic culture, legacy procedures Micro-level: necessary conditions--trust, need, help, urgency

Ownership

Array of shareholders Agents of reform: saints, demons, wizards

Purpose

Policy driven: domestic versus foreign Technique driven: plateaus versus summits

Strategy

Reform processes: recognize, improve, change, sustain

Sources of Stephen Peterson (2011) Reforming Public Financial Management in Africa CONTEXT Virtually every account of public financial management reform begins and ends with the observation that public financial management reform is contextual. Context has three dimensions: macro-level (political, social, economic), mid-level (administrative structure, bureaucratic culture, legacy procedures), and micro-level (necessary but not sufficient conditions for reform—trust, need, help and urgency). In the case of Somaliland, In December 2013, the government was approved Somaliland’s PFM Reform Strategy19 which was designed the government of Somaliland Political Stakeholders and Development Partners (DPs); while in Ethiopia; for example the Civil Service Reform Program CSRP20 was quietly developed within the Prime Minister’s Office with no input from foreign aid agencies. The macro social context reinforced the political context (Peterson, 2011). So the macro political context for the public financial management reform was not just firmly in place, there was no higher priority because it was core to the political strategy of control and in turn, the government policy of decentralization. The second level of context, mid-level, is comprised of the administrative structure, the bureaucratic culture, and the legacy administrative procedures. For instance the case of 19

Somaliland PFM Reform Strategy is made up of five pillars: Economic and Budget Management; Resource Mobilization; Financial Management and Reporting; Public Procurement. 20 The Civil Service Reform Program (CSRP) had five sub-programs: Top Systems Management, Human Resources, Service Delivery, Ethics and Expenditure Management and Control (EMCP) (Task Force for Civil Service Reform, 1996).

19

Somaliland management’s role in the budget and expenditure management is minimal. Financial management responsibilities are centralized and shared between the department of finance and administration, the accountant general and the budget department. Directors and perhaps the heads of work units may have the opportunity to request additional resources in preparation for the new budget year (i.e. based on an incremental approach as opposed to zero-based budgeting) or to address in-year pressures, but all decisions are ultimately centralized; but in Ethiopia during the reform the administrative structure includes the tiers of government and the structures within tiers (federal, region, administrative area, zone, wereda and kabelle) and the structures within tiers (e.g. the organization of the Ministry of Finance). During the public financial management reform the structure of government and administration underwent dramatic change. Overnight in 2002, the government delegated financial responsibility over the bulk of regional resources to weredas. Also during the reform, the separate organizations of finance and planning were merged. Into these dramatic changes the public financial management reform was introduced and how this turbulence promoted or hindered reform was the key contextual dynamic shaping the reform (Peterson, 2011). The third or micro-level of context is whether there are the necessary, though not sufficient conditions in place for a reform program and supporting technical assistance to succeed. The necessary conditions of reform are trust, need, help, and urgency (Peterson, 1996a). Trust was secured from the start in large part, because it was a government not foreign aid designed reform. In Somaliland the purpose of the terms of reference is to set out DFIDs requirements for a Supplier to implement the project,21 thus only the team of the international experts and Somaliland public finance management coordinator set up the design to implement a work plan that will support the achievement of sustainable revenue increases in the short term, as well as laying the policy, procedural and administrative foundation for devolution to a semi-autonomous revenue authority in the near future, but establishing a high-performing semi-autonomous revenue agency will require superior capacities in human resources management which at this time is not available in the Ministry of Finance or the government of Somaliland, although there may be pockets of more modern human resources management practices scattered across the public service including myself who registered as an employee in the ministry of education and 21

The DFID Somalia Operational Plan of Strengthening Revenue Policy and Administration In Somaliland: 20142016

20

high studies with undefined position; but in the case of Ethiopia the government ensured its trust in the design of the reform by dividing and managing the foreign technical assistance it requested to implement its reform. While there was trust in the design of the reform, the various technical assistance projects had to build trust at the operational, and at times, strategic level. Trust is built through a momentum of success and small wins achieved with close partnership of government staff. That was the formula for success of the Decentralization Support Activity project.22 OWNERSHIP Ownership can be viewed in terms of who are the shareholders (i.e. government, donor/lenders, contractors, and others—civil society) and how many shares they hold (i.e. majority, minority). Owners in turn can be further understood in terms of the roles they play in reform: saints (government staffs who lead and protect the reform), demons (any actor that actively obstructs or passively fails to support the reform), and wizards23 (the technical and managerial resource persons assisting the saints in implementing the reform) (Peterson, 1998). In case of Somaliland; Department for International Development (DFID) has partner with the government of Somaliland to support Pillar 2: Resource Mobilization and will be implemented by Adam Smith International. Progress against the public finance management reform plan is overseen by the Joint public finance management reform steering committee, chaired by the Minister of Finance and comprised of both government officials and development partners. The reforms will be coordinated by the public finance management reform unit, which has been recently established within the Ministry of Finance; but ownership is a fundamental concept in public sector reform in general and in public finance management reform specifically. The management of public money goes to the heart of sovereignty and it is not appropriate for foreign aid to demand, much less manage, the change of public finance management systems. Government can request support, which is very different from foreign aid ‘inviting them selves’ which is often the case and indeed is often part of the first paragraph of their mission reports. Beyond the critical issue 22

The DSA project implemented five components under the EMCP: budget planning, budgeting, accounting, cash management, and information systems. 23 Wizards are not an undifferentiated good. Wizards can be classified as appropriate (they have relevant and lengthy experience and judgment), inappropriate (they lack the qualities of the appropriate wizards) and imposters. A common fault of wizards is that they tend to view a new assignment as a continuation of the previous country they worked in and tend not to adequately see the new context. They also tend to take the easy path of installing a cookie cutter approach of technique and sequencing and pay little attention to context.

21

of impinging on the sovereignty of African states, the core problem of public finance management on the continent is execution not systems (Linert, 2002; Stevens, 2005). Foreign aid largely concerns itself with the establishment of new systems and the dilemmas of sequencing techniques. The execution task tends to be neglected in large part because it is fundamentally an issue of improving government management of staffs and organizations and thus outside the purview of foreign aid. The decision by foreign aid agencies to use government financial systems for the delivery of loans and grants should be binary with appropriate benchmarks as necessary. In the case of Ethiopia the political context confirmed sole government ownership over this reform. The Prime Minister personally chaired the Civil Service Reform Program (CSRP) from its inception and was quietly developed within the Prime Minister’s Office with no input from foreign aid agencies and with only one expatriate advisor24 (Peterson, 2011). PURPOSE Ownership is the defining variable of a financial reform and it connotes property physical and intellectual. Public financial management is a constituent part of sovereignty, which defines the state as property. To capture the qualities of property and ownership it is useful to think about a public financial management system as a plateau. A well-established public financial management plateau has an array of the basic systems of financial control (e.g. budgets, accounts, and audit) that are adequately executed and sustained with government resources (funding, staffing, and institutions). A plateau should be the wholly owned property of a government and it extends beyond public financial management and encompasses the structure of government and administration. Successful public financial management reform is aligned with the needs of government and fits within these structures (Peterson, 2011). But in the case of Somaliland Adam Smith International deployed an international public financial management adviser stated objective to increase domestic revenue, as well as establish a RA in the near future, a number of recommendations for activities within each Output are summarized below (Table 4), thus only by formality they interview the intellectuals of Somaliland who deeply committed to revenue reform issues and has extensive knowledge of public financial management include myself; but until Somaliland government can execute the basics of financial 24

The lone expatriate advisor, Peter Silkin, was directly contracted by the government who learned about him through its own connections established during the civil war. An auditor from the National Audit Office in the U.K., the CSRP was Peter’s first foray in providing technical assistance to a foreign government. The focus on control in the design of the EMCP was clearly influenced by Peter’s audit background.

22

control, it is not prudent for them to attempt to scale the higher elevations of financial technique. Even if Somaliland government attempt and reach these higher summits with intensive support (oxygen?) of their foreign guides, the path is risky and unsustainable. Table 4 Recommendation by Output Output 1: SLRA Roadmap

2: Tax Policy Reform

3: Tax and Customs Administration Reform

Recommendations  Review and recommendations provided on the revenue legislation  Consideration and decision concerning the policy choices for the governance structure of the RA  Consideration and decision concerning the policy choices for the transition of staff to the RA  Consideration and decision concerning the policy choices regarding the financing of the RA  Organization of stakeholder consultations prior to and during the transition to the RA  Develop revenue data analysis function and capacity  Develop revenue forecasting function and capacity  Develop a Tax Policy Unit  Develop a mechanism to include off-budget revenue  Undertake an analysis of exemptions and account for the ‘fiscal loss’  Establish a tax policy unit at the Ministry of Finance  Consider future tax policy options  Consolidate taxpayer registration (to expand tax base) with the introduction of a fully operational »TIN system, becoming a part of the single window initiative for taxpayer registration and integrated with ASYCUDA  Implement a fully functional SLRA domestic tax administration structure  Create and implement a Medium Taxpayer Office and a Small & Micro Taxpayer Office within the Inland Revenue Department  Re-engineer processes and procedures in preparation for the implementation of an Integrated Tax »Software System (ITAS), reducing the number of steps to encourage taxpayer compliance  Acquire ITAS , customize and implement  Implement a fully developed risk based tax audit system  Implement a robust tax debt recovery system

23

Management and organizational Reform

 Implement electronic fling and electronic payment, expanding ITAS  Introduce VAT law to replace the current sales tax  Implement VAT law supported by taxpayer education programme  Build a robust taxpayer education programme  Identify and document all roles and responsibilities in order to identify gaps  Design and embed code of conduct  Re-engineer processes and procedures in preparation for the implementation of ASYCUDA  Acquire ASYCUDA , customize and implement  Develop a fully operational risk analysis system for selective controls and post clearance audit  Link ASYCUDA with ITAS and share information  Review enforcement role (sanctions, right to prosecute etc.)  Make full preparation for membership of World Customs Organization (WCO), assuming Somaliland obtains international recognition  Develop border controls and facilitate trade with neighbouring countries  Implement a fully reliable valuation system  Develop a fully operational post clearance audit system  Evaluate the impact of implementing the successive reforms on revenue  Build capacity to effectively implement the above reform initiatives.  Undertake functional review of the Department of Inland Revenue and the Department of Customs  Develop Revenue Authority Preparation Plan (RAPP) in preparation for Revenue Bill promulgation  Develop Workforce Capacity Development Plan  Establish Revenue Authority Preparedness Unit (RAPU) as part of the RAPP to anchor human »resources management, information technology, communication, change management and other corporate functions of importance to the establishment of the RA  Carry out human resources reform as part of the RAPP to begin with putting in place manpower »control (i.e. linking costs with employees), organizational design practices (e.g. clarify mandates, functions), position description writing capacities and position administration systems to account for any structural, position and employee changes  Assist the PFM Office to develop a Corporate Governance 24

Guide for the new RA with information »on Board operations, member performance expectations, a Revenue Authority overview, Code of Conduct/Conflict of Interest policies, Board Sub-Committees and other matters important to the proper functioning of the Board.  Develop simple Programme Management Guide to monitor key operational objective performance  Produce a Change Management Plan to support programme activities  Conduct one-on-one change management training with MoF members in the Revenue Reform »Working Group and two Regional Coordinators and in groups with Customs and Inland Revenue Department managers Sources of 6879: Inception Report.

The plateau metaphor illustrates one of the principal causes of failure of public sector and public financial management reforms in Africa—the focus on summits of sophisticated techniques (international best practice) rather than improving the bedrock of plateaus—basic systems and their execution (appropriate to locale). Many public financial management reforms in Africa have been akin to the attempts of climbers to summit high altitude peaks. Few make it, those that do don’t stay long, and most fatalities occur in the descent25. In recent years, financial summits in Africa have included several techniques that governments and their technical advisors have slipped on—Medium Term Expenditure Frameworks, performance/program budgeting, Integrated Financial Management Information Systems, accrual accounting, and business process reengineering to name just a few. The summit strategy and the techniques that underpin it are inappropriate for most if not all African governments. Financial summiting is costly, risky and not needed. An appropriate strategy for guiding African governments that promotes ownership, and establishes good financial control is to build solid financial plateaus. Only once a government has a financial plateau that is stable and sustainable, should it consider building a higher (more sophisticated) plateau. Governments should make such a decision alone and it should be justified

25

Krakauer (1997). Unlike climbers that make mistakes and perish on Himalayan peaks, technical assistance advisors and their firms often live another day and another contract. There is a serious issue of accountability in this profession with two year wonders (contracts) which mean technical assistance leaves before it is held accountable for results and negative results rarely if ever follow them.

25

in terms of improving outcomes—does it promote growth and social wellbeing.26There is considerable evidence from developed and developing country experience that advanced public financial techniques, assuming they can be made to work, do not improve the outcomes of public expenditure (Mellet et al., 2009). Financial summits are attractive to donor/lenders for several reasons: they are a source of benchmarks for conditionality, many involve significant commodity dumps (e.g. Integrated Financial Management Systems) which allow rapid disbursement of funds, they facilitate the placement of technical assistance personnel within core government functions who can monitor their funding and financial risk, they extend their influence over government, and they support the business model of technical assistance departments of foreign aid agencies. Governments are attracted to summits in part, because they are told to do so as part of conditionality and the need to be modern with international best practice. But a more telling reason for government support of summits is that they often generate handsome rents—especially in large-scale financial automation projects. Plateaus are built on domestic not foreign resources, which limit the opportunities for rents. Plateaus do place a demand on the domestic budget, but that is good, because supporting systems are principally about better management of government staff and organizations, which are existing (and modest) budget commitments. Financial summits also provide cash cows for contractors because implementing projects with a complex scope within government bureaucracies that have limited capacity to manage even simple contracts makes it highly likely that project schedule and budget will be extended (Peterson, 2011). Plateau as a Perspective on public financial management Plateaus provide a nuanced perspective of a country’s PFM. As territory, it stresses the sovereignty of public financial management, which is a core function of the state. As a point of departure or destination, it stresses the need to recognize what exists and the need to establish the basics of financial control. As geology, it reminds one of the difficulty of changing administrative structures and the virtue of the stability of those structures. As landscape, it presents how the parts relate to the whole (financial systems and their execution as well public 26

Beyond the risk, cost and dubious return, governments, and hopefully their foreign aid partners need to recognize that it takes a long time to implement a PFM reform—SIDA estimates it takes twelve to fifteen years (SIDA, 2007).

26

financial management and the broader structure of government and administration). As place, it reminds one that public financial management is a cultural, linguistic, and historical artifact27. As environmental quality, it brings focus to the quality (output) of PFM—sustainability, sectoral allocation, composition of expenditure, and risk. As elevation, it shows how decentralized systems have different plateaus at varying elevations (capacity and sophistication) and the challenge of maintaining communication with diverse terrain. As layers, it brings focus to what is bedrock and what are the fault lines. Finally, a plateau perspective illustrates the virtues of a public financial management reform sequence of recognize-improve sustain, rather than defining reform as only change. Reform as building a plateau not a summit accords with what Karl Weick calls the ‘small wins’ approach to successful implementation of public policy. A series of small wins is also more structurally sound than a large win because small wins are stable building blocks….A small win is a concrete, complete, implemented outcome of moderate importance. By itself, one small win may seem unimportant. A series of small wins at small but significant tasks, however, reveals a pattern that may attract allies, deter opponents, and lower resistance to subsequent proposals. Small wins are controllable opportunities that produce visible results (Weick, 1984, p. 43). Small wins also reduce the risk of reform. Small wins are like short stacks. They preserve gains, they cannot unravel, each one requires less coordination to execute, interruptions such as might occur when there is a change in political administration have limited effects, and subparts can be assembled into different configurations [they can be adapted to different plateaus] (Weick, 1984, p. 44). Building public financial management plateaus does not imply that a country’s public financial management has ‘plateaued’ meaning that further reform is not possible or even needed. The plateau metaphor reinforces the oft-heard recommendation in the public financial management field of the need to first ‘get the basics right’, which means having a firm and coherent base—a plateau.

27

The first offering of the Harvard Executive Program in Public Financial Management was held at Egerton College Kenya in 1987 and Harris Mule, then Permanent Secretary of the Ministry of Finance of the government of Kenya, opined that budgeting is a language and a ‘fragile tradition’ that one should take care in changing.

27

Plateaus and Platforms Given the increasing use of the term ‘platform’ in public financial management, it is important to distinguish it from the concept of a plateau. Public financial management can be viewed as having three platforms: transaction, policy and legal (Table 5). The transaction platform includes the systems of financial administration—revenue, budgeting, accounting, cash management and information systems and is a recognition that public financial management systems are fundamentally about managing transactions. The policy platform includes fiscal management (macroeconomic framework, fiscal architecture) as well as the outputs and outcome of the transactions (performance and program budgeting). The legal oversight platform includes the legal framework (constitution, organic budget law, financial regulations) and external audit and the legislative oversight. Effective public financial management systems require that all three platforms perform well. The sequence of public financial management reform is the array of reform to the components within and between these three platforms. A later and very different formulation of the platform concept emerges from the literature on policy-based foreign aid and the conditionality’s needed to reduce fiduciary risk. Peter Brooke notes that the value of the platform approach is ‘in providing greater clarity to both the Governments and donors about the rules of engagement and disbursement and what each expects of the other in the partnership.28 The Brooke platform approach views public financial management reform as driven by the real or perceived needs of foreign aid. Our platform approach is generic and locates reform within the drivers of the reform (COPS) with sequencing emerging from the strategy of reform (the processes of recognize, improve, change and sustain) discussed next.

28

Brooke (2003, p. 2). See also Department for International Development (2005).

28

Table 529: The Platforms of Public Financial Management Policy Platform Macroeconomic framework Fiscal architecture Performance/program budgets Regulation

Transaction Platform Revenue administration Budget Accounts Cash management Input management Internal audit Financial Information Systems Legal Platform Legal framework External audit Legislative oversight

Sources of Stephen Peterson (2011) Reforming Public Financial Management in Africa

29

The three platforms of PFM were introduced in the Executive Program in Public Financial Management at the Kennedy School of Government in summer 2002. I am indebted to Perran Penrose for his inputs into this framework.

29

STRATEGY There are four tasks in public sector reform: recognize, improve, change and sustain. The Ethiopian pathway of reform was a sequence of recognize-improve-sustain and only if absolutely necessary, change. It was a pathway of building a plateau. This reform sequence was appropriate given the priority of supporting rapid decentralization (Peterson, 2011). One can argue, that the principal problem with public sector reform in general and public financial management reform specifically in Africa, such as my country (Somaliland) is defining reform as change, often perpetual, and ignoring the other tasks of reform. Recognize Recognizing meaning understanding and respecting what exists is the first and often most neglected step in reform. Recognizing also means not dismissing, indeed denigrating, ongoing reforms to impose one’s own.30 All too often, governments in developing countries do not understand the strengths of their systems and are too quick to change them, often on the advice of others. The practice of a cursory review of existing systems, if at all, before leaping to major reform in public financial management in Africa is akin to the reengineering movement of the 1990s which advocated only a brief look at business processes on the grounds that they were going to be replaced (Hammer, 1995, p. 19). Taking the Linert/Stevens argument that public financial management systems in Africa are robust but not executed, one must thoroughly understand them to have confidence that they are indeed robust and if so, what if anything, should be done31. Adequate description is especially critical in countries that are decentralized because they often have multiple systems (plateaus) changing on different timetables. While one can find detailed descriptions of the parts of a country’s public financial management system (e.g. accounting), it is rare to find comprehensive description of a country’s public financial management he nuances across its plateaus thus In countries that are decentralized, or even in countries that are centralized but have extremes in capacity, there is not ‘one’ public financial management system and public financial 30

A problem with PFM reforms in Africa is having too many cooks in the kitchen.

31

An early diagnostic of Ethiopia’s PFM system was conducted by Peterson through the European Commission funded Social Expenditure Review conducted by Oxford University in 1996 (Peterson, 1996b). See also Peterson (1996c).

30

management reformers with a focus on technique, often fail to appreciate these differences that do matter. In Ethiopia for example, there was considerable confusion by foreign aid over the impact of their grants and loans on the offset of the fiscal transfer from federal to regional governments. For years, this confusion rippled throughout public financial management (e.g. medium term planning, budgeting, and accounting). Decentralization poses considerable complexity to the delivery of foreign aid and donor/lenders need to particularly recognize the nuances and the impact of their funding. Again in order to consolidate recent achievements in the area of revenue administration, in 2011 the government of Somaliland (GoSL) made a strategic decision to set up an autonomous Somaliland Revenue Authority (SLRA). Since then, a substantial number of background documents setting the stage for the formation of the SLRA have been produced with donor support. The most prominent of these are the Somaliland public financial management reform strategy prepared by the Ministry of Finance, and the “Organization Structure and Roadmap for Phased Implementation” presented by the East African School of Taxation in 2011. Many other documents and papers have also been produced, but little progress has been made in terms of implementation. The recognition task of reform is significant for it focuses one on the definition of the problem rather than the leap to the solution. Unfortunately, public financial management reforms is much about technique and sequence and little about hard questions—why change and how does it improve outcomes. Again, the strategy of small wins makes sense. The strategy of small wins addresses social problems by working directly on their construction and indirectly on their resolution…. A shift of attention away from outcomes toward inputs is not trivial, because the content of appropriate solutions is often implied by the definition of what needs to be solved. To focus on the process of problem definition is to incorporate a more substantial portion of psychology, specifically, its understanding of processes of appraisal, social construction of reality, problem finding, and definition of the situation (Weick, 1984, p. 40). Richard Allen, echoes this need for better recognition of the problem before leaping to the toolbox of techniques by noting that IMF finance advisors ‘are too technical (Allen, 2010)’ and that they need to ‘fumble around in the dark (Allen, 2009, p. 25).’ Overlooking the recognition task of reform is promoted by the dominant public financial management diagnostic of the day, the Public Expenditure Financial Accountability Framework (PEFA), which is used to assess a 31

country’s public financial management (PEFA, 2005). A comprehensive assessment of a country’s public financial management should cover the inputs (systems, execution), throughput (performance of process), and outputs (the quality and impact of expenditure). The PEFA examines only throughput and as such, principally functions like a temperature gauge in a vehicle telling the driver whether the engine is overheating. Taking the analogy further, a PEFA does not tell the driver where the car is going, or the capabilities of the engine and transmission, or the maintenance schedule. Perhaps the principal weakness of a PEFA is that is does not allow for a cursory inspection under the hood (examining the inputs of public financial management systems and the elements of execution) to see the cause of overheating. 32 At best, the PEFA is a very partial recognition of a country’s public financial management system. By focusing on throughput, and thus funds flow, the PEFA diagnostic is of particular interest to donor/lenders and their concern for the flow of their funds, while most governments tend to be focused on outputs if only to promote their legitimacy (and control) through the delivery of services. A promising new public financial management diagnostic is the Country Integrated Fiduciary Assessment (CIFA), which examines all three attributes of a public financial management system (inputs, throughputs and outputs). The bottom line of a CIFA unlike a PEFA is that it presents an action plan for public financial management reform going forward which is meant to address the risks of a country’s public financial management, which is of importance to governments and foreign aid alike. Improve Improving what exists is the essence of a small win. Improving also fits with the reality found in most African governments of systems being robust but not adequately executed. Toning up what exists and focusing on strengthening execution (i.e. in-service training, schemes of service, organization of finance functions) can have significant payoffs at modest cost. Reforms that focus on improving rather than changing, are faster, cheaper, less risky, and are less disruptive of daily operations. Most important, reform as improvement ensures that government remains in the driver’s seat of management and operation of public financial management (Peterson, 2011).

32

Since PEFAs don’t look at the inputs and their link to outputs, they cannot specify an action plan because they do not address causality. A further deficiency of the PEFA is that it does not cover decentralized systems although there is an initiative to elaborate guidelines for multitier assessments.

32

In the case of the government of Somaliland expected to complete revision of revenue related laws, Ministerial and Presidential decrees, regulations and other tax policy documents is required, in addition to that the main challenge to developing the capacity of the Ministry of Finance for revenue forecasting is lack of accurate data; so the government of Somaliland needs to be able to anticipate the level of revenues that will be available in order to finance their budget expenditures. A number of factors need to be carefully taken into account when considering revenue forecasting. One of the basic pre-requisites is to have the necessary revenue data (historical/actual tax collection information). Other important factors include defining the methodology that will be used for calculating future revenues, and also the underlying general and more tax-specific assumptions that will be used for the forecasts. Also the ministry of finance of Somaliland does not currently have a tax policy unit or a department exclusively dedicated to tax policy; but in most countries there is a tax policy department that deals with the economic aspects of the tax system. These departments generally analyze how tax legislation and amendments affect saving, consumption, investments and labour, and especially, how tax revenues respond when regulations are changed or new taxes are introduced. A tax policy department is also responsible for preparing annual tax forecasts and proposals outlined in the budget; whereas International trade plays an important part in the development of economies. Facilitating trade is therefore a natural concern for policy makers. Currently traders in Hargeisa must fill out and submit six documents to export and seven to import. The government should explore ways to streamline these requirements. Exports from good practice countries require only four documents: bill of lading, commercial invoice, customs declaration and packing list. Fewer documents reduce the hassle and delay for importers and exporters; also publicizing customs procedures, fees and documents would promote transparency and help importers and exporters save time. In addition to that, adoption of electronic systems for filing, transferring, and processing and exchanging customs information has become an important tool for managing flows of information. It has been one of the most popular trade reforms in recent years. So it is difficult to make businesses pay taxes if they are used to avoiding them. Communications efforts can help raise awareness and diffuse potential resistance. Somaliland can learn from other post conflict economies that have successfully improved compliance through communications; for instances the government must held business forums to gather feed-back on improving the tax system, and they communicated directly with businesses to encourage tax compliance. Often the 33

biggest obstacles to paying taxes for small businesses are the lack of basic accounting skills as well as the gap between the tax law and its interpretation in practice. The government can fill this gap by providing capacity-building and training for small businesses. The benefits to tax training are mutual: if well-trained entrepreneurs are able to file returns and pay taxes more efficiently, the government can improve compliance with the tax regulation. Change Change should be done judiciously and justified in terms of improving the quality of public financial management outputs (e.g. sectoral allocation, composition of expenditure). Returning to Everest, the key to surviving the climb, whether one summits or not, is the establishment and adherence to ironclad decision rules before the climb (Krakauer, 1997, pp. 284-288). On Everest, while summiting may be an option, survival should not be. Most fatalities occur on the descent, often by those who successfully reached the summit but did not follow the decision rules of when to turn back.33 As with mountaineering, the significantly higher risk of changing rather than improving public financial management reform requires clear rules to manage risk and avoid failure. Sustain Sustaining is the orphan of reform. As noted above, it lacks constituencies. Sustaining a reform, however, is the key to effective execution of systems, and that is the weak link in public financial management reform in Africa. Sustaining is the ‘operating and maintenance (o&m)’ of reform. Governments the world over underfunds o&m and gives priority to new expenditure (capital) and required expenditure (recurrent wage and recurrent statutory). O&M is discretionary. The dearth of resources for sustaining existing public financial management systems is a reflection of the broader public financial management problem—poor composition of expenditure.34 Developing a strategy for sustaining a reform, especially for in-service training, should not be an after-thought or a last minute agenda as a technical assistance project is closing down, much less, the glimmer for a follow-on technical assistance project. From the start of the Ethiopian 33

The failure to follow the rules of the climb is often the result of ‘summit fever’ and severe degradation of judgment due to oxygen deprivation. I shall leave it to the reader to decide the degree and causes of impairment of judgment, which many public financial management reforms on the continent reflect. 34

The composition of expenditure refers to the respective balance of expenditure (recurrent wage, recurrent o&m, recurrent statutory and capital) for a particular activity.

34

reform, the project defined the key constraint as training and indeed one of the first initiatives was to establish a training infrastructure at the federal and regional levels that could prepare finance officials for the reform and train future staffs to sustain it Peterson (1997). The virtue of a small wins plateau approach to public financial management reform is that sustainability is developed all along. Finally a brief example of a PFM reform that has not worked in Somaliland can be explained in 2011 the government of Somaliland made a strategic decision to set up an autonomous Somaliland Revenue Authority (SLRA). Since then, a substantial number of background documents setting the stage for the formation of the SLRA have been produced with donor support prepared by the Ministry of Finance, and the “Organization Structure and Roadmap for Phased Implementation” presented by the East African School of Taxation in 2011. Many other documents and papers have also been produced, but little progress has been made in terms of implementation. The lack of alignment with the drivers of public sector reform explains the failure of this initiative. Whereas the present inception report failed to respect the understanding realities on the ground and has lost its way, because it has been much about summits that has not worked and has caused considerable harm and seems to be a school of thought and unfortunately practice because summits can be done quickly and risk can be managed; but the most important alignment of the reform was building financial plateaus at the lowest and weakest tier of government, rather than summiting from the highest and most capable tier. The Ethiopian experience shows a pathway of reform that built plateaus and managed risk. It also reaffirms that there are no shortcuts to progress in Africa (or elsewhere) (Hyden, 1983). The government of Somaliland must remain in the driver’s seat of management and operation of public financial management reform and then to put in place the infrastructure needed to support existing systems (e.g. in-service training facilities and programs) which could then be used later to support new systems that recommended strengthening revenue policy and administration. Having the infrastructure in place for supporting public financial management means that these systems can adapt to rapid changes in government policy; Given that public financial management reforms are very long-term affairs, they must be designed for the inevitable and often disruptive political and administrative changes. It is far easier to weather a violent storm from a firm plateau, than from an exposed summit.

35

REFERENCES Accountable Public Governance. Shah A (ed.). The World Bank: Washington, D.C. African Economies, Oxford University. AIDCO, European Commission, Brussels. Budgeting, Budget Institutions, and Budget Reforms for a More Responsive and Child Policy Forum: Addis Ababa. Doing Business in Hargeisa at www. Doing business.org/sub national Faculty Research Working Paper Series GGACC - UNDP Somaliland National Corruption Perception Survey Pp23 Ministry of Finance, Federal Republic of Ethiopia by the Centre for the Study of of Ethiopia. In Ethiopia Social Sector Review (PER II). Stephen Peterson (2011) Reforming Public Financial Management in Africa IMF Working Paper 03/2. Washington, D.C. Hyden G. 1983. No Shortcuts to Progress: African Development Management in Perspective. Heinemann: London. Weick K. 1984. Small wins redefining the scale of social problems. American Psychologist Peterson S. 1996a. Making IT work: implementing effective financial information systems in bureaucracies of developing countries. In Information Technology and Innovation in Tax Administration. Jenkins G (ed.). Kluwer Law International: The Hague. Peterson S. 1996b. Generic issues in financial management for the central government Peterson S. 1996c. Improving financial management in the Government of Ethiopia. Peterson S. 1997. The training strategy for accounts, budget and financial planning under civil service reform. Paper prepared for the Prime Minister’s Office and the Ministry of Finance, Federal Republic of Ethiopia, Addis Ababa. Peterson S. 1998. Saints, demons, wizards and systems: why information technology 36

Cohen J, Peterson S. 1999. Administrative Decentralization: Strategies for Developing Countries. Kumarian: West Hartford. CT. Linert I. 2002. Comparison between two public expenditure management systems in Brooke, Peter. 2003. Study of Measures used to Address Weaknesses in Public Financial Management Systems in the Context of Policy Based Support. Report prepared for PEFA. Harvard University. 2004. In-Service Training Program in Financial Management. Proposal United States Agency for International Development, Mission to Ethiopia, Addis Ababa PEFA. 2005. Public Financial Management Performance Measurement Framework. PEFA Secretariat, World Bank. Washington, D.C. Peterson S. 2007. Automating public financial management in developing countries; In Allen R. 2009. The challenge of reforming budgetary institutions in developing countries IMF Working Paper 96 Mellet H, Marriot N, Macniven L. 2009. Devolution, accruals accounting and asset management in NHS Wales. Financial Accountability and Management. 25: 435- 450. Quist R. 2009. On the PEFA reform sequence model framework. Paper submitted to Hedger E, de Renzio P. 2010 what do public financial management assessments tell us about PFM reform? ODI Background Note Peterson S. 2010. Rethinking the Millennium Development Goals for Africa. African ILO Labour Force Survey 2012. 6879: Somaliland – Strengthening Revenue Policy and Administration- DRAFT Inception Report

37